diff options
Diffstat (limited to 'main/linux-grsec/grsecurity-2.2.2-2.6.38.3-201104171745.patch')
| -rw-r--r-- | main/linux-grsec/grsecurity-2.2.2-2.6.38.3-201104171745.patch | 60737 |
1 files changed, 60737 insertions, 0 deletions
diff --git a/main/linux-grsec/grsecurity-2.2.2-2.6.38.3-201104171745.patch b/main/linux-grsec/grsecurity-2.2.2-2.6.38.3-201104171745.patch new file mode 100644 index 0000000000..98ae043e80 --- /dev/null +++ b/main/linux-grsec/grsecurity-2.2.2-2.6.38.3-201104171745.patch @@ -0,0 +1,60737 @@ +diff -urNp linux-2.6.38.3/arch/alpha/include/asm/dma-mapping.h linux-2.6.38.3/arch/alpha/include/asm/dma-mapping.h +--- linux-2.6.38.3/arch/alpha/include/asm/dma-mapping.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/alpha/include/asm/dma-mapping.h 2011-04-17 15:57:32.000000000 -0400 +@@ -3,9 +3,9 @@ + + #include <linux/dma-attrs.h> + +-extern struct dma_map_ops *dma_ops; ++extern const struct dma_map_ops *dma_ops; + +-static inline struct dma_map_ops *get_dma_ops(struct device *dev) ++static inline const struct dma_map_ops *get_dma_ops(struct device *dev) + { + return dma_ops; + } +diff -urNp linux-2.6.38.3/arch/alpha/include/asm/elf.h linux-2.6.38.3/arch/alpha/include/asm/elf.h +--- linux-2.6.38.3/arch/alpha/include/asm/elf.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/alpha/include/asm/elf.h 2011-04-17 15:57:32.000000000 -0400 +@@ -90,6 +90,13 @@ typedef elf_fpreg_t elf_fpregset_t[ELF_N + + #define ELF_ET_DYN_BASE (TASK_UNMAPPED_BASE + 0x1000000) + ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE (current->personality & ADDR_LIMIT_32BIT ? 0x10000 : 0x120000000UL) ++ ++#define PAX_DELTA_MMAP_LEN (current->personality & ADDR_LIMIT_32BIT ? 14 : 28) ++#define PAX_DELTA_STACK_LEN (current->personality & ADDR_LIMIT_32BIT ? 14 : 19) ++#endif ++ + /* $0 is set by ld.so to a pointer to a function which might be + registered using atexit. This provides a mean for the dynamic + linker to call DT_FINI functions for shared libraries that have +diff -urNp linux-2.6.38.3/arch/alpha/include/asm/pgtable.h linux-2.6.38.3/arch/alpha/include/asm/pgtable.h +--- linux-2.6.38.3/arch/alpha/include/asm/pgtable.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/alpha/include/asm/pgtable.h 2011-04-17 15:57:32.000000000 -0400 +@@ -101,6 +101,17 @@ struct vm_area_struct; + #define PAGE_SHARED __pgprot(_PAGE_VALID | __ACCESS_BITS) + #define PAGE_COPY __pgprot(_PAGE_VALID | __ACCESS_BITS | _PAGE_FOW) + #define PAGE_READONLY __pgprot(_PAGE_VALID | __ACCESS_BITS | _PAGE_FOW) ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++# define PAGE_SHARED_NOEXEC __pgprot(_PAGE_VALID | __ACCESS_BITS | _PAGE_FOE) ++# define PAGE_COPY_NOEXEC __pgprot(_PAGE_VALID | __ACCESS_BITS | _PAGE_FOW | _PAGE_FOE) ++# define PAGE_READONLY_NOEXEC __pgprot(_PAGE_VALID | __ACCESS_BITS | _PAGE_FOW | _PAGE_FOE) ++#else ++# define PAGE_SHARED_NOEXEC PAGE_SHARED ++# define PAGE_COPY_NOEXEC PAGE_COPY ++# define PAGE_READONLY_NOEXEC PAGE_READONLY ++#endif ++ + #define PAGE_KERNEL __pgprot(_PAGE_VALID | _PAGE_ASM | _PAGE_KRE | _PAGE_KWE) + + #define _PAGE_NORMAL(x) __pgprot(_PAGE_VALID | __ACCESS_BITS | (x)) +diff -urNp linux-2.6.38.3/arch/alpha/kernel/module.c linux-2.6.38.3/arch/alpha/kernel/module.c +--- linux-2.6.38.3/arch/alpha/kernel/module.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/alpha/kernel/module.c 2011-04-17 15:57:32.000000000 -0400 +@@ -182,7 +182,7 @@ apply_relocate_add(Elf64_Shdr *sechdrs, + + /* The small sections were sorted to the end of the segment. + The following should definitely cover them. */ +- gp = (u64)me->module_core + me->core_size - 0x8000; ++ gp = (u64)me->module_core_rw + me->core_size_rw - 0x8000; + got = sechdrs[me->arch.gotsecindex].sh_addr; + + for (i = 0; i < n; i++) { +diff -urNp linux-2.6.38.3/arch/alpha/kernel/osf_sys.c linux-2.6.38.3/arch/alpha/kernel/osf_sys.c +--- linux-2.6.38.3/arch/alpha/kernel/osf_sys.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/alpha/kernel/osf_sys.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1162,7 +1162,7 @@ arch_get_unmapped_area_1(unsigned long a + /* At this point: (!vma || addr < vma->vm_end). */ + if (limit - len < addr) + return -ENOMEM; +- if (!vma || addr + len <= vma->vm_start) ++ if (check_heap_stack_gap(vma, addr, len)) + return addr; + addr = vma->vm_end; + vma = vma->vm_next; +@@ -1198,6 +1198,10 @@ arch_get_unmapped_area(struct file *filp + merely specific addresses, but regions of memory -- perhaps + this feature should be incorporated into all ports? */ + ++#ifdef CONFIG_PAX_RANDMMAP ++ if (!(current->mm->pax_flags & MF_PAX_RANDMMAP)) ++#endif ++ + if (addr) { + addr = arch_get_unmapped_area_1 (PAGE_ALIGN(addr), len, limit); + if (addr != (unsigned long) -ENOMEM) +@@ -1205,8 +1209,8 @@ arch_get_unmapped_area(struct file *filp + } + + /* Next, try allocating at TASK_UNMAPPED_BASE. */ +- addr = arch_get_unmapped_area_1 (PAGE_ALIGN(TASK_UNMAPPED_BASE), +- len, limit); ++ addr = arch_get_unmapped_area_1 (PAGE_ALIGN(current->mm->mmap_base), len, limit); ++ + if (addr != (unsigned long) -ENOMEM) + return addr; + +diff -urNp linux-2.6.38.3/arch/alpha/kernel/pci_iommu.c linux-2.6.38.3/arch/alpha/kernel/pci_iommu.c +--- linux-2.6.38.3/arch/alpha/kernel/pci_iommu.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/alpha/kernel/pci_iommu.c 2011-04-17 15:57:32.000000000 -0400 +@@ -950,7 +950,7 @@ static int alpha_pci_set_mask(struct dev + return 0; + } + +-struct dma_map_ops alpha_pci_ops = { ++const struct dma_map_ops alpha_pci_ops = { + .alloc_coherent = alpha_pci_alloc_coherent, + .free_coherent = alpha_pci_free_coherent, + .map_page = alpha_pci_map_page, +@@ -962,5 +962,5 @@ struct dma_map_ops alpha_pci_ops = { + .set_dma_mask = alpha_pci_set_mask, + }; + +-struct dma_map_ops *dma_ops = &alpha_pci_ops; ++const struct dma_map_ops *dma_ops = &alpha_pci_ops; + EXPORT_SYMBOL(dma_ops); +diff -urNp linux-2.6.38.3/arch/alpha/kernel/pci-noop.c linux-2.6.38.3/arch/alpha/kernel/pci-noop.c +--- linux-2.6.38.3/arch/alpha/kernel/pci-noop.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/alpha/kernel/pci-noop.c 2011-04-17 15:57:32.000000000 -0400 +@@ -173,7 +173,7 @@ static int alpha_noop_set_mask(struct de + return 0; + } + +-struct dma_map_ops alpha_noop_ops = { ++const struct dma_map_ops alpha_noop_ops = { + .alloc_coherent = alpha_noop_alloc_coherent, + .free_coherent = alpha_noop_free_coherent, + .map_page = alpha_noop_map_page, +@@ -183,7 +183,7 @@ struct dma_map_ops alpha_noop_ops = { + .set_dma_mask = alpha_noop_set_mask, + }; + +-struct dma_map_ops *dma_ops = &alpha_noop_ops; ++const struct dma_map_ops *dma_ops = &alpha_noop_ops; + EXPORT_SYMBOL(dma_ops); + + void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen) +diff -urNp linux-2.6.38.3/arch/alpha/mm/fault.c linux-2.6.38.3/arch/alpha/mm/fault.c +--- linux-2.6.38.3/arch/alpha/mm/fault.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/alpha/mm/fault.c 2011-04-17 15:57:32.000000000 -0400 +@@ -54,6 +54,124 @@ __load_new_mm_context(struct mm_struct * + __reload_thread(pcb); + } + ++#ifdef CONFIG_PAX_PAGEEXEC ++/* ++ * PaX: decide what to do with offenders (regs->pc = fault address) ++ * ++ * returns 1 when task should be killed ++ * 2 when patched PLT trampoline was detected ++ * 3 when unpatched PLT trampoline was detected ++ */ ++static int pax_handle_fetch_fault(struct pt_regs *regs) ++{ ++ ++#ifdef CONFIG_PAX_EMUPLT ++ int err; ++ ++ do { /* PaX: patched PLT emulation #1 */ ++ unsigned int ldah, ldq, jmp; ++ ++ err = get_user(ldah, (unsigned int *)regs->pc); ++ err |= get_user(ldq, (unsigned int *)(regs->pc+4)); ++ err |= get_user(jmp, (unsigned int *)(regs->pc+8)); ++ ++ if (err) ++ break; ++ ++ if ((ldah & 0xFFFF0000U) == 0x277B0000U && ++ (ldq & 0xFFFF0000U) == 0xA77B0000U && ++ jmp == 0x6BFB0000U) ++ { ++ unsigned long r27, addr; ++ unsigned long addrh = (ldah | 0xFFFFFFFFFFFF0000UL) << 16; ++ unsigned long addrl = ldq | 0xFFFFFFFFFFFF0000UL; ++ ++ addr = regs->r27 + ((addrh ^ 0x80000000UL) + 0x80000000UL) + ((addrl ^ 0x8000UL) + 0x8000UL); ++ err = get_user(r27, (unsigned long *)addr); ++ if (err) ++ break; ++ ++ regs->r27 = r27; ++ regs->pc = r27; ++ return 2; ++ } ++ } while (0); ++ ++ do { /* PaX: patched PLT emulation #2 */ ++ unsigned int ldah, lda, br; ++ ++ err = get_user(ldah, (unsigned int *)regs->pc); ++ err |= get_user(lda, (unsigned int *)(regs->pc+4)); ++ err |= get_user(br, (unsigned int *)(regs->pc+8)); ++ ++ if (err) ++ break; ++ ++ if ((ldah & 0xFFFF0000U) == 0x277B0000U && ++ (lda & 0xFFFF0000U) == 0xA77B0000U && ++ (br & 0xFFE00000U) == 0xC3E00000U) ++ { ++ unsigned long addr = br | 0xFFFFFFFFFFE00000UL; ++ unsigned long addrh = (ldah | 0xFFFFFFFFFFFF0000UL) << 16; ++ unsigned long addrl = lda | 0xFFFFFFFFFFFF0000UL; ++ ++ regs->r27 += ((addrh ^ 0x80000000UL) + 0x80000000UL) + ((addrl ^ 0x8000UL) + 0x8000UL); ++ regs->pc += 12 + (((addr ^ 0x00100000UL) + 0x00100000UL) << 2); ++ return 2; ++ } ++ } while (0); ++ ++ do { /* PaX: unpatched PLT emulation */ ++ unsigned int br; ++ ++ err = get_user(br, (unsigned int *)regs->pc); ++ ++ if (!err && (br & 0xFFE00000U) == 0xC3800000U) { ++ unsigned int br2, ldq, nop, jmp; ++ unsigned long addr = br | 0xFFFFFFFFFFE00000UL, resolver; ++ ++ addr = regs->pc + 4 + (((addr ^ 0x00100000UL) + 0x00100000UL) << 2); ++ err = get_user(br2, (unsigned int *)addr); ++ err |= get_user(ldq, (unsigned int *)(addr+4)); ++ err |= get_user(nop, (unsigned int *)(addr+8)); ++ err |= get_user(jmp, (unsigned int *)(addr+12)); ++ err |= get_user(resolver, (unsigned long *)(addr+16)); ++ ++ if (err) ++ break; ++ ++ if (br2 == 0xC3600000U && ++ ldq == 0xA77B000CU && ++ nop == 0x47FF041FU && ++ jmp == 0x6B7B0000U) ++ { ++ regs->r28 = regs->pc+4; ++ regs->r27 = addr+16; ++ regs->pc = resolver; ++ return 3; ++ } ++ } ++ } while (0); ++#endif ++ ++ return 1; ++} ++ ++void pax_report_insns(void *pc, void *sp) ++{ ++ unsigned long i; ++ ++ printk(KERN_ERR "PAX: bytes at PC: "); ++ for (i = 0; i < 5; i++) { ++ unsigned int c; ++ if (get_user(c, (unsigned int *)pc+i)) ++ printk(KERN_CONT "???????? "); ++ else ++ printk(KERN_CONT "%08x ", c); ++ } ++ printk("\n"); ++} ++#endif + + /* + * This routine handles page faults. It determines the address, +@@ -131,8 +249,29 @@ do_page_fault(unsigned long address, uns + good_area: + si_code = SEGV_ACCERR; + if (cause < 0) { +- if (!(vma->vm_flags & VM_EXEC)) ++ if (!(vma->vm_flags & VM_EXEC)) { ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if (!(mm->pax_flags & MF_PAX_PAGEEXEC) || address != regs->pc) ++ goto bad_area; ++ ++ up_read(&mm->mmap_sem); ++ switch (pax_handle_fetch_fault(regs)) { ++ ++#ifdef CONFIG_PAX_EMUPLT ++ case 2: ++ case 3: ++ return; ++#endif ++ ++ } ++ pax_report_fault(regs, (void *)regs->pc, (void *)rdusp()); ++ do_group_exit(SIGKILL); ++#else + goto bad_area; ++#endif ++ ++ } + } else if (!cause) { + /* Allow reads even for write-only mappings */ + if (!(vma->vm_flags & (VM_READ | VM_WRITE))) +diff -urNp linux-2.6.38.3/arch/arm/include/asm/elf.h linux-2.6.38.3/arch/arm/include/asm/elf.h +--- linux-2.6.38.3/arch/arm/include/asm/elf.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/arm/include/asm/elf.h 2011-04-17 15:57:32.000000000 -0400 +@@ -115,7 +115,14 @@ int dump_task_regs(struct task_struct *t + the loader. We need to make sure that it is out of the way of the program + that it will "exec", and that there is sufficient room for the brk. */ + +-#define ELF_ET_DYN_BASE (2 * TASK_SIZE / 3) ++#define ELF_ET_DYN_BASE (TASK_SIZE / 3 * 2) ++ ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE 0x00008000UL ++ ++#define PAX_DELTA_MMAP_LEN ((current->personality == PER_LINUX_32BIT) ? 16 : 10) ++#define PAX_DELTA_STACK_LEN ((current->personality == PER_LINUX_32BIT) ? 16 : 10) ++#endif + + /* When the program starts, a1 contains a pointer to a function to be + registered with atexit, as per the SVR4 ABI. A value of 0 means we +@@ -125,10 +132,6 @@ int dump_task_regs(struct task_struct *t + extern void elf_set_personality(const struct elf32_hdr *); + #define SET_PERSONALITY(ex) elf_set_personality(&(ex)) + +-struct mm_struct; +-extern unsigned long arch_randomize_brk(struct mm_struct *mm); +-#define arch_randomize_brk arch_randomize_brk +- + extern int vectors_user_mapping(void); + #define arch_setup_additional_pages(bprm, uses_interp) vectors_user_mapping() + #define ARCH_HAS_SETUP_ADDITIONAL_PAGES +diff -urNp linux-2.6.38.3/arch/arm/include/asm/kmap_types.h linux-2.6.38.3/arch/arm/include/asm/kmap_types.h +--- linux-2.6.38.3/arch/arm/include/asm/kmap_types.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/arm/include/asm/kmap_types.h 2011-04-17 15:57:32.000000000 -0400 +@@ -21,6 +21,7 @@ enum km_type { + KM_L1_CACHE, + KM_L2_CACHE, + KM_KDB, ++ KM_CLEARPAGE, + KM_TYPE_NR + }; + +diff -urNp linux-2.6.38.3/arch/arm/include/asm/uaccess.h linux-2.6.38.3/arch/arm/include/asm/uaccess.h +--- linux-2.6.38.3/arch/arm/include/asm/uaccess.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/arm/include/asm/uaccess.h 2011-04-17 15:57:32.000000000 -0400 +@@ -403,6 +403,9 @@ extern unsigned long __must_check __strn + + static inline unsigned long __must_check copy_from_user(void *to, const void __user *from, unsigned long n) + { ++ if ((long)n < 0) ++ return n; ++ + if (access_ok(VERIFY_READ, from, n)) + n = __copy_from_user(to, from, n); + else /* security hole - plug it */ +@@ -412,6 +415,9 @@ static inline unsigned long __must_check + + static inline unsigned long __must_check copy_to_user(void __user *to, const void *from, unsigned long n) + { ++ if ((long)n < 0) ++ return n; ++ + if (access_ok(VERIFY_WRITE, to, n)) + n = __copy_to_user(to, from, n); + return n; +diff -urNp linux-2.6.38.3/arch/arm/kernel/kgdb.c linux-2.6.38.3/arch/arm/kernel/kgdb.c +--- linux-2.6.38.3/arch/arm/kernel/kgdb.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/arm/kernel/kgdb.c 2011-04-17 15:57:32.000000000 -0400 +@@ -246,7 +246,7 @@ void kgdb_arch_exit(void) + * and we handle the normal undef case within the do_undefinstr + * handler. + */ +-struct kgdb_arch arch_kgdb_ops = { ++const struct kgdb_arch arch_kgdb_ops = { + #ifndef __ARMEB__ + .gdb_bpt_instr = {0xfe, 0xde, 0xff, 0xe7} + #else /* ! __ARMEB__ */ +diff -urNp linux-2.6.38.3/arch/arm/kernel/process.c linux-2.6.38.3/arch/arm/kernel/process.c +--- linux-2.6.38.3/arch/arm/kernel/process.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/arm/kernel/process.c 2011-04-17 15:57:32.000000000 -0400 +@@ -28,7 +28,6 @@ + #include <linux/tick.h> + #include <linux/utsname.h> + #include <linux/uaccess.h> +-#include <linux/random.h> + #include <linux/hw_breakpoint.h> + + #include <asm/cacheflush.h> +@@ -477,12 +476,6 @@ unsigned long get_wchan(struct task_stru + return 0; + } + +-unsigned long arch_randomize_brk(struct mm_struct *mm) +-{ +- unsigned long range_end = mm->brk + 0x02000000; +- return randomize_range(mm->brk, range_end, 0) ? : mm->brk; +-} +- + #ifdef CONFIG_MMU + /* + * The vectors page is always readable from user space for the +diff -urNp linux-2.6.38.3/arch/arm/mach-msm/last_radio_log.c linux-2.6.38.3/arch/arm/mach-msm/last_radio_log.c +--- linux-2.6.38.3/arch/arm/mach-msm/last_radio_log.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/arm/mach-msm/last_radio_log.c 2011-04-17 15:57:32.000000000 -0400 +@@ -47,7 +47,7 @@ static ssize_t last_radio_log_read(struc + return count; + } + +-static struct file_operations last_radio_log_fops = { ++static struct file_operations last_radio_log_fops = { /* cannot be const, see msm_init_last_radio_log */ + .read = last_radio_log_read, + .llseek = default_llseek, + }; +diff -urNp linux-2.6.38.3/arch/arm/mach-ux500/mbox-db5500.c linux-2.6.38.3/arch/arm/mach-ux500/mbox-db5500.c +--- linux-2.6.38.3/arch/arm/mach-ux500/mbox-db5500.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/arm/mach-ux500/mbox-db5500.c 2011-04-17 15:57:32.000000000 -0400 +@@ -168,7 +168,7 @@ static ssize_t mbox_read_fifo(struct dev + return sprintf(buf, "0x%X\n", mbox_value); + } + +-static DEVICE_ATTR(fifo, S_IWUGO | S_IRUGO, mbox_read_fifo, mbox_write_fifo); ++static DEVICE_ATTR(fifo, S_IWUSR | S_IRUGO, mbox_read_fifo, mbox_write_fifo); + + static int mbox_show(struct seq_file *s, void *data) + { +diff -urNp linux-2.6.38.3/arch/arm/mm/fault.c linux-2.6.38.3/arch/arm/mm/fault.c +--- linux-2.6.38.3/arch/arm/mm/fault.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/arm/mm/fault.c 2011-04-17 15:57:32.000000000 -0400 +@@ -167,6 +167,13 @@ __do_user_fault(struct task_struct *tsk, + } + #endif + ++#ifdef CONFIG_PAX_PAGEEXEC ++ if (fsr & FSR_LNX_PF) { ++ pax_report_fault(regs, (void *)regs->ARM_pc, (void *)regs->ARM_sp); ++ do_group_exit(SIGKILL); ++ } ++#endif ++ + tsk->thread.address = addr; + tsk->thread.error_code = fsr; + tsk->thread.trap_no = 14; +@@ -364,6 +371,33 @@ do_page_fault(unsigned long addr, unsign + } + #endif /* CONFIG_MMU */ + ++#ifdef CONFIG_PAX_PAGEEXEC ++void pax_report_insns(void *pc, void *sp) ++{ ++ long i; ++ ++ printk(KERN_ERR "PAX: bytes at PC: "); ++ for (i = 0; i < 20; i++) { ++ unsigned char c; ++ if (get_user(c, (__force unsigned char __user *)pc+i)) ++ printk(KERN_CONT "?? "); ++ else ++ printk(KERN_CONT "%02x ", c); ++ } ++ printk("\n"); ++ ++ printk(KERN_ERR "PAX: bytes at SP-4: "); ++ for (i = -1; i < 20; i++) { ++ unsigned long c; ++ if (get_user(c, (__force unsigned long __user *)sp+i)) ++ printk(KERN_CONT "???????? "); ++ else ++ printk(KERN_CONT "%08lx ", c); ++ } ++ printk("\n"); ++} ++#endif ++ + /* + * First Level Translation Fault Handler + * +diff -urNp linux-2.6.38.3/arch/arm/mm/mmap.c linux-2.6.38.3/arch/arm/mm/mmap.c +--- linux-2.6.38.3/arch/arm/mm/mmap.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/arm/mm/mmap.c 2011-04-17 15:57:32.000000000 -0400 +@@ -64,6 +64,10 @@ arch_get_unmapped_area(struct file *filp + if (len > TASK_SIZE) + return -ENOMEM; + ++#ifdef CONFIG_PAX_RANDMMAP ++ if (!(mm->pax_flags & MF_PAX_RANDMMAP)) ++#endif ++ + if (addr) { + if (do_align) + addr = COLOUR_ALIGN(addr, pgoff); +@@ -71,15 +75,14 @@ arch_get_unmapped_area(struct file *filp + addr = PAGE_ALIGN(addr); + + vma = find_vma(mm, addr); +- if (TASK_SIZE - len >= addr && +- (!vma || addr + len <= vma->vm_start)) ++ if (TASK_SIZE - len >= addr && check_heap_stack_gap(vma, addr, len)) + return addr; + } + if (len > mm->cached_hole_size) { +- start_addr = addr = mm->free_area_cache; ++ start_addr = addr = mm->free_area_cache; + } else { +- start_addr = addr = TASK_UNMAPPED_BASE; +- mm->cached_hole_size = 0; ++ start_addr = addr = mm->mmap_base; ++ mm->cached_hole_size = 0; + } + /* 8 bits of randomness in 20 address space bits */ + if (current->flags & PF_RANDOMIZE) +@@ -98,14 +101,14 @@ full_search: + * Start a new search - just in case we missed + * some holes. + */ +- if (start_addr != TASK_UNMAPPED_BASE) { +- start_addr = addr = TASK_UNMAPPED_BASE; ++ if (start_addr != mm->mmap_base) { ++ start_addr = addr = mm->mmap_base; + mm->cached_hole_size = 0; + goto full_search; + } + return -ENOMEM; + } +- if (!vma || addr + len <= vma->vm_start) { ++ if (check_heap_stack_gap(vma, addr, len)) { + /* + * Remember the place where we stopped the search: + */ +diff -urNp linux-2.6.38.3/arch/avr32/include/asm/elf.h linux-2.6.38.3/arch/avr32/include/asm/elf.h +--- linux-2.6.38.3/arch/avr32/include/asm/elf.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/avr32/include/asm/elf.h 2011-04-17 15:57:32.000000000 -0400 +@@ -84,8 +84,14 @@ typedef struct user_fpu_struct elf_fpreg + the loader. We need to make sure that it is out of the way of the program + that it will "exec", and that there is sufficient room for the brk. */ + +-#define ELF_ET_DYN_BASE (2 * TASK_SIZE / 3) ++#define ELF_ET_DYN_BASE (TASK_SIZE / 3 * 2) + ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE 0x00001000UL ++ ++#define PAX_DELTA_MMAP_LEN 15 ++#define PAX_DELTA_STACK_LEN 15 ++#endif + + /* This yields a mask that user programs can use to figure out what + instruction set this CPU supports. This could be done in user space, +diff -urNp linux-2.6.38.3/arch/avr32/include/asm/kmap_types.h linux-2.6.38.3/arch/avr32/include/asm/kmap_types.h +--- linux-2.6.38.3/arch/avr32/include/asm/kmap_types.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/avr32/include/asm/kmap_types.h 2011-04-17 15:57:32.000000000 -0400 +@@ -22,7 +22,8 @@ D(10) KM_IRQ0, + D(11) KM_IRQ1, + D(12) KM_SOFTIRQ0, + D(13) KM_SOFTIRQ1, +-D(14) KM_TYPE_NR ++D(14) KM_CLEARPAGE, ++D(15) KM_TYPE_NR + }; + + #undef D +diff -urNp linux-2.6.38.3/arch/avr32/mm/fault.c linux-2.6.38.3/arch/avr32/mm/fault.c +--- linux-2.6.38.3/arch/avr32/mm/fault.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/avr32/mm/fault.c 2011-04-17 15:57:32.000000000 -0400 +@@ -41,6 +41,23 @@ static inline int notify_page_fault(stru + + int exception_trace = 1; + ++#ifdef CONFIG_PAX_PAGEEXEC ++void pax_report_insns(void *pc, void *sp) ++{ ++ unsigned long i; ++ ++ printk(KERN_ERR "PAX: bytes at PC: "); ++ for (i = 0; i < 20; i++) { ++ unsigned char c; ++ if (get_user(c, (unsigned char *)pc+i)) ++ printk(KERN_CONT "???????? "); ++ else ++ printk(KERN_CONT "%02x ", c); ++ } ++ printk("\n"); ++} ++#endif ++ + /* + * This routine handles page faults. It determines the address and the + * problem, and then passes it off to one of the appropriate routines. +@@ -156,6 +173,16 @@ bad_area: + up_read(&mm->mmap_sem); + + if (user_mode(regs)) { ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if (mm->pax_flags & MF_PAX_PAGEEXEC) { ++ if (ecr == ECR_PROTECTION_X || ecr == ECR_TLB_MISS_X) { ++ pax_report_fault(regs, (void *)regs->pc, (void *)regs->sp); ++ do_group_exit(SIGKILL); ++ } ++ } ++#endif ++ + if (exception_trace && printk_ratelimit()) + printk("%s%s[%d]: segfault at %08lx pc %08lx " + "sp %08lx ecr %lu\n", +diff -urNp linux-2.6.38.3/arch/blackfin/kernel/kgdb.c linux-2.6.38.3/arch/blackfin/kernel/kgdb.c +--- linux-2.6.38.3/arch/blackfin/kernel/kgdb.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/blackfin/kernel/kgdb.c 2011-04-17 15:57:32.000000000 -0400 +@@ -420,7 +420,7 @@ int kgdb_arch_handle_exception(int vecto + return -1; /* this means that we do not want to exit from the handler */ + } + +-struct kgdb_arch arch_kgdb_ops = { ++const struct kgdb_arch arch_kgdb_ops = { + .gdb_bpt_instr = {0xa1}, + #ifdef CONFIG_SMP + .flags = KGDB_HW_BREAKPOINT|KGDB_THR_PROC_SWAP, +diff -urNp linux-2.6.38.3/arch/blackfin/mm/maccess.c linux-2.6.38.3/arch/blackfin/mm/maccess.c +--- linux-2.6.38.3/arch/blackfin/mm/maccess.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/blackfin/mm/maccess.c 2011-04-17 15:57:32.000000000 -0400 +@@ -16,7 +16,7 @@ static int validate_memory_access_addres + return bfin_mem_access_type(addr, size); + } + +-long probe_kernel_read(void *dst, void *src, size_t size) ++long probe_kernel_read(void *dst, const void *src, size_t size) + { + unsigned long lsrc = (unsigned long)src; + int mem_type; +@@ -55,7 +55,7 @@ long probe_kernel_read(void *dst, void * + return -EFAULT; + } + +-long probe_kernel_write(void *dst, void *src, size_t size) ++long probe_kernel_write(void *dst, const void *src, size_t size) + { + unsigned long ldst = (unsigned long)dst; + int mem_type; +diff -urNp linux-2.6.38.3/arch/frv/include/asm/kmap_types.h linux-2.6.38.3/arch/frv/include/asm/kmap_types.h +--- linux-2.6.38.3/arch/frv/include/asm/kmap_types.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/frv/include/asm/kmap_types.h 2011-04-17 15:57:32.000000000 -0400 +@@ -23,6 +23,7 @@ enum km_type { + KM_IRQ1, + KM_SOFTIRQ0, + KM_SOFTIRQ1, ++ KM_CLEARPAGE, + KM_TYPE_NR + }; + +diff -urNp linux-2.6.38.3/arch/frv/mm/elf-fdpic.c linux-2.6.38.3/arch/frv/mm/elf-fdpic.c +--- linux-2.6.38.3/arch/frv/mm/elf-fdpic.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/frv/mm/elf-fdpic.c 2011-04-17 15:57:32.000000000 -0400 +@@ -73,8 +73,7 @@ unsigned long arch_get_unmapped_area(str + if (addr) { + addr = PAGE_ALIGN(addr); + vma = find_vma(current->mm, addr); +- if (TASK_SIZE - len >= addr && +- (!vma || addr + len <= vma->vm_start)) ++ if (TASK_SIZE - len >= addr && check_heap_stack_gap(vma, addr, len)) + goto success; + } + +@@ -89,7 +88,7 @@ unsigned long arch_get_unmapped_area(str + for (; vma; vma = vma->vm_next) { + if (addr > limit) + break; +- if (addr + len <= vma->vm_start) ++ if (check_heap_stack_gap(vma, addr, len)) + goto success; + addr = vma->vm_end; + } +@@ -104,7 +103,7 @@ unsigned long arch_get_unmapped_area(str + for (; vma; vma = vma->vm_next) { + if (addr > limit) + break; +- if (addr + len <= vma->vm_start) ++ if (check_heap_stack_gap(vma, addr, len)) + goto success; + addr = vma->vm_end; + } +diff -urNp linux-2.6.38.3/arch/ia64/hp/common/hwsw_iommu.c linux-2.6.38.3/arch/ia64/hp/common/hwsw_iommu.c +--- linux-2.6.38.3/arch/ia64/hp/common/hwsw_iommu.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/ia64/hp/common/hwsw_iommu.c 2011-04-17 15:57:32.000000000 -0400 +@@ -17,7 +17,7 @@ + #include <linux/swiotlb.h> + #include <asm/machvec.h> + +-extern struct dma_map_ops sba_dma_ops, swiotlb_dma_ops; ++extern const struct dma_map_ops sba_dma_ops, swiotlb_dma_ops; + + /* swiotlb declarations & definitions: */ + extern int swiotlb_late_init_with_default_size (size_t size); +@@ -33,7 +33,7 @@ static inline int use_swiotlb(struct dev + !sba_dma_ops.dma_supported(dev, *dev->dma_mask); + } + +-struct dma_map_ops *hwsw_dma_get_ops(struct device *dev) ++const struct dma_map_ops *hwsw_dma_get_ops(struct device *dev) + { + if (use_swiotlb(dev)) + return &swiotlb_dma_ops; +diff -urNp linux-2.6.38.3/arch/ia64/hp/common/sba_iommu.c linux-2.6.38.3/arch/ia64/hp/common/sba_iommu.c +--- linux-2.6.38.3/arch/ia64/hp/common/sba_iommu.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/ia64/hp/common/sba_iommu.c 2011-04-17 15:57:32.000000000 -0400 +@@ -2097,7 +2097,7 @@ static struct acpi_driver acpi_sba_ioc_d + }, + }; + +-extern struct dma_map_ops swiotlb_dma_ops; ++extern const struct dma_map_ops swiotlb_dma_ops; + + static int __init + sba_init(void) +@@ -2211,7 +2211,7 @@ sba_page_override(char *str) + + __setup("sbapagesize=",sba_page_override); + +-struct dma_map_ops sba_dma_ops = { ++const struct dma_map_ops sba_dma_ops = { + .alloc_coherent = sba_alloc_coherent, + .free_coherent = sba_free_coherent, + .map_page = sba_map_page, +diff -urNp linux-2.6.38.3/arch/ia64/include/asm/dma-mapping.h linux-2.6.38.3/arch/ia64/include/asm/dma-mapping.h +--- linux-2.6.38.3/arch/ia64/include/asm/dma-mapping.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/ia64/include/asm/dma-mapping.h 2011-04-17 15:57:32.000000000 -0400 +@@ -12,7 +12,7 @@ + + #define ARCH_HAS_DMA_GET_REQUIRED_MASK + +-extern struct dma_map_ops *dma_ops; ++extern const struct dma_map_ops *dma_ops; + extern struct ia64_machine_vector ia64_mv; + extern void set_iommu_machvec(void); + +@@ -24,7 +24,7 @@ extern void machvec_dma_sync_sg(struct d + static inline void *dma_alloc_coherent(struct device *dev, size_t size, + dma_addr_t *daddr, gfp_t gfp) + { +- struct dma_map_ops *ops = platform_dma_get_ops(dev); ++ const struct dma_map_ops *ops = platform_dma_get_ops(dev); + void *caddr; + + caddr = ops->alloc_coherent(dev, size, daddr, gfp); +@@ -35,7 +35,7 @@ static inline void *dma_alloc_coherent(s + static inline void dma_free_coherent(struct device *dev, size_t size, + void *caddr, dma_addr_t daddr) + { +- struct dma_map_ops *ops = platform_dma_get_ops(dev); ++ const struct dma_map_ops *ops = platform_dma_get_ops(dev); + debug_dma_free_coherent(dev, size, caddr, daddr); + ops->free_coherent(dev, size, caddr, daddr); + } +@@ -49,13 +49,13 @@ static inline void dma_free_coherent(str + + static inline int dma_mapping_error(struct device *dev, dma_addr_t daddr) + { +- struct dma_map_ops *ops = platform_dma_get_ops(dev); ++ const struct dma_map_ops *ops = platform_dma_get_ops(dev); + return ops->mapping_error(dev, daddr); + } + + static inline int dma_supported(struct device *dev, u64 mask) + { +- struct dma_map_ops *ops = platform_dma_get_ops(dev); ++ const struct dma_map_ops *ops = platform_dma_get_ops(dev); + return ops->dma_supported(dev, mask); + } + +diff -urNp linux-2.6.38.3/arch/ia64/include/asm/elf.h linux-2.6.38.3/arch/ia64/include/asm/elf.h +--- linux-2.6.38.3/arch/ia64/include/asm/elf.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/ia64/include/asm/elf.h 2011-04-17 15:57:32.000000000 -0400 +@@ -42,6 +42,13 @@ + */ + #define ELF_ET_DYN_BASE (TASK_UNMAPPED_BASE + 0x800000000UL) + ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE (current->personality == PER_LINUX32 ? 0x08048000UL : 0x4000000000000000UL) ++ ++#define PAX_DELTA_MMAP_LEN (current->personality == PER_LINUX32 ? 16 : 3*PAGE_SHIFT - 13) ++#define PAX_DELTA_STACK_LEN (current->personality == PER_LINUX32 ? 16 : 3*PAGE_SHIFT - 13) ++#endif ++ + #define PT_IA_64_UNWIND 0x70000001 + + /* IA-64 relocations: */ +diff -urNp linux-2.6.38.3/arch/ia64/include/asm/machvec.h linux-2.6.38.3/arch/ia64/include/asm/machvec.h +--- linux-2.6.38.3/arch/ia64/include/asm/machvec.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/ia64/include/asm/machvec.h 2011-04-17 15:57:32.000000000 -0400 +@@ -45,7 +45,7 @@ typedef void ia64_mv_kernel_launch_event + /* DMA-mapping interface: */ + typedef void ia64_mv_dma_init (void); + typedef u64 ia64_mv_dma_get_required_mask (struct device *); +-typedef struct dma_map_ops *ia64_mv_dma_get_ops(struct device *); ++typedef const struct dma_map_ops *ia64_mv_dma_get_ops(struct device *); + + /* + * WARNING: The legacy I/O space is _architected_. Platforms are +@@ -251,7 +251,7 @@ extern void machvec_init_from_cmdline(co + # endif /* CONFIG_IA64_GENERIC */ + + extern void swiotlb_dma_init(void); +-extern struct dma_map_ops *dma_get_ops(struct device *); ++extern const struct dma_map_ops *dma_get_ops(struct device *); + + /* + * Define default versions so we can extend machvec for new platforms without having +diff -urNp linux-2.6.38.3/arch/ia64/include/asm/pgtable.h linux-2.6.38.3/arch/ia64/include/asm/pgtable.h +--- linux-2.6.38.3/arch/ia64/include/asm/pgtable.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/ia64/include/asm/pgtable.h 2011-04-17 15:57:32.000000000 -0400 +@@ -12,7 +12,7 @@ + * David Mosberger-Tang <davidm@hpl.hp.com> + */ + +- ++#include <linux/const.h> + #include <asm/mman.h> + #include <asm/page.h> + #include <asm/processor.h> +@@ -143,6 +143,17 @@ + #define PAGE_READONLY __pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_R) + #define PAGE_COPY __pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_R) + #define PAGE_COPY_EXEC __pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_RX) ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++# define PAGE_SHARED_NOEXEC __pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_RW) ++# define PAGE_READONLY_NOEXEC __pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_R) ++# define PAGE_COPY_NOEXEC __pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_R) ++#else ++# define PAGE_SHARED_NOEXEC PAGE_SHARED ++# define PAGE_READONLY_NOEXEC PAGE_READONLY ++# define PAGE_COPY_NOEXEC PAGE_COPY ++#endif ++ + #define PAGE_GATE __pgprot(__ACCESS_BITS | _PAGE_PL_0 | _PAGE_AR_X_RX) + #define PAGE_KERNEL __pgprot(__DIRTY_BITS | _PAGE_PL_0 | _PAGE_AR_RWX) + #define PAGE_KERNELRX __pgprot(__ACCESS_BITS | _PAGE_PL_0 | _PAGE_AR_RX) +diff -urNp linux-2.6.38.3/arch/ia64/include/asm/spinlock.h linux-2.6.38.3/arch/ia64/include/asm/spinlock.h +--- linux-2.6.38.3/arch/ia64/include/asm/spinlock.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/ia64/include/asm/spinlock.h 2011-04-17 15:57:32.000000000 -0400 +@@ -72,7 +72,7 @@ static __always_inline void __ticket_spi + unsigned short *p = (unsigned short *)&lock->lock + 1, tmp; + + asm volatile ("ld2.bias %0=[%1]" : "=r"(tmp) : "r"(p)); +- ACCESS_ONCE(*p) = (tmp + 2) & ~1; ++ ACCESS_ONCE_RW(*p) = (tmp + 2) & ~1; + } + + static __always_inline void __ticket_spin_unlock_wait(arch_spinlock_t *lock) +diff -urNp linux-2.6.38.3/arch/ia64/include/asm/uaccess.h linux-2.6.38.3/arch/ia64/include/asm/uaccess.h +--- linux-2.6.38.3/arch/ia64/include/asm/uaccess.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/ia64/include/asm/uaccess.h 2011-04-17 15:57:32.000000000 -0400 +@@ -257,7 +257,7 @@ __copy_from_user (void *to, const void _ + const void *__cu_from = (from); \ + long __cu_len = (n); \ + \ +- if (__access_ok(__cu_to, __cu_len, get_fs())) \ ++ if (__cu_len > 0 && __cu_len <= INT_MAX && __access_ok(__cu_to, __cu_len, get_fs())) \ + __cu_len = __copy_user(__cu_to, (__force void __user *) __cu_from, __cu_len); \ + __cu_len; \ + }) +@@ -269,7 +269,7 @@ __copy_from_user (void *to, const void _ + long __cu_len = (n); \ + \ + __chk_user_ptr(__cu_from); \ +- if (__access_ok(__cu_from, __cu_len, get_fs())) \ ++ if (__cu_len > 0 && __cu_len <= INT_MAX && __access_ok(__cu_from, __cu_len, get_fs())) \ + __cu_len = __copy_user((__force void __user *) __cu_to, __cu_from, __cu_len); \ + __cu_len; \ + }) +diff -urNp linux-2.6.38.3/arch/ia64/kernel/dma-mapping.c linux-2.6.38.3/arch/ia64/kernel/dma-mapping.c +--- linux-2.6.38.3/arch/ia64/kernel/dma-mapping.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/ia64/kernel/dma-mapping.c 2011-04-17 15:57:32.000000000 -0400 +@@ -3,7 +3,7 @@ + /* Set this to 1 if there is a HW IOMMU in the system */ + int iommu_detected __read_mostly; + +-struct dma_map_ops *dma_ops; ++const struct dma_map_ops *dma_ops; + EXPORT_SYMBOL(dma_ops); + + #define PREALLOC_DMA_DEBUG_ENTRIES (1 << 16) +@@ -16,7 +16,7 @@ static int __init dma_init(void) + } + fs_initcall(dma_init); + +-struct dma_map_ops *dma_get_ops(struct device *dev) ++const struct dma_map_ops *dma_get_ops(struct device *dev) + { + return dma_ops; + } +diff -urNp linux-2.6.38.3/arch/ia64/kernel/module.c linux-2.6.38.3/arch/ia64/kernel/module.c +--- linux-2.6.38.3/arch/ia64/kernel/module.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/ia64/kernel/module.c 2011-04-17 15:57:32.000000000 -0400 +@@ -315,8 +315,7 @@ module_alloc (unsigned long size) + void + module_free (struct module *mod, void *module_region) + { +- if (mod && mod->arch.init_unw_table && +- module_region == mod->module_init) { ++ if (mod && mod->arch.init_unw_table && module_region == mod->module_init_rx) { + unw_remove_unwind_table(mod->arch.init_unw_table); + mod->arch.init_unw_table = NULL; + } +@@ -502,15 +501,39 @@ module_frob_arch_sections (Elf_Ehdr *ehd + } + + static inline int ++in_init_rx (const struct module *mod, uint64_t addr) ++{ ++ return addr - (uint64_t) mod->module_init_rx < mod->init_size_rx; ++} ++ ++static inline int ++in_init_rw (const struct module *mod, uint64_t addr) ++{ ++ return addr - (uint64_t) mod->module_init_rw < mod->init_size_rw; ++} ++ ++static inline int + in_init (const struct module *mod, uint64_t addr) + { +- return addr - (uint64_t) mod->module_init < mod->init_size; ++ return in_init_rx(mod, addr) || in_init_rw(mod, addr); ++} ++ ++static inline int ++in_core_rx (const struct module *mod, uint64_t addr) ++{ ++ return addr - (uint64_t) mod->module_core_rx < mod->core_size_rx; ++} ++ ++static inline int ++in_core_rw (const struct module *mod, uint64_t addr) ++{ ++ return addr - (uint64_t) mod->module_core_rw < mod->core_size_rw; + } + + static inline int + in_core (const struct module *mod, uint64_t addr) + { +- return addr - (uint64_t) mod->module_core < mod->core_size; ++ return in_core_rx(mod, addr) || in_core_rw(mod, addr); + } + + static inline int +@@ -693,7 +716,14 @@ do_reloc (struct module *mod, uint8_t r_ + break; + + case RV_BDREL: +- val -= (uint64_t) (in_init(mod, val) ? mod->module_init : mod->module_core); ++ if (in_init_rx(mod, val)) ++ val -= (uint64_t) mod->module_init_rx; ++ else if (in_init_rw(mod, val)) ++ val -= (uint64_t) mod->module_init_rw; ++ else if (in_core_rx(mod, val)) ++ val -= (uint64_t) mod->module_core_rx; ++ else if (in_core_rw(mod, val)) ++ val -= (uint64_t) mod->module_core_rw; + break; + + case RV_LTV: +@@ -828,15 +858,15 @@ apply_relocate_add (Elf64_Shdr *sechdrs, + * addresses have been selected... + */ + uint64_t gp; +- if (mod->core_size > MAX_LTOFF) ++ if (mod->core_size_rx + mod->core_size_rw > MAX_LTOFF) + /* + * This takes advantage of fact that SHF_ARCH_SMALL gets allocated + * at the end of the module. + */ +- gp = mod->core_size - MAX_LTOFF / 2; ++ gp = mod->core_size_rx + mod->core_size_rw - MAX_LTOFF / 2; + else +- gp = mod->core_size / 2; +- gp = (uint64_t) mod->module_core + ((gp + 7) & -8); ++ gp = (mod->core_size_rx + mod->core_size_rw) / 2; ++ gp = (uint64_t) mod->module_core_rx + ((gp + 7) & -8); + mod->arch.gp = gp; + DEBUGP("%s: placing gp at 0x%lx\n", __func__, gp); + } +diff -urNp linux-2.6.38.3/arch/ia64/kernel/pci-dma.c linux-2.6.38.3/arch/ia64/kernel/pci-dma.c +--- linux-2.6.38.3/arch/ia64/kernel/pci-dma.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/ia64/kernel/pci-dma.c 2011-04-17 15:57:32.000000000 -0400 +@@ -43,7 +43,7 @@ struct device fallback_dev = { + .dma_mask = &fallback_dev.coherent_dma_mask, + }; + +-extern struct dma_map_ops intel_dma_ops; ++extern const struct dma_map_ops intel_dma_ops; + + static int __init pci_iommu_init(void) + { +diff -urNp linux-2.6.38.3/arch/ia64/kernel/pci-swiotlb.c linux-2.6.38.3/arch/ia64/kernel/pci-swiotlb.c +--- linux-2.6.38.3/arch/ia64/kernel/pci-swiotlb.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/ia64/kernel/pci-swiotlb.c 2011-04-17 15:57:32.000000000 -0400 +@@ -22,7 +22,7 @@ static void *ia64_swiotlb_alloc_coherent + return swiotlb_alloc_coherent(dev, size, dma_handle, gfp); + } + +-struct dma_map_ops swiotlb_dma_ops = { ++const struct dma_map_ops swiotlb_dma_ops = { + .alloc_coherent = ia64_swiotlb_alloc_coherent, + .free_coherent = swiotlb_free_coherent, + .map_page = swiotlb_map_page, +diff -urNp linux-2.6.38.3/arch/ia64/kernel/sys_ia64.c linux-2.6.38.3/arch/ia64/kernel/sys_ia64.c +--- linux-2.6.38.3/arch/ia64/kernel/sys_ia64.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/ia64/kernel/sys_ia64.c 2011-04-17 15:57:32.000000000 -0400 +@@ -43,6 +43,13 @@ arch_get_unmapped_area (struct file *fil + if (REGION_NUMBER(addr) == RGN_HPAGE) + addr = 0; + #endif ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ addr = mm->free_area_cache; ++ else ++#endif ++ + if (!addr) + addr = mm->free_area_cache; + +@@ -61,14 +68,14 @@ arch_get_unmapped_area (struct file *fil + for (vma = find_vma(mm, addr); ; vma = vma->vm_next) { + /* At this point: (!vma || addr < vma->vm_end). */ + if (TASK_SIZE - len < addr || RGN_MAP_LIMIT - len < REGION_OFFSET(addr)) { +- if (start_addr != TASK_UNMAPPED_BASE) { ++ if (start_addr != mm->mmap_base) { + /* Start a new search --- just in case we missed some holes. */ +- addr = TASK_UNMAPPED_BASE; ++ addr = mm->mmap_base; + goto full_search; + } + return -ENOMEM; + } +- if (!vma || addr + len <= vma->vm_start) { ++ if (check_heap_stack_gap(vma, addr, len)) { + /* Remember the address where we stopped this search: */ + mm->free_area_cache = addr + len; + return addr; +diff -urNp linux-2.6.38.3/arch/ia64/kernel/vmlinux.lds.S linux-2.6.38.3/arch/ia64/kernel/vmlinux.lds.S +--- linux-2.6.38.3/arch/ia64/kernel/vmlinux.lds.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/ia64/kernel/vmlinux.lds.S 2011-04-17 15:57:32.000000000 -0400 +@@ -199,7 +199,7 @@ SECTIONS { + /* Per-cpu data: */ + . = ALIGN(PERCPU_PAGE_SIZE); + PERCPU_VADDR(PERCPU_ADDR, :percpu) +- __phys_per_cpu_start = __per_cpu_load; ++ __phys_per_cpu_start = per_cpu_load; + /* + * ensure percpu data fits + * into percpu page size +diff -urNp linux-2.6.38.3/arch/ia64/mm/fault.c linux-2.6.38.3/arch/ia64/mm/fault.c +--- linux-2.6.38.3/arch/ia64/mm/fault.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/ia64/mm/fault.c 2011-04-17 15:57:32.000000000 -0400 +@@ -72,6 +72,23 @@ mapped_kernel_page_is_present (unsigned + return pte_present(pte); + } + ++#ifdef CONFIG_PAX_PAGEEXEC ++void pax_report_insns(void *pc, void *sp) ++{ ++ unsigned long i; ++ ++ printk(KERN_ERR "PAX: bytes at PC: "); ++ for (i = 0; i < 8; i++) { ++ unsigned int c; ++ if (get_user(c, (unsigned int *)pc+i)) ++ printk(KERN_CONT "???????? "); ++ else ++ printk(KERN_CONT "%08x ", c); ++ } ++ printk("\n"); ++} ++#endif ++ + void __kprobes + ia64_do_page_fault (unsigned long address, unsigned long isr, struct pt_regs *regs) + { +@@ -145,9 +162,23 @@ ia64_do_page_fault (unsigned long addres + mask = ( (((isr >> IA64_ISR_X_BIT) & 1UL) << VM_EXEC_BIT) + | (((isr >> IA64_ISR_W_BIT) & 1UL) << VM_WRITE_BIT)); + +- if ((vma->vm_flags & mask) != mask) ++ if ((vma->vm_flags & mask) != mask) { ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if (!(vma->vm_flags & VM_EXEC) && (mask & VM_EXEC)) { ++ if (!(mm->pax_flags & MF_PAX_PAGEEXEC) || address != regs->cr_iip) ++ goto bad_area; ++ ++ up_read(&mm->mmap_sem); ++ pax_report_fault(regs, (void *)regs->cr_iip, (void *)regs->r12); ++ do_group_exit(SIGKILL); ++ } ++#endif ++ + goto bad_area; + ++ } ++ + /* + * If for any reason at all we couldn't handle the fault, make + * sure we exit gracefully rather than endlessly redo the +diff -urNp linux-2.6.38.3/arch/ia64/mm/hugetlbpage.c linux-2.6.38.3/arch/ia64/mm/hugetlbpage.c +--- linux-2.6.38.3/arch/ia64/mm/hugetlbpage.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/ia64/mm/hugetlbpage.c 2011-04-17 15:57:32.000000000 -0400 +@@ -171,7 +171,7 @@ unsigned long hugetlb_get_unmapped_area( + /* At this point: (!vmm || addr < vmm->vm_end). */ + if (REGION_OFFSET(addr) + len > RGN_MAP_LIMIT) + return -ENOMEM; +- if (!vmm || (addr + len) <= vmm->vm_start) ++ if (check_heap_stack_gap(vmm, addr, len)) + return addr; + addr = ALIGN(vmm->vm_end, HPAGE_SIZE); + } +diff -urNp linux-2.6.38.3/arch/ia64/mm/init.c linux-2.6.38.3/arch/ia64/mm/init.c +--- linux-2.6.38.3/arch/ia64/mm/init.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/ia64/mm/init.c 2011-04-17 15:57:32.000000000 -0400 +@@ -122,6 +122,19 @@ ia64_init_addr_space (void) + vma->vm_start = current->thread.rbs_bot & PAGE_MASK; + vma->vm_end = vma->vm_start + PAGE_SIZE; + vma->vm_flags = VM_DATA_DEFAULT_FLAGS|VM_GROWSUP|VM_ACCOUNT; ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if (current->mm->pax_flags & MF_PAX_PAGEEXEC) { ++ vma->vm_flags &= ~VM_EXEC; ++ ++#ifdef CONFIG_PAX_MPROTECT ++ if (current->mm->pax_flags & MF_PAX_MPROTECT) ++ vma->vm_flags &= ~VM_MAYEXEC; ++#endif ++ ++ } ++#endif ++ + vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); + down_write(¤t->mm->mmap_sem); + if (insert_vm_struct(current->mm, vma)) { +diff -urNp linux-2.6.38.3/arch/ia64/sn/pci/pci_dma.c linux-2.6.38.3/arch/ia64/sn/pci/pci_dma.c +--- linux-2.6.38.3/arch/ia64/sn/pci/pci_dma.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/ia64/sn/pci/pci_dma.c 2011-04-17 15:57:32.000000000 -0400 +@@ -465,7 +465,7 @@ int sn_pci_legacy_write(struct pci_bus * + return ret; + } + +-static struct dma_map_ops sn_dma_ops = { ++static const struct dma_map_ops sn_dma_ops = { + .alloc_coherent = sn_dma_alloc_coherent, + .free_coherent = sn_dma_free_coherent, + .map_page = sn_dma_map_page, +diff -urNp linux-2.6.38.3/arch/m32r/lib/usercopy.c linux-2.6.38.3/arch/m32r/lib/usercopy.c +--- linux-2.6.38.3/arch/m32r/lib/usercopy.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/m32r/lib/usercopy.c 2011-04-17 15:57:32.000000000 -0400 +@@ -14,6 +14,9 @@ + unsigned long + __generic_copy_to_user(void __user *to, const void *from, unsigned long n) + { ++ if ((long)n < 0) ++ return n; ++ + prefetch(from); + if (access_ok(VERIFY_WRITE, to, n)) + __copy_user(to,from,n); +@@ -23,6 +26,9 @@ __generic_copy_to_user(void __user *to, + unsigned long + __generic_copy_from_user(void *to, const void __user *from, unsigned long n) + { ++ if ((long)n < 0) ++ return n; ++ + prefetchw(to); + if (access_ok(VERIFY_READ, from, n)) + __copy_user_zeroing(to,from,n); +diff -urNp linux-2.6.38.3/arch/microblaze/include/asm/device.h linux-2.6.38.3/arch/microblaze/include/asm/device.h +--- linux-2.6.38.3/arch/microblaze/include/asm/device.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/microblaze/include/asm/device.h 2011-04-17 15:57:32.000000000 -0400 +@@ -13,7 +13,7 @@ struct device_node; + + struct dev_archdata { + /* DMA operations on that device */ +- struct dma_map_ops *dma_ops; ++ const struct dma_map_ops *dma_ops; + void *dma_data; + }; + +diff -urNp linux-2.6.38.3/arch/microblaze/include/asm/dma-mapping.h linux-2.6.38.3/arch/microblaze/include/asm/dma-mapping.h +--- linux-2.6.38.3/arch/microblaze/include/asm/dma-mapping.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/microblaze/include/asm/dma-mapping.h 2011-04-17 15:57:32.000000000 -0400 +@@ -43,14 +43,14 @@ static inline unsigned long device_to_ma + return 0xfffffffful; + } + +-extern struct dma_map_ops *dma_ops; ++extern const struct dma_map_ops *dma_ops; + + /* + * Available generic sets of operations + */ +-extern struct dma_map_ops dma_direct_ops; ++extern const struct dma_map_ops dma_direct_ops; + +-static inline struct dma_map_ops *get_dma_ops(struct device *dev) ++static inline const struct dma_map_ops *get_dma_ops(struct device *dev) + { + /* We don't handle the NULL dev case for ISA for now. We could + * do it via an out of line call but it is not needed for now. The +@@ -63,14 +63,14 @@ static inline struct dma_map_ops *get_dm + return dev->archdata.dma_ops; + } + +-static inline void set_dma_ops(struct device *dev, struct dma_map_ops *ops) ++static inline void set_dma_ops(struct device *dev, const struct dma_map_ops *ops) + { + dev->archdata.dma_ops = ops; + } + + static inline int dma_supported(struct device *dev, u64 mask) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + + if (unlikely(!ops)) + return 0; +@@ -81,7 +81,7 @@ static inline int dma_supported(struct d + + static inline int dma_set_mask(struct device *dev, u64 dma_mask) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + + if (unlikely(ops == NULL)) + return -EIO; +@@ -97,7 +97,7 @@ static inline int dma_set_mask(struct de + + static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + if (ops->mapping_error) + return ops->mapping_error(dev, dma_addr); + +@@ -110,7 +110,7 @@ static inline int dma_mapping_error(stru + static inline void *dma_alloc_coherent(struct device *dev, size_t size, + dma_addr_t *dma_handle, gfp_t flag) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + void *memory; + + BUG_ON(!ops); +@@ -124,7 +124,7 @@ static inline void *dma_alloc_coherent(s + static inline void dma_free_coherent(struct device *dev, size_t size, + void *cpu_addr, dma_addr_t dma_handle) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + + BUG_ON(!ops); + debug_dma_free_coherent(dev, size, cpu_addr, dma_handle); +diff -urNp linux-2.6.38.3/arch/microblaze/include/asm/pci.h linux-2.6.38.3/arch/microblaze/include/asm/pci.h +--- linux-2.6.38.3/arch/microblaze/include/asm/pci.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/microblaze/include/asm/pci.h 2011-04-17 15:57:32.000000000 -0400 +@@ -54,8 +54,8 @@ static inline void pcibios_penalize_isa_ + } + + #ifdef CONFIG_PCI +-extern void set_pci_dma_ops(struct dma_map_ops *dma_ops); +-extern struct dma_map_ops *get_pci_dma_ops(void); ++extern void set_pci_dma_ops(const struct dma_map_ops *dma_ops); ++extern const struct dma_map_ops *get_pci_dma_ops(void); + #else /* CONFIG_PCI */ + #define set_pci_dma_ops(d) + #define get_pci_dma_ops() NULL +diff -urNp linux-2.6.38.3/arch/microblaze/kernel/dma.c linux-2.6.38.3/arch/microblaze/kernel/dma.c +--- linux-2.6.38.3/arch/microblaze/kernel/dma.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/microblaze/kernel/dma.c 2011-04-17 15:57:32.000000000 -0400 +@@ -133,7 +133,7 @@ static inline void dma_direct_unmap_page + __dma_sync_page(dma_address, 0 , size, direction); + } + +-struct dma_map_ops dma_direct_ops = { ++const struct dma_map_ops dma_direct_ops = { + .alloc_coherent = dma_direct_alloc_coherent, + .free_coherent = dma_direct_free_coherent, + .map_sg = dma_direct_map_sg, +diff -urNp linux-2.6.38.3/arch/microblaze/kernel/kgdb.c linux-2.6.38.3/arch/microblaze/kernel/kgdb.c +--- linux-2.6.38.3/arch/microblaze/kernel/kgdb.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/microblaze/kernel/kgdb.c 2011-04-17 15:57:32.000000000 -0400 +@@ -141,10 +141,11 @@ void kgdb_arch_exit(void) + /* + * Global data + */ +-struct kgdb_arch arch_kgdb_ops = { ++const struct kgdb_arch arch_kgdb_ops = { + #ifdef __MICROBLAZEEL__ + .gdb_bpt_instr = {0x18, 0x00, 0x0c, 0xba}, /* brki r16, 0x18 */ + #else ++>>>>>>> master + .gdb_bpt_instr = {0xba, 0x0c, 0x00, 0x18}, /* brki r16, 0x18 */ + #endif + }; +diff -urNp linux-2.6.38.3/arch/microblaze/pci/pci-common.c linux-2.6.38.3/arch/microblaze/pci/pci-common.c +--- linux-2.6.38.3/arch/microblaze/pci/pci-common.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/microblaze/pci/pci-common.c 2011-04-17 15:57:32.000000000 -0400 +@@ -47,14 +47,14 @@ resource_size_t isa_mem_base; + /* Default PCI flags is 0 on ppc32, modified at boot on ppc64 */ + unsigned int pci_flags; + +-static struct dma_map_ops *pci_dma_ops = &dma_direct_ops; ++static const struct dma_map_ops *pci_dma_ops = &dma_direct_ops; + +-void set_pci_dma_ops(struct dma_map_ops *dma_ops) ++void set_pci_dma_ops(const struct dma_map_ops *dma_ops) + { + pci_dma_ops = dma_ops; + } + +-struct dma_map_ops *get_pci_dma_ops(void) ++const struct dma_map_ops *get_pci_dma_ops(void) + { + return pci_dma_ops; + } +diff -urNp linux-2.6.38.3/arch/mips/cavium-octeon/dma-octeon.c linux-2.6.38.3/arch/mips/cavium-octeon/dma-octeon.c +--- linux-2.6.38.3/arch/mips/cavium-octeon/dma-octeon.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/mips/cavium-octeon/dma-octeon.c 2011-04-17 15:57:32.000000000 -0400 +@@ -202,7 +202,7 @@ static phys_addr_t octeon_unity_dma_to_p + } + + struct octeon_dma_map_ops { +- struct dma_map_ops dma_map_ops; ++ const struct dma_map_ops dma_map_ops; + dma_addr_t (*phys_to_dma)(struct device *dev, phys_addr_t paddr); + phys_addr_t (*dma_to_phys)(struct device *dev, dma_addr_t daddr); + }; +@@ -324,7 +324,7 @@ static struct octeon_dma_map_ops _octeon + }, + }; + +-struct dma_map_ops *octeon_pci_dma_map_ops; ++const struct dma_map_ops *octeon_pci_dma_map_ops; + + void __init octeon_pci_dma_init(void) + { +diff -urNp linux-2.6.38.3/arch/mips/include/asm/device.h linux-2.6.38.3/arch/mips/include/asm/device.h +--- linux-2.6.38.3/arch/mips/include/asm/device.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/mips/include/asm/device.h 2011-04-17 15:57:32.000000000 -0400 +@@ -10,7 +10,7 @@ struct dma_map_ops; + + struct dev_archdata { + /* DMA operations on that device */ +- struct dma_map_ops *dma_ops; ++ const struct dma_map_ops *dma_ops; + }; + + struct pdev_archdata { +diff -urNp linux-2.6.38.3/arch/mips/include/asm/dma-mapping.h linux-2.6.38.3/arch/mips/include/asm/dma-mapping.h +--- linux-2.6.38.3/arch/mips/include/asm/dma-mapping.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/mips/include/asm/dma-mapping.h 2011-04-17 15:57:32.000000000 -0400 +@@ -7,9 +7,9 @@ + + #include <dma-coherence.h> + +-extern struct dma_map_ops *mips_dma_map_ops; ++extern const struct dma_map_ops *mips_dma_map_ops; + +-static inline struct dma_map_ops *get_dma_ops(struct device *dev) ++static inline const struct dma_map_ops *get_dma_ops(struct device *dev) + { + if (dev && dev->archdata.dma_ops) + return dev->archdata.dma_ops; +@@ -31,13 +31,13 @@ static inline void dma_mark_clean(void * + + static inline int dma_supported(struct device *dev, u64 mask) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + return ops->dma_supported(dev, mask); + } + + static inline int dma_mapping_error(struct device *dev, u64 mask) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + return ops->mapping_error(dev, mask); + } + +@@ -59,7 +59,7 @@ static inline void *dma_alloc_coherent(s + dma_addr_t *dma_handle, gfp_t gfp) + { + void *ret; +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + + ret = ops->alloc_coherent(dev, size, dma_handle, gfp); + +@@ -71,7 +71,7 @@ static inline void *dma_alloc_coherent(s + static inline void dma_free_coherent(struct device *dev, size_t size, + void *vaddr, dma_addr_t dma_handle) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + + ops->free_coherent(dev, size, vaddr, dma_handle); + +diff -urNp linux-2.6.38.3/arch/mips/include/asm/elf.h linux-2.6.38.3/arch/mips/include/asm/elf.h +--- linux-2.6.38.3/arch/mips/include/asm/elf.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/mips/include/asm/elf.h 2011-04-17 15:57:32.000000000 -0400 +@@ -372,13 +372,16 @@ extern const char *__elf_platform; + #define ELF_ET_DYN_BASE (TASK_SIZE / 3 * 2) + #endif + ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE (test_thread_flag(TIF_32BIT_ADDR) ? 0x00400000UL : 0x00400000UL) ++ ++#define PAX_DELTA_MMAP_LEN (test_thread_flag(TIF_32BIT_ADDR) ? 27-PAGE_SHIFT : 36-PAGE_SHIFT) ++#define PAX_DELTA_STACK_LEN (test_thread_flag(TIF_32BIT_ADDR) ? 27-PAGE_SHIFT : 36-PAGE_SHIFT) ++#endif ++ + #define ARCH_HAS_SETUP_ADDITIONAL_PAGES 1 + struct linux_binprm; + extern int arch_setup_additional_pages(struct linux_binprm *bprm, + int uses_interp); + +-struct mm_struct; +-extern unsigned long arch_randomize_brk(struct mm_struct *mm); +-#define arch_randomize_brk arch_randomize_brk +- + #endif /* _ASM_ELF_H */ +diff -urNp linux-2.6.38.3/arch/mips/include/asm/mach-cavium-octeon/dma-coherence.h linux-2.6.38.3/arch/mips/include/asm/mach-cavium-octeon/dma-coherence.h +--- linux-2.6.38.3/arch/mips/include/asm/mach-cavium-octeon/dma-coherence.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/mips/include/asm/mach-cavium-octeon/dma-coherence.h 2011-04-17 15:57:32.000000000 -0400 +@@ -66,7 +66,7 @@ dma_addr_t phys_to_dma(struct device *de + phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr); + + struct dma_map_ops; +-extern struct dma_map_ops *octeon_pci_dma_map_ops; ++extern const struct dma_map_ops *octeon_pci_dma_map_ops; + extern char *octeon_swiotlb; + + #endif /* __ASM_MACH_CAVIUM_OCTEON_DMA_COHERENCE_H */ +diff -urNp linux-2.6.38.3/arch/mips/include/asm/page.h linux-2.6.38.3/arch/mips/include/asm/page.h +--- linux-2.6.38.3/arch/mips/include/asm/page.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/mips/include/asm/page.h 2011-04-17 15:57:32.000000000 -0400 +@@ -93,7 +93,7 @@ extern void copy_user_highpage(struct pa + #ifdef CONFIG_CPU_MIPS32 + typedef struct { unsigned long pte_low, pte_high; } pte_t; + #define pte_val(x) ((x).pte_low | ((unsigned long long)(x).pte_high << 32)) +- #define __pte(x) ({ pte_t __pte = {(x), ((unsigned long long)(x)) >> 32}; __pte; }) ++ #define __pte(x) ({ pte_t __pte = {(x), (x) >> 32}; __pte; }) + #else + typedef struct { unsigned long long pte; } pte_t; + #define pte_val(x) ((x).pte) +diff -urNp linux-2.6.38.3/arch/mips/include/asm/system.h linux-2.6.38.3/arch/mips/include/asm/system.h +--- linux-2.6.38.3/arch/mips/include/asm/system.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/mips/include/asm/system.h 2011-04-17 15:57:32.000000000 -0400 +@@ -230,6 +230,6 @@ extern void per_cpu_trap_init(void); + */ + #define __ARCH_WANT_UNLOCKED_CTXSW + +-extern unsigned long arch_align_stack(unsigned long sp); ++#define arch_align_stack(x) ((x) & ~0xfUL) + + #endif /* _ASM_SYSTEM_H */ +diff -urNp linux-2.6.38.3/arch/mips/kernel/binfmt_elfn32.c linux-2.6.38.3/arch/mips/kernel/binfmt_elfn32.c +--- linux-2.6.38.3/arch/mips/kernel/binfmt_elfn32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/mips/kernel/binfmt_elfn32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -50,6 +50,13 @@ typedef elf_fpreg_t elf_fpregset_t[ELF_N + #undef ELF_ET_DYN_BASE + #define ELF_ET_DYN_BASE (TASK32_SIZE / 3 * 2) + ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE (test_thread_flag(TIF_32BIT_ADDR) ? 0x00400000UL : 0x00400000UL) ++ ++#define PAX_DELTA_MMAP_LEN (test_thread_flag(TIF_32BIT_ADDR) ? 27-PAGE_SHIFT : 36-PAGE_SHIFT) ++#define PAX_DELTA_STACK_LEN (test_thread_flag(TIF_32BIT_ADDR) ? 27-PAGE_SHIFT : 36-PAGE_SHIFT) ++#endif ++ + #include <asm/processor.h> + #include <linux/module.h> + #include <linux/elfcore.h> +diff -urNp linux-2.6.38.3/arch/mips/kernel/binfmt_elfo32.c linux-2.6.38.3/arch/mips/kernel/binfmt_elfo32.c +--- linux-2.6.38.3/arch/mips/kernel/binfmt_elfo32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/mips/kernel/binfmt_elfo32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -52,6 +52,13 @@ typedef elf_fpreg_t elf_fpregset_t[ELF_N + #undef ELF_ET_DYN_BASE + #define ELF_ET_DYN_BASE (TASK32_SIZE / 3 * 2) + ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE (test_thread_flag(TIF_32BIT_ADDR) ? 0x00400000UL : 0x00400000UL) ++ ++#define PAX_DELTA_MMAP_LEN (test_thread_flag(TIF_32BIT_ADDR) ? 27-PAGE_SHIFT : 36-PAGE_SHIFT) ++#define PAX_DELTA_STACK_LEN (test_thread_flag(TIF_32BIT_ADDR) ? 27-PAGE_SHIFT : 36-PAGE_SHIFT) ++#endif ++ + #include <asm/processor.h> + + /* +diff -urNp linux-2.6.38.3/arch/mips/kernel/kgdb.c linux-2.6.38.3/arch/mips/kernel/kgdb.c +--- linux-2.6.38.3/arch/mips/kernel/kgdb.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/mips/kernel/kgdb.c 2011-04-17 15:57:32.000000000 -0400 +@@ -351,7 +351,7 @@ int kgdb_arch_handle_exception(int vecto + return -1; + } + +-struct kgdb_arch arch_kgdb_ops; ++struct kgdb_arch arch_kgdb_ops; /* cannot be const, see kgdb_arch_init */ + + /* + * We use kgdb_early_setup so that functions we need to call now don't +diff -urNp linux-2.6.38.3/arch/mips/kernel/process.c linux-2.6.38.3/arch/mips/kernel/process.c +--- linux-2.6.38.3/arch/mips/kernel/process.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/mips/kernel/process.c 2011-04-17 15:57:32.000000000 -0400 +@@ -473,15 +473,3 @@ unsigned long get_wchan(struct task_stru + out: + return pc; + } +- +-/* +- * Don't forget that the stack pointer must be aligned on a 8 bytes +- * boundary for 32-bits ABI and 16 bytes for 64-bits ABI. +- */ +-unsigned long arch_align_stack(unsigned long sp) +-{ +- if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space) +- sp -= get_random_int() & ~PAGE_MASK; +- +- return sp & ALMASK; +-} +diff -urNp linux-2.6.38.3/arch/mips/kernel/syscall.c linux-2.6.38.3/arch/mips/kernel/syscall.c +--- linux-2.6.38.3/arch/mips/kernel/syscall.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/mips/kernel/syscall.c 2011-04-17 15:57:32.000000000 -0400 +@@ -108,14 +108,18 @@ unsigned long arch_get_unmapped_area(str + do_color_align = 0; + if (filp || (flags & MAP_SHARED)) + do_color_align = 1; ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (!(current->mm->pax_flags & MF_PAX_RANDMMAP)) ++#endif ++ + if (addr) { + if (do_color_align) + addr = COLOUR_ALIGN(addr, pgoff); + else + addr = PAGE_ALIGN(addr); + vmm = find_vma(current->mm, addr); +- if (task_size - len >= addr && +- (!vmm || addr + len <= vmm->vm_start)) ++ if (task_size - len >= addr && check_heap_stack_gap(vmm, addr, len)) + return addr; + } + addr = current->mm->mmap_base; +@@ -128,7 +132,7 @@ unsigned long arch_get_unmapped_area(str + /* At this point: (!vmm || addr < vmm->vm_end). */ + if (task_size - len < addr) + return -ENOMEM; +- if (!vmm || addr + len <= vmm->vm_start) ++ if (check_heap_stack_gap(vmm, addr, len)) + return addr; + addr = vmm->vm_end; + if (do_color_align) +@@ -168,19 +172,6 @@ static inline unsigned long brk_rnd(void + return rnd; + } + +-unsigned long arch_randomize_brk(struct mm_struct *mm) +-{ +- unsigned long base = mm->brk; +- unsigned long ret; +- +- ret = PAGE_ALIGN(base + brk_rnd()); +- +- if (ret < mm->brk) +- return mm->brk; +- +- return ret; +-} +- + SYSCALL_DEFINE6(mips_mmap, unsigned long, addr, unsigned long, len, + unsigned long, prot, unsigned long, flags, unsigned long, + fd, off_t, offset) +diff -urNp linux-2.6.38.3/arch/mips/mm/dma-default.c linux-2.6.38.3/arch/mips/mm/dma-default.c +--- linux-2.6.38.3/arch/mips/mm/dma-default.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/mips/mm/dma-default.c 2011-04-17 15:57:32.000000000 -0400 +@@ -300,7 +300,7 @@ void dma_cache_sync(struct device *dev, + + EXPORT_SYMBOL(dma_cache_sync); + +-static struct dma_map_ops mips_default_dma_map_ops = { ++static const struct dma_map_ops mips_default_dma_map_ops = { + .alloc_coherent = mips_dma_alloc_coherent, + .free_coherent = mips_dma_free_coherent, + .map_page = mips_dma_map_page, +@@ -315,7 +315,7 @@ static struct dma_map_ops mips_default_d + .dma_supported = mips_dma_supported + }; + +-struct dma_map_ops *mips_dma_map_ops = &mips_default_dma_map_ops; ++const struct dma_map_ops *mips_dma_map_ops = &mips_default_dma_map_ops; + EXPORT_SYMBOL(mips_dma_map_ops); + + #define PREALLOC_DMA_DEBUG_ENTRIES (1 << 16) +diff -urNp linux-2.6.38.3/arch/mips/mm/fault.c linux-2.6.38.3/arch/mips/mm/fault.c +--- linux-2.6.38.3/arch/mips/mm/fault.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/mips/mm/fault.c 2011-04-17 15:57:32.000000000 -0400 +@@ -28,6 +28,23 @@ + #include <asm/highmem.h> /* For VMALLOC_END */ + #include <linux/kdebug.h> + ++#ifdef CONFIG_PAX_PAGEEXEC ++void pax_report_insns(void *pc, void *sp) ++{ ++ unsigned long i; ++ ++ printk(KERN_ERR "PAX: bytes at PC: "); ++ for (i = 0; i < 5; i++) { ++ unsigned int c; ++ if (get_user(c, (unsigned int *)pc+i)) ++ printk(KERN_CONT "???????? "); ++ else ++ printk(KERN_CONT "%08x ", c); ++ } ++ printk("\n"); ++} ++#endif ++ + /* + * This routine handles page faults. It determines the address, + * and the problem, and then passes it off to one of the appropriate +diff -urNp linux-2.6.38.3/arch/parisc/include/asm/elf.h linux-2.6.38.3/arch/parisc/include/asm/elf.h +--- linux-2.6.38.3/arch/parisc/include/asm/elf.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/parisc/include/asm/elf.h 2011-04-17 15:57:32.000000000 -0400 +@@ -342,6 +342,13 @@ struct pt_regs; /* forward declaration.. + + #define ELF_ET_DYN_BASE (TASK_UNMAPPED_BASE + 0x01000000) + ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE 0x10000UL ++ ++#define PAX_DELTA_MMAP_LEN 16 ++#define PAX_DELTA_STACK_LEN 16 ++#endif ++ + /* This yields a mask that user programs can use to figure out what + instruction set this CPU supports. This could be done in user space, + but it's not easy, and we've already done it here. */ +diff -urNp linux-2.6.38.3/arch/parisc/include/asm/pgtable.h linux-2.6.38.3/arch/parisc/include/asm/pgtable.h +--- linux-2.6.38.3/arch/parisc/include/asm/pgtable.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/parisc/include/asm/pgtable.h 2011-04-17 15:57:32.000000000 -0400 +@@ -209,6 +209,17 @@ struct vm_area_struct; + #define PAGE_EXECREAD __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_READ | _PAGE_EXEC |_PAGE_ACCESSED) + #define PAGE_COPY PAGE_EXECREAD + #define PAGE_RWX __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_READ | _PAGE_WRITE | _PAGE_EXEC |_PAGE_ACCESSED) ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++# define PAGE_SHARED_NOEXEC __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_READ | _PAGE_WRITE | _PAGE_ACCESSED) ++# define PAGE_COPY_NOEXEC __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_READ | _PAGE_ACCESSED) ++# define PAGE_READONLY_NOEXEC __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_READ | _PAGE_ACCESSED) ++#else ++# define PAGE_SHARED_NOEXEC PAGE_SHARED ++# define PAGE_COPY_NOEXEC PAGE_COPY ++# define PAGE_READONLY_NOEXEC PAGE_READONLY ++#endif ++ + #define PAGE_KERNEL __pgprot(_PAGE_KERNEL) + #define PAGE_KERNEL_RO __pgprot(_PAGE_KERNEL & ~_PAGE_WRITE) + #define PAGE_KERNEL_UNC __pgprot(_PAGE_KERNEL | _PAGE_NO_CACHE) +diff -urNp linux-2.6.38.3/arch/parisc/kernel/module.c linux-2.6.38.3/arch/parisc/kernel/module.c +--- linux-2.6.38.3/arch/parisc/kernel/module.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/parisc/kernel/module.c 2011-04-17 15:57:32.000000000 -0400 +@@ -96,16 +96,38 @@ + + /* three functions to determine where in the module core + * or init pieces the location is */ ++static inline int in_init_rx(struct module *me, void *loc) ++{ ++ return (loc >= me->module_init_rx && ++ loc < (me->module_init_rx + me->init_size_rx)); ++} ++ ++static inline int in_init_rw(struct module *me, void *loc) ++{ ++ return (loc >= me->module_init_rw && ++ loc < (me->module_init_rw + me->init_size_rw)); ++} ++ + static inline int in_init(struct module *me, void *loc) + { +- return (loc >= me->module_init && +- loc <= (me->module_init + me->init_size)); ++ return in_init_rx(me, loc) || in_init_rw(me, loc); ++} ++ ++static inline int in_core_rx(struct module *me, void *loc) ++{ ++ return (loc >= me->module_core_rx && ++ loc < (me->module_core_rx + me->core_size_rx)); ++} ++ ++static inline int in_core_rw(struct module *me, void *loc) ++{ ++ return (loc >= me->module_core_rw && ++ loc < (me->module_core_rw + me->core_size_rw)); + } + + static inline int in_core(struct module *me, void *loc) + { +- return (loc >= me->module_core && +- loc <= (me->module_core + me->core_size)); ++ return in_core_rx(me, loc) || in_core_rw(me, loc); + } + + static inline int in_local(struct module *me, void *loc) +@@ -365,13 +387,13 @@ int module_frob_arch_sections(CONST Elf_ + } + + /* align things a bit */ +- me->core_size = ALIGN(me->core_size, 16); +- me->arch.got_offset = me->core_size; +- me->core_size += gots * sizeof(struct got_entry); +- +- me->core_size = ALIGN(me->core_size, 16); +- me->arch.fdesc_offset = me->core_size; +- me->core_size += fdescs * sizeof(Elf_Fdesc); ++ me->core_size_rw = ALIGN(me->core_size_rw, 16); ++ me->arch.got_offset = me->core_size_rw; ++ me->core_size_rw += gots * sizeof(struct got_entry); ++ ++ me->core_size_rw = ALIGN(me->core_size_rw, 16); ++ me->arch.fdesc_offset = me->core_size_rw; ++ me->core_size_rw += fdescs * sizeof(Elf_Fdesc); + + me->arch.got_max = gots; + me->arch.fdesc_max = fdescs; +@@ -389,7 +411,7 @@ static Elf64_Word get_got(struct module + + BUG_ON(value == 0); + +- got = me->module_core + me->arch.got_offset; ++ got = me->module_core_rw + me->arch.got_offset; + for (i = 0; got[i].addr; i++) + if (got[i].addr == value) + goto out; +@@ -407,7 +429,7 @@ static Elf64_Word get_got(struct module + #ifdef CONFIG_64BIT + static Elf_Addr get_fdesc(struct module *me, unsigned long value) + { +- Elf_Fdesc *fdesc = me->module_core + me->arch.fdesc_offset; ++ Elf_Fdesc *fdesc = me->module_core_rw + me->arch.fdesc_offset; + + if (!value) { + printk(KERN_ERR "%s: zero OPD requested!\n", me->name); +@@ -425,7 +447,7 @@ static Elf_Addr get_fdesc(struct module + + /* Create new one */ + fdesc->addr = value; +- fdesc->gp = (Elf_Addr)me->module_core + me->arch.got_offset; ++ fdesc->gp = (Elf_Addr)me->module_core_rw + me->arch.got_offset; + return (Elf_Addr)fdesc; + } + #endif /* CONFIG_64BIT */ +@@ -849,7 +871,7 @@ register_unwind_table(struct module *me, + + table = (unsigned char *)sechdrs[me->arch.unwind_section].sh_addr; + end = table + sechdrs[me->arch.unwind_section].sh_size; +- gp = (Elf_Addr)me->module_core + me->arch.got_offset; ++ gp = (Elf_Addr)me->module_core_rw + me->arch.got_offset; + + DEBUGP("register_unwind_table(), sect = %d at 0x%p - 0x%p (gp=0x%lx)\n", + me->arch.unwind_section, table, end, gp); +diff -urNp linux-2.6.38.3/arch/parisc/kernel/sys_parisc.c linux-2.6.38.3/arch/parisc/kernel/sys_parisc.c +--- linux-2.6.38.3/arch/parisc/kernel/sys_parisc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/parisc/kernel/sys_parisc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -43,7 +43,7 @@ static unsigned long get_unshared_area(u + /* At this point: (!vma || addr < vma->vm_end). */ + if (TASK_SIZE - len < addr) + return -ENOMEM; +- if (!vma || addr + len <= vma->vm_start) ++ if (check_heap_stack_gap(vma, addr, len)) + return addr; + addr = vma->vm_end; + } +@@ -79,7 +79,7 @@ static unsigned long get_shared_area(str + /* At this point: (!vma || addr < vma->vm_end). */ + if (TASK_SIZE - len < addr) + return -ENOMEM; +- if (!vma || addr + len <= vma->vm_start) ++ if (check_heap_stack_gap(vma, addr, len)) + return addr; + addr = DCACHE_ALIGN(vma->vm_end - offset) + offset; + if (addr < vma->vm_end) /* handle wraparound */ +@@ -98,7 +98,7 @@ unsigned long arch_get_unmapped_area(str + if (flags & MAP_FIXED) + return addr; + if (!addr) +- addr = TASK_UNMAPPED_BASE; ++ addr = current->mm->mmap_base; + + if (filp) { + addr = get_shared_area(filp->f_mapping, addr, len, pgoff); +diff -urNp linux-2.6.38.3/arch/parisc/kernel/traps.c linux-2.6.38.3/arch/parisc/kernel/traps.c +--- linux-2.6.38.3/arch/parisc/kernel/traps.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/parisc/kernel/traps.c 2011-04-17 15:57:32.000000000 -0400 +@@ -733,9 +733,7 @@ void notrace handle_interruption(int cod + + down_read(¤t->mm->mmap_sem); + vma = find_vma(current->mm,regs->iaoq[0]); +- if (vma && (regs->iaoq[0] >= vma->vm_start) +- && (vma->vm_flags & VM_EXEC)) { +- ++ if (vma && (regs->iaoq[0] >= vma->vm_start)) { + fault_address = regs->iaoq[0]; + fault_space = regs->iasq[0]; + +diff -urNp linux-2.6.38.3/arch/parisc/mm/fault.c linux-2.6.38.3/arch/parisc/mm/fault.c +--- linux-2.6.38.3/arch/parisc/mm/fault.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/parisc/mm/fault.c 2011-04-17 15:57:32.000000000 -0400 +@@ -15,6 +15,7 @@ + #include <linux/sched.h> + #include <linux/interrupt.h> + #include <linux/module.h> ++#include <linux/unistd.h> + + #include <asm/uaccess.h> + #include <asm/traps.h> +@@ -52,7 +53,7 @@ DEFINE_PER_CPU(struct exception_data, ex + static unsigned long + parisc_acctyp(unsigned long code, unsigned int inst) + { +- if (code == 6 || code == 16) ++ if (code == 6 || code == 7 || code == 16) + return VM_EXEC; + + switch (inst & 0xf0000000) { +@@ -138,6 +139,116 @@ parisc_acctyp(unsigned long code, unsign + } + #endif + ++#ifdef CONFIG_PAX_PAGEEXEC ++/* ++ * PaX: decide what to do with offenders (instruction_pointer(regs) = fault address) ++ * ++ * returns 1 when task should be killed ++ * 2 when rt_sigreturn trampoline was detected ++ * 3 when unpatched PLT trampoline was detected ++ */ ++static int pax_handle_fetch_fault(struct pt_regs *regs) ++{ ++ ++#ifdef CONFIG_PAX_EMUPLT ++ int err; ++ ++ do { /* PaX: unpatched PLT emulation */ ++ unsigned int bl, depwi; ++ ++ err = get_user(bl, (unsigned int *)instruction_pointer(regs)); ++ err |= get_user(depwi, (unsigned int *)(instruction_pointer(regs)+4)); ++ ++ if (err) ++ break; ++ ++ if (bl == 0xEA9F1FDDU && depwi == 0xD6801C1EU) { ++ unsigned int ldw, bv, ldw2, addr = instruction_pointer(regs)-12; ++ ++ err = get_user(ldw, (unsigned int *)addr); ++ err |= get_user(bv, (unsigned int *)(addr+4)); ++ err |= get_user(ldw2, (unsigned int *)(addr+8)); ++ ++ if (err) ++ break; ++ ++ if (ldw == 0x0E801096U && ++ bv == 0xEAC0C000U && ++ ldw2 == 0x0E881095U) ++ { ++ unsigned int resolver, map; ++ ++ err = get_user(resolver, (unsigned int *)(instruction_pointer(regs)+8)); ++ err |= get_user(map, (unsigned int *)(instruction_pointer(regs)+12)); ++ if (err) ++ break; ++ ++ regs->gr[20] = instruction_pointer(regs)+8; ++ regs->gr[21] = map; ++ regs->gr[22] = resolver; ++ regs->iaoq[0] = resolver | 3UL; ++ regs->iaoq[1] = regs->iaoq[0] + 4; ++ return 3; ++ } ++ } ++ } while (0); ++#endif ++ ++#ifdef CONFIG_PAX_EMUTRAMP ++ ++#ifndef CONFIG_PAX_EMUSIGRT ++ if (!(current->mm->pax_flags & MF_PAX_EMUTRAMP)) ++ return 1; ++#endif ++ ++ do { /* PaX: rt_sigreturn emulation */ ++ unsigned int ldi1, ldi2, bel, nop; ++ ++ err = get_user(ldi1, (unsigned int *)instruction_pointer(regs)); ++ err |= get_user(ldi2, (unsigned int *)(instruction_pointer(regs)+4)); ++ err |= get_user(bel, (unsigned int *)(instruction_pointer(regs)+8)); ++ err |= get_user(nop, (unsigned int *)(instruction_pointer(regs)+12)); ++ ++ if (err) ++ break; ++ ++ if ((ldi1 == 0x34190000U || ldi1 == 0x34190002U) && ++ ldi2 == 0x3414015AU && ++ bel == 0xE4008200U && ++ nop == 0x08000240U) ++ { ++ regs->gr[25] = (ldi1 & 2) >> 1; ++ regs->gr[20] = __NR_rt_sigreturn; ++ regs->gr[31] = regs->iaoq[1] + 16; ++ regs->sr[0] = regs->iasq[1]; ++ regs->iaoq[0] = 0x100UL; ++ regs->iaoq[1] = regs->iaoq[0] + 4; ++ regs->iasq[0] = regs->sr[2]; ++ regs->iasq[1] = regs->sr[2]; ++ return 2; ++ } ++ } while (0); ++#endif ++ ++ return 1; ++} ++ ++void pax_report_insns(void *pc, void *sp) ++{ ++ unsigned long i; ++ ++ printk(KERN_ERR "PAX: bytes at PC: "); ++ for (i = 0; i < 5; i++) { ++ unsigned int c; ++ if (get_user(c, (unsigned int *)pc+i)) ++ printk(KERN_CONT "???????? "); ++ else ++ printk(KERN_CONT "%08x ", c); ++ } ++ printk("\n"); ++} ++#endif ++ + int fixup_exception(struct pt_regs *regs) + { + const struct exception_table_entry *fix; +@@ -192,8 +303,33 @@ good_area: + + acc_type = parisc_acctyp(code,regs->iir); + +- if ((vma->vm_flags & acc_type) != acc_type) ++ if ((vma->vm_flags & acc_type) != acc_type) { ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if ((mm->pax_flags & MF_PAX_PAGEEXEC) && (acc_type & VM_EXEC) && ++ (address & ~3UL) == instruction_pointer(regs)) ++ { ++ up_read(&mm->mmap_sem); ++ switch (pax_handle_fetch_fault(regs)) { ++ ++#ifdef CONFIG_PAX_EMUPLT ++ case 3: ++ return; ++#endif ++ ++#ifdef CONFIG_PAX_EMUTRAMP ++ case 2: ++ return; ++#endif ++ ++ } ++ pax_report_fault(regs, (void *)instruction_pointer(regs), (void *)regs->gr[30]); ++ do_group_exit(SIGKILL); ++ } ++#endif ++ + goto bad_area; ++ } + + /* + * If for any reason at all we couldn't handle the fault, make +diff -urNp linux-2.6.38.3/arch/powerpc/include/asm/device.h linux-2.6.38.3/arch/powerpc/include/asm/device.h +--- linux-2.6.38.3/arch/powerpc/include/asm/device.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/include/asm/device.h 2011-04-17 15:57:32.000000000 -0400 +@@ -17,7 +17,7 @@ struct device_node; + */ + struct dev_archdata { + /* DMA operations on that device */ +- struct dma_map_ops *dma_ops; ++ const struct dma_map_ops *dma_ops; + + /* + * When an iommu is in use, dma_data is used as a ptr to the base of the +diff -urNp linux-2.6.38.3/arch/powerpc/include/asm/dma-mapping.h linux-2.6.38.3/arch/powerpc/include/asm/dma-mapping.h +--- linux-2.6.38.3/arch/powerpc/include/asm/dma-mapping.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/include/asm/dma-mapping.h 2011-04-17 15:57:32.000000000 -0400 +@@ -66,12 +66,13 @@ static inline unsigned long device_to_ma + /* + * Available generic sets of operations + */ ++/* cannot be const */ + #ifdef CONFIG_PPC64 +-extern struct dma_map_ops dma_iommu_ops; ++extern const struct dma_map_ops dma_iommu_ops; + #endif +-extern struct dma_map_ops dma_direct_ops; ++extern const struct dma_map_ops dma_direct_ops; + +-static inline struct dma_map_ops *get_dma_ops(struct device *dev) ++static inline const struct dma_map_ops *get_dma_ops(struct device *dev) + { + /* We don't handle the NULL dev case for ISA for now. We could + * do it via an out of line call but it is not needed for now. The +@@ -84,7 +85,7 @@ static inline struct dma_map_ops *get_dm + return dev->archdata.dma_ops; + } + +-static inline void set_dma_ops(struct device *dev, struct dma_map_ops *ops) ++static inline void set_dma_ops(struct device *dev, const struct dma_map_ops *ops) + { + dev->archdata.dma_ops = ops; + } +@@ -118,7 +119,7 @@ static inline void set_dma_offset(struct + + static inline int dma_supported(struct device *dev, u64 mask) + { +- struct dma_map_ops *dma_ops = get_dma_ops(dev); ++ const struct dma_map_ops *dma_ops = get_dma_ops(dev); + + if (unlikely(dma_ops == NULL)) + return 0; +@@ -132,7 +133,7 @@ extern int dma_set_mask(struct device *d + static inline void *dma_alloc_coherent(struct device *dev, size_t size, + dma_addr_t *dma_handle, gfp_t flag) + { +- struct dma_map_ops *dma_ops = get_dma_ops(dev); ++ const struct dma_map_ops *dma_ops = get_dma_ops(dev); + void *cpu_addr; + + BUG_ON(!dma_ops); +@@ -147,7 +148,7 @@ static inline void *dma_alloc_coherent(s + static inline void dma_free_coherent(struct device *dev, size_t size, + void *cpu_addr, dma_addr_t dma_handle) + { +- struct dma_map_ops *dma_ops = get_dma_ops(dev); ++ const struct dma_map_ops *dma_ops = get_dma_ops(dev); + + BUG_ON(!dma_ops); + +@@ -158,7 +159,7 @@ static inline void dma_free_coherent(str + + static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr) + { +- struct dma_map_ops *dma_ops = get_dma_ops(dev); ++ const struct dma_map_ops *dma_ops = get_dma_ops(dev); + + if (dma_ops->mapping_error) + return dma_ops->mapping_error(dev, dma_addr); +diff -urNp linux-2.6.38.3/arch/powerpc/include/asm/elf.h linux-2.6.38.3/arch/powerpc/include/asm/elf.h +--- linux-2.6.38.3/arch/powerpc/include/asm/elf.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/include/asm/elf.h 2011-04-17 15:57:32.000000000 -0400 +@@ -178,8 +178,19 @@ typedef elf_fpreg_t elf_vsrreghalf_t32[E + the loader. We need to make sure that it is out of the way of the program + that it will "exec", and that there is sufficient room for the brk. */ + +-extern unsigned long randomize_et_dyn(unsigned long base); +-#define ELF_ET_DYN_BASE (randomize_et_dyn(0x20000000)) ++#define ELF_ET_DYN_BASE (0x20000000) ++ ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE (0x10000000UL) ++ ++#ifdef __powerpc64__ ++#define PAX_DELTA_MMAP_LEN (is_32bit_task() ? 16 : 28) ++#define PAX_DELTA_STACK_LEN (is_32bit_task() ? 16 : 28) ++#else ++#define PAX_DELTA_MMAP_LEN 15 ++#define PAX_DELTA_STACK_LEN 15 ++#endif ++#endif + + /* + * Our registers are always unsigned longs, whether we're a 32 bit +@@ -274,9 +285,6 @@ extern int arch_setup_additional_pages(s + (0x7ff >> (PAGE_SHIFT - 12)) : \ + (0x3ffff >> (PAGE_SHIFT - 12))) + +-extern unsigned long arch_randomize_brk(struct mm_struct *mm); +-#define arch_randomize_brk arch_randomize_brk +- + #endif /* __KERNEL__ */ + + /* +diff -urNp linux-2.6.38.3/arch/powerpc/include/asm/iommu.h linux-2.6.38.3/arch/powerpc/include/asm/iommu.h +--- linux-2.6.38.3/arch/powerpc/include/asm/iommu.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/include/asm/iommu.h 2011-04-17 15:57:32.000000000 -0400 +@@ -116,6 +116,9 @@ extern void iommu_init_early_iSeries(voi + extern void iommu_init_early_dart(void); + extern void iommu_init_early_pasemi(void); + ++/* dma-iommu.c */ ++extern int dma_iommu_dma_supported(struct device *dev, u64 mask); ++ + #ifdef CONFIG_PCI + extern void pci_iommu_init(void); + extern void pci_direct_iommu_init(void); +diff -urNp linux-2.6.38.3/arch/powerpc/include/asm/kmap_types.h linux-2.6.38.3/arch/powerpc/include/asm/kmap_types.h +--- linux-2.6.38.3/arch/powerpc/include/asm/kmap_types.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/include/asm/kmap_types.h 2011-04-17 15:57:32.000000000 -0400 +@@ -27,6 +27,7 @@ enum km_type { + KM_PPC_SYNC_PAGE, + KM_PPC_SYNC_ICACHE, + KM_KDB, ++ KM_CLEARPAGE, + KM_TYPE_NR + }; + +diff -urNp linux-2.6.38.3/arch/powerpc/include/asm/page_64.h linux-2.6.38.3/arch/powerpc/include/asm/page_64.h +--- linux-2.6.38.3/arch/powerpc/include/asm/page_64.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/include/asm/page_64.h 2011-04-17 15:57:32.000000000 -0400 +@@ -172,15 +172,18 @@ do { \ + * stack by default, so in the absense of a PT_GNU_STACK program header + * we turn execute permission off. + */ +-#define VM_STACK_DEFAULT_FLAGS32 (VM_READ | VM_WRITE | VM_EXEC | \ +- VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) ++#define VM_STACK_DEFAULT_FLAGS32 \ ++ (((current->personality & READ_IMPLIES_EXEC) ? VM_EXEC : 0) | \ ++ VM_READ | VM_WRITE | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) + + #define VM_STACK_DEFAULT_FLAGS64 (VM_READ | VM_WRITE | \ + VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) + ++#ifndef CONFIG_PAX_PAGEEXEC + #define VM_STACK_DEFAULT_FLAGS \ + (is_32bit_task() ? \ + VM_STACK_DEFAULT_FLAGS32 : VM_STACK_DEFAULT_FLAGS64) ++#endif + + #include <asm-generic/getorder.h> + +diff -urNp linux-2.6.38.3/arch/powerpc/include/asm/page.h linux-2.6.38.3/arch/powerpc/include/asm/page.h +--- linux-2.6.38.3/arch/powerpc/include/asm/page.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/include/asm/page.h 2011-04-17 15:57:32.000000000 -0400 +@@ -129,8 +129,9 @@ extern phys_addr_t kernstart_addr; + * and needs to be executable. This means the whole heap ends + * up being executable. + */ +-#define VM_DATA_DEFAULT_FLAGS32 (VM_READ | VM_WRITE | VM_EXEC | \ +- VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) ++#define VM_DATA_DEFAULT_FLAGS32 \ ++ (((current->personality & READ_IMPLIES_EXEC) ? VM_EXEC : 0) | \ ++ VM_READ | VM_WRITE | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) + + #define VM_DATA_DEFAULT_FLAGS64 (VM_READ | VM_WRITE | \ + VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) +@@ -158,6 +159,9 @@ extern phys_addr_t kernstart_addr; + #define is_kernel_addr(x) ((x) >= PAGE_OFFSET) + #endif + ++#define ktla_ktva(addr) (addr) ++#define ktva_ktla(addr) (addr) ++ + #ifndef __ASSEMBLY__ + + #undef STRICT_MM_TYPECHECKS +diff -urNp linux-2.6.38.3/arch/powerpc/include/asm/pci.h linux-2.6.38.3/arch/powerpc/include/asm/pci.h +--- linux-2.6.38.3/arch/powerpc/include/asm/pci.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/include/asm/pci.h 2011-04-17 15:57:32.000000000 -0400 +@@ -65,8 +65,8 @@ static inline int pci_get_legacy_ide_irq + } + + #ifdef CONFIG_PCI +-extern void set_pci_dma_ops(struct dma_map_ops *dma_ops); +-extern struct dma_map_ops *get_pci_dma_ops(void); ++extern void set_pci_dma_ops(const struct dma_map_ops *dma_ops); ++extern const struct dma_map_ops *get_pci_dma_ops(void); + #else /* CONFIG_PCI */ + #define set_pci_dma_ops(d) + #define get_pci_dma_ops() NULL +diff -urNp linux-2.6.38.3/arch/powerpc/include/asm/pgtable.h linux-2.6.38.3/arch/powerpc/include/asm/pgtable.h +--- linux-2.6.38.3/arch/powerpc/include/asm/pgtable.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/include/asm/pgtable.h 2011-04-17 15:57:32.000000000 -0400 +@@ -2,6 +2,7 @@ + #define _ASM_POWERPC_PGTABLE_H + #ifdef __KERNEL__ + ++#include <linux/const.h> + #ifndef __ASSEMBLY__ + #include <asm/processor.h> /* For TASK_SIZE */ + #include <asm/mmu.h> +diff -urNp linux-2.6.38.3/arch/powerpc/include/asm/pte-hash32.h linux-2.6.38.3/arch/powerpc/include/asm/pte-hash32.h +--- linux-2.6.38.3/arch/powerpc/include/asm/pte-hash32.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/include/asm/pte-hash32.h 2011-04-17 15:57:32.000000000 -0400 +@@ -21,6 +21,7 @@ + #define _PAGE_FILE 0x004 /* when !present: nonlinear file mapping */ + #define _PAGE_USER 0x004 /* usermode access allowed */ + #define _PAGE_GUARDED 0x008 /* G: prohibit speculative access */ ++#define _PAGE_EXEC _PAGE_GUARDED + #define _PAGE_COHERENT 0x010 /* M: enforce memory coherence (SMP systems) */ + #define _PAGE_NO_CACHE 0x020 /* I: cache inhibit */ + #define _PAGE_WRITETHRU 0x040 /* W: cache write-through */ +diff -urNp linux-2.6.38.3/arch/powerpc/include/asm/reg.h linux-2.6.38.3/arch/powerpc/include/asm/reg.h +--- linux-2.6.38.3/arch/powerpc/include/asm/reg.h 2011-03-23 17:20:06.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/include/asm/reg.h 2011-04-17 15:57:32.000000000 -0400 +@@ -191,6 +191,7 @@ + #define SPRN_DBCR 0x136 /* e300 Data Breakpoint Control Reg */ + #define SPRN_DSISR 0x012 /* Data Storage Interrupt Status Register */ + #define DSISR_NOHPTE 0x40000000 /* no translation found */ ++#define DSISR_GUARDED 0x10000000 /* fetch from guarded storage */ + #define DSISR_PROTFAULT 0x08000000 /* protection fault */ + #define DSISR_ISSTORE 0x02000000 /* access was a store */ + #define DSISR_DABRMATCH 0x00400000 /* hit data breakpoint */ +diff -urNp linux-2.6.38.3/arch/powerpc/include/asm/swiotlb.h linux-2.6.38.3/arch/powerpc/include/asm/swiotlb.h +--- linux-2.6.38.3/arch/powerpc/include/asm/swiotlb.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/include/asm/swiotlb.h 2011-04-17 15:57:32.000000000 -0400 +@@ -13,7 +13,7 @@ + + #include <linux/swiotlb.h> + +-extern struct dma_map_ops swiotlb_dma_ops; ++extern const struct dma_map_ops swiotlb_dma_ops; + + static inline void dma_mark_clean(void *addr, size_t size) {} + +diff -urNp linux-2.6.38.3/arch/powerpc/include/asm/system.h linux-2.6.38.3/arch/powerpc/include/asm/system.h +--- linux-2.6.38.3/arch/powerpc/include/asm/system.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/include/asm/system.h 2011-04-17 15:57:32.000000000 -0400 +@@ -533,7 +533,7 @@ __cmpxchg_local(volatile void *ptr, unsi + #define cmpxchg64_local(ptr, o, n) __cmpxchg64_local_generic((ptr), (o), (n)) + #endif + +-extern unsigned long arch_align_stack(unsigned long sp); ++#define arch_align_stack(x) ((x) & ~0xfUL) + + /* Used in very early kernel initialization. */ + extern unsigned long reloc_offset(void); +diff -urNp linux-2.6.38.3/arch/powerpc/include/asm/uaccess.h linux-2.6.38.3/arch/powerpc/include/asm/uaccess.h +--- linux-2.6.38.3/arch/powerpc/include/asm/uaccess.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/include/asm/uaccess.h 2011-04-17 15:57:32.000000000 -0400 +@@ -13,6 +13,8 @@ + #define VERIFY_READ 0 + #define VERIFY_WRITE 1 + ++extern void check_object_size(const void *ptr, unsigned long n, bool to); ++ + /* + * The fs value determines whether argument validity checking should be + * performed or not. If get_fs() == USER_DS, checking is performed, with +@@ -327,52 +329,6 @@ do { \ + extern unsigned long __copy_tofrom_user(void __user *to, + const void __user *from, unsigned long size); + +-#ifndef __powerpc64__ +- +-static inline unsigned long copy_from_user(void *to, +- const void __user *from, unsigned long n) +-{ +- unsigned long over; +- +- if (access_ok(VERIFY_READ, from, n)) +- return __copy_tofrom_user((__force void __user *)to, from, n); +- if ((unsigned long)from < TASK_SIZE) { +- over = (unsigned long)from + n - TASK_SIZE; +- return __copy_tofrom_user((__force void __user *)to, from, +- n - over) + over; +- } +- return n; +-} +- +-static inline unsigned long copy_to_user(void __user *to, +- const void *from, unsigned long n) +-{ +- unsigned long over; +- +- if (access_ok(VERIFY_WRITE, to, n)) +- return __copy_tofrom_user(to, (__force void __user *)from, n); +- if ((unsigned long)to < TASK_SIZE) { +- over = (unsigned long)to + n - TASK_SIZE; +- return __copy_tofrom_user(to, (__force void __user *)from, +- n - over) + over; +- } +- return n; +-} +- +-#else /* __powerpc64__ */ +- +-#define __copy_in_user(to, from, size) \ +- __copy_tofrom_user((to), (from), (size)) +- +-extern unsigned long copy_from_user(void *to, const void __user *from, +- unsigned long n); +-extern unsigned long copy_to_user(void __user *to, const void *from, +- unsigned long n); +-extern unsigned long copy_in_user(void __user *to, const void __user *from, +- unsigned long n); +- +-#endif /* __powerpc64__ */ +- + static inline unsigned long __copy_from_user_inatomic(void *to, + const void __user *from, unsigned long n) + { +@@ -396,6 +352,10 @@ static inline unsigned long __copy_from_ + if (ret == 0) + return 0; + } ++ ++ if (!__builtin_constant_p(n)) ++ check_object_size(to, n, false); ++ + return __copy_tofrom_user((__force void __user *)to, from, n); + } + +@@ -422,6 +382,10 @@ static inline unsigned long __copy_to_us + if (ret == 0) + return 0; + } ++ ++ if (!__builtin_constant_p(n)) ++ check_object_size(from, n, true); ++ + return __copy_tofrom_user(to, (__force const void __user *)from, n); + } + +@@ -439,6 +403,92 @@ static inline unsigned long __copy_to_us + return __copy_to_user_inatomic(to, from, size); + } + ++#ifndef __powerpc64__ ++ ++static inline unsigned long __must_check copy_from_user(void *to, ++ const void __user *from, unsigned long n) ++{ ++ unsigned long over; ++ ++ if ((long)n < 0) ++ return n; ++ ++ if (access_ok(VERIFY_READ, from, n)) { ++ if (!__builtin_constant_p(n)) ++ check_object_size(to, n, false); ++ return __copy_tofrom_user((__force void __user *)to, from, n); ++ } ++ if ((unsigned long)from < TASK_SIZE) { ++ over = (unsigned long)from + n - TASK_SIZE; ++ if (!__builtin_constant_p(n - over)) ++ check_object_size(to, n - over, false); ++ return __copy_tofrom_user((__force void __user *)to, from, ++ n - over) + over; ++ } ++ return n; ++} ++ ++static inline unsigned long __must_check copy_to_user(void __user *to, ++ const void *from, unsigned long n) ++{ ++ unsigned long over; ++ ++ if ((long)n < 0) ++ return n; ++ ++ if (access_ok(VERIFY_WRITE, to, n)) { ++ if (!__builtin_constant_p(n)) ++ check_object_size(from, n, true); ++ return __copy_tofrom_user(to, (__force void __user *)from, n); ++ } ++ if ((unsigned long)to < TASK_SIZE) { ++ over = (unsigned long)to + n - TASK_SIZE; ++ if (!__builtin_constant_p(n)) ++ check_object_size(from, n - over, true); ++ return __copy_tofrom_user(to, (__force void __user *)from, ++ n - over) + over; ++ } ++ return n; ++} ++ ++#else /* __powerpc64__ */ ++ ++#define __copy_in_user(to, from, size) \ ++ __copy_tofrom_user((to), (from), (size)) ++ ++static inline unsigned long __must_check copy_from_user(void *to, const void __user *from, unsigned long n) ++{ ++ if ((long)n < 0 || n > INT_MAX) ++ return n; ++ ++ if (!__builtin_constant_p(n)) ++ check_object_size(to, n, false); ++ ++ if (likely(access_ok(VERIFY_READ, from, n))) ++ n = __copy_from_user(to, from, n); ++ else ++ memset(to, 0, n); ++ return n; ++} ++ ++static inline unsigned long __must_check copy_to_user(void __user *to, const void *from, unsigned long n) ++{ ++ if ((long)n < 0 || n > INT_MAX) ++ return n; ++ ++ if (likely(access_ok(VERIFY_WRITE, to, n))) { ++ if (!__builtin_constant_p(n)) ++ check_object_size(from, n, true); ++ n = __copy_to_user(to, from, n); ++ } ++ return n; ++} ++ ++extern unsigned long copy_in_user(void __user *to, const void __user *from, ++ unsigned long n); ++ ++#endif /* __powerpc64__ */ ++ + extern unsigned long __clear_user(void __user *addr, unsigned long size); + + static inline unsigned long clear_user(void __user *addr, unsigned long size) +diff -urNp linux-2.6.38.3/arch/powerpc/kernel/dma.c linux-2.6.38.3/arch/powerpc/kernel/dma.c +--- linux-2.6.38.3/arch/powerpc/kernel/dma.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/kernel/dma.c 2011-04-17 15:57:32.000000000 -0400 +@@ -136,7 +136,7 @@ static inline void dma_direct_sync_singl + } + #endif + +-struct dma_map_ops dma_direct_ops = { ++const struct dma_map_ops dma_direct_ops = { + .alloc_coherent = dma_direct_alloc_coherent, + .free_coherent = dma_direct_free_coherent, + .map_sg = dma_direct_map_sg, +@@ -157,7 +157,7 @@ EXPORT_SYMBOL(dma_direct_ops); + + int dma_set_mask(struct device *dev, u64 dma_mask) + { +- struct dma_map_ops *dma_ops = get_dma_ops(dev); ++ const struct dma_map_ops *dma_ops = get_dma_ops(dev); + + if (ppc_md.dma_set_mask) + return ppc_md.dma_set_mask(dev, dma_mask); +diff -urNp linux-2.6.38.3/arch/powerpc/kernel/dma-iommu.c linux-2.6.38.3/arch/powerpc/kernel/dma-iommu.c +--- linux-2.6.38.3/arch/powerpc/kernel/dma-iommu.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/kernel/dma-iommu.c 2011-04-17 15:57:32.000000000 -0400 +@@ -70,7 +70,7 @@ static void dma_iommu_unmap_sg(struct de + } + + /* We support DMA to/from any memory page via the iommu */ +-static int dma_iommu_dma_supported(struct device *dev, u64 mask) ++int dma_iommu_dma_supported(struct device *dev, u64 mask) + { + struct iommu_table *tbl = get_iommu_table_base(dev); + +@@ -90,7 +90,7 @@ static int dma_iommu_dma_supported(struc + return 1; + } + +-struct dma_map_ops dma_iommu_ops = { ++struct dma_map_ops dma_iommu_ops = { /* cannot be const, see arch/powerpc/platforms/cell/iommu.c */ + .alloc_coherent = dma_iommu_alloc_coherent, + .free_coherent = dma_iommu_free_coherent, + .map_sg = dma_iommu_map_sg, +diff -urNp linux-2.6.38.3/arch/powerpc/kernel/dma-swiotlb.c linux-2.6.38.3/arch/powerpc/kernel/dma-swiotlb.c +--- linux-2.6.38.3/arch/powerpc/kernel/dma-swiotlb.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/kernel/dma-swiotlb.c 2011-04-17 15:57:32.000000000 -0400 +@@ -31,7 +31,7 @@ unsigned int ppc_swiotlb_enable; + * map_page, and unmap_page on highmem, use normal dma_ops + * for everything else. + */ +-struct dma_map_ops swiotlb_dma_ops = { ++const struct dma_map_ops swiotlb_dma_ops = { + .alloc_coherent = dma_direct_alloc_coherent, + .free_coherent = dma_direct_free_coherent, + .map_sg = swiotlb_map_sg_attrs, +diff -urNp linux-2.6.38.3/arch/powerpc/kernel/exceptions-64e.S linux-2.6.38.3/arch/powerpc/kernel/exceptions-64e.S +--- linux-2.6.38.3/arch/powerpc/kernel/exceptions-64e.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/kernel/exceptions-64e.S 2011-04-17 15:57:32.000000000 -0400 +@@ -495,6 +495,7 @@ storage_fault_common: + std r14,_DAR(r1) + std r15,_DSISR(r1) + addi r3,r1,STACK_FRAME_OVERHEAD ++ bl .save_nvgprs + mr r4,r14 + mr r5,r15 + ld r14,PACA_EXGEN+EX_R14(r13) +@@ -504,8 +505,7 @@ storage_fault_common: + cmpdi r3,0 + bne- 1f + b .ret_from_except_lite +-1: bl .save_nvgprs +- mr r5,r3 ++1: mr r5,r3 + addi r3,r1,STACK_FRAME_OVERHEAD + ld r4,_DAR(r1) + bl .bad_page_fault +diff -urNp linux-2.6.38.3/arch/powerpc/kernel/exceptions-64s.S linux-2.6.38.3/arch/powerpc/kernel/exceptions-64s.S +--- linux-2.6.38.3/arch/powerpc/kernel/exceptions-64s.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/kernel/exceptions-64s.S 2011-04-17 15:57:32.000000000 -0400 +@@ -848,10 +848,10 @@ handle_page_fault: + 11: ld r4,_DAR(r1) + ld r5,_DSISR(r1) + addi r3,r1,STACK_FRAME_OVERHEAD ++ bl .save_nvgprs + bl .do_page_fault + cmpdi r3,0 + beq+ 13f +- bl .save_nvgprs + mr r5,r3 + addi r3,r1,STACK_FRAME_OVERHEAD + lwz r4,_DAR(r1) +diff -urNp linux-2.6.38.3/arch/powerpc/kernel/ibmebus.c linux-2.6.38.3/arch/powerpc/kernel/ibmebus.c +--- linux-2.6.38.3/arch/powerpc/kernel/ibmebus.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/kernel/ibmebus.c 2011-04-17 15:57:32.000000000 -0400 +@@ -128,7 +128,7 @@ static int ibmebus_dma_supported(struct + return 1; + } + +-static struct dma_map_ops ibmebus_dma_ops = { ++static const struct dma_map_ops ibmebus_dma_ops = { + .alloc_coherent = ibmebus_alloc_coherent, + .free_coherent = ibmebus_free_coherent, + .map_sg = ibmebus_map_sg, +diff -urNp linux-2.6.38.3/arch/powerpc/kernel/kgdb.c linux-2.6.38.3/arch/powerpc/kernel/kgdb.c +--- linux-2.6.38.3/arch/powerpc/kernel/kgdb.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/kernel/kgdb.c 2011-04-17 15:57:32.000000000 -0400 +@@ -422,7 +422,7 @@ int kgdb_arch_handle_exception(int vecto + /* + * Global data + */ +-struct kgdb_arch arch_kgdb_ops = { ++const struct kgdb_arch arch_kgdb_ops = { + .gdb_bpt_instr = {0x7d, 0x82, 0x10, 0x08}, + }; + +diff -urNp linux-2.6.38.3/arch/powerpc/kernel/module_32.c linux-2.6.38.3/arch/powerpc/kernel/module_32.c +--- linux-2.6.38.3/arch/powerpc/kernel/module_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/kernel/module_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -162,7 +162,7 @@ int module_frob_arch_sections(Elf32_Ehdr + me->arch.core_plt_section = i; + } + if (!me->arch.core_plt_section || !me->arch.init_plt_section) { +- printk("Module doesn't contain .plt or .init.plt sections.\n"); ++ printk("Module %s doesn't contain .plt or .init.plt sections.\n", me->name); + return -ENOEXEC; + } + +@@ -203,11 +203,16 @@ static uint32_t do_plt_call(void *locati + + DEBUGP("Doing plt for call to 0x%x at 0x%x\n", val, (unsigned int)location); + /* Init, or core PLT? */ +- if (location >= mod->module_core +- && location < mod->module_core + mod->core_size) ++ if ((location >= mod->module_core_rx && location < mod->module_core_rx + mod->core_size_rx) || ++ (location >= mod->module_core_rw && location < mod->module_core_rw + mod->core_size_rw)) + entry = (void *)sechdrs[mod->arch.core_plt_section].sh_addr; +- else ++ else if ((location >= mod->module_init_rx && location < mod->module_init_rx + mod->init_size_rx) || ++ (location >= mod->module_init_rw && location < mod->module_init_rw + mod->init_size_rw)) + entry = (void *)sechdrs[mod->arch.init_plt_section].sh_addr; ++ else { ++ printk(KERN_ERR "%s: invalid R_PPC_REL24 entry found\n", mod->name); ++ return ~0UL; ++ } + + /* Find this entry, or if that fails, the next avail. entry */ + while (entry->jump[0]) { +diff -urNp linux-2.6.38.3/arch/powerpc/kernel/module.c linux-2.6.38.3/arch/powerpc/kernel/module.c +--- linux-2.6.38.3/arch/powerpc/kernel/module.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/kernel/module.c 2011-04-17 15:57:32.000000000 -0400 +@@ -31,11 +31,24 @@ + + LIST_HEAD(module_bug_list); + ++#ifdef CONFIG_PAX_KERNEXEC + void *module_alloc(unsigned long size) + { + if (size == 0) + return NULL; + ++ return vmalloc(size); ++} ++ ++void *module_alloc_exec(unsigned long size) ++#else ++void *module_alloc(unsigned long size) ++#endif ++ ++{ ++ if (size == 0) ++ return NULL; ++ + return vmalloc_exec(size); + } + +@@ -45,6 +58,13 @@ void module_free(struct module *mod, voi + vfree(module_region); + } + ++#ifdef CONFIG_PAX_KERNEXEC ++void module_free_exec(struct module *mod, void *module_region) ++{ ++ module_free(mod, module_region); ++} ++#endif ++ + static const Elf_Shdr *find_section(const Elf_Ehdr *hdr, + const Elf_Shdr *sechdrs, + const char *name) +diff -urNp linux-2.6.38.3/arch/powerpc/kernel/pci-common.c linux-2.6.38.3/arch/powerpc/kernel/pci-common.c +--- linux-2.6.38.3/arch/powerpc/kernel/pci-common.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/kernel/pci-common.c 2011-04-17 15:57:32.000000000 -0400 +@@ -52,14 +52,14 @@ resource_size_t isa_mem_base; + unsigned int ppc_pci_flags = 0; + + +-static struct dma_map_ops *pci_dma_ops = &dma_direct_ops; ++static const struct dma_map_ops *pci_dma_ops = &dma_direct_ops; + +-void set_pci_dma_ops(struct dma_map_ops *dma_ops) ++void set_pci_dma_ops(const struct dma_map_ops *dma_ops) + { + pci_dma_ops = dma_ops; + } + +-struct dma_map_ops *get_pci_dma_ops(void) ++const struct dma_map_ops *get_pci_dma_ops(void) + { + return pci_dma_ops; + } +diff -urNp linux-2.6.38.3/arch/powerpc/kernel/process.c linux-2.6.38.3/arch/powerpc/kernel/process.c +--- linux-2.6.38.3/arch/powerpc/kernel/process.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/kernel/process.c 2011-04-17 15:57:32.000000000 -0400 +@@ -655,8 +655,8 @@ void show_regs(struct pt_regs * regs) + * Lookup NIP late so we have the best change of getting the + * above info out without failing + */ +- printk("NIP ["REG"] %pS\n", regs->nip, (void *)regs->nip); +- printk("LR ["REG"] %pS\n", regs->link, (void *)regs->link); ++ printk("NIP ["REG"] %pA\n", regs->nip, (void *)regs->nip); ++ printk("LR ["REG"] %pA\n", regs->link, (void *)regs->link); + #endif + show_stack(current, (unsigned long *) regs->gpr[1]); + if (!user_mode(regs)) +@@ -1146,10 +1146,10 @@ void show_stack(struct task_struct *tsk, + newsp = stack[0]; + ip = stack[STACK_FRAME_LR_SAVE]; + if (!firstframe || ip != lr) { +- printk("["REG"] ["REG"] %pS", sp, ip, (void *)ip); ++ printk("["REG"] ["REG"] %pA", sp, ip, (void *)ip); + #ifdef CONFIG_FUNCTION_GRAPH_TRACER + if ((ip == rth || ip == mrth) && curr_frame >= 0) { +- printk(" (%pS)", ++ printk(" (%pA)", + (void *)current->ret_stack[curr_frame].ret); + curr_frame--; + } +@@ -1169,7 +1169,7 @@ void show_stack(struct task_struct *tsk, + struct pt_regs *regs = (struct pt_regs *) + (sp + STACK_FRAME_OVERHEAD); + lr = regs->link; +- printk("--- Exception: %lx at %pS\n LR = %pS\n", ++ printk("--- Exception: %lx at %pA\n LR = %pA\n", + regs->trap, (void *)regs->nip, (void *)lr); + firstframe = 1; + } +@@ -1244,58 +1244,3 @@ void thread_info_cache_init(void) + } + + #endif /* THREAD_SHIFT < PAGE_SHIFT */ +- +-unsigned long arch_align_stack(unsigned long sp) +-{ +- if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space) +- sp -= get_random_int() & ~PAGE_MASK; +- return sp & ~0xf; +-} +- +-static inline unsigned long brk_rnd(void) +-{ +- unsigned long rnd = 0; +- +- /* 8MB for 32bit, 1GB for 64bit */ +- if (is_32bit_task()) +- rnd = (long)(get_random_int() % (1<<(23-PAGE_SHIFT))); +- else +- rnd = (long)(get_random_int() % (1<<(30-PAGE_SHIFT))); +- +- return rnd << PAGE_SHIFT; +-} +- +-unsigned long arch_randomize_brk(struct mm_struct *mm) +-{ +- unsigned long base = mm->brk; +- unsigned long ret; +- +-#ifdef CONFIG_PPC_STD_MMU_64 +- /* +- * If we are using 1TB segments and we are allowed to randomise +- * the heap, we can put it above 1TB so it is backed by a 1TB +- * segment. Otherwise the heap will be in the bottom 1TB +- * which always uses 256MB segments and this may result in a +- * performance penalty. +- */ +- if (!is_32bit_task() && (mmu_highuser_ssize == MMU_SEGSIZE_1T)) +- base = max_t(unsigned long, mm->brk, 1UL << SID_SHIFT_1T); +-#endif +- +- ret = PAGE_ALIGN(base + brk_rnd()); +- +- if (ret < mm->brk) +- return mm->brk; +- +- return ret; +-} +- +-unsigned long randomize_et_dyn(unsigned long base) +-{ +- unsigned long ret = PAGE_ALIGN(base + brk_rnd()); +- +- if (ret < base) +- return base; +- +- return ret; +-} +diff -urNp linux-2.6.38.3/arch/powerpc/kernel/signal_32.c linux-2.6.38.3/arch/powerpc/kernel/signal_32.c +--- linux-2.6.38.3/arch/powerpc/kernel/signal_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/kernel/signal_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -858,7 +858,7 @@ int handle_rt_signal32(unsigned long sig + /* Save user registers on the stack */ + frame = &rt_sf->uc.uc_mcontext; + addr = frame; +- if (vdso32_rt_sigtramp && current->mm->context.vdso_base) { ++ if (vdso32_rt_sigtramp && current->mm->context.vdso_base != ~0UL) { + if (save_user_regs(regs, frame, 0, 1)) + goto badframe; + regs->link = current->mm->context.vdso_base + vdso32_rt_sigtramp; +diff -urNp linux-2.6.38.3/arch/powerpc/kernel/signal_64.c linux-2.6.38.3/arch/powerpc/kernel/signal_64.c +--- linux-2.6.38.3/arch/powerpc/kernel/signal_64.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/kernel/signal_64.c 2011-04-17 15:57:32.000000000 -0400 +@@ -429,7 +429,7 @@ int handle_rt_signal64(int signr, struct + current->thread.fpscr.val = 0; + + /* Set up to return from userspace. */ +- if (vdso64_rt_sigtramp && current->mm->context.vdso_base) { ++ if (vdso64_rt_sigtramp && current->mm->context.vdso_base != ~0UL) { + regs->link = current->mm->context.vdso_base + vdso64_rt_sigtramp; + } else { + err |= setup_trampoline(__NR_rt_sigreturn, &frame->tramp[0]); +diff -urNp linux-2.6.38.3/arch/powerpc/kernel/vdso.c linux-2.6.38.3/arch/powerpc/kernel/vdso.c +--- linux-2.6.38.3/arch/powerpc/kernel/vdso.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/kernel/vdso.c 2011-04-17 15:57:32.000000000 -0400 +@@ -36,6 +36,7 @@ + #include <asm/firmware.h> + #include <asm/vdso.h> + #include <asm/vdso_datapage.h> ++#include <asm/mman.h> + + #include "setup.h" + +@@ -220,7 +221,7 @@ int arch_setup_additional_pages(struct l + vdso_base = VDSO32_MBASE; + #endif + +- current->mm->context.vdso_base = 0; ++ current->mm->context.vdso_base = ~0UL; + + /* vDSO has a problem and was disabled, just don't "enable" it for the + * process +@@ -240,7 +241,7 @@ int arch_setup_additional_pages(struct l + vdso_base = get_unmapped_area(NULL, vdso_base, + (vdso_pages << PAGE_SHIFT) + + ((VDSO_ALIGNMENT - 1) & PAGE_MASK), +- 0, 0); ++ 0, MAP_PRIVATE | MAP_EXECUTABLE); + if (IS_ERR_VALUE(vdso_base)) { + rc = vdso_base; + goto fail_mmapsem; +diff -urNp linux-2.6.38.3/arch/powerpc/kernel/vio.c linux-2.6.38.3/arch/powerpc/kernel/vio.c +--- linux-2.6.38.3/arch/powerpc/kernel/vio.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/kernel/vio.c 2011-04-17 15:57:32.000000000 -0400 +@@ -605,11 +605,12 @@ static int vio_dma_iommu_dma_supported(s + return dma_iommu_ops.dma_supported(dev, mask); + } + +-struct dma_map_ops vio_dma_mapping_ops = { ++const struct dma_map_ops vio_dma_mapping_ops = { + .alloc_coherent = vio_dma_iommu_alloc_coherent, + .free_coherent = vio_dma_iommu_free_coherent, + .map_sg = vio_dma_iommu_map_sg, + .unmap_sg = vio_dma_iommu_unmap_sg, ++ .dma_supported = dma_iommu_dma_supported, + .map_page = vio_dma_iommu_map_page, + .unmap_page = vio_dma_iommu_unmap_page, + .dma_supported = vio_dma_iommu_dma_supported, +diff -urNp linux-2.6.38.3/arch/powerpc/lib/usercopy_64.c linux-2.6.38.3/arch/powerpc/lib/usercopy_64.c +--- linux-2.6.38.3/arch/powerpc/lib/usercopy_64.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/lib/usercopy_64.c 2011-04-17 15:57:32.000000000 -0400 +@@ -9,22 +9,6 @@ + #include <linux/module.h> + #include <asm/uaccess.h> + +-unsigned long copy_from_user(void *to, const void __user *from, unsigned long n) +-{ +- if (likely(access_ok(VERIFY_READ, from, n))) +- n = __copy_from_user(to, from, n); +- else +- memset(to, 0, n); +- return n; +-} +- +-unsigned long copy_to_user(void __user *to, const void *from, unsigned long n) +-{ +- if (likely(access_ok(VERIFY_WRITE, to, n))) +- n = __copy_to_user(to, from, n); +- return n; +-} +- + unsigned long copy_in_user(void __user *to, const void __user *from, + unsigned long n) + { +@@ -35,7 +19,5 @@ unsigned long copy_in_user(void __user * + return n; + } + +-EXPORT_SYMBOL(copy_from_user); +-EXPORT_SYMBOL(copy_to_user); + EXPORT_SYMBOL(copy_in_user); + +diff -urNp linux-2.6.38.3/arch/powerpc/mm/fault.c linux-2.6.38.3/arch/powerpc/mm/fault.c +--- linux-2.6.38.3/arch/powerpc/mm/fault.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/mm/fault.c 2011-04-17 15:57:32.000000000 -0400 +@@ -31,6 +31,10 @@ + #include <linux/kdebug.h> + #include <linux/perf_event.h> + #include <linux/magic.h> ++#include <linux/slab.h> ++#include <linux/pagemap.h> ++#include <linux/compiler.h> ++#include <linux/unistd.h> + + #include <asm/firmware.h> + #include <asm/page.h> +@@ -42,6 +46,7 @@ + #include <asm/tlbflush.h> + #include <asm/siginfo.h> + #include <mm/mmu_decl.h> ++#include <asm/ptrace.h> + + #ifdef CONFIG_KPROBES + static inline int notify_page_fault(struct pt_regs *regs) +@@ -65,6 +70,33 @@ static inline int notify_page_fault(stru + } + #endif + ++#ifdef CONFIG_PAX_PAGEEXEC ++/* ++ * PaX: decide what to do with offenders (regs->nip = fault address) ++ * ++ * returns 1 when task should be killed ++ */ ++static int pax_handle_fetch_fault(struct pt_regs *regs) ++{ ++ return 1; ++} ++ ++void pax_report_insns(void *pc, void *sp) ++{ ++ unsigned long i; ++ ++ printk(KERN_ERR "PAX: bytes at PC: "); ++ for (i = 0; i < 5; i++) { ++ unsigned int c; ++ if (get_user(c, (unsigned int __user *)pc+i)) ++ printk(KERN_CONT "???????? "); ++ else ++ printk(KERN_CONT "%08x ", c); ++ } ++ printk("\n"); ++} ++#endif ++ + /* + * Check whether the instruction at regs->nip is a store using + * an update addressing form which will update r1. +@@ -135,7 +167,7 @@ int __kprobes do_page_fault(struct pt_re + * indicate errors in DSISR but can validly be set in SRR1. + */ + if (trap == 0x400) +- error_code &= 0x48200000; ++ error_code &= 0x58200000; + else + is_write = error_code & DSISR_ISSTORE; + #else +@@ -258,7 +290,7 @@ good_area: + * "undefined". Of those that can be set, this is the only + * one which seems bad. + */ +- if (error_code & 0x10000000) ++ if (error_code & DSISR_GUARDED) + /* Guarded storage error. */ + goto bad_area; + #endif /* CONFIG_8xx */ +@@ -273,7 +305,7 @@ good_area: + * processors use the same I/D cache coherency mechanism + * as embedded. + */ +- if (error_code & DSISR_PROTFAULT) ++ if (error_code & (DSISR_PROTFAULT | DSISR_GUARDED)) + goto bad_area; + #endif /* CONFIG_PPC_STD_MMU */ + +@@ -342,6 +374,23 @@ bad_area: + bad_area_nosemaphore: + /* User mode accesses cause a SIGSEGV */ + if (user_mode(regs)) { ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if (mm->pax_flags & MF_PAX_PAGEEXEC) { ++#ifdef CONFIG_PPC_STD_MMU ++ if (is_exec && (error_code & (DSISR_PROTFAULT | DSISR_GUARDED))) { ++#else ++ if (is_exec && regs->nip == address) { ++#endif ++ switch (pax_handle_fetch_fault(regs)) { ++ } ++ ++ pax_report_fault(regs, (void *)regs->nip, (void *)regs->gpr[PT_R1]); ++ do_group_exit(SIGKILL); ++ } ++ } ++#endif ++ + _exception(SIGSEGV, regs, code, address); + return 0; + } +diff -urNp linux-2.6.38.3/arch/powerpc/mm/mmap_64.c linux-2.6.38.3/arch/powerpc/mm/mmap_64.c +--- linux-2.6.38.3/arch/powerpc/mm/mmap_64.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/mm/mmap_64.c 2011-04-17 15:57:32.000000000 -0400 +@@ -99,10 +99,22 @@ void arch_pick_mmap_layout(struct mm_str + */ + if (mmap_is_legacy()) { + mm->mmap_base = TASK_UNMAPPED_BASE; ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base += mm->delta_mmap; ++#endif ++ + mm->get_unmapped_area = arch_get_unmapped_area; + mm->unmap_area = arch_unmap_area; + } else { + mm->mmap_base = mmap_base(); ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base -= mm->delta_mmap + mm->delta_stack; ++#endif ++ + mm->get_unmapped_area = arch_get_unmapped_area_topdown; + mm->unmap_area = arch_unmap_area_topdown; + } +diff -urNp linux-2.6.38.3/arch/powerpc/mm/slice.c linux-2.6.38.3/arch/powerpc/mm/slice.c +--- linux-2.6.38.3/arch/powerpc/mm/slice.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/mm/slice.c 2011-04-17 15:57:32.000000000 -0400 +@@ -98,7 +98,7 @@ static int slice_area_is_free(struct mm_ + if ((mm->task_size - len) < addr) + return 0; + vma = find_vma(mm, addr); +- return (!vma || (addr + len) <= vma->vm_start); ++ return check_heap_stack_gap(vma, addr, len); + } + + static int slice_low_has_vma(struct mm_struct *mm, unsigned long slice) +@@ -256,7 +256,7 @@ full_search: + addr = _ALIGN_UP(addr + 1, 1ul << SLICE_HIGH_SHIFT); + continue; + } +- if (!vma || addr + len <= vma->vm_start) { ++ if (check_heap_stack_gap(vma, addr, len)) { + /* + * Remember the place where we stopped the search: + */ +@@ -313,10 +313,14 @@ static unsigned long slice_find_area_top + } + } + +- addr = mm->mmap_base; +- while (addr > len) { ++ if (mm->mmap_base < len) ++ addr = -ENOMEM; ++ else ++ addr = mm->mmap_base - len; ++ ++ while (!IS_ERR_VALUE(addr)) { + /* Go down by chunk size */ +- addr = _ALIGN_DOWN(addr - len, 1ul << pshift); ++ addr = _ALIGN_DOWN(addr, 1ul << pshift); + + /* Check for hit with different page size */ + mask = slice_range_to_mask(addr, len); +@@ -336,7 +340,7 @@ static unsigned long slice_find_area_top + * return with success: + */ + vma = find_vma(mm, addr); +- if (!vma || (addr + len) <= vma->vm_start) { ++ if (check_heap_stack_gap(vma, addr, len)) { + /* remember the address as a hint for next time */ + if (use_cache) + mm->free_area_cache = addr; +@@ -348,7 +352,7 @@ static unsigned long slice_find_area_top + mm->cached_hole_size = vma->vm_start - addr; + + /* try just below the current vma->vm_start */ +- addr = vma->vm_start; ++ addr = skip_heap_stack_gap(vma, len); + } + + /* +@@ -426,6 +430,11 @@ unsigned long slice_get_unmapped_area(un + if (fixed && addr > (mm->task_size - len)) + return -EINVAL; + ++#ifdef CONFIG_PAX_RANDMMAP ++ if (!fixed && (mm->pax_flags & MF_PAX_RANDMMAP)) ++ addr = 0; ++#endif ++ + /* If hint, make sure it matches our alignment restrictions */ + if (!fixed && addr) { + addr = _ALIGN_UP(addr, 1ul << pshift); +diff -urNp linux-2.6.38.3/arch/powerpc/platforms/cell/iommu.c linux-2.6.38.3/arch/powerpc/platforms/cell/iommu.c +--- linux-2.6.38.3/arch/powerpc/platforms/cell/iommu.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/platforms/cell/iommu.c 2011-04-17 15:57:32.000000000 -0400 +@@ -642,7 +642,7 @@ static int dma_fixed_dma_supported(struc + + static int dma_set_mask_and_switch(struct device *dev, u64 dma_mask); + +-struct dma_map_ops dma_iommu_fixed_ops = { ++const struct dma_map_ops dma_iommu_fixed_ops = { + .alloc_coherent = dma_fixed_alloc_coherent, + .free_coherent = dma_fixed_free_coherent, + .map_sg = dma_fixed_map_sg, +diff -urNp linux-2.6.38.3/arch/powerpc/platforms/ps3/system-bus.c linux-2.6.38.3/arch/powerpc/platforms/ps3/system-bus.c +--- linux-2.6.38.3/arch/powerpc/platforms/ps3/system-bus.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/platforms/ps3/system-bus.c 2011-04-17 15:57:32.000000000 -0400 +@@ -695,7 +695,7 @@ static int ps3_dma_supported(struct devi + return mask >= DMA_BIT_MASK(32); + } + +-static struct dma_map_ops ps3_sb_dma_ops = { ++static const struct dma_map_ops ps3_sb_dma_ops = { + .alloc_coherent = ps3_alloc_coherent, + .free_coherent = ps3_free_coherent, + .map_sg = ps3_sb_map_sg, +@@ -705,7 +705,7 @@ static struct dma_map_ops ps3_sb_dma_ops + .unmap_page = ps3_unmap_page, + }; + +-static struct dma_map_ops ps3_ioc0_dma_ops = { ++static const struct dma_map_ops ps3_ioc0_dma_ops = { + .alloc_coherent = ps3_alloc_coherent, + .free_coherent = ps3_free_coherent, + .map_sg = ps3_ioc0_map_sg, +diff -urNp linux-2.6.38.3/arch/powerpc/sysdev/ppc4xx_cpm.c linux-2.6.38.3/arch/powerpc/sysdev/ppc4xx_cpm.c +--- linux-2.6.38.3/arch/powerpc/sysdev/ppc4xx_cpm.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/powerpc/sysdev/ppc4xx_cpm.c 2011-04-17 15:57:32.000000000 -0400 +@@ -240,7 +240,7 @@ static int cpm_suspend_enter(suspend_sta + return 0; + } + +-static struct platform_suspend_ops cpm_suspend_ops = { ++static const struct platform_suspend_ops cpm_suspend_ops = { + .valid = cpm_suspend_valid, + .enter = cpm_suspend_enter, + }; +diff -urNp linux-2.6.38.3/arch/s390/include/asm/elf.h linux-2.6.38.3/arch/s390/include/asm/elf.h +--- linux-2.6.38.3/arch/s390/include/asm/elf.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/s390/include/asm/elf.h 2011-04-17 15:57:32.000000000 -0400 +@@ -162,8 +162,14 @@ extern unsigned int vdso_enabled; + the loader. We need to make sure that it is out of the way of the program + that it will "exec", and that there is sufficient room for the brk. */ + +-extern unsigned long randomize_et_dyn(unsigned long base); +-#define ELF_ET_DYN_BASE (randomize_et_dyn(STACK_TOP / 3 * 2)) ++#define ELF_ET_DYN_BASE (STACK_TOP / 3 * 2) ++ ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE (test_thread_flag(TIF_31BIT) ? 0x10000UL : 0x80000000UL) ++ ++#define PAX_DELTA_MMAP_LEN (test_thread_flag(TIF_31BIT) ? 15 : 26 ) ++#define PAX_DELTA_STACK_LEN (test_thread_flag(TIF_31BIT) ? 15 : 26 ) ++#endif + + /* This yields a mask that user programs can use to figure out what + instruction set this CPU supports. */ +@@ -222,7 +228,4 @@ struct linux_binprm; + #define ARCH_HAS_SETUP_ADDITIONAL_PAGES 1 + int arch_setup_additional_pages(struct linux_binprm *, int); + +-extern unsigned long arch_randomize_brk(struct mm_struct *mm); +-#define arch_randomize_brk arch_randomize_brk +- + #endif +diff -urNp linux-2.6.38.3/arch/s390/include/asm/system.h linux-2.6.38.3/arch/s390/include/asm/system.h +--- linux-2.6.38.3/arch/s390/include/asm/system.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/s390/include/asm/system.h 2011-04-17 15:57:32.000000000 -0400 +@@ -449,7 +449,7 @@ extern void (*_machine_restart)(char *co + extern void (*_machine_halt)(void); + extern void (*_machine_power_off)(void); + +-extern unsigned long arch_align_stack(unsigned long sp); ++#define arch_align_stack(x) ((x) & ~0xfUL) + + static inline int tprot(unsigned long addr) + { +diff -urNp linux-2.6.38.3/arch/s390/include/asm/uaccess.h linux-2.6.38.3/arch/s390/include/asm/uaccess.h +--- linux-2.6.38.3/arch/s390/include/asm/uaccess.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/s390/include/asm/uaccess.h 2011-04-17 15:57:32.000000000 -0400 +@@ -234,6 +234,10 @@ static inline unsigned long __must_check + copy_to_user(void __user *to, const void *from, unsigned long n) + { + might_fault(); ++ ++ if ((long)n < 0) ++ return n; ++ + if (access_ok(VERIFY_WRITE, to, n)) + n = __copy_to_user(to, from, n); + return n; +@@ -259,6 +263,9 @@ copy_to_user(void __user *to, const void + static inline unsigned long __must_check + __copy_from_user(void *to, const void __user *from, unsigned long n) + { ++ if ((long)n < 0) ++ return n; ++ + if (__builtin_constant_p(n) && (n <= 256)) + return uaccess.copy_from_user_small(n, from, to); + else +@@ -293,6 +300,10 @@ copy_from_user(void *to, const void __us + unsigned int sz = __compiletime_object_size(to); + + might_fault(); ++ ++ if ((long)n < 0) ++ return n; ++ + if (unlikely(sz != -1 && sz < n)) { + copy_from_user_overflow(); + return n; +diff -urNp linux-2.6.38.3/arch/s390/Kconfig linux-2.6.38.3/arch/s390/Kconfig +--- linux-2.6.38.3/arch/s390/Kconfig 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/s390/Kconfig 2011-04-17 15:57:32.000000000 -0400 +@@ -233,11 +233,9 @@ config S390_EXEC_PROTECT + prompt "Data execute protection" + help + This option allows to enable a buffer overflow protection for user +- space programs and it also selects the addressing mode option above. +- The kernel parameter noexec=on will enable this feature and also +- switch the addressing modes, default is disabled. Enabling this (via +- kernel parameter) on machines earlier than IBM System z9 this will +- reduce system performance. ++ space programs. ++ Enabling this (via kernel parameter) on machines earlier than IBM ++ System z9 this will reduce system performance. + + comment "Code generation options" + +diff -urNp linux-2.6.38.3/arch/s390/kernel/module.c linux-2.6.38.3/arch/s390/kernel/module.c +--- linux-2.6.38.3/arch/s390/kernel/module.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/s390/kernel/module.c 2011-04-17 15:57:32.000000000 -0400 +@@ -168,11 +168,11 @@ module_frob_arch_sections(Elf_Ehdr *hdr, + + /* Increase core size by size of got & plt and set start + offsets for got and plt. */ +- me->core_size = ALIGN(me->core_size, 4); +- me->arch.got_offset = me->core_size; +- me->core_size += me->arch.got_size; +- me->arch.plt_offset = me->core_size; +- me->core_size += me->arch.plt_size; ++ me->core_size_rw = ALIGN(me->core_size_rw, 4); ++ me->arch.got_offset = me->core_size_rw; ++ me->core_size_rw += me->arch.got_size; ++ me->arch.plt_offset = me->core_size_rx; ++ me->core_size_rx += me->arch.plt_size; + return 0; + } + +@@ -258,7 +258,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base + if (info->got_initialized == 0) { + Elf_Addr *gotent; + +- gotent = me->module_core + me->arch.got_offset + ++ gotent = me->module_core_rw + me->arch.got_offset + + info->got_offset; + *gotent = val; + info->got_initialized = 1; +@@ -282,7 +282,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base + else if (r_type == R_390_GOTENT || + r_type == R_390_GOTPLTENT) + *(unsigned int *) loc = +- (val + (Elf_Addr) me->module_core - loc) >> 1; ++ (val + (Elf_Addr) me->module_core_rw - loc) >> 1; + else if (r_type == R_390_GOT64 || + r_type == R_390_GOTPLT64) + *(unsigned long *) loc = val; +@@ -296,7 +296,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base + case R_390_PLTOFF64: /* 16 bit offset from GOT to PLT. */ + if (info->plt_initialized == 0) { + unsigned int *ip; +- ip = me->module_core + me->arch.plt_offset + ++ ip = me->module_core_rx + me->arch.plt_offset + + info->plt_offset; + #ifndef CONFIG_64BIT + ip[0] = 0x0d105810; /* basr 1,0; l 1,6(1); br 1 */ +@@ -321,7 +321,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base + val - loc + 0xffffUL < 0x1ffffeUL) || + (r_type == R_390_PLT32DBL && + val - loc + 0xffffffffULL < 0x1fffffffeULL))) +- val = (Elf_Addr) me->module_core + ++ val = (Elf_Addr) me->module_core_rx + + me->arch.plt_offset + + info->plt_offset; + val += rela->r_addend - loc; +@@ -343,7 +343,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base + case R_390_GOTOFF32: /* 32 bit offset to GOT. */ + case R_390_GOTOFF64: /* 64 bit offset to GOT. */ + val = val + rela->r_addend - +- ((Elf_Addr) me->module_core + me->arch.got_offset); ++ ((Elf_Addr) me->module_core_rw + me->arch.got_offset); + if (r_type == R_390_GOTOFF16) + *(unsigned short *) loc = val; + else if (r_type == R_390_GOTOFF32) +@@ -353,7 +353,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base + break; + case R_390_GOTPC: /* 32 bit PC relative offset to GOT. */ + case R_390_GOTPCDBL: /* 32 bit PC rel. off. to GOT shifted by 1. */ +- val = (Elf_Addr) me->module_core + me->arch.got_offset + ++ val = (Elf_Addr) me->module_core_rw + me->arch.got_offset + + rela->r_addend - loc; + if (r_type == R_390_GOTPC) + *(unsigned int *) loc = val; +diff -urNp linux-2.6.38.3/arch/s390/kernel/process.c linux-2.6.38.3/arch/s390/kernel/process.c +--- linux-2.6.38.3/arch/s390/kernel/process.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/s390/kernel/process.c 2011-04-17 15:57:32.000000000 -0400 +@@ -334,39 +334,3 @@ unsigned long get_wchan(struct task_stru + } + return 0; + } +- +-unsigned long arch_align_stack(unsigned long sp) +-{ +- if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space) +- sp -= get_random_int() & ~PAGE_MASK; +- return sp & ~0xf; +-} +- +-static inline unsigned long brk_rnd(void) +-{ +- /* 8MB for 32bit, 1GB for 64bit */ +- if (is_32bit_task()) +- return (get_random_int() & 0x7ffUL) << PAGE_SHIFT; +- else +- return (get_random_int() & 0x3ffffUL) << PAGE_SHIFT; +-} +- +-unsigned long arch_randomize_brk(struct mm_struct *mm) +-{ +- unsigned long ret = PAGE_ALIGN(mm->brk + brk_rnd()); +- +- if (ret < mm->brk) +- return mm->brk; +- return ret; +-} +- +-unsigned long randomize_et_dyn(unsigned long base) +-{ +- unsigned long ret = PAGE_ALIGN(base + brk_rnd()); +- +- if (!(current->flags & PF_RANDOMIZE)) +- return base; +- if (ret < base) +- return base; +- return ret; +-} +diff -urNp linux-2.6.38.3/arch/s390/kernel/setup.c linux-2.6.38.3/arch/s390/kernel/setup.c +--- linux-2.6.38.3/arch/s390/kernel/setup.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/s390/kernel/setup.c 2011-04-17 15:57:32.000000000 -0400 +@@ -281,7 +281,7 @@ static int __init early_parse_mem(char * + } + early_param("mem", early_parse_mem); + +-unsigned int user_mode = HOME_SPACE_MODE; ++unsigned int user_mode = SECONDARY_SPACE_MODE; + EXPORT_SYMBOL_GPL(user_mode); + + static int set_amode_and_uaccess(unsigned long user_amode, +@@ -310,17 +310,6 @@ static int set_amode_and_uaccess(unsigne + } + } + +-/* +- * Switch kernel/user addressing modes? +- */ +-static int __init early_parse_switch_amode(char *p) +-{ +- if (user_mode != SECONDARY_SPACE_MODE) +- user_mode = PRIMARY_SPACE_MODE; +- return 0; +-} +-early_param("switch_amode", early_parse_switch_amode); +- + static int __init early_parse_user_mode(char *p) + { + if (p && strcmp(p, "primary") == 0) +@@ -337,20 +326,6 @@ static int __init early_parse_user_mode( + } + early_param("user_mode", early_parse_user_mode); + +-#ifdef CONFIG_S390_EXEC_PROTECT +-/* +- * Enable execute protection? +- */ +-static int __init early_parse_noexec(char *p) +-{ +- if (!strncmp(p, "off", 3)) +- return 0; +- user_mode = SECONDARY_SPACE_MODE; +- return 0; +-} +-early_param("noexec", early_parse_noexec); +-#endif /* CONFIG_S390_EXEC_PROTECT */ +- + static void setup_addressing_mode(void) + { + if (user_mode == SECONDARY_SPACE_MODE) { +diff -urNp linux-2.6.38.3/arch/s390/mm/maccess.c linux-2.6.38.3/arch/s390/mm/maccess.c +--- linux-2.6.38.3/arch/s390/mm/maccess.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/s390/mm/maccess.c 2011-04-17 15:57:32.000000000 -0400 +@@ -45,7 +45,7 @@ static long probe_kernel_write_odd(void + return rc ? rc : count; + } + +-long probe_kernel_write(void *dst, void *src, size_t size) ++long probe_kernel_write(void *dst, const void *src, size_t size) + { + long copied = 0; + +diff -urNp linux-2.6.38.3/arch/s390/mm/mmap.c linux-2.6.38.3/arch/s390/mm/mmap.c +--- linux-2.6.38.3/arch/s390/mm/mmap.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/s390/mm/mmap.c 2011-04-17 15:57:32.000000000 -0400 +@@ -91,10 +91,22 @@ void arch_pick_mmap_layout(struct mm_str + */ + if (mmap_is_legacy()) { + mm->mmap_base = TASK_UNMAPPED_BASE; ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base += mm->delta_mmap; ++#endif ++ + mm->get_unmapped_area = arch_get_unmapped_area; + mm->unmap_area = arch_unmap_area; + } else { + mm->mmap_base = mmap_base(); ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base -= mm->delta_mmap + mm->delta_stack; ++#endif ++ + mm->get_unmapped_area = arch_get_unmapped_area_topdown; + mm->unmap_area = arch_unmap_area_topdown; + } +@@ -166,10 +178,22 @@ void arch_pick_mmap_layout(struct mm_str + */ + if (mmap_is_legacy()) { + mm->mmap_base = TASK_UNMAPPED_BASE; ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base += mm->delta_mmap; ++#endif ++ + mm->get_unmapped_area = s390_get_unmapped_area; + mm->unmap_area = arch_unmap_area; + } else { + mm->mmap_base = mmap_base(); ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base -= mm->delta_mmap + mm->delta_stack; ++#endif ++ + mm->get_unmapped_area = s390_get_unmapped_area_topdown; + mm->unmap_area = arch_unmap_area_topdown; + } +diff -urNp linux-2.6.38.3/arch/score/include/asm/system.h linux-2.6.38.3/arch/score/include/asm/system.h +--- linux-2.6.38.3/arch/score/include/asm/system.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/score/include/asm/system.h 2011-04-17 15:57:32.000000000 -0400 +@@ -17,7 +17,7 @@ do { \ + #define finish_arch_switch(prev) do {} while (0) + + typedef void (*vi_handler_t)(void); +-extern unsigned long arch_align_stack(unsigned long sp); ++#define arch_align_stack(x) (x) + + #define mb() barrier() + #define rmb() barrier() +diff -urNp linux-2.6.38.3/arch/score/kernel/process.c linux-2.6.38.3/arch/score/kernel/process.c +--- linux-2.6.38.3/arch/score/kernel/process.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/score/kernel/process.c 2011-04-17 15:57:32.000000000 -0400 +@@ -161,8 +161,3 @@ unsigned long get_wchan(struct task_stru + + return task_pt_regs(task)->cp0_epc; + } +- +-unsigned long arch_align_stack(unsigned long sp) +-{ +- return sp; +-} +diff -urNp linux-2.6.38.3/arch/sh/include/asm/dma-mapping.h linux-2.6.38.3/arch/sh/include/asm/dma-mapping.h +--- linux-2.6.38.3/arch/sh/include/asm/dma-mapping.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sh/include/asm/dma-mapping.h 2011-04-17 15:57:32.000000000 -0400 +@@ -1,10 +1,10 @@ + #ifndef __ASM_SH_DMA_MAPPING_H + #define __ASM_SH_DMA_MAPPING_H + +-extern struct dma_map_ops *dma_ops; ++extern const struct dma_map_ops *dma_ops; + extern void no_iommu_init(void); + +-static inline struct dma_map_ops *get_dma_ops(struct device *dev) ++static inline const struct dma_map_ops *get_dma_ops(struct device *dev) + { + return dma_ops; + } +@@ -14,7 +14,7 @@ static inline struct dma_map_ops *get_dm + + static inline int dma_supported(struct device *dev, u64 mask) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + + if (ops->dma_supported) + return ops->dma_supported(dev, mask); +@@ -24,7 +24,7 @@ static inline int dma_supported(struct d + + static inline int dma_set_mask(struct device *dev, u64 mask) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + + if (!dev->dma_mask || !dma_supported(dev, mask)) + return -EIO; +@@ -44,7 +44,7 @@ void dma_cache_sync(struct device *dev, + + static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + + if (ops->mapping_error) + return ops->mapping_error(dev, dma_addr); +@@ -55,7 +55,7 @@ static inline int dma_mapping_error(stru + static inline void *dma_alloc_coherent(struct device *dev, size_t size, + dma_addr_t *dma_handle, gfp_t gfp) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + void *memory; + + if (dma_alloc_from_coherent(dev, size, dma_handle, &memory)) +@@ -72,7 +72,7 @@ static inline void *dma_alloc_coherent(s + static inline void dma_free_coherent(struct device *dev, size_t size, + void *vaddr, dma_addr_t dma_handle) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + + if (dma_release_from_coherent(dev, get_order(size), vaddr)) + return; +diff -urNp linux-2.6.38.3/arch/sh/kernel/dma-nommu.c linux-2.6.38.3/arch/sh/kernel/dma-nommu.c +--- linux-2.6.38.3/arch/sh/kernel/dma-nommu.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sh/kernel/dma-nommu.c 2011-04-17 15:57:32.000000000 -0400 +@@ -62,7 +62,7 @@ static void nommu_sync_sg(struct device + } + #endif + +-struct dma_map_ops nommu_dma_ops = { ++const struct dma_map_ops nommu_dma_ops = { + .alloc_coherent = dma_generic_alloc_coherent, + .free_coherent = dma_generic_free_coherent, + .map_page = nommu_map_page, +diff -urNp linux-2.6.38.3/arch/sh/kernel/kgdb.c linux-2.6.38.3/arch/sh/kernel/kgdb.c +--- linux-2.6.38.3/arch/sh/kernel/kgdb.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sh/kernel/kgdb.c 2011-04-17 15:57:32.000000000 -0400 +@@ -319,7 +319,7 @@ void kgdb_arch_exit(void) + unregister_die_notifier(&kgdb_notifier); + } + +-struct kgdb_arch arch_kgdb_ops = { ++const struct kgdb_arch arch_kgdb_ops = { + /* Breakpoint instruction: trapa #0x3c */ + #ifdef CONFIG_CPU_LITTLE_ENDIAN + .gdb_bpt_instr = { 0x3c, 0xc3 }, +diff -urNp linux-2.6.38.3/arch/sh/mm/consistent.c linux-2.6.38.3/arch/sh/mm/consistent.c +--- linux-2.6.38.3/arch/sh/mm/consistent.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sh/mm/consistent.c 2011-04-17 15:57:32.000000000 -0400 +@@ -22,7 +22,7 @@ + + #define PREALLOC_DMA_DEBUG_ENTRIES 4096 + +-struct dma_map_ops *dma_ops; ++const struct dma_map_ops *dma_ops; + EXPORT_SYMBOL(dma_ops); + + static int __init dma_init(void) +diff -urNp linux-2.6.38.3/arch/sh/mm/mmap.c linux-2.6.38.3/arch/sh/mm/mmap.c +--- linux-2.6.38.3/arch/sh/mm/mmap.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sh/mm/mmap.c 2011-04-17 15:57:32.000000000 -0400 +@@ -74,8 +74,7 @@ unsigned long arch_get_unmapped_area(str + addr = PAGE_ALIGN(addr); + + vma = find_vma(mm, addr); +- if (TASK_SIZE - len >= addr && +- (!vma || addr + len <= vma->vm_start)) ++ if (TASK_SIZE - len >= addr && check_heap_stack_gap(vma, addr, len)) + return addr; + } + +@@ -106,7 +105,7 @@ full_search: + } + return -ENOMEM; + } +- if (likely(!vma || addr + len <= vma->vm_start)) { ++ if (likely(check_heap_stack_gap(vma, addr, len))) { + /* + * Remember the place where we stopped the search: + */ +@@ -157,8 +156,7 @@ arch_get_unmapped_area_topdown(struct fi + addr = PAGE_ALIGN(addr); + + vma = find_vma(mm, addr); +- if (TASK_SIZE - len >= addr && +- (!vma || addr + len <= vma->vm_start)) ++ if (TASK_SIZE - len >= addr && check_heap_stack_gap(vma, addr, len)) + return addr; + } + +@@ -179,7 +177,7 @@ arch_get_unmapped_area_topdown(struct fi + /* make sure it can fit in the remaining address space */ + if (likely(addr > len)) { + vma = find_vma(mm, addr-len); +- if (!vma || addr <= vma->vm_start) { ++ if (check_heap_stack_gap(vma, addr - len, len)) { + /* remember the address as a hint for next time */ + return (mm->free_area_cache = addr-len); + } +@@ -188,18 +186,18 @@ arch_get_unmapped_area_topdown(struct fi + if (unlikely(mm->mmap_base < len)) + goto bottomup; + +- addr = mm->mmap_base-len; +- if (do_colour_align) +- addr = COLOUR_ALIGN_DOWN(addr, pgoff); ++ addr = mm->mmap_base - len; + + do { ++ if (do_colour_align) ++ addr = COLOUR_ALIGN_DOWN(addr, pgoff); + /* + * Lookup failure means no vma is above this address, + * else if new region fits below vma->vm_start, + * return with success: + */ + vma = find_vma(mm, addr); +- if (likely(!vma || addr+len <= vma->vm_start)) { ++ if (likely(check_heap_stack_gap(vma, addr, len))) { + /* remember the address as a hint for next time */ + return (mm->free_area_cache = addr); + } +@@ -209,10 +207,8 @@ arch_get_unmapped_area_topdown(struct fi + mm->cached_hole_size = vma->vm_start - addr; + + /* try just below the current vma->vm_start */ +- addr = vma->vm_start-len; +- if (do_colour_align) +- addr = COLOUR_ALIGN_DOWN(addr, pgoff); +- } while (likely(len < vma->vm_start)); ++ addr = skip_heap_stack_gap(vma, len); ++ } while (!IS_ERR_VALUE(addr)); + + bottomup: + /* +diff -urNp linux-2.6.38.3/arch/sparc/include/asm/atomic_64.h linux-2.6.38.3/arch/sparc/include/asm/atomic_64.h +--- linux-2.6.38.3/arch/sparc/include/asm/atomic_64.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/include/asm/atomic_64.h 2011-04-17 15:57:32.000000000 -0400 +@@ -14,18 +14,40 @@ + #define ATOMIC64_INIT(i) { (i) } + + #define atomic_read(v) (*(volatile int *)&(v)->counter) ++static inline int atomic_read_unchecked(const atomic_unchecked_t *v) ++{ ++ return v->counter; ++} + #define atomic64_read(v) (*(volatile long *)&(v)->counter) ++static inline long atomic64_read_unchecked(const atomic64_unchecked_t *v) ++{ ++ return v->counter; ++} + + #define atomic_set(v, i) (((v)->counter) = i) ++static inline void atomic_set_unchecked(atomic_unchecked_t *v, int i) ++{ ++ v->counter = i; ++} + #define atomic64_set(v, i) (((v)->counter) = i) ++static inline void atomic64_set_unchecked(atomic64_unchecked_t *v, long i) ++{ ++ v->counter = i; ++} + + extern void atomic_add(int, atomic_t *); ++extern void atomic_add_unchecked(int, atomic_unchecked_t *); + extern void atomic64_add(long, atomic64_t *); ++extern void atomic64_add_unchecked(long, atomic64_unchecked_t *); + extern void atomic_sub(int, atomic_t *); ++extern void atomic_sub_unchecked(int, atomic_unchecked_t *); + extern void atomic64_sub(long, atomic64_t *); ++extern void atomic64_sub_unchecked(long, atomic64_unchecked_t *); + + extern int atomic_add_ret(int, atomic_t *); ++extern int atomic_add_ret_unchecked(int, atomic_unchecked_t *); + extern long atomic64_add_ret(long, atomic64_t *); ++extern long atomic64_add_ret_unchecked(long, atomic64_unchecked_t *); + extern int atomic_sub_ret(int, atomic_t *); + extern long atomic64_sub_ret(long, atomic64_t *); + +@@ -33,12 +55,24 @@ extern long atomic64_sub_ret(long, atomi + #define atomic64_dec_return(v) atomic64_sub_ret(1, v) + + #define atomic_inc_return(v) atomic_add_ret(1, v) ++static inline int atomic_inc_return_unchecked(atomic_unchecked_t *v) ++{ ++ return atomic_add_ret_unchecked(1, v); ++} + #define atomic64_inc_return(v) atomic64_add_ret(1, v) ++static inline long atomic64_inc_return_unchecked(atomic64_unchecked_t *v) ++{ ++ return atomic64_add_ret_unchecked(1, v); ++} + + #define atomic_sub_return(i, v) atomic_sub_ret(i, v) + #define atomic64_sub_return(i, v) atomic64_sub_ret(i, v) + + #define atomic_add_return(i, v) atomic_add_ret(i, v) ++static inline int atomic_add_return_unchecked(int i, atomic_unchecked_t *v) ++{ ++ return atomic_add_ret_unchecked(i, v); ++} + #define atomic64_add_return(i, v) atomic64_add_ret(i, v) + + /* +@@ -59,10 +93,26 @@ extern long atomic64_sub_ret(long, atomi + #define atomic64_dec_and_test(v) (atomic64_sub_ret(1, v) == 0) + + #define atomic_inc(v) atomic_add(1, v) ++static inline void atomic_inc_unchecked(atomic_unchecked_t *v) ++{ ++ atomic_add_unchecked(1, v); ++} + #define atomic64_inc(v) atomic64_add(1, v) ++static inline void atomic64_inc_unchecked(atomic64_unchecked_t *v) ++{ ++ atomic64_add_unchecked(1, v); ++} + + #define atomic_dec(v) atomic_sub(1, v) ++static inline void atomic_dec_unchecked(atomic_unchecked_t *v) ++{ ++ atomic_sub_unchecked(1, v); ++} + #define atomic64_dec(v) atomic64_sub(1, v) ++static inline void atomic64_dec_unchecked(atomic64_unchecked_t *v) ++{ ++ atomic64_sub_unchecked(1, v); ++} + + #define atomic_add_negative(i, v) (atomic_add_ret(i, v) < 0) + #define atomic64_add_negative(i, v) (atomic64_add_ret(i, v) < 0) +@@ -72,17 +122,28 @@ extern long atomic64_sub_ret(long, atomi + + static inline int atomic_add_unless(atomic_t *v, int a, int u) + { +- int c, old; ++ int c, old, new; + c = atomic_read(v); + for (;;) { +- if (unlikely(c == (u))) ++ if (unlikely(c == u)) + break; +- old = atomic_cmpxchg((v), c, c + (a)); ++ ++ asm volatile("addcc %2, %0, %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "tvs %%icc, 6\n" ++#endif ++ ++ : "=r" (new) ++ : "0" (c), "ir" (a) ++ : "cc"); ++ ++ old = atomic_cmpxchg(v, c, new); + if (likely(old == c)) + break; + c = old; + } +- return c != (u); ++ return c != u; + } + + #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0) +@@ -93,17 +154,28 @@ static inline int atomic_add_unless(atom + + static inline long atomic64_add_unless(atomic64_t *v, long a, long u) + { +- long c, old; ++ long c, old, new; + c = atomic64_read(v); + for (;;) { +- if (unlikely(c == (u))) ++ if (unlikely(c == u)) + break; +- old = atomic64_cmpxchg((v), c, c + (a)); ++ ++ asm volatile("addcc %2, %0, %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "tvs %%xcc, 6\n" ++#endif ++ ++ : "=r" (new) ++ : "0" (c), "ir" (a) ++ : "cc"); ++ ++ old = atomic64_cmpxchg(v, c, new); + if (likely(old == c)) + break; + c = old; + } +- return c != (u); ++ return c != u; + } + + #define atomic64_inc_not_zero(v) atomic64_add_unless((v), 1, 0) +diff -urNp linux-2.6.38.3/arch/sparc/include/asm/dma-mapping.h linux-2.6.38.3/arch/sparc/include/asm/dma-mapping.h +--- linux-2.6.38.3/arch/sparc/include/asm/dma-mapping.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/include/asm/dma-mapping.h 2011-04-17 15:57:32.000000000 -0400 +@@ -12,10 +12,10 @@ extern int dma_supported(struct device * + #define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f) + #define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h) + +-extern struct dma_map_ops *dma_ops, pci32_dma_ops; ++extern const struct dma_map_ops *dma_ops, pci32_dma_ops; + extern struct bus_type pci_bus_type; + +-static inline struct dma_map_ops *get_dma_ops(struct device *dev) ++static inline const struct dma_map_ops *get_dma_ops(struct device *dev) + { + #if defined(CONFIG_SPARC32) && defined(CONFIG_PCI) + if (dev->bus == &pci_bus_type) +@@ -29,7 +29,7 @@ static inline struct dma_map_ops *get_dm + static inline void *dma_alloc_coherent(struct device *dev, size_t size, + dma_addr_t *dma_handle, gfp_t flag) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + void *cpu_addr; + + cpu_addr = ops->alloc_coherent(dev, size, dma_handle, flag); +@@ -40,7 +40,7 @@ static inline void *dma_alloc_coherent(s + static inline void dma_free_coherent(struct device *dev, size_t size, + void *cpu_addr, dma_addr_t dma_handle) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + + debug_dma_free_coherent(dev, size, cpu_addr, dma_handle); + ops->free_coherent(dev, size, cpu_addr, dma_handle); +diff -urNp linux-2.6.38.3/arch/sparc/include/asm/elf_32.h linux-2.6.38.3/arch/sparc/include/asm/elf_32.h +--- linux-2.6.38.3/arch/sparc/include/asm/elf_32.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/include/asm/elf_32.h 2011-04-17 15:57:32.000000000 -0400 +@@ -114,6 +114,13 @@ typedef struct { + + #define ELF_ET_DYN_BASE (TASK_UNMAPPED_BASE) + ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE 0x10000UL ++ ++#define PAX_DELTA_MMAP_LEN 16 ++#define PAX_DELTA_STACK_LEN 16 ++#endif ++ + /* This yields a mask that user programs can use to figure out what + instruction set this cpu supports. This can NOT be done in userspace + on Sparc. */ +diff -urNp linux-2.6.38.3/arch/sparc/include/asm/elf_64.h linux-2.6.38.3/arch/sparc/include/asm/elf_64.h +--- linux-2.6.38.3/arch/sparc/include/asm/elf_64.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/include/asm/elf_64.h 2011-04-17 15:57:32.000000000 -0400 +@@ -162,6 +162,12 @@ typedef struct { + #define ELF_ET_DYN_BASE 0x0000010000000000UL + #define COMPAT_ELF_ET_DYN_BASE 0x0000000070000000UL + ++#ifdef CONFIG_PAX_ASLR ++#define PAX_ELF_ET_DYN_BASE (test_thread_flag(TIF_32BIT) ? 0x10000UL : 0x100000UL) ++ ++#define PAX_DELTA_MMAP_LEN (test_thread_flag(TIF_32BIT) ? 14 : 28) ++#define PAX_DELTA_STACK_LEN (test_thread_flag(TIF_32BIT) ? 15 : 29) ++#endif + + /* This yields a mask that user programs can use to figure out what + instruction set this cpu supports. */ +diff -urNp linux-2.6.38.3/arch/sparc/include/asm/pgtable_32.h linux-2.6.38.3/arch/sparc/include/asm/pgtable_32.h +--- linux-2.6.38.3/arch/sparc/include/asm/pgtable_32.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/include/asm/pgtable_32.h 2011-04-17 15:57:32.000000000 -0400 +@@ -43,6 +43,13 @@ BTFIXUPDEF_SIMM13(user_ptrs_per_pgd) + BTFIXUPDEF_INT(page_none) + BTFIXUPDEF_INT(page_copy) + BTFIXUPDEF_INT(page_readonly) ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++BTFIXUPDEF_INT(page_shared_noexec) ++BTFIXUPDEF_INT(page_copy_noexec) ++BTFIXUPDEF_INT(page_readonly_noexec) ++#endif ++ + BTFIXUPDEF_INT(page_kernel) + + #define PMD_SHIFT SUN4C_PMD_SHIFT +@@ -64,6 +71,16 @@ extern pgprot_t PAGE_SHARED; + #define PAGE_COPY __pgprot(BTFIXUP_INT(page_copy)) + #define PAGE_READONLY __pgprot(BTFIXUP_INT(page_readonly)) + ++#ifdef CONFIG_PAX_PAGEEXEC ++extern pgprot_t PAGE_SHARED_NOEXEC; ++# define PAGE_COPY_NOEXEC __pgprot(BTFIXUP_INT(page_copy_noexec)) ++# define PAGE_READONLY_NOEXEC __pgprot(BTFIXUP_INT(page_readonly_noexec)) ++#else ++# define PAGE_SHARED_NOEXEC PAGE_SHARED ++# define PAGE_COPY_NOEXEC PAGE_COPY ++# define PAGE_READONLY_NOEXEC PAGE_READONLY ++#endif ++ + extern unsigned long page_kernel; + + #ifdef MODULE +diff -urNp linux-2.6.38.3/arch/sparc/include/asm/pgtsrmmu.h linux-2.6.38.3/arch/sparc/include/asm/pgtsrmmu.h +--- linux-2.6.38.3/arch/sparc/include/asm/pgtsrmmu.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/include/asm/pgtsrmmu.h 2011-04-17 15:57:32.000000000 -0400 +@@ -115,6 +115,13 @@ + SRMMU_EXEC | SRMMU_REF) + #define SRMMU_PAGE_RDONLY __pgprot(SRMMU_VALID | SRMMU_CACHE | \ + SRMMU_EXEC | SRMMU_REF) ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++#define SRMMU_PAGE_SHARED_NOEXEC __pgprot(SRMMU_VALID | SRMMU_CACHE | SRMMU_WRITE | SRMMU_REF) ++#define SRMMU_PAGE_COPY_NOEXEC __pgprot(SRMMU_VALID | SRMMU_CACHE | SRMMU_REF) ++#define SRMMU_PAGE_RDONLY_NOEXEC __pgprot(SRMMU_VALID | SRMMU_CACHE | SRMMU_REF) ++#endif ++ + #define SRMMU_PAGE_KERNEL __pgprot(SRMMU_VALID | SRMMU_CACHE | SRMMU_PRIV | \ + SRMMU_DIRTY | SRMMU_REF) + +diff -urNp linux-2.6.38.3/arch/sparc/include/asm/spinlock_64.h linux-2.6.38.3/arch/sparc/include/asm/spinlock_64.h +--- linux-2.6.38.3/arch/sparc/include/asm/spinlock_64.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/include/asm/spinlock_64.h 2011-04-17 15:57:32.000000000 -0400 +@@ -99,7 +99,12 @@ static void inline arch_read_lock(arch_r + __asm__ __volatile__ ( + "1: ldsw [%2], %0\n" + " brlz,pn %0, 2f\n" +-"4: add %0, 1, %1\n" ++"4: addcc %0, 1, %1\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++" tvs %%icc, 6\n" ++#endif ++ + " cas [%2], %0, %1\n" + " cmp %0, %1\n" + " bne,pn %%icc, 1b\n" +@@ -112,7 +117,7 @@ static void inline arch_read_lock(arch_r + " .previous" + : "=&r" (tmp1), "=&r" (tmp2) + : "r" (lock) +- : "memory"); ++ : "memory", "cc"); + } + + static int inline arch_read_trylock(arch_rwlock_t *lock) +@@ -123,7 +128,12 @@ static int inline arch_read_trylock(arch + "1: ldsw [%2], %0\n" + " brlz,a,pn %0, 2f\n" + " mov 0, %0\n" +-" add %0, 1, %1\n" ++" addcc %0, 1, %1\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++" tvs %%icc, 6\n" ++#endif ++ + " cas [%2], %0, %1\n" + " cmp %0, %1\n" + " bne,pn %%icc, 1b\n" +@@ -142,7 +152,12 @@ static void inline arch_read_unlock(arch + + __asm__ __volatile__( + "1: lduw [%2], %0\n" +-" sub %0, 1, %1\n" ++" subcc %0, 1, %1\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++" tvs %%icc, 6\n" ++#endif ++ + " cas [%2], %0, %1\n" + " cmp %0, %1\n" + " bne,pn %%xcc, 1b\n" +diff -urNp linux-2.6.38.3/arch/sparc/include/asm/uaccess_32.h linux-2.6.38.3/arch/sparc/include/asm/uaccess_32.h +--- linux-2.6.38.3/arch/sparc/include/asm/uaccess_32.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/include/asm/uaccess_32.h 2011-04-17 15:57:32.000000000 -0400 +@@ -249,27 +249,46 @@ extern unsigned long __copy_user(void __ + + static inline unsigned long copy_to_user(void __user *to, const void *from, unsigned long n) + { +- if (n && __access_ok((unsigned long) to, n)) ++ if ((long)n < 0) ++ return n; ++ ++ if (n && __access_ok((unsigned long) to, n)) { ++ if (!__builtin_constant_p(n)) ++ check_object_size(from, n, true); + return __copy_user(to, (__force void __user *) from, n); +- else ++ } else + return n; + } + + static inline unsigned long __copy_to_user(void __user *to, const void *from, unsigned long n) + { ++ if ((long)n < 0) ++ return n; ++ ++ if (!__builtin_constant_p(n)) ++ check_object_size(from, n, true); ++ + return __copy_user(to, (__force void __user *) from, n); + } + + static inline unsigned long copy_from_user(void *to, const void __user *from, unsigned long n) + { +- if (n && __access_ok((unsigned long) from, n)) ++ if ((long)n < 0) ++ return n; ++ ++ if (n && __access_ok((unsigned long) from, n)) { ++ if (!__builtin_constant_p(n)) ++ check_object_size(to, n, false); + return __copy_user((__force void __user *) to, from, n); +- else ++ } else + return n; + } + + static inline unsigned long __copy_from_user(void *to, const void __user *from, unsigned long n) + { ++ if ((long)n < 0) ++ return n; ++ + return __copy_user((__force void __user *) to, from, n); + } + +diff -urNp linux-2.6.38.3/arch/sparc/include/asm/uaccess_64.h linux-2.6.38.3/arch/sparc/include/asm/uaccess_64.h +--- linux-2.6.38.3/arch/sparc/include/asm/uaccess_64.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/include/asm/uaccess_64.h 2011-04-17 15:57:32.000000000 -0400 +@@ -10,6 +10,7 @@ + #include <linux/compiler.h> + #include <linux/string.h> + #include <linux/thread_info.h> ++#include <linux/kernel.h> + #include <asm/asi.h> + #include <asm/system.h> + #include <asm/spitfire.h> +@@ -213,8 +214,15 @@ extern unsigned long copy_from_user_fixu + static inline unsigned long __must_check + copy_from_user(void *to, const void __user *from, unsigned long size) + { +- unsigned long ret = ___copy_from_user(to, from, size); ++ unsigned long ret; + ++ if ((long)size < 0 || size > INT_MAX) ++ return size; ++ ++ if (!__builtin_constant_p(size)) ++ check_object_size(to, size, false); ++ ++ ret = ___copy_from_user(to, from, size); + if (unlikely(ret)) + ret = copy_from_user_fixup(to, from, size); + +@@ -230,8 +238,15 @@ extern unsigned long copy_to_user_fixup( + static inline unsigned long __must_check + copy_to_user(void __user *to, const void *from, unsigned long size) + { +- unsigned long ret = ___copy_to_user(to, from, size); ++ unsigned long ret; ++ ++ if ((long)size < 0 || size > INT_MAX) ++ return size; ++ ++ if (!__builtin_constant_p(size)) ++ check_object_size(from, size, true); + ++ ret = ___copy_to_user(to, from, size); + if (unlikely(ret)) + ret = copy_to_user_fixup(to, from, size); + return ret; +diff -urNp linux-2.6.38.3/arch/sparc/include/asm/uaccess.h linux-2.6.38.3/arch/sparc/include/asm/uaccess.h +--- linux-2.6.38.3/arch/sparc/include/asm/uaccess.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/include/asm/uaccess.h 2011-04-17 15:57:32.000000000 -0400 +@@ -1,5 +1,13 @@ + #ifndef ___ASM_SPARC_UACCESS_H + #define ___ASM_SPARC_UACCESS_H ++ ++#ifdef __KERNEL__ ++#ifndef __ASSEMBLY__ ++#include <linux/types.h> ++extern void check_object_size(const void *ptr, unsigned long n, bool to); ++#endif ++#endif ++ + #if defined(__sparc__) && defined(__arch64__) + #include <asm/uaccess_64.h> + #else +diff -urNp linux-2.6.38.3/arch/sparc/kernel/iommu.c linux-2.6.38.3/arch/sparc/kernel/iommu.c +--- linux-2.6.38.3/arch/sparc/kernel/iommu.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/kernel/iommu.c 2011-04-17 15:57:32.000000000 -0400 +@@ -827,7 +827,7 @@ static void dma_4u_sync_sg_for_cpu(struc + spin_unlock_irqrestore(&iommu->lock, flags); + } + +-static struct dma_map_ops sun4u_dma_ops = { ++static const struct dma_map_ops sun4u_dma_ops = { + .alloc_coherent = dma_4u_alloc_coherent, + .free_coherent = dma_4u_free_coherent, + .map_page = dma_4u_map_page, +@@ -838,7 +838,7 @@ static struct dma_map_ops sun4u_dma_ops + .sync_sg_for_cpu = dma_4u_sync_sg_for_cpu, + }; + +-struct dma_map_ops *dma_ops = &sun4u_dma_ops; ++const struct dma_map_ops *dma_ops = &sun4u_dma_ops; + EXPORT_SYMBOL(dma_ops); + + extern int pci64_dma_supported(struct pci_dev *pdev, u64 device_mask); +diff -urNp linux-2.6.38.3/arch/sparc/kernel/ioport.c linux-2.6.38.3/arch/sparc/kernel/ioport.c +--- linux-2.6.38.3/arch/sparc/kernel/ioport.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/kernel/ioport.c 2011-04-17 15:57:32.000000000 -0400 +@@ -397,7 +397,7 @@ static void sbus_sync_sg_for_device(stru + BUG(); + } + +-struct dma_map_ops sbus_dma_ops = { ++const struct dma_map_ops sbus_dma_ops = { + .alloc_coherent = sbus_alloc_coherent, + .free_coherent = sbus_free_coherent, + .map_page = sbus_map_page, +@@ -408,7 +408,7 @@ struct dma_map_ops sbus_dma_ops = { + .sync_sg_for_device = sbus_sync_sg_for_device, + }; + +-struct dma_map_ops *dma_ops = &sbus_dma_ops; ++const struct dma_map_ops *dma_ops = &sbus_dma_ops; + EXPORT_SYMBOL(dma_ops); + + static int __init sparc_register_ioport(void) +@@ -645,7 +645,7 @@ static void pci32_sync_sg_for_device(str + } + } + +-struct dma_map_ops pci32_dma_ops = { ++const struct dma_map_ops pci32_dma_ops = { + .alloc_coherent = pci32_alloc_coherent, + .free_coherent = pci32_free_coherent, + .map_page = pci32_map_page, +diff -urNp linux-2.6.38.3/arch/sparc/kernel/kgdb_32.c linux-2.6.38.3/arch/sparc/kernel/kgdb_32.c +--- linux-2.6.38.3/arch/sparc/kernel/kgdb_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/kernel/kgdb_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -164,7 +164,7 @@ void kgdb_arch_set_pc(struct pt_regs *re + regs->npc = regs->pc + 4; + } + +-struct kgdb_arch arch_kgdb_ops = { ++const struct kgdb_arch arch_kgdb_ops = { + /* Breakpoint instruction: ta 0x7d */ + .gdb_bpt_instr = { 0x91, 0xd0, 0x20, 0x7d }, + }; +diff -urNp linux-2.6.38.3/arch/sparc/kernel/kgdb_64.c linux-2.6.38.3/arch/sparc/kernel/kgdb_64.c +--- linux-2.6.38.3/arch/sparc/kernel/kgdb_64.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/kernel/kgdb_64.c 2011-04-17 15:57:32.000000000 -0400 +@@ -187,7 +187,7 @@ void kgdb_arch_set_pc(struct pt_regs *re + regs->tnpc = regs->tpc + 4; + } + +-struct kgdb_arch arch_kgdb_ops = { ++const struct kgdb_arch arch_kgdb_ops = { + /* Breakpoint instruction: ta 0x72 */ + .gdb_bpt_instr = { 0x91, 0xd0, 0x20, 0x72 }, + }; +diff -urNp linux-2.6.38.3/arch/sparc/kernel/Makefile linux-2.6.38.3/arch/sparc/kernel/Makefile +--- linux-2.6.38.3/arch/sparc/kernel/Makefile 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/kernel/Makefile 2011-04-17 15:57:32.000000000 -0400 +@@ -3,7 +3,7 @@ + # + + asflags-y := -ansi +-ccflags-y := -Werror ++#ccflags-y := -Werror + + extra-y := head_$(BITS).o + extra-y += init_task.o +diff -urNp linux-2.6.38.3/arch/sparc/kernel/pci_sun4v.c linux-2.6.38.3/arch/sparc/kernel/pci_sun4v.c +--- linux-2.6.38.3/arch/sparc/kernel/pci_sun4v.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/kernel/pci_sun4v.c 2011-04-17 15:57:32.000000000 -0400 +@@ -525,7 +525,7 @@ static void dma_4v_unmap_sg(struct devic + spin_unlock_irqrestore(&iommu->lock, flags); + } + +-static struct dma_map_ops sun4v_dma_ops = { ++static const struct dma_map_ops sun4v_dma_ops = { + .alloc_coherent = dma_4v_alloc_coherent, + .free_coherent = dma_4v_free_coherent, + .map_page = dma_4v_map_page, +diff -urNp linux-2.6.38.3/arch/sparc/kernel/process_32.c linux-2.6.38.3/arch/sparc/kernel/process_32.c +--- linux-2.6.38.3/arch/sparc/kernel/process_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/kernel/process_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -196,7 +196,7 @@ void __show_backtrace(unsigned long fp) + rw->ins[4], rw->ins[5], + rw->ins[6], + rw->ins[7]); +- printk("%pS\n", (void *) rw->ins[7]); ++ printk("%pA\n", (void *) rw->ins[7]); + rw = (struct reg_window32 *) rw->ins[6]; + } + spin_unlock_irqrestore(&sparc_backtrace_lock, flags); +@@ -263,14 +263,14 @@ void show_regs(struct pt_regs *r) + + printk("PSR: %08lx PC: %08lx NPC: %08lx Y: %08lx %s\n", + r->psr, r->pc, r->npc, r->y, print_tainted()); +- printk("PC: <%pS>\n", (void *) r->pc); ++ printk("PC: <%pA>\n", (void *) r->pc); + printk("%%G: %08lx %08lx %08lx %08lx %08lx %08lx %08lx %08lx\n", + r->u_regs[0], r->u_regs[1], r->u_regs[2], r->u_regs[3], + r->u_regs[4], r->u_regs[5], r->u_regs[6], r->u_regs[7]); + printk("%%O: %08lx %08lx %08lx %08lx %08lx %08lx %08lx %08lx\n", + r->u_regs[8], r->u_regs[9], r->u_regs[10], r->u_regs[11], + r->u_regs[12], r->u_regs[13], r->u_regs[14], r->u_regs[15]); +- printk("RPC: <%pS>\n", (void *) r->u_regs[15]); ++ printk("RPC: <%pA>\n", (void *) r->u_regs[15]); + + printk("%%L: %08lx %08lx %08lx %08lx %08lx %08lx %08lx %08lx\n", + rw->locals[0], rw->locals[1], rw->locals[2], rw->locals[3], +@@ -305,7 +305,7 @@ void show_stack(struct task_struct *tsk, + rw = (struct reg_window32 *) fp; + pc = rw->ins[7]; + printk("[%08lx : ", pc); +- printk("%pS ] ", (void *) pc); ++ printk("%pA ] ", (void *) pc); + fp = rw->ins[6]; + } while (++count < 16); + printk("\n"); +diff -urNp linux-2.6.38.3/arch/sparc/kernel/process_64.c linux-2.6.38.3/arch/sparc/kernel/process_64.c +--- linux-2.6.38.3/arch/sparc/kernel/process_64.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/kernel/process_64.c 2011-04-17 15:57:32.000000000 -0400 +@@ -180,14 +180,14 @@ static void show_regwindow(struct pt_reg + printk("i4: %016lx i5: %016lx i6: %016lx i7: %016lx\n", + rwk->ins[4], rwk->ins[5], rwk->ins[6], rwk->ins[7]); + if (regs->tstate & TSTATE_PRIV) +- printk("I7: <%pS>\n", (void *) rwk->ins[7]); ++ printk("I7: <%pA>\n", (void *) rwk->ins[7]); + } + + void show_regs(struct pt_regs *regs) + { + printk("TSTATE: %016lx TPC: %016lx TNPC: %016lx Y: %08x %s\n", regs->tstate, + regs->tpc, regs->tnpc, regs->y, print_tainted()); +- printk("TPC: <%pS>\n", (void *) regs->tpc); ++ printk("TPC: <%pA>\n", (void *) regs->tpc); + printk("g0: %016lx g1: %016lx g2: %016lx g3: %016lx\n", + regs->u_regs[0], regs->u_regs[1], regs->u_regs[2], + regs->u_regs[3]); +@@ -200,7 +200,7 @@ void show_regs(struct pt_regs *regs) + printk("o4: %016lx o5: %016lx sp: %016lx ret_pc: %016lx\n", + regs->u_regs[12], regs->u_regs[13], regs->u_regs[14], + regs->u_regs[15]); +- printk("RPC: <%pS>\n", (void *) regs->u_regs[15]); ++ printk("RPC: <%pA>\n", (void *) regs->u_regs[15]); + show_regwindow(regs); + show_stack(current, (unsigned long *) regs->u_regs[UREG_FP]); + } +@@ -285,7 +285,7 @@ void arch_trigger_all_cpu_backtrace(void + ((tp && tp->task) ? tp->task->pid : -1)); + + if (gp->tstate & TSTATE_PRIV) { +- printk(" TPC[%pS] O7[%pS] I7[%pS] RPC[%pS]\n", ++ printk(" TPC[%pA] O7[%pA] I7[%pA] RPC[%pA]\n", + (void *) gp->tpc, + (void *) gp->o7, + (void *) gp->i7, +diff -urNp linux-2.6.38.3/arch/sparc/kernel/sys_sparc_32.c linux-2.6.38.3/arch/sparc/kernel/sys_sparc_32.c +--- linux-2.6.38.3/arch/sparc/kernel/sys_sparc_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/kernel/sys_sparc_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -56,7 +56,7 @@ unsigned long arch_get_unmapped_area(str + if (ARCH_SUN4C && len > 0x20000000) + return -ENOMEM; + if (!addr) +- addr = TASK_UNMAPPED_BASE; ++ addr = current->mm->mmap_base; + + if (flags & MAP_SHARED) + addr = COLOUR_ALIGN(addr); +@@ -71,7 +71,7 @@ unsigned long arch_get_unmapped_area(str + } + if (TASK_SIZE - PAGE_SIZE - len < addr) + return -ENOMEM; +- if (!vmm || addr + len <= vmm->vm_start) ++ if (check_heap_stack_gap(vmm, addr, len)) + return addr; + addr = vmm->vm_end; + if (flags & MAP_SHARED) +diff -urNp linux-2.6.38.3/arch/sparc/kernel/sys_sparc_64.c linux-2.6.38.3/arch/sparc/kernel/sys_sparc_64.c +--- linux-2.6.38.3/arch/sparc/kernel/sys_sparc_64.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/kernel/sys_sparc_64.c 2011-04-17 15:57:32.000000000 -0400 +@@ -124,7 +124,7 @@ unsigned long arch_get_unmapped_area(str + /* We do not accept a shared mapping if it would violate + * cache aliasing constraints. + */ +- if ((flags & MAP_SHARED) && ++ if ((filp || (flags & MAP_SHARED)) && + ((addr - (pgoff << PAGE_SHIFT)) & (SHMLBA - 1))) + return -EINVAL; + return addr; +@@ -139,6 +139,10 @@ unsigned long arch_get_unmapped_area(str + if (filp || (flags & MAP_SHARED)) + do_color_align = 1; + ++#ifdef CONFIG_PAX_RANDMMAP ++ if (!(mm->pax_flags & MF_PAX_RANDMMAP)) ++#endif ++ + if (addr) { + if (do_color_align) + addr = COLOUR_ALIGN(addr, pgoff); +@@ -146,15 +150,14 @@ unsigned long arch_get_unmapped_area(str + addr = PAGE_ALIGN(addr); + + vma = find_vma(mm, addr); +- if (task_size - len >= addr && +- (!vma || addr + len <= vma->vm_start)) ++ if (task_size - len >= addr && check_heap_stack_gap(vma, addr, len)) + return addr; + } + + if (len > mm->cached_hole_size) { +- start_addr = addr = mm->free_area_cache; ++ start_addr = addr = mm->free_area_cache; + } else { +- start_addr = addr = TASK_UNMAPPED_BASE; ++ start_addr = addr = mm->mmap_base; + mm->cached_hole_size = 0; + } + +@@ -174,14 +177,14 @@ full_search: + vma = find_vma(mm, VA_EXCLUDE_END); + } + if (unlikely(task_size < addr)) { +- if (start_addr != TASK_UNMAPPED_BASE) { +- start_addr = addr = TASK_UNMAPPED_BASE; ++ if (start_addr != mm->mmap_base) { ++ start_addr = addr = mm->mmap_base; + mm->cached_hole_size = 0; + goto full_search; + } + return -ENOMEM; + } +- if (likely(!vma || addr + len <= vma->vm_start)) { ++ if (likely(check_heap_stack_gap(vma, addr, len))) { + /* + * Remember the place where we stopped the search: + */ +@@ -215,7 +218,7 @@ arch_get_unmapped_area_topdown(struct fi + /* We do not accept a shared mapping if it would violate + * cache aliasing constraints. + */ +- if ((flags & MAP_SHARED) && ++ if ((filp || (flags & MAP_SHARED)) && + ((addr - (pgoff << PAGE_SHIFT)) & (SHMLBA - 1))) + return -EINVAL; + return addr; +@@ -236,8 +239,7 @@ arch_get_unmapped_area_topdown(struct fi + addr = PAGE_ALIGN(addr); + + vma = find_vma(mm, addr); +- if (task_size - len >= addr && +- (!vma || addr + len <= vma->vm_start)) ++ if (task_size - len >= addr && check_heap_stack_gap(vma, addr, len)) + return addr; + } + +@@ -258,7 +260,7 @@ arch_get_unmapped_area_topdown(struct fi + /* make sure it can fit in the remaining address space */ + if (likely(addr > len)) { + vma = find_vma(mm, addr-len); +- if (!vma || addr <= vma->vm_start) { ++ if (check_heap_stack_gap(vma, addr - len, len)) { + /* remember the address as a hint for next time */ + return (mm->free_area_cache = addr-len); + } +@@ -267,18 +269,18 @@ arch_get_unmapped_area_topdown(struct fi + if (unlikely(mm->mmap_base < len)) + goto bottomup; + +- addr = mm->mmap_base-len; +- if (do_color_align) +- addr = COLOUR_ALIGN_DOWN(addr, pgoff); ++ addr = mm->mmap_base - len; + + do { ++ if (do_color_align) ++ addr = COLOUR_ALIGN_DOWN(addr, pgoff); + /* + * Lookup failure means no vma is above this address, + * else if new region fits below vma->vm_start, + * return with success: + */ + vma = find_vma(mm, addr); +- if (likely(!vma || addr+len <= vma->vm_start)) { ++ if (likely(check_heap_stack_gap(vma, addr, len))) { + /* remember the address as a hint for next time */ + return (mm->free_area_cache = addr); + } +@@ -288,10 +290,8 @@ arch_get_unmapped_area_topdown(struct fi + mm->cached_hole_size = vma->vm_start - addr; + + /* try just below the current vma->vm_start */ +- addr = vma->vm_start-len; +- if (do_color_align) +- addr = COLOUR_ALIGN_DOWN(addr, pgoff); +- } while (likely(len < vma->vm_start)); ++ addr = skip_heap_stack_gap(vma, len); ++ } while (!IS_ERR_VALUE(addr)); + + bottomup: + /* +@@ -385,6 +385,12 @@ void arch_pick_mmap_layout(struct mm_str + gap == RLIM_INFINITY || + sysctl_legacy_va_layout) { + mm->mmap_base = TASK_UNMAPPED_BASE + random_factor; ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base += mm->delta_mmap; ++#endif ++ + mm->get_unmapped_area = arch_get_unmapped_area; + mm->unmap_area = arch_unmap_area; + } else { +@@ -397,6 +403,12 @@ void arch_pick_mmap_layout(struct mm_str + gap = (task_size / 6 * 5); + + mm->mmap_base = PAGE_ALIGN(task_size - gap - random_factor); ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base -= mm->delta_mmap + mm->delta_stack; ++#endif ++ + mm->get_unmapped_area = arch_get_unmapped_area_topdown; + mm->unmap_area = arch_unmap_area_topdown; + } +diff -urNp linux-2.6.38.3/arch/sparc/kernel/traps_32.c linux-2.6.38.3/arch/sparc/kernel/traps_32.c +--- linux-2.6.38.3/arch/sparc/kernel/traps_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/kernel/traps_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -76,7 +76,7 @@ void die_if_kernel(char *str, struct pt_ + count++ < 30 && + (((unsigned long) rw) >= PAGE_OFFSET) && + !(((unsigned long) rw) & 0x7)) { +- printk("Caller[%08lx]: %pS\n", rw->ins[7], ++ printk("Caller[%08lx]: %pA\n", rw->ins[7], + (void *) rw->ins[7]); + rw = (struct reg_window32 *)rw->ins[6]; + } +diff -urNp linux-2.6.38.3/arch/sparc/kernel/traps_64.c linux-2.6.38.3/arch/sparc/kernel/traps_64.c +--- linux-2.6.38.3/arch/sparc/kernel/traps_64.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/kernel/traps_64.c 2011-04-17 15:57:32.000000000 -0400 +@@ -75,7 +75,7 @@ static void dump_tl1_traplog(struct tl1_ + i + 1, + p->trapstack[i].tstate, p->trapstack[i].tpc, + p->trapstack[i].tnpc, p->trapstack[i].tt); +- printk("TRAPLOG: TPC<%pS>\n", (void *) p->trapstack[i].tpc); ++ printk("TRAPLOG: TPC<%pA>\n", (void *) p->trapstack[i].tpc); + } + } + +@@ -95,6 +95,12 @@ void bad_trap(struct pt_regs *regs, long + + lvl -= 0x100; + if (regs->tstate & TSTATE_PRIV) { ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ if (lvl == 6) ++ pax_report_refcount_overflow(regs); ++#endif ++ + sprintf(buffer, "Kernel bad sw trap %lx", lvl); + die_if_kernel(buffer, regs); + } +@@ -113,11 +119,16 @@ void bad_trap(struct pt_regs *regs, long + void bad_trap_tl1(struct pt_regs *regs, long lvl) + { + char buffer[32]; +- ++ + if (notify_die(DIE_TRAP_TL1, "bad trap tl1", regs, + 0, lvl, SIGTRAP) == NOTIFY_STOP) + return; + ++#ifdef CONFIG_PAX_REFCOUNT ++ if (lvl == 6) ++ pax_report_refcount_overflow(regs); ++#endif ++ + dump_tl1_traplog((struct tl1_traplog *)(regs + 1)); + + sprintf (buffer, "Bad trap %lx at tl>0", lvl); +@@ -1141,7 +1152,7 @@ static void cheetah_log_errors(struct pt + regs->tpc, regs->tnpc, regs->u_regs[UREG_I7], regs->tstate); + printk("%s" "ERROR(%d): ", + (recoverable ? KERN_WARNING : KERN_CRIT), smp_processor_id()); +- printk("TPC<%pS>\n", (void *) regs->tpc); ++ printk("TPC<%pA>\n", (void *) regs->tpc); + printk("%s" "ERROR(%d): M_SYND(%lx), E_SYND(%lx)%s%s\n", + (recoverable ? KERN_WARNING : KERN_CRIT), smp_processor_id(), + (afsr & CHAFSR_M_SYNDROME) >> CHAFSR_M_SYNDROME_SHIFT, +@@ -1748,7 +1759,7 @@ void cheetah_plus_parity_error(int type, + smp_processor_id(), + (type & 0x1) ? 'I' : 'D', + regs->tpc); +- printk(KERN_EMERG "TPC<%pS>\n", (void *) regs->tpc); ++ printk(KERN_EMERG "TPC<%pA>\n", (void *) regs->tpc); + panic("Irrecoverable Cheetah+ parity error."); + } + +@@ -1756,7 +1767,7 @@ void cheetah_plus_parity_error(int type, + smp_processor_id(), + (type & 0x1) ? 'I' : 'D', + regs->tpc); +- printk(KERN_WARNING "TPC<%pS>\n", (void *) regs->tpc); ++ printk(KERN_WARNING "TPC<%pA>\n", (void *) regs->tpc); + } + + struct sun4v_error_entry { +@@ -1963,9 +1974,9 @@ void sun4v_itlb_error_report(struct pt_r + + printk(KERN_EMERG "SUN4V-ITLB: Error at TPC[%lx], tl %d\n", + regs->tpc, tl); +- printk(KERN_EMERG "SUN4V-ITLB: TPC<%pS>\n", (void *) regs->tpc); ++ printk(KERN_EMERG "SUN4V-ITLB: TPC<%pA>\n", (void *) regs->tpc); + printk(KERN_EMERG "SUN4V-ITLB: O7[%lx]\n", regs->u_regs[UREG_I7]); +- printk(KERN_EMERG "SUN4V-ITLB: O7<%pS>\n", ++ printk(KERN_EMERG "SUN4V-ITLB: O7<%pA>\n", + (void *) regs->u_regs[UREG_I7]); + printk(KERN_EMERG "SUN4V-ITLB: vaddr[%lx] ctx[%lx] " + "pte[%lx] error[%lx]\n", +@@ -1987,9 +1998,9 @@ void sun4v_dtlb_error_report(struct pt_r + + printk(KERN_EMERG "SUN4V-DTLB: Error at TPC[%lx], tl %d\n", + regs->tpc, tl); +- printk(KERN_EMERG "SUN4V-DTLB: TPC<%pS>\n", (void *) regs->tpc); ++ printk(KERN_EMERG "SUN4V-DTLB: TPC<%pA>\n", (void *) regs->tpc); + printk(KERN_EMERG "SUN4V-DTLB: O7[%lx]\n", regs->u_regs[UREG_I7]); +- printk(KERN_EMERG "SUN4V-DTLB: O7<%pS>\n", ++ printk(KERN_EMERG "SUN4V-DTLB: O7<%pA>\n", + (void *) regs->u_regs[UREG_I7]); + printk(KERN_EMERG "SUN4V-DTLB: vaddr[%lx] ctx[%lx] " + "pte[%lx] error[%lx]\n", +@@ -2196,13 +2207,13 @@ void show_stack(struct task_struct *tsk, + fp = (unsigned long)sf->fp + STACK_BIAS; + } + +- printk(" [%016lx] %pS\n", pc, (void *) pc); ++ printk(" [%016lx] %pA\n", pc, (void *) pc); + #ifdef CONFIG_FUNCTION_GRAPH_TRACER + if ((pc + 8UL) == (unsigned long) &return_to_handler) { + int index = tsk->curr_ret_stack; + if (tsk->ret_stack && index >= graph) { + pc = tsk->ret_stack[index - graph].ret; +- printk(" [%016lx] %pS\n", pc, (void *) pc); ++ printk(" [%016lx] %pA\n", pc, (void *) pc); + graph++; + } + } +@@ -2255,7 +2266,7 @@ void die_if_kernel(char *str, struct pt_ + while (rw && + count++ < 30 && + kstack_valid(tp, (unsigned long) rw)) { +- printk("Caller[%016lx]: %pS\n", rw->ins[7], ++ printk("Caller[%016lx]: %pA\n", rw->ins[7], + (void *) rw->ins[7]); + + rw = kernel_stack_up(rw); +diff -urNp linux-2.6.38.3/arch/sparc/kernel/unaligned_64.c linux-2.6.38.3/arch/sparc/kernel/unaligned_64.c +--- linux-2.6.38.3/arch/sparc/kernel/unaligned_64.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/kernel/unaligned_64.c 2011-04-17 15:57:32.000000000 -0400 +@@ -278,7 +278,7 @@ static void log_unaligned(struct pt_regs + static DEFINE_RATELIMIT_STATE(ratelimit, 5 * HZ, 5); + + if (__ratelimit(&ratelimit)) { +- printk("Kernel unaligned access at TPC[%lx] %pS\n", ++ printk("Kernel unaligned access at TPC[%lx] %pA\n", + regs->tpc, (void *) regs->tpc); + } + } +diff -urNp linux-2.6.38.3/arch/sparc/lib/atomic_64.S linux-2.6.38.3/arch/sparc/lib/atomic_64.S +--- linux-2.6.38.3/arch/sparc/lib/atomic_64.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/lib/atomic_64.S 2011-04-17 15:57:32.000000000 -0400 +@@ -18,7 +18,12 @@ + atomic_add: /* %o0 = increment, %o1 = atomic_ptr */ + BACKOFF_SETUP(%o2) + 1: lduw [%o1], %g1 +- add %g1, %o0, %g7 ++ addcc %g1, %o0, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %icc, 6 ++#endif ++ + cas [%o1], %g1, %g7 + cmp %g1, %g7 + bne,pn %icc, BACKOFF_LABEL(2f, 1b) +@@ -28,12 +33,32 @@ atomic_add: /* %o0 = increment, %o1 = at + 2: BACKOFF_SPIN(%o2, %o3, 1b) + .size atomic_add, .-atomic_add + ++ .globl atomic_add_unchecked ++ .type atomic_add_unchecked,#function ++atomic_add_unchecked: /* %o0 = increment, %o1 = atomic_ptr */ ++ BACKOFF_SETUP(%o2) ++1: lduw [%o1], %g1 ++ add %g1, %o0, %g7 ++ cas [%o1], %g1, %g7 ++ cmp %g1, %g7 ++ bne,pn %icc, 2f ++ nop ++ retl ++ nop ++2: BACKOFF_SPIN(%o2, %o3, 1b) ++ .size atomic_add_unchecked, .-atomic_add_unchecked ++ + .globl atomic_sub + .type atomic_sub,#function + atomic_sub: /* %o0 = decrement, %o1 = atomic_ptr */ + BACKOFF_SETUP(%o2) + 1: lduw [%o1], %g1 +- sub %g1, %o0, %g7 ++ subcc %g1, %o0, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %icc, 6 ++#endif ++ + cas [%o1], %g1, %g7 + cmp %g1, %g7 + bne,pn %icc, BACKOFF_LABEL(2f, 1b) +@@ -43,12 +68,32 @@ atomic_sub: /* %o0 = decrement, %o1 = at + 2: BACKOFF_SPIN(%o2, %o3, 1b) + .size atomic_sub, .-atomic_sub + ++ .globl atomic_sub_unchecked ++ .type atomic_sub_unchecked,#function ++atomic_sub_unchecked: /* %o0 = decrement, %o1 = atomic_ptr */ ++ BACKOFF_SETUP(%o2) ++1: lduw [%o1], %g1 ++ sub %g1, %o0, %g7 ++ cas [%o1], %g1, %g7 ++ cmp %g1, %g7 ++ bne,pn %icc, 2f ++ nop ++ retl ++ nop ++2: BACKOFF_SPIN(%o2, %o3, 1b) ++ .size atomic_sub_unchecked, .-atomic_sub_unchecked ++ + .globl atomic_add_ret + .type atomic_add_ret,#function + atomic_add_ret: /* %o0 = increment, %o1 = atomic_ptr */ + BACKOFF_SETUP(%o2) + 1: lduw [%o1], %g1 +- add %g1, %o0, %g7 ++ addcc %g1, %o0, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %icc, 6 ++#endif ++ + cas [%o1], %g1, %g7 + cmp %g1, %g7 + bne,pn %icc, BACKOFF_LABEL(2f, 1b) +@@ -58,12 +103,33 @@ atomic_add_ret: /* %o0 = increment, %o1 + 2: BACKOFF_SPIN(%o2, %o3, 1b) + .size atomic_add_ret, .-atomic_add_ret + ++ .globl atomic_add_ret_unchecked ++ .type atomic_add_ret_unchecked,#function ++atomic_add_ret_unchecked: /* %o0 = increment, %o1 = atomic_ptr */ ++ BACKOFF_SETUP(%o2) ++1: lduw [%o1], %g1 ++ addcc %g1, %o0, %g7 ++ cas [%o1], %g1, %g7 ++ cmp %g1, %g7 ++ bne,pn %icc, 2f ++ add %g7, %o0, %g7 ++ sra %g7, 0, %o0 ++ retl ++ nop ++2: BACKOFF_SPIN(%o2, %o3, 1b) ++ .size atomic_add_ret_unchecked, .-atomic_add_ret_unchecked ++ + .globl atomic_sub_ret + .type atomic_sub_ret,#function + atomic_sub_ret: /* %o0 = decrement, %o1 = atomic_ptr */ + BACKOFF_SETUP(%o2) + 1: lduw [%o1], %g1 +- sub %g1, %o0, %g7 ++ subcc %g1, %o0, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %icc, 6 ++#endif ++ + cas [%o1], %g1, %g7 + cmp %g1, %g7 + bne,pn %icc, BACKOFF_LABEL(2f, 1b) +@@ -78,7 +144,12 @@ atomic_sub_ret: /* %o0 = decrement, %o1 + atomic64_add: /* %o0 = increment, %o1 = atomic_ptr */ + BACKOFF_SETUP(%o2) + 1: ldx [%o1], %g1 +- add %g1, %o0, %g7 ++ addcc %g1, %o0, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %xcc, 6 ++#endif ++ + casx [%o1], %g1, %g7 + cmp %g1, %g7 + bne,pn %xcc, BACKOFF_LABEL(2f, 1b) +@@ -88,12 +159,32 @@ atomic64_add: /* %o0 = increment, %o1 = + 2: BACKOFF_SPIN(%o2, %o3, 1b) + .size atomic64_add, .-atomic64_add + ++ .globl atomic64_add_unchecked ++ .type atomic64_add_unchecked,#function ++atomic64_add_unchecked: /* %o0 = increment, %o1 = atomic_ptr */ ++ BACKOFF_SETUP(%o2) ++1: ldx [%o1], %g1 ++ addcc %g1, %o0, %g7 ++ casx [%o1], %g1, %g7 ++ cmp %g1, %g7 ++ bne,pn %xcc, 2f ++ nop ++ retl ++ nop ++2: BACKOFF_SPIN(%o2, %o3, 1b) ++ .size atomic64_add_unchecked, .-atomic64_add_unchecked ++ + .globl atomic64_sub + .type atomic64_sub,#function + atomic64_sub: /* %o0 = decrement, %o1 = atomic_ptr */ + BACKOFF_SETUP(%o2) + 1: ldx [%o1], %g1 +- sub %g1, %o0, %g7 ++ subcc %g1, %o0, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %xcc, 6 ++#endif ++ + casx [%o1], %g1, %g7 + cmp %g1, %g7 + bne,pn %xcc, BACKOFF_LABEL(2f, 1b) +@@ -103,12 +194,32 @@ atomic64_sub: /* %o0 = decrement, %o1 = + 2: BACKOFF_SPIN(%o2, %o3, 1b) + .size atomic64_sub, .-atomic64_sub + ++ .globl atomic64_sub_unchecked ++ .type atomic64_sub_unchecked,#function ++atomic64_sub_unchecked: /* %o0 = decrement, %o1 = atomic_ptr */ ++ BACKOFF_SETUP(%o2) ++1: ldx [%o1], %g1 ++ subcc %g1, %o0, %g7 ++ casx [%o1], %g1, %g7 ++ cmp %g1, %g7 ++ bne,pn %xcc, 2f ++ nop ++ retl ++ nop ++2: BACKOFF_SPIN(%o2, %o3, 1b) ++ .size atomic64_sub_unchecked, .-atomic64_sub_unchecked ++ + .globl atomic64_add_ret + .type atomic64_add_ret,#function + atomic64_add_ret: /* %o0 = increment, %o1 = atomic_ptr */ + BACKOFF_SETUP(%o2) + 1: ldx [%o1], %g1 +- add %g1, %o0, %g7 ++ addcc %g1, %o0, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %xcc, 6 ++#endif ++ + casx [%o1], %g1, %g7 + cmp %g1, %g7 + bne,pn %xcc, BACKOFF_LABEL(2f, 1b) +@@ -118,12 +229,33 @@ atomic64_add_ret: /* %o0 = increment, %o + 2: BACKOFF_SPIN(%o2, %o3, 1b) + .size atomic64_add_ret, .-atomic64_add_ret + ++ .globl atomic64_add_ret_unchecked ++ .type atomic64_add_ret_unchecked,#function ++atomic64_add_ret_unchecked: /* %o0 = increment, %o1 = atomic_ptr */ ++ BACKOFF_SETUP(%o2) ++1: ldx [%o1], %g1 ++ addcc %g1, %o0, %g7 ++ casx [%o1], %g1, %g7 ++ cmp %g1, %g7 ++ bne,pn %xcc, 2f ++ add %g7, %o0, %g7 ++ mov %g7, %o0 ++ retl ++ nop ++2: BACKOFF_SPIN(%o2, %o3, 1b) ++ .size atomic64_add_ret_unchecked, .-atomic64_add_ret_unchecked ++ + .globl atomic64_sub_ret + .type atomic64_sub_ret,#function + atomic64_sub_ret: /* %o0 = decrement, %o1 = atomic_ptr */ + BACKOFF_SETUP(%o2) + 1: ldx [%o1], %g1 +- sub %g1, %o0, %g7 ++ subcc %g1, %o0, %g7 ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ tvs %xcc, 6 ++#endif ++ + casx [%o1], %g1, %g7 + cmp %g1, %g7 + bne,pn %xcc, BACKOFF_LABEL(2f, 1b) +diff -urNp linux-2.6.38.3/arch/sparc/lib/ksyms.c linux-2.6.38.3/arch/sparc/lib/ksyms.c +--- linux-2.6.38.3/arch/sparc/lib/ksyms.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/lib/ksyms.c 2011-04-17 15:57:32.000000000 -0400 +@@ -142,12 +142,17 @@ EXPORT_SYMBOL(__downgrade_write); + + /* Atomic counter implementation. */ + EXPORT_SYMBOL(atomic_add); ++EXPORT_SYMBOL(atomic_add_unchecked); + EXPORT_SYMBOL(atomic_add_ret); + EXPORT_SYMBOL(atomic_sub); ++EXPORT_SYMBOL(atomic_sub_unchecked); + EXPORT_SYMBOL(atomic_sub_ret); + EXPORT_SYMBOL(atomic64_add); ++EXPORT_SYMBOL(atomic64_add_unchecked); + EXPORT_SYMBOL(atomic64_add_ret); ++EXPORT_SYMBOL(atomic64_add_ret_unchecked); + EXPORT_SYMBOL(atomic64_sub); ++EXPORT_SYMBOL(atomic64_sub_unchecked); + EXPORT_SYMBOL(atomic64_sub_ret); + + /* Atomic bit operations. */ +diff -urNp linux-2.6.38.3/arch/sparc/Makefile linux-2.6.38.3/arch/sparc/Makefile +--- linux-2.6.38.3/arch/sparc/Makefile 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/Makefile 2011-04-17 15:57:32.000000000 -0400 +@@ -75,7 +75,7 @@ drivers-$(CONFIG_OPROFILE) += arch/sparc + # Export what is needed by arch/sparc/boot/Makefile + export VMLINUX_INIT VMLINUX_MAIN + VMLINUX_INIT := $(head-y) $(init-y) +-VMLINUX_MAIN := $(core-y) kernel/ mm/ fs/ ipc/ security/ crypto/ block/ ++VMLINUX_MAIN := $(core-y) kernel/ mm/ fs/ ipc/ security/ crypto/ block/ grsecurity/ + VMLINUX_MAIN += $(patsubst %/, %/lib.a, $(libs-y)) $(libs-y) + VMLINUX_MAIN += $(drivers-y) $(net-y) + +diff -urNp linux-2.6.38.3/arch/sparc/mm/fault_32.c linux-2.6.38.3/arch/sparc/mm/fault_32.c +--- linux-2.6.38.3/arch/sparc/mm/fault_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/mm/fault_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -22,6 +22,9 @@ + #include <linux/interrupt.h> + #include <linux/module.h> + #include <linux/kdebug.h> ++#include <linux/slab.h> ++#include <linux/pagemap.h> ++#include <linux/compiler.h> + + #include <asm/system.h> + #include <asm/page.h> +@@ -209,6 +212,268 @@ static unsigned long compute_si_addr(str + return safe_compute_effective_address(regs, insn); + } + ++#ifdef CONFIG_PAX_PAGEEXEC ++#ifdef CONFIG_PAX_DLRESOLVE ++static void pax_emuplt_close(struct vm_area_struct *vma) ++{ ++ vma->vm_mm->call_dl_resolve = 0UL; ++} ++ ++static int pax_emuplt_fault(struct vm_area_struct *vma, struct vm_fault *vmf) ++{ ++ unsigned int *kaddr; ++ ++ vmf->page = alloc_page(GFP_HIGHUSER); ++ if (!vmf->page) ++ return VM_FAULT_OOM; ++ ++ kaddr = kmap(vmf->page); ++ memset(kaddr, 0, PAGE_SIZE); ++ kaddr[0] = 0x9DE3BFA8U; /* save */ ++ flush_dcache_page(vmf->page); ++ kunmap(vmf->page); ++ return VM_FAULT_MAJOR; ++} ++ ++static const struct vm_operations_struct pax_vm_ops = { ++ .close = pax_emuplt_close, ++ .fault = pax_emuplt_fault ++}; ++ ++static int pax_insert_vma(struct vm_area_struct *vma, unsigned long addr) ++{ ++ int ret; ++ ++ INIT_LIST_HEAD(&vma->anon_vma_chain); ++ vma->vm_mm = current->mm; ++ vma->vm_start = addr; ++ vma->vm_end = addr + PAGE_SIZE; ++ vma->vm_flags = VM_READ | VM_EXEC | VM_MAYREAD | VM_MAYEXEC; ++ vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); ++ vma->vm_ops = &pax_vm_ops; ++ ++ ret = insert_vm_struct(current->mm, vma); ++ if (ret) ++ return ret; ++ ++ ++current->mm->total_vm; ++ return 0; ++} ++#endif ++ ++/* ++ * PaX: decide what to do with offenders (regs->pc = fault address) ++ * ++ * returns 1 when task should be killed ++ * 2 when patched PLT trampoline was detected ++ * 3 when unpatched PLT trampoline was detected ++ */ ++static int pax_handle_fetch_fault(struct pt_regs *regs) ++{ ++ ++#ifdef CONFIG_PAX_EMUPLT ++ int err; ++ ++ do { /* PaX: patched PLT emulation #1 */ ++ unsigned int sethi1, sethi2, jmpl; ++ ++ err = get_user(sethi1, (unsigned int *)regs->pc); ++ err |= get_user(sethi2, (unsigned int *)(regs->pc+4)); ++ err |= get_user(jmpl, (unsigned int *)(regs->pc+8)); ++ ++ if (err) ++ break; ++ ++ if ((sethi1 & 0xFFC00000U) == 0x03000000U && ++ (sethi2 & 0xFFC00000U) == 0x03000000U && ++ (jmpl & 0xFFFFE000U) == 0x81C06000U) ++ { ++ unsigned int addr; ++ ++ regs->u_regs[UREG_G1] = (sethi2 & 0x003FFFFFU) << 10; ++ addr = regs->u_regs[UREG_G1]; ++ addr += (((jmpl | 0xFFFFE000U) ^ 0x00001000U) + 0x00001000U); ++ regs->pc = addr; ++ regs->npc = addr+4; ++ return 2; ++ } ++ } while (0); ++ ++ { /* PaX: patched PLT emulation #2 */ ++ unsigned int ba; ++ ++ err = get_user(ba, (unsigned int *)regs->pc); ++ ++ if (!err && (ba & 0xFFC00000U) == 0x30800000U) { ++ unsigned int addr; ++ ++ addr = regs->pc + ((((ba | 0xFFC00000U) ^ 0x00200000U) + 0x00200000U) << 2); ++ regs->pc = addr; ++ regs->npc = addr+4; ++ return 2; ++ } ++ } ++ ++ do { /* PaX: patched PLT emulation #3 */ ++ unsigned int sethi, jmpl, nop; ++ ++ err = get_user(sethi, (unsigned int *)regs->pc); ++ err |= get_user(jmpl, (unsigned int *)(regs->pc+4)); ++ err |= get_user(nop, (unsigned int *)(regs->pc+8)); ++ ++ if (err) ++ break; ++ ++ if ((sethi & 0xFFC00000U) == 0x03000000U && ++ (jmpl & 0xFFFFE000U) == 0x81C06000U && ++ nop == 0x01000000U) ++ { ++ unsigned int addr; ++ ++ addr = (sethi & 0x003FFFFFU) << 10; ++ regs->u_regs[UREG_G1] = addr; ++ addr += (((jmpl | 0xFFFFE000U) ^ 0x00001000U) + 0x00001000U); ++ regs->pc = addr; ++ regs->npc = addr+4; ++ return 2; ++ } ++ } while (0); ++ ++ do { /* PaX: unpatched PLT emulation step 1 */ ++ unsigned int sethi, ba, nop; ++ ++ err = get_user(sethi, (unsigned int *)regs->pc); ++ err |= get_user(ba, (unsigned int *)(regs->pc+4)); ++ err |= get_user(nop, (unsigned int *)(regs->pc+8)); ++ ++ if (err) ++ break; ++ ++ if ((sethi & 0xFFC00000U) == 0x03000000U && ++ ((ba & 0xFFC00000U) == 0x30800000U || (ba & 0xFFF80000U) == 0x30680000U) && ++ nop == 0x01000000U) ++ { ++ unsigned int addr, save, call; ++ ++ if ((ba & 0xFFC00000U) == 0x30800000U) ++ addr = regs->pc + 4 + ((((ba | 0xFFC00000U) ^ 0x00200000U) + 0x00200000U) << 2); ++ else ++ addr = regs->pc + 4 + ((((ba | 0xFFF80000U) ^ 0x00040000U) + 0x00040000U) << 2); ++ ++ err = get_user(save, (unsigned int *)addr); ++ err |= get_user(call, (unsigned int *)(addr+4)); ++ err |= get_user(nop, (unsigned int *)(addr+8)); ++ if (err) ++ break; ++ ++#ifdef CONFIG_PAX_DLRESOLVE ++ if (save == 0x9DE3BFA8U && ++ (call & 0xC0000000U) == 0x40000000U && ++ nop == 0x01000000U) ++ { ++ struct vm_area_struct *vma; ++ unsigned long call_dl_resolve; ++ ++ down_read(¤t->mm->mmap_sem); ++ call_dl_resolve = current->mm->call_dl_resolve; ++ up_read(¤t->mm->mmap_sem); ++ if (likely(call_dl_resolve)) ++ goto emulate; ++ ++ vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); ++ ++ down_write(¤t->mm->mmap_sem); ++ if (current->mm->call_dl_resolve) { ++ call_dl_resolve = current->mm->call_dl_resolve; ++ up_write(¤t->mm->mmap_sem); ++ if (vma) ++ kmem_cache_free(vm_area_cachep, vma); ++ goto emulate; ++ } ++ ++ call_dl_resolve = get_unmapped_area(NULL, 0UL, PAGE_SIZE, 0UL, MAP_PRIVATE); ++ if (!vma || (call_dl_resolve & ~PAGE_MASK)) { ++ up_write(¤t->mm->mmap_sem); ++ if (vma) ++ kmem_cache_free(vm_area_cachep, vma); ++ return 1; ++ } ++ ++ if (pax_insert_vma(vma, call_dl_resolve)) { ++ up_write(¤t->mm->mmap_sem); ++ kmem_cache_free(vm_area_cachep, vma); ++ return 1; ++ } ++ ++ current->mm->call_dl_resolve = call_dl_resolve; ++ up_write(¤t->mm->mmap_sem); ++ ++emulate: ++ regs->u_regs[UREG_G1] = (sethi & 0x003FFFFFU) << 10; ++ regs->pc = call_dl_resolve; ++ regs->npc = addr+4; ++ return 3; ++ } ++#endif ++ ++ /* PaX: glibc 2.4+ generates sethi/jmpl instead of save/call */ ++ if ((save & 0xFFC00000U) == 0x05000000U && ++ (call & 0xFFFFE000U) == 0x85C0A000U && ++ nop == 0x01000000U) ++ { ++ regs->u_regs[UREG_G1] = (sethi & 0x003FFFFFU) << 10; ++ regs->u_regs[UREG_G2] = addr + 4; ++ addr = (save & 0x003FFFFFU) << 10; ++ addr += (((call | 0xFFFFE000U) ^ 0x00001000U) + 0x00001000U); ++ regs->pc = addr; ++ regs->npc = addr+4; ++ return 3; ++ } ++ } ++ } while (0); ++ ++ do { /* PaX: unpatched PLT emulation step 2 */ ++ unsigned int save, call, nop; ++ ++ err = get_user(save, (unsigned int *)(regs->pc-4)); ++ err |= get_user(call, (unsigned int *)regs->pc); ++ err |= get_user(nop, (unsigned int *)(regs->pc+4)); ++ if (err) ++ break; ++ ++ if (save == 0x9DE3BFA8U && ++ (call & 0xC0000000U) == 0x40000000U && ++ nop == 0x01000000U) ++ { ++ unsigned int dl_resolve = regs->pc + ((((call | 0xC0000000U) ^ 0x20000000U) + 0x20000000U) << 2); ++ ++ regs->u_regs[UREG_RETPC] = regs->pc; ++ regs->pc = dl_resolve; ++ regs->npc = dl_resolve+4; ++ return 3; ++ } ++ } while (0); ++#endif ++ ++ return 1; ++} ++ ++void pax_report_insns(void *pc, void *sp) ++{ ++ unsigned long i; ++ ++ printk(KERN_ERR "PAX: bytes at PC: "); ++ for (i = 0; i < 8; i++) { ++ unsigned int c; ++ if (get_user(c, (unsigned int *)pc+i)) ++ printk(KERN_CONT "???????? "); ++ else ++ printk(KERN_CONT "%08x ", c); ++ } ++ printk("\n"); ++} ++#endif ++ + static noinline void do_fault_siginfo(int code, int sig, struct pt_regs *regs, + int text_fault) + { +@@ -282,6 +547,24 @@ good_area: + if(!(vma->vm_flags & VM_WRITE)) + goto bad_area; + } else { ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if ((mm->pax_flags & MF_PAX_PAGEEXEC) && text_fault && !(vma->vm_flags & VM_EXEC)) { ++ up_read(&mm->mmap_sem); ++ switch (pax_handle_fetch_fault(regs)) { ++ ++#ifdef CONFIG_PAX_EMUPLT ++ case 2: ++ case 3: ++ return; ++#endif ++ ++ } ++ pax_report_fault(regs, (void *)regs->pc, (void *)regs->u_regs[UREG_FP]); ++ do_group_exit(SIGKILL); ++ } ++#endif ++ + /* Allow reads even for write-only mappings */ + if(!(vma->vm_flags & (VM_READ | VM_EXEC))) + goto bad_area; +diff -urNp linux-2.6.38.3/arch/sparc/mm/fault_64.c linux-2.6.38.3/arch/sparc/mm/fault_64.c +--- linux-2.6.38.3/arch/sparc/mm/fault_64.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/mm/fault_64.c 2011-04-17 15:57:32.000000000 -0400 +@@ -21,6 +21,9 @@ + #include <linux/kprobes.h> + #include <linux/kdebug.h> + #include <linux/percpu.h> ++#include <linux/slab.h> ++#include <linux/pagemap.h> ++#include <linux/compiler.h> + + #include <asm/page.h> + #include <asm/pgtable.h> +@@ -74,7 +77,7 @@ static void __kprobes bad_kernel_pc(stru + printk(KERN_CRIT "OOPS: Bogus kernel PC [%016lx] in fault handler\n", + regs->tpc); + printk(KERN_CRIT "OOPS: RPC [%016lx]\n", regs->u_regs[15]); +- printk("OOPS: RPC <%pS>\n", (void *) regs->u_regs[15]); ++ printk("OOPS: RPC <%pA>\n", (void *) regs->u_regs[15]); + printk(KERN_CRIT "OOPS: Fault was to vaddr[%lx]\n", vaddr); + dump_stack(); + unhandled_fault(regs->tpc, current, regs); +@@ -272,6 +275,457 @@ static void noinline __kprobes bogus_32b + show_regs(regs); + } + ++#ifdef CONFIG_PAX_PAGEEXEC ++#ifdef CONFIG_PAX_DLRESOLVE ++static void pax_emuplt_close(struct vm_area_struct *vma) ++{ ++ vma->vm_mm->call_dl_resolve = 0UL; ++} ++ ++static int pax_emuplt_fault(struct vm_area_struct *vma, struct vm_fault *vmf) ++{ ++ unsigned int *kaddr; ++ ++ vmf->page = alloc_page(GFP_HIGHUSER); ++ if (!vmf->page) ++ return VM_FAULT_OOM; ++ ++ kaddr = kmap(vmf->page); ++ memset(kaddr, 0, PAGE_SIZE); ++ kaddr[0] = 0x9DE3BFA8U; /* save */ ++ flush_dcache_page(vmf->page); ++ kunmap(vmf->page); ++ return VM_FAULT_MAJOR; ++} ++ ++static const struct vm_operations_struct pax_vm_ops = { ++ .close = pax_emuplt_close, ++ .fault = pax_emuplt_fault ++}; ++ ++static int pax_insert_vma(struct vm_area_struct *vma, unsigned long addr) ++{ ++ int ret; ++ ++ INIT_LIST_HEAD(&vma->anon_vma_chain); ++ vma->vm_mm = current->mm; ++ vma->vm_start = addr; ++ vma->vm_end = addr + PAGE_SIZE; ++ vma->vm_flags = VM_READ | VM_EXEC | VM_MAYREAD | VM_MAYEXEC; ++ vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); ++ vma->vm_ops = &pax_vm_ops; ++ ++ ret = insert_vm_struct(current->mm, vma); ++ if (ret) ++ return ret; ++ ++ ++current->mm->total_vm; ++ return 0; ++} ++#endif ++ ++/* ++ * PaX: decide what to do with offenders (regs->tpc = fault address) ++ * ++ * returns 1 when task should be killed ++ * 2 when patched PLT trampoline was detected ++ * 3 when unpatched PLT trampoline was detected ++ */ ++static int pax_handle_fetch_fault(struct pt_regs *regs) ++{ ++ ++#ifdef CONFIG_PAX_EMUPLT ++ int err; ++ ++ do { /* PaX: patched PLT emulation #1 */ ++ unsigned int sethi1, sethi2, jmpl; ++ ++ err = get_user(sethi1, (unsigned int *)regs->tpc); ++ err |= get_user(sethi2, (unsigned int *)(regs->tpc+4)); ++ err |= get_user(jmpl, (unsigned int *)(regs->tpc+8)); ++ ++ if (err) ++ break; ++ ++ if ((sethi1 & 0xFFC00000U) == 0x03000000U && ++ (sethi2 & 0xFFC00000U) == 0x03000000U && ++ (jmpl & 0xFFFFE000U) == 0x81C06000U) ++ { ++ unsigned long addr; ++ ++ regs->u_regs[UREG_G1] = (sethi2 & 0x003FFFFFU) << 10; ++ addr = regs->u_regs[UREG_G1]; ++ addr += (((jmpl | 0xFFFFFFFFFFFFE000UL) ^ 0x00001000UL) + 0x00001000UL); ++ ++ if (test_thread_flag(TIF_32BIT)) ++ addr &= 0xFFFFFFFFUL; ++ ++ regs->tpc = addr; ++ regs->tnpc = addr+4; ++ return 2; ++ } ++ } while (0); ++ ++ { /* PaX: patched PLT emulation #2 */ ++ unsigned int ba; ++ ++ err = get_user(ba, (unsigned int *)regs->tpc); ++ ++ if (!err && (ba & 0xFFC00000U) == 0x30800000U) { ++ unsigned long addr; ++ ++ addr = regs->tpc + ((((ba | 0xFFFFFFFFFFC00000UL) ^ 0x00200000UL) + 0x00200000UL) << 2); ++ ++ if (test_thread_flag(TIF_32BIT)) ++ addr &= 0xFFFFFFFFUL; ++ ++ regs->tpc = addr; ++ regs->tnpc = addr+4; ++ return 2; ++ } ++ } ++ ++ do { /* PaX: patched PLT emulation #3 */ ++ unsigned int sethi, jmpl, nop; ++ ++ err = get_user(sethi, (unsigned int *)regs->tpc); ++ err |= get_user(jmpl, (unsigned int *)(regs->tpc+4)); ++ err |= get_user(nop, (unsigned int *)(regs->tpc+8)); ++ ++ if (err) ++ break; ++ ++ if ((sethi & 0xFFC00000U) == 0x03000000U && ++ (jmpl & 0xFFFFE000U) == 0x81C06000U && ++ nop == 0x01000000U) ++ { ++ unsigned long addr; ++ ++ addr = (sethi & 0x003FFFFFU) << 10; ++ regs->u_regs[UREG_G1] = addr; ++ addr += (((jmpl | 0xFFFFFFFFFFFFE000UL) ^ 0x00001000UL) + 0x00001000UL); ++ ++ if (test_thread_flag(TIF_32BIT)) ++ addr &= 0xFFFFFFFFUL; ++ ++ regs->tpc = addr; ++ regs->tnpc = addr+4; ++ return 2; ++ } ++ } while (0); ++ ++ do { /* PaX: patched PLT emulation #4 */ ++ unsigned int sethi, mov1, call, mov2; ++ ++ err = get_user(sethi, (unsigned int *)regs->tpc); ++ err |= get_user(mov1, (unsigned int *)(regs->tpc+4)); ++ err |= get_user(call, (unsigned int *)(regs->tpc+8)); ++ err |= get_user(mov2, (unsigned int *)(regs->tpc+12)); ++ ++ if (err) ++ break; ++ ++ if ((sethi & 0xFFC00000U) == 0x03000000U && ++ mov1 == 0x8210000FU && ++ (call & 0xC0000000U) == 0x40000000U && ++ mov2 == 0x9E100001U) ++ { ++ unsigned long addr; ++ ++ regs->u_regs[UREG_G1] = regs->u_regs[UREG_RETPC]; ++ addr = regs->tpc + 4 + ((((call | 0xFFFFFFFFC0000000UL) ^ 0x20000000UL) + 0x20000000UL) << 2); ++ ++ if (test_thread_flag(TIF_32BIT)) ++ addr &= 0xFFFFFFFFUL; ++ ++ regs->tpc = addr; ++ regs->tnpc = addr+4; ++ return 2; ++ } ++ } while (0); ++ ++ do { /* PaX: patched PLT emulation #5 */ ++ unsigned int sethi, sethi1, sethi2, or1, or2, sllx, jmpl, nop; ++ ++ err = get_user(sethi, (unsigned int *)regs->tpc); ++ err |= get_user(sethi1, (unsigned int *)(regs->tpc+4)); ++ err |= get_user(sethi2, (unsigned int *)(regs->tpc+8)); ++ err |= get_user(or1, (unsigned int *)(regs->tpc+12)); ++ err |= get_user(or2, (unsigned int *)(regs->tpc+16)); ++ err |= get_user(sllx, (unsigned int *)(regs->tpc+20)); ++ err |= get_user(jmpl, (unsigned int *)(regs->tpc+24)); ++ err |= get_user(nop, (unsigned int *)(regs->tpc+28)); ++ ++ if (err) ++ break; ++ ++ if ((sethi & 0xFFC00000U) == 0x03000000U && ++ (sethi1 & 0xFFC00000U) == 0x03000000U && ++ (sethi2 & 0xFFC00000U) == 0x0B000000U && ++ (or1 & 0xFFFFE000U) == 0x82106000U && ++ (or2 & 0xFFFFE000U) == 0x8A116000U && ++ sllx == 0x83287020U && ++ jmpl == 0x81C04005U && ++ nop == 0x01000000U) ++ { ++ unsigned long addr; ++ ++ regs->u_regs[UREG_G1] = ((sethi1 & 0x003FFFFFU) << 10) | (or1 & 0x000003FFU); ++ regs->u_regs[UREG_G1] <<= 32; ++ regs->u_regs[UREG_G5] = ((sethi2 & 0x003FFFFFU) << 10) | (or2 & 0x000003FFU); ++ addr = regs->u_regs[UREG_G1] + regs->u_regs[UREG_G5]; ++ regs->tpc = addr; ++ regs->tnpc = addr+4; ++ return 2; ++ } ++ } while (0); ++ ++ do { /* PaX: patched PLT emulation #6 */ ++ unsigned int sethi, sethi1, sethi2, sllx, or, jmpl, nop; ++ ++ err = get_user(sethi, (unsigned int *)regs->tpc); ++ err |= get_user(sethi1, (unsigned int *)(regs->tpc+4)); ++ err |= get_user(sethi2, (unsigned int *)(regs->tpc+8)); ++ err |= get_user(sllx, (unsigned int *)(regs->tpc+12)); ++ err |= get_user(or, (unsigned int *)(regs->tpc+16)); ++ err |= get_user(jmpl, (unsigned int *)(regs->tpc+20)); ++ err |= get_user(nop, (unsigned int *)(regs->tpc+24)); ++ ++ if (err) ++ break; ++ ++ if ((sethi & 0xFFC00000U) == 0x03000000U && ++ (sethi1 & 0xFFC00000U) == 0x03000000U && ++ (sethi2 & 0xFFC00000U) == 0x0B000000U && ++ sllx == 0x83287020U && ++ (or & 0xFFFFE000U) == 0x8A116000U && ++ jmpl == 0x81C04005U && ++ nop == 0x01000000U) ++ { ++ unsigned long addr; ++ ++ regs->u_regs[UREG_G1] = (sethi1 & 0x003FFFFFU) << 10; ++ regs->u_regs[UREG_G1] <<= 32; ++ regs->u_regs[UREG_G5] = ((sethi2 & 0x003FFFFFU) << 10) | (or & 0x3FFU); ++ addr = regs->u_regs[UREG_G1] + regs->u_regs[UREG_G5]; ++ regs->tpc = addr; ++ regs->tnpc = addr+4; ++ return 2; ++ } ++ } while (0); ++ ++ do { /* PaX: unpatched PLT emulation step 1 */ ++ unsigned int sethi, ba, nop; ++ ++ err = get_user(sethi, (unsigned int *)regs->tpc); ++ err |= get_user(ba, (unsigned int *)(regs->tpc+4)); ++ err |= get_user(nop, (unsigned int *)(regs->tpc+8)); ++ ++ if (err) ++ break; ++ ++ if ((sethi & 0xFFC00000U) == 0x03000000U && ++ ((ba & 0xFFC00000U) == 0x30800000U || (ba & 0xFFF80000U) == 0x30680000U) && ++ nop == 0x01000000U) ++ { ++ unsigned long addr; ++ unsigned int save, call; ++ unsigned int sethi1, sethi2, or1, or2, sllx, add, jmpl; ++ ++ if ((ba & 0xFFC00000U) == 0x30800000U) ++ addr = regs->tpc + 4 + ((((ba | 0xFFFFFFFFFFC00000UL) ^ 0x00200000UL) + 0x00200000UL) << 2); ++ else ++ addr = regs->tpc + 4 + ((((ba | 0xFFFFFFFFFFF80000UL) ^ 0x00040000UL) + 0x00040000UL) << 2); ++ ++ if (test_thread_flag(TIF_32BIT)) ++ addr &= 0xFFFFFFFFUL; ++ ++ err = get_user(save, (unsigned int *)addr); ++ err |= get_user(call, (unsigned int *)(addr+4)); ++ err |= get_user(nop, (unsigned int *)(addr+8)); ++ if (err) ++ break; ++ ++#ifdef CONFIG_PAX_DLRESOLVE ++ if (save == 0x9DE3BFA8U && ++ (call & 0xC0000000U) == 0x40000000U && ++ nop == 0x01000000U) ++ { ++ struct vm_area_struct *vma; ++ unsigned long call_dl_resolve; ++ ++ down_read(¤t->mm->mmap_sem); ++ call_dl_resolve = current->mm->call_dl_resolve; ++ up_read(¤t->mm->mmap_sem); ++ if (likely(call_dl_resolve)) ++ goto emulate; ++ ++ vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); ++ ++ down_write(¤t->mm->mmap_sem); ++ if (current->mm->call_dl_resolve) { ++ call_dl_resolve = current->mm->call_dl_resolve; ++ up_write(¤t->mm->mmap_sem); ++ if (vma) ++ kmem_cache_free(vm_area_cachep, vma); ++ goto emulate; ++ } ++ ++ call_dl_resolve = get_unmapped_area(NULL, 0UL, PAGE_SIZE, 0UL, MAP_PRIVATE); ++ if (!vma || (call_dl_resolve & ~PAGE_MASK)) { ++ up_write(¤t->mm->mmap_sem); ++ if (vma) ++ kmem_cache_free(vm_area_cachep, vma); ++ return 1; ++ } ++ ++ if (pax_insert_vma(vma, call_dl_resolve)) { ++ up_write(¤t->mm->mmap_sem); ++ kmem_cache_free(vm_area_cachep, vma); ++ return 1; ++ } ++ ++ current->mm->call_dl_resolve = call_dl_resolve; ++ up_write(¤t->mm->mmap_sem); ++ ++emulate: ++ regs->u_regs[UREG_G1] = (sethi & 0x003FFFFFU) << 10; ++ regs->tpc = call_dl_resolve; ++ regs->tnpc = addr+4; ++ return 3; ++ } ++#endif ++ ++ /* PaX: glibc 2.4+ generates sethi/jmpl instead of save/call */ ++ if ((save & 0xFFC00000U) == 0x05000000U && ++ (call & 0xFFFFE000U) == 0x85C0A000U && ++ nop == 0x01000000U) ++ { ++ regs->u_regs[UREG_G1] = (sethi & 0x003FFFFFU) << 10; ++ regs->u_regs[UREG_G2] = addr + 4; ++ addr = (save & 0x003FFFFFU) << 10; ++ addr += (((call | 0xFFFFFFFFFFFFE000UL) ^ 0x00001000UL) + 0x00001000UL); ++ ++ if (test_thread_flag(TIF_32BIT)) ++ addr &= 0xFFFFFFFFUL; ++ ++ regs->tpc = addr; ++ regs->tnpc = addr+4; ++ return 3; ++ } ++ ++ /* PaX: 64-bit PLT stub */ ++ err = get_user(sethi1, (unsigned int *)addr); ++ err |= get_user(sethi2, (unsigned int *)(addr+4)); ++ err |= get_user(or1, (unsigned int *)(addr+8)); ++ err |= get_user(or2, (unsigned int *)(addr+12)); ++ err |= get_user(sllx, (unsigned int *)(addr+16)); ++ err |= get_user(add, (unsigned int *)(addr+20)); ++ err |= get_user(jmpl, (unsigned int *)(addr+24)); ++ err |= get_user(nop, (unsigned int *)(addr+28)); ++ if (err) ++ break; ++ ++ if ((sethi1 & 0xFFC00000U) == 0x09000000U && ++ (sethi2 & 0xFFC00000U) == 0x0B000000U && ++ (or1 & 0xFFFFE000U) == 0x88112000U && ++ (or2 & 0xFFFFE000U) == 0x8A116000U && ++ sllx == 0x89293020U && ++ add == 0x8A010005U && ++ jmpl == 0x89C14000U && ++ nop == 0x01000000U) ++ { ++ regs->u_regs[UREG_G1] = (sethi & 0x003FFFFFU) << 10; ++ regs->u_regs[UREG_G4] = ((sethi1 & 0x003FFFFFU) << 10) | (or1 & 0x000003FFU); ++ regs->u_regs[UREG_G4] <<= 32; ++ regs->u_regs[UREG_G5] = ((sethi2 & 0x003FFFFFU) << 10) | (or2 & 0x000003FFU); ++ regs->u_regs[UREG_G5] += regs->u_regs[UREG_G4]; ++ regs->u_regs[UREG_G4] = addr + 24; ++ addr = regs->u_regs[UREG_G5]; ++ regs->tpc = addr; ++ regs->tnpc = addr+4; ++ return 3; ++ } ++ } ++ } while (0); ++ ++#ifdef CONFIG_PAX_DLRESOLVE ++ do { /* PaX: unpatched PLT emulation step 2 */ ++ unsigned int save, call, nop; ++ ++ err = get_user(save, (unsigned int *)(regs->tpc-4)); ++ err |= get_user(call, (unsigned int *)regs->tpc); ++ err |= get_user(nop, (unsigned int *)(regs->tpc+4)); ++ if (err) ++ break; ++ ++ if (save == 0x9DE3BFA8U && ++ (call & 0xC0000000U) == 0x40000000U && ++ nop == 0x01000000U) ++ { ++ unsigned long dl_resolve = regs->tpc + ((((call | 0xFFFFFFFFC0000000UL) ^ 0x20000000UL) + 0x20000000UL) << 2); ++ ++ if (test_thread_flag(TIF_32BIT)) ++ dl_resolve &= 0xFFFFFFFFUL; ++ ++ regs->u_regs[UREG_RETPC] = regs->tpc; ++ regs->tpc = dl_resolve; ++ regs->tnpc = dl_resolve+4; ++ return 3; ++ } ++ } while (0); ++#endif ++ ++ do { /* PaX: patched PLT emulation #7, must be AFTER the unpatched PLT emulation */ ++ unsigned int sethi, ba, nop; ++ ++ err = get_user(sethi, (unsigned int *)regs->tpc); ++ err |= get_user(ba, (unsigned int *)(regs->tpc+4)); ++ err |= get_user(nop, (unsigned int *)(regs->tpc+8)); ++ ++ if (err) ++ break; ++ ++ if ((sethi & 0xFFC00000U) == 0x03000000U && ++ (ba & 0xFFF00000U) == 0x30600000U && ++ nop == 0x01000000U) ++ { ++ unsigned long addr; ++ ++ addr = (sethi & 0x003FFFFFU) << 10; ++ regs->u_regs[UREG_G1] = addr; ++ addr = regs->tpc + ((((ba | 0xFFFFFFFFFFF80000UL) ^ 0x00040000UL) + 0x00040000UL) << 2); ++ ++ if (test_thread_flag(TIF_32BIT)) ++ addr &= 0xFFFFFFFFUL; ++ ++ regs->tpc = addr; ++ regs->tnpc = addr+4; ++ return 2; ++ } ++ } while (0); ++ ++#endif ++ ++ return 1; ++} ++ ++void pax_report_insns(void *pc, void *sp) ++{ ++ unsigned long i; ++ ++ printk(KERN_ERR "PAX: bytes at PC: "); ++ for (i = 0; i < 8; i++) { ++ unsigned int c; ++ if (get_user(c, (unsigned int *)pc+i)) ++ printk(KERN_CONT "???????? "); ++ else ++ printk(KERN_CONT "%08x ", c); ++ } ++ printk("\n"); ++} ++#endif ++ + asmlinkage void __kprobes do_sparc64_fault(struct pt_regs *regs) + { + struct mm_struct *mm = current->mm; +@@ -340,6 +794,29 @@ asmlinkage void __kprobes do_sparc64_fau + if (!vma) + goto bad_area; + ++#ifdef CONFIG_PAX_PAGEEXEC ++ /* PaX: detect ITLB misses on non-exec pages */ ++ if ((mm->pax_flags & MF_PAX_PAGEEXEC) && vma->vm_start <= address && ++ !(vma->vm_flags & VM_EXEC) && (fault_code & FAULT_CODE_ITLB)) ++ { ++ if (address != regs->tpc) ++ goto good_area; ++ ++ up_read(&mm->mmap_sem); ++ switch (pax_handle_fetch_fault(regs)) { ++ ++#ifdef CONFIG_PAX_EMUPLT ++ case 2: ++ case 3: ++ return; ++#endif ++ ++ } ++ pax_report_fault(regs, (void *)regs->tpc, (void *)(regs->u_regs[UREG_FP] + STACK_BIAS)); ++ do_group_exit(SIGKILL); ++ } ++#endif ++ + /* Pure DTLB misses do not tell us whether the fault causing + * load/store/atomic was a write or not, it only says that there + * was no match. So in such a case we (carefully) read the +diff -urNp linux-2.6.38.3/arch/sparc/mm/hugetlbpage.c linux-2.6.38.3/arch/sparc/mm/hugetlbpage.c +--- linux-2.6.38.3/arch/sparc/mm/hugetlbpage.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/mm/hugetlbpage.c 2011-04-17 15:57:32.000000000 -0400 +@@ -68,7 +68,7 @@ full_search: + } + return -ENOMEM; + } +- if (likely(!vma || addr + len <= vma->vm_start)) { ++ if (likely(check_heap_stack_gap(vma, addr, len))) { + /* + * Remember the place where we stopped the search: + */ +@@ -107,7 +107,7 @@ hugetlb_get_unmapped_area_topdown(struct + /* make sure it can fit in the remaining address space */ + if (likely(addr > len)) { + vma = find_vma(mm, addr-len); +- if (!vma || addr <= vma->vm_start) { ++ if (check_heap_stack_gap(vma, addr - len, len)) { + /* remember the address as a hint for next time */ + return (mm->free_area_cache = addr-len); + } +@@ -116,16 +116,17 @@ hugetlb_get_unmapped_area_topdown(struct + if (unlikely(mm->mmap_base < len)) + goto bottomup; + +- addr = (mm->mmap_base-len) & HPAGE_MASK; ++ addr = mm->mmap_base - len; + + do { ++ addr &= HPAGE_MASK; + /* + * Lookup failure means no vma is above this address, + * else if new region fits below vma->vm_start, + * return with success: + */ + vma = find_vma(mm, addr); +- if (likely(!vma || addr+len <= vma->vm_start)) { ++ if (likely(check_heap_stack_gap(vma, addr, len))) { + /* remember the address as a hint for next time */ + return (mm->free_area_cache = addr); + } +@@ -135,8 +136,8 @@ hugetlb_get_unmapped_area_topdown(struct + mm->cached_hole_size = vma->vm_start - addr; + + /* try just below the current vma->vm_start */ +- addr = (vma->vm_start-len) & HPAGE_MASK; +- } while (likely(len < vma->vm_start)); ++ addr = skip_heap_stack_gap(vma, len); ++ } while (!IS_ERR_VALUE(addr)); + + bottomup: + /* +@@ -182,8 +183,7 @@ hugetlb_get_unmapped_area(struct file *f + if (addr) { + addr = ALIGN(addr, HPAGE_SIZE); + vma = find_vma(mm, addr); +- if (task_size - len >= addr && +- (!vma || addr + len <= vma->vm_start)) ++ if (task_size - len >= addr && check_heap_stack_gap(vma, addr, len)) + return addr; + } + if (mm->get_unmapped_area == arch_get_unmapped_area) +diff -urNp linux-2.6.38.3/arch/sparc/mm/init_32.c linux-2.6.38.3/arch/sparc/mm/init_32.c +--- linux-2.6.38.3/arch/sparc/mm/init_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/mm/init_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -318,6 +318,9 @@ extern void device_scan(void); + pgprot_t PAGE_SHARED __read_mostly; + EXPORT_SYMBOL(PAGE_SHARED); + ++pgprot_t PAGE_SHARED_NOEXEC __read_mostly; ++EXPORT_SYMBOL(PAGE_SHARED_NOEXEC); ++ + void __init paging_init(void) + { + switch(sparc_cpu_model) { +@@ -346,17 +349,17 @@ void __init paging_init(void) + + /* Initialize the protection map with non-constant, MMU dependent values. */ + protection_map[0] = PAGE_NONE; +- protection_map[1] = PAGE_READONLY; +- protection_map[2] = PAGE_COPY; +- protection_map[3] = PAGE_COPY; ++ protection_map[1] = PAGE_READONLY_NOEXEC; ++ protection_map[2] = PAGE_COPY_NOEXEC; ++ protection_map[3] = PAGE_COPY_NOEXEC; + protection_map[4] = PAGE_READONLY; + protection_map[5] = PAGE_READONLY; + protection_map[6] = PAGE_COPY; + protection_map[7] = PAGE_COPY; + protection_map[8] = PAGE_NONE; +- protection_map[9] = PAGE_READONLY; +- protection_map[10] = PAGE_SHARED; +- protection_map[11] = PAGE_SHARED; ++ protection_map[9] = PAGE_READONLY_NOEXEC; ++ protection_map[10] = PAGE_SHARED_NOEXEC; ++ protection_map[11] = PAGE_SHARED_NOEXEC; + protection_map[12] = PAGE_READONLY; + protection_map[13] = PAGE_READONLY; + protection_map[14] = PAGE_SHARED; +diff -urNp linux-2.6.38.3/arch/sparc/mm/Makefile linux-2.6.38.3/arch/sparc/mm/Makefile +--- linux-2.6.38.3/arch/sparc/mm/Makefile 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/mm/Makefile 2011-04-17 15:57:32.000000000 -0400 +@@ -2,7 +2,7 @@ + # + + asflags-y := -ansi +-ccflags-y := -Werror ++#ccflags-y := -Werror + + obj-$(CONFIG_SPARC64) += ultra.o tlb.o tsb.o + obj-y += fault_$(BITS).o +diff -urNp linux-2.6.38.3/arch/sparc/mm/srmmu.c linux-2.6.38.3/arch/sparc/mm/srmmu.c +--- linux-2.6.38.3/arch/sparc/mm/srmmu.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/sparc/mm/srmmu.c 2011-04-17 15:57:32.000000000 -0400 +@@ -2200,6 +2200,13 @@ void __init ld_mmu_srmmu(void) + PAGE_SHARED = pgprot_val(SRMMU_PAGE_SHARED); + BTFIXUPSET_INT(page_copy, pgprot_val(SRMMU_PAGE_COPY)); + BTFIXUPSET_INT(page_readonly, pgprot_val(SRMMU_PAGE_RDONLY)); ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ PAGE_SHARED_NOEXEC = pgprot_val(SRMMU_PAGE_SHARED_NOEXEC); ++ BTFIXUPSET_INT(page_copy_noexec, pgprot_val(SRMMU_PAGE_COPY_NOEXEC)); ++ BTFIXUPSET_INT(page_readonly_noexec, pgprot_val(SRMMU_PAGE_RDONLY_NOEXEC)); ++#endif ++ + BTFIXUPSET_INT(page_kernel, pgprot_val(SRMMU_PAGE_KERNEL)); + page_kernel = pgprot_val(SRMMU_PAGE_KERNEL); + +diff -urNp linux-2.6.38.3/arch/um/include/asm/kmap_types.h linux-2.6.38.3/arch/um/include/asm/kmap_types.h +--- linux-2.6.38.3/arch/um/include/asm/kmap_types.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/um/include/asm/kmap_types.h 2011-04-17 15:57:32.000000000 -0400 +@@ -23,6 +23,7 @@ enum km_type { + KM_IRQ1, + KM_SOFTIRQ0, + KM_SOFTIRQ1, ++ KM_CLEARPAGE, + KM_TYPE_NR + }; + +diff -urNp linux-2.6.38.3/arch/um/include/asm/page.h linux-2.6.38.3/arch/um/include/asm/page.h +--- linux-2.6.38.3/arch/um/include/asm/page.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/um/include/asm/page.h 2011-04-17 15:57:32.000000000 -0400 +@@ -14,6 +14,9 @@ + #define PAGE_SIZE (_AC(1, UL) << PAGE_SHIFT) + #define PAGE_MASK (~(PAGE_SIZE-1)) + ++#define ktla_ktva(addr) (addr) ++#define ktva_ktla(addr) (addr) ++ + #ifndef __ASSEMBLY__ + + struct page; +diff -urNp linux-2.6.38.3/arch/um/kernel/process.c linux-2.6.38.3/arch/um/kernel/process.c +--- linux-2.6.38.3/arch/um/kernel/process.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/um/kernel/process.c 2011-04-17 15:57:32.000000000 -0400 +@@ -404,22 +404,6 @@ int singlestepping(void * t) + return 2; + } + +-/* +- * Only x86 and x86_64 have an arch_align_stack(). +- * All other arches have "#define arch_align_stack(x) (x)" +- * in their asm/system.h +- * As this is included in UML from asm-um/system-generic.h, +- * we can use it to behave as the subarch does. +- */ +-#ifndef arch_align_stack +-unsigned long arch_align_stack(unsigned long sp) +-{ +- if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space) +- sp -= get_random_int() % 8192; +- return sp & ~0xf; +-} +-#endif +- + unsigned long get_wchan(struct task_struct *p) + { + unsigned long stack_page, sp, ip; +diff -urNp linux-2.6.38.3/arch/um/sys-i386/syscalls.c linux-2.6.38.3/arch/um/sys-i386/syscalls.c +--- linux-2.6.38.3/arch/um/sys-i386/syscalls.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/um/sys-i386/syscalls.c 2011-04-17 15:57:32.000000000 -0400 +@@ -11,6 +11,21 @@ + #include "asm/uaccess.h" + #include "asm/unistd.h" + ++int i386_mmap_check(unsigned long addr, unsigned long len, unsigned long flags) ++{ ++ unsigned long pax_task_size = TASK_SIZE; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (current->mm->pax_flags & MF_PAX_SEGMEXEC) ++ pax_task_size = SEGMEXEC_TASK_SIZE; ++#endif ++ ++ if (len > pax_task_size || addr > pax_task_size - len) ++ return -EINVAL; ++ ++ return 0; ++} ++ + /* + * The prototype on i386 is: + * +diff -urNp linux-2.6.38.3/arch/x86/boot/bitops.h linux-2.6.38.3/arch/x86/boot/bitops.h +--- linux-2.6.38.3/arch/x86/boot/bitops.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/boot/bitops.h 2011-04-17 15:57:32.000000000 -0400 +@@ -26,7 +26,7 @@ static inline int variable_test_bit(int + u8 v; + const u32 *p = (const u32 *)addr; + +- asm("btl %2,%1; setc %0" : "=qm" (v) : "m" (*p), "Ir" (nr)); ++ asm volatile("btl %2,%1; setc %0" : "=qm" (v) : "m" (*p), "Ir" (nr)); + return v; + } + +@@ -37,7 +37,7 @@ static inline int variable_test_bit(int + + static inline void set_bit(int nr, void *addr) + { +- asm("btsl %1,%0" : "+m" (*(u32 *)addr) : "Ir" (nr)); ++ asm volatile("btsl %1,%0" : "+m" (*(u32 *)addr) : "Ir" (nr)); + } + + #endif /* BOOT_BITOPS_H */ +diff -urNp linux-2.6.38.3/arch/x86/boot/boot.h linux-2.6.38.3/arch/x86/boot/boot.h +--- linux-2.6.38.3/arch/x86/boot/boot.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/boot/boot.h 2011-04-17 15:57:32.000000000 -0400 +@@ -85,7 +85,7 @@ static inline void io_delay(void) + static inline u16 ds(void) + { + u16 seg; +- asm("movw %%ds,%0" : "=rm" (seg)); ++ asm volatile("movw %%ds,%0" : "=rm" (seg)); + return seg; + } + +@@ -181,7 +181,7 @@ static inline void wrgs32(u32 v, addr_t + static inline int memcmp(const void *s1, const void *s2, size_t len) + { + u8 diff; +- asm("repe; cmpsb; setnz %0" ++ asm volatile("repe; cmpsb; setnz %0" + : "=qm" (diff), "+D" (s1), "+S" (s2), "+c" (len)); + return diff; + } +diff -urNp linux-2.6.38.3/arch/x86/boot/compressed/head_32.S linux-2.6.38.3/arch/x86/boot/compressed/head_32.S +--- linux-2.6.38.3/arch/x86/boot/compressed/head_32.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/boot/compressed/head_32.S 2011-04-17 15:57:32.000000000 -0400 +@@ -76,7 +76,7 @@ ENTRY(startup_32) + notl %eax + andl %eax, %ebx + #else +- movl $LOAD_PHYSICAL_ADDR, %ebx ++ movl $____LOAD_PHYSICAL_ADDR, %ebx + #endif + + /* Target address to relocate to for decompression */ +@@ -162,7 +162,7 @@ relocated: + * and where it was actually loaded. + */ + movl %ebp, %ebx +- subl $LOAD_PHYSICAL_ADDR, %ebx ++ subl $____LOAD_PHYSICAL_ADDR, %ebx + jz 2f /* Nothing to be done if loaded at compiled addr. */ + /* + * Process relocations. +@@ -170,8 +170,7 @@ relocated: + + 1: subl $4, %edi + movl (%edi), %ecx +- testl %ecx, %ecx +- jz 2f ++ jecxz 2f + addl %ebx, -__PAGE_OFFSET(%ebx, %ecx) + jmp 1b + 2: +diff -urNp linux-2.6.38.3/arch/x86/boot/compressed/head_64.S linux-2.6.38.3/arch/x86/boot/compressed/head_64.S +--- linux-2.6.38.3/arch/x86/boot/compressed/head_64.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/boot/compressed/head_64.S 2011-04-17 15:57:32.000000000 -0400 +@@ -91,7 +91,7 @@ ENTRY(startup_32) + notl %eax + andl %eax, %ebx + #else +- movl $LOAD_PHYSICAL_ADDR, %ebx ++ movl $____LOAD_PHYSICAL_ADDR, %ebx + #endif + + /* Target address to relocate to for decompression */ +@@ -233,7 +233,7 @@ ENTRY(startup_64) + notq %rax + andq %rax, %rbp + #else +- movq $LOAD_PHYSICAL_ADDR, %rbp ++ movq $____LOAD_PHYSICAL_ADDR, %rbp + #endif + + /* Target address to relocate to for decompression */ +diff -urNp linux-2.6.38.3/arch/x86/boot/compressed/misc.c linux-2.6.38.3/arch/x86/boot/compressed/misc.c +--- linux-2.6.38.3/arch/x86/boot/compressed/misc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/boot/compressed/misc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -310,7 +310,7 @@ static void parse_elf(void *output) + case PT_LOAD: + #ifdef CONFIG_RELOCATABLE + dest = output; +- dest += (phdr->p_paddr - LOAD_PHYSICAL_ADDR); ++ dest += (phdr->p_paddr - ____LOAD_PHYSICAL_ADDR); + #else + dest = (void *)(phdr->p_paddr); + #endif +@@ -363,7 +363,7 @@ asmlinkage void decompress_kernel(void * + error("Destination address too large"); + #endif + #ifndef CONFIG_RELOCATABLE +- if ((unsigned long)output != LOAD_PHYSICAL_ADDR) ++ if ((unsigned long)output != ____LOAD_PHYSICAL_ADDR) + error("Wrong destination address"); + #endif + +diff -urNp linux-2.6.38.3/arch/x86/boot/compressed/relocs.c linux-2.6.38.3/arch/x86/boot/compressed/relocs.c +--- linux-2.6.38.3/arch/x86/boot/compressed/relocs.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/boot/compressed/relocs.c 2011-04-17 15:57:32.000000000 -0400 +@@ -13,8 +13,11 @@ + + static void die(char *fmt, ...); + ++#include "../../../../include/generated/autoconf.h" ++ + #define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0])) + static Elf32_Ehdr ehdr; ++static Elf32_Phdr *phdr; + static unsigned long reloc_count, reloc_idx; + static unsigned long *relocs; + +@@ -270,9 +273,39 @@ static void read_ehdr(FILE *fp) + } + } + ++static void read_phdrs(FILE *fp) ++{ ++ unsigned int i; ++ ++ phdr = calloc(ehdr.e_phnum, sizeof(Elf32_Phdr)); ++ if (!phdr) { ++ die("Unable to allocate %d program headers\n", ++ ehdr.e_phnum); ++ } ++ if (fseek(fp, ehdr.e_phoff, SEEK_SET) < 0) { ++ die("Seek to %d failed: %s\n", ++ ehdr.e_phoff, strerror(errno)); ++ } ++ if (fread(phdr, sizeof(*phdr), ehdr.e_phnum, fp) != ehdr.e_phnum) { ++ die("Cannot read ELF program headers: %s\n", ++ strerror(errno)); ++ } ++ for(i = 0; i < ehdr.e_phnum; i++) { ++ phdr[i].p_type = elf32_to_cpu(phdr[i].p_type); ++ phdr[i].p_offset = elf32_to_cpu(phdr[i].p_offset); ++ phdr[i].p_vaddr = elf32_to_cpu(phdr[i].p_vaddr); ++ phdr[i].p_paddr = elf32_to_cpu(phdr[i].p_paddr); ++ phdr[i].p_filesz = elf32_to_cpu(phdr[i].p_filesz); ++ phdr[i].p_memsz = elf32_to_cpu(phdr[i].p_memsz); ++ phdr[i].p_flags = elf32_to_cpu(phdr[i].p_flags); ++ phdr[i].p_align = elf32_to_cpu(phdr[i].p_align); ++ } ++ ++} ++ + static void read_shdrs(FILE *fp) + { +- int i; ++ unsigned int i; + Elf32_Shdr shdr; + + secs = calloc(ehdr.e_shnum, sizeof(struct section)); +@@ -307,7 +340,7 @@ static void read_shdrs(FILE *fp) + + static void read_strtabs(FILE *fp) + { +- int i; ++ unsigned int i; + for (i = 0; i < ehdr.e_shnum; i++) { + struct section *sec = &secs[i]; + if (sec->shdr.sh_type != SHT_STRTAB) { +@@ -332,7 +365,7 @@ static void read_strtabs(FILE *fp) + + static void read_symtabs(FILE *fp) + { +- int i,j; ++ unsigned int i,j; + for (i = 0; i < ehdr.e_shnum; i++) { + struct section *sec = &secs[i]; + if (sec->shdr.sh_type != SHT_SYMTAB) { +@@ -365,7 +398,9 @@ static void read_symtabs(FILE *fp) + + static void read_relocs(FILE *fp) + { +- int i,j; ++ unsigned int i,j; ++ uint32_t base; ++ + for (i = 0; i < ehdr.e_shnum; i++) { + struct section *sec = &secs[i]; + if (sec->shdr.sh_type != SHT_REL) { +@@ -385,9 +420,18 @@ static void read_relocs(FILE *fp) + die("Cannot read symbol table: %s\n", + strerror(errno)); + } ++ base = 0; ++ for (j = 0; j < ehdr.e_phnum; j++) { ++ if (phdr[j].p_type != PT_LOAD ) ++ continue; ++ if (secs[sec->shdr.sh_info].shdr.sh_offset < phdr[j].p_offset || secs[sec->shdr.sh_info].shdr.sh_offset >= phdr[j].p_offset + phdr[j].p_filesz) ++ continue; ++ base = CONFIG_PAGE_OFFSET + phdr[j].p_paddr - phdr[j].p_vaddr; ++ break; ++ } + for (j = 0; j < sec->shdr.sh_size/sizeof(Elf32_Rel); j++) { + Elf32_Rel *rel = &sec->reltab[j]; +- rel->r_offset = elf32_to_cpu(rel->r_offset); ++ rel->r_offset = elf32_to_cpu(rel->r_offset) + base; + rel->r_info = elf32_to_cpu(rel->r_info); + } + } +@@ -396,14 +440,14 @@ static void read_relocs(FILE *fp) + + static void print_absolute_symbols(void) + { +- int i; ++ unsigned int i; + printf("Absolute symbols\n"); + printf(" Num: Value Size Type Bind Visibility Name\n"); + for (i = 0; i < ehdr.e_shnum; i++) { + struct section *sec = &secs[i]; + char *sym_strtab; + Elf32_Sym *sh_symtab; +- int j; ++ unsigned int j; + + if (sec->shdr.sh_type != SHT_SYMTAB) { + continue; +@@ -431,14 +475,14 @@ static void print_absolute_symbols(void) + + static void print_absolute_relocs(void) + { +- int i, printed = 0; ++ unsigned int i, printed = 0; + + for (i = 0; i < ehdr.e_shnum; i++) { + struct section *sec = &secs[i]; + struct section *sec_applies, *sec_symtab; + char *sym_strtab; + Elf32_Sym *sh_symtab; +- int j; ++ unsigned int j; + if (sec->shdr.sh_type != SHT_REL) { + continue; + } +@@ -499,13 +543,13 @@ static void print_absolute_relocs(void) + + static void walk_relocs(void (*visit)(Elf32_Rel *rel, Elf32_Sym *sym)) + { +- int i; ++ unsigned int i; + /* Walk through the relocations */ + for (i = 0; i < ehdr.e_shnum; i++) { + char *sym_strtab; + Elf32_Sym *sh_symtab; + struct section *sec_applies, *sec_symtab; +- int j; ++ unsigned int j; + struct section *sec = &secs[i]; + + if (sec->shdr.sh_type != SHT_REL) { +@@ -530,6 +574,22 @@ static void walk_relocs(void (*visit)(El + !is_rel_reloc(sym_name(sym_strtab, sym))) { + continue; + } ++ /* Don't relocate actual per-cpu variables, they are absolute indices, not addresses */ ++ if (!strcmp(sec_name(sym->st_shndx), ".data..percpu") && strcmp(sym_name(sym_strtab, sym), "__per_cpu_load")) ++ continue; ++ ++#if defined(CONFIG_PAX_KERNEXEC) && defined(CONFIG_X86_32) ++ /* Don't relocate actual code, they are relocated implicitly by the base address of KERNEL_CS */ ++ if (!strcmp(sec_name(sym->st_shndx), ".module.text") && !strcmp(sym_name(sym_strtab, sym), "_etext")) ++ continue; ++ if (!strcmp(sec_name(sym->st_shndx), ".init.text")) ++ continue; ++ if (!strcmp(sec_name(sym->st_shndx), ".exit.text")) ++ continue; ++ if (!strcmp(sec_name(sym->st_shndx), ".text") && strcmp(sym_name(sym_strtab, sym), "__LOAD_PHYSICAL_ADDR")) ++ continue; ++#endif ++ + switch (r_type) { + case R_386_NONE: + case R_386_PC32: +@@ -571,7 +631,7 @@ static int cmp_relocs(const void *va, co + + static void emit_relocs(int as_text) + { +- int i; ++ unsigned int i; + /* Count how many relocations I have and allocate space for them. */ + reloc_count = 0; + walk_relocs(count_reloc); +@@ -665,6 +725,7 @@ int main(int argc, char **argv) + fname, strerror(errno)); + } + read_ehdr(fp); ++ read_phdrs(fp); + read_shdrs(fp); + read_strtabs(fp); + read_symtabs(fp); +diff -urNp linux-2.6.38.3/arch/x86/boot/cpucheck.c linux-2.6.38.3/arch/x86/boot/cpucheck.c +--- linux-2.6.38.3/arch/x86/boot/cpucheck.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/boot/cpucheck.c 2011-04-17 15:57:32.000000000 -0400 +@@ -74,7 +74,7 @@ static int has_fpu(void) + u16 fcw = -1, fsw = -1; + u32 cr0; + +- asm("movl %%cr0,%0" : "=r" (cr0)); ++ asm volatile("movl %%cr0,%0" : "=r" (cr0)); + if (cr0 & (X86_CR0_EM|X86_CR0_TS)) { + cr0 &= ~(X86_CR0_EM|X86_CR0_TS); + asm volatile("movl %0,%%cr0" : : "r" (cr0)); +@@ -90,7 +90,7 @@ static int has_eflag(u32 mask) + { + u32 f0, f1; + +- asm("pushfl ; " ++ asm volatile("pushfl ; " + "pushfl ; " + "popl %0 ; " + "movl %0,%1 ; " +@@ -115,7 +115,7 @@ static void get_flags(void) + set_bit(X86_FEATURE_FPU, cpu.flags); + + if (has_eflag(X86_EFLAGS_ID)) { +- asm("cpuid" ++ asm volatile("cpuid" + : "=a" (max_intel_level), + "=b" (cpu_vendor[0]), + "=d" (cpu_vendor[1]), +@@ -124,7 +124,7 @@ static void get_flags(void) + + if (max_intel_level >= 0x00000001 && + max_intel_level <= 0x0000ffff) { +- asm("cpuid" ++ asm volatile("cpuid" + : "=a" (tfms), + "=c" (cpu.flags[4]), + "=d" (cpu.flags[0]) +@@ -136,7 +136,7 @@ static void get_flags(void) + cpu.model += ((tfms >> 16) & 0xf) << 4; + } + +- asm("cpuid" ++ asm volatile("cpuid" + : "=a" (max_amd_level) + : "a" (0x80000000) + : "ebx", "ecx", "edx"); +@@ -144,7 +144,7 @@ static void get_flags(void) + if (max_amd_level >= 0x80000001 && + max_amd_level <= 0x8000ffff) { + u32 eax = 0x80000001; +- asm("cpuid" ++ asm volatile("cpuid" + : "+a" (eax), + "=c" (cpu.flags[6]), + "=d" (cpu.flags[1]) +@@ -203,9 +203,9 @@ int check_cpu(int *cpu_level_ptr, int *r + u32 ecx = MSR_K7_HWCR; + u32 eax, edx; + +- asm("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx)); ++ asm volatile("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx)); + eax &= ~(1 << 15); +- asm("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx)); ++ asm volatile("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx)); + + get_flags(); /* Make sure it really did something */ + err = check_flags(); +@@ -218,9 +218,9 @@ int check_cpu(int *cpu_level_ptr, int *r + u32 ecx = MSR_VIA_FCR; + u32 eax, edx; + +- asm("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx)); ++ asm volatile("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx)); + eax |= (1<<1)|(1<<7); +- asm("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx)); ++ asm volatile("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx)); + + set_bit(X86_FEATURE_CX8, cpu.flags); + err = check_flags(); +@@ -231,12 +231,12 @@ int check_cpu(int *cpu_level_ptr, int *r + u32 eax, edx; + u32 level = 1; + +- asm("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx)); +- asm("wrmsr" : : "a" (~0), "d" (edx), "c" (ecx)); +- asm("cpuid" ++ asm volatile("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx)); ++ asm volatile("wrmsr" : : "a" (~0), "d" (edx), "c" (ecx)); ++ asm volatile("cpuid" + : "+a" (level), "=d" (cpu.flags[0]) + : : "ecx", "ebx"); +- asm("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx)); ++ asm volatile("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx)); + + err = check_flags(); + } +diff -urNp linux-2.6.38.3/arch/x86/boot/header.S linux-2.6.38.3/arch/x86/boot/header.S +--- linux-2.6.38.3/arch/x86/boot/header.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/boot/header.S 2011-04-17 15:57:32.000000000 -0400 +@@ -224,7 +224,7 @@ setup_data: .quad 0 # 64-bit physical + # single linked list of + # struct setup_data + +-pref_address: .quad LOAD_PHYSICAL_ADDR # preferred load addr ++pref_address: .quad ____LOAD_PHYSICAL_ADDR # preferred load addr + + #define ZO_INIT_SIZE (ZO__end - ZO_startup_32 + ZO_z_extract_offset) + #define VO_INIT_SIZE (VO__end - VO__text) +diff -urNp linux-2.6.38.3/arch/x86/boot/memory.c linux-2.6.38.3/arch/x86/boot/memory.c +--- linux-2.6.38.3/arch/x86/boot/memory.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/boot/memory.c 2011-04-17 15:57:32.000000000 -0400 +@@ -19,7 +19,7 @@ + + static int detect_memory_e820(void) + { +- int count = 0; ++ unsigned int count = 0; + struct biosregs ireg, oreg; + struct e820entry *desc = boot_params.e820_map; + static struct e820entry buf; /* static so it is zeroed */ +diff -urNp linux-2.6.38.3/arch/x86/boot/video.c linux-2.6.38.3/arch/x86/boot/video.c +--- linux-2.6.38.3/arch/x86/boot/video.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/boot/video.c 2011-04-17 15:57:32.000000000 -0400 +@@ -96,7 +96,7 @@ static void store_mode_params(void) + static unsigned int get_entry(void) + { + char entry_buf[4]; +- int i, len = 0; ++ unsigned int i, len = 0; + int key; + unsigned int v; + +diff -urNp linux-2.6.38.3/arch/x86/boot/video-vesa.c linux-2.6.38.3/arch/x86/boot/video-vesa.c +--- linux-2.6.38.3/arch/x86/boot/video-vesa.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/boot/video-vesa.c 2011-04-17 15:57:32.000000000 -0400 +@@ -200,6 +200,7 @@ static void vesa_store_pm_info(void) + + boot_params.screen_info.vesapm_seg = oreg.es; + boot_params.screen_info.vesapm_off = oreg.di; ++ boot_params.screen_info.vesapm_size = oreg.cx; + } + + /* +diff -urNp linux-2.6.38.3/arch/x86/ia32/ia32_aout.c linux-2.6.38.3/arch/x86/ia32/ia32_aout.c +--- linux-2.6.38.3/arch/x86/ia32/ia32_aout.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/ia32/ia32_aout.c 2011-04-17 15:57:32.000000000 -0400 +@@ -162,6 +162,8 @@ static int aout_core_dump(long signr, st + unsigned long dump_start, dump_size; + struct user32 dump; + ++ memset(&dump, 0, sizeof(dump)); ++ + fs = get_fs(); + set_fs(KERNEL_DS); + has_dumped = 1; +diff -urNp linux-2.6.38.3/arch/x86/ia32/ia32entry.S linux-2.6.38.3/arch/x86/ia32/ia32entry.S +--- linux-2.6.38.3/arch/x86/ia32/ia32entry.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/ia32/ia32entry.S 2011-04-17 15:57:32.000000000 -0400 +@@ -13,6 +13,7 @@ + #include <asm/thread_info.h> + #include <asm/segment.h> + #include <asm/irqflags.h> ++#include <asm/pgtable.h> + #include <linux/linkage.h> + + /* Avoid __ASSEMBLER__'ifying <linux/audit.h> just for this. */ +@@ -93,6 +94,23 @@ ENTRY(native_irq_enable_sysexit) + ENDPROC(native_irq_enable_sysexit) + #endif + ++ .macro pax_enter_kernel_user ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ call pax_enter_kernel_user ++#endif ++ .endm ++ ++ .macro pax_exit_kernel_user ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ call pax_exit_kernel_user ++#endif ++#ifdef CONFIG_PAX_RANDKSTACK ++ push %rax ++ call pax_randomize_kstack ++ pop %rax ++#endif ++ .endm ++ + /* + * 32bit SYSENTER instruction entry. + * +@@ -119,7 +137,7 @@ ENTRY(ia32_sysenter_target) + CFI_REGISTER rsp,rbp + SWAPGS_UNSAFE_STACK + movq PER_CPU_VAR(kernel_stack), %rsp +- addq $(KERNEL_STACK_OFFSET),%rsp ++ pax_enter_kernel_user + /* + * No need to follow this irqs on/off section: the syscall + * disabled irqs, here we enable it straight after entry: +@@ -135,7 +153,8 @@ ENTRY(ia32_sysenter_target) + pushfq + CFI_ADJUST_CFA_OFFSET 8 + /*CFI_REL_OFFSET rflags,0*/ +- movl 8*3-THREAD_SIZE+TI_sysenter_return(%rsp), %r10d ++ GET_THREAD_INFO(%r10) ++ movl TI_sysenter_return(%r10), %r10d + CFI_REGISTER rip,r10 + pushq $__USER32_CS + CFI_ADJUST_CFA_OFFSET 8 +@@ -150,6 +169,12 @@ ENTRY(ia32_sysenter_target) + SAVE_ARGS 0,0,1 + /* no need to do an access_ok check here because rbp has been + 32bit zero extended */ ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ mov $PAX_USER_SHADOW_BASE,%r10 ++ add %r10,%rbp ++#endif ++ + 1: movl (%rbp),%ebp + .section __ex_table,"a" + .quad 1b,ia32_badarg +@@ -172,6 +197,7 @@ sysenter_dispatch: + testl $_TIF_ALLWORK_MASK,TI_flags(%r10) + jnz sysexit_audit + sysexit_from_sys_call: ++ pax_exit_kernel_user + andl $~TS_COMPAT,TI_status(%r10) + /* clear IF, that popfq doesn't enable interrupts early */ + andl $~0x200,EFLAGS-R11(%rsp) +@@ -283,19 +309,24 @@ ENDPROC(ia32_sysenter_target) + ENTRY(ia32_cstar_target) + CFI_STARTPROC32 simple + CFI_SIGNAL_FRAME +- CFI_DEF_CFA rsp,KERNEL_STACK_OFFSET ++ CFI_DEF_CFA rsp,0 + CFI_REGISTER rip,rcx + /*CFI_REGISTER rflags,r11*/ + SWAPGS_UNSAFE_STACK + movl %esp,%r8d + CFI_REGISTER rsp,r8 + movq PER_CPU_VAR(kernel_stack),%rsp ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ pax_enter_kernel_user ++#endif ++ + /* + * No need to follow this irqs on/off section: the syscall + * disabled irqs and here we enable it straight after entry: + */ + ENABLE_INTERRUPTS(CLBR_NONE) +- SAVE_ARGS 8,1,1 ++ SAVE_ARGS 8*6,1,1 + movl %eax,%eax /* zero extension */ + movq %rax,ORIG_RAX-ARGOFFSET(%rsp) + movq %rcx,RIP-ARGOFFSET(%rsp) +@@ -311,6 +342,12 @@ ENTRY(ia32_cstar_target) + /* no need to do an access_ok check here because r8 has been + 32bit zero extended */ + /* hardware stack frame is complete now */ ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ mov $PAX_USER_SHADOW_BASE,%r10 ++ add %r10,%r8 ++#endif ++ + 1: movl (%r8),%r9d + .section __ex_table,"a" + .quad 1b,ia32_badarg +@@ -323,7 +360,7 @@ ENTRY(ia32_cstar_target) + cmpq $IA32_NR_syscalls-1,%rax + ja ia32_badsys + cstar_do_call: +- IA32_ARG_FIXUP 1 ++ IA32_ARG_FIXUP 0 + cstar_dispatch: + call *ia32_sys_call_table(,%rax,8) + movq %rax,RAX-ARGOFFSET(%rsp) +@@ -333,6 +370,7 @@ cstar_dispatch: + testl $_TIF_ALLWORK_MASK,TI_flags(%r10) + jnz sysretl_audit + sysretl_from_sys_call: ++ pax_exit_kernel_user + andl $~TS_COMPAT,TI_status(%r10) + RESTORE_ARGS 1,-ARG_SKIP,1,1,1 + movl RIP-ARGOFFSET(%rsp),%ecx +@@ -415,6 +453,7 @@ ENTRY(ia32_syscall) + CFI_REL_OFFSET rip,RIP-RIP + PARAVIRT_ADJUST_EXCEPTION_FRAME + SWAPGS ++ pax_enter_kernel_user + /* + * No need to follow this irqs on/off section: the syscall + * disabled irqs and here we enable it straight after entry: +diff -urNp linux-2.6.38.3/arch/x86/ia32/ia32_signal.c linux-2.6.38.3/arch/x86/ia32/ia32_signal.c +--- linux-2.6.38.3/arch/x86/ia32/ia32_signal.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/ia32/ia32_signal.c 2011-04-17 15:57:32.000000000 -0400 +@@ -403,7 +403,7 @@ static void __user *get_sigframe(struct + sp -= frame_size; + /* Align the stack pointer according to the i386 ABI, + * i.e. so that on function entry ((sp + 4) & 15) == 0. */ +- sp = ((sp + 4) & -16ul) - 4; ++ sp = ((sp - 12) & -16ul) - 4; + return (void __user *) sp; + } + +@@ -461,7 +461,7 @@ int ia32_setup_frame(int sig, struct k_s + * These are actually not used anymore, but left because some + * gdb versions depend on them as a marker. + */ +- put_user_ex(*((u64 *)&code), (u64 *)frame->retcode); ++ put_user_ex(*((const u64 *)&code), (u64 *)frame->retcode); + } put_user_catch(err); + + if (err) +@@ -503,7 +503,7 @@ int ia32_setup_rt_frame(int sig, struct + 0xb8, + __NR_ia32_rt_sigreturn, + 0x80cd, +- 0, ++ 0 + }; + + frame = get_sigframe(ka, regs, sizeof(*frame), &fpstate); +@@ -533,16 +533,18 @@ int ia32_setup_rt_frame(int sig, struct + + if (ka->sa.sa_flags & SA_RESTORER) + restorer = ka->sa.sa_restorer; ++ else if (current->mm->context.vdso) ++ /* Return stub is in 32bit vsyscall page */ ++ restorer = VDSO32_SYMBOL(current->mm->context.vdso, rt_sigreturn); + else +- restorer = VDSO32_SYMBOL(current->mm->context.vdso, +- rt_sigreturn); ++ restorer = &frame->retcode; + put_user_ex(ptr_to_compat(restorer), &frame->pretcode); + + /* + * Not actually used anymore, but left because some gdb + * versions need it. + */ +- put_user_ex(*((u64 *)&code), (u64 *)frame->retcode); ++ put_user_ex(*((const u64 *)&code), (u64 *)frame->retcode); + } put_user_catch(err); + + if (err) +diff -urNp linux-2.6.38.3/arch/x86/include/asm/alternative.h linux-2.6.38.3/arch/x86/include/asm/alternative.h +--- linux-2.6.38.3/arch/x86/include/asm/alternative.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/alternative.h 2011-04-17 15:57:32.000000000 -0400 +@@ -94,7 +94,7 @@ static inline int alternatives_text_rese + ".section .discard,\"aw\",@progbits\n" \ + " .byte 0xff + (664f-663f) - (662b-661b)\n" /* rlen <= slen */ \ + ".previous\n" \ +- ".section .altinstr_replacement, \"ax\"\n" \ ++ ".section .altinstr_replacement, \"a\"\n" \ + "663:\n\t" newinstr "\n664:\n" /* replacement */ \ + ".previous" + +diff -urNp linux-2.6.38.3/arch/x86/include/asm/apm.h linux-2.6.38.3/arch/x86/include/asm/apm.h +--- linux-2.6.38.3/arch/x86/include/asm/apm.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/apm.h 2011-04-17 15:57:32.000000000 -0400 +@@ -34,7 +34,7 @@ static inline void apm_bios_call_asm(u32 + __asm__ __volatile__(APM_DO_ZERO_SEGS + "pushl %%edi\n\t" + "pushl %%ebp\n\t" +- "lcall *%%cs:apm_bios_entry\n\t" ++ "lcall *%%ss:apm_bios_entry\n\t" + "setc %%al\n\t" + "popl %%ebp\n\t" + "popl %%edi\n\t" +@@ -58,7 +58,7 @@ static inline u8 apm_bios_call_simple_as + __asm__ __volatile__(APM_DO_ZERO_SEGS + "pushl %%edi\n\t" + "pushl %%ebp\n\t" +- "lcall *%%cs:apm_bios_entry\n\t" ++ "lcall *%%ss:apm_bios_entry\n\t" + "setc %%bl\n\t" + "popl %%ebp\n\t" + "popl %%edi\n\t" +diff -urNp linux-2.6.38.3/arch/x86/include/asm/atomic64_32.h linux-2.6.38.3/arch/x86/include/asm/atomic64_32.h +--- linux-2.6.38.3/arch/x86/include/asm/atomic64_32.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/atomic64_32.h 2011-04-17 15:57:32.000000000 -0400 +@@ -12,6 +12,14 @@ typedef struct { + u64 __aligned(8) counter; + } atomic64_t; + ++#ifdef CONFIG_PAX_REFCOUNT ++typedef struct { ++ u64 __aligned(8) counter; ++} atomic64_unchecked_t; ++#else ++typedef atomic64_t atomic64_unchecked_t; ++#endif ++ + #define ATOMIC64_INIT(val) { (val) } + + #ifdef CONFIG_X86_CMPXCHG64 +diff -urNp linux-2.6.38.3/arch/x86/include/asm/atomic64_64.h linux-2.6.38.3/arch/x86/include/asm/atomic64_64.h +--- linux-2.6.38.3/arch/x86/include/asm/atomic64_64.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/atomic64_64.h 2011-04-17 15:57:32.000000000 -0400 +@@ -18,7 +18,19 @@ + */ + static inline long atomic64_read(const atomic64_t *v) + { +- return (*(volatile long *)&(v)->counter); ++ return (*(volatile const long *)&(v)->counter); ++} ++ ++/** ++ * atomic64_read_unchecked - read atomic64 variable ++ * @v: pointer of type atomic64_unchecked_t ++ * ++ * Atomically reads the value of @v. ++ * Doesn't imply a read memory barrier. ++ */ ++static inline long atomic64_read_unchecked(const atomic64_unchecked_t *v) ++{ ++ return (*(volatile const long *)&(v)->counter); + } + + /** +@@ -34,6 +46,18 @@ static inline void atomic64_set(atomic64 + } + + /** ++ * atomic64_set_unchecked - set atomic64 variable ++ * @v: pointer to type atomic64_unchecked_t ++ * @i: required value ++ * ++ * Atomically sets the value of @v to @i. ++ */ ++static inline void atomic64_set_unchecked(atomic64_unchecked_t *v, long i) ++{ ++ v->counter = i; ++} ++ ++/** + * atomic64_add - add integer to atomic64 variable + * @i: integer value to add + * @v: pointer to type atomic64_t +@@ -42,6 +66,28 @@ static inline void atomic64_set(atomic64 + */ + static inline void atomic64_add(long i, atomic64_t *v) + { ++ asm volatile(LOCK_PREFIX "addq %1,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "subq %1,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ : "=m" (v->counter) ++ : "er" (i), "m" (v->counter)); ++} ++ ++/** ++ * atomic64_add_unchecked - add integer to atomic64 variable ++ * @i: integer value to add ++ * @v: pointer to type atomic64_unchecked_t ++ * ++ * Atomically adds @i to @v. ++ */ ++static inline void atomic64_add_unchecked(long i, atomic64_unchecked_t *v) ++{ + asm volatile(LOCK_PREFIX "addq %1,%0" + : "=m" (v->counter) + : "er" (i), "m" (v->counter)); +@@ -56,7 +102,29 @@ static inline void atomic64_add(long i, + */ + static inline void atomic64_sub(long i, atomic64_t *v) + { +- asm volatile(LOCK_PREFIX "subq %1,%0" ++ asm volatile(LOCK_PREFIX "subq %1,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "addq %1,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ : "=m" (v->counter) ++ : "er" (i), "m" (v->counter)); ++} ++ ++/** ++ * atomic64_sub_unchecked - subtract the atomic64 variable ++ * @i: integer value to subtract ++ * @v: pointer to type atomic64_unchecked_t ++ * ++ * Atomically subtracts @i from @v. ++ */ ++static inline void atomic64_sub_unchecked(long i, atomic64_unchecked_t *v) ++{ ++ asm volatile(LOCK_PREFIX "subq %1,%0\n" + : "=m" (v->counter) + : "er" (i), "m" (v->counter)); + } +@@ -74,7 +142,16 @@ static inline int atomic64_sub_and_test( + { + unsigned char c; + +- asm volatile(LOCK_PREFIX "subq %2,%0; sete %1" ++ asm volatile(LOCK_PREFIX "subq %2,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "addq %2,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ "sete %1\n" + : "=m" (v->counter), "=qm" (c) + : "er" (i), "m" (v->counter) : "memory"); + return c; +@@ -88,6 +165,27 @@ static inline int atomic64_sub_and_test( + */ + static inline void atomic64_inc(atomic64_t *v) + { ++ asm volatile(LOCK_PREFIX "incq %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "decq %0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ : "=m" (v->counter) ++ : "m" (v->counter)); ++} ++ ++/** ++ * atomic64_inc_unchecked - increment atomic64 variable ++ * @v: pointer to type atomic64_unchecked_t ++ * ++ * Atomically increments @v by 1. ++ */ ++static inline void atomic64_inc_unchecked(atomic64_unchecked_t *v) ++{ + asm volatile(LOCK_PREFIX "incq %0" + : "=m" (v->counter) + : "m" (v->counter)); +@@ -101,7 +199,28 @@ static inline void atomic64_inc(atomic64 + */ + static inline void atomic64_dec(atomic64_t *v) + { +- asm volatile(LOCK_PREFIX "decq %0" ++ asm volatile(LOCK_PREFIX "decq %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "incq %0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ : "=m" (v->counter) ++ : "m" (v->counter)); ++} ++ ++/** ++ * atomic64_dec_unchecked - decrement atomic64 variable ++ * @v: pointer to type atomic64_t ++ * ++ * Atomically decrements @v by 1. ++ */ ++static inline void atomic64_dec_unchecked(atomic64_unchecked_t *v) ++{ ++ asm volatile(LOCK_PREFIX "decq %0\n" + : "=m" (v->counter) + : "m" (v->counter)); + } +@@ -118,7 +237,16 @@ static inline int atomic64_dec_and_test( + { + unsigned char c; + +- asm volatile(LOCK_PREFIX "decq %0; sete %1" ++ asm volatile(LOCK_PREFIX "decq %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "incq %0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ "sete %1\n" + : "=m" (v->counter), "=qm" (c) + : "m" (v->counter) : "memory"); + return c != 0; +@@ -136,7 +264,16 @@ static inline int atomic64_inc_and_test( + { + unsigned char c; + +- asm volatile(LOCK_PREFIX "incq %0; sete %1" ++ asm volatile(LOCK_PREFIX "incq %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "decq %0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ "sete %1\n" + : "=m" (v->counter), "=qm" (c) + : "m" (v->counter) : "memory"); + return c != 0; +@@ -155,7 +292,16 @@ static inline int atomic64_add_negative( + { + unsigned char c; + +- asm volatile(LOCK_PREFIX "addq %2,%0; sets %1" ++ asm volatile(LOCK_PREFIX "addq %2,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "subq %2,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ "sets %1\n" + : "=m" (v->counter), "=qm" (c) + : "er" (i), "m" (v->counter) : "memory"); + return c; +@@ -171,7 +317,31 @@ static inline int atomic64_add_negative( + static inline long atomic64_add_return(long i, atomic64_t *v) + { + long __i = i; +- asm volatile(LOCK_PREFIX "xaddq %0, %1;" ++ asm volatile(LOCK_PREFIX "xaddq %0, %1\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "movq %0, %1\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ : "+r" (i), "+m" (v->counter) ++ : : "memory"); ++ return i + __i; ++} ++ ++/** ++ * atomic64_add_return_unchecked - add and return ++ * @i: integer value to add ++ * @v: pointer to type atomic64_unchecked_t ++ * ++ * Atomically adds @i to @v and returns @i + @v ++ */ ++static inline long atomic64_add_return_unchecked(long i, atomic64_unchecked_t *v) ++{ ++ long __i = i; ++ asm volatile(LOCK_PREFIX "xaddq %0, %1" + : "+r" (i), "+m" (v->counter) + : : "memory"); + return i + __i; +@@ -183,6 +353,10 @@ static inline long atomic64_sub_return(l + } + + #define atomic64_inc_return(v) (atomic64_add_return(1, (v))) ++static inline long atomic64_inc_return_unchecked(atomic64_unchecked_t *v) ++{ ++ return atomic64_add_return_unchecked(1, v); ++} + #define atomic64_dec_return(v) (atomic64_sub_return(1, (v))) + + static inline long atomic64_cmpxchg(atomic64_t *v, long old, long new) +@@ -206,17 +380,30 @@ static inline long atomic64_xchg(atomic6 + */ + static inline int atomic64_add_unless(atomic64_t *v, long a, long u) + { +- long c, old; ++ long c, old, new; + c = atomic64_read(v); + for (;;) { +- if (unlikely(c == (u))) ++ if (unlikely(c == u)) + break; +- old = atomic64_cmpxchg((v), c, c + (a)); ++ ++ asm volatile("add %2,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "sub %2,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ : "=r" (new) ++ : "0" (c), "ir" (a)); ++ ++ old = atomic64_cmpxchg(v, c, new); + if (likely(old == c)) + break; + c = old; + } +- return c != (u); ++ return c != u; + } + + #define atomic64_inc_not_zero(v) atomic64_add_unless((v), 1, 0) +diff -urNp linux-2.6.38.3/arch/x86/include/asm/atomic.h linux-2.6.38.3/arch/x86/include/asm/atomic.h +--- linux-2.6.38.3/arch/x86/include/asm/atomic.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/atomic.h 2011-04-17 15:57:32.000000000 -0400 +@@ -22,7 +22,18 @@ + */ + static inline int atomic_read(const atomic_t *v) + { +- return (*(volatile int *)&(v)->counter); ++ return (*(volatile const int *)&(v)->counter); ++} ++ ++/** ++ * atomic_read_unchecked - read atomic variable ++ * @v: pointer of type atomic_unchecked_t ++ * ++ * Atomically reads the value of @v. ++ */ ++static inline int atomic_read_unchecked(const atomic_unchecked_t *v) ++{ ++ return (*(volatile const int *)&(v)->counter); + } + + /** +@@ -38,6 +49,18 @@ static inline void atomic_set(atomic_t * + } + + /** ++ * atomic_set_unchecked - set atomic variable ++ * @v: pointer of type atomic_unchecked_t ++ * @i: required value ++ * ++ * Atomically sets the value of @v to @i. ++ */ ++static inline void atomic_set_unchecked(atomic_unchecked_t *v, int i) ++{ ++ v->counter = i; ++} ++ ++/** + * atomic_add - add integer to atomic variable + * @i: integer value to add + * @v: pointer of type atomic_t +@@ -46,7 +69,29 @@ static inline void atomic_set(atomic_t * + */ + static inline void atomic_add(int i, atomic_t *v) + { +- asm volatile(LOCK_PREFIX "addl %1,%0" ++ asm volatile(LOCK_PREFIX "addl %1,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "subl %1,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ : "+m" (v->counter) ++ : "ir" (i)); ++} ++ ++/** ++ * atomic_add_unchecked - add integer to atomic variable ++ * @i: integer value to add ++ * @v: pointer of type atomic_unchecked_t ++ * ++ * Atomically adds @i to @v. ++ */ ++static inline void atomic_add_unchecked(int i, atomic_unchecked_t *v) ++{ ++ asm volatile(LOCK_PREFIX "addl %1,%0\n" + : "+m" (v->counter) + : "ir" (i)); + } +@@ -60,7 +105,29 @@ static inline void atomic_add(int i, ato + */ + static inline void atomic_sub(int i, atomic_t *v) + { +- asm volatile(LOCK_PREFIX "subl %1,%0" ++ asm volatile(LOCK_PREFIX "subl %1,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "addl %1,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ : "+m" (v->counter) ++ : "ir" (i)); ++} ++ ++/** ++ * atomic_sub_unchecked - subtract integer from atomic variable ++ * @i: integer value to subtract ++ * @v: pointer of type atomic_t ++ * ++ * Atomically subtracts @i from @v. ++ */ ++static inline void atomic_sub_unchecked(int i, atomic_unchecked_t *v) ++{ ++ asm volatile(LOCK_PREFIX "subl %1,%0\n" + : "+m" (v->counter) + : "ir" (i)); + } +@@ -78,7 +145,16 @@ static inline int atomic_sub_and_test(in + { + unsigned char c; + +- asm volatile(LOCK_PREFIX "subl %2,%0; sete %1" ++ asm volatile(LOCK_PREFIX "subl %2,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "addl %2,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ "sete %1\n" + : "+m" (v->counter), "=qm" (c) + : "ir" (i) : "memory"); + return c; +@@ -92,7 +168,27 @@ static inline int atomic_sub_and_test(in + */ + static inline void atomic_inc(atomic_t *v) + { +- asm volatile(LOCK_PREFIX "incl %0" ++ asm volatile(LOCK_PREFIX "incl %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "decl %0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ : "+m" (v->counter)); ++} ++ ++/** ++ * atomic_inc_unchecked - increment atomic variable ++ * @v: pointer of type atomic_unchecked_t ++ * ++ * Atomically increments @v by 1. ++ */ ++static inline void atomic_inc_unchecked(atomic_unchecked_t *v) ++{ ++ asm volatile(LOCK_PREFIX "incl %0\n" + : "+m" (v->counter)); + } + +@@ -104,7 +200,27 @@ static inline void atomic_inc(atomic_t * + */ + static inline void atomic_dec(atomic_t *v) + { +- asm volatile(LOCK_PREFIX "decl %0" ++ asm volatile(LOCK_PREFIX "decl %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "incl %0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ : "+m" (v->counter)); ++} ++ ++/** ++ * atomic_dec_unchecked - decrement atomic variable ++ * @v: pointer of type atomic_t ++ * ++ * Atomically decrements @v by 1. ++ */ ++static inline void atomic_dec_unchecked(atomic_unchecked_t *v) ++{ ++ asm volatile(LOCK_PREFIX "decl %0\n" + : "+m" (v->counter)); + } + +@@ -120,7 +236,16 @@ static inline int atomic_dec_and_test(at + { + unsigned char c; + +- asm volatile(LOCK_PREFIX "decl %0; sete %1" ++ asm volatile(LOCK_PREFIX "decl %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "incl %0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ "sete %1\n" + : "+m" (v->counter), "=qm" (c) + : : "memory"); + return c != 0; +@@ -138,7 +263,16 @@ static inline int atomic_inc_and_test(at + { + unsigned char c; + +- asm volatile(LOCK_PREFIX "incl %0; sete %1" ++ asm volatile(LOCK_PREFIX "incl %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "decl %0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ "sete %1\n" + : "+m" (v->counter), "=qm" (c) + : : "memory"); + return c != 0; +@@ -157,7 +291,16 @@ static inline int atomic_add_negative(in + { + unsigned char c; + +- asm volatile(LOCK_PREFIX "addl %2,%0; sets %1" ++ asm volatile(LOCK_PREFIX "addl %2,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "subl %2,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ "sets %1\n" + : "+m" (v->counter), "=qm" (c) + : "ir" (i) : "memory"); + return c; +@@ -180,6 +323,46 @@ static inline int atomic_add_return(int + #endif + /* Modern 486+ processor */ + __i = i; ++ asm volatile(LOCK_PREFIX "xaddl %0, %1\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "movl %0, %1\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ : "+r" (i), "+m" (v->counter) ++ : : "memory"); ++ return i + __i; ++ ++#ifdef CONFIG_M386 ++no_xadd: /* Legacy 386 processor */ ++ local_irq_save(flags); ++ __i = atomic_read(v); ++ atomic_set(v, i + __i); ++ local_irq_restore(flags); ++ return i + __i; ++#endif ++} ++ ++/** ++ * atomic_add_return_unchecked - add integer and return ++ * @v: pointer of type atomic_unchecked_t ++ * @i: integer value to add ++ * ++ * Atomically adds @i to @v and returns @i + @v ++ */ ++static inline int atomic_add_return_unchecked(int i, atomic_unchecked_t *v) ++{ ++ int __i; ++#ifdef CONFIG_M386 ++ unsigned long flags; ++ if (unlikely(boot_cpu_data.x86 <= 3)) ++ goto no_xadd; ++#endif ++ /* Modern 486+ processor */ ++ __i = i; + asm volatile(LOCK_PREFIX "xaddl %0, %1" + : "+r" (i), "+m" (v->counter) + : : "memory"); +@@ -208,6 +391,10 @@ static inline int atomic_sub_return(int + } + + #define atomic_inc_return(v) (atomic_add_return(1, v)) ++static inline int atomic_inc_return_unchecked(atomic_unchecked_t *v) ++{ ++ return atomic_add_return_unchecked(1, v); ++} + #define atomic_dec_return(v) (atomic_sub_return(1, v)) + + static inline int atomic_cmpxchg(atomic_t *v, int old, int new) +@@ -231,21 +418,77 @@ static inline int atomic_xchg(atomic_t * + */ + static inline int atomic_add_unless(atomic_t *v, int a, int u) + { +- int c, old; ++ int c, old, new; + c = atomic_read(v); + for (;;) { +- if (unlikely(c == (u))) ++ if (unlikely(c == u)) + break; +- old = atomic_cmpxchg((v), c, c + (a)); ++ ++ asm volatile("addl %2,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "subl %2,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ : "=r" (new) ++ : "0" (c), "ir" (a)); ++ ++ old = atomic_cmpxchg(v, c, new); + if (likely(old == c)) + break; + c = old; + } +- return c != (u); ++ return c != u; + } + + #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0) + ++/** ++ * atomic_inc_not_zero_hint - increment if not null ++ * @v: pointer of type atomic_t ++ * @hint: probable value of the atomic before the increment ++ * ++ * This version of atomic_inc_not_zero() gives a hint of probable ++ * value of the atomic. This helps processor to not read the memory ++ * before doing the atomic read/modify/write cycle, lowering ++ * number of bus transactions on some arches. ++ * ++ * Returns: 0 if increment was not done, 1 otherwise. ++ */ ++#define atomic_inc_not_zero_hint atomic_inc_not_zero_hint ++static inline int atomic_inc_not_zero_hint(atomic_t *v, int hint) ++{ ++ int val, c = hint, new; ++ ++ /* sanity test, should be removed by compiler if hint is a constant */ ++ if (!hint) ++ return atomic_inc_not_zero(v); ++ ++ do { ++ asm volatile("incl %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "decl %0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ : "=r" (new) ++ : "0" (c)); ++ ++ val = atomic_cmpxchg(v, c, new); ++ if (val == c) ++ return 1; ++ c = val; ++ } while (c); ++ ++ return 0; ++} ++ + /* + * atomic_dec_if_positive - decrement by 1 if old value positive + * @v: pointer of type atomic_t +diff -urNp linux-2.6.38.3/arch/x86/include/asm/bitops.h linux-2.6.38.3/arch/x86/include/asm/bitops.h +--- linux-2.6.38.3/arch/x86/include/asm/bitops.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/bitops.h 2011-04-17 15:57:32.000000000 -0400 +@@ -38,7 +38,7 @@ + * a mask operation on a byte. + */ + #define IS_IMMEDIATE(nr) (__builtin_constant_p(nr)) +-#define CONST_MASK_ADDR(nr, addr) BITOP_ADDR((void *)(addr) + ((nr)>>3)) ++#define CONST_MASK_ADDR(nr, addr) BITOP_ADDR((volatile void *)(addr) + ((nr)>>3)) + #define CONST_MASK(nr) (1 << ((nr) & 7)) + + /** +diff -urNp linux-2.6.38.3/arch/x86/include/asm/boot.h linux-2.6.38.3/arch/x86/include/asm/boot.h +--- linux-2.6.38.3/arch/x86/include/asm/boot.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/boot.h 2011-04-17 15:57:32.000000000 -0400 +@@ -11,10 +11,15 @@ + #include <asm/pgtable_types.h> + + /* Physical address where kernel should be loaded. */ +-#define LOAD_PHYSICAL_ADDR ((CONFIG_PHYSICAL_START \ ++#define ____LOAD_PHYSICAL_ADDR ((CONFIG_PHYSICAL_START \ + + (CONFIG_PHYSICAL_ALIGN - 1)) \ + & ~(CONFIG_PHYSICAL_ALIGN - 1)) + ++#ifndef __ASSEMBLY__ ++extern unsigned char __LOAD_PHYSICAL_ADDR[]; ++#define LOAD_PHYSICAL_ADDR ((unsigned long)__LOAD_PHYSICAL_ADDR) ++#endif ++ + /* Minimum kernel alignment, as a power of two */ + #ifdef CONFIG_X86_64 + #define MIN_KERNEL_ALIGN_LG2 PMD_SHIFT +diff -urNp linux-2.6.38.3/arch/x86/include/asm/cacheflush.h linux-2.6.38.3/arch/x86/include/asm/cacheflush.h +--- linux-2.6.38.3/arch/x86/include/asm/cacheflush.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/cacheflush.h 2011-04-17 15:57:32.000000000 -0400 +@@ -26,7 +26,7 @@ static inline unsigned long get_page_mem + unsigned long pg_flags = pg->flags & _PGMT_MASK; + + if (pg_flags == _PGMT_DEFAULT) +- return -1; ++ return ~0UL; + else if (pg_flags == _PGMT_WC) + return _PAGE_CACHE_WC; + else if (pg_flags == _PGMT_UC_MINUS) +diff -urNp linux-2.6.38.3/arch/x86/include/asm/cache.h linux-2.6.38.3/arch/x86/include/asm/cache.h +--- linux-2.6.38.3/arch/x86/include/asm/cache.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/cache.h 2011-04-17 15:57:32.000000000 -0400 +@@ -8,6 +8,7 @@ + #define L1_CACHE_BYTES (1 << L1_CACHE_SHIFT) + + #define __read_mostly __attribute__((__section__(".data..read_mostly"))) ++#define __read_only __attribute__((__section__(".data..read_only"))) + + #define INTERNODE_CACHE_SHIFT CONFIG_X86_INTERNODE_CACHE_SHIFT + #define INTERNODE_CACHE_BYTES (1 << INTERNODE_CACHE_SHIFT) +diff -urNp linux-2.6.38.3/arch/x86/include/asm/checksum_32.h linux-2.6.38.3/arch/x86/include/asm/checksum_32.h +--- linux-2.6.38.3/arch/x86/include/asm/checksum_32.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/checksum_32.h 2011-04-17 15:57:32.000000000 -0400 +@@ -31,6 +31,14 @@ asmlinkage __wsum csum_partial_copy_gene + int len, __wsum sum, + int *src_err_ptr, int *dst_err_ptr); + ++asmlinkage __wsum csum_partial_copy_generic_to_user(const void *src, void *dst, ++ int len, __wsum sum, ++ int *src_err_ptr, int *dst_err_ptr); ++ ++asmlinkage __wsum csum_partial_copy_generic_from_user(const void *src, void *dst, ++ int len, __wsum sum, ++ int *src_err_ptr, int *dst_err_ptr); ++ + /* + * Note: when you get a NULL pointer exception here this means someone + * passed in an incorrect kernel address to one of these functions. +@@ -50,7 +58,7 @@ static inline __wsum csum_partial_copy_f + int *err_ptr) + { + might_sleep(); +- return csum_partial_copy_generic((__force void *)src, dst, ++ return csum_partial_copy_generic_from_user((__force void *)src, dst, + len, sum, err_ptr, NULL); + } + +@@ -178,7 +186,7 @@ static inline __wsum csum_and_copy_to_us + { + might_sleep(); + if (access_ok(VERIFY_WRITE, dst, len)) +- return csum_partial_copy_generic(src, (__force void *)dst, ++ return csum_partial_copy_generic_to_user(src, (__force void *)dst, + len, sum, NULL, err_ptr); + + if (len) +diff -urNp linux-2.6.38.3/arch/x86/include/asm/cpufeature.h linux-2.6.38.3/arch/x86/include/asm/cpufeature.h +--- linux-2.6.38.3/arch/x86/include/asm/cpufeature.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/cpufeature.h 2011-04-17 15:57:32.000000000 -0400 +@@ -349,7 +349,7 @@ static __always_inline __pure bool __sta + ".section .discard,\"aw\",@progbits\n" + " .byte 0xff + (4f-3f) - (2b-1b)\n" /* size check */ + ".previous\n" +- ".section .altinstr_replacement,\"ax\"\n" ++ ".section .altinstr_replacement,\"a\"\n" + "3: movb $1,%0\n" + "4:\n" + ".previous\n" +diff -urNp linux-2.6.38.3/arch/x86/include/asm/desc_defs.h linux-2.6.38.3/arch/x86/include/asm/desc_defs.h +--- linux-2.6.38.3/arch/x86/include/asm/desc_defs.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/desc_defs.h 2011-04-17 15:57:32.000000000 -0400 +@@ -31,6 +31,12 @@ struct desc_struct { + unsigned base1: 8, type: 4, s: 1, dpl: 2, p: 1; + unsigned limit: 4, avl: 1, l: 1, d: 1, g: 1, base2: 8; + }; ++ struct { ++ u16 offset_low; ++ u16 seg; ++ unsigned reserved: 8, type: 4, s: 1, dpl: 2, p: 1; ++ unsigned offset_high: 16; ++ } gate; + }; + } __attribute__((packed)); + +diff -urNp linux-2.6.38.3/arch/x86/include/asm/desc.h linux-2.6.38.3/arch/x86/include/asm/desc.h +--- linux-2.6.38.3/arch/x86/include/asm/desc.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/desc.h 2011-04-17 15:57:32.000000000 -0400 +@@ -4,6 +4,7 @@ + #include <asm/desc_defs.h> + #include <asm/ldt.h> + #include <asm/mmu.h> ++#include <asm/pgtable.h> + #include <linux/smp.h> + + static inline void fill_ldt(struct desc_struct *desc, +@@ -15,6 +16,7 @@ static inline void fill_ldt(struct desc_ + desc->base1 = (info->base_addr & 0x00ff0000) >> 16; + desc->type = (info->read_exec_only ^ 1) << 1; + desc->type |= info->contents << 2; ++ desc->type |= info->seg_not_present ^ 1; + desc->s = 1; + desc->dpl = 0x3; + desc->p = info->seg_not_present ^ 1; +@@ -31,16 +33,12 @@ static inline void fill_ldt(struct desc_ + } + + extern struct desc_ptr idt_descr; +-extern gate_desc idt_table[]; +- +-struct gdt_page { +- struct desc_struct gdt[GDT_ENTRIES]; +-} __attribute__((aligned(PAGE_SIZE))); +-DECLARE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page); ++extern gate_desc idt_table[256]; + ++extern struct desc_struct cpu_gdt_table[NR_CPUS][PAGE_SIZE / sizeof(struct desc_struct)]; + static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu) + { +- return per_cpu(gdt_page, cpu).gdt; ++ return cpu_gdt_table[cpu]; + } + + #ifdef CONFIG_X86_64 +@@ -65,9 +63,14 @@ static inline void pack_gate(gate_desc * + unsigned long base, unsigned dpl, unsigned flags, + unsigned short seg) + { +- gate->a = (seg << 16) | (base & 0xffff); +- gate->b = (base & 0xffff0000) | +- (((0x80 | type | (dpl << 5)) & 0xff) << 8); ++ gate->gate.offset_low = base; ++ gate->gate.seg = seg; ++ gate->gate.reserved = 0; ++ gate->gate.type = type; ++ gate->gate.s = 0; ++ gate->gate.dpl = dpl; ++ gate->gate.p = 1; ++ gate->gate.offset_high = base >> 16; + } + + #endif +@@ -115,19 +118,24 @@ static inline void paravirt_free_ldt(str + static inline void native_write_idt_entry(gate_desc *idt, int entry, + const gate_desc *gate) + { ++ pax_open_kernel(); + memcpy(&idt[entry], gate, sizeof(*gate)); ++ pax_close_kernel(); + } + + static inline void native_write_ldt_entry(struct desc_struct *ldt, int entry, + const void *desc) + { ++ pax_open_kernel(); + memcpy(&ldt[entry], desc, 8); ++ pax_close_kernel(); + } + + static inline void native_write_gdt_entry(struct desc_struct *gdt, int entry, + const void *desc, int type) + { + unsigned int size; ++ + switch (type) { + case DESC_TSS: + size = sizeof(tss_desc); +@@ -139,7 +147,10 @@ static inline void native_write_gdt_entr + size = sizeof(struct desc_struct); + break; + } ++ ++ pax_open_kernel(); + memcpy(&gdt[entry], desc, size); ++ pax_close_kernel(); + } + + static inline void pack_descriptor(struct desc_struct *desc, unsigned long base, +@@ -211,7 +222,9 @@ static inline void native_set_ldt(const + + static inline void native_load_tr_desc(void) + { ++ pax_open_kernel(); + asm volatile("ltr %w0"::"q" (GDT_ENTRY_TSS*8)); ++ pax_close_kernel(); + } + + static inline void native_load_gdt(const struct desc_ptr *dtr) +@@ -246,8 +259,10 @@ static inline void native_load_tls(struc + unsigned int i; + struct desc_struct *gdt = get_cpu_gdt_table(cpu); + ++ pax_open_kernel(); + for (i = 0; i < GDT_ENTRY_TLS_ENTRIES; i++) + gdt[GDT_ENTRY_TLS_MIN + i] = t->tls_array[i]; ++ pax_close_kernel(); + } + + #define _LDT_empty(info) \ +@@ -309,7 +324,7 @@ static inline void set_desc_limit(struct + desc->limit = (limit >> 16) & 0xf; + } + +-static inline void _set_gate(int gate, unsigned type, void *addr, ++static inline void _set_gate(int gate, unsigned type, const void *addr, + unsigned dpl, unsigned ist, unsigned seg) + { + gate_desc s; +@@ -327,7 +342,7 @@ static inline void _set_gate(int gate, u + * Pentium F0 0F bugfix can have resulted in the mapped + * IDT being write-protected. + */ +-static inline void set_intr_gate(unsigned int n, void *addr) ++static inline void set_intr_gate(unsigned int n, const void *addr) + { + BUG_ON((unsigned)n > 0xFF); + _set_gate(n, GATE_INTERRUPT, addr, 0, 0, __KERNEL_CS); +@@ -356,19 +371,19 @@ static inline void alloc_intr_gate(unsig + /* + * This routine sets up an interrupt gate at directory privilege level 3. + */ +-static inline void set_system_intr_gate(unsigned int n, void *addr) ++static inline void set_system_intr_gate(unsigned int n, const void *addr) + { + BUG_ON((unsigned)n > 0xFF); + _set_gate(n, GATE_INTERRUPT, addr, 0x3, 0, __KERNEL_CS); + } + +-static inline void set_system_trap_gate(unsigned int n, void *addr) ++static inline void set_system_trap_gate(unsigned int n, const void *addr) + { + BUG_ON((unsigned)n > 0xFF); + _set_gate(n, GATE_TRAP, addr, 0x3, 0, __KERNEL_CS); + } + +-static inline void set_trap_gate(unsigned int n, void *addr) ++static inline void set_trap_gate(unsigned int n, const void *addr) + { + BUG_ON((unsigned)n > 0xFF); + _set_gate(n, GATE_TRAP, addr, 0, 0, __KERNEL_CS); +@@ -377,19 +392,31 @@ static inline void set_trap_gate(unsigne + static inline void set_task_gate(unsigned int n, unsigned int gdt_entry) + { + BUG_ON((unsigned)n > 0xFF); +- _set_gate(n, GATE_TASK, (void *)0, 0, 0, (gdt_entry<<3)); ++ _set_gate(n, GATE_TASK, (const void *)0, 0, 0, (gdt_entry<<3)); + } + +-static inline void set_intr_gate_ist(int n, void *addr, unsigned ist) ++static inline void set_intr_gate_ist(int n, const void *addr, unsigned ist) + { + BUG_ON((unsigned)n > 0xFF); + _set_gate(n, GATE_INTERRUPT, addr, 0, ist, __KERNEL_CS); + } + +-static inline void set_system_intr_gate_ist(int n, void *addr, unsigned ist) ++static inline void set_system_intr_gate_ist(int n, const void *addr, unsigned ist) + { + BUG_ON((unsigned)n > 0xFF); + _set_gate(n, GATE_INTERRUPT, addr, 0x3, ist, __KERNEL_CS); + } + ++#ifdef CONFIG_X86_32 ++static inline void set_user_cs(unsigned long base, unsigned long limit, int cpu) ++{ ++ struct desc_struct d; ++ ++ if (likely(limit)) ++ limit = (limit - 1UL) >> PAGE_SHIFT; ++ pack_descriptor(&d, base, limit, 0xFB, 0xC); ++ write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_DEFAULT_USER_CS, &d, DESCTYPE_S); ++} ++#endif ++ + #endif /* _ASM_X86_DESC_H */ +diff -urNp linux-2.6.38.3/arch/x86/include/asm/device.h linux-2.6.38.3/arch/x86/include/asm/device.h +--- linux-2.6.38.3/arch/x86/include/asm/device.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/device.h 2011-04-17 15:57:32.000000000 -0400 +@@ -6,7 +6,7 @@ struct dev_archdata { + void *acpi_handle; + #endif + #ifdef CONFIG_X86_64 +-struct dma_map_ops *dma_ops; ++ const struct dma_map_ops *dma_ops; + #endif + #if defined(CONFIG_DMAR) || defined(CONFIG_AMD_IOMMU) + void *iommu; /* hook for IOMMU specific extension */ +diff -urNp linux-2.6.38.3/arch/x86/include/asm/dma-mapping.h linux-2.6.38.3/arch/x86/include/asm/dma-mapping.h +--- linux-2.6.38.3/arch/x86/include/asm/dma-mapping.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/dma-mapping.h 2011-04-17 15:57:32.000000000 -0400 +@@ -26,9 +26,9 @@ extern int iommu_merge; + extern struct device x86_dma_fallback_dev; + extern int panic_on_overflow; + +-extern struct dma_map_ops *dma_ops; ++extern const struct dma_map_ops *dma_ops; + +-static inline struct dma_map_ops *get_dma_ops(struct device *dev) ++static inline const struct dma_map_ops *get_dma_ops(struct device *dev) + { + #ifdef CONFIG_X86_32 + return dma_ops; +@@ -45,7 +45,7 @@ static inline struct dma_map_ops *get_dm + /* Make sure we keep the same behaviour */ + static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + if (ops->mapping_error) + return ops->mapping_error(dev, dma_addr); + +@@ -115,7 +115,7 @@ static inline void * + dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle, + gfp_t gfp) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + void *memory; + + gfp &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32); +@@ -142,7 +142,7 @@ dma_alloc_coherent(struct device *dev, s + static inline void dma_free_coherent(struct device *dev, size_t size, + void *vaddr, dma_addr_t bus) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + + WARN_ON(irqs_disabled()); /* for portability */ + +diff -urNp linux-2.6.38.3/arch/x86/include/asm/e820.h linux-2.6.38.3/arch/x86/include/asm/e820.h +--- linux-2.6.38.3/arch/x86/include/asm/e820.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/e820.h 2011-04-17 15:57:32.000000000 -0400 +@@ -69,7 +69,7 @@ struct e820map { + #define ISA_START_ADDRESS 0xa0000 + #define ISA_END_ADDRESS 0x100000 + +-#define BIOS_BEGIN 0x000a0000 ++#define BIOS_BEGIN 0x000c0000 + #define BIOS_END 0x00100000 + + #define BIOS_ROM_BASE 0xffe00000 +diff -urNp linux-2.6.38.3/arch/x86/include/asm/elf.h linux-2.6.38.3/arch/x86/include/asm/elf.h +--- linux-2.6.38.3/arch/x86/include/asm/elf.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/elf.h 2011-04-17 15:57:32.000000000 -0400 +@@ -237,7 +237,25 @@ extern int force_personality32; + the loader. We need to make sure that it is out of the way of the program + that it will "exec", and that there is sufficient room for the brk. */ + ++#ifdef CONFIG_PAX_SEGMEXEC ++#define ELF_ET_DYN_BASE ((current->mm->pax_flags & MF_PAX_SEGMEXEC) ? SEGMEXEC_TASK_SIZE/3*2 : TASK_SIZE/3*2) ++#else + #define ELF_ET_DYN_BASE (TASK_SIZE / 3 * 2) ++#endif ++ ++#ifdef CONFIG_PAX_ASLR ++#ifdef CONFIG_X86_32 ++#define PAX_ELF_ET_DYN_BASE 0x10000000UL ++ ++#define PAX_DELTA_MMAP_LEN (current->mm->pax_flags & MF_PAX_SEGMEXEC ? 15 : 16) ++#define PAX_DELTA_STACK_LEN (current->mm->pax_flags & MF_PAX_SEGMEXEC ? 15 : 16) ++#else ++#define PAX_ELF_ET_DYN_BASE 0x400000UL ++ ++#define PAX_DELTA_MMAP_LEN ((test_thread_flag(TIF_IA32)) ? 16 : TASK_SIZE_MAX_SHIFT - PAGE_SHIFT - 3) ++#define PAX_DELTA_STACK_LEN ((test_thread_flag(TIF_IA32)) ? 16 : TASK_SIZE_MAX_SHIFT - PAGE_SHIFT - 3) ++#endif ++#endif + + /* This yields a mask that user programs can use to figure out what + instruction set this CPU supports. This could be done in user space, +@@ -291,8 +309,7 @@ do { \ + #define ARCH_DLINFO \ + do { \ + if (vdso_enabled) \ +- NEW_AUX_ENT(AT_SYSINFO_EHDR, \ +- (unsigned long)current->mm->context.vdso); \ ++ NEW_AUX_ENT(AT_SYSINFO_EHDR, current->mm->context.vdso);\ + } while (0) + + #define AT_SYSINFO 32 +@@ -303,7 +320,7 @@ do { \ + + #endif /* !CONFIG_X86_32 */ + +-#define VDSO_CURRENT_BASE ((unsigned long)current->mm->context.vdso) ++#define VDSO_CURRENT_BASE (current->mm->context.vdso) + + #define VDSO_ENTRY \ + ((unsigned long)VDSO32_SYMBOL(VDSO_CURRENT_BASE, vsyscall)) +@@ -317,7 +334,4 @@ extern int arch_setup_additional_pages(s + extern int syscall32_setup_pages(struct linux_binprm *, int exstack); + #define compat_arch_setup_additional_pages syscall32_setup_pages + +-extern unsigned long arch_randomize_brk(struct mm_struct *mm); +-#define arch_randomize_brk arch_randomize_brk +- + #endif /* _ASM_X86_ELF_H */ +diff -urNp linux-2.6.38.3/arch/x86/include/asm/futex.h linux-2.6.38.3/arch/x86/include/asm/futex.h +--- linux-2.6.38.3/arch/x86/include/asm/futex.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/futex.h 2011-04-17 15:57:32.000000000 -0400 +@@ -12,16 +12,18 @@ + #include <asm/system.h> + + #define __futex_atomic_op1(insn, ret, oldval, uaddr, oparg) \ ++ typecheck(u32 *, uaddr); \ + asm volatile("1:\t" insn "\n" \ + "2:\t.section .fixup,\"ax\"\n" \ + "3:\tmov\t%3, %1\n" \ + "\tjmp\t2b\n" \ + "\t.previous\n" \ + _ASM_EXTABLE(1b, 3b) \ +- : "=r" (oldval), "=r" (ret), "+m" (*uaddr) \ ++ : "=r" (oldval), "=r" (ret), "+m" (*(u32 *)____m(uaddr))\ + : "i" (-EFAULT), "0" (oparg), "1" (0)) + + #define __futex_atomic_op2(insn, ret, oldval, uaddr, oparg) \ ++ typecheck(u32 *, uaddr); \ + asm volatile("1:\tmovl %2, %0\n" \ + "\tmovl\t%0, %3\n" \ + "\t" insn "\n" \ +@@ -34,10 +36,10 @@ + _ASM_EXTABLE(1b, 4b) \ + _ASM_EXTABLE(2b, 4b) \ + : "=&a" (oldval), "=&r" (ret), \ +- "+m" (*uaddr), "=&r" (tem) \ ++ "+m" (*(u32 *)____m(uaddr)), "=&r" (tem) \ + : "r" (oparg), "i" (-EFAULT), "1" (0)) + +-static inline int futex_atomic_op_inuser(int encoded_op, int __user *uaddr) ++static inline int futex_atomic_op_inuser(int encoded_op, u32 __user *uaddr) + { + int op = (encoded_op >> 28) & 7; + int cmp = (encoded_op >> 24) & 15; +@@ -61,10 +63,10 @@ static inline int futex_atomic_op_inuser + + switch (op) { + case FUTEX_OP_SET: +- __futex_atomic_op1("xchgl %0, %2", ret, oldval, uaddr, oparg); ++ __futex_atomic_op1(__copyuser_seg"xchgl %0, %2", ret, oldval, uaddr, oparg); + break; + case FUTEX_OP_ADD: +- __futex_atomic_op1(LOCK_PREFIX "xaddl %0, %2", ret, oldval, ++ __futex_atomic_op1(LOCK_PREFIX __copyuser_seg"xaddl %0, %2", ret, oldval, + uaddr, oparg); + break; + case FUTEX_OP_OR: +@@ -109,7 +111,7 @@ static inline int futex_atomic_op_inuser + return ret; + } + +-static inline int futex_atomic_cmpxchg_inatomic(int __user *uaddr, int oldval, ++static inline int futex_atomic_cmpxchg_inatomic(u32 __user *uaddr, int oldval, + int newval) + { + +@@ -119,16 +121,16 @@ static inline int futex_atomic_cmpxchg_i + return -ENOSYS; + #endif + +- if (!access_ok(VERIFY_WRITE, uaddr, sizeof(int))) ++ if (!access_ok(VERIFY_WRITE, uaddr, sizeof(u32))) + return -EFAULT; + +- asm volatile("1:\t" LOCK_PREFIX "cmpxchgl %3, %1\n" ++ asm volatile("1:\t" LOCK_PREFIX __copyuser_seg"cmpxchgl %3, %1\n" + "2:\t.section .fixup, \"ax\"\n" + "3:\tmov %2, %0\n" + "\tjmp 2b\n" + "\t.previous\n" + _ASM_EXTABLE(1b, 3b) +- : "=a" (oldval), "+m" (*uaddr) ++ : "=a" (oldval), "+m" (*(u32 *)____m(uaddr)) + : "i" (-EFAULT), "r" (newval), "0" (oldval) + : "memory" + ); +diff -urNp linux-2.6.38.3/arch/x86/include/asm/i387.h linux-2.6.38.3/arch/x86/include/asm/i387.h +--- linux-2.6.38.3/arch/x86/include/asm/i387.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/i387.h 2011-04-17 15:57:32.000000000 -0400 +@@ -92,6 +92,11 @@ static inline int fxrstor_checking(struc + { + int err; + ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) ++ if ((unsigned long)fx < PAX_USER_SHADOW_BASE) ++ fx = (struct i387_fxsave_struct *)((void *)fx + PAX_USER_SHADOW_BASE); ++#endif ++ + /* See comment in fxsave() below. */ + #ifdef CONFIG_AS_FXSAVEQ + asm volatile("1: fxrstorq %[fx]\n\t" +@@ -121,6 +126,11 @@ static inline int fxsave_user(struct i38 + { + int err; + ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) ++ if ((unsigned long)fx < PAX_USER_SHADOW_BASE) ++ fx = (struct i387_fxsave_struct __user *)((void __user *)fx + PAX_USER_SHADOW_BASE); ++#endif ++ + /* + * Clear the bytes not touched by the fxsave and reserved + * for the SW usage. +@@ -213,13 +223,8 @@ static inline void fpu_fxsave(struct fpu + #endif /* CONFIG_X86_64 */ + + /* We need a safe address that is cheap to find and that is already +- in L1 during context switch. The best choices are unfortunately +- different for UP and SMP */ +-#ifdef CONFIG_SMP +-#define safe_address (__per_cpu_offset[0]) +-#else +-#define safe_address (kstat_cpu(0).cpustat.user) +-#endif ++ in L1 during context switch. */ ++#define safe_address (init_tss[smp_processor_id()].x86_tss.sp0) + + /* + * These must be called with preempt disabled +@@ -237,7 +242,7 @@ static inline void fpu_save_init(struct + } else if (use_fxsr()) { + fpu_fxsave(fpu); + } else { +- asm volatile("fsave %[fx]; fwait" ++ asm volatile("fnsave %[fx]; fwait" + : [fx] "=m" (fpu->state->fsave)); + return; + } +@@ -312,7 +317,7 @@ static inline void kernel_fpu_begin(void + struct thread_info *me = current_thread_info(); + preempt_disable(); + if (me->status & TS_USEDFPU) +- __save_init_fpu(me->task); ++ __save_init_fpu(current); + else + clts(); + } +diff -urNp linux-2.6.38.3/arch/x86/include/asm/io.h linux-2.6.38.3/arch/x86/include/asm/io.h +--- linux-2.6.38.3/arch/x86/include/asm/io.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/io.h 2011-04-17 15:57:32.000000000 -0400 +@@ -216,6 +216,17 @@ extern void set_iounmap_nonlazy(void); + + #include <linux/vmalloc.h> + ++#define ARCH_HAS_VALID_PHYS_ADDR_RANGE ++static inline int valid_phys_addr_range(unsigned long addr, size_t count) ++{ ++ return ((addr + count + PAGE_SIZE - 1) >> PAGE_SHIFT) < (1ULL << (boot_cpu_data.x86_phys_bits - PAGE_SHIFT)) ? 1 : 0; ++} ++ ++static inline int valid_mmap_phys_addr_range(unsigned long pfn, size_t count) ++{ ++ return (pfn + (count >> PAGE_SHIFT)) < (1ULL << (boot_cpu_data.x86_phys_bits - PAGE_SHIFT)) ? 1 : 0; ++} ++ + /* + * Convert a virtual cached pointer to an uncached pointer + */ +diff -urNp linux-2.6.38.3/arch/x86/include/asm/iommu.h linux-2.6.38.3/arch/x86/include/asm/iommu.h +--- linux-2.6.38.3/arch/x86/include/asm/iommu.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/iommu.h 2011-04-17 15:57:32.000000000 -0400 +@@ -1,7 +1,7 @@ + #ifndef _ASM_X86_IOMMU_H + #define _ASM_X86_IOMMU_H + +-extern struct dma_map_ops nommu_dma_ops; ++extern const struct dma_map_ops nommu_dma_ops; + extern int force_iommu, no_iommu; + extern int iommu_detected; + extern int iommu_pass_through; +diff -urNp linux-2.6.38.3/arch/x86/include/asm/irqflags.h linux-2.6.38.3/arch/x86/include/asm/irqflags.h +--- linux-2.6.38.3/arch/x86/include/asm/irqflags.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/irqflags.h 2011-04-17 15:57:32.000000000 -0400 +@@ -140,6 +140,11 @@ static inline unsigned long arch_local_i + sti; \ + sysexit + ++#define GET_CR0_INTO_RDI mov %cr0, %rdi ++#define SET_RDI_INTO_CR0 mov %rdi, %cr0 ++#define GET_CR3_INTO_RDI mov %cr3, %rdi ++#define SET_RDI_INTO_CR3 mov %rdi, %cr3 ++ + #else + #define INTERRUPT_RETURN iret + #define ENABLE_INTERRUPTS_SYSEXIT sti; sysexit +diff -urNp linux-2.6.38.3/arch/x86/include/asm/kvm_host.h linux-2.6.38.3/arch/x86/include/asm/kvm_host.h +--- linux-2.6.38.3/arch/x86/include/asm/kvm_host.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/kvm_host.h 2011-04-17 15:57:32.000000000 -0400 +@@ -603,7 +603,7 @@ struct kvm_arch_async_pf { + bool direct_map; + }; + +-extern struct kvm_x86_ops *kvm_x86_ops; ++extern const struct kvm_x86_ops *kvm_x86_ops; + + int kvm_mmu_module_init(void); + void kvm_mmu_module_exit(void); +diff -urNp linux-2.6.38.3/arch/x86/include/asm/local.h linux-2.6.38.3/arch/x86/include/asm/local.h +--- linux-2.6.38.3/arch/x86/include/asm/local.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/local.h 2011-04-17 15:57:32.000000000 -0400 +@@ -18,26 +18,58 @@ typedef struct { + + static inline void local_inc(local_t *l) + { +- asm volatile(_ASM_INC "%0" ++ asm volatile(_ASM_INC "%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ _ASM_DEC "%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ + : "+m" (l->a.counter)); + } + + static inline void local_dec(local_t *l) + { +- asm volatile(_ASM_DEC "%0" ++ asm volatile(_ASM_DEC "%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ _ASM_INC "%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ + : "+m" (l->a.counter)); + } + + static inline void local_add(long i, local_t *l) + { +- asm volatile(_ASM_ADD "%1,%0" ++ asm volatile(_ASM_ADD "%1,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ _ASM_SUB "%1,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ + : "+m" (l->a.counter) + : "ir" (i)); + } + + static inline void local_sub(long i, local_t *l) + { +- asm volatile(_ASM_SUB "%1,%0" ++ asm volatile(_ASM_SUB "%1,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ _ASM_ADD "%1,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ + : "+m" (l->a.counter) + : "ir" (i)); + } +@@ -55,7 +87,16 @@ static inline int local_sub_and_test(lon + { + unsigned char c; + +- asm volatile(_ASM_SUB "%2,%0; sete %1" ++ asm volatile(_ASM_SUB "%2,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ _ASM_ADD "%2,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ "sete %1\n" + : "+m" (l->a.counter), "=qm" (c) + : "ir" (i) : "memory"); + return c; +@@ -73,7 +114,16 @@ static inline int local_dec_and_test(loc + { + unsigned char c; + +- asm volatile(_ASM_DEC "%0; sete %1" ++ asm volatile(_ASM_DEC "%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ _ASM_INC "%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ "sete %1\n" + : "+m" (l->a.counter), "=qm" (c) + : : "memory"); + return c != 0; +@@ -91,7 +141,16 @@ static inline int local_inc_and_test(loc + { + unsigned char c; + +- asm volatile(_ASM_INC "%0; sete %1" ++ asm volatile(_ASM_INC "%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ _ASM_DEC "%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ "sete %1\n" + : "+m" (l->a.counter), "=qm" (c) + : : "memory"); + return c != 0; +@@ -110,7 +169,16 @@ static inline int local_add_negative(lon + { + unsigned char c; + +- asm volatile(_ASM_ADD "%2,%0; sets %1" ++ asm volatile(_ASM_ADD "%2,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ _ASM_SUB "%2,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ "sets %1\n" + : "+m" (l->a.counter), "=qm" (c) + : "ir" (i) : "memory"); + return c; +@@ -133,7 +201,15 @@ static inline long local_add_return(long + #endif + /* Modern 486+ processor */ + __i = i; +- asm volatile(_ASM_XADD "%0, %1;" ++ asm volatile(_ASM_XADD "%0, %1\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ _ASM_MOV "%0,%1\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ + : "+r" (i), "+m" (l->a.counter) + : : "memory"); + return i + __i; +diff -urNp linux-2.6.38.3/arch/x86/include/asm/mc146818rtc.h linux-2.6.38.3/arch/x86/include/asm/mc146818rtc.h +--- linux-2.6.38.3/arch/x86/include/asm/mc146818rtc.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/mc146818rtc.h 2011-04-17 15:57:32.000000000 -0400 +@@ -81,8 +81,8 @@ static inline unsigned char current_lock + #else + #define lock_cmos_prefix(reg) do {} while (0) + #define lock_cmos_suffix(reg) do {} while (0) +-#define lock_cmos(reg) +-#define unlock_cmos() ++#define lock_cmos(reg) do {} while (0) ++#define unlock_cmos() do {} while (0) + #define do_i_have_lock_cmos() 0 + #define current_lock_cmos_reg() 0 + #endif +diff -urNp linux-2.6.38.3/arch/x86/include/asm/mce.h linux-2.6.38.3/arch/x86/include/asm/mce.h +--- linux-2.6.38.3/arch/x86/include/asm/mce.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/mce.h 2011-04-17 15:57:32.000000000 -0400 +@@ -198,7 +198,7 @@ int mce_notify_irq(void); + void mce_notify_process(void); + + DECLARE_PER_CPU(struct mce, injectm); +-extern struct file_operations mce_chrdev_ops; ++extern struct file_operations mce_chrdev_ops; /* cannot be const, see arch/x86/kernel/cpu/mcheck/mce. */ + + /* + * Exception handler +diff -urNp linux-2.6.38.3/arch/x86/include/asm/microcode.h linux-2.6.38.3/arch/x86/include/asm/microcode.h +--- linux-2.6.38.3/arch/x86/include/asm/microcode.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/microcode.h 2011-04-17 15:57:32.000000000 -0400 +@@ -12,13 +12,13 @@ struct device; + enum ucode_state { UCODE_ERROR, UCODE_OK, UCODE_NFOUND }; + + struct microcode_ops { +- enum ucode_state (*request_microcode_user) (int cpu, ++ enum ucode_state (* const request_microcode_user) (int cpu, + const void __user *buf, size_t size); + +- enum ucode_state (*request_microcode_fw) (int cpu, ++ enum ucode_state (* const request_microcode_fw) (int cpu, + struct device *device); + +- void (*microcode_fini_cpu) (int cpu); ++ void (* const microcode_fini_cpu) (int cpu); + + /* + * The generic 'microcode_core' part guarantees that +@@ -38,16 +38,16 @@ struct ucode_cpu_info { + extern struct ucode_cpu_info ucode_cpu_info[]; + + #ifdef CONFIG_MICROCODE_INTEL +-extern struct microcode_ops * __init init_intel_microcode(void); ++extern const struct microcode_ops * __init init_intel_microcode(void); + #else +-static inline struct microcode_ops * __init init_intel_microcode(void) ++static inline const struct microcode_ops * __init init_intel_microcode(void) + { + return NULL; + } + #endif /* CONFIG_MICROCODE_INTEL */ + + #ifdef CONFIG_MICROCODE_AMD +-extern struct microcode_ops * __init init_amd_microcode(void); ++extern const struct microcode_ops * __init init_amd_microcode(void); + + static inline void get_ucode_data(void *to, const u8 *from, size_t n) + { +@@ -55,7 +55,7 @@ static inline void get_ucode_data(void * + } + + #else +-static inline struct microcode_ops * __init init_amd_microcode(void) ++static inline const struct microcode_ops * __init init_amd_microcode(void) + { + return NULL; + } +diff -urNp linux-2.6.38.3/arch/x86/include/asm/mman.h linux-2.6.38.3/arch/x86/include/asm/mman.h +--- linux-2.6.38.3/arch/x86/include/asm/mman.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/mman.h 2011-04-17 15:57:32.000000000 -0400 +@@ -5,4 +5,14 @@ + + #include <asm-generic/mman.h> + ++#ifdef __KERNEL__ ++#ifndef __ASSEMBLY__ ++#ifdef CONFIG_X86_32 ++#define arch_mmap_check i386_mmap_check ++int i386_mmap_check(unsigned long addr, unsigned long len, ++ unsigned long flags); ++#endif ++#endif ++#endif ++ + #endif /* _ASM_X86_MMAN_H */ +diff -urNp linux-2.6.38.3/arch/x86/include/asm/mmu_context.h linux-2.6.38.3/arch/x86/include/asm/mmu_context.h +--- linux-2.6.38.3/arch/x86/include/asm/mmu_context.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/mmu_context.h 2011-04-17 15:57:32.000000000 -0400 +@@ -24,6 +24,21 @@ void destroy_context(struct mm_struct *m + + static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk) + { ++ ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) ++ unsigned int i; ++ pgd_t *pgd; ++ ++ pax_open_kernel(); ++ pgd = get_cpu_pgd(smp_processor_id()); ++ for (i = USER_PGD_PTRS; i < 2 * USER_PGD_PTRS; ++i) ++ if (paravirt_enabled()) ++ set_pgd(pgd+i, native_make_pgd(0)); ++ else ++ pgd[i] = native_make_pgd(0); ++ pax_close_kernel(); ++#endif ++ + #ifdef CONFIG_SMP + if (percpu_read(cpu_tlbstate.state) == TLBSTATE_OK) + percpu_write(cpu_tlbstate.state, TLBSTATE_LAZY); +@@ -34,16 +49,30 @@ static inline void switch_mm(struct mm_s + struct task_struct *tsk) + { + unsigned cpu = smp_processor_id(); ++#if defined(CONFIG_X86_32) && defined(CONFIG_SMP) ++ int tlbstate = TLBSTATE_OK; ++#endif + + if (likely(prev != next)) { + #ifdef CONFIG_SMP ++#ifdef CONFIG_X86_32 ++ tlbstate = percpu_read(cpu_tlbstate.state); ++#endif + percpu_write(cpu_tlbstate.state, TLBSTATE_OK); + percpu_write(cpu_tlbstate.active_mm, next); + #endif + cpumask_set_cpu(cpu, mm_cpumask(next)); + + /* Re-load page tables */ ++#ifdef CONFIG_PAX_PER_CPU_PGD ++ pax_open_kernel(); ++ __clone_user_pgds(get_cpu_pgd(cpu), next->pgd, USER_PGD_PTRS); ++ __shadow_user_pgds(get_cpu_pgd(cpu) + USER_PGD_PTRS, next->pgd, USER_PGD_PTRS); ++ pax_close_kernel(); ++ load_cr3(get_cpu_pgd(cpu)); ++#else + load_cr3(next->pgd); ++#endif + + /* stop flush ipis for the previous mm */ + cpumask_clear_cpu(cpu, mm_cpumask(prev)); +@@ -53,9 +82,38 @@ static inline void switch_mm(struct mm_s + */ + if (unlikely(prev->context.ldt != next->context.ldt)) + load_LDT_nolock(&next->context); +- } ++ ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_PAGEEXEC) && defined(CONFIG_SMP) ++ if (!(__supported_pte_mask & _PAGE_NX)) { ++ smp_mb__before_clear_bit(); ++ cpu_clear(cpu, prev->context.cpu_user_cs_mask); ++ smp_mb__after_clear_bit(); ++ cpu_set(cpu, next->context.cpu_user_cs_mask); ++ } ++#endif ++ ++#if defined(CONFIG_X86_32) && (defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC)) ++ if (unlikely(prev->context.user_cs_base != next->context.user_cs_base || ++ prev->context.user_cs_limit != next->context.user_cs_limit)) ++ set_user_cs(next->context.user_cs_base, next->context.user_cs_limit, cpu); + #ifdef CONFIG_SMP ++ else if (unlikely(tlbstate != TLBSTATE_OK)) ++ set_user_cs(next->context.user_cs_base, next->context.user_cs_limit, cpu); ++#endif ++#endif ++ ++ } + else { ++ ++#ifdef CONFIG_PAX_PER_CPU_PGD ++ pax_open_kernel(); ++ __clone_user_pgds(get_cpu_pgd(cpu), next->pgd, USER_PGD_PTRS); ++ __shadow_user_pgds(get_cpu_pgd(cpu) + USER_PGD_PTRS, next->pgd, USER_PGD_PTRS); ++ pax_close_kernel(); ++ load_cr3(get_cpu_pgd(cpu)); ++#endif ++ ++#ifdef CONFIG_SMP + percpu_write(cpu_tlbstate.state, TLBSTATE_OK); + BUG_ON(percpu_read(cpu_tlbstate.active_mm) != next); + +@@ -64,11 +122,28 @@ static inline void switch_mm(struct mm_s + * tlb flush IPI delivery. We must reload CR3 + * to make sure to use no freed page tables. + */ ++ ++#ifndef CONFIG_PAX_PER_CPU_PGD + load_cr3(next->pgd); ++#endif ++ + load_LDT_nolock(&next->context); ++ ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_PAGEEXEC) ++ if (!(__supported_pte_mask & _PAGE_NX)) ++ cpu_set(cpu, next->context.cpu_user_cs_mask); ++#endif ++ ++#if defined(CONFIG_X86_32) && (defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC)) ++#ifdef CONFIG_PAX_PAGEEXEC ++ if (!((next->pax_flags & MF_PAX_PAGEEXEC) && (__supported_pte_mask & _PAGE_NX))) ++#endif ++ set_user_cs(next->context.user_cs_base, next->context.user_cs_limit, cpu); ++#endif ++ + } +- } + #endif ++ } + } + + #define activate_mm(prev, next) \ +diff -urNp linux-2.6.38.3/arch/x86/include/asm/mmu.h linux-2.6.38.3/arch/x86/include/asm/mmu.h +--- linux-2.6.38.3/arch/x86/include/asm/mmu.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/mmu.h 2011-04-17 15:57:32.000000000 -0400 +@@ -9,10 +9,23 @@ + * we put the segment information here. + */ + typedef struct { +- void *ldt; ++ struct desc_struct *ldt; + int size; + struct mutex lock; +- void *vdso; ++ unsigned long vdso; ++ ++#ifdef CONFIG_X86_32 ++#if defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) ++ unsigned long user_cs_base; ++ unsigned long user_cs_limit; ++ ++#if defined(CONFIG_PAX_PAGEEXEC) && defined(CONFIG_SMP) ++ cpumask_t cpu_user_cs_mask; ++#endif ++ ++#endif ++#endif ++ + } mm_context_t; + + #ifdef CONFIG_SMP +diff -urNp linux-2.6.38.3/arch/x86/include/asm/module.h linux-2.6.38.3/arch/x86/include/asm/module.h +--- linux-2.6.38.3/arch/x86/include/asm/module.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/module.h 2011-04-17 15:57:32.000000000 -0400 +@@ -59,8 +59,26 @@ + #error unknown processor family + #endif + ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++#define MODULE_PAX_UDEREF "UDEREF " ++#else ++#define MODULE_PAX_UDEREF "" ++#endif ++ + #ifdef CONFIG_X86_32 +-# define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY ++# ifdef CONFIG_PAX_KERNEXEC ++# define MODULE_PAX_KERNEXEC "KERNEXEC " ++# else ++# define MODULE_PAX_KERNEXEC "" ++# endif ++# ifdef CONFIG_GRKERNSEC ++# define MODULE_GRSEC "GRSECURITY " ++# else ++# define MODULE_GRSEC "" ++# endif ++# define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY MODULE_GRSEC MODULE_PAX_KERNEXEC MODULE_PAX_UDEREF ++#else ++# define MODULE_ARCH_VERMAGIC MODULE_PAX_UDEREF + #endif + + #endif /* _ASM_X86_MODULE_H */ +diff -urNp linux-2.6.38.3/arch/x86/include/asm/page_64_types.h linux-2.6.38.3/arch/x86/include/asm/page_64_types.h +--- linux-2.6.38.3/arch/x86/include/asm/page_64_types.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/page_64_types.h 2011-04-17 15:57:32.000000000 -0400 +@@ -56,7 +56,7 @@ void copy_page(void *to, void *from); + + /* duplicated to the one in bootmem.h */ + extern unsigned long max_pfn; +-extern unsigned long phys_base; ++extern const unsigned long phys_base; + + extern unsigned long __phys_addr(unsigned long); + #define __phys_reloc_hide(x) (x) +diff -urNp linux-2.6.38.3/arch/x86/include/asm/paravirt.h linux-2.6.38.3/arch/x86/include/asm/paravirt.h +--- linux-2.6.38.3/arch/x86/include/asm/paravirt.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/paravirt.h 2011-04-17 15:57:32.000000000 -0400 +@@ -739,6 +739,21 @@ static inline void __set_fixmap(unsigned + pv_mmu_ops.set_fixmap(idx, phys, flags); + } + ++#ifdef CONFIG_PAX_KERNEXEC ++static inline unsigned long pax_open_kernel(void) ++{ ++ return PVOP_CALL0(unsigned long, pv_mmu_ops.pax_open_kernel); ++} ++ ++static inline unsigned long pax_close_kernel(void) ++{ ++ return PVOP_CALL0(unsigned long, pv_mmu_ops.pax_close_kernel); ++} ++#else ++static inline unsigned long pax_open_kernel(void) { return 0; } ++static inline unsigned long pax_close_kernel(void) { return 0; } ++#endif ++ + #if defined(CONFIG_SMP) && defined(CONFIG_PARAVIRT_SPINLOCKS) + + static inline int arch_spin_is_locked(struct arch_spinlock *lock) +@@ -955,7 +970,7 @@ extern void default_banner(void); + + #define PARA_PATCH(struct, off) ((PARAVIRT_PATCH_##struct + (off)) / 4) + #define PARA_SITE(ptype, clobbers, ops) _PVSITE(ptype, clobbers, ops, .long, 4) +-#define PARA_INDIRECT(addr) *%cs:addr ++#define PARA_INDIRECT(addr) *%ss:addr + #endif + + #define INTERRUPT_RETURN \ +@@ -1032,6 +1047,21 @@ extern void default_banner(void); + PARA_SITE(PARA_PATCH(pv_cpu_ops, PV_CPU_irq_enable_sysexit), \ + CLBR_NONE, \ + jmp PARA_INDIRECT(pv_cpu_ops+PV_CPU_irq_enable_sysexit)) ++ ++#define GET_CR0_INTO_RDI \ ++ call PARA_INDIRECT(pv_cpu_ops+PV_CPU_read_cr0); \ ++ mov %rax,%rdi ++ ++#define SET_RDI_INTO_CR0 \ ++ call PARA_INDIRECT(pv_cpu_ops+PV_CPU_write_cr0) ++ ++#define GET_CR3_INTO_RDI \ ++ call PARA_INDIRECT(pv_mmu_ops+PV_MMU_read_cr3); \ ++ mov %rax,%rdi ++ ++#define SET_RDI_INTO_CR3 \ ++ call PARA_INDIRECT(pv_mmu_ops+PV_MMU_write_cr3) ++ + #endif /* CONFIG_X86_32 */ + + #endif /* __ASSEMBLY__ */ +diff -urNp linux-2.6.38.3/arch/x86/include/asm/paravirt_types.h linux-2.6.38.3/arch/x86/include/asm/paravirt_types.h +--- linux-2.6.38.3/arch/x86/include/asm/paravirt_types.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/paravirt_types.h 2011-04-17 15:57:32.000000000 -0400 +@@ -317,6 +317,12 @@ struct pv_mmu_ops { + an mfn. We can tell which is which from the index. */ + void (*set_fixmap)(unsigned /* enum fixed_addresses */ idx, + phys_addr_t phys, pgprot_t flags); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ unsigned long (*pax_open_kernel)(void); ++ unsigned long (*pax_close_kernel)(void); ++#endif ++ + }; + + struct arch_spinlock; +diff -urNp linux-2.6.38.3/arch/x86/include/asm/pci_x86.h linux-2.6.38.3/arch/x86/include/asm/pci_x86.h +--- linux-2.6.38.3/arch/x86/include/asm/pci_x86.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/pci_x86.h 2011-04-17 15:57:32.000000000 -0400 +@@ -93,16 +93,16 @@ extern int (*pcibios_enable_irq)(struct + extern void (*pcibios_disable_irq)(struct pci_dev *dev); + + struct pci_raw_ops { +- int (*read)(unsigned int domain, unsigned int bus, unsigned int devfn, ++ int (* const read)(unsigned int domain, unsigned int bus, unsigned int devfn, + int reg, int len, u32 *val); +- int (*write)(unsigned int domain, unsigned int bus, unsigned int devfn, ++ int (* const write)(unsigned int domain, unsigned int bus, unsigned int devfn, + int reg, int len, u32 val); + }; + +-extern struct pci_raw_ops *raw_pci_ops; +-extern struct pci_raw_ops *raw_pci_ext_ops; ++extern const struct pci_raw_ops *raw_pci_ops; ++extern const struct pci_raw_ops *raw_pci_ext_ops; + +-extern struct pci_raw_ops pci_direct_conf1; ++extern const struct pci_raw_ops pci_direct_conf1; + extern bool port_cf9_safe; + + /* arch_initcall level */ +diff -urNp linux-2.6.38.3/arch/x86/include/asm/pgalloc.h linux-2.6.38.3/arch/x86/include/asm/pgalloc.h +--- linux-2.6.38.3/arch/x86/include/asm/pgalloc.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/pgalloc.h 2011-04-17 15:57:32.000000000 -0400 +@@ -63,6 +63,13 @@ static inline void pmd_populate_kernel(s + pmd_t *pmd, pte_t *pte) + { + paravirt_alloc_pte(mm, __pa(pte) >> PAGE_SHIFT); ++ set_pmd(pmd, __pmd(__pa(pte) | _KERNPG_TABLE)); ++} ++ ++static inline void pmd_populate_user(struct mm_struct *mm, ++ pmd_t *pmd, pte_t *pte) ++{ ++ paravirt_alloc_pte(mm, __pa(pte) >> PAGE_SHIFT); + set_pmd(pmd, __pmd(__pa(pte) | _PAGE_TABLE)); + } + +diff -urNp linux-2.6.38.3/arch/x86/include/asm/pgtable-2level.h linux-2.6.38.3/arch/x86/include/asm/pgtable-2level.h +--- linux-2.6.38.3/arch/x86/include/asm/pgtable-2level.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/pgtable-2level.h 2011-04-17 15:57:32.000000000 -0400 +@@ -18,7 +18,9 @@ static inline void native_set_pte(pte_t + + static inline void native_set_pmd(pmd_t *pmdp, pmd_t pmd) + { ++ pax_open_kernel(); + *pmdp = pmd; ++ pax_close_kernel(); + } + + static inline void native_set_pte_atomic(pte_t *ptep, pte_t pte) +diff -urNp linux-2.6.38.3/arch/x86/include/asm/pgtable_32.h linux-2.6.38.3/arch/x86/include/asm/pgtable_32.h +--- linux-2.6.38.3/arch/x86/include/asm/pgtable_32.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/pgtable_32.h 2011-04-17 15:57:32.000000000 -0400 +@@ -25,9 +25,6 @@ + struct mm_struct; + struct vm_area_struct; + +-extern pgd_t swapper_pg_dir[1024]; +-extern pgd_t initial_page_table[1024]; +- + static inline void pgtable_cache_init(void) { } + static inline void check_pgt_cache(void) { } + void paging_init(void); +@@ -48,6 +45,12 @@ extern void set_pmd_pfn(unsigned long, u + # include <asm/pgtable-2level.h> + #endif + ++extern pgd_t swapper_pg_dir[PTRS_PER_PGD]; ++extern pgd_t initial_page_table[PTRS_PER_PGD]; ++#ifdef CONFIG_X86_PAE ++extern pmd_t swapper_pm_dir[PTRS_PER_PGD][PTRS_PER_PMD]; ++#endif ++ + #if defined(CONFIG_HIGHPTE) + #define pte_offset_map(dir, address) \ + ((pte_t *)kmap_atomic(pmd_page(*(dir))) + \ +@@ -62,7 +65,9 @@ extern void set_pmd_pfn(unsigned long, u + /* Clear a kernel PTE and flush it from the TLB */ + #define kpte_clear_flush(ptep, vaddr) \ + do { \ ++ pax_open_kernel(); \ + pte_clear(&init_mm, (vaddr), (ptep)); \ ++ pax_close_kernel(); \ + __flush_tlb_one((vaddr)); \ + } while (0) + +@@ -74,6 +79,9 @@ do { \ + + #endif /* !__ASSEMBLY__ */ + ++#define HAVE_ARCH_UNMAPPED_AREA ++#define HAVE_ARCH_UNMAPPED_AREA_TOPDOWN ++ + /* + * kern_addr_valid() is (1) for FLATMEM and (0) for + * SPARSEMEM and DISCONTIGMEM +diff -urNp linux-2.6.38.3/arch/x86/include/asm/pgtable_32_types.h linux-2.6.38.3/arch/x86/include/asm/pgtable_32_types.h +--- linux-2.6.38.3/arch/x86/include/asm/pgtable_32_types.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/pgtable_32_types.h 2011-04-17 15:57:32.000000000 -0400 +@@ -8,7 +8,7 @@ + */ + #ifdef CONFIG_X86_PAE + # include <asm/pgtable-3level_types.h> +-# define PMD_SIZE (1UL << PMD_SHIFT) ++# define PMD_SIZE (_AC(1, UL) << PMD_SHIFT) + # define PMD_MASK (~(PMD_SIZE - 1)) + #else + # include <asm/pgtable-2level_types.h> +@@ -46,6 +46,19 @@ extern bool __vmalloc_start_set; /* set + # define VMALLOC_END (FIXADDR_START - 2 * PAGE_SIZE) + #endif + ++#ifdef CONFIG_PAX_KERNEXEC ++#ifndef __ASSEMBLY__ ++extern unsigned char MODULES_EXEC_VADDR[]; ++extern unsigned char MODULES_EXEC_END[]; ++#endif ++#include <asm/boot.h> ++#define ktla_ktva(addr) (addr + LOAD_PHYSICAL_ADDR + PAGE_OFFSET) ++#define ktva_ktla(addr) (addr - LOAD_PHYSICAL_ADDR - PAGE_OFFSET) ++#else ++#define ktla_ktva(addr) (addr) ++#define ktva_ktla(addr) (addr) ++#endif ++ + #define MODULES_VADDR VMALLOC_START + #define MODULES_END VMALLOC_END + #define MODULES_LEN (MODULES_VADDR - MODULES_END) +diff -urNp linux-2.6.38.3/arch/x86/include/asm/pgtable-3level.h linux-2.6.38.3/arch/x86/include/asm/pgtable-3level.h +--- linux-2.6.38.3/arch/x86/include/asm/pgtable-3level.h 2011-03-23 17:20:06.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/pgtable-3level.h 2011-04-17 15:57:32.000000000 -0400 +@@ -38,12 +38,16 @@ static inline void native_set_pte_atomic + + static inline void native_set_pmd(pmd_t *pmdp, pmd_t pmd) + { ++ pax_open_kernel(); + set_64bit((unsigned long long *)(pmdp), native_pmd_val(pmd)); ++ pax_close_kernel(); + } + + static inline void native_set_pud(pud_t *pudp, pud_t pud) + { ++ pax_open_kernel(); + set_64bit((unsigned long long *)(pudp), native_pud_val(pud)); ++ pax_close_kernel(); + } + + /* +diff -urNp linux-2.6.38.3/arch/x86/include/asm/pgtable_64.h linux-2.6.38.3/arch/x86/include/asm/pgtable_64.h +--- linux-2.6.38.3/arch/x86/include/asm/pgtable_64.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/pgtable_64.h 2011-04-17 15:57:32.000000000 -0400 +@@ -16,10 +16,13 @@ + + extern pud_t level3_kernel_pgt[512]; + extern pud_t level3_ident_pgt[512]; ++extern pud_t level3_vmalloc_pgt[512]; ++extern pud_t level3_vmemmap_pgt[512]; ++extern pud_t level2_vmemmap_pgt[512]; + extern pmd_t level2_kernel_pgt[512]; + extern pmd_t level2_fixmap_pgt[512]; +-extern pmd_t level2_ident_pgt[512]; +-extern pgd_t init_level4_pgt[]; ++extern pmd_t level2_ident_pgt[512*2]; ++extern pgd_t init_level4_pgt[512]; + + #define swapper_pg_dir init_level4_pgt + +@@ -61,7 +64,9 @@ static inline void native_set_pte_atomic + + static inline void native_set_pmd(pmd_t *pmdp, pmd_t pmd) + { ++ pax_open_kernel(); + *pmdp = pmd; ++ pax_close_kernel(); + } + + static inline void native_pmd_clear(pmd_t *pmd) +@@ -107,7 +112,9 @@ static inline void native_pud_clear(pud_ + + static inline void native_set_pgd(pgd_t *pgdp, pgd_t pgd) + { ++ pax_open_kernel(); + *pgdp = pgd; ++ pax_close_kernel(); + } + + static inline void native_pgd_clear(pgd_t *pgd) +diff -urNp linux-2.6.38.3/arch/x86/include/asm/pgtable_64_types.h linux-2.6.38.3/arch/x86/include/asm/pgtable_64_types.h +--- linux-2.6.38.3/arch/x86/include/asm/pgtable_64_types.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/pgtable_64_types.h 2011-04-17 15:57:32.000000000 -0400 +@@ -59,5 +59,10 @@ typedef struct { pteval_t pte; } pte_t; + #define MODULES_VADDR _AC(0xffffffffa0000000, UL) + #define MODULES_END _AC(0xffffffffff000000, UL) + #define MODULES_LEN (MODULES_END - MODULES_VADDR) ++#define MODULES_EXEC_VADDR MODULES_VADDR ++#define MODULES_EXEC_END MODULES_END ++ ++#define ktla_ktva(addr) (addr) ++#define ktva_ktla(addr) (addr) + + #endif /* _ASM_X86_PGTABLE_64_DEFS_H */ +diff -urNp linux-2.6.38.3/arch/x86/include/asm/pgtable.h linux-2.6.38.3/arch/x86/include/asm/pgtable.h +--- linux-2.6.38.3/arch/x86/include/asm/pgtable.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/pgtable.h 2011-04-17 15:57:32.000000000 -0400 +@@ -81,12 +81,51 @@ extern struct mm_struct *pgd_page_get_mm + + #define arch_end_context_switch(prev) do {} while(0) + ++#define pax_open_kernel() native_pax_open_kernel() ++#define pax_close_kernel() native_pax_close_kernel() + #endif /* CONFIG_PARAVIRT */ + ++#define __HAVE_ARCH_PAX_OPEN_KERNEL ++#define __HAVE_ARCH_PAX_CLOSE_KERNEL ++ ++#ifdef CONFIG_PAX_KERNEXEC ++static inline unsigned long native_pax_open_kernel(void) ++{ ++ unsigned long cr0; ++ ++ preempt_disable(); ++ barrier(); ++ cr0 = read_cr0() ^ X86_CR0_WP; ++ BUG_ON(unlikely(cr0 & X86_CR0_WP)); ++ write_cr0(cr0); ++ return cr0 ^ X86_CR0_WP; ++} ++ ++static inline unsigned long native_pax_close_kernel(void) ++{ ++ unsigned long cr0; ++ ++ cr0 = read_cr0() ^ X86_CR0_WP; ++ BUG_ON(unlikely(!(cr0 & X86_CR0_WP))); ++ write_cr0(cr0); ++ barrier(); ++ preempt_enable_no_resched(); ++ return cr0 ^ X86_CR0_WP; ++} ++#else ++static inline unsigned long native_pax_open_kernel(void) { return 0; } ++static inline unsigned long native_pax_close_kernel(void) { return 0; } ++#endif ++ + /* + * The following only work if pte_present() is true. + * Undefined behaviour if not.. + */ ++static inline int pte_user(pte_t pte) ++{ ++ return pte_val(pte) & _PAGE_USER; ++} ++ + static inline int pte_dirty(pte_t pte) + { + return pte_flags(pte) & _PAGE_DIRTY; +@@ -196,9 +235,29 @@ static inline pte_t pte_wrprotect(pte_t + return pte_clear_flags(pte, _PAGE_RW); + } + ++static inline pte_t pte_mkread(pte_t pte) ++{ ++ return __pte(pte_val(pte) | _PAGE_USER); ++} ++ + static inline pte_t pte_mkexec(pte_t pte) + { +- return pte_clear_flags(pte, _PAGE_NX); ++#ifdef CONFIG_X86_PAE ++ if (__supported_pte_mask & _PAGE_NX) ++ return pte_clear_flags(pte, _PAGE_NX); ++ else ++#endif ++ return pte_set_flags(pte, _PAGE_USER); ++} ++ ++static inline pte_t pte_exprotect(pte_t pte) ++{ ++#ifdef CONFIG_X86_PAE ++ if (__supported_pte_mask & _PAGE_NX) ++ return pte_set_flags(pte, _PAGE_NX); ++ else ++#endif ++ return pte_clear_flags(pte, _PAGE_USER); + } + + static inline pte_t pte_mkdirty(pte_t pte) +@@ -390,6 +449,15 @@ pte_t *populate_extra_pte(unsigned long + #endif + + #ifndef __ASSEMBLY__ ++ ++#ifdef CONFIG_PAX_PER_CPU_PGD ++extern pgd_t cpu_pgd[NR_CPUS][PTRS_PER_PGD]; ++static inline pgd_t *get_cpu_pgd(unsigned int cpu) ++{ ++ return cpu_pgd[cpu]; ++} ++#endif ++ + #include <linux/mm_types.h> + + static inline int pte_none(pte_t pte) +@@ -560,7 +628,7 @@ static inline pud_t *pud_offset(pgd_t *p + + static inline int pgd_bad(pgd_t pgd) + { +- return (pgd_flags(pgd) & ~_PAGE_USER) != _KERNPG_TABLE; ++ return (pgd_flags(pgd) & ~(_PAGE_USER | _PAGE_NX)) != _KERNPG_TABLE; + } + + static inline int pgd_none(pgd_t pgd) +@@ -583,7 +651,12 @@ static inline int pgd_none(pgd_t pgd) + * pgd_offset() returns a (pgd_t *) + * pgd_index() is used get the offset into the pgd page's array of pgd_t's; + */ +-#define pgd_offset(mm, address) ((mm)->pgd + pgd_index((address))) ++#define pgd_offset(mm, address) ((mm)->pgd + pgd_index(address)) ++ ++#ifdef CONFIG_PAX_PER_CPU_PGD ++#define pgd_offset_cpu(cpu, address) (get_cpu_pgd(cpu) + pgd_index(address)) ++#endif ++ + /* + * a shortcut which implies the use of the kernel's pgd, instead + * of a process's +@@ -594,6 +667,20 @@ static inline int pgd_none(pgd_t pgd) + #define KERNEL_PGD_BOUNDARY pgd_index(PAGE_OFFSET) + #define KERNEL_PGD_PTRS (PTRS_PER_PGD - KERNEL_PGD_BOUNDARY) + ++#ifdef CONFIG_X86_32 ++#define USER_PGD_PTRS KERNEL_PGD_BOUNDARY ++#else ++#define TASK_SIZE_MAX_SHIFT CONFIG_TASK_SIZE_MAX_SHIFT ++#define USER_PGD_PTRS (_AC(1,UL) << (TASK_SIZE_MAX_SHIFT - PGDIR_SHIFT)) ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++#define PAX_USER_SHADOW_BASE (_AC(1,UL) << TASK_SIZE_MAX_SHIFT) ++#else ++#define PAX_USER_SHADOW_BASE (_AC(0,UL)) ++#endif ++ ++#endif ++ + #ifndef __ASSEMBLY__ + + extern int direct_gbpages; +@@ -758,11 +845,23 @@ static inline void pmdp_set_wrprotect(st + * dst and src can be on the same page, but the range must not overlap, + * and must not cross a page boundary. + */ +-static inline void clone_pgd_range(pgd_t *dst, pgd_t *src, int count) ++static inline void clone_pgd_range(pgd_t *dst, const pgd_t *src, int count) + { +- memcpy(dst, src, count * sizeof(pgd_t)); ++ pax_open_kernel(); ++ while (count--) ++ *dst++ = *src++; ++ pax_close_kernel(); + } + ++#ifdef CONFIG_PAX_PER_CPU_PGD ++extern void __clone_user_pgds(pgd_t *dst, const pgd_t *src, int count); ++#endif ++ ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) ++extern void __shadow_user_pgds(pgd_t *dst, const pgd_t *src, int count); ++#else ++static inline void __shadow_user_pgds(pgd_t *dst, const pgd_t *src, int count) {} ++#endif + + #include <asm-generic/pgtable.h> + #endif /* __ASSEMBLY__ */ +diff -urNp linux-2.6.38.3/arch/x86/include/asm/pgtable_types.h linux-2.6.38.3/arch/x86/include/asm/pgtable_types.h +--- linux-2.6.38.3/arch/x86/include/asm/pgtable_types.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/pgtable_types.h 2011-04-17 15:57:32.000000000 -0400 +@@ -16,13 +16,12 @@ + #define _PAGE_BIT_PSE 7 /* 4 MB (or 2MB) page */ + #define _PAGE_BIT_PAT 7 /* on 4KB pages */ + #define _PAGE_BIT_GLOBAL 8 /* Global TLB entry PPro+ */ +-#define _PAGE_BIT_UNUSED1 9 /* available for programmer */ ++#define _PAGE_BIT_SPECIAL 9 /* special mappings, no associated struct page */ + #define _PAGE_BIT_IOMAP 10 /* flag used to indicate IO mapping */ + #define _PAGE_BIT_HIDDEN 11 /* hidden by kmemcheck */ + #define _PAGE_BIT_PAT_LARGE 12 /* On 2MB or 1GB pages */ +-#define _PAGE_BIT_SPECIAL _PAGE_BIT_UNUSED1 +-#define _PAGE_BIT_CPA_TEST _PAGE_BIT_UNUSED1 +-#define _PAGE_BIT_SPLITTING _PAGE_BIT_UNUSED1 /* only valid on a PSE pmd */ ++#define _PAGE_BIT_CPA_TEST _PAGE_BIT_SPECIAL ++#define _PAGE_BIT_SPLITTING _PAGE_BIT_SPECIAL /* only valid on a PSE pmd */ + #define _PAGE_BIT_NX 63 /* No execute: only valid after cpuid check */ + + /* If _PAGE_BIT_PRESENT is clear, we use these: */ +@@ -40,7 +39,6 @@ + #define _PAGE_DIRTY (_AT(pteval_t, 1) << _PAGE_BIT_DIRTY) + #define _PAGE_PSE (_AT(pteval_t, 1) << _PAGE_BIT_PSE) + #define _PAGE_GLOBAL (_AT(pteval_t, 1) << _PAGE_BIT_GLOBAL) +-#define _PAGE_UNUSED1 (_AT(pteval_t, 1) << _PAGE_BIT_UNUSED1) + #define _PAGE_IOMAP (_AT(pteval_t, 1) << _PAGE_BIT_IOMAP) + #define _PAGE_PAT (_AT(pteval_t, 1) << _PAGE_BIT_PAT) + #define _PAGE_PAT_LARGE (_AT(pteval_t, 1) << _PAGE_BIT_PAT_LARGE) +@@ -57,8 +55,10 @@ + + #if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE) + #define _PAGE_NX (_AT(pteval_t, 1) << _PAGE_BIT_NX) +-#else ++#elif defined(CONFIG_KMEMCHECK) + #define _PAGE_NX (_AT(pteval_t, 0)) ++#else ++#define _PAGE_NX (_AT(pteval_t, 1) << _PAGE_BIT_HIDDEN) + #endif + + #define _PAGE_FILE (_AT(pteval_t, 1) << _PAGE_BIT_FILE) +@@ -96,6 +96,9 @@ + #define PAGE_READONLY_EXEC __pgprot(_PAGE_PRESENT | _PAGE_USER | \ + _PAGE_ACCESSED) + ++#define PAGE_READONLY_NOEXEC PAGE_READONLY ++#define PAGE_SHARED_NOEXEC PAGE_SHARED ++ + #define __PAGE_KERNEL_EXEC \ + (_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED | _PAGE_GLOBAL) + #define __PAGE_KERNEL (__PAGE_KERNEL_EXEC | _PAGE_NX) +@@ -106,8 +109,8 @@ + #define __PAGE_KERNEL_WC (__PAGE_KERNEL | _PAGE_CACHE_WC) + #define __PAGE_KERNEL_NOCACHE (__PAGE_KERNEL | _PAGE_PCD | _PAGE_PWT) + #define __PAGE_KERNEL_UC_MINUS (__PAGE_KERNEL | _PAGE_PCD) +-#define __PAGE_KERNEL_VSYSCALL (__PAGE_KERNEL_RX | _PAGE_USER) +-#define __PAGE_KERNEL_VSYSCALL_NOCACHE (__PAGE_KERNEL_VSYSCALL | _PAGE_PCD | _PAGE_PWT) ++#define __PAGE_KERNEL_VSYSCALL (__PAGE_KERNEL_RO | _PAGE_USER) ++#define __PAGE_KERNEL_VSYSCALL_NOCACHE (__PAGE_KERNEL_RO | _PAGE_PCD | _PAGE_PWT | _PAGE_USER) + #define __PAGE_KERNEL_LARGE (__PAGE_KERNEL | _PAGE_PSE) + #define __PAGE_KERNEL_LARGE_NOCACHE (__PAGE_KERNEL | _PAGE_CACHE_UC | _PAGE_PSE) + #define __PAGE_KERNEL_LARGE_EXEC (__PAGE_KERNEL_EXEC | _PAGE_PSE) +@@ -166,8 +169,8 @@ + * bits are combined, this will alow user to access the high address mapped + * VDSO in the presence of CONFIG_COMPAT_VDSO + */ +-#define PTE_IDENT_ATTR 0x003 /* PRESENT+RW */ +-#define PDE_IDENT_ATTR 0x067 /* PRESENT+RW+USER+DIRTY+ACCESSED */ ++#define PTE_IDENT_ATTR 0x063 /* PRESENT+RW+DIRTY+ACCESSED */ ++#define PDE_IDENT_ATTR 0x063 /* PRESENT+RW+DIRTY+ACCESSED */ + #define PGD_IDENT_ATTR 0x001 /* PRESENT (no other attributes) */ + #endif + +@@ -205,7 +208,17 @@ static inline pgdval_t pgd_flags(pgd_t p + { + return native_pgd_val(pgd) & PTE_FLAGS_MASK; + } ++#endif + ++#if PAGETABLE_LEVELS == 3 ++#include <asm-generic/pgtable-nopud.h> ++#endif ++ ++#if PAGETABLE_LEVELS == 2 ++#include <asm-generic/pgtable-nopmd.h> ++#endif ++ ++#ifndef __ASSEMBLY__ + #if PAGETABLE_LEVELS > 3 + typedef struct { pudval_t pud; } pud_t; + +@@ -219,8 +232,6 @@ static inline pudval_t native_pud_val(pu + return pud.pud; + } + #else +-#include <asm-generic/pgtable-nopud.h> +- + static inline pudval_t native_pud_val(pud_t pud) + { + return native_pgd_val(pud.pgd); +@@ -240,8 +251,6 @@ static inline pmdval_t native_pmd_val(pm + return pmd.pmd; + } + #else +-#include <asm-generic/pgtable-nopmd.h> +- + static inline pmdval_t native_pmd_val(pmd_t pmd) + { + return native_pgd_val(pmd.pud.pgd); +@@ -281,7 +290,6 @@ typedef struct page *pgtable_t; + + extern pteval_t __supported_pte_mask; + extern void set_nx(void); +-extern int nx_enabled; + + #define pgprot_writecombine pgprot_writecombine + extern pgprot_t pgprot_writecombine(pgprot_t prot); +diff -urNp linux-2.6.38.3/arch/x86/include/asm/processor.h linux-2.6.38.3/arch/x86/include/asm/processor.h +--- linux-2.6.38.3/arch/x86/include/asm/processor.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/processor.h 2011-04-17 15:57:32.000000000 -0400 +@@ -270,7 +270,7 @@ struct tss_struct { + + } ____cacheline_aligned; + +-DECLARE_PER_CPU_SHARED_ALIGNED(struct tss_struct, init_tss); ++extern struct tss_struct init_tss[NR_CPUS]; + + /* + * Save the original ist values for checking stack pointers during debugging +@@ -864,8 +864,15 @@ static inline void spin_lock_prefetch(co + */ + #define TASK_SIZE PAGE_OFFSET + #define TASK_SIZE_MAX TASK_SIZE ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++#define SEGMEXEC_TASK_SIZE (TASK_SIZE / 2) ++#define STACK_TOP ((current->mm->pax_flags & MF_PAX_SEGMEXEC)?SEGMEXEC_TASK_SIZE:TASK_SIZE) ++#else + #define STACK_TOP TASK_SIZE +-#define STACK_TOP_MAX STACK_TOP ++#endif ++ ++#define STACK_TOP_MAX TASK_SIZE + + #define INIT_THREAD { \ + .sp0 = sizeof(init_stack) + (long)&init_stack, \ +@@ -882,7 +889,7 @@ static inline void spin_lock_prefetch(co + */ + #define INIT_TSS { \ + .x86_tss = { \ +- .sp0 = sizeof(init_stack) + (long)&init_stack, \ ++ .sp0 = sizeof(init_stack) + (long)&init_stack - 8, \ + .ss0 = __KERNEL_DS, \ + .ss1 = __KERNEL_CS, \ + .io_bitmap_base = INVALID_IO_BITMAP_OFFSET, \ +@@ -893,11 +900,7 @@ static inline void spin_lock_prefetch(co + extern unsigned long thread_saved_pc(struct task_struct *tsk); + + #define THREAD_SIZE_LONGS (THREAD_SIZE/sizeof(unsigned long)) +-#define KSTK_TOP(info) \ +-({ \ +- unsigned long *__ptr = (unsigned long *)(info); \ +- (unsigned long)(&__ptr[THREAD_SIZE_LONGS]); \ +-}) ++#define KSTK_TOP(info) ((info)->task.thread.sp0) + + /* + * The below -8 is to reserve 8 bytes on top of the ring0 stack. +@@ -912,7 +915,7 @@ extern unsigned long thread_saved_pc(str + #define task_pt_regs(task) \ + ({ \ + struct pt_regs *__regs__; \ +- __regs__ = (struct pt_regs *)(KSTK_TOP(task_stack_page(task))-8); \ ++ __regs__ = (struct pt_regs *)((task)->thread.sp0); \ + __regs__ - 1; \ + }) + +@@ -922,13 +925,13 @@ extern unsigned long thread_saved_pc(str + /* + * User space process size. 47bits minus one guard page. + */ +-#define TASK_SIZE_MAX ((1UL << 47) - PAGE_SIZE) ++#define TASK_SIZE_MAX ((1UL << TASK_SIZE_MAX_SHIFT) - PAGE_SIZE) + + /* This decides where the kernel will search for a free chunk of vm + * space during mmap's. + */ + #define IA32_PAGE_OFFSET ((current->personality & ADDR_LIMIT_3GB) ? \ +- 0xc0000000 : 0xFFFFe000) ++ 0xc0000000 : 0xFFFFf000) + + #define TASK_SIZE (test_thread_flag(TIF_IA32) ? \ + IA32_PAGE_OFFSET : TASK_SIZE_MAX) +@@ -965,6 +968,10 @@ extern void start_thread(struct pt_regs + */ + #define TASK_UNMAPPED_BASE (PAGE_ALIGN(TASK_SIZE / 3)) + ++#ifdef CONFIG_PAX_SEGMEXEC ++#define SEGMEXEC_TASK_UNMAPPED_BASE (PAGE_ALIGN(SEGMEXEC_TASK_SIZE / 3)) ++#endif ++ + #define KSTK_EIP(task) (task_pt_regs(task)->ip) + + /* Get/set a process' ability to use the timestamp counter instruction */ +diff -urNp linux-2.6.38.3/arch/x86/include/asm/ptrace.h linux-2.6.38.3/arch/x86/include/asm/ptrace.h +--- linux-2.6.38.3/arch/x86/include/asm/ptrace.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/ptrace.h 2011-04-17 15:57:32.000000000 -0400 +@@ -152,28 +152,29 @@ static inline unsigned long regs_return_ + } + + /* +- * user_mode_vm(regs) determines whether a register set came from user mode. ++ * user_mode(regs) determines whether a register set came from user mode. + * This is true if V8086 mode was enabled OR if the register set was from + * protected mode with RPL-3 CS value. This tricky test checks that with + * one comparison. Many places in the kernel can bypass this full check +- * if they have already ruled out V8086 mode, so user_mode(regs) can be used. ++ * if they have already ruled out V8086 mode, so user_mode_novm(regs) can ++ * be used. + */ +-static inline int user_mode(struct pt_regs *regs) ++static inline int user_mode_novm(struct pt_regs *regs) + { + #ifdef CONFIG_X86_32 + return (regs->cs & SEGMENT_RPL_MASK) == USER_RPL; + #else +- return !!(regs->cs & 3); ++ return !!(regs->cs & SEGMENT_RPL_MASK); + #endif + } + +-static inline int user_mode_vm(struct pt_regs *regs) ++static inline int user_mode(struct pt_regs *regs) + { + #ifdef CONFIG_X86_32 + return ((regs->cs & SEGMENT_RPL_MASK) | (regs->flags & X86_VM_MASK)) >= + USER_RPL; + #else +- return user_mode(regs); ++ return user_mode_novm(regs); + #endif + } + +diff -urNp linux-2.6.38.3/arch/x86/include/asm/reboot.h linux-2.6.38.3/arch/x86/include/asm/reboot.h +--- linux-2.6.38.3/arch/x86/include/asm/reboot.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/reboot.h 2011-04-17 15:57:32.000000000 -0400 +@@ -18,7 +18,7 @@ extern struct machine_ops machine_ops; + + void native_machine_crash_shutdown(struct pt_regs *regs); + void native_machine_shutdown(void); +-void machine_real_restart(const unsigned char *code, int length); ++void machine_real_restart(const unsigned char *code, unsigned int length); + + typedef void (*nmi_shootdown_cb)(int, struct die_args*); + void nmi_shootdown_cpus(nmi_shootdown_cb callback); +diff -urNp linux-2.6.38.3/arch/x86/include/asm/rwsem.h linux-2.6.38.3/arch/x86/include/asm/rwsem.h +--- linux-2.6.38.3/arch/x86/include/asm/rwsem.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/rwsem.h 2011-04-17 15:57:32.000000000 -0400 +@@ -118,6 +118,14 @@ static inline void __down_read(struct rw + { + asm volatile("# beginning down_read\n\t" + LOCK_PREFIX _ASM_INC "(%1)\n\t" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX _ASM_DEC "(%1)\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ + /* adds 0x00000001 */ + " jns 1f\n" + " call call_rwsem_down_read_failed\n" +@@ -139,6 +147,14 @@ static inline int __down_read_trylock(st + "1:\n\t" + " mov %1,%2\n\t" + " add %3,%2\n\t" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "sub %3,%2\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ + " jle 2f\n\t" + LOCK_PREFIX " cmpxchg %2,%0\n\t" + " jnz 1b\n\t" +@@ -158,6 +174,14 @@ static inline void __down_write_nested(s + rwsem_count_t tmp; + asm volatile("# beginning down_write\n\t" + LOCK_PREFIX " xadd %1,(%2)\n\t" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "mov %1,(%2)\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ + /* adds 0xffff0001, returns the old value */ + " test %1,%1\n\t" + /* was the count 0 before? */ +@@ -196,6 +220,14 @@ static inline void __up_read(struct rw_s + rwsem_count_t tmp; + asm volatile("# beginning __up_read\n\t" + LOCK_PREFIX " xadd %1,(%2)\n\t" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "mov %1,(%2)\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ + /* subtracts 1, returns the old value */ + " jns 1f\n\t" + " call call_rwsem_wake\n" /* expects old value in %edx */ +@@ -214,6 +246,14 @@ static inline void __up_write(struct rw_ + rwsem_count_t tmp; + asm volatile("# beginning __up_write\n\t" + LOCK_PREFIX " xadd %1,(%2)\n\t" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "mov %1,(%2)\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ + /* subtracts 0xffff0001, returns the old value */ + " jns 1f\n\t" + " call call_rwsem_wake\n" /* expects old value in %edx */ +@@ -231,6 +271,14 @@ static inline void __downgrade_write(str + { + asm volatile("# beginning __downgrade_write\n\t" + LOCK_PREFIX _ASM_ADD "%2,(%1)\n\t" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX _ASM_SUB "%2,(%1)\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ + /* + * transitions 0xZZZZ0001 -> 0xYYYY0001 (i386) + * 0xZZZZZZZZ00000001 -> 0xYYYYYYYY00000001 (x86_64) +@@ -250,7 +298,15 @@ static inline void __downgrade_write(str + static inline void rwsem_atomic_add(rwsem_count_t delta, + struct rw_semaphore *sem) + { +- asm volatile(LOCK_PREFIX _ASM_ADD "%1,%0" ++ asm volatile(LOCK_PREFIX _ASM_ADD "%1,%0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX _ASM_SUB "%1,%0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ + : "+m" (sem->count) + : "er" (delta)); + } +@@ -263,7 +319,15 @@ static inline rwsem_count_t rwsem_atomic + { + rwsem_count_t tmp = delta; + +- asm volatile(LOCK_PREFIX "xadd %0,%1" ++ asm volatile(LOCK_PREFIX "xadd %0,%1\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ "mov %0,%1\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ + : "+r" (tmp), "+m" (sem->count) + : : "memory"); + +diff -urNp linux-2.6.38.3/arch/x86/include/asm/segment.h linux-2.6.38.3/arch/x86/include/asm/segment.h +--- linux-2.6.38.3/arch/x86/include/asm/segment.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/segment.h 2011-04-17 15:57:32.000000000 -0400 +@@ -62,8 +62,8 @@ + * 26 - ESPFIX small SS + * 27 - per-cpu [ offset to per-cpu data area ] + * 28 - stack_canary-20 [ for stack protector ] +- * 29 - unused +- * 30 - unused ++ * 29 - PCI BIOS CS ++ * 30 - PCI BIOS DS + * 31 - TSS for double fault handler + */ + #define GDT_ENTRY_TLS_MIN 6 +@@ -77,6 +77,8 @@ + + #define GDT_ENTRY_KERNEL_CS (GDT_ENTRY_KERNEL_BASE+0) + ++#define GDT_ENTRY_KERNEXEC_KERNEL_CS (4) ++ + #define GDT_ENTRY_KERNEL_DS (GDT_ENTRY_KERNEL_BASE+1) + + #define GDT_ENTRY_TSS (GDT_ENTRY_KERNEL_BASE+4) +@@ -102,6 +104,12 @@ + #define __KERNEL_STACK_CANARY 0 + #endif + ++#define GDT_ENTRY_PCIBIOS_CS (GDT_ENTRY_KERNEL_BASE+17) ++#define __PCIBIOS_CS (GDT_ENTRY_PCIBIOS_CS * 8) ++ ++#define GDT_ENTRY_PCIBIOS_DS (GDT_ENTRY_KERNEL_BASE+18) ++#define __PCIBIOS_DS (GDT_ENTRY_PCIBIOS_DS * 8) ++ + #define GDT_ENTRY_DOUBLEFAULT_TSS 31 + + /* +@@ -139,7 +147,7 @@ + */ + + /* Matches PNP_CS32 and PNP_CS16 (they must be consecutive) */ +-#define SEGMENT_IS_PNP_CODE(x) (((x) & 0xf4) == GDT_ENTRY_PNPBIOS_BASE * 8) ++#define SEGMENT_IS_PNP_CODE(x) (((x) & 0xFFFCU) == PNP_CS32 || ((x) & 0xFFFCU) == PNP_CS16) + + + #else +@@ -163,6 +171,8 @@ + #define __USER32_CS (GDT_ENTRY_DEFAULT_USER32_CS * 8 + 3) + #define __USER32_DS __USER_DS + ++#define GDT_ENTRY_KERNEXEC_KERNEL_CS 7 ++ + #define GDT_ENTRY_TSS 8 /* needs two entries */ + #define GDT_ENTRY_LDT 10 /* needs two entries */ + #define GDT_ENTRY_TLS_MIN 12 +@@ -183,6 +193,7 @@ + #endif + + #define __KERNEL_CS (GDT_ENTRY_KERNEL_CS*8) ++#define __KERNEXEC_KERNEL_CS (GDT_ENTRY_KERNEXEC_KERNEL_CS*8) + #define __KERNEL_DS (GDT_ENTRY_KERNEL_DS*8) + #define __USER_DS (GDT_ENTRY_DEFAULT_USER_DS*8+3) + #define __USER_CS (GDT_ENTRY_DEFAULT_USER_CS*8+3) +diff -urNp linux-2.6.38.3/arch/x86/include/asm/smp.h linux-2.6.38.3/arch/x86/include/asm/smp.h +--- linux-2.6.38.3/arch/x86/include/asm/smp.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/smp.h 2011-04-17 15:57:32.000000000 -0400 +@@ -24,7 +24,7 @@ extern unsigned int num_processors; + DECLARE_PER_CPU(cpumask_var_t, cpu_sibling_map); + DECLARE_PER_CPU(cpumask_var_t, cpu_core_map); + DECLARE_PER_CPU(u16, cpu_llc_id); +-DECLARE_PER_CPU(int, cpu_number); ++DECLARE_PER_CPU(unsigned int, cpu_number); + + static inline struct cpumask *cpu_sibling_mask(int cpu) + { +@@ -172,14 +172,8 @@ extern unsigned disabled_cpus __cpuinitd + extern int safe_smp_processor_id(void); + + #elif defined(CONFIG_X86_64_SMP) +-#define raw_smp_processor_id() (percpu_read(cpu_number)) +- +-#define stack_smp_processor_id() \ +-({ \ +- struct thread_info *ti; \ +- __asm__("andq %%rsp,%0; ":"=r" (ti) : "0" (CURRENT_MASK)); \ +- ti->cpu; \ +-}) ++#define raw_smp_processor_id() (percpu_read(cpu_number)) ++#define stack_smp_processor_id() raw_smp_processor_id() + #define safe_smp_processor_id() smp_processor_id() + + #endif +diff -urNp linux-2.6.38.3/arch/x86/include/asm/spinlock.h linux-2.6.38.3/arch/x86/include/asm/spinlock.h +--- linux-2.6.38.3/arch/x86/include/asm/spinlock.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/spinlock.h 2011-04-17 15:57:32.000000000 -0400 +@@ -249,6 +249,14 @@ static inline int arch_write_can_lock(ar + static inline void arch_read_lock(arch_rwlock_t *rw) + { + asm volatile(LOCK_PREFIX " subl $1,(%0)\n\t" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX " addl $1,(%0)\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ + "jns 1f\n" + "call __read_lock_failed\n\t" + "1:\n" +@@ -258,6 +266,14 @@ static inline void arch_read_lock(arch_r + static inline void arch_write_lock(arch_rwlock_t *rw) + { + asm volatile(LOCK_PREFIX " subl %1,(%0)\n\t" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX " addl %1,(%0)\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ + "jz 1f\n" + "call __write_lock_failed\n\t" + "1:\n" +@@ -286,12 +302,29 @@ static inline int arch_write_trylock(arc + + static inline void arch_read_unlock(arch_rwlock_t *rw) + { +- asm volatile(LOCK_PREFIX "incl %0" :"+m" (rw->lock) : : "memory"); ++ asm volatile(LOCK_PREFIX "incl %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "decl %0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ ++ :"+m" (rw->lock) : : "memory"); + } + + static inline void arch_write_unlock(arch_rwlock_t *rw) + { +- asm volatile(LOCK_PREFIX "addl %1, %0" ++ asm volatile(LOCK_PREFIX "addl %1, %0\n" ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ "jno 0f\n" ++ LOCK_PREFIX "subl %1, %0\n" ++ "int $4\n0:\n" ++ _ASM_EXTABLE(0b, 0b) ++#endif ++ + : "+m" (rw->lock) : "i" (RW_LOCK_BIAS) : "memory"); + } + +diff -urNp linux-2.6.38.3/arch/x86/include/asm/stackprotector.h linux-2.6.38.3/arch/x86/include/asm/stackprotector.h +--- linux-2.6.38.3/arch/x86/include/asm/stackprotector.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/stackprotector.h 2011-04-17 15:57:32.000000000 -0400 +@@ -113,7 +113,7 @@ static inline void setup_stack_canary_se + + static inline void load_stack_canary_segment(void) + { +-#ifdef CONFIG_X86_32 ++#if defined(CONFIG_X86_32) && !defined(CONFIG_PAX_MEMORY_UDEREF) + asm volatile ("mov %0, %%gs" : : "r" (0)); + #endif + } +diff -urNp linux-2.6.38.3/arch/x86/include/asm/stacktrace.h linux-2.6.38.3/arch/x86/include/asm/stacktrace.h +--- linux-2.6.38.3/arch/x86/include/asm/stacktrace.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/stacktrace.h 2011-04-17 15:57:32.000000000 -0400 +@@ -11,28 +11,20 @@ + + extern int kstack_depth_to_print; + +-struct thread_info; ++struct task_struct; + struct stacktrace_ops; + +-typedef unsigned long (*walk_stack_t)(struct thread_info *tinfo, +- unsigned long *stack, +- unsigned long bp, +- const struct stacktrace_ops *ops, +- void *data, +- unsigned long *end, +- int *graph); +- +-extern unsigned long +-print_context_stack(struct thread_info *tinfo, +- unsigned long *stack, unsigned long bp, +- const struct stacktrace_ops *ops, void *data, +- unsigned long *end, int *graph); +- +-extern unsigned long +-print_context_stack_bp(struct thread_info *tinfo, +- unsigned long *stack, unsigned long bp, +- const struct stacktrace_ops *ops, void *data, +- unsigned long *end, int *graph); ++typedef unsigned long walk_stack_t(struct task_struct *task, ++ void *stack_start, ++ unsigned long *stack, ++ unsigned long bp, ++ const struct stacktrace_ops *ops, ++ void *data, ++ unsigned long *end, ++ int *graph); ++ ++extern walk_stack_t print_context_stack; ++extern walk_stack_t print_context_stack_bp; + + /* Generic stack tracer with callbacks */ + +@@ -43,7 +35,7 @@ struct stacktrace_ops { + void (*address)(void *data, unsigned long address, int reliable); + /* On negative return stop dumping */ + int (*stack)(void *data, char *name); +- walk_stack_t walk_stack; ++ walk_stack_t *walk_stack; + }; + + void dump_trace(struct task_struct *tsk, struct pt_regs *regs, +diff -urNp linux-2.6.38.3/arch/x86/include/asm/system.h linux-2.6.38.3/arch/x86/include/asm/system.h +--- linux-2.6.38.3/arch/x86/include/asm/system.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/system.h 2011-04-17 15:57:32.000000000 -0400 +@@ -131,7 +131,7 @@ do { \ + "call __switch_to\n\t" \ + "movq "__percpu_arg([current_task])",%%rsi\n\t" \ + __switch_canary \ +- "movq %P[thread_info](%%rsi),%%r8\n\t" \ ++ "movq "__percpu_arg([thread_info])",%%r8\n\t" \ + "movq %%rax,%%rdi\n\t" \ + "testl %[_tif_fork],%P[ti_flags](%%r8)\n\t" \ + "jnz ret_from_fork\n\t" \ +@@ -142,7 +142,7 @@ do { \ + [threadrsp] "i" (offsetof(struct task_struct, thread.sp)), \ + [ti_flags] "i" (offsetof(struct thread_info, flags)), \ + [_tif_fork] "i" (_TIF_FORK), \ +- [thread_info] "i" (offsetof(struct task_struct, stack)), \ ++ [thread_info] "m" (current_tinfo), \ + [current_task] "m" (current_task) \ + __switch_canary_iparam \ + : "memory", "cc" __EXTRA_CLOBBER) +@@ -202,7 +202,7 @@ static inline unsigned long get_limit(un + { + unsigned long __limit; + asm("lsll %1,%0" : "=r" (__limit) : "r" (segment)); +- return __limit + 1; ++ return __limit; + } + + static inline void native_clts(void) +@@ -342,7 +342,7 @@ void enable_hlt(void); + + void cpu_idle_wait(void); + +-extern unsigned long arch_align_stack(unsigned long sp); ++#define arch_align_stack(x) ((x) & ~0xfUL) + extern void free_init_pages(char *what, unsigned long begin, unsigned long end); + + void default_idle(void); +diff -urNp linux-2.6.38.3/arch/x86/include/asm/thread_info.h linux-2.6.38.3/arch/x86/include/asm/thread_info.h +--- linux-2.6.38.3/arch/x86/include/asm/thread_info.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/thread_info.h 2011-04-17 15:57:32.000000000 -0400 +@@ -10,6 +10,7 @@ + #include <linux/compiler.h> + #include <asm/page.h> + #include <asm/types.h> ++#include <asm/percpu.h> + + /* + * low level task data that entry.S needs immediate access to +@@ -24,7 +25,6 @@ struct exec_domain; + #include <asm/atomic.h> + + struct thread_info { +- struct task_struct *task; /* main task structure */ + struct exec_domain *exec_domain; /* execution domain */ + __u32 flags; /* low level flags */ + __u32 status; /* thread synchronous flags */ +@@ -34,18 +34,11 @@ struct thread_info { + mm_segment_t addr_limit; + struct restart_block restart_block; + void __user *sysenter_return; +-#ifdef CONFIG_X86_32 +- unsigned long previous_esp; /* ESP of the previous stack in +- case of nested (IRQ) stacks +- */ +- __u8 supervisor_stack[0]; +-#endif + int uaccess_err; + }; + +-#define INIT_THREAD_INFO(tsk) \ ++#define INIT_THREAD_INFO \ + { \ +- .task = &tsk, \ + .exec_domain = &default_exec_domain, \ + .flags = 0, \ + .cpu = 0, \ +@@ -56,7 +49,7 @@ struct thread_info { + }, \ + } + +-#define init_thread_info (init_thread_union.thread_info) ++#define init_thread_info (init_task.tinfo) + #define init_stack (init_thread_union.stack) + + #else /* !__ASSEMBLY__ */ +@@ -164,6 +157,23 @@ struct thread_info { + #define alloc_thread_info(tsk) \ + ((struct thread_info *)__get_free_pages(THREAD_FLAGS, THREAD_ORDER)) + ++#ifdef __ASSEMBLY__ ++/* how to get the thread information struct from ASM */ ++#define GET_THREAD_INFO(reg) \ ++ mov PER_CPU_VAR(current_tinfo), reg ++ ++/* use this one if reg already contains %esp */ ++#define GET_THREAD_INFO_WITH_ESP(reg) GET_THREAD_INFO(reg) ++#else ++/* how to get the thread information struct from C */ ++DECLARE_PER_CPU(struct thread_info *, current_tinfo); ++ ++static __always_inline struct thread_info *current_thread_info(void) ++{ ++ return percpu_read_stable(current_tinfo); ++} ++#endif ++ + #ifdef CONFIG_X86_32 + + #define STACK_WARN (THREAD_SIZE/8) +@@ -174,35 +184,13 @@ struct thread_info { + */ + #ifndef __ASSEMBLY__ + +- + /* how to get the current stack pointer from C */ + register unsigned long current_stack_pointer asm("esp") __used; + +-/* how to get the thread information struct from C */ +-static inline struct thread_info *current_thread_info(void) +-{ +- return (struct thread_info *) +- (current_stack_pointer & ~(THREAD_SIZE - 1)); +-} +- +-#else /* !__ASSEMBLY__ */ +- +-/* how to get the thread information struct from ASM */ +-#define GET_THREAD_INFO(reg) \ +- movl $-THREAD_SIZE, reg; \ +- andl %esp, reg +- +-/* use this one if reg already contains %esp */ +-#define GET_THREAD_INFO_WITH_ESP(reg) \ +- andl $-THREAD_SIZE, reg +- + #endif + + #else /* X86_32 */ + +-#include <asm/percpu.h> +-#define KERNEL_STACK_OFFSET (5*8) +- + /* + * macros/functions for gaining access to the thread information structure + * preempt_count needs to be 1 initially, until the scheduler is functional. +@@ -210,21 +198,6 @@ static inline struct thread_info *curren + #ifndef __ASSEMBLY__ + DECLARE_PER_CPU(unsigned long, kernel_stack); + +-static inline struct thread_info *current_thread_info(void) +-{ +- struct thread_info *ti; +- ti = (void *)(percpu_read_stable(kernel_stack) + +- KERNEL_STACK_OFFSET - THREAD_SIZE); +- return ti; +-} +- +-#else /* !__ASSEMBLY__ */ +- +-/* how to get the thread information struct from ASM */ +-#define GET_THREAD_INFO(reg) \ +- movq PER_CPU_VAR(kernel_stack),reg ; \ +- subq $(THREAD_SIZE-KERNEL_STACK_OFFSET),reg +- + #endif + + #endif /* !X86_32 */ +@@ -260,5 +233,16 @@ extern void arch_task_cache_init(void); + extern void free_thread_info(struct thread_info *ti); + extern int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src); + #define arch_task_cache_init arch_task_cache_init ++ ++#define __HAVE_THREAD_FUNCTIONS ++#define task_thread_info(task) (&(task)->tinfo) ++#define task_stack_page(task) ((task)->stack) ++#define setup_thread_stack(p, org) do {} while (0) ++#define end_of_stack(p) ((unsigned long *)task_stack_page(p) + 1) ++ ++#define __HAVE_ARCH_TASK_STRUCT_ALLOCATOR ++extern struct task_struct *alloc_task_struct(void); ++extern void free_task_struct(struct task_struct *); ++ + #endif + #endif /* _ASM_X86_THREAD_INFO_H */ +diff -urNp linux-2.6.38.3/arch/x86/include/asm/uaccess_32.h linux-2.6.38.3/arch/x86/include/asm/uaccess_32.h +--- linux-2.6.38.3/arch/x86/include/asm/uaccess_32.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/uaccess_32.h 2011-04-17 15:57:32.000000000 -0400 +@@ -44,6 +44,9 @@ unsigned long __must_check __copy_from_u + static __always_inline unsigned long __must_check + __copy_to_user_inatomic(void __user *to, const void *from, unsigned long n) + { ++ if ((long)n < 0) ++ return n; ++ + if (__builtin_constant_p(n)) { + unsigned long ret; + +@@ -62,6 +65,8 @@ __copy_to_user_inatomic(void __user *to, + return ret; + } + } ++ if (!__builtin_constant_p(n)) ++ check_object_size(from, n, true); + return __copy_to_user_ll(to, from, n); + } + +@@ -89,6 +94,9 @@ __copy_to_user(void __user *to, const vo + static __always_inline unsigned long + __copy_from_user_inatomic(void *to, const void __user *from, unsigned long n) + { ++ if ((long)n < 0) ++ return n; ++ + /* Avoid zeroing the tail if the copy fails.. + * If 'n' is constant and 1, 2, or 4, we do still zero on a failure, + * but as the zeroing behaviour is only significant when n is not +@@ -138,6 +146,10 @@ static __always_inline unsigned long + __copy_from_user(void *to, const void __user *from, unsigned long n) + { + might_fault(); ++ ++ if ((long)n < 0) ++ return n; ++ + if (__builtin_constant_p(n)) { + unsigned long ret; + +@@ -153,6 +165,8 @@ __copy_from_user(void *to, const void __ + return ret; + } + } ++ if (!__builtin_constant_p(n)) ++ check_object_size(to, n, false); + return __copy_from_user_ll(to, from, n); + } + +@@ -160,6 +174,10 @@ static __always_inline unsigned long __c + const void __user *from, unsigned long n) + { + might_fault(); ++ ++ if ((long)n < 0) ++ return n; ++ + if (__builtin_constant_p(n)) { + unsigned long ret; + +@@ -182,15 +200,19 @@ static __always_inline unsigned long + __copy_from_user_inatomic_nocache(void *to, const void __user *from, + unsigned long n) + { +- return __copy_from_user_ll_nocache_nozero(to, from, n); +-} ++ if ((long)n < 0) ++ return n; + +-unsigned long __must_check copy_to_user(void __user *to, +- const void *from, unsigned long n); +-unsigned long __must_check _copy_from_user(void *to, +- const void __user *from, +- unsigned long n); ++ return __copy_from_user_ll_nocache_nozero(to, from, n); ++} + ++extern void copy_to_user_overflow(void) ++#ifdef CONFIG_DEBUG_STRICT_USER_COPY_CHECKS ++ __compiletime_error("copy_to_user() buffer size is not provably correct") ++#else ++ __compiletime_warning("copy_to_user() buffer size is not provably correct") ++#endif ++; + + extern void copy_from_user_overflow(void) + #ifdef CONFIG_DEBUG_STRICT_USER_COPY_CHECKS +@@ -200,17 +222,61 @@ extern void copy_from_user_overflow(void + #endif + ; + +-static inline unsigned long __must_check copy_from_user(void *to, +- const void __user *from, +- unsigned long n) ++/** ++ * copy_to_user: - Copy a block of data into user space. ++ * @to: Destination address, in user space. ++ * @from: Source address, in kernel space. ++ * @n: Number of bytes to copy. ++ * ++ * Context: User context only. This function may sleep. ++ * ++ * Copy data from kernel space to user space. ++ * ++ * Returns number of bytes that could not be copied. ++ * On success, this will be zero. ++ */ ++static inline unsigned long __must_check ++copy_to_user(void __user *to, const void *from, unsigned long n) ++{ ++ int sz = __compiletime_object_size(from); ++ ++ if (unlikely(sz != -1 && sz < n)) ++ copy_to_user_overflow(); ++ else if (access_ok(VERIFY_WRITE, to, n)) ++ n = __copy_to_user(to, from, n); ++ return n; ++} ++ ++/** ++ * copy_from_user: - Copy a block of data from user space. ++ * @to: Destination address, in kernel space. ++ * @from: Source address, in user space. ++ * @n: Number of bytes to copy. ++ * ++ * Context: User context only. This function may sleep. ++ * ++ * Copy data from user space to kernel space. ++ * ++ * Returns number of bytes that could not be copied. ++ * On success, this will be zero. ++ * ++ * If some data could not be copied, this function will pad the copied ++ * data to the requested size using zero bytes. ++ */ ++static inline unsigned long __must_check ++copy_from_user(void *to, const void __user *from, unsigned long n) + { + int sz = __compiletime_object_size(to); + +- if (likely(sz == -1 || sz >= n)) +- n = _copy_from_user(to, from, n); +- else ++ if (unlikely(sz != -1 && sz < n)) + copy_from_user_overflow(); +- ++ else if (access_ok(VERIFY_READ, from, n)) ++ n = __copy_from_user(to, from, n); ++ else if ((long)n > 0) { ++ if (!__builtin_constant_p(n)) ++ check_object_size(to, n, false); ++ memset(to, 0, n); ++ } + return n; + } + +diff -urNp linux-2.6.38.3/arch/x86/include/asm/uaccess_64.h linux-2.6.38.3/arch/x86/include/asm/uaccess_64.h +--- linux-2.6.38.3/arch/x86/include/asm/uaccess_64.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/uaccess_64.h 2011-04-17 15:57:32.000000000 -0400 +@@ -11,6 +11,9 @@ + #include <asm/alternative.h> + #include <asm/cpufeature.h> + #include <asm/page.h> ++#include <asm/pgtable.h> ++ ++#define set_fs(x) (current_thread_info()->addr_limit = (x)) + + /* + * Copy To/From Userspace +@@ -37,26 +40,26 @@ copy_user_generic(void *to, const void * + return ret; + } + +-__must_check unsigned long +-_copy_to_user(void __user *to, const void *from, unsigned len); +-__must_check unsigned long +-_copy_from_user(void *to, const void __user *from, unsigned len); ++static __always_inline __must_check unsigned long ++__copy_to_user(void __user *to, const void *from, unsigned len); ++static __always_inline __must_check unsigned long ++__copy_from_user(void *to, const void __user *from, unsigned len); + __must_check unsigned long + copy_in_user(void __user *to, const void __user *from, unsigned len); + + static inline unsigned long __must_check copy_from_user(void *to, + const void __user *from, +- unsigned long n) ++ unsigned n) + { +- int sz = __compiletime_object_size(to); +- + might_fault(); +- if (likely(sz == -1 || sz >= n)) +- n = _copy_from_user(to, from, n); +-#ifdef CONFIG_DEBUG_VM +- else +- WARN(1, "Buffer overflow detected!\n"); +-#endif ++ ++ if (access_ok(VERIFY_READ, from, n)) ++ n = __copy_from_user(to, from, n); ++ else if ((int)n > 0) { ++ if (!__builtin_constant_p(n)) ++ check_object_size(to, n, false); ++ memset(to, 0, n); ++ } + return n; + } + +@@ -65,110 +68,174 @@ int copy_to_user(void __user *dst, const + { + might_fault(); + +- return _copy_to_user(dst, src, size); ++ if (access_ok(VERIFY_WRITE, dst, size)) ++ size = __copy_to_user(dst, src, size); ++ return size; + } + + static __always_inline __must_check +-int __copy_from_user(void *dst, const void __user *src, unsigned size) ++unsigned long __copy_from_user(void *dst, const void __user *src, unsigned size) + { +- int ret = 0; ++ int sz = __compiletime_object_size(dst); ++ unsigned ret = 0; + + might_fault(); +- if (!__builtin_constant_p(size)) +- return copy_user_generic(dst, (__force void *)src, size); ++ ++ if ((int)size < 0) ++ return size; ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ if (!__access_ok(VERIFY_READ, src, size)) ++ return size; ++#endif ++ ++ if (unlikely(sz != -1 && sz < size)) { ++#ifdef CONFIG_DEBUG_VM ++ WARN(1, "Buffer overflow detected!\n"); ++#endif ++ return size; ++ } ++ ++ if (!__builtin_constant_p(size)) { ++ check_object_size(dst, size, false); ++ if ((unsigned long)src < PAX_USER_SHADOW_BASE) ++ src += PAX_USER_SHADOW_BASE; ++ return copy_user_generic(dst, (__force const void *)src, size); ++ } + switch (size) { +- case 1:__get_user_asm(*(u8 *)dst, (u8 __user *)src, ++ case 1:__get_user_asm(*(u8 *)dst, (const u8 __user *)src, + ret, "b", "b", "=q", 1); + return ret; +- case 2:__get_user_asm(*(u16 *)dst, (u16 __user *)src, ++ case 2:__get_user_asm(*(u16 *)dst, (const u16 __user *)src, + ret, "w", "w", "=r", 2); + return ret; +- case 4:__get_user_asm(*(u32 *)dst, (u32 __user *)src, ++ case 4:__get_user_asm(*(u32 *)dst, (const u32 __user *)src, + ret, "l", "k", "=r", 4); + return ret; +- case 8:__get_user_asm(*(u64 *)dst, (u64 __user *)src, ++ case 8:__get_user_asm(*(u64 *)dst, (const u64 __user *)src, + ret, "q", "", "=r", 8); + return ret; + case 10: +- __get_user_asm(*(u64 *)dst, (u64 __user *)src, ++ __get_user_asm(*(u64 *)dst, (const u64 __user *)src, + ret, "q", "", "=r", 10); + if (unlikely(ret)) + return ret; + __get_user_asm(*(u16 *)(8 + (char *)dst), +- (u16 __user *)(8 + (char __user *)src), ++ (const u16 __user *)(8 + (const char __user *)src), + ret, "w", "w", "=r", 2); + return ret; + case 16: +- __get_user_asm(*(u64 *)dst, (u64 __user *)src, ++ __get_user_asm(*(u64 *)dst, (const u64 __user *)src, + ret, "q", "", "=r", 16); + if (unlikely(ret)) + return ret; + __get_user_asm(*(u64 *)(8 + (char *)dst), +- (u64 __user *)(8 + (char __user *)src), ++ (const u64 __user *)(8 + (const char __user *)src), + ret, "q", "", "=r", 8); + return ret; + default: +- return copy_user_generic(dst, (__force void *)src, size); ++ if ((unsigned long)src < PAX_USER_SHADOW_BASE) ++ src += PAX_USER_SHADOW_BASE; ++ return copy_user_generic(dst, (__force const void *)src, size); + } + } + + static __always_inline __must_check +-int __copy_to_user(void __user *dst, const void *src, unsigned size) ++unsigned long __copy_to_user(void __user *dst, const void *src, unsigned size) + { +- int ret = 0; ++ int sz = __compiletime_object_size(src); ++ unsigned ret = 0; + + might_fault(); +- if (!__builtin_constant_p(size)) ++ ++ if ((int)size < 0) ++ return size; ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ if (!__access_ok(VERIFY_WRITE, dst, size)) ++ return size; ++#endif ++ ++ if (unlikely(sz != -1 && sz < size)) { ++#ifdef CONFIG_DEBUG_VM ++ WARN(1, "Buffer overflow detected!\n"); ++#endif ++ return size; ++ } ++ ++ if (!__builtin_constant_p(size)) { ++ check_object_size(src, size, true); ++ if ((unsigned long)dst < PAX_USER_SHADOW_BASE) ++ dst += PAX_USER_SHADOW_BASE; + return copy_user_generic((__force void *)dst, src, size); ++ } + switch (size) { +- case 1:__put_user_asm(*(u8 *)src, (u8 __user *)dst, ++ case 1:__put_user_asm(*(const u8 *)src, (u8 __user *)dst, + ret, "b", "b", "iq", 1); + return ret; +- case 2:__put_user_asm(*(u16 *)src, (u16 __user *)dst, ++ case 2:__put_user_asm(*(const u16 *)src, (u16 __user *)dst, + ret, "w", "w", "ir", 2); + return ret; +- case 4:__put_user_asm(*(u32 *)src, (u32 __user *)dst, ++ case 4:__put_user_asm(*(const u32 *)src, (u32 __user *)dst, + ret, "l", "k", "ir", 4); + return ret; +- case 8:__put_user_asm(*(u64 *)src, (u64 __user *)dst, ++ case 8:__put_user_asm(*(const u64 *)src, (u64 __user *)dst, + ret, "q", "", "er", 8); + return ret; + case 10: +- __put_user_asm(*(u64 *)src, (u64 __user *)dst, ++ __put_user_asm(*(const u64 *)src, (u64 __user *)dst, + ret, "q", "", "er", 10); + if (unlikely(ret)) + return ret; + asm("":::"memory"); +- __put_user_asm(4[(u16 *)src], 4 + (u16 __user *)dst, ++ __put_user_asm(4[(const u16 *)src], 4 + (u16 __user *)dst, + ret, "w", "w", "ir", 2); + return ret; + case 16: +- __put_user_asm(*(u64 *)src, (u64 __user *)dst, ++ __put_user_asm(*(const u64 *)src, (u64 __user *)dst, + ret, "q", "", "er", 16); + if (unlikely(ret)) + return ret; + asm("":::"memory"); +- __put_user_asm(1[(u64 *)src], 1 + (u64 __user *)dst, ++ __put_user_asm(1[(const u64 *)src], 1 + (u64 __user *)dst, + ret, "q", "", "er", 8); + return ret; + default: ++ if ((unsigned long)dst < PAX_USER_SHADOW_BASE) ++ dst += PAX_USER_SHADOW_BASE; + return copy_user_generic((__force void *)dst, src, size); + } + } + + static __always_inline __must_check +-int __copy_in_user(void __user *dst, const void __user *src, unsigned size) ++unsigned long __copy_in_user(void __user *dst, const void __user *src, unsigned size) + { +- int ret = 0; ++ unsigned ret = 0; + + might_fault(); +- if (!__builtin_constant_p(size)) ++ ++ if ((int)size < 0) ++ return size; ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ if (!__access_ok(VERIFY_READ, src, size)) ++ return size; ++ if (!__access_ok(VERIFY_WRITE, dst, size)) ++ return size; ++#endif ++ ++ if (!__builtin_constant_p(size)) { ++ if ((unsigned long)src < PAX_USER_SHADOW_BASE) ++ src += PAX_USER_SHADOW_BASE; ++ if ((unsigned long)dst < PAX_USER_SHADOW_BASE) ++ dst += PAX_USER_SHADOW_BASE; + return copy_user_generic((__force void *)dst, +- (__force void *)src, size); ++ (__force const void *)src, size); ++ } + switch (size) { + case 1: { + u8 tmp; +- __get_user_asm(tmp, (u8 __user *)src, ++ __get_user_asm(tmp, (const u8 __user *)src, + ret, "b", "b", "=q", 1); + if (likely(!ret)) + __put_user_asm(tmp, (u8 __user *)dst, +@@ -177,7 +244,7 @@ int __copy_in_user(void __user *dst, con + } + case 2: { + u16 tmp; +- __get_user_asm(tmp, (u16 __user *)src, ++ __get_user_asm(tmp, (const u16 __user *)src, + ret, "w", "w", "=r", 2); + if (likely(!ret)) + __put_user_asm(tmp, (u16 __user *)dst, +@@ -187,7 +254,7 @@ int __copy_in_user(void __user *dst, con + + case 4: { + u32 tmp; +- __get_user_asm(tmp, (u32 __user *)src, ++ __get_user_asm(tmp, (const u32 __user *)src, + ret, "l", "k", "=r", 4); + if (likely(!ret)) + __put_user_asm(tmp, (u32 __user *)dst, +@@ -196,7 +263,7 @@ int __copy_in_user(void __user *dst, con + } + case 8: { + u64 tmp; +- __get_user_asm(tmp, (u64 __user *)src, ++ __get_user_asm(tmp, (const u64 __user *)src, + ret, "q", "", "=r", 8); + if (likely(!ret)) + __put_user_asm(tmp, (u64 __user *)dst, +@@ -204,8 +271,12 @@ int __copy_in_user(void __user *dst, con + return ret; + } + default: ++ if ((unsigned long)src < PAX_USER_SHADOW_BASE) ++ src += PAX_USER_SHADOW_BASE; ++ if ((unsigned long)dst < PAX_USER_SHADOW_BASE) ++ dst += PAX_USER_SHADOW_BASE; + return copy_user_generic((__force void *)dst, +- (__force void *)src, size); ++ (__force const void *)src, size); + } + } + +@@ -222,33 +293,68 @@ __must_check unsigned long __clear_user( + static __must_check __always_inline int + __copy_from_user_inatomic(void *dst, const void __user *src, unsigned size) + { ++ if ((int)size < 0) ++ return size; ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ if (!__access_ok(VERIFY_READ, src, size)) ++ return size; ++#endif ++ ++ if ((unsigned long)src < PAX_USER_SHADOW_BASE) ++ src += PAX_USER_SHADOW_BASE; + return copy_user_generic(dst, (__force const void *)src, size); + } + +-static __must_check __always_inline int ++static __must_check __always_inline unsigned long + __copy_to_user_inatomic(void __user *dst, const void *src, unsigned size) + { ++ if ((int)size < 0) ++ return size; ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ if (!__access_ok(VERIFY_WRITE, dst, size)) ++ return size; ++#endif ++ ++ if ((unsigned long)dst < PAX_USER_SHADOW_BASE) ++ dst += PAX_USER_SHADOW_BASE; + return copy_user_generic((__force void *)dst, src, size); + } + +-extern long __copy_user_nocache(void *dst, const void __user *src, ++extern unsigned long __copy_user_nocache(void *dst, const void __user *src, + unsigned size, int zerorest); + +-static inline int +-__copy_from_user_nocache(void *dst, const void __user *src, unsigned size) ++static inline unsigned long __copy_from_user_nocache(void *dst, const void __user *src, unsigned size) + { + might_sleep(); ++ ++ if ((int)size < 0) ++ return size; ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ if (!__access_ok(VERIFY_READ, src, size)) ++ return size; ++#endif ++ + return __copy_user_nocache(dst, src, size, 1); + } + +-static inline int +-__copy_from_user_inatomic_nocache(void *dst, const void __user *src, ++static inline unsigned long __copy_from_user_inatomic_nocache(void *dst, const void __user *src, + unsigned size) + { ++ if ((int)size < 0) ++ return size; ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ if (!__access_ok(VERIFY_READ, src, size)) ++ return size; ++#endif ++ + return __copy_user_nocache(dst, src, size, 0); + } + +-unsigned long ++extern unsigned long + copy_user_handle_tail(char *to, char *from, unsigned len, unsigned zerorest); + + #endif /* _ASM_X86_UACCESS_64_H */ +diff -urNp linux-2.6.38.3/arch/x86/include/asm/uaccess.h linux-2.6.38.3/arch/x86/include/asm/uaccess.h +--- linux-2.6.38.3/arch/x86/include/asm/uaccess.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/uaccess.h 2011-04-17 15:57:32.000000000 -0400 +@@ -8,12 +8,15 @@ + #include <linux/thread_info.h> + #include <linux/prefetch.h> + #include <linux/string.h> ++#include <linux/sched.h> + #include <asm/asm.h> + #include <asm/page.h> + + #define VERIFY_READ 0 + #define VERIFY_WRITE 1 + ++extern void check_object_size(const void *ptr, unsigned long n, bool to); ++ + /* + * The fs value determines whether argument validity checking should be + * performed or not. If get_fs() == USER_DS, checking is performed, with +@@ -29,7 +32,12 @@ + + #define get_ds() (KERNEL_DS) + #define get_fs() (current_thread_info()->addr_limit) ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_MEMORY_UDEREF) ++void __set_fs(mm_segment_t x); ++void set_fs(mm_segment_t x); ++#else + #define set_fs(x) (current_thread_info()->addr_limit = (x)) ++#endif + + #define segment_eq(a, b) ((a).seg == (b).seg) + +@@ -77,7 +85,33 @@ + * checks that the pointer is in the user space range - after calling + * this function, memory access functions may still return -EFAULT. + */ +-#define access_ok(type, addr, size) (likely(__range_not_ok(addr, size) == 0)) ++#define __access_ok(type, addr, size) (likely(__range_not_ok(addr, size) == 0)) ++#define access_ok(type, addr, size) \ ++({ \ ++ long __size = size; \ ++ unsigned long __addr = (unsigned long)addr; \ ++ unsigned long __addr_ao = __addr & PAGE_MASK; \ ++ unsigned long __end_ao = __addr + __size - 1; \ ++ bool __ret_ao = __range_not_ok(__addr, __size) == 0; \ ++ if (__ret_ao && unlikely((__end_ao ^ __addr_ao) & PAGE_MASK)) { \ ++ while(__addr_ao <= __end_ao) { \ ++ char __c_ao; \ ++ __addr_ao += PAGE_SIZE; \ ++ if (__size > PAGE_SIZE) \ ++ cond_resched(); \ ++ if (__get_user(__c_ao, (char __user *)__addr)) \ ++ break; \ ++ if (type != VERIFY_WRITE) { \ ++ __addr = __addr_ao; \ ++ continue; \ ++ } \ ++ if (__put_user(__c_ao, (char __user *)__addr)) \ ++ break; \ ++ __addr = __addr_ao; \ ++ } \ ++ } \ ++ __ret_ao; \ ++}) + + /* + * The exception table consists of pairs of addresses: the first is the +@@ -183,12 +217,20 @@ extern int __get_user_bad(void); + asm volatile("call __put_user_" #size : "=a" (__ret_pu) \ + : "0" ((typeof(*(ptr)))(x)), "c" (ptr) : "ebx") + +- ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_MEMORY_UDEREF) ++#define __copyuser_seg "gs;" ++#define __COPYUSER_SET_ES "pushl %%gs; popl %%es\n" ++#define __COPYUSER_RESTORE_ES "pushl %%ss; popl %%es\n" ++#else ++#define __copyuser_seg ++#define __COPYUSER_SET_ES ++#define __COPYUSER_RESTORE_ES ++#endif + + #ifdef CONFIG_X86_32 + #define __put_user_asm_u64(x, addr, err, errret) \ +- asm volatile("1: movl %%eax,0(%2)\n" \ +- "2: movl %%edx,4(%2)\n" \ ++ asm volatile("1: "__copyuser_seg"movl %%eax,0(%2)\n" \ ++ "2: "__copyuser_seg"movl %%edx,4(%2)\n" \ + "3:\n" \ + ".section .fixup,\"ax\"\n" \ + "4: movl %3,%0\n" \ +@@ -200,8 +242,8 @@ extern int __get_user_bad(void); + : "A" (x), "r" (addr), "i" (errret), "0" (err)) + + #define __put_user_asm_ex_u64(x, addr) \ +- asm volatile("1: movl %%eax,0(%1)\n" \ +- "2: movl %%edx,4(%1)\n" \ ++ asm volatile("1: "__copyuser_seg"movl %%eax,0(%1)\n" \ ++ "2: "__copyuser_seg"movl %%edx,4(%1)\n" \ + "3:\n" \ + _ASM_EXTABLE(1b, 2b - 1b) \ + _ASM_EXTABLE(2b, 3b - 2b) \ +@@ -374,7 +416,7 @@ do { \ + } while (0) + + #define __get_user_asm(x, addr, err, itype, rtype, ltype, errret) \ +- asm volatile("1: mov"itype" %2,%"rtype"1\n" \ ++ asm volatile("1: "__copyuser_seg"mov"itype" %2,%"rtype"1\n"\ + "2:\n" \ + ".section .fixup,\"ax\"\n" \ + "3: mov %3,%0\n" \ +@@ -382,7 +424,7 @@ do { \ + " jmp 2b\n" \ + ".previous\n" \ + _ASM_EXTABLE(1b, 3b) \ +- : "=r" (err), ltype(x) \ ++ : "=r" (err), ltype (x) \ + : "m" (__m(addr)), "i" (errret), "0" (err)) + + #define __get_user_size_ex(x, ptr, size) \ +@@ -407,7 +449,7 @@ do { \ + } while (0) + + #define __get_user_asm_ex(x, addr, itype, rtype, ltype) \ +- asm volatile("1: mov"itype" %1,%"rtype"0\n" \ ++ asm volatile("1: "__copyuser_seg"mov"itype" %1,%"rtype"0\n"\ + "2:\n" \ + _ASM_EXTABLE(1b, 2b - 1b) \ + : ltype(x) : "m" (__m(addr))) +@@ -424,13 +466,24 @@ do { \ + int __gu_err; \ + unsigned long __gu_val; \ + __get_user_size(__gu_val, (ptr), (size), __gu_err, -EFAULT); \ +- (x) = (__force __typeof__(*(ptr)))__gu_val; \ ++ (x) = (__typeof__(*(ptr)))__gu_val; \ + __gu_err; \ + }) + + /* FIXME: this hack is definitely wrong -AK */ + struct __large_struct { unsigned long buf[100]; }; +-#define __m(x) (*(struct __large_struct __user *)(x)) ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) ++#define ____m(x) \ ++({ \ ++ unsigned long ____x = (unsigned long)(x); \ ++ if (____x < PAX_USER_SHADOW_BASE) \ ++ ____x += PAX_USER_SHADOW_BASE; \ ++ (void __user *)____x; \ ++}) ++#else ++#define ____m(x) (x) ++#endif ++#define __m(x) (*(struct __large_struct __user *)____m(x)) + + /* + * Tell gcc we read from memory instead of writing: this is because +@@ -438,7 +491,7 @@ struct __large_struct { unsigned long bu + * aliasing issues. + */ + #define __put_user_asm(x, addr, err, itype, rtype, ltype, errret) \ +- asm volatile("1: mov"itype" %"rtype"1,%2\n" \ ++ asm volatile("1: "__copyuser_seg"mov"itype" %"rtype"1,%2\n"\ + "2:\n" \ + ".section .fixup,\"ax\"\n" \ + "3: mov %3,%0\n" \ +@@ -446,10 +499,10 @@ struct __large_struct { unsigned long bu + ".previous\n" \ + _ASM_EXTABLE(1b, 3b) \ + : "=r"(err) \ +- : ltype(x), "m" (__m(addr)), "i" (errret), "0" (err)) ++ : ltype (x), "m" (__m(addr)), "i" (errret), "0" (err)) + + #define __put_user_asm_ex(x, addr, itype, rtype, ltype) \ +- asm volatile("1: mov"itype" %"rtype"0,%1\n" \ ++ asm volatile("1: "__copyuser_seg"mov"itype" %"rtype"0,%1\n"\ + "2:\n" \ + _ASM_EXTABLE(1b, 2b - 1b) \ + : : ltype(x), "m" (__m(addr))) +@@ -488,8 +541,12 @@ struct __large_struct { unsigned long bu + * On error, the variable @x is set to zero. + */ + ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) ++#define __get_user(x, ptr) get_user((x), (ptr)) ++#else + #define __get_user(x, ptr) \ + __get_user_nocheck((x), (ptr), sizeof(*(ptr))) ++#endif + + /** + * __put_user: - Write a simple value into user space, with less checking. +@@ -511,8 +568,12 @@ struct __large_struct { unsigned long bu + * Returns zero on success, or -EFAULT on error. + */ + ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) ++#define __put_user(x, ptr) put_user((x), (ptr)) ++#else + #define __put_user(x, ptr) \ + __put_user_nocheck((__typeof__(*(ptr)))(x), (ptr), sizeof(*(ptr))) ++#endif + + #define __get_user_unaligned __get_user + #define __put_user_unaligned __put_user +@@ -530,7 +591,7 @@ struct __large_struct { unsigned long bu + #define get_user_ex(x, ptr) do { \ + unsigned long __gue_val; \ + __get_user_size_ex((__gue_val), (ptr), (sizeof(*(ptr)))); \ +- (x) = (__force __typeof__(*(ptr)))__gue_val; \ ++ (x) = (__typeof__(*(ptr)))__gue_val; \ + } while (0) + + #ifdef CONFIG_X86_WP_WORKS_OK +@@ -567,6 +628,7 @@ extern struct movsl_mask { + + #define ARCH_HAS_NOCACHE_UACCESS 1 + ++#define ARCH_HAS_SORT_EXTABLE + #ifdef CONFIG_X86_32 + # include "uaccess_32.h" + #else +diff -urNp linux-2.6.38.3/arch/x86/include/asm/vgtod.h linux-2.6.38.3/arch/x86/include/asm/vgtod.h +--- linux-2.6.38.3/arch/x86/include/asm/vgtod.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/vgtod.h 2011-04-17 15:57:32.000000000 -0400 +@@ -14,6 +14,7 @@ struct vsyscall_gtod_data { + int sysctl_enabled; + struct timezone sys_tz; + struct { /* extract of a clocksource struct */ ++ char name[8]; + cycle_t (*vread)(void); + cycle_t cycle_last; + cycle_t mask; +diff -urNp linux-2.6.38.3/arch/x86/include/asm/vsyscall.h linux-2.6.38.3/arch/x86/include/asm/vsyscall.h +--- linux-2.6.38.3/arch/x86/include/asm/vsyscall.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/vsyscall.h 2011-04-17 15:57:32.000000000 -0400 +@@ -15,9 +15,10 @@ enum vsyscall_num { + + #ifdef __KERNEL__ + #include <linux/seqlock.h> ++#include <linux/getcpu.h> ++#include <linux/time.h> + + #define __section_vgetcpu_mode __attribute__ ((unused, __section__ (".vgetcpu_mode"), aligned(16))) +-#define __section_jiffies __attribute__ ((unused, __section__ (".jiffies"), aligned(16))) + + /* Definitions for CONFIG_GENERIC_TIME definitions */ + #define __section_vsyscall_gtod_data __attribute__ \ +@@ -31,7 +32,6 @@ enum vsyscall_num { + #define VGETCPU_LSL 2 + + extern int __vgetcpu_mode; +-extern volatile unsigned long __jiffies; + + /* kernel space (writeable) */ + extern int vgetcpu_mode; +@@ -39,6 +39,9 @@ extern struct timezone sys_tz; + + extern void map_vsyscall(void); + ++extern int vgettimeofday(struct timeval * tv, struct timezone * tz); ++extern time_t vtime(time_t *t); ++extern long vgetcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *tcache); + #endif /* __KERNEL__ */ + + #endif /* _ASM_X86_VSYSCALL_H */ +diff -urNp linux-2.6.38.3/arch/x86/include/asm/xsave.h linux-2.6.38.3/arch/x86/include/asm/xsave.h +--- linux-2.6.38.3/arch/x86/include/asm/xsave.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/include/asm/xsave.h 2011-04-17 15:57:32.000000000 -0400 +@@ -65,6 +65,11 @@ static inline int xsave_user(struct xsav + { + int err; + ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) ++ if ((unsigned long)buf < PAX_USER_SHADOW_BASE) ++ buf = (struct xsave_struct __user *)((void __user*)buf + PAX_USER_SHADOW_BASE); ++#endif ++ + /* + * Clear the xsave header first, so that reserved fields are + * initialized to zero. +@@ -100,6 +105,11 @@ static inline int xrestore_user(struct x + u32 lmask = mask; + u32 hmask = mask >> 32; + ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) ++ if ((unsigned long)xstate < PAX_USER_SHADOW_BASE) ++ xstate = (struct xsave_struct *)((void *)xstate + PAX_USER_SHADOW_BASE); ++#endif ++ + __asm__ __volatile__("1: .byte " REX_PREFIX "0x0f,0xae,0x2f\n" + "2:\n" + ".section .fixup,\"ax\"\n" +diff -urNp linux-2.6.38.3/arch/x86/Kconfig linux-2.6.38.3/arch/x86/Kconfig +--- linux-2.6.38.3/arch/x86/Kconfig 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/Kconfig 2011-04-17 15:57:32.000000000 -0400 +@@ -223,7 +223,7 @@ config X86_TRAMPOLINE + + config X86_32_LAZY_GS + def_bool y +- depends on X86_32 && !CC_STACKPROTECTOR ++ depends on X86_32 && !CC_STACKPROTECTOR && !PAX_MEMORY_UDEREF + + config ARCH_HWEIGHT_CFLAGS + string +@@ -1019,7 +1019,7 @@ choice + + config NOHIGHMEM + bool "off" +- depends on !X86_NUMAQ ++ depends on !X86_NUMAQ && !(PAX_PAGEEXEC && PAX_ENABLE_PAE) + ---help--- + Linux can use up to 64 Gigabytes of physical memory on x86 systems. + However, the address space of 32-bit x86 processors is only 4 +@@ -1056,7 +1056,7 @@ config NOHIGHMEM + + config HIGHMEM4G + bool "4GB" +- depends on !X86_NUMAQ ++ depends on !X86_NUMAQ && !(PAX_PAGEEXEC && PAX_ENABLE_PAE) + ---help--- + Select this if you have a 32-bit processor and between 1 and 4 + gigabytes of physical RAM. +@@ -1110,7 +1110,7 @@ config PAGE_OFFSET + hex + default 0xB0000000 if VMSPLIT_3G_OPT + default 0x80000000 if VMSPLIT_2G +- default 0x78000000 if VMSPLIT_2G_OPT ++ default 0x70000000 if VMSPLIT_2G_OPT + default 0x40000000 if VMSPLIT_1G + default 0xC0000000 + depends on X86_32 +@@ -1454,7 +1454,7 @@ config ARCH_USES_PG_UNCACHED + + config EFI + bool "EFI runtime service support" +- depends on ACPI ++ depends on ACPI && !PAX_KERNEXEC + ---help--- + This enables the kernel to use EFI runtime services that are + available (such as the EFI variable services). +@@ -1484,6 +1484,7 @@ config SECCOMP + + config CC_STACKPROTECTOR + bool "Enable -fstack-protector buffer overflow detection (EXPERIMENTAL)" ++ depends on X86_64 || !PAX_MEMORY_UDEREF + ---help--- + This option turns on the -fstack-protector GCC feature. This + feature puts, at the beginning of functions, a canary value on +@@ -1541,6 +1542,7 @@ config KEXEC_JUMP + config PHYSICAL_START + hex "Physical address where the kernel is loaded" if (EXPERT || CRASH_DUMP) + default "0x1000000" ++ range 0x400000 0x40000000 + ---help--- + This gives the physical address where the kernel is loaded. + +@@ -1604,6 +1606,7 @@ config X86_NEED_RELOCS + config PHYSICAL_ALIGN + hex "Alignment value to which kernel should be aligned" if X86_32 + default "0x1000000" ++ range 0x400000 0x1000000 if PAX_KERNEXEC + range 0x2000 0x1000000 + ---help--- + This value puts the alignment restrictions on physical address +@@ -1635,9 +1638,10 @@ config HOTPLUG_CPU + Say N if you want to disable CPU hotplug. + + config COMPAT_VDSO +- def_bool y ++ def_bool n + prompt "Compat VDSO support" + depends on X86_32 || IA32_EMULATION ++ depends on !PAX_NOEXEC && !PAX_MEMORY_UDEREF + ---help--- + Map the 32-bit VDSO to the predictable old-style address too. + +diff -urNp linux-2.6.38.3/arch/x86/Kconfig.cpu linux-2.6.38.3/arch/x86/Kconfig.cpu +--- linux-2.6.38.3/arch/x86/Kconfig.cpu 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/Kconfig.cpu 2011-04-17 15:57:32.000000000 -0400 +@@ -339,7 +339,7 @@ config X86_PPRO_FENCE + + config X86_F00F_BUG + def_bool y +- depends on M586MMX || M586TSC || M586 || M486 || M386 ++ depends on (M586MMX || M586TSC || M586 || M486 || M386) && !PAX_KERNEXEC + + config X86_INVD_BUG + def_bool y +@@ -363,7 +363,7 @@ config X86_POPAD_OK + + config X86_ALIGNMENT_16 + def_bool y +- depends on MWINCHIP3D || MWINCHIPC6 || MCYRIXIII || X86_ELAN || MK6 || M586MMX || M586TSC || M586 || M486 || MVIAC3_2 || MGEODEGX1 ++ depends on MWINCHIP3D || MWINCHIPC6 || MCYRIXIII || X86_ELAN || MK8 || MK7 || MK6 || MCORE2 || MPENTIUM4 || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || M586 || M486 || MVIAC3_2 || MGEODEGX1 + + config X86_INTEL_USERCOPY + def_bool y +@@ -409,7 +409,7 @@ config X86_CMPXCHG64 + # generates cmov. + config X86_CMOV + def_bool y +- depends on (MK8 || MK7 || MCORE2 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MVIAC3_2 || MVIAC7 || MCRUSOE || MEFFICEON || X86_64 || MATOM || MGEODE_LX) ++ depends on (MK8 || MK7 || MCORE2 || MPSC || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MVIAC3_2 || MVIAC7 || MCRUSOE || MEFFICEON || X86_64 || MATOM || MGEODE_LX) + + config X86_MINIMUM_CPU_FAMILY + int +diff -urNp linux-2.6.38.3/arch/x86/Kconfig.debug linux-2.6.38.3/arch/x86/Kconfig.debug +--- linux-2.6.38.3/arch/x86/Kconfig.debug 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/Kconfig.debug 2011-04-17 15:57:32.000000000 -0400 +@@ -101,7 +101,7 @@ config X86_PTDUMP + config DEBUG_RODATA + bool "Write protect kernel read-only data structures" + default y +- depends on DEBUG_KERNEL ++ depends on DEBUG_KERNEL && BROKEN + ---help--- + Mark the kernel read-only data as write-protected in the pagetables, + in order to catch accidental (and incorrect) writes to such const +@@ -119,7 +119,7 @@ config DEBUG_RODATA_TEST + + config DEBUG_SET_MODULE_RONX + bool "Set loadable kernel module data as NX and text as RO" +- depends on MODULES ++ depends on MODULES && BROKEN + ---help--- + This option helps catch unintended modifications to loadable + kernel module's text and read-only data. It also prevents execution +diff -urNp linux-2.6.38.3/arch/x86/kernel/acpi/boot.c linux-2.6.38.3/arch/x86/kernel/acpi/boot.c +--- linux-2.6.38.3/arch/x86/kernel/acpi/boot.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/acpi/boot.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1472,7 +1472,7 @@ static struct dmi_system_id __initdata a + DMI_MATCH(DMI_PRODUCT_NAME, "HP Compaq 6715b"), + }, + }, +- {} ++ { NULL, NULL, {{0, {0}}}, NULL} + }; + + /* +diff -urNp linux-2.6.38.3/arch/x86/kernel/acpi/sleep.c linux-2.6.38.3/arch/x86/kernel/acpi/sleep.c +--- linux-2.6.38.3/arch/x86/kernel/acpi/sleep.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/acpi/sleep.c 2011-04-17 15:57:32.000000000 -0400 +@@ -18,7 +18,7 @@ + #include "realmode/wakeup.h" + #include "sleep.h" + +-unsigned long acpi_wakeup_address; ++unsigned long acpi_wakeup_address = 0x2000; + unsigned long acpi_realmode_flags; + + /* address in low memory of the wakeup routine. */ +@@ -99,8 +99,12 @@ int acpi_save_state_mem(void) + header->trampoline_segment = setup_trampoline() >> 4; + #ifdef CONFIG_SMP + stack_start = (unsigned long)temp_stack + sizeof(temp_stack); ++ ++ pax_open_kernel(); + early_gdt_descr.address = + (unsigned long)get_cpu_gdt_table(smp_processor_id()); ++ pax_close_kernel(); ++ + initial_gs = per_cpu_offset(smp_processor_id()); + #endif + initial_code = (unsigned long)wakeup_long64; +diff -urNp linux-2.6.38.3/arch/x86/kernel/acpi/wakeup_32.S linux-2.6.38.3/arch/x86/kernel/acpi/wakeup_32.S +--- linux-2.6.38.3/arch/x86/kernel/acpi/wakeup_32.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/acpi/wakeup_32.S 2011-04-17 15:57:32.000000000 -0400 +@@ -30,13 +30,11 @@ wakeup_pmode_return: + # and restore the stack ... but you need gdt for this to work + movl saved_context_esp, %esp + +- movl %cs:saved_magic, %eax +- cmpl $0x12345678, %eax ++ cmpl $0x12345678, saved_magic + jne bogus_magic + + # jump to place where we left off +- movl saved_eip, %eax +- jmp *%eax ++ jmp *(saved_eip) + + bogus_magic: + jmp bogus_magic +diff -urNp linux-2.6.38.3/arch/x86/kernel/alternative.c linux-2.6.38.3/arch/x86/kernel/alternative.c +--- linux-2.6.38.3/arch/x86/kernel/alternative.c 2011-03-23 17:20:06.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/alternative.c 2011-04-17 15:57:32.000000000 -0400 +@@ -248,7 +248,7 @@ static void alternatives_smp_lock(const + if (!*poff || ptr < text || ptr >= text_end) + continue; + /* turn DS segment override prefix into lock prefix */ +- if (*ptr == 0x3e) ++ if (*ktla_ktva(ptr) == 0x3e) + text_poke(ptr, ((unsigned char []){0xf0}), 1); + }; + mutex_unlock(&text_mutex); +@@ -269,7 +269,7 @@ static void alternatives_smp_unlock(cons + if (!*poff || ptr < text || ptr >= text_end) + continue; + /* turn lock prefix into DS segment override prefix */ +- if (*ptr == 0xf0) ++ if (*ktla_ktva(ptr) == 0xf0) + text_poke(ptr, ((unsigned char []){0x3E}), 1); + }; + mutex_unlock(&text_mutex); +@@ -438,7 +438,7 @@ void __init_or_module apply_paravirt(str + + BUG_ON(p->len > MAX_PATCH_LEN); + /* prep the buffer with the original instructions */ +- memcpy(insnbuf, p->instr, p->len); ++ memcpy(insnbuf, ktla_ktva(p->instr), p->len); + used = pv_init_ops.patch(p->instrtype, p->clobbers, insnbuf, + (unsigned long)p->instr, p->len); + +@@ -506,7 +506,7 @@ void __init alternative_instructions(voi + if (smp_alt_once) + free_init_pages("SMP alternatives", + (unsigned long)__smp_locks, +- (unsigned long)__smp_locks_end); ++ PAGE_ALIGN((unsigned long)__smp_locks_end)); + + restart_nmi(); + } +@@ -523,13 +523,17 @@ void __init alternative_instructions(voi + * instructions. And on the local CPU you need to be protected again NMI or MCE + * handlers seeing an inconsistent instruction while you patch. + */ +-void *__init_or_module text_poke_early(void *addr, const void *opcode, ++void *__kprobes text_poke_early(void *addr, const void *opcode, + size_t len) + { + unsigned long flags; + local_irq_save(flags); +- memcpy(addr, opcode, len); ++ ++ pax_open_kernel(); ++ memcpy(ktla_ktva(addr), opcode, len); + sync_core(); ++ pax_close_kernel(); ++ + local_irq_restore(flags); + /* Could also do a CLFLUSH here to speed up CPU recovery; but + that causes hangs on some VIA CPUs. */ +@@ -551,36 +555,22 @@ void *__init_or_module text_poke_early(v + */ + void *__kprobes text_poke(void *addr, const void *opcode, size_t len) + { +- unsigned long flags; +- char *vaddr; ++ unsigned char *vaddr = ktla_ktva(addr); + struct page *pages[2]; +- int i; ++ size_t i; + + if (!core_kernel_text((unsigned long)addr)) { +- pages[0] = vmalloc_to_page(addr); +- pages[1] = vmalloc_to_page(addr + PAGE_SIZE); ++ pages[0] = vmalloc_to_page(vaddr); ++ pages[1] = vmalloc_to_page(vaddr + PAGE_SIZE); + } else { +- pages[0] = virt_to_page(addr); ++ pages[0] = virt_to_page(vaddr); + WARN_ON(!PageReserved(pages[0])); +- pages[1] = virt_to_page(addr + PAGE_SIZE); ++ pages[1] = virt_to_page(vaddr + PAGE_SIZE); + } + BUG_ON(!pages[0]); +- local_irq_save(flags); +- set_fixmap(FIX_TEXT_POKE0, page_to_phys(pages[0])); +- if (pages[1]) +- set_fixmap(FIX_TEXT_POKE1, page_to_phys(pages[1])); +- vaddr = (char *)fix_to_virt(FIX_TEXT_POKE0); +- memcpy(&vaddr[(unsigned long)addr & ~PAGE_MASK], opcode, len); +- clear_fixmap(FIX_TEXT_POKE0); +- if (pages[1]) +- clear_fixmap(FIX_TEXT_POKE1); +- local_flush_tlb(); +- sync_core(); +- /* Could also do a CLFLUSH here to speed up CPU recovery; but +- that causes hangs on some VIA CPUs. */ ++ text_poke_early(addr, opcode, len); + for (i = 0; i < len; i++) +- BUG_ON(((char *)addr)[i] != ((char *)opcode)[i]); +- local_irq_restore(flags); ++ BUG_ON((vaddr)[i] != ((const unsigned char *)opcode)[i]); + return addr; + } + +@@ -682,9 +672,9 @@ void __kprobes text_poke_smp_batch(struc + #if defined(CONFIG_DYNAMIC_FTRACE) || defined(HAVE_JUMP_LABEL) + + #ifdef CONFIG_X86_64 +-unsigned char ideal_nop5[5] = { 0x66, 0x66, 0x66, 0x66, 0x90 }; ++unsigned char ideal_nop5[5] __read_only = { 0x66, 0x66, 0x66, 0x66, 0x90 }; + #else +-unsigned char ideal_nop5[5] = { 0x3e, 0x8d, 0x74, 0x26, 0x00 }; ++unsigned char ideal_nop5[5] __read_only = { 0x3e, 0x8d, 0x74, 0x26, 0x00 }; + #endif + + void __init arch_init_ideal_nop5(void) +diff -urNp linux-2.6.38.3/arch/x86/kernel/amd_iommu.c linux-2.6.38.3/arch/x86/kernel/amd_iommu.c +--- linux-2.6.38.3/arch/x86/kernel/amd_iommu.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/amd_iommu.c 2011-04-17 15:57:32.000000000 -0400 +@@ -2286,7 +2286,7 @@ static void prealloc_protection_domains( + } + } + +-static struct dma_map_ops amd_iommu_dma_ops = { ++static const struct dma_map_ops amd_iommu_dma_ops = { + .alloc_coherent = alloc_coherent, + .free_coherent = free_coherent, + .map_page = map_page, +diff -urNp linux-2.6.38.3/arch/x86/kernel/apic/io_apic.c linux-2.6.38.3/arch/x86/kernel/apic/io_apic.c +--- linux-2.6.38.3/arch/x86/kernel/apic/io_apic.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/apic/io_apic.c 2011-04-17 15:57:32.000000000 -0400 +@@ -617,7 +617,7 @@ struct IO_APIC_route_entry **alloc_ioapi + ioapic_entries = kzalloc(sizeof(*ioapic_entries) * nr_ioapics, + GFP_KERNEL); + if (!ioapic_entries) +- return 0; ++ return NULL; + + for (apic = 0; apic < nr_ioapics; apic++) { + ioapic_entries[apic] = +@@ -634,7 +634,7 @@ nomem: + kfree(ioapic_entries[apic]); + kfree(ioapic_entries); + +- return 0; ++ return NULL; + } + + /* +@@ -1044,7 +1044,7 @@ int IO_APIC_get_PCI_irq_vector(int bus, + } + EXPORT_SYMBOL(IO_APIC_get_PCI_irq_vector); + +-void lock_vector_lock(void) ++void lock_vector_lock(void) __acquires(vector_lock) + { + /* Used to the online set of cpus does not change + * during assign_irq_vector. +@@ -1052,7 +1052,7 @@ void lock_vector_lock(void) + raw_spin_lock(&vector_lock); + } + +-void unlock_vector_lock(void) ++void unlock_vector_lock(void) __releases(vector_lock) + { + raw_spin_unlock(&vector_lock); + } +diff -urNp linux-2.6.38.3/arch/x86/kernel/apm_32.c linux-2.6.38.3/arch/x86/kernel/apm_32.c +--- linux-2.6.38.3/arch/x86/kernel/apm_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/apm_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -410,7 +410,7 @@ static DEFINE_MUTEX(apm_mutex); + * This is for buggy BIOS's that refer to (real mode) segment 0x40 + * even though they are called in protected mode. + */ +-static struct desc_struct bad_bios_desc = GDT_ENTRY_INIT(0x4092, ++static const struct desc_struct bad_bios_desc = GDT_ENTRY_INIT(0x4093, + (unsigned long)__va(0x400UL), PAGE_SIZE - 0x400 - 1); + + static const char driver_version[] = "1.16ac"; /* no spaces */ +@@ -588,7 +588,10 @@ static long __apm_bios_call(void *_call) + BUG_ON(cpu != 0); + gdt = get_cpu_gdt_table(cpu); + save_desc_40 = gdt[0x40 / 8]; ++ ++ pax_open_kernel(); + gdt[0x40 / 8] = bad_bios_desc; ++ pax_close_kernel(); + + apm_irq_save(flags); + APM_DO_SAVE_SEGS; +@@ -597,7 +600,11 @@ static long __apm_bios_call(void *_call) + &call->esi); + APM_DO_RESTORE_SEGS; + apm_irq_restore(flags); ++ ++ pax_open_kernel(); + gdt[0x40 / 8] = save_desc_40; ++ pax_close_kernel(); ++ + put_cpu(); + + return call->eax & 0xff; +@@ -664,7 +671,10 @@ static long __apm_bios_call_simple(void + BUG_ON(cpu != 0); + gdt = get_cpu_gdt_table(cpu); + save_desc_40 = gdt[0x40 / 8]; ++ ++ pax_open_kernel(); + gdt[0x40 / 8] = bad_bios_desc; ++ pax_close_kernel(); + + apm_irq_save(flags); + APM_DO_SAVE_SEGS; +@@ -672,7 +682,11 @@ static long __apm_bios_call_simple(void + &call->eax); + APM_DO_RESTORE_SEGS; + apm_irq_restore(flags); ++ ++ pax_open_kernel(); + gdt[0x40 / 8] = save_desc_40; ++ pax_close_kernel(); ++ + put_cpu(); + return error; + } +@@ -975,7 +989,7 @@ recalc: + + static void apm_power_off(void) + { +- unsigned char po_bios_call[] = { ++ const unsigned char po_bios_call[] = { + 0xb8, 0x00, 0x10, /* movw $0x1000,ax */ + 0x8e, 0xd0, /* movw ax,ss */ + 0xbc, 0x00, 0xf0, /* movw $0xf000,sp */ +@@ -1932,7 +1946,10 @@ static const struct file_operations apm_ + static struct miscdevice apm_device = { + APM_MINOR_DEV, + "apm_bios", +- &apm_bios_fops ++ &apm_bios_fops, ++ {NULL, NULL}, ++ NULL, ++ NULL + }; + + +@@ -2253,7 +2270,7 @@ static struct dmi_system_id __initdata a + { DMI_MATCH(DMI_SYS_VENDOR, "IBM"), }, + }, + +- { } ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL} + }; + + /* +@@ -2356,12 +2373,15 @@ static int __init apm_init(void) + * code to that CPU. + */ + gdt = get_cpu_gdt_table(0); ++ ++ pax_open_kernel(); + set_desc_base(&gdt[APM_CS >> 3], + (unsigned long)__va((unsigned long)apm_info.bios.cseg << 4)); + set_desc_base(&gdt[APM_CS_16 >> 3], + (unsigned long)__va((unsigned long)apm_info.bios.cseg_16 << 4)); + set_desc_base(&gdt[APM_DS >> 3], + (unsigned long)__va((unsigned long)apm_info.bios.dseg << 4)); ++ pax_close_kernel(); + + proc_create("apm", 0, NULL, &apm_file_ops); + +diff -urNp linux-2.6.38.3/arch/x86/kernel/asm-offsets_32.c linux-2.6.38.3/arch/x86/kernel/asm-offsets_32.c +--- linux-2.6.38.3/arch/x86/kernel/asm-offsets_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/asm-offsets_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -51,7 +51,6 @@ void foo(void) + OFFSET(CPUINFO_x86_vendor_id, cpuinfo_x86, x86_vendor_id); + BLANK(); + +- OFFSET(TI_task, thread_info, task); + OFFSET(TI_exec_domain, thread_info, exec_domain); + OFFSET(TI_flags, thread_info, flags); + OFFSET(TI_status, thread_info, status); +@@ -113,6 +112,11 @@ void foo(void) + OFFSET(PV_CPU_iret, pv_cpu_ops, iret); + OFFSET(PV_CPU_irq_enable_sysexit, pv_cpu_ops, irq_enable_sysexit); + OFFSET(PV_CPU_read_cr0, pv_cpu_ops, read_cr0); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ OFFSET(PV_CPU_write_cr0, pv_cpu_ops, write_cr0); ++#endif ++ + #endif + + #ifdef CONFIG_XEN +diff -urNp linux-2.6.38.3/arch/x86/kernel/asm-offsets_64.c linux-2.6.38.3/arch/x86/kernel/asm-offsets_64.c +--- linux-2.6.38.3/arch/x86/kernel/asm-offsets_64.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/asm-offsets_64.c 2011-04-17 15:57:32.000000000 -0400 +@@ -63,6 +63,18 @@ int main(void) + OFFSET(PV_CPU_irq_enable_sysexit, pv_cpu_ops, irq_enable_sysexit); + OFFSET(PV_CPU_swapgs, pv_cpu_ops, swapgs); + OFFSET(PV_MMU_read_cr2, pv_mmu_ops, read_cr2); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ OFFSET(PV_CPU_read_cr0, pv_cpu_ops, read_cr0); ++ OFFSET(PV_CPU_write_cr0, pv_cpu_ops, write_cr0); ++#endif ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ OFFSET(PV_MMU_read_cr3, pv_mmu_ops, read_cr3); ++ OFFSET(PV_MMU_write_cr3, pv_mmu_ops, write_cr3); ++ OFFSET(PV_MMU_set_pgd, pv_mmu_ops, set_pgd); ++#endif ++ + #endif + + +@@ -115,6 +127,7 @@ int main(void) + ENTRY(cr8); + BLANK(); + #undef ENTRY ++ DEFINE(TSS_size, sizeof(struct tss_struct)); + DEFINE(TSS_ist, offsetof(struct tss_struct, x86_tss.ist)); + BLANK(); + DEFINE(crypto_tfm_ctx_offset, offsetof(struct crypto_tfm, __crt_ctx)); +diff -urNp linux-2.6.38.3/arch/x86/kernel/cpu/amd.c linux-2.6.38.3/arch/x86/kernel/cpu/amd.c +--- linux-2.6.38.3/arch/x86/kernel/cpu/amd.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/cpu/amd.c 2011-04-17 15:57:32.000000000 -0400 +@@ -601,7 +601,7 @@ static unsigned int __cpuinit amd_size_c + unsigned int size) + { + /* AMD errata T13 (order #21922) */ +- if ((c->x86 == 6)) { ++ if (c->x86 == 6) { + /* Duron Rev A0 */ + if (c->x86_model == 3 && c->x86_mask == 0) + size = 64; +diff -urNp linux-2.6.38.3/arch/x86/kernel/cpu/common.c linux-2.6.38.3/arch/x86/kernel/cpu/common.c +--- linux-2.6.38.3/arch/x86/kernel/cpu/common.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/cpu/common.c 2011-04-17 15:57:32.000000000 -0400 +@@ -83,60 +83,6 @@ static const struct cpu_dev __cpuinitcon + + static const struct cpu_dev *this_cpu __cpuinitdata = &default_cpu; + +-DEFINE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page) = { .gdt = { +-#ifdef CONFIG_X86_64 +- /* +- * We need valid kernel segments for data and code in long mode too +- * IRET will check the segment types kkeil 2000/10/28 +- * Also sysret mandates a special GDT layout +- * +- * TLS descriptors are currently at a different place compared to i386. +- * Hopefully nobody expects them at a fixed place (Wine?) +- */ +- [GDT_ENTRY_KERNEL32_CS] = GDT_ENTRY_INIT(0xc09b, 0, 0xfffff), +- [GDT_ENTRY_KERNEL_CS] = GDT_ENTRY_INIT(0xa09b, 0, 0xfffff), +- [GDT_ENTRY_KERNEL_DS] = GDT_ENTRY_INIT(0xc093, 0, 0xfffff), +- [GDT_ENTRY_DEFAULT_USER32_CS] = GDT_ENTRY_INIT(0xc0fb, 0, 0xfffff), +- [GDT_ENTRY_DEFAULT_USER_DS] = GDT_ENTRY_INIT(0xc0f3, 0, 0xfffff), +- [GDT_ENTRY_DEFAULT_USER_CS] = GDT_ENTRY_INIT(0xa0fb, 0, 0xfffff), +-#else +- [GDT_ENTRY_KERNEL_CS] = GDT_ENTRY_INIT(0xc09a, 0, 0xfffff), +- [GDT_ENTRY_KERNEL_DS] = GDT_ENTRY_INIT(0xc092, 0, 0xfffff), +- [GDT_ENTRY_DEFAULT_USER_CS] = GDT_ENTRY_INIT(0xc0fa, 0, 0xfffff), +- [GDT_ENTRY_DEFAULT_USER_DS] = GDT_ENTRY_INIT(0xc0f2, 0, 0xfffff), +- /* +- * Segments used for calling PnP BIOS have byte granularity. +- * They code segments and data segments have fixed 64k limits, +- * the transfer segment sizes are set at run time. +- */ +- /* 32-bit code */ +- [GDT_ENTRY_PNPBIOS_CS32] = GDT_ENTRY_INIT(0x409a, 0, 0xffff), +- /* 16-bit code */ +- [GDT_ENTRY_PNPBIOS_CS16] = GDT_ENTRY_INIT(0x009a, 0, 0xffff), +- /* 16-bit data */ +- [GDT_ENTRY_PNPBIOS_DS] = GDT_ENTRY_INIT(0x0092, 0, 0xffff), +- /* 16-bit data */ +- [GDT_ENTRY_PNPBIOS_TS1] = GDT_ENTRY_INIT(0x0092, 0, 0), +- /* 16-bit data */ +- [GDT_ENTRY_PNPBIOS_TS2] = GDT_ENTRY_INIT(0x0092, 0, 0), +- /* +- * The APM segments have byte granularity and their bases +- * are set at run time. All have 64k limits. +- */ +- /* 32-bit code */ +- [GDT_ENTRY_APMBIOS_BASE] = GDT_ENTRY_INIT(0x409a, 0, 0xffff), +- /* 16-bit code */ +- [GDT_ENTRY_APMBIOS_BASE+1] = GDT_ENTRY_INIT(0x009a, 0, 0xffff), +- /* data */ +- [GDT_ENTRY_APMBIOS_BASE+2] = GDT_ENTRY_INIT(0x4092, 0, 0xffff), +- +- [GDT_ENTRY_ESPFIX_SS] = GDT_ENTRY_INIT(0xc092, 0, 0xfffff), +- [GDT_ENTRY_PERCPU] = GDT_ENTRY_INIT(0xc092, 0, 0xfffff), +- GDT_STACK_CANARY_INIT +-#endif +-} }; +-EXPORT_PER_CPU_SYMBOL_GPL(gdt_page); +- + static int __init x86_xsave_setup(char *s) + { + setup_clear_cpu_cap(X86_FEATURE_XSAVE); +@@ -352,7 +298,7 @@ void switch_to_new_gdt(int cpu) + { + struct desc_ptr gdt_descr; + +- gdt_descr.address = (long)get_cpu_gdt_table(cpu); ++ gdt_descr.address = (unsigned long)get_cpu_gdt_table(cpu); + gdt_descr.size = GDT_SIZE - 1; + load_gdt(&gdt_descr); + /* Reload the per-cpu base */ +@@ -825,6 +771,10 @@ static void __cpuinit identify_cpu(struc + /* Filter out anything that depends on CPUID levels we don't have */ + filter_cpuid_features(c, true); + ++#if defined(CONFIG_PAX_SEGMEXEC) || defined(CONFIG_PAX_KERNEXEC) || (defined(CONFIG_PAX_MEMORY_UDEREF) && defined(CONFIG_X86_32)) ++ setup_clear_cpu_cap(X86_FEATURE_SEP); ++#endif ++ + /* If the model name is still unset, do table lookup. */ + if (!c->x86_model_id[0]) { + const char *p; +@@ -1004,6 +954,9 @@ static __init int setup_disablecpuid(cha + } + __setup("clearcpuid=", setup_disablecpuid); + ++DEFINE_PER_CPU(struct thread_info *, current_tinfo) = &init_task.tinfo; ++EXPORT_PER_CPU_SYMBOL(current_tinfo); ++ + #ifdef CONFIG_X86_64 + struct desc_ptr idt_descr = { NR_VECTORS * 16 - 1, (unsigned long) idt_table }; + +@@ -1019,7 +972,7 @@ DEFINE_PER_CPU(struct task_struct *, cur + EXPORT_PER_CPU_SYMBOL(current_task); + + DEFINE_PER_CPU(unsigned long, kernel_stack) = +- (unsigned long)&init_thread_union - KERNEL_STACK_OFFSET + THREAD_SIZE; ++ (unsigned long)&init_thread_union - 8 + THREAD_SIZE; + EXPORT_PER_CPU_SYMBOL(kernel_stack); + + DEFINE_PER_CPU(char *, irq_stack_ptr) = +@@ -1084,7 +1037,7 @@ struct pt_regs * __cpuinit idle_regs(str + { + memset(regs, 0, sizeof(struct pt_regs)); + regs->fs = __KERNEL_PERCPU; +- regs->gs = __KERNEL_STACK_CANARY; ++ savesegment(gs, regs->gs); + + return regs; + } +@@ -1139,7 +1092,7 @@ void __cpuinit cpu_init(void) + int i; + + cpu = stack_smp_processor_id(); +- t = &per_cpu(init_tss, cpu); ++ t = init_tss + cpu; + oist = &per_cpu(orig_ist, cpu); + + #ifdef CONFIG_NUMA +@@ -1165,7 +1118,7 @@ void __cpuinit cpu_init(void) + switch_to_new_gdt(cpu); + loadsegment(fs, 0); + +- load_idt((const struct desc_ptr *)&idt_descr); ++ load_idt(&idt_descr); + + memset(me->thread.tls_array, 0, GDT_ENTRY_TLS_ENTRIES * 8); + syscall_init(); +@@ -1174,7 +1127,6 @@ void __cpuinit cpu_init(void) + wrmsrl(MSR_KERNEL_GS_BASE, 0); + barrier(); + +- x86_configure_nx(); + if (cpu != 0) + enable_x2apic(); + +@@ -1228,7 +1180,7 @@ void __cpuinit cpu_init(void) + { + int cpu = smp_processor_id(); + struct task_struct *curr = current; +- struct tss_struct *t = &per_cpu(init_tss, cpu); ++ struct tss_struct *t = init_tss + cpu; + struct thread_struct *thread = &curr->thread; + + if (cpumask_test_and_set_cpu(cpu, cpu_initialized_mask)) { +diff -urNp linux-2.6.38.3/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c linux-2.6.38.3/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c +--- linux-2.6.38.3/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c 2011-04-17 15:57:32.000000000 -0400 +@@ -481,7 +481,7 @@ static const struct dmi_system_id sw_any + DMI_MATCH(DMI_PRODUCT_NAME, "X6DLP"), + }, + }, +- { } ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } + }; + + static int acpi_cpufreq_blacklist(struct cpuinfo_x86 *c) +diff -urNp linux-2.6.38.3/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c linux-2.6.38.3/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c +--- linux-2.6.38.3/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c 2011-04-17 15:57:32.000000000 -0400 +@@ -226,7 +226,7 @@ static struct cpu_model models[] = + { &cpu_ids[CPU_MP4HT_D0], NULL, 0, NULL }, + { &cpu_ids[CPU_MP4HT_E0], NULL, 0, NULL }, + +- { NULL, } ++ { NULL, NULL, 0, NULL} + }; + #undef _BANIAS + #undef BANIAS +diff -urNp linux-2.6.38.3/arch/x86/kernel/cpu/intel.c linux-2.6.38.3/arch/x86/kernel/cpu/intel.c +--- linux-2.6.38.3/arch/x86/kernel/cpu/intel.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/cpu/intel.c 2011-04-17 15:57:32.000000000 -0400 +@@ -161,7 +161,7 @@ static void __cpuinit trap_init_f00f_bug + * Update the IDT descriptor and reload the IDT so that + * it uses the read-only mapped virtual address. + */ +- idt_descr.address = fix_to_virt(FIX_F00F_IDT); ++ idt_descr.address = (struct desc_struct *)fix_to_virt(FIX_F00F_IDT); + load_idt(&idt_descr); + } + #endif +diff -urNp linux-2.6.38.3/arch/x86/kernel/cpu/Makefile linux-2.6.38.3/arch/x86/kernel/cpu/Makefile +--- linux-2.6.38.3/arch/x86/kernel/cpu/Makefile 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/cpu/Makefile 2011-04-17 15:57:32.000000000 -0400 +@@ -8,10 +8,6 @@ CFLAGS_REMOVE_common.o = -pg + CFLAGS_REMOVE_perf_event.o = -pg + endif + +-# Make sure load_percpu_segment has no stackprotector +-nostackp := $(call cc-option, -fno-stack-protector) +-CFLAGS_common.o := $(nostackp) +- + obj-y := intel_cacheinfo.o scattered.o topology.o + obj-y += proc.o capflags.o powerflags.o common.o + obj-y += vmware.o hypervisor.o sched.o mshyperv.o +diff -urNp linux-2.6.38.3/arch/x86/kernel/cpu/mcheck/mce.c linux-2.6.38.3/arch/x86/kernel/cpu/mcheck/mce.c +--- linux-2.6.38.3/arch/x86/kernel/cpu/mcheck/mce.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/cpu/mcheck/mce.c 2011-04-17 15:57:32.000000000 -0400 +@@ -45,6 +45,7 @@ + #include <asm/ipi.h> + #include <asm/mce.h> + #include <asm/msr.h> ++#include <asm/local.h> + + #include "mce-internal.h" + +@@ -219,7 +220,7 @@ static void print_mce(struct mce *m) + !(m->mcgstatus & MCG_STATUS_EIPV) ? " !INEXACT!" : "", + m->cs, m->ip); + +- if (m->cs == __KERNEL_CS) ++ if (m->cs == __KERNEL_CS || m->cs == __KERNEXEC_KERNEL_CS) + print_symbol("{%s}", m->ip); + pr_cont("\n"); + } +@@ -1460,14 +1461,14 @@ void __cpuinit mcheck_cpu_init(struct cp + */ + + static DEFINE_SPINLOCK(mce_state_lock); +-static int open_count; /* #times opened */ ++static local_t open_count; /* #times opened */ + static int open_exclu; /* already open exclusive? */ + + static int mce_open(struct inode *inode, struct file *file) + { + spin_lock(&mce_state_lock); + +- if (open_exclu || (open_count && (file->f_flags & O_EXCL))) { ++ if (open_exclu || (local_read(&open_count) && (file->f_flags & O_EXCL))) { + spin_unlock(&mce_state_lock); + + return -EBUSY; +@@ -1475,7 +1476,7 @@ static int mce_open(struct inode *inode, + + if (file->f_flags & O_EXCL) + open_exclu = 1; +- open_count++; ++ local_inc(&open_count); + + spin_unlock(&mce_state_lock); + +@@ -1486,7 +1487,7 @@ static int mce_release(struct inode *ino + { + spin_lock(&mce_state_lock); + +- open_count--; ++ local_dec(&open_count); + open_exclu = 0; + + spin_unlock(&mce_state_lock); +@@ -1658,8 +1659,7 @@ static long mce_ioctl(struct file *f, un + } + } + +-/* Modified in mce-inject.c, so not static or const */ +-struct file_operations mce_chrdev_ops = { ++struct file_operations mce_chrdev_ops = { /* Modified in mce-inject.c, so not static or const */ + .open = mce_open, + .release = mce_release, + .read = mce_read, +@@ -1673,6 +1673,7 @@ static struct miscdevice mce_log_device + MISC_MCELOG_MINOR, + "mcelog", + &mce_chrdev_ops, ++ {NULL, NULL}, NULL, NULL + }; + + /* +diff -urNp linux-2.6.38.3/arch/x86/kernel/cpu/mtrr/generic.c linux-2.6.38.3/arch/x86/kernel/cpu/mtrr/generic.c +--- linux-2.6.38.3/arch/x86/kernel/cpu/mtrr/generic.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/cpu/mtrr/generic.c 2011-04-17 15:57:32.000000000 -0400 +@@ -28,7 +28,7 @@ static struct fixed_range_block fixed_ra + { MSR_MTRRfix64K_00000, 1 }, /* one 64k MTRR */ + { MSR_MTRRfix16K_80000, 2 }, /* two 16k MTRRs */ + { MSR_MTRRfix4K_C0000, 8 }, /* eight 4k MTRRs */ +- {} ++ { 0, 0 } + }; + + static unsigned long smp_changes_mask; +diff -urNp linux-2.6.38.3/arch/x86/kernel/cpu/mtrr/main.c linux-2.6.38.3/arch/x86/kernel/cpu/mtrr/main.c +--- linux-2.6.38.3/arch/x86/kernel/cpu/mtrr/main.c 2011-04-17 16:53:01.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/cpu/mtrr/main.c 2011-04-17 16:53:16.000000000 -0400 +@@ -61,7 +61,7 @@ static DEFINE_MUTEX(mtrr_mutex); + u64 size_or_mask, size_and_mask; + static bool mtrr_aps_delayed_init; + +-static const struct mtrr_ops *mtrr_ops[X86_VENDOR_NUM]; ++static const struct mtrr_ops *mtrr_ops[X86_VENDOR_NUM] __read_only; + + const struct mtrr_ops *mtrr_if; + +diff -urNp linux-2.6.38.3/arch/x86/kernel/cpu/mtrr/mtrr.h linux-2.6.38.3/arch/x86/kernel/cpu/mtrr/mtrr.h +--- linux-2.6.38.3/arch/x86/kernel/cpu/mtrr/mtrr.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/cpu/mtrr/mtrr.h 2011-04-17 15:57:32.000000000 -0400 +@@ -12,19 +12,19 @@ + extern unsigned int mtrr_usage_table[MTRR_MAX_VAR_RANGES]; + + struct mtrr_ops { +- u32 vendor; +- u32 use_intel_if; +- void (*set)(unsigned int reg, unsigned long base, ++ const u32 vendor; ++ const u32 use_intel_if; ++ void (* const set)(unsigned int reg, unsigned long base, + unsigned long size, mtrr_type type); +- void (*set_all)(void); ++ void (* const set_all)(void); + +- void (*get)(unsigned int reg, unsigned long *base, ++ void (* const get)(unsigned int reg, unsigned long *base, + unsigned long *size, mtrr_type *type); +- int (*get_free_region)(unsigned long base, unsigned long size, ++ int (* const get_free_region)(unsigned long base, unsigned long size, + int replace_reg); +- int (*validate_add_page)(unsigned long base, unsigned long size, ++ int (* const validate_add_page)(unsigned long base, unsigned long size, + unsigned int type); +- int (*have_wrcomb)(void); ++ int (* const have_wrcomb)(void); + }; + + extern int generic_get_free_region(unsigned long base, unsigned long size, +diff -urNp linux-2.6.38.3/arch/x86/kernel/cpu/perf_event.c linux-2.6.38.3/arch/x86/kernel/cpu/perf_event.c +--- linux-2.6.38.3/arch/x86/kernel/cpu/perf_event.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/cpu/perf_event.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1781,7 +1781,7 @@ perf_callchain_user(struct perf_callchai + break; + + perf_callchain_store(entry, frame.return_address); +- fp = frame.next_frame; ++ fp = (__force const void __user *)frame.next_frame; + } + } + +diff -urNp linux-2.6.38.3/arch/x86/kernel/crash.c linux-2.6.38.3/arch/x86/kernel/crash.c +--- linux-2.6.38.3/arch/x86/kernel/crash.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/crash.c 2011-04-17 15:57:32.000000000 -0400 +@@ -42,7 +42,7 @@ static void kdump_nmi_callback(int cpu, + regs = args->regs; + + #ifdef CONFIG_X86_32 +- if (!user_mode_vm(regs)) { ++ if (!user_mode(regs)) { + crash_fixup_ss_esp(&fixed_regs, regs); + regs = &fixed_regs; + } +diff -urNp linux-2.6.38.3/arch/x86/kernel/doublefault_32.c linux-2.6.38.3/arch/x86/kernel/doublefault_32.c +--- linux-2.6.38.3/arch/x86/kernel/doublefault_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/doublefault_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -11,7 +11,7 @@ + + #define DOUBLEFAULT_STACKSIZE (1024) + static unsigned long doublefault_stack[DOUBLEFAULT_STACKSIZE]; +-#define STACK_START (unsigned long)(doublefault_stack+DOUBLEFAULT_STACKSIZE) ++#define STACK_START (unsigned long)(doublefault_stack+DOUBLEFAULT_STACKSIZE-2) + + #define ptr_ok(x) ((x) > PAGE_OFFSET && (x) < PAGE_OFFSET + MAXMEM) + +@@ -21,7 +21,7 @@ static void doublefault_fn(void) + unsigned long gdt, tss; + + store_gdt(&gdt_desc); +- gdt = gdt_desc.address; ++ gdt = (unsigned long)gdt_desc.address; + + printk(KERN_EMERG "PANIC: double fault, gdt at %08lx [%d bytes]\n", gdt, gdt_desc.size); + +@@ -58,10 +58,10 @@ struct tss_struct doublefault_tss __cach + /* 0x2 bit is always set */ + .flags = X86_EFLAGS_SF | 0x2, + .sp = STACK_START, +- .es = __USER_DS, ++ .es = __KERNEL_DS, + .cs = __KERNEL_CS, + .ss = __KERNEL_DS, +- .ds = __USER_DS, ++ .ds = __KERNEL_DS, + .fs = __KERNEL_PERCPU, + + .__cr3 = __pa_nodebug(swapper_pg_dir), +diff -urNp linux-2.6.38.3/arch/x86/kernel/dumpstack_32.c linux-2.6.38.3/arch/x86/kernel/dumpstack_32.c +--- linux-2.6.38.3/arch/x86/kernel/dumpstack_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/dumpstack_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -37,15 +37,12 @@ void dump_trace(struct task_struct *task + + bp = stack_frame(task, regs); + for (;;) { +- struct thread_info *context; ++ void *stack_start = (void *)((unsigned long)stack & ~(THREAD_SIZE-1)); ++ bp = ops->walk_stack(task, stack_start, stack, bp, ops, data, NULL, &graph); + +- context = (struct thread_info *) +- ((unsigned long)stack & (~(THREAD_SIZE - 1))); +- bp = ops->walk_stack(context, stack, bp, ops, data, NULL, &graph); +- +- stack = (unsigned long *)context->previous_esp; +- if (!stack) ++ if (stack_start == task_stack_page(task)) + break; ++ stack = *(unsigned long **)stack_start; + if (ops->stack(data, "IRQ") < 0) + break; + touch_nmi_watchdog(); +@@ -95,21 +92,22 @@ void show_registers(struct pt_regs *regs + * When in-kernel, we also print out the stack and code at the + * time of the fault.. + */ +- if (!user_mode_vm(regs)) { ++ if (!user_mode(regs)) { + unsigned int code_prologue = code_bytes * 43 / 64; + unsigned int code_len = code_bytes; + unsigned char c; + u8 *ip; ++ unsigned long cs_base = get_desc_base(&get_cpu_gdt_table(smp_processor_id())[(0xffff & regs->cs) >> 3]); + + printk(KERN_EMERG "Stack:\n"); + show_stack_log_lvl(NULL, regs, ®s->sp, KERN_EMERG); + + printk(KERN_EMERG "Code: "); + +- ip = (u8 *)regs->ip - code_prologue; ++ ip = (u8 *)regs->ip - code_prologue + cs_base; + if (ip < (u8 *)PAGE_OFFSET || probe_kernel_address(ip, c)) { + /* try starting at IP */ +- ip = (u8 *)regs->ip; ++ ip = (u8 *)regs->ip + cs_base; + code_len = code_len - code_prologue + 1; + } + for (i = 0; i < code_len; i++, ip++) { +@@ -118,7 +116,7 @@ void show_registers(struct pt_regs *regs + printk(" Bad EIP value."); + break; + } +- if (ip == (u8 *)regs->ip) ++ if (ip == (u8 *)regs->ip + cs_base) + printk("<%02x> ", c); + else + printk("%02x ", c); +@@ -131,6 +129,7 @@ int is_valid_bugaddr(unsigned long ip) + { + unsigned short ud2; + ++ ip = ktla_ktva(ip); + if (ip < PAGE_OFFSET) + return 0; + if (probe_kernel_address((unsigned short *)ip, ud2)) +diff -urNp linux-2.6.38.3/arch/x86/kernel/dumpstack_64.c linux-2.6.38.3/arch/x86/kernel/dumpstack_64.c +--- linux-2.6.38.3/arch/x86/kernel/dumpstack_64.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/dumpstack_64.c 2011-04-17 15:57:32.000000000 -0400 +@@ -147,10 +147,10 @@ void dump_trace(struct task_struct *task + unsigned long *irq_stack_end = + (unsigned long *)per_cpu(irq_stack_ptr, cpu); + unsigned used = 0; +- struct thread_info *tinfo; + int graph = 0; + unsigned long dummy; + unsigned long bp; ++ void *stack_start; + + if (!task) + task = current; +@@ -167,10 +167,10 @@ void dump_trace(struct task_struct *task + * current stack address. If the stacks consist of nested + * exceptions + */ +- tinfo = task_thread_info(task); + for (;;) { + char *id; + unsigned long *estack_end; ++ + estack_end = in_exception_stack(cpu, (unsigned long)stack, + &used, &id); + +@@ -178,7 +178,7 @@ void dump_trace(struct task_struct *task + if (ops->stack(data, id) < 0) + break; + +- bp = ops->walk_stack(tinfo, stack, bp, ops, ++ bp = ops->walk_stack(task, estack_end - EXCEPTION_STKSZ, stack, bp, ops, + data, estack_end, &graph); + ops->stack(data, "<EOE>"); + /* +@@ -197,7 +197,7 @@ void dump_trace(struct task_struct *task + if (in_irq_stack(stack, irq_stack, irq_stack_end)) { + if (ops->stack(data, "IRQ") < 0) + break; +- bp = ops->walk_stack(tinfo, stack, bp, ++ bp = ops->walk_stack(task, irq_stack, stack, bp, + ops, data, irq_stack_end, &graph); + /* + * We link to the next stack (which would be +@@ -218,7 +218,8 @@ void dump_trace(struct task_struct *task + /* + * This handles the process stack: + */ +- bp = ops->walk_stack(tinfo, stack, bp, ops, data, NULL, &graph); ++ stack_start = (void *)((unsigned long)stack & ~(THREAD_SIZE-1)); ++ bp = ops->walk_stack(task, stack_start, stack, bp, ops, data, NULL, &graph); + put_cpu(); + } + EXPORT_SYMBOL(dump_trace); +diff -urNp linux-2.6.38.3/arch/x86/kernel/dumpstack.c linux-2.6.38.3/arch/x86/kernel/dumpstack.c +--- linux-2.6.38.3/arch/x86/kernel/dumpstack.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/dumpstack.c 2011-04-17 15:57:32.000000000 -0400 +@@ -2,6 +2,9 @@ + * Copyright (C) 1991, 1992 Linus Torvalds + * Copyright (C) 2000, 2001, 2002 Andi Kleen, SuSE Labs + */ ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++#define __INCLUDED_BY_HIDESYM 1 ++#endif + #include <linux/kallsyms.h> + #include <linux/kprobes.h> + #include <linux/uaccess.h> +@@ -27,7 +30,7 @@ static int die_counter; + + void printk_address(unsigned long address, int reliable) + { +- printk(" [<%p>] %s%pS\n", (void *) address, ++ printk(" [<%p>] %s%pA\n", (void *) address, + reliable ? "" : "? ", (void *) address); + } + +@@ -35,9 +38,8 @@ void printk_address(unsigned long addres + static void + print_ftrace_graph_addr(unsigned long addr, void *data, + const struct stacktrace_ops *ops, +- struct thread_info *tinfo, int *graph) ++ struct task_struct *task, int *graph) + { +- struct task_struct *task = tinfo->task; + unsigned long ret_addr; + int index = task->curr_ret_stack; + +@@ -58,7 +60,7 @@ print_ftrace_graph_addr(unsigned long ad + static inline void + print_ftrace_graph_addr(unsigned long addr, void *data, + const struct stacktrace_ops *ops, +- struct thread_info *tinfo, int *graph) ++ struct task_struct *task, int *graph) + { } + #endif + +@@ -69,10 +71,8 @@ print_ftrace_graph_addr(unsigned long ad + * severe exception (double fault, nmi, stack fault, debug, mce) hardware stack + */ + +-static inline int valid_stack_ptr(struct thread_info *tinfo, +- void *p, unsigned int size, void *end) ++static inline int valid_stack_ptr(void *t, void *p, unsigned int size, void *end) + { +- void *t = tinfo; + if (end) { + if (p < end && p >= (end-THREAD_SIZE)) + return 1; +@@ -83,14 +83,14 @@ static inline int valid_stack_ptr(struct + } + + unsigned long +-print_context_stack(struct thread_info *tinfo, ++print_context_stack(struct task_struct *task, void *stack_start, + unsigned long *stack, unsigned long bp, + const struct stacktrace_ops *ops, void *data, + unsigned long *end, int *graph) + { + struct stack_frame *frame = (struct stack_frame *)bp; + +- while (valid_stack_ptr(tinfo, stack, sizeof(*stack), end)) { ++ while (valid_stack_ptr(stack_start, stack, sizeof(*stack), end)) { + unsigned long addr; + + addr = *stack; +@@ -102,7 +102,7 @@ print_context_stack(struct thread_info * + } else { + ops->address(data, addr, 0); + } +- print_ftrace_graph_addr(addr, data, ops, tinfo, graph); ++ print_ftrace_graph_addr(addr, data, ops, task, graph); + } + stack++; + } +@@ -111,7 +111,7 @@ print_context_stack(struct thread_info * + EXPORT_SYMBOL_GPL(print_context_stack); + + unsigned long +-print_context_stack_bp(struct thread_info *tinfo, ++print_context_stack_bp(struct task_struct *task, void *stack_start, + unsigned long *stack, unsigned long bp, + const struct stacktrace_ops *ops, void *data, + unsigned long *end, int *graph) +@@ -119,7 +119,7 @@ print_context_stack_bp(struct thread_inf + struct stack_frame *frame = (struct stack_frame *)bp; + unsigned long *ret_addr = &frame->return_address; + +- while (valid_stack_ptr(tinfo, ret_addr, sizeof(*ret_addr), end)) { ++ while (valid_stack_ptr(stack_start, ret_addr, sizeof(*ret_addr), end)) { + unsigned long addr = *ret_addr; + + if (!__kernel_text_address(addr)) +@@ -128,7 +128,7 @@ print_context_stack_bp(struct thread_inf + ops->address(data, addr, 1); + frame = frame->next_frame; + ret_addr = &frame->return_address; +- print_ftrace_graph_addr(addr, data, ops, tinfo, graph); ++ print_ftrace_graph_addr(addr, data, ops, task, graph); + } + + return (unsigned long)frame; +@@ -200,7 +200,7 @@ void dump_stack(void) + unsigned long stack; + + printk("Pid: %d, comm: %.20s %s %s %.*s\n", +- current->pid, current->comm, print_tainted(), ++ task_pid_nr(current), current->comm, print_tainted(), + init_utsname()->release, + (int)strcspn(init_utsname()->version, " "), + init_utsname()->version); +@@ -236,6 +236,8 @@ unsigned __kprobes long oops_begin(void) + } + EXPORT_SYMBOL_GPL(oops_begin); + ++extern void gr_handle_kernel_exploit(void); ++ + void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr) + { + if (regs && kexec_should_crash(current)) +@@ -257,7 +259,10 @@ void __kprobes oops_end(unsigned long fl + panic("Fatal exception in interrupt"); + if (panic_on_oops) + panic("Fatal exception"); +- do_exit(signr); ++ ++ gr_handle_kernel_exploit(); ++ ++ do_group_exit(signr); + } + + int __kprobes __die(const char *str, struct pt_regs *regs, long err) +@@ -284,7 +289,7 @@ int __kprobes __die(const char *str, str + + show_registers(regs); + #ifdef CONFIG_X86_32 +- if (user_mode_vm(regs)) { ++ if (user_mode(regs)) { + sp = regs->sp; + ss = regs->ss & 0xffff; + } else { +@@ -312,7 +317,7 @@ void die(const char *str, struct pt_regs + unsigned long flags = oops_begin(); + int sig = SIGSEGV; + +- if (!user_mode_vm(regs)) ++ if (!user_mode(regs)) + report_bug(regs->ip, regs); + + if (__die(str, regs, err)) +diff -urNp linux-2.6.38.3/arch/x86/kernel/entry_32.S linux-2.6.38.3/arch/x86/kernel/entry_32.S +--- linux-2.6.38.3/arch/x86/kernel/entry_32.S 2011-03-28 17:42:40.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/entry_32.S 2011-04-17 15:57:32.000000000 -0400 +@@ -183,13 +183,81 @@ + /*CFI_REL_OFFSET gs, PT_GS*/ + .endm + .macro SET_KERNEL_GS reg ++ ++#ifdef CONFIG_CC_STACKPROTECTOR + movl $(__KERNEL_STACK_CANARY), \reg ++#elif defined(CONFIG_PAX_MEMORY_UDEREF) ++ movl $(__USER_DS), \reg ++#else ++ xorl \reg, \reg ++#endif ++ + movl \reg, %gs + .endm + + #endif /* CONFIG_X86_32_LAZY_GS */ + +-.macro SAVE_ALL ++.macro PAX_EXIT_KERNEL ++#ifdef CONFIG_PAX_KERNEXEC ++#ifdef CONFIG_PARAVIRT ++ push %eax; push %ecx ++#endif ++ mov %cs, %esi ++ cmp $__KERNEXEC_KERNEL_CS, %esi ++ jnz 2f ++#ifdef CONFIG_PARAVIRT ++ call PARA_INDIRECT(pv_cpu_ops+PV_CPU_read_cr0); ++ mov %eax, %esi ++#else ++ mov %cr0, %esi ++#endif ++ btr $16, %esi ++ ljmp $__KERNEL_CS, $1f ++1: ++#ifdef CONFIG_PARAVIRT ++ mov %esi, %eax ++ call PARA_INDIRECT(pv_cpu_ops+PV_CPU_write_cr0); ++#else ++ mov %esi, %cr0 ++#endif ++2: ++#ifdef CONFIG_PARAVIRT ++ pop %ecx; pop %eax ++#endif ++#endif ++.endm ++ ++.macro PAX_ENTER_KERNEL ++#ifdef CONFIG_PAX_KERNEXEC ++#ifdef CONFIG_PARAVIRT ++ push %eax; push %ecx ++ call PARA_INDIRECT(pv_cpu_ops+PV_CPU_read_cr0) ++ mov %eax, %esi ++#else ++ mov %cr0, %esi ++#endif ++ bts $16, %esi ++ jnc 1f ++ mov %cs, %esi ++ cmp $__KERNEL_CS, %esi ++ jz 3f ++ ljmp $__KERNEL_CS, $3f ++1: ljmp $__KERNEXEC_KERNEL_CS, $2f ++2: ++#ifdef CONFIG_PARAVIRT ++ mov %esi, %eax ++ call PARA_INDIRECT(pv_cpu_ops+PV_CPU_write_cr0) ++#else ++ mov %esi, %cr0 ++#endif ++3: ++#ifdef CONFIG_PARAVIRT ++ pop %ecx; pop %eax ++#endif ++#endif ++.endm ++ ++.macro __SAVE_ALL _DS + cld + PUSH_GS + pushl_cfi %fs +@@ -212,7 +280,7 @@ + CFI_REL_OFFSET ecx, 0 + pushl_cfi %ebx + CFI_REL_OFFSET ebx, 0 +- movl $(__USER_DS), %edx ++ movl $\_DS, %edx + movl %edx, %ds + movl %edx, %es + movl $(__KERNEL_PERCPU), %edx +@@ -220,6 +288,15 @@ + SET_KERNEL_GS %edx + .endm + ++.macro SAVE_ALL ++#if defined(CONFIG_PAX_KERNEXEC) || defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) || defined(CONFIG_PAX_MEMORY_UDEREF) ++ __SAVE_ALL __KERNEL_DS ++ PAX_ENTER_KERNEL ++#else ++ __SAVE_ALL __USER_DS ++#endif ++.endm ++ + .macro RESTORE_INT_REGS + popl_cfi %ebx + CFI_RESTORE ebx +@@ -330,7 +407,15 @@ check_userspace: + movb PT_CS(%esp), %al + andl $(X86_EFLAGS_VM | SEGMENT_RPL_MASK), %eax + cmpl $USER_RPL, %eax ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ jae resume_userspace ++ ++ PAX_EXIT_KERNEL ++ jmp resume_kernel ++#else + jb resume_kernel # not returning to v8086 or userspace ++#endif + + ENTRY(resume_userspace) + LOCKDEP_SYS_EXIT +@@ -392,10 +477,9 @@ sysenter_past_esp: + /*CFI_REL_OFFSET cs, 0*/ + /* + * Push current_thread_info()->sysenter_return to the stack. +- * A tiny bit of offset fixup is necessary - 4*4 means the 4 words +- * pushed above; +8 corresponds to copy_thread's esp0 setting. + */ +- pushl_cfi ((TI_sysenter_return)-THREAD_SIZE_asm+8+4*4)(%esp) ++ GET_THREAD_INFO(%ebp) ++ pushl_cfi TI_sysenter_return(%ebp) + CFI_REL_OFFSET eip, 0 + + pushl_cfi %eax +@@ -406,9 +490,19 @@ sysenter_past_esp: + * Load the potential sixth argument from user stack. + * Careful about security. + */ ++ movl PT_OLDESP(%esp),%ebp ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ mov PT_OLDSS(%esp),%ds ++1: movl %ds:(%ebp),%ebp ++ push %ss ++ pop %ds ++#else + cmpl $__PAGE_OFFSET-3,%ebp + jae syscall_fault + 1: movl (%ebp),%ebp ++#endif ++ + movl %ebp,PT_EBP(%esp) + .section __ex_table,"a" + .align 4 +@@ -431,12 +525,23 @@ sysenter_do_call: + testl $_TIF_ALLWORK_MASK, %ecx + jne sysexit_audit + sysenter_exit: ++ ++#ifdef CONFIG_PAX_RANDKSTACK ++ pushl %eax ++ CFI_ADJUST_CFA_OFFSET 4 ++ call pax_randomize_kstack ++ popl %eax ++ CFI_ADJUST_CFA_OFFSET -4 ++#endif ++ + /* if something modifies registers it must also disable sysexit */ + movl PT_EIP(%esp), %edx + movl PT_OLDESP(%esp), %ecx + xorl %ebp,%ebp + TRACE_IRQS_ON + 1: mov PT_FS(%esp), %fs ++2: mov PT_DS(%esp), %ds ++3: mov PT_ES(%esp), %es + PTGS_TO_GS + ENABLE_INTERRUPTS_SYSEXIT + +@@ -479,11 +584,17 @@ sysexit_audit: + + CFI_ENDPROC + .pushsection .fixup,"ax" +-2: movl $0,PT_FS(%esp) ++4: movl $0,PT_FS(%esp) ++ jmp 1b ++5: movl $0,PT_DS(%esp) ++ jmp 1b ++6: movl $0,PT_ES(%esp) + jmp 1b + .section __ex_table,"a" + .align 4 +- .long 1b,2b ++ .long 1b,4b ++ .long 2b,5b ++ .long 3b,6b + .popsection + PTGS_TO_GS_EX + ENDPROC(ia32_sysenter_target) +@@ -516,6 +627,10 @@ syscall_exit: + testl $_TIF_ALLWORK_MASK, %ecx # current->work + jne syscall_exit_work + ++#ifdef CONFIG_PAX_RANDKSTACK ++ call pax_randomize_kstack ++#endif ++ + restore_all: + TRACE_IRQS_IRET + restore_all_notrace: +@@ -575,14 +690,21 @@ ldt_ss: + * compensating for the offset by changing to the ESPFIX segment with + * a base address that matches for the difference. + */ +-#define GDT_ESPFIX_SS PER_CPU_VAR(gdt_page) + (GDT_ENTRY_ESPFIX_SS * 8) ++#define GDT_ESPFIX_SS (GDT_ENTRY_ESPFIX_SS * 8)(%ebx) + mov %esp, %edx /* load kernel esp */ + mov PT_OLDESP(%esp), %eax /* load userspace esp */ + mov %dx, %ax /* eax: new kernel esp */ + sub %eax, %edx /* offset (low word is 0) */ ++#ifdef CONFIG_SMP ++ movl PER_CPU_VAR(cpu_number), %ebx ++ shll $PAGE_SHIFT_asm, %ebx ++ addl $cpu_gdt_table, %ebx ++#else ++ movl $cpu_gdt_table, %ebx ++#endif + shr $16, %edx +- mov %dl, GDT_ESPFIX_SS + 4 /* bits 16..23 */ +- mov %dh, GDT_ESPFIX_SS + 7 /* bits 24..31 */ ++ mov %dl, 4 + GDT_ESPFIX_SS /* bits 16..23 */ ++ mov %dh, 7 + GDT_ESPFIX_SS /* bits 24..31 */ + pushl_cfi $__ESPFIX_SS + pushl_cfi %eax /* new kernel esp */ + /* Disable interrupts, but do not irqtrace this section: we +@@ -617,23 +739,17 @@ work_resched: + + work_notifysig: # deal with pending signals and + # notify-resume requests ++ movl %esp, %eax + #ifdef CONFIG_VM86 + testl $X86_EFLAGS_VM, PT_EFLAGS(%esp) +- movl %esp, %eax +- jne work_notifysig_v86 # returning to kernel-space or ++ jz 1f # returning to kernel-space or + # vm86-space +- xorl %edx, %edx +- call do_notify_resume +- jmp resume_userspace_sig + +- ALIGN +-work_notifysig_v86: + pushl_cfi %ecx # save ti_flags for do_notify_resume + call save_v86_state # %eax contains pt_regs pointer + popl_cfi %ecx + movl %eax, %esp +-#else +- movl %esp, %eax ++1: + #endif + xorl %edx, %edx + call do_notify_resume +@@ -668,6 +784,10 @@ END(syscall_exit_work) + + RING0_INT_FRAME # can't unwind into user space anyway + syscall_fault: ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ push %ss ++ pop %ds ++#endif + GET_THREAD_INFO(%ebp) + movl $-EFAULT,PT_EAX(%esp) + jmp resume_userspace +@@ -750,6 +870,36 @@ ptregs_clone: + CFI_ENDPROC + ENDPROC(ptregs_clone) + ++ ALIGN; ++ENTRY(kernel_execve) ++ CFI_STARTPROC ++ pushl_cfi %ebp ++ sub $PT_OLDSS+4,%esp ++ pushl_cfi %edi ++ pushl_cfi %ecx ++ pushl_cfi %eax ++ lea 3*4(%esp),%edi ++ mov $PT_OLDSS/4+1,%ecx ++ xorl %eax,%eax ++ rep stosl ++ popl_cfi %eax ++ popl_cfi %ecx ++ popl_cfi %edi ++ movl $X86_EFLAGS_IF,PT_EFLAGS(%esp) ++ pushl_cfi %esp ++ call sys_execve ++ add $4,%esp ++ CFI_ADJUST_CFA_OFFSET -4 ++ GET_THREAD_INFO(%ebp) ++ test %eax,%eax ++ jz syscall_exit ++ add $PT_OLDSS+4,%esp ++ CFI_ADJUST_CFA_OFFSET -PT_OLDSS-4 ++ popl_cfi %ebp ++ ret ++ CFI_ENDPROC ++ENDPROC(kernel_execve) ++ + .macro FIXUP_ESPFIX_STACK + /* + * Switch back for ESPFIX stack to the normal zerobased stack +@@ -759,8 +909,15 @@ ENDPROC(ptregs_clone) + * normal stack and adjusts ESP with the matching offset. + */ + /* fixup the stack */ +- mov GDT_ESPFIX_SS + 4, %al /* bits 16..23 */ +- mov GDT_ESPFIX_SS + 7, %ah /* bits 24..31 */ ++#ifdef CONFIG_SMP ++ movl PER_CPU_VAR(cpu_number), %ebx ++ shll $PAGE_SHIFT_asm, %ebx ++ addl $cpu_gdt_table, %ebx ++#else ++ movl $cpu_gdt_table, %ebx ++#endif ++ mov 4 + GDT_ESPFIX_SS, %al /* bits 16..23 */ ++ mov 7 + GDT_ESPFIX_SS, %ah /* bits 24..31 */ + shl $16, %eax + addl %esp, %eax /* the adjusted stack pointer */ + pushl_cfi $__KERNEL_DS +@@ -1211,7 +1368,6 @@ return_to_handler: + jmp *%ecx + #endif + +-.section .rodata,"a" + #include "syscall_table_32.S" + + syscall_table_size=(.-sys_call_table) +@@ -1257,9 +1413,12 @@ error_code: + movl $-1, PT_ORIG_EAX(%esp) # no syscall to restart + REG_TO_PTGS %ecx + SET_KERNEL_GS %ecx +- movl $(__USER_DS), %ecx ++ movl $(__KERNEL_DS), %ecx + movl %ecx, %ds + movl %ecx, %es ++ ++ PAX_ENTER_KERNEL ++ + TRACE_IRQS_OFF + movl %esp,%eax # pt_regs pointer + call *%edi +@@ -1344,6 +1503,9 @@ nmi_stack_correct: + xorl %edx,%edx # zero error code + movl %esp,%eax # pt_regs pointer + call do_nmi ++ ++ PAX_EXIT_KERNEL ++ + jmp restore_all_notrace + CFI_ENDPROC + +@@ -1380,6 +1542,9 @@ nmi_espfix_stack: + FIXUP_ESPFIX_STACK # %eax == %esp + xorl %edx,%edx # zero error code + call do_nmi ++ ++ PAX_EXIT_KERNEL ++ + RESTORE_REGS + lss 12+4(%esp), %esp # back to espfix stack + CFI_ADJUST_CFA_OFFSET -24 +diff -urNp linux-2.6.38.3/arch/x86/kernel/entry_64.S linux-2.6.38.3/arch/x86/kernel/entry_64.S +--- linux-2.6.38.3/arch/x86/kernel/entry_64.S 2011-03-23 17:20:06.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/entry_64.S 2011-04-17 15:57:32.000000000 -0400 +@@ -53,6 +53,7 @@ + #include <asm/paravirt.h> + #include <asm/ftrace.h> + #include <asm/percpu.h> ++#include <asm/pgtable.h> + + /* Avoid __ASSEMBLER__'ifying <linux/audit.h> just for this. */ + #include <linux/elf-em.h> +@@ -174,6 +175,206 @@ ENTRY(native_usergs_sysret64) + ENDPROC(native_usergs_sysret64) + #endif /* CONFIG_PARAVIRT */ + ++ .macro ljmpq sel, off ++#if defined(CONFIG_MCORE2) || defined (CONFIG_MATOM) ++ .byte 0x48; ljmp *1234f(%rip) ++ .pushsection .rodata ++ .align 16 ++ 1234: .quad \off; .word \sel ++ .popsection ++#else ++ push $\sel ++ push $\off ++ lretq ++#endif ++ .endm ++ ++ .macro pax_enter_kernel ++#ifdef CONFIG_PAX_KERNEXEC ++ call pax_enter_kernel ++#endif ++ .endm ++ ++ .macro pax_exit_kernel ++#ifdef CONFIG_PAX_KERNEXEC ++ call pax_exit_kernel ++#endif ++ .endm ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ENTRY(pax_enter_kernel) ++ push %rdi ++ ++#ifdef CONFIG_PARAVIRT ++ PV_SAVE_REGS(CLBR_RDI) ++#endif ++ ++ GET_CR0_INTO_RDI ++ bts $16,%rdi ++ jnc 1f ++ mov %cs,%edi ++ cmp $__KERNEL_CS,%edi ++ jz 3f ++ ljmpq __KERNEL_CS,3f ++1: ljmpq __KERNEXEC_KERNEL_CS,2f ++2: SET_RDI_INTO_CR0 ++3: ++ ++#ifdef CONFIG_PARAVIRT ++ PV_RESTORE_REGS(CLBR_RDI) ++#endif ++ ++ pop %rdi ++ retq ++ENDPROC(pax_enter_kernel) ++ ++ENTRY(pax_exit_kernel) ++ push %rdi ++ ++#ifdef CONFIG_PARAVIRT ++ PV_SAVE_REGS(CLBR_RDI) ++#endif ++ ++ mov %cs,%rdi ++ cmp $__KERNEXEC_KERNEL_CS,%edi ++ jnz 2f ++ GET_CR0_INTO_RDI ++ btr $16,%rdi ++ ljmpq __KERNEL_CS,1f ++1: SET_RDI_INTO_CR0 ++2: ++ ++#ifdef CONFIG_PARAVIRT ++ PV_RESTORE_REGS(CLBR_RDI); ++#endif ++ ++ pop %rdi ++ retq ++ENDPROC(pax_exit_kernel) ++#endif ++ ++ .macro pax_enter_kernel_user ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ call pax_enter_kernel_user ++#endif ++ .endm ++ ++ .macro pax_exit_kernel_user ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ call pax_exit_kernel_user ++#endif ++#ifdef CONFIG_PAX_RANDKSTACK ++ push %rax ++ call pax_randomize_kstack ++ pop %rax ++#endif ++ .endm ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ENTRY(pax_enter_kernel_user) ++ push %rdi ++ push %rbx ++ ++#ifdef CONFIG_PARAVIRT ++ PV_SAVE_REGS(CLBR_RDI) ++#endif ++ ++ GET_CR3_INTO_RDI ++ mov %rdi,%rbx ++ add $__START_KERNEL_map,%rbx ++ sub phys_base(%rip),%rbx ++ ++#ifdef CONFIG_PARAVIRT ++ push %rdi ++ cmpl $0, pv_info+PARAVIRT_enabled ++ jz 1f ++ i = 0 ++ .rept USER_PGD_PTRS ++ mov i*8(%rbx),%rsi ++ mov $0,%sil ++ lea i*8(%rbx),%rdi ++ call PARA_INDIRECT(pv_mmu_ops+PV_MMU_set_pgd) ++ i = i + 1 ++ .endr ++ jmp 2f ++1: ++#endif ++ ++ i = 0 ++ .rept USER_PGD_PTRS ++ movb $0,i*8(%rbx) ++ i = i + 1 ++ .endr ++ ++#ifdef CONFIG_PARAVIRT ++2: pop %rdi ++#endif ++ SET_RDI_INTO_CR3 ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ GET_CR0_INTO_RDI ++ bts $16,%rdi ++ SET_RDI_INTO_CR0 ++#endif ++ ++#ifdef CONFIG_PARAVIRT ++ PV_RESTORE_REGS(CLBR_RDI) ++#endif ++ ++ pop %rbx ++ pop %rdi ++ retq ++ENDPROC(pax_enter_kernel_user) ++ ++ENTRY(pax_exit_kernel_user) ++ push %rdi ++ ++#ifdef CONFIG_PARAVIRT ++ push %rbx ++ PV_SAVE_REGS(CLBR_RDI) ++#endif ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ GET_CR0_INTO_RDI ++ btr $16,%rdi ++ SET_RDI_INTO_CR0 ++#endif ++ ++ GET_CR3_INTO_RDI ++ add $__START_KERNEL_map,%rdi ++ sub phys_base(%rip),%rdi ++ ++#ifdef CONFIG_PARAVIRT ++ cmpl $0, pv_info+PARAVIRT_enabled ++ jz 1f ++ mov %rdi,%rbx ++ i = 0 ++ .rept USER_PGD_PTRS ++ mov i*8(%rbx),%rsi ++ mov $0x67,%sil ++ lea i*8(%rbx),%rdi ++ call PARA_INDIRECT(pv_mmu_ops+PV_MMU_set_pgd) ++ i = i + 1 ++ .endr ++ jmp 2f ++1: ++#endif ++ ++ i = 0 ++ .rept USER_PGD_PTRS ++ movb $0x67,i*8(%rdi) ++ i = i + 1 ++ .endr ++ ++#ifdef CONFIG_PARAVIRT ++2: PV_RESTORE_REGS(CLBR_RDI) ++ pop %rbx ++#endif ++ ++ pop %rdi ++ retq ++ENDPROC(pax_exit_kernel_user) ++#endif + + .macro TRACE_IRQS_IRETQ offset=ARGOFFSET + #ifdef CONFIG_TRACE_IRQFLAGS +@@ -316,7 +517,7 @@ ENTRY(save_args) + leaq -RBP+8(%rsp),%rdi /* arg1 for handler */ + movq_cfi rbp, 8 /* push %rbp */ + leaq 8(%rsp), %rbp /* mov %rsp, %ebp */ +- testl $3, CS(%rdi) ++ testb $3, CS(%rdi) + je 1f + SWAPGS + /* +@@ -407,7 +608,7 @@ ENTRY(ret_from_fork) + + RESTORE_REST + +- testl $3, CS-ARGOFFSET(%rsp) # from kernel_thread? ++ testb $3, CS-ARGOFFSET(%rsp) # from kernel_thread? + je int_ret_from_sys_call + + testl $_TIF_IA32, TI_flags(%rcx) # 32-bit compat task needs IRET +@@ -453,7 +654,7 @@ END(ret_from_fork) + ENTRY(system_call) + CFI_STARTPROC simple + CFI_SIGNAL_FRAME +- CFI_DEF_CFA rsp,KERNEL_STACK_OFFSET ++ CFI_DEF_CFA rsp,0 + CFI_REGISTER rip,rcx + /*CFI_REGISTER rflags,r11*/ + SWAPGS_UNSAFE_STACK +@@ -466,12 +667,13 @@ ENTRY(system_call_after_swapgs) + + movq %rsp,PER_CPU_VAR(old_rsp) + movq PER_CPU_VAR(kernel_stack),%rsp ++ pax_enter_kernel_user + /* + * No need to follow this irqs off/on section - it's straight + * and short: + */ + ENABLE_INTERRUPTS(CLBR_NONE) +- SAVE_ARGS 8,1 ++ SAVE_ARGS 8*6,1 + movq %rax,ORIG_RAX-ARGOFFSET(%rsp) + movq %rcx,RIP-ARGOFFSET(%rsp) + CFI_REL_OFFSET rip,RIP-ARGOFFSET +@@ -500,6 +702,7 @@ sysret_check: + andl %edi,%edx + jnz sysret_careful + CFI_REMEMBER_STATE ++ pax_exit_kernel_user + /* + * sysretq will re-enable interrupts: + */ +@@ -609,7 +812,7 @@ tracesys: + GLOBAL(int_ret_from_sys_call) + DISABLE_INTERRUPTS(CLBR_NONE) + TRACE_IRQS_OFF +- testl $3,CS-ARGOFFSET(%rsp) ++ testb $3,CS-ARGOFFSET(%rsp) + je retint_restore_args + movl $_TIF_ALLWORK_MASK,%edi + /* edi: mask to check */ +@@ -791,6 +994,16 @@ END(interrupt) + CFI_ADJUST_CFA_OFFSET ORIG_RAX-RBP + call save_args + PARTIAL_FRAME 0 ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ testb $3, CS(%rdi) ++ jnz 1f ++ pax_enter_kernel ++ jmp 2f ++1: pax_enter_kernel_user ++2: ++#else ++ pax_enter_kernel ++#endif + call \func + .endm + +@@ -823,7 +1036,7 @@ ret_from_intr: + CFI_ADJUST_CFA_OFFSET -8 + exit_intr: + GET_THREAD_INFO(%rcx) +- testl $3,CS-ARGOFFSET(%rsp) ++ testb $3,CS-ARGOFFSET(%rsp) + je retint_kernel + + /* Interrupt came from user space */ +@@ -845,12 +1058,14 @@ retint_swapgs: /* return to user-space + * The iretq could re-enable interrupts: + */ + DISABLE_INTERRUPTS(CLBR_ANY) ++ pax_exit_kernel_user + TRACE_IRQS_IRETQ + SWAPGS + jmp restore_args + + retint_restore_args: /* return to kernel space */ + DISABLE_INTERRUPTS(CLBR_ANY) ++ pax_exit_kernel + /* + * The iretq could re-enable interrupts: + */ +@@ -1022,6 +1237,16 @@ ENTRY(\sym) + CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15 + call error_entry + DEFAULT_FRAME 0 ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ testb $3, CS(%rsp) ++ jnz 1f ++ pax_enter_kernel ++ jmp 2f ++1: pax_enter_kernel_user ++2: ++#else ++ pax_enter_kernel ++#endif + movq %rsp,%rdi /* pt_regs pointer */ + xorl %esi,%esi /* no error code */ + call \do_sym +@@ -1039,6 +1264,16 @@ ENTRY(\sym) + CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15 + call save_paranoid + TRACE_IRQS_OFF ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ testb $3, CS(%rsp) ++ jnz 1f ++ pax_enter_kernel ++ jmp 2f ++1: pax_enter_kernel_user ++2: ++#else ++ pax_enter_kernel ++#endif + movq %rsp,%rdi /* pt_regs pointer */ + xorl %esi,%esi /* no error code */ + call \do_sym +@@ -1047,7 +1282,7 @@ ENTRY(\sym) + END(\sym) + .endm + +-#define INIT_TSS_IST(x) PER_CPU_VAR(init_tss) + (TSS_ist + ((x) - 1) * 8) ++#define INIT_TSS_IST(x) (TSS_ist + ((x) - 1) * 8)(%r12) + .macro paranoidzeroentry_ist sym do_sym ist + ENTRY(\sym) + INTR_FRAME +@@ -1057,8 +1292,24 @@ ENTRY(\sym) + CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15 + call save_paranoid + TRACE_IRQS_OFF ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ testb $3, CS(%rsp) ++ jnz 1f ++ pax_enter_kernel ++ jmp 2f ++1: pax_enter_kernel_user ++2: ++#else ++ pax_enter_kernel ++#endif + movq %rsp,%rdi /* pt_regs pointer */ + xorl %esi,%esi /* no error code */ ++#ifdef CONFIG_SMP ++ imul $TSS_size, PER_CPU_VAR(cpu_number), %r12d ++ lea init_tss(%r12), %r12 ++#else ++ lea init_tss(%rip), %r12 ++#endif + subq $EXCEPTION_STKSZ, INIT_TSS_IST(\ist) + call \do_sym + addq $EXCEPTION_STKSZ, INIT_TSS_IST(\ist) +@@ -1075,6 +1326,16 @@ ENTRY(\sym) + CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15 + call error_entry + DEFAULT_FRAME 0 ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ testb $3, CS(%rsp) ++ jnz 1f ++ pax_enter_kernel ++ jmp 2f ++1: pax_enter_kernel_user ++2: ++#else ++ pax_enter_kernel ++#endif + movq %rsp,%rdi /* pt_regs pointer */ + movq ORIG_RAX(%rsp),%rsi /* get error code */ + movq $-1,ORIG_RAX(%rsp) /* no syscall to restart */ +@@ -1094,6 +1355,16 @@ ENTRY(\sym) + call save_paranoid + DEFAULT_FRAME 0 + TRACE_IRQS_OFF ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ testb $3, CS(%rsp) ++ jnz 1f ++ pax_enter_kernel ++ jmp 2f ++1: pax_enter_kernel_user ++2: ++#else ++ pax_enter_kernel ++#endif + movq %rsp,%rdi /* pt_regs pointer */ + movq ORIG_RAX(%rsp),%rsi /* get error code */ + movq $-1,ORIG_RAX(%rsp) /* no syscall to restart */ +@@ -1356,14 +1627,27 @@ ENTRY(paranoid_exit) + TRACE_IRQS_OFF + testl %ebx,%ebx /* swapgs needed? */ + jnz paranoid_restore +- testl $3,CS(%rsp) ++ testb $3,CS(%rsp) + jnz paranoid_userspace ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ pax_exit_kernel ++ TRACE_IRQS_IRETQ 0 ++ SWAPGS_UNSAFE_STACK ++ RESTORE_ALL 8 ++ jmp irq_return ++#endif + paranoid_swapgs: ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ pax_exit_kernel_user ++#else ++ pax_exit_kernel ++#endif + TRACE_IRQS_IRETQ 0 + SWAPGS_UNSAFE_STACK + RESTORE_ALL 8 + jmp irq_return + paranoid_restore: ++ pax_exit_kernel + TRACE_IRQS_IRETQ 0 + RESTORE_ALL 8 + jmp irq_return +@@ -1421,7 +1705,7 @@ ENTRY(error_entry) + movq_cfi r14, R14+8 + movq_cfi r15, R15+8 + xorl %ebx,%ebx +- testl $3,CS+8(%rsp) ++ testb $3,CS+8(%rsp) + je error_kernelspace + error_swapgs: + SWAPGS +@@ -1485,6 +1769,16 @@ ENTRY(nmi) + CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15 + call save_paranoid + DEFAULT_FRAME 0 ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ testb $3, CS(%rsp) ++ jnz 1f ++ pax_enter_kernel ++ jmp 2f ++1: pax_enter_kernel_user ++2: ++#else ++ pax_enter_kernel ++#endif + /* paranoidentry do_nmi, 0; without TRACE_IRQS_OFF */ + movq %rsp,%rdi + movq $-1,%rsi +@@ -1495,11 +1789,25 @@ ENTRY(nmi) + DISABLE_INTERRUPTS(CLBR_NONE) + testl %ebx,%ebx /* swapgs needed? */ + jnz nmi_restore +- testl $3,CS(%rsp) ++ testb $3,CS(%rsp) + jnz nmi_userspace ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ pax_exit_kernel ++ SWAPGS_UNSAFE_STACK ++ RESTORE_ALL 8 ++ jmp irq_return ++#endif + nmi_swapgs: ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ pax_exit_kernel_user ++#else ++ pax_exit_kernel ++#endif + SWAPGS_UNSAFE_STACK ++ RESTORE_ALL 8 ++ jmp irq_return + nmi_restore: ++ pax_exit_kernel + RESTORE_ALL 8 + jmp irq_return + nmi_userspace: +diff -urNp linux-2.6.38.3/arch/x86/kernel/ftrace.c linux-2.6.38.3/arch/x86/kernel/ftrace.c +--- linux-2.6.38.3/arch/x86/kernel/ftrace.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/ftrace.c 2011-04-17 15:57:32.000000000 -0400 +@@ -177,7 +177,9 @@ void ftrace_nmi_enter(void) + + if (atomic_inc_return(&nmi_running) & MOD_CODE_WRITE_FLAG) { + smp_rmb(); ++ pax_open_kernel(); + ftrace_mod_code(); ++ pax_close_kernel(); + atomic_inc(&nmi_update_count); + } + /* Must have previous changes seen before executions */ +@@ -271,6 +273,8 @@ ftrace_modify_code(unsigned long ip, uns + { + unsigned char replaced[MCOUNT_INSN_SIZE]; + ++ ip = ktla_ktva(ip); ++ + /* + * Note: Due to modules and __init, code can + * disappear and change, we need to protect against faulting +@@ -327,7 +331,7 @@ int ftrace_update_ftrace_func(ftrace_fun + unsigned char old[MCOUNT_INSN_SIZE], *new; + int ret; + +- memcpy(old, &ftrace_call, MCOUNT_INSN_SIZE); ++ memcpy(old, (void *)ktla_ktva((unsigned long)ftrace_call), MCOUNT_INSN_SIZE); + new = ftrace_call_replace(ip, (unsigned long)func); + ret = ftrace_modify_code(ip, old, new); + +@@ -353,6 +357,8 @@ static int ftrace_mod_jmp(unsigned long + { + unsigned char code[MCOUNT_INSN_SIZE]; + ++ ip = ktla_ktva(ip); ++ + if (probe_kernel_read(code, (void *)ip, MCOUNT_INSN_SIZE)) + return -EFAULT; + +diff -urNp linux-2.6.38.3/arch/x86/kernel/head32.c linux-2.6.38.3/arch/x86/kernel/head32.c +--- linux-2.6.38.3/arch/x86/kernel/head32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/head32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -19,6 +19,7 @@ + #include <asm/io_apic.h> + #include <asm/bios_ebda.h> + #include <asm/tlbflush.h> ++#include <asm/boot.h> + + static void __init i386_default_early_setup(void) + { +@@ -43,7 +44,7 @@ void __init i386_start_kernel(void) + memblock_x86_reserve_range(PAGE_SIZE, PAGE_SIZE + PAGE_SIZE, "EX TRAMPOLINE"); + #endif + +- memblock_x86_reserve_range(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS"); ++ memblock_x86_reserve_range(LOAD_PHYSICAL_ADDR, __pa_symbol(&__bss_stop), "TEXT DATA BSS"); + + #ifdef CONFIG_BLK_DEV_INITRD + /* Reserve INITRD */ +diff -urNp linux-2.6.38.3/arch/x86/kernel/head_32.S linux-2.6.38.3/arch/x86/kernel/head_32.S +--- linux-2.6.38.3/arch/x86/kernel/head_32.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/head_32.S 2011-04-17 16:02:16.000000000 -0400 +@@ -25,6 +25,12 @@ + /* Physical address */ + #define pa(X) ((X) - __PAGE_OFFSET) + ++#ifdef CONFIG_PAX_KERNEXEC ++#define ta(X) (X) ++#else ++#define ta(X) ((X) - __PAGE_OFFSET) ++#endif ++ + /* + * References to members of the new_cpu_data structure. + */ +@@ -54,11 +60,7 @@ + * and small than max_low_pfn, otherwise will waste some page table entries + */ + +-#if PTRS_PER_PMD > 1 +-#define PAGE_TABLE_SIZE(pages) (((pages) / PTRS_PER_PMD) + PTRS_PER_PGD) +-#else +-#define PAGE_TABLE_SIZE(pages) ((pages) / PTRS_PER_PGD) +-#endif ++#define PAGE_TABLE_SIZE(pages) ((pages) / PTRS_PER_PTE) + + /* Number of possible pages in the lowmem region */ + LOWMEM_PAGES = (((1<<32) - __PAGE_OFFSET) >> PAGE_SHIFT) +@@ -77,6 +79,12 @@ INIT_MAP_SIZE = PAGE_TABLE_SIZE(KERNEL_P + RESERVE_BRK(pagetables, INIT_MAP_SIZE) + + /* ++ * Real beginning of normal "text" segment ++ */ ++ENTRY(stext) ++ENTRY(_stext) ++ ++/* + * 32-bit kernel entrypoint; only used by the boot CPU. On entry, + * %esi points to the real-mode code as a 32-bit pointer. + * CS and DS must be 4 GB flat segments, but we don't depend on +@@ -84,6 +92,13 @@ RESERVE_BRK(pagetables, INIT_MAP_SIZE) + * can. + */ + __HEAD ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ jmp startup_32 ++/* PaX: fill first page in .text with int3 to catch NULL derefs in kernel mode */ ++.fill PAGE_SIZE-5,1,0xcc ++#endif ++ + ENTRY(startup_32) + movl pa(stack_start),%ecx + +@@ -105,6 +120,57 @@ ENTRY(startup_32) + 2: + leal -__PAGE_OFFSET(%ecx),%esp + ++#ifdef CONFIG_SMP ++ movl $pa(cpu_gdt_table),%edi ++ movl $__per_cpu_load,%eax ++ movw %ax,__KERNEL_PERCPU + 2(%edi) ++ rorl $16,%eax ++ movb %al,__KERNEL_PERCPU + 4(%edi) ++ movb %ah,__KERNEL_PERCPU + 7(%edi) ++ movl $__per_cpu_end - 1,%eax ++ subl $__per_cpu_start,%eax ++ movw %ax,__KERNEL_PERCPU + 0(%edi) ++#endif ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ movl $NR_CPUS,%ecx ++ movl $pa(cpu_gdt_table),%edi ++1: ++ movl $((((__PAGE_OFFSET-1) & 0xf0000000) >> 12) | 0x00c09700),GDT_ENTRY_KERNEL_DS * 8 + 4(%edi) ++ movl $((((__PAGE_OFFSET-1) & 0xf0000000) >> 12) | 0x00c0fb00),GDT_ENTRY_DEFAULT_USER_CS * 8 + 4(%edi) ++ movl $((((__PAGE_OFFSET-1) & 0xf0000000) >> 12) | 0x00c0f300),GDT_ENTRY_DEFAULT_USER_DS * 8 + 4(%edi) ++ addl $PAGE_SIZE_asm,%edi ++ loop 1b ++#endif ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ movl $pa(boot_gdt),%edi ++ movl $__LOAD_PHYSICAL_ADDR,%eax ++ movw %ax,__BOOT_CS + 2(%edi) ++ rorl $16,%eax ++ movb %al,__BOOT_CS + 4(%edi) ++ movb %ah,__BOOT_CS + 7(%edi) ++ rorl $16,%eax ++ ++ ljmp $(__BOOT_CS),$1f ++1: ++ ++ movl $NR_CPUS,%ecx ++ movl $pa(cpu_gdt_table),%edi ++ addl $__PAGE_OFFSET,%eax ++1: ++ movw %ax,__KERNEL_CS + 2(%edi) ++ movw %ax,__KERNEXEC_KERNEL_CS + 2(%edi) ++ rorl $16,%eax ++ movb %al,__KERNEL_CS + 4(%edi) ++ movb %al,__KERNEXEC_KERNEL_CS + 4(%edi) ++ movb %ah,__KERNEL_CS + 7(%edi) ++ movb %ah,__KERNEXEC_KERNEL_CS + 7(%edi) ++ rorl $16,%eax ++ addl $PAGE_SIZE_asm,%edi ++ loop 1b ++#endif ++ + /* + * Clear BSS first so that there are no surprises... + */ +@@ -195,8 +261,11 @@ ENTRY(startup_32) + movl %eax, pa(max_pfn_mapped) + + /* Do early initialization of the fixmap area */ +- movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR,%eax +- movl %eax,pa(initial_pg_pmd+0x1000*KPMDS-8) ++#ifdef CONFIG_COMPAT_VDSO ++ movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR+_PAGE_USER,pa(initial_pg_pmd+0x1000*KPMDS-8) ++#else ++ movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR,pa(initial_pg_pmd+0x1000*KPMDS-8) ++#endif + #else /* Not PAE */ + + page_pde_offset = (__PAGE_OFFSET >> 20); +@@ -226,8 +295,11 @@ page_pde_offset = (__PAGE_OFFSET >> 20); + movl %eax, pa(max_pfn_mapped) + + /* Do early initialization of the fixmap area */ +- movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR,%eax +- movl %eax,pa(initial_page_table+0xffc) ++#ifdef CONFIG_COMPAT_VDSO ++ movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR+_PAGE_USER,pa(initial_page_table+0xffc) ++#else ++ movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR,pa(initial_page_table+0xffc) ++#endif + #endif + + #ifdef CONFIG_PARAVIRT +@@ -241,9 +313,7 @@ page_pde_offset = (__PAGE_OFFSET >> 20); + cmpl $num_subarch_entries, %eax + jae bad_subarch + +- movl pa(subarch_entries)(,%eax,4), %eax +- subl $__PAGE_OFFSET, %eax +- jmp *%eax ++ jmp *pa(subarch_entries)(,%eax,4) + + bad_subarch: + WEAK(lguest_entry) +@@ -255,10 +325,10 @@ WEAK(xen_entry) + __INITDATA + + subarch_entries: +- .long default_entry /* normal x86/PC */ +- .long lguest_entry /* lguest hypervisor */ +- .long xen_entry /* Xen hypervisor */ +- .long default_entry /* Moorestown MID */ ++ .long pa(default_entry) /* normal x86/PC */ ++ .long pa(lguest_entry) /* lguest hypervisor */ ++ .long pa(xen_entry) /* Xen hypervisor */ ++ .long pa(default_entry) /* Moorestown MID */ + num_subarch_entries = (. - subarch_entries) / 4 + .previous + #else +@@ -312,6 +382,7 @@ default_entry: + orl %edx,%eax + movl %eax,%cr4 + ++#ifdef CONFIG_X86_PAE + testb $X86_CR4_PAE, %al # check if PAE is enabled + jz 6f + +@@ -340,6 +411,9 @@ default_entry: + /* Make changes effective */ + wrmsr + ++ btsl $_PAGE_BIT_NX-32,pa(__supported_pte_mask+4) ++#endif ++ + 6: + + /* +@@ -443,7 +517,7 @@ is386: movl $2,%ecx # set MP + 1: movl $(__KERNEL_DS),%eax # reload all the segment registers + movl %eax,%ss # after changing gdt. + +- movl $(__USER_DS),%eax # DS/ES contains default USER segment ++# movl $(__KERNEL_DS),%eax # DS/ES contains default KERNEL segment + movl %eax,%ds + movl %eax,%es + +@@ -457,15 +531,22 @@ is386: movl $2,%ecx # set MP + */ + cmpb $0,ready + jne 1f +- movl $gdt_page,%eax ++ movl $cpu_gdt_table,%eax + movl $stack_canary,%ecx ++#ifdef CONFIG_SMP ++ addl $__per_cpu_load,%ecx ++#endif + movw %cx, 8 * GDT_ENTRY_STACK_CANARY + 2(%eax) + shrl $16, %ecx + movb %cl, 8 * GDT_ENTRY_STACK_CANARY + 4(%eax) + movb %ch, 8 * GDT_ENTRY_STACK_CANARY + 7(%eax) + 1: +-#endif + movl $(__KERNEL_STACK_CANARY),%eax ++#elif defined(CONFIG_PAX_MEMORY_UDEREF) ++ movl $(__USER_DS),%eax ++#else ++ xorl %eax,%eax ++#endif + movl %eax,%gs + + xorl %eax,%eax # Clear LDT +@@ -558,22 +639,22 @@ early_page_fault: + jmp early_fault + + early_fault: +- cld + #ifdef CONFIG_PRINTK ++ cmpl $1,%ss:early_recursion_flag ++ je hlt_loop ++ incl %ss:early_recursion_flag ++ cld + pusha + movl $(__KERNEL_DS),%eax + movl %eax,%ds + movl %eax,%es +- cmpl $2,early_recursion_flag +- je hlt_loop +- incl early_recursion_flag + movl %cr2,%eax + pushl %eax + pushl %edx /* trapno */ + pushl $fault_msg + call printk ++; call dump_stack + #endif +- call dump_stack + hlt_loop: + hlt + jmp hlt_loop +@@ -581,8 +662,11 @@ hlt_loop: + /* This is the default interrupt "handler" :-) */ + ALIGN + ignore_int: +- cld + #ifdef CONFIG_PRINTK ++ cmpl $2,%ss:early_recursion_flag ++ je hlt_loop ++ incl %ss:early_recursion_flag ++ cld + pushl %eax + pushl %ecx + pushl %edx +@@ -591,9 +675,6 @@ ignore_int: + movl $(__KERNEL_DS),%eax + movl %eax,%ds + movl %eax,%es +- cmpl $2,early_recursion_flag +- je hlt_loop +- incl early_recursion_flag + pushl 16(%esp) + pushl 24(%esp) + pushl 32(%esp) +@@ -622,29 +703,43 @@ ENTRY(initial_code) + /* + * BSS section + */ +-__PAGE_ALIGNED_BSS +- .align PAGE_SIZE_asm + #ifdef CONFIG_X86_PAE ++.section .initial_pg_pmd,"a",@progbits + initial_pg_pmd: + .fill 1024*KPMDS,4,0 + #else ++.section .initial_page_table,"a",@progbits + ENTRY(initial_page_table) + .fill 1024,4,0 + #endif ++.section .initial_pg_fixmap,"a",@progbits + initial_pg_fixmap: + .fill 1024,4,0 ++.section .empty_zero_page,"a",@progbits + ENTRY(empty_zero_page) + .fill 4096,1,0 ++.section .swapper_pg_dir,"a",@progbits + ENTRY(swapper_pg_dir) ++#ifdef CONFIG_X86_PAE ++ .fill 4,8,0 ++#else + .fill 1024,4,0 ++#endif ++ ++/* ++ * The IDT has to be page-aligned to simplify the Pentium ++ * F0 0F bug workaround.. We have a special link segment ++ * for this. ++ */ ++.section .idt,"a",@progbits ++ENTRY(idt_table) ++ .fill 256,8,0 + + /* + * This starts the data section. + */ + #ifdef CONFIG_X86_PAE +-__PAGE_ALIGNED_DATA +- /* Page-aligned for the benefit of paravirt? */ +- .align PAGE_SIZE_asm ++.section .initial_page_table,"a",@progbits + ENTRY(initial_page_table) + .long pa(initial_pg_pmd+PGD_IDENT_ATTR),0 /* low identity map */ + # if KPMDS == 3 +@@ -663,18 +758,27 @@ ENTRY(initial_page_table) + # error "Kernel PMDs should be 1, 2 or 3" + # endif + .align PAGE_SIZE_asm /* needs to be page-sized too */ ++ ++#ifdef CONFIG_PAX_PER_CPU_PGD ++ENTRY(cpu_pgd) ++ .rept NR_CPUS ++ .fill 4,8,0 ++ .endr ++#endif ++ + #endif + + .data + .balign 4 + ENTRY(stack_start) +- .long init_thread_union+THREAD_SIZE ++ .long init_thread_union+THREAD_SIZE-8 ++ ++ready: .byte 0 + ++.section .rodata,"a",@progbits + early_recursion_flag: + .long 0 + +-ready: .byte 0 +- + int_msg: + .asciz "Unknown interrupt or fault at: %p %p %p\n" + +@@ -707,7 +811,7 @@ fault_msg: + .word 0 # 32 bit align gdt_desc.address + boot_gdt_descr: + .word __BOOT_DS+7 +- .long boot_gdt - __PAGE_OFFSET ++ .long pa(boot_gdt) + + .word 0 # 32-bit align idt_desc.address + idt_descr: +@@ -718,7 +822,7 @@ idt_descr: + .word 0 # 32 bit align gdt_desc.address + ENTRY(early_gdt_descr) + .word GDT_ENTRIES*8-1 +- .long gdt_page /* Overwritten for secondary CPUs */ ++ .long cpu_gdt_table /* Overwritten for secondary CPUs */ + + /* + * The boot_gdt must mirror the equivalent in setup.S and is +@@ -727,5 +831,65 @@ ENTRY(early_gdt_descr) + .align L1_CACHE_BYTES + ENTRY(boot_gdt) + .fill GDT_ENTRY_BOOT_CS,8,0 +- .quad 0x00cf9a000000ffff /* kernel 4GB code at 0x00000000 */ +- .quad 0x00cf92000000ffff /* kernel 4GB data at 0x00000000 */ ++ .quad 0x00cf9b000000ffff /* kernel 4GB code at 0x00000000 */ ++ .quad 0x00cf93000000ffff /* kernel 4GB data at 0x00000000 */ ++ ++ .align PAGE_SIZE_asm ++ENTRY(cpu_gdt_table) ++ .rept NR_CPUS ++ .quad 0x0000000000000000 /* NULL descriptor */ ++ .quad 0x0000000000000000 /* 0x0b reserved */ ++ .quad 0x0000000000000000 /* 0x13 reserved */ ++ .quad 0x0000000000000000 /* 0x1b reserved */ ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ .quad 0x00cf9b000000ffff /* 0x20 alternate kernel 4GB code at 0x00000000 */ ++#else ++ .quad 0x0000000000000000 /* 0x20 unused */ ++#endif ++ ++ .quad 0x0000000000000000 /* 0x28 unused */ ++ .quad 0x0000000000000000 /* 0x33 TLS entry 1 */ ++ .quad 0x0000000000000000 /* 0x3b TLS entry 2 */ ++ .quad 0x0000000000000000 /* 0x43 TLS entry 3 */ ++ .quad 0x0000000000000000 /* 0x4b reserved */ ++ .quad 0x0000000000000000 /* 0x53 reserved */ ++ .quad 0x0000000000000000 /* 0x5b reserved */ ++ ++ .quad 0x00cf9b000000ffff /* 0x60 kernel 4GB code at 0x00000000 */ ++ .quad 0x00cf93000000ffff /* 0x68 kernel 4GB data at 0x00000000 */ ++ .quad 0x00cffb000000ffff /* 0x73 user 4GB code at 0x00000000 */ ++ .quad 0x00cff3000000ffff /* 0x7b user 4GB data at 0x00000000 */ ++ ++ .quad 0x0000000000000000 /* 0x80 TSS descriptor */ ++ .quad 0x0000000000000000 /* 0x88 LDT descriptor */ ++ ++ /* ++ * Segments used for calling PnP BIOS have byte granularity. ++ * The code segments and data segments have fixed 64k limits, ++ * the transfer segment sizes are set at run time. ++ */ ++ .quad 0x00409b000000ffff /* 0x90 32-bit code */ ++ .quad 0x00009b000000ffff /* 0x98 16-bit code */ ++ .quad 0x000093000000ffff /* 0xa0 16-bit data */ ++ .quad 0x0000930000000000 /* 0xa8 16-bit data */ ++ .quad 0x0000930000000000 /* 0xb0 16-bit data */ ++ ++ /* ++ * The APM segments have byte granularity and their bases ++ * are set at run time. All have 64k limits. ++ */ ++ .quad 0x00409b000000ffff /* 0xb8 APM CS code */ ++ .quad 0x00009b000000ffff /* 0xc0 APM CS 16 code (16 bit) */ ++ .quad 0x004093000000ffff /* 0xc8 APM DS data */ ++ ++ .quad 0x00c0930000000000 /* 0xd0 - ESPFIX SS */ ++ .quad 0x0040930000000000 /* 0xd8 - PERCPU */ ++ .quad 0x0040910000000018 /* 0xe0 - STACK_CANARY */ ++ .quad 0x0000000000000000 /* 0xe8 - PCIBIOS_CS */ ++ .quad 0x0000000000000000 /* 0xf0 - PCIBIOS_DS */ ++ .quad 0x0000000000000000 /* 0xf8 - GDT entry 31: double-fault TSS */ ++ ++ /* Be sure this is zeroed to avoid false validations in Xen */ ++ .fill PAGE_SIZE_asm - GDT_SIZE,1,0 ++ .endr +diff -urNp linux-2.6.38.3/arch/x86/kernel/head_64.S linux-2.6.38.3/arch/x86/kernel/head_64.S +--- linux-2.6.38.3/arch/x86/kernel/head_64.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/head_64.S 2011-04-17 15:57:32.000000000 -0400 +@@ -19,6 +19,7 @@ + #include <asm/cache.h> + #include <asm/processor-flags.h> + #include <asm/percpu.h> ++#include <asm/cpufeature.h> + + #ifdef CONFIG_PARAVIRT + #include <asm/asm-offsets.h> +@@ -38,6 +39,10 @@ L4_PAGE_OFFSET = pgd_index(__PAGE_OFFSET + L3_PAGE_OFFSET = pud_index(__PAGE_OFFSET) + L4_START_KERNEL = pgd_index(__START_KERNEL_map) + L3_START_KERNEL = pud_index(__START_KERNEL_map) ++L4_VMALLOC_START = pgd_index(VMALLOC_START) ++L3_VMALLOC_START = pud_index(VMALLOC_START) ++L4_VMEMMAP_START = pgd_index(VMEMMAP_START) ++L3_VMEMMAP_START = pud_index(VMEMMAP_START) + + .text + __HEAD +@@ -85,35 +90,22 @@ startup_64: + */ + addq %rbp, init_level4_pgt + 0(%rip) + addq %rbp, init_level4_pgt + (L4_PAGE_OFFSET*8)(%rip) ++ addq %rbp, init_level4_pgt + (L4_VMALLOC_START*8)(%rip) ++ addq %rbp, init_level4_pgt + (L4_VMEMMAP_START*8)(%rip) + addq %rbp, init_level4_pgt + (L4_START_KERNEL*8)(%rip) + + addq %rbp, level3_ident_pgt + 0(%rip) ++#ifndef CONFIG_XEN ++ addq %rbp, level3_ident_pgt + 8(%rip) ++#endif + +- addq %rbp, level3_kernel_pgt + (510*8)(%rip) +- addq %rbp, level3_kernel_pgt + (511*8)(%rip) ++ addq %rbp, level3_vmemmap_pgt + (L3_VMEMMAP_START*8)(%rip) + +- addq %rbp, level2_fixmap_pgt + (506*8)(%rip) ++ addq %rbp, level3_kernel_pgt + (L3_START_KERNEL*8)(%rip) ++ addq %rbp, level3_kernel_pgt + (L3_START_KERNEL*8+8)(%rip) + +- /* Add an Identity mapping if I am above 1G */ +- leaq _text(%rip), %rdi +- andq $PMD_PAGE_MASK, %rdi +- +- movq %rdi, %rax +- shrq $PUD_SHIFT, %rax +- andq $(PTRS_PER_PUD - 1), %rax +- jz ident_complete +- +- leaq (level2_spare_pgt - __START_KERNEL_map + _KERNPG_TABLE)(%rbp), %rdx +- leaq level3_ident_pgt(%rip), %rbx +- movq %rdx, 0(%rbx, %rax, 8) +- +- movq %rdi, %rax +- shrq $PMD_SHIFT, %rax +- andq $(PTRS_PER_PMD - 1), %rax +- leaq __PAGE_KERNEL_IDENT_LARGE_EXEC(%rdi), %rdx +- leaq level2_spare_pgt(%rip), %rbx +- movq %rdx, 0(%rbx, %rax, 8) +-ident_complete: ++ addq %rbp, level2_fixmap_pgt + (506*8)(%rip) ++ addq %rbp, level2_fixmap_pgt + (507*8)(%rip) + + /* + * Fixup the kernel text+data virtual addresses. Note that +@@ -161,8 +153,8 @@ ENTRY(secondary_startup_64) + * after the boot processor executes this code. + */ + +- /* Enable PAE mode and PGE */ +- movl $(X86_CR4_PAE | X86_CR4_PGE), %eax ++ /* Enable PAE mode and PSE/PGE */ ++ movl $(X86_CR4_PSE | X86_CR4_PAE | X86_CR4_PGE), %eax + movq %rax, %cr4 + + /* Setup early boot stage 4 level pagetables. */ +@@ -184,9 +176,14 @@ ENTRY(secondary_startup_64) + movl $MSR_EFER, %ecx + rdmsr + btsl $_EFER_SCE, %eax /* Enable System Call */ +- btl $20,%edi /* No Execute supported? */ ++ btl $(X86_FEATURE_NX & 31),%edi /* No Execute supported? */ + jnc 1f + btsl $_EFER_NX, %eax ++ leaq init_level4_pgt(%rip), %rdi ++ btsq $_PAGE_BIT_NX, 8*L4_PAGE_OFFSET(%rdi) ++ btsq $_PAGE_BIT_NX, 8*L4_VMALLOC_START(%rdi) ++ btsq $_PAGE_BIT_NX, 8*L4_VMEMMAP_START(%rdi) ++ btsq $_PAGE_BIT_NX, __supported_pte_mask(%rip) + 1: wrmsr /* Make changes effective */ + + /* Setup cr0 */ +@@ -270,7 +267,7 @@ ENTRY(secondary_startup_64) + bad_address: + jmp bad_address + +- .section ".init.text","ax" ++ __INIT + #ifdef CONFIG_EARLY_PRINTK + .globl early_idt_handlers + early_idt_handlers: +@@ -315,18 +312,23 @@ ENTRY(early_idt_handler) + #endif /* EARLY_PRINTK */ + 1: hlt + jmp 1b ++ .previous + + #ifdef CONFIG_EARLY_PRINTK ++ __INITDATA + early_recursion_flag: + .long 0 ++ .previous + ++ .section .rodata,"a",@progbits + early_idt_msg: + .asciz "PANIC: early exception %02lx rip %lx:%lx error %lx cr2 %lx\n" + early_idt_ripmsg: + .asciz "RIP %s\n" +-#endif /* CONFIG_EARLY_PRINTK */ + .previous ++#endif /* CONFIG_EARLY_PRINTK */ + ++ .section .rodata,"a",@progbits + #define NEXT_PAGE(name) \ + .balign PAGE_SIZE; \ + ENTRY(name) +@@ -339,7 +341,6 @@ ENTRY(name) + i = i + 1 ; \ + .endr + +- .data + /* + * This default setting generates an ident mapping at address 0x100000 + * and a mapping for the kernel that precisely maps virtual address +@@ -350,13 +351,36 @@ NEXT_PAGE(init_level4_pgt) + .quad level3_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE + .org init_level4_pgt + L4_PAGE_OFFSET*8, 0 + .quad level3_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE ++ .org init_level4_pgt + L4_VMALLOC_START*8, 0 ++ .quad level3_vmalloc_pgt - __START_KERNEL_map + _KERNPG_TABLE ++ .org init_level4_pgt + L4_VMEMMAP_START*8, 0 ++ .quad level3_vmemmap_pgt - __START_KERNEL_map + _KERNPG_TABLE + .org init_level4_pgt + L4_START_KERNEL*8, 0 + /* (2^48-(2*1024*1024*1024))/(2^39) = 511 */ + .quad level3_kernel_pgt - __START_KERNEL_map + _PAGE_TABLE + ++#ifdef CONFIG_PAX_PER_CPU_PGD ++NEXT_PAGE(cpu_pgd) ++ .rept NR_CPUS ++ .fill 512,8,0 ++ .endr ++#endif ++ + NEXT_PAGE(level3_ident_pgt) + .quad level2_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE ++#ifdef CONFIG_XEN + .fill 511,8,0 ++#else ++ .quad level2_ident_pgt + PAGE_SIZE - __START_KERNEL_map + _KERNPG_TABLE ++ .fill 510,8,0 ++#endif ++ ++NEXT_PAGE(level3_vmalloc_pgt) ++ .fill 512,8,0 ++ ++NEXT_PAGE(level3_vmemmap_pgt) ++ .fill L3_VMEMMAP_START,8,0 ++ .quad level2_vmemmap_pgt - __START_KERNEL_map + _KERNPG_TABLE + + NEXT_PAGE(level3_kernel_pgt) + .fill L3_START_KERNEL,8,0 +@@ -364,20 +388,23 @@ NEXT_PAGE(level3_kernel_pgt) + .quad level2_kernel_pgt - __START_KERNEL_map + _KERNPG_TABLE + .quad level2_fixmap_pgt - __START_KERNEL_map + _PAGE_TABLE + ++NEXT_PAGE(level2_vmemmap_pgt) ++ .fill 512,8,0 ++ + NEXT_PAGE(level2_fixmap_pgt) +- .fill 506,8,0 +- .quad level1_fixmap_pgt - __START_KERNEL_map + _PAGE_TABLE +- /* 8MB reserved for vsyscalls + a 2MB hole = 4 + 1 entries */ +- .fill 5,8,0 ++ .fill 507,8,0 ++ .quad level1_vsyscall_pgt - __START_KERNEL_map + _PAGE_TABLE ++ /* 6MB reserved for vsyscalls + a 2MB hole = 3 + 1 entries */ ++ .fill 4,8,0 + +-NEXT_PAGE(level1_fixmap_pgt) ++NEXT_PAGE(level1_vsyscall_pgt) + .fill 512,8,0 + +-NEXT_PAGE(level2_ident_pgt) +- /* Since I easily can, map the first 1G. ++ /* Since I easily can, map the first 2G. + * Don't set NX because code runs from these pages. + */ +- PMDS(0, __PAGE_KERNEL_IDENT_LARGE_EXEC, PTRS_PER_PMD) ++NEXT_PAGE(level2_ident_pgt) ++ PMDS(0, __PAGE_KERNEL_IDENT_LARGE_EXEC, 2*PTRS_PER_PMD) + + NEXT_PAGE(level2_kernel_pgt) + /* +@@ -390,33 +417,55 @@ NEXT_PAGE(level2_kernel_pgt) + * If you want to increase this then increase MODULES_VADDR + * too.) + */ +- PMDS(0, __PAGE_KERNEL_LARGE_EXEC, +- KERNEL_IMAGE_SIZE/PMD_SIZE) +- +-NEXT_PAGE(level2_spare_pgt) +- .fill 512, 8, 0 ++ PMDS(0, __PAGE_KERNEL_LARGE_EXEC, KERNEL_IMAGE_SIZE/PMD_SIZE) + + #undef PMDS + #undef NEXT_PAGE + +- .data ++ .align PAGE_SIZE ++ENTRY(cpu_gdt_table) ++ .rept NR_CPUS ++ .quad 0x0000000000000000 /* NULL descriptor */ ++ .quad 0x00cf9b000000ffff /* __KERNEL32_CS */ ++ .quad 0x00af9b000000ffff /* __KERNEL_CS */ ++ .quad 0x00cf93000000ffff /* __KERNEL_DS */ ++ .quad 0x00cffb000000ffff /* __USER32_CS */ ++ .quad 0x00cff3000000ffff /* __USER_DS, __USER32_DS */ ++ .quad 0x00affb000000ffff /* __USER_CS */ ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ .quad 0x00af9b000000ffff /* __KERNEXEC_KERNEL_CS */ ++#else ++ .quad 0x0 /* unused */ ++#endif ++ ++ .quad 0,0 /* TSS */ ++ .quad 0,0 /* LDT */ ++ .quad 0,0,0 /* three TLS descriptors */ ++ .quad 0x0000f40000000000 /* node/CPU stored in limit */ ++ /* asm/segment.h:GDT_ENTRIES must match this */ ++ ++ /* zero the remaining page */ ++ .fill PAGE_SIZE / 8 - GDT_ENTRIES,8,0 ++ .endr ++ + .align 16 + .globl early_gdt_descr + early_gdt_descr: + .word GDT_ENTRIES*8-1 + early_gdt_descr_base: +- .quad INIT_PER_CPU_VAR(gdt_page) ++ .quad cpu_gdt_table + + ENTRY(phys_base) + /* This must match the first entry in level2_kernel_pgt */ + .quad 0x0000000000000000 + + #include "../../x86/xen/xen-head.S" +- +- .section .bss, "aw", @nobits ++ ++ .section .rodata,"a",@progbits + .align L1_CACHE_BYTES + ENTRY(idt_table) +- .skip IDT_ENTRIES * 16 ++ .fill 512,8,0 + + __PAGE_ALIGNED_BSS + .align PAGE_SIZE +diff -urNp linux-2.6.38.3/arch/x86/kernel/i386_ksyms_32.c linux-2.6.38.3/arch/x86/kernel/i386_ksyms_32.c +--- linux-2.6.38.3/arch/x86/kernel/i386_ksyms_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/i386_ksyms_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -20,8 +20,12 @@ extern void cmpxchg8b_emu(void); + EXPORT_SYMBOL(cmpxchg8b_emu); + #endif + ++EXPORT_SYMBOL_GPL(cpu_gdt_table); ++ + /* Networking helper routines. */ + EXPORT_SYMBOL(csum_partial_copy_generic); ++EXPORT_SYMBOL(csum_partial_copy_generic_to_user); ++EXPORT_SYMBOL(csum_partial_copy_generic_from_user); + + EXPORT_SYMBOL(__get_user_1); + EXPORT_SYMBOL(__get_user_2); +@@ -36,3 +40,7 @@ EXPORT_SYMBOL(strstr); + + EXPORT_SYMBOL(csum_partial); + EXPORT_SYMBOL(empty_zero_page); ++ ++#ifdef CONFIG_PAX_KERNEXEC ++EXPORT_SYMBOL(__LOAD_PHYSICAL_ADDR); ++#endif +diff -urNp linux-2.6.38.3/arch/x86/kernel/init_task.c linux-2.6.38.3/arch/x86/kernel/init_task.c +--- linux-2.6.38.3/arch/x86/kernel/init_task.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/init_task.c 2011-04-17 15:57:32.000000000 -0400 +@@ -20,8 +20,7 @@ static struct sighand_struct init_sighan + * way process stacks are handled. This is done by having a special + * "init_task" linker map entry.. + */ +-union thread_union init_thread_union __init_task_data = +- { INIT_THREAD_INFO(init_task) }; ++union thread_union init_thread_union __init_task_data; + + /* + * Initial task structure. +@@ -38,5 +37,5 @@ EXPORT_SYMBOL(init_task); + * section. Since TSS's are completely CPU-local, we want them + * on exact cacheline boundaries, to eliminate cacheline ping-pong. + */ +-DEFINE_PER_CPU_SHARED_ALIGNED(struct tss_struct, init_tss) = INIT_TSS; +- ++struct tss_struct init_tss[NR_CPUS] ____cacheline_internodealigned_in_smp = { [0 ... NR_CPUS-1] = INIT_TSS }; ++EXPORT_SYMBOL(init_tss); +diff -urNp linux-2.6.38.3/arch/x86/kernel/ioport.c linux-2.6.38.3/arch/x86/kernel/ioport.c +--- linux-2.6.38.3/arch/x86/kernel/ioport.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/ioport.c 2011-04-17 15:57:32.000000000 -0400 +@@ -6,6 +6,7 @@ + #include <linux/sched.h> + #include <linux/kernel.h> + #include <linux/capability.h> ++#include <linux/security.h> + #include <linux/errno.h> + #include <linux/types.h> + #include <linux/ioport.h> +@@ -41,6 +42,12 @@ asmlinkage long sys_ioperm(unsigned long + + if ((from + num <= from) || (from + num > IO_BITMAP_BITS)) + return -EINVAL; ++#ifdef CONFIG_GRKERNSEC_IO ++ if (turn_on && grsec_disable_privio) { ++ gr_handle_ioperm(); ++ return -EPERM; ++ } ++#endif + if (turn_on && !capable(CAP_SYS_RAWIO)) + return -EPERM; + +@@ -67,7 +74,7 @@ asmlinkage long sys_ioperm(unsigned long + * because the ->io_bitmap_max value must match the bitmap + * contents: + */ +- tss = &per_cpu(init_tss, get_cpu()); ++ tss = init_tss + get_cpu(); + + set_bitmap(t->io_bitmap_ptr, from, num, !turn_on); + +@@ -112,6 +119,12 @@ long sys_iopl(unsigned int level, struct + return -EINVAL; + /* Trying to gain more privileges? */ + if (level > old) { ++#ifdef CONFIG_GRKERNSEC_IO ++ if (grsec_disable_privio) { ++ gr_handle_iopl(); ++ return -EPERM; ++ } ++#endif + if (!capable(CAP_SYS_RAWIO)) + return -EPERM; + } +diff -urNp linux-2.6.38.3/arch/x86/kernel/irq_32.c linux-2.6.38.3/arch/x86/kernel/irq_32.c +--- linux-2.6.38.3/arch/x86/kernel/irq_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/irq_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -36,7 +36,7 @@ static int check_stack_overflow(void) + __asm__ __volatile__("andl %%esp,%0" : + "=r" (sp) : "0" (THREAD_SIZE - 1)); + +- return sp < (sizeof(struct thread_info) + STACK_WARN); ++ return sp < STACK_WARN; + } + + static void print_stack_overflow(void) +@@ -54,8 +54,8 @@ static inline void print_stack_overflow( + * per-CPU IRQ handling contexts (thread information and stack) + */ + union irq_ctx { +- struct thread_info tinfo; +- u32 stack[THREAD_SIZE/sizeof(u32)]; ++ unsigned long previous_esp; ++ u32 stack[THREAD_SIZE/sizeof(u32)]; + } __attribute__((aligned(THREAD_SIZE))); + + static DEFINE_PER_CPU(union irq_ctx *, hardirq_ctx); +@@ -75,10 +75,9 @@ static void call_on_stack(void *func, vo + static inline int + execute_on_irq_stack(int overflow, struct irq_desc *desc, int irq) + { +- union irq_ctx *curctx, *irqctx; ++ union irq_ctx *irqctx; + u32 *isp, arg1, arg2; + +- curctx = (union irq_ctx *) current_thread_info(); + irqctx = __this_cpu_read(hardirq_ctx); + + /* +@@ -87,21 +86,17 @@ execute_on_irq_stack(int overflow, struc + * handler) we can't do that and just have to keep using the + * current stack (which is the irq stack already after all) + */ +- if (unlikely(curctx == irqctx)) ++ if (unlikely((void *)current_stack_pointer - (void *)irqctx < THREAD_SIZE)) + return 0; + + /* build the stack frame on the IRQ stack */ +- isp = (u32 *) ((char *)irqctx + sizeof(*irqctx)); +- irqctx->tinfo.task = curctx->tinfo.task; +- irqctx->tinfo.previous_esp = current_stack_pointer; ++ isp = (u32 *) ((char *)irqctx + sizeof(*irqctx) - 8); ++ irqctx->previous_esp = current_stack_pointer; ++ add_preempt_count(HARDIRQ_OFFSET); + +- /* +- * Copy the softirq bits in preempt_count so that the +- * softirq checks work in the hardirq context. +- */ +- irqctx->tinfo.preempt_count = +- (irqctx->tinfo.preempt_count & ~SOFTIRQ_MASK) | +- (curctx->tinfo.preempt_count & SOFTIRQ_MASK); ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ __set_fs(MAKE_MM_SEG(0)); ++#endif + + if (unlikely(overflow)) + call_on_stack(print_stack_overflow, isp); +@@ -113,6 +108,12 @@ execute_on_irq_stack(int overflow, struc + : "0" (irq), "1" (desc), "2" (isp), + "D" (desc->handle_irq) + : "memory", "cc", "ecx"); ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ __set_fs(current_thread_info()->addr_limit); ++#endif ++ ++ sub_preempt_count(HARDIRQ_OFFSET); + return 1; + } + +@@ -121,29 +122,11 @@ execute_on_irq_stack(int overflow, struc + */ + void __cpuinit irq_ctx_init(int cpu) + { +- union irq_ctx *irqctx; +- + if (per_cpu(hardirq_ctx, cpu)) + return; + +- irqctx = page_address(alloc_pages_node(cpu_to_node(cpu), +- THREAD_FLAGS, +- THREAD_ORDER)); +- memset(&irqctx->tinfo, 0, sizeof(struct thread_info)); +- irqctx->tinfo.cpu = cpu; +- irqctx->tinfo.preempt_count = HARDIRQ_OFFSET; +- irqctx->tinfo.addr_limit = MAKE_MM_SEG(0); +- +- per_cpu(hardirq_ctx, cpu) = irqctx; +- +- irqctx = page_address(alloc_pages_node(cpu_to_node(cpu), +- THREAD_FLAGS, +- THREAD_ORDER)); +- memset(&irqctx->tinfo, 0, sizeof(struct thread_info)); +- irqctx->tinfo.cpu = cpu; +- irqctx->tinfo.addr_limit = MAKE_MM_SEG(0); +- +- per_cpu(softirq_ctx, cpu) = irqctx; ++ per_cpu(hardirq_ctx, cpu) = page_address(alloc_pages_node(cpu_to_node(cpu), THREAD_FLAGS, THREAD_ORDER)); ++ per_cpu(softirq_ctx, cpu) = page_address(alloc_pages_node(cpu_to_node(cpu), THREAD_FLAGS, THREAD_ORDER)); + + printk(KERN_DEBUG "CPU %u irqstacks, hard=%p soft=%p\n", + cpu, per_cpu(hardirq_ctx, cpu), per_cpu(softirq_ctx, cpu)); +@@ -152,7 +135,6 @@ void __cpuinit irq_ctx_init(int cpu) + asmlinkage void do_softirq(void) + { + unsigned long flags; +- struct thread_info *curctx; + union irq_ctx *irqctx; + u32 *isp; + +@@ -162,15 +144,22 @@ asmlinkage void do_softirq(void) + local_irq_save(flags); + + if (local_softirq_pending()) { +- curctx = current_thread_info(); + irqctx = __this_cpu_read(softirq_ctx); +- irqctx->tinfo.task = curctx->task; +- irqctx->tinfo.previous_esp = current_stack_pointer; ++ irqctx->previous_esp = current_stack_pointer; + + /* build the stack frame on the softirq stack */ +- isp = (u32 *) ((char *)irqctx + sizeof(*irqctx)); ++ isp = (u32 *) ((char *)irqctx + sizeof(*irqctx) - 8); ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ __set_fs(MAKE_MM_SEG(0)); ++#endif + + call_on_stack(__do_softirq, isp); ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ __set_fs(current_thread_info()->addr_limit); ++#endif ++ + /* + * Shouldnt happen, we returned above if in_interrupt(): + */ +diff -urNp linux-2.6.38.3/arch/x86/kernel/kgdb.c linux-2.6.38.3/arch/x86/kernel/kgdb.c +--- linux-2.6.38.3/arch/x86/kernel/kgdb.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/kgdb.c 2011-04-17 15:57:32.000000000 -0400 +@@ -124,11 +124,11 @@ char *dbg_get_reg(int regno, void *mem, + switch (regno) { + #ifdef CONFIG_X86_32 + case GDB_SS: +- if (!user_mode_vm(regs)) ++ if (!user_mode(regs)) + *(unsigned long *)mem = __KERNEL_DS; + break; + case GDB_SP: +- if (!user_mode_vm(regs)) ++ if (!user_mode(regs)) + *(unsigned long *)mem = kernel_stack_pointer(regs); + break; + case GDB_GS: +@@ -719,7 +719,7 @@ void kgdb_arch_set_pc(struct pt_regs *re + regs->ip = ip; + } + +-struct kgdb_arch arch_kgdb_ops = { ++const struct kgdb_arch arch_kgdb_ops = { + /* Breakpoint instruction: */ + .gdb_bpt_instr = { 0xcc }, + .flags = KGDB_HW_BREAKPOINT, +diff -urNp linux-2.6.38.3/arch/x86/kernel/kprobes.c linux-2.6.38.3/arch/x86/kernel/kprobes.c +--- linux-2.6.38.3/arch/x86/kernel/kprobes.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/kprobes.c 2011-04-17 15:57:32.000000000 -0400 +@@ -115,8 +115,11 @@ static void __kprobes __synthesize_relat + } __attribute__((packed)) *insn; + + insn = (struct __arch_relative_insn *)from; ++ ++ pax_open_kernel(); + insn->raddr = (s32)((long)(to) - ((long)(from) + 5)); + insn->op = op; ++ pax_close_kernel(); + } + + /* Insert a jump instruction at address 'from', which jumps to address 'to'.*/ +@@ -153,7 +156,7 @@ static int __kprobes can_boost(kprobe_op + kprobe_opcode_t opcode; + kprobe_opcode_t *orig_opcodes = opcodes; + +- if (search_exception_tables((unsigned long)opcodes)) ++ if (search_exception_tables(ktva_ktla((unsigned long)opcodes))) + return 0; /* Page fault may occur on this address. */ + + retry: +@@ -314,7 +317,9 @@ static int __kprobes __copy_instruction( + } + } + insn_get_length(&insn); ++ pax_open_kernel(); + memcpy(dest, insn.kaddr, insn.length); ++ pax_close_kernel(); + + #ifdef CONFIG_X86_64 + if (insn_rip_relative(&insn)) { +@@ -338,7 +343,9 @@ static int __kprobes __copy_instruction( + (u8 *) dest; + BUG_ON((s64) (s32) newdisp != newdisp); /* Sanity check. */ + disp = (u8 *) dest + insn_offset_displacement(&insn); ++ pax_open_kernel(); + *(s32 *) disp = (s32) newdisp; ++ pax_close_kernel(); + } + #endif + return insn.length; +@@ -352,12 +359,12 @@ static void __kprobes arch_copy_kprobe(s + */ + __copy_instruction(p->ainsn.insn, p->addr, 0); + +- if (can_boost(p->addr)) ++ if (can_boost(ktla_ktva(p->addr))) + p->ainsn.boostable = 0; + else + p->ainsn.boostable = -1; + +- p->opcode = *p->addr; ++ p->opcode = *(ktla_ktva(p->addr)); + } + + int __kprobes arch_prepare_kprobe(struct kprobe *p) +@@ -474,7 +481,7 @@ static void __kprobes setup_singlestep(s + * nor set current_kprobe, because it doesn't use single + * stepping. + */ +- regs->ip = (unsigned long)p->ainsn.insn; ++ regs->ip = ktva_ktla((unsigned long)p->ainsn.insn); + preempt_enable_no_resched(); + return; + } +@@ -493,7 +500,7 @@ static void __kprobes setup_singlestep(s + if (p->opcode == BREAKPOINT_INSTRUCTION) + regs->ip = (unsigned long)p->addr; + else +- regs->ip = (unsigned long)p->ainsn.insn; ++ regs->ip = ktva_ktla((unsigned long)p->ainsn.insn); + } + + /* +@@ -572,7 +579,7 @@ static int __kprobes kprobe_handler(stru + setup_singlestep(p, regs, kcb, 0); + return 1; + } +- } else if (*addr != BREAKPOINT_INSTRUCTION) { ++ } else if (*(kprobe_opcode_t *)ktla_ktva((unsigned long)addr) != BREAKPOINT_INSTRUCTION) { + /* + * The breakpoint instruction was removed right + * after we hit it. Another cpu has removed +@@ -817,7 +824,7 @@ static void __kprobes resume_execution(s + struct pt_regs *regs, struct kprobe_ctlblk *kcb) + { + unsigned long *tos = stack_addr(regs); +- unsigned long copy_ip = (unsigned long)p->ainsn.insn; ++ unsigned long copy_ip = ktva_ktla((unsigned long)p->ainsn.insn); + unsigned long orig_ip = (unsigned long)p->addr; + kprobe_opcode_t *insn = p->ainsn.insn; + +@@ -999,7 +1006,7 @@ int __kprobes kprobe_exceptions_notify(s + struct die_args *args = data; + int ret = NOTIFY_DONE; + +- if (args->regs && user_mode_vm(args->regs)) ++ if (args->regs && user_mode(args->regs)) + return ret; + + switch (val) { +@@ -1372,7 +1379,7 @@ int __kprobes arch_prepare_optimized_kpr + * Verify if the address gap is in 2GB range, because this uses + * a relative jump. + */ +- rel = (long)op->optinsn.insn - (long)op->kp.addr + RELATIVEJUMP_SIZE; ++ rel = (long)op->optinsn.insn - ktla_ktva((long)op->kp.addr) + RELATIVEJUMP_SIZE; + if (abs(rel) > 0x7fffffff) + return -ERANGE; + +@@ -1393,11 +1400,11 @@ int __kprobes arch_prepare_optimized_kpr + synthesize_set_arg1(buf + TMPL_MOVE_IDX, (unsigned long)op); + + /* Set probe function call */ +- synthesize_relcall(buf + TMPL_CALL_IDX, optimized_callback); ++ synthesize_relcall(buf + TMPL_CALL_IDX, ktla_ktva(optimized_callback)); + + /* Set returning jmp instruction at the tail of out-of-line buffer */ + synthesize_reljump(buf + TMPL_END_IDX + op->optinsn.size, +- (u8 *)op->kp.addr + op->optinsn.size); ++ (u8 *)ktla_ktva(op->kp.addr) + op->optinsn.size); + + flush_icache_range((unsigned long) buf, + (unsigned long) buf + TMPL_END_IDX + +@@ -1419,7 +1426,7 @@ static void __kprobes setup_optimize_kpr + ((long)op->kp.addr + RELATIVEJUMP_SIZE)); + + /* Backup instructions which will be replaced by jump address */ +- memcpy(op->optinsn.copied_insn, op->kp.addr + INT3_SIZE, ++ memcpy(op->optinsn.copied_insn, ktla_ktva(op->kp.addr) + INT3_SIZE, + RELATIVE_ADDR_SIZE); + + insn_buf[0] = RELATIVEJUMP_OPCODE; +diff -urNp linux-2.6.38.3/arch/x86/kernel/ldt.c linux-2.6.38.3/arch/x86/kernel/ldt.c +--- linux-2.6.38.3/arch/x86/kernel/ldt.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/ldt.c 2011-04-17 15:57:32.000000000 -0400 +@@ -67,13 +67,13 @@ static int alloc_ldt(mm_context_t *pc, i + if (reload) { + #ifdef CONFIG_SMP + preempt_disable(); +- load_LDT(pc); ++ load_LDT_nolock(pc); + if (!cpumask_equal(mm_cpumask(current->mm), + cpumask_of(smp_processor_id()))) + smp_call_function(flush_ldt, current->mm, 1); + preempt_enable(); + #else +- load_LDT(pc); ++ load_LDT_nolock(pc); + #endif + } + if (oldsize) { +@@ -95,7 +95,7 @@ static inline int copy_ldt(mm_context_t + return err; + + for (i = 0; i < old->size; i++) +- write_ldt_entry(new->ldt, i, old->ldt + i * LDT_ENTRY_SIZE); ++ write_ldt_entry(new->ldt, i, old->ldt + i); + return 0; + } + +@@ -116,6 +116,24 @@ int init_new_context(struct task_struct + retval = copy_ldt(&mm->context, &old_mm->context); + mutex_unlock(&old_mm->context.lock); + } ++ ++ if (tsk == current) { ++ mm->context.vdso = 0; ++ ++#ifdef CONFIG_X86_32 ++#if defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) ++ mm->context.user_cs_base = 0UL; ++ mm->context.user_cs_limit = ~0UL; ++ ++#if defined(CONFIG_PAX_PAGEEXEC) && defined(CONFIG_SMP) ++ cpus_clear(mm->context.cpu_user_cs_mask); ++#endif ++ ++#endif ++#endif ++ ++ } ++ + return retval; + } + +@@ -230,6 +248,13 @@ static int write_ldt(void __user *ptr, u + } + } + ++#ifdef CONFIG_PAX_SEGMEXEC ++ if ((mm->pax_flags & MF_PAX_SEGMEXEC) && (ldt_info.contents & MODIFY_LDT_CONTENTS_CODE)) { ++ error = -EINVAL; ++ goto out_unlock; ++ } ++#endif ++ + fill_ldt(&ldt, &ldt_info); + if (oldmode) + ldt.avl = 0; +diff -urNp linux-2.6.38.3/arch/x86/kernel/machine_kexec_32.c linux-2.6.38.3/arch/x86/kernel/machine_kexec_32.c +--- linux-2.6.38.3/arch/x86/kernel/machine_kexec_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/machine_kexec_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -27,7 +27,7 @@ + #include <asm/cacheflush.h> + #include <asm/debugreg.h> + +-static void set_idt(void *newidt, __u16 limit) ++static void set_idt(struct desc_struct *newidt, __u16 limit) + { + struct desc_ptr curidt; + +@@ -39,7 +39,7 @@ static void set_idt(void *newidt, __u16 + } + + +-static void set_gdt(void *newgdt, __u16 limit) ++static void set_gdt(struct desc_struct *newgdt, __u16 limit) + { + struct desc_ptr curgdt; + +@@ -217,7 +217,7 @@ void machine_kexec(struct kimage *image) + } + + control_page = page_address(image->control_code_page); +- memcpy(control_page, relocate_kernel, KEXEC_CONTROL_CODE_MAX_SIZE); ++ memcpy(control_page, (void *)ktla_ktva((unsigned long)relocate_kernel), KEXEC_CONTROL_CODE_MAX_SIZE); + + relocate_kernel_ptr = control_page; + page_list[PA_CONTROL_PAGE] = __pa(control_page); +diff -urNp linux-2.6.38.3/arch/x86/kernel/microcode_amd.c linux-2.6.38.3/arch/x86/kernel/microcode_amd.c +--- linux-2.6.38.3/arch/x86/kernel/microcode_amd.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/microcode_amd.c 2011-04-17 15:57:32.000000000 -0400 +@@ -317,7 +317,7 @@ static void microcode_fini_cpu_amd(int c + uci->mc = NULL; + } + +-static struct microcode_ops microcode_amd_ops = { ++static const struct microcode_ops microcode_amd_ops = { + .request_microcode_user = request_microcode_user, + .request_microcode_fw = request_microcode_fw, + .collect_cpu_info = collect_cpu_info_amd, +@@ -325,7 +325,7 @@ static struct microcode_ops microcode_am + .microcode_fini_cpu = microcode_fini_cpu_amd, + }; + +-struct microcode_ops * __init init_amd_microcode(void) ++const struct microcode_ops * __init init_amd_microcode(void) + { + return µcode_amd_ops; + } +diff -urNp linux-2.6.38.3/arch/x86/kernel/microcode_core.c linux-2.6.38.3/arch/x86/kernel/microcode_core.c +--- linux-2.6.38.3/arch/x86/kernel/microcode_core.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/microcode_core.c 2011-04-17 15:57:32.000000000 -0400 +@@ -92,7 +92,7 @@ MODULE_LICENSE("GPL"); + + #define MICROCODE_VERSION "2.00" + +-static struct microcode_ops *microcode_ops; ++static const struct microcode_ops *microcode_ops; + + /* + * Synchronization. +diff -urNp linux-2.6.38.3/arch/x86/kernel/microcode_intel.c linux-2.6.38.3/arch/x86/kernel/microcode_intel.c +--- linux-2.6.38.3/arch/x86/kernel/microcode_intel.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/microcode_intel.c 2011-04-17 15:57:32.000000000 -0400 +@@ -440,13 +440,13 @@ static enum ucode_state request_microcod + + static int get_ucode_user(void *to, const void *from, size_t n) + { +- return copy_from_user(to, from, n); ++ return copy_from_user(to, (__force const void __user *)from, n); + } + + static enum ucode_state + request_microcode_user(int cpu, const void __user *buf, size_t size) + { +- return generic_load_microcode(cpu, (void *)buf, size, &get_ucode_user); ++ return generic_load_microcode(cpu, (__force void *)buf, size, &get_ucode_user); + } + + static void microcode_fini_cpu(int cpu) +@@ -457,7 +457,7 @@ static void microcode_fini_cpu(int cpu) + uci->mc = NULL; + } + +-static struct microcode_ops microcode_intel_ops = { ++static const struct microcode_ops microcode_intel_ops = { + .request_microcode_user = request_microcode_user, + .request_microcode_fw = request_microcode_fw, + .collect_cpu_info = collect_cpu_info, +@@ -465,7 +465,7 @@ static struct microcode_ops microcode_in + .microcode_fini_cpu = microcode_fini_cpu, + }; + +-struct microcode_ops * __init init_intel_microcode(void) ++const struct microcode_ops * __init init_intel_microcode(void) + { + return µcode_intel_ops; + } +diff -urNp linux-2.6.38.3/arch/x86/kernel/module.c linux-2.6.38.3/arch/x86/kernel/module.c +--- linux-2.6.38.3/arch/x86/kernel/module.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/module.c 2011-04-17 15:57:32.000000000 -0400 +@@ -35,21 +35,66 @@ + #define DEBUGP(fmt...) + #endif + +-void *module_alloc(unsigned long size) ++static inline void *__module_alloc(unsigned long size, pgprot_t prot) + { + if (PAGE_ALIGN(size) > MODULES_LEN) + return NULL; + return __vmalloc_node_range(size, 1, MODULES_VADDR, MODULES_END, +- GFP_KERNEL | __GFP_HIGHMEM, PAGE_KERNEL_EXEC, ++ GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO, prot, + -1, __builtin_return_address(0)); + } + ++void *module_alloc(unsigned long size) ++{ ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ return __module_alloc(size, PAGE_KERNEL); ++#else ++ return __module_alloc(size, PAGE_KERNEL_EXEC); ++#endif ++ ++} ++ + /* Free memory returned from module_alloc */ + void module_free(struct module *mod, void *module_region) + { + vfree(module_region); + } + ++#ifdef CONFIG_PAX_KERNEXEC ++#ifdef CONFIG_X86_32 ++void *module_alloc_exec(unsigned long size) ++{ ++ struct vm_struct *area; ++ ++ if (size == 0) ++ return NULL; ++ ++ area = __get_vm_area(size, VM_ALLOC, (unsigned long)&MODULES_EXEC_VADDR, (unsigned long)&MODULES_EXEC_END); ++ return area ? area->addr : NULL; ++} ++EXPORT_SYMBOL(module_alloc_exec); ++ ++void module_free_exec(struct module *mod, void *module_region) ++{ ++ vunmap(module_region); ++} ++EXPORT_SYMBOL(module_free_exec); ++#else ++void module_free_exec(struct module *mod, void *module_region) ++{ ++ module_free(mod, module_region); ++} ++EXPORT_SYMBOL(module_free_exec); ++ ++void *module_alloc_exec(unsigned long size) ++{ ++ return __module_alloc(size, PAGE_KERNEL_RX); ++} ++EXPORT_SYMBOL(module_alloc_exec); ++#endif ++#endif ++ + /* We don't need anything special. */ + int module_frob_arch_sections(Elf_Ehdr *hdr, + Elf_Shdr *sechdrs, +@@ -69,14 +114,16 @@ int apply_relocate(Elf32_Shdr *sechdrs, + unsigned int i; + Elf32_Rel *rel = (void *)sechdrs[relsec].sh_addr; + Elf32_Sym *sym; +- uint32_t *location; ++ uint32_t *plocation, location; + + DEBUGP("Applying relocate section %u to %u\n", relsec, + sechdrs[relsec].sh_info); + for (i = 0; i < sechdrs[relsec].sh_size / sizeof(*rel); i++) { + /* This is where to make the change */ +- location = (void *)sechdrs[sechdrs[relsec].sh_info].sh_addr +- + rel[i].r_offset; ++ plocation = (void *)sechdrs[sechdrs[relsec].sh_info].sh_addr + rel[i].r_offset; ++ location = (uint32_t)plocation; ++ if (sechdrs[sechdrs[relsec].sh_info].sh_flags & SHF_EXECINSTR) ++ plocation = ktla_ktva((void *)plocation); + /* This is the symbol it is referring to. Note that all + undefined symbols have been resolved. */ + sym = (Elf32_Sym *)sechdrs[symindex].sh_addr +@@ -85,11 +132,15 @@ int apply_relocate(Elf32_Shdr *sechdrs, + switch (ELF32_R_TYPE(rel[i].r_info)) { + case R_386_32: + /* We add the value into the location given */ +- *location += sym->st_value; ++ pax_open_kernel(); ++ *plocation += sym->st_value; ++ pax_close_kernel(); + break; + case R_386_PC32: + /* Add the value, subtract its postition */ +- *location += sym->st_value - (uint32_t)location; ++ pax_open_kernel(); ++ *plocation += sym->st_value - location; ++ pax_close_kernel(); + break; + default: + printk(KERN_ERR "module %s: Unknown relocation: %u\n", +@@ -145,21 +196,30 @@ int apply_relocate_add(Elf64_Shdr *sechd + case R_X86_64_NONE: + break; + case R_X86_64_64: ++ pax_open_kernel(); + *(u64 *)loc = val; ++ pax_close_kernel(); + break; + case R_X86_64_32: ++ pax_open_kernel(); + *(u32 *)loc = val; ++ pax_close_kernel(); + if (val != *(u32 *)loc) + goto overflow; + break; + case R_X86_64_32S: ++ pax_open_kernel(); + *(s32 *)loc = val; ++ pax_close_kernel(); + if ((s64)val != *(s32 *)loc) + goto overflow; + break; + case R_X86_64_PC32: + val -= (u64)loc; ++ pax_open_kernel(); + *(u32 *)loc = val; ++ pax_close_kernel(); ++ + #if 0 + if ((s64)val != *(s32 *)loc) + goto overflow; +diff -urNp linux-2.6.38.3/arch/x86/kernel/paravirt.c linux-2.6.38.3/arch/x86/kernel/paravirt.c +--- linux-2.6.38.3/arch/x86/kernel/paravirt.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/paravirt.c 2011-04-17 15:57:32.000000000 -0400 +@@ -122,7 +122,7 @@ unsigned paravirt_patch_jmp(void *insnbu + * corresponding structure. */ + static void *get_call_destination(u8 type) + { +- struct paravirt_patch_template tmpl = { ++ const struct paravirt_patch_template tmpl = { + .pv_init_ops = pv_init_ops, + .pv_time_ops = pv_time_ops, + .pv_cpu_ops = pv_cpu_ops, +@@ -145,14 +145,14 @@ unsigned paravirt_patch_default(u8 type, + if (opfunc == NULL) + /* If there's no function, patch it with a ud2a (BUG) */ + ret = paravirt_patch_insns(insnbuf, len, ud2a, ud2a+sizeof(ud2a)); +- else if (opfunc == _paravirt_nop) ++ else if (opfunc == (void *)_paravirt_nop) + /* If the operation is a nop, then nop the callsite */ + ret = paravirt_patch_nop(); + + /* identity functions just return their single argument */ +- else if (opfunc == _paravirt_ident_32) ++ else if (opfunc == (void *)_paravirt_ident_32) + ret = paravirt_patch_ident_32(insnbuf, len); +- else if (opfunc == _paravirt_ident_64) ++ else if (opfunc == (void *)_paravirt_ident_64) + ret = paravirt_patch_ident_64(insnbuf, len); + + else if (type == PARAVIRT_PATCH(pv_cpu_ops.iret) || +@@ -178,7 +178,7 @@ unsigned paravirt_patch_insns(void *insn + if (insn_len > len || start == NULL) + insn_len = len; + else +- memcpy(insnbuf, start, insn_len); ++ memcpy(insnbuf, ktla_ktva(start), insn_len); + + return insn_len; + } +@@ -294,22 +294,22 @@ void arch_flush_lazy_mmu_mode(void) + preempt_enable(); + } + +-struct pv_info pv_info = { ++struct pv_info pv_info __read_only = { + .name = "bare hardware", + .paravirt_enabled = 0, + .kernel_rpl = 0, + .shared_kernel_pmd = 1, /* Only used when CONFIG_X86_PAE is set */ + }; + +-struct pv_init_ops pv_init_ops = { ++struct pv_init_ops pv_init_ops __read_only = { + .patch = native_patch, + }; + +-struct pv_time_ops pv_time_ops = { ++struct pv_time_ops pv_time_ops __read_only = { + .sched_clock = native_sched_clock, + }; + +-struct pv_irq_ops pv_irq_ops = { ++struct pv_irq_ops pv_irq_ops __read_only = { + .save_fl = __PV_IS_CALLEE_SAVE(native_save_fl), + .restore_fl = __PV_IS_CALLEE_SAVE(native_restore_fl), + .irq_disable = __PV_IS_CALLEE_SAVE(native_irq_disable), +@@ -321,7 +321,7 @@ struct pv_irq_ops pv_irq_ops = { + #endif + }; + +-struct pv_cpu_ops pv_cpu_ops = { ++struct pv_cpu_ops pv_cpu_ops __read_only = { + .cpuid = native_cpuid, + .get_debugreg = native_get_debugreg, + .set_debugreg = native_set_debugreg, +@@ -382,7 +382,7 @@ struct pv_cpu_ops pv_cpu_ops = { + .end_context_switch = paravirt_nop, + }; + +-struct pv_apic_ops pv_apic_ops = { ++struct pv_apic_ops pv_apic_ops __read_only = { + #ifdef CONFIG_X86_LOCAL_APIC + .startup_ipi_hook = paravirt_nop, + #endif +@@ -396,7 +396,7 @@ struct pv_apic_ops pv_apic_ops = { + #define PTE_IDENT __PV_IS_CALLEE_SAVE(_paravirt_ident_64) + #endif + +-struct pv_mmu_ops pv_mmu_ops = { ++struct pv_mmu_ops pv_mmu_ops __read_only = { + + .read_cr2 = native_read_cr2, + .write_cr2 = native_write_cr2, +@@ -465,6 +465,12 @@ struct pv_mmu_ops pv_mmu_ops = { + }, + + .set_fixmap = native_set_fixmap, ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ .pax_open_kernel = native_pax_open_kernel, ++ .pax_close_kernel = native_pax_close_kernel, ++#endif ++ + }; + + EXPORT_SYMBOL_GPL(pv_time_ops); +diff -urNp linux-2.6.38.3/arch/x86/kernel/paravirt-spinlocks.c linux-2.6.38.3/arch/x86/kernel/paravirt-spinlocks.c +--- linux-2.6.38.3/arch/x86/kernel/paravirt-spinlocks.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/paravirt-spinlocks.c 2011-04-17 15:57:32.000000000 -0400 +@@ -13,7 +13,7 @@ default_spin_lock_flags(arch_spinlock_t + arch_spin_lock(lock); + } + +-struct pv_lock_ops pv_lock_ops = { ++struct pv_lock_ops pv_lock_ops __read_only = { + #ifdef CONFIG_SMP + .spin_is_locked = __ticket_spin_is_locked, + .spin_is_contended = __ticket_spin_is_contended, +diff -urNp linux-2.6.38.3/arch/x86/kernel/pci-calgary_64.c linux-2.6.38.3/arch/x86/kernel/pci-calgary_64.c +--- linux-2.6.38.3/arch/x86/kernel/pci-calgary_64.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/pci-calgary_64.c 2011-04-17 15:57:32.000000000 -0400 +@@ -476,7 +476,7 @@ static void calgary_free_coherent(struct + free_pages((unsigned long)vaddr, get_order(size)); + } + +-static struct dma_map_ops calgary_dma_ops = { ++static const struct dma_map_ops calgary_dma_ops = { + .alloc_coherent = calgary_alloc_coherent, + .free_coherent = calgary_free_coherent, + .map_sg = calgary_map_sg, +diff -urNp linux-2.6.38.3/arch/x86/kernel/pci-dma.c linux-2.6.38.3/arch/x86/kernel/pci-dma.c +--- linux-2.6.38.3/arch/x86/kernel/pci-dma.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/pci-dma.c 2011-04-17 15:57:32.000000000 -0400 +@@ -16,7 +16,7 @@ + + static int forbid_dac __read_mostly; + +-struct dma_map_ops *dma_ops = &nommu_dma_ops; ++const struct dma_map_ops *dma_ops = &nommu_dma_ops; + EXPORT_SYMBOL(dma_ops); + + static int iommu_sac_force __read_mostly; +@@ -250,7 +250,7 @@ early_param("iommu", iommu_setup); + + int dma_supported(struct device *dev, u64 mask) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + + #ifdef CONFIG_PCI + if (mask > 0xffffffff && forbid_dac > 0) { +diff -urNp linux-2.6.38.3/arch/x86/kernel/pci-gart_64.c linux-2.6.38.3/arch/x86/kernel/pci-gart_64.c +--- linux-2.6.38.3/arch/x86/kernel/pci-gart_64.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/pci-gart_64.c 2011-04-17 15:57:32.000000000 -0400 +@@ -706,7 +706,7 @@ static __init int init_amd_gatt(struct a + return -1; + } + +-static struct dma_map_ops gart_dma_ops = { ++static const struct dma_map_ops gart_dma_ops = { + .map_sg = gart_map_sg, + .unmap_sg = gart_unmap_sg, + .map_page = gart_map_page, +diff -urNp linux-2.6.38.3/arch/x86/kernel/pci-nommu.c linux-2.6.38.3/arch/x86/kernel/pci-nommu.c +--- linux-2.6.38.3/arch/x86/kernel/pci-nommu.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/pci-nommu.c 2011-04-17 15:57:32.000000000 -0400 +@@ -95,7 +95,7 @@ static void nommu_sync_sg_for_device(str + flush_write_buffers(); + } + +-struct dma_map_ops nommu_dma_ops = { ++const struct dma_map_ops nommu_dma_ops = { + .alloc_coherent = dma_generic_alloc_coherent, + .free_coherent = nommu_free_coherent, + .map_sg = nommu_map_sg, +diff -urNp linux-2.6.38.3/arch/x86/kernel/pci-swiotlb.c linux-2.6.38.3/arch/x86/kernel/pci-swiotlb.c +--- linux-2.6.38.3/arch/x86/kernel/pci-swiotlb.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/pci-swiotlb.c 2011-04-17 15:57:32.000000000 -0400 +@@ -26,7 +26,7 @@ static void *x86_swiotlb_alloc_coherent( + return swiotlb_alloc_coherent(hwdev, size, dma_handle, flags); + } + +-static struct dma_map_ops swiotlb_dma_ops = { ++static const struct dma_map_ops swiotlb_dma_ops = { + .mapping_error = swiotlb_dma_mapping_error, + .alloc_coherent = x86_swiotlb_alloc_coherent, + .free_coherent = swiotlb_free_coherent, +diff -urNp linux-2.6.38.3/arch/x86/kernel/process_32.c linux-2.6.38.3/arch/x86/kernel/process_32.c +--- linux-2.6.38.3/arch/x86/kernel/process_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/process_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -65,6 +65,7 @@ asmlinkage void ret_from_fork(void) __as + unsigned long thread_saved_pc(struct task_struct *tsk) + { + return ((unsigned long *)tsk->thread.sp)[3]; ++//XXX return tsk->thread.eip; + } + + #ifndef CONFIG_SMP +@@ -126,15 +127,14 @@ void __show_regs(struct pt_regs *regs, i + unsigned long sp; + unsigned short ss, gs; + +- if (user_mode_vm(regs)) { ++ if (user_mode(regs)) { + sp = regs->sp; + ss = regs->ss & 0xffff; +- gs = get_user_gs(regs); + } else { + sp = kernel_stack_pointer(regs); + savesegment(ss, ss); +- savesegment(gs, gs); + } ++ gs = get_user_gs(regs); + + show_regs_common(); + +@@ -196,7 +196,7 @@ int copy_thread(unsigned long clone_flag + struct task_struct *tsk; + int err; + +- childregs = task_pt_regs(p); ++ childregs = task_stack_page(p) + THREAD_SIZE - sizeof(struct pt_regs) - 8; + *childregs = *regs; + childregs->ax = 0; + childregs->sp = sp; +@@ -293,7 +293,7 @@ __switch_to(struct task_struct *prev_p, + struct thread_struct *prev = &prev_p->thread, + *next = &next_p->thread; + int cpu = smp_processor_id(); +- struct tss_struct *tss = &per_cpu(init_tss, cpu); ++ struct tss_struct *tss = init_tss + cpu; + bool preload_fpu; + + /* never put a printk in __switch_to... printk() calls wake_up*() indirectly */ +@@ -328,6 +328,10 @@ __switch_to(struct task_struct *prev_p, + */ + lazy_save_gs(prev->gs); + ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ __set_fs(task_thread_info(next_p)->addr_limit); ++#endif ++ + /* + * Load the per-thread Thread-Local Storage descriptor. + */ +@@ -363,6 +367,9 @@ __switch_to(struct task_struct *prev_p, + */ + arch_end_context_switch(next_p); + ++ percpu_write(current_task, next_p); ++ percpu_write(current_tinfo, &next_p->tinfo); ++ + if (preload_fpu) + __math_state_restore(); + +@@ -372,8 +379,6 @@ __switch_to(struct task_struct *prev_p, + if (prev->gs | next->gs) + lazy_load_gs(next->gs); + +- percpu_write(current_task, next_p); +- + return prev_p; + } + +@@ -403,4 +408,3 @@ unsigned long get_wchan(struct task_stru + } while (count++ < 16); + return 0; + } +- +diff -urNp linux-2.6.38.3/arch/x86/kernel/process_64.c linux-2.6.38.3/arch/x86/kernel/process_64.c +--- linux-2.6.38.3/arch/x86/kernel/process_64.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/process_64.c 2011-04-17 15:57:32.000000000 -0400 +@@ -87,7 +87,7 @@ static void __exit_idle(void) + void exit_idle(void) + { + /* idle loop has pid 0 */ +- if (current->pid) ++ if (task_pid_nr(current)) + return; + __exit_idle(); + } +@@ -260,8 +260,7 @@ int copy_thread(unsigned long clone_flag + struct pt_regs *childregs; + struct task_struct *me = current; + +- childregs = ((struct pt_regs *) +- (THREAD_SIZE + task_stack_page(p))) - 1; ++ childregs = task_stack_page(p) + THREAD_SIZE - sizeof(struct pt_regs) - 8; + *childregs = *regs; + + childregs->ax = 0; +@@ -376,7 +375,7 @@ __switch_to(struct task_struct *prev_p, + struct thread_struct *prev = &prev_p->thread; + struct thread_struct *next = &next_p->thread; + int cpu = smp_processor_id(); +- struct tss_struct *tss = &per_cpu(init_tss, cpu); ++ struct tss_struct *tss = init_tss + cpu; + unsigned fsindex, gsindex; + bool preload_fpu; + +@@ -472,10 +471,9 @@ __switch_to(struct task_struct *prev_p, + prev->usersp = percpu_read(old_rsp); + percpu_write(old_rsp, next->usersp); + percpu_write(current_task, next_p); ++ percpu_write(current_tinfo, &next_p->tinfo); + +- percpu_write(kernel_stack, +- (unsigned long)task_stack_page(next_p) + +- THREAD_SIZE - KERNEL_STACK_OFFSET); ++ percpu_write(kernel_stack, next->sp0); + + /* + * Now maybe reload the debug registers and handle I/O bitmaps +@@ -529,12 +527,11 @@ unsigned long get_wchan(struct task_stru + if (!p || p == current || p->state == TASK_RUNNING) + return 0; + stack = (unsigned long)task_stack_page(p); +- if (p->thread.sp < stack || p->thread.sp >= stack+THREAD_SIZE) ++ if (p->thread.sp < stack || p->thread.sp > stack+THREAD_SIZE-8-sizeof(u64)) + return 0; + fp = *(u64 *)(p->thread.sp); + do { +- if (fp < (unsigned long)stack || +- fp >= (unsigned long)stack+THREAD_SIZE) ++ if (fp < stack || fp > stack+THREAD_SIZE-8-sizeof(u64)) + return 0; + ip = *(u64 *)(fp+8); + if (!in_sched_functions(ip)) +diff -urNp linux-2.6.38.3/arch/x86/kernel/process.c linux-2.6.38.3/arch/x86/kernel/process.c +--- linux-2.6.38.3/arch/x86/kernel/process.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/process.c 2011-04-17 15:57:32.000000000 -0400 +@@ -48,16 +48,33 @@ void free_thread_xstate(struct task_stru + + void free_thread_info(struct thread_info *ti) + { +- free_thread_xstate(ti->task); + free_pages((unsigned long)ti, get_order(THREAD_SIZE)); + } + ++static struct kmem_cache *task_struct_cachep; ++ + void arch_task_cache_init(void) + { +- task_xstate_cachep = +- kmem_cache_create("task_xstate", xstate_size, ++ /* create a slab on which task_structs can be allocated */ ++ task_struct_cachep = ++ kmem_cache_create("task_struct", sizeof(struct task_struct), ++ ARCH_MIN_TASKALIGN, SLAB_PANIC | SLAB_NOTRACK, NULL); ++ ++ task_xstate_cachep = ++ kmem_cache_create("task_xstate", xstate_size, + __alignof__(union thread_xstate), +- SLAB_PANIC | SLAB_NOTRACK, NULL); ++ SLAB_PANIC | SLAB_NOTRACK | SLAB_USERCOPY, NULL); ++} ++ ++struct task_struct *alloc_task_struct(void) ++{ ++ return kmem_cache_alloc(task_struct_cachep, GFP_KERNEL); ++} ++ ++void free_task_struct(struct task_struct *task) ++{ ++ free_thread_xstate(task); ++ kmem_cache_free(task_struct_cachep, task); + } + + /* +@@ -70,7 +87,7 @@ void exit_thread(void) + unsigned long *bp = t->io_bitmap_ptr; + + if (bp) { +- struct tss_struct *tss = &per_cpu(init_tss, get_cpu()); ++ struct tss_struct *tss = init_tss + get_cpu(); + + t->io_bitmap_ptr = NULL; + clear_thread_flag(TIF_IO_BITMAP); +@@ -106,7 +123,7 @@ void show_regs_common(void) + + printk(KERN_CONT "\n"); + printk(KERN_DEFAULT "Pid: %d, comm: %.20s %s %s %.*s", +- current->pid, current->comm, print_tainted(), ++ task_pid_nr(current), current->comm, print_tainted(), + init_utsname()->release, + (int)strcspn(init_utsname()->version, " "), + init_utsname()->version); +@@ -123,6 +140,9 @@ void flush_thread(void) + { + struct task_struct *tsk = current; + ++#if defined(CONFIG_X86_32) && !defined(CONFIG_CC_STACKPROTECTOR) && !defined(CONFIG_PAX_MEMORY_UDEREF) ++ loadsegment(gs, 0); ++#endif + flush_ptrace_hw_breakpoint(tsk); + memset(tsk->thread.tls_array, 0, sizeof(tsk->thread.tls_array)); + /* +@@ -285,10 +305,10 @@ int kernel_thread(int (*fn)(void *), voi + regs.di = (unsigned long) arg; + + #ifdef CONFIG_X86_32 +- regs.ds = __USER_DS; +- regs.es = __USER_DS; ++ regs.ds = __KERNEL_DS; ++ regs.es = __KERNEL_DS; + regs.fs = __KERNEL_PERCPU; +- regs.gs = __KERNEL_STACK_CANARY; ++ savesegment(gs, regs.gs); + #else + regs.ss = __KERNEL_DS; + #endif +@@ -668,16 +688,31 @@ static int __init idle_setup(char *str) + } + early_param("idle", idle_setup); + +-unsigned long arch_align_stack(unsigned long sp) ++#ifdef CONFIG_PAX_RANDKSTACK ++asmlinkage void pax_randomize_kstack(void) + { +- if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space) +- sp -= get_random_int() % 8192; +- return sp & ~0xf; +-} ++ struct thread_struct *thread = ¤t->thread; ++ unsigned long time; + +-unsigned long arch_randomize_brk(struct mm_struct *mm) +-{ +- unsigned long range_end = mm->brk + 0x02000000; +- return randomize_range(mm->brk, range_end, 0) ? : mm->brk; +-} ++ if (!randomize_va_space) ++ return; ++ ++ rdtscl(time); ++ ++ /* P4 seems to return a 0 LSB, ignore it */ ++#ifdef CONFIG_MPENTIUM4 ++ time &= 0x1EUL; ++ time <<= 2; ++#else ++ time &= 0xFUL; ++ time <<= 3; ++#endif ++ ++ thread->sp0 ^= time; ++ load_sp0(init_tss + smp_processor_id(), thread); + ++#ifdef CONFIG_X86_64 ++ percpu_write(kernel_stack, thread->sp0); ++#endif ++} ++#endif +diff -urNp linux-2.6.38.3/arch/x86/kernel/ptrace.c linux-2.6.38.3/arch/x86/kernel/ptrace.c +--- linux-2.6.38.3/arch/x86/kernel/ptrace.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/ptrace.c 2011-04-17 15:57:32.000000000 -0400 +@@ -805,7 +805,7 @@ long arch_ptrace(struct task_struct *chi + unsigned long addr, unsigned long data) + { + int ret; +- unsigned long __user *datap = (unsigned long __user *)data; ++ unsigned long __user *datap = (__force unsigned long __user *)data; + + switch (request) { + /* read the word at location addr in the USER area. */ +@@ -890,14 +890,14 @@ long arch_ptrace(struct task_struct *chi + if ((int) addr < 0) + return -EIO; + ret = do_get_thread_area(child, addr, +- (struct user_desc __user *)data); ++ (__force struct user_desc __user *) data); + break; + + case PTRACE_SET_THREAD_AREA: + if ((int) addr < 0) + return -EIO; + ret = do_set_thread_area(child, addr, +- (struct user_desc __user *)data, 0); ++ (__force struct user_desc __user *) data, 0); + break; + #endif + +@@ -1314,7 +1314,7 @@ static void fill_sigtrap_info(struct tas + memset(info, 0, sizeof(*info)); + info->si_signo = SIGTRAP; + info->si_code = si_code; +- info->si_addr = user_mode_vm(regs) ? (void __user *)regs->ip : NULL; ++ info->si_addr = user_mode(regs) ? (__force void __user *)regs->ip : NULL; + } + + void user_single_step_siginfo(struct task_struct *tsk, +@@ -1347,7 +1347,7 @@ void send_sigtrap(struct task_struct *ts + * We must return the syscall number to actually look up in the table. + * This can be -1L to skip running any syscall at all. + */ +-asmregparm long syscall_trace_enter(struct pt_regs *regs) ++long syscall_trace_enter(struct pt_regs *regs) + { + long ret = 0; + +@@ -1392,7 +1392,7 @@ asmregparm long syscall_trace_enter(stru + return ret ?: regs->orig_ax; + } + +-asmregparm void syscall_trace_leave(struct pt_regs *regs) ++void syscall_trace_leave(struct pt_regs *regs) + { + bool step; + +diff -urNp linux-2.6.38.3/arch/x86/kernel/reboot.c linux-2.6.38.3/arch/x86/kernel/reboot.c +--- linux-2.6.38.3/arch/x86/kernel/reboot.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/reboot.c 2011-04-17 15:57:32.000000000 -0400 +@@ -34,7 +34,7 @@ void (*pm_power_off)(void); + EXPORT_SYMBOL(pm_power_off); + + static const struct desc_ptr no_idt = {}; +-static int reboot_mode; ++static unsigned short reboot_mode; + enum reboot_type reboot_type = BOOT_KBD; + int reboot_force; + +@@ -293,7 +293,7 @@ static struct dmi_system_id __initdata r + DMI_MATCH(DMI_BOARD_NAME, "VersaLogic Menlow board"), + }, + }, +- { } ++ { NULL, NULL, {{0, {0}}}, NULL} + }; + + static int __init reboot_init(void) +@@ -309,12 +309,12 @@ core_initcall(reboot_init); + controller to pulse the CPU reset line, which is more thorough, but + doesn't work with at least one type of 486 motherboard. It is easy + to stop this code working; hence the copious comments. */ +-static const unsigned long long +-real_mode_gdt_entries [3] = ++static struct desc_struct ++real_mode_gdt_entries [3] __read_only = + { +- 0x0000000000000000ULL, /* Null descriptor */ +- 0x00009b000000ffffULL, /* 16-bit real-mode 64k code at 0x00000000 */ +- 0x000093000100ffffULL /* 16-bit real-mode 64k data at 0x00000100 */ ++ GDT_ENTRY_INIT(0, 0, 0), /* Null descriptor */ ++ GDT_ENTRY_INIT(0x9b, 0, 0xffff), /* 16-bit real-mode 64k code at 0x00000000 */ ++ GDT_ENTRY_INIT(0x93, 0x100, 0xffff) /* 16-bit real-mode 64k data at 0x00000100 */ + }; + + static const struct desc_ptr +@@ -363,7 +363,7 @@ static const unsigned char jump_to_bios + * specified by the code and length parameters. + * We assume that length will aways be less that 100! + */ +-void machine_real_restart(const unsigned char *code, int length) ++void machine_real_restart(const unsigned char *code, unsigned int length) + { + local_irq_disable(); + +@@ -390,16 +390,15 @@ void machine_real_restart(const unsigned + boot)". This seems like a fairly standard thing that gets set by + REBOOT.COM programs, and the previous reset routine did this + too. */ +- *((unsigned short *)0x472) = reboot_mode; ++ *(unsigned short *)(__va(0x472)) = reboot_mode; + + /* For the switch to real mode, copy some code to low memory. It has + to be in the first 64k because it is running in 16-bit mode, and it + has to have the same physical and virtual address, because it turns + off paging. Copy it near the end of the first page, out of the way + of BIOS variables. */ +- memcpy((void *)(0x1000 - sizeof(real_mode_switch) - 100), +- real_mode_switch, sizeof (real_mode_switch)); +- memcpy((void *)(0x1000 - 100), code, length); ++ memcpy(__va(0x1000 - sizeof (real_mode_switch) - 100), real_mode_switch, sizeof (real_mode_switch)); ++ memcpy(__va(0x1000 - 100), code, length); + + /* Set up the IDT for real mode. */ + load_idt(&real_mode_idt); +diff -urNp linux-2.6.38.3/arch/x86/kernel/setup.c linux-2.6.38.3/arch/x86/kernel/setup.c +--- linux-2.6.38.3/arch/x86/kernel/setup.c 2011-04-17 16:53:01.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/setup.c 2011-04-17 16:53:16.000000000 -0400 +@@ -654,7 +654,7 @@ static void __init trim_bios_range(void) + * area (640->1Mb) as ram even though it is not. + * take them out. + */ +- e820_remove_range(BIOS_BEGIN, BIOS_END - BIOS_BEGIN, E820_RAM, 1); ++ e820_remove_range(ISA_START_ADDRESS, ISA_END_ADDRESS - ISA_START_ADDRESS, E820_RAM, 1); + sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map); + } + +@@ -790,14 +790,14 @@ void __init setup_arch(char **cmdline_p) + + if (!boot_params.hdr.root_flags) + root_mountflags &= ~MS_RDONLY; +- init_mm.start_code = (unsigned long) _text; +- init_mm.end_code = (unsigned long) _etext; ++ init_mm.start_code = ktla_ktva((unsigned long) _text); ++ init_mm.end_code = ktla_ktva((unsigned long) _etext); + init_mm.end_data = (unsigned long) _edata; + init_mm.brk = _brk_end; + +- code_resource.start = virt_to_phys(_text); +- code_resource.end = virt_to_phys(_etext)-1; +- data_resource.start = virt_to_phys(_etext); ++ code_resource.start = virt_to_phys(ktla_ktva(_text)); ++ code_resource.end = virt_to_phys(ktla_ktva(_etext))-1; ++ data_resource.start = virt_to_phys(_sdata); + data_resource.end = virt_to_phys(_edata)-1; + bss_resource.start = virt_to_phys(&__bss_start); + bss_resource.end = virt_to_phys(&__bss_stop)-1; +diff -urNp linux-2.6.38.3/arch/x86/kernel/setup_percpu.c linux-2.6.38.3/arch/x86/kernel/setup_percpu.c +--- linux-2.6.38.3/arch/x86/kernel/setup_percpu.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/setup_percpu.c 2011-04-17 15:57:32.000000000 -0400 +@@ -21,19 +21,17 @@ + #include <asm/cpu.h> + #include <asm/stackprotector.h> + +-DEFINE_PER_CPU(int, cpu_number); ++#ifdef CONFIG_SMP ++DEFINE_PER_CPU(unsigned int, cpu_number); + EXPORT_PER_CPU_SYMBOL(cpu_number); ++#endif + +-#ifdef CONFIG_X86_64 + #define BOOT_PERCPU_OFFSET ((unsigned long)__per_cpu_load) +-#else +-#define BOOT_PERCPU_OFFSET 0 +-#endif + + DEFINE_PER_CPU(unsigned long, this_cpu_off) = BOOT_PERCPU_OFFSET; + EXPORT_PER_CPU_SYMBOL(this_cpu_off); + +-unsigned long __per_cpu_offset[NR_CPUS] __read_mostly = { ++unsigned long __per_cpu_offset[NR_CPUS] __read_only = { + [0 ... NR_CPUS-1] = BOOT_PERCPU_OFFSET, + }; + EXPORT_SYMBOL(__per_cpu_offset); +@@ -155,10 +153,10 @@ static inline void setup_percpu_segment( + { + #ifdef CONFIG_X86_32 + struct desc_struct gdt; ++ unsigned long base = per_cpu_offset(cpu); + +- pack_descriptor(&gdt, per_cpu_offset(cpu), 0xFFFFF, +- 0x2 | DESCTYPE_S, 0x8); +- gdt.s = 1; ++ pack_descriptor(&gdt, base, (VMALLOC_END - base - 1) >> PAGE_SHIFT, ++ 0x83 | DESCTYPE_S, 0xC); + write_gdt_entry(get_cpu_gdt_table(cpu), + GDT_ENTRY_PERCPU, &gdt, DESCTYPE_S); + #endif +@@ -207,6 +205,11 @@ void __init setup_per_cpu_areas(void) + /* alrighty, percpu areas up and running */ + delta = (unsigned long)pcpu_base_addr - (unsigned long)__per_cpu_start; + for_each_possible_cpu(cpu) { ++#ifdef CONFIG_CC_STACKPROTECTOR ++#ifdef CONFIG_x86_32 ++ unsigned long canary = per_cpu(stack_canary, cpu); ++#endif ++#endif + per_cpu_offset(cpu) = delta + pcpu_unit_offsets[cpu]; + per_cpu(this_cpu_off, cpu) = per_cpu_offset(cpu); + per_cpu(cpu_number, cpu) = cpu; +@@ -243,6 +246,12 @@ void __init setup_per_cpu_areas(void) + set_cpu_numa_node(cpu, early_cpu_to_node(cpu)); + #endif + #endif ++#ifdef CONFIG_CC_STACKPROTECTOR ++#ifdef CONFIG_x86_32 ++ if (cpu == boot_cpu_id) ++ per_cpu(stack_canary, cpu) = canary; ++#endif ++#endif + /* + * Up to this point, the boot CPU has been using .init.data + * area. Reload any changed state for the boot CPU. +diff -urNp linux-2.6.38.3/arch/x86/kernel/signal.c linux-2.6.38.3/arch/x86/kernel/signal.c +--- linux-2.6.38.3/arch/x86/kernel/signal.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/signal.c 2011-04-17 15:57:32.000000000 -0400 +@@ -198,7 +198,7 @@ static unsigned long align_sigframe(unsi + * Align the stack pointer according to the i386 ABI, + * i.e. so that on function entry ((sp + 4) & 15) == 0. + */ +- sp = ((sp + 4) & -16ul) - 4; ++ sp = ((sp - 12) & -16ul) - 4; + #else /* !CONFIG_X86_32 */ + sp = round_down(sp, 16) - 8; + #endif +@@ -249,11 +249,11 @@ get_sigframe(struct k_sigaction *ka, str + * Return an always-bogus address instead so we will die with SIGSEGV. + */ + if (onsigstack && !likely(on_sig_stack(sp))) +- return (void __user *)-1L; ++ return (__force void __user *)-1L; + + /* save i387 state */ + if (used_math() && save_i387_xstate(*fpstate) < 0) +- return (void __user *)-1L; ++ return (__force void __user *)-1L; + + return (void __user *)sp; + } +@@ -308,9 +308,9 @@ __setup_frame(int sig, struct k_sigactio + } + + if (current->mm->context.vdso) +- restorer = VDSO32_SYMBOL(current->mm->context.vdso, sigreturn); ++ restorer = (__force void __user *)VDSO32_SYMBOL(current->mm->context.vdso, sigreturn); + else +- restorer = &frame->retcode; ++ restorer = (void __user *)&frame->retcode; + if (ka->sa.sa_flags & SA_RESTORER) + restorer = ka->sa.sa_restorer; + +@@ -324,7 +324,7 @@ __setup_frame(int sig, struct k_sigactio + * reasons and because gdb uses it as a signature to notice + * signal handler stack frames. + */ +- err |= __put_user(*((u64 *)&retcode), (u64 *)frame->retcode); ++ err |= __put_user(*((u64 *)&retcode), (u64 __user *)frame->retcode); + + if (err) + return -EFAULT; +@@ -378,7 +378,10 @@ static int __setup_rt_frame(int sig, str + err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)); + + /* Set up to return from userspace. */ +- restorer = VDSO32_SYMBOL(current->mm->context.vdso, rt_sigreturn); ++ if (current->mm->context.vdso) ++ restorer = (__force void __user *)VDSO32_SYMBOL(current->mm->context.vdso, rt_sigreturn); ++ else ++ restorer = (void __user *)&frame->retcode; + if (ka->sa.sa_flags & SA_RESTORER) + restorer = ka->sa.sa_restorer; + put_user_ex(restorer, &frame->pretcode); +@@ -390,7 +393,7 @@ static int __setup_rt_frame(int sig, str + * reasons and because gdb uses it as a signature to notice + * signal handler stack frames. + */ +- put_user_ex(*((u64 *)&rt_retcode), (u64 *)frame->retcode); ++ put_user_ex(*((u64 *)&rt_retcode), (u64 __user *)frame->retcode); + } put_user_catch(err); + + if (err) +@@ -780,7 +783,7 @@ static void do_signal(struct pt_regs *re + * X86_32: vm86 regs switched out by assembly code before reaching + * here, so testing against kernel CS suffices. + */ +- if (!user_mode(regs)) ++ if (!user_mode_novm(regs)) + return; + + if (current_thread_info()->status & TS_RESTORE_SIGMASK) +diff -urNp linux-2.6.38.3/arch/x86/kernel/smpboot.c linux-2.6.38.3/arch/x86/kernel/smpboot.c +--- linux-2.6.38.3/arch/x86/kernel/smpboot.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/smpboot.c 2011-04-17 15:57:32.000000000 -0400 +@@ -773,17 +773,20 @@ static int __cpuinit do_boot_cpu(int api + set_idle_for_cpu(cpu, c_idle.idle); + do_rest: + per_cpu(current_task, cpu) = c_idle.idle; ++ per_cpu(current_tinfo, cpu) = &c_idle.idle->tinfo; + #ifdef CONFIG_X86_32 + /* Stack for startup_32 can be just as for start_secondary onwards */ + irq_ctx_init(cpu); + #else + clear_tsk_thread_flag(c_idle.idle, TIF_FORK); + initial_gs = per_cpu_offset(cpu); +- per_cpu(kernel_stack, cpu) = +- (unsigned long)task_stack_page(c_idle.idle) - +- KERNEL_STACK_OFFSET + THREAD_SIZE; ++ per_cpu(kernel_stack, cpu) = (unsigned long)task_stack_page(c_idle.idle) - 8 + THREAD_SIZE; + #endif ++ ++ pax_open_kernel(); + early_gdt_descr.address = (unsigned long)get_cpu_gdt_table(cpu); ++ pax_close_kernel(); ++ + initial_code = (unsigned long)start_secondary; + stack_start = c_idle.idle->thread.sp; + +@@ -923,6 +926,12 @@ int __cpuinit native_cpu_up(unsigned int + + per_cpu(cpu_state, cpu) = CPU_UP_PREPARE; + ++#ifdef CONFIG_PAX_PER_CPU_PGD ++ clone_pgd_range(get_cpu_pgd(cpu) + KERNEL_PGD_BOUNDARY, ++ swapper_pg_dir + KERNEL_PGD_BOUNDARY, ++ KERNEL_PGD_PTRS); ++#endif ++ + err = do_boot_cpu(apicid, cpu); + if (err) { + pr_debug("do_boot_cpu failed %d\n", err); +diff -urNp linux-2.6.38.3/arch/x86/kernel/step.c linux-2.6.38.3/arch/x86/kernel/step.c +--- linux-2.6.38.3/arch/x86/kernel/step.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/step.c 2011-04-17 15:57:32.000000000 -0400 +@@ -27,10 +27,10 @@ unsigned long convert_ip_to_linear(struc + struct desc_struct *desc; + unsigned long base; + +- seg &= ~7UL; ++ seg >>= 3; + + mutex_lock(&child->mm->context.lock); +- if (unlikely((seg >> 3) >= child->mm->context.size)) ++ if (unlikely(seg >= child->mm->context.size)) + addr = -1L; /* bogus selector, access would fault */ + else { + desc = child->mm->context.ldt + seg; +@@ -42,7 +42,8 @@ unsigned long convert_ip_to_linear(struc + addr += base; + } + mutex_unlock(&child->mm->context.lock); +- } ++ } else if (seg == __KERNEL_CS || seg == __KERNEXEC_KERNEL_CS) ++ addr = ktla_ktva(addr); + + return addr; + } +@@ -53,6 +54,9 @@ static int is_setting_trap_flag(struct t + unsigned char opcode[15]; + unsigned long addr = convert_ip_to_linear(child, regs); + ++ if (addr == -EINVAL) ++ return 0; ++ + copied = access_process_vm(child, addr, opcode, sizeof(opcode), 0); + for (i = 0; i < copied; i++) { + switch (opcode[i]) { +@@ -74,7 +78,7 @@ static int is_setting_trap_flag(struct t + + #ifdef CONFIG_X86_64 + case 0x40 ... 0x4f: +- if (regs->cs != __USER_CS) ++ if ((regs->cs & 0xffff) != __USER_CS) + /* 32-bit mode: register increment */ + return 0; + /* 64-bit mode: REX prefix */ +diff -urNp linux-2.6.38.3/arch/x86/kernel/syscall_table_32.S linux-2.6.38.3/arch/x86/kernel/syscall_table_32.S +--- linux-2.6.38.3/arch/x86/kernel/syscall_table_32.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/syscall_table_32.S 2011-04-17 15:57:32.000000000 -0400 +@@ -1,3 +1,4 @@ ++.section .rodata,"a",@progbits + ENTRY(sys_call_table) + .long sys_restart_syscall /* 0 - old "setup()" system call, used for restarting */ + .long sys_exit +diff -urNp linux-2.6.38.3/arch/x86/kernel/sys_i386_32.c linux-2.6.38.3/arch/x86/kernel/sys_i386_32.c +--- linux-2.6.38.3/arch/x86/kernel/sys_i386_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/sys_i386_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -24,17 +24,224 @@ + + #include <asm/syscalls.h> + +-/* +- * Do a system call from kernel instead of calling sys_execve so we +- * end up with proper pt_regs. +- */ +-int kernel_execve(const char *filename, +- const char *const argv[], +- const char *const envp[]) ++int i386_mmap_check(unsigned long addr, unsigned long len, unsigned long flags) + { +- long __res; +- asm volatile ("int $0x80" +- : "=a" (__res) +- : "0" (__NR_execve), "b" (filename), "c" (argv), "d" (envp) : "memory"); +- return __res; ++ unsigned long pax_task_size = TASK_SIZE; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (current->mm->pax_flags & MF_PAX_SEGMEXEC) ++ pax_task_size = SEGMEXEC_TASK_SIZE; ++#endif ++ ++ if (len > pax_task_size || addr > pax_task_size - len) ++ return -EINVAL; ++ ++ return 0; ++} ++ ++unsigned long ++arch_get_unmapped_area(struct file *filp, unsigned long addr, ++ unsigned long len, unsigned long pgoff, unsigned long flags) ++{ ++ struct mm_struct *mm = current->mm; ++ struct vm_area_struct *vma; ++ unsigned long start_addr, pax_task_size = TASK_SIZE; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (mm->pax_flags & MF_PAX_SEGMEXEC) ++ pax_task_size = SEGMEXEC_TASK_SIZE; ++#endif ++ ++ pax_task_size -= PAGE_SIZE; ++ ++ if (len > pax_task_size) ++ return -ENOMEM; ++ ++ if (flags & MAP_FIXED) ++ return addr; ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (!(mm->pax_flags & MF_PAX_RANDMMAP)) ++#endif ++ ++ if (addr) { ++ addr = PAGE_ALIGN(addr); ++ if (pax_task_size - len >= addr) { ++ vma = find_vma(mm, addr); ++ if (check_heap_stack_gap(vma, addr, len)) ++ return addr; ++ } ++ } ++ if (len > mm->cached_hole_size) { ++ start_addr = addr = mm->free_area_cache; ++ } else { ++ start_addr = addr = mm->mmap_base; ++ mm->cached_hole_size = 0; ++ } ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if (!(__supported_pte_mask & _PAGE_NX) && (mm->pax_flags & MF_PAX_PAGEEXEC) && (flags & MAP_EXECUTABLE) && start_addr >= mm->mmap_base) { ++ start_addr = 0x00110000UL; ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ start_addr += mm->delta_mmap & 0x03FFF000UL; ++#endif ++ ++ if (mm->start_brk <= start_addr && start_addr < mm->mmap_base) ++ start_addr = addr = mm->mmap_base; ++ else ++ addr = start_addr; ++ } ++#endif ++ ++full_search: ++ for (vma = find_vma(mm, addr); ; vma = vma->vm_next) { ++ /* At this point: (!vma || addr < vma->vm_end). */ ++ if (pax_task_size - len < addr) { ++ /* ++ * Start a new search - just in case we missed ++ * some holes. ++ */ ++ if (start_addr != mm->mmap_base) { ++ start_addr = addr = mm->mmap_base; ++ mm->cached_hole_size = 0; ++ goto full_search; ++ } ++ return -ENOMEM; ++ } ++ if (check_heap_stack_gap(vma, addr, len)) ++ break; ++ if (addr + mm->cached_hole_size < vma->vm_start) ++ mm->cached_hole_size = vma->vm_start - addr; ++ addr = vma->vm_end; ++ if (mm->start_brk <= addr && addr < mm->mmap_base) { ++ start_addr = addr = mm->mmap_base; ++ mm->cached_hole_size = 0; ++ goto full_search; ++ } ++ } ++ ++ /* ++ * Remember the place where we stopped the search: ++ */ ++ mm->free_area_cache = addr + len; ++ return addr; ++} ++ ++unsigned long ++arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, ++ const unsigned long len, const unsigned long pgoff, ++ const unsigned long flags) ++{ ++ struct vm_area_struct *vma; ++ struct mm_struct *mm = current->mm; ++ unsigned long base = mm->mmap_base, addr = addr0, pax_task_size = TASK_SIZE; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (mm->pax_flags & MF_PAX_SEGMEXEC) ++ pax_task_size = SEGMEXEC_TASK_SIZE; ++#endif ++ ++ pax_task_size -= PAGE_SIZE; ++ ++ /* requested length too big for entire address space */ ++ if (len > pax_task_size) ++ return -ENOMEM; ++ ++ if (flags & MAP_FIXED) ++ return addr; ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if (!(__supported_pte_mask & _PAGE_NX) && (mm->pax_flags & MF_PAX_PAGEEXEC) && (flags & MAP_EXECUTABLE)) ++ goto bottomup; ++#endif ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (!(mm->pax_flags & MF_PAX_RANDMMAP)) ++#endif ++ ++ /* requesting a specific address */ ++ if (addr) { ++ addr = PAGE_ALIGN(addr); ++ if (pax_task_size - len >= addr) { ++ vma = find_vma(mm, addr); ++ if (check_heap_stack_gap(vma, addr, len)) ++ return addr; ++ } ++ } ++ ++ /* check if free_area_cache is useful for us */ ++ if (len <= mm->cached_hole_size) { ++ mm->cached_hole_size = 0; ++ mm->free_area_cache = mm->mmap_base; ++ } ++ ++ /* either no address requested or can't fit in requested address hole */ ++ addr = mm->free_area_cache; ++ ++ /* make sure it can fit in the remaining address space */ ++ if (addr > len) { ++ vma = find_vma(mm, addr-len); ++ if (check_heap_stack_gap(vma, addr - len, len)) ++ /* remember the address as a hint for next time */ ++ return (mm->free_area_cache = addr-len); ++ } ++ ++ if (mm->mmap_base < len) ++ goto bottomup; ++ ++ addr = mm->mmap_base-len; ++ ++ do { ++ /* ++ * Lookup failure means no vma is above this address, ++ * else if new region fits below vma->vm_start, ++ * return with success: ++ */ ++ vma = find_vma(mm, addr); ++ if (check_heap_stack_gap(vma, addr, len)) ++ /* remember the address as a hint for next time */ ++ return (mm->free_area_cache = addr); ++ ++ /* remember the largest hole we saw so far */ ++ if (addr + mm->cached_hole_size < vma->vm_start) ++ mm->cached_hole_size = vma->vm_start - addr; ++ ++ /* try just below the current vma->vm_start */ ++ addr = skip_heap_stack_gap(vma, len); ++ } while (!IS_ERR_VALUE(addr)); ++ ++bottomup: ++ /* ++ * A failed mmap() very likely causes application failure, ++ * so fall back to the bottom-up function here. This scenario ++ * can happen with large stack limits and large mmap() ++ * allocations. ++ */ ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (mm->pax_flags & MF_PAX_SEGMEXEC) ++ mm->mmap_base = SEGMEXEC_TASK_UNMAPPED_BASE; ++ else ++#endif ++ ++ mm->mmap_base = TASK_UNMAPPED_BASE; ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base += mm->delta_mmap; ++#endif ++ ++ mm->free_area_cache = mm->mmap_base; ++ mm->cached_hole_size = ~0UL; ++ addr = arch_get_unmapped_area(filp, addr0, len, pgoff, flags); ++ /* ++ * Restore the topdown base: ++ */ ++ mm->mmap_base = base; ++ mm->free_area_cache = base; ++ mm->cached_hole_size = ~0UL; ++ ++ return addr; + } +diff -urNp linux-2.6.38.3/arch/x86/kernel/sys_x86_64.c linux-2.6.38.3/arch/x86/kernel/sys_x86_64.c +--- linux-2.6.38.3/arch/x86/kernel/sys_x86_64.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/sys_x86_64.c 2011-04-17 15:57:32.000000000 -0400 +@@ -32,8 +32,8 @@ out: + return error; + } + +-static void find_start_end(unsigned long flags, unsigned long *begin, +- unsigned long *end) ++static void find_start_end(struct mm_struct *mm, unsigned long flags, ++ unsigned long *begin, unsigned long *end) + { + if (!test_thread_flag(TIF_IA32) && (flags & MAP_32BIT)) { + unsigned long new_begin; +@@ -52,7 +52,7 @@ static void find_start_end(unsigned long + *begin = new_begin; + } + } else { +- *begin = TASK_UNMAPPED_BASE; ++ *begin = mm->mmap_base; + *end = TASK_SIZE; + } + } +@@ -69,16 +69,19 @@ arch_get_unmapped_area(struct file *filp + if (flags & MAP_FIXED) + return addr; + +- find_start_end(flags, &begin, &end); ++ find_start_end(mm, flags, &begin, &end); + + if (len > end) + return -ENOMEM; + ++#ifdef CONFIG_PAX_RANDMMAP ++ if (!(mm->pax_flags & MF_PAX_RANDMMAP)) ++#endif ++ + if (addr) { + addr = PAGE_ALIGN(addr); + vma = find_vma(mm, addr); +- if (end - len >= addr && +- (!vma || addr + len <= vma->vm_start)) ++ if (end - len >= addr && check_heap_stack_gap(vma, addr, len)) + return addr; + } + if (((flags & MAP_32BIT) || test_thread_flag(TIF_IA32)) +@@ -106,7 +109,7 @@ full_search: + } + return -ENOMEM; + } +- if (!vma || addr + len <= vma->vm_start) { ++ if (check_heap_stack_gap(vma, addr, len)) { + /* + * Remember the place where we stopped the search: + */ +@@ -128,7 +131,7 @@ arch_get_unmapped_area_topdown(struct fi + { + struct vm_area_struct *vma; + struct mm_struct *mm = current->mm; +- unsigned long addr = addr0; ++ unsigned long base = mm->mmap_base, addr = addr0; + + /* requested length too big for entire address space */ + if (len > TASK_SIZE) +@@ -141,13 +144,18 @@ arch_get_unmapped_area_topdown(struct fi + if (!test_thread_flag(TIF_IA32) && (flags & MAP_32BIT)) + goto bottomup; + ++#ifdef CONFIG_PAX_RANDMMAP ++ if (!(mm->pax_flags & MF_PAX_RANDMMAP)) ++#endif ++ + /* requesting a specific address */ + if (addr) { + addr = PAGE_ALIGN(addr); +- vma = find_vma(mm, addr); +- if (TASK_SIZE - len >= addr && +- (!vma || addr + len <= vma->vm_start)) +- return addr; ++ if (TASK_SIZE - len >= addr) { ++ vma = find_vma(mm, addr); ++ if (check_heap_stack_gap(vma, addr, len)) ++ return addr; ++ } + } + + /* check if free_area_cache is useful for us */ +@@ -162,7 +170,7 @@ arch_get_unmapped_area_topdown(struct fi + /* make sure it can fit in the remaining address space */ + if (addr > len) { + vma = find_vma(mm, addr-len); +- if (!vma || addr <= vma->vm_start) ++ if (check_heap_stack_gap(vma, addr - len, len)) + /* remember the address as a hint for next time */ + return mm->free_area_cache = addr-len; + } +@@ -179,7 +187,7 @@ arch_get_unmapped_area_topdown(struct fi + * return with success: + */ + vma = find_vma(mm, addr); +- if (!vma || addr+len <= vma->vm_start) ++ if (check_heap_stack_gap(vma, addr, len)) + /* remember the address as a hint for next time */ + return mm->free_area_cache = addr; + +@@ -188,8 +196,8 @@ arch_get_unmapped_area_topdown(struct fi + mm->cached_hole_size = vma->vm_start - addr; + + /* try just below the current vma->vm_start */ +- addr = vma->vm_start-len; +- } while (len < vma->vm_start); ++ addr = skip_heap_stack_gap(vma, len); ++ } while (!IS_ERR_VALUE(addr)); + + bottomup: + /* +@@ -198,13 +206,21 @@ bottomup: + * can happen with large stack limits and large mmap() + * allocations. + */ ++ mm->mmap_base = TASK_UNMAPPED_BASE; ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base += mm->delta_mmap; ++#endif ++ ++ mm->free_area_cache = mm->mmap_base; + mm->cached_hole_size = ~0UL; +- mm->free_area_cache = TASK_UNMAPPED_BASE; + addr = arch_get_unmapped_area(filp, addr0, len, pgoff, flags); + /* + * Restore the topdown base: + */ +- mm->free_area_cache = mm->mmap_base; ++ mm->mmap_base = base; ++ mm->free_area_cache = base; + mm->cached_hole_size = ~0UL; + + return addr; +diff -urNp linux-2.6.38.3/arch/x86/kernel/time.c linux-2.6.38.3/arch/x86/kernel/time.c +--- linux-2.6.38.3/arch/x86/kernel/time.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/time.c 2011-04-17 15:57:32.000000000 -0400 +@@ -22,17 +22,13 @@ + #include <asm/hpet.h> + #include <asm/time.h> + +-#ifdef CONFIG_X86_64 +-volatile unsigned long __jiffies __section_jiffies = INITIAL_JIFFIES; +-#endif +- + unsigned long profile_pc(struct pt_regs *regs) + { + unsigned long pc = instruction_pointer(regs); + +- if (!user_mode_vm(regs) && in_lock_functions(pc)) { ++ if (!user_mode(regs) && in_lock_functions(pc)) { + #ifdef CONFIG_FRAME_POINTER +- return *(unsigned long *)(regs->bp + sizeof(long)); ++ return ktla_ktva(*(unsigned long *)(regs->bp + sizeof(long))); + #else + unsigned long *sp = + (unsigned long *)kernel_stack_pointer(regs); +@@ -41,11 +37,17 @@ unsigned long profile_pc(struct pt_regs + * or above a saved flags. Eflags has bits 22-31 zero, + * kernel addresses don't. + */ ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ return ktla_ktva(sp[0]); ++#else + if (sp[0] >> 22) + return sp[0]; + if (sp[1] >> 22) + return sp[1]; + #endif ++ ++#endif + } + return pc; + } +diff -urNp linux-2.6.38.3/arch/x86/kernel/tls.c linux-2.6.38.3/arch/x86/kernel/tls.c +--- linux-2.6.38.3/arch/x86/kernel/tls.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/tls.c 2011-04-17 15:57:32.000000000 -0400 +@@ -85,6 +85,11 @@ int do_set_thread_area(struct task_struc + if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX) + return -EINVAL; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ if ((p->mm->pax_flags & MF_PAX_SEGMEXEC) && (info.contents & MODIFY_LDT_CONTENTS_CODE)) ++ return -EINVAL; ++#endif ++ + set_tls_desc(p, idx, &info, 1); + + return 0; +diff -urNp linux-2.6.38.3/arch/x86/kernel/trampoline_32.S linux-2.6.38.3/arch/x86/kernel/trampoline_32.S +--- linux-2.6.38.3/arch/x86/kernel/trampoline_32.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/trampoline_32.S 2011-04-17 15:57:32.000000000 -0400 +@@ -32,6 +32,12 @@ + #include <asm/segment.h> + #include <asm/page_types.h> + ++#ifdef CONFIG_PAX_KERNEXEC ++#define ta(X) (X) ++#else ++#define ta(X) ((X) - __PAGE_OFFSET) ++#endif ++ + /* We can free up trampoline after bootup if cpu hotplug is not supported. */ + __CPUINITRODATA + .code16 +@@ -60,7 +66,7 @@ r_base = . + inc %ax # protected mode (PE) bit + lmsw %ax # into protected mode + # flush prefetch and jump to startup_32_smp in arch/i386/kernel/head.S +- ljmpl $__BOOT_CS, $(startup_32_smp-__PAGE_OFFSET) ++ ljmpl $__BOOT_CS, $ta(startup_32_smp) + + # These need to be in the same 64K segment as the above; + # hence we don't use the boot_gdt_descr defined in head.S +diff -urNp linux-2.6.38.3/arch/x86/kernel/trampoline_64.S linux-2.6.38.3/arch/x86/kernel/trampoline_64.S +--- linux-2.6.38.3/arch/x86/kernel/trampoline_64.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/trampoline_64.S 2011-04-17 15:57:32.000000000 -0400 +@@ -91,7 +91,7 @@ startup_32: + movl $__KERNEL_DS, %eax # Initialize the %ds segment register + movl %eax, %ds + +- movl $X86_CR4_PAE, %eax ++ movl $(X86_CR4_PSE | X86_CR4_PAE | X86_CR4_PGE), %eax + movl %eax, %cr4 # Enable PAE mode + + # Setup trampoline 4 level pagetables +@@ -138,7 +138,7 @@ tidt: + # so the kernel can live anywhere + .balign 4 + tgdt: +- .short tgdt_end - tgdt # gdt limit ++ .short tgdt_end - tgdt - 1 # gdt limit + .long tgdt - r_base + .short 0 + .quad 0x00cf9b000000ffff # __KERNEL32_CS +diff -urNp linux-2.6.38.3/arch/x86/kernel/traps.c linux-2.6.38.3/arch/x86/kernel/traps.c +--- linux-2.6.38.3/arch/x86/kernel/traps.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/traps.c 2011-04-17 15:57:32.000000000 -0400 +@@ -70,12 +70,6 @@ asmlinkage int system_call(void); + + /* Do we ignore FPU interrupts ? */ + char ignore_fpu_irq; +- +-/* +- * The IDT has to be page-aligned to simplify the Pentium +- * F0 0F bug workaround. +- */ +-gate_desc idt_table[NR_VECTORS] __page_aligned_data = { { { { 0, 0 } } }, }; + #endif + + DECLARE_BITMAP(used_vectors, NR_VECTORS); +@@ -117,13 +111,13 @@ static inline void preempt_conditional_c + } + + static void __kprobes +-do_trap(int trapnr, int signr, char *str, struct pt_regs *regs, ++do_trap(int trapnr, int signr, const char *str, struct pt_regs *regs, + long error_code, siginfo_t *info) + { + struct task_struct *tsk = current; + + #ifdef CONFIG_X86_32 +- if (regs->flags & X86_VM_MASK) { ++ if (v8086_mode(regs)) { + /* + * traps 0, 1, 3, 4, and 5 should be forwarded to vm86. + * On nmi (interrupt 2), do_trap should not be called. +@@ -134,7 +128,7 @@ do_trap(int trapnr, int signr, char *str + } + #endif + +- if (!user_mode(regs)) ++ if (!user_mode_novm(regs)) + goto kernel_trap; + + #ifdef CONFIG_X86_32 +@@ -157,7 +151,7 @@ trap_signal: + printk_ratelimit()) { + printk(KERN_INFO + "%s[%d] trap %s ip:%lx sp:%lx error:%lx", +- tsk->comm, tsk->pid, str, ++ tsk->comm, task_pid_nr(tsk), str, + regs->ip, regs->sp, error_code); + print_vma_addr(" in ", regs->ip); + printk("\n"); +@@ -174,8 +168,20 @@ kernel_trap: + if (!fixup_exception(regs)) { + tsk->thread.error_code = error_code; + tsk->thread.trap_no = trapnr; ++ ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) ++ if (trapnr == 12 && ((regs->cs & 0xFFFF) == __KERNEL_CS || (regs->cs & 0xFFFF) == __KERNEXEC_KERNEL_CS)) ++ str = "PAX: suspicious stack segment fault"; ++#endif ++ + die(str, regs, error_code); + } ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ if (trapnr == 4) ++ pax_report_refcount_overflow(regs); ++#endif ++ + return; + + #ifdef CONFIG_X86_32 +@@ -264,14 +270,30 @@ do_general_protection(struct pt_regs *re + conditional_sti(regs); + + #ifdef CONFIG_X86_32 +- if (regs->flags & X86_VM_MASK) ++ if (v8086_mode(regs)) + goto gp_in_vm86; + #endif + + tsk = current; +- if (!user_mode(regs)) ++ if (!user_mode_novm(regs)) + goto gp_in_kernel; + ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_PAGEEXEC) ++ if (!(__supported_pte_mask & _PAGE_NX) && tsk->mm && (tsk->mm->pax_flags & MF_PAX_PAGEEXEC)) { ++ struct mm_struct *mm = tsk->mm; ++ unsigned long limit; ++ ++ down_write(&mm->mmap_sem); ++ limit = mm->context.user_cs_limit; ++ if (limit < TASK_SIZE) { ++ track_exec_limit(mm, limit, TASK_SIZE, VM_EXEC); ++ up_write(&mm->mmap_sem); ++ return; ++ } ++ up_write(&mm->mmap_sem); ++ } ++#endif ++ + tsk->thread.error_code = error_code; + tsk->thread.trap_no = 13; + +@@ -304,6 +326,13 @@ gp_in_kernel: + if (notify_die(DIE_GPF, "general protection fault", regs, + error_code, 13, SIGSEGV) == NOTIFY_STOP) + return; ++ ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) ++ if ((regs->cs & 0xFFFF) == __KERNEL_CS || (regs->cs & 0xFFFF) == __KERNEXEC_KERNEL_CS) ++ die("PAX: suspicious general protection fault", regs, error_code); ++ else ++#endif ++ + die("general protection fault", regs, error_code); + } + +@@ -569,7 +598,7 @@ dotraplinkage void __kprobes do_debug(st + /* It's safe to allow irq's after DR6 has been saved */ + preempt_conditional_sti(regs); + +- if (regs->flags & X86_VM_MASK) { ++ if (v8086_mode(regs)) { + handle_vm86_trap((struct kernel_vm86_regs *) regs, + error_code, 1); + preempt_conditional_cli(regs); +@@ -583,7 +612,7 @@ dotraplinkage void __kprobes do_debug(st + * We already checked v86 mode above, so we can check for kernel mode + * by just checking the CPL of CS. + */ +- if ((dr6 & DR_STEP) && !user_mode(regs)) { ++ if ((dr6 & DR_STEP) && !user_mode_novm(regs)) { + tsk->thread.debugreg6 &= ~DR_STEP; + set_tsk_thread_flag(tsk, TIF_SINGLESTEP); + regs->flags &= ~X86_EFLAGS_TF; +@@ -612,7 +641,7 @@ void math_error(struct pt_regs *regs, in + return; + conditional_sti(regs); + +- if (!user_mode_vm(regs)) ++ if (!user_mode(regs)) + { + if (!fixup_exception(regs)) { + task->thread.error_code = error_code; +@@ -723,7 +752,7 @@ asmlinkage void __attribute__((weak)) sm + void __math_state_restore(void) + { + struct thread_info *thread = current_thread_info(); +- struct task_struct *tsk = thread->task; ++ struct task_struct *tsk = current; + + /* + * Paranoid restore. send a SIGSEGV if we fail to restore the state. +@@ -750,8 +779,7 @@ void __math_state_restore(void) + */ + asmlinkage void math_state_restore(void) + { +- struct thread_info *thread = current_thread_info(); +- struct task_struct *tsk = thread->task; ++ struct task_struct *tsk = current; + + if (!tsk_used_math(tsk)) { + local_irq_enable(); +diff -urNp linux-2.6.38.3/arch/x86/kernel/tsc.c linux-2.6.38.3/arch/x86/kernel/tsc.c +--- linux-2.6.38.3/arch/x86/kernel/tsc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/tsc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -837,7 +837,7 @@ static struct dmi_system_id __initdata b + DMI_MATCH(DMI_BOARD_NAME, "2635FA0"), + }, + }, +- {} ++ { NULL, NULL, {{0, {0}}}, NULL} + }; + + static void __init check_system_tsc_reliable(void) +diff -urNp linux-2.6.38.3/arch/x86/kernel/vm86_32.c linux-2.6.38.3/arch/x86/kernel/vm86_32.c +--- linux-2.6.38.3/arch/x86/kernel/vm86_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/vm86_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -41,6 +41,7 @@ + #include <linux/ptrace.h> + #include <linux/audit.h> + #include <linux/stddef.h> ++#include <linux/grsecurity.h> + + #include <asm/uaccess.h> + #include <asm/io.h> +@@ -148,7 +149,7 @@ struct pt_regs *save_v86_state(struct ke + do_exit(SIGSEGV); + } + +- tss = &per_cpu(init_tss, get_cpu()); ++ tss = init_tss + get_cpu(); + current->thread.sp0 = current->thread.saved_sp0; + current->thread.sysenter_cs = __KERNEL_CS; + load_sp0(tss, ¤t->thread); +@@ -208,6 +209,13 @@ int sys_vm86old(struct vm86_struct __use + struct task_struct *tsk; + int tmp, ret = -EPERM; + ++#ifdef CONFIG_GRKERNSEC_VM86 ++ if (!capable(CAP_SYS_RAWIO)) { ++ gr_handle_vm86(); ++ goto out; ++ } ++#endif ++ + tsk = current; + if (tsk->thread.saved_sp0) + goto out; +@@ -238,6 +246,14 @@ int sys_vm86(unsigned long cmd, unsigned + int tmp, ret; + struct vm86plus_struct __user *v86; + ++#ifdef CONFIG_GRKERNSEC_VM86 ++ if (!capable(CAP_SYS_RAWIO)) { ++ gr_handle_vm86(); ++ ret = -EPERM; ++ goto out; ++ } ++#endif ++ + tsk = current; + switch (cmd) { + case VM86_REQUEST_IRQ: +@@ -324,7 +340,7 @@ static void do_sys_vm86(struct kernel_vm + tsk->thread.saved_fs = info->regs32->fs; + tsk->thread.saved_gs = get_user_gs(info->regs32); + +- tss = &per_cpu(init_tss, get_cpu()); ++ tss = init_tss + get_cpu(); + tsk->thread.sp0 = (unsigned long) &info->VM86_TSS_ESP0; + if (cpu_has_sep) + tsk->thread.sysenter_cs = 0; +@@ -529,7 +545,7 @@ static void do_int(struct kernel_vm86_re + goto cannot_handle; + if (i == 0x21 && is_revectored(AH(regs), &KVM86->int21_revectored)) + goto cannot_handle; +- intr_ptr = (unsigned long __user *) (i << 2); ++ intr_ptr = (__force unsigned long __user *) (i << 2); + if (get_user(segoffs, intr_ptr)) + goto cannot_handle; + if ((segoffs >> 16) == BIOSSEG) +diff -urNp linux-2.6.38.3/arch/x86/kernel/vmlinux.lds.S linux-2.6.38.3/arch/x86/kernel/vmlinux.lds.S +--- linux-2.6.38.3/arch/x86/kernel/vmlinux.lds.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/vmlinux.lds.S 2011-04-17 15:57:32.000000000 -0400 +@@ -26,6 +26,13 @@ + #include <asm/page_types.h> + #include <asm/cache.h> + #include <asm/boot.h> ++#include <asm/segment.h> ++ ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) ++#define __KERNEL_TEXT_OFFSET (LOAD_OFFSET + ____LOAD_PHYSICAL_ADDR) ++#else ++#define __KERNEL_TEXT_OFFSET 0 ++#endif + + #undef i386 /* in case the preprocessor is a 32bit one */ + +@@ -34,11 +41,9 @@ OUTPUT_FORMAT(CONFIG_OUTPUT_FORMAT, CONF + #ifdef CONFIG_X86_32 + OUTPUT_ARCH(i386) + ENTRY(phys_startup_32) +-jiffies = jiffies_64; + #else + OUTPUT_ARCH(i386:x86-64) + ENTRY(phys_startup_64) +-jiffies_64 = jiffies; + #endif + + #if defined(CONFIG_X86_64) && defined(CONFIG_DEBUG_RODATA) +@@ -69,31 +74,46 @@ jiffies_64 = jiffies; + + PHDRS { + text PT_LOAD FLAGS(5); /* R_E */ ++#ifdef CONFIG_X86_32 ++ module PT_LOAD FLAGS(5); /* R_E */ ++#endif ++#ifdef CONFIG_XEN ++ rodata PT_LOAD FLAGS(5); /* R_E */ ++#else ++ rodata PT_LOAD FLAGS(4); /* R__ */ ++#endif + data PT_LOAD FLAGS(6); /* RW_ */ + #ifdef CONFIG_X86_64 + user PT_LOAD FLAGS(5); /* R_E */ ++#endif ++ init.begin PT_LOAD FLAGS(6); /* RW_ */ + #ifdef CONFIG_SMP + percpu PT_LOAD FLAGS(6); /* RW_ */ + #endif ++ text.init PT_LOAD FLAGS(5); /* R_E */ ++ text.exit PT_LOAD FLAGS(5); /* R_E */ + init PT_LOAD FLAGS(7); /* RWE */ +-#endif + note PT_NOTE FLAGS(0); /* ___ */ + } + + SECTIONS + { + #ifdef CONFIG_X86_32 +- . = LOAD_OFFSET + LOAD_PHYSICAL_ADDR; +- phys_startup_32 = startup_32 - LOAD_OFFSET; ++ . = LOAD_OFFSET + ____LOAD_PHYSICAL_ADDR; + #else +- . = __START_KERNEL; +- phys_startup_64 = startup_64 - LOAD_OFFSET; ++ . = __START_KERNEL; + #endif + + /* Text and read-only data */ +- .text : AT(ADDR(.text) - LOAD_OFFSET) { +- _text = .; ++ .text (. - __KERNEL_TEXT_OFFSET): AT(ADDR(.text) - LOAD_OFFSET + __KERNEL_TEXT_OFFSET) { + /* bootstrapping code */ ++#ifdef CONFIG_X86_32 ++ phys_startup_32 = startup_32 - LOAD_OFFSET + __KERNEL_TEXT_OFFSET; ++#else ++ phys_startup_64 = startup_64 - LOAD_OFFSET + __KERNEL_TEXT_OFFSET; ++#endif ++ __LOAD_PHYSICAL_ADDR = . - LOAD_OFFSET + __KERNEL_TEXT_OFFSET; ++ _text = .; + HEAD_TEXT + #ifdef CONFIG_X86_32 + . = ALIGN(PAGE_SIZE); +@@ -108,13 +128,47 @@ SECTIONS + IRQENTRY_TEXT + *(.fixup) + *(.gnu.warning) +- /* End of text section */ +- _etext = .; + } :text = 0x9090 + +- NOTES :text :note ++ . += __KERNEL_TEXT_OFFSET; ++ ++#ifdef CONFIG_X86_32 ++ . = ALIGN(PAGE_SIZE); ++ .module.text : AT(ADDR(.module.text) - LOAD_OFFSET) { ++ ++#if defined(CONFIG_PAX_KERNEXEC) && defined(CONFIG_MODULES) ++ MODULES_EXEC_VADDR = .; ++ BYTE(0) ++ . += (CONFIG_PAX_KERNEXEC_MODULE_TEXT * 1024 * 1024); ++ . = ALIGN(HPAGE_SIZE); ++ MODULES_EXEC_END = . - 1; ++#endif ++ ++ } :module ++#endif ++ ++ .text.end : AT(ADDR(.text.end) - LOAD_OFFSET) { ++ /* End of text section */ ++ _etext = . - __KERNEL_TEXT_OFFSET; ++ } + +- EXCEPTION_TABLE(16) :text = 0x9090 ++#ifdef CONFIG_X86_32 ++ . = ALIGN(PAGE_SIZE); ++ .rodata.page_aligned : AT(ADDR(.rodata.page_aligned) - LOAD_OFFSET) { ++ *(.idt) ++ . = ALIGN(PAGE_SIZE); ++ *(.empty_zero_page) ++ *(.initial_pg_fixmap) ++ *(.initial_pg_pmd) ++ *(.initial_page_table) ++ *(.swapper_pg_dir) ++ } :rodata ++#endif ++ ++ . = ALIGN(PAGE_SIZE); ++ NOTES :rodata :note ++ ++ EXCEPTION_TABLE(16) :rodata + + #if defined(CONFIG_DEBUG_RODATA) + /* .text should occupy whole number of pages */ +@@ -126,16 +180,20 @@ SECTIONS + + /* Data */ + .data : AT(ADDR(.data) - LOAD_OFFSET) { ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ . = ALIGN(HPAGE_SIZE); ++#else ++ . = ALIGN(PAGE_SIZE); ++#endif ++ + /* Start of data section */ + _sdata = .; + + /* init_task */ + INIT_TASK_DATA(THREAD_SIZE) + +-#ifdef CONFIG_X86_32 +- /* 32 bit has nosave before _edata */ + NOSAVE_DATA +-#endif + + PAGE_ALIGNED_DATA(PAGE_SIZE) + +@@ -144,6 +202,8 @@ SECTIONS + DATA_DATA + CONSTRUCTORS + ++ jiffies = jiffies_64; ++ + /* rarely changed data like cpu maps */ + READ_MOSTLY_DATA(INTERNODE_CACHE_BYTES) + +@@ -198,12 +258,6 @@ SECTIONS + } + vgetcpu_mode = VVIRT(.vgetcpu_mode); + +- . = ALIGN(L1_CACHE_BYTES); +- .jiffies : AT(VLOAD(.jiffies)) { +- *(.jiffies) +- } +- jiffies = VVIRT(.jiffies); +- + .vsyscall_3 ADDR(.vsyscall_0) + 3072: AT(VLOAD(.vsyscall_3)) { + *(.vsyscall_3) + } +@@ -219,12 +273,19 @@ SECTIONS + #endif /* CONFIG_X86_64 */ + + /* Init code and data - will be freed after init */ +- . = ALIGN(PAGE_SIZE); + .init.begin : AT(ADDR(.init.begin) - LOAD_OFFSET) { ++ BYTE(0) ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ . = ALIGN(HPAGE_SIZE); ++#else ++ . = ALIGN(PAGE_SIZE); ++#endif ++ + __init_begin = .; /* paired with __init_end */ +- } ++ } :init.begin + +-#if defined(CONFIG_X86_64) && defined(CONFIG_SMP) ++#ifdef CONFIG_SMP + /* + * percpu offsets are zero-based on SMP. PERCPU_VADDR() changes the + * output PHDR, so the next output section - .init.text - should +@@ -233,12 +294,27 @@ SECTIONS + PERCPU_VADDR(0, :percpu) + #endif + +- INIT_TEXT_SECTION(PAGE_SIZE) +-#ifdef CONFIG_X86_64 +- :init +-#endif ++ . = ALIGN(PAGE_SIZE); ++ init_begin = .; ++ .init.text (. - __KERNEL_TEXT_OFFSET): AT(init_begin - LOAD_OFFSET) { ++ VMLINUX_SYMBOL(_sinittext) = .; ++ INIT_TEXT ++ VMLINUX_SYMBOL(_einittext) = .; ++ . = ALIGN(PAGE_SIZE); ++ } :text.init + +- INIT_DATA_SECTION(16) ++ /* ++ * .exit.text is discard at runtime, not link time, to deal with ++ * references from .altinstructions and .eh_frame ++ */ ++ .exit.text : AT(ADDR(.exit.text) - LOAD_OFFSET + __KERNEL_TEXT_OFFSET) { ++ EXIT_TEXT ++ . = ALIGN(16); ++ } :text.exit ++ . = init_begin + SIZEOF(.init.text) + SIZEOF(.exit.text); ++ ++ . = ALIGN(PAGE_SIZE); ++ INIT_DATA_SECTION(16) :init + + .x86_cpu_dev.init : AT(ADDR(.x86_cpu_dev.init) - LOAD_OFFSET) { + __x86_cpu_dev_start = .; +@@ -292,19 +368,12 @@ SECTIONS + __iommu_table_end = .; + } + . = ALIGN(8); +- /* +- * .exit.text is discard at runtime, not link time, to deal with +- * references from .altinstructions and .eh_frame +- */ +- .exit.text : AT(ADDR(.exit.text) - LOAD_OFFSET) { +- EXIT_TEXT +- } + + .exit.data : AT(ADDR(.exit.data) - LOAD_OFFSET) { + EXIT_DATA + } + +-#if !defined(CONFIG_X86_64) || !defined(CONFIG_SMP) ++#ifndef CONFIG_SMP + PERCPU(THREAD_SIZE) + #endif + +@@ -323,16 +392,10 @@ SECTIONS + .smp_locks : AT(ADDR(.smp_locks) - LOAD_OFFSET) { + __smp_locks = .; + *(.smp_locks) +- . = ALIGN(PAGE_SIZE); + __smp_locks_end = .; ++ . = ALIGN(PAGE_SIZE); + } + +-#ifdef CONFIG_X86_64 +- .data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) { +- NOSAVE_DATA +- } +-#endif +- + /* BSS */ + . = ALIGN(PAGE_SIZE); + .bss : AT(ADDR(.bss) - LOAD_OFFSET) { +@@ -348,6 +411,7 @@ SECTIONS + __brk_base = .; + . += 64 * 1024; /* 64k alignment slop space */ + *(.brk_reservation) /* areas brk users have reserved */ ++ . = ALIGN(HPAGE_SIZE); + __brk_limit = .; + } + +@@ -374,13 +438,12 @@ SECTIONS + * for the boot processor. + */ + #define INIT_PER_CPU(x) init_per_cpu__##x = x + __per_cpu_load +-INIT_PER_CPU(gdt_page); + INIT_PER_CPU(irq_stack_union); + + /* + * Build-time check on the image size: + */ +-. = ASSERT((_end - _text <= KERNEL_IMAGE_SIZE), ++. = ASSERT((_end - _text - __KERNEL_TEXT_OFFSET <= KERNEL_IMAGE_SIZE), + "kernel image bigger than KERNEL_IMAGE_SIZE"); + + #ifdef CONFIG_SMP +diff -urNp linux-2.6.38.3/arch/x86/kernel/vsyscall_64.c linux-2.6.38.3/arch/x86/kernel/vsyscall_64.c +--- linux-2.6.38.3/arch/x86/kernel/vsyscall_64.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/vsyscall_64.c 2011-04-17 15:57:32.000000000 -0400 +@@ -80,6 +80,7 @@ void update_vsyscall(struct timespec *wa + + write_seqlock_irqsave(&vsyscall_gtod_data.lock, flags); + /* copy vsyscall data */ ++ strlcpy(vsyscall_gtod_data.clock.name, clock->name, sizeof vsyscall_gtod_data.clock.name); + vsyscall_gtod_data.clock.vread = clock->vread; + vsyscall_gtod_data.clock.cycle_last = clock->cycle_last; + vsyscall_gtod_data.clock.mask = clock->mask; +@@ -208,7 +209,7 @@ vgetcpu(unsigned *cpu, unsigned *node, s + We do this here because otherwise user space would do it on + its own in a likely inferior way (no access to jiffies). + If you don't like it pass NULL. */ +- if (tcache && tcache->blob[0] == (j = __jiffies)) { ++ if (tcache && tcache->blob[0] == (j = jiffies)) { + p = tcache->blob[1]; + } else if (__vgetcpu_mode == VGETCPU_RDTSCP) { + /* Load per CPU data from RDTSCP */ +diff -urNp linux-2.6.38.3/arch/x86/kernel/x8664_ksyms_64.c linux-2.6.38.3/arch/x86/kernel/x8664_ksyms_64.c +--- linux-2.6.38.3/arch/x86/kernel/x8664_ksyms_64.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/x8664_ksyms_64.c 2011-04-17 15:57:32.000000000 -0400 +@@ -29,8 +29,6 @@ EXPORT_SYMBOL(__put_user_8); + EXPORT_SYMBOL(copy_user_generic_string); + EXPORT_SYMBOL(copy_user_generic_unrolled); + EXPORT_SYMBOL(__copy_user_nocache); +-EXPORT_SYMBOL(_copy_from_user); +-EXPORT_SYMBOL(_copy_to_user); + + EXPORT_SYMBOL(copy_page); + EXPORT_SYMBOL(clear_page); +diff -urNp linux-2.6.38.3/arch/x86/kernel/xsave.c linux-2.6.38.3/arch/x86/kernel/xsave.c +--- linux-2.6.38.3/arch/x86/kernel/xsave.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kernel/xsave.c 2011-04-17 15:57:32.000000000 -0400 +@@ -130,7 +130,7 @@ int check_for_xstate(struct i387_fxsave_ + fx_sw_user->xstate_size > fx_sw_user->extended_size) + return -EINVAL; + +- err = __get_user(magic2, (__u32 *) (((void *)fpstate) + ++ err = __get_user(magic2, (__u32 __user *) (((void __user *)fpstate) + + fx_sw_user->extended_size - + FP_XSTATE_MAGIC2_SIZE)); + if (err) +@@ -267,7 +267,7 @@ fx_only: + * the other extended state. + */ + xrstor_state(init_xstate_buf, pcntxt_mask & ~XSTATE_FPSSE); +- return fxrstor_checking((__force struct i387_fxsave_struct *)buf); ++ return fxrstor_checking((struct i387_fxsave_struct __user *)buf); + } + + /* +@@ -299,7 +299,7 @@ int restore_i387_xstate(void __user *buf + if (use_xsave()) + err = restore_user_xstate(buf); + else +- err = fxrstor_checking((__force struct i387_fxsave_struct *) ++ err = fxrstor_checking((struct i387_fxsave_struct __user *) + buf); + if (unlikely(err)) { + /* +diff -urNp linux-2.6.38.3/arch/x86/kvm/emulate.c linux-2.6.38.3/arch/x86/kvm/emulate.c +--- linux-2.6.38.3/arch/x86/kvm/emulate.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kvm/emulate.c 2011-04-17 15:57:32.000000000 -0400 +@@ -88,7 +88,7 @@ + #define Src2ImmByte (2<<29) + #define Src2One (3<<29) + #define Src2Imm (4<<29) +-#define Src2Mask (7<<29) ++#define Src2Mask (7U<<29) + + #define X2(x...) x, x + #define X3(x...) X2(x), x +@@ -189,6 +189,7 @@ struct group_dual { + + #define ____emulate_2op(_op, _src, _dst, _eflags, _x, _y, _suffix, _dsttype) \ + do { \ ++ unsigned long _tmp; \ + __asm__ __volatile__ ( \ + _PRE_EFLAGS("0", "4", "2") \ + _op _suffix " %"_x"3,%1; " \ +@@ -202,8 +203,6 @@ struct group_dual { + /* Raw emulation: instruction has two explicit operands. */ + #define __emulate_2op_nobyte(_op,_src,_dst,_eflags,_wx,_wy,_lx,_ly,_qx,_qy) \ + do { \ +- unsigned long _tmp; \ +- \ + switch ((_dst).bytes) { \ + case 2: \ + ____emulate_2op(_op,_src,_dst,_eflags,_wx,_wy,"w",u16);\ +@@ -219,7 +218,6 @@ struct group_dual { + + #define __emulate_2op(_op,_src,_dst,_eflags,_bx,_by,_wx,_wy,_lx,_ly,_qx,_qy) \ + do { \ +- unsigned long _tmp; \ + switch ((_dst).bytes) { \ + case 1: \ + ____emulate_2op(_op,_src,_dst,_eflags,_bx,_by,"b",u8); \ +diff -urNp linux-2.6.38.3/arch/x86/kvm/lapic.c linux-2.6.38.3/arch/x86/kvm/lapic.c +--- linux-2.6.38.3/arch/x86/kvm/lapic.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kvm/lapic.c 2011-04-17 15:57:32.000000000 -0400 +@@ -53,7 +53,7 @@ + #define APIC_BUS_CYCLE_NS 1 + + /* #define apic_debug(fmt,arg...) printk(KERN_WARNING fmt,##arg) */ +-#define apic_debug(fmt, arg...) ++#define apic_debug(fmt, arg...) do {} while (0) + + #define APIC_LVT_NUM 6 + /* 14 is the version for Xeon and Pentium 8.4.8*/ +diff -urNp linux-2.6.38.3/arch/x86/kvm/svm.c linux-2.6.38.3/arch/x86/kvm/svm.c +--- linux-2.6.38.3/arch/x86/kvm/svm.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kvm/svm.c 2011-04-17 15:57:32.000000000 -0400 +@@ -3273,7 +3273,11 @@ static void reload_tss(struct kvm_vcpu * + int cpu = raw_smp_processor_id(); + + struct svm_cpu_data *sd = per_cpu(svm_data, cpu); ++ ++ pax_open_kernel(); + sd->tss_desc->type = 9; /* available 32/64-bit TSS */ ++ pax_close_kernel(); ++ + load_TR_desc(); + } + +@@ -3850,7 +3854,7 @@ static void svm_fpu_deactivate(struct kv + update_cr0_intercept(svm); + } + +-static struct kvm_x86_ops svm_x86_ops = { ++static const struct kvm_x86_ops svm_x86_ops = { + .cpu_has_kvm_support = has_svm, + .disabled_by_bios = is_disabled, + .hardware_setup = svm_hardware_setup, +diff -urNp linux-2.6.38.3/arch/x86/kvm/vmx.c linux-2.6.38.3/arch/x86/kvm/vmx.c +--- linux-2.6.38.3/arch/x86/kvm/vmx.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kvm/vmx.c 2011-04-17 15:57:32.000000000 -0400 +@@ -725,7 +725,11 @@ static void reload_tss(void) + struct desc_struct *descs; + + descs = (void *)gdt->address; ++ ++ pax_open_kernel(); + descs[GDT_ENTRY_TSS].type = 9; /* available TSS */ ++ pax_close_kernel(); ++ + load_TR_desc(); + } + +@@ -1642,8 +1646,11 @@ static __init int hardware_setup(void) + if (!cpu_has_vmx_flexpriority()) + flexpriority_enabled = 0; + +- if (!cpu_has_vmx_tpr_shadow()) +- kvm_x86_ops->update_cr8_intercept = NULL; ++ if (!cpu_has_vmx_tpr_shadow()) { ++ pax_open_kernel(); ++ *(void **)&kvm_x86_ops->update_cr8_intercept = NULL; ++ pax_close_kernel(); ++ } + + if (enable_ept && !cpu_has_vmx_ept_2m_page()) + kvm_disable_largepages(); +@@ -2640,7 +2647,7 @@ static int vmx_vcpu_setup(struct vcpu_vm + vmcs_writel(HOST_IDTR_BASE, dt.address); /* 22.2.4 */ + + asm("mov $.Lkvm_vmx_return, %0" : "=r"(kvm_vmx_return)); +- vmcs_writel(HOST_RIP, kvm_vmx_return); /* 22.2.5 */ ++ vmcs_writel(HOST_RIP, ktla_ktva(kvm_vmx_return)); /* 22.2.5 */ + vmcs_write32(VM_EXIT_MSR_STORE_COUNT, 0); + vmcs_write32(VM_EXIT_MSR_LOAD_COUNT, 0); + vmcs_write64(VM_EXIT_MSR_LOAD_ADDR, __pa(vmx->msr_autoload.host)); +@@ -4031,6 +4038,12 @@ static void vmx_vcpu_run(struct kvm_vcpu + "jmp .Lkvm_vmx_return \n\t" + ".Llaunched: " __ex(ASM_VMX_VMRESUME) "\n\t" + ".Lkvm_vmx_return: " ++ ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) ++ "ljmp %[cs],$.Lkvm_vmx_return2\n\t" ++ ".Lkvm_vmx_return2: " ++#endif ++ + /* Save guest registers, load host registers, keep flags */ + "xchg %0, (%%"R"sp) \n\t" + "mov %%"R"ax, %c[rax](%0) \n\t" +@@ -4077,6 +4090,11 @@ static void vmx_vcpu_run(struct kvm_vcpu + [r15]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R15])), + #endif + [cr2]"i"(offsetof(struct vcpu_vmx, vcpu.arch.cr2)) ++ ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) ++ ,[cs]"i"(__KERNEL_CS) ++#endif ++ + : "cc", "memory" + , R"ax", R"bx", R"di", R"si" + #ifdef CONFIG_X86_64 +@@ -4091,7 +4109,7 @@ static void vmx_vcpu_run(struct kvm_vcpu + + vmx->idt_vectoring_info = vmcs_read32(IDT_VECTORING_INFO_FIELD); + +- asm("mov %0, %%ds; mov %0, %%es" : : "r"(__USER_DS)); ++ asm("mov %0, %%ds; mov %0, %%es; mov %0, %%ss" : : "r"(__KERNEL_DS)); + vmx->launched = 1; + + vmx->exit_reason = vmcs_read32(VM_EXIT_REASON); +@@ -4326,7 +4344,7 @@ static void vmx_set_supported_cpuid(u32 + { + } + +-static struct kvm_x86_ops vmx_x86_ops = { ++static const struct kvm_x86_ops vmx_x86_ops = { + .cpu_has_kvm_support = cpu_has_kvm_support, + .disabled_by_bios = vmx_disabled_by_bios, + .hardware_setup = hardware_setup, +diff -urNp linux-2.6.38.3/arch/x86/kvm/x86.c linux-2.6.38.3/arch/x86/kvm/x86.c +--- linux-2.6.38.3/arch/x86/kvm/x86.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/kvm/x86.c 2011-04-17 15:57:32.000000000 -0400 +@@ -93,7 +93,7 @@ static void update_cr8_intercept(struct + static int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid, + struct kvm_cpuid_entry2 __user *entries); + +-struct kvm_x86_ops *kvm_x86_ops; ++const struct kvm_x86_ops *kvm_x86_ops; + EXPORT_SYMBOL_GPL(kvm_x86_ops); + + int ignore_msrs = 0; +@@ -119,38 +119,38 @@ static struct kvm_shared_msrs_global __r + static DEFINE_PER_CPU(struct kvm_shared_msrs, shared_msrs); + + struct kvm_stats_debugfs_item debugfs_entries[] = { +- { "pf_fixed", VCPU_STAT(pf_fixed) }, +- { "pf_guest", VCPU_STAT(pf_guest) }, +- { "tlb_flush", VCPU_STAT(tlb_flush) }, +- { "invlpg", VCPU_STAT(invlpg) }, +- { "exits", VCPU_STAT(exits) }, +- { "io_exits", VCPU_STAT(io_exits) }, +- { "mmio_exits", VCPU_STAT(mmio_exits) }, +- { "signal_exits", VCPU_STAT(signal_exits) }, +- { "irq_window", VCPU_STAT(irq_window_exits) }, +- { "nmi_window", VCPU_STAT(nmi_window_exits) }, +- { "halt_exits", VCPU_STAT(halt_exits) }, +- { "halt_wakeup", VCPU_STAT(halt_wakeup) }, +- { "hypercalls", VCPU_STAT(hypercalls) }, +- { "request_irq", VCPU_STAT(request_irq_exits) }, +- { "irq_exits", VCPU_STAT(irq_exits) }, +- { "host_state_reload", VCPU_STAT(host_state_reload) }, +- { "efer_reload", VCPU_STAT(efer_reload) }, +- { "fpu_reload", VCPU_STAT(fpu_reload) }, +- { "insn_emulation", VCPU_STAT(insn_emulation) }, +- { "insn_emulation_fail", VCPU_STAT(insn_emulation_fail) }, +- { "irq_injections", VCPU_STAT(irq_injections) }, +- { "nmi_injections", VCPU_STAT(nmi_injections) }, +- { "mmu_shadow_zapped", VM_STAT(mmu_shadow_zapped) }, +- { "mmu_pte_write", VM_STAT(mmu_pte_write) }, +- { "mmu_pte_updated", VM_STAT(mmu_pte_updated) }, +- { "mmu_pde_zapped", VM_STAT(mmu_pde_zapped) }, +- { "mmu_flooded", VM_STAT(mmu_flooded) }, +- { "mmu_recycled", VM_STAT(mmu_recycled) }, +- { "mmu_cache_miss", VM_STAT(mmu_cache_miss) }, +- { "mmu_unsync", VM_STAT(mmu_unsync) }, +- { "remote_tlb_flush", VM_STAT(remote_tlb_flush) }, +- { "largepages", VM_STAT(lpages) }, ++ { "pf_fixed", VCPU_STAT(pf_fixed), NULL }, ++ { "pf_guest", VCPU_STAT(pf_guest), NULL }, ++ { "tlb_flush", VCPU_STAT(tlb_flush), NULL }, ++ { "invlpg", VCPU_STAT(invlpg), NULL }, ++ { "exits", VCPU_STAT(exits), NULL }, ++ { "io_exits", VCPU_STAT(io_exits), NULL }, ++ { "mmio_exits", VCPU_STAT(mmio_exits), NULL }, ++ { "signal_exits", VCPU_STAT(signal_exits), NULL }, ++ { "irq_window", VCPU_STAT(irq_window_exits), NULL }, ++ { "nmi_window", VCPU_STAT(nmi_window_exits), NULL }, ++ { "halt_exits", VCPU_STAT(halt_exits), NULL }, ++ { "halt_wakeup", VCPU_STAT(halt_wakeup), NULL }, ++ { "hypercalls", VCPU_STAT(hypercalls), NULL }, ++ { "request_irq", VCPU_STAT(request_irq_exits), NULL }, ++ { "irq_exits", VCPU_STAT(irq_exits), NULL }, ++ { "host_state_reload", VCPU_STAT(host_state_reload), NULL }, ++ { "efer_reload", VCPU_STAT(efer_reload), NULL }, ++ { "fpu_reload", VCPU_STAT(fpu_reload), NULL }, ++ { "insn_emulation", VCPU_STAT(insn_emulation), NULL }, ++ { "insn_emulation_fail", VCPU_STAT(insn_emulation_fail), NULL }, ++ { "irq_injections", VCPU_STAT(irq_injections), NULL }, ++ { "nmi_injections", VCPU_STAT(nmi_injections), NULL }, ++ { "mmu_shadow_zapped", VM_STAT(mmu_shadow_zapped), NULL }, ++ { "mmu_pte_write", VM_STAT(mmu_pte_write), NULL }, ++ { "mmu_pte_updated", VM_STAT(mmu_pte_updated), NULL }, ++ { "mmu_pde_zapped", VM_STAT(mmu_pde_zapped), NULL }, ++ { "mmu_flooded", VM_STAT(mmu_flooded), NULL }, ++ { "mmu_recycled", VM_STAT(mmu_recycled), NULL }, ++ { "mmu_cache_miss", VM_STAT(mmu_cache_miss), NULL }, ++ { "mmu_unsync", VM_STAT(mmu_unsync), NULL }, ++ { "remote_tlb_flush", VM_STAT(remote_tlb_flush), NULL }, ++ { "largepages", VM_STAT(lpages), NULL }, + { NULL } + }; + +@@ -2023,6 +2023,8 @@ long kvm_arch_dev_ioctl(struct file *fil + if (n < msr_list.nmsrs) + goto out; + r = -EFAULT; ++ if (num_msrs_to_save > ARRAY_SIZE(msrs_to_save)) ++ goto out; + if (copy_to_user(user_msr_list->indices, &msrs_to_save, + num_msrs_to_save * sizeof(u32))) + goto out; +@@ -2499,7 +2501,7 @@ static int kvm_vcpu_ioctl_set_lapic(stru + static int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu, + struct kvm_interrupt *irq) + { +- if (irq->irq < 0 || irq->irq >= 256) ++ if (irq->irq >= 256) + return -EINVAL; + if (irqchip_in_kernel(vcpu->kvm)) + return -ENXIO; +@@ -4687,10 +4689,10 @@ void kvm_after_handle_nmi(struct kvm_vcp + } + EXPORT_SYMBOL_GPL(kvm_after_handle_nmi); + +-int kvm_arch_init(void *opaque) ++int kvm_arch_init(const void *opaque) + { + int r; +- struct kvm_x86_ops *ops = (struct kvm_x86_ops *)opaque; ++ const struct kvm_x86_ops *ops = (const struct kvm_x86_ops *)opaque; + + if (kvm_x86_ops) { + printk(KERN_ERR "kvm: already loaded the other module\n"); +diff -urNp linux-2.6.38.3/arch/x86/lib/atomic64_cx8_32.S linux-2.6.38.3/arch/x86/lib/atomic64_cx8_32.S +--- linux-2.6.38.3/arch/x86/lib/atomic64_cx8_32.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/lib/atomic64_cx8_32.S 2011-04-17 15:57:32.000000000 -0400 +@@ -86,13 +86,23 @@ ENTRY(atomic64_\func\()_return_cx8) + movl %edx, %ecx + \ins\()l %esi, %ebx + \insc\()l %edi, %ecx ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ into ++2: ++ _ASM_EXTABLE(2b, 3f) ++#endif ++ + LOCK_PREFIX + cmpxchg8b (%ebp) + jne 1b +- +-10: + movl %ebx, %eax + movl %ecx, %edx ++ ++#ifdef CONFIG_PAX_REFCOUNT ++3: ++#endif ++ + RESTORE edi + RESTORE esi + RESTORE ebx +@@ -116,13 +126,24 @@ ENTRY(atomic64_\func\()_return_cx8) + movl %edx, %ecx + \ins\()l $1, %ebx + \insc\()l $0, %ecx ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ into ++2: ++ _ASM_EXTABLE(2b, 3f) ++#endif ++ + LOCK_PREFIX + cmpxchg8b (%esi) + jne 1b + +-10: + movl %ebx, %eax + movl %ecx, %edx ++ ++#ifdef CONFIG_PAX_REFCOUNT ++3: ++#endif ++ + RESTORE ebx + ret + CFI_ENDPROC +@@ -176,6 +197,13 @@ ENTRY(atomic64_add_unless_cx8) + movl %edx, %ecx + addl %esi, %ebx + adcl %edi, %ecx ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ into ++1234: ++ _ASM_EXTABLE(1234b, 1234b) ++#endif ++ + LOCK_PREFIX + cmpxchg8b (%ebp) + jne 1b +@@ -208,6 +236,13 @@ ENTRY(atomic64_inc_not_zero_cx8) + movl %edx, %ecx + addl $1, %ebx + adcl $0, %ecx ++ ++#ifdef CONFIG_PAX_REFCOUNT ++ into ++1234: ++ _ASM_EXTABLE(1234b, 1234b) ++#endif ++ + LOCK_PREFIX + cmpxchg8b (%esi) + jne 1b +diff -urNp linux-2.6.38.3/arch/x86/lib/checksum_32.S linux-2.6.38.3/arch/x86/lib/checksum_32.S +--- linux-2.6.38.3/arch/x86/lib/checksum_32.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/lib/checksum_32.S 2011-04-17 15:57:32.000000000 -0400 +@@ -28,7 +28,8 @@ + #include <linux/linkage.h> + #include <asm/dwarf2.h> + #include <asm/errno.h> +- ++#include <asm/segment.h> ++ + /* + * computes a partial checksum, e.g. for TCP/UDP fragments + */ +@@ -304,9 +305,28 @@ unsigned int csum_partial_copy_generic ( + + #define ARGBASE 16 + #define FP 12 +- +-ENTRY(csum_partial_copy_generic) ++ ++ENTRY(csum_partial_copy_generic_to_user) + CFI_STARTPROC ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ pushl %gs ++ CFI_ADJUST_CFA_OFFSET 4 ++ popl %es ++ CFI_ADJUST_CFA_OFFSET -4 ++ jmp csum_partial_copy_generic ++#endif ++ ++ENTRY(csum_partial_copy_generic_from_user) ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ pushl %gs ++ CFI_ADJUST_CFA_OFFSET 4 ++ popl %ds ++ CFI_ADJUST_CFA_OFFSET -4 ++#endif ++ ++ENTRY(csum_partial_copy_generic) + subl $4,%esp + CFI_ADJUST_CFA_OFFSET 4 + pushl %edi +@@ -331,7 +351,7 @@ ENTRY(csum_partial_copy_generic) + jmp 4f + SRC(1: movw (%esi), %bx ) + addl $2, %esi +-DST( movw %bx, (%edi) ) ++DST( movw %bx, %es:(%edi) ) + addl $2, %edi + addw %bx, %ax + adcl $0, %eax +@@ -343,30 +363,30 @@ DST( movw %bx, (%edi) ) + SRC(1: movl (%esi), %ebx ) + SRC( movl 4(%esi), %edx ) + adcl %ebx, %eax +-DST( movl %ebx, (%edi) ) ++DST( movl %ebx, %es:(%edi) ) + adcl %edx, %eax +-DST( movl %edx, 4(%edi) ) ++DST( movl %edx, %es:4(%edi) ) + + SRC( movl 8(%esi), %ebx ) + SRC( movl 12(%esi), %edx ) + adcl %ebx, %eax +-DST( movl %ebx, 8(%edi) ) ++DST( movl %ebx, %es:8(%edi) ) + adcl %edx, %eax +-DST( movl %edx, 12(%edi) ) ++DST( movl %edx, %es:12(%edi) ) + + SRC( movl 16(%esi), %ebx ) + SRC( movl 20(%esi), %edx ) + adcl %ebx, %eax +-DST( movl %ebx, 16(%edi) ) ++DST( movl %ebx, %es:16(%edi) ) + adcl %edx, %eax +-DST( movl %edx, 20(%edi) ) ++DST( movl %edx, %es:20(%edi) ) + + SRC( movl 24(%esi), %ebx ) + SRC( movl 28(%esi), %edx ) + adcl %ebx, %eax +-DST( movl %ebx, 24(%edi) ) ++DST( movl %ebx, %es:24(%edi) ) + adcl %edx, %eax +-DST( movl %edx, 28(%edi) ) ++DST( movl %edx, %es:28(%edi) ) + + lea 32(%esi), %esi + lea 32(%edi), %edi +@@ -380,7 +400,7 @@ DST( movl %edx, 28(%edi) ) + shrl $2, %edx # This clears CF + SRC(3: movl (%esi), %ebx ) + adcl %ebx, %eax +-DST( movl %ebx, (%edi) ) ++DST( movl %ebx, %es:(%edi) ) + lea 4(%esi), %esi + lea 4(%edi), %edi + dec %edx +@@ -392,12 +412,12 @@ DST( movl %ebx, (%edi) ) + jb 5f + SRC( movw (%esi), %cx ) + leal 2(%esi), %esi +-DST( movw %cx, (%edi) ) ++DST( movw %cx, %es:(%edi) ) + leal 2(%edi), %edi + je 6f + shll $16,%ecx + SRC(5: movb (%esi), %cl ) +-DST( movb %cl, (%edi) ) ++DST( movb %cl, %es:(%edi) ) + 6: addl %ecx, %eax + adcl $0, %eax + 7: +@@ -408,7 +428,7 @@ DST( movb %cl, (%edi) ) + + 6001: + movl ARGBASE+20(%esp), %ebx # src_err_ptr +- movl $-EFAULT, (%ebx) ++ movl $-EFAULT, %ss:(%ebx) + + # zero the complete destination - computing the rest + # is too much work +@@ -421,11 +441,19 @@ DST( movb %cl, (%edi) ) + + 6002: + movl ARGBASE+24(%esp), %ebx # dst_err_ptr +- movl $-EFAULT,(%ebx) ++ movl $-EFAULT,%ss:(%ebx) + jmp 5000b + + .previous + ++ pushl %ss ++ CFI_ADJUST_CFA_OFFSET 4 ++ popl %ds ++ CFI_ADJUST_CFA_OFFSET -4 ++ pushl %ss ++ CFI_ADJUST_CFA_OFFSET 4 ++ popl %es ++ CFI_ADJUST_CFA_OFFSET -4 + popl %ebx + CFI_ADJUST_CFA_OFFSET -4 + CFI_RESTORE ebx +@@ -439,26 +467,47 @@ DST( movb %cl, (%edi) ) + CFI_ADJUST_CFA_OFFSET -4 + ret + CFI_ENDPROC +-ENDPROC(csum_partial_copy_generic) ++ENDPROC(csum_partial_copy_generic_to_user) + + #else + + /* Version for PentiumII/PPro */ + + #define ROUND1(x) \ ++ nop; nop; nop; \ + SRC(movl x(%esi), %ebx ) ; \ + addl %ebx, %eax ; \ +- DST(movl %ebx, x(%edi) ) ; ++ DST(movl %ebx, %es:x(%edi)) ; + + #define ROUND(x) \ ++ nop; nop; nop; \ + SRC(movl x(%esi), %ebx ) ; \ + adcl %ebx, %eax ; \ +- DST(movl %ebx, x(%edi) ) ; ++ DST(movl %ebx, %es:x(%edi)) ; + + #define ARGBASE 12 +- +-ENTRY(csum_partial_copy_generic) ++ ++ENTRY(csum_partial_copy_generic_to_user) + CFI_STARTPROC ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ pushl %gs ++ CFI_ADJUST_CFA_OFFSET 4 ++ popl %es ++ CFI_ADJUST_CFA_OFFSET -4 ++ jmp csum_partial_copy_generic ++#endif ++ ++ENTRY(csum_partial_copy_generic_from_user) ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ pushl %gs ++ CFI_ADJUST_CFA_OFFSET 4 ++ popl %ds ++ CFI_ADJUST_CFA_OFFSET -4 ++#endif ++ ++ENTRY(csum_partial_copy_generic) + pushl %ebx + CFI_ADJUST_CFA_OFFSET 4 + CFI_REL_OFFSET ebx, 0 +@@ -482,7 +531,7 @@ ENTRY(csum_partial_copy_generic) + subl %ebx, %edi + lea -1(%esi),%edx + andl $-32,%edx +- lea 3f(%ebx,%ebx), %ebx ++ lea 3f(%ebx,%ebx,2), %ebx + testl %esi, %esi + jmp *%ebx + 1: addl $64,%esi +@@ -503,19 +552,19 @@ ENTRY(csum_partial_copy_generic) + jb 5f + SRC( movw (%esi), %dx ) + leal 2(%esi), %esi +-DST( movw %dx, (%edi) ) ++DST( movw %dx, %es:(%edi) ) + leal 2(%edi), %edi + je 6f + shll $16,%edx + 5: + SRC( movb (%esi), %dl ) +-DST( movb %dl, (%edi) ) ++DST( movb %dl, %es:(%edi) ) + 6: addl %edx, %eax + adcl $0, %eax + 7: + .section .fixup, "ax" + 6001: movl ARGBASE+20(%esp), %ebx # src_err_ptr +- movl $-EFAULT, (%ebx) ++ movl $-EFAULT, %ss:(%ebx) + # zero the complete destination (computing the rest is too much work) + movl ARGBASE+8(%esp),%edi # dst + movl ARGBASE+12(%esp),%ecx # len +@@ -523,10 +572,21 @@ DST( movb %dl, (%edi) ) + rep; stosb + jmp 7b + 6002: movl ARGBASE+24(%esp), %ebx # dst_err_ptr +- movl $-EFAULT, (%ebx) ++ movl $-EFAULT, %ss:(%ebx) + jmp 7b + .previous + ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ pushl %ss ++ CFI_ADJUST_CFA_OFFSET 4 ++ popl %ds ++ CFI_ADJUST_CFA_OFFSET -4 ++ pushl %ss ++ CFI_ADJUST_CFA_OFFSET 4 ++ popl %es ++ CFI_ADJUST_CFA_OFFSET -4 ++#endif ++ + popl %esi + CFI_ADJUST_CFA_OFFSET -4 + CFI_RESTORE esi +@@ -538,7 +598,7 @@ DST( movb %dl, (%edi) ) + CFI_RESTORE ebx + ret + CFI_ENDPROC +-ENDPROC(csum_partial_copy_generic) ++ENDPROC(csum_partial_copy_generic_to_user) + + #undef ROUND + #undef ROUND1 +diff -urNp linux-2.6.38.3/arch/x86/lib/clear_page_64.S linux-2.6.38.3/arch/x86/lib/clear_page_64.S +--- linux-2.6.38.3/arch/x86/lib/clear_page_64.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/lib/clear_page_64.S 2011-04-17 15:57:32.000000000 -0400 +@@ -43,7 +43,7 @@ ENDPROC(clear_page) + + #include <asm/cpufeature.h> + +- .section .altinstr_replacement,"ax" ++ .section .altinstr_replacement,"a" + 1: .byte 0xeb /* jmp <disp8> */ + .byte (clear_page_c - clear_page) - (2f - 1b) /* offset */ + 2: +diff -urNp linux-2.6.38.3/arch/x86/lib/copy_page_64.S linux-2.6.38.3/arch/x86/lib/copy_page_64.S +--- linux-2.6.38.3/arch/x86/lib/copy_page_64.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/lib/copy_page_64.S 2011-04-17 15:57:32.000000000 -0400 +@@ -104,7 +104,7 @@ ENDPROC(copy_page) + + #include <asm/cpufeature.h> + +- .section .altinstr_replacement,"ax" ++ .section .altinstr_replacement,"a" + 1: .byte 0xeb /* jmp <disp8> */ + .byte (copy_page_c - copy_page) - (2f - 1b) /* offset */ + 2: +diff -urNp linux-2.6.38.3/arch/x86/lib/copy_user_64.S linux-2.6.38.3/arch/x86/lib/copy_user_64.S +--- linux-2.6.38.3/arch/x86/lib/copy_user_64.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/lib/copy_user_64.S 2011-04-17 15:57:32.000000000 -0400 +@@ -15,13 +15,14 @@ + #include <asm/asm-offsets.h> + #include <asm/thread_info.h> + #include <asm/cpufeature.h> ++#include <asm/pgtable.h> + + .macro ALTERNATIVE_JUMP feature,orig,alt + 0: + .byte 0xe9 /* 32bit jump */ + .long \orig-1f /* by default jump to orig */ + 1: +- .section .altinstr_replacement,"ax" ++ .section .altinstr_replacement,"a" + 2: .byte 0xe9 /* near jump with 32bit immediate */ + .long \alt-1b /* offset */ /* or alternatively to alt */ + .previous +@@ -64,37 +65,13 @@ + #endif + .endm + +-/* Standard copy_to_user with segment limit checking */ +-ENTRY(_copy_to_user) +- CFI_STARTPROC +- GET_THREAD_INFO(%rax) +- movq %rdi,%rcx +- addq %rdx,%rcx +- jc bad_to_user +- cmpq TI_addr_limit(%rax),%rcx +- jae bad_to_user +- ALTERNATIVE_JUMP X86_FEATURE_REP_GOOD,copy_user_generic_unrolled,copy_user_generic_string +- CFI_ENDPROC +-ENDPROC(_copy_to_user) +- +-/* Standard copy_from_user with segment limit checking */ +-ENTRY(_copy_from_user) +- CFI_STARTPROC +- GET_THREAD_INFO(%rax) +- movq %rsi,%rcx +- addq %rdx,%rcx +- jc bad_from_user +- cmpq TI_addr_limit(%rax),%rcx +- jae bad_from_user +- ALTERNATIVE_JUMP X86_FEATURE_REP_GOOD,copy_user_generic_unrolled,copy_user_generic_string +- CFI_ENDPROC +-ENDPROC(_copy_from_user) +- + .section .fixup,"ax" + /* must zero dest */ + ENTRY(bad_from_user) + bad_from_user: + CFI_STARTPROC ++ testl %edx,%edx ++ js bad_to_user + movl %edx,%ecx + xorl %eax,%eax + rep +diff -urNp linux-2.6.38.3/arch/x86/lib/copy_user_nocache_64.S linux-2.6.38.3/arch/x86/lib/copy_user_nocache_64.S +--- linux-2.6.38.3/arch/x86/lib/copy_user_nocache_64.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/lib/copy_user_nocache_64.S 2011-04-17 15:57:32.000000000 -0400 +@@ -14,6 +14,7 @@ + #include <asm/current.h> + #include <asm/asm-offsets.h> + #include <asm/thread_info.h> ++#include <asm/pgtable.h> + + .macro ALIGN_DESTINATION + #ifdef FIX_ALIGNMENT +@@ -50,6 +51,15 @@ + */ + ENTRY(__copy_user_nocache) + CFI_STARTPROC ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ mov $PAX_USER_SHADOW_BASE,%rcx ++ cmp %rcx,%rsi ++ jae 1f ++ add %rcx,%rsi ++1: ++#endif ++ + cmpl $8,%edx + jb 20f /* less then 8 bytes, go to byte copy loop */ + ALIGN_DESTINATION +diff -urNp linux-2.6.38.3/arch/x86/lib/csum-wrappers_64.c linux-2.6.38.3/arch/x86/lib/csum-wrappers_64.c +--- linux-2.6.38.3/arch/x86/lib/csum-wrappers_64.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/lib/csum-wrappers_64.c 2011-04-17 15:57:32.000000000 -0400 +@@ -52,6 +52,8 @@ csum_partial_copy_from_user(const void _ + len -= 2; + } + } ++ if ((unsigned long)src < PAX_USER_SHADOW_BASE) ++ src += PAX_USER_SHADOW_BASE; + isum = csum_partial_copy_generic((__force const void *)src, + dst, len, isum, errp, NULL); + if (unlikely(*errp)) +@@ -105,6 +107,8 @@ csum_partial_copy_to_user(const void *sr + } + + *errp = 0; ++ if ((unsigned long)dst < PAX_USER_SHADOW_BASE) ++ dst += PAX_USER_SHADOW_BASE; + return csum_partial_copy_generic(src, (void __force *)dst, + len, isum, NULL, errp); + } +diff -urNp linux-2.6.38.3/arch/x86/lib/getuser.S linux-2.6.38.3/arch/x86/lib/getuser.S +--- linux-2.6.38.3/arch/x86/lib/getuser.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/lib/getuser.S 2011-04-17 15:57:32.000000000 -0400 +@@ -33,14 +33,35 @@ + #include <asm/asm-offsets.h> + #include <asm/thread_info.h> + #include <asm/asm.h> ++#include <asm/segment.h> ++#include <asm/pgtable.h> ++ ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_MEMORY_UDEREF) ++#define __copyuser_seg gs; ++#else ++#define __copyuser_seg ++#endif + + .text + ENTRY(__get_user_1) + CFI_STARTPROC ++ ++#if !defined(CONFIG_X86_32) || !defined(CONFIG_PAX_MEMORY_UDEREF) + GET_THREAD_INFO(%_ASM_DX) + cmp TI_addr_limit(%_ASM_DX),%_ASM_AX + jae bad_get_user +-1: movzb (%_ASM_AX),%edx ++ ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) ++ mov $PAX_USER_SHADOW_BASE,%_ASM_DX ++ cmp %_ASM_DX,%_ASM_AX ++ jae 1234f ++ add %_ASM_DX,%_ASM_AX ++1234: ++#endif ++ ++#endif ++ ++1: __copyuser_seg movzb (%_ASM_AX),%edx + xor %eax,%eax + ret + CFI_ENDPROC +@@ -49,11 +70,24 @@ ENDPROC(__get_user_1) + ENTRY(__get_user_2) + CFI_STARTPROC + add $1,%_ASM_AX ++ ++#if !defined(CONFIG_X86_32) || !defined(CONFIG_PAX_MEMORY_UDEREF) + jc bad_get_user + GET_THREAD_INFO(%_ASM_DX) + cmp TI_addr_limit(%_ASM_DX),%_ASM_AX + jae bad_get_user +-2: movzwl -1(%_ASM_AX),%edx ++ ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) ++ mov $PAX_USER_SHADOW_BASE,%_ASM_DX ++ cmp %_ASM_DX,%_ASM_AX ++ jae 1234f ++ add %_ASM_DX,%_ASM_AX ++1234: ++#endif ++ ++#endif ++ ++2: __copyuser_seg movzwl -1(%_ASM_AX),%edx + xor %eax,%eax + ret + CFI_ENDPROC +@@ -62,11 +96,24 @@ ENDPROC(__get_user_2) + ENTRY(__get_user_4) + CFI_STARTPROC + add $3,%_ASM_AX ++ ++#if !defined(CONFIG_X86_32) || !defined(CONFIG_PAX_MEMORY_UDEREF) + jc bad_get_user + GET_THREAD_INFO(%_ASM_DX) + cmp TI_addr_limit(%_ASM_DX),%_ASM_AX + jae bad_get_user +-3: mov -3(%_ASM_AX),%edx ++ ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) ++ mov $PAX_USER_SHADOW_BASE,%_ASM_DX ++ cmp %_ASM_DX,%_ASM_AX ++ jae 1234f ++ add %_ASM_DX,%_ASM_AX ++1234: ++#endif ++ ++#endif ++ ++3: __copyuser_seg mov -3(%_ASM_AX),%edx + xor %eax,%eax + ret + CFI_ENDPROC +@@ -80,6 +127,15 @@ ENTRY(__get_user_8) + GET_THREAD_INFO(%_ASM_DX) + cmp TI_addr_limit(%_ASM_DX),%_ASM_AX + jae bad_get_user ++ ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++ mov $PAX_USER_SHADOW_BASE,%_ASM_DX ++ cmp %_ASM_DX,%_ASM_AX ++ jae 1234f ++ add %_ASM_DX,%_ASM_AX ++1234: ++#endif ++ + 4: movq -7(%_ASM_AX),%_ASM_DX + xor %eax,%eax + ret +diff -urNp linux-2.6.38.3/arch/x86/lib/insn.c linux-2.6.38.3/arch/x86/lib/insn.c +--- linux-2.6.38.3/arch/x86/lib/insn.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/lib/insn.c 2011-04-17 15:57:32.000000000 -0400 +@@ -21,6 +21,11 @@ + #include <linux/string.h> + #include <asm/inat.h> + #include <asm/insn.h> ++#ifdef __KERNEL__ ++#include <asm/pgtable_types.h> ++#else ++#define ktla_ktva(addr) addr ++#endif + + #define get_next(t, insn) \ + ({t r; r = *(t*)insn->next_byte; insn->next_byte += sizeof(t); r; }) +@@ -40,8 +45,8 @@ + void insn_init(struct insn *insn, const void *kaddr, int x86_64) + { + memset(insn, 0, sizeof(*insn)); +- insn->kaddr = kaddr; +- insn->next_byte = kaddr; ++ insn->kaddr = ktla_ktva(kaddr); ++ insn->next_byte = ktla_ktva(kaddr); + insn->x86_64 = x86_64 ? 1 : 0; + insn->opnd_bytes = 4; + if (x86_64) +diff -urNp linux-2.6.38.3/arch/x86/lib/mmx_32.c linux-2.6.38.3/arch/x86/lib/mmx_32.c +--- linux-2.6.38.3/arch/x86/lib/mmx_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/lib/mmx_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -29,6 +29,7 @@ void *_mmx_memcpy(void *to, const void * + { + void *p; + int i; ++ unsigned long cr0; + + if (unlikely(in_interrupt())) + return __memcpy(to, from, len); +@@ -39,44 +40,72 @@ void *_mmx_memcpy(void *to, const void * + kernel_fpu_begin(); + + __asm__ __volatile__ ( +- "1: prefetch (%0)\n" /* This set is 28 bytes */ +- " prefetch 64(%0)\n" +- " prefetch 128(%0)\n" +- " prefetch 192(%0)\n" +- " prefetch 256(%0)\n" ++ "1: prefetch (%1)\n" /* This set is 28 bytes */ ++ " prefetch 64(%1)\n" ++ " prefetch 128(%1)\n" ++ " prefetch 192(%1)\n" ++ " prefetch 256(%1)\n" + "2: \n" + ".section .fixup, \"ax\"\n" +- "3: movw $0x1AEB, 1b\n" /* jmp on 26 bytes */ ++ "3: \n" ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %%cr0, %0\n" ++ " movl %0, %%eax\n" ++ " andl $0xFFFEFFFF, %%eax\n" ++ " movl %%eax, %%cr0\n" ++#endif ++ ++ " movw $0x1AEB, 1b\n" /* jmp on 26 bytes */ ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %0, %%cr0\n" ++#endif ++ + " jmp 2b\n" + ".previous\n" + _ASM_EXTABLE(1b, 3b) +- : : "r" (from)); ++ : "=&r" (cr0) : "r" (from) : "ax"); + + for ( ; i > 5; i--) { + __asm__ __volatile__ ( +- "1: prefetch 320(%0)\n" +- "2: movq (%0), %%mm0\n" +- " movq 8(%0), %%mm1\n" +- " movq 16(%0), %%mm2\n" +- " movq 24(%0), %%mm3\n" +- " movq %%mm0, (%1)\n" +- " movq %%mm1, 8(%1)\n" +- " movq %%mm2, 16(%1)\n" +- " movq %%mm3, 24(%1)\n" +- " movq 32(%0), %%mm0\n" +- " movq 40(%0), %%mm1\n" +- " movq 48(%0), %%mm2\n" +- " movq 56(%0), %%mm3\n" +- " movq %%mm0, 32(%1)\n" +- " movq %%mm1, 40(%1)\n" +- " movq %%mm2, 48(%1)\n" +- " movq %%mm3, 56(%1)\n" ++ "1: prefetch 320(%1)\n" ++ "2: movq (%1), %%mm0\n" ++ " movq 8(%1), %%mm1\n" ++ " movq 16(%1), %%mm2\n" ++ " movq 24(%1), %%mm3\n" ++ " movq %%mm0, (%2)\n" ++ " movq %%mm1, 8(%2)\n" ++ " movq %%mm2, 16(%2)\n" ++ " movq %%mm3, 24(%2)\n" ++ " movq 32(%1), %%mm0\n" ++ " movq 40(%1), %%mm1\n" ++ " movq 48(%1), %%mm2\n" ++ " movq 56(%1), %%mm3\n" ++ " movq %%mm0, 32(%2)\n" ++ " movq %%mm1, 40(%2)\n" ++ " movq %%mm2, 48(%2)\n" ++ " movq %%mm3, 56(%2)\n" + ".section .fixup, \"ax\"\n" +- "3: movw $0x05EB, 1b\n" /* jmp on 5 bytes */ ++ "3:\n" ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %%cr0, %0\n" ++ " movl %0, %%eax\n" ++ " andl $0xFFFEFFFF, %%eax\n" ++ " movl %%eax, %%cr0\n" ++#endif ++ ++ " movw $0x05EB, 1b\n" /* jmp on 5 bytes */ ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %0, %%cr0\n" ++#endif ++ + " jmp 2b\n" + ".previous\n" + _ASM_EXTABLE(1b, 3b) +- : : "r" (from), "r" (to) : "memory"); ++ : "=&r" (cr0) : "r" (from), "r" (to) : "memory", "ax"); + + from += 64; + to += 64; +@@ -158,6 +187,7 @@ static void fast_clear_page(void *page) + static void fast_copy_page(void *to, void *from) + { + int i; ++ unsigned long cr0; + + kernel_fpu_begin(); + +@@ -166,42 +196,70 @@ static void fast_copy_page(void *to, voi + * but that is for later. -AV + */ + __asm__ __volatile__( +- "1: prefetch (%0)\n" +- " prefetch 64(%0)\n" +- " prefetch 128(%0)\n" +- " prefetch 192(%0)\n" +- " prefetch 256(%0)\n" ++ "1: prefetch (%1)\n" ++ " prefetch 64(%1)\n" ++ " prefetch 128(%1)\n" ++ " prefetch 192(%1)\n" ++ " prefetch 256(%1)\n" + "2: \n" + ".section .fixup, \"ax\"\n" +- "3: movw $0x1AEB, 1b\n" /* jmp on 26 bytes */ ++ "3: \n" ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %%cr0, %0\n" ++ " movl %0, %%eax\n" ++ " andl $0xFFFEFFFF, %%eax\n" ++ " movl %%eax, %%cr0\n" ++#endif ++ ++ " movw $0x1AEB, 1b\n" /* jmp on 26 bytes */ ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %0, %%cr0\n" ++#endif ++ + " jmp 2b\n" + ".previous\n" +- _ASM_EXTABLE(1b, 3b) : : "r" (from)); ++ _ASM_EXTABLE(1b, 3b) : "=&r" (cr0) : "r" (from) : "ax"); + + for (i = 0; i < (4096-320)/64; i++) { + __asm__ __volatile__ ( +- "1: prefetch 320(%0)\n" +- "2: movq (%0), %%mm0\n" +- " movntq %%mm0, (%1)\n" +- " movq 8(%0), %%mm1\n" +- " movntq %%mm1, 8(%1)\n" +- " movq 16(%0), %%mm2\n" +- " movntq %%mm2, 16(%1)\n" +- " movq 24(%0), %%mm3\n" +- " movntq %%mm3, 24(%1)\n" +- " movq 32(%0), %%mm4\n" +- " movntq %%mm4, 32(%1)\n" +- " movq 40(%0), %%mm5\n" +- " movntq %%mm5, 40(%1)\n" +- " movq 48(%0), %%mm6\n" +- " movntq %%mm6, 48(%1)\n" +- " movq 56(%0), %%mm7\n" +- " movntq %%mm7, 56(%1)\n" ++ "1: prefetch 320(%1)\n" ++ "2: movq (%1), %%mm0\n" ++ " movntq %%mm0, (%2)\n" ++ " movq 8(%1), %%mm1\n" ++ " movntq %%mm1, 8(%2)\n" ++ " movq 16(%1), %%mm2\n" ++ " movntq %%mm2, 16(%2)\n" ++ " movq 24(%1), %%mm3\n" ++ " movntq %%mm3, 24(%2)\n" ++ " movq 32(%1), %%mm4\n" ++ " movntq %%mm4, 32(%2)\n" ++ " movq 40(%1), %%mm5\n" ++ " movntq %%mm5, 40(%2)\n" ++ " movq 48(%1), %%mm6\n" ++ " movntq %%mm6, 48(%2)\n" ++ " movq 56(%1), %%mm7\n" ++ " movntq %%mm7, 56(%2)\n" + ".section .fixup, \"ax\"\n" +- "3: movw $0x05EB, 1b\n" /* jmp on 5 bytes */ ++ "3:\n" ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %%cr0, %0\n" ++ " movl %0, %%eax\n" ++ " andl $0xFFFEFFFF, %%eax\n" ++ " movl %%eax, %%cr0\n" ++#endif ++ ++ " movw $0x05EB, 1b\n" /* jmp on 5 bytes */ ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %0, %%cr0\n" ++#endif ++ + " jmp 2b\n" + ".previous\n" +- _ASM_EXTABLE(1b, 3b) : : "r" (from), "r" (to) : "memory"); ++ _ASM_EXTABLE(1b, 3b) : "=&r" (cr0) : "r" (from), "r" (to) : "memory", "ax"); + + from += 64; + to += 64; +@@ -280,47 +338,76 @@ static void fast_clear_page(void *page) + static void fast_copy_page(void *to, void *from) + { + int i; ++ unsigned long cr0; + + kernel_fpu_begin(); + + __asm__ __volatile__ ( +- "1: prefetch (%0)\n" +- " prefetch 64(%0)\n" +- " prefetch 128(%0)\n" +- " prefetch 192(%0)\n" +- " prefetch 256(%0)\n" ++ "1: prefetch (%1)\n" ++ " prefetch 64(%1)\n" ++ " prefetch 128(%1)\n" ++ " prefetch 192(%1)\n" ++ " prefetch 256(%1)\n" + "2: \n" + ".section .fixup, \"ax\"\n" +- "3: movw $0x1AEB, 1b\n" /* jmp on 26 bytes */ ++ "3: \n" ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %%cr0, %0\n" ++ " movl %0, %%eax\n" ++ " andl $0xFFFEFFFF, %%eax\n" ++ " movl %%eax, %%cr0\n" ++#endif ++ ++ " movw $0x1AEB, 1b\n" /* jmp on 26 bytes */ ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %0, %%cr0\n" ++#endif ++ + " jmp 2b\n" + ".previous\n" +- _ASM_EXTABLE(1b, 3b) : : "r" (from)); ++ _ASM_EXTABLE(1b, 3b) : "=&r" (cr0) : "r" (from) : "ax"); + + for (i = 0; i < 4096/64; i++) { + __asm__ __volatile__ ( +- "1: prefetch 320(%0)\n" +- "2: movq (%0), %%mm0\n" +- " movq 8(%0), %%mm1\n" +- " movq 16(%0), %%mm2\n" +- " movq 24(%0), %%mm3\n" +- " movq %%mm0, (%1)\n" +- " movq %%mm1, 8(%1)\n" +- " movq %%mm2, 16(%1)\n" +- " movq %%mm3, 24(%1)\n" +- " movq 32(%0), %%mm0\n" +- " movq 40(%0), %%mm1\n" +- " movq 48(%0), %%mm2\n" +- " movq 56(%0), %%mm3\n" +- " movq %%mm0, 32(%1)\n" +- " movq %%mm1, 40(%1)\n" +- " movq %%mm2, 48(%1)\n" +- " movq %%mm3, 56(%1)\n" ++ "1: prefetch 320(%1)\n" ++ "2: movq (%1), %%mm0\n" ++ " movq 8(%1), %%mm1\n" ++ " movq 16(%1), %%mm2\n" ++ " movq 24(%1), %%mm3\n" ++ " movq %%mm0, (%2)\n" ++ " movq %%mm1, 8(%2)\n" ++ " movq %%mm2, 16(%2)\n" ++ " movq %%mm3, 24(%2)\n" ++ " movq 32(%1), %%mm0\n" ++ " movq 40(%1), %%mm1\n" ++ " movq 48(%1), %%mm2\n" ++ " movq 56(%1), %%mm3\n" ++ " movq %%mm0, 32(%2)\n" ++ " movq %%mm1, 40(%2)\n" ++ " movq %%mm2, 48(%2)\n" ++ " movq %%mm3, 56(%2)\n" + ".section .fixup, \"ax\"\n" +- "3: movw $0x05EB, 1b\n" /* jmp on 5 bytes */ ++ "3:\n" ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %%cr0, %0\n" ++ " movl %0, %%eax\n" ++ " andl $0xFFFEFFFF, %%eax\n" ++ " movl %%eax, %%cr0\n" ++#endif ++ ++ " movw $0x05EB, 1b\n" /* jmp on 5 bytes */ ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ " movl %0, %%cr0\n" ++#endif ++ + " jmp 2b\n" + ".previous\n" + _ASM_EXTABLE(1b, 3b) +- : : "r" (from), "r" (to) : "memory"); ++ : "=&r" (cr0) : "r" (from), "r" (to) : "memory", "ax"); + + from += 64; + to += 64; +diff -urNp linux-2.6.38.3/arch/x86/lib/putuser.S linux-2.6.38.3/arch/x86/lib/putuser.S +--- linux-2.6.38.3/arch/x86/lib/putuser.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/lib/putuser.S 2011-04-17 15:57:32.000000000 -0400 +@@ -15,7 +15,8 @@ + #include <asm/thread_info.h> + #include <asm/errno.h> + #include <asm/asm.h> +- ++#include <asm/segment.h> ++#include <asm/pgtable.h> + + /* + * __put_user_X +@@ -29,52 +30,119 @@ + * as they get called from within inline assembly. + */ + +-#define ENTER CFI_STARTPROC ; \ +- GET_THREAD_INFO(%_ASM_BX) ++#define ENTER CFI_STARTPROC + #define EXIT ret ; \ + CFI_ENDPROC + ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) ++#define _DEST %_ASM_CX,%_ASM_BX ++#else ++#define _DEST %_ASM_CX ++#endif ++ ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_MEMORY_UDEREF) ++#define __copyuser_seg gs; ++#else ++#define __copyuser_seg ++#endif ++ + .text + ENTRY(__put_user_1) + ENTER ++ ++#if !defined(CONFIG_X86_32) || !defined(CONFIG_PAX_MEMORY_UDEREF) ++ GET_THREAD_INFO(%_ASM_BX) + cmp TI_addr_limit(%_ASM_BX),%_ASM_CX + jae bad_put_user +-1: movb %al,(%_ASM_CX) ++ ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) ++ mov $PAX_USER_SHADOW_BASE,%_ASM_BX ++ cmp %_ASM_BX,%_ASM_CX ++ jb 1234f ++ xor %ebx,%ebx ++1234: ++#endif ++ ++#endif ++ ++1: __copyuser_seg movb %al,(_DEST) + xor %eax,%eax + EXIT + ENDPROC(__put_user_1) + + ENTRY(__put_user_2) + ENTER ++ ++#if !defined(CONFIG_X86_32) || !defined(CONFIG_PAX_MEMORY_UDEREF) ++ GET_THREAD_INFO(%_ASM_BX) + mov TI_addr_limit(%_ASM_BX),%_ASM_BX + sub $1,%_ASM_BX + cmp %_ASM_BX,%_ASM_CX + jae bad_put_user +-2: movw %ax,(%_ASM_CX) ++ ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) ++ mov $PAX_USER_SHADOW_BASE,%_ASM_BX ++ cmp %_ASM_BX,%_ASM_CX ++ jb 1234f ++ xor %ebx,%ebx ++1234: ++#endif ++ ++#endif ++ ++2: __copyuser_seg movw %ax,(_DEST) + xor %eax,%eax + EXIT + ENDPROC(__put_user_2) + + ENTRY(__put_user_4) + ENTER ++ ++#if !defined(CONFIG_X86_32) || !defined(CONFIG_PAX_MEMORY_UDEREF) ++ GET_THREAD_INFO(%_ASM_BX) + mov TI_addr_limit(%_ASM_BX),%_ASM_BX + sub $3,%_ASM_BX + cmp %_ASM_BX,%_ASM_CX + jae bad_put_user +-3: movl %eax,(%_ASM_CX) ++ ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) ++ mov $PAX_USER_SHADOW_BASE,%_ASM_BX ++ cmp %_ASM_BX,%_ASM_CX ++ jb 1234f ++ xor %ebx,%ebx ++1234: ++#endif ++ ++#endif ++ ++3: __copyuser_seg movl %eax,(_DEST) + xor %eax,%eax + EXIT + ENDPROC(__put_user_4) + + ENTRY(__put_user_8) + ENTER ++ ++#if !defined(CONFIG_X86_32) || !defined(CONFIG_PAX_MEMORY_UDEREF) ++ GET_THREAD_INFO(%_ASM_BX) + mov TI_addr_limit(%_ASM_BX),%_ASM_BX + sub $7,%_ASM_BX + cmp %_ASM_BX,%_ASM_CX + jae bad_put_user +-4: mov %_ASM_AX,(%_ASM_CX) ++ ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) ++ mov $PAX_USER_SHADOW_BASE,%_ASM_BX ++ cmp %_ASM_BX,%_ASM_CX ++ jb 1234f ++ xor %ebx,%ebx ++1234: ++#endif ++ ++#endif ++ ++4: __copyuser_seg mov %_ASM_AX,(_DEST) + #ifdef CONFIG_X86_32 +-5: movl %edx,4(%_ASM_CX) ++5: __copyuser_seg movl %edx,4(_DEST) + #endif + xor %eax,%eax + EXIT +diff -urNp linux-2.6.38.3/arch/x86/lib/usercopy_32.c linux-2.6.38.3/arch/x86/lib/usercopy_32.c +--- linux-2.6.38.3/arch/x86/lib/usercopy_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/lib/usercopy_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -43,7 +43,7 @@ do { \ + __asm__ __volatile__( \ + " testl %1,%1\n" \ + " jz 2f\n" \ +- "0: lodsb\n" \ ++ "0: "__copyuser_seg"lodsb\n" \ + " stosb\n" \ + " testb %%al,%%al\n" \ + " jz 1f\n" \ +@@ -128,10 +128,12 @@ do { \ + int __d0; \ + might_fault(); \ + __asm__ __volatile__( \ ++ __COPYUSER_SET_ES \ + "0: rep; stosl\n" \ + " movl %2,%0\n" \ + "1: rep; stosb\n" \ + "2:\n" \ ++ __COPYUSER_RESTORE_ES \ + ".section .fixup,\"ax\"\n" \ + "3: lea 0(%2,%0,4),%0\n" \ + " jmp 2b\n" \ +@@ -200,6 +202,7 @@ long strnlen_user(const char __user *s, + might_fault(); + + __asm__ __volatile__( ++ __COPYUSER_SET_ES + " testl %0, %0\n" + " jz 3f\n" + " andl %0,%%ecx\n" +@@ -208,6 +211,7 @@ long strnlen_user(const char __user *s, + " subl %%ecx,%0\n" + " addl %0,%%eax\n" + "1:\n" ++ __COPYUSER_RESTORE_ES + ".section .fixup,\"ax\"\n" + "2: xorl %%eax,%%eax\n" + " jmp 1b\n" +@@ -227,7 +231,7 @@ EXPORT_SYMBOL(strnlen_user); + + #ifdef CONFIG_X86_INTEL_USERCOPY + static unsigned long +-__copy_user_intel(void __user *to, const void *from, unsigned long size) ++__generic_copy_to_user_intel(void __user *to, const void *from, unsigned long size) + { + int d0, d1; + __asm__ __volatile__( +@@ -239,36 +243,36 @@ __copy_user_intel(void __user *to, const + " .align 2,0x90\n" + "3: movl 0(%4), %%eax\n" + "4: movl 4(%4), %%edx\n" +- "5: movl %%eax, 0(%3)\n" +- "6: movl %%edx, 4(%3)\n" ++ "5: "__copyuser_seg" movl %%eax, 0(%3)\n" ++ "6: "__copyuser_seg" movl %%edx, 4(%3)\n" + "7: movl 8(%4), %%eax\n" + "8: movl 12(%4),%%edx\n" +- "9: movl %%eax, 8(%3)\n" +- "10: movl %%edx, 12(%3)\n" ++ "9: "__copyuser_seg" movl %%eax, 8(%3)\n" ++ "10: "__copyuser_seg" movl %%edx, 12(%3)\n" + "11: movl 16(%4), %%eax\n" + "12: movl 20(%4), %%edx\n" +- "13: movl %%eax, 16(%3)\n" +- "14: movl %%edx, 20(%3)\n" ++ "13: "__copyuser_seg" movl %%eax, 16(%3)\n" ++ "14: "__copyuser_seg" movl %%edx, 20(%3)\n" + "15: movl 24(%4), %%eax\n" + "16: movl 28(%4), %%edx\n" +- "17: movl %%eax, 24(%3)\n" +- "18: movl %%edx, 28(%3)\n" ++ "17: "__copyuser_seg" movl %%eax, 24(%3)\n" ++ "18: "__copyuser_seg" movl %%edx, 28(%3)\n" + "19: movl 32(%4), %%eax\n" + "20: movl 36(%4), %%edx\n" +- "21: movl %%eax, 32(%3)\n" +- "22: movl %%edx, 36(%3)\n" ++ "21: "__copyuser_seg" movl %%eax, 32(%3)\n" ++ "22: "__copyuser_seg" movl %%edx, 36(%3)\n" + "23: movl 40(%4), %%eax\n" + "24: movl 44(%4), %%edx\n" +- "25: movl %%eax, 40(%3)\n" +- "26: movl %%edx, 44(%3)\n" ++ "25: "__copyuser_seg" movl %%eax, 40(%3)\n" ++ "26: "__copyuser_seg" movl %%edx, 44(%3)\n" + "27: movl 48(%4), %%eax\n" + "28: movl 52(%4), %%edx\n" +- "29: movl %%eax, 48(%3)\n" +- "30: movl %%edx, 52(%3)\n" ++ "29: "__copyuser_seg" movl %%eax, 48(%3)\n" ++ "30: "__copyuser_seg" movl %%edx, 52(%3)\n" + "31: movl 56(%4), %%eax\n" + "32: movl 60(%4), %%edx\n" +- "33: movl %%eax, 56(%3)\n" +- "34: movl %%edx, 60(%3)\n" ++ "33: "__copyuser_seg" movl %%eax, 56(%3)\n" ++ "34: "__copyuser_seg" movl %%edx, 60(%3)\n" + " addl $-64, %0\n" + " addl $64, %4\n" + " addl $64, %3\n" +@@ -278,10 +282,119 @@ __copy_user_intel(void __user *to, const + " shrl $2, %0\n" + " andl $3, %%eax\n" + " cld\n" ++ __COPYUSER_SET_ES + "99: rep; movsl\n" + "36: movl %%eax, %0\n" + "37: rep; movsb\n" + "100:\n" ++ __COPYUSER_RESTORE_ES ++ ".section .fixup,\"ax\"\n" ++ "101: lea 0(%%eax,%0,4),%0\n" ++ " jmp 100b\n" ++ ".previous\n" ++ ".section __ex_table,\"a\"\n" ++ " .align 4\n" ++ " .long 1b,100b\n" ++ " .long 2b,100b\n" ++ " .long 3b,100b\n" ++ " .long 4b,100b\n" ++ " .long 5b,100b\n" ++ " .long 6b,100b\n" ++ " .long 7b,100b\n" ++ " .long 8b,100b\n" ++ " .long 9b,100b\n" ++ " .long 10b,100b\n" ++ " .long 11b,100b\n" ++ " .long 12b,100b\n" ++ " .long 13b,100b\n" ++ " .long 14b,100b\n" ++ " .long 15b,100b\n" ++ " .long 16b,100b\n" ++ " .long 17b,100b\n" ++ " .long 18b,100b\n" ++ " .long 19b,100b\n" ++ " .long 20b,100b\n" ++ " .long 21b,100b\n" ++ " .long 22b,100b\n" ++ " .long 23b,100b\n" ++ " .long 24b,100b\n" ++ " .long 25b,100b\n" ++ " .long 26b,100b\n" ++ " .long 27b,100b\n" ++ " .long 28b,100b\n" ++ " .long 29b,100b\n" ++ " .long 30b,100b\n" ++ " .long 31b,100b\n" ++ " .long 32b,100b\n" ++ " .long 33b,100b\n" ++ " .long 34b,100b\n" ++ " .long 35b,100b\n" ++ " .long 36b,100b\n" ++ " .long 37b,100b\n" ++ " .long 99b,101b\n" ++ ".previous" ++ : "=&c"(size), "=&D" (d0), "=&S" (d1) ++ : "1"(to), "2"(from), "0"(size) ++ : "eax", "edx", "memory"); ++ return size; ++} ++ ++static unsigned long ++__generic_copy_from_user_intel(void *to, const void __user *from, unsigned long size) ++{ ++ int d0, d1; ++ __asm__ __volatile__( ++ " .align 2,0x90\n" ++ "1: "__copyuser_seg" movl 32(%4), %%eax\n" ++ " cmpl $67, %0\n" ++ " jbe 3f\n" ++ "2: "__copyuser_seg" movl 64(%4), %%eax\n" ++ " .align 2,0x90\n" ++ "3: "__copyuser_seg" movl 0(%4), %%eax\n" ++ "4: "__copyuser_seg" movl 4(%4), %%edx\n" ++ "5: movl %%eax, 0(%3)\n" ++ "6: movl %%edx, 4(%3)\n" ++ "7: "__copyuser_seg" movl 8(%4), %%eax\n" ++ "8: "__copyuser_seg" movl 12(%4),%%edx\n" ++ "9: movl %%eax, 8(%3)\n" ++ "10: movl %%edx, 12(%3)\n" ++ "11: "__copyuser_seg" movl 16(%4), %%eax\n" ++ "12: "__copyuser_seg" movl 20(%4), %%edx\n" ++ "13: movl %%eax, 16(%3)\n" ++ "14: movl %%edx, 20(%3)\n" ++ "15: "__copyuser_seg" movl 24(%4), %%eax\n" ++ "16: "__copyuser_seg" movl 28(%4), %%edx\n" ++ "17: movl %%eax, 24(%3)\n" ++ "18: movl %%edx, 28(%3)\n" ++ "19: "__copyuser_seg" movl 32(%4), %%eax\n" ++ "20: "__copyuser_seg" movl 36(%4), %%edx\n" ++ "21: movl %%eax, 32(%3)\n" ++ "22: movl %%edx, 36(%3)\n" ++ "23: "__copyuser_seg" movl 40(%4), %%eax\n" ++ "24: "__copyuser_seg" movl 44(%4), %%edx\n" ++ "25: movl %%eax, 40(%3)\n" ++ "26: movl %%edx, 44(%3)\n" ++ "27: "__copyuser_seg" movl 48(%4), %%eax\n" ++ "28: "__copyuser_seg" movl 52(%4), %%edx\n" ++ "29: movl %%eax, 48(%3)\n" ++ "30: movl %%edx, 52(%3)\n" ++ "31: "__copyuser_seg" movl 56(%4), %%eax\n" ++ "32: "__copyuser_seg" movl 60(%4), %%edx\n" ++ "33: movl %%eax, 56(%3)\n" ++ "34: movl %%edx, 60(%3)\n" ++ " addl $-64, %0\n" ++ " addl $64, %4\n" ++ " addl $64, %3\n" ++ " cmpl $63, %0\n" ++ " ja 1b\n" ++ "35: movl %0, %%eax\n" ++ " shrl $2, %0\n" ++ " andl $3, %%eax\n" ++ " cld\n" ++ "99: rep; "__copyuser_seg" movsl\n" ++ "36: movl %%eax, %0\n" ++ "37: rep; "__copyuser_seg" movsb\n" ++ "100:\n" + ".section .fixup,\"ax\"\n" + "101: lea 0(%%eax,%0,4),%0\n" + " jmp 100b\n" +@@ -339,41 +452,41 @@ __copy_user_zeroing_intel(void *to, cons + int d0, d1; + __asm__ __volatile__( + " .align 2,0x90\n" +- "0: movl 32(%4), %%eax\n" ++ "0: "__copyuser_seg" movl 32(%4), %%eax\n" + " cmpl $67, %0\n" + " jbe 2f\n" +- "1: movl 64(%4), %%eax\n" ++ "1: "__copyuser_seg" movl 64(%4), %%eax\n" + " .align 2,0x90\n" +- "2: movl 0(%4), %%eax\n" +- "21: movl 4(%4), %%edx\n" ++ "2: "__copyuser_seg" movl 0(%4), %%eax\n" ++ "21: "__copyuser_seg" movl 4(%4), %%edx\n" + " movl %%eax, 0(%3)\n" + " movl %%edx, 4(%3)\n" +- "3: movl 8(%4), %%eax\n" +- "31: movl 12(%4),%%edx\n" ++ "3: "__copyuser_seg" movl 8(%4), %%eax\n" ++ "31: "__copyuser_seg" movl 12(%4),%%edx\n" + " movl %%eax, 8(%3)\n" + " movl %%edx, 12(%3)\n" +- "4: movl 16(%4), %%eax\n" +- "41: movl 20(%4), %%edx\n" ++ "4: "__copyuser_seg" movl 16(%4), %%eax\n" ++ "41: "__copyuser_seg" movl 20(%4), %%edx\n" + " movl %%eax, 16(%3)\n" + " movl %%edx, 20(%3)\n" +- "10: movl 24(%4), %%eax\n" +- "51: movl 28(%4), %%edx\n" ++ "10: "__copyuser_seg" movl 24(%4), %%eax\n" ++ "51: "__copyuser_seg" movl 28(%4), %%edx\n" + " movl %%eax, 24(%3)\n" + " movl %%edx, 28(%3)\n" +- "11: movl 32(%4), %%eax\n" +- "61: movl 36(%4), %%edx\n" ++ "11: "__copyuser_seg" movl 32(%4), %%eax\n" ++ "61: "__copyuser_seg" movl 36(%4), %%edx\n" + " movl %%eax, 32(%3)\n" + " movl %%edx, 36(%3)\n" +- "12: movl 40(%4), %%eax\n" +- "71: movl 44(%4), %%edx\n" ++ "12: "__copyuser_seg" movl 40(%4), %%eax\n" ++ "71: "__copyuser_seg" movl 44(%4), %%edx\n" + " movl %%eax, 40(%3)\n" + " movl %%edx, 44(%3)\n" +- "13: movl 48(%4), %%eax\n" +- "81: movl 52(%4), %%edx\n" ++ "13: "__copyuser_seg" movl 48(%4), %%eax\n" ++ "81: "__copyuser_seg" movl 52(%4), %%edx\n" + " movl %%eax, 48(%3)\n" + " movl %%edx, 52(%3)\n" +- "14: movl 56(%4), %%eax\n" +- "91: movl 60(%4), %%edx\n" ++ "14: "__copyuser_seg" movl 56(%4), %%eax\n" ++ "91: "__copyuser_seg" movl 60(%4), %%edx\n" + " movl %%eax, 56(%3)\n" + " movl %%edx, 60(%3)\n" + " addl $-64, %0\n" +@@ -385,9 +498,9 @@ __copy_user_zeroing_intel(void *to, cons + " shrl $2, %0\n" + " andl $3, %%eax\n" + " cld\n" +- "6: rep; movsl\n" ++ "6: rep; "__copyuser_seg" movsl\n" + " movl %%eax,%0\n" +- "7: rep; movsb\n" ++ "7: rep; "__copyuser_seg" movsb\n" + "8:\n" + ".section .fixup,\"ax\"\n" + "9: lea 0(%%eax,%0,4),%0\n" +@@ -440,41 +553,41 @@ static unsigned long __copy_user_zeroing + + __asm__ __volatile__( + " .align 2,0x90\n" +- "0: movl 32(%4), %%eax\n" ++ "0: "__copyuser_seg" movl 32(%4), %%eax\n" + " cmpl $67, %0\n" + " jbe 2f\n" +- "1: movl 64(%4), %%eax\n" ++ "1: "__copyuser_seg" movl 64(%4), %%eax\n" + " .align 2,0x90\n" +- "2: movl 0(%4), %%eax\n" +- "21: movl 4(%4), %%edx\n" ++ "2: "__copyuser_seg" movl 0(%4), %%eax\n" ++ "21: "__copyuser_seg" movl 4(%4), %%edx\n" + " movnti %%eax, 0(%3)\n" + " movnti %%edx, 4(%3)\n" +- "3: movl 8(%4), %%eax\n" +- "31: movl 12(%4),%%edx\n" ++ "3: "__copyuser_seg" movl 8(%4), %%eax\n" ++ "31: "__copyuser_seg" movl 12(%4),%%edx\n" + " movnti %%eax, 8(%3)\n" + " movnti %%edx, 12(%3)\n" +- "4: movl 16(%4), %%eax\n" +- "41: movl 20(%4), %%edx\n" ++ "4: "__copyuser_seg" movl 16(%4), %%eax\n" ++ "41: "__copyuser_seg" movl 20(%4), %%edx\n" + " movnti %%eax, 16(%3)\n" + " movnti %%edx, 20(%3)\n" +- "10: movl 24(%4), %%eax\n" +- "51: movl 28(%4), %%edx\n" ++ "10: "__copyuser_seg" movl 24(%4), %%eax\n" ++ "51: "__copyuser_seg" movl 28(%4), %%edx\n" + " movnti %%eax, 24(%3)\n" + " movnti %%edx, 28(%3)\n" +- "11: movl 32(%4), %%eax\n" +- "61: movl 36(%4), %%edx\n" ++ "11: "__copyuser_seg" movl 32(%4), %%eax\n" ++ "61: "__copyuser_seg" movl 36(%4), %%edx\n" + " movnti %%eax, 32(%3)\n" + " movnti %%edx, 36(%3)\n" +- "12: movl 40(%4), %%eax\n" +- "71: movl 44(%4), %%edx\n" ++ "12: "__copyuser_seg" movl 40(%4), %%eax\n" ++ "71: "__copyuser_seg" movl 44(%4), %%edx\n" + " movnti %%eax, 40(%3)\n" + " movnti %%edx, 44(%3)\n" +- "13: movl 48(%4), %%eax\n" +- "81: movl 52(%4), %%edx\n" ++ "13: "__copyuser_seg" movl 48(%4), %%eax\n" ++ "81: "__copyuser_seg" movl 52(%4), %%edx\n" + " movnti %%eax, 48(%3)\n" + " movnti %%edx, 52(%3)\n" +- "14: movl 56(%4), %%eax\n" +- "91: movl 60(%4), %%edx\n" ++ "14: "__copyuser_seg" movl 56(%4), %%eax\n" ++ "91: "__copyuser_seg" movl 60(%4), %%edx\n" + " movnti %%eax, 56(%3)\n" + " movnti %%edx, 60(%3)\n" + " addl $-64, %0\n" +@@ -487,9 +600,9 @@ static unsigned long __copy_user_zeroing + " shrl $2, %0\n" + " andl $3, %%eax\n" + " cld\n" +- "6: rep; movsl\n" ++ "6: rep; "__copyuser_seg" movsl\n" + " movl %%eax,%0\n" +- "7: rep; movsb\n" ++ "7: rep; "__copyuser_seg" movsb\n" + "8:\n" + ".section .fixup,\"ax\"\n" + "9: lea 0(%%eax,%0,4),%0\n" +@@ -537,41 +650,41 @@ static unsigned long __copy_user_intel_n + + __asm__ __volatile__( + " .align 2,0x90\n" +- "0: movl 32(%4), %%eax\n" ++ "0: "__copyuser_seg" movl 32(%4), %%eax\n" + " cmpl $67, %0\n" + " jbe 2f\n" +- "1: movl 64(%4), %%eax\n" ++ "1: "__copyuser_seg" movl 64(%4), %%eax\n" + " .align 2,0x90\n" +- "2: movl 0(%4), %%eax\n" +- "21: movl 4(%4), %%edx\n" ++ "2: "__copyuser_seg" movl 0(%4), %%eax\n" ++ "21: "__copyuser_seg" movl 4(%4), %%edx\n" + " movnti %%eax, 0(%3)\n" + " movnti %%edx, 4(%3)\n" +- "3: movl 8(%4), %%eax\n" +- "31: movl 12(%4),%%edx\n" ++ "3: "__copyuser_seg" movl 8(%4), %%eax\n" ++ "31: "__copyuser_seg" movl 12(%4),%%edx\n" + " movnti %%eax, 8(%3)\n" + " movnti %%edx, 12(%3)\n" +- "4: movl 16(%4), %%eax\n" +- "41: movl 20(%4), %%edx\n" ++ "4: "__copyuser_seg" movl 16(%4), %%eax\n" ++ "41: "__copyuser_seg" movl 20(%4), %%edx\n" + " movnti %%eax, 16(%3)\n" + " movnti %%edx, 20(%3)\n" +- "10: movl 24(%4), %%eax\n" +- "51: movl 28(%4), %%edx\n" ++ "10: "__copyuser_seg" movl 24(%4), %%eax\n" ++ "51: "__copyuser_seg" movl 28(%4), %%edx\n" + " movnti %%eax, 24(%3)\n" + " movnti %%edx, 28(%3)\n" +- "11: movl 32(%4), %%eax\n" +- "61: movl 36(%4), %%edx\n" ++ "11: "__copyuser_seg" movl 32(%4), %%eax\n" ++ "61: "__copyuser_seg" movl 36(%4), %%edx\n" + " movnti %%eax, 32(%3)\n" + " movnti %%edx, 36(%3)\n" +- "12: movl 40(%4), %%eax\n" +- "71: movl 44(%4), %%edx\n" ++ "12: "__copyuser_seg" movl 40(%4), %%eax\n" ++ "71: "__copyuser_seg" movl 44(%4), %%edx\n" + " movnti %%eax, 40(%3)\n" + " movnti %%edx, 44(%3)\n" +- "13: movl 48(%4), %%eax\n" +- "81: movl 52(%4), %%edx\n" ++ "13: "__copyuser_seg" movl 48(%4), %%eax\n" ++ "81: "__copyuser_seg" movl 52(%4), %%edx\n" + " movnti %%eax, 48(%3)\n" + " movnti %%edx, 52(%3)\n" +- "14: movl 56(%4), %%eax\n" +- "91: movl 60(%4), %%edx\n" ++ "14: "__copyuser_seg" movl 56(%4), %%eax\n" ++ "91: "__copyuser_seg" movl 60(%4), %%edx\n" + " movnti %%eax, 56(%3)\n" + " movnti %%edx, 60(%3)\n" + " addl $-64, %0\n" +@@ -584,9 +697,9 @@ static unsigned long __copy_user_intel_n + " shrl $2, %0\n" + " andl $3, %%eax\n" + " cld\n" +- "6: rep; movsl\n" ++ "6: rep; "__copyuser_seg" movsl\n" + " movl %%eax,%0\n" +- "7: rep; movsb\n" ++ "7: rep; "__copyuser_seg" movsb\n" + "8:\n" + ".section .fixup,\"ax\"\n" + "9: lea 0(%%eax,%0,4),%0\n" +@@ -629,32 +742,36 @@ static unsigned long __copy_user_intel_n + */ + unsigned long __copy_user_zeroing_intel(void *to, const void __user *from, + unsigned long size); +-unsigned long __copy_user_intel(void __user *to, const void *from, ++unsigned long __generic_copy_to_user_intel(void __user *to, const void *from, ++ unsigned long size); ++unsigned long __generic_copy_from_user_intel(void *to, const void __user *from, + unsigned long size); + unsigned long __copy_user_zeroing_intel_nocache(void *to, + const void __user *from, unsigned long size); + #endif /* CONFIG_X86_INTEL_USERCOPY */ + + /* Generic arbitrary sized copy. */ +-#define __copy_user(to, from, size) \ ++#define __copy_user(to, from, size, prefix, set, restore) \ + do { \ + int __d0, __d1, __d2; \ + __asm__ __volatile__( \ ++ set \ + " cmp $7,%0\n" \ + " jbe 1f\n" \ + " movl %1,%0\n" \ + " negl %0\n" \ + " andl $7,%0\n" \ + " subl %0,%3\n" \ +- "4: rep; movsb\n" \ ++ "4: rep; "prefix"movsb\n" \ + " movl %3,%0\n" \ + " shrl $2,%0\n" \ + " andl $3,%3\n" \ + " .align 2,0x90\n" \ +- "0: rep; movsl\n" \ ++ "0: rep; "prefix"movsl\n" \ + " movl %3,%0\n" \ +- "1: rep; movsb\n" \ ++ "1: rep; "prefix"movsb\n" \ + "2:\n" \ ++ restore \ + ".section .fixup,\"ax\"\n" \ + "5: addl %3,%0\n" \ + " jmp 2b\n" \ +@@ -682,14 +799,14 @@ do { \ + " negl %0\n" \ + " andl $7,%0\n" \ + " subl %0,%3\n" \ +- "4: rep; movsb\n" \ ++ "4: rep; "__copyuser_seg"movsb\n" \ + " movl %3,%0\n" \ + " shrl $2,%0\n" \ + " andl $3,%3\n" \ + " .align 2,0x90\n" \ +- "0: rep; movsl\n" \ ++ "0: rep; "__copyuser_seg"movsl\n" \ + " movl %3,%0\n" \ +- "1: rep; movsb\n" \ ++ "1: rep; "__copyuser_seg"movsb\n" \ + "2:\n" \ + ".section .fixup,\"ax\"\n" \ + "5: addl %3,%0\n" \ +@@ -775,9 +892,9 @@ survive: + } + #endif + if (movsl_is_ok(to, from, n)) +- __copy_user(to, from, n); ++ __copy_user(to, from, n, "", __COPYUSER_SET_ES, __COPYUSER_RESTORE_ES); + else +- n = __copy_user_intel(to, from, n); ++ n = __generic_copy_to_user_intel(to, from, n); + return n; + } + EXPORT_SYMBOL(__copy_to_user_ll); +@@ -797,10 +914,9 @@ unsigned long __copy_from_user_ll_nozero + unsigned long n) + { + if (movsl_is_ok(to, from, n)) +- __copy_user(to, from, n); ++ __copy_user(to, from, n, __copyuser_seg, "", ""); + else +- n = __copy_user_intel((void __user *)to, +- (const void *)from, n); ++ n = __generic_copy_from_user_intel(to, from, n); + return n; + } + EXPORT_SYMBOL(__copy_from_user_ll_nozero); +@@ -827,65 +943,49 @@ unsigned long __copy_from_user_ll_nocach + if (n > 64 && cpu_has_xmm2) + n = __copy_user_intel_nocache(to, from, n); + else +- __copy_user(to, from, n); ++ __copy_user(to, from, n, __copyuser_seg, "", ""); + #else +- __copy_user(to, from, n); ++ __copy_user(to, from, n, __copyuser_seg, "", ""); + #endif + return n; + } + EXPORT_SYMBOL(__copy_from_user_ll_nocache_nozero); + +-/** +- * copy_to_user: - Copy a block of data into user space. +- * @to: Destination address, in user space. +- * @from: Source address, in kernel space. +- * @n: Number of bytes to copy. +- * +- * Context: User context only. This function may sleep. +- * +- * Copy data from kernel space to user space. +- * +- * Returns number of bytes that could not be copied. +- * On success, this will be zero. +- */ +-unsigned long +-copy_to_user(void __user *to, const void *from, unsigned long n) ++void copy_from_user_overflow(void) + { +- if (access_ok(VERIFY_WRITE, to, n)) +- n = __copy_to_user(to, from, n); +- return n; ++ WARN(1, "Buffer overflow detected!\n"); + } +-EXPORT_SYMBOL(copy_to_user); ++EXPORT_SYMBOL(copy_from_user_overflow); + +-/** +- * copy_from_user: - Copy a block of data from user space. +- * @to: Destination address, in kernel space. +- * @from: Source address, in user space. +- * @n: Number of bytes to copy. +- * +- * Context: User context only. This function may sleep. +- * +- * Copy data from user space to kernel space. +- * +- * Returns number of bytes that could not be copied. +- * On success, this will be zero. +- * +- * If some data could not be copied, this function will pad the copied +- * data to the requested size using zero bytes. +- */ +-unsigned long +-_copy_from_user(void *to, const void __user *from, unsigned long n) ++void copy_to_user_overflow(void) + { +- if (access_ok(VERIFY_READ, from, n)) +- n = __copy_from_user(to, from, n); +- else +- memset(to, 0, n); +- return n; ++ WARN(1, "Buffer overflow detected!\n"); + } +-EXPORT_SYMBOL(_copy_from_user); ++EXPORT_SYMBOL(copy_to_user_overflow); + +-void copy_from_user_overflow(void) ++#ifdef CONFIG_PAX_MEMORY_UDEREF ++void __set_fs(mm_segment_t x) + { +- WARN(1, "Buffer overflow detected!\n"); ++ switch (x.seg) { ++ case 0: ++ loadsegment(gs, 0); ++ break; ++ case TASK_SIZE_MAX: ++ loadsegment(gs, __USER_DS); ++ break; ++ case -1UL: ++ loadsegment(gs, __KERNEL_DS); ++ break; ++ default: ++ BUG(); ++ } ++ return; + } +-EXPORT_SYMBOL(copy_from_user_overflow); ++ ++void set_fs(mm_segment_t x) ++{ ++ current_thread_info()->addr_limit = x; ++ __set_fs(x); ++} ++EXPORT_SYMBOL(set_fs); ++#endif +diff -urNp linux-2.6.38.3/arch/x86/lib/usercopy_64.c linux-2.6.38.3/arch/x86/lib/usercopy_64.c +--- linux-2.6.38.3/arch/x86/lib/usercopy_64.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/lib/usercopy_64.c 2011-04-17 15:57:32.000000000 -0400 +@@ -42,6 +42,8 @@ long + __strncpy_from_user(char *dst, const char __user *src, long count) + { + long res; ++ if ((unsigned long)src < PAX_USER_SHADOW_BASE) ++ src += PAX_USER_SHADOW_BASE; + __do_strncpy_from_user(dst, src, count, res); + return res; + } +@@ -65,6 +67,8 @@ unsigned long __clear_user(void __user * + { + long __d0; + might_fault(); ++ if ((unsigned long)addr < PAX_USER_SHADOW_BASE) ++ addr += PAX_USER_SHADOW_BASE; + /* no memory constraint because it doesn't change any memory gcc knows + about */ + asm volatile( +@@ -151,10 +155,14 @@ EXPORT_SYMBOL(strlen_user); + + unsigned long copy_in_user(void __user *to, const void __user *from, unsigned len) + { +- if (access_ok(VERIFY_WRITE, to, len) && access_ok(VERIFY_READ, from, len)) { ++ if (access_ok(VERIFY_WRITE, to, len) && access_ok(VERIFY_READ, from, len)) { ++ if ((unsigned long)to < PAX_USER_SHADOW_BASE) ++ to += PAX_USER_SHADOW_BASE; ++ if ((unsigned long)from < PAX_USER_SHADOW_BASE) ++ from += PAX_USER_SHADOW_BASE; + return copy_user_generic((__force void *)to, (__force void *)from, len); +- } +- return len; ++ } ++ return len; + } + EXPORT_SYMBOL(copy_in_user); + +diff -urNp linux-2.6.38.3/arch/x86/Makefile linux-2.6.38.3/arch/x86/Makefile +--- linux-2.6.38.3/arch/x86/Makefile 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/Makefile 2011-04-17 15:57:32.000000000 -0400 +@@ -195,3 +195,12 @@ define archhelp + echo ' FDARGS="..." arguments for the booted kernel' + echo ' FDINITRD=file initrd for the booted kernel' + endef ++ ++define OLD_LD ++ ++*** ${VERSION}.${PATCHLEVEL} PaX kernels no longer build correctly with old versions of binutils. ++*** Please upgrade your binutils to 2.18 or newer ++endef ++ ++archprepare: ++ $(if $(LDFLAGS_BUILD_ID),,$(error $(OLD_LD))) +diff -urNp linux-2.6.38.3/arch/x86/mm/extable.c linux-2.6.38.3/arch/x86/mm/extable.c +--- linux-2.6.38.3/arch/x86/mm/extable.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/mm/extable.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1,14 +1,71 @@ + #include <linux/module.h> + #include <linux/spinlock.h> ++#include <linux/sort.h> + #include <asm/uaccess.h> ++#include <asm/pgtable.h> + ++/* ++ * The exception table needs to be sorted so that the binary ++ * search that we use to find entries in it works properly. ++ * This is used both for the kernel exception table and for ++ * the exception tables of modules that get loaded. ++ */ ++static int cmp_ex(const void *a, const void *b) ++{ ++ const struct exception_table_entry *x = a, *y = b; ++ ++ /* avoid overflow */ ++ if (x->insn > y->insn) ++ return 1; ++ if (x->insn < y->insn) ++ return -1; ++ return 0; ++} ++ ++static void swap_ex(void *a, void *b, int size) ++{ ++ struct exception_table_entry t, *x = a, *y = b; ++ ++ t = *x; ++ ++ pax_open_kernel(); ++ *x = *y; ++ *y = t; ++ pax_close_kernel(); ++} ++ ++void sort_extable(struct exception_table_entry *start, ++ struct exception_table_entry *finish) ++{ ++ sort(start, finish - start, sizeof(struct exception_table_entry), ++ cmp_ex, swap_ex); ++} ++ ++#ifdef CONFIG_MODULES ++/* ++ * If the exception table is sorted, any referring to the module init ++ * will be at the beginning or the end. ++ */ ++void trim_init_extable(struct module *m) ++{ ++ /*trim the beginning*/ ++ while (m->num_exentries && within_module_init(m->extable[0].insn, m)) { ++ m->extable++; ++ m->num_exentries--; ++ } ++ /*trim the end*/ ++ while (m->num_exentries && ++ within_module_init(m->extable[m->num_exentries-1].insn, m)) ++ m->num_exentries--; ++} ++#endif /* CONFIG_MODULES */ + + int fixup_exception(struct pt_regs *regs) + { + const struct exception_table_entry *fixup; + + #ifdef CONFIG_PNPBIOS +- if (unlikely(SEGMENT_IS_PNP_CODE(regs->cs))) { ++ if (unlikely(!v8086_mode(regs) && SEGMENT_IS_PNP_CODE(regs->cs))) { + extern u32 pnp_bios_fault_eip, pnp_bios_fault_esp; + extern u32 pnp_bios_is_utter_crap; + pnp_bios_is_utter_crap = 1; +diff -urNp linux-2.6.38.3/arch/x86/mm/fault.c linux-2.6.38.3/arch/x86/mm/fault.c +--- linux-2.6.38.3/arch/x86/mm/fault.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/mm/fault.c 2011-04-17 15:57:32.000000000 -0400 +@@ -12,10 +12,18 @@ + #include <linux/mmiotrace.h> /* kmmio_handler, ... */ + #include <linux/perf_event.h> /* perf_sw_event */ + #include <linux/hugetlb.h> /* hstate_index_to_shift */ ++#include <linux/unistd.h> ++#include <linux/compiler.h> + + #include <asm/traps.h> /* dotraplinkage, ... */ + #include <asm/pgalloc.h> /* pgd_*(), ... */ + #include <asm/kmemcheck.h> /* kmemcheck_*(), ... */ ++#include <asm/vsyscall.h> ++#include <asm/tlbflush.h> ++ ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) ++#include <asm/stacktrace.h> ++#endif + + /* + * Page fault error code bits: +@@ -53,7 +61,7 @@ static inline int __kprobes notify_page_ + int ret = 0; + + /* kprobe_running() needs smp_processor_id() */ +- if (kprobes_built_in() && !user_mode_vm(regs)) { ++ if (kprobes_built_in() && !user_mode(regs)) { + preempt_disable(); + if (kprobe_running() && kprobe_fault_handler(regs, 14)) + ret = 1; +@@ -114,7 +122,10 @@ check_prefetch_opcode(struct pt_regs *re + return !instr_lo || (instr_lo>>1) == 1; + case 0x00: + /* Prefetch instruction is 0x0F0D or 0x0F18 */ +- if (probe_kernel_address(instr, opcode)) ++ if (user_mode(regs)) { ++ if (__copy_from_user_inatomic(&opcode, (__force unsigned char __user *)(instr), 1)) ++ return 0; ++ } else if (probe_kernel_address(instr, opcode)) + return 0; + + *prefetch = (instr_lo == 0xF) && +@@ -148,7 +159,10 @@ is_prefetch(struct pt_regs *regs, unsign + while (instr < max_instr) { + unsigned char opcode; + +- if (probe_kernel_address(instr, opcode)) ++ if (user_mode(regs)) { ++ if (__copy_from_user_inatomic(&opcode, (__force unsigned char __user *)(instr), 1)) ++ break; ++ } else if (probe_kernel_address(instr, opcode)) + break; + + instr++; +@@ -179,6 +193,30 @@ force_sig_info_fault(int si_signo, int s + force_sig_info(si_signo, &info, tsk); + } + ++#ifdef CONFIG_PAX_EMUTRAMP ++static int pax_handle_fetch_fault(struct pt_regs *regs); ++#endif ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++static inline pmd_t * pax_get_pmd(struct mm_struct *mm, unsigned long address) ++{ ++ pgd_t *pgd; ++ pud_t *pud; ++ pmd_t *pmd; ++ ++ pgd = pgd_offset(mm, address); ++ if (!pgd_present(*pgd)) ++ return NULL; ++ pud = pud_offset(pgd, address); ++ if (!pud_present(*pud)) ++ return NULL; ++ pmd = pmd_offset(pud, address); ++ if (!pmd_present(*pmd)) ++ return NULL; ++ return pmd; ++} ++#endif ++ + DEFINE_SPINLOCK(pgd_lock); + LIST_HEAD(pgd_list); + +@@ -229,10 +267,22 @@ void vmalloc_sync_all(void) + for (address = VMALLOC_START & PMD_MASK; + address >= TASK_SIZE && address < FIXADDR_TOP; + address += PMD_SIZE) { ++ ++#ifdef CONFIG_PAX_PER_CPU_PGD ++ unsigned long cpu; ++#else + struct page *page; ++#endif + + spin_lock(&pgd_lock); ++ ++#ifdef CONFIG_PAX_PER_CPU_PGD ++ for (cpu = 0; cpu < NR_CPUS; ++cpu) { ++ pgd_t *pgd = get_cpu_pgd(cpu); ++ pmd_t *ret; ++#else + list_for_each_entry(page, &pgd_list, lru) { ++ pgd_t *pgd = page_address(page); + spinlock_t *pgt_lock; + pmd_t *ret; + +@@ -240,8 +290,13 @@ void vmalloc_sync_all(void) + pgt_lock = &pgd_page_get_mm(page)->page_table_lock; + + spin_lock(pgt_lock); +- ret = vmalloc_sync_one(page_address(page), address); ++#endif ++ ++ ret = vmalloc_sync_one(pgd, address); ++ ++#ifndef CONFIG_PAX_PER_CPU_PGD + spin_unlock(pgt_lock); ++#endif + + if (!ret) + break; +@@ -275,6 +330,11 @@ static noinline __kprobes int vmalloc_fa + * an interrupt in the middle of a task switch.. + */ + pgd_paddr = read_cr3(); ++ ++#ifdef CONFIG_PAX_PER_CPU_PGD ++ BUG_ON(__pa(get_cpu_pgd(smp_processor_id())) != (pgd_paddr & PHYSICAL_PAGE_MASK)); ++#endif ++ + pmd_k = vmalloc_sync_one(__va(pgd_paddr), address); + if (!pmd_k) + return -1; +@@ -370,7 +430,14 @@ static noinline __kprobes int vmalloc_fa + * happen within a race in page table update. In the later + * case just flush: + */ ++ ++#ifdef CONFIG_PAX_PER_CPU_PGD ++ BUG_ON(__pa(get_cpu_pgd(smp_processor_id())) != (read_cr3() & PHYSICAL_PAGE_MASK)); ++ pgd = pgd_offset_cpu(smp_processor_id(), address); ++#else + pgd = pgd_offset(current->active_mm, address); ++#endif ++ + pgd_ref = pgd_offset_k(address); + if (pgd_none(*pgd_ref)) + return -1; +@@ -532,7 +599,7 @@ static int is_errata93(struct pt_regs *r + static int is_errata100(struct pt_regs *regs, unsigned long address) + { + #ifdef CONFIG_X86_64 +- if ((regs->cs == __USER32_CS || (regs->cs & (1<<2))) && (address >> 32)) ++ if ((regs->cs == __USER32_CS || (regs->cs & SEGMENT_LDT)) && (address >> 32)) + return 1; + #endif + return 0; +@@ -559,7 +626,7 @@ static int is_f00f_bug(struct pt_regs *r + } + + static const char nx_warning[] = KERN_CRIT +-"kernel tried to execute NX-protected page - exploit attempt? (uid: %d)\n"; ++"kernel tried to execute NX-protected page - exploit attempt? (uid: %d, task: %s, pid: %d)\n"; + + static void + show_fault_oops(struct pt_regs *regs, unsigned long error_code, +@@ -568,15 +635,26 @@ show_fault_oops(struct pt_regs *regs, un + if (!oops_may_print()) + return; + +- if (error_code & PF_INSTR) { ++ if ((__supported_pte_mask & _PAGE_NX) && (error_code & PF_INSTR)) { + unsigned int level; + + pte_t *pte = lookup_address(address, &level); + + if (pte && pte_present(*pte) && !pte_exec(*pte)) +- printk(nx_warning, current_uid()); ++ printk(nx_warning, current_uid(), current->comm, task_pid_nr(current)); + } + ++#ifdef CONFIG_PAX_KERNEXEC ++ if (init_mm.start_code <= address && address < init_mm.end_code) { ++ if (current->signal->curr_ip) ++ printk(KERN_ERR "PAX: From %pI4: %s:%d, uid/euid: %u/%u, attempted to modify kernel code\n", ++ ¤t->signal->curr_ip, current->comm, task_pid_nr(current), current_uid(), current_euid()); ++ else ++ printk(KERN_ERR "PAX: %s:%d, uid/euid: %u/%u, attempted to modify kernel code\n", ++ current->comm, task_pid_nr(current), current_uid(), current_euid()); ++ } ++#endif ++ + printk(KERN_ALERT "BUG: unable to handle kernel "); + if (address < PAGE_SIZE) + printk(KERN_CONT "NULL pointer dereference"); +@@ -701,6 +779,68 @@ __bad_area_nosemaphore(struct pt_regs *r + unsigned long address, int si_code) + { + struct task_struct *tsk = current; ++ struct mm_struct *mm = tsk->mm; ++ ++#ifdef CONFIG_X86_64 ++ if (mm && (error_code & PF_INSTR) && mm->context.vdso) { ++ if (regs->ip == (unsigned long)vgettimeofday) { ++ regs->ip = (unsigned long)VDSO64_SYMBOL(mm->context.vdso, fallback_gettimeofday); ++ return; ++ } else if (regs->ip == (unsigned long)vtime) { ++ regs->ip = (unsigned long)VDSO64_SYMBOL(mm->context.vdso, fallback_time); ++ return; ++ } else if (regs->ip == (unsigned long)vgetcpu) { ++ regs->ip = (unsigned long)VDSO64_SYMBOL(mm->context.vdso, getcpu); ++ return; ++ } ++ } ++#endif ++ ++#if defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) ++ if (mm && (error_code & PF_USER)) { ++ unsigned long ip = regs->ip; ++ ++ if (v8086_mode(regs)) ++ ip = ((regs->cs & 0xffff) << 4) + (ip & 0xffff); ++ ++ /* ++ * It's possible to have interrupts off here: ++ */ ++ local_irq_enable(); ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if ((mm->pax_flags & MF_PAX_PAGEEXEC) && ++ (((__supported_pte_mask & _PAGE_NX) && (error_code & PF_INSTR)) || (!(error_code & (PF_PROT | PF_WRITE)) && ip == address))) { ++ ++#ifdef CONFIG_PAX_EMUTRAMP ++ switch (pax_handle_fetch_fault(regs)) { ++ case 2: ++ return; ++ } ++#endif ++ ++ pax_report_fault(regs, (void *)ip, (void *)regs->sp); ++ do_group_exit(SIGKILL); ++ } ++#endif ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if ((mm->pax_flags & MF_PAX_SEGMEXEC) && !(error_code & (PF_PROT | PF_WRITE)) && (ip + SEGMEXEC_TASK_SIZE == address)) { ++ ++#ifdef CONFIG_PAX_EMUTRAMP ++ switch (pax_handle_fetch_fault(regs)) { ++ case 2: ++ return; ++ } ++#endif ++ ++ pax_report_fault(regs, (void *)ip, (void *)regs->sp); ++ do_group_exit(SIGKILL); ++ } ++#endif ++ ++ } ++#endif + + /* User mode accesses just cause a SIGSEGV */ + if (error_code & PF_USER) { +@@ -855,6 +995,99 @@ static int spurious_fault_check(unsigned + return 1; + } + ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_PAGEEXEC) ++static int pax_handle_pageexec_fault(struct pt_regs *regs, struct mm_struct *mm, unsigned long address, unsigned long error_code) ++{ ++ pte_t *pte; ++ pmd_t *pmd; ++ spinlock_t *ptl; ++ unsigned char pte_mask; ++ ++ if ((__supported_pte_mask & _PAGE_NX) || (error_code & (PF_PROT|PF_USER)) != (PF_PROT|PF_USER) || v8086_mode(regs) || ++ !(mm->pax_flags & MF_PAX_PAGEEXEC)) ++ return 0; ++ ++ /* PaX: it's our fault, let's handle it if we can */ ++ ++ /* PaX: take a look at read faults before acquiring any locks */ ++ if (unlikely(!(error_code & PF_WRITE) && (regs->ip == address))) { ++ /* instruction fetch attempt from a protected page in user mode */ ++ up_read(&mm->mmap_sem); ++ ++#ifdef CONFIG_PAX_EMUTRAMP ++ switch (pax_handle_fetch_fault(regs)) { ++ case 2: ++ return 1; ++ } ++#endif ++ ++ pax_report_fault(regs, (void *)regs->ip, (void *)regs->sp); ++ do_group_exit(SIGKILL); ++ } ++ ++ pmd = pax_get_pmd(mm, address); ++ if (unlikely(!pmd)) ++ return 0; ++ ++ pte = pte_offset_map_lock(mm, pmd, address, &ptl); ++ if (unlikely(!(pte_val(*pte) & _PAGE_PRESENT) || pte_user(*pte))) { ++ pte_unmap_unlock(pte, ptl); ++ return 0; ++ } ++ ++ if (unlikely((error_code & PF_WRITE) && !pte_write(*pte))) { ++ /* write attempt to a protected page in user mode */ ++ pte_unmap_unlock(pte, ptl); ++ return 0; ++ } ++ ++#ifdef CONFIG_SMP ++ if (likely(address > get_limit(regs->cs) && cpu_isset(smp_processor_id(), mm->context.cpu_user_cs_mask))) ++#else ++ if (likely(address > get_limit(regs->cs))) ++#endif ++ { ++ set_pte(pte, pte_mkread(*pte)); ++ __flush_tlb_one(address); ++ pte_unmap_unlock(pte, ptl); ++ up_read(&mm->mmap_sem); ++ return 1; ++ } ++ ++ pte_mask = _PAGE_ACCESSED | _PAGE_USER | ((error_code & PF_WRITE) << (_PAGE_BIT_DIRTY-1)); ++ ++ /* ++ * PaX: fill DTLB with user rights and retry ++ */ ++ __asm__ __volatile__ ( ++ "orb %2,(%1)\n" ++#if defined(CONFIG_M586) || defined(CONFIG_M586TSC) ++/* ++ * PaX: let this uncommented 'invlpg' remind us on the behaviour of Intel's ++ * (and AMD's) TLBs. namely, they do not cache PTEs that would raise *any* ++ * page fault when examined during a TLB load attempt. this is true not only ++ * for PTEs holding a non-present entry but also present entries that will ++ * raise a page fault (such as those set up by PaX, or the copy-on-write ++ * mechanism). in effect it means that we do *not* need to flush the TLBs ++ * for our target pages since their PTEs are simply not in the TLBs at all. ++ ++ * the best thing in omitting it is that we gain around 15-20% speed in the ++ * fast path of the page fault handler and can get rid of tracing since we ++ * can no longer flush unintended entries. ++ */ ++ "invlpg (%0)\n" ++#endif ++ __copyuser_seg"testb $0,(%0)\n" ++ "xorb %3,(%1)\n" ++ : ++ : "r" (address), "r" (pte), "q" (pte_mask), "i" (_PAGE_USER) ++ : "memory", "cc"); ++ pte_unmap_unlock(pte, ptl); ++ up_read(&mm->mmap_sem); ++ return 1; ++} ++#endif ++ + /* + * Handle a spurious fault caused by a stale TLB entry. + * +@@ -927,6 +1160,9 @@ int show_unhandled_signals = 1; + static inline int + access_error(unsigned long error_code, struct vm_area_struct *vma) + { ++ if ((__supported_pte_mask & _PAGE_NX) && (error_code & PF_INSTR) && !(vma->vm_flags & VM_EXEC)) ++ return 1; ++ + if (error_code & PF_WRITE) { + /* write, present and write, not present: */ + if (unlikely(!(vma->vm_flags & VM_WRITE))) +@@ -960,19 +1196,33 @@ do_page_fault(struct pt_regs *regs, unsi + { + struct vm_area_struct *vma; + struct task_struct *tsk; +- unsigned long address; + struct mm_struct *mm; + int fault; + int write = error_code & PF_WRITE; + unsigned int flags = FAULT_FLAG_ALLOW_RETRY | + (write ? FAULT_FLAG_WRITE : 0); + ++ /* Get the faulting address: */ ++ unsigned long address = read_cr2(); ++ ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) ++ if (!user_mode(regs) && address < 2 * PAX_USER_SHADOW_BASE) { ++ if (!search_exception_tables(regs->ip)) { ++ bad_area_nosemaphore(regs, error_code, address); ++ return; ++ } ++ if (address < PAX_USER_SHADOW_BASE) { ++ printk(KERN_ERR "PAX: please report this to pageexec@freemail.hu\n"); ++ printk(KERN_ERR "PAX: faulting IP: %pA\n", (void *)regs->ip); ++ show_trace_log_lvl(NULL, NULL, (void *)regs->sp, KERN_ERR); ++ } else ++ address -= PAX_USER_SHADOW_BASE; ++ } ++#endif ++ + tsk = current; + mm = tsk->mm; + +- /* Get the faulting address: */ +- address = read_cr2(); +- + /* + * Detect and handle instructions that would cause a page fault for + * both a tracked kernel page and a userspace page. +@@ -1032,7 +1282,7 @@ do_page_fault(struct pt_regs *regs, unsi + * User-mode registers count as a user access even for any + * potential system fault or CPU buglet: + */ +- if (user_mode_vm(regs)) { ++ if (user_mode(regs)) { + local_irq_enable(); + error_code |= PF_USER; + } else { +@@ -1087,6 +1337,11 @@ retry: + might_sleep(); + } + ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_PAGEEXEC) ++ if (pax_handle_pageexec_fault(regs, mm, address, error_code)) ++ return; ++#endif ++ + vma = find_vma(mm, address); + if (unlikely(!vma)) { + bad_area(regs, error_code, address); +@@ -1098,18 +1353,24 @@ retry: + bad_area(regs, error_code, address); + return; + } +- if (error_code & PF_USER) { +- /* +- * Accessing the stack below %sp is always a bug. +- * The large cushion allows instructions like enter +- * and pusha to work. ("enter $65535, $31" pushes +- * 32 pointers and then decrements %sp by 65535.) +- */ +- if (unlikely(address + 65536 + 32 * sizeof(unsigned long) < regs->sp)) { +- bad_area(regs, error_code, address); +- return; +- } ++ /* ++ * Accessing the stack below %sp is always a bug. ++ * The large cushion allows instructions like enter ++ * and pusha to work. ("enter $65535, $31" pushes ++ * 32 pointers and then decrements %sp by 65535.) ++ */ ++ if (unlikely(address + 65536 + 32 * sizeof(unsigned long) < task_pt_regs(tsk)->sp)) { ++ bad_area(regs, error_code, address); ++ return; + } ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (unlikely((mm->pax_flags & MF_PAX_SEGMEXEC) && vma->vm_end - SEGMEXEC_TASK_SIZE - 1 < address - SEGMEXEC_TASK_SIZE - 1)) { ++ bad_area(regs, error_code, address); ++ return; ++ } ++#endif ++ + if (unlikely(expand_stack(vma, address))) { + bad_area(regs, error_code, address); + return; +@@ -1164,3 +1425,199 @@ good_area: + + up_read(&mm->mmap_sem); + } ++ ++#ifdef CONFIG_PAX_EMUTRAMP ++static int pax_handle_fetch_fault_32(struct pt_regs *regs) ++{ ++ int err; ++ ++ do { /* PaX: gcc trampoline emulation #1 */ ++ unsigned char mov1, mov2; ++ unsigned short jmp; ++ unsigned int addr1, addr2; ++ ++#ifdef CONFIG_X86_64 ++ if ((regs->ip + 11) >> 32) ++ break; ++#endif ++ ++ err = get_user(mov1, (unsigned char __user *)regs->ip); ++ err |= get_user(addr1, (unsigned int __user *)(regs->ip + 1)); ++ err |= get_user(mov2, (unsigned char __user *)(regs->ip + 5)); ++ err |= get_user(addr2, (unsigned int __user *)(regs->ip + 6)); ++ err |= get_user(jmp, (unsigned short __user *)(regs->ip + 10)); ++ ++ if (err) ++ break; ++ ++ if (mov1 == 0xB9 && mov2 == 0xB8 && jmp == 0xE0FF) { ++ regs->cx = addr1; ++ regs->ax = addr2; ++ regs->ip = addr2; ++ return 2; ++ } ++ } while (0); ++ ++ do { /* PaX: gcc trampoline emulation #2 */ ++ unsigned char mov, jmp; ++ unsigned int addr1, addr2; ++ ++#ifdef CONFIG_X86_64 ++ if ((regs->ip + 9) >> 32) ++ break; ++#endif ++ ++ err = get_user(mov, (unsigned char __user *)regs->ip); ++ err |= get_user(addr1, (unsigned int __user *)(regs->ip + 1)); ++ err |= get_user(jmp, (unsigned char __user *)(regs->ip + 5)); ++ err |= get_user(addr2, (unsigned int __user *)(regs->ip + 6)); ++ ++ if (err) ++ break; ++ ++ if (mov == 0xB9 && jmp == 0xE9) { ++ regs->cx = addr1; ++ regs->ip = (unsigned int)(regs->ip + addr2 + 10); ++ return 2; ++ } ++ } while (0); ++ ++ return 1; /* PaX in action */ ++} ++ ++#ifdef CONFIG_X86_64 ++static int pax_handle_fetch_fault_64(struct pt_regs *regs) ++{ ++ int err; ++ ++ do { /* PaX: gcc trampoline emulation #1 */ ++ unsigned short mov1, mov2, jmp1; ++ unsigned char jmp2; ++ unsigned int addr1; ++ unsigned long addr2; ++ ++ err = get_user(mov1, (unsigned short __user *)regs->ip); ++ err |= get_user(addr1, (unsigned int __user *)(regs->ip + 2)); ++ err |= get_user(mov2, (unsigned short __user *)(regs->ip + 6)); ++ err |= get_user(addr2, (unsigned long __user *)(regs->ip + 8)); ++ err |= get_user(jmp1, (unsigned short __user *)(regs->ip + 16)); ++ err |= get_user(jmp2, (unsigned char __user *)(regs->ip + 18)); ++ ++ if (err) ++ break; ++ ++ if (mov1 == 0xBB41 && mov2 == 0xBA49 && jmp1 == 0xFF49 && jmp2 == 0xE3) { ++ regs->r11 = addr1; ++ regs->r10 = addr2; ++ regs->ip = addr1; ++ return 2; ++ } ++ } while (0); ++ ++ do { /* PaX: gcc trampoline emulation #2 */ ++ unsigned short mov1, mov2, jmp1; ++ unsigned char jmp2; ++ unsigned long addr1, addr2; ++ ++ err = get_user(mov1, (unsigned short __user *)regs->ip); ++ err |= get_user(addr1, (unsigned long __user *)(regs->ip + 2)); ++ err |= get_user(mov2, (unsigned short __user *)(regs->ip + 10)); ++ err |= get_user(addr2, (unsigned long __user *)(regs->ip + 12)); ++ err |= get_user(jmp1, (unsigned short __user *)(regs->ip + 20)); ++ err |= get_user(jmp2, (unsigned char __user *)(regs->ip + 22)); ++ ++ if (err) ++ break; ++ ++ if (mov1 == 0xBB49 && mov2 == 0xBA49 && jmp1 == 0xFF49 && jmp2 == 0xE3) { ++ regs->r11 = addr1; ++ regs->r10 = addr2; ++ regs->ip = addr1; ++ return 2; ++ } ++ } while (0); ++ ++ return 1; /* PaX in action */ ++} ++#endif ++ ++/* ++ * PaX: decide what to do with offenders (regs->ip = fault address) ++ * ++ * returns 1 when task should be killed ++ * 2 when gcc trampoline was detected ++ */ ++static int pax_handle_fetch_fault(struct pt_regs *regs) ++{ ++ if (v8086_mode(regs)) ++ return 1; ++ ++ if (!(current->mm->pax_flags & MF_PAX_EMUTRAMP)) ++ return 1; ++ ++#ifdef CONFIG_X86_32 ++ return pax_handle_fetch_fault_32(regs); ++#else ++ if (regs->cs == __USER32_CS || (regs->cs & SEGMENT_LDT)) ++ return pax_handle_fetch_fault_32(regs); ++ else ++ return pax_handle_fetch_fault_64(regs); ++#endif ++} ++#endif ++ ++#if defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) ++void pax_report_insns(void *pc, void *sp) ++{ ++ long i; ++ ++ printk(KERN_ERR "PAX: bytes at PC: "); ++ for (i = 0; i < 20; i++) { ++ unsigned char c; ++ if (get_user(c, (__force unsigned char __user *)pc+i)) ++ printk(KERN_CONT "?? "); ++ else ++ printk(KERN_CONT "%02x ", c); ++ } ++ printk("\n"); ++ ++ printk(KERN_ERR "PAX: bytes at SP-%lu: ", (unsigned long)sizeof(long)); ++ for (i = -1; i < 80 / (long)sizeof(long); i++) { ++ unsigned long c; ++ if (get_user(c, (__force unsigned long __user *)sp+i)) ++#ifdef CONFIG_X86_32 ++ printk(KERN_CONT "???????? "); ++#else ++ printk(KERN_CONT "???????????????? "); ++#endif ++ else ++ printk(KERN_CONT "%0*lx ", 2 * (int)sizeof(long), c); ++ } ++ printk("\n"); ++} ++#endif ++ ++/** ++ * probe_kernel_write(): safely attempt to write to a location ++ * @dst: address to write to ++ * @src: pointer to the data that shall be written ++ * @size: size of the data chunk ++ * ++ * Safely write to address @dst from the buffer at @src. If a kernel fault ++ * happens, handle that and return -EFAULT. ++ */ ++long notrace probe_kernel_write(void *dst, const void *src, size_t size) ++{ ++ long ret; ++ mm_segment_t old_fs = get_fs(); ++ ++ set_fs(KERNEL_DS); ++ pagefault_disable(); ++ pax_open_kernel(); ++ ret = __copy_to_user_inatomic((__force void __user *)dst, src, size); ++ pax_close_kernel(); ++ pagefault_enable(); ++ set_fs(old_fs); ++ ++ return ret ? -EFAULT : 0; ++} +diff -urNp linux-2.6.38.3/arch/x86/mm/gup.c linux-2.6.38.3/arch/x86/mm/gup.c +--- linux-2.6.38.3/arch/x86/mm/gup.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/mm/gup.c 2011-04-17 15:57:32.000000000 -0400 +@@ -263,7 +263,7 @@ int __get_user_pages_fast(unsigned long + addr = start; + len = (unsigned long) nr_pages << PAGE_SHIFT; + end = start + len; +- if (unlikely(!access_ok(write ? VERIFY_WRITE : VERIFY_READ, ++ if (unlikely(!__access_ok(write ? VERIFY_WRITE : VERIFY_READ, + (void __user *)start, len))) + return 0; + +diff -urNp linux-2.6.38.3/arch/x86/mm/highmem_32.c linux-2.6.38.3/arch/x86/mm/highmem_32.c +--- linux-2.6.38.3/arch/x86/mm/highmem_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/mm/highmem_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -44,7 +44,10 @@ void *kmap_atomic_prot(struct page *page + idx = type + KM_TYPE_NR*smp_processor_id(); + vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx); + BUG_ON(!pte_none(*(kmap_pte-idx))); ++ ++ pax_open_kernel(); + set_pte(kmap_pte-idx, mk_pte(page, prot)); ++ pax_close_kernel(); + + return (void *)vaddr; + } +diff -urNp linux-2.6.38.3/arch/x86/mm/hugetlbpage.c linux-2.6.38.3/arch/x86/mm/hugetlbpage.c +--- linux-2.6.38.3/arch/x86/mm/hugetlbpage.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/mm/hugetlbpage.c 2011-04-17 15:57:32.000000000 -0400 +@@ -266,13 +266,20 @@ static unsigned long hugetlb_get_unmappe + struct hstate *h = hstate_file(file); + struct mm_struct *mm = current->mm; + struct vm_area_struct *vma; +- unsigned long start_addr; ++ unsigned long start_addr, pax_task_size = TASK_SIZE; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (mm->pax_flags & MF_PAX_SEGMEXEC) ++ pax_task_size = SEGMEXEC_TASK_SIZE; ++#endif ++ ++ pax_task_size -= PAGE_SIZE; + + if (len > mm->cached_hole_size) { +- start_addr = mm->free_area_cache; ++ start_addr = mm->free_area_cache; + } else { +- start_addr = TASK_UNMAPPED_BASE; +- mm->cached_hole_size = 0; ++ start_addr = mm->mmap_base; ++ mm->cached_hole_size = 0; + } + + full_search: +@@ -280,26 +287,27 @@ full_search: + + for (vma = find_vma(mm, addr); ; vma = vma->vm_next) { + /* At this point: (!vma || addr < vma->vm_end). */ +- if (TASK_SIZE - len < addr) { ++ if (pax_task_size - len < addr) { + /* + * Start a new search - just in case we missed + * some holes. + */ +- if (start_addr != TASK_UNMAPPED_BASE) { +- start_addr = TASK_UNMAPPED_BASE; ++ if (start_addr != mm->mmap_base) { ++ start_addr = mm->mmap_base; + mm->cached_hole_size = 0; + goto full_search; + } + return -ENOMEM; + } +- if (!vma || addr + len <= vma->vm_start) { +- mm->free_area_cache = addr + len; +- return addr; +- } ++ if (check_heap_stack_gap(vma, addr, len)) ++ break; + if (addr + mm->cached_hole_size < vma->vm_start) + mm->cached_hole_size = vma->vm_start - addr; + addr = ALIGN(vma->vm_end, huge_page_size(h)); + } ++ ++ mm->free_area_cache = addr + len; ++ return addr; + } + + static unsigned long hugetlb_get_unmapped_area_topdown(struct file *file, +@@ -308,10 +316,9 @@ static unsigned long hugetlb_get_unmappe + { + struct hstate *h = hstate_file(file); + struct mm_struct *mm = current->mm; +- struct vm_area_struct *vma, *prev_vma; +- unsigned long base = mm->mmap_base, addr = addr0; ++ struct vm_area_struct *vma; ++ unsigned long base = mm->mmap_base, addr; + unsigned long largest_hole = mm->cached_hole_size; +- int first_time = 1; + + /* don't allow allocations above current base */ + if (mm->free_area_cache > base) +@@ -321,64 +328,63 @@ static unsigned long hugetlb_get_unmappe + largest_hole = 0; + mm->free_area_cache = base; + } +-try_again: ++ + /* make sure it can fit in the remaining address space */ + if (mm->free_area_cache < len) + goto fail; + + /* either no address requested or cant fit in requested address hole */ +- addr = (mm->free_area_cache - len) & huge_page_mask(h); ++ addr = (mm->free_area_cache - len); + do { ++ addr &= huge_page_mask(h); ++ vma = find_vma(mm, addr); + /* + * Lookup failure means no vma is above this address, + * i.e. return with success: +- */ +- if (!(vma = find_vma_prev(mm, addr, &prev_vma))) +- return addr; +- +- /* + * new region fits between prev_vma->vm_end and + * vma->vm_start, use it: + */ +- if (addr + len <= vma->vm_start && +- (!prev_vma || (addr >= prev_vma->vm_end))) { ++ if (check_heap_stack_gap(vma, addr, len)) { + /* remember the address as a hint for next time */ +- mm->cached_hole_size = largest_hole; +- return (mm->free_area_cache = addr); +- } else { +- /* pull free_area_cache down to the first hole */ +- if (mm->free_area_cache == vma->vm_end) { +- mm->free_area_cache = vma->vm_start; +- mm->cached_hole_size = largest_hole; +- } ++ mm->cached_hole_size = largest_hole; ++ return (mm->free_area_cache = addr); ++ } ++ /* pull free_area_cache down to the first hole */ ++ if (mm->free_area_cache == vma->vm_end) { ++ mm->free_area_cache = vma->vm_start; ++ mm->cached_hole_size = largest_hole; + } + + /* remember the largest hole we saw so far */ + if (addr + largest_hole < vma->vm_start) +- largest_hole = vma->vm_start - addr; ++ largest_hole = vma->vm_start - addr; + + /* try just below the current vma->vm_start */ +- addr = (vma->vm_start - len) & huge_page_mask(h); +- } while (len <= vma->vm_start); ++ addr = skip_heap_stack_gap(vma, len); ++ } while (!IS_ERR_VALUE(addr)); + + fail: + /* +- * if hint left us with no space for the requested +- * mapping then try again: +- */ +- if (first_time) { +- mm->free_area_cache = base; +- largest_hole = 0; +- first_time = 0; +- goto try_again; +- } +- /* + * A failed mmap() very likely causes application failure, + * so fall back to the bottom-up function here. This scenario + * can happen with large stack limits and large mmap() + * allocations. + */ +- mm->free_area_cache = TASK_UNMAPPED_BASE; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (mm->pax_flags & MF_PAX_SEGMEXEC) ++ mm->mmap_base = SEGMEXEC_TASK_UNMAPPED_BASE; ++ else ++#endif ++ ++ mm->mmap_base = TASK_UNMAPPED_BASE; ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base += mm->delta_mmap; ++#endif ++ ++ mm->free_area_cache = mm->mmap_base; + mm->cached_hole_size = ~0UL; + addr = hugetlb_get_unmapped_area_bottomup(file, addr0, + len, pgoff, flags); +@@ -386,6 +392,7 @@ fail: + /* + * Restore the topdown base: + */ ++ mm->mmap_base = base; + mm->free_area_cache = base; + mm->cached_hole_size = ~0UL; + +@@ -399,10 +406,19 @@ hugetlb_get_unmapped_area(struct file *f + struct hstate *h = hstate_file(file); + struct mm_struct *mm = current->mm; + struct vm_area_struct *vma; ++ unsigned long pax_task_size = TASK_SIZE; + + if (len & ~huge_page_mask(h)) + return -EINVAL; +- if (len > TASK_SIZE) ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (mm->pax_flags & MF_PAX_SEGMEXEC) ++ pax_task_size = SEGMEXEC_TASK_SIZE; ++#endif ++ ++ pax_task_size -= PAGE_SIZE; ++ ++ if (len > pax_task_size) + return -ENOMEM; + + if (flags & MAP_FIXED) { +@@ -414,8 +430,7 @@ hugetlb_get_unmapped_area(struct file *f + if (addr) { + addr = ALIGN(addr, huge_page_size(h)); + vma = find_vma(mm, addr); +- if (TASK_SIZE - len >= addr && +- (!vma || addr + len <= vma->vm_start)) ++ if (pax_task_size - len >= addr && check_heap_stack_gap(vma, addr, len)) + return addr; + } + if (mm->get_unmapped_area == arch_get_unmapped_area) +diff -urNp linux-2.6.38.3/arch/x86/mm/init_32.c linux-2.6.38.3/arch/x86/mm/init_32.c +--- linux-2.6.38.3/arch/x86/mm/init_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/mm/init_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -74,36 +74,6 @@ static __init void *alloc_low_page(void) + } + + /* +- * Creates a middle page table and puts a pointer to it in the +- * given global directory entry. This only returns the gd entry +- * in non-PAE compilation mode, since the middle layer is folded. +- */ +-static pmd_t * __init one_md_table_init(pgd_t *pgd) +-{ +- pud_t *pud; +- pmd_t *pmd_table; +- +-#ifdef CONFIG_X86_PAE +- if (!(pgd_val(*pgd) & _PAGE_PRESENT)) { +- if (after_bootmem) +- pmd_table = (pmd_t *)alloc_bootmem_pages(PAGE_SIZE); +- else +- pmd_table = (pmd_t *)alloc_low_page(); +- paravirt_alloc_pmd(&init_mm, __pa(pmd_table) >> PAGE_SHIFT); +- set_pgd(pgd, __pgd(__pa(pmd_table) | _PAGE_PRESENT)); +- pud = pud_offset(pgd, 0); +- BUG_ON(pmd_table != pmd_offset(pud, 0)); +- +- return pmd_table; +- } +-#endif +- pud = pud_offset(pgd, 0); +- pmd_table = pmd_offset(pud, 0); +- +- return pmd_table; +-} +- +-/* + * Create a page table and place a pointer to it in a middle page + * directory entry: + */ +@@ -123,13 +93,28 @@ static pte_t * __init one_page_table_ini + page_table = (pte_t *)alloc_low_page(); + + paravirt_alloc_pte(&init_mm, __pa(page_table) >> PAGE_SHIFT); ++#if defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) ++ set_pmd(pmd, __pmd(__pa(page_table) | _KERNPG_TABLE)); ++#else + set_pmd(pmd, __pmd(__pa(page_table) | _PAGE_TABLE)); ++#endif + BUG_ON(page_table != pte_offset_kernel(pmd, 0)); + } + + return pte_offset_kernel(pmd, 0); + } + ++static pmd_t * __init one_md_table_init(pgd_t *pgd) ++{ ++ pud_t *pud; ++ pmd_t *pmd_table; ++ ++ pud = pud_offset(pgd, 0); ++ pmd_table = pmd_offset(pud, 0); ++ ++ return pmd_table; ++} ++ + pmd_t * __init populate_extra_pmd(unsigned long vaddr) + { + int pgd_idx = pgd_index(vaddr); +@@ -203,6 +188,7 @@ page_table_range_init(unsigned long star + int pgd_idx, pmd_idx; + unsigned long vaddr; + pgd_t *pgd; ++ pud_t *pud; + pmd_t *pmd; + pte_t *pte = NULL; + +@@ -212,8 +198,13 @@ page_table_range_init(unsigned long star + pgd = pgd_base + pgd_idx; + + for ( ; (pgd_idx < PTRS_PER_PGD) && (vaddr != end); pgd++, pgd_idx++) { +- pmd = one_md_table_init(pgd); +- pmd = pmd + pmd_index(vaddr); ++ pud = pud_offset(pgd, vaddr); ++ pmd = pmd_offset(pud, vaddr); ++ ++#ifdef CONFIG_X86_PAE ++ paravirt_alloc_pmd(&init_mm, __pa(pmd) >> PAGE_SHIFT); ++#endif ++ + for (; (pmd_idx < PTRS_PER_PMD) && (vaddr != end); + pmd++, pmd_idx++) { + pte = page_table_kmap_check(one_page_table_init(pmd), +@@ -225,11 +216,20 @@ page_table_range_init(unsigned long star + } + } + +-static inline int is_kernel_text(unsigned long addr) ++static inline int is_kernel_text(unsigned long start, unsigned long end) + { +- if (addr >= (unsigned long)_text && addr <= (unsigned long)__init_end) +- return 1; +- return 0; ++ if ((start > ktla_ktva((unsigned long)_etext) || ++ end <= ktla_ktva((unsigned long)_stext)) && ++ (start > ktla_ktva((unsigned long)_einittext) || ++ end <= ktla_ktva((unsigned long)_sinittext)) && ++ ++#ifdef CONFIG_ACPI_SLEEP ++ (start > (unsigned long)__va(acpi_wakeup_address) + 0x4000 || end <= (unsigned long)__va(acpi_wakeup_address)) && ++#endif ++ ++ (start > (unsigned long)__va(0xfffff) || end <= (unsigned long)__va(0xc0000))) ++ return 0; ++ return 1; + } + + /* +@@ -246,9 +246,10 @@ kernel_physical_mapping_init(unsigned lo + unsigned long last_map_addr = end; + unsigned long start_pfn, end_pfn; + pgd_t *pgd_base = swapper_pg_dir; +- int pgd_idx, pmd_idx, pte_ofs; ++ unsigned int pgd_idx, pmd_idx, pte_ofs; + unsigned long pfn; + pgd_t *pgd; ++ pud_t *pud; + pmd_t *pmd; + pte_t *pte; + unsigned pages_2m, pages_4k; +@@ -281,8 +282,13 @@ repeat: + pfn = start_pfn; + pgd_idx = pgd_index((pfn<<PAGE_SHIFT) + PAGE_OFFSET); + pgd = pgd_base + pgd_idx; +- for (; pgd_idx < PTRS_PER_PGD; pgd++, pgd_idx++) { +- pmd = one_md_table_init(pgd); ++ for (; pgd_idx < PTRS_PER_PGD && pfn < max_low_pfn; pgd++, pgd_idx++) { ++ pud = pud_offset(pgd, 0); ++ pmd = pmd_offset(pud, 0); ++ ++#ifdef CONFIG_X86_PAE ++ paravirt_alloc_pmd(&init_mm, __pa(pmd) >> PAGE_SHIFT); ++#endif + + if (pfn >= end_pfn) + continue; +@@ -294,14 +300,13 @@ repeat: + #endif + for (; pmd_idx < PTRS_PER_PMD && pfn < end_pfn; + pmd++, pmd_idx++) { +- unsigned int addr = pfn * PAGE_SIZE + PAGE_OFFSET; ++ unsigned long address = pfn * PAGE_SIZE + PAGE_OFFSET; + + /* + * Map with big pages if possible, otherwise + * create normal page tables: + */ + if (use_pse) { +- unsigned int addr2; + pgprot_t prot = PAGE_KERNEL_LARGE; + /* + * first pass will use the same initial +@@ -311,11 +316,7 @@ repeat: + __pgprot(PTE_IDENT_ATTR | + _PAGE_PSE); + +- addr2 = (pfn + PTRS_PER_PTE-1) * PAGE_SIZE + +- PAGE_OFFSET + PAGE_SIZE-1; +- +- if (is_kernel_text(addr) || +- is_kernel_text(addr2)) ++ if (is_kernel_text(address, address + PMD_SIZE)) + prot = PAGE_KERNEL_LARGE_EXEC; + + pages_2m++; +@@ -332,7 +333,7 @@ repeat: + pte_ofs = pte_index((pfn<<PAGE_SHIFT) + PAGE_OFFSET); + pte += pte_ofs; + for (; pte_ofs < PTRS_PER_PTE && pfn < end_pfn; +- pte++, pfn++, pte_ofs++, addr += PAGE_SIZE) { ++ pte++, pfn++, pte_ofs++, address += PAGE_SIZE) { + pgprot_t prot = PAGE_KERNEL; + /* + * first pass will use the same initial +@@ -340,7 +341,7 @@ repeat: + */ + pgprot_t init_prot = __pgprot(PTE_IDENT_ATTR); + +- if (is_kernel_text(addr)) ++ if (is_kernel_text(address, address + PAGE_SIZE)) + prot = PAGE_KERNEL_EXEC; + + pages_4k++; +@@ -472,7 +473,7 @@ void __init native_pagetable_setup_start + + pud = pud_offset(pgd, va); + pmd = pmd_offset(pud, va); +- if (!pmd_present(*pmd)) ++ if (!pmd_present(*pmd) || pmd_huge(*pmd)) + break; + + pte = pte_offset_kernel(pmd, va); +@@ -524,12 +525,10 @@ void __init early_ioremap_page_table_ran + + static void __init pagetable_init(void) + { +- pgd_t *pgd_base = swapper_pg_dir; +- +- permanent_kmaps_init(pgd_base); ++ permanent_kmaps_init(swapper_pg_dir); + } + +-pteval_t __supported_pte_mask __read_mostly = ~(_PAGE_NX | _PAGE_GLOBAL | _PAGE_IOMAP); ++pteval_t __supported_pte_mask __read_only = ~(_PAGE_NX | _PAGE_GLOBAL | _PAGE_IOMAP); + EXPORT_SYMBOL_GPL(__supported_pte_mask); + + /* user-defined highmem size */ +@@ -755,6 +754,12 @@ void __init mem_init(void) + + pci_iommu_alloc(); + ++#ifdef CONFIG_PAX_PER_CPU_PGD ++ clone_pgd_range(get_cpu_pgd(0) + KERNEL_PGD_BOUNDARY, ++ swapper_pg_dir + KERNEL_PGD_BOUNDARY, ++ KERNEL_PGD_PTRS); ++#endif ++ + #ifdef CONFIG_FLATMEM + BUG_ON(!mem_map); + #endif +@@ -772,7 +777,7 @@ void __init mem_init(void) + set_highmem_pages_init(); + + codesize = (unsigned long) &_etext - (unsigned long) &_text; +- datasize = (unsigned long) &_edata - (unsigned long) &_etext; ++ datasize = (unsigned long) &_edata - (unsigned long) &_sdata; + initsize = (unsigned long) &__init_end - (unsigned long) &__init_begin; + + printk(KERN_INFO "Memory: %luk/%luk available (%dk kernel code, " +@@ -813,10 +818,10 @@ void __init mem_init(void) + ((unsigned long)&__init_end - + (unsigned long)&__init_begin) >> 10, + +- (unsigned long)&_etext, (unsigned long)&_edata, +- ((unsigned long)&_edata - (unsigned long)&_etext) >> 10, ++ (unsigned long)&_sdata, (unsigned long)&_edata, ++ ((unsigned long)&_edata - (unsigned long)&_sdata) >> 10, + +- (unsigned long)&_text, (unsigned long)&_etext, ++ ktla_ktva((unsigned long)&_text), ktla_ktva((unsigned long)&_etext), + ((unsigned long)&_etext - (unsigned long)&_text) >> 10); + + /* +@@ -894,6 +899,7 @@ void set_kernel_text_rw(void) + if (!kernel_set_to_readonly) + return; + ++ start = ktla_ktva(start); + pr_debug("Set kernel text: %lx - %lx for read write\n", + start, start+size); + +@@ -908,6 +914,7 @@ void set_kernel_text_ro(void) + if (!kernel_set_to_readonly) + return; + ++ start = ktla_ktva(start); + pr_debug("Set kernel text: %lx - %lx for read only\n", + start, start+size); + +@@ -936,6 +943,7 @@ void mark_rodata_ro(void) + unsigned long start = PFN_ALIGN(_text); + unsigned long size = PFN_ALIGN(_etext) - start; + ++ start = ktla_ktva(start); + set_pages_ro(virt_to_page(start), size >> PAGE_SHIFT); + printk(KERN_INFO "Write protecting the kernel text: %luk\n", + size >> 10); +diff -urNp linux-2.6.38.3/arch/x86/mm/init_64.c linux-2.6.38.3/arch/x86/mm/init_64.c +--- linux-2.6.38.3/arch/x86/mm/init_64.c 2011-04-17 16:53:01.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/mm/init_64.c 2011-04-17 16:53:16.000000000 -0400 +@@ -72,7 +72,7 @@ early_param("gbpages", parse_direct_gbpa + * around without checking the pgd every time. + */ + +-pteval_t __supported_pte_mask __read_mostly = ~_PAGE_IOMAP; ++pteval_t __supported_pte_mask __read_only = ~(_PAGE_NX | _PAGE_IOMAP); + EXPORT_SYMBOL_GPL(__supported_pte_mask); + + int force_personality32; +@@ -105,12 +105,22 @@ void sync_global_pgds(unsigned long star + + for (address = start; address <= end; address += PGDIR_SIZE) { + const pgd_t *pgd_ref = pgd_offset_k(address); ++ ++#ifdef CONFIG_PAX_PER_CPU_PGD ++ unsigned long cpu; ++#else + struct page *page; ++#endif + + if (pgd_none(*pgd_ref)) + continue; + + spin_lock(&pgd_lock); ++ ++#ifdef CONFIG_PAX_PER_CPU_PGD ++ for (cpu = 0; cpu < NR_CPUS; ++cpu) { ++ pgd_t *pgd = pgd_offset_cpu(cpu, address); ++#else + list_for_each_entry(page, &pgd_list, lru) { + pgd_t *pgd; + spinlock_t *pgt_lock; +@@ -119,6 +129,7 @@ void sync_global_pgds(unsigned long star + /* the pgt_lock only for Xen */ + pgt_lock = &pgd_page_get_mm(page)->page_table_lock; + spin_lock(pgt_lock); ++#endif + + if (pgd_none(*pgd)) + set_pgd(pgd, *pgd_ref); +@@ -126,7 +137,10 @@ void sync_global_pgds(unsigned long star + BUG_ON(pgd_page_vaddr(*pgd) + != pgd_page_vaddr(*pgd_ref)); + ++#ifndef CONFIG_PAX_PER_CPU_PGD + spin_unlock(pgt_lock); ++#endif ++ + } + spin_unlock(&pgd_lock); + } +@@ -200,7 +214,9 @@ void set_pte_vaddr_pud(pud_t *pud_page, + pmd = fill_pmd(pud, vaddr); + pte = fill_pte(pmd, vaddr); + ++ pax_open_kernel(); + set_pte(pte, new_pte); ++ pax_close_kernel(); + + /* + * It's enough to flush this one mapping. +@@ -259,14 +275,12 @@ static void __init __init_extra_mapping( + pgd = pgd_offset_k((unsigned long)__va(phys)); + if (pgd_none(*pgd)) { + pud = (pud_t *) spp_getpage(); +- set_pgd(pgd, __pgd(__pa(pud) | _KERNPG_TABLE | +- _PAGE_USER)); ++ set_pgd(pgd, __pgd(__pa(pud) | _PAGE_TABLE)); + } + pud = pud_offset(pgd, (unsigned long)__va(phys)); + if (pud_none(*pud)) { + pmd = (pmd_t *) spp_getpage(); +- set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE | +- _PAGE_USER)); ++ set_pud(pud, __pud(__pa(pmd) | _PAGE_TABLE)); + } + pmd = pmd_offset(pud, phys); + BUG_ON(!pmd_none(*pmd)); +@@ -706,6 +720,12 @@ void __init mem_init(void) + + pci_iommu_alloc(); + ++#ifdef CONFIG_PAX_PER_CPU_PGD ++ clone_pgd_range(get_cpu_pgd(0) + KERNEL_PGD_BOUNDARY, ++ swapper_pg_dir + KERNEL_PGD_BOUNDARY, ++ KERNEL_PGD_PTRS); ++#endif ++ + /* clear_bss() already clear the empty_zero_page */ + + reservedpages = 0; +@@ -866,8 +886,8 @@ int kern_addr_valid(unsigned long addr) + static struct vm_area_struct gate_vma = { + .vm_start = VSYSCALL_START, + .vm_end = VSYSCALL_START + (VSYSCALL_MAPPED_PAGES * PAGE_SIZE), +- .vm_page_prot = PAGE_READONLY_EXEC, +- .vm_flags = VM_READ | VM_EXEC ++ .vm_page_prot = PAGE_READONLY, ++ .vm_flags = VM_READ + }; + + struct vm_area_struct *get_gate_vma(struct task_struct *tsk) +@@ -901,7 +921,7 @@ int in_gate_area_no_task(unsigned long a + + const char *arch_vma_name(struct vm_area_struct *vma) + { +- if (vma->vm_mm && vma->vm_start == (long)vma->vm_mm->context.vdso) ++ if (vma->vm_mm && vma->vm_start == vma->vm_mm->context.vdso) + return "[vdso]"; + if (vma == &gate_vma) + return "[vsyscall]"; +diff -urNp linux-2.6.38.3/arch/x86/mm/init.c linux-2.6.38.3/arch/x86/mm/init.c +--- linux-2.6.38.3/arch/x86/mm/init.c 2011-04-17 16:53:01.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/mm/init.c 2011-04-17 16:53:16.000000000 -0400 +@@ -72,11 +72,7 @@ static void __init find_early_table_spac + * cause a hotspot and fill up ZONE_DMA. The page tables + * need roughly 0.5KB per GB. + */ +-#ifdef CONFIG_X86_32 +- start = 0x7000; +-#else +- start = 0x8000; +-#endif ++ start = 0x100000; + base = memblock_find_in_range(start, max_pfn_mapped<<PAGE_SHIFT, + tables, PAGE_SIZE); + if (base == MEMBLOCK_ERROR) +@@ -323,7 +319,16 @@ unsigned long __init_refok init_memory_m + */ + int devmem_is_allowed(unsigned long pagenr) + { +- if (pagenr <= 256) ++#ifndef CONFIG_GRKERNSEC_KMEM ++ if (!pagenr) ++ return 1; ++#ifdef CONFIG_VM86 ++ if (pagenr < (ISA_START_ADDRESS >> PAGE_SHIFT)) ++ return 1; ++#endif ++#endif ++ ++ if ((ISA_START_ADDRESS >> PAGE_SHIFT) <= pagenr && pagenr < (ISA_END_ADDRESS >> PAGE_SHIFT)) + return 1; + if (iomem_is_exclusive(pagenr << PAGE_SHIFT)) + return 0; +@@ -383,6 +388,86 @@ void free_init_pages(char *what, unsigne + + void free_initmem(void) + { ++ ++#ifdef CONFIG_PAX_KERNEXEC ++#ifdef CONFIG_X86_32 ++ /* PaX: limit KERNEL_CS to actual size */ ++ unsigned long addr, limit; ++ struct desc_struct d; ++ int cpu; ++ ++ limit = paravirt_enabled() ? ktva_ktla(0xffffffff) : (unsigned long)&_etext; ++ limit = (limit - 1UL) >> PAGE_SHIFT; ++ ++ memset(__LOAD_PHYSICAL_ADDR + PAGE_OFFSET, POISON_FREE_INITMEM, PAGE_SIZE); ++ for (cpu = 0; cpu < NR_CPUS; cpu++) { ++ pack_descriptor(&d, get_desc_base(&get_cpu_gdt_table(cpu)[GDT_ENTRY_KERNEL_CS]), limit, 0x9B, 0xC); ++ write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_KERNEL_CS, &d, DESCTYPE_S); ++ } ++ ++ /* PaX: make KERNEL_CS read-only */ ++ addr = PFN_ALIGN(ktla_ktva((unsigned long)&_text)); ++ if (!paravirt_enabled()) ++ set_memory_ro(addr, (PFN_ALIGN(_sdata) - addr) >> PAGE_SHIFT); ++/* ++ for (addr = ktla_ktva((unsigned long)&_text); addr < (unsigned long)&_sdata; addr += PMD_SIZE) { ++ pgd = pgd_offset_k(addr); ++ pud = pud_offset(pgd, addr); ++ pmd = pmd_offset(pud, addr); ++ set_pmd(pmd, __pmd(pmd_val(*pmd) & ~_PAGE_RW)); ++ } ++*/ ++#ifdef CONFIG_X86_PAE ++ set_memory_nx(PFN_ALIGN(__init_begin), (PFN_ALIGN(__init_end) - PFN_ALIGN(__init_begin)) >> PAGE_SHIFT); ++/* ++ for (addr = (unsigned long)&__init_begin; addr < (unsigned long)&__init_end; addr += PMD_SIZE) { ++ pgd = pgd_offset_k(addr); ++ pud = pud_offset(pgd, addr); ++ pmd = pmd_offset(pud, addr); ++ set_pmd(pmd, __pmd(pmd_val(*pmd) | (_PAGE_NX & __supported_pte_mask))); ++ } ++*/ ++#endif ++ ++#ifdef CONFIG_MODULES ++ set_memory_4k((unsigned long)MODULES_EXEC_VADDR, (MODULES_EXEC_END - MODULES_EXEC_VADDR) >> PAGE_SHIFT); ++#endif ++ ++#else ++ pgd_t *pgd; ++ pud_t *pud; ++ pmd_t *pmd; ++ unsigned long addr, end; ++ ++ /* PaX: make kernel code/rodata read-only, rest non-executable */ ++ for (addr = __START_KERNEL_map; addr < __START_KERNEL_map + KERNEL_IMAGE_SIZE; addr += PMD_SIZE) { ++ pgd = pgd_offset_k(addr); ++ pud = pud_offset(pgd, addr); ++ pmd = pmd_offset(pud, addr); ++ if (!pmd_present(*pmd)) ++ continue; ++ if ((unsigned long)_text <= addr && addr < (unsigned long)_sdata) ++ set_pmd(pmd, __pmd(pmd_val(*pmd) & ~_PAGE_RW)); ++ else ++ set_pmd(pmd, __pmd(pmd_val(*pmd) | (_PAGE_NX & __supported_pte_mask))); ++ } ++ ++ addr = (unsigned long)__va(__pa(__START_KERNEL_map)); ++ end = addr + KERNEL_IMAGE_SIZE; ++ for (; addr < end; addr += PMD_SIZE) { ++ pgd = pgd_offset_k(addr); ++ pud = pud_offset(pgd, addr); ++ pmd = pmd_offset(pud, addr); ++ if (!pmd_present(*pmd)) ++ continue; ++ if ((unsigned long)__va(__pa(_text)) <= addr && addr < (unsigned long)__va(__pa(_sdata))) ++ set_pmd(pmd, __pmd(pmd_val(*pmd) & ~_PAGE_RW)); ++ } ++#endif ++ ++ flush_tlb_all(); ++#endif ++ + free_init_pages("unused kernel memory", + (unsigned long)(&__init_begin), + (unsigned long)(&__init_end)); +diff -urNp linux-2.6.38.3/arch/x86/mm/iomap_32.c linux-2.6.38.3/arch/x86/mm/iomap_32.c +--- linux-2.6.38.3/arch/x86/mm/iomap_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/mm/iomap_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -64,7 +64,11 @@ void *kmap_atomic_prot_pfn(unsigned long + type = kmap_atomic_idx_push(); + idx = type + KM_TYPE_NR * smp_processor_id(); + vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx); ++ ++ pax_open_kernel(); + set_pte(kmap_pte - idx, pfn_pte(pfn, prot)); ++ pax_close_kernel(); ++ + arch_flush_lazy_mmu_mode(); + + return (void *)vaddr; +diff -urNp linux-2.6.38.3/arch/x86/mm/ioremap.c linux-2.6.38.3/arch/x86/mm/ioremap.c +--- linux-2.6.38.3/arch/x86/mm/ioremap.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/mm/ioremap.c 2011-04-17 15:57:32.000000000 -0400 +@@ -104,7 +104,7 @@ static void __iomem *__ioremap_caller(re + for (pfn = phys_addr >> PAGE_SHIFT; pfn <= last_pfn; pfn++) { + int is_ram = page_is_ram(pfn); + +- if (is_ram && pfn_valid(pfn) && !PageReserved(pfn_to_page(pfn))) ++ if (is_ram && pfn_valid(pfn) && (pfn >= 0x100 || !PageReserved(pfn_to_page(pfn)))) + return NULL; + WARN_ON_ONCE(is_ram); + } +@@ -344,7 +344,7 @@ static int __init early_ioremap_debug_se + early_param("early_ioremap_debug", early_ioremap_debug_setup); + + static __initdata int after_paging_init; +-static pte_t bm_pte[PAGE_SIZE/sizeof(pte_t)] __page_aligned_bss; ++static pte_t bm_pte[PAGE_SIZE/sizeof(pte_t)] __read_only __aligned(PAGE_SIZE); + + static inline pmd_t * __init early_ioremap_pmd(unsigned long addr) + { +@@ -381,8 +381,7 @@ void __init early_ioremap_init(void) + slot_virt[i] = __fix_to_virt(FIX_BTMAP_BEGIN - NR_FIX_BTMAPS*i); + + pmd = early_ioremap_pmd(fix_to_virt(FIX_BTMAP_BEGIN)); +- memset(bm_pte, 0, sizeof(bm_pte)); +- pmd_populate_kernel(&init_mm, pmd, bm_pte); ++ pmd_populate_user(&init_mm, pmd, bm_pte); + + /* + * The boot-ioremap range spans multiple pmds, for which +diff -urNp linux-2.6.38.3/arch/x86/mm/kmemcheck/kmemcheck.c linux-2.6.38.3/arch/x86/mm/kmemcheck/kmemcheck.c +--- linux-2.6.38.3/arch/x86/mm/kmemcheck/kmemcheck.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/mm/kmemcheck/kmemcheck.c 2011-04-17 15:57:32.000000000 -0400 +@@ -622,9 +622,9 @@ bool kmemcheck_fault(struct pt_regs *reg + * memory (e.g. tracked pages)? For now, we need this to avoid + * invoking kmemcheck for PnP BIOS calls. + */ +- if (regs->flags & X86_VM_MASK) ++ if (v8086_mode(regs)) + return false; +- if (regs->cs != __KERNEL_CS) ++ if (regs->cs != __KERNEL_CS && regs->cs != __KERNEXEC_KERNEL_CS) + return false; + + pte = kmemcheck_pte_lookup(address); +diff -urNp linux-2.6.38.3/arch/x86/mm/mmap.c linux-2.6.38.3/arch/x86/mm/mmap.c +--- linux-2.6.38.3/arch/x86/mm/mmap.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/mm/mmap.c 2011-04-17 15:57:32.000000000 -0400 +@@ -49,7 +49,7 @@ static unsigned int stack_maxrandom_size + * Leave an at least ~128 MB hole with possible stack randomization. + */ + #define MIN_GAP (128*1024*1024UL + stack_maxrandom_size()) +-#define MAX_GAP (TASK_SIZE/6*5) ++#define MAX_GAP (pax_task_size/6*5) + + /* + * True on X86_32 or when emulating IA32 on X86_64 +@@ -94,27 +94,40 @@ static unsigned long mmap_rnd(void) + return rnd << PAGE_SHIFT; + } + +-static unsigned long mmap_base(void) ++static unsigned long mmap_base(struct mm_struct *mm) + { + unsigned long gap = rlimit(RLIMIT_STACK); ++ unsigned long pax_task_size = TASK_SIZE; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (mm->pax_flags & MF_PAX_SEGMEXEC) ++ pax_task_size = SEGMEXEC_TASK_SIZE; ++#endif + + if (gap < MIN_GAP) + gap = MIN_GAP; + else if (gap > MAX_GAP) + gap = MAX_GAP; + +- return PAGE_ALIGN(TASK_SIZE - gap - mmap_rnd()); ++ return PAGE_ALIGN(pax_task_size - gap - mmap_rnd()); + } + + /* + * Bottom-up (legacy) layout on X86_32 did not support randomization, X86_64 + * does, but not when emulating X86_32 + */ +-static unsigned long mmap_legacy_base(void) ++static unsigned long mmap_legacy_base(struct mm_struct *mm) + { +- if (mmap_is_ia32()) ++ if (mmap_is_ia32()) { ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (mm->pax_flags & MF_PAX_SEGMEXEC) ++ return SEGMEXEC_TASK_UNMAPPED_BASE; ++ else ++#endif ++ + return TASK_UNMAPPED_BASE; +- else ++ } else + return TASK_UNMAPPED_BASE + mmap_rnd(); + } + +@@ -125,11 +138,23 @@ static unsigned long mmap_legacy_base(vo + void arch_pick_mmap_layout(struct mm_struct *mm) + { + if (mmap_is_legacy()) { +- mm->mmap_base = mmap_legacy_base(); ++ mm->mmap_base = mmap_legacy_base(mm); ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base += mm->delta_mmap; ++#endif ++ + mm->get_unmapped_area = arch_get_unmapped_area; + mm->unmap_area = arch_unmap_area; + } else { +- mm->mmap_base = mmap_base(); ++ mm->mmap_base = mmap_base(mm); ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base -= mm->delta_mmap + mm->delta_stack; ++#endif ++ + mm->get_unmapped_area = arch_get_unmapped_area_topdown; + mm->unmap_area = arch_unmap_area_topdown; + } +diff -urNp linux-2.6.38.3/arch/x86/mm/numa_32.c linux-2.6.38.3/arch/x86/mm/numa_32.c +--- linux-2.6.38.3/arch/x86/mm/numa_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/mm/numa_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -99,7 +99,6 @@ unsigned long node_memmap_size_bytes(int + } + #endif + +-extern unsigned long find_max_low_pfn(void); + extern unsigned long highend_pfn, highstart_pfn; + + #define LARGE_PAGE_BYTES (PTRS_PER_PTE * PAGE_SIZE) +diff -urNp linux-2.6.38.3/arch/x86/mm/pageattr.c linux-2.6.38.3/arch/x86/mm/pageattr.c +--- linux-2.6.38.3/arch/x86/mm/pageattr.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/mm/pageattr.c 2011-04-17 15:57:32.000000000 -0400 +@@ -261,7 +261,7 @@ static inline pgprot_t static_protection + */ + #ifdef CONFIG_PCI_BIOS + if (pcibios_enabled && within(pfn, BIOS_BEGIN >> PAGE_SHIFT, BIOS_END >> PAGE_SHIFT)) +- pgprot_val(forbidden) |= _PAGE_NX; ++ pgprot_val(forbidden) |= _PAGE_NX & __supported_pte_mask; + #endif + + /* +@@ -269,9 +269,10 @@ static inline pgprot_t static_protection + * Does not cover __inittext since that is gone later on. On + * 64bit we do not enforce !NX on the low mapping + */ +- if (within(address, (unsigned long)_text, (unsigned long)_etext)) +- pgprot_val(forbidden) |= _PAGE_NX; ++ if (within(address, ktla_ktva((unsigned long)_text), ktla_ktva((unsigned long)_etext))) ++ pgprot_val(forbidden) |= _PAGE_NX & __supported_pte_mask; + ++#ifdef CONFIG_DEBUG_RODATA + /* + * The .rodata section needs to be read-only. Using the pfn + * catches all aliases. +@@ -279,6 +280,7 @@ static inline pgprot_t static_protection + if (within(pfn, __pa((unsigned long)__start_rodata) >> PAGE_SHIFT, + __pa((unsigned long)__end_rodata) >> PAGE_SHIFT)) + pgprot_val(forbidden) |= _PAGE_RW; ++#endif + + #if defined(CONFIG_X86_64) && defined(CONFIG_DEBUG_RODATA) + /* +@@ -317,6 +319,13 @@ static inline pgprot_t static_protection + } + #endif + ++#ifdef CONFIG_PAX_KERNEXEC ++ if (within(pfn, __pa((unsigned long)&_text), __pa((unsigned long)&_sdata))) { ++ pgprot_val(forbidden) |= _PAGE_RW; ++ pgprot_val(forbidden) |= _PAGE_NX & __supported_pte_mask; ++ } ++#endif ++ + prot = __pgprot(pgprot_val(prot) & ~pgprot_val(forbidden)); + + return prot; +@@ -369,23 +378,37 @@ EXPORT_SYMBOL_GPL(lookup_address); + static void __set_pmd_pte(pte_t *kpte, unsigned long address, pte_t pte) + { + /* change init_mm */ ++ pax_open_kernel(); + set_pte_atomic(kpte, pte); ++ + #ifdef CONFIG_X86_32 + if (!SHARED_KERNEL_PMD) { ++ ++#ifdef CONFIG_PAX_PER_CPU_PGD ++ unsigned long cpu; ++#else + struct page *page; ++#endif + ++#ifdef CONFIG_PAX_PER_CPU_PGD ++ for (cpu = 0; cpu < NR_CPUS; ++cpu) { ++ pgd_t *pgd = get_cpu_pgd(cpu); ++#else + list_for_each_entry(page, &pgd_list, lru) { +- pgd_t *pgd; ++ pgd_t *pgd = (pgd_t *)page_address(page); ++#endif ++ + pud_t *pud; + pmd_t *pmd; + +- pgd = (pgd_t *)page_address(page) + pgd_index(address); ++ pgd += pgd_index(address); + pud = pud_offset(pgd, address); + pmd = pmd_offset(pud, address); + set_pte_atomic((pte_t *)pmd, pte); + } + } + #endif ++ pax_close_kernel(); + } + + static int +diff -urNp linux-2.6.38.3/arch/x86/mm/pageattr-test.c linux-2.6.38.3/arch/x86/mm/pageattr-test.c +--- linux-2.6.38.3/arch/x86/mm/pageattr-test.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/mm/pageattr-test.c 2011-04-17 15:57:32.000000000 -0400 +@@ -36,7 +36,7 @@ enum { + + static int pte_testbit(pte_t pte) + { +- return pte_flags(pte) & _PAGE_UNUSED1; ++ return pte_flags(pte) & _PAGE_CPA_TEST; + } + + struct split_state { +diff -urNp linux-2.6.38.3/arch/x86/mm/pat.c linux-2.6.38.3/arch/x86/mm/pat.c +--- linux-2.6.38.3/arch/x86/mm/pat.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/mm/pat.c 2011-04-17 15:57:32.000000000 -0400 +@@ -361,7 +361,7 @@ int free_memtype(u64 start, u64 end) + + if (!entry) { + printk(KERN_INFO "%s:%d freeing invalid memtype %Lx-%Lx\n", +- current->comm, current->pid, start, end); ++ current->comm, task_pid_nr(current), start, end); + return -EINVAL; + } + +@@ -492,8 +492,8 @@ static inline int range_is_allowed(unsig + while (cursor < to) { + if (!devmem_is_allowed(pfn)) { + printk(KERN_INFO +- "Program %s tried to access /dev/mem between %Lx->%Lx.\n", +- current->comm, from, to); ++ "Program %s tried to access /dev/mem between %Lx->%Lx (%Lx).\n", ++ current->comm, from, to, cursor); + return 0; + } + cursor += PAGE_SIZE; +@@ -557,7 +557,7 @@ int kernel_map_sync_memtype(u64 base, un + printk(KERN_INFO + "%s:%d ioremap_change_attr failed %s " + "for %Lx-%Lx\n", +- current->comm, current->pid, ++ current->comm, task_pid_nr(current), + cattr_name(flags), + base, (unsigned long long)(base + size)); + return -EINVAL; +@@ -593,7 +593,7 @@ static int reserve_pfn_range(u64 paddr, + if (want_flags != flags) { + printk(KERN_WARNING + "%s:%d map pfn RAM range req %s for %Lx-%Lx, got %s\n", +- current->comm, current->pid, ++ current->comm, task_pid_nr(current), + cattr_name(want_flags), + (unsigned long long)paddr, + (unsigned long long)(paddr + size), +@@ -615,7 +615,7 @@ static int reserve_pfn_range(u64 paddr, + free_memtype(paddr, paddr + size); + printk(KERN_ERR "%s:%d map pfn expected mapping type %s" + " for %Lx-%Lx, got %s\n", +- current->comm, current->pid, ++ current->comm, task_pid_nr(current), + cattr_name(want_flags), + (unsigned long long)paddr, + (unsigned long long)(paddr + size), +diff -urNp linux-2.6.38.3/arch/x86/mm/pgtable_32.c linux-2.6.38.3/arch/x86/mm/pgtable_32.c +--- linux-2.6.38.3/arch/x86/mm/pgtable_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/mm/pgtable_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -48,10 +48,13 @@ void set_pte_vaddr(unsigned long vaddr, + return; + } + pte = pte_offset_kernel(pmd, vaddr); ++ ++ pax_open_kernel(); + if (pte_val(pteval)) + set_pte_at(&init_mm, vaddr, pte, pteval); + else + pte_clear(&init_mm, vaddr, pte); ++ pax_close_kernel(); + + /* + * It's enough to flush this one mapping. +diff -urNp linux-2.6.38.3/arch/x86/mm/pgtable.c linux-2.6.38.3/arch/x86/mm/pgtable.c +--- linux-2.6.38.3/arch/x86/mm/pgtable.c 2011-03-23 17:20:06.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/mm/pgtable.c 2011-04-17 15:57:32.000000000 -0400 +@@ -84,9 +84,58 @@ static inline void pgd_list_del(pgd_t *p + list_del(&page->lru); + } + +-#define UNSHARED_PTRS_PER_PGD \ +- (SHARED_KERNEL_PMD ? KERNEL_PGD_BOUNDARY : PTRS_PER_PGD) ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) ++pgdval_t clone_pgd_mask __read_only = ~_PAGE_PRESENT; + ++void __shadow_user_pgds(pgd_t *dst, const pgd_t *src, int count) ++{ ++ while (count--) ++ *dst++ = __pgd((pgd_val(*src++) | (_PAGE_NX & __supported_pte_mask)) & ~_PAGE_USER); ++} ++#endif ++ ++#ifdef CONFIG_PAX_PER_CPU_PGD ++void __clone_user_pgds(pgd_t *dst, const pgd_t *src, int count) ++{ ++ while (count--) ++ ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) ++ *dst++ = __pgd(pgd_val(*src++) & clone_pgd_mask); ++#else ++ *dst++ = *src++; ++#endif ++ ++} ++#endif ++ ++#ifdef CONFIG_PAX_PER_CPU_PGD ++static inline void pgd_ctor(struct mm_struct *mm, pgd_t *pgd) {} ++static inline void pgd_dtor(pgd_t *pgd) {} ++#ifdef CONFIG_X86_64 ++#define pxd_t pud_t ++#define pyd_t pgd_t ++#define paravirt_release_pxd(pfn) paravirt_release_pud(pfn) ++#define pxd_free(mm, pud) pud_free((mm), (pud)) ++#define pyd_populate(mm, pgd, pud) pgd_populate((mm), (pgd), (pud)) ++#define pyd_offset(mm ,address) pgd_offset((mm), (address)) ++#define PYD_SIZE PGDIR_SIZE ++#else ++#define pxd_t pmd_t ++#define pyd_t pud_t ++#define paravirt_release_pxd(pfn) paravirt_release_pmd(pfn) ++#define pxd_free(mm, pud) pmd_free((mm), (pud)) ++#define pyd_populate(mm, pgd, pud) pud_populate((mm), (pgd), (pud)) ++#define pyd_offset(mm ,address) pud_offset((mm), (address)) ++#define PYD_SIZE PUD_SIZE ++#endif ++#else ++#define pxd_t pmd_t ++#define pyd_t pud_t ++#define paravirt_release_pxd(pfn) paravirt_release_pmd(pfn) ++#define pxd_free(mm, pmd) pmd_free((mm), (pmd)) ++#define pyd_populate(mm, pud, pmd) pud_populate((mm), (pud), (pmd)) ++#define pyd_offset(mm ,address) pud_offset((mm), (address)) ++#define PYD_SIZE PUD_SIZE + + static void pgd_set_mm(pgd_t *pgd, struct mm_struct *mm) + { +@@ -128,6 +177,7 @@ static void pgd_dtor(pgd_t *pgd) + pgd_list_del(pgd); + spin_unlock(&pgd_lock); + } ++#endif + + /* + * List of all pgd's needed for non-PAE so it can invalidate entries +@@ -140,7 +190,7 @@ static void pgd_dtor(pgd_t *pgd) + * -- wli + */ + +-#ifdef CONFIG_X86_PAE ++#if defined(CONFIG_X86_32) && defined(CONFIG_X86_PAE) + /* + * In PAE mode, we need to do a cr3 reload (=tlb flush) when + * updating the top-level pagetable entries to guarantee the +@@ -152,7 +202,7 @@ static void pgd_dtor(pgd_t *pgd) + * not shared between pagetables (!SHARED_KERNEL_PMDS), we allocate + * and initialize the kernel pmds here. + */ +-#define PREALLOCATED_PMDS UNSHARED_PTRS_PER_PGD ++#define PREALLOCATED_PXDS (SHARED_KERNEL_PMD ? KERNEL_PGD_BOUNDARY : PTRS_PER_PGD) + + void pud_populate(struct mm_struct *mm, pud_t *pudp, pmd_t *pmd) + { +@@ -170,36 +220,38 @@ void pud_populate(struct mm_struct *mm, + */ + flush_tlb_mm(mm); + } ++#elif defined(CONFIG_X86_64) && defined(CONFIG_PAX_PER_CPU_PGD) ++#define PREALLOCATED_PXDS USER_PGD_PTRS + #else /* !CONFIG_X86_PAE */ + + /* No need to prepopulate any pagetable entries in non-PAE modes. */ +-#define PREALLOCATED_PMDS 0 ++#define PREALLOCATED_PXDS 0 + + #endif /* CONFIG_X86_PAE */ + +-static void free_pmds(pmd_t *pmds[]) ++static void free_pxds(pxd_t *pxds[]) + { + int i; + +- for(i = 0; i < PREALLOCATED_PMDS; i++) +- if (pmds[i]) +- free_page((unsigned long)pmds[i]); ++ for(i = 0; i < PREALLOCATED_PXDS; i++) ++ if (pxds[i]) ++ free_page((unsigned long)pxds[i]); + } + +-static int preallocate_pmds(pmd_t *pmds[]) ++static int preallocate_pxds(pxd_t *pxds[]) + { + int i; + bool failed = false; + +- for(i = 0; i < PREALLOCATED_PMDS; i++) { +- pmd_t *pmd = (pmd_t *)__get_free_page(PGALLOC_GFP); +- if (pmd == NULL) ++ for(i = 0; i < PREALLOCATED_PXDS; i++) { ++ pxd_t *pxd = (pxd_t *)__get_free_page(PGALLOC_GFP); ++ if (pxd == NULL) + failed = true; +- pmds[i] = pmd; ++ pxds[i] = pxd; + } + + if (failed) { +- free_pmds(pmds); ++ free_pxds(pxds); + return -ENOMEM; + } + +@@ -212,51 +264,55 @@ static int preallocate_pmds(pmd_t *pmds[ + * preallocate which never got a corresponding vma will need to be + * freed manually. + */ +-static void pgd_mop_up_pmds(struct mm_struct *mm, pgd_t *pgdp) ++static void pgd_mop_up_pxds(struct mm_struct *mm, pgd_t *pgdp) + { + int i; + +- for(i = 0; i < PREALLOCATED_PMDS; i++) { ++ for(i = 0; i < PREALLOCATED_PXDS; i++) { + pgd_t pgd = pgdp[i]; + + if (pgd_val(pgd) != 0) { +- pmd_t *pmd = (pmd_t *)pgd_page_vaddr(pgd); ++ pxd_t *pxd = (pxd_t *)pgd_page_vaddr(pgd); + +- pgdp[i] = native_make_pgd(0); ++ set_pgd(pgdp + i, native_make_pgd(0)); + +- paravirt_release_pmd(pgd_val(pgd) >> PAGE_SHIFT); +- pmd_free(mm, pmd); ++ paravirt_release_pxd(pgd_val(pgd) >> PAGE_SHIFT); ++ pxd_free(mm, pxd); + } + } + } + +-static void pgd_prepopulate_pmd(struct mm_struct *mm, pgd_t *pgd, pmd_t *pmds[]) ++static void pgd_prepopulate_pxd(struct mm_struct *mm, pgd_t *pgd, pxd_t *pxds[]) + { +- pud_t *pud; ++ pyd_t *pyd; + unsigned long addr; + int i; + +- if (PREALLOCATED_PMDS == 0) /* Work around gcc-3.4.x bug */ ++ if (PREALLOCATED_PXDS == 0) /* Work around gcc-3.4.x bug */ + return; + +- pud = pud_offset(pgd, 0); ++#ifdef CONFIG_X86_64 ++ pyd = pyd_offset(mm, 0L); ++#else ++ pyd = pyd_offset(pgd, 0L); ++#endif + +- for (addr = i = 0; i < PREALLOCATED_PMDS; +- i++, pud++, addr += PUD_SIZE) { +- pmd_t *pmd = pmds[i]; ++ for (addr = i = 0; i < PREALLOCATED_PXDS; ++ i++, pyd++, addr += PYD_SIZE) { ++ pxd_t *pxd = pxds[i]; + + if (i >= KERNEL_PGD_BOUNDARY) +- memcpy(pmd, (pmd_t *)pgd_page_vaddr(swapper_pg_dir[i]), +- sizeof(pmd_t) * PTRS_PER_PMD); ++ memcpy(pxd, (pxd_t *)pgd_page_vaddr(swapper_pg_dir[i]), ++ sizeof(pxd_t) * PTRS_PER_PMD); + +- pud_populate(mm, pud, pmd); ++ pyd_populate(mm, pyd, pxd); + } + } + + pgd_t *pgd_alloc(struct mm_struct *mm) + { + pgd_t *pgd; +- pmd_t *pmds[PREALLOCATED_PMDS]; ++ pxd_t *pxds[PREALLOCATED_PXDS]; + + pgd = (pgd_t *)__get_free_page(PGALLOC_GFP); + +@@ -265,11 +321,11 @@ pgd_t *pgd_alloc(struct mm_struct *mm) + + mm->pgd = pgd; + +- if (preallocate_pmds(pmds) != 0) ++ if (preallocate_pxds(pxds) != 0) + goto out_free_pgd; + + if (paravirt_pgd_alloc(mm) != 0) +- goto out_free_pmds; ++ goto out_free_pxds; + + /* + * Make sure that pre-populating the pmds is atomic with +@@ -279,14 +335,14 @@ pgd_t *pgd_alloc(struct mm_struct *mm) + spin_lock(&pgd_lock); + + pgd_ctor(mm, pgd); +- pgd_prepopulate_pmd(mm, pgd, pmds); ++ pgd_prepopulate_pxd(mm, pgd, pxds); + + spin_unlock(&pgd_lock); + + return pgd; + +-out_free_pmds: +- free_pmds(pmds); ++out_free_pxds: ++ free_pxds(pxds); + out_free_pgd: + free_page((unsigned long)pgd); + out: +@@ -295,7 +351,7 @@ out: + + void pgd_free(struct mm_struct *mm, pgd_t *pgd) + { +- pgd_mop_up_pmds(mm, pgd); ++ pgd_mop_up_pxds(mm, pgd); + pgd_dtor(pgd); + paravirt_pgd_free(mm, pgd); + free_page((unsigned long)pgd); +diff -urNp linux-2.6.38.3/arch/x86/mm/setup_nx.c linux-2.6.38.3/arch/x86/mm/setup_nx.c +--- linux-2.6.38.3/arch/x86/mm/setup_nx.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/mm/setup_nx.c 2011-04-17 15:57:32.000000000 -0400 +@@ -5,8 +5,10 @@ + #include <asm/pgtable.h> + #include <asm/proto.h> + ++#if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE) + static int disable_nx __cpuinitdata; + ++#ifndef CONFIG_PAX_PAGEEXEC + /* + * noexec = on|off + * +@@ -28,12 +30,17 @@ static int __init noexec_setup(char *str + return 0; + } + early_param("noexec", noexec_setup); ++#endif ++ ++#endif + + void __cpuinit x86_configure_nx(void) + { ++#if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE) + if (cpu_has_nx && !disable_nx) + __supported_pte_mask |= _PAGE_NX; + else ++#endif + __supported_pte_mask &= ~_PAGE_NX; + } + +diff -urNp linux-2.6.38.3/arch/x86/mm/tlb.c linux-2.6.38.3/arch/x86/mm/tlb.c +--- linux-2.6.38.3/arch/x86/mm/tlb.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/mm/tlb.c 2011-04-17 15:57:32.000000000 -0400 +@@ -14,7 +14,7 @@ + #include <asm/uv/uv.h> + + DEFINE_PER_CPU_SHARED_ALIGNED(struct tlb_state, cpu_tlbstate) +- = { &init_mm, 0, }; ++ = { &init_mm, 0 }; + + /* + * Smarter SMP flushing macros. +@@ -65,7 +65,11 @@ void leave_mm(int cpu) + BUG(); + cpumask_clear_cpu(cpu, + mm_cpumask(percpu_read(cpu_tlbstate.active_mm))); ++ ++#ifndef CONFIG_PAX_PER_CPU_PGD + load_cr3(swapper_pg_dir); ++#endif ++ + } + EXPORT_SYMBOL_GPL(leave_mm); + +diff -urNp linux-2.6.38.3/arch/x86/oprofile/backtrace.c linux-2.6.38.3/arch/x86/oprofile/backtrace.c +--- linux-2.6.38.3/arch/x86/oprofile/backtrace.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/oprofile/backtrace.c 2011-04-17 15:57:32.000000000 -0400 +@@ -57,7 +57,7 @@ dump_user_backtrace_32(struct stack_fram + struct stack_frame_ia32 *fp; + + /* Also check accessibility of one struct frame_head beyond */ +- if (!access_ok(VERIFY_READ, head, sizeof(bufhead))) ++ if (!__access_ok(VERIFY_READ, head, sizeof(bufhead))) + return NULL; + if (__copy_from_user_inatomic(bufhead, head, sizeof(bufhead))) + return NULL; +@@ -123,7 +123,7 @@ x86_backtrace(struct pt_regs * const reg + { + struct stack_frame *head = (struct stack_frame *)frame_pointer(regs); + +- if (!user_mode_vm(regs)) { ++ if (!user_mode(regs)) { + unsigned long stack = kernel_stack_pointer(regs); + if (depth) + dump_trace(NULL, regs, (unsigned long *)stack, +diff -urNp linux-2.6.38.3/arch/x86/oprofile/op_model_p4.c linux-2.6.38.3/arch/x86/oprofile/op_model_p4.c +--- linux-2.6.38.3/arch/x86/oprofile/op_model_p4.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/oprofile/op_model_p4.c 2011-04-17 15:57:32.000000000 -0400 +@@ -50,7 +50,7 @@ static inline void setup_num_counters(vo + #endif + } + +-static int inline addr_increment(void) ++static inline int addr_increment(void) + { + #ifdef CONFIG_SMP + return smp_num_siblings == 2 ? 2 : 1; +diff -urNp linux-2.6.38.3/arch/x86/pci/ce4100.c linux-2.6.38.3/arch/x86/pci/ce4100.c +--- linux-2.6.38.3/arch/x86/pci/ce4100.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/pci/ce4100.c 2011-04-17 15:57:32.000000000 -0400 +@@ -302,7 +302,7 @@ static int ce4100_conf_write(unsigned in + return pci_direct_conf1.write(seg, bus, devfn, reg, len, value); + } + +-struct pci_raw_ops ce4100_pci_conf = { ++const struct pci_raw_ops ce4100_pci_conf = { + .read = ce4100_conf_read, + .write = ce4100_conf_write, + }; +diff -urNp linux-2.6.38.3/arch/x86/pci/common.c linux-2.6.38.3/arch/x86/pci/common.c +--- linux-2.6.38.3/arch/x86/pci/common.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/pci/common.c 2011-04-17 15:57:32.000000000 -0400 +@@ -33,8 +33,8 @@ int noioapicreroute = 1; + int pcibios_last_bus = -1; + unsigned long pirq_table_addr; + struct pci_bus *pci_root_bus; +-struct pci_raw_ops *raw_pci_ops; +-struct pci_raw_ops *raw_pci_ext_ops; ++const struct pci_raw_ops *raw_pci_ops; ++const struct pci_raw_ops *raw_pci_ext_ops; + + int raw_pci_read(unsigned int domain, unsigned int bus, unsigned int devfn, + int reg, int len, u32 *val) +@@ -423,7 +423,7 @@ static const struct dmi_system_id __devi + DMI_MATCH(DMI_PRODUCT_NAME, "ProLiant DL585 G2"), + }, + }, +- {} ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL} + }; + + void __init dmi_check_pciprobe(void) +diff -urNp linux-2.6.38.3/arch/x86/pci/direct.c linux-2.6.38.3/arch/x86/pci/direct.c +--- linux-2.6.38.3/arch/x86/pci/direct.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/pci/direct.c 2011-04-17 15:57:32.000000000 -0400 +@@ -79,7 +79,7 @@ static int pci_conf1_write(unsigned int + + #undef PCI_CONF1_ADDRESS + +-struct pci_raw_ops pci_direct_conf1 = { ++const struct pci_raw_ops pci_direct_conf1 = { + .read = pci_conf1_read, + .write = pci_conf1_write, + }; +@@ -173,7 +173,7 @@ static int pci_conf2_write(unsigned int + + #undef PCI_CONF2_ADDRESS + +-struct pci_raw_ops pci_direct_conf2 = { ++const struct pci_raw_ops pci_direct_conf2 = { + .read = pci_conf2_read, + .write = pci_conf2_write, + }; +@@ -189,7 +189,7 @@ struct pci_raw_ops pci_direct_conf2 = { + * This should be close to trivial, but it isn't, because there are buggy + * chipsets (yes, you guessed it, by Intel and Compaq) that have no class ID. + */ +-static int __init pci_sanity_check(struct pci_raw_ops *o) ++static int __init pci_sanity_check(const struct pci_raw_ops *o) + { + u32 x = 0; + int year, devfn; +diff -urNp linux-2.6.38.3/arch/x86/pci/fixup.c linux-2.6.38.3/arch/x86/pci/fixup.c +--- linux-2.6.38.3/arch/x86/pci/fixup.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/pci/fixup.c 2011-04-17 15:57:32.000000000 -0400 +@@ -364,7 +364,7 @@ static const struct dmi_system_id __devi + DMI_MATCH(DMI_PRODUCT_NAME, "MS-6702E"), + }, + }, +- {} ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } + }; + + /* +@@ -435,7 +435,7 @@ static const struct dmi_system_id __devi + DMI_MATCH(DMI_PRODUCT_VERSION, "PSA40U"), + }, + }, +- { } ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } + }; + + static void __devinit pci_pre_fixup_toshiba_ohci1394(struct pci_dev *dev) +diff -urNp linux-2.6.38.3/arch/x86/pci/irq.c linux-2.6.38.3/arch/x86/pci/irq.c +--- linux-2.6.38.3/arch/x86/pci/irq.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/pci/irq.c 2011-04-17 15:57:32.000000000 -0400 +@@ -542,7 +542,7 @@ static __init int intel_router_probe(str + static struct pci_device_id __initdata pirq_440gx[] = { + { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82443GX_0) }, + { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82443GX_2) }, +- { }, ++ { PCI_DEVICE(0, 0) } + }; + + /* 440GX has a proprietary PIRQ router -- don't use it */ +@@ -1115,7 +1115,7 @@ static struct dmi_system_id __initdata p + DMI_MATCH(DMI_PRODUCT_NAME, "TravelMate 360"), + }, + }, +- { } ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } + }; + + void __init pcibios_irq_init(void) +diff -urNp linux-2.6.38.3/arch/x86/pci/mmconfig_32.c linux-2.6.38.3/arch/x86/pci/mmconfig_32.c +--- linux-2.6.38.3/arch/x86/pci/mmconfig_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/pci/mmconfig_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -117,7 +117,7 @@ static int pci_mmcfg_write(unsigned int + return 0; + } + +-static struct pci_raw_ops pci_mmcfg = { ++static const struct pci_raw_ops pci_mmcfg = { + .read = pci_mmcfg_read, + .write = pci_mmcfg_write, + }; +diff -urNp linux-2.6.38.3/arch/x86/pci/mmconfig_64.c linux-2.6.38.3/arch/x86/pci/mmconfig_64.c +--- linux-2.6.38.3/arch/x86/pci/mmconfig_64.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/pci/mmconfig_64.c 2011-04-17 15:57:32.000000000 -0400 +@@ -81,7 +81,7 @@ static int pci_mmcfg_write(unsigned int + return 0; + } + +-static struct pci_raw_ops pci_mmcfg = { ++static const struct pci_raw_ops pci_mmcfg = { + .read = pci_mmcfg_read, + .write = pci_mmcfg_write, + }; +diff -urNp linux-2.6.38.3/arch/x86/pci/numaq_32.c linux-2.6.38.3/arch/x86/pci/numaq_32.c +--- linux-2.6.38.3/arch/x86/pci/numaq_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/pci/numaq_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -108,7 +108,7 @@ static int pci_conf1_mq_write(unsigned i + + #undef PCI_CONF1_MQ_ADDRESS + +-static struct pci_raw_ops pci_direct_conf1_mq = { ++static const struct pci_raw_ops pci_direct_conf1_mq = { + .read = pci_conf1_mq_read, + .write = pci_conf1_mq_write + }; +diff -urNp linux-2.6.38.3/arch/x86/pci/olpc.c linux-2.6.38.3/arch/x86/pci/olpc.c +--- linux-2.6.38.3/arch/x86/pci/olpc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/pci/olpc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -297,7 +297,7 @@ static int pci_olpc_write(unsigned int s + return 0; + } + +-static struct pci_raw_ops pci_olpc_conf = { ++static const struct pci_raw_ops pci_olpc_conf = { + .read = pci_olpc_read, + .write = pci_olpc_write, + }; +diff -urNp linux-2.6.38.3/arch/x86/pci/pcbios.c linux-2.6.38.3/arch/x86/pci/pcbios.c +--- linux-2.6.38.3/arch/x86/pci/pcbios.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/pci/pcbios.c 2011-04-17 15:57:32.000000000 -0400 +@@ -79,50 +79,93 @@ union bios32 { + static struct { + unsigned long address; + unsigned short segment; +-} bios32_indirect = { 0, __KERNEL_CS }; ++} bios32_indirect __read_only = { 0, __PCIBIOS_CS }; + + /* + * Returns the entry point for the given service, NULL on error + */ + +-static unsigned long bios32_service(unsigned long service) ++static unsigned long __devinit bios32_service(unsigned long service) + { + unsigned char return_code; /* %al */ + unsigned long address; /* %ebx */ + unsigned long length; /* %ecx */ + unsigned long entry; /* %edx */ + unsigned long flags; ++ struct desc_struct d, *gdt; + + local_irq_save(flags); +- __asm__("lcall *(%%edi); cld" ++ ++ gdt = get_cpu_gdt_table(smp_processor_id()); ++ ++ pack_descriptor(&d, 0UL, 0xFFFFFUL, 0x9B, 0xC); ++ write_gdt_entry(gdt, GDT_ENTRY_PCIBIOS_CS, &d, DESCTYPE_S); ++ pack_descriptor(&d, 0UL, 0xFFFFFUL, 0x93, 0xC); ++ write_gdt_entry(gdt, GDT_ENTRY_PCIBIOS_DS, &d, DESCTYPE_S); ++ ++ __asm__("movw %w7, %%ds; lcall *(%%edi); push %%ss; pop %%ds; cld" + : "=a" (return_code), + "=b" (address), + "=c" (length), + "=d" (entry) + : "0" (service), + "1" (0), +- "D" (&bios32_indirect)); ++ "D" (&bios32_indirect), ++ "r"(__PCIBIOS_DS) ++ : "memory"); ++ ++ pax_open_kernel(); ++ gdt[GDT_ENTRY_PCIBIOS_CS].a = 0; ++ gdt[GDT_ENTRY_PCIBIOS_CS].b = 0; ++ gdt[GDT_ENTRY_PCIBIOS_DS].a = 0; ++ gdt[GDT_ENTRY_PCIBIOS_DS].b = 0; ++ pax_close_kernel(); ++ + local_irq_restore(flags); + + switch (return_code) { +- case 0: +- return address + entry; +- case 0x80: /* Not present */ +- printk(KERN_WARNING "bios32_service(0x%lx): not present\n", service); +- return 0; +- default: /* Shouldn't happen */ +- printk(KERN_WARNING "bios32_service(0x%lx): returned 0x%x -- BIOS bug!\n", +- service, return_code); ++ case 0: { ++ int cpu; ++ unsigned char flags; ++ ++ printk(KERN_INFO "bios32_service: base:%08lx length:%08lx entry:%08lx\n", address, length, entry); ++ if (address >= 0xFFFF0 || length > 0x100000 - address || length <= entry) { ++ printk(KERN_WARNING "bios32_service: not valid\n"); + return 0; ++ } ++ address = address + PAGE_OFFSET; ++ length += 16UL; /* some BIOSs underreport this... */ ++ flags = 4; ++ if (length >= 64*1024*1024) { ++ length >>= PAGE_SHIFT; ++ flags |= 8; ++ } ++ ++ for (cpu = 0; cpu < NR_CPUS; cpu++) { ++ gdt = get_cpu_gdt_table(cpu); ++ pack_descriptor(&d, address, length, 0x9b, flags); ++ write_gdt_entry(gdt, GDT_ENTRY_PCIBIOS_CS, &d, DESCTYPE_S); ++ pack_descriptor(&d, address, length, 0x93, flags); ++ write_gdt_entry(gdt, GDT_ENTRY_PCIBIOS_DS, &d, DESCTYPE_S); ++ } ++ return entry; ++ } ++ case 0x80: /* Not present */ ++ printk(KERN_WARNING "bios32_service(0x%lx): not present\n", service); ++ return 0; ++ default: /* Shouldn't happen */ ++ printk(KERN_WARNING "bios32_service(0x%lx): returned 0x%x -- BIOS bug!\n", ++ service, return_code); ++ return 0; + } + } + + static struct { + unsigned long address; + unsigned short segment; +-} pci_indirect = { 0, __KERNEL_CS }; ++} pci_indirect __read_only = { 0, __PCIBIOS_CS }; + +-static int pci_bios_present; ++static int pci_bios_present __read_only; + + static int __devinit check_pcibios(void) + { +@@ -131,11 +174,13 @@ static int __devinit check_pcibios(void) + unsigned long flags, pcibios_entry; + + if ((pcibios_entry = bios32_service(PCI_SERVICE))) { +- pci_indirect.address = pcibios_entry + PAGE_OFFSET; ++ pci_indirect.address = pcibios_entry; + + local_irq_save(flags); +- __asm__( +- "lcall *(%%edi); cld\n\t" ++ __asm__("movw %w6, %%ds\n\t" ++ "lcall *%%ss:(%%edi); cld\n\t" ++ "push %%ss\n\t" ++ "pop %%ds\n\t" + "jc 1f\n\t" + "xor %%ah, %%ah\n" + "1:" +@@ -144,7 +189,8 @@ static int __devinit check_pcibios(void) + "=b" (ebx), + "=c" (ecx) + : "1" (PCIBIOS_PCI_BIOS_PRESENT), +- "D" (&pci_indirect) ++ "D" (&pci_indirect), ++ "r" (__PCIBIOS_DS) + : "memory"); + local_irq_restore(flags); + +@@ -188,7 +234,10 @@ static int pci_bios_read(unsigned int se + + switch (len) { + case 1: +- __asm__("lcall *(%%esi); cld\n\t" ++ __asm__("movw %w6, %%ds\n\t" ++ "lcall *%%ss:(%%esi); cld\n\t" ++ "push %%ss\n\t" ++ "pop %%ds\n\t" + "jc 1f\n\t" + "xor %%ah, %%ah\n" + "1:" +@@ -197,7 +246,8 @@ static int pci_bios_read(unsigned int se + : "1" (PCIBIOS_READ_CONFIG_BYTE), + "b" (bx), + "D" ((long)reg), +- "S" (&pci_indirect)); ++ "S" (&pci_indirect), ++ "r" (__PCIBIOS_DS)); + /* + * Zero-extend the result beyond 8 bits, do not trust the + * BIOS having done it: +@@ -205,7 +255,10 @@ static int pci_bios_read(unsigned int se + *value &= 0xff; + break; + case 2: +- __asm__("lcall *(%%esi); cld\n\t" ++ __asm__("movw %w6, %%ds\n\t" ++ "lcall *%%ss:(%%esi); cld\n\t" ++ "push %%ss\n\t" ++ "pop %%ds\n\t" + "jc 1f\n\t" + "xor %%ah, %%ah\n" + "1:" +@@ -214,7 +267,8 @@ static int pci_bios_read(unsigned int se + : "1" (PCIBIOS_READ_CONFIG_WORD), + "b" (bx), + "D" ((long)reg), +- "S" (&pci_indirect)); ++ "S" (&pci_indirect), ++ "r" (__PCIBIOS_DS)); + /* + * Zero-extend the result beyond 16 bits, do not trust the + * BIOS having done it: +@@ -222,7 +276,10 @@ static int pci_bios_read(unsigned int se + *value &= 0xffff; + break; + case 4: +- __asm__("lcall *(%%esi); cld\n\t" ++ __asm__("movw %w6, %%ds\n\t" ++ "lcall *%%ss:(%%esi); cld\n\t" ++ "push %%ss\n\t" ++ "pop %%ds\n\t" + "jc 1f\n\t" + "xor %%ah, %%ah\n" + "1:" +@@ -231,7 +288,8 @@ static int pci_bios_read(unsigned int se + : "1" (PCIBIOS_READ_CONFIG_DWORD), + "b" (bx), + "D" ((long)reg), +- "S" (&pci_indirect)); ++ "S" (&pci_indirect), ++ "r" (__PCIBIOS_DS)); + break; + } + +@@ -254,7 +312,10 @@ static int pci_bios_write(unsigned int s + + switch (len) { + case 1: +- __asm__("lcall *(%%esi); cld\n\t" ++ __asm__("movw %w6, %%ds\n\t" ++ "lcall *%%ss:(%%esi); cld\n\t" ++ "push %%ss\n\t" ++ "pop %%ds\n\t" + "jc 1f\n\t" + "xor %%ah, %%ah\n" + "1:" +@@ -263,10 +324,14 @@ static int pci_bios_write(unsigned int s + "c" (value), + "b" (bx), + "D" ((long)reg), +- "S" (&pci_indirect)); ++ "S" (&pci_indirect), ++ "r" (__PCIBIOS_DS)); + break; + case 2: +- __asm__("lcall *(%%esi); cld\n\t" ++ __asm__("movw %w6, %%ds\n\t" ++ "lcall *%%ss:(%%esi); cld\n\t" ++ "push %%ss\n\t" ++ "pop %%ds\n\t" + "jc 1f\n\t" + "xor %%ah, %%ah\n" + "1:" +@@ -275,10 +340,14 @@ static int pci_bios_write(unsigned int s + "c" (value), + "b" (bx), + "D" ((long)reg), +- "S" (&pci_indirect)); ++ "S" (&pci_indirect), ++ "r" (__PCIBIOS_DS)); + break; + case 4: +- __asm__("lcall *(%%esi); cld\n\t" ++ __asm__("movw %w6, %%ds\n\t" ++ "lcall *%%ss:(%%esi); cld\n\t" ++ "push %%ss\n\t" ++ "pop %%ds\n\t" + "jc 1f\n\t" + "xor %%ah, %%ah\n" + "1:" +@@ -287,7 +356,8 @@ static int pci_bios_write(unsigned int s + "c" (value), + "b" (bx), + "D" ((long)reg), +- "S" (&pci_indirect)); ++ "S" (&pci_indirect), ++ "r" (__PCIBIOS_DS)); + break; + } + +@@ -301,7 +371,7 @@ static int pci_bios_write(unsigned int s + * Function table for BIOS32 access + */ + +-static struct pci_raw_ops pci_bios_access = { ++static const struct pci_raw_ops pci_bios_access = { + .read = pci_bios_read, + .write = pci_bios_write + }; +@@ -310,7 +380,7 @@ static struct pci_raw_ops pci_bios_acces + * Try to find PCI BIOS. + */ + +-static struct pci_raw_ops * __devinit pci_find_bios(void) ++static const struct pci_raw_ops * __devinit pci_find_bios(void) + { + union bios32 *check; + unsigned char sum; +@@ -392,10 +462,13 @@ struct irq_routing_table * pcibios_get_i + + DBG("PCI: Fetching IRQ routing table... "); + __asm__("push %%es\n\t" ++ "movw %w8, %%ds\n\t" + "push %%ds\n\t" + "pop %%es\n\t" +- "lcall *(%%esi); cld\n\t" ++ "lcall *%%ss:(%%esi); cld\n\t" + "pop %%es\n\t" ++ "push %%ss\n\t" ++ "pop %%ds\n" + "jc 1f\n\t" + "xor %%ah, %%ah\n" + "1:" +@@ -406,7 +479,8 @@ struct irq_routing_table * pcibios_get_i + "1" (0), + "D" ((long) &opt), + "S" (&pci_indirect), +- "m" (opt) ++ "m" (opt), ++ "r" (__PCIBIOS_DS) + : "memory"); + DBG("OK ret=%d, size=%d, map=%x\n", ret, opt.size, map); + if (ret & 0xff00) +@@ -430,7 +504,10 @@ int pcibios_set_irq_routing(struct pci_d + { + int ret; + +- __asm__("lcall *(%%esi); cld\n\t" ++ __asm__("movw %w5, %%ds\n\t" ++ "lcall *%%ss:(%%esi); cld\n\t" ++ "push %%ss\n\t" ++ "pop %%ds\n" + "jc 1f\n\t" + "xor %%ah, %%ah\n" + "1:" +@@ -438,7 +515,8 @@ int pcibios_set_irq_routing(struct pci_d + : "0" (PCIBIOS_SET_PCI_HW_INT), + "b" ((dev->bus->number << 8) | dev->devfn), + "c" ((irq << 8) | (pin + 10)), +- "S" (&pci_indirect)); ++ "S" (&pci_indirect), ++ "r" (__PCIBIOS_DS)); + return !(ret & 0xff00); + } + EXPORT_SYMBOL(pcibios_set_irq_routing); +diff -urNp linux-2.6.38.3/arch/x86/platform/efi/efi_32.c linux-2.6.38.3/arch/x86/platform/efi/efi_32.c +--- linux-2.6.38.3/arch/x86/platform/efi/efi_32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/platform/efi/efi_32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -38,70 +38,37 @@ + */ + + static unsigned long efi_rt_eflags; +-static pgd_t efi_bak_pg_dir_pointer[2]; ++static pgd_t __initdata efi_bak_pg_dir_pointer[KERNEL_PGD_PTRS]; + +-void efi_call_phys_prelog(void) ++void __init efi_call_phys_prelog(void) + { +- unsigned long cr4; +- unsigned long temp; + struct desc_ptr gdt_descr; + + local_irq_save(efi_rt_eflags); + +- /* +- * If I don't have PAE, I should just duplicate two entries in page +- * directory. If I have PAE, I just need to duplicate one entry in +- * page directory. +- */ +- cr4 = read_cr4_safe(); +- +- if (cr4 & X86_CR4_PAE) { +- efi_bak_pg_dir_pointer[0].pgd = +- swapper_pg_dir[pgd_index(0)].pgd; +- swapper_pg_dir[0].pgd = +- swapper_pg_dir[pgd_index(PAGE_OFFSET)].pgd; +- } else { +- efi_bak_pg_dir_pointer[0].pgd = +- swapper_pg_dir[pgd_index(0)].pgd; +- efi_bak_pg_dir_pointer[1].pgd = +- swapper_pg_dir[pgd_index(0x400000)].pgd; +- swapper_pg_dir[pgd_index(0)].pgd = +- swapper_pg_dir[pgd_index(PAGE_OFFSET)].pgd; +- temp = PAGE_OFFSET + 0x400000; +- swapper_pg_dir[pgd_index(0x400000)].pgd = +- swapper_pg_dir[pgd_index(temp)].pgd; +- } ++ clone_pgd_range(efi_bak_pg_dir_pointer, swapper_pg_dir, KERNEL_PGD_PTRS); ++ clone_pgd_range(swapper_pg_dir, swapper_pg_dir + KERNEL_PGD_BOUNDARY, ++ min_t(unsigned long, KERNEL_PGD_PTRS, KERNEL_PGD_BOUNDARY)); + + /* + * After the lock is released, the original page table is restored. + */ + __flush_tlb_all(); + +- gdt_descr.address = __pa(get_cpu_gdt_table(0)); ++ gdt_descr.address = (struct desc_struct *)__pa(get_cpu_gdt_table(0)); + gdt_descr.size = GDT_SIZE - 1; + load_gdt(&gdt_descr); + } + +-void efi_call_phys_epilog(void) ++void __init efi_call_phys_epilog(void) + { +- unsigned long cr4; + struct desc_ptr gdt_descr; + +- gdt_descr.address = (unsigned long)get_cpu_gdt_table(0); ++ gdt_descr.address = get_cpu_gdt_table(0); + gdt_descr.size = GDT_SIZE - 1; + load_gdt(&gdt_descr); + +- cr4 = read_cr4_safe(); +- +- if (cr4 & X86_CR4_PAE) { +- swapper_pg_dir[pgd_index(0)].pgd = +- efi_bak_pg_dir_pointer[0].pgd; +- } else { +- swapper_pg_dir[pgd_index(0)].pgd = +- efi_bak_pg_dir_pointer[0].pgd; +- swapper_pg_dir[pgd_index(0x400000)].pgd = +- efi_bak_pg_dir_pointer[1].pgd; +- } ++ clone_pgd_range(swapper_pg_dir, efi_bak_pg_dir_pointer, KERNEL_PGD_PTRS); + + /* + * After the lock is released, the original page table is restored. +diff -urNp linux-2.6.38.3/arch/x86/platform/efi/efi_stub_32.S linux-2.6.38.3/arch/x86/platform/efi/efi_stub_32.S +--- linux-2.6.38.3/arch/x86/platform/efi/efi_stub_32.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/platform/efi/efi_stub_32.S 2011-04-17 15:57:32.000000000 -0400 +@@ -6,6 +6,7 @@ + */ + + #include <linux/linkage.h> ++#include <linux/init.h> + #include <asm/page_types.h> + + /* +@@ -20,7 +21,7 @@ + * service functions will comply with gcc calling convention, too. + */ + +-.text ++__INIT + ENTRY(efi_call_phys) + /* + * 0. The function can only be called in Linux kernel. So CS has been +@@ -36,9 +37,7 @@ ENTRY(efi_call_phys) + * The mapping of lower virtual memory has been created in prelog and + * epilog. + */ +- movl $1f, %edx +- subl $__PAGE_OFFSET, %edx +- jmp *%edx ++ jmp 1f-__PAGE_OFFSET + 1: + + /* +@@ -47,14 +46,8 @@ ENTRY(efi_call_phys) + * parameter 2, ..., param n. To make things easy, we save the return + * address of efi_call_phys in a global variable. + */ +- popl %edx +- movl %edx, saved_return_addr +- /* get the function pointer into ECX*/ +- popl %ecx +- movl %ecx, efi_rt_function_ptr +- movl $2f, %edx +- subl $__PAGE_OFFSET, %edx +- pushl %edx ++ popl (saved_return_addr) ++ popl (efi_rt_function_ptr) + + /* + * 3. Clear PG bit in %CR0. +@@ -73,9 +66,8 @@ ENTRY(efi_call_phys) + /* + * 5. Call the physical function. + */ +- jmp *%ecx ++ call *(efi_rt_function_ptr-__PAGE_OFFSET) + +-2: + /* + * 6. After EFI runtime service returns, control will return to + * following instruction. We'd better readjust stack pointer first. +@@ -88,35 +80,28 @@ ENTRY(efi_call_phys) + movl %cr0, %edx + orl $0x80000000, %edx + movl %edx, %cr0 +- jmp 1f +-1: ++ + /* + * 8. Now restore the virtual mode from flat mode by + * adding EIP with PAGE_OFFSET. + */ +- movl $1f, %edx +- jmp *%edx ++ jmp 1f+__PAGE_OFFSET + 1: + + /* + * 9. Balance the stack. And because EAX contain the return value, + * we'd better not clobber it. + */ +- leal efi_rt_function_ptr, %edx +- movl (%edx), %ecx +- pushl %ecx ++ pushl (efi_rt_function_ptr) + + /* +- * 10. Push the saved return address onto the stack and return. ++ * 10. Return to the saved return address. + */ +- leal saved_return_addr, %edx +- movl (%edx), %ecx +- pushl %ecx +- ret ++ jmpl *(saved_return_addr) + ENDPROC(efi_call_phys) + .previous + +-.data ++__INITDATA + saved_return_addr: + .long 0 + efi_rt_function_ptr: +diff -urNp linux-2.6.38.3/arch/x86/power/cpu.c linux-2.6.38.3/arch/x86/power/cpu.c +--- linux-2.6.38.3/arch/x86/power/cpu.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/power/cpu.c 2011-04-17 15:57:32.000000000 -0400 +@@ -130,7 +130,7 @@ static void do_fpu_end(void) + static void fix_processor_context(void) + { + int cpu = smp_processor_id(); +- struct tss_struct *t = &per_cpu(init_tss, cpu); ++ struct tss_struct *t = init_tss + cpu; + + set_tss_desc(cpu, t); /* + * This just modifies memory; should not be +@@ -140,7 +140,9 @@ static void fix_processor_context(void) + */ + + #ifdef CONFIG_X86_64 ++ pax_open_kernel(); + get_cpu_gdt_table(cpu)[GDT_ENTRY_TSS].type = 9; ++ pax_close_kernel(); + + syscall_init(); /* This sets MSR_*STAR and related */ + #endif +diff -urNp linux-2.6.38.3/arch/x86/vdso/Makefile linux-2.6.38.3/arch/x86/vdso/Makefile +--- linux-2.6.38.3/arch/x86/vdso/Makefile 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/vdso/Makefile 2011-04-17 15:57:32.000000000 -0400 +@@ -123,7 +123,7 @@ quiet_cmd_vdso = VDSO $@ + -Wl,-T,$(filter %.lds,$^) $(filter %.o,$^) && \ + sh $(srctree)/$(src)/checkundef.sh '$(NM)' '$@' + +-VDSO_LDFLAGS = -fPIC -shared $(call cc-ldoption, -Wl$(comma)--hash-style=sysv) ++VDSO_LDFLAGS = -fPIC -shared -Wl,--no-undefined $(call cc-ldoption, -Wl$(comma)--hash-style=sysv) + GCOV_PROFILE := n + + # +diff -urNp linux-2.6.38.3/arch/x86/vdso/vclock_gettime.c linux-2.6.38.3/arch/x86/vdso/vclock_gettime.c +--- linux-2.6.38.3/arch/x86/vdso/vclock_gettime.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/vdso/vclock_gettime.c 2011-04-17 15:57:32.000000000 -0400 +@@ -22,24 +22,48 @@ + #include <asm/hpet.h> + #include <asm/unistd.h> + #include <asm/io.h> ++#include <asm/fixmap.h> + #include "vextern.h" + + #define gtod vdso_vsyscall_gtod_data + ++notrace noinline long __vdso_fallback_time(long *t) ++{ ++ long secs; ++ asm volatile("syscall" ++ : "=a" (secs) ++ : "0" (__NR_time),"D" (t) : "r11", "cx", "memory"); ++ return secs; ++} ++ + notrace static long vdso_fallback_gettime(long clock, struct timespec *ts) + { + long ret; + asm("syscall" : "=a" (ret) : +- "0" (__NR_clock_gettime),"D" (clock), "S" (ts) : "memory"); ++ "0" (__NR_clock_gettime),"D" (clock), "S" (ts) : "r11", "cx", "memory"); + return ret; + } + ++notrace static inline cycle_t __vdso_vread_hpet(void) ++{ ++ return readl((const void __iomem *)fix_to_virt(VSYSCALL_HPET) + 0xf0); ++} ++ ++notrace static inline cycle_t __vdso_vread_tsc(void) ++{ ++ cycle_t ret = (cycle_t)vget_cycles(); ++ ++ return ret >= gtod->clock.cycle_last ? ret : gtod->clock.cycle_last; ++} ++ + notrace static inline long vgetns(void) + { + long v; +- cycles_t (*vread)(void); +- vread = gtod->clock.vread; +- v = (vread() - gtod->clock.cycle_last) & gtod->clock.mask; ++ if (gtod->clock.name[0] == 't' && gtod->clock.name[1] == 's' && gtod->clock.name[2] == 'c' && !gtod->clock.name[3]) ++ v = __vdso_vread_tsc(); ++ else ++ v = __vdso_vread_hpet(); ++ v = (v - gtod->clock.cycle_last) & gtod->clock.mask; + return (v * gtod->clock.mult) >> gtod->clock.shift; + } + +@@ -113,7 +137,9 @@ notrace static noinline int do_monotonic + + notrace int __vdso_clock_gettime(clockid_t clock, struct timespec *ts) + { +- if (likely(gtod->sysctl_enabled)) ++ if (likely(gtod->sysctl_enabled && ++ ((gtod->clock.name[0] == 'h' && gtod->clock.name[1] == 'p' && gtod->clock.name[2] == 'e' && gtod->clock.name[3] == 't' && !gtod->clock.name[4]) || ++ (gtod->clock.name[0] == 't' && gtod->clock.name[1] == 's' && gtod->clock.name[2] == 'c' && !gtod->clock.name[3])))) + switch (clock) { + case CLOCK_REALTIME: + if (likely(gtod->clock.vread)) +@@ -133,10 +159,20 @@ notrace int __vdso_clock_gettime(clockid + int clock_gettime(clockid_t, struct timespec *) + __attribute__((weak, alias("__vdso_clock_gettime"))); + +-notrace int __vdso_gettimeofday(struct timeval *tv, struct timezone *tz) ++notrace noinline int __vdso_fallback_gettimeofday(struct timeval *tv, struct timezone *tz) + { + long ret; +- if (likely(gtod->sysctl_enabled && gtod->clock.vread)) { ++ asm("syscall" : "=a" (ret) : ++ "0" (__NR_gettimeofday), "D" (tv), "S" (tz) : "r11", "cx", "memory"); ++ return ret; ++} ++ ++notrace int __vdso_gettimeofday(struct timeval *tv, struct timezone *tz) ++{ ++ if (likely(gtod->sysctl_enabled && ++ ((gtod->clock.name[0] == 'h' && gtod->clock.name[1] == 'p' && gtod->clock.name[2] == 'e' && gtod->clock.name[3] == 't' && !gtod->clock.name[4]) || ++ (gtod->clock.name[0] == 't' && gtod->clock.name[1] == 's' && gtod->clock.name[2] == 'c' && !gtod->clock.name[3])))) ++ { + if (likely(tv != NULL)) { + BUILD_BUG_ON(offsetof(struct timeval, tv_usec) != + offsetof(struct timespec, tv_nsec) || +@@ -151,9 +187,7 @@ notrace int __vdso_gettimeofday(struct t + } + return 0; + } +- asm("syscall" : "=a" (ret) : +- "0" (__NR_gettimeofday), "D" (tv), "S" (tz) : "memory"); +- return ret; ++ return __vdso_fallback_gettimeofday(tv, tz); + } + int gettimeofday(struct timeval *, struct timezone *) + __attribute__((weak, alias("__vdso_gettimeofday"))); +diff -urNp linux-2.6.38.3/arch/x86/vdso/vdso32-setup.c linux-2.6.38.3/arch/x86/vdso/vdso32-setup.c +--- linux-2.6.38.3/arch/x86/vdso/vdso32-setup.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/vdso/vdso32-setup.c 2011-04-17 15:57:32.000000000 -0400 +@@ -25,6 +25,7 @@ + #include <asm/tlbflush.h> + #include <asm/vdso.h> + #include <asm/proto.h> ++#include <asm/mman.h> + + enum { + VDSO_DISABLED = 0, +@@ -226,7 +227,7 @@ static inline void map_compat_vdso(int m + void enable_sep_cpu(void) + { + int cpu = get_cpu(); +- struct tss_struct *tss = &per_cpu(init_tss, cpu); ++ struct tss_struct *tss = init_tss + cpu; + + if (!boot_cpu_has(X86_FEATURE_SEP)) { + put_cpu(); +@@ -249,7 +250,7 @@ static int __init gate_vma_init(void) + gate_vma.vm_start = FIXADDR_USER_START; + gate_vma.vm_end = FIXADDR_USER_END; + gate_vma.vm_flags = VM_READ | VM_MAYREAD | VM_EXEC | VM_MAYEXEC; +- gate_vma.vm_page_prot = __P101; ++ gate_vma.vm_page_prot = vm_get_page_prot(gate_vma.vm_flags); + /* + * Make sure the vDSO gets into every core dump. + * Dumping its contents makes post-mortem fully interpretable later +@@ -331,14 +332,14 @@ int arch_setup_additional_pages(struct l + if (compat) + addr = VDSO_HIGH_BASE; + else { +- addr = get_unmapped_area(NULL, 0, PAGE_SIZE, 0, 0); ++ addr = get_unmapped_area(NULL, 0, PAGE_SIZE, 0, MAP_EXECUTABLE); + if (IS_ERR_VALUE(addr)) { + ret = addr; + goto up_fail; + } + } + +- current->mm->context.vdso = (void *)addr; ++ current->mm->context.vdso = addr; + + if (compat_uses_vma || !compat) { + /* +@@ -361,11 +362,11 @@ int arch_setup_additional_pages(struct l + } + + current_thread_info()->sysenter_return = +- VDSO32_SYMBOL(addr, SYSENTER_RETURN); ++ (__force void __user *)VDSO32_SYMBOL(addr, SYSENTER_RETURN); + + up_fail: + if (ret) +- current->mm->context.vdso = NULL; ++ current->mm->context.vdso = 0; + + up_write(&mm->mmap_sem); + +@@ -412,8 +413,14 @@ __initcall(ia32_binfmt_init); + + const char *arch_vma_name(struct vm_area_struct *vma) + { +- if (vma->vm_mm && vma->vm_start == (long)vma->vm_mm->context.vdso) ++ if (vma->vm_mm && vma->vm_start == vma->vm_mm->context.vdso) + return "[vdso]"; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (vma->vm_mm && vma->vm_mirror && vma->vm_mirror->vm_start == vma->vm_mm->context.vdso) ++ return "[vdso]"; ++#endif ++ + return NULL; + } + +@@ -422,7 +429,7 @@ struct vm_area_struct *get_gate_vma(stru + struct mm_struct *mm = tsk->mm; + + /* Check to see if this task was created in compat vdso mode */ +- if (mm && mm->context.vdso == (void *)VDSO_HIGH_BASE) ++ if (mm && mm->context.vdso == VDSO_HIGH_BASE) + return &gate_vma; + return NULL; + } +diff -urNp linux-2.6.38.3/arch/x86/vdso/vdso.lds.S linux-2.6.38.3/arch/x86/vdso/vdso.lds.S +--- linux-2.6.38.3/arch/x86/vdso/vdso.lds.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/vdso/vdso.lds.S 2011-04-17 15:57:32.000000000 -0400 +@@ -35,3 +35,9 @@ VDSO64_PRELINK = VDSO_PRELINK; + #define VEXTERN(x) VDSO64_ ## x = vdso_ ## x; + #include "vextern.h" + #undef VEXTERN ++ ++#define VEXTERN(x) VDSO64_ ## x = __vdso_ ## x; ++VEXTERN(fallback_gettimeofday) ++VEXTERN(fallback_time) ++VEXTERN(getcpu) ++#undef VEXTERN +diff -urNp linux-2.6.38.3/arch/x86/vdso/vextern.h linux-2.6.38.3/arch/x86/vdso/vextern.h +--- linux-2.6.38.3/arch/x86/vdso/vextern.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/vdso/vextern.h 2011-04-17 15:57:32.000000000 -0400 +@@ -11,6 +11,5 @@ + put into vextern.h and be referenced as a pointer with vdso prefix. + The main kernel later fills in the values. */ + +-VEXTERN(jiffies) + VEXTERN(vgetcpu_mode) + VEXTERN(vsyscall_gtod_data) +diff -urNp linux-2.6.38.3/arch/x86/vdso/vma.c linux-2.6.38.3/arch/x86/vdso/vma.c +--- linux-2.6.38.3/arch/x86/vdso/vma.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/vdso/vma.c 2011-04-17 15:57:32.000000000 -0400 +@@ -58,7 +58,7 @@ static int __init init_vdso_vars(void) + if (!vbase) + goto oom; + +- if (memcmp(vbase, "\177ELF", 4)) { ++ if (memcmp(vbase, ELFMAG, SELFMAG)) { + printk("VDSO: I'm broken; not ELF\n"); + vdso_enabled = 0; + } +@@ -118,7 +118,7 @@ int arch_setup_additional_pages(struct l + goto up_fail; + } + +- current->mm->context.vdso = (void *)addr; ++ current->mm->context.vdso = addr; + + ret = install_special_mapping(mm, addr, vdso_size, + VM_READ|VM_EXEC| +@@ -126,7 +126,7 @@ int arch_setup_additional_pages(struct l + VM_ALWAYSDUMP, + vdso_pages); + if (ret) { +- current->mm->context.vdso = NULL; ++ current->mm->context.vdso = 0; + goto up_fail; + } + +@@ -134,10 +134,3 @@ up_fail: + up_write(&mm->mmap_sem); + return ret; + } +- +-static __init int vdso_setup(char *s) +-{ +- vdso_enabled = simple_strtoul(s, NULL, 0); +- return 0; +-} +-__setup("vdso=", vdso_setup); +diff -urNp linux-2.6.38.3/arch/x86/xen/enlighten.c linux-2.6.38.3/arch/x86/xen/enlighten.c +--- linux-2.6.38.3/arch/x86/xen/enlighten.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/xen/enlighten.c 2011-04-17 15:57:32.000000000 -0400 +@@ -85,8 +85,6 @@ EXPORT_SYMBOL_GPL(xen_start_info); + + struct shared_info xen_dummy_shared_info; + +-void *xen_initial_gdt; +- + RESERVE_BRK(shared_info_page_brk, PAGE_SIZE); + __read_mostly int xen_have_vector_callback; + EXPORT_SYMBOL_GPL(xen_have_vector_callback); +@@ -1134,7 +1132,17 @@ asmlinkage void __init xen_start_kernel( + __userpte_alloc_gfp &= ~__GFP_HIGHMEM; + + /* Work out if we support NX */ +- x86_configure_nx(); ++#if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE) ++ if ((cpuid_eax(0x80000000) & 0xffff0000) == 0x80000000 && ++ (cpuid_edx(0x80000001) & (1U << (X86_FEATURE_NX & 31)))) { ++ unsigned l, h; ++ ++ __supported_pte_mask |= _PAGE_NX; ++ rdmsr(MSR_EFER, l, h); ++ l |= EFER_NX; ++ wrmsr(MSR_EFER, l, h); ++ } ++#endif + + xen_setup_features(); + +@@ -1165,13 +1173,6 @@ asmlinkage void __init xen_start_kernel( + + machine_ops = xen_machine_ops; + +- /* +- * The only reliable way to retain the initial address of the +- * percpu gdt_page is to remember it here, so we can go and +- * mark it RW later, when the initial percpu area is freed. +- */ +- xen_initial_gdt = &per_cpu(gdt_page, 0); +- + xen_smp_init(); + + #ifdef CONFIG_ACPI_NUMA +diff -urNp linux-2.6.38.3/arch/x86/xen/mmu.c linux-2.6.38.3/arch/x86/xen/mmu.c +--- linux-2.6.38.3/arch/x86/xen/mmu.c 2011-03-28 17:42:40.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/xen/mmu.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1721,6 +1721,8 @@ __init pgd_t *xen_setup_kernel_pagetable + convert_pfn_mfn(init_level4_pgt); + convert_pfn_mfn(level3_ident_pgt); + convert_pfn_mfn(level3_kernel_pgt); ++ convert_pfn_mfn(level3_vmalloc_pgt); ++ convert_pfn_mfn(level3_vmemmap_pgt); + + l3 = m2v(pgd[pgd_index(__START_KERNEL_map)].pgd); + l2 = m2v(l3[pud_index(__START_KERNEL_map)].pud); +@@ -1739,7 +1741,10 @@ __init pgd_t *xen_setup_kernel_pagetable + set_page_prot(init_level4_pgt, PAGE_KERNEL_RO); + set_page_prot(level3_ident_pgt, PAGE_KERNEL_RO); + set_page_prot(level3_kernel_pgt, PAGE_KERNEL_RO); ++ set_page_prot(level3_vmalloc_pgt, PAGE_KERNEL_RO); ++ set_page_prot(level3_vmemmap_pgt, PAGE_KERNEL_RO); + set_page_prot(level3_user_vsyscall, PAGE_KERNEL_RO); ++ set_page_prot(level2_vmemmap_pgt, PAGE_KERNEL_RO); + set_page_prot(level2_kernel_pgt, PAGE_KERNEL_RO); + set_page_prot(level2_fixmap_pgt, PAGE_KERNEL_RO); + +diff -urNp linux-2.6.38.3/arch/x86/xen/pci-swiotlb-xen.c linux-2.6.38.3/arch/x86/xen/pci-swiotlb-xen.c +--- linux-2.6.38.3/arch/x86/xen/pci-swiotlb-xen.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/xen/pci-swiotlb-xen.c 2011-04-17 15:57:32.000000000 -0400 +@@ -10,7 +10,7 @@ + + int xen_swiotlb __read_mostly; + +-static struct dma_map_ops xen_swiotlb_dma_ops = { ++static const struct dma_map_ops xen_swiotlb_dma_ops = { + .mapping_error = xen_swiotlb_dma_mapping_error, + .alloc_coherent = xen_swiotlb_alloc_coherent, + .free_coherent = xen_swiotlb_free_coherent, +diff -urNp linux-2.6.38.3/arch/x86/xen/smp.c linux-2.6.38.3/arch/x86/xen/smp.c +--- linux-2.6.38.3/arch/x86/xen/smp.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/xen/smp.c 2011-04-17 15:57:32.000000000 -0400 +@@ -194,11 +194,6 @@ static void __init xen_smp_prepare_boot_ + { + BUG_ON(smp_processor_id() != 0); + native_smp_prepare_boot_cpu(); +- +- /* We've switched to the "real" per-cpu gdt, so make sure the +- old memory can be recycled */ +- make_lowmem_page_readwrite(xen_initial_gdt); +- + xen_filter_cpu_maps(); + xen_setup_vcpu_info_placement(); + } +@@ -259,12 +254,12 @@ cpu_initialize_context(unsigned int cpu, + gdt = get_cpu_gdt_table(cpu); + + ctxt->flags = VGCF_IN_KERNEL; +- ctxt->user_regs.ds = __USER_DS; +- ctxt->user_regs.es = __USER_DS; ++ ctxt->user_regs.ds = __KERNEL_DS; ++ ctxt->user_regs.es = __KERNEL_DS; + ctxt->user_regs.ss = __KERNEL_DS; + #ifdef CONFIG_X86_32 + ctxt->user_regs.fs = __KERNEL_PERCPU; +- ctxt->user_regs.gs = __KERNEL_STACK_CANARY; ++ savesegment(gs, ctxt->user_regs.gs); + #else + ctxt->gs_base_kernel = per_cpu_offset(cpu); + #endif +@@ -315,13 +310,12 @@ static int __cpuinit xen_cpu_up(unsigned + int rc; + + per_cpu(current_task, cpu) = idle; ++ per_cpu(current_tinfo, cpu) = &idle->tinfo; + #ifdef CONFIG_X86_32 + irq_ctx_init(cpu); + #else + clear_tsk_thread_flag(idle, TIF_FORK); +- per_cpu(kernel_stack, cpu) = +- (unsigned long)task_stack_page(idle) - +- KERNEL_STACK_OFFSET + THREAD_SIZE; ++ per_cpu(kernel_stack, cpu) = (unsigned long)task_stack_page(idle) - 8 + THREAD_SIZE; + #endif + xen_setup_runstate_info(cpu); + xen_setup_timer(cpu); +diff -urNp linux-2.6.38.3/arch/x86/xen/xen-head.S linux-2.6.38.3/arch/x86/xen/xen-head.S +--- linux-2.6.38.3/arch/x86/xen/xen-head.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/xen/xen-head.S 2011-04-17 15:57:32.000000000 -0400 +@@ -19,6 +19,17 @@ ENTRY(startup_xen) + #ifdef CONFIG_X86_32 + mov %esi,xen_start_info + mov $init_thread_union+THREAD_SIZE,%esp ++#ifdef CONFIG_SMP ++ movl $cpu_gdt_table,%edi ++ movl $__per_cpu_load,%eax ++ movw %ax,__KERNEL_PERCPU + 2(%edi) ++ rorl $16,%eax ++ movb %al,__KERNEL_PERCPU + 4(%edi) ++ movb %ah,__KERNEL_PERCPU + 7(%edi) ++ movl $__per_cpu_end - 1,%eax ++ subl $__per_cpu_start,%eax ++ movw %ax,__KERNEL_PERCPU + 0(%edi) ++#endif + #else + mov %rsi,xen_start_info + mov $init_thread_union+THREAD_SIZE,%rsp +diff -urNp linux-2.6.38.3/arch/x86/xen/xen-ops.h linux-2.6.38.3/arch/x86/xen/xen-ops.h +--- linux-2.6.38.3/arch/x86/xen/xen-ops.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/arch/x86/xen/xen-ops.h 2011-04-17 15:57:32.000000000 -0400 +@@ -10,8 +10,6 @@ + extern const char xen_hypervisor_callback[]; + extern const char xen_failsafe_callback[]; + +-extern void *xen_initial_gdt; +- + struct trap_info; + void xen_copy_trap_info(struct trap_info *traps); + +diff -urNp linux-2.6.38.3/block/blk-iopoll.c linux-2.6.38.3/block/blk-iopoll.c +--- linux-2.6.38.3/block/blk-iopoll.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/block/blk-iopoll.c 2011-04-17 15:57:32.000000000 -0400 +@@ -77,7 +77,7 @@ void blk_iopoll_complete(struct blk_iopo + } + EXPORT_SYMBOL(blk_iopoll_complete); + +-static void blk_iopoll_softirq(struct softirq_action *h) ++static void blk_iopoll_softirq(void) + { + struct list_head *list = &__get_cpu_var(blk_cpu_iopoll); + int rearm = 0, budget = blk_iopoll_budget; +diff -urNp linux-2.6.38.3/block/blk-map.c linux-2.6.38.3/block/blk-map.c +--- linux-2.6.38.3/block/blk-map.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/block/blk-map.c 2011-04-17 15:57:32.000000000 -0400 +@@ -301,7 +301,7 @@ int blk_rq_map_kern(struct request_queue + if (!len || !kbuf) + return -EINVAL; + +- do_copy = !blk_rq_aligned(q, addr, len) || object_is_on_stack(kbuf); ++ do_copy = !blk_rq_aligned(q, addr, len) || object_starts_on_stack(kbuf); + if (do_copy) + bio = bio_copy_kern(q, kbuf, len, gfp_mask, reading); + else +diff -urNp linux-2.6.38.3/block/blk-softirq.c linux-2.6.38.3/block/blk-softirq.c +--- linux-2.6.38.3/block/blk-softirq.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/block/blk-softirq.c 2011-04-17 15:57:32.000000000 -0400 +@@ -17,7 +17,7 @@ static DEFINE_PER_CPU(struct list_head, + * Softirq action handler - move entries to local list and loop over them + * while passing them to the queue registered handler. + */ +-static void blk_done_softirq(struct softirq_action *h) ++static void blk_done_softirq(void) + { + struct list_head *cpu_list, local_list; + +diff -urNp linux-2.6.38.3/block/bsg.c linux-2.6.38.3/block/bsg.c +--- linux-2.6.38.3/block/bsg.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/block/bsg.c 2011-04-17 15:57:32.000000000 -0400 +@@ -176,16 +176,24 @@ static int blk_fill_sgv4_hdr_rq(struct r + struct sg_io_v4 *hdr, struct bsg_device *bd, + fmode_t has_write_perm) + { ++ unsigned char tmpcmd[sizeof(rq->__cmd)]; ++ unsigned char *cmdptr; ++ + if (hdr->request_len > BLK_MAX_CDB) { + rq->cmd = kzalloc(hdr->request_len, GFP_KERNEL); + if (!rq->cmd) + return -ENOMEM; +- } ++ cmdptr = rq->cmd; ++ } else ++ cmdptr = tmpcmd; + +- if (copy_from_user(rq->cmd, (void *)(unsigned long)hdr->request, ++ if (copy_from_user(cmdptr, (void *)(unsigned long)hdr->request, + hdr->request_len)) + return -EFAULT; + ++ if (cmdptr != rq->cmd) ++ memcpy(rq->cmd, cmdptr, hdr->request_len); ++ + if (hdr->subprotocol == BSG_SUB_PROTOCOL_SCSI_CMD) { + if (blk_verify_command(rq->cmd, has_write_perm)) + return -EPERM; +diff -urNp linux-2.6.38.3/block/scsi_ioctl.c linux-2.6.38.3/block/scsi_ioctl.c +--- linux-2.6.38.3/block/scsi_ioctl.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/block/scsi_ioctl.c 2011-04-17 15:57:32.000000000 -0400 +@@ -222,8 +222,20 @@ EXPORT_SYMBOL(blk_verify_command); + static int blk_fill_sghdr_rq(struct request_queue *q, struct request *rq, + struct sg_io_hdr *hdr, fmode_t mode) + { +- if (copy_from_user(rq->cmd, hdr->cmdp, hdr->cmd_len)) ++ unsigned char tmpcmd[sizeof(rq->__cmd)]; ++ unsigned char *cmdptr; ++ ++ if (rq->cmd != rq->__cmd) ++ cmdptr = rq->cmd; ++ else ++ cmdptr = tmpcmd; ++ ++ if (copy_from_user(cmdptr, hdr->cmdp, hdr->cmd_len)) + return -EFAULT; ++ ++ if (cmdptr != rq->cmd) ++ memcpy(rq->cmd, cmdptr, hdr->cmd_len); ++ + if (blk_verify_command(rq->cmd, mode & FMODE_WRITE)) + return -EPERM; + +@@ -432,6 +444,8 @@ int sg_scsi_ioctl(struct request_queue * + int err; + unsigned int in_len, out_len, bytes, opcode, cmdlen; + char *buffer = NULL, sense[SCSI_SENSE_BUFFERSIZE]; ++ unsigned char tmpcmd[sizeof(rq->__cmd)]; ++ unsigned char *cmdptr; + + if (!sic) + return -EINVAL; +@@ -465,9 +479,18 @@ int sg_scsi_ioctl(struct request_queue * + */ + err = -EFAULT; + rq->cmd_len = cmdlen; +- if (copy_from_user(rq->cmd, sic->data, cmdlen)) ++ ++ if (rq->cmd != rq->__cmd) ++ cmdptr = rq->cmd; ++ else ++ cmdptr = tmpcmd; ++ ++ if (copy_from_user(cmdptr, sic->data, cmdlen)) + goto error; + ++ if (rq->cmd != cmdptr) ++ memcpy(rq->cmd, cmdptr, cmdlen); ++ + if (in_len && copy_from_user(buffer, sic->data + cmdlen, in_len)) + goto error; + +diff -urNp linux-2.6.38.3/crypto/lrw.c linux-2.6.38.3/crypto/lrw.c +--- linux-2.6.38.3/crypto/lrw.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/crypto/lrw.c 2011-04-17 15:57:32.000000000 -0400 +@@ -60,7 +60,7 @@ static int setkey(struct crypto_tfm *par + struct priv *ctx = crypto_tfm_ctx(parent); + struct crypto_cipher *child = ctx->child; + int err, i; +- be128 tmp = { 0 }; ++ be128 tmp = { 0, 0 }; + int bsize = crypto_cipher_blocksize(child); + + crypto_cipher_clear_flags(child, CRYPTO_TFM_REQ_MASK); +diff -urNp linux-2.6.38.3/Documentation/dontdiff linux-2.6.38.3/Documentation/dontdiff +--- linux-2.6.38.3/Documentation/dontdiff 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/Documentation/dontdiff 2011-04-17 15:57:32.000000000 -0400 +@@ -3,6 +3,7 @@ + *.bin + *.cpio + *.csp ++*.dbg + *.dsp + *.dvi + *.elf +@@ -38,8 +39,10 @@ + *.tab.h + *.tex + *.ver ++*.vim + *.xml + *_MODULES ++*_reg_safe.h + *_vga16.c + *~ + *.9 +@@ -49,11 +52,16 @@ + 53c700_d.h + CVS + ChangeSet ++GPATH ++GRTAGS ++GSYMS ++GTAGS + Image + Kerntypes + Module.markers + Module.symvers + PENDING ++PERF* + SCCS + System.map* + TAGS +@@ -82,6 +90,8 @@ bvmlinux + bzImage* + capflags.c + classlist.h* ++clut_vga16.c ++common-cmds.h + comp*.log + compile.h* + conf +@@ -106,16 +116,19 @@ fore200e_mkfirm + fore200e_pca_fw.c* + gconf + gen-devlist ++gen-kdb_cmds.c + gen_crc32table + gen_init_cpio + generated + genheaders + genksyms + *_gray256.c ++hash + ihex2fw + ikconfig.h* + inat-tables.c + initramfs_data.cpio ++initramfs_data.cpio.bz2 + initramfs_data.cpio.gz + initramfs_list + int16.c +@@ -125,7 +138,6 @@ int32.c + int4.c + int8.c + kallsyms +-kconfig + keywords.c + ksym.c* + ksym.h* +@@ -149,7 +161,9 @@ mkboot + mkbugboot + mkcpustr + mkdep ++mkpiggy + mkprep ++mkregtable + mktables + mktree + modpost +@@ -165,6 +179,7 @@ parse.h + patches* + pca200e.bin + pca200e_ecd.bin2 ++perf-archive + piggy.gz + piggyback + piggy.S +@@ -180,6 +195,7 @@ r600_reg_safe.h + raid6altivec*.c + raid6int*.c + raid6tables.c ++regdb.c + relocs + rn50_reg_safe.h + rs600_reg_safe.h +@@ -189,6 +205,7 @@ setup + setup.bin + setup.elf + sImage ++slabinfo + sm_tbl* + split-include + syscalltab.h +@@ -213,13 +230,17 @@ version.h* + vmlinux + vmlinux-* + vmlinux.aout ++vmlinux.bin.all ++vmlinux.bin.bz2 + vmlinux.lds ++vmlinux.relocs + voffset.h + vsyscall.lds + vsyscall_32.lds + wanxlfw.inc + uImage + unifdef ++utsrelease.h + wakeup.bin + wakeup.elf + wakeup.lds +diff -urNp linux-2.6.38.3/Documentation/filesystems/sysfs.txt linux-2.6.38.3/Documentation/filesystems/sysfs.txt +--- linux-2.6.38.3/Documentation/filesystems/sysfs.txt 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/Documentation/filesystems/sysfs.txt 2011-04-17 15:57:32.000000000 -0400 +@@ -123,8 +123,8 @@ set of sysfs operations for forwarding r + show and store methods of the attribute owners. + + struct sysfs_ops { +- ssize_t (*show)(struct kobject *, struct attribute *, char *); +- ssize_t (*store)(struct kobject *, struct attribute *, const char *, size_t); ++ ssize_t (* const show)(struct kobject *, struct attribute *, char *); ++ ssize_t (* const store)(struct kobject *, struct attribute *, const char *, size_t); + }; + + [ Subsystems should have already defined a struct kobj_type as a +diff -urNp linux-2.6.38.3/Documentation/kernel-parameters.txt linux-2.6.38.3/Documentation/kernel-parameters.txt +--- linux-2.6.38.3/Documentation/kernel-parameters.txt 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/Documentation/kernel-parameters.txt 2011-04-17 15:57:32.000000000 -0400 +@@ -1853,6 +1853,13 @@ bytes respectively. Such letter suffixes + the specified number of seconds. This is to be used if + your oopses keep scrolling off the screen. + ++ pax_nouderef [X86] disables UDEREF. Most likely needed under certain ++ virtualization environments that don't cope well with the ++ expand down segment used by UDEREF on X86-32 or the frequent ++ page table updates on X86-64. ++ ++ pax_softmode= 0/1 to disable/enable PaX softmode on boot already. ++ + pcbit= [HW,ISDN] + + pcd. [PARIDE] +diff -urNp linux-2.6.38.3/drivers/acpi/battery.c linux-2.6.38.3/drivers/acpi/battery.c +--- linux-2.6.38.3/drivers/acpi/battery.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/acpi/battery.c 2011-04-17 15:57:32.000000000 -0400 +@@ -862,7 +862,7 @@ DECLARE_FILE_FUNCTIONS(alarm); + } + + static struct battery_file { +- struct file_operations ops; ++ const struct file_operations ops; + mode_t mode; + const char *name; + } acpi_battery_file[] = { +diff -urNp linux-2.6.38.3/drivers/acpi/blacklist.c linux-2.6.38.3/drivers/acpi/blacklist.c +--- linux-2.6.38.3/drivers/acpi/blacklist.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/acpi/blacklist.c 2011-04-17 15:57:32.000000000 -0400 +@@ -73,7 +73,7 @@ static struct acpi_blacklist_item acpi_b + {"IBM ", "TP600E ", 0x00000105, ACPI_SIG_DSDT, less_than_or_equal, + "Incorrect _ADR", 1}, + +- {""} ++ {"", "", 0, NULL, all_versions, NULL, 0} + }; + + #if CONFIG_ACPI_BLACKLIST_YEAR +diff -urNp linux-2.6.38.3/drivers/acpi/dock.c linux-2.6.38.3/drivers/acpi/dock.c +--- linux-2.6.38.3/drivers/acpi/dock.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/acpi/dock.c 2011-04-17 15:57:32.000000000 -0400 +@@ -77,7 +77,7 @@ struct dock_dependent_device { + struct list_head list; + struct list_head hotplug_list; + acpi_handle handle; +- struct acpi_dock_ops *ops; ++ const struct acpi_dock_ops *ops; + void *context; + }; + +@@ -589,7 +589,7 @@ EXPORT_SYMBOL_GPL(unregister_dock_notifi + * the dock driver after _DCK is executed. + */ + int +-register_hotplug_dock_device(acpi_handle handle, struct acpi_dock_ops *ops, ++register_hotplug_dock_device(acpi_handle handle, const struct acpi_dock_ops *ops, + void *context) + { + struct dock_dependent_device *dd; +diff -urNp linux-2.6.38.3/drivers/acpi/ec_sys.c linux-2.6.38.3/drivers/acpi/ec_sys.c +--- linux-2.6.38.3/drivers/acpi/ec_sys.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/acpi/ec_sys.c 2011-04-17 15:57:32.000000000 -0400 +@@ -96,7 +96,7 @@ static ssize_t acpi_ec_write_io(struct f + return count; + } + +-static struct file_operations acpi_ec_io_ops = { ++static const struct file_operations acpi_ec_io_ops = { + .owner = THIS_MODULE, + .open = acpi_ec_open_io, + .read = acpi_ec_read_io, +diff -urNp linux-2.6.38.3/drivers/acpi/power_meter.c linux-2.6.38.3/drivers/acpi/power_meter.c +--- linux-2.6.38.3/drivers/acpi/power_meter.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/acpi/power_meter.c 2011-04-17 15:57:32.000000000 -0400 +@@ -316,8 +316,6 @@ static ssize_t set_trip(struct device *d + return res; + + temp /= 1000; +- if (temp < 0) +- return -EINVAL; + + mutex_lock(&resource->lock); + resource->trip[attr->index - 7] = temp; +diff -urNp linux-2.6.38.3/drivers/acpi/proc.c linux-2.6.38.3/drivers/acpi/proc.c +--- linux-2.6.38.3/drivers/acpi/proc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/acpi/proc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -342,19 +342,13 @@ acpi_system_write_wakeup_device(struct f + size_t count, loff_t * ppos) + { + struct list_head *node, *next; +- char strbuf[5]; +- char str[5] = ""; +- unsigned int len = count; +- +- if (len > 4) +- len = 4; +- if (len < 0) +- return -EFAULT; ++ char strbuf[5] = {0}; + +- if (copy_from_user(strbuf, buffer, len)) ++ if (count > 4) ++ count = 4; ++ if (copy_from_user(strbuf, buffer, count)) + return -EFAULT; +- strbuf[len] = '\0'; +- sscanf(strbuf, "%s", str); ++ strbuf[count] = '\0'; + + mutex_lock(&acpi_device_lock); + list_for_each_safe(node, next, &acpi_wakeup_device_list) { +@@ -363,7 +357,7 @@ acpi_system_write_wakeup_device(struct f + if (!dev->wakeup.flags.valid) + continue; + +- if (!strncmp(dev->pnp.bus_id, str, 4)) { ++ if (!strncmp(dev->pnp.bus_id, strbuf, 4)) { + if (device_can_wakeup(&dev->dev)) { + bool enable = !device_may_wakeup(&dev->dev); + device_set_wakeup_enable(&dev->dev, enable); +diff -urNp linux-2.6.38.3/drivers/acpi/processor_driver.c linux-2.6.38.3/drivers/acpi/processor_driver.c +--- linux-2.6.38.3/drivers/acpi/processor_driver.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/acpi/processor_driver.c 2011-04-17 15:57:32.000000000 -0400 +@@ -473,7 +473,7 @@ static int __cpuinit acpi_processor_add( + return 0; + #endif + +- BUG_ON((pr->id >= nr_cpu_ids) || (pr->id < 0)); ++ BUG_ON(pr->id >= nr_cpu_ids); + + /* + * Buggy BIOS check +diff -urNp linux-2.6.38.3/drivers/acpi/processor_idle.c linux-2.6.38.3/drivers/acpi/processor_idle.c +--- linux-2.6.38.3/drivers/acpi/processor_idle.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/acpi/processor_idle.c 2011-04-17 15:57:32.000000000 -0400 +@@ -121,7 +121,7 @@ static struct dmi_system_id __cpuinitdat + DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK Computer Inc."), + DMI_MATCH(DMI_PRODUCT_NAME,"L8400B series Notebook PC")}, + (void *)1}, +- {}, ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL}, + }; + + +diff -urNp linux-2.6.38.3/drivers/ata/acard-ahci.c linux-2.6.38.3/drivers/ata/acard-ahci.c +--- linux-2.6.38.3/drivers/ata/acard-ahci.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/acard-ahci.c 2011-04-17 15:57:32.000000000 -0400 +@@ -87,7 +87,7 @@ static struct scsi_host_template acard_a + AHCI_SHT("acard-ahci"), + }; + +-static struct ata_port_operations acard_ops = { ++static const struct ata_port_operations acard_ops = { + .inherits = &ahci_ops, + .qc_prep = acard_ahci_qc_prep, + .qc_fill_rtf = acard_ahci_qc_fill_rtf, +diff -urNp linux-2.6.38.3/drivers/ata/ahci.c linux-2.6.38.3/drivers/ata/ahci.c +--- linux-2.6.38.3/drivers/ata/ahci.c 2011-03-23 17:20:06.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/ahci.c 2011-04-17 15:57:32.000000000 -0400 +@@ -94,17 +94,17 @@ static struct scsi_host_template ahci_sh + AHCI_SHT("ahci"), + }; + +-static struct ata_port_operations ahci_vt8251_ops = { ++static const struct ata_port_operations ahci_vt8251_ops = { + .inherits = &ahci_ops, + .hardreset = ahci_vt8251_hardreset, + }; + +-static struct ata_port_operations ahci_p5wdh_ops = { ++static const struct ata_port_operations ahci_p5wdh_ops = { + .inherits = &ahci_ops, + .hardreset = ahci_p5wdh_hardreset, + }; + +-static struct ata_port_operations ahci_sb600_ops = { ++static const struct ata_port_operations ahci_sb600_ops = { + .inherits = &ahci_ops, + .softreset = ahci_sb600_softreset, + .pmp_softreset = ahci_sb600_softreset, +@@ -394,7 +394,7 @@ static const struct pci_device_id ahci_p + { PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID, + PCI_CLASS_STORAGE_SATA_AHCI, 0xffffff, board_ahci }, + +- { } /* terminate list */ ++ { 0, 0, 0, 0, 0, 0, 0 } /* terminate list */ + }; + + +diff -urNp linux-2.6.38.3/drivers/ata/ahci.h linux-2.6.38.3/drivers/ata/ahci.h +--- linux-2.6.38.3/drivers/ata/ahci.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/ahci.h 2011-04-17 15:57:32.000000000 -0400 +@@ -309,7 +309,7 @@ extern struct device_attribute *ahci_sde + .shost_attrs = ahci_shost_attrs, \ + .sdev_attrs = ahci_sdev_attrs + +-extern struct ata_port_operations ahci_ops; ++extern const struct ata_port_operations ahci_ops; + + void ahci_fill_cmd_slot(struct ahci_port_priv *pp, unsigned int tag, + u32 opts); +diff -urNp linux-2.6.38.3/drivers/ata/ata_generic.c linux-2.6.38.3/drivers/ata/ata_generic.c +--- linux-2.6.38.3/drivers/ata/ata_generic.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/ata_generic.c 2011-04-17 15:57:32.000000000 -0400 +@@ -101,7 +101,7 @@ static struct scsi_host_template generic + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations generic_port_ops = { ++static const struct ata_port_operations generic_port_ops = { + .inherits = &ata_bmdma_port_ops, + .cable_detect = ata_cable_unknown, + .set_mode = generic_set_mode, +diff -urNp linux-2.6.38.3/drivers/ata/ata_piix.c linux-2.6.38.3/drivers/ata/ata_piix.c +--- linux-2.6.38.3/drivers/ata/ata_piix.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/ata_piix.c 2011-04-17 15:57:32.000000000 -0400 +@@ -309,7 +309,7 @@ static const struct pci_device_id piix_p + { 0x8086, 0x1d00, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich8_sata }, + /* SATA Controller IDE (PBG) */ + { 0x8086, 0x1d08, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich8_2port_sata }, +- { } /* terminate list */ ++ { 0, 0, 0, 0, 0, 0, 0 } /* terminate list */ + }; + + static struct pci_driver piix_pci_driver = { +@@ -327,12 +327,12 @@ static struct scsi_host_template piix_sh + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations piix_sata_ops = { ++static const struct ata_port_operations piix_sata_ops = { + .inherits = &ata_bmdma32_port_ops, + .sff_irq_check = piix_irq_check, + }; + +-static struct ata_port_operations piix_pata_ops = { ++static const struct ata_port_operations piix_pata_ops = { + .inherits = &piix_sata_ops, + .cable_detect = ata_cable_40wire, + .set_piomode = piix_set_piomode, +@@ -340,12 +340,12 @@ static struct ata_port_operations piix_p + .prereset = piix_pata_prereset, + }; + +-static struct ata_port_operations piix_vmw_ops = { ++static const struct ata_port_operations piix_vmw_ops = { + .inherits = &piix_pata_ops, + .bmdma_status = piix_vmw_bmdma_status, + }; + +-static struct ata_port_operations ich_pata_ops = { ++static const struct ata_port_operations ich_pata_ops = { + .inherits = &piix_pata_ops, + .cable_detect = ich_pata_cable_detect, + .set_dmamode = ich_set_dmamode, +@@ -361,7 +361,7 @@ static struct scsi_host_template piix_si + .shost_attrs = piix_sidpr_shost_attrs, + }; + +-static struct ata_port_operations piix_sidpr_sata_ops = { ++static const struct ata_port_operations piix_sidpr_sata_ops = { + .inherits = &piix_sata_ops, + .hardreset = sata_std_hardreset, + .scr_read = piix_sidpr_scr_read, +@@ -638,7 +638,7 @@ static const struct ich_laptop ich_lapto + { 0x2653, 0x1043, 0x82D8 }, /* ICH6M on Asus Eee 701 */ + { 0x27df, 0x104d, 0x900e }, /* ICH7 on Sony TZ-90 */ + /* end marker */ +- { 0, } ++ { 0, 0, 0 } + }; + + /** +@@ -1130,7 +1130,7 @@ static int piix_broken_suspend(void) + }, + }, + +- { } /* terminate list */ ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } /* terminate list */ + }; + static const char *oemstrs[] = { + "Tecra M3,", +diff -urNp linux-2.6.38.3/drivers/ata/libahci.c linux-2.6.38.3/drivers/ata/libahci.c +--- linux-2.6.38.3/drivers/ata/libahci.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/libahci.c 2011-04-17 15:57:32.000000000 -0400 +@@ -137,7 +137,7 @@ struct device_attribute *ahci_sdev_attrs + }; + EXPORT_SYMBOL_GPL(ahci_sdev_attrs); + +-struct ata_port_operations ahci_ops = { ++const struct ata_port_operations ahci_ops = { + .inherits = &sata_pmp_port_ops, + + .qc_defer = ahci_pmp_qc_defer, +diff -urNp linux-2.6.38.3/drivers/ata/libata-acpi.c linux-2.6.38.3/drivers/ata/libata-acpi.c +--- linux-2.6.38.3/drivers/ata/libata-acpi.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/libata-acpi.c 2011-04-17 15:57:32.000000000 -0400 +@@ -218,12 +218,12 @@ static void ata_acpi_dev_uevent(acpi_han + ata_acpi_uevent(dev->link->ap, dev, event); + } + +-static struct acpi_dock_ops ata_acpi_dev_dock_ops = { ++static const struct acpi_dock_ops ata_acpi_dev_dock_ops = { + .handler = ata_acpi_dev_notify_dock, + .uevent = ata_acpi_dev_uevent, + }; + +-static struct acpi_dock_ops ata_acpi_ap_dock_ops = { ++static const struct acpi_dock_ops ata_acpi_ap_dock_ops = { + .handler = ata_acpi_ap_notify_dock, + .uevent = ata_acpi_ap_uevent, + }; +diff -urNp linux-2.6.38.3/drivers/ata/libata-core.c linux-2.6.38.3/drivers/ata/libata-core.c +--- linux-2.6.38.3/drivers/ata/libata-core.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/libata-core.c 2011-04-17 15:57:32.000000000 -0400 +@@ -897,7 +897,7 @@ static const struct ata_xfer_ent { + { ATA_SHIFT_PIO, ATA_NR_PIO_MODES, XFER_PIO_0 }, + { ATA_SHIFT_MWDMA, ATA_NR_MWDMA_MODES, XFER_MW_DMA_0 }, + { ATA_SHIFT_UDMA, ATA_NR_UDMA_MODES, XFER_UDMA_0 }, +- { -1, }, ++ { -1, 0, 0 } + }; + + /** +@@ -2885,7 +2885,7 @@ static const struct ata_timing ata_timin + { XFER_UDMA_5, 0, 0, 0, 0, 0, 0, 0, 0, 20 }, + { XFER_UDMA_6, 0, 0, 0, 0, 0, 0, 0, 0, 15 }, + +- { 0xFF } ++ { 0xFF, 0, 0, 0, 0, 0, 0, 0, 0 } + }; + + #define ENOUGH(v, unit) (((v)-1)/(unit)+1) +@@ -4141,7 +4141,7 @@ static const struct ata_blacklist_entry + { "PIONEER DVD-RW DVR-212D", "1.28", ATA_HORKAGE_NOSETXFER }, + + /* End Marker */ +- { } ++ { NULL, NULL, 0 } + }; + + /** +@@ -4746,7 +4746,7 @@ void ata_qc_free(struct ata_queued_cmd * + struct ata_port *ap; + unsigned int tag; + +- WARN_ON_ONCE(qc == NULL); /* ata_qc_from_tag _might_ return NULL */ ++ BUG_ON(qc == NULL); /* ata_qc_from_tag _might_ return NULL */ + ap = qc->ap; + + qc->flags = 0; +@@ -4762,7 +4762,7 @@ void __ata_qc_complete(struct ata_queued + struct ata_port *ap; + struct ata_link *link; + +- WARN_ON_ONCE(qc == NULL); /* ata_qc_from_tag _might_ return NULL */ ++ BUG_ON(qc == NULL); /* ata_qc_from_tag _might_ return NULL */ + WARN_ON_ONCE(!(qc->flags & ATA_QCFLAG_ACTIVE)); + ap = qc->ap; + link = qc->dev->link; +@@ -5755,7 +5755,7 @@ static void ata_host_stop(struct device + * LOCKING: + * None. + */ +-static void ata_finalize_port_ops(struct ata_port_operations *ops) ++static void ata_finalize_port_ops(const struct ata_port_operations *ops) + { + static DEFINE_SPINLOCK(lock); + const struct ata_port_operations *cur; +@@ -5767,6 +5767,7 @@ static void ata_finalize_port_ops(struct + return; + + spin_lock(&lock); ++ pax_open_kernel(); + + for (cur = ops->inherits; cur; cur = cur->inherits) { + void **inherit = (void **)cur; +@@ -5780,8 +5781,9 @@ static void ata_finalize_port_ops(struct + if (IS_ERR(*pp)) + *pp = NULL; + +- ops->inherits = NULL; ++ ((struct ata_port_operations *)ops)->inherits = NULL; + ++ pax_close_kernel(); + spin_unlock(&lock); + } + +@@ -5878,7 +5880,7 @@ int ata_host_start(struct ata_host *host + */ + /* KILLME - the only user left is ipr */ + void ata_host_init(struct ata_host *host, struct device *dev, +- unsigned long flags, struct ata_port_operations *ops) ++ unsigned long flags, const struct ata_port_operations *ops) + { + spin_lock_init(&host->lock); + mutex_init(&host->eh_mutex); +@@ -6584,7 +6586,7 @@ static void ata_dummy_error_handler(stru + /* truly dummy */ + } + +-struct ata_port_operations ata_dummy_port_ops = { ++const struct ata_port_operations ata_dummy_port_ops = { + .qc_prep = ata_noop_qc_prep, + .qc_issue = ata_dummy_qc_issue, + .error_handler = ata_dummy_error_handler, +diff -urNp linux-2.6.38.3/drivers/ata/libata-eh.c linux-2.6.38.3/drivers/ata/libata-eh.c +--- linux-2.6.38.3/drivers/ata/libata-eh.c 2011-03-23 17:20:06.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/libata-eh.c 2011-04-17 15:57:32.000000000 -0400 +@@ -3880,7 +3880,7 @@ void ata_do_eh(struct ata_port *ap, ata_ + */ + void ata_std_error_handler(struct ata_port *ap) + { +- struct ata_port_operations *ops = ap->ops; ++ const struct ata_port_operations *ops = ap->ops; + ata_reset_fn_t hardreset = ops->hardreset; + + /* ignore built-in hardreset if SCR access is not available */ +diff -urNp linux-2.6.38.3/drivers/ata/libata-pmp.c linux-2.6.38.3/drivers/ata/libata-pmp.c +--- linux-2.6.38.3/drivers/ata/libata-pmp.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/libata-pmp.c 2011-04-17 15:57:32.000000000 -0400 +@@ -912,7 +912,7 @@ static int sata_pmp_handle_link_fail(str + */ + static int sata_pmp_eh_recover(struct ata_port *ap) + { +- struct ata_port_operations *ops = ap->ops; ++ const struct ata_port_operations *ops = ap->ops; + int pmp_tries, link_tries[SATA_PMP_MAX_PORTS]; + struct ata_link *pmp_link = &ap->link; + struct ata_device *pmp_dev = pmp_link->device; +diff -urNp linux-2.6.38.3/drivers/ata/pata_acpi.c linux-2.6.38.3/drivers/ata/pata_acpi.c +--- linux-2.6.38.3/drivers/ata/pata_acpi.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_acpi.c 2011-04-17 15:57:32.000000000 -0400 +@@ -216,7 +216,7 @@ static struct scsi_host_template pacpi_s + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations pacpi_ops = { ++static const struct ata_port_operations pacpi_ops = { + .inherits = &ata_bmdma_port_ops, + .qc_issue = pacpi_qc_issue, + .cable_detect = pacpi_cable_detect, +diff -urNp linux-2.6.38.3/drivers/ata/pata_ali.c linux-2.6.38.3/drivers/ata/pata_ali.c +--- linux-2.6.38.3/drivers/ata/pata_ali.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_ali.c 2011-04-17 15:57:32.000000000 -0400 +@@ -363,7 +363,7 @@ static struct scsi_host_template ali_sht + * Port operations for PIO only ALi + */ + +-static struct ata_port_operations ali_early_port_ops = { ++static const struct ata_port_operations ali_early_port_ops = { + .inherits = &ata_sff_port_ops, + .cable_detect = ata_cable_40wire, + .set_piomode = ali_set_piomode, +@@ -380,7 +380,7 @@ static const struct ata_port_operations + * Port operations for DMA capable ALi without cable + * detect + */ +-static struct ata_port_operations ali_20_port_ops = { ++static const struct ata_port_operations ali_20_port_ops = { + .inherits = &ali_dma_base_ops, + .cable_detect = ata_cable_40wire, + .mode_filter = ali_20_filter, +@@ -391,7 +391,7 @@ static struct ata_port_operations ali_20 + /* + * Port operations for DMA capable ALi with cable detect + */ +-static struct ata_port_operations ali_c2_port_ops = { ++static const struct ata_port_operations ali_c2_port_ops = { + .inherits = &ali_dma_base_ops, + .check_atapi_dma = ali_check_atapi_dma, + .cable_detect = ali_c2_cable_detect, +@@ -402,7 +402,7 @@ static struct ata_port_operations ali_c2 + /* + * Port operations for DMA capable ALi with cable detect + */ +-static struct ata_port_operations ali_c4_port_ops = { ++static const struct ata_port_operations ali_c4_port_ops = { + .inherits = &ali_dma_base_ops, + .check_atapi_dma = ali_check_atapi_dma, + .cable_detect = ali_c2_cable_detect, +@@ -412,7 +412,7 @@ static struct ata_port_operations ali_c4 + /* + * Port operations for DMA capable ALi with cable detect and LBA48 + */ +-static struct ata_port_operations ali_c5_port_ops = { ++static const struct ata_port_operations ali_c5_port_ops = { + .inherits = &ali_dma_base_ops, + .check_atapi_dma = ali_check_atapi_dma, + .dev_config = ali_warn_atapi_dma, +diff -urNp linux-2.6.38.3/drivers/ata/pata_amd.c linux-2.6.38.3/drivers/ata/pata_amd.c +--- linux-2.6.38.3/drivers/ata/pata_amd.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_amd.c 2011-04-17 15:57:32.000000000 -0400 +@@ -397,28 +397,28 @@ static const struct ata_port_operations + .prereset = amd_pre_reset, + }; + +-static struct ata_port_operations amd33_port_ops = { ++static const struct ata_port_operations amd33_port_ops = { + .inherits = &amd_base_port_ops, + .cable_detect = ata_cable_40wire, + .set_piomode = amd33_set_piomode, + .set_dmamode = amd33_set_dmamode, + }; + +-static struct ata_port_operations amd66_port_ops = { ++static const struct ata_port_operations amd66_port_ops = { + .inherits = &amd_base_port_ops, + .cable_detect = ata_cable_unknown, + .set_piomode = amd66_set_piomode, + .set_dmamode = amd66_set_dmamode, + }; + +-static struct ata_port_operations amd100_port_ops = { ++static const struct ata_port_operations amd100_port_ops = { + .inherits = &amd_base_port_ops, + .cable_detect = ata_cable_unknown, + .set_piomode = amd100_set_piomode, + .set_dmamode = amd100_set_dmamode, + }; + +-static struct ata_port_operations amd133_port_ops = { ++static const struct ata_port_operations amd133_port_ops = { + .inherits = &amd_base_port_ops, + .cable_detect = amd_cable_detect, + .set_piomode = amd133_set_piomode, +@@ -433,13 +433,13 @@ static const struct ata_port_operations + .host_stop = nv_host_stop, + }; + +-static struct ata_port_operations nv100_port_ops = { ++static const struct ata_port_operations nv100_port_ops = { + .inherits = &nv_base_port_ops, + .set_piomode = nv100_set_piomode, + .set_dmamode = nv100_set_dmamode, + }; + +-static struct ata_port_operations nv133_port_ops = { ++static const struct ata_port_operations nv133_port_ops = { + .inherits = &nv_base_port_ops, + .set_piomode = nv133_set_piomode, + .set_dmamode = nv133_set_dmamode, +diff -urNp linux-2.6.38.3/drivers/ata/pata_artop.c linux-2.6.38.3/drivers/ata/pata_artop.c +--- linux-2.6.38.3/drivers/ata/pata_artop.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_artop.c 2011-04-17 15:57:32.000000000 -0400 +@@ -312,7 +312,7 @@ static struct scsi_host_template artop_s + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations artop6210_ops = { ++static const struct ata_port_operations artop6210_ops = { + .inherits = &ata_bmdma_port_ops, + .cable_detect = ata_cable_40wire, + .set_piomode = artop6210_set_piomode, +@@ -321,7 +321,7 @@ static struct ata_port_operations artop6 + .qc_defer = artop6210_qc_defer, + }; + +-static struct ata_port_operations artop6260_ops = { ++static const struct ata_port_operations artop6260_ops = { + .inherits = &ata_bmdma_port_ops, + .cable_detect = artop6260_cable_detect, + .set_piomode = artop6260_set_piomode, +diff -urNp linux-2.6.38.3/drivers/ata/pata_at32.c linux-2.6.38.3/drivers/ata/pata_at32.c +--- linux-2.6.38.3/drivers/ata/pata_at32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_at32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -173,7 +173,7 @@ static struct scsi_host_template at32_sh + ATA_PIO_SHT(DRV_NAME), + }; + +-static struct ata_port_operations at32_port_ops = { ++static const struct ata_port_operations at32_port_ops = { + .inherits = &ata_sff_port_ops, + .cable_detect = ata_cable_40wire, + .set_piomode = pata_at32_set_piomode, +diff -urNp linux-2.6.38.3/drivers/ata/pata_at91.c linux-2.6.38.3/drivers/ata/pata_at91.c +--- linux-2.6.38.3/drivers/ata/pata_at91.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_at91.c 2011-04-17 15:57:32.000000000 -0400 +@@ -196,7 +196,7 @@ static struct scsi_host_template pata_at + ATA_PIO_SHT(DRV_NAME), + }; + +-static struct ata_port_operations pata_at91_port_ops = { ++static const struct ata_port_operations pata_at91_port_ops = { + .inherits = &ata_sff_port_ops, + + .sff_data_xfer = pata_at91_data_xfer_noirq, +diff -urNp linux-2.6.38.3/drivers/ata/pata_atiixp.c linux-2.6.38.3/drivers/ata/pata_atiixp.c +--- linux-2.6.38.3/drivers/ata/pata_atiixp.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_atiixp.c 2011-04-17 15:57:32.000000000 -0400 +@@ -214,7 +214,7 @@ static struct scsi_host_template atiixp_ + .sg_tablesize = LIBATA_DUMB_MAX_PRD, + }; + +-static struct ata_port_operations atiixp_port_ops = { ++static const struct ata_port_operations atiixp_port_ops = { + .inherits = &ata_bmdma_port_ops, + + .qc_prep = ata_bmdma_dumb_qc_prep, +diff -urNp linux-2.6.38.3/drivers/ata/pata_atp867x.c linux-2.6.38.3/drivers/ata/pata_atp867x.c +--- linux-2.6.38.3/drivers/ata/pata_atp867x.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_atp867x.c 2011-04-17 15:57:32.000000000 -0400 +@@ -275,7 +275,7 @@ static struct scsi_host_template atp867x + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations atp867x_ops = { ++static const struct ata_port_operations atp867x_ops = { + .inherits = &ata_bmdma_port_ops, + .cable_detect = atp867x_cable_detect, + .set_piomode = atp867x_set_piomode, +diff -urNp linux-2.6.38.3/drivers/ata/pata_bf54x.c linux-2.6.38.3/drivers/ata/pata_bf54x.c +--- linux-2.6.38.3/drivers/ata/pata_bf54x.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_bf54x.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1420,7 +1420,7 @@ static struct scsi_host_template bfin_sh + .dma_boundary = ATA_DMA_BOUNDARY, + }; + +-static struct ata_port_operations bfin_pata_ops = { ++static const struct ata_port_operations bfin_pata_ops = { + .inherits = &ata_bmdma_port_ops, + + .set_piomode = bfin_set_piomode, +diff -urNp linux-2.6.38.3/drivers/ata/pata_cmd640.c linux-2.6.38.3/drivers/ata/pata_cmd640.c +--- linux-2.6.38.3/drivers/ata/pata_cmd640.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_cmd640.c 2011-04-17 15:57:32.000000000 -0400 +@@ -176,7 +176,7 @@ static struct scsi_host_template cmd640_ + ATA_PIO_SHT(DRV_NAME), + }; + +-static struct ata_port_operations cmd640_port_ops = { ++static const struct ata_port_operations cmd640_port_ops = { + .inherits = &ata_sff_port_ops, + /* In theory xfer_noirq is not needed once we kill the prefetcher */ + .sff_data_xfer = ata_sff_data_xfer_noirq, +diff -urNp linux-2.6.38.3/drivers/ata/pata_cmd64x.c linux-2.6.38.3/drivers/ata/pata_cmd64x.c +--- linux-2.6.38.3/drivers/ata/pata_cmd64x.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_cmd64x.c 2011-04-17 15:57:32.000000000 -0400 +@@ -268,18 +268,18 @@ static const struct ata_port_operations + .set_dmamode = cmd64x_set_dmamode, + }; + +-static struct ata_port_operations cmd64x_port_ops = { ++static const struct ata_port_operations cmd64x_port_ops = { + .inherits = &cmd64x_base_ops, + .cable_detect = ata_cable_40wire, + }; + +-static struct ata_port_operations cmd646r1_port_ops = { ++static const struct ata_port_operations cmd646r1_port_ops = { + .inherits = &cmd64x_base_ops, + .bmdma_stop = cmd646r1_bmdma_stop, + .cable_detect = ata_cable_40wire, + }; + +-static struct ata_port_operations cmd648_port_ops = { ++static const struct ata_port_operations cmd648_port_ops = { + .inherits = &cmd64x_base_ops, + .bmdma_stop = cmd648_bmdma_stop, + .cable_detect = cmd648_cable_detect, +diff -urNp linux-2.6.38.3/drivers/ata/pata_cs5520.c linux-2.6.38.3/drivers/ata/pata_cs5520.c +--- linux-2.6.38.3/drivers/ata/pata_cs5520.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_cs5520.c 2011-04-17 15:57:32.000000000 -0400 +@@ -108,7 +108,7 @@ static struct scsi_host_template cs5520_ + .sg_tablesize = LIBATA_DUMB_MAX_PRD, + }; + +-static struct ata_port_operations cs5520_port_ops = { ++static const struct ata_port_operations cs5520_port_ops = { + .inherits = &ata_bmdma_port_ops, + .qc_prep = ata_bmdma_dumb_qc_prep, + .cable_detect = ata_cable_40wire, +diff -urNp linux-2.6.38.3/drivers/ata/pata_cs5530.c linux-2.6.38.3/drivers/ata/pata_cs5530.c +--- linux-2.6.38.3/drivers/ata/pata_cs5530.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_cs5530.c 2011-04-17 15:57:32.000000000 -0400 +@@ -164,7 +164,7 @@ static struct scsi_host_template cs5530_ + .sg_tablesize = LIBATA_DUMB_MAX_PRD, + }; + +-static struct ata_port_operations cs5530_port_ops = { ++static const struct ata_port_operations cs5530_port_ops = { + .inherits = &ata_bmdma_port_ops, + + .qc_prep = ata_bmdma_dumb_qc_prep, +diff -urNp linux-2.6.38.3/drivers/ata/pata_cs5535.c linux-2.6.38.3/drivers/ata/pata_cs5535.c +--- linux-2.6.38.3/drivers/ata/pata_cs5535.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_cs5535.c 2011-04-17 15:57:32.000000000 -0400 +@@ -160,7 +160,7 @@ static struct scsi_host_template cs5535_ + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations cs5535_port_ops = { ++static const struct ata_port_operations cs5535_port_ops = { + .inherits = &ata_bmdma_port_ops, + .cable_detect = cs5535_cable_detect, + .set_piomode = cs5535_set_piomode, +diff -urNp linux-2.6.38.3/drivers/ata/pata_cs5536.c linux-2.6.38.3/drivers/ata/pata_cs5536.c +--- linux-2.6.38.3/drivers/ata/pata_cs5536.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_cs5536.c 2011-04-17 15:57:32.000000000 -0400 +@@ -233,7 +233,7 @@ static struct scsi_host_template cs5536_ + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations cs5536_port_ops = { ++static const struct ata_port_operations cs5536_port_ops = { + .inherits = &ata_bmdma32_port_ops, + .cable_detect = cs5536_cable_detect, + .set_piomode = cs5536_set_piomode, +diff -urNp linux-2.6.38.3/drivers/ata/pata_cypress.c linux-2.6.38.3/drivers/ata/pata_cypress.c +--- linux-2.6.38.3/drivers/ata/pata_cypress.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_cypress.c 2011-04-17 15:57:32.000000000 -0400 +@@ -115,7 +115,7 @@ static struct scsi_host_template cy82c69 + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations cy82c693_port_ops = { ++static const struct ata_port_operations cy82c693_port_ops = { + .inherits = &ata_bmdma_port_ops, + .cable_detect = ata_cable_40wire, + .set_piomode = cy82c693_set_piomode, +diff -urNp linux-2.6.38.3/drivers/ata/pata_efar.c linux-2.6.38.3/drivers/ata/pata_efar.c +--- linux-2.6.38.3/drivers/ata/pata_efar.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_efar.c 2011-04-17 15:57:32.000000000 -0400 +@@ -238,7 +238,7 @@ static struct scsi_host_template efar_sh + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations efar_ops = { ++static const struct ata_port_operations efar_ops = { + .inherits = &ata_bmdma_port_ops, + .cable_detect = efar_cable_detect, + .set_piomode = efar_set_piomode, +diff -urNp linux-2.6.38.3/drivers/ata/pata_hpt366.c linux-2.6.38.3/drivers/ata/pata_hpt366.c +--- linux-2.6.38.3/drivers/ata/pata_hpt366.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_hpt366.c 2011-04-17 15:57:32.000000000 -0400 +@@ -275,7 +275,7 @@ static struct scsi_host_template hpt36x_ + * Configuration for HPT366/68 + */ + +-static struct ata_port_operations hpt366_port_ops = { ++static const struct ata_port_operations hpt366_port_ops = { + .inherits = &ata_bmdma_port_ops, + .cable_detect = hpt36x_cable_detect, + .mode_filter = hpt366_filter, +diff -urNp linux-2.6.38.3/drivers/ata/pata_hpt37x.c linux-2.6.38.3/drivers/ata/pata_hpt37x.c +--- linux-2.6.38.3/drivers/ata/pata_hpt37x.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_hpt37x.c 2011-04-17 15:57:32.000000000 -0400 +@@ -587,7 +587,7 @@ static struct scsi_host_template hpt37x_ + * Configuration for HPT370 + */ + +-static struct ata_port_operations hpt370_port_ops = { ++static const struct ata_port_operations hpt370_port_ops = { + .inherits = &ata_bmdma_port_ops, + + .bmdma_stop = hpt370_bmdma_stop, +@@ -603,7 +603,7 @@ static struct ata_port_operations hpt370 + * Configuration for HPT370A. Close to 370 but less filters + */ + +-static struct ata_port_operations hpt370a_port_ops = { ++static const struct ata_port_operations hpt370a_port_ops = { + .inherits = &hpt370_port_ops, + .mode_filter = hpt370a_filter, + }; +@@ -613,7 +613,7 @@ static struct ata_port_operations hpt370 + * mode setting functionality. + */ + +-static struct ata_port_operations hpt302_port_ops = { ++static const struct ata_port_operations hpt302_port_ops = { + .inherits = &ata_bmdma_port_ops, + + .bmdma_stop = hpt37x_bmdma_stop, +@@ -629,7 +629,7 @@ static struct ata_port_operations hpt302 + * but we have a mode filter. + */ + +-static struct ata_port_operations hpt372_port_ops = { ++static const struct ata_port_operations hpt372_port_ops = { + .inherits = &hpt302_port_ops, + .mode_filter = hpt372_filter, + }; +@@ -639,7 +639,7 @@ static struct ata_port_operations hpt372 + * but we have a different cable detection procedure for function 1. + */ + +-static struct ata_port_operations hpt374_fn1_port_ops = { ++static const struct ata_port_operations hpt374_fn1_port_ops = { + .inherits = &hpt372_port_ops, + .cable_detect = hpt374_fn1_cable_detect, + }; +diff -urNp linux-2.6.38.3/drivers/ata/pata_hpt3x2n.c linux-2.6.38.3/drivers/ata/pata_hpt3x2n.c +--- linux-2.6.38.3/drivers/ata/pata_hpt3x2n.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_hpt3x2n.c 2011-04-17 15:57:32.000000000 -0400 +@@ -348,7 +348,7 @@ static struct scsi_host_template hpt3x2n + * Configuration for HPT302N/371N. + */ + +-static struct ata_port_operations hpt3xxn_port_ops = { ++static const struct ata_port_operations hpt3xxn_port_ops = { + .inherits = &ata_bmdma_port_ops, + + .bmdma_stop = hpt3x2n_bmdma_stop, +@@ -366,7 +366,7 @@ static struct ata_port_operations hpt3xx + * Configuration for HPT372N. Same as 302N/371N but we have a mode filter. + */ + +-static struct ata_port_operations hpt372n_port_ops = { ++static const struct ata_port_operations hpt372n_port_ops = { + .inherits = &hpt3xxn_port_ops, + .mode_filter = &hpt372n_filter, + }; +diff -urNp linux-2.6.38.3/drivers/ata/pata_hpt3x3.c linux-2.6.38.3/drivers/ata/pata_hpt3x3.c +--- linux-2.6.38.3/drivers/ata/pata_hpt3x3.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_hpt3x3.c 2011-04-17 15:57:32.000000000 -0400 +@@ -141,7 +141,7 @@ static struct scsi_host_template hpt3x3_ + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations hpt3x3_port_ops = { ++static const struct ata_port_operations hpt3x3_port_ops = { + .inherits = &ata_bmdma_port_ops, + .cable_detect = ata_cable_40wire, + .set_piomode = hpt3x3_set_piomode, +diff -urNp linux-2.6.38.3/drivers/ata/pata_icside.c linux-2.6.38.3/drivers/ata/pata_icside.c +--- linux-2.6.38.3/drivers/ata/pata_icside.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_icside.c 2011-04-17 15:57:32.000000000 -0400 +@@ -320,7 +320,7 @@ static void pata_icside_postreset(struct + } + } + +-static struct ata_port_operations pata_icside_port_ops = { ++static const struct ata_port_operations pata_icside_port_ops = { + .inherits = &ata_bmdma_port_ops, + /* no need to build any PRD tables for DMA */ + .qc_prep = ata_noop_qc_prep, +diff -urNp linux-2.6.38.3/drivers/ata/pata_isapnp.c linux-2.6.38.3/drivers/ata/pata_isapnp.c +--- linux-2.6.38.3/drivers/ata/pata_isapnp.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_isapnp.c 2011-04-17 15:57:32.000000000 -0400 +@@ -23,12 +23,12 @@ static struct scsi_host_template isapnp_ + ATA_PIO_SHT(DRV_NAME), + }; + +-static struct ata_port_operations isapnp_port_ops = { ++static const struct ata_port_operations isapnp_port_ops = { + .inherits = &ata_sff_port_ops, + .cable_detect = ata_cable_40wire, + }; + +-static struct ata_port_operations isapnp_noalt_port_ops = { ++static const struct ata_port_operations isapnp_noalt_port_ops = { + .inherits = &ata_sff_port_ops, + .cable_detect = ata_cable_40wire, + /* No altstatus so we don't want to use the lost interrupt poll */ +diff -urNp linux-2.6.38.3/drivers/ata/pata_it8213.c linux-2.6.38.3/drivers/ata/pata_it8213.c +--- linux-2.6.38.3/drivers/ata/pata_it8213.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_it8213.c 2011-04-17 15:57:32.000000000 -0400 +@@ -233,7 +233,7 @@ static struct scsi_host_template it8213_ + }; + + +-static struct ata_port_operations it8213_ops = { ++static const struct ata_port_operations it8213_ops = { + .inherits = &ata_bmdma_port_ops, + .cable_detect = it8213_cable_detect, + .set_piomode = it8213_set_piomode, +diff -urNp linux-2.6.38.3/drivers/ata/pata_it821x.c linux-2.6.38.3/drivers/ata/pata_it821x.c +--- linux-2.6.38.3/drivers/ata/pata_it821x.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_it821x.c 2011-04-17 15:57:32.000000000 -0400 +@@ -801,7 +801,7 @@ static struct scsi_host_template it821x_ + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations it821x_smart_port_ops = { ++static const struct ata_port_operations it821x_smart_port_ops = { + .inherits = &ata_bmdma_port_ops, + + .check_atapi_dma= it821x_check_atapi_dma, +@@ -815,7 +815,7 @@ static struct ata_port_operations it821x + .port_start = it821x_port_start, + }; + +-static struct ata_port_operations it821x_passthru_port_ops = { ++static const struct ata_port_operations it821x_passthru_port_ops = { + .inherits = &ata_bmdma_port_ops, + + .check_atapi_dma= it821x_check_atapi_dma, +@@ -831,7 +831,7 @@ static struct ata_port_operations it821x + .port_start = it821x_port_start, + }; + +-static struct ata_port_operations it821x_rdc_port_ops = { ++static const struct ata_port_operations it821x_rdc_port_ops = { + .inherits = &ata_bmdma_port_ops, + + .check_atapi_dma= it821x_check_atapi_dma, +diff -urNp linux-2.6.38.3/drivers/ata/pata_ixp4xx_cf.c linux-2.6.38.3/drivers/ata/pata_ixp4xx_cf.c +--- linux-2.6.38.3/drivers/ata/pata_ixp4xx_cf.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_ixp4xx_cf.c 2011-04-17 15:57:32.000000000 -0400 +@@ -89,7 +89,7 @@ static struct scsi_host_template ixp4xx_ + ATA_PIO_SHT(DRV_NAME), + }; + +-static struct ata_port_operations ixp4xx_port_ops = { ++static const struct ata_port_operations ixp4xx_port_ops = { + .inherits = &ata_sff_port_ops, + .sff_data_xfer = ixp4xx_mmio_data_xfer, + .cable_detect = ata_cable_40wire, +diff -urNp linux-2.6.38.3/drivers/ata/pata_jmicron.c linux-2.6.38.3/drivers/ata/pata_jmicron.c +--- linux-2.6.38.3/drivers/ata/pata_jmicron.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_jmicron.c 2011-04-17 15:57:32.000000000 -0400 +@@ -111,7 +111,7 @@ static struct scsi_host_template jmicron + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations jmicron_ops = { ++static const struct ata_port_operations jmicron_ops = { + .inherits = &ata_bmdma_port_ops, + .prereset = jmicron_pre_reset, + }; +diff -urNp linux-2.6.38.3/drivers/ata/pata_legacy.c linux-2.6.38.3/drivers/ata/pata_legacy.c +--- linux-2.6.38.3/drivers/ata/pata_legacy.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_legacy.c 2011-04-17 15:57:32.000000000 -0400 +@@ -116,7 +116,7 @@ struct legacy_probe { + + struct legacy_controller { + const char *name; +- struct ata_port_operations *ops; ++ const struct ata_port_operations *ops; + unsigned int pio_mask; + unsigned int flags; + unsigned int pflags; +@@ -239,12 +239,12 @@ static const struct ata_port_operations + * pio_mask as well. + */ + +-static struct ata_port_operations simple_port_ops = { ++static const struct ata_port_operations simple_port_ops = { + .inherits = &legacy_base_port_ops, + .sff_data_xfer = ata_sff_data_xfer_noirq, + }; + +-static struct ata_port_operations legacy_port_ops = { ++static const struct ata_port_operations legacy_port_ops = { + .inherits = &legacy_base_port_ops, + .sff_data_xfer = ata_sff_data_xfer_noirq, + .set_mode = legacy_set_mode, +@@ -340,7 +340,7 @@ static unsigned int pdc_data_xfer_vlb(st + return buflen; + } + +-static struct ata_port_operations pdc20230_port_ops = { ++static const struct ata_port_operations pdc20230_port_ops = { + .inherits = &legacy_base_port_ops, + .set_piomode = pdc20230_set_piomode, + .sff_data_xfer = pdc_data_xfer_vlb, +@@ -373,7 +373,7 @@ static void ht6560a_set_piomode(struct a + ioread8(ap->ioaddr.status_addr); + } + +-static struct ata_port_operations ht6560a_port_ops = { ++static const struct ata_port_operations ht6560a_port_ops = { + .inherits = &legacy_base_port_ops, + .set_piomode = ht6560a_set_piomode, + }; +@@ -416,7 +416,7 @@ static void ht6560b_set_piomode(struct a + ioread8(ap->ioaddr.status_addr); + } + +-static struct ata_port_operations ht6560b_port_ops = { ++static const struct ata_port_operations ht6560b_port_ops = { + .inherits = &legacy_base_port_ops, + .set_piomode = ht6560b_set_piomode, + }; +@@ -515,7 +515,7 @@ static void opti82c611a_set_piomode(stru + } + + +-static struct ata_port_operations opti82c611a_port_ops = { ++static const struct ata_port_operations opti82c611a_port_ops = { + .inherits = &legacy_base_port_ops, + .set_piomode = opti82c611a_set_piomode, + }; +@@ -625,7 +625,7 @@ static unsigned int opti82c46x_qc_issue( + return ata_sff_qc_issue(qc); + } + +-static struct ata_port_operations opti82c46x_port_ops = { ++static const struct ata_port_operations opti82c46x_port_ops = { + .inherits = &legacy_base_port_ops, + .set_piomode = opti82c46x_set_piomode, + .qc_issue = opti82c46x_qc_issue, +@@ -787,20 +787,20 @@ static int qdi_port(struct platform_devi + return 0; + } + +-static struct ata_port_operations qdi6500_port_ops = { ++static const struct ata_port_operations qdi6500_port_ops = { + .inherits = &legacy_base_port_ops, + .set_piomode = qdi6500_set_piomode, + .qc_issue = qdi_qc_issue, + .sff_data_xfer = vlb32_data_xfer, + }; + +-static struct ata_port_operations qdi6580_port_ops = { ++static const struct ata_port_operations qdi6580_port_ops = { + .inherits = &legacy_base_port_ops, + .set_piomode = qdi6580_set_piomode, + .sff_data_xfer = vlb32_data_xfer, + }; + +-static struct ata_port_operations qdi6580dp_port_ops = { ++static const struct ata_port_operations qdi6580dp_port_ops = { + .inherits = &legacy_base_port_ops, + .set_piomode = qdi6580dp_set_piomode, + .qc_issue = qdi_qc_issue, +@@ -872,7 +872,7 @@ static int winbond_port(struct platform_ + return 0; + } + +-static struct ata_port_operations winbond_port_ops = { ++static const struct ata_port_operations winbond_port_ops = { + .inherits = &legacy_base_port_ops, + .set_piomode = winbond_set_piomode, + .sff_data_xfer = vlb32_data_xfer, +@@ -995,7 +995,7 @@ static __init int legacy_init_one(struct + int pio_modes = controller->pio_mask; + unsigned long io = probe->port; + u32 mask = (1 << probe->slot); +- struct ata_port_operations *ops = controller->ops; ++ const struct ata_port_operations *ops = controller->ops; + struct legacy_data *ld = &legacy_data[probe->slot]; + struct ata_host *host = NULL; + struct ata_port *ap; +diff -urNp linux-2.6.38.3/drivers/ata/pata_macio.c linux-2.6.38.3/drivers/ata/pata_macio.c +--- linux-2.6.38.3/drivers/ata/pata_macio.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_macio.c 2011-04-17 15:57:32.000000000 -0400 +@@ -918,9 +918,8 @@ static struct scsi_host_template pata_ma + .slave_configure = pata_macio_slave_config, + }; + +-static struct ata_port_operations pata_macio_ops = { ++static const struct ata_port_operations pata_macio_ops = { + .inherits = &ata_bmdma_port_ops, +- + .freeze = pata_macio_freeze, + .set_piomode = pata_macio_set_timings, + .set_dmamode = pata_macio_set_timings, +diff -urNp linux-2.6.38.3/drivers/ata/pata_marvell.c linux-2.6.38.3/drivers/ata/pata_marvell.c +--- linux-2.6.38.3/drivers/ata/pata_marvell.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_marvell.c 2011-04-17 15:57:32.000000000 -0400 +@@ -100,7 +100,7 @@ static struct scsi_host_template marvell + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations marvell_ops = { ++static const struct ata_port_operations marvell_ops = { + .inherits = &ata_bmdma_port_ops, + .cable_detect = marvell_cable_detect, + .prereset = marvell_pre_reset, +diff -urNp linux-2.6.38.3/drivers/ata/pata_mpc52xx.c linux-2.6.38.3/drivers/ata/pata_mpc52xx.c +--- linux-2.6.38.3/drivers/ata/pata_mpc52xx.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_mpc52xx.c 2011-04-17 15:57:32.000000000 -0400 +@@ -609,7 +609,7 @@ static struct scsi_host_template mpc52xx + ATA_PIO_SHT(DRV_NAME), + }; + +-static struct ata_port_operations mpc52xx_ata_port_ops = { ++static const struct ata_port_operations mpc52xx_ata_port_ops = { + .inherits = &ata_bmdma_port_ops, + .sff_dev_select = mpc52xx_ata_dev_select, + .set_piomode = mpc52xx_ata_set_piomode, +diff -urNp linux-2.6.38.3/drivers/ata/pata_mpiix.c linux-2.6.38.3/drivers/ata/pata_mpiix.c +--- linux-2.6.38.3/drivers/ata/pata_mpiix.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_mpiix.c 2011-04-17 15:57:32.000000000 -0400 +@@ -140,7 +140,7 @@ static struct scsi_host_template mpiix_s + ATA_PIO_SHT(DRV_NAME), + }; + +-static struct ata_port_operations mpiix_port_ops = { ++static const struct ata_port_operations mpiix_port_ops = { + .inherits = &ata_sff_port_ops, + .qc_issue = mpiix_qc_issue, + .cable_detect = ata_cable_40wire, +diff -urNp linux-2.6.38.3/drivers/ata/pata_netcell.c linux-2.6.38.3/drivers/ata/pata_netcell.c +--- linux-2.6.38.3/drivers/ata/pata_netcell.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_netcell.c 2011-04-17 15:57:32.000000000 -0400 +@@ -34,7 +34,7 @@ static struct scsi_host_template netcell + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations netcell_ops = { ++static const struct ata_port_operations netcell_ops = { + .inherits = &ata_bmdma_port_ops, + .cable_detect = ata_cable_80wire, + .read_id = netcell_read_id, +diff -urNp linux-2.6.38.3/drivers/ata/pata_ninja32.c linux-2.6.38.3/drivers/ata/pata_ninja32.c +--- linux-2.6.38.3/drivers/ata/pata_ninja32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_ninja32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -81,7 +81,7 @@ static struct scsi_host_template ninja32 + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations ninja32_port_ops = { ++static const struct ata_port_operations ninja32_port_ops = { + .inherits = &ata_bmdma_port_ops, + .sff_dev_select = ninja32_dev_select, + .cable_detect = ata_cable_40wire, +diff -urNp linux-2.6.38.3/drivers/ata/pata_ns87410.c linux-2.6.38.3/drivers/ata/pata_ns87410.c +--- linux-2.6.38.3/drivers/ata/pata_ns87410.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_ns87410.c 2011-04-17 15:57:32.000000000 -0400 +@@ -132,7 +132,7 @@ static struct scsi_host_template ns87410 + ATA_PIO_SHT(DRV_NAME), + }; + +-static struct ata_port_operations ns87410_port_ops = { ++static const struct ata_port_operations ns87410_port_ops = { + .inherits = &ata_sff_port_ops, + .qc_issue = ns87410_qc_issue, + .cable_detect = ata_cable_40wire, +diff -urNp linux-2.6.38.3/drivers/ata/pata_ns87415.c linux-2.6.38.3/drivers/ata/pata_ns87415.c +--- linux-2.6.38.3/drivers/ata/pata_ns87415.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_ns87415.c 2011-04-17 15:57:32.000000000 -0400 +@@ -299,7 +299,7 @@ static u8 ns87560_bmdma_status(struct at + } + #endif /* 87560 SuperIO Support */ + +-static struct ata_port_operations ns87415_pata_ops = { ++static const struct ata_port_operations ns87415_pata_ops = { + .inherits = &ata_bmdma_port_ops, + + .check_atapi_dma = ns87415_check_atapi_dma, +@@ -313,7 +313,7 @@ static struct ata_port_operations ns8741 + }; + + #if defined(CONFIG_SUPERIO) +-static struct ata_port_operations ns87560_pata_ops = { ++static const struct ata_port_operations ns87560_pata_ops = { + .inherits = &ns87415_pata_ops, + .sff_tf_read = ns87560_tf_read, + .sff_check_status = ns87560_check_status, +diff -urNp linux-2.6.38.3/drivers/ata/pata_octeon_cf.c linux-2.6.38.3/drivers/ata/pata_octeon_cf.c +--- linux-2.6.38.3/drivers/ata/pata_octeon_cf.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_octeon_cf.c 2011-04-17 15:57:32.000000000 -0400 +@@ -780,7 +780,7 @@ static unsigned int octeon_cf_qc_issue(s + return 0; + } + +-static struct ata_port_operations octeon_cf_ops = { ++static struct ata_port_operations octeon_cf_ops = { /* cannot be const */ + .inherits = &ata_sff_port_ops, + .check_atapi_dma = octeon_cf_check_atapi_dma, + .qc_prep = ata_noop_qc_prep, +diff -urNp linux-2.6.38.3/drivers/ata/pata_oldpiix.c linux-2.6.38.3/drivers/ata/pata_oldpiix.c +--- linux-2.6.38.3/drivers/ata/pata_oldpiix.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_oldpiix.c 2011-04-17 15:57:32.000000000 -0400 +@@ -208,7 +208,7 @@ static struct scsi_host_template oldpiix + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations oldpiix_pata_ops = { ++static const struct ata_port_operations oldpiix_pata_ops = { + .inherits = &ata_bmdma_port_ops, + .qc_issue = oldpiix_qc_issue, + .cable_detect = ata_cable_40wire, +diff -urNp linux-2.6.38.3/drivers/ata/pata_opti.c linux-2.6.38.3/drivers/ata/pata_opti.c +--- linux-2.6.38.3/drivers/ata/pata_opti.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_opti.c 2011-04-17 15:57:32.000000000 -0400 +@@ -152,7 +152,7 @@ static struct scsi_host_template opti_sh + ATA_PIO_SHT(DRV_NAME), + }; + +-static struct ata_port_operations opti_port_ops = { ++static const struct ata_port_operations opti_port_ops = { + .inherits = &ata_sff_port_ops, + .cable_detect = ata_cable_40wire, + .set_piomode = opti_set_piomode, +diff -urNp linux-2.6.38.3/drivers/ata/pata_optidma.c linux-2.6.38.3/drivers/ata/pata_optidma.c +--- linux-2.6.38.3/drivers/ata/pata_optidma.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_optidma.c 2011-04-17 15:57:32.000000000 -0400 +@@ -337,7 +337,7 @@ static struct scsi_host_template optidma + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations optidma_port_ops = { ++static const struct ata_port_operations optidma_port_ops = { + .inherits = &ata_bmdma_port_ops, + .cable_detect = ata_cable_40wire, + .set_piomode = optidma_set_pio_mode, +@@ -346,7 +346,7 @@ static struct ata_port_operations optidm + .prereset = optidma_pre_reset, + }; + +-static struct ata_port_operations optiplus_port_ops = { ++static const struct ata_port_operations optiplus_port_ops = { + .inherits = &optidma_port_ops, + .set_piomode = optiplus_set_pio_mode, + .set_dmamode = optiplus_set_dma_mode, +diff -urNp linux-2.6.38.3/drivers/ata/pata_palmld.c linux-2.6.38.3/drivers/ata/pata_palmld.c +--- linux-2.6.38.3/drivers/ata/pata_palmld.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_palmld.c 2011-04-17 15:57:32.000000000 -0400 +@@ -37,7 +37,7 @@ static struct scsi_host_template palmld_ + ATA_PIO_SHT(DRV_NAME), + }; + +-static struct ata_port_operations palmld_port_ops = { ++static const struct ata_port_operations palmld_port_ops = { + .inherits = &ata_sff_port_ops, + .sff_data_xfer = ata_sff_data_xfer_noirq, + .cable_detect = ata_cable_40wire, +diff -urNp linux-2.6.38.3/drivers/ata/pata_pcmcia.c linux-2.6.38.3/drivers/ata/pata_pcmcia.c +--- linux-2.6.38.3/drivers/ata/pata_pcmcia.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_pcmcia.c 2011-04-17 15:57:32.000000000 -0400 +@@ -151,14 +151,14 @@ static struct scsi_host_template pcmcia_ + ATA_PIO_SHT(DRV_NAME), + }; + +-static struct ata_port_operations pcmcia_port_ops = { ++static const struct ata_port_operations pcmcia_port_ops = { + .inherits = &ata_sff_port_ops, + .sff_data_xfer = ata_sff_data_xfer_noirq, + .cable_detect = ata_cable_40wire, + .set_mode = pcmcia_set_mode, + }; + +-static struct ata_port_operations pcmcia_8bit_port_ops = { ++static const struct ata_port_operations pcmcia_8bit_port_ops = { + .inherits = &ata_sff_port_ops, + .sff_data_xfer = ata_data_xfer_8bit, + .cable_detect = ata_cable_40wire, +@@ -205,7 +205,7 @@ static int pcmcia_init_one(struct pcmcia + unsigned long io_base, ctl_base; + void __iomem *io_addr, *ctl_addr; + int n_ports = 1; +- struct ata_port_operations *ops = &pcmcia_port_ops; ++ const struct ata_port_operations *ops = &pcmcia_port_ops; + + /* Set up attributes in order to probe card and get resources */ + pdev->config_flags |= CONF_ENABLE_IRQ | CONF_AUTO_SET_IO | +diff -urNp linux-2.6.38.3/drivers/ata/pata_pdc2027x.c linux-2.6.38.3/drivers/ata/pata_pdc2027x.c +--- linux-2.6.38.3/drivers/ata/pata_pdc2027x.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_pdc2027x.c 2011-04-17 15:57:32.000000000 -0400 +@@ -132,14 +132,14 @@ static struct scsi_host_template pdc2027 + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations pdc2027x_pata100_ops = { ++static const struct ata_port_operations pdc2027x_pata100_ops = { + .inherits = &ata_bmdma_port_ops, + .check_atapi_dma = pdc2027x_check_atapi_dma, + .cable_detect = pdc2027x_cable_detect, + .prereset = pdc2027x_prereset, + }; + +-static struct ata_port_operations pdc2027x_pata133_ops = { ++static const struct ata_port_operations pdc2027x_pata133_ops = { + .inherits = &pdc2027x_pata100_ops, + .mode_filter = pdc2027x_mode_filter, + .set_piomode = pdc2027x_set_piomode, +diff -urNp linux-2.6.38.3/drivers/ata/pata_pdc202xx_old.c linux-2.6.38.3/drivers/ata/pata_pdc202xx_old.c +--- linux-2.6.38.3/drivers/ata/pata_pdc202xx_old.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_pdc202xx_old.c 2011-04-17 15:57:32.000000000 -0400 +@@ -295,7 +295,7 @@ static struct scsi_host_template pdc202x + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations pdc2024x_port_ops = { ++static const struct ata_port_operations pdc2024x_port_ops = { + .inherits = &ata_bmdma_port_ops, + + .cable_detect = ata_cable_40wire, +@@ -306,7 +306,7 @@ static struct ata_port_operations pdc202 + .sff_irq_check = pdc202xx_irq_check, + }; + +-static struct ata_port_operations pdc2026x_port_ops = { ++static const struct ata_port_operations pdc2026x_port_ops = { + .inherits = &pdc2024x_port_ops, + + .check_atapi_dma = pdc2026x_check_atapi_dma, +diff -urNp linux-2.6.38.3/drivers/ata/pata_piccolo.c linux-2.6.38.3/drivers/ata/pata_piccolo.c +--- linux-2.6.38.3/drivers/ata/pata_piccolo.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_piccolo.c 2011-04-17 15:57:32.000000000 -0400 +@@ -67,7 +67,7 @@ static struct scsi_host_template tosh_sh + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations tosh_port_ops = { ++static const struct ata_port_operations tosh_port_ops = { + .inherits = &ata_bmdma_port_ops, + .cable_detect = ata_cable_unknown, + .set_piomode = tosh_set_piomode, +diff -urNp linux-2.6.38.3/drivers/ata/pata_platform.c linux-2.6.38.3/drivers/ata/pata_platform.c +--- linux-2.6.38.3/drivers/ata/pata_platform.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_platform.c 2011-04-17 15:57:32.000000000 -0400 +@@ -48,7 +48,7 @@ static struct scsi_host_template pata_pl + ATA_PIO_SHT(DRV_NAME), + }; + +-static struct ata_port_operations pata_platform_port_ops = { ++static const struct ata_port_operations pata_platform_port_ops = { + .inherits = &ata_sff_port_ops, + .sff_data_xfer = ata_sff_data_xfer_noirq, + .cable_detect = ata_cable_unknown, +diff -urNp linux-2.6.38.3/drivers/ata/pata_pxa.c linux-2.6.38.3/drivers/ata/pata_pxa.c +--- linux-2.6.38.3/drivers/ata/pata_pxa.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_pxa.c 2011-04-17 15:57:32.000000000 -0400 +@@ -198,7 +198,7 @@ static struct scsi_host_template pxa_ata + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations pxa_ata_port_ops = { ++static const struct ata_port_operations pxa_ata_port_ops = { + .inherits = &ata_bmdma_port_ops, + .cable_detect = ata_cable_40wire, + +diff -urNp linux-2.6.38.3/drivers/ata/pata_qdi.c linux-2.6.38.3/drivers/ata/pata_qdi.c +--- linux-2.6.38.3/drivers/ata/pata_qdi.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_qdi.c 2011-04-17 15:57:32.000000000 -0400 +@@ -157,7 +157,7 @@ static struct scsi_host_template qdi_sht + ATA_PIO_SHT(DRV_NAME), + }; + +-static struct ata_port_operations qdi6500_port_ops = { ++static const struct ata_port_operations qdi6500_port_ops = { + .inherits = &ata_sff_port_ops, + .qc_issue = qdi_qc_issue, + .sff_data_xfer = qdi_data_xfer, +@@ -165,7 +165,7 @@ static struct ata_port_operations qdi650 + .set_piomode = qdi6500_set_piomode, + }; + +-static struct ata_port_operations qdi6580_port_ops = { ++static const struct ata_port_operations qdi6580_port_ops = { + .inherits = &qdi6500_port_ops, + .set_piomode = qdi6580_set_piomode, + }; +diff -urNp linux-2.6.38.3/drivers/ata/pata_radisys.c linux-2.6.38.3/drivers/ata/pata_radisys.c +--- linux-2.6.38.3/drivers/ata/pata_radisys.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_radisys.c 2011-04-17 15:57:32.000000000 -0400 +@@ -187,7 +187,7 @@ static struct scsi_host_template radisys + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations radisys_pata_ops = { ++static const struct ata_port_operations radisys_pata_ops = { + .inherits = &ata_bmdma_port_ops, + .qc_issue = radisys_qc_issue, + .cable_detect = ata_cable_unknown, +diff -urNp linux-2.6.38.3/drivers/ata/pata_rb532_cf.c linux-2.6.38.3/drivers/ata/pata_rb532_cf.c +--- linux-2.6.38.3/drivers/ata/pata_rb532_cf.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_rb532_cf.c 2011-04-17 15:57:32.000000000 -0400 +@@ -69,7 +69,7 @@ static irqreturn_t rb532_pata_irq_handle + return IRQ_HANDLED; + } + +-static struct ata_port_operations rb532_pata_port_ops = { ++static const struct ata_port_operations rb532_pata_port_ops = { + .inherits = &ata_sff_port_ops, + .sff_data_xfer = ata_sff_data_xfer32, + }; +diff -urNp linux-2.6.38.3/drivers/ata/pata_rdc.c linux-2.6.38.3/drivers/ata/pata_rdc.c +--- linux-2.6.38.3/drivers/ata/pata_rdc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_rdc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -273,7 +273,7 @@ static void rdc_set_dmamode(struct ata_p + pci_write_config_byte(dev, 0x48, udma_enable); + } + +-static struct ata_port_operations rdc_pata_ops = { ++static const struct ata_port_operations rdc_pata_ops = { + .inherits = &ata_bmdma32_port_ops, + .cable_detect = rdc_pata_cable_detect, + .set_piomode = rdc_set_piomode, +diff -urNp linux-2.6.38.3/drivers/ata/pata_rz1000.c linux-2.6.38.3/drivers/ata/pata_rz1000.c +--- linux-2.6.38.3/drivers/ata/pata_rz1000.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_rz1000.c 2011-04-17 15:57:32.000000000 -0400 +@@ -54,7 +54,7 @@ static struct scsi_host_template rz1000_ + ATA_PIO_SHT(DRV_NAME), + }; + +-static struct ata_port_operations rz1000_port_ops = { ++static const struct ata_port_operations rz1000_port_ops = { + .inherits = &ata_sff_port_ops, + .cable_detect = ata_cable_40wire, + .set_mode = rz1000_set_mode, +diff -urNp linux-2.6.38.3/drivers/ata/pata_samsung_cf.c linux-2.6.38.3/drivers/ata/pata_samsung_cf.c +--- linux-2.6.38.3/drivers/ata/pata_samsung_cf.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_samsung_cf.c 2011-04-17 15:57:32.000000000 -0400 +@@ -399,7 +399,7 @@ static struct scsi_host_template pata_s3 + ATA_PIO_SHT(DRV_NAME), + }; + +-static struct ata_port_operations pata_s3c_port_ops = { ++static const struct ata_port_operations pata_s3c_port_ops = { + .inherits = &ata_sff_port_ops, + .sff_check_status = pata_s3c_check_status, + .sff_check_altstatus = pata_s3c_check_altstatus, +@@ -413,7 +413,7 @@ static struct ata_port_operations pata_s + .set_piomode = pata_s3c_set_piomode, + }; + +-static struct ata_port_operations pata_s5p_port_ops = { ++static const struct ata_port_operations pata_s5p_port_ops = { + .inherits = &ata_sff_port_ops, + .set_piomode = pata_s3c_set_piomode, + }; +diff -urNp linux-2.6.38.3/drivers/ata/pata_sc1200.c linux-2.6.38.3/drivers/ata/pata_sc1200.c +--- linux-2.6.38.3/drivers/ata/pata_sc1200.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_sc1200.c 2011-04-17 15:57:32.000000000 -0400 +@@ -207,7 +207,7 @@ static struct scsi_host_template sc1200_ + .sg_tablesize = LIBATA_DUMB_MAX_PRD, + }; + +-static struct ata_port_operations sc1200_port_ops = { ++static const struct ata_port_operations sc1200_port_ops = { + .inherits = &ata_bmdma_port_ops, + .qc_prep = ata_bmdma_dumb_qc_prep, + .qc_issue = sc1200_qc_issue, +diff -urNp linux-2.6.38.3/drivers/ata/pata_scc.c linux-2.6.38.3/drivers/ata/pata_scc.c +--- linux-2.6.38.3/drivers/ata/pata_scc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_scc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -926,7 +926,7 @@ static struct scsi_host_template scc_sht + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations scc_pata_ops = { ++static const struct ata_port_operations scc_pata_ops = { + .inherits = &ata_bmdma_port_ops, + + .set_piomode = scc_set_piomode, +diff -urNp linux-2.6.38.3/drivers/ata/pata_sch.c linux-2.6.38.3/drivers/ata/pata_sch.c +--- linux-2.6.38.3/drivers/ata/pata_sch.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_sch.c 2011-04-17 15:57:32.000000000 -0400 +@@ -75,7 +75,7 @@ static struct scsi_host_template sch_sht + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations sch_pata_ops = { ++static const struct ata_port_operations sch_pata_ops = { + .inherits = &ata_bmdma_port_ops, + .cable_detect = ata_cable_unknown, + .set_piomode = sch_set_piomode, +diff -urNp linux-2.6.38.3/drivers/ata/pata_serverworks.c linux-2.6.38.3/drivers/ata/pata_serverworks.c +--- linux-2.6.38.3/drivers/ata/pata_serverworks.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_serverworks.c 2011-04-17 15:57:32.000000000 -0400 +@@ -300,7 +300,7 @@ static struct scsi_host_template serverw + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations serverworks_osb4_port_ops = { ++static const struct ata_port_operations serverworks_osb4_port_ops = { + .inherits = &ata_bmdma_port_ops, + .cable_detect = serverworks_cable_detect, + .mode_filter = serverworks_osb4_filter, +@@ -308,7 +308,7 @@ static struct ata_port_operations server + .set_dmamode = serverworks_set_dmamode, + }; + +-static struct ata_port_operations serverworks_csb_port_ops = { ++static const struct ata_port_operations serverworks_csb_port_ops = { + .inherits = &serverworks_osb4_port_ops, + .mode_filter = serverworks_csb_filter, + }; +diff -urNp linux-2.6.38.3/drivers/ata/pata_sil680.c linux-2.6.38.3/drivers/ata/pata_sil680.c +--- linux-2.6.38.3/drivers/ata/pata_sil680.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_sil680.c 2011-04-17 15:57:32.000000000 -0400 +@@ -225,8 +225,7 @@ static struct scsi_host_template sil680_ + ATA_BMDMA_SHT(DRV_NAME), + }; + +- +-static struct ata_port_operations sil680_port_ops = { ++static const struct ata_port_operations sil680_port_ops = { + .inherits = &ata_bmdma32_port_ops, + .sff_exec_command = sil680_sff_exec_command, + .sff_irq_check = sil680_sff_irq_check, +diff -urNp linux-2.6.38.3/drivers/ata/pata_sis.c linux-2.6.38.3/drivers/ata/pata_sis.c +--- linux-2.6.38.3/drivers/ata/pata_sis.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_sis.c 2011-04-17 15:57:32.000000000 -0400 +@@ -503,47 +503,47 @@ static struct scsi_host_template sis_sht + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations sis_133_for_sata_ops = { ++static const struct ata_port_operations sis_133_for_sata_ops = { + .inherits = &ata_bmdma_port_ops, + .set_piomode = sis_133_set_piomode, + .set_dmamode = sis_133_set_dmamode, + .cable_detect = sis_133_cable_detect, + }; + +-static struct ata_port_operations sis_base_ops = { ++static const struct ata_port_operations sis_base_ops = { + .inherits = &ata_bmdma_port_ops, + .prereset = sis_pre_reset, + }; + +-static struct ata_port_operations sis_133_ops = { ++static const struct ata_port_operations sis_133_ops = { + .inherits = &sis_base_ops, + .set_piomode = sis_133_set_piomode, + .set_dmamode = sis_133_set_dmamode, + .cable_detect = sis_133_cable_detect, + }; + +-static struct ata_port_operations sis_133_early_ops = { ++static const struct ata_port_operations sis_133_early_ops = { + .inherits = &sis_base_ops, + .set_piomode = sis_100_set_piomode, + .set_dmamode = sis_133_early_set_dmamode, + .cable_detect = sis_66_cable_detect, + }; + +-static struct ata_port_operations sis_100_ops = { ++static const struct ata_port_operations sis_100_ops = { + .inherits = &sis_base_ops, + .set_piomode = sis_100_set_piomode, + .set_dmamode = sis_100_set_dmamode, + .cable_detect = sis_66_cable_detect, + }; + +-static struct ata_port_operations sis_66_ops = { ++static const struct ata_port_operations sis_66_ops = { + .inherits = &sis_base_ops, + .set_piomode = sis_old_set_piomode, + .set_dmamode = sis_66_set_dmamode, + .cable_detect = sis_66_cable_detect, + }; + +-static struct ata_port_operations sis_old_ops = { ++static const struct ata_port_operations sis_old_ops = { + .inherits = &sis_base_ops, + .set_piomode = sis_old_set_piomode, + .set_dmamode = sis_old_set_dmamode, +diff -urNp linux-2.6.38.3/drivers/ata/pata_sl82c105.c linux-2.6.38.3/drivers/ata/pata_sl82c105.c +--- linux-2.6.38.3/drivers/ata/pata_sl82c105.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_sl82c105.c 2011-04-17 15:57:32.000000000 -0400 +@@ -241,7 +241,7 @@ static struct scsi_host_template sl82c10 + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations sl82c105_port_ops = { ++static const struct ata_port_operations sl82c105_port_ops = { + .inherits = &ata_bmdma_port_ops, + .qc_defer = sl82c105_qc_defer, + .bmdma_start = sl82c105_bmdma_start, +diff -urNp linux-2.6.38.3/drivers/ata/pata_triflex.c linux-2.6.38.3/drivers/ata/pata_triflex.c +--- linux-2.6.38.3/drivers/ata/pata_triflex.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_triflex.c 2011-04-17 15:57:32.000000000 -0400 +@@ -178,7 +178,7 @@ static struct scsi_host_template triflex + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations triflex_port_ops = { ++static const struct ata_port_operations triflex_port_ops = { + .inherits = &ata_bmdma_port_ops, + .bmdma_start = triflex_bmdma_start, + .bmdma_stop = triflex_bmdma_stop, +diff -urNp linux-2.6.38.3/drivers/ata/pata_via.c linux-2.6.38.3/drivers/ata/pata_via.c +--- linux-2.6.38.3/drivers/ata/pata_via.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pata_via.c 2011-04-17 15:57:32.000000000 -0400 +@@ -441,7 +441,7 @@ static struct scsi_host_template via_sht + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations via_port_ops = { ++static const struct ata_port_operations via_port_ops = { + .inherits = &ata_bmdma_port_ops, + .cable_detect = via_cable_detect, + .set_piomode = via_set_piomode, +@@ -452,7 +452,7 @@ static struct ata_port_operations via_po + .mode_filter = via_mode_filter, + }; + +-static struct ata_port_operations via_port_ops_noirq = { ++static const struct ata_port_operations via_port_ops_noirq = { + .inherits = &via_port_ops, + .sff_data_xfer = ata_sff_data_xfer_noirq, + }; +diff -urNp linux-2.6.38.3/drivers/ata/pdc_adma.c linux-2.6.38.3/drivers/ata/pdc_adma.c +--- linux-2.6.38.3/drivers/ata/pdc_adma.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/pdc_adma.c 2011-04-17 15:57:32.000000000 -0400 +@@ -146,7 +146,7 @@ static struct scsi_host_template adma_at + .dma_boundary = ADMA_DMA_BOUNDARY, + }; + +-static struct ata_port_operations adma_ata_ops = { ++static const struct ata_port_operations adma_ata_ops = { + .inherits = &ata_sff_port_ops, + + .lost_interrupt = ATA_OP_NULL, +diff -urNp linux-2.6.38.3/drivers/ata/sata_dwc_460ex.c linux-2.6.38.3/drivers/ata/sata_dwc_460ex.c +--- linux-2.6.38.3/drivers/ata/sata_dwc_460ex.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/sata_dwc_460ex.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1560,7 +1560,7 @@ static struct scsi_host_template sata_dw + .dma_boundary = ATA_DMA_BOUNDARY, + }; + +-static struct ata_port_operations sata_dwc_ops = { ++static const struct ata_port_operations sata_dwc_ops = { + .inherits = &ata_sff_port_ops, + + .error_handler = sata_dwc_error_handler, +diff -urNp linux-2.6.38.3/drivers/ata/sata_fsl.c linux-2.6.38.3/drivers/ata/sata_fsl.c +--- linux-2.6.38.3/drivers/ata/sata_fsl.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/sata_fsl.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1258,7 +1258,7 @@ static struct scsi_host_template sata_fs + .dma_boundary = ATA_DMA_BOUNDARY, + }; + +-static struct ata_port_operations sata_fsl_ops = { ++static const struct ata_port_operations sata_fsl_ops = { + .inherits = &sata_pmp_port_ops, + + .qc_defer = ata_std_qc_defer, +diff -urNp linux-2.6.38.3/drivers/ata/sata_inic162x.c linux-2.6.38.3/drivers/ata/sata_inic162x.c +--- linux-2.6.38.3/drivers/ata/sata_inic162x.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/sata_inic162x.c 2011-04-17 15:57:32.000000000 -0400 +@@ -705,7 +705,7 @@ static int inic_port_start(struct ata_po + return 0; + } + +-static struct ata_port_operations inic_port_ops = { ++static const struct ata_port_operations inic_port_ops = { + .inherits = &sata_port_ops, + + .check_atapi_dma = inic_check_atapi_dma, +diff -urNp linux-2.6.38.3/drivers/ata/sata_mv.c linux-2.6.38.3/drivers/ata/sata_mv.c +--- linux-2.6.38.3/drivers/ata/sata_mv.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/sata_mv.c 2011-04-17 15:57:32.000000000 -0400 +@@ -663,7 +663,7 @@ static struct scsi_host_template mv6_sht + .dma_boundary = MV_DMA_BOUNDARY, + }; + +-static struct ata_port_operations mv5_ops = { ++static const struct ata_port_operations mv5_ops = { + .inherits = &ata_sff_port_ops, + + .lost_interrupt = ATA_OP_NULL, +@@ -683,7 +683,7 @@ static struct ata_port_operations mv5_op + .port_stop = mv_port_stop, + }; + +-static struct ata_port_operations mv6_ops = { ++static const struct ata_port_operations mv6_ops = { + .inherits = &ata_bmdma_port_ops, + + .lost_interrupt = ATA_OP_NULL, +@@ -717,7 +717,7 @@ static struct ata_port_operations mv6_op + .port_stop = mv_port_stop, + }; + +-static struct ata_port_operations mv_iie_ops = { ++static const struct ata_port_operations mv_iie_ops = { + .inherits = &mv6_ops, + .dev_config = ATA_OP_NULL, + .qc_prep = mv_qc_prep_iie, +diff -urNp linux-2.6.38.3/drivers/ata/sata_nv.c linux-2.6.38.3/drivers/ata/sata_nv.c +--- linux-2.6.38.3/drivers/ata/sata_nv.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/sata_nv.c 2011-04-17 15:57:32.000000000 -0400 +@@ -465,7 +465,7 @@ static struct scsi_host_template nv_swnc + * cases. Define nv_hardreset() which only kicks in for post-boot + * probing and use it for all variants. + */ +-static struct ata_port_operations nv_generic_ops = { ++static const struct ata_port_operations nv_generic_ops = { + .inherits = &ata_bmdma_port_ops, + .lost_interrupt = ATA_OP_NULL, + .scr_read = nv_scr_read, +@@ -473,20 +473,20 @@ static struct ata_port_operations nv_gen + .hardreset = nv_hardreset, + }; + +-static struct ata_port_operations nv_nf2_ops = { ++static const struct ata_port_operations nv_nf2_ops = { + .inherits = &nv_generic_ops, + .freeze = nv_nf2_freeze, + .thaw = nv_nf2_thaw, + }; + +-static struct ata_port_operations nv_ck804_ops = { ++static const struct ata_port_operations nv_ck804_ops = { + .inherits = &nv_generic_ops, + .freeze = nv_ck804_freeze, + .thaw = nv_ck804_thaw, + .host_stop = nv_ck804_host_stop, + }; + +-static struct ata_port_operations nv_adma_ops = { ++static const struct ata_port_operations nv_adma_ops = { + .inherits = &nv_ck804_ops, + + .check_atapi_dma = nv_adma_check_atapi_dma, +@@ -510,7 +510,7 @@ static struct ata_port_operations nv_adm + .host_stop = nv_adma_host_stop, + }; + +-static struct ata_port_operations nv_swncq_ops = { ++static const struct ata_port_operations nv_swncq_ops = { + .inherits = &nv_generic_ops, + + .qc_defer = ata_std_qc_defer, +diff -urNp linux-2.6.38.3/drivers/ata/sata_promise.c linux-2.6.38.3/drivers/ata/sata_promise.c +--- linux-2.6.38.3/drivers/ata/sata_promise.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/sata_promise.c 2011-04-17 15:57:32.000000000 -0400 +@@ -196,7 +196,7 @@ static const struct ata_port_operations + .error_handler = pdc_error_handler, + }; + +-static struct ata_port_operations pdc_sata_ops = { ++static const struct ata_port_operations pdc_sata_ops = { + .inherits = &pdc_common_ops, + .cable_detect = pdc_sata_cable_detect, + .freeze = pdc_sata_freeze, +@@ -209,14 +209,14 @@ static struct ata_port_operations pdc_sa + + /* First-generation chips need a more restrictive ->check_atapi_dma op, + and ->freeze/thaw that ignore the hotplug controls. */ +-static struct ata_port_operations pdc_old_sata_ops = { ++static const struct ata_port_operations pdc_old_sata_ops = { + .inherits = &pdc_sata_ops, + .freeze = pdc_freeze, + .thaw = pdc_thaw, + .check_atapi_dma = pdc_old_sata_check_atapi_dma, + }; + +-static struct ata_port_operations pdc_pata_ops = { ++static const struct ata_port_operations pdc_pata_ops = { + .inherits = &pdc_common_ops, + .cable_detect = pdc_pata_cable_detect, + .freeze = pdc_freeze, +diff -urNp linux-2.6.38.3/drivers/ata/sata_qstor.c linux-2.6.38.3/drivers/ata/sata_qstor.c +--- linux-2.6.38.3/drivers/ata/sata_qstor.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/sata_qstor.c 2011-04-17 15:57:32.000000000 -0400 +@@ -131,7 +131,7 @@ static struct scsi_host_template qs_ata_ + .dma_boundary = QS_DMA_BOUNDARY, + }; + +-static struct ata_port_operations qs_ata_ops = { ++static const struct ata_port_operations qs_ata_ops = { + .inherits = &ata_sff_port_ops, + + .check_atapi_dma = qs_check_atapi_dma, +diff -urNp linux-2.6.38.3/drivers/ata/sata_sil24.c linux-2.6.38.3/drivers/ata/sata_sil24.c +--- linux-2.6.38.3/drivers/ata/sata_sil24.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/sata_sil24.c 2011-04-17 15:57:32.000000000 -0400 +@@ -389,7 +389,7 @@ static struct scsi_host_template sil24_s + .dma_boundary = ATA_DMA_BOUNDARY, + }; + +-static struct ata_port_operations sil24_ops = { ++static const struct ata_port_operations sil24_ops = { + .inherits = &sata_pmp_port_ops, + + .qc_defer = sil24_qc_defer, +diff -urNp linux-2.6.38.3/drivers/ata/sata_sil.c linux-2.6.38.3/drivers/ata/sata_sil.c +--- linux-2.6.38.3/drivers/ata/sata_sil.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/sata_sil.c 2011-04-17 15:57:32.000000000 -0400 +@@ -182,7 +182,7 @@ static struct scsi_host_template sil_sht + .sg_tablesize = ATA_MAX_PRD + }; + +-static struct ata_port_operations sil_ops = { ++static const struct ata_port_operations sil_ops = { + .inherits = &ata_bmdma32_port_ops, + .dev_config = sil_dev_config, + .set_mode = sil_set_mode, +diff -urNp linux-2.6.38.3/drivers/ata/sata_sis.c linux-2.6.38.3/drivers/ata/sata_sis.c +--- linux-2.6.38.3/drivers/ata/sata_sis.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/sata_sis.c 2011-04-17 15:57:32.000000000 -0400 +@@ -89,7 +89,7 @@ static struct scsi_host_template sis_sht + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations sis_ops = { ++static const struct ata_port_operations sis_ops = { + .inherits = &ata_bmdma_port_ops, + .scr_read = sis_scr_read, + .scr_write = sis_scr_write, +diff -urNp linux-2.6.38.3/drivers/ata/sata_svw.c linux-2.6.38.3/drivers/ata/sata_svw.c +--- linux-2.6.38.3/drivers/ata/sata_svw.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/sata_svw.c 2011-04-17 15:57:32.000000000 -0400 +@@ -344,7 +344,7 @@ static struct scsi_host_template k2_sata + }; + + +-static struct ata_port_operations k2_sata_ops = { ++static const struct ata_port_operations k2_sata_ops = { + .inherits = &ata_bmdma_port_ops, + .sff_tf_load = k2_sata_tf_load, + .sff_tf_read = k2_sata_tf_read, +diff -urNp linux-2.6.38.3/drivers/ata/sata_sx4.c linux-2.6.38.3/drivers/ata/sata_sx4.c +--- linux-2.6.38.3/drivers/ata/sata_sx4.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/sata_sx4.c 2011-04-17 15:57:32.000000000 -0400 +@@ -249,7 +249,7 @@ static struct scsi_host_template pdc_sat + }; + + /* TODO: inherit from base port_ops after converting to new EH */ +-static struct ata_port_operations pdc_20621_ops = { ++static const struct ata_port_operations pdc_20621_ops = { + .inherits = &ata_sff_port_ops, + + .check_atapi_dma = pdc_check_atapi_dma, +diff -urNp linux-2.6.38.3/drivers/ata/sata_uli.c linux-2.6.38.3/drivers/ata/sata_uli.c +--- linux-2.6.38.3/drivers/ata/sata_uli.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/sata_uli.c 2011-04-17 15:57:32.000000000 -0400 +@@ -80,7 +80,7 @@ static struct scsi_host_template uli_sht + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations uli_ops = { ++static const struct ata_port_operations uli_ops = { + .inherits = &ata_bmdma_port_ops, + .scr_read = uli_scr_read, + .scr_write = uli_scr_write, +diff -urNp linux-2.6.38.3/drivers/ata/sata_via.c linux-2.6.38.3/drivers/ata/sata_via.c +--- linux-2.6.38.3/drivers/ata/sata_via.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/sata_via.c 2011-04-17 15:57:32.000000000 -0400 +@@ -115,32 +115,32 @@ static struct scsi_host_template svia_sh + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations svia_base_ops = { ++static const struct ata_port_operations svia_base_ops = { + .inherits = &ata_bmdma_port_ops, + .sff_tf_load = svia_tf_load, + }; + +-static struct ata_port_operations vt6420_sata_ops = { ++static const struct ata_port_operations vt6420_sata_ops = { + .inherits = &svia_base_ops, + .freeze = svia_noop_freeze, + .prereset = vt6420_prereset, + .bmdma_start = vt6420_bmdma_start, + }; + +-static struct ata_port_operations vt6421_pata_ops = { ++static const struct ata_port_operations vt6421_pata_ops = { + .inherits = &svia_base_ops, + .cable_detect = vt6421_pata_cable_detect, + .set_piomode = vt6421_set_pio_mode, + .set_dmamode = vt6421_set_dma_mode, + }; + +-static struct ata_port_operations vt6421_sata_ops = { ++static const struct ata_port_operations vt6421_sata_ops = { + .inherits = &svia_base_ops, + .scr_read = svia_scr_read, + .scr_write = svia_scr_write, + }; + +-static struct ata_port_operations vt8251_ops = { ++static const struct ata_port_operations vt8251_ops = { + .inherits = &svia_base_ops, + .hardreset = sata_std_hardreset, + .scr_read = vt8251_scr_read, +diff -urNp linux-2.6.38.3/drivers/ata/sata_vsc.c linux-2.6.38.3/drivers/ata/sata_vsc.c +--- linux-2.6.38.3/drivers/ata/sata_vsc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ata/sata_vsc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -300,7 +300,7 @@ static struct scsi_host_template vsc_sat + }; + + +-static struct ata_port_operations vsc_sata_ops = { ++static const struct ata_port_operations vsc_sata_ops = { + .inherits = &ata_bmdma_port_ops, + /* The IRQ handling is not quite standard SFF behaviour so we + cannot use the default lost interrupt handler */ +diff -urNp linux-2.6.38.3/drivers/atm/adummy.c linux-2.6.38.3/drivers/atm/adummy.c +--- linux-2.6.38.3/drivers/atm/adummy.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/atm/adummy.c 2011-04-17 15:57:32.000000000 -0400 +@@ -114,7 +114,7 @@ adummy_send(struct atm_vcc *vcc, struct + vcc->pop(vcc, skb); + else + dev_kfree_skb_any(skb); +- atomic_inc(&vcc->stats->tx); ++ atomic_inc_unchecked(&vcc->stats->tx); + + return 0; + } +diff -urNp linux-2.6.38.3/drivers/atm/ambassador.c linux-2.6.38.3/drivers/atm/ambassador.c +--- linux-2.6.38.3/drivers/atm/ambassador.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/atm/ambassador.c 2011-04-17 15:57:32.000000000 -0400 +@@ -454,7 +454,7 @@ static void tx_complete (amb_dev * dev, + PRINTD (DBG_FLOW|DBG_TX, "tx_complete %p %p", dev, tx); + + // VC layer stats +- atomic_inc(&ATM_SKB(skb)->vcc->stats->tx); ++ atomic_inc_unchecked(&ATM_SKB(skb)->vcc->stats->tx); + + // free the descriptor + kfree (tx_descr); +@@ -495,7 +495,7 @@ static void rx_complete (amb_dev * dev, + dump_skb ("<<<", vc, skb); + + // VC layer stats +- atomic_inc(&atm_vcc->stats->rx); ++ atomic_inc_unchecked(&atm_vcc->stats->rx); + __net_timestamp(skb); + // end of our responsability + atm_vcc->push (atm_vcc, skb); +@@ -510,7 +510,7 @@ static void rx_complete (amb_dev * dev, + } else { + PRINTK (KERN_INFO, "dropped over-size frame"); + // should we count this? +- atomic_inc(&atm_vcc->stats->rx_drop); ++ atomic_inc_unchecked(&atm_vcc->stats->rx_drop); + } + + } else { +@@ -1342,7 +1342,7 @@ static int amb_send (struct atm_vcc * at + } + + if (check_area (skb->data, skb->len)) { +- atomic_inc(&atm_vcc->stats->tx_err); ++ atomic_inc_unchecked(&atm_vcc->stats->tx_err); + return -ENOMEM; // ? + } + +diff -urNp linux-2.6.38.3/drivers/atm/atmtcp.c linux-2.6.38.3/drivers/atm/atmtcp.c +--- linux-2.6.38.3/drivers/atm/atmtcp.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/atm/atmtcp.c 2011-04-17 15:57:32.000000000 -0400 +@@ -207,7 +207,7 @@ static int atmtcp_v_send(struct atm_vcc + if (vcc->pop) vcc->pop(vcc,skb); + else dev_kfree_skb(skb); + if (dev_data) return 0; +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + return -ENOLINK; + } + size = skb->len+sizeof(struct atmtcp_hdr); +@@ -215,7 +215,7 @@ static int atmtcp_v_send(struct atm_vcc + if (!new_skb) { + if (vcc->pop) vcc->pop(vcc,skb); + else dev_kfree_skb(skb); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + return -ENOBUFS; + } + hdr = (void *) skb_put(new_skb,sizeof(struct atmtcp_hdr)); +@@ -226,8 +226,8 @@ static int atmtcp_v_send(struct atm_vcc + if (vcc->pop) vcc->pop(vcc,skb); + else dev_kfree_skb(skb); + out_vcc->push(out_vcc,new_skb); +- atomic_inc(&vcc->stats->tx); +- atomic_inc(&out_vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->tx); ++ atomic_inc_unchecked(&out_vcc->stats->rx); + return 0; + } + +@@ -301,7 +301,7 @@ static int atmtcp_c_send(struct atm_vcc + out_vcc = find_vcc(dev, ntohs(hdr->vpi), ntohs(hdr->vci)); + read_unlock(&vcc_sklist_lock); + if (!out_vcc) { +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + goto done; + } + skb_pull(skb,sizeof(struct atmtcp_hdr)); +@@ -313,8 +313,8 @@ static int atmtcp_c_send(struct atm_vcc + __net_timestamp(new_skb); + skb_copy_from_linear_data(skb, skb_put(new_skb, skb->len), skb->len); + out_vcc->push(out_vcc,new_skb); +- atomic_inc(&vcc->stats->tx); +- atomic_inc(&out_vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->tx); ++ atomic_inc_unchecked(&out_vcc->stats->rx); + done: + if (vcc->pop) vcc->pop(vcc,skb); + else dev_kfree_skb(skb); +diff -urNp linux-2.6.38.3/drivers/atm/eni.c linux-2.6.38.3/drivers/atm/eni.c +--- linux-2.6.38.3/drivers/atm/eni.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/atm/eni.c 2011-04-17 15:57:32.000000000 -0400 +@@ -526,7 +526,7 @@ static int rx_aal0(struct atm_vcc *vcc) + DPRINTK(DEV_LABEL "(itf %d): trashing empty cell\n", + vcc->dev->number); + length = 0; +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + } + else { + length = ATM_CELL_SIZE-1; /* no HEC */ +@@ -581,7 +581,7 @@ static int rx_aal5(struct atm_vcc *vcc) + size); + } + eff = length = 0; +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + } + else { + size = (descr & MID_RED_COUNT)*(ATM_CELL_PAYLOAD >> 2); +@@ -598,7 +598,7 @@ static int rx_aal5(struct atm_vcc *vcc) + "(VCI=%d,length=%ld,size=%ld (descr 0x%lx))\n", + vcc->dev->number,vcc->vci,length,size << 2,descr); + length = eff = 0; +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + } + } + skb = eff ? atm_alloc_charge(vcc,eff << 2,GFP_ATOMIC) : NULL; +@@ -771,7 +771,7 @@ rx_dequeued++; + vcc->push(vcc,skb); + pushed++; + } +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + } + wake_up(&eni_dev->rx_wait); + } +@@ -1228,7 +1228,7 @@ static void dequeue_tx(struct atm_dev *d + PCI_DMA_TODEVICE); + if (vcc->pop) vcc->pop(vcc,skb); + else dev_kfree_skb_irq(skb); +- atomic_inc(&vcc->stats->tx); ++ atomic_inc_unchecked(&vcc->stats->tx); + wake_up(&eni_dev->tx_wait); + dma_complete++; + } +diff -urNp linux-2.6.38.3/drivers/atm/firestream.c linux-2.6.38.3/drivers/atm/firestream.c +--- linux-2.6.38.3/drivers/atm/firestream.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/atm/firestream.c 2011-04-17 15:57:32.000000000 -0400 +@@ -749,7 +749,7 @@ static void process_txdone_queue (struct + } + } + +- atomic_inc(&ATM_SKB(skb)->vcc->stats->tx); ++ atomic_inc_unchecked(&ATM_SKB(skb)->vcc->stats->tx); + + fs_dprintk (FS_DEBUG_TXMEM, "i"); + fs_dprintk (FS_DEBUG_ALLOC, "Free t-skb: %p\n", skb); +@@ -816,7 +816,7 @@ static void process_incoming (struct fs_ + #endif + skb_put (skb, qe->p1 & 0xffff); + ATM_SKB(skb)->vcc = atm_vcc; +- atomic_inc(&atm_vcc->stats->rx); ++ atomic_inc_unchecked(&atm_vcc->stats->rx); + __net_timestamp(skb); + fs_dprintk (FS_DEBUG_ALLOC, "Free rec-skb: %p (pushed)\n", skb); + atm_vcc->push (atm_vcc, skb); +@@ -837,12 +837,12 @@ static void process_incoming (struct fs_ + kfree (pe); + } + if (atm_vcc) +- atomic_inc(&atm_vcc->stats->rx_drop); ++ atomic_inc_unchecked(&atm_vcc->stats->rx_drop); + break; + case 0x1f: /* Reassembly abort: no buffers. */ + /* Silently increment error counter. */ + if (atm_vcc) +- atomic_inc(&atm_vcc->stats->rx_drop); ++ atomic_inc_unchecked(&atm_vcc->stats->rx_drop); + break; + default: /* Hmm. Haven't written the code to handle the others yet... -- REW */ + printk (KERN_WARNING "Don't know what to do with RX status %x: %s.\n", +diff -urNp linux-2.6.38.3/drivers/atm/fore200e.c linux-2.6.38.3/drivers/atm/fore200e.c +--- linux-2.6.38.3/drivers/atm/fore200e.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/atm/fore200e.c 2011-04-17 15:57:32.000000000 -0400 +@@ -933,9 +933,9 @@ fore200e_tx_irq(struct fore200e* fore200 + #endif + /* check error condition */ + if (*entry->status & STATUS_ERROR) +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + else +- atomic_inc(&vcc->stats->tx); ++ atomic_inc_unchecked(&vcc->stats->tx); + } + } + +@@ -1084,7 +1084,7 @@ fore200e_push_rpd(struct fore200e* fore2 + if (skb == NULL) { + DPRINTK(2, "unable to alloc new skb, rx PDU length = %d\n", pdu_len); + +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + return -ENOMEM; + } + +@@ -1127,14 +1127,14 @@ fore200e_push_rpd(struct fore200e* fore2 + + dev_kfree_skb_any(skb); + +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + return -ENOMEM; + } + + ASSERT(atomic_read(&sk_atm(vcc)->sk_wmem_alloc) >= 0); + + vcc->push(vcc, skb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + + ASSERT(atomic_read(&sk_atm(vcc)->sk_wmem_alloc) >= 0); + +@@ -1212,7 +1212,7 @@ fore200e_rx_irq(struct fore200e* fore200 + DPRINTK(2, "damaged PDU on %d.%d.%d\n", + fore200e->atm_dev->number, + entry->rpd->atm_header.vpi, entry->rpd->atm_header.vci); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + } + } + +@@ -1657,7 +1657,7 @@ fore200e_send(struct atm_vcc *vcc, struc + goto retry_here; + } + +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + + fore200e->tx_sat++; + DPRINTK(2, "tx queue of device %s is saturated, PDU dropped - heartbeat is %08x\n", +diff -urNp linux-2.6.38.3/drivers/atm/he.c linux-2.6.38.3/drivers/atm/he.c +--- linux-2.6.38.3/drivers/atm/he.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/atm/he.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1709,7 +1709,7 @@ he_service_rbrq(struct he_dev *he_dev, i + + if (RBRQ_HBUF_ERR(he_dev->rbrq_head)) { + hprintk("HBUF_ERR! (cid 0x%x)\n", cid); +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + goto return_host_buffers; + } + +@@ -1736,7 +1736,7 @@ he_service_rbrq(struct he_dev *he_dev, i + RBRQ_LEN_ERR(he_dev->rbrq_head) + ? "LEN_ERR" : "", + vcc->vpi, vcc->vci); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + goto return_host_buffers; + } + +@@ -1788,7 +1788,7 @@ he_service_rbrq(struct he_dev *he_dev, i + vcc->push(vcc, skb); + spin_lock(&he_dev->global_lock); + +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + + return_host_buffers: + ++pdus_assembled; +@@ -2114,7 +2114,7 @@ __enqueue_tpd(struct he_dev *he_dev, str + tpd->vcc->pop(tpd->vcc, tpd->skb); + else + dev_kfree_skb_any(tpd->skb); +- atomic_inc(&tpd->vcc->stats->tx_err); ++ atomic_inc_unchecked(&tpd->vcc->stats->tx_err); + } + pci_pool_free(he_dev->tpd_pool, tpd, TPD_ADDR(tpd->status)); + return; +@@ -2526,7 +2526,7 @@ he_send(struct atm_vcc *vcc, struct sk_b + vcc->pop(vcc, skb); + else + dev_kfree_skb_any(skb); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + return -EINVAL; + } + +@@ -2537,7 +2537,7 @@ he_send(struct atm_vcc *vcc, struct sk_b + vcc->pop(vcc, skb); + else + dev_kfree_skb_any(skb); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + return -EINVAL; + } + #endif +@@ -2549,7 +2549,7 @@ he_send(struct atm_vcc *vcc, struct sk_b + vcc->pop(vcc, skb); + else + dev_kfree_skb_any(skb); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + spin_unlock_irqrestore(&he_dev->global_lock, flags); + return -ENOMEM; + } +@@ -2591,7 +2591,7 @@ he_send(struct atm_vcc *vcc, struct sk_b + vcc->pop(vcc, skb); + else + dev_kfree_skb_any(skb); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + spin_unlock_irqrestore(&he_dev->global_lock, flags); + return -ENOMEM; + } +@@ -2622,7 +2622,7 @@ he_send(struct atm_vcc *vcc, struct sk_b + __enqueue_tpd(he_dev, tpd, cid); + spin_unlock_irqrestore(&he_dev->global_lock, flags); + +- atomic_inc(&vcc->stats->tx); ++ atomic_inc_unchecked(&vcc->stats->tx); + + return 0; + } +diff -urNp linux-2.6.38.3/drivers/atm/horizon.c linux-2.6.38.3/drivers/atm/horizon.c +--- linux-2.6.38.3/drivers/atm/horizon.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/atm/horizon.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1034,7 +1034,7 @@ static void rx_schedule (hrz_dev * dev, + { + struct atm_vcc * vcc = ATM_SKB(skb)->vcc; + // VC layer stats +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + __net_timestamp(skb); + // end of our responsability + vcc->push (vcc, skb); +@@ -1186,7 +1186,7 @@ static void tx_schedule (hrz_dev * const + dev->tx_iovec = NULL; + + // VC layer stats +- atomic_inc(&ATM_SKB(skb)->vcc->stats->tx); ++ atomic_inc_unchecked(&ATM_SKB(skb)->vcc->stats->tx); + + // free the skb + hrz_kfree_skb (skb); +diff -urNp linux-2.6.38.3/drivers/atm/idt77252.c linux-2.6.38.3/drivers/atm/idt77252.c +--- linux-2.6.38.3/drivers/atm/idt77252.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/atm/idt77252.c 2011-04-17 15:57:32.000000000 -0400 +@@ -811,7 +811,7 @@ drain_scq(struct idt77252_dev *card, str + else + dev_kfree_skb(skb); + +- atomic_inc(&vcc->stats->tx); ++ atomic_inc_unchecked(&vcc->stats->tx); + } + + atomic_dec(&scq->used); +@@ -1074,13 +1074,13 @@ dequeue_rx(struct idt77252_dev *card, st + if ((sb = dev_alloc_skb(64)) == NULL) { + printk("%s: Can't allocate buffers for aal0.\n", + card->name); +- atomic_add(i, &vcc->stats->rx_drop); ++ atomic_add_unchecked(i, &vcc->stats->rx_drop); + break; + } + if (!atm_charge(vcc, sb->truesize)) { + RXPRINTK("%s: atm_charge() dropped aal0 packets.\n", + card->name); +- atomic_add(i - 1, &vcc->stats->rx_drop); ++ atomic_add_unchecked(i - 1, &vcc->stats->rx_drop); + dev_kfree_skb(sb); + break; + } +@@ -1097,7 +1097,7 @@ dequeue_rx(struct idt77252_dev *card, st + ATM_SKB(sb)->vcc = vcc; + __net_timestamp(sb); + vcc->push(vcc, sb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + + cell += ATM_CELL_PAYLOAD; + } +@@ -1134,13 +1134,13 @@ dequeue_rx(struct idt77252_dev *card, st + "(CDC: %08x)\n", + card->name, len, rpp->len, readl(SAR_REG_CDC)); + recycle_rx_pool_skb(card, rpp); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + return; + } + if (stat & SAR_RSQE_CRC) { + RXPRINTK("%s: AAL5 CRC error.\n", card->name); + recycle_rx_pool_skb(card, rpp); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + return; + } + if (skb_queue_len(&rpp->queue) > 1) { +@@ -1151,7 +1151,7 @@ dequeue_rx(struct idt77252_dev *card, st + RXPRINTK("%s: Can't alloc RX skb.\n", + card->name); + recycle_rx_pool_skb(card, rpp); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + return; + } + if (!atm_charge(vcc, skb->truesize)) { +@@ -1170,7 +1170,7 @@ dequeue_rx(struct idt77252_dev *card, st + __net_timestamp(skb); + + vcc->push(vcc, skb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + + return; + } +@@ -1192,7 +1192,7 @@ dequeue_rx(struct idt77252_dev *card, st + __net_timestamp(skb); + + vcc->push(vcc, skb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + + if (skb->truesize > SAR_FB_SIZE_3) + add_rx_skb(card, 3, SAR_FB_SIZE_3, 1); +@@ -1304,14 +1304,14 @@ idt77252_rx_raw(struct idt77252_dev *car + if (vcc->qos.aal != ATM_AAL0) { + RPRINTK("%s: raw cell for non AAL0 vc %u.%u\n", + card->name, vpi, vci); +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + goto drop; + } + + if ((sb = dev_alloc_skb(64)) == NULL) { + printk("%s: Can't allocate buffers for AAL0.\n", + card->name); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + goto drop; + } + +@@ -1330,7 +1330,7 @@ idt77252_rx_raw(struct idt77252_dev *car + ATM_SKB(sb)->vcc = vcc; + __net_timestamp(sb); + vcc->push(vcc, sb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + + drop: + skb_pull(queue, 64); +@@ -1955,13 +1955,13 @@ idt77252_send_skb(struct atm_vcc *vcc, s + + if (vc == NULL) { + printk("%s: NULL connection in send().\n", card->name); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + dev_kfree_skb(skb); + return -EINVAL; + } + if (!test_bit(VCF_TX, &vc->flags)) { + printk("%s: Trying to transmit on a non-tx VC.\n", card->name); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + dev_kfree_skb(skb); + return -EINVAL; + } +@@ -1973,14 +1973,14 @@ idt77252_send_skb(struct atm_vcc *vcc, s + break; + default: + printk("%s: Unsupported AAL: %d\n", card->name, vcc->qos.aal); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + dev_kfree_skb(skb); + return -EINVAL; + } + + if (skb_shinfo(skb)->nr_frags != 0) { + printk("%s: No scatter-gather yet.\n", card->name); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + dev_kfree_skb(skb); + return -EINVAL; + } +@@ -1988,7 +1988,7 @@ idt77252_send_skb(struct atm_vcc *vcc, s + + err = queue_skb(card, vc, skb, oam); + if (err) { +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + dev_kfree_skb(skb); + return err; + } +@@ -2011,7 +2011,7 @@ idt77252_send_oam(struct atm_vcc *vcc, v + skb = dev_alloc_skb(64); + if (!skb) { + printk("%s: Out of memory in send_oam().\n", card->name); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + return -ENOMEM; + } + atomic_add(skb->truesize, &sk_atm(vcc)->sk_wmem_alloc); +diff -urNp linux-2.6.38.3/drivers/atm/iphase.c linux-2.6.38.3/drivers/atm/iphase.c +--- linux-2.6.38.3/drivers/atm/iphase.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/atm/iphase.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1124,7 +1124,7 @@ static int rx_pkt(struct atm_dev *dev) + status = (u_short) (buf_desc_ptr->desc_mode); + if (status & (RX_CER | RX_PTE | RX_OFL)) + { +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + IF_ERR(printk("IA: bad packet, dropping it");) + if (status & RX_CER) { + IF_ERR(printk(" cause: packet CRC error\n");) +@@ -1147,7 +1147,7 @@ static int rx_pkt(struct atm_dev *dev) + len = dma_addr - buf_addr; + if (len > iadev->rx_buf_sz) { + printk("Over %d bytes sdu received, dropped!!!\n", iadev->rx_buf_sz); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + goto out_free_desc; + } + +@@ -1297,7 +1297,7 @@ static void rx_dle_intr(struct atm_dev * + ia_vcc = INPH_IA_VCC(vcc); + if (ia_vcc == NULL) + { +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + dev_kfree_skb_any(skb); + atm_return(vcc, atm_guess_pdu2truesize(len)); + goto INCR_DLE; +@@ -1309,7 +1309,7 @@ static void rx_dle_intr(struct atm_dev * + if ((length > iadev->rx_buf_sz) || (length > + (skb->len - sizeof(struct cpcs_trailer)))) + { +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + IF_ERR(printk("rx_dle_intr: Bad AAL5 trailer %d (skb len %d)", + length, skb->len);) + dev_kfree_skb_any(skb); +@@ -1325,7 +1325,7 @@ static void rx_dle_intr(struct atm_dev * + + IF_RX(printk("rx_dle_intr: skb push");) + vcc->push(vcc,skb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + iadev->rx_pkt_cnt++; + } + INCR_DLE: +@@ -2807,15 +2807,15 @@ static int ia_ioctl(struct atm_dev *dev, + { + struct k_sonet_stats *stats; + stats = &PRIV(_ia_dev[board])->sonet_stats; +- printk("section_bip: %d\n", atomic_read(&stats->section_bip)); +- printk("line_bip : %d\n", atomic_read(&stats->line_bip)); +- printk("path_bip : %d\n", atomic_read(&stats->path_bip)); +- printk("line_febe : %d\n", atomic_read(&stats->line_febe)); +- printk("path_febe : %d\n", atomic_read(&stats->path_febe)); +- printk("corr_hcs : %d\n", atomic_read(&stats->corr_hcs)); +- printk("uncorr_hcs : %d\n", atomic_read(&stats->uncorr_hcs)); +- printk("tx_cells : %d\n", atomic_read(&stats->tx_cells)); +- printk("rx_cells : %d\n", atomic_read(&stats->rx_cells)); ++ printk("section_bip: %d\n", atomic_read_unchecked(&stats->section_bip)); ++ printk("line_bip : %d\n", atomic_read_unchecked(&stats->line_bip)); ++ printk("path_bip : %d\n", atomic_read_unchecked(&stats->path_bip)); ++ printk("line_febe : %d\n", atomic_read_unchecked(&stats->line_febe)); ++ printk("path_febe : %d\n", atomic_read_unchecked(&stats->path_febe)); ++ printk("corr_hcs : %d\n", atomic_read_unchecked(&stats->corr_hcs)); ++ printk("uncorr_hcs : %d\n", atomic_read_unchecked(&stats->uncorr_hcs)); ++ printk("tx_cells : %d\n", atomic_read_unchecked(&stats->tx_cells)); ++ printk("rx_cells : %d\n", atomic_read_unchecked(&stats->rx_cells)); + } + ia_cmds.status = 0; + break; +@@ -2920,7 +2920,7 @@ static int ia_pkt_tx (struct atm_vcc *vc + if ((desc == 0) || (desc > iadev->num_tx_desc)) + { + IF_ERR(printk(DEV_LABEL "invalid desc for send: %d\n", desc);) +- atomic_inc(&vcc->stats->tx); ++ atomic_inc_unchecked(&vcc->stats->tx); + if (vcc->pop) + vcc->pop(vcc, skb); + else +@@ -3025,14 +3025,14 @@ static int ia_pkt_tx (struct atm_vcc *vc + ATM_DESC(skb) = vcc->vci; + skb_queue_tail(&iadev->tx_dma_q, skb); + +- atomic_inc(&vcc->stats->tx); ++ atomic_inc_unchecked(&vcc->stats->tx); + iadev->tx_pkt_cnt++; + /* Increment transaction counter */ + writel(2, iadev->dma+IPHASE5575_TX_COUNTER); + + #if 0 + /* add flow control logic */ +- if (atomic_read(&vcc->stats->tx) % 20 == 0) { ++ if (atomic_read_unchecked(&vcc->stats->tx) % 20 == 0) { + if (iavcc->vc_desc_cnt > 10) { + vcc->tx_quota = vcc->tx_quota * 3 / 4; + printk("Tx1: vcc->tx_quota = %d \n", (u32)vcc->tx_quota ); +diff -urNp linux-2.6.38.3/drivers/atm/lanai.c linux-2.6.38.3/drivers/atm/lanai.c +--- linux-2.6.38.3/drivers/atm/lanai.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/atm/lanai.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1303,7 +1303,7 @@ static void lanai_send_one_aal5(struct l + vcc_tx_add_aal5_trailer(lvcc, skb->len, 0, 0); + lanai_endtx(lanai, lvcc); + lanai_free_skb(lvcc->tx.atmvcc, skb); +- atomic_inc(&lvcc->tx.atmvcc->stats->tx); ++ atomic_inc_unchecked(&lvcc->tx.atmvcc->stats->tx); + } + + /* Try to fill the buffer - don't call unless there is backlog */ +@@ -1426,7 +1426,7 @@ static void vcc_rx_aal5(struct lanai_vcc + ATM_SKB(skb)->vcc = lvcc->rx.atmvcc; + __net_timestamp(skb); + lvcc->rx.atmvcc->push(lvcc->rx.atmvcc, skb); +- atomic_inc(&lvcc->rx.atmvcc->stats->rx); ++ atomic_inc_unchecked(&lvcc->rx.atmvcc->stats->rx); + out: + lvcc->rx.buf.ptr = end; + cardvcc_write(lvcc, endptr, vcc_rxreadptr); +@@ -1668,7 +1668,7 @@ static int handle_service(struct lanai_d + DPRINTK("(itf %d) got RX service entry 0x%X for non-AAL5 " + "vcc %d\n", lanai->number, (unsigned int) s, vci); + lanai->stats.service_rxnotaal5++; +- atomic_inc(&lvcc->rx.atmvcc->stats->rx_err); ++ atomic_inc_unchecked(&lvcc->rx.atmvcc->stats->rx_err); + return 0; + } + if (likely(!(s & (SERVICE_TRASH | SERVICE_STREAM | SERVICE_CRCERR)))) { +@@ -1680,7 +1680,7 @@ static int handle_service(struct lanai_d + int bytes; + read_unlock(&vcc_sklist_lock); + DPRINTK("got trashed rx pdu on vci %d\n", vci); +- atomic_inc(&lvcc->rx.atmvcc->stats->rx_err); ++ atomic_inc_unchecked(&lvcc->rx.atmvcc->stats->rx_err); + lvcc->stats.x.aal5.service_trash++; + bytes = (SERVICE_GET_END(s) * 16) - + (((unsigned long) lvcc->rx.buf.ptr) - +@@ -1692,7 +1692,7 @@ static int handle_service(struct lanai_d + } + if (s & SERVICE_STREAM) { + read_unlock(&vcc_sklist_lock); +- atomic_inc(&lvcc->rx.atmvcc->stats->rx_err); ++ atomic_inc_unchecked(&lvcc->rx.atmvcc->stats->rx_err); + lvcc->stats.x.aal5.service_stream++; + printk(KERN_ERR DEV_LABEL "(itf %d): Got AAL5 stream " + "PDU on VCI %d!\n", lanai->number, vci); +@@ -1700,7 +1700,7 @@ static int handle_service(struct lanai_d + return 0; + } + DPRINTK("got rx crc error on vci %d\n", vci); +- atomic_inc(&lvcc->rx.atmvcc->stats->rx_err); ++ atomic_inc_unchecked(&lvcc->rx.atmvcc->stats->rx_err); + lvcc->stats.x.aal5.service_rxcrc++; + lvcc->rx.buf.ptr = &lvcc->rx.buf.start[SERVICE_GET_END(s) * 4]; + cardvcc_write(lvcc, SERVICE_GET_END(s), vcc_rxreadptr); +diff -urNp linux-2.6.38.3/drivers/atm/nicstar.c linux-2.6.38.3/drivers/atm/nicstar.c +--- linux-2.6.38.3/drivers/atm/nicstar.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/atm/nicstar.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1654,7 +1654,7 @@ static int ns_send(struct atm_vcc *vcc, + if ((vc = (vc_map *) vcc->dev_data) == NULL) { + printk("nicstar%d: vcc->dev_data == NULL on ns_send().\n", + card->index); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + dev_kfree_skb_any(skb); + return -EINVAL; + } +@@ -1662,7 +1662,7 @@ static int ns_send(struct atm_vcc *vcc, + if (!vc->tx) { + printk("nicstar%d: Trying to transmit on a non-tx VC.\n", + card->index); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + dev_kfree_skb_any(skb); + return -EINVAL; + } +@@ -1670,14 +1670,14 @@ static int ns_send(struct atm_vcc *vcc, + if (vcc->qos.aal != ATM_AAL5 && vcc->qos.aal != ATM_AAL0) { + printk("nicstar%d: Only AAL0 and AAL5 are supported.\n", + card->index); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + dev_kfree_skb_any(skb); + return -EINVAL; + } + + if (skb_shinfo(skb)->nr_frags != 0) { + printk("nicstar%d: No scatter-gather yet.\n", card->index); +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + dev_kfree_skb_any(skb); + return -EINVAL; + } +@@ -1725,11 +1725,11 @@ static int ns_send(struct atm_vcc *vcc, + } + + if (push_scqe(card, vc, scq, &scqe, skb) != 0) { +- atomic_inc(&vcc->stats->tx_err); ++ atomic_inc_unchecked(&vcc->stats->tx_err); + dev_kfree_skb_any(skb); + return -EIO; + } +- atomic_inc(&vcc->stats->tx); ++ atomic_inc_unchecked(&vcc->stats->tx); + + return 0; + } +@@ -2046,14 +2046,14 @@ static void dequeue_rx(ns_dev * card, ns + printk + ("nicstar%d: Can't allocate buffers for aal0.\n", + card->index); +- atomic_add(i, &vcc->stats->rx_drop); ++ atomic_add_unchecked(i, &vcc->stats->rx_drop); + break; + } + if (!atm_charge(vcc, sb->truesize)) { + RXPRINTK + ("nicstar%d: atm_charge() dropped aal0 packets.\n", + card->index); +- atomic_add(i - 1, &vcc->stats->rx_drop); /* already increased by 1 */ ++ atomic_add_unchecked(i - 1, &vcc->stats->rx_drop); /* already increased by 1 */ + dev_kfree_skb_any(sb); + break; + } +@@ -2068,7 +2068,7 @@ static void dequeue_rx(ns_dev * card, ns + ATM_SKB(sb)->vcc = vcc; + __net_timestamp(sb); + vcc->push(vcc, sb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + cell += ATM_CELL_PAYLOAD; + } + +@@ -2085,7 +2085,7 @@ static void dequeue_rx(ns_dev * card, ns + if (iovb == NULL) { + printk("nicstar%d: Out of iovec buffers.\n", + card->index); +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + recycle_rx_buf(card, skb); + return; + } +@@ -2109,7 +2109,7 @@ static void dequeue_rx(ns_dev * card, ns + small or large buffer itself. */ + } else if (NS_PRV_IOVCNT(iovb) >= NS_MAX_IOVECS) { + printk("nicstar%d: received too big AAL5 SDU.\n", card->index); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + recycle_iovec_rx_bufs(card, (struct iovec *)iovb->data, + NS_MAX_IOVECS); + NS_PRV_IOVCNT(iovb) = 0; +@@ -2129,7 +2129,7 @@ static void dequeue_rx(ns_dev * card, ns + ("nicstar%d: Expected a small buffer, and this is not one.\n", + card->index); + which_list(card, skb); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + recycle_rx_buf(card, skb); + vc->rx_iov = NULL; + recycle_iov_buf(card, iovb); +@@ -2142,7 +2142,7 @@ static void dequeue_rx(ns_dev * card, ns + ("nicstar%d: Expected a large buffer, and this is not one.\n", + card->index); + which_list(card, skb); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + recycle_iovec_rx_bufs(card, (struct iovec *)iovb->data, + NS_PRV_IOVCNT(iovb)); + vc->rx_iov = NULL; +@@ -2165,7 +2165,7 @@ static void dequeue_rx(ns_dev * card, ns + printk(" - PDU size mismatch.\n"); + else + printk(".\n"); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + recycle_iovec_rx_bufs(card, (struct iovec *)iovb->data, + NS_PRV_IOVCNT(iovb)); + vc->rx_iov = NULL; +@@ -2179,7 +2179,7 @@ static void dequeue_rx(ns_dev * card, ns + /* skb points to a small buffer */ + if (!atm_charge(vcc, skb->truesize)) { + push_rxbufs(card, skb); +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + } else { + skb_put(skb, len); + dequeue_sm_buf(card, skb); +@@ -2189,7 +2189,7 @@ static void dequeue_rx(ns_dev * card, ns + ATM_SKB(skb)->vcc = vcc; + __net_timestamp(skb); + vcc->push(vcc, skb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + } + } else if (NS_PRV_IOVCNT(iovb) == 2) { /* One small plus one large buffer */ + struct sk_buff *sb; +@@ -2200,7 +2200,7 @@ static void dequeue_rx(ns_dev * card, ns + if (len <= NS_SMBUFSIZE) { + if (!atm_charge(vcc, sb->truesize)) { + push_rxbufs(card, sb); +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + } else { + skb_put(sb, len); + dequeue_sm_buf(card, sb); +@@ -2210,7 +2210,7 @@ static void dequeue_rx(ns_dev * card, ns + ATM_SKB(sb)->vcc = vcc; + __net_timestamp(sb); + vcc->push(vcc, sb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + } + + push_rxbufs(card, skb); +@@ -2219,7 +2219,7 @@ static void dequeue_rx(ns_dev * card, ns + + if (!atm_charge(vcc, skb->truesize)) { + push_rxbufs(card, skb); +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + } else { + dequeue_lg_buf(card, skb); + #ifdef NS_USE_DESTRUCTORS +@@ -2232,7 +2232,7 @@ static void dequeue_rx(ns_dev * card, ns + ATM_SKB(skb)->vcc = vcc; + __net_timestamp(skb); + vcc->push(vcc, skb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + } + + push_rxbufs(card, sb); +@@ -2253,7 +2253,7 @@ static void dequeue_rx(ns_dev * card, ns + printk + ("nicstar%d: Out of huge buffers.\n", + card->index); +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + recycle_iovec_rx_bufs(card, + (struct iovec *) + iovb->data, +@@ -2304,7 +2304,7 @@ static void dequeue_rx(ns_dev * card, ns + card->hbpool.count++; + } else + dev_kfree_skb_any(hb); +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + } else { + /* Copy the small buffer to the huge buffer */ + sb = (struct sk_buff *)iov->iov_base; +@@ -2341,7 +2341,7 @@ static void dequeue_rx(ns_dev * card, ns + #endif /* NS_USE_DESTRUCTORS */ + __net_timestamp(hb); + vcc->push(vcc, hb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + } + } + +diff -urNp linux-2.6.38.3/drivers/atm/solos-pci.c linux-2.6.38.3/drivers/atm/solos-pci.c +--- linux-2.6.38.3/drivers/atm/solos-pci.c 2011-04-17 16:53:01.000000000 -0400 ++++ linux-2.6.38.3/drivers/atm/solos-pci.c 2011-04-17 16:53:16.000000000 -0400 +@@ -715,7 +715,7 @@ void solos_bh(unsigned long card_arg) + } + atm_charge(vcc, skb->truesize); + vcc->push(vcc, skb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + break; + + case PKT_STATUS: +@@ -1009,7 +1009,7 @@ static uint32_t fpga_tx(struct solos_car + vcc = SKB_CB(oldskb)->vcc; + + if (vcc) { +- atomic_inc(&vcc->stats->tx); ++ atomic_inc_unchecked(&vcc->stats->tx); + solos_pop(vcc, oldskb); + } else + dev_kfree_skb_irq(oldskb); +diff -urNp linux-2.6.38.3/drivers/atm/suni.c linux-2.6.38.3/drivers/atm/suni.c +--- linux-2.6.38.3/drivers/atm/suni.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/atm/suni.c 2011-04-17 15:57:32.000000000 -0400 +@@ -50,8 +50,8 @@ static DEFINE_SPINLOCK(sunis_lock); + + + #define ADD_LIMITED(s,v) \ +- atomic_add((v),&stats->s); \ +- if (atomic_read(&stats->s) < 0) atomic_set(&stats->s,INT_MAX); ++ atomic_add_unchecked((v),&stats->s); \ ++ if (atomic_read_unchecked(&stats->s) < 0) atomic_set_unchecked(&stats->s,INT_MAX); + + + static void suni_hz(unsigned long from_timer) +diff -urNp linux-2.6.38.3/drivers/atm/uPD98402.c linux-2.6.38.3/drivers/atm/uPD98402.c +--- linux-2.6.38.3/drivers/atm/uPD98402.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/atm/uPD98402.c 2011-04-17 15:57:32.000000000 -0400 +@@ -42,7 +42,7 @@ static int fetch_stats(struct atm_dev *d + struct sonet_stats tmp; + int error = 0; + +- atomic_add(GET(HECCT),&PRIV(dev)->sonet_stats.uncorr_hcs); ++ atomic_add_unchecked(GET(HECCT),&PRIV(dev)->sonet_stats.uncorr_hcs); + sonet_copy_stats(&PRIV(dev)->sonet_stats,&tmp); + if (arg) error = copy_to_user(arg,&tmp,sizeof(tmp)); + if (zero && !error) { +@@ -161,9 +161,9 @@ static int uPD98402_ioctl(struct atm_dev + + + #define ADD_LIMITED(s,v) \ +- { atomic_add(GET(v),&PRIV(dev)->sonet_stats.s); \ +- if (atomic_read(&PRIV(dev)->sonet_stats.s) < 0) \ +- atomic_set(&PRIV(dev)->sonet_stats.s,INT_MAX); } ++ { atomic_add_unchecked(GET(v),&PRIV(dev)->sonet_stats.s); \ ++ if (atomic_read_unchecked(&PRIV(dev)->sonet_stats.s) < 0) \ ++ atomic_set_unchecked(&PRIV(dev)->sonet_stats.s,INT_MAX); } + + + static void stat_event(struct atm_dev *dev) +@@ -194,7 +194,7 @@ static void uPD98402_int(struct atm_dev + if (reason & uPD98402_INT_PFM) stat_event(dev); + if (reason & uPD98402_INT_PCO) { + (void) GET(PCOCR); /* clear interrupt cause */ +- atomic_add(GET(HECCT), ++ atomic_add_unchecked(GET(HECCT), + &PRIV(dev)->sonet_stats.uncorr_hcs); + } + if ((reason & uPD98402_INT_RFO) && +@@ -222,9 +222,9 @@ static int uPD98402_start(struct atm_dev + PUT(~(uPD98402_INT_PFM | uPD98402_INT_ALM | uPD98402_INT_RFO | + uPD98402_INT_LOS),PIMR); /* enable them */ + (void) fetch_stats(dev,NULL,1); /* clear kernel counters */ +- atomic_set(&PRIV(dev)->sonet_stats.corr_hcs,-1); +- atomic_set(&PRIV(dev)->sonet_stats.tx_cells,-1); +- atomic_set(&PRIV(dev)->sonet_stats.rx_cells,-1); ++ atomic_set_unchecked(&PRIV(dev)->sonet_stats.corr_hcs,-1); ++ atomic_set_unchecked(&PRIV(dev)->sonet_stats.tx_cells,-1); ++ atomic_set_unchecked(&PRIV(dev)->sonet_stats.rx_cells,-1); + return 0; + } + +diff -urNp linux-2.6.38.3/drivers/atm/zatm.c linux-2.6.38.3/drivers/atm/zatm.c +--- linux-2.6.38.3/drivers/atm/zatm.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/atm/zatm.c 2011-04-17 15:57:32.000000000 -0400 +@@ -459,7 +459,7 @@ printk("dummy: 0x%08lx, 0x%08lx\n",dummy + } + if (!size) { + dev_kfree_skb_irq(skb); +- if (vcc) atomic_inc(&vcc->stats->rx_err); ++ if (vcc) atomic_inc_unchecked(&vcc->stats->rx_err); + continue; + } + if (!atm_charge(vcc,skb->truesize)) { +@@ -469,7 +469,7 @@ printk("dummy: 0x%08lx, 0x%08lx\n",dummy + skb->len = size; + ATM_SKB(skb)->vcc = vcc; + vcc->push(vcc,skb); +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + } + zout(pos & 0xffff,MTA(mbx)); + #if 0 /* probably a stupid idea */ +@@ -733,7 +733,7 @@ if (*ZATM_PRV_DSC(skb) != (uPD98401_TXPD + skb_queue_head(&zatm_vcc->backlog,skb); + break; + } +- atomic_inc(&vcc->stats->tx); ++ atomic_inc_unchecked(&vcc->stats->tx); + wake_up(&zatm_vcc->tx_wait); + } + +diff -urNp linux-2.6.38.3/drivers/block/cciss.c linux-2.6.38.3/drivers/block/cciss.c +--- linux-2.6.38.3/drivers/block/cciss.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/block/cciss.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1112,6 +1112,8 @@ static int cciss_ioctl32_passthru(struct + int err; + u32 cp; + ++ memset(&arg64, 0, sizeof(arg64)); ++ + err = 0; + err |= + copy_from_user(&arg64.LUN_info, &arg32->LUN_info, +diff -urNp linux-2.6.38.3/drivers/char/agp/frontend.c linux-2.6.38.3/drivers/char/agp/frontend.c +--- linux-2.6.38.3/drivers/char/agp/frontend.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/char/agp/frontend.c 2011-04-17 15:57:32.000000000 -0400 +@@ -817,7 +817,7 @@ static int agpioc_reserve_wrap(struct ag + if (copy_from_user(&reserve, arg, sizeof(struct agp_region))) + return -EFAULT; + +- if ((unsigned) reserve.seg_count >= ~0U/sizeof(struct agp_segment)) ++ if ((unsigned) reserve.seg_count >= ~0U/sizeof(struct agp_segment_priv)) + return -EFAULT; + + client = agp_find_client_by_pid(reserve.pid); +diff -urNp linux-2.6.38.3/drivers/char/agp/intel-agp.c linux-2.6.38.3/drivers/char/agp/intel-agp.c +--- linux-2.6.38.3/drivers/char/agp/intel-agp.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/char/agp/intel-agp.c 2011-04-17 15:57:32.000000000 -0400 +@@ -903,7 +903,7 @@ static struct pci_device_id agp_intel_pc + ID(PCI_DEVICE_ID_INTEL_SANDYBRIDGE_HB), + ID(PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_HB), + ID(PCI_DEVICE_ID_INTEL_SANDYBRIDGE_S_HB), +- { } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(pci, agp_intel_pci_table); +diff -urNp linux-2.6.38.3/drivers/char/hpet.c linux-2.6.38.3/drivers/char/hpet.c +--- linux-2.6.38.3/drivers/char/hpet.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/char/hpet.c 2011-04-17 15:57:32.000000000 -0400 +@@ -553,7 +553,7 @@ static inline unsigned long hpet_time_di + } + + static int +-hpet_ioctl_common(struct hpet_dev *devp, int cmd, unsigned long arg, ++hpet_ioctl_common(struct hpet_dev *devp, unsigned int cmd, unsigned long arg, + struct hpet_info *info) + { + struct hpet_timer __iomem *timer; +@@ -1043,7 +1043,7 @@ static struct acpi_driver hpet_acpi_driv + }, + }; + +-static struct miscdevice hpet_misc = { HPET_MINOR, "hpet", &hpet_fops }; ++static struct miscdevice hpet_misc = { HPET_MINOR, "hpet", &hpet_fops, {NULL, NULL}, NULL, NULL }; + + static int __init hpet_init(void) + { +diff -urNp linux-2.6.38.3/drivers/char/ipmi/ipmi_msghandler.c linux-2.6.38.3/drivers/char/ipmi/ipmi_msghandler.c +--- linux-2.6.38.3/drivers/char/ipmi/ipmi_msghandler.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/char/ipmi/ipmi_msghandler.c 2011-04-17 15:57:32.000000000 -0400 +@@ -414,7 +414,7 @@ struct ipmi_smi { + struct proc_dir_entry *proc_dir; + char proc_dir_name[10]; + +- atomic_t stats[IPMI_NUM_STATS]; ++ atomic_unchecked_t stats[IPMI_NUM_STATS]; + + /* + * run_to_completion duplicate of smb_info, smi_info +@@ -447,9 +447,9 @@ static DEFINE_MUTEX(smi_watchers_mutex); + + + #define ipmi_inc_stat(intf, stat) \ +- atomic_inc(&(intf)->stats[IPMI_STAT_ ## stat]) ++ atomic_inc_unchecked(&(intf)->stats[IPMI_STAT_ ## stat]) + #define ipmi_get_stat(intf, stat) \ +- ((unsigned int) atomic_read(&(intf)->stats[IPMI_STAT_ ## stat])) ++ ((unsigned int) atomic_read_unchecked(&(intf)->stats[IPMI_STAT_ ## stat])) + + static int is_lan_addr(struct ipmi_addr *addr) + { +@@ -2844,7 +2844,7 @@ int ipmi_register_smi(struct ipmi_smi_ha + INIT_LIST_HEAD(&intf->cmd_rcvrs); + init_waitqueue_head(&intf->waitq); + for (i = 0; i < IPMI_NUM_STATS; i++) +- atomic_set(&intf->stats[i], 0); ++ atomic_set_unchecked(&intf->stats[i], 0); + + intf->proc_dir = NULL; + +diff -urNp linux-2.6.38.3/drivers/char/ipmi/ipmi_si_intf.c linux-2.6.38.3/drivers/char/ipmi/ipmi_si_intf.c +--- linux-2.6.38.3/drivers/char/ipmi/ipmi_si_intf.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/char/ipmi/ipmi_si_intf.c 2011-04-17 15:57:32.000000000 -0400 +@@ -285,7 +285,7 @@ struct smi_info { + unsigned char slave_addr; + + /* Counters and things for the proc filesystem. */ +- atomic_t stats[SI_NUM_STATS]; ++ atomic_unchecked_t stats[SI_NUM_STATS]; + + struct task_struct *thread; + +@@ -294,9 +294,9 @@ struct smi_info { + }; + + #define smi_inc_stat(smi, stat) \ +- atomic_inc(&(smi)->stats[SI_STAT_ ## stat]) ++ atomic_inc_unchecked(&(smi)->stats[SI_STAT_ ## stat]) + #define smi_get_stat(smi, stat) \ +- ((unsigned int) atomic_read(&(smi)->stats[SI_STAT_ ## stat])) ++ ((unsigned int) atomic_read_unchecked(&(smi)->stats[SI_STAT_ ## stat])) + + #define SI_MAX_PARMS 4 + +@@ -3202,7 +3202,7 @@ static int try_smi_init(struct smi_info + atomic_set(&new_smi->req_events, 0); + new_smi->run_to_completion = 0; + for (i = 0; i < SI_NUM_STATS; i++) +- atomic_set(&new_smi->stats[i], 0); ++ atomic_set_unchecked(&new_smi->stats[i], 0); + + new_smi->interrupt_disabled = 1; + atomic_set(&new_smi->stop_operation, 0); +diff -urNp linux-2.6.38.3/drivers/char/Kconfig linux-2.6.38.3/drivers/char/Kconfig +--- linux-2.6.38.3/drivers/char/Kconfig 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/char/Kconfig 2011-04-17 15:57:32.000000000 -0400 +@@ -90,7 +90,8 @@ config VT_HW_CONSOLE_BINDING + + config DEVKMEM + bool "/dev/kmem virtual device support" +- default y ++ default n ++ depends on !GRKERNSEC_KMEM + help + Say Y here if you want to support the /dev/kmem device. The + /dev/kmem device is rarely used, but can be used for certain +@@ -1130,9 +1131,10 @@ config TELCLOCK + + config DEVPORT + bool ++ default n + depends on !M68K + depends on ISA || PCI +- default y ++ depends on !GRKERNSEC_KMEM + + source "drivers/s390/char/Kconfig" + +diff -urNp linux-2.6.38.3/drivers/char/mem.c linux-2.6.38.3/drivers/char/mem.c +--- linux-2.6.38.3/drivers/char/mem.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/char/mem.c 2011-04-17 15:57:32.000000000 -0400 +@@ -18,6 +18,7 @@ + #include <linux/raw.h> + #include <linux/tty.h> + #include <linux/capability.h> ++#include <linux/security.h> + #include <linux/ptrace.h> + #include <linux/device.h> + #include <linux/highmem.h> +@@ -34,6 +35,10 @@ + # include <linux/efi.h> + #endif + ++#if defined(CONFIG_GRKERNSEC) && !defined(CONFIG_GRKERNSEC_NO_RBAC) ++extern struct file_operations grsec_fops; ++#endif ++ + static inline unsigned long size_inside_page(unsigned long start, + unsigned long size) + { +@@ -68,9 +73,13 @@ static inline int range_is_allowed(unsig + + while (cursor < to) { + if (!devmem_is_allowed(pfn)) { ++#ifdef CONFIG_GRKERNSEC_KMEM ++ gr_handle_mem_readwrite(from, to); ++#else + printk(KERN_INFO + "Program %s tried to access /dev/mem between %Lx->%Lx.\n", + current->comm, from, to); ++#endif + return 0; + } + cursor += PAGE_SIZE; +@@ -78,6 +87,11 @@ static inline int range_is_allowed(unsig + } + return 1; + } ++#elif defined(CONFIG_GRKERNSEC_KMEM) ++static inline int range_is_allowed(unsigned long pfn, unsigned long size) ++{ ++ return 0; ++} + #else + static inline int range_is_allowed(unsigned long pfn, unsigned long size) + { +@@ -120,6 +134,7 @@ static ssize_t read_mem(struct file *fil + + while (count > 0) { + unsigned long remaining; ++ char *temp; + + sz = size_inside_page(p, count); + +@@ -135,7 +150,23 @@ static ssize_t read_mem(struct file *fil + if (!ptr) + return -EFAULT; + +- remaining = copy_to_user(buf, ptr, sz); ++#ifdef CONFIG_PAX_USERCOPY ++ temp = kmalloc(sz, GFP_KERNEL); ++ if (!temp) { ++ unxlate_dev_mem_ptr(p, ptr); ++ return -ENOMEM; ++ } ++ memcpy(temp, ptr, sz); ++#else ++ temp = ptr; ++#endif ++ ++ remaining = copy_to_user(buf, temp, sz); ++ ++#ifdef CONFIG_PAX_USERCOPY ++ kfree(temp); ++#endif ++ + unxlate_dev_mem_ptr(p, ptr); + if (remaining) + return -EFAULT; +@@ -398,9 +429,8 @@ static ssize_t read_kmem(struct file *fi + size_t count, loff_t *ppos) + { + unsigned long p = *ppos; +- ssize_t low_count, read, sz; ++ ssize_t low_count, read, sz, err = 0; + char * kbuf; /* k-addr because vread() takes vmlist_lock rwlock */ +- int err = 0; + + read = 0; + if (p < (unsigned long) high_memory) { +@@ -422,6 +452,8 @@ static ssize_t read_kmem(struct file *fi + } + #endif + while (low_count > 0) { ++ char *temp; ++ + sz = size_inside_page(p, low_count); + + /* +@@ -431,7 +463,22 @@ static ssize_t read_kmem(struct file *fi + */ + kbuf = xlate_dev_kmem_ptr((char *)p); + +- if (copy_to_user(buf, kbuf, sz)) ++#ifdef CONFIG_PAX_USERCOPY ++ temp = kmalloc(sz, GFP_KERNEL); ++ if (!temp) ++ return -ENOMEM; ++ memcpy(temp, kbuf, sz); ++#else ++ temp = kbuf; ++#endif ++ ++ err = copy_to_user(buf, temp, sz); ++ ++#ifdef CONFIG_PAX_USERCOPY ++ kfree(temp); ++#endif ++ ++ if (err) + return -EFAULT; + buf += sz; + p += sz; +@@ -857,6 +904,9 @@ static const struct memdev { + #ifdef CONFIG_CRASH_DUMP + [12] = { "oldmem", 0, &oldmem_fops, NULL }, + #endif ++#if defined(CONFIG_GRKERNSEC) && !defined(CONFIG_GRKERNSEC_NO_RBAC) ++ [13] = { "grsec",S_IRUSR | S_IWUGO, &grsec_fops, NULL }, ++#endif + }; + + static int memory_open(struct inode *inode, struct file *filp) +diff -urNp linux-2.6.38.3/drivers/char/nvram.c linux-2.6.38.3/drivers/char/nvram.c +--- linux-2.6.38.3/drivers/char/nvram.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/char/nvram.c 2011-04-17 15:57:32.000000000 -0400 +@@ -246,7 +246,7 @@ static ssize_t nvram_read(struct file *f + + spin_unlock_irq(&rtc_lock); + +- if (copy_to_user(buf, contents, tmp - contents)) ++ if (tmp - contents > sizeof(contents) || copy_to_user(buf, contents, tmp - contents)) + return -EFAULT; + + *ppos = i; +@@ -435,7 +435,10 @@ static const struct file_operations nvra + static struct miscdevice nvram_dev = { + NVRAM_MINOR, + "nvram", +- &nvram_fops ++ &nvram_fops, ++ {NULL, NULL}, ++ NULL, ++ NULL + }; + + static int __init nvram_init(void) +diff -urNp linux-2.6.38.3/drivers/char/pcmcia/ipwireless/tty.c linux-2.6.38.3/drivers/char/pcmcia/ipwireless/tty.c +--- linux-2.6.38.3/drivers/char/pcmcia/ipwireless/tty.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/char/pcmcia/ipwireless/tty.c 2011-04-17 15:57:32.000000000 -0400 +@@ -29,6 +29,7 @@ + #include <linux/tty_driver.h> + #include <linux/tty_flip.h> + #include <linux/uaccess.h> ++#include <asm/local.h> + + #include "tty.h" + #include "network.h" +@@ -51,7 +52,7 @@ struct ipw_tty { + int tty_type; + struct ipw_network *network; + struct tty_struct *linux_tty; +- int open_count; ++ local_t open_count; + unsigned int control_lines; + struct mutex ipw_tty_mutex; + int tx_bytes_queued; +@@ -127,10 +128,10 @@ static int ipw_open(struct tty_struct *l + mutex_unlock(&tty->ipw_tty_mutex); + return -ENODEV; + } +- if (tty->open_count == 0) ++ if (local_read(&tty->open_count) == 0) + tty->tx_bytes_queued = 0; + +- tty->open_count++; ++ local_inc(&tty->open_count); + + tty->linux_tty = linux_tty; + linux_tty->driver_data = tty; +@@ -146,9 +147,7 @@ static int ipw_open(struct tty_struct *l + + static void do_ipw_close(struct ipw_tty *tty) + { +- tty->open_count--; +- +- if (tty->open_count == 0) { ++ if (local_dec_return(&tty->open_count) == 0) { + struct tty_struct *linux_tty = tty->linux_tty; + + if (linux_tty != NULL) { +@@ -169,7 +168,7 @@ static void ipw_hangup(struct tty_struct + return; + + mutex_lock(&tty->ipw_tty_mutex); +- if (tty->open_count == 0) { ++ if (local_read(&tty->open_count) == 0) { + mutex_unlock(&tty->ipw_tty_mutex); + return; + } +@@ -198,7 +197,7 @@ void ipwireless_tty_received(struct ipw_ + return; + } + +- if (!tty->open_count) { ++ if (!local_read(&tty->open_count)) { + mutex_unlock(&tty->ipw_tty_mutex); + return; + } +@@ -240,7 +239,7 @@ static int ipw_write(struct tty_struct * + return -ENODEV; + + mutex_lock(&tty->ipw_tty_mutex); +- if (!tty->open_count) { ++ if (!local_read(&tty->open_count)) { + mutex_unlock(&tty->ipw_tty_mutex); + return -EINVAL; + } +@@ -280,7 +279,7 @@ static int ipw_write_room(struct tty_str + if (!tty) + return -ENODEV; + +- if (!tty->open_count) ++ if (!local_read(&tty->open_count)) + return -EINVAL; + + room = IPWIRELESS_TX_QUEUE_SIZE - tty->tx_bytes_queued; +@@ -322,7 +321,7 @@ static int ipw_chars_in_buffer(struct tt + if (!tty) + return 0; + +- if (!tty->open_count) ++ if (!local_read(&tty->open_count)) + return 0; + + return tty->tx_bytes_queued; +@@ -403,7 +402,7 @@ static int ipw_tiocmget(struct tty_struc + if (!tty) + return -ENODEV; + +- if (!tty->open_count) ++ if (!local_read(&tty->open_count)) + return -EINVAL; + + return get_control_lines(tty); +@@ -419,7 +418,7 @@ ipw_tiocmset(struct tty_struct *linux_tt + if (!tty) + return -ENODEV; + +- if (!tty->open_count) ++ if (!local_read(&tty->open_count)) + return -EINVAL; + + return set_control_lines(tty, set, clear); +@@ -433,7 +432,7 @@ static int ipw_ioctl(struct tty_struct * + if (!tty) + return -ENODEV; + +- if (!tty->open_count) ++ if (!local_read(&tty->open_count)) + return -EINVAL; + + /* FIXME: Exactly how is the tty object locked here .. */ +@@ -582,7 +581,7 @@ void ipwireless_tty_free(struct ipw_tty + against a parallel ioctl etc */ + mutex_lock(&ttyj->ipw_tty_mutex); + } +- while (ttyj->open_count) ++ while (local_read(&ttyj->open_count)) + do_ipw_close(ttyj); + ipwireless_disassociate_network_ttys(network, + ttyj->channel_idx); +diff -urNp linux-2.6.38.3/drivers/char/random.c linux-2.6.38.3/drivers/char/random.c +--- linux-2.6.38.3/drivers/char/random.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/char/random.c 2011-04-17 15:57:32.000000000 -0400 +@@ -254,8 +254,13 @@ + /* + * Configuration information + */ ++#ifdef CONFIG_GRKERNSEC_RANDNET ++#define INPUT_POOL_WORDS 512 ++#define OUTPUT_POOL_WORDS 128 ++#else + #define INPUT_POOL_WORDS 128 + #define OUTPUT_POOL_WORDS 32 ++#endif + #define SEC_XFER_SIZE 512 + #define EXTRACT_SIZE 10 + +@@ -293,10 +298,17 @@ static struct poolinfo { + int poolwords; + int tap1, tap2, tap3, tap4, tap5; + } poolinfo_table[] = { ++#ifdef CONFIG_GRKERNSEC_RANDNET ++ /* x^512 + x^411 + x^308 + x^208 +x^104 + x + 1 -- 225 */ ++ { 512, 411, 308, 208, 104, 1 }, ++ /* x^128 + x^103 + x^76 + x^51 + x^25 + x + 1 -- 105 */ ++ { 128, 103, 76, 51, 25, 1 }, ++#else + /* x^128 + x^103 + x^76 + x^51 +x^25 + x + 1 -- 105 */ + { 128, 103, 76, 51, 25, 1 }, + /* x^32 + x^26 + x^20 + x^14 + x^7 + x + 1 -- 15 */ + { 32, 26, 20, 14, 7, 1 }, ++#endif + #if 0 + /* x^2048 + x^1638 + x^1231 + x^819 + x^411 + x + 1 -- 115 */ + { 2048, 1638, 1231, 819, 411, 1 }, +@@ -902,7 +914,7 @@ static ssize_t extract_entropy_user(stru + + extract_buf(r, tmp); + i = min_t(int, nbytes, EXTRACT_SIZE); +- if (copy_to_user(buf, tmp, i)) { ++ if (i > sizeof(tmp) || copy_to_user(buf, tmp, i)) { + ret = -EFAULT; + break; + } +@@ -1207,7 +1219,7 @@ EXPORT_SYMBOL(generate_random_uuid); + #include <linux/sysctl.h> + + static int min_read_thresh = 8, min_write_thresh; +-static int max_read_thresh = INPUT_POOL_WORDS * 32; ++static int max_read_thresh = OUTPUT_POOL_WORDS * 32; + static int max_write_thresh = INPUT_POOL_WORDS * 32; + static char sysctl_bootid[16]; + +diff -urNp linux-2.6.38.3/drivers/char/sonypi.c linux-2.6.38.3/drivers/char/sonypi.c +--- linux-2.6.38.3/drivers/char/sonypi.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/char/sonypi.c 2011-04-17 15:57:32.000000000 -0400 +@@ -55,6 +55,7 @@ + #include <asm/uaccess.h> + #include <asm/io.h> + #include <asm/system.h> ++#include <asm/local.h> + + #include <linux/sonypi.h> + +@@ -491,7 +492,7 @@ static struct sonypi_device { + spinlock_t fifo_lock; + wait_queue_head_t fifo_proc_list; + struct fasync_struct *fifo_async; +- int open_count; ++ local_t open_count; + int model; + struct input_dev *input_jog_dev; + struct input_dev *input_key_dev; +@@ -898,7 +899,7 @@ static int sonypi_misc_fasync(int fd, st + static int sonypi_misc_release(struct inode *inode, struct file *file) + { + mutex_lock(&sonypi_device.lock); +- sonypi_device.open_count--; ++ local_dec(&sonypi_device.open_count); + mutex_unlock(&sonypi_device.lock); + return 0; + } +@@ -907,9 +908,9 @@ static int sonypi_misc_open(struct inode + { + mutex_lock(&sonypi_device.lock); + /* Flush input queue on first open */ +- if (!sonypi_device.open_count) ++ if (!local_read(&sonypi_device.open_count)) + kfifo_reset(&sonypi_device.fifo); +- sonypi_device.open_count++; ++ local_inc(&sonypi_device.open_count); + mutex_unlock(&sonypi_device.lock); + + return 0; +diff -urNp linux-2.6.38.3/drivers/char/tpm/tpm_bios.c linux-2.6.38.3/drivers/char/tpm/tpm_bios.c +--- linux-2.6.38.3/drivers/char/tpm/tpm_bios.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/char/tpm/tpm_bios.c 2011-04-17 15:57:32.000000000 -0400 +@@ -173,7 +173,7 @@ static void *tpm_bios_measurements_start + event = addr; + + if ((event->event_type == 0 && event->event_size == 0) || +- ((addr + sizeof(struct tcpa_event) + event->event_size) >= limit)) ++ (event->event_size >= limit - addr - sizeof(struct tcpa_event))) + return NULL; + + return addr; +@@ -198,7 +198,7 @@ static void *tpm_bios_measurements_next( + return NULL; + + if ((event->event_type == 0 && event->event_size == 0) || +- ((v + sizeof(struct tcpa_event) + event->event_size) >= limit)) ++ (event->event_size >= limit - v - sizeof(struct tcpa_event))) + return NULL; + + (*pos)++; +@@ -291,7 +291,8 @@ static int tpm_binary_bios_measurements_ + int i; + + for (i = 0; i < sizeof(struct tcpa_event) + event->event_size; i++) +- seq_putc(m, data[i]); ++ if (!seq_putc(m, data[i])) ++ return -EFAULT; + + return 0; + } +@@ -410,6 +411,11 @@ static int read_log(struct tpm_bios_log + log->bios_event_log_end = log->bios_event_log + len; + + virt = acpi_os_map_memory(start, len); ++ if (!virt) { ++ kfree(log->bios_event_log); ++ log->bios_event_log = NULL; ++ return -EFAULT; ++ } + + memcpy(log->bios_event_log, virt, len); + +diff -urNp linux-2.6.38.3/drivers/char/tpm/tpm.c linux-2.6.38.3/drivers/char/tpm/tpm.c +--- linux-2.6.38.3/drivers/char/tpm/tpm.c 2011-04-17 16:53:01.000000000 -0400 ++++ linux-2.6.38.3/drivers/char/tpm/tpm.c 2011-04-17 16:53:16.000000000 -0400 +@@ -411,7 +411,7 @@ static ssize_t tpm_transmit(struct tpm_c + chip->vendor.req_complete_val) + goto out_recv; + +- if ((status == chip->vendor.req_canceled)) { ++ if (status == chip->vendor.req_canceled) { + dev_err(chip->dev, "Operation Canceled\n"); + rc = -ECANCELED; + goto out; +diff -urNp linux-2.6.38.3/drivers/cpuidle/sysfs.c linux-2.6.38.3/drivers/cpuidle/sysfs.c +--- linux-2.6.38.3/drivers/cpuidle/sysfs.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/cpuidle/sysfs.c 2011-04-17 15:57:32.000000000 -0400 +@@ -300,7 +300,7 @@ static struct kobj_type ktype_state_cpui + .release = cpuidle_state_sysfs_release, + }; + +-static void inline cpuidle_free_state_kobj(struct cpuidle_device *device, int i) ++static inline void cpuidle_free_state_kobj(struct cpuidle_device *device, int i) + { + kobject_put(&device->kobjs[i]->kobj); + wait_for_completion(&device->kobjs[i]->kobj_unregister); +diff -urNp linux-2.6.38.3/drivers/edac/edac_core.h linux-2.6.38.3/drivers/edac/edac_core.h +--- linux-2.6.38.3/drivers/edac/edac_core.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/edac/edac_core.h 2011-04-17 15:57:32.000000000 -0400 +@@ -88,11 +88,11 @@ extern int edac_debug_level; + + #else /* !CONFIG_EDAC_DEBUG */ + +-#define debugf0( ... ) +-#define debugf1( ... ) +-#define debugf2( ... ) +-#define debugf3( ... ) +-#define debugf4( ... ) ++#define debugf0( ... ) do {} while (0) ++#define debugf1( ... ) do {} while (0) ++#define debugf2( ... ) do {} while (0) ++#define debugf3( ... ) do {} while (0) ++#define debugf4( ... ) do {} while (0) + + #endif /* !CONFIG_EDAC_DEBUG */ + +diff -urNp linux-2.6.38.3/drivers/edac/edac_mc_sysfs.c linux-2.6.38.3/drivers/edac/edac_mc_sysfs.c +--- linux-2.6.38.3/drivers/edac/edac_mc_sysfs.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/edac/edac_mc_sysfs.c 2011-04-17 15:57:32.000000000 -0400 +@@ -761,7 +761,7 @@ static void edac_inst_grp_release(struct + } + + /* Intermediate show/store table */ +-static struct sysfs_ops inst_grp_ops = { ++static const struct sysfs_ops inst_grp_ops = { + .show = inst_grp_show, + .store = inst_grp_store + }; +diff -urNp linux-2.6.38.3/drivers/firewire/core-cdev.c linux-2.6.38.3/drivers/firewire/core-cdev.c +--- linux-2.6.38.3/drivers/firewire/core-cdev.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/firewire/core-cdev.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1329,8 +1329,7 @@ static int init_iso_resource(struct clie + int ret; + + if ((request->channels == 0 && request->bandwidth == 0) || +- request->bandwidth > BANDWIDTH_AVAILABLE_INITIAL || +- request->bandwidth < 0) ++ request->bandwidth > BANDWIDTH_AVAILABLE_INITIAL) + return -EINVAL; + + r = kmalloc(sizeof(*r), GFP_KERNEL); +diff -urNp linux-2.6.38.3/drivers/firmware/dmi_scan.c linux-2.6.38.3/drivers/firmware/dmi_scan.c +--- linux-2.6.38.3/drivers/firmware/dmi_scan.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/firmware/dmi_scan.c 2011-04-17 15:57:32.000000000 -0400 +@@ -449,11 +449,6 @@ void __init dmi_scan_machine(void) + } + } + else { +- /* +- * no iounmap() for that ioremap(); it would be a no-op, but +- * it's so early in setup that sucker gets confused into doing +- * what it shouldn't if we actually call it. +- */ + p = dmi_ioremap(0xF0000, 0x10000); + if (p == NULL) + goto error; +diff -urNp linux-2.6.38.3/drivers/gpu/drm/drm_crtc_helper.c linux-2.6.38.3/drivers/gpu/drm/drm_crtc_helper.c +--- linux-2.6.38.3/drivers/gpu/drm/drm_crtc_helper.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/drm_crtc_helper.c 2011-04-17 15:57:32.000000000 -0400 +@@ -276,7 +276,7 @@ static bool drm_encoder_crtc_ok(struct d + struct drm_crtc *tmp; + int crtc_mask = 1; + +- WARN(!crtc, "checking null crtc?\n"); ++ BUG_ON(!crtc); + + dev = crtc->dev; + +diff -urNp linux-2.6.38.3/drivers/gpu/drm/drm_drv.c linux-2.6.38.3/drivers/gpu/drm/drm_drv.c +--- linux-2.6.38.3/drivers/gpu/drm/drm_drv.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/drm_drv.c 2011-04-17 15:57:32.000000000 -0400 +@@ -425,7 +425,7 @@ long drm_ioctl(struct file *filp, + + dev = file_priv->minor->dev; + atomic_inc(&dev->ioctl_count); +- atomic_inc(&dev->counts[_DRM_STAT_IOCTLS]); ++ atomic_inc_unchecked(&dev->counts[_DRM_STAT_IOCTLS]); + ++file_priv->ioctl_count; + + DRM_DEBUG("pid=%d, cmd=0x%02x, nr=0x%02x, dev 0x%lx, auth=%d\n", +diff -urNp linux-2.6.38.3/drivers/gpu/drm/drm_fops.c linux-2.6.38.3/drivers/gpu/drm/drm_fops.c +--- linux-2.6.38.3/drivers/gpu/drm/drm_fops.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/drm_fops.c 2011-04-17 15:57:32.000000000 -0400 +@@ -70,7 +70,7 @@ static int drm_setup(struct drm_device * + } + + for (i = 0; i < ARRAY_SIZE(dev->counts); i++) +- atomic_set(&dev->counts[i], 0); ++ atomic_set_unchecked(&dev->counts[i], 0); + + dev->sigdata.lock = NULL; + +@@ -134,8 +134,8 @@ int drm_open(struct inode *inode, struct + + retcode = drm_open_helper(inode, filp, dev); + if (!retcode) { +- atomic_inc(&dev->counts[_DRM_STAT_OPENS]); +- if (!dev->open_count++) ++ atomic_inc_unchecked(&dev->counts[_DRM_STAT_OPENS]); ++ if (local_inc_return(&dev->open_count) == 1) + retcode = drm_setup(dev); + } + if (!retcode) { +@@ -472,7 +472,7 @@ int drm_release(struct inode *inode, str + + mutex_lock(&drm_global_mutex); + +- DRM_DEBUG("open_count = %d\n", dev->open_count); ++ DRM_DEBUG("open_count = %d\n", local_read(&dev->open_count)); + + if (dev->driver->preclose) + dev->driver->preclose(dev, file_priv); +@@ -484,7 +484,7 @@ int drm_release(struct inode *inode, str + DRM_DEBUG("pid = %d, device = 0x%lx, open_count = %d\n", + task_pid_nr(current), + (long)old_encode_dev(file_priv->minor->device), +- dev->open_count); ++ local_read(&dev->open_count)); + + /* if the master has gone away we can't do anything with the lock */ + if (file_priv->minor->master) +@@ -565,8 +565,8 @@ int drm_release(struct inode *inode, str + * End inline drm_release + */ + +- atomic_inc(&dev->counts[_DRM_STAT_CLOSES]); +- if (!--dev->open_count) { ++ atomic_inc_unchecked(&dev->counts[_DRM_STAT_CLOSES]); ++ if (local_dec_and_test(&dev->open_count)) { + if (atomic_read(&dev->ioctl_count)) { + DRM_ERROR("Device busy: %d\n", + atomic_read(&dev->ioctl_count)); +diff -urNp linux-2.6.38.3/drivers/gpu/drm/drm_global.c linux-2.6.38.3/drivers/gpu/drm/drm_global.c +--- linux-2.6.38.3/drivers/gpu/drm/drm_global.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/drm_global.c 2011-04-17 15:57:32.000000000 -0400 +@@ -36,7 +36,7 @@ + struct drm_global_item { + struct mutex mutex; + void *object; +- int refcount; ++ atomic_t refcount; + }; + + static struct drm_global_item glob[DRM_GLOBAL_NUM]; +@@ -49,7 +49,7 @@ void drm_global_init(void) + struct drm_global_item *item = &glob[i]; + mutex_init(&item->mutex); + item->object = NULL; +- item->refcount = 0; ++ atomic_set(&item->refcount, 0); + } + } + +@@ -59,7 +59,7 @@ void drm_global_release(void) + for (i = 0; i < DRM_GLOBAL_NUM; ++i) { + struct drm_global_item *item = &glob[i]; + BUG_ON(item->object != NULL); +- BUG_ON(item->refcount != 0); ++ BUG_ON(atomic_read(&item->refcount) != 0); + } + } + +@@ -70,7 +70,7 @@ int drm_global_item_ref(struct drm_globa + void *object; + + mutex_lock(&item->mutex); +- if (item->refcount == 0) { ++ if (atomic_read(&item->refcount) == 0) { + item->object = kzalloc(ref->size, GFP_KERNEL); + if (unlikely(item->object == NULL)) { + ret = -ENOMEM; +@@ -83,7 +83,7 @@ int drm_global_item_ref(struct drm_globa + goto out_err; + + } +- ++item->refcount; ++ atomic_inc(&item->refcount); + ref->object = item->object; + object = item->object; + mutex_unlock(&item->mutex); +@@ -100,9 +100,9 @@ void drm_global_item_unref(struct drm_gl + struct drm_global_item *item = &glob[ref->global_type]; + + mutex_lock(&item->mutex); +- BUG_ON(item->refcount == 0); ++ BUG_ON(atomic_read(&item->refcount) == 0); + BUG_ON(ref->object != item->object); +- if (--item->refcount == 0) { ++ if (atomic_dec_and_test(&item->refcount)) { + ref->release(ref); + item->object = NULL; + } +diff -urNp linux-2.6.38.3/drivers/gpu/drm/drm_info.c linux-2.6.38.3/drivers/gpu/drm/drm_info.c +--- linux-2.6.38.3/drivers/gpu/drm/drm_info.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/drm_info.c 2011-04-17 15:57:32.000000000 -0400 +@@ -86,10 +86,14 @@ int drm_vm_info(struct seq_file *m, void + struct drm_local_map *map; + struct drm_map_list *r_list; + +- /* Hardcoded from _DRM_FRAME_BUFFER, +- _DRM_REGISTERS, _DRM_SHM, _DRM_AGP, and +- _DRM_SCATTER_GATHER and _DRM_CONSISTENT */ +- const char *types[] = { "FB", "REG", "SHM", "AGP", "SG", "PCI" }; ++ static const char * const types[] = { ++ [_DRM_FRAME_BUFFER] = "FB", ++ [_DRM_REGISTERS] = "REG", ++ [_DRM_SHM] = "SHM", ++ [_DRM_AGP] = "AGP", ++ [_DRM_SCATTER_GATHER] = "SG", ++ [_DRM_CONSISTENT] = "PCI", ++ [_DRM_GEM] = "GEM" }; + const char *type; + int i; + +@@ -100,7 +104,7 @@ int drm_vm_info(struct seq_file *m, void + map = r_list->map; + if (!map) + continue; +- if (map->type < 0 || map->type > 5) ++ if (map->type >= ARRAY_SIZE(types)) + type = "??"; + else + type = types[map->type]; +@@ -301,7 +305,11 @@ int drm_vma_info(struct seq_file *m, voi + vma->vm_flags & VM_MAYSHARE ? 's' : 'p', + vma->vm_flags & VM_LOCKED ? 'l' : '-', + vma->vm_flags & VM_IO ? 'i' : '-', ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ 0); ++#else + vma->vm_pgoff); ++#endif + + #if defined(__i386__) + pgprot = pgprot_val(vma->vm_page_prot); +diff -urNp linux-2.6.38.3/drivers/gpu/drm/drm_ioctl.c linux-2.6.38.3/drivers/gpu/drm/drm_ioctl.c +--- linux-2.6.38.3/drivers/gpu/drm/drm_ioctl.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/drm_ioctl.c 2011-04-17 15:57:32.000000000 -0400 +@@ -353,7 +353,7 @@ int drm_getstats(struct drm_device *dev, + stats->data[i].value = + (file_priv->master->lock.hw_lock ? file_priv->master->lock.hw_lock->lock : 0); + else +- stats->data[i].value = atomic_read(&dev->counts[i]); ++ stats->data[i].value = atomic_read_unchecked(&dev->counts[i]); + stats->data[i].type = dev->types[i]; + } + +diff -urNp linux-2.6.38.3/drivers/gpu/drm/drm_lock.c linux-2.6.38.3/drivers/gpu/drm/drm_lock.c +--- linux-2.6.38.3/drivers/gpu/drm/drm_lock.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/drm_lock.c 2011-04-17 15:57:32.000000000 -0400 +@@ -89,7 +89,7 @@ int drm_lock(struct drm_device *dev, voi + if (drm_lock_take(&master->lock, lock->context)) { + master->lock.file_priv = file_priv; + master->lock.lock_time = jiffies; +- atomic_inc(&dev->counts[_DRM_STAT_LOCKS]); ++ atomic_inc_unchecked(&dev->counts[_DRM_STAT_LOCKS]); + break; /* Got lock */ + } + +@@ -160,7 +160,7 @@ int drm_unlock(struct drm_device *dev, v + return -EINVAL; + } + +- atomic_inc(&dev->counts[_DRM_STAT_UNLOCKS]); ++ atomic_inc_unchecked(&dev->counts[_DRM_STAT_UNLOCKS]); + + if (drm_lock_free(&master->lock, lock->context)) { + /* FIXME: Should really bail out here. */ +diff -urNp linux-2.6.38.3/drivers/gpu/drm/i810/i810_dma.c linux-2.6.38.3/drivers/gpu/drm/i810/i810_dma.c +--- linux-2.6.38.3/drivers/gpu/drm/i810/i810_dma.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/i810/i810_dma.c 2011-04-17 15:57:32.000000000 -0400 +@@ -953,8 +953,8 @@ static int i810_dma_vertex(struct drm_de + dma->buflist[vertex->idx], + vertex->discard, vertex->used); + +- atomic_add(vertex->used, &dev->counts[_DRM_STAT_SECONDARY]); +- atomic_inc(&dev->counts[_DRM_STAT_DMA]); ++ atomic_add_unchecked(vertex->used, &dev->counts[_DRM_STAT_SECONDARY]); ++ atomic_inc_unchecked(&dev->counts[_DRM_STAT_DMA]); + sarea_priv->last_enqueue = dev_priv->counter - 1; + sarea_priv->last_dispatch = (int)hw_status[5]; + +@@ -1114,8 +1114,8 @@ static int i810_dma_mc(struct drm_device + i810_dma_dispatch_mc(dev, dma->buflist[mc->idx], mc->used, + mc->last_render); + +- atomic_add(mc->used, &dev->counts[_DRM_STAT_SECONDARY]); +- atomic_inc(&dev->counts[_DRM_STAT_DMA]); ++ atomic_add_unchecked(mc->used, &dev->counts[_DRM_STAT_SECONDARY]); ++ atomic_inc_unchecked(&dev->counts[_DRM_STAT_DMA]); + sarea_priv->last_enqueue = dev_priv->counter - 1; + sarea_priv->last_dispatch = (int)hw_status[5]; + +diff -urNp linux-2.6.38.3/drivers/gpu/drm/i915/dvo_ch7017.c linux-2.6.38.3/drivers/gpu/drm/i915/dvo_ch7017.c +--- linux-2.6.38.3/drivers/gpu/drm/i915/dvo_ch7017.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/i915/dvo_ch7017.c 2011-04-17 15:57:32.000000000 -0400 +@@ -390,7 +390,7 @@ static void ch7017_destroy(struct intel_ + } + } + +-struct intel_dvo_dev_ops ch7017_ops = { ++const struct intel_dvo_dev_ops ch7017_ops = { + .init = ch7017_init, + .detect = ch7017_detect, + .mode_valid = ch7017_mode_valid, +diff -urNp linux-2.6.38.3/drivers/gpu/drm/i915/dvo_ch7xxx.c linux-2.6.38.3/drivers/gpu/drm/i915/dvo_ch7xxx.c +--- linux-2.6.38.3/drivers/gpu/drm/i915/dvo_ch7xxx.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/i915/dvo_ch7xxx.c 2011-04-17 15:57:32.000000000 -0400 +@@ -320,7 +320,7 @@ static void ch7xxx_destroy(struct intel_ + } + } + +-struct intel_dvo_dev_ops ch7xxx_ops = { ++const struct intel_dvo_dev_ops ch7xxx_ops = { + .init = ch7xxx_init, + .detect = ch7xxx_detect, + .mode_valid = ch7xxx_mode_valid, +diff -urNp linux-2.6.38.3/drivers/gpu/drm/i915/dvo.h linux-2.6.38.3/drivers/gpu/drm/i915/dvo.h +--- linux-2.6.38.3/drivers/gpu/drm/i915/dvo.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/i915/dvo.h 2011-04-17 15:57:32.000000000 -0400 +@@ -122,23 +122,23 @@ struct intel_dvo_dev_ops { + * + * \return singly-linked list of modes or NULL if no modes found. + */ +- struct drm_display_mode *(*get_modes)(struct intel_dvo_device *dvo); ++ struct drm_display_mode *(* const get_modes)(struct intel_dvo_device *dvo); + + /** + * Clean up driver-specific bits of the output + */ +- void (*destroy) (struct intel_dvo_device *dvo); ++ void (* const destroy) (struct intel_dvo_device *dvo); + + /** + * Debugging hook to dump device registers to log file + */ +- void (*dump_regs)(struct intel_dvo_device *dvo); ++ void (* const dump_regs)(struct intel_dvo_device *dvo); + }; + +-extern struct intel_dvo_dev_ops sil164_ops; +-extern struct intel_dvo_dev_ops ch7xxx_ops; +-extern struct intel_dvo_dev_ops ivch_ops; +-extern struct intel_dvo_dev_ops tfp410_ops; +-extern struct intel_dvo_dev_ops ch7017_ops; ++extern const struct intel_dvo_dev_ops sil164_ops; ++extern const struct intel_dvo_dev_ops ch7xxx_ops; ++extern const struct intel_dvo_dev_ops ivch_ops; ++extern const struct intel_dvo_dev_ops tfp410_ops; ++extern const struct intel_dvo_dev_ops ch7017_ops; + + #endif /* _INTEL_DVO_H */ +diff -urNp linux-2.6.38.3/drivers/gpu/drm/i915/dvo_ivch.c linux-2.6.38.3/drivers/gpu/drm/i915/dvo_ivch.c +--- linux-2.6.38.3/drivers/gpu/drm/i915/dvo_ivch.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/i915/dvo_ivch.c 2011-04-17 15:57:32.000000000 -0400 +@@ -410,7 +410,7 @@ static void ivch_destroy(struct intel_dv + } + } + +-struct intel_dvo_dev_ops ivch_ops= { ++const struct intel_dvo_dev_ops ivch_ops= { + .init = ivch_init, + .dpms = ivch_dpms, + .mode_valid = ivch_mode_valid, +diff -urNp linux-2.6.38.3/drivers/gpu/drm/i915/dvo_sil164.c linux-2.6.38.3/drivers/gpu/drm/i915/dvo_sil164.c +--- linux-2.6.38.3/drivers/gpu/drm/i915/dvo_sil164.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/i915/dvo_sil164.c 2011-04-17 15:57:32.000000000 -0400 +@@ -252,7 +252,7 @@ static void sil164_destroy(struct intel_ + } + } + +-struct intel_dvo_dev_ops sil164_ops = { ++const struct intel_dvo_dev_ops sil164_ops = { + .init = sil164_init, + .detect = sil164_detect, + .mode_valid = sil164_mode_valid, +diff -urNp linux-2.6.38.3/drivers/gpu/drm/i915/dvo_tfp410.c linux-2.6.38.3/drivers/gpu/drm/i915/dvo_tfp410.c +--- linux-2.6.38.3/drivers/gpu/drm/i915/dvo_tfp410.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/i915/dvo_tfp410.c 2011-04-17 15:57:32.000000000 -0400 +@@ -293,7 +293,7 @@ static void tfp410_destroy(struct intel_ + } + } + +-struct intel_dvo_dev_ops tfp410_ops = { ++const struct intel_dvo_dev_ops tfp410_ops = { + .init = tfp410_init, + .detect = tfp410_detect, + .mode_valid = tfp410_mode_valid, +diff -urNp linux-2.6.38.3/drivers/gpu/drm/i915/i915_dma.c linux-2.6.38.3/drivers/gpu/drm/i915/i915_dma.c +--- linux-2.6.38.3/drivers/gpu/drm/i915/i915_dma.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/i915/i915_dma.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1159,7 +1159,7 @@ static bool i915_switcheroo_can_switch(s + bool can_switch; + + spin_lock(&dev->count_lock); +- can_switch = (dev->open_count == 0); ++ can_switch = (local_read(&dev->open_count) == 0); + spin_unlock(&dev->count_lock); + return can_switch; + } +diff -urNp linux-2.6.38.3/drivers/gpu/drm/i915/i915_drv.c linux-2.6.38.3/drivers/gpu/drm/i915/i915_drv.c +--- linux-2.6.38.3/drivers/gpu/drm/i915/i915_drv.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/i915/i915_drv.c 2011-04-17 15:57:32.000000000 -0400 +@@ -673,7 +673,7 @@ static const struct dev_pm_ops i915_pm_o + .restore = i915_pm_resume, + }; + +-static struct vm_operations_struct i915_gem_vm_ops = { ++static const struct vm_operations_struct i915_gem_vm_ops = { + .fault = i915_gem_fault, + .open = drm_gem_vm_open, + .close = drm_gem_vm_close, +diff -urNp linux-2.6.38.3/drivers/gpu/drm/i915/i915_drv.h linux-2.6.38.3/drivers/gpu/drm/i915/i915_drv.h +--- linux-2.6.38.3/drivers/gpu/drm/i915/i915_drv.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/i915/i915_drv.h 2011-04-17 15:57:32.000000000 -0400 +@@ -1229,7 +1229,7 @@ extern int intel_setup_gmbus(struct drm_ + extern void intel_teardown_gmbus(struct drm_device *dev); + extern void intel_gmbus_set_speed(struct i2c_adapter *adapter, int speed); + extern void intel_gmbus_force_bit(struct i2c_adapter *adapter, bool force_bit); +-extern inline bool intel_gmbus_is_forced_bit(struct i2c_adapter *adapter) ++static inline bool intel_gmbus_is_forced_bit(struct i2c_adapter *adapter) + { + return container_of(adapter, struct intel_gmbus, adapter)->force_bit; + } +diff -urNp linux-2.6.38.3/drivers/gpu/drm/nouveau/nouveau_state.c linux-2.6.38.3/drivers/gpu/drm/nouveau/nouveau_state.c +--- linux-2.6.38.3/drivers/gpu/drm/nouveau/nouveau_state.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/nouveau/nouveau_state.c 2011-04-17 15:57:32.000000000 -0400 +@@ -621,7 +621,7 @@ static bool nouveau_switcheroo_can_switc + bool can_switch; + + spin_lock(&dev->count_lock); +- can_switch = (dev->open_count == 0); ++ can_switch = (local_read(&dev->open_count) == 0); + spin_unlock(&dev->count_lock); + return can_switch; + } +diff -urNp linux-2.6.38.3/drivers/gpu/drm/radeon/mkregtable.c linux-2.6.38.3/drivers/gpu/drm/radeon/mkregtable.c +--- linux-2.6.38.3/drivers/gpu/drm/radeon/mkregtable.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/radeon/mkregtable.c 2011-04-17 15:57:32.000000000 -0400 +@@ -637,14 +637,14 @@ static int parser_auth(struct table *t, + regex_t mask_rex; + regmatch_t match[4]; + char buf[1024]; +- size_t end; ++ long end; + int len; + int done = 0; + int r; + unsigned o; + struct offset *offset; + char last_reg_s[10]; +- int last_reg; ++ unsigned long last_reg; + + if (regcomp + (&mask_rex, "(0x[0-9a-fA-F]*) *([_a-zA-Z0-9]*)", REG_EXTENDED)) { +diff -urNp linux-2.6.38.3/drivers/gpu/drm/radeon/radeon_device.c linux-2.6.38.3/drivers/gpu/drm/radeon/radeon_device.c +--- linux-2.6.38.3/drivers/gpu/drm/radeon/radeon_device.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/radeon/radeon_device.c 2011-04-17 15:57:32.000000000 -0400 +@@ -673,7 +673,7 @@ static bool radeon_switcheroo_can_switch + bool can_switch; + + spin_lock(&dev->count_lock); +- can_switch = (dev->open_count == 0); ++ can_switch = (local_read(&dev->open_count) == 0); + spin_unlock(&dev->count_lock); + return can_switch; + } +diff -urNp linux-2.6.38.3/drivers/gpu/drm/radeon/radeon_ioc32.c linux-2.6.38.3/drivers/gpu/drm/radeon/radeon_ioc32.c +--- linux-2.6.38.3/drivers/gpu/drm/radeon/radeon_ioc32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/radeon/radeon_ioc32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -359,7 +359,7 @@ static int compat_radeon_cp_setparam(str + request = compat_alloc_user_space(sizeof(*request)); + if (!access_ok(VERIFY_WRITE, request, sizeof(*request)) + || __put_user(req32.param, &request->param) +- || __put_user((void __user *)(unsigned long)req32.value, ++ || __put_user((unsigned long)req32.value, + &request->value)) + return -EFAULT; + +diff -urNp linux-2.6.38.3/drivers/gpu/drm/radeon/radeon_state.c linux-2.6.38.3/drivers/gpu/drm/radeon/radeon_state.c +--- linux-2.6.38.3/drivers/gpu/drm/radeon/radeon_state.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/radeon/radeon_state.c 2011-04-17 15:57:32.000000000 -0400 +@@ -2168,7 +2168,7 @@ static int radeon_cp_clear(struct drm_de + if (sarea_priv->nbox > RADEON_NR_SAREA_CLIPRECTS) + sarea_priv->nbox = RADEON_NR_SAREA_CLIPRECTS; + +- if (DRM_COPY_FROM_USER(&depth_boxes, clear->depth_boxes, ++ if (sarea_priv->nbox > RADEON_NR_SAREA_CLIPRECTS || DRM_COPY_FROM_USER(&depth_boxes, clear->depth_boxes, + sarea_priv->nbox * sizeof(depth_boxes[0]))) + return -EFAULT; + +@@ -3031,7 +3031,7 @@ static int radeon_cp_getparam(struct drm + { + drm_radeon_private_t *dev_priv = dev->dev_private; + drm_radeon_getparam_t *param = data; +- int value; ++ int value = 0; + + DRM_DEBUG("pid=%d\n", DRM_CURRENTPID); + +diff -urNp linux-2.6.38.3/drivers/gpu/drm/radeon/radeon_ttm.c linux-2.6.38.3/drivers/gpu/drm/radeon/radeon_ttm.c +--- linux-2.6.38.3/drivers/gpu/drm/radeon/radeon_ttm.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/radeon/radeon_ttm.c 2011-04-17 15:57:32.000000000 -0400 +@@ -603,8 +603,9 @@ void radeon_ttm_set_active_vram_size(str + man->size = size >> PAGE_SHIFT; + } + +-static struct vm_operations_struct radeon_ttm_vm_ops; +-static const struct vm_operations_struct *ttm_vm_ops = NULL; ++extern int ttm_bo_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf); ++extern void ttm_bo_vm_open(struct vm_area_struct *vma); ++extern void ttm_bo_vm_close(struct vm_area_struct *vma); + + static int radeon_ttm_fault(struct vm_area_struct *vma, struct vm_fault *vmf) + { +@@ -612,17 +613,22 @@ static int radeon_ttm_fault(struct vm_ar + struct radeon_device *rdev; + int r; + +- bo = (struct ttm_buffer_object *)vma->vm_private_data; +- if (bo == NULL) { ++ bo = (struct ttm_buffer_object *)vma->vm_private_data; ++ if (!bo) + return VM_FAULT_NOPAGE; +- } + rdev = radeon_get_rdev(bo->bdev); + mutex_lock(&rdev->vram_mutex); +- r = ttm_vm_ops->fault(vma, vmf); ++ r = ttm_bo_vm_fault(vma, vmf); + mutex_unlock(&rdev->vram_mutex); + return r; + } + ++static const struct vm_operations_struct radeon_ttm_vm_ops = { ++ .fault = radeon_ttm_fault, ++ .open = ttm_bo_vm_open, ++ .close = ttm_bo_vm_close ++}; ++ + int radeon_mmap(struct file *filp, struct vm_area_struct *vma) + { + struct drm_file *file_priv; +@@ -635,18 +641,11 @@ int radeon_mmap(struct file *filp, struc + + file_priv = filp->private_data; + rdev = file_priv->minor->dev->dev_private; +- if (rdev == NULL) { ++ if (!rdev) + return -EINVAL; +- } + r = ttm_bo_mmap(filp, vma, &rdev->mman.bdev); +- if (unlikely(r != 0)) { ++ if (r) + return r; +- } +- if (unlikely(ttm_vm_ops == NULL)) { +- ttm_vm_ops = vma->vm_ops; +- radeon_ttm_vm_ops = *ttm_vm_ops; +- radeon_ttm_vm_ops.fault = &radeon_ttm_fault; +- } + vma->vm_ops = &radeon_ttm_vm_ops; + return 0; + } +diff -urNp linux-2.6.38.3/drivers/gpu/drm/radeon/rs690.c linux-2.6.38.3/drivers/gpu/drm/radeon/rs690.c +--- linux-2.6.38.3/drivers/gpu/drm/radeon/rs690.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/radeon/rs690.c 2011-04-17 15:57:32.000000000 -0400 +@@ -304,9 +304,11 @@ void rs690_crtc_bandwidth_compute(struct + if (rdev->pm.max_bandwidth.full > rdev->pm.sideport_bandwidth.full && + rdev->pm.sideport_bandwidth.full) + rdev->pm.max_bandwidth = rdev->pm.sideport_bandwidth; +- read_delay_latency.full = dfixed_const(370 * 800 * 1000); ++ read_delay_latency.full = dfixed_const(800 * 1000); + read_delay_latency.full = dfixed_div(read_delay_latency, + rdev->pm.igp_sideport_mclk); ++ a.full = dfixed_const(370); ++ read_delay_latency.full = dfixed_mul(read_delay_latency, a); + } else { + if (rdev->pm.max_bandwidth.full > rdev->pm.k8_bandwidth.full && + rdev->pm.k8_bandwidth.full) +diff -urNp linux-2.6.38.3/drivers/gpu/drm/ttm/ttm_bo.c linux-2.6.38.3/drivers/gpu/drm/ttm/ttm_bo.c +--- linux-2.6.38.3/drivers/gpu/drm/ttm/ttm_bo.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/ttm/ttm_bo.c 2011-04-17 15:57:32.000000000 -0400 +@@ -40,7 +40,7 @@ + #include <asm/atomic.h> + + #define TTM_ASSERT_LOCKED(param) +-#define TTM_DEBUG(fmt, arg...) ++#define TTM_DEBUG(fmt, arg...) do {} while (0) + #define TTM_BO_HASH_ORDER 13 + + static int ttm_bo_setup_vm(struct ttm_buffer_object *bo); +diff -urNp linux-2.6.38.3/drivers/gpu/drm/ttm/ttm_bo_vm.c linux-2.6.38.3/drivers/gpu/drm/ttm/ttm_bo_vm.c +--- linux-2.6.38.3/drivers/gpu/drm/ttm/ttm_bo_vm.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/ttm/ttm_bo_vm.c 2011-04-17 15:57:32.000000000 -0400 +@@ -69,11 +69,11 @@ static struct ttm_buffer_object *ttm_bo_ + return best_bo; + } + +-static int ttm_bo_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf) ++int ttm_bo_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf) + { + struct ttm_buffer_object *bo = (struct ttm_buffer_object *) + vma->vm_private_data; +- struct ttm_bo_device *bdev = bo->bdev; ++ struct ttm_bo_device *bdev; + unsigned long page_offset; + unsigned long page_last; + unsigned long pfn; +@@ -83,8 +83,12 @@ static int ttm_bo_vm_fault(struct vm_are + int i; + unsigned long address = (unsigned long)vmf->virtual_address; + int retval = VM_FAULT_NOPAGE; +- struct ttm_mem_type_manager *man = +- &bdev->man[bo->mem.mem_type]; ++ struct ttm_mem_type_manager *man; ++ ++ if (!bo) ++ return VM_FAULT_NOPAGE; ++ bdev = bo->bdev; ++ man = &bdev->man[bo->mem.mem_type]; + + /* + * Work around locking order reversal in fault / nopfn +@@ -219,22 +223,25 @@ out_unlock: + ttm_bo_unreserve(bo); + return retval; + } ++EXPORT_SYMBOL(ttm_bo_vm_fault); + +-static void ttm_bo_vm_open(struct vm_area_struct *vma) ++void ttm_bo_vm_open(struct vm_area_struct *vma) + { + struct ttm_buffer_object *bo = + (struct ttm_buffer_object *)vma->vm_private_data; + + (void)ttm_bo_reference(bo); + } ++EXPORT_SYMBOL(ttm_bo_vm_open); + +-static void ttm_bo_vm_close(struct vm_area_struct *vma) ++void ttm_bo_vm_close(struct vm_area_struct *vma) + { + struct ttm_buffer_object *bo = (struct ttm_buffer_object *)vma->vm_private_data; + + ttm_bo_unref(&bo); + vma->vm_private_data = NULL; + } ++EXPORT_SYMBOL(ttm_bo_vm_close); + + static const struct vm_operations_struct ttm_bo_vm_ops = { + .fault = ttm_bo_vm_fault, +diff -urNp linux-2.6.38.3/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c linux-2.6.38.3/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c +--- linux-2.6.38.3/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c 2011-04-17 15:57:32.000000000 -0400 +@@ -534,7 +534,7 @@ static int vmw_fifo_vm_fault(struct vm_a + return VM_FAULT_SIGBUS; + } + +-static struct vm_operations_struct vmw_fifo_vm_ops = { ++static const struct vm_operations_struct vmw_fifo_vm_ops = { + .fault = vmw_fifo_vm_fault, + .open = NULL, + .close = NULL +diff -urNp linux-2.6.38.3/drivers/hid/usbhid/hiddev.c linux-2.6.38.3/drivers/hid/usbhid/hiddev.c +--- linux-2.6.38.3/drivers/hid/usbhid/hiddev.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/hid/usbhid/hiddev.c 2011-04-17 15:57:32.000000000 -0400 +@@ -613,7 +613,7 @@ static long hiddev_ioctl(struct file *fi + break; + + case HIDIOCAPPLICATION: +- if (arg < 0 || arg >= hid->maxapplication) ++ if (arg >= hid->maxapplication) + break; + + for (i = 0; i < hid->maxcollection; i++) +diff -urNp linux-2.6.38.3/drivers/hwmon/k8temp.c linux-2.6.38.3/drivers/hwmon/k8temp.c +--- linux-2.6.38.3/drivers/hwmon/k8temp.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/hwmon/k8temp.c 2011-04-17 15:57:32.000000000 -0400 +@@ -138,7 +138,7 @@ static DEVICE_ATTR(name, S_IRUGO, show_n + + static const struct pci_device_id k8temp_ids[] = { + { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_K8_NB_MISC) }, +- { 0 }, ++ { 0, 0, 0, 0, 0, 0, 0 }, + }; + + MODULE_DEVICE_TABLE(pci, k8temp_ids); +diff -urNp linux-2.6.38.3/drivers/hwmon/sis5595.c linux-2.6.38.3/drivers/hwmon/sis5595.c +--- linux-2.6.38.3/drivers/hwmon/sis5595.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/hwmon/sis5595.c 2011-04-17 15:57:32.000000000 -0400 +@@ -701,7 +701,7 @@ static struct sis5595_data *sis5595_upda + + static const struct pci_device_id sis5595_pci_ids[] = { + { PCI_DEVICE(PCI_VENDOR_ID_SI, PCI_DEVICE_ID_SI_503) }, +- { 0, } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(pci, sis5595_pci_ids); +diff -urNp linux-2.6.38.3/drivers/hwmon/via686a.c linux-2.6.38.3/drivers/hwmon/via686a.c +--- linux-2.6.38.3/drivers/hwmon/via686a.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/hwmon/via686a.c 2011-04-17 15:57:32.000000000 -0400 +@@ -779,7 +779,7 @@ static struct via686a_data *via686a_upda + + static const struct pci_device_id via686a_pci_ids[] = { + { PCI_DEVICE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C686_4) }, +- { 0, } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(pci, via686a_pci_ids); +diff -urNp linux-2.6.38.3/drivers/hwmon/vt8231.c linux-2.6.38.3/drivers/hwmon/vt8231.c +--- linux-2.6.38.3/drivers/hwmon/vt8231.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/hwmon/vt8231.c 2011-04-17 15:57:32.000000000 -0400 +@@ -701,7 +701,7 @@ static struct platform_driver vt8231_dri + + static const struct pci_device_id vt8231_pci_ids[] = { + { PCI_DEVICE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8231_4) }, +- { 0, } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(pci, vt8231_pci_ids); +diff -urNp linux-2.6.38.3/drivers/hwmon/w83791d.c linux-2.6.38.3/drivers/hwmon/w83791d.c +--- linux-2.6.38.3/drivers/hwmon/w83791d.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/hwmon/w83791d.c 2011-04-17 15:57:32.000000000 -0400 +@@ -329,8 +329,8 @@ static int w83791d_detect(struct i2c_cli + struct i2c_board_info *info); + static int w83791d_remove(struct i2c_client *client); + +-static int w83791d_read(struct i2c_client *client, u8 register); +-static int w83791d_write(struct i2c_client *client, u8 register, u8 value); ++static int w83791d_read(struct i2c_client *client, u8 reg); ++static int w83791d_write(struct i2c_client *client, u8 reg, u8 value); + static struct w83791d_data *w83791d_update_device(struct device *dev); + + #ifdef DEBUG +diff -urNp linux-2.6.38.3/drivers/i2c/busses/i2c-i801.c linux-2.6.38.3/drivers/i2c/busses/i2c-i801.c +--- linux-2.6.38.3/drivers/i2c/busses/i2c-i801.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/i2c/busses/i2c-i801.c 2011-04-17 15:57:32.000000000 -0400 +@@ -621,7 +621,7 @@ static const struct pci_device_id i801_i + { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_PATSBURG_SMBUS_IDF0) }, + { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_PATSBURG_SMBUS_IDF1) }, + { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_PATSBURG_SMBUS_IDF2) }, +- { 0, } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(pci, i801_ids); +diff -urNp linux-2.6.38.3/drivers/i2c/busses/i2c-piix4.c linux-2.6.38.3/drivers/i2c/busses/i2c-piix4.c +--- linux-2.6.38.3/drivers/i2c/busses/i2c-piix4.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/i2c/busses/i2c-piix4.c 2011-04-17 15:57:32.000000000 -0400 +@@ -124,7 +124,7 @@ static struct dmi_system_id __devinitdat + .ident = "IBM", + .matches = { DMI_MATCH(DMI_SYS_VENDOR, "IBM"), }, + }, +- { }, ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } + }; + + static int __devinit piix4_setup(struct pci_dev *PIIX4_dev, +@@ -491,7 +491,7 @@ static const struct pci_device_id piix4_ + PCI_DEVICE_ID_SERVERWORKS_HT1000SB) }, + { PCI_DEVICE(PCI_VENDOR_ID_SERVERWORKS, + PCI_DEVICE_ID_SERVERWORKS_HT1100LD) }, +- { 0, } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE (pci, piix4_ids); +diff -urNp linux-2.6.38.3/drivers/i2c/busses/i2c-sis630.c linux-2.6.38.3/drivers/i2c/busses/i2c-sis630.c +--- linux-2.6.38.3/drivers/i2c/busses/i2c-sis630.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/i2c/busses/i2c-sis630.c 2011-04-17 15:57:32.000000000 -0400 +@@ -471,7 +471,7 @@ static struct i2c_adapter sis630_adapter + static const struct pci_device_id sis630_ids[] __devinitconst = { + { PCI_DEVICE(PCI_VENDOR_ID_SI, PCI_DEVICE_ID_SI_503) }, + { PCI_DEVICE(PCI_VENDOR_ID_SI, PCI_DEVICE_ID_SI_LPC) }, +- { 0, } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE (pci, sis630_ids); +diff -urNp linux-2.6.38.3/drivers/i2c/busses/i2c-sis96x.c linux-2.6.38.3/drivers/i2c/busses/i2c-sis96x.c +--- linux-2.6.38.3/drivers/i2c/busses/i2c-sis96x.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/i2c/busses/i2c-sis96x.c 2011-04-17 15:57:32.000000000 -0400 +@@ -247,7 +247,7 @@ static struct i2c_adapter sis96x_adapter + + static const struct pci_device_id sis96x_ids[] = { + { PCI_DEVICE(PCI_VENDOR_ID_SI, PCI_DEVICE_ID_SI_SMBUS) }, +- { 0, } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE (pci, sis96x_ids); +diff -urNp linux-2.6.38.3/drivers/ide/ide-cd.c linux-2.6.38.3/drivers/ide/ide-cd.c +--- linux-2.6.38.3/drivers/ide/ide-cd.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/ide/ide-cd.c 2011-04-17 15:57:32.000000000 -0400 +@@ -776,7 +776,7 @@ static void cdrom_do_block_pc(ide_drive_ + alignment = queue_dma_alignment(q) | q->dma_pad_mask; + if ((unsigned long)buf & alignment + || blk_rq_bytes(rq) & q->dma_pad_mask +- || object_is_on_stack(buf)) ++ || object_starts_on_stack(buf)) + drive->dma = 0; + } + } +diff -urNp linux-2.6.38.3/drivers/infiniband/core/cm.c linux-2.6.38.3/drivers/infiniband/core/cm.c +--- linux-2.6.38.3/drivers/infiniband/core/cm.c 2011-03-23 17:20:07.000000000 -0400 ++++ linux-2.6.38.3/drivers/infiniband/core/cm.c 2011-04-17 15:57:32.000000000 -0400 +@@ -113,7 +113,7 @@ static char const counter_group_names[CM + + struct cm_counter_group { + struct kobject obj; +- atomic_long_t counter[CM_ATTR_COUNT]; ++ atomic_long_unchecked_t counter[CM_ATTR_COUNT]; + }; + + struct cm_counter_attribute { +@@ -1387,7 +1387,7 @@ static void cm_dup_req_handler(struct cm + struct ib_mad_send_buf *msg = NULL; + int ret; + +- atomic_long_inc(&work->port->counter_group[CM_RECV_DUPLICATES]. ++ atomic_long_inc_unchecked(&work->port->counter_group[CM_RECV_DUPLICATES]. + counter[CM_REQ_COUNTER]); + + /* Quick state check to discard duplicate REQs. */ +@@ -1765,7 +1765,7 @@ static void cm_dup_rep_handler(struct cm + if (!cm_id_priv) + return; + +- atomic_long_inc(&work->port->counter_group[CM_RECV_DUPLICATES]. ++ atomic_long_inc_unchecked(&work->port->counter_group[CM_RECV_DUPLICATES]. + counter[CM_REP_COUNTER]); + ret = cm_alloc_response_msg(work->port, work->mad_recv_wc, &msg); + if (ret) +@@ -1932,7 +1932,7 @@ static int cm_rtu_handler(struct cm_work + if (cm_id_priv->id.state != IB_CM_REP_SENT && + cm_id_priv->id.state != IB_CM_MRA_REP_RCVD) { + spin_unlock_irq(&cm_id_priv->lock); +- atomic_long_inc(&work->port->counter_group[CM_RECV_DUPLICATES]. ++ atomic_long_inc_unchecked(&work->port->counter_group[CM_RECV_DUPLICATES]. + counter[CM_RTU_COUNTER]); + goto out; + } +@@ -2111,7 +2111,7 @@ static int cm_dreq_handler(struct cm_wor + cm_id_priv = cm_acquire_id(dreq_msg->remote_comm_id, + dreq_msg->local_comm_id); + if (!cm_id_priv) { +- atomic_long_inc(&work->port->counter_group[CM_RECV_DUPLICATES]. ++ atomic_long_inc_unchecked(&work->port->counter_group[CM_RECV_DUPLICATES]. + counter[CM_DREQ_COUNTER]); + cm_issue_drep(work->port, work->mad_recv_wc); + return -EINVAL; +@@ -2132,7 +2132,7 @@ static int cm_dreq_handler(struct cm_wor + case IB_CM_MRA_REP_RCVD: + break; + case IB_CM_TIMEWAIT: +- atomic_long_inc(&work->port->counter_group[CM_RECV_DUPLICATES]. ++ atomic_long_inc_unchecked(&work->port->counter_group[CM_RECV_DUPLICATES]. + counter[CM_DREQ_COUNTER]); + if (cm_alloc_response_msg(work->port, work->mad_recv_wc, &msg)) + goto unlock; +@@ -2146,7 +2146,7 @@ static int cm_dreq_handler(struct cm_wor + cm_free_msg(msg); + goto deref; + case IB_CM_DREQ_RCVD: +- atomic_long_inc(&work->port->counter_group[CM_RECV_DUPLICATES]. ++ atomic_long_inc_unchecked(&work->port->counter_group[CM_RECV_DUPLICATES]. + counter[CM_DREQ_COUNTER]); + goto unlock; + default: +@@ -2504,7 +2504,7 @@ static int cm_mra_handler(struct cm_work + ib_modify_mad(cm_id_priv->av.port->mad_agent, + cm_id_priv->msg, timeout)) { + if (cm_id_priv->id.lap_state == IB_CM_MRA_LAP_RCVD) +- atomic_long_inc(&work->port-> ++ atomic_long_inc_unchecked(&work->port-> + counter_group[CM_RECV_DUPLICATES]. + counter[CM_MRA_COUNTER]); + goto out; +@@ -2513,7 +2513,7 @@ static int cm_mra_handler(struct cm_work + break; + case IB_CM_MRA_REQ_RCVD: + case IB_CM_MRA_REP_RCVD: +- atomic_long_inc(&work->port->counter_group[CM_RECV_DUPLICATES]. ++ atomic_long_inc_unchecked(&work->port->counter_group[CM_RECV_DUPLICATES]. + counter[CM_MRA_COUNTER]); + /* fall through */ + default: +@@ -2675,7 +2675,7 @@ static int cm_lap_handler(struct cm_work + case IB_CM_LAP_IDLE: + break; + case IB_CM_MRA_LAP_SENT: +- atomic_long_inc(&work->port->counter_group[CM_RECV_DUPLICATES]. ++ atomic_long_inc_unchecked(&work->port->counter_group[CM_RECV_DUPLICATES]. + counter[CM_LAP_COUNTER]); + if (cm_alloc_response_msg(work->port, work->mad_recv_wc, &msg)) + goto unlock; +@@ -2691,7 +2691,7 @@ static int cm_lap_handler(struct cm_work + cm_free_msg(msg); + goto deref; + case IB_CM_LAP_RCVD: +- atomic_long_inc(&work->port->counter_group[CM_RECV_DUPLICATES]. ++ atomic_long_inc_unchecked(&work->port->counter_group[CM_RECV_DUPLICATES]. + counter[CM_LAP_COUNTER]); + goto unlock; + default: +@@ -2975,7 +2975,7 @@ static int cm_sidr_req_handler(struct cm + cur_cm_id_priv = cm_insert_remote_sidr(cm_id_priv); + if (cur_cm_id_priv) { + spin_unlock_irq(&cm.lock); +- atomic_long_inc(&work->port->counter_group[CM_RECV_DUPLICATES]. ++ atomic_long_inc_unchecked(&work->port->counter_group[CM_RECV_DUPLICATES]. + counter[CM_SIDR_REQ_COUNTER]); + goto out; /* Duplicate message. */ + } +@@ -3187,10 +3187,10 @@ static void cm_send_handler(struct ib_ma + if (!msg->context[0] && (attr_index != CM_REJ_COUNTER)) + msg->retries = 1; + +- atomic_long_add(1 + msg->retries, ++ atomic_long_add_unchecked(1 + msg->retries, + &port->counter_group[CM_XMIT].counter[attr_index]); + if (msg->retries) +- atomic_long_add(msg->retries, ++ atomic_long_add_unchecked(msg->retries, + &port->counter_group[CM_XMIT_RETRIES]. + counter[attr_index]); + +@@ -3400,7 +3400,7 @@ static void cm_recv_handler(struct ib_ma + } + + attr_id = be16_to_cpu(mad_recv_wc->recv_buf.mad->mad_hdr.attr_id); +- atomic_long_inc(&port->counter_group[CM_RECV]. ++ atomic_long_inc_unchecked(&port->counter_group[CM_RECV]. + counter[attr_id - CM_ATTR_ID_OFFSET]); + + work = kmalloc(sizeof *work + sizeof(struct ib_sa_path_rec) * paths, +@@ -3598,7 +3598,7 @@ static ssize_t cm_show_counter(struct ko + cm_attr = container_of(attr, struct cm_counter_attribute, attr); + + return sprintf(buf, "%ld\n", +- atomic_long_read(&group->counter[cm_attr->index])); ++ atomic_long_read_unchecked(&group->counter[cm_attr->index])); + } + + static const struct sysfs_ops cm_counter_ops = { +diff -urNp linux-2.6.38.3/drivers/infiniband/hw/qib/qib.h linux-2.6.38.3/drivers/infiniband/hw/qib/qib.h +--- linux-2.6.38.3/drivers/infiniband/hw/qib/qib.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/infiniband/hw/qib/qib.h 2011-04-17 15:57:32.000000000 -0400 +@@ -51,6 +51,7 @@ + #include <linux/completion.h> + #include <linux/kref.h> + #include <linux/sched.h> ++#include <linux/slab.h> + + #include "qib_common.h" + #include "qib_verbs.h" +diff -urNp linux-2.6.38.3/drivers/input/keyboard/atkbd.c linux-2.6.38.3/drivers/input/keyboard/atkbd.c +--- linux-2.6.38.3/drivers/input/keyboard/atkbd.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/input/keyboard/atkbd.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1250,7 +1250,7 @@ static struct serio_device_id atkbd_seri + .id = SERIO_ANY, + .extra = SERIO_ANY, + }, +- { 0 } ++ { 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(serio, atkbd_serio_ids); +diff -urNp linux-2.6.38.3/drivers/input/mouse/lifebook.c linux-2.6.38.3/drivers/input/mouse/lifebook.c +--- linux-2.6.38.3/drivers/input/mouse/lifebook.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/input/mouse/lifebook.c 2011-04-17 15:57:32.000000000 -0400 +@@ -123,7 +123,7 @@ static const struct dmi_system_id __init + DMI_MATCH(DMI_PRODUCT_NAME, "LifeBook B142"), + }, + }, +- { } ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL} + }; + + void __init lifebook_module_init(void) +diff -urNp linux-2.6.38.3/drivers/input/mouse/psmouse-base.c linux-2.6.38.3/drivers/input/mouse/psmouse-base.c +--- linux-2.6.38.3/drivers/input/mouse/psmouse-base.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/input/mouse/psmouse-base.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1462,7 +1462,7 @@ static struct serio_device_id psmouse_se + .id = SERIO_ANY, + .extra = SERIO_ANY, + }, +- { 0 } ++ { 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(serio, psmouse_serio_ids); +diff -urNp linux-2.6.38.3/drivers/input/mouse/synaptics.c linux-2.6.38.3/drivers/input/mouse/synaptics.c +--- linux-2.6.38.3/drivers/input/mouse/synaptics.c 2011-04-17 16:53:01.000000000 -0400 ++++ linux-2.6.38.3/drivers/input/mouse/synaptics.c 2011-04-17 16:53:16.000000000 -0400 +@@ -559,7 +559,7 @@ static void synaptics_process_packet(str + break; + case 2: + if (SYN_MODEL_PEN(priv->model_id)) +- ; /* Nothing, treat a pen as a single finger */ ++ break; /* Nothing, treat a pen as a single finger */ + break; + case 4 ... 15: + if (SYN_CAP_PALMDETECT(priv->capabilities)) +@@ -825,7 +825,6 @@ static const struct dmi_system_id __init + DMI_MATCH(DMI_SYS_VENDOR, "TOSHIBA"), + DMI_MATCH(DMI_PRODUCT_NAME, "PORTEGE M300"), + }, +- + }, + { + /* Toshiba Portege M300 */ +@@ -834,10 +833,9 @@ static const struct dmi_system_id __init + DMI_MATCH(DMI_PRODUCT_NAME, "Portable PC"), + DMI_MATCH(DMI_PRODUCT_VERSION, "Version 1.0"), + }, +- + }, ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } + #endif +- { } + }; + + static bool broken_olpc_ec; +diff -urNp linux-2.6.38.3/drivers/input/mousedev.c linux-2.6.38.3/drivers/input/mousedev.c +--- linux-2.6.38.3/drivers/input/mousedev.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/input/mousedev.c 2011-04-17 15:57:32.000000000 -0400 +@@ -764,7 +764,7 @@ static ssize_t mousedev_read(struct file + + spin_unlock_irq(&client->packet_lock); + +- if (copy_to_user(buffer, data, count)) ++ if (count > sizeof(data) || copy_to_user(buffer, data, count)) + return -EFAULT; + + return count; +@@ -1067,7 +1067,7 @@ static struct input_handler mousedev_han + + #ifdef CONFIG_INPUT_MOUSEDEV_PSAUX + static struct miscdevice psaux_mouse = { +- PSMOUSE_MINOR, "psaux", &mousedev_fops ++ PSMOUSE_MINOR, "psaux", &mousedev_fops, {NULL, NULL}, NULL, NULL + }; + static int psaux_registered; + #endif +diff -urNp linux-2.6.38.3/drivers/input/serio/i8042-x86ia64io.h linux-2.6.38.3/drivers/input/serio/i8042-x86ia64io.h +--- linux-2.6.38.3/drivers/input/serio/i8042-x86ia64io.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/input/serio/i8042-x86ia64io.h 2011-04-17 15:57:32.000000000 -0400 +@@ -183,7 +183,7 @@ static const struct dmi_system_id __init + DMI_MATCH(DMI_PRODUCT_VERSION, "Rev 1"), + }, + }, +- { } ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } + }; + + /* +@@ -431,7 +431,7 @@ static const struct dmi_system_id __init + DMI_MATCH(DMI_PRODUCT_NAME, "Vostro V13"), + }, + }, +- { } ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } + }; + + static const struct dmi_system_id __initconst i8042_dmi_reset_table[] = { +@@ -505,7 +505,7 @@ static const struct dmi_system_id __init + DMI_MATCH(DMI_PRODUCT_NAME, "Vostro 1720"), + }, + }, +- { } ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } + }; + + #ifdef CONFIG_PNP +@@ -524,7 +524,7 @@ static const struct dmi_system_id __init + DMI_MATCH(DMI_BOARD_VENDOR, "MICRO-STAR INTERNATIONAL CO., LTD"), + }, + }, +- { } ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } + }; + + static const struct dmi_system_id __initconst i8042_dmi_laptop_table[] = { +@@ -548,7 +548,7 @@ static const struct dmi_system_id __init + DMI_MATCH(DMI_CHASSIS_TYPE, "14"), /* Sub-Notebook */ + }, + }, +- { } ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } + }; + #endif + +@@ -640,7 +640,7 @@ static const struct dmi_system_id __init + DMI_MATCH(DMI_PRODUCT_NAME, "TravelMate 4280"), + }, + }, +- { } ++ { NULL, NULL, {DMI_MATCH(DMI_NONE, {0})}, NULL } + }; + + #endif /* CONFIG_X86 */ +diff -urNp linux-2.6.38.3/drivers/input/serio/serio_raw.c linux-2.6.38.3/drivers/input/serio/serio_raw.c +--- linux-2.6.38.3/drivers/input/serio/serio_raw.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/input/serio/serio_raw.c 2011-04-17 15:57:32.000000000 -0400 +@@ -376,7 +376,7 @@ static struct serio_device_id serio_raw_ + .id = SERIO_ANY, + .extra = SERIO_ANY, + }, +- { 0 } ++ { 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(serio, serio_raw_serio_ids); +diff -urNp linux-2.6.38.3/drivers/isdn/gigaset/common.c linux-2.6.38.3/drivers/isdn/gigaset/common.c +--- linux-2.6.38.3/drivers/isdn/gigaset/common.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/isdn/gigaset/common.c 2011-04-17 15:57:32.000000000 -0400 +@@ -723,7 +723,7 @@ struct cardstate *gigaset_initcs(struct + cs->commands_pending = 0; + cs->cur_at_seq = 0; + cs->gotfwver = -1; +- cs->open_count = 0; ++ local_set(&cs->open_count, 0); + cs->dev = NULL; + cs->tty = NULL; + cs->tty_dev = NULL; +diff -urNp linux-2.6.38.3/drivers/isdn/gigaset/gigaset.h linux-2.6.38.3/drivers/isdn/gigaset/gigaset.h +--- linux-2.6.38.3/drivers/isdn/gigaset/gigaset.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/isdn/gigaset/gigaset.h 2011-04-17 15:57:32.000000000 -0400 +@@ -35,6 +35,7 @@ + #include <linux/tty_driver.h> + #include <linux/list.h> + #include <asm/atomic.h> ++#include <asm/local.h> + + #define GIG_VERSION {0, 5, 0, 0} + #define GIG_COMPAT {0, 4, 0, 0} +@@ -433,7 +434,7 @@ struct cardstate { + spinlock_t cmdlock; + unsigned curlen, cmdbytes; + +- unsigned open_count; ++ local_t open_count; + struct tty_struct *tty; + struct tasklet_struct if_wake_tasklet; + unsigned control_state; +diff -urNp linux-2.6.38.3/drivers/isdn/gigaset/interface.c linux-2.6.38.3/drivers/isdn/gigaset/interface.c +--- linux-2.6.38.3/drivers/isdn/gigaset/interface.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/isdn/gigaset/interface.c 2011-04-17 15:57:32.000000000 -0400 +@@ -160,9 +160,7 @@ static int if_open(struct tty_struct *tt + return -ERESTARTSYS; + tty->driver_data = cs; + +- ++cs->open_count; +- +- if (cs->open_count == 1) { ++ if (local_inc_return(&cs->open_count) == 1) { + spin_lock_irqsave(&cs->lock, flags); + cs->tty = tty; + spin_unlock_irqrestore(&cs->lock, flags); +@@ -190,10 +188,10 @@ static void if_close(struct tty_struct * + + if (!cs->connected) + gig_dbg(DEBUG_IF, "not connected"); /* nothing to do */ +- else if (!cs->open_count) ++ else if (!local_read(&cs->open_count)) + dev_warn(cs->dev, "%s: device not opened\n", __func__); + else { +- if (!--cs->open_count) { ++ if (!local_dec_return(&cs->open_count)) { + spin_lock_irqsave(&cs->lock, flags); + cs->tty = NULL; + spin_unlock_irqrestore(&cs->lock, flags); +@@ -228,7 +226,7 @@ static int if_ioctl(struct tty_struct *t + if (!cs->connected) { + gig_dbg(DEBUG_IF, "not connected"); + retval = -ENODEV; +- } else if (!cs->open_count) ++ } else if (!local_read(&cs->open_count)) + dev_warn(cs->dev, "%s: device not opened\n", __func__); + else { + retval = 0; +@@ -358,7 +356,7 @@ static int if_write(struct tty_struct *t + retval = -ENODEV; + goto done; + } +- if (!cs->open_count) { ++ if (!local_read(&cs->open_count)) { + dev_warn(cs->dev, "%s: device not opened\n", __func__); + retval = -ENODEV; + goto done; +@@ -411,7 +409,7 @@ static int if_write_room(struct tty_stru + if (!cs->connected) { + gig_dbg(DEBUG_IF, "not connected"); + retval = -ENODEV; +- } else if (!cs->open_count) ++ } else if (!local_read(&cs->open_count)) + dev_warn(cs->dev, "%s: device not opened\n", __func__); + else if (cs->mstate != MS_LOCKED) { + dev_warn(cs->dev, "can't write to unlocked device\n"); +@@ -441,7 +439,7 @@ static int if_chars_in_buffer(struct tty + + if (!cs->connected) + gig_dbg(DEBUG_IF, "not connected"); +- else if (!cs->open_count) ++ else if (!local_read(&cs->open_count)) + dev_warn(cs->dev, "%s: device not opened\n", __func__); + else if (cs->mstate != MS_LOCKED) + dev_warn(cs->dev, "can't write to unlocked device\n"); +@@ -469,7 +467,7 @@ static void if_throttle(struct tty_struc + + if (!cs->connected) + gig_dbg(DEBUG_IF, "not connected"); /* nothing to do */ +- else if (!cs->open_count) ++ else if (!local_read(&cs->open_count)) + dev_warn(cs->dev, "%s: device not opened\n", __func__); + else + gig_dbg(DEBUG_IF, "%s: not implemented\n", __func__); +@@ -493,7 +491,7 @@ static void if_unthrottle(struct tty_str + + if (!cs->connected) + gig_dbg(DEBUG_IF, "not connected"); /* nothing to do */ +- else if (!cs->open_count) ++ else if (!local_read(&cs->open_count)) + dev_warn(cs->dev, "%s: device not opened\n", __func__); + else + gig_dbg(DEBUG_IF, "%s: not implemented\n", __func__); +@@ -524,7 +522,7 @@ static void if_set_termios(struct tty_st + goto out; + } + +- if (!cs->open_count) { ++ if (!local_read(&cs->open_count)) { + dev_warn(cs->dev, "%s: device not opened\n", __func__); + goto out; + } +diff -urNp linux-2.6.38.3/drivers/isdn/hardware/avm/b1.c linux-2.6.38.3/drivers/isdn/hardware/avm/b1.c +--- linux-2.6.38.3/drivers/isdn/hardware/avm/b1.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/isdn/hardware/avm/b1.c 2011-04-17 15:57:32.000000000 -0400 +@@ -176,7 +176,7 @@ int b1_load_t4file(avmcard *card, capilo + } + if (left) { + if (t4file->user) { +- if (copy_from_user(buf, dp, left)) ++ if (left > sizeof buf || copy_from_user(buf, dp, left)) + return -EFAULT; + } else { + memcpy(buf, dp, left); +@@ -224,7 +224,7 @@ int b1_load_config(avmcard *card, capilo + } + if (left) { + if (config->user) { +- if (copy_from_user(buf, dp, left)) ++ if (left > sizeof buf || copy_from_user(buf, dp, left)) + return -EFAULT; + } else { + memcpy(buf, dp, left); +diff -urNp linux-2.6.38.3/drivers/isdn/icn/icn.c linux-2.6.38.3/drivers/isdn/icn/icn.c +--- linux-2.6.38.3/drivers/isdn/icn/icn.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/isdn/icn/icn.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1045,7 +1045,7 @@ icn_writecmd(const u_char * buf, int len + if (count > len) + count = len; + if (user) { +- if (copy_from_user(msg, buf, count)) ++ if (count > sizeof msg || copy_from_user(msg, buf, count)) + return -EFAULT; + } else + memcpy(msg, buf, count); +diff -urNp linux-2.6.38.3/drivers/lguest/core.c linux-2.6.38.3/drivers/lguest/core.c +--- linux-2.6.38.3/drivers/lguest/core.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/lguest/core.c 2011-04-17 15:57:32.000000000 -0400 +@@ -92,9 +92,17 @@ static __init int map_switcher(void) + * it's worked so far. The end address needs +1 because __get_vm_area + * allocates an extra guard page, so we need space for that. + */ ++ ++#if defined(CONFIG_MODULES) && defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) ++ switcher_vma = __get_vm_area(TOTAL_SWITCHER_PAGES * PAGE_SIZE, ++ VM_ALLOC | VM_KERNEXEC, SWITCHER_ADDR, SWITCHER_ADDR ++ + (TOTAL_SWITCHER_PAGES+1) * PAGE_SIZE); ++#else + switcher_vma = __get_vm_area(TOTAL_SWITCHER_PAGES * PAGE_SIZE, + VM_ALLOC, SWITCHER_ADDR, SWITCHER_ADDR + + (TOTAL_SWITCHER_PAGES+1) * PAGE_SIZE); ++#endif ++ + if (!switcher_vma) { + err = -ENOMEM; + printk("lguest: could not map switcher pages high\n"); +@@ -119,7 +127,7 @@ static __init int map_switcher(void) + * Now the Switcher is mapped at the right address, we can't fail! + * Copy in the compiled-in Switcher code (from <arch>_switcher.S). + */ +- memcpy(switcher_vma->addr, start_switcher_text, ++ memcpy(switcher_vma->addr, ktla_ktva(start_switcher_text), + end_switcher_text - start_switcher_text); + + printk(KERN_INFO "lguest: mapped switcher at %p\n", +diff -urNp linux-2.6.38.3/drivers/lguest/x86/core.c linux-2.6.38.3/drivers/lguest/x86/core.c +--- linux-2.6.38.3/drivers/lguest/x86/core.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/lguest/x86/core.c 2011-04-17 15:57:32.000000000 -0400 +@@ -59,7 +59,7 @@ static struct { + /* Offset from where switcher.S was compiled to where we've copied it */ + static unsigned long switcher_offset(void) + { +- return SWITCHER_ADDR - (unsigned long)start_switcher_text; ++ return SWITCHER_ADDR - (unsigned long)ktla_ktva(start_switcher_text); + } + + /* This cpu's struct lguest_pages. */ +@@ -100,7 +100,13 @@ static void copy_in_guest_info(struct lg + * These copies are pretty cheap, so we do them unconditionally: */ + /* Save the current Host top-level page directory. + */ ++ ++#ifdef CONFIG_PAX_PER_CPU_PGD ++ pages->state.host_cr3 = read_cr3(); ++#else + pages->state.host_cr3 = __pa(current->mm->pgd); ++#endif ++ + /* + * Set up the Guest's page tables to see this CPU's pages (and no + * other CPU's pages). +@@ -547,7 +553,7 @@ void __init lguest_arch_host_init(void) + * compiled-in switcher code and the high-mapped copy we just made. + */ + for (i = 0; i < IDT_ENTRIES; i++) +- default_idt_entries[i] += switcher_offset(); ++ default_idt_entries[i] = ktla_ktva(default_idt_entries[i]) + switcher_offset(); + + /* + * Set up the Switcher's per-cpu areas. +@@ -630,7 +636,7 @@ void __init lguest_arch_host_init(void) + * it will be undisturbed when we switch. To change %cs and jump we + * need this structure to feed to Intel's "lcall" instruction. + */ +- lguest_entry.offset = (long)switch_to_guest + switcher_offset(); ++ lguest_entry.offset = (long)ktla_ktva(switch_to_guest) + switcher_offset(); + lguest_entry.segment = LGUEST_CS; + + /* +diff -urNp linux-2.6.38.3/drivers/lguest/x86/switcher_32.S linux-2.6.38.3/drivers/lguest/x86/switcher_32.S +--- linux-2.6.38.3/drivers/lguest/x86/switcher_32.S 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/lguest/x86/switcher_32.S 2011-04-17 15:57:32.000000000 -0400 +@@ -87,6 +87,7 @@ + #include <asm/page.h> + #include <asm/segment.h> + #include <asm/lguest.h> ++#include <asm/processor-flags.h> + + // We mark the start of the code to copy + // It's placed in .text tho it's never run here +@@ -149,6 +150,13 @@ ENTRY(switch_to_guest) + // Changes type when we load it: damn Intel! + // For after we switch over our page tables + // That entry will be read-only: we'd crash. ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ mov %cr0, %edx ++ xor $X86_CR0_WP, %edx ++ mov %edx, %cr0 ++#endif ++ + movl $(GDT_ENTRY_TSS*8), %edx + ltr %dx + +@@ -157,9 +165,15 @@ ENTRY(switch_to_guest) + // Let's clear it again for our return. + // The GDT descriptor of the Host + // Points to the table after two "size" bytes +- movl (LGUEST_PAGES_host_gdt_desc+2)(%eax), %edx ++ movl (LGUEST_PAGES_host_gdt_desc+2)(%eax), %eax + // Clear "used" from type field (byte 5, bit 2) +- andb $0xFD, (GDT_ENTRY_TSS*8 + 5)(%edx) ++ andb $0xFD, (GDT_ENTRY_TSS*8 + 5)(%eax) ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ mov %cr0, %eax ++ xor $X86_CR0_WP, %eax ++ mov %eax, %cr0 ++#endif + + // Once our page table's switched, the Guest is live! + // The Host fades as we run this final step. +@@ -295,13 +309,12 @@ deliver_to_host: + // I consulted gcc, and it gave + // These instructions, which I gladly credit: + leal (%edx,%ebx,8), %eax +- movzwl (%eax),%edx +- movl 4(%eax), %eax +- xorw %ax, %ax +- orl %eax, %edx ++ movl 4(%eax), %edx ++ movw (%eax), %dx + // Now the address of the handler's in %edx + // We call it now: its "iret" drops us home. +- jmp *%edx ++ ljmp $__KERNEL_CS, $1f ++1: jmp *%edx + + // Every interrupt can come to us here + // But we must truly tell each apart. +diff -urNp linux-2.6.38.3/drivers/md/bitmap.c linux-2.6.38.3/drivers/md/bitmap.c +--- linux-2.6.38.3/drivers/md/bitmap.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/md/bitmap.c 2011-04-17 15:57:32.000000000 -0400 +@@ -55,7 +55,7 @@ + # if DEBUG > 0 + # define PRINTK(x...) printk(KERN_DEBUG x) + # else +-# define PRINTK(x...) ++# define PRINTK(x...) do {} while (0) + # endif + #endif + +diff -urNp linux-2.6.38.3/drivers/md/dm-ioctl.c linux-2.6.38.3/drivers/md/dm-ioctl.c +--- linux-2.6.38.3/drivers/md/dm-ioctl.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/md/dm-ioctl.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1541,7 +1541,7 @@ static int validate_params(uint cmd, str + cmd == DM_LIST_VERSIONS_CMD) + return 0; + +- if ((cmd == DM_DEV_CREATE_CMD)) { ++ if (cmd == DM_DEV_CREATE_CMD) { + if (!*param->name) { + DMWARN("name not supplied when creating device"); + return -EINVAL; +diff -urNp linux-2.6.38.3/drivers/md/dm-table.c linux-2.6.38.3/drivers/md/dm-table.c +--- linux-2.6.38.3/drivers/md/dm-table.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/md/dm-table.c 2011-04-17 15:57:32.000000000 -0400 +@@ -372,7 +372,7 @@ static int device_area_is_invalid(struct + if (!dev_size) + return 0; + +- if ((start >= dev_size) || (start + len > dev_size)) { ++ if ((start >= dev_size) || (len > dev_size - start)) { + DMWARN("%s: %s too small for target: " + "start=%llu, len=%llu, dev_size=%llu", + dm_device_name(ti->table->md), bdevname(bdev, b), +diff -urNp linux-2.6.38.3/drivers/md/md.c linux-2.6.38.3/drivers/md/md.c +--- linux-2.6.38.3/drivers/md/md.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/md/md.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1889,7 +1889,7 @@ static int bind_rdev_to_array(mdk_rdev_t + + ko = &part_to_dev(rdev->bdev->bd_part)->kobj; + if (sysfs_create_link(&rdev->kobj, ko, "block")) +- /* failure here is OK */; ++ /* failure here is OK */{} + rdev->sysfs_state = sysfs_get_dirent_safe(rdev->kobj.sd, "state"); + + list_add_rcu(&rdev->same_set, &mddev->disks); +@@ -2499,7 +2499,7 @@ slot_store(mdk_rdev_t *rdev, const char + sysfs_notify_dirent_safe(rdev->sysfs_state); + sprintf(nm, "rd%d", rdev->raid_disk); + if (sysfs_create_link(&rdev->mddev->kobj, &rdev->kobj, nm)) +- /* failure here is OK */; ++ /* failure here is OK */{} + /* don't wakeup anyone, leave that to userspace. */ + } else { + if (slot >= rdev->mddev->raid_disks && +@@ -4594,7 +4594,7 @@ int md_run(mddev_t *mddev) + char nm[20]; + sprintf(nm, "rd%d", rdev->raid_disk); + if (sysfs_create_link(&mddev->kobj, &rdev->kobj, nm)) +- /* failure here is OK */; ++ /* failure here is OK */{} + } + + set_bit(MD_RECOVERY_NEEDED, &mddev->recovery); +@@ -6462,7 +6462,7 @@ static int md_seq_show(struct seq_file * + chunk_kb ? "KB" : "B"); + if (bitmap->file) { + seq_printf(seq, ", file: "); +- seq_path(seq, &bitmap->file->f_path, " \t\n"); ++ seq_path(seq, &bitmap->file->f_path, " \t\n\\"); + } + + seq_printf(seq, "\n"); +@@ -6556,7 +6556,7 @@ static int is_mddev_idle(mddev_t *mddev, + struct gendisk *disk = rdev->bdev->bd_contains->bd_disk; + curr_events = (int)part_stat_read(&disk->part0, sectors[0]) + + (int)part_stat_read(&disk->part0, sectors[1]) - +- atomic_read(&disk->sync_io); ++ atomic_read_unchecked(&disk->sync_io); + /* sync IO will cause sync_io to increase before the disk_stats + * as sync_io is counted when a request starts, and + * disk_stats is counted when it completes. +@@ -7070,7 +7070,7 @@ static int remove_and_add_spares(mddev_t + sprintf(nm, "rd%d", rdev->raid_disk); + if (sysfs_create_link(&mddev->kobj, + &rdev->kobj, nm)) +- /* failure here is OK */; ++ /* failure here is OK */{} + spares++; + md_new_event(mddev); + set_bit(MD_CHANGE_DEVS, &mddev->flags); +diff -urNp linux-2.6.38.3/drivers/md/md.h linux-2.6.38.3/drivers/md/md.h +--- linux-2.6.38.3/drivers/md/md.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/md/md.h 2011-04-17 15:57:32.000000000 -0400 +@@ -360,7 +360,7 @@ static inline void rdev_dec_pending(mdk_ + + static inline void md_sync_acct(struct block_device *bdev, unsigned long nr_sectors) + { +- atomic_add(nr_sectors, &bdev->bd_contains->bd_disk->sync_io); ++ atomic_add_unchecked(nr_sectors, &bdev->bd_contains->bd_disk->sync_io); + } + + struct mdk_personality +diff -urNp linux-2.6.38.3/drivers/media/dvb/dvb-core/dvbdev.c linux-2.6.38.3/drivers/media/dvb/dvb-core/dvbdev.c +--- linux-2.6.38.3/drivers/media/dvb/dvb-core/dvbdev.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/media/dvb/dvb-core/dvbdev.c 2011-04-17 15:57:32.000000000 -0400 +@@ -192,7 +192,7 @@ int dvb_register_device(struct dvb_adapt + const struct dvb_device *template, void *priv, int type) + { + struct dvb_device *dvbdev; +- struct file_operations *dvbdevfops; ++ struct file_operations *dvbdevfops; /* cannot be const, see this function */ + struct device *clsdev; + int minor; + int id; +diff -urNp linux-2.6.38.3/drivers/media/radio/radio-cadet.c linux-2.6.38.3/drivers/media/radio/radio-cadet.c +--- linux-2.6.38.3/drivers/media/radio/radio-cadet.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/media/radio/radio-cadet.c 2011-04-17 15:57:32.000000000 -0400 +@@ -349,7 +349,7 @@ static ssize_t cadet_read(struct file *f + readbuf[i++] = dev->rdsbuf[dev->rdsout++]; + mutex_unlock(&dev->lock); + +- if (copy_to_user(data, readbuf, i)) ++ if (i > sizeof readbuf || copy_to_user(data, readbuf, i)) + return -EFAULT; + return i; + } +diff -urNp linux-2.6.38.3/drivers/media/rc/ir-lirc-codec.c linux-2.6.38.3/drivers/media/rc/ir-lirc-codec.c +--- linux-2.6.38.3/drivers/media/rc/ir-lirc-codec.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/media/rc/ir-lirc-codec.c 2011-04-17 15:57:32.000000000 -0400 +@@ -277,7 +277,7 @@ static void ir_lirc_close(void *data) + return; + } + +-static struct file_operations lirc_fops = { ++static const struct file_operations lirc_fops = { + .owner = THIS_MODULE, + .write = ir_lirc_transmit_ir, + .unlocked_ioctl = ir_lirc_ioctl, +diff -urNp linux-2.6.38.3/drivers/media/rc/lirc_dev.c linux-2.6.38.3/drivers/media/rc/lirc_dev.c +--- linux-2.6.38.3/drivers/media/rc/lirc_dev.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/media/rc/lirc_dev.c 2011-04-17 15:57:32.000000000 -0400 +@@ -151,7 +151,7 @@ static int lirc_thread(void *irctl) + } + + +-static struct file_operations lirc_dev_fops = { ++static const struct file_operations lirc_dev_fops = { + .owner = THIS_MODULE, + .read = lirc_dev_fop_read, + .write = lirc_dev_fop_write, +diff -urNp linux-2.6.38.3/drivers/media/video/sn9c102/sn9c102_core.c linux-2.6.38.3/drivers/media/video/sn9c102/sn9c102_core.c +--- linux-2.6.38.3/drivers/media/video/sn9c102/sn9c102_core.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/media/video/sn9c102/sn9c102_core.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1430,9 +1430,9 @@ static DEVICE_ATTR(i2c_reg, S_IRUGO | S_ + sn9c102_show_i2c_reg, sn9c102_store_i2c_reg); + static DEVICE_ATTR(i2c_val, S_IRUGO | S_IWUSR, + sn9c102_show_i2c_val, sn9c102_store_i2c_val); +-static DEVICE_ATTR(green, S_IWUGO, NULL, sn9c102_store_green); +-static DEVICE_ATTR(blue, S_IWUGO, NULL, sn9c102_store_blue); +-static DEVICE_ATTR(red, S_IWUGO, NULL, sn9c102_store_red); ++static DEVICE_ATTR(green, S_IWUSR, NULL, sn9c102_store_green); ++static DEVICE_ATTR(blue, S_IWUSR, NULL, sn9c102_store_blue); ++static DEVICE_ATTR(red, S_IWUSR, NULL, sn9c102_store_red); + static DEVICE_ATTR(frame_header, S_IRUGO, sn9c102_show_frame_header, NULL); + + +diff -urNp linux-2.6.38.3/drivers/message/fusion/mptbase.c linux-2.6.38.3/drivers/message/fusion/mptbase.c +--- linux-2.6.38.3/drivers/message/fusion/mptbase.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/message/fusion/mptbase.c 2011-04-17 15:57:32.000000000 -0400 +@@ -6683,8 +6683,13 @@ static int mpt_iocinfo_proc_show(struct + seq_printf(m, " MaxChainDepth = 0x%02x frames\n", ioc->facts.MaxChainDepth); + seq_printf(m, " MinBlockSize = 0x%02x bytes\n", 4*ioc->facts.BlockSize); + ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ seq_printf(m, " RequestFrames @ 0x%p (Dma @ 0x%p)\n", NULL, NULL); ++#else + seq_printf(m, " RequestFrames @ 0x%p (Dma @ 0x%p)\n", + (void *)ioc->req_frames, (void *)(ulong)ioc->req_frames_dma); ++#endif ++ + /* + * Rounding UP to nearest 4-kB boundary here... + */ +diff -urNp linux-2.6.38.3/drivers/message/fusion/mptdebug.h linux-2.6.38.3/drivers/message/fusion/mptdebug.h +--- linux-2.6.38.3/drivers/message/fusion/mptdebug.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/message/fusion/mptdebug.h 2011-04-17 15:57:32.000000000 -0400 +@@ -71,7 +71,7 @@ + CMD; \ + } + #else +-#define MPT_CHECK_LOGGING(IOC, CMD, BITS) ++#define MPT_CHECK_LOGGING(IOC, CMD, BITS) do {} while (0) + #endif + + +diff -urNp linux-2.6.38.3/drivers/message/fusion/mptsas.c linux-2.6.38.3/drivers/message/fusion/mptsas.c +--- linux-2.6.38.3/drivers/message/fusion/mptsas.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/message/fusion/mptsas.c 2011-04-17 15:57:32.000000000 -0400 +@@ -439,6 +439,23 @@ mptsas_is_end_device(struct mptsas_devin + return 0; + } + ++static inline void ++mptsas_set_rphy(MPT_ADAPTER *ioc, struct mptsas_phyinfo *phy_info, struct sas_rphy *rphy) ++{ ++ if (phy_info->port_details) { ++ phy_info->port_details->rphy = rphy; ++ dsaswideprintk(ioc, printk(MYIOC_s_DEBUG_FMT "sas_rphy_add: rphy=%p\n", ++ ioc->name, rphy)); ++ } ++ ++ if (rphy) { ++ dsaswideprintk(ioc, dev_printk(KERN_DEBUG, ++ &rphy->dev, MYIOC_s_FMT "add:", ioc->name)); ++ dsaswideprintk(ioc, printk(MYIOC_s_DEBUG_FMT "rphy=%p release=%p\n", ++ ioc->name, rphy, rphy->dev.release)); ++ } ++} ++ + /* no mutex */ + static void + mptsas_port_delete(MPT_ADAPTER *ioc, struct mptsas_portinfo_details * port_details) +@@ -477,23 +494,6 @@ mptsas_get_rphy(struct mptsas_phyinfo *p + return NULL; + } + +-static inline void +-mptsas_set_rphy(MPT_ADAPTER *ioc, struct mptsas_phyinfo *phy_info, struct sas_rphy *rphy) +-{ +- if (phy_info->port_details) { +- phy_info->port_details->rphy = rphy; +- dsaswideprintk(ioc, printk(MYIOC_s_DEBUG_FMT "sas_rphy_add: rphy=%p\n", +- ioc->name, rphy)); +- } +- +- if (rphy) { +- dsaswideprintk(ioc, dev_printk(KERN_DEBUG, +- &rphy->dev, MYIOC_s_FMT "add:", ioc->name)); +- dsaswideprintk(ioc, printk(MYIOC_s_DEBUG_FMT "rphy=%p release=%p\n", +- ioc->name, rphy, rphy->dev.release)); +- } +-} +- + static inline struct sas_port * + mptsas_get_port(struct mptsas_phyinfo *phy_info) + { +diff -urNp linux-2.6.38.3/drivers/message/fusion/mptscsih.c linux-2.6.38.3/drivers/message/fusion/mptscsih.c +--- linux-2.6.38.3/drivers/message/fusion/mptscsih.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/message/fusion/mptscsih.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1268,15 +1268,16 @@ mptscsih_info(struct Scsi_Host *SChost) + + h = shost_priv(SChost); + +- if (h) { +- if (h->info_kbuf == NULL) +- if ((h->info_kbuf = kmalloc(0x1000 /* 4Kb */, GFP_KERNEL)) == NULL) +- return h->info_kbuf; +- h->info_kbuf[0] = '\0'; ++ if (!h) ++ return NULL; + +- mpt_print_ioc_summary(h->ioc, h->info_kbuf, &size, 0, 0); +- h->info_kbuf[size-1] = '\0'; +- } ++ if (h->info_kbuf == NULL) ++ if ((h->info_kbuf = kmalloc(0x1000 /* 4Kb */, GFP_KERNEL)) == NULL) ++ return h->info_kbuf; ++ h->info_kbuf[0] = '\0'; ++ ++ mpt_print_ioc_summary(h->ioc, h->info_kbuf, &size, 0, 0); ++ h->info_kbuf[size-1] = '\0'; + + return h->info_kbuf; + } +diff -urNp linux-2.6.38.3/drivers/message/i2o/i2o_proc.c linux-2.6.38.3/drivers/message/i2o/i2o_proc.c +--- linux-2.6.38.3/drivers/message/i2o/i2o_proc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/message/i2o/i2o_proc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -255,13 +255,6 @@ static char *scsi_devices[] = { + "Array Controller Device" + }; + +-static char *chtostr(u8 * chars, int n) +-{ +- char tmp[256]; +- tmp[0] = 0; +- return strncat(tmp, (char *)chars, n); +-} +- + static int i2o_report_query_status(struct seq_file *seq, int block_status, + char *group) + { +@@ -838,8 +831,7 @@ static int i2o_seq_show_ddm_table(struct + + seq_printf(seq, "%-#7x", ddm_table.i2o_vendor_id); + seq_printf(seq, "%-#8x", ddm_table.module_id); +- seq_printf(seq, "%-29s", +- chtostr(ddm_table.module_name_version, 28)); ++ seq_printf(seq, "%-.28s", ddm_table.module_name_version); + seq_printf(seq, "%9d ", ddm_table.data_size); + seq_printf(seq, "%8d", ddm_table.code_size); + +@@ -940,8 +932,8 @@ static int i2o_seq_show_drivers_stored(s + + seq_printf(seq, "%-#7x", dst->i2o_vendor_id); + seq_printf(seq, "%-#8x", dst->module_id); +- seq_printf(seq, "%-29s", chtostr(dst->module_name_version, 28)); +- seq_printf(seq, "%-9s", chtostr(dst->date, 8)); ++ seq_printf(seq, "%-.28s", dst->module_name_version); ++ seq_printf(seq, "%-.8s", dst->date); + seq_printf(seq, "%8d ", dst->module_size); + seq_printf(seq, "%8d ", dst->mpb_size); + seq_printf(seq, "0x%04x", dst->module_flags); +@@ -1272,14 +1264,10 @@ static int i2o_seq_show_dev_identity(str + seq_printf(seq, "Device Class : %s\n", i2o_get_class_name(work16[0])); + seq_printf(seq, "Owner TID : %0#5x\n", work16[2]); + seq_printf(seq, "Parent TID : %0#5x\n", work16[3]); +- seq_printf(seq, "Vendor info : %s\n", +- chtostr((u8 *) (work32 + 2), 16)); +- seq_printf(seq, "Product info : %s\n", +- chtostr((u8 *) (work32 + 6), 16)); +- seq_printf(seq, "Description : %s\n", +- chtostr((u8 *) (work32 + 10), 16)); +- seq_printf(seq, "Product rev. : %s\n", +- chtostr((u8 *) (work32 + 14), 8)); ++ seq_printf(seq, "Vendor info : %.16s\n", (u8 *) (work32 + 2)); ++ seq_printf(seq, "Product info : %.16s\n", (u8 *) (work32 + 6)); ++ seq_printf(seq, "Description : %.16s\n", (u8 *) (work32 + 10)); ++ seq_printf(seq, "Product rev. : %.8s\n", (u8 *) (work32 + 14)); + + seq_printf(seq, "Serial number : "); + print_serial_number(seq, (u8 *) (work32 + 16), +@@ -1324,10 +1312,8 @@ static int i2o_seq_show_ddm_identity(str + } + + seq_printf(seq, "Registering DDM TID : 0x%03x\n", result.ddm_tid); +- seq_printf(seq, "Module name : %s\n", +- chtostr(result.module_name, 24)); +- seq_printf(seq, "Module revision : %s\n", +- chtostr(result.module_rev, 8)); ++ seq_printf(seq, "Module name : %.24s\n", result.module_name); ++ seq_printf(seq, "Module revision : %.8s\n", result.module_rev); + + seq_printf(seq, "Serial number : "); + print_serial_number(seq, result.serial_number, sizeof(result) - 36); +@@ -1358,14 +1344,10 @@ static int i2o_seq_show_uinfo(struct seq + return 0; + } + +- seq_printf(seq, "Device name : %s\n", +- chtostr(result.device_name, 64)); +- seq_printf(seq, "Service name : %s\n", +- chtostr(result.service_name, 64)); +- seq_printf(seq, "Physical name : %s\n", +- chtostr(result.physical_location, 64)); +- seq_printf(seq, "Instance number : %s\n", +- chtostr(result.instance_number, 4)); ++ seq_printf(seq, "Device name : %.64s\n", result.device_name); ++ seq_printf(seq, "Service name : %.64s\n", result.service_name); ++ seq_printf(seq, "Physical name : %.64s\n", result.physical_location); ++ seq_printf(seq, "Instance number : %.4s\n", result.instance_number); + + return 0; + } +diff -urNp linux-2.6.38.3/drivers/mfd/janz-cmodio.c linux-2.6.38.3/drivers/mfd/janz-cmodio.c +--- linux-2.6.38.3/drivers/mfd/janz-cmodio.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/mfd/janz-cmodio.c 2011-04-17 15:57:32.000000000 -0400 +@@ -13,6 +13,7 @@ + + #include <linux/kernel.h> + #include <linux/module.h> ++#include <linux/slab.h> + #include <linux/init.h> + #include <linux/pci.h> + #include <linux/interrupt.h> +diff -urNp linux-2.6.38.3/drivers/misc/kgdbts.c linux-2.6.38.3/drivers/misc/kgdbts.c +--- linux-2.6.38.3/drivers/misc/kgdbts.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/misc/kgdbts.c 2011-04-17 15:57:32.000000000 -0400 +@@ -118,7 +118,7 @@ + } while (0) + #define MAX_CONFIG_LEN 40 + +-static struct kgdb_io kgdbts_io_ops; ++static const struct kgdb_io kgdbts_io_ops; + static char get_buf[BUFMAX]; + static int get_buf_cnt; + static char put_buf[BUFMAX]; +@@ -1103,7 +1103,7 @@ static void kgdbts_post_exp_handler(void + module_put(THIS_MODULE); + } + +-static struct kgdb_io kgdbts_io_ops = { ++static const struct kgdb_io kgdbts_io_ops = { + .name = "kgdbts", + .read_char = kgdbts_get_char, + .write_char = kgdbts_put_char, +diff -urNp linux-2.6.38.3/drivers/misc/sgi-gru/gruhandles.c linux-2.6.38.3/drivers/misc/sgi-gru/gruhandles.c +--- linux-2.6.38.3/drivers/misc/sgi-gru/gruhandles.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/misc/sgi-gru/gruhandles.c 2011-04-17 15:57:32.000000000 -0400 +@@ -44,8 +44,8 @@ static void update_mcs_stats(enum mcs_op + unsigned long nsec; + + nsec = CLKS2NSEC(clks); +- atomic_long_inc(&mcs_op_statistics[op].count); +- atomic_long_add(nsec, &mcs_op_statistics[op].total); ++ atomic_long_inc_unchecked(&mcs_op_statistics[op].count); ++ atomic_long_add_unchecked(nsec, &mcs_op_statistics[op].total); + if (mcs_op_statistics[op].max < nsec) + mcs_op_statistics[op].max = nsec; + } +diff -urNp linux-2.6.38.3/drivers/misc/sgi-gru/gruprocfs.c linux-2.6.38.3/drivers/misc/sgi-gru/gruprocfs.c +--- linux-2.6.38.3/drivers/misc/sgi-gru/gruprocfs.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/misc/sgi-gru/gruprocfs.c 2011-04-17 15:57:32.000000000 -0400 +@@ -32,9 +32,9 @@ + + #define printstat(s, f) printstat_val(s, &gru_stats.f, #f) + +-static void printstat_val(struct seq_file *s, atomic_long_t *v, char *id) ++static void printstat_val(struct seq_file *s, atomic_long_unchecked_t *v, char *id) + { +- unsigned long val = atomic_long_read(v); ++ unsigned long val = atomic_long_read_unchecked(v); + + seq_printf(s, "%16lu %s\n", val, id); + } +@@ -134,8 +134,8 @@ static int mcs_statistics_show(struct se + + seq_printf(s, "%-20s%12s%12s%12s\n", "#id", "count", "aver-clks", "max-clks"); + for (op = 0; op < mcsop_last; op++) { +- count = atomic_long_read(&mcs_op_statistics[op].count); +- total = atomic_long_read(&mcs_op_statistics[op].total); ++ count = atomic_long_read_unchecked(&mcs_op_statistics[op].count); ++ total = atomic_long_read_unchecked(&mcs_op_statistics[op].total); + max = mcs_op_statistics[op].max; + seq_printf(s, "%-20s%12ld%12ld%12ld\n", id[op], count, + count ? total / count : 0, max); +diff -urNp linux-2.6.38.3/drivers/misc/sgi-gru/grutables.h linux-2.6.38.3/drivers/misc/sgi-gru/grutables.h +--- linux-2.6.38.3/drivers/misc/sgi-gru/grutables.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/misc/sgi-gru/grutables.h 2011-04-17 15:57:32.000000000 -0400 +@@ -167,82 +167,82 @@ extern unsigned int gru_max_gids; + * GRU statistics. + */ + struct gru_stats_s { +- atomic_long_t vdata_alloc; +- atomic_long_t vdata_free; +- atomic_long_t gts_alloc; +- atomic_long_t gts_free; +- atomic_long_t gms_alloc; +- atomic_long_t gms_free; +- atomic_long_t gts_double_allocate; +- atomic_long_t assign_context; +- atomic_long_t assign_context_failed; +- atomic_long_t free_context; +- atomic_long_t load_user_context; +- atomic_long_t load_kernel_context; +- atomic_long_t lock_kernel_context; +- atomic_long_t unlock_kernel_context; +- atomic_long_t steal_user_context; +- atomic_long_t steal_kernel_context; +- atomic_long_t steal_context_failed; +- atomic_long_t nopfn; +- atomic_long_t asid_new; +- atomic_long_t asid_next; +- atomic_long_t asid_wrap; +- atomic_long_t asid_reuse; +- atomic_long_t intr; +- atomic_long_t intr_cbr; +- atomic_long_t intr_tfh; +- atomic_long_t intr_spurious; +- atomic_long_t intr_mm_lock_failed; +- atomic_long_t call_os; +- atomic_long_t call_os_wait_queue; +- atomic_long_t user_flush_tlb; +- atomic_long_t user_unload_context; +- atomic_long_t user_exception; +- atomic_long_t set_context_option; +- atomic_long_t check_context_retarget_intr; +- atomic_long_t check_context_unload; +- atomic_long_t tlb_dropin; +- atomic_long_t tlb_preload_page; +- atomic_long_t tlb_dropin_fail_no_asid; +- atomic_long_t tlb_dropin_fail_upm; +- atomic_long_t tlb_dropin_fail_invalid; +- atomic_long_t tlb_dropin_fail_range_active; +- atomic_long_t tlb_dropin_fail_idle; +- atomic_long_t tlb_dropin_fail_fmm; +- atomic_long_t tlb_dropin_fail_no_exception; +- atomic_long_t tfh_stale_on_fault; +- atomic_long_t mmu_invalidate_range; +- atomic_long_t mmu_invalidate_page; +- atomic_long_t flush_tlb; +- atomic_long_t flush_tlb_gru; +- atomic_long_t flush_tlb_gru_tgh; +- atomic_long_t flush_tlb_gru_zero_asid; +- +- atomic_long_t copy_gpa; +- atomic_long_t read_gpa; +- +- atomic_long_t mesq_receive; +- atomic_long_t mesq_receive_none; +- atomic_long_t mesq_send; +- atomic_long_t mesq_send_failed; +- atomic_long_t mesq_noop; +- atomic_long_t mesq_send_unexpected_error; +- atomic_long_t mesq_send_lb_overflow; +- atomic_long_t mesq_send_qlimit_reached; +- atomic_long_t mesq_send_amo_nacked; +- atomic_long_t mesq_send_put_nacked; +- atomic_long_t mesq_page_overflow; +- atomic_long_t mesq_qf_locked; +- atomic_long_t mesq_qf_noop_not_full; +- atomic_long_t mesq_qf_switch_head_failed; +- atomic_long_t mesq_qf_unexpected_error; +- atomic_long_t mesq_noop_unexpected_error; +- atomic_long_t mesq_noop_lb_overflow; +- atomic_long_t mesq_noop_qlimit_reached; +- atomic_long_t mesq_noop_amo_nacked; +- atomic_long_t mesq_noop_put_nacked; +- atomic_long_t mesq_noop_page_overflow; ++ atomic_long_unchecked_t vdata_alloc; ++ atomic_long_unchecked_t vdata_free; ++ atomic_long_unchecked_t gts_alloc; ++ atomic_long_unchecked_t gts_free; ++ atomic_long_unchecked_t gms_alloc; ++ atomic_long_unchecked_t gms_free; ++ atomic_long_unchecked_t gts_double_allocate; ++ atomic_long_unchecked_t assign_context; ++ atomic_long_unchecked_t assign_context_failed; ++ atomic_long_unchecked_t free_context; ++ atomic_long_unchecked_t load_user_context; ++ atomic_long_unchecked_t load_kernel_context; ++ atomic_long_unchecked_t lock_kernel_context; ++ atomic_long_unchecked_t unlock_kernel_context; ++ atomic_long_unchecked_t steal_user_context; ++ atomic_long_unchecked_t steal_kernel_context; ++ atomic_long_unchecked_t steal_context_failed; ++ atomic_long_unchecked_t nopfn; ++ atomic_long_unchecked_t asid_new; ++ atomic_long_unchecked_t asid_next; ++ atomic_long_unchecked_t asid_wrap; ++ atomic_long_unchecked_t asid_reuse; ++ atomic_long_unchecked_t intr; ++ atomic_long_unchecked_t intr_cbr; ++ atomic_long_unchecked_t intr_tfh; ++ atomic_long_unchecked_t intr_spurious; ++ atomic_long_unchecked_t intr_mm_lock_failed; ++ atomic_long_unchecked_t call_os; ++ atomic_long_unchecked_t call_os_wait_queue; ++ atomic_long_unchecked_t user_flush_tlb; ++ atomic_long_unchecked_t user_unload_context; ++ atomic_long_unchecked_t user_exception; ++ atomic_long_unchecked_t set_context_option; ++ atomic_long_unchecked_t check_context_retarget_intr; ++ atomic_long_unchecked_t check_context_unload; ++ atomic_long_unchecked_t tlb_dropin; ++ atomic_long_unchecked_t tlb_preload_page; ++ atomic_long_unchecked_t tlb_dropin_fail_no_asid; ++ atomic_long_unchecked_t tlb_dropin_fail_upm; ++ atomic_long_unchecked_t tlb_dropin_fail_invalid; ++ atomic_long_unchecked_t tlb_dropin_fail_range_active; ++ atomic_long_unchecked_t tlb_dropin_fail_idle; ++ atomic_long_unchecked_t tlb_dropin_fail_fmm; ++ atomic_long_unchecked_t tlb_dropin_fail_no_exception; ++ atomic_long_unchecked_t tfh_stale_on_fault; ++ atomic_long_unchecked_t mmu_invalidate_range; ++ atomic_long_unchecked_t mmu_invalidate_page; ++ atomic_long_unchecked_t flush_tlb; ++ atomic_long_unchecked_t flush_tlb_gru; ++ atomic_long_unchecked_t flush_tlb_gru_tgh; ++ atomic_long_unchecked_t flush_tlb_gru_zero_asid; ++ ++ atomic_long_unchecked_t copy_gpa; ++ atomic_long_unchecked_t read_gpa; ++ ++ atomic_long_unchecked_t mesq_receive; ++ atomic_long_unchecked_t mesq_receive_none; ++ atomic_long_unchecked_t mesq_send; ++ atomic_long_unchecked_t mesq_send_failed; ++ atomic_long_unchecked_t mesq_noop; ++ atomic_long_unchecked_t mesq_send_unexpected_error; ++ atomic_long_unchecked_t mesq_send_lb_overflow; ++ atomic_long_unchecked_t mesq_send_qlimit_reached; ++ atomic_long_unchecked_t mesq_send_amo_nacked; ++ atomic_long_unchecked_t mesq_send_put_nacked; ++ atomic_long_unchecked_t mesq_page_overflow; ++ atomic_long_unchecked_t mesq_qf_locked; ++ atomic_long_unchecked_t mesq_qf_noop_not_full; ++ atomic_long_unchecked_t mesq_qf_switch_head_failed; ++ atomic_long_unchecked_t mesq_qf_unexpected_error; ++ atomic_long_unchecked_t mesq_noop_unexpected_error; ++ atomic_long_unchecked_t mesq_noop_lb_overflow; ++ atomic_long_unchecked_t mesq_noop_qlimit_reached; ++ atomic_long_unchecked_t mesq_noop_amo_nacked; ++ atomic_long_unchecked_t mesq_noop_put_nacked; ++ atomic_long_unchecked_t mesq_noop_page_overflow; + + }; + +@@ -251,8 +251,8 @@ enum mcs_op {cchop_allocate, cchop_start + tghop_invalidate, mcsop_last}; + + struct mcs_op_statistic { +- atomic_long_t count; +- atomic_long_t total; ++ atomic_long_unchecked_t count; ++ atomic_long_unchecked_t total; + unsigned long max; + }; + +@@ -275,7 +275,7 @@ extern struct mcs_op_statistic mcs_op_st + + #define STAT(id) do { \ + if (gru_options & OPT_STATS) \ +- atomic_long_inc(&gru_stats.id); \ ++ atomic_long_inc_unchecked(&gru_stats.id); \ + } while (0) + + #ifdef CONFIG_SGI_GRU_DEBUG +diff -urNp linux-2.6.38.3/drivers/mtd/devices/doc2000.c linux-2.6.38.3/drivers/mtd/devices/doc2000.c +--- linux-2.6.38.3/drivers/mtd/devices/doc2000.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/mtd/devices/doc2000.c 2011-04-17 15:57:32.000000000 -0400 +@@ -776,7 +776,7 @@ static int doc_write(struct mtd_info *mt + + /* The ECC will not be calculated correctly if less than 512 is written */ + /* DBB- +- if (len != 0x200 && eccbuf) ++ if (len != 0x200) + printk(KERN_WARNING + "ECC needs a full sector write (adr: %lx size %lx)\n", + (long) to, (long) len); +diff -urNp linux-2.6.38.3/drivers/mtd/devices/doc2001.c linux-2.6.38.3/drivers/mtd/devices/doc2001.c +--- linux-2.6.38.3/drivers/mtd/devices/doc2001.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/mtd/devices/doc2001.c 2011-04-17 15:57:32.000000000 -0400 +@@ -393,7 +393,7 @@ static int doc_read (struct mtd_info *mt + struct Nand *mychip = &this->chips[from >> (this->chipshift)]; + + /* Don't allow read past end of device */ +- if (from >= this->totlen) ++ if (from >= this->totlen || !len) + return -EINVAL; + + /* Don't allow a single read to cross a 512-byte block boundary */ +diff -urNp linux-2.6.38.3/drivers/mtd/nand/denali.c linux-2.6.38.3/drivers/mtd/nand/denali.c +--- linux-2.6.38.3/drivers/mtd/nand/denali.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/mtd/nand/denali.c 2011-04-17 15:57:32.000000000 -0400 +@@ -25,6 +25,7 @@ + #include <linux/pci.h> + #include <linux/mtd/mtd.h> + #include <linux/module.h> ++#include <linux/slab.h> + + #include "denali.h" + +diff -urNp linux-2.6.38.3/drivers/mtd/ubi/build.c linux-2.6.38.3/drivers/mtd/ubi/build.c +--- linux-2.6.38.3/drivers/mtd/ubi/build.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/mtd/ubi/build.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1285,7 +1285,7 @@ module_exit(ubi_exit); + static int __init bytes_str_to_int(const char *str) + { + char *endp; +- unsigned long result; ++ unsigned long result, scale = 1; + + result = simple_strtoul(str, &endp, 0); + if (str == endp || result >= INT_MAX) { +@@ -1296,11 +1296,11 @@ static int __init bytes_str_to_int(const + + switch (*endp) { + case 'G': +- result *= 1024; ++ scale *= 1024; + case 'M': +- result *= 1024; ++ scale *= 1024; + case 'K': +- result *= 1024; ++ scale *= 1024; + if (endp[1] == 'i' && endp[2] == 'B') + endp += 2; + case '\0': +@@ -1311,7 +1311,13 @@ static int __init bytes_str_to_int(const + return -EINVAL; + } + +- return result; ++ if ((intoverflow_t)result*scale >= INT_MAX) { ++ printk(KERN_ERR "UBI error: incorrect bytes count: \"%s\"\n", ++ str); ++ return -EINVAL; ++ } ++ ++ return result*scale; + } + + /** +diff -urNp linux-2.6.38.3/drivers/net/e1000e/82571.c linux-2.6.38.3/drivers/net/e1000e/82571.c +--- linux-2.6.38.3/drivers/net/e1000e/82571.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/net/e1000e/82571.c 2011-04-17 15:57:32.000000000 -0400 +@@ -239,7 +239,7 @@ static s32 e1000_init_mac_params_82571(s + { + struct e1000_hw *hw = &adapter->hw; + struct e1000_mac_info *mac = &hw->mac; +- struct e1000_mac_operations *func = &mac->ops; ++ struct e1000_mac_operations *func = &mac->ops; /* cannot be const */ + u32 swsm = 0; + u32 swsm2 = 0; + bool force_clear_smbi = false; +@@ -1930,7 +1930,7 @@ static void e1000_clear_hw_cntrs_82571(s + er32(ICRXDMTC); + } + +-static struct e1000_mac_operations e82571_mac_ops = { ++static const struct e1000_mac_operations e82571_mac_ops = { + /* .check_mng_mode: mac type dependent */ + /* .check_for_link: media type dependent */ + .id_led_init = e1000e_id_led_init, +@@ -1952,7 +1952,7 @@ static struct e1000_mac_operations e8257 + .read_mac_addr = e1000_read_mac_addr_82571, + }; + +-static struct e1000_phy_operations e82_phy_ops_igp = { ++static const struct e1000_phy_operations e82_phy_ops_igp = { + .acquire = e1000_get_hw_semaphore_82571, + .check_polarity = e1000_check_polarity_igp, + .check_reset_block = e1000e_check_reset_block_generic, +@@ -1970,7 +1970,7 @@ static struct e1000_phy_operations e82_p + .cfg_on_link_up = NULL, + }; + +-static struct e1000_phy_operations e82_phy_ops_m88 = { ++static const struct e1000_phy_operations e82_phy_ops_m88 = { + .acquire = e1000_get_hw_semaphore_82571, + .check_polarity = e1000_check_polarity_m88, + .check_reset_block = e1000e_check_reset_block_generic, +@@ -1988,7 +1988,7 @@ static struct e1000_phy_operations e82_p + .cfg_on_link_up = NULL, + }; + +-static struct e1000_phy_operations e82_phy_ops_bm = { ++static const struct e1000_phy_operations e82_phy_ops_bm = { + .acquire = e1000_get_hw_semaphore_82571, + .check_polarity = e1000_check_polarity_m88, + .check_reset_block = e1000e_check_reset_block_generic, +@@ -2006,7 +2006,7 @@ static struct e1000_phy_operations e82_p + .cfg_on_link_up = NULL, + }; + +-static struct e1000_nvm_operations e82571_nvm_ops = { ++static const struct e1000_nvm_operations e82571_nvm_ops = { + .acquire = e1000_acquire_nvm_82571, + .read = e1000e_read_nvm_eerd, + .release = e1000_release_nvm_82571, +diff -urNp linux-2.6.38.3/drivers/net/e1000e/e1000.h linux-2.6.38.3/drivers/net/e1000e/e1000.h +--- linux-2.6.38.3/drivers/net/e1000e/e1000.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/net/e1000e/e1000.h 2011-04-17 15:57:32.000000000 -0400 +@@ -408,9 +408,9 @@ struct e1000_info { + u32 pba; + u32 max_hw_frame_size; + s32 (*get_variants)(struct e1000_adapter *); +- struct e1000_mac_operations *mac_ops; +- struct e1000_phy_operations *phy_ops; +- struct e1000_nvm_operations *nvm_ops; ++ const struct e1000_mac_operations *mac_ops; ++ const struct e1000_phy_operations *phy_ops; ++ const struct e1000_nvm_operations *nvm_ops; + }; + + /* hardware capability, feature, and workaround flags */ +diff -urNp linux-2.6.38.3/drivers/net/e1000e/es2lan.c linux-2.6.38.3/drivers/net/e1000e/es2lan.c +--- linux-2.6.38.3/drivers/net/e1000e/es2lan.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/net/e1000e/es2lan.c 2011-04-17 15:57:32.000000000 -0400 +@@ -205,7 +205,7 @@ static s32 e1000_init_mac_params_80003es + { + struct e1000_hw *hw = &adapter->hw; + struct e1000_mac_info *mac = &hw->mac; +- struct e1000_mac_operations *func = &mac->ops; ++ struct e1000_mac_operations *func = &mac->ops; /* cannot be const */ + + /* Set media type */ + switch (adapter->pdev->device) { +@@ -1431,7 +1431,7 @@ static void e1000_clear_hw_cntrs_80003es + er32(ICRXDMTC); + } + +-static struct e1000_mac_operations es2_mac_ops = { ++static const struct e1000_mac_operations es2_mac_ops = { + .read_mac_addr = e1000_read_mac_addr_80003es2lan, + .id_led_init = e1000e_id_led_init, + .check_mng_mode = e1000e_check_mng_mode_generic, +@@ -1453,7 +1453,7 @@ static struct e1000_mac_operations es2_m + .setup_led = e1000e_setup_led_generic, + }; + +-static struct e1000_phy_operations es2_phy_ops = { ++static const struct e1000_phy_operations es2_phy_ops = { + .acquire = e1000_acquire_phy_80003es2lan, + .check_polarity = e1000_check_polarity_m88, + .check_reset_block = e1000e_check_reset_block_generic, +@@ -1471,7 +1471,7 @@ static struct e1000_phy_operations es2_p + .cfg_on_link_up = e1000_cfg_on_link_up_80003es2lan, + }; + +-static struct e1000_nvm_operations es2_nvm_ops = { ++static const struct e1000_nvm_operations es2_nvm_ops = { + .acquire = e1000_acquire_nvm_80003es2lan, + .read = e1000e_read_nvm_eerd, + .release = e1000_release_nvm_80003es2lan, +diff -urNp linux-2.6.38.3/drivers/net/e1000e/hw.h linux-2.6.38.3/drivers/net/e1000e/hw.h +--- linux-2.6.38.3/drivers/net/e1000e/hw.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/net/e1000e/hw.h 2011-04-17 15:57:32.000000000 -0400 +@@ -801,16 +801,17 @@ struct e1000_phy_operations { + + /* Function pointers for the NVM. */ + struct e1000_nvm_operations { +- s32 (*acquire)(struct e1000_hw *); +- s32 (*read)(struct e1000_hw *, u16, u16, u16 *); +- void (*release)(struct e1000_hw *); +- s32 (*update)(struct e1000_hw *); +- s32 (*valid_led_default)(struct e1000_hw *, u16 *); +- s32 (*validate)(struct e1000_hw *); +- s32 (*write)(struct e1000_hw *, u16, u16, u16 *); ++ s32 (* acquire)(struct e1000_hw *); /* cannot be const, see drivers/net/e1000e/82571.c e1000_init_nvm_params_82571() */ ++ s32 (* const read)(struct e1000_hw *, u16, u16, u16 *); ++ void (* release)(struct e1000_hw *); /* cannot be const, see drivers/net/e1000e/82571.c e1000_init_nvm_params_82571() */ ++ s32 (* const update)(struct e1000_hw *); ++ s32 (* const valid_led_default)(struct e1000_hw *, u16 *); ++ s32 (* const validate)(struct e1000_hw *); ++ s32 (* const write)(struct e1000_hw *, u16, u16, u16 *); + }; + + struct e1000_mac_info { ++ /* cannot be const see e1000_init_mac_params_ich8lan */ + struct e1000_mac_operations ops; + + u8 addr[6]; +@@ -853,6 +854,7 @@ struct e1000_mac_info { + }; + + struct e1000_phy_info { ++ /* Cannot be const see e1000_init_phy_params_82571() */ + struct e1000_phy_operations ops; + + enum e1000_phy_type type; +@@ -887,6 +889,7 @@ struct e1000_phy_info { + }; + + struct e1000_nvm_info { ++ /* cannot be const */ + struct e1000_nvm_operations ops; + + enum e1000_nvm_type type; +diff -urNp linux-2.6.38.3/drivers/net/e1000e/ich8lan.c linux-2.6.38.3/drivers/net/e1000e/ich8lan.c +--- linux-2.6.38.3/drivers/net/e1000e/ich8lan.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/net/e1000e/ich8lan.c 2011-04-17 15:57:32.000000000 -0400 +@@ -3840,7 +3840,7 @@ static void e1000_clear_hw_cntrs_ich8lan + } + } + +-static struct e1000_mac_operations ich8_mac_ops = { ++static const struct e1000_mac_operations ich8_mac_ops = { + .id_led_init = e1000e_id_led_init, + /* check_mng_mode dependent on mac type */ + .check_for_link = e1000_check_for_copper_link_ich8lan, +@@ -3859,7 +3859,7 @@ static struct e1000_mac_operations ich8_ + /* id_led_init dependent on mac type */ + }; + +-static struct e1000_phy_operations ich8_phy_ops = { ++static const struct e1000_phy_operations ich8_phy_ops = { + .acquire = e1000_acquire_swflag_ich8lan, + .check_reset_block = e1000_check_reset_block_ich8lan, + .commit = NULL, +@@ -3873,7 +3873,7 @@ static struct e1000_phy_operations ich8_ + .write_reg = e1000e_write_phy_reg_igp, + }; + +-static struct e1000_nvm_operations ich8_nvm_ops = { ++static const struct e1000_nvm_operations ich8_nvm_ops = { + .acquire = e1000_acquire_nvm_ich8lan, + .read = e1000_read_nvm_ich8lan, + .release = e1000_release_nvm_ich8lan, +diff -urNp linux-2.6.38.3/drivers/net/igb/e1000_82575.c linux-2.6.38.3/drivers/net/igb/e1000_82575.c +--- linux-2.6.38.3/drivers/net/igb/e1000_82575.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/net/igb/e1000_82575.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1747,7 +1747,7 @@ u16 igb_rxpbs_adjust_82580(u32 data) + return ret_val; + } + +-static struct e1000_mac_operations e1000_mac_ops_82575 = { ++static const struct e1000_mac_operations e1000_mac_ops_82575 = { + .init_hw = igb_init_hw_82575, + .check_for_link = igb_check_for_link_82575, + .rar_set = igb_rar_set, +@@ -1755,13 +1755,13 @@ static struct e1000_mac_operations e1000 + .get_speed_and_duplex = igb_get_speed_and_duplex_copper, + }; + +-static struct e1000_phy_operations e1000_phy_ops_82575 = { ++static const struct e1000_phy_operations e1000_phy_ops_82575 = { + .acquire = igb_acquire_phy_82575, + .get_cfg_done = igb_get_cfg_done_82575, + .release = igb_release_phy_82575, + }; + +-static struct e1000_nvm_operations e1000_nvm_ops_82575 = { ++static const struct e1000_nvm_operations e1000_nvm_ops_82575 = { + .acquire = igb_acquire_nvm_82575, + .read = igb_read_nvm_eerd, + .release = igb_release_nvm_82575, +diff -urNp linux-2.6.38.3/drivers/net/igb/e1000_hw.h linux-2.6.38.3/drivers/net/igb/e1000_hw.h +--- linux-2.6.38.3/drivers/net/igb/e1000_hw.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/net/igb/e1000_hw.h 2011-04-17 15:57:32.000000000 -0400 +@@ -327,22 +327,23 @@ struct e1000_phy_operations { + }; + + struct e1000_nvm_operations { +- s32 (*acquire)(struct e1000_hw *); +- s32 (*read)(struct e1000_hw *, u16, u16, u16 *); +- void (*release)(struct e1000_hw *); +- s32 (*write)(struct e1000_hw *, u16, u16, u16 *); ++ s32 (* const acquire)(struct e1000_hw *); ++ s32 (* const read)(struct e1000_hw *, u16, u16, u16 *); ++ void (* const release)(struct e1000_hw *); ++ s32 (* const write)(struct e1000_hw *, u16, u16, u16 *); + }; + + struct e1000_info { + s32 (*get_invariants)(struct e1000_hw *); +- struct e1000_mac_operations *mac_ops; +- struct e1000_phy_operations *phy_ops; +- struct e1000_nvm_operations *nvm_ops; ++ const struct e1000_mac_operations *mac_ops; ++ const struct e1000_phy_operations *phy_ops; ++ const struct e1000_nvm_operations *nvm_ops; + }; + + extern const struct e1000_info e1000_82575_info; + + struct e1000_mac_info { ++ /* cannot be const see igb_get_invariants_82575() */ + struct e1000_mac_operations ops; + + u8 addr[6]; +@@ -381,6 +382,7 @@ struct e1000_mac_info { + }; + + struct e1000_phy_info { ++ /* cannot be const see igb_get_invariants_82575() */ + struct e1000_phy_operations ops; + + enum e1000_phy_type type; +@@ -416,6 +418,7 @@ struct e1000_phy_info { + }; + + struct e1000_nvm_info { ++ /* cannot be const */ + struct e1000_nvm_operations ops; + + enum e1000_nvm_type type; +diff -urNp linux-2.6.38.3/drivers/net/igbvf/vf.h linux-2.6.38.3/drivers/net/igbvf/vf.h +--- linux-2.6.38.3/drivers/net/igbvf/vf.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/net/igbvf/vf.h 2011-04-17 15:57:32.000000000 -0400 +@@ -191,6 +191,7 @@ struct e1000_mac_operations { + }; + + struct e1000_mac_info { ++ /* cannot be const see e1000_init_mac_params_vf() */ + struct e1000_mac_operations ops; + u8 addr[6]; + u8 perm_addr[6]; +diff -urNp linux-2.6.38.3/drivers/net/irda/vlsi_ir.c linux-2.6.38.3/drivers/net/irda/vlsi_ir.c +--- linux-2.6.38.3/drivers/net/irda/vlsi_ir.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/net/irda/vlsi_ir.c 2011-04-17 15:57:32.000000000 -0400 +@@ -907,13 +907,12 @@ static netdev_tx_t vlsi_hard_start_xmit( + /* no race - tx-ring already empty */ + vlsi_set_baud(idev, iobase); + netif_wake_queue(ndev); +- } +- else +- ; ++ } else { + /* keep the speed change pending like it would + * for any len>0 packet. tx completion interrupt + * will apply it when the tx ring becomes empty. + */ ++ } + spin_unlock_irqrestore(&idev->lock, flags); + dev_kfree_skb_any(skb); + return NETDEV_TX_OK; +diff -urNp linux-2.6.38.3/drivers/net/pcnet32.c linux-2.6.38.3/drivers/net/pcnet32.c +--- linux-2.6.38.3/drivers/net/pcnet32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/net/pcnet32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -82,7 +82,7 @@ static int cards_found; + /* + * VLB I/O addresses + */ +-static unsigned int pcnet32_portlist[] __initdata = ++static unsigned int pcnet32_portlist[] __devinitdata = + { 0x300, 0x320, 0x340, 0x360, 0 }; + + static int pcnet32_debug; +diff -urNp linux-2.6.38.3/drivers/net/ppp_generic.c linux-2.6.38.3/drivers/net/ppp_generic.c +--- linux-2.6.38.3/drivers/net/ppp_generic.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/net/ppp_generic.c 2011-04-17 15:57:32.000000000 -0400 +@@ -986,7 +986,6 @@ ppp_net_ioctl(struct net_device *dev, st + void __user *addr = (void __user *) ifr->ifr_ifru.ifru_data; + struct ppp_stats stats; + struct ppp_comp_stats cstats; +- char *vers; + + switch (cmd) { + case SIOCGPPPSTATS: +@@ -1008,8 +1007,7 @@ ppp_net_ioctl(struct net_device *dev, st + break; + + case SIOCGPPPVER: +- vers = PPP_VERSION; +- if (copy_to_user(addr, vers, strlen(vers) + 1)) ++ if (copy_to_user(addr, PPP_VERSION, sizeof(PPP_VERSION))) + break; + err = 0; + break; +diff -urNp linux-2.6.38.3/drivers/net/tg3.h linux-2.6.38.3/drivers/net/tg3.h +--- linux-2.6.38.3/drivers/net/tg3.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/net/tg3.h 2011-04-17 15:57:32.000000000 -0400 +@@ -131,6 +131,7 @@ + #define CHIPREV_ID_5750_A0 0x4000 + #define CHIPREV_ID_5750_A1 0x4001 + #define CHIPREV_ID_5750_A3 0x4003 ++#define CHIPREV_ID_5750_C1 0x4201 + #define CHIPREV_ID_5750_C2 0x4202 + #define CHIPREV_ID_5752_A0_HW 0x5000 + #define CHIPREV_ID_5752_A0 0x6000 +diff -urNp linux-2.6.38.3/drivers/net/tulip/de4x5.c linux-2.6.38.3/drivers/net/tulip/de4x5.c +--- linux-2.6.38.3/drivers/net/tulip/de4x5.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/net/tulip/de4x5.c 2011-04-17 15:57:32.000000000 -0400 +@@ -5401,7 +5401,7 @@ de4x5_ioctl(struct net_device *dev, stru + for (i=0; i<ETH_ALEN; i++) { + tmp.addr[i] = dev->dev_addr[i]; + } +- if (copy_to_user(ioc->data, tmp.addr, ioc->len)) return -EFAULT; ++ if (ioc->len > sizeof tmp.addr || copy_to_user(ioc->data, tmp.addr, ioc->len)) return -EFAULT; + break; + + case DE4X5_SET_HWADDR: /* Set the hardware address */ +@@ -5441,7 +5441,7 @@ de4x5_ioctl(struct net_device *dev, stru + spin_lock_irqsave(&lp->lock, flags); + memcpy(&statbuf, &lp->pktStats, ioc->len); + spin_unlock_irqrestore(&lp->lock, flags); +- if (copy_to_user(ioc->data, &statbuf, ioc->len)) ++ if (ioc->len > sizeof statbuf || copy_to_user(ioc->data, &statbuf, ioc->len)) + return -EFAULT; + break; + } +diff -urNp linux-2.6.38.3/drivers/net/usb/hso.c linux-2.6.38.3/drivers/net/usb/hso.c +--- linux-2.6.38.3/drivers/net/usb/hso.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/net/usb/hso.c 2011-04-17 15:57:32.000000000 -0400 +@@ -71,7 +71,7 @@ + #include <asm/byteorder.h> + #include <linux/serial_core.h> + #include <linux/serial.h> +- ++#include <asm/local.h> + + #define MOD_AUTHOR "Option Wireless" + #define MOD_DESCRIPTION "USB High Speed Option driver" +@@ -257,7 +257,7 @@ struct hso_serial { + + /* from usb_serial_port */ + struct tty_struct *tty; +- int open_count; ++ local_t open_count; + spinlock_t serial_lock; + + int (*write_data) (struct hso_serial *serial); +@@ -1190,7 +1190,7 @@ static void put_rxbuf_data_and_resubmit_ + struct urb *urb; + + urb = serial->rx_urb[0]; +- if (serial->open_count > 0) { ++ if (local_read(&serial->open_count) > 0) { + count = put_rxbuf_data(urb, serial); + if (count == -1) + return; +@@ -1226,7 +1226,7 @@ static void hso_std_serial_read_bulk_cal + DUMP1(urb->transfer_buffer, urb->actual_length); + + /* Anyone listening? */ +- if (serial->open_count == 0) ++ if (local_read(&serial->open_count) == 0) + return; + + if (status == 0) { +@@ -1311,8 +1311,7 @@ static int hso_serial_open(struct tty_st + spin_unlock_irq(&serial->serial_lock); + + /* check for port already opened, if not set the termios */ +- serial->open_count++; +- if (serial->open_count == 1) { ++ if (local_inc_return(&serial->open_count) == 1) { + serial->rx_state = RX_IDLE; + /* Force default termio settings */ + _hso_serial_set_termios(tty, NULL); +@@ -1324,7 +1323,7 @@ static int hso_serial_open(struct tty_st + result = hso_start_serial_device(serial->parent, GFP_KERNEL); + if (result) { + hso_stop_serial_device(serial->parent); +- serial->open_count--; ++ local_dec(&serial->open_count); + kref_put(&serial->parent->ref, hso_serial_ref_free); + } + } else { +@@ -1361,10 +1360,10 @@ static void hso_serial_close(struct tty_ + + /* reset the rts and dtr */ + /* do the actual close */ +- serial->open_count--; ++ local_dec(&serial->open_count); + +- if (serial->open_count <= 0) { +- serial->open_count = 0; ++ if (local_read(&serial->open_count) <= 0) { ++ local_set(&serial->open_count, 0); + spin_lock_irq(&serial->serial_lock); + if (serial->tty == tty) { + serial->tty->driver_data = NULL; +@@ -1446,7 +1445,7 @@ static void hso_serial_set_termios(struc + + /* the actual setup */ + spin_lock_irqsave(&serial->serial_lock, flags); +- if (serial->open_count) ++ if (local_read(&serial->open_count)) + _hso_serial_set_termios(tty, old); + else + tty->termios = old; +@@ -1905,7 +1904,7 @@ static void intr_callback(struct urb *ur + D1("Pending read interrupt on port %d\n", i); + spin_lock(&serial->serial_lock); + if (serial->rx_state == RX_IDLE && +- serial->open_count > 0) { ++ local_read(&serial->open_count) > 0) { + /* Setup and send a ctrl req read on + * port i */ + if (!serial->rx_urb_filled[0]) { +@@ -3097,7 +3096,7 @@ static int hso_resume(struct usb_interfa + /* Start all serial ports */ + for (i = 0; i < HSO_SERIAL_TTY_MINORS; i++) { + if (serial_table[i] && (serial_table[i]->interface == iface)) { +- if (dev2ser(serial_table[i])->open_count) { ++ if (local_read(&dev2ser(serial_table[i])->open_count)) { + result = + hso_start_serial_device(serial_table[i], GFP_NOIO); + hso_kick_transmit(dev2ser(serial_table[i])); +diff -urNp linux-2.6.38.3/drivers/net/vmxnet3/vmxnet3_ethtool.c linux-2.6.38.3/drivers/net/vmxnet3/vmxnet3_ethtool.c +--- linux-2.6.38.3/drivers/net/vmxnet3/vmxnet3_ethtool.c 2011-04-17 16:53:01.000000000 -0400 ++++ linux-2.6.38.3/drivers/net/vmxnet3/vmxnet3_ethtool.c 2011-04-17 16:53:41.000000000 -0400 +@@ -628,8 +628,7 @@ vmxnet3_set_rss_indir(struct net_device + * Return with error code if any of the queue indices + * is out of range + */ +- if (p->ring_index[i] < 0 || +- p->ring_index[i] >= adapter->num_rx_queues) ++ if (p->ring_index[i] >= adapter->num_rx_queues) + return -EINVAL; + } + +diff -urNp linux-2.6.38.3/drivers/net/wireless/b43/debugfs.c linux-2.6.38.3/drivers/net/wireless/b43/debugfs.c +--- linux-2.6.38.3/drivers/net/wireless/b43/debugfs.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/net/wireless/b43/debugfs.c 2011-04-17 15:57:32.000000000 -0400 +@@ -43,7 +43,7 @@ static struct dentry *rootdir; + struct b43_debugfs_fops { + ssize_t (*read)(struct b43_wldev *dev, char *buf, size_t bufsize); + int (*write)(struct b43_wldev *dev, const char *buf, size_t count); +- struct file_operations fops; ++ const struct file_operations fops; + /* Offset of struct b43_dfs_file in struct b43_dfsentry */ + size_t file_struct_offset; + }; +diff -urNp linux-2.6.38.3/drivers/net/wireless/b43legacy/debugfs.c linux-2.6.38.3/drivers/net/wireless/b43legacy/debugfs.c +--- linux-2.6.38.3/drivers/net/wireless/b43legacy/debugfs.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/net/wireless/b43legacy/debugfs.c 2011-04-17 15:57:32.000000000 -0400 +@@ -44,7 +44,7 @@ static struct dentry *rootdir; + struct b43legacy_debugfs_fops { + ssize_t (*read)(struct b43legacy_wldev *dev, char *buf, size_t bufsize); + int (*write)(struct b43legacy_wldev *dev, const char *buf, size_t count); +- struct file_operations fops; ++ const struct file_operations fops; + /* Offset of struct b43legacy_dfs_file in struct b43legacy_dfsentry */ + size_t file_struct_offset; + /* Take wl->irq_lock before calling read/write? */ +diff -urNp linux-2.6.38.3/drivers/net/wireless/iwlwifi/iwl-debug.h linux-2.6.38.3/drivers/net/wireless/iwlwifi/iwl-debug.h +--- linux-2.6.38.3/drivers/net/wireless/iwlwifi/iwl-debug.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/net/wireless/iwlwifi/iwl-debug.h 2011-04-17 15:57:32.000000000 -0400 +@@ -68,8 +68,8 @@ do { + } while (0) + + #else +-#define IWL_DEBUG(__priv, level, fmt, args...) +-#define IWL_DEBUG_LIMIT(__priv, level, fmt, args...) ++#define IWL_DEBUG(__priv, level, fmt, args...) do {} while (0) ++#define IWL_DEBUG_LIMIT(__priv, level, fmt, args...) do {} while (0) + static inline void iwl_print_hex_dump(struct iwl_priv *priv, int level, + const void *p, u32 len) + {} +diff -urNp linux-2.6.38.3/drivers/net/wireless/libertas/debugfs.c linux-2.6.38.3/drivers/net/wireless/libertas/debugfs.c +--- linux-2.6.38.3/drivers/net/wireless/libertas/debugfs.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/net/wireless/libertas/debugfs.c 2011-04-17 15:57:32.000000000 -0400 +@@ -702,7 +702,7 @@ out_unlock: + struct lbs_debugfs_files { + const char *name; + int perm; +- struct file_operations fops; ++ const struct file_operations fops; + }; + + static const struct lbs_debugfs_files debugfs_files[] = { +diff -urNp linux-2.6.38.3/drivers/net/wireless/rndis_wlan.c linux-2.6.38.3/drivers/net/wireless/rndis_wlan.c +--- linux-2.6.38.3/drivers/net/wireless/rndis_wlan.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/net/wireless/rndis_wlan.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1277,7 +1277,7 @@ static int set_rts_threshold(struct usbn + + netdev_dbg(usbdev->net, "%s(): %i\n", __func__, rts_threshold); + +- if (rts_threshold < 0 || rts_threshold > 2347) ++ if (rts_threshold > 2347) + rts_threshold = 2347; + + tmp = cpu_to_le32(rts_threshold); +diff -urNp linux-2.6.38.3/drivers/oprofile/buffer_sync.c linux-2.6.38.3/drivers/oprofile/buffer_sync.c +--- linux-2.6.38.3/drivers/oprofile/buffer_sync.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/oprofile/buffer_sync.c 2011-04-17 15:57:32.000000000 -0400 +@@ -342,7 +342,7 @@ static void add_data(struct op_entry *en + if (cookie == NO_COOKIE) + offset = pc; + if (cookie == INVALID_COOKIE) { +- atomic_inc(&oprofile_stats.sample_lost_no_mapping); ++ atomic_inc_unchecked(&oprofile_stats.sample_lost_no_mapping); + offset = pc; + } + if (cookie != last_cookie) { +@@ -386,14 +386,14 @@ add_sample(struct mm_struct *mm, struct + /* add userspace sample */ + + if (!mm) { +- atomic_inc(&oprofile_stats.sample_lost_no_mm); ++ atomic_inc_unchecked(&oprofile_stats.sample_lost_no_mm); + return 0; + } + + cookie = lookup_dcookie(mm, s->eip, &offset); + + if (cookie == INVALID_COOKIE) { +- atomic_inc(&oprofile_stats.sample_lost_no_mapping); ++ atomic_inc_unchecked(&oprofile_stats.sample_lost_no_mapping); + return 0; + } + +@@ -562,7 +562,7 @@ void sync_buffer(int cpu) + /* ignore backtraces if failed to add a sample */ + if (state == sb_bt_start) { + state = sb_bt_ignore; +- atomic_inc(&oprofile_stats.bt_lost_no_mapping); ++ atomic_inc_unchecked(&oprofile_stats.bt_lost_no_mapping); + } + } + release_mm(mm); +diff -urNp linux-2.6.38.3/drivers/oprofile/event_buffer.c linux-2.6.38.3/drivers/oprofile/event_buffer.c +--- linux-2.6.38.3/drivers/oprofile/event_buffer.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/oprofile/event_buffer.c 2011-04-17 15:57:32.000000000 -0400 +@@ -53,7 +53,7 @@ void add_event_entry(unsigned long value + } + + if (buffer_pos == buffer_size) { +- atomic_inc(&oprofile_stats.event_lost_overflow); ++ atomic_inc_unchecked(&oprofile_stats.event_lost_overflow); + return; + } + +diff -urNp linux-2.6.38.3/drivers/oprofile/oprof.c linux-2.6.38.3/drivers/oprofile/oprof.c +--- linux-2.6.38.3/drivers/oprofile/oprof.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/oprofile/oprof.c 2011-04-17 15:57:32.000000000 -0400 +@@ -110,7 +110,7 @@ static void switch_worker(struct work_st + if (oprofile_ops.switch_events()) + return; + +- atomic_inc(&oprofile_stats.multiplex_counter); ++ atomic_inc_unchecked(&oprofile_stats.multiplex_counter); + start_switch_worker(); + } + +diff -urNp linux-2.6.38.3/drivers/oprofile/oprofilefs.c linux-2.6.38.3/drivers/oprofile/oprofilefs.c +--- linux-2.6.38.3/drivers/oprofile/oprofilefs.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/oprofile/oprofilefs.c 2011-04-17 15:57:32.000000000 -0400 +@@ -186,7 +186,7 @@ static const struct file_operations atom + + + int oprofilefs_create_ro_atomic(struct super_block *sb, struct dentry *root, +- char const *name, atomic_t *val) ++ char const *name, atomic_unchecked_t *val) + { + return __oprofilefs_create_file(sb, root, name, + &atomic_ro_fops, 0444, val); +diff -urNp linux-2.6.38.3/drivers/oprofile/oprofile_stats.c linux-2.6.38.3/drivers/oprofile/oprofile_stats.c +--- linux-2.6.38.3/drivers/oprofile/oprofile_stats.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/oprofile/oprofile_stats.c 2011-04-17 15:57:32.000000000 -0400 +@@ -30,11 +30,11 @@ void oprofile_reset_stats(void) + cpu_buf->sample_invalid_eip = 0; + } + +- atomic_set(&oprofile_stats.sample_lost_no_mm, 0); +- atomic_set(&oprofile_stats.sample_lost_no_mapping, 0); +- atomic_set(&oprofile_stats.event_lost_overflow, 0); +- atomic_set(&oprofile_stats.bt_lost_no_mapping, 0); +- atomic_set(&oprofile_stats.multiplex_counter, 0); ++ atomic_set_unchecked(&oprofile_stats.sample_lost_no_mm, 0); ++ atomic_set_unchecked(&oprofile_stats.sample_lost_no_mapping, 0); ++ atomic_set_unchecked(&oprofile_stats.event_lost_overflow, 0); ++ atomic_set_unchecked(&oprofile_stats.bt_lost_no_mapping, 0); ++ atomic_set_unchecked(&oprofile_stats.multiplex_counter, 0); + } + + +diff -urNp linux-2.6.38.3/drivers/oprofile/oprofile_stats.h linux-2.6.38.3/drivers/oprofile/oprofile_stats.h +--- linux-2.6.38.3/drivers/oprofile/oprofile_stats.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/oprofile/oprofile_stats.h 2011-04-17 15:57:32.000000000 -0400 +@@ -13,11 +13,11 @@ + #include <asm/atomic.h> + + struct oprofile_stat_struct { +- atomic_t sample_lost_no_mm; +- atomic_t sample_lost_no_mapping; +- atomic_t bt_lost_no_mapping; +- atomic_t event_lost_overflow; +- atomic_t multiplex_counter; ++ atomic_unchecked_t sample_lost_no_mm; ++ atomic_unchecked_t sample_lost_no_mapping; ++ atomic_unchecked_t bt_lost_no_mapping; ++ atomic_unchecked_t event_lost_overflow; ++ atomic_unchecked_t multiplex_counter; + }; + + extern struct oprofile_stat_struct oprofile_stats; +diff -urNp linux-2.6.38.3/drivers/parport/procfs.c linux-2.6.38.3/drivers/parport/procfs.c +--- linux-2.6.38.3/drivers/parport/procfs.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/parport/procfs.c 2011-04-17 15:57:32.000000000 -0400 +@@ -64,7 +64,7 @@ static int do_active_device(ctl_table *t + + *ppos += len; + +- return copy_to_user(result, buffer, len) ? -EFAULT : 0; ++ return (len > sizeof buffer || copy_to_user(result, buffer, len)) ? -EFAULT : 0; + } + + #ifdef CONFIG_PARPORT_1284 +@@ -106,7 +106,7 @@ static int do_autoprobe(ctl_table *table + + *ppos += len; + +- return copy_to_user (result, buffer, len) ? -EFAULT : 0; ++ return (len > sizeof buffer || copy_to_user (result, buffer, len)) ? -EFAULT : 0; + } + #endif /* IEEE1284.3 support. */ + +diff -urNp linux-2.6.38.3/drivers/pci/hotplug/acpiphp_glue.c linux-2.6.38.3/drivers/pci/hotplug/acpiphp_glue.c +--- linux-2.6.38.3/drivers/pci/hotplug/acpiphp_glue.c 2011-03-28 17:42:40.000000000 -0400 ++++ linux-2.6.38.3/drivers/pci/hotplug/acpiphp_glue.c 2011-04-17 15:57:32.000000000 -0400 +@@ -110,7 +110,7 @@ static int post_dock_fixups(struct notif + } + + +-static struct acpi_dock_ops acpiphp_dock_ops = { ++static const struct acpi_dock_ops acpiphp_dock_ops = { + .handler = handle_hotplug_event_func, + }; + +diff -urNp linux-2.6.38.3/drivers/pci/hotplug/cpqphp_nvram.c linux-2.6.38.3/drivers/pci/hotplug/cpqphp_nvram.c +--- linux-2.6.38.3/drivers/pci/hotplug/cpqphp_nvram.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/pci/hotplug/cpqphp_nvram.c 2011-04-17 15:57:32.000000000 -0400 +@@ -428,9 +428,13 @@ static u32 store_HRT (void __iomem *rom_ + + void compaq_nvram_init (void __iomem *rom_start) + { ++ ++#ifndef CONFIG_PAX_KERNEXEC + if (rom_start) { + compaq_int15_entry_point = (rom_start + ROM_INT15_PHY_ADDR - ROM_PHY_ADDR); + } ++#endif ++ + dbg("int15 entry = %p\n", compaq_int15_entry_point); + + /* initialize our int15 lock */ +diff -urNp linux-2.6.38.3/drivers/pci/intel-iommu.c linux-2.6.38.3/drivers/pci/intel-iommu.c +--- linux-2.6.38.3/drivers/pci/intel-iommu.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/pci/intel-iommu.c 2011-04-17 15:57:32.000000000 -0400 +@@ -2934,7 +2934,7 @@ static int intel_mapping_error(struct de + return !dma_addr; + } + +-struct dma_map_ops intel_dma_ops = { ++const struct dma_map_ops intel_dma_ops = { + .alloc_coherent = intel_alloc_coherent, + .free_coherent = intel_free_coherent, + .map_sg = intel_map_sg, +diff -urNp linux-2.6.38.3/drivers/pci/pcie/aspm.c linux-2.6.38.3/drivers/pci/pcie/aspm.c +--- linux-2.6.38.3/drivers/pci/pcie/aspm.c 2011-04-17 16:53:01.000000000 -0400 ++++ linux-2.6.38.3/drivers/pci/pcie/aspm.c 2011-04-17 16:53:41.000000000 -0400 +@@ -27,9 +27,9 @@ + #define MODULE_PARAM_PREFIX "pcie_aspm." + + /* Note: those are not register definitions */ +-#define ASPM_STATE_L0S_UP (1) /* Upstream direction L0s state */ +-#define ASPM_STATE_L0S_DW (2) /* Downstream direction L0s state */ +-#define ASPM_STATE_L1 (4) /* L1 state */ ++#define ASPM_STATE_L0S_UP (1U) /* Upstream direction L0s state */ ++#define ASPM_STATE_L0S_DW (2U) /* Downstream direction L0s state */ ++#define ASPM_STATE_L1 (4U) /* L1 state */ + #define ASPM_STATE_L0S (ASPM_STATE_L0S_UP | ASPM_STATE_L0S_DW) + #define ASPM_STATE_ALL (ASPM_STATE_L0S | ASPM_STATE_L1) + +diff -urNp linux-2.6.38.3/drivers/pci/pcie/portdrv_pci.c linux-2.6.38.3/drivers/pci/pcie/portdrv_pci.c +--- linux-2.6.38.3/drivers/pci/pcie/portdrv_pci.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/pci/pcie/portdrv_pci.c 2011-04-17 15:57:32.000000000 -0400 +@@ -307,7 +307,7 @@ static void pcie_portdrv_err_resume(stru + static const struct pci_device_id port_pci_ids[] = { { + /* handle any PCI-Express port */ + PCI_DEVICE_CLASS(((PCI_CLASS_BRIDGE_PCI << 8) | 0x00), ~0), +- }, { /* end: all zeroes */ } ++ }, { 0, 0, 0, 0, 0, 0, 0 } + }; + MODULE_DEVICE_TABLE(pci, port_pci_ids); + +diff -urNp linux-2.6.38.3/drivers/pci/probe.c linux-2.6.38.3/drivers/pci/probe.c +--- linux-2.6.38.3/drivers/pci/probe.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/pci/probe.c 2011-04-17 15:57:32.000000000 -0400 +@@ -62,14 +62,14 @@ static ssize_t pci_bus_show_cpuaffinity( + return ret; + } + +-static ssize_t inline pci_bus_show_cpumaskaffinity(struct device *dev, ++static inline ssize_t pci_bus_show_cpumaskaffinity(struct device *dev, + struct device_attribute *attr, + char *buf) + { + return pci_bus_show_cpuaffinity(dev, 0, attr, buf); + } + +-static ssize_t inline pci_bus_show_cpulistaffinity(struct device *dev, ++static inline ssize_t pci_bus_show_cpulistaffinity(struct device *dev, + struct device_attribute *attr, + char *buf) + { +@@ -165,7 +165,7 @@ int __pci_read_base(struct pci_dev *dev, + u32 l, sz, mask; + u16 orig_cmd; + +- mask = type ? PCI_ROM_ADDRESS_MASK : ~0; ++ mask = type ? (u32)PCI_ROM_ADDRESS_MASK : ~0; + + if (!dev->mmio_always_on) { + pci_read_config_word(dev, PCI_COMMAND, &orig_cmd); +diff -urNp linux-2.6.38.3/drivers/pci/proc.c linux-2.6.38.3/drivers/pci/proc.c +--- linux-2.6.38.3/drivers/pci/proc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/pci/proc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -476,7 +476,16 @@ static const struct file_operations proc + static int __init pci_proc_init(void) + { + struct pci_dev *dev = NULL; ++ ++#ifdef CONFIG_GRKERNSEC_PROC_ADD ++#ifdef CONFIG_GRKERNSEC_PROC_USER ++ proc_bus_pci_dir = proc_mkdir_mode("bus/pci", S_IRUSR | S_IXUSR, NULL); ++#elif defined(CONFIG_GRKERNSEC_PROC_USERGROUP) ++ proc_bus_pci_dir = proc_mkdir_mode("bus/pci", S_IRUSR | S_IXUSR | S_IRGRP | S_IXGRP, NULL); ++#endif ++#else + proc_bus_pci_dir = proc_mkdir("bus/pci", NULL); ++#endif + proc_create("devices", 0, proc_bus_pci_dir, + &proc_bus_pci_dev_operations); + proc_initialized = 1; +diff -urNp linux-2.6.38.3/drivers/pcmcia/ti113x.h linux-2.6.38.3/drivers/pcmcia/ti113x.h +--- linux-2.6.38.3/drivers/pcmcia/ti113x.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/pcmcia/ti113x.h 2011-04-17 15:57:32.000000000 -0400 +@@ -936,7 +936,7 @@ static struct pci_device_id ene_tune_tbl + DEVID(PCI_VENDOR_ID_MOTOROLA, 0x3410, 0xECC0, PCI_ANY_ID, + ENE_TEST_C9_TLTENABLE | ENE_TEST_C9_PFENABLE, ENE_TEST_C9_TLTENABLE), + +- {} ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + static void ene_tune_bridge(struct pcmcia_socket *sock, struct pci_bus *bus) +diff -urNp linux-2.6.38.3/drivers/pcmcia/yenta_socket.c linux-2.6.38.3/drivers/pcmcia/yenta_socket.c +--- linux-2.6.38.3/drivers/pcmcia/yenta_socket.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/pcmcia/yenta_socket.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1426,7 +1426,7 @@ static struct pci_device_id yenta_table[ + + /* match any cardbus bridge */ + CB_ID(PCI_ANY_ID, PCI_ANY_ID, DEFAULT), +- { /* all zeroes */ } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + MODULE_DEVICE_TABLE(pci, yenta_table); + +diff -urNp linux-2.6.38.3/drivers/platform/x86/asus-laptop.c linux-2.6.38.3/drivers/platform/x86/asus-laptop.c +--- linux-2.6.38.3/drivers/platform/x86/asus-laptop.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/platform/x86/asus-laptop.c 2011-04-17 15:57:32.000000000 -0400 +@@ -243,7 +243,6 @@ struct asus_laptop { + struct asus_led gled; + struct asus_led kled; + struct workqueue_struct *led_workqueue; +- + int wireless_status; + bool have_rsts; + int lcd_state; +diff -urNp linux-2.6.38.3/drivers/pnp/pnpbios/bioscalls.c linux-2.6.38.3/drivers/pnp/pnpbios/bioscalls.c +--- linux-2.6.38.3/drivers/pnp/pnpbios/bioscalls.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/pnp/pnpbios/bioscalls.c 2011-04-17 15:57:32.000000000 -0400 +@@ -59,7 +59,7 @@ do { \ + set_desc_limit(&gdt[(selname) >> 3], (size) - 1); \ + } while(0) + +-static struct desc_struct bad_bios_desc = GDT_ENTRY_INIT(0x4092, ++static const struct desc_struct bad_bios_desc = GDT_ENTRY_INIT(0x4093, + (unsigned long)__va(0x400UL), PAGE_SIZE - 0x400 - 1); + + /* +@@ -96,7 +96,10 @@ static inline u16 call_pnp_bios(u16 func + + cpu = get_cpu(); + save_desc_40 = get_cpu_gdt_table(cpu)[0x40 / 8]; ++ ++ pax_open_kernel(); + get_cpu_gdt_table(cpu)[0x40 / 8] = bad_bios_desc; ++ pax_close_kernel(); + + /* On some boxes IRQ's during PnP BIOS calls are deadly. */ + spin_lock_irqsave(&pnp_bios_lock, flags); +@@ -134,7 +137,10 @@ static inline u16 call_pnp_bios(u16 func + :"memory"); + spin_unlock_irqrestore(&pnp_bios_lock, flags); + ++ pax_open_kernel(); + get_cpu_gdt_table(cpu)[0x40 / 8] = save_desc_40; ++ pax_close_kernel(); ++ + put_cpu(); + + /* If we get here and this is set then the PnP BIOS faulted on us. */ +@@ -468,7 +474,7 @@ int pnp_bios_read_escd(char *data, u32 n + return status; + } + +-void pnpbios_calls_init(union pnp_bios_install_struct *header) ++void __init pnpbios_calls_init(union pnp_bios_install_struct *header) + { + int i; + +@@ -476,6 +482,8 @@ void pnpbios_calls_init(union pnp_bios_i + pnp_bios_callpoint.offset = header->fields.pm16offset; + pnp_bios_callpoint.segment = PNP_CS16; + ++ pax_open_kernel(); ++ + for_each_possible_cpu(i) { + struct desc_struct *gdt = get_cpu_gdt_table(i); + if (!gdt) +@@ -487,4 +495,6 @@ void pnpbios_calls_init(union pnp_bios_i + set_desc_base(&gdt[GDT_ENTRY_PNPBIOS_DS], + (unsigned long)__va(header->fields.pm16dseg)); + } ++ ++ pax_close_kernel(); + } +diff -urNp linux-2.6.38.3/drivers/pnp/quirks.c linux-2.6.38.3/drivers/pnp/quirks.c +--- linux-2.6.38.3/drivers/pnp/quirks.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/pnp/quirks.c 2011-04-17 15:57:32.000000000 -0400 +@@ -322,7 +322,7 @@ static struct pnp_fixup pnp_fixups[] = { + /* PnP resources that might overlap PCI BARs */ + {"PNP0c01", quirk_system_pci_resources}, + {"PNP0c02", quirk_system_pci_resources}, +- {""} ++ {"", NULL} + }; + + void pnp_fixup_device(struct pnp_dev *dev) +diff -urNp linux-2.6.38.3/drivers/pnp/resource.c linux-2.6.38.3/drivers/pnp/resource.c +--- linux-2.6.38.3/drivers/pnp/resource.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/pnp/resource.c 2011-04-17 15:57:32.000000000 -0400 +@@ -360,7 +360,7 @@ int pnp_check_irq(struct pnp_dev *dev, s + return 1; + + /* check if the resource is valid */ +- if (*irq < 0 || *irq > 15) ++ if (*irq > 15) + return 0; + + /* check if the resource is reserved */ +@@ -424,7 +424,7 @@ int pnp_check_dma(struct pnp_dev *dev, s + return 1; + + /* check if the resource is valid */ +- if (*dma < 0 || *dma == 4 || *dma > 7) ++ if (*dma == 4 || *dma > 7) + return 0; + + /* check if the resource is reserved */ +diff -urNp linux-2.6.38.3/drivers/rtc/rtc-dev.c linux-2.6.38.3/drivers/rtc/rtc-dev.c +--- linux-2.6.38.3/drivers/rtc/rtc-dev.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/rtc/rtc-dev.c 2011-04-17 15:57:32.000000000 -0400 +@@ -14,6 +14,7 @@ + #include <linux/module.h> + #include <linux/rtc.h> + #include <linux/sched.h> ++#include <linux/grsecurity.h> + #include "rtc-core.h" + + static dev_t rtc_devt; +@@ -345,6 +346,8 @@ static long rtc_dev_ioctl(struct file *f + if (copy_from_user(&tm, uarg, sizeof(tm))) + return -EFAULT; + ++ gr_log_timechange(); ++ + return rtc_set_time(rtc, &tm); + + case RTC_PIE_ON: +diff -urNp linux-2.6.38.3/drivers/s390/cio/qdio_debug.c linux-2.6.38.3/drivers/s390/cio/qdio_debug.c +--- linux-2.6.38.3/drivers/s390/cio/qdio_debug.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/s390/cio/qdio_debug.c 2011-04-17 15:57:32.000000000 -0400 +@@ -225,7 +225,7 @@ static int qperf_seq_open(struct inode * + filp->f_path.dentry->d_inode->i_private); + } + +-static struct file_operations debugfs_perf_fops = { ++static const struct file_operations debugfs_perf_fops = { + .owner = THIS_MODULE, + .open = qperf_seq_open, + .read = seq_read, +diff -urNp linux-2.6.38.3/drivers/scsi/aic94xx/aic94xx_init.c linux-2.6.38.3/drivers/scsi/aic94xx/aic94xx_init.c +--- linux-2.6.38.3/drivers/scsi/aic94xx/aic94xx_init.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/scsi/aic94xx/aic94xx_init.c 2011-04-17 15:57:32.000000000 -0400 +@@ -486,7 +486,7 @@ static ssize_t asd_show_update_bios(stru + flash_error_table[i].reason); + } + +-static DEVICE_ATTR(update_bios, S_IRUGO|S_IWUGO, ++static DEVICE_ATTR(update_bios, S_IRUGO|S_IWUSR, + asd_show_update_bios, asd_store_update_bios); + + static int asd_create_dev_attrs(struct asd_ha_struct *asd_ha) +diff -urNp linux-2.6.38.3/drivers/scsi/hpsa.c linux-2.6.38.3/drivers/scsi/hpsa.c +--- linux-2.6.38.3/drivers/scsi/hpsa.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/scsi/hpsa.c 2011-04-17 15:57:32.000000000 -0400 +@@ -2281,6 +2281,8 @@ static int hpsa_ioctl32_passthru(struct + int err; + u32 cp; + ++ memset(&arg64, 0, sizeof(arg64)); ++ + err = 0; + err |= copy_from_user(&arg64.LUN_info, &arg32->LUN_info, + sizeof(arg64.LUN_info)); +diff -urNp linux-2.6.38.3/drivers/scsi/ipr.c linux-2.6.38.3/drivers/scsi/ipr.c +--- linux-2.6.38.3/drivers/scsi/ipr.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/scsi/ipr.c 2011-04-17 15:57:32.000000000 -0400 +@@ -6207,7 +6207,7 @@ static bool ipr_qc_fill_rtf(struct ata_q + return true; + } + +-static struct ata_port_operations ipr_sata_ops = { ++static const struct ata_port_operations ipr_sata_ops = { + .phy_reset = ipr_ata_phy_reset, + .hardreset = ipr_sata_reset, + .post_internal_cmd = ipr_ata_post_internal, +diff -urNp linux-2.6.38.3/drivers/scsi/libfc/fc_exch.c linux-2.6.38.3/drivers/scsi/libfc/fc_exch.c +--- linux-2.6.38.3/drivers/scsi/libfc/fc_exch.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/scsi/libfc/fc_exch.c 2011-04-17 15:57:32.000000000 -0400 +@@ -105,12 +105,12 @@ struct fc_exch_mgr { + * all together if not used XXX + */ + struct { +- atomic_t no_free_exch; +- atomic_t no_free_exch_xid; +- atomic_t xid_not_found; +- atomic_t xid_busy; +- atomic_t seq_not_found; +- atomic_t non_bls_resp; ++ atomic_unchecked_t no_free_exch; ++ atomic_unchecked_t no_free_exch_xid; ++ atomic_unchecked_t xid_not_found; ++ atomic_unchecked_t xid_busy; ++ atomic_unchecked_t seq_not_found; ++ atomic_unchecked_t non_bls_resp; + } stats; + }; + +@@ -687,7 +687,7 @@ static struct fc_exch *fc_exch_em_alloc( + /* allocate memory for exchange */ + ep = mempool_alloc(mp->ep_pool, GFP_ATOMIC); + if (!ep) { +- atomic_inc(&mp->stats.no_free_exch); ++ atomic_inc_unchecked(&mp->stats.no_free_exch); + goto out; + } + memset(ep, 0, sizeof(*ep)); +@@ -748,7 +748,7 @@ out: + return ep; + err: + spin_unlock_bh(&pool->lock); +- atomic_inc(&mp->stats.no_free_exch_xid); ++ atomic_inc_unchecked(&mp->stats.no_free_exch_xid); + mempool_free(ep, mp->ep_pool); + return NULL; + } +@@ -893,7 +893,7 @@ static enum fc_pf_rjt_reason fc_seq_look + xid = ntohs(fh->fh_ox_id); /* we originated exch */ + ep = fc_exch_find(mp, xid); + if (!ep) { +- atomic_inc(&mp->stats.xid_not_found); ++ atomic_inc_unchecked(&mp->stats.xid_not_found); + reject = FC_RJT_OX_ID; + goto out; + } +@@ -923,7 +923,7 @@ static enum fc_pf_rjt_reason fc_seq_look + ep = fc_exch_find(mp, xid); + if ((f_ctl & FC_FC_FIRST_SEQ) && fc_sof_is_init(fr_sof(fp))) { + if (ep) { +- atomic_inc(&mp->stats.xid_busy); ++ atomic_inc_unchecked(&mp->stats.xid_busy); + reject = FC_RJT_RX_ID; + goto rel; + } +@@ -934,7 +934,7 @@ static enum fc_pf_rjt_reason fc_seq_look + } + xid = ep->xid; /* get our XID */ + } else if (!ep) { +- atomic_inc(&mp->stats.xid_not_found); ++ atomic_inc_unchecked(&mp->stats.xid_not_found); + reject = FC_RJT_RX_ID; /* XID not found */ + goto out; + } +@@ -951,7 +951,7 @@ static enum fc_pf_rjt_reason fc_seq_look + } else { + sp = &ep->seq; + if (sp->id != fh->fh_seq_id) { +- atomic_inc(&mp->stats.seq_not_found); ++ atomic_inc_unchecked(&mp->stats.seq_not_found); + reject = FC_RJT_SEQ_ID; /* sequence/exch should exist */ + goto rel; + } +@@ -1368,22 +1368,22 @@ static void fc_exch_recv_seq_resp(struct + + ep = fc_exch_find(mp, ntohs(fh->fh_ox_id)); + if (!ep) { +- atomic_inc(&mp->stats.xid_not_found); ++ atomic_inc_unchecked(&mp->stats.xid_not_found); + goto out; + } + if (ep->esb_stat & ESB_ST_COMPLETE) { +- atomic_inc(&mp->stats.xid_not_found); ++ atomic_inc_unchecked(&mp->stats.xid_not_found); + goto rel; + } + if (ep->rxid == FC_XID_UNKNOWN) + ep->rxid = ntohs(fh->fh_rx_id); + if (ep->sid != 0 && ep->sid != ntoh24(fh->fh_d_id)) { +- atomic_inc(&mp->stats.xid_not_found); ++ atomic_inc_unchecked(&mp->stats.xid_not_found); + goto rel; + } + if (ep->did != ntoh24(fh->fh_s_id) && + ep->did != FC_FID_FLOGI) { +- atomic_inc(&mp->stats.xid_not_found); ++ atomic_inc_unchecked(&mp->stats.xid_not_found); + goto rel; + } + sof = fr_sof(fp); +@@ -1392,7 +1392,7 @@ static void fc_exch_recv_seq_resp(struct + sp->ssb_stat |= SSB_ST_RESP; + sp->id = fh->fh_seq_id; + } else if (sp->id != fh->fh_seq_id) { +- atomic_inc(&mp->stats.seq_not_found); ++ atomic_inc_unchecked(&mp->stats.seq_not_found); + goto rel; + } + +@@ -1455,9 +1455,9 @@ static void fc_exch_recv_resp(struct fc_ + sp = fc_seq_lookup_orig(mp, fp); /* doesn't hold sequence */ + + if (!sp) +- atomic_inc(&mp->stats.xid_not_found); ++ atomic_inc_unchecked(&mp->stats.xid_not_found); + else +- atomic_inc(&mp->stats.non_bls_resp); ++ atomic_inc_unchecked(&mp->stats.non_bls_resp); + + fc_frame_free(fp); + } +diff -urNp linux-2.6.38.3/drivers/scsi/libsas/sas_ata.c linux-2.6.38.3/drivers/scsi/libsas/sas_ata.c +--- linux-2.6.38.3/drivers/scsi/libsas/sas_ata.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/scsi/libsas/sas_ata.c 2011-04-17 15:57:32.000000000 -0400 +@@ -348,10 +348,10 @@ static int sas_ata_scr_read(struct ata_l + } + } + +-static struct ata_port_operations sas_sata_ops = { ++static const struct ata_port_operations sas_sata_ops = { + .phy_reset = sas_ata_phy_reset, + .post_internal_cmd = sas_ata_post_internal, +- .qc_defer = ata_std_qc_defer, ++ .qc_defer = ata_std_qc_defer, + .qc_prep = ata_noop_qc_prep, + .qc_issue = sas_ata_qc_issue, + .qc_fill_rtf = sas_ata_qc_fill_rtf, +diff -urNp linux-2.6.38.3/drivers/scsi/mpt2sas/mpt2sas_debug.h linux-2.6.38.3/drivers/scsi/mpt2sas/mpt2sas_debug.h +--- linux-2.6.38.3/drivers/scsi/mpt2sas/mpt2sas_debug.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/scsi/mpt2sas/mpt2sas_debug.h 2011-04-17 15:57:32.000000000 -0400 +@@ -79,7 +79,7 @@ + CMD; \ + } + #else +-#define MPT_CHECK_LOGGING(IOC, CMD, BITS) ++#define MPT_CHECK_LOGGING(IOC, CMD, BITS) do {} while (0) + #endif /* CONFIG_SCSI_MPT2SAS_LOGGING */ + + +diff -urNp linux-2.6.38.3/drivers/scsi/qla2xxx/qla_os.c linux-2.6.38.3/drivers/scsi/qla2xxx/qla_os.c +--- linux-2.6.38.3/drivers/scsi/qla2xxx/qla_os.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/scsi/qla2xxx/qla_os.c 2011-04-17 15:57:32.000000000 -0400 +@@ -4096,7 +4096,7 @@ static struct pci_driver qla2xxx_pci_dri + .err_handler = &qla2xxx_err_handler, + }; + +-static struct file_operations apidev_fops = { ++static const struct file_operations apidev_fops = { + .owner = THIS_MODULE, + .llseek = noop_llseek, + }; +diff -urNp linux-2.6.38.3/drivers/scsi/scsi_logging.h linux-2.6.38.3/drivers/scsi/scsi_logging.h +--- linux-2.6.38.3/drivers/scsi/scsi_logging.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/scsi/scsi_logging.h 2011-04-17 15:57:32.000000000 -0400 +@@ -51,7 +51,7 @@ do { \ + } while (0); \ + } while (0) + #else +-#define SCSI_CHECK_LOGGING(SHIFT, BITS, LEVEL, CMD) ++#define SCSI_CHECK_LOGGING(SHIFT, BITS, LEVEL, CMD) do {} while (0) + #endif /* CONFIG_SCSI_LOGGING */ + + /* +diff -urNp linux-2.6.38.3/drivers/scsi/scsi_transport_fc.c linux-2.6.38.3/drivers/scsi/scsi_transport_fc.c +--- linux-2.6.38.3/drivers/scsi/scsi_transport_fc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/scsi/scsi_transport_fc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -836,7 +836,7 @@ static int fc_str_to_dev_loss(const char + char *cp; + + *val = simple_strtoul(buf, &cp, 0); +- if ((*cp && (*cp != '\n')) || (*val < 0)) ++ if (*cp && (*cp != '\n')) + return -EINVAL; + /* + * Check for overflow; dev_loss_tmo is u32 +diff -urNp linux-2.6.38.3/drivers/scsi/sg.c linux-2.6.38.3/drivers/scsi/sg.c +--- linux-2.6.38.3/drivers/scsi/sg.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/scsi/sg.c 2011-04-17 15:57:32.000000000 -0400 +@@ -2310,7 +2310,7 @@ struct sg_proc_leaf { + const struct file_operations * fops; + }; + +-static struct sg_proc_leaf sg_proc_leaf_arr[] = { ++static const struct sg_proc_leaf sg_proc_leaf_arr[] = { + {"allow_dio", &adio_fops}, + {"debug", &debug_fops}, + {"def_reserved_size", &dressz_fops}, +@@ -2325,7 +2325,7 @@ sg_proc_init(void) + { + int k, mask; + int num_leaves = ARRAY_SIZE(sg_proc_leaf_arr); +- struct sg_proc_leaf * leaf; ++ const struct sg_proc_leaf * leaf; + + sg_proc_sgp = proc_mkdir(sg_proc_sg_dirname, NULL); + if (!sg_proc_sgp) +diff -urNp linux-2.6.38.3/drivers/staging/autofs/root.c linux-2.6.38.3/drivers/staging/autofs/root.c +--- linux-2.6.38.3/drivers/staging/autofs/root.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/staging/autofs/root.c 2011-04-17 15:57:32.000000000 -0400 +@@ -311,7 +311,8 @@ static int autofs_root_symlink(struct in + set_bit(n,sbi->symlink_bitmap); + sl = &sbi->symlink[n]; + sl->len = strlen(symname); +- sl->data = kmalloc(slsize = sl->len+1, GFP_KERNEL); ++ slsize = sl->len+1; ++ sl->data = kmalloc(slsize, GFP_KERNEL); + if (!sl->data) { + clear_bit(n,sbi->symlink_bitmap); + unlock_kernel(); +diff -urNp linux-2.6.38.3/drivers/staging/bcm/Bcmchar.c linux-2.6.38.3/drivers/staging/bcm/Bcmchar.c +--- linux-2.6.38.3/drivers/staging/bcm/Bcmchar.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/staging/bcm/Bcmchar.c 2011-04-17 15:57:32.000000000 -0400 +@@ -2093,7 +2093,7 @@ static long bcm_char_ioctl(struct file * + } + + +-static struct file_operations bcm_fops = { ++static const struct file_operations bcm_fops = { + .owner = THIS_MODULE, + .open = bcm_char_open, + .release = bcm_char_release, +diff -urNp linux-2.6.38.3/drivers/staging/brcm80211/brcmfmac/dhd_linux.c linux-2.6.38.3/drivers/staging/brcm80211/brcmfmac/dhd_linux.c +--- linux-2.6.38.3/drivers/staging/brcm80211/brcmfmac/dhd_linux.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/staging/brcm80211/brcmfmac/dhd_linux.c 2011-04-17 15:57:32.000000000 -0400 +@@ -863,14 +863,14 @@ static void dhd_op_if(dhd_if_t *ifp) + free_netdev(ifp->net); + } + /* Allocate etherdev, including space for private structure */ +- ifp->net = alloc_etherdev(sizeof(dhd)); ++ ifp->net = alloc_etherdev(sizeof(*dhd)); + if (!ifp->net) { + DHD_ERROR(("%s: OOM - alloc_etherdev\n", __func__)); + ret = -ENOMEM; + } + if (ret == 0) { + strcpy(ifp->net->name, ifp->name); +- memcpy(netdev_priv(ifp->net), &dhd, sizeof(dhd)); ++ memcpy(netdev_priv(ifp->net), dhd, sizeof(*dhd)); + err = dhd_net_attach(&dhd->pub, ifp->idx); + if (err != 0) { + DHD_ERROR(("%s: dhd_net_attach failed, " +@@ -1969,25 +1969,23 @@ dhd_pub_t *dhd_attach(struct osl_info *o + strcpy(nv_path, nvram_path); + + /* Allocate etherdev, including space for private structure */ +- net = alloc_etherdev(sizeof(dhd)); ++ net = alloc_etherdev(sizeof(*dhd)); + if (!net) { + DHD_ERROR(("%s: OOM - alloc_etherdev\n", __func__)); + goto fail; + } + + /* Allocate primary dhd_info */ +- dhd = kmalloc(sizeof(dhd_info_t), GFP_ATOMIC); ++ dhd = kzalloc(sizeof(dhd_info_t), GFP_ATOMIC); + if (!dhd) { + DHD_ERROR(("%s: OOM - alloc dhd_info\n", __func__)); + goto fail; + } + +- memset(dhd, 0, sizeof(dhd_info_t)); +- + /* + * Save the dhd_info into the priv + */ +- memcpy(netdev_priv(net), &dhd, sizeof(dhd)); ++ memcpy(netdev_priv(net), dhd, sizeof(*dhd)); + dhd->pub.osh = osh; + + /* Set network interface name if it was provided as module parameter */ +@@ -2105,7 +2103,7 @@ dhd_pub_t *dhd_attach(struct osl_info *o + /* + * Save the dhd_info into the priv + */ +- memcpy(netdev_priv(net), &dhd, sizeof(dhd)); ++ memcpy(netdev_priv(net), dhd, sizeof(*dhd)); + + #if defined(CUSTOMER_HW2) && defined(CONFIG_WIFI_CONTROL_FUNC) + g_bus = bus; +diff -urNp linux-2.6.38.3/drivers/staging/brcm80211/brcmfmac/wl_iw.c linux-2.6.38.3/drivers/staging/brcm80211/brcmfmac/wl_iw.c +--- linux-2.6.38.3/drivers/staging/brcm80211/brcmfmac/wl_iw.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/staging/brcm80211/brcmfmac/wl_iw.c 2011-04-17 15:57:32.000000000 -0400 +@@ -513,7 +513,7 @@ wl_iw_get_range(struct net_device *dev, + list = (wl_u32_list_t *) channels; + + dwrq->length = sizeof(struct iw_range); +- memset(range, 0, sizeof(range)); ++ memset(range, 0, sizeof(*range)); + + range->min_nwid = range->max_nwid = 0; + +diff -urNp linux-2.6.38.3/drivers/staging/comedi/comedi_fops.c linux-2.6.38.3/drivers/staging/comedi/comedi_fops.c +--- linux-2.6.38.3/drivers/staging/comedi/comedi_fops.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/staging/comedi/comedi_fops.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1426,7 +1426,7 @@ static void comedi_unmap(struct vm_area_ + mutex_unlock(&dev->mutex); + } + +-static struct vm_operations_struct comedi_vm_ops = { ++static const struct vm_operations_struct comedi_vm_ops = { + .close = comedi_unmap, + }; + +diff -urNp linux-2.6.38.3/drivers/staging/ft1000/ft1000-usb/ft1000_debug.c linux-2.6.38.3/drivers/staging/ft1000/ft1000-usb/ft1000_debug.c +--- linux-2.6.38.3/drivers/staging/ft1000/ft1000-usb/ft1000_debug.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/staging/ft1000/ft1000-usb/ft1000_debug.c 2011-04-17 15:57:32.000000000 -0400 +@@ -55,7 +55,7 @@ int numofmsgbuf = 0; + // + // Table of entry-point routines for char device + // +-static struct file_operations ft1000fops = ++static const struct file_operations ft1000fops = + { + .unlocked_ioctl = ft1000_ioctl, + .poll = ft1000_poll_dev, +diff -urNp linux-2.6.38.3/drivers/staging/go7007/go7007-v4l2.c linux-2.6.38.3/drivers/staging/go7007/go7007-v4l2.c +--- linux-2.6.38.3/drivers/staging/go7007/go7007-v4l2.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/staging/go7007/go7007-v4l2.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1672,7 +1672,7 @@ static int go7007_vm_fault(struct vm_are + return 0; + } + +-static struct vm_operations_struct go7007_vm_ops = { ++static const struct vm_operations_struct go7007_vm_ops = { + .open = go7007_vm_open, + .close = go7007_vm_close, + .fault = go7007_vm_fault, +diff -urNp linux-2.6.38.3/drivers/staging/hv/hv.c linux-2.6.38.3/drivers/staging/hv/hv.c +--- linux-2.6.38.3/drivers/staging/hv/hv.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/staging/hv/hv.c 2011-04-17 15:57:32.000000000 -0400 +@@ -163,7 +163,7 @@ static u64 do_hypercall(u64 control, voi + u64 output_address = (output) ? virt_to_phys(output) : 0; + u32 output_address_hi = output_address >> 32; + u32 output_address_lo = output_address & 0xFFFFFFFF; +- volatile void *hypercall_page = hv_context.hypercall_page; ++ volatile void *hypercall_page = ktva_ktla(hv_context.hypercall_page); + + DPRINT_DBG(VMBUS, "Hypercall <control %llx input %p output %p>", + control, input, output); +diff -urNp linux-2.6.38.3/drivers/staging/phison/phison.c linux-2.6.38.3/drivers/staging/phison/phison.c +--- linux-2.6.38.3/drivers/staging/phison/phison.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/staging/phison/phison.c 2011-04-17 15:57:32.000000000 -0400 +@@ -43,7 +43,7 @@ static struct scsi_host_template phison_ + ATA_BMDMA_SHT(DRV_NAME), + }; + +-static struct ata_port_operations phison_ops = { ++static const struct ata_port_operations phison_ops = { + .inherits = &ata_bmdma_port_ops, + .prereset = phison_pre_reset, + }; +diff -urNp linux-2.6.38.3/drivers/staging/pohmelfs/inode.c linux-2.6.38.3/drivers/staging/pohmelfs/inode.c +--- linux-2.6.38.3/drivers/staging/pohmelfs/inode.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/staging/pohmelfs/inode.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1855,7 +1855,7 @@ static int pohmelfs_fill_super(struct su + mutex_init(&psb->mcache_lock); + psb->mcache_root = RB_ROOT; + psb->mcache_timeout = msecs_to_jiffies(5000); +- atomic_long_set(&psb->mcache_gen, 0); ++ atomic_long_set_unchecked(&psb->mcache_gen, 0); + + psb->trans_max_pages = 100; + +diff -urNp linux-2.6.38.3/drivers/staging/pohmelfs/mcache.c linux-2.6.38.3/drivers/staging/pohmelfs/mcache.c +--- linux-2.6.38.3/drivers/staging/pohmelfs/mcache.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/staging/pohmelfs/mcache.c 2011-04-17 15:57:32.000000000 -0400 +@@ -121,7 +121,7 @@ struct pohmelfs_mcache *pohmelfs_mcache_ + m->data = data; + m->start = start; + m->size = size; +- m->gen = atomic_long_inc_return(&psb->mcache_gen); ++ m->gen = atomic_long_inc_return_unchecked(&psb->mcache_gen); + + mutex_lock(&psb->mcache_lock); + err = pohmelfs_mcache_insert(psb, m); +diff -urNp linux-2.6.38.3/drivers/staging/pohmelfs/netfs.h linux-2.6.38.3/drivers/staging/pohmelfs/netfs.h +--- linux-2.6.38.3/drivers/staging/pohmelfs/netfs.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/staging/pohmelfs/netfs.h 2011-04-17 15:57:32.000000000 -0400 +@@ -571,7 +571,7 @@ struct pohmelfs_config; + struct pohmelfs_sb { + struct rb_root mcache_root; + struct mutex mcache_lock; +- atomic_long_t mcache_gen; ++ atomic_long_unchecked_t mcache_gen; + unsigned long mcache_timeout; + + unsigned int idx; +diff -urNp linux-2.6.38.3/drivers/staging/rtl8192u/ieee80211/proc.c linux-2.6.38.3/drivers/staging/rtl8192u/ieee80211/proc.c +--- linux-2.6.38.3/drivers/staging/rtl8192u/ieee80211/proc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/staging/rtl8192u/ieee80211/proc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -99,7 +99,7 @@ static int crypto_info_open(struct inode + return seq_open(file, &crypto_seq_ops); + } + +-static struct file_operations proc_crypto_ops = { ++static const struct file_operations proc_crypto_ops = { + .open = crypto_info_open, + .read = seq_read, + .llseek = seq_lseek, +diff -urNp linux-2.6.38.3/drivers/staging/spectra/ffsport.c linux-2.6.38.3/drivers/staging/spectra/ffsport.c +--- linux-2.6.38.3/drivers/staging/spectra/ffsport.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/staging/spectra/ffsport.c 2011-04-17 15:57:32.000000000 -0400 +@@ -604,7 +604,7 @@ int GLOB_SBD_unlocked_ioctl(struct block + return ret; + } + +-static struct block_device_operations GLOB_SBD_ops = { ++static const struct block_device_operations GLOB_SBD_ops = { + .owner = THIS_MODULE, + .open = GLOB_SBD_open, + .release = GLOB_SBD_release, +diff -urNp linux-2.6.38.3/drivers/staging/vme/devices/vme_user.c linux-2.6.38.3/drivers/staging/vme/devices/vme_user.c +--- linux-2.6.38.3/drivers/staging/vme/devices/vme_user.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/staging/vme/devices/vme_user.c 2011-04-17 15:57:32.000000000 -0400 +@@ -138,7 +138,7 @@ static long vme_user_unlocked_ioctl(stru + static int __devinit vme_user_probe(struct device *, int, int); + static int __devexit vme_user_remove(struct device *, int, int); + +-static struct file_operations vme_user_fops = { ++static const struct file_operations vme_user_fops = { + .open = vme_user_open, + .release = vme_user_release, + .read = vme_user_read, +diff -urNp linux-2.6.38.3/drivers/staging/westbridge/astoria/block/cyasblkdev_block.c linux-2.6.38.3/drivers/staging/westbridge/astoria/block/cyasblkdev_block.c +--- linux-2.6.38.3/drivers/staging/westbridge/astoria/block/cyasblkdev_block.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/staging/westbridge/astoria/block/cyasblkdev_block.c 2011-04-17 15:57:32.000000000 -0400 +@@ -426,7 +426,7 @@ int cyasblkdev_revalidate_disk(struct ge + + + /*standard block device driver interface */ +-static struct block_device_operations cyasblkdev_bdops = { ++static const struct block_device_operations cyasblkdev_bdops = { + .open = cyasblkdev_blk_open, + .release = cyasblkdev_blk_release, + .ioctl = cyasblkdev_blk_ioctl, +diff -urNp linux-2.6.38.3/drivers/tty/hvc/hvc_console.h linux-2.6.38.3/drivers/tty/hvc/hvc_console.h +--- linux-2.6.38.3/drivers/tty/hvc/hvc_console.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/tty/hvc/hvc_console.h 2011-04-17 15:57:32.000000000 -0400 +@@ -82,6 +82,7 @@ extern int hvc_instantiate(uint32_t vter + /* register a vterm for hvc tty operation (module_init or hotplug add) */ + extern struct hvc_struct * hvc_alloc(uint32_t vtermno, int data, + const struct hv_ops *ops, int outbuf_size); ++ + /* remove a vterm from hvc tty operation (module_exit or hotplug remove) */ + extern int hvc_remove(struct hvc_struct *hp); + +diff -urNp linux-2.6.38.3/drivers/tty/hvc/hvcs.c linux-2.6.38.3/drivers/tty/hvc/hvcs.c +--- linux-2.6.38.3/drivers/tty/hvc/hvcs.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/tty/hvc/hvcs.c 2011-04-17 15:57:32.000000000 -0400 +@@ -83,6 +83,7 @@ + #include <asm/hvcserver.h> + #include <asm/uaccess.h> + #include <asm/vio.h> ++#include <asm/local.h> + + /* + * 1.3.0 -> 1.3.1 In hvcs_open memset(..,0x00,..) instead of memset(..,0x3F,00). +@@ -270,7 +271,7 @@ struct hvcs_struct { + unsigned int index; + + struct tty_struct *tty; +- int open_count; ++ local_t open_count; + + /* + * Used to tell the driver kernel_thread what operations need to take +@@ -420,7 +421,7 @@ static ssize_t hvcs_vterm_state_store(st + + spin_lock_irqsave(&hvcsd->lock, flags); + +- if (hvcsd->open_count > 0) { ++ if (local_read(&hvcsd->open_count) > 0) { + spin_unlock_irqrestore(&hvcsd->lock, flags); + printk(KERN_INFO "HVCS: vterm state unchanged. " + "The hvcs device node is still in use.\n"); +@@ -1136,7 +1137,7 @@ static int hvcs_open(struct tty_struct * + if ((retval = hvcs_partner_connect(hvcsd))) + goto error_release; + +- hvcsd->open_count = 1; ++ local_set(&hvcsd->open_count, 1); + hvcsd->tty = tty; + tty->driver_data = hvcsd; + +@@ -1170,7 +1171,7 @@ fast_open: + + spin_lock_irqsave(&hvcsd->lock, flags); + kref_get(&hvcsd->kref); +- hvcsd->open_count++; ++ local_inc(&hvcsd->open_count); + hvcsd->todo_mask |= HVCS_SCHED_READ; + spin_unlock_irqrestore(&hvcsd->lock, flags); + +@@ -1214,7 +1215,7 @@ static void hvcs_close(struct tty_struct + hvcsd = tty->driver_data; + + spin_lock_irqsave(&hvcsd->lock, flags); +- if (--hvcsd->open_count == 0) { ++ if (local_dec_and_test(&hvcsd->open_count)) { + + vio_disable_interrupts(hvcsd->vdev); + +@@ -1240,10 +1241,10 @@ static void hvcs_close(struct tty_struct + free_irq(irq, hvcsd); + kref_put(&hvcsd->kref, destroy_hvcs_struct); + return; +- } else if (hvcsd->open_count < 0) { ++ } else if (local_read(&hvcsd->open_count) < 0) { + printk(KERN_ERR "HVCS: vty-server@%X open_count: %d" + " is missmanaged.\n", +- hvcsd->vdev->unit_address, hvcsd->open_count); ++ hvcsd->vdev->unit_address, local_read(&hvcsd->open_count)); + } + + spin_unlock_irqrestore(&hvcsd->lock, flags); +@@ -1259,7 +1260,7 @@ static void hvcs_hangup(struct tty_struc + + spin_lock_irqsave(&hvcsd->lock, flags); + /* Preserve this so that we know how many kref refs to put */ +- temp_open_count = hvcsd->open_count; ++ temp_open_count = local_read(&hvcsd->open_count); + + /* + * Don't kref put inside the spinlock because the destruction +@@ -1274,7 +1275,7 @@ static void hvcs_hangup(struct tty_struc + hvcsd->tty->driver_data = NULL; + hvcsd->tty = NULL; + +- hvcsd->open_count = 0; ++ local_set(&hvcsd->open_count, 0); + + /* This will drop any buffered data on the floor which is OK in a hangup + * scenario. */ +@@ -1345,7 +1346,7 @@ static int hvcs_write(struct tty_struct + * the middle of a write operation? This is a crummy place to do this + * but we want to keep it all in the spinlock. + */ +- if (hvcsd->open_count <= 0) { ++ if (local_read(&hvcsd->open_count) <= 0) { + spin_unlock_irqrestore(&hvcsd->lock, flags); + return -ENODEV; + } +@@ -1419,7 +1420,7 @@ static int hvcs_write_room(struct tty_st + { + struct hvcs_struct *hvcsd = tty->driver_data; + +- if (!hvcsd || hvcsd->open_count <= 0) ++ if (!hvcsd || local_read(&hvcsd->open_count) <= 0) + return 0; + + return HVCS_BUFF_LEN - hvcsd->chars_in_buffer; +diff -urNp linux-2.6.38.3/drivers/tty/hvc/hvc_xen.c linux-2.6.38.3/drivers/tty/hvc/hvc_xen.c +--- linux-2.6.38.3/drivers/tty/hvc/hvc_xen.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/tty/hvc/hvc_xen.c 2011-04-17 15:57:32.000000000 -0400 +@@ -123,7 +123,7 @@ static int domU_read_console(uint32_t vt + return recv; + } + +-static struct hv_ops domU_hvc_ops = { ++static const struct hv_ops domU_hvc_ops = { + .get_chars = domU_read_console, + .put_chars = domU_write_console, + .notifier_add = notifier_add_irq, +@@ -149,7 +149,7 @@ static int dom0_write_console(uint32_t v + return len; + } + +-static struct hv_ops dom0_hvc_ops = { ++static const struct hv_ops dom0_hvc_ops = { + .get_chars = dom0_read_console, + .put_chars = dom0_write_console, + .notifier_add = notifier_add_irq, +@@ -160,7 +160,7 @@ static struct hv_ops dom0_hvc_ops = { + static int __init xen_hvc_init(void) + { + struct hvc_struct *hp; +- struct hv_ops *ops; ++ const struct hv_ops *ops; + + if (!xen_pv_domain()) + return -ENODEV; +@@ -203,7 +203,7 @@ static void __exit xen_hvc_fini(void) + + static int xen_cons_init(void) + { +- struct hv_ops *ops; ++ const struct hv_ops *ops; + + if (!xen_pv_domain()) + return 0; +diff -urNp linux-2.6.38.3/drivers/tty/n_gsm.c linux-2.6.38.3/drivers/tty/n_gsm.c +--- linux-2.6.38.3/drivers/tty/n_gsm.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/tty/n_gsm.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1589,7 +1589,7 @@ static struct gsm_dlci *gsm_dlci_alloc(s + return NULL; + spin_lock_init(&dlci->lock); + dlci->fifo = &dlci->_fifo; +- if (kfifo_alloc(&dlci->_fifo, 4096, GFP_KERNEL) < 0) { ++ if (kfifo_alloc(&dlci->_fifo, 4096, GFP_KERNEL)) { + kfree(dlci); + return NULL; + } +diff -urNp linux-2.6.38.3/drivers/tty/n_tty.c linux-2.6.38.3/drivers/tty/n_tty.c +--- linux-2.6.38.3/drivers/tty/n_tty.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/tty/n_tty.c 2011-04-17 15:57:32.000000000 -0400 +@@ -2116,6 +2116,7 @@ void n_tty_inherit_ops(struct tty_ldisc_ + { + *ops = tty_ldisc_N_TTY; + ops->owner = NULL; +- ops->refcount = ops->flags = 0; ++ atomic_set(&ops->refcount, 0); ++ ops->flags = 0; + } + EXPORT_SYMBOL_GPL(n_tty_inherit_ops); +diff -urNp linux-2.6.38.3/drivers/tty/pty.c linux-2.6.38.3/drivers/tty/pty.c +--- linux-2.6.38.3/drivers/tty/pty.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/tty/pty.c 2011-04-17 15:57:32.000000000 -0400 +@@ -700,7 +700,18 @@ out: + return retval; + } + +-static struct file_operations ptmx_fops; ++static const struct file_operations ptmx_fops = { ++ .llseek = no_llseek, ++ .read = tty_read, ++ .write = tty_write, ++ .poll = tty_poll, ++ .unlocked_ioctl = tty_ioctl, ++ .compat_ioctl = tty_compat_ioctl, ++ .open = ptmx_open, ++ .release = tty_release, ++ .fasync = tty_fasync, ++}; ++ + + static void __init unix98_pty_init(void) + { +@@ -753,10 +764,6 @@ static void __init unix98_pty_init(void) + + register_sysctl_table(pty_root_table); + +- /* Now create the /dev/ptmx special device */ +- tty_default_fops(&ptmx_fops); +- ptmx_fops.open = ptmx_open; +- + cdev_init(&ptmx_cdev, &ptmx_fops); + if (cdev_add(&ptmx_cdev, MKDEV(TTYAUX_MAJOR, 2), 1) || + register_chrdev_region(MKDEV(TTYAUX_MAJOR, 2), 1, "/dev/ptmx") < 0) +diff -urNp linux-2.6.38.3/drivers/tty/serial/8250_pci.c linux-2.6.38.3/drivers/tty/serial/8250_pci.c +--- linux-2.6.38.3/drivers/tty/serial/8250_pci.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/tty/serial/8250_pci.c 2011-04-17 15:57:32.000000000 -0400 +@@ -3818,7 +3818,7 @@ static struct pci_device_id serial_pci_t + PCI_ANY_ID, PCI_ANY_ID, + PCI_CLASS_COMMUNICATION_MULTISERIAL << 8, + 0xffff00, pbn_default }, +- { 0, } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + static struct pci_driver serial_pci_driver = { +diff -urNp linux-2.6.38.3/drivers/tty/serial/kgdboc.c linux-2.6.38.3/drivers/tty/serial/kgdboc.c +--- linux-2.6.38.3/drivers/tty/serial/kgdboc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/tty/serial/kgdboc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -22,7 +22,7 @@ + + #define MAX_CONFIG_LEN 40 + +-static struct kgdb_io kgdboc_io_ops; ++static struct kgdb_io kgdboc_io_ops; /* cannot be const, see configure_kgdboc() */ + + /* -1 = init not run yet, 0 = unconfigured, 1 = configured. */ + static int configured = -1; +@@ -293,7 +293,7 @@ static void kgdboc_post_exp_handler(void + kgdboc_restore_input(); + } + +-static struct kgdb_io kgdboc_io_ops = { ++static struct kgdb_io kgdboc_io_ops = { /* cannot be const, see configure_kgdboc() */ + .name = "kgdboc", + .read_char = kgdboc_get_char, + .write_char = kgdboc_put_char, +diff -urNp linux-2.6.38.3/drivers/tty/tty_io.c linux-2.6.38.3/drivers/tty/tty_io.c +--- linux-2.6.38.3/drivers/tty/tty_io.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/tty/tty_io.c 2011-04-17 15:57:32.000000000 -0400 +@@ -140,21 +140,11 @@ EXPORT_SYMBOL(tty_mutex); + /* Spinlock to protect the tty->tty_files list */ + DEFINE_SPINLOCK(tty_files_lock); + +-static ssize_t tty_read(struct file *, char __user *, size_t, loff_t *); +-static ssize_t tty_write(struct file *, const char __user *, size_t, loff_t *); + ssize_t redirected_tty_write(struct file *, const char __user *, + size_t, loff_t *); +-static unsigned int tty_poll(struct file *, poll_table *); + static int tty_open(struct inode *, struct file *); + long tty_ioctl(struct file *file, unsigned int cmd, unsigned long arg); +-#ifdef CONFIG_COMPAT +-static long tty_compat_ioctl(struct file *file, unsigned int cmd, +- unsigned long arg); +-#else +-#define tty_compat_ioctl NULL +-#endif + static int __tty_fasync(int fd, struct file *filp, int on); +-static int tty_fasync(int fd, struct file *filp, int on); + static void release_tty(struct tty_struct *tty, int idx); + static void __proc_set_tty(struct task_struct *tsk, struct tty_struct *tty); + static void proc_set_tty(struct task_struct *tsk, struct tty_struct *tty); +@@ -938,7 +928,7 @@ EXPORT_SYMBOL(start_tty); + * read calls may be outstanding in parallel. + */ + +-static ssize_t tty_read(struct file *file, char __user *buf, size_t count, ++ssize_t tty_read(struct file *file, char __user *buf, size_t count, + loff_t *ppos) + { + int i; +@@ -964,6 +954,8 @@ static ssize_t tty_read(struct file *fil + return i; + } + ++EXPORT_SYMBOL(tty_read); ++ + void tty_write_unlock(struct tty_struct *tty) + { + mutex_unlock(&tty->atomic_write_lock); +@@ -1113,7 +1105,7 @@ void tty_write_message(struct tty_struct + * write method will not be invoked in parallel for each device. + */ + +-static ssize_t tty_write(struct file *file, const char __user *buf, ++ssize_t tty_write(struct file *file, const char __user *buf, + size_t count, loff_t *ppos) + { + struct inode *inode = file->f_path.dentry->d_inode; +@@ -1139,6 +1131,8 @@ static ssize_t tty_write(struct file *fi + return ret; + } + ++EXPORT_SYMBOL(tty_write); ++ + ssize_t redirected_tty_write(struct file *file, const char __user *buf, + size_t count, loff_t *ppos) + { +@@ -1778,6 +1772,8 @@ int tty_release(struct inode *inode, str + return 0; + } + ++EXPORT_SYMBOL(tty_release); ++ + /** + * tty_open - open a tty device + * @inode: inode of device file +@@ -1969,7 +1965,7 @@ got_driver: + * may be re-entered freely by other callers. + */ + +-static unsigned int tty_poll(struct file *filp, poll_table *wait) ++unsigned int tty_poll(struct file *filp, poll_table *wait) + { + struct tty_struct *tty = file_tty(filp); + struct tty_ldisc *ld; +@@ -1985,6 +1981,8 @@ static unsigned int tty_poll(struct file + return ret; + } + ++EXPORT_SYMBOL(tty_poll); ++ + static int __tty_fasync(int fd, struct file *filp, int on) + { + struct tty_struct *tty = file_tty(filp); +@@ -2026,7 +2024,7 @@ out: + return retval; + } + +-static int tty_fasync(int fd, struct file *filp, int on) ++int tty_fasync(int fd, struct file *filp, int on) + { + int retval; + tty_lock(); +@@ -2035,6 +2033,8 @@ static int tty_fasync(int fd, struct fil + return retval; + } + ++EXPORT_SYMBOL(tty_fasync); ++ + /** + * tiocsti - fake input character + * @tty: tty to fake input into +@@ -2692,8 +2692,10 @@ long tty_ioctl(struct file *file, unsign + return retval; + } + ++EXPORT_SYMBOL(tty_ioctl); ++ + #ifdef CONFIG_COMPAT +-static long tty_compat_ioctl(struct file *file, unsigned int cmd, ++long tty_compat_ioctl(struct file *file, unsigned int cmd, + unsigned long arg) + { + struct inode *inode = file->f_dentry->d_inode; +@@ -2717,6 +2719,9 @@ static long tty_compat_ioctl(struct file + + return retval; + } ++ ++EXPORT_SYMBOL(tty_compat_ioctl); ++ + #endif + + /* +@@ -3195,11 +3200,6 @@ struct tty_struct *get_current_tty(void) + } + EXPORT_SYMBOL_GPL(get_current_tty); + +-void tty_default_fops(struct file_operations *fops) +-{ +- *fops = tty_fops; +-} +- + /* + * Initialize the console device. This is called *early*, so + * we can't necessarily depend on lots of kernel help here. +diff -urNp linux-2.6.38.3/drivers/tty/tty_ldisc.c linux-2.6.38.3/drivers/tty/tty_ldisc.c +--- linux-2.6.38.3/drivers/tty/tty_ldisc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/tty/tty_ldisc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -76,7 +76,7 @@ static void put_ldisc(struct tty_ldisc * + if (atomic_dec_and_lock(&ld->users, &tty_ldisc_lock)) { + struct tty_ldisc_ops *ldo = ld->ops; + +- ldo->refcount--; ++ atomic_dec(&ldo->refcount); + module_put(ldo->owner); + spin_unlock_irqrestore(&tty_ldisc_lock, flags); + +@@ -111,7 +111,7 @@ int tty_register_ldisc(int disc, struct + spin_lock_irqsave(&tty_ldisc_lock, flags); + tty_ldiscs[disc] = new_ldisc; + new_ldisc->num = disc; +- new_ldisc->refcount = 0; ++ atomic_set(&new_ldisc->refcount, 0); + spin_unlock_irqrestore(&tty_ldisc_lock, flags); + + return ret; +@@ -139,7 +139,7 @@ int tty_unregister_ldisc(int disc) + return -EINVAL; + + spin_lock_irqsave(&tty_ldisc_lock, flags); +- if (tty_ldiscs[disc]->refcount) ++ if (atomic_read(&tty_ldiscs[disc]->refcount)) + ret = -EBUSY; + else + tty_ldiscs[disc] = NULL; +@@ -160,7 +160,7 @@ static struct tty_ldisc_ops *get_ldops(i + if (ldops) { + ret = ERR_PTR(-EAGAIN); + if (try_module_get(ldops->owner)) { +- ldops->refcount++; ++ atomic_inc(&ldops->refcount); + ret = ldops; + } + } +@@ -173,7 +173,7 @@ static void put_ldops(struct tty_ldisc_o + unsigned long flags; + + spin_lock_irqsave(&tty_ldisc_lock, flags); +- ldops->refcount--; ++ atomic_dec(&ldops->refcount); + module_put(ldops->owner); + spin_unlock_irqrestore(&tty_ldisc_lock, flags); + } +diff -urNp linux-2.6.38.3/drivers/tty/vt/keyboard.c linux-2.6.38.3/drivers/tty/vt/keyboard.c +--- linux-2.6.38.3/drivers/tty/vt/keyboard.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/tty/vt/keyboard.c 2011-04-17 15:57:32.000000000 -0400 +@@ -657,6 +657,16 @@ static void k_spec(struct vc_data *vc, u + kbd->kbdmode == VC_MEDIUMRAW) && + value != KVAL(K_SAK)) + return; /* SAK is allowed even in raw mode */ ++ ++#if defined(CONFIG_GRKERNSEC_PROC) || defined(CONFIG_GRKERNSEC_PROC_MEMMAP) ++ { ++ void *func = fn_handler[value]; ++ if (func == fn_show_state || func == fn_show_ptregs || ++ func == fn_show_mem) ++ return; ++ } ++#endif ++ + fn_handler[value](vc); + } + +@@ -1413,7 +1423,7 @@ static const struct input_device_id kbd_ + .evbit = { BIT_MASK(EV_SND) }, + }, + +- { }, /* Terminating entry */ ++ { 0 }, /* Terminating entry */ + }; + + MODULE_DEVICE_TABLE(input, kbd_ids); +diff -urNp linux-2.6.38.3/drivers/tty/vt/vt.c linux-2.6.38.3/drivers/tty/vt/vt.c +--- linux-2.6.38.3/drivers/tty/vt/vt.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/tty/vt/vt.c 2011-04-17 15:57:32.000000000 -0400 +@@ -262,7 +262,7 @@ EXPORT_SYMBOL_GPL(unregister_vt_notifier + + static void notify_write(struct vc_data *vc, unsigned int unicode) + { +- struct vt_notifier_param param = { .vc = vc, unicode = unicode }; ++ struct vt_notifier_param param = { .vc = vc, .c = unicode }; + atomic_notifier_call_chain(&vt_notifier_list, VT_WRITE, ¶m); + } + +diff -urNp linux-2.6.38.3/drivers/tty/vt/vt_ioctl.c linux-2.6.38.3/drivers/tty/vt/vt_ioctl.c +--- linux-2.6.38.3/drivers/tty/vt/vt_ioctl.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/tty/vt/vt_ioctl.c 2011-04-17 15:57:32.000000000 -0400 +@@ -210,9 +210,6 @@ do_kdsk_ioctl(int cmd, struct kbentry __ + if (copy_from_user(&tmp, user_kbe, sizeof(struct kbentry))) + return -EFAULT; + +- if (!capable(CAP_SYS_TTY_CONFIG)) +- perm = 0; +- + switch (cmd) { + case KDGKBENT: + key_map = key_maps[s]; +@@ -224,6 +221,9 @@ do_kdsk_ioctl(int cmd, struct kbentry __ + val = (i ? K_HOLE : K_NOSUCHMAP); + return put_user(val, &user_kbe->kb_value); + case KDSKBENT: ++ if (!capable(CAP_SYS_TTY_CONFIG)) ++ perm = 0; ++ + if (!perm) + return -EPERM; + if (!i && v == K_NOSUCHMAP) { +@@ -325,9 +325,6 @@ do_kdgkb_ioctl(int cmd, struct kbsentry + int i, j, k; + int ret; + +- if (!capable(CAP_SYS_TTY_CONFIG)) +- perm = 0; +- + kbs = kmalloc(sizeof(*kbs), GFP_KERNEL); + if (!kbs) { + ret = -ENOMEM; +@@ -361,6 +358,9 @@ do_kdgkb_ioctl(int cmd, struct kbsentry + kfree(kbs); + return ((p && *p) ? -EOVERFLOW : 0); + case KDSKBSENT: ++ if (!capable(CAP_SYS_TTY_CONFIG)) ++ perm = 0; ++ + if (!perm) { + ret = -EPERM; + goto reterr; +diff -urNp linux-2.6.38.3/drivers/uio/uio.c linux-2.6.38.3/drivers/uio/uio.c +--- linux-2.6.38.3/drivers/uio/uio.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/uio/uio.c 2011-04-17 15:57:32.000000000 -0400 +@@ -25,6 +25,7 @@ + #include <linux/kobject.h> + #include <linux/cdev.h> + #include <linux/uio_driver.h> ++#include <asm/local.h> + + #define UIO_MAX_DEVICES (1U << MINORBITS) + +@@ -35,7 +36,7 @@ struct uio_device { + atomic_t event; + struct fasync_struct *async_queue; + wait_queue_head_t wait; +- int vma_count; ++ local_t vma_count; + struct uio_info *info; + struct kobject *map_dir; + struct kobject *portio_dir; +@@ -602,13 +603,13 @@ static int uio_find_mem_index(struct vm_ + static void uio_vma_open(struct vm_area_struct *vma) + { + struct uio_device *idev = vma->vm_private_data; +- idev->vma_count++; ++ local_inc(&idev->vma_count); + } + + static void uio_vma_close(struct vm_area_struct *vma) + { + struct uio_device *idev = vma->vm_private_data; +- idev->vma_count--; ++ local_dec(&idev->vma_count); + } + + static int uio_vma_fault(struct vm_area_struct *vma, struct vm_fault *vmf) +diff -urNp linux-2.6.38.3/drivers/usb/atm/cxacru.c linux-2.6.38.3/drivers/usb/atm/cxacru.c +--- linux-2.6.38.3/drivers/usb/atm/cxacru.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/usb/atm/cxacru.c 2011-04-17 15:57:32.000000000 -0400 +@@ -473,7 +473,7 @@ static ssize_t cxacru_sysfs_store_adsl_c + ret = sscanf(buf + pos, "%x=%x%n", &index, &value, &tmp); + if (ret < 2) + return -EINVAL; +- if (index < 0 || index > 0x7f) ++ if (index > 0x7f) + return -EINVAL; + pos += tmp; + +diff -urNp linux-2.6.38.3/drivers/usb/atm/usbatm.c linux-2.6.38.3/drivers/usb/atm/usbatm.c +--- linux-2.6.38.3/drivers/usb/atm/usbatm.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/usb/atm/usbatm.c 2011-04-17 15:57:32.000000000 -0400 +@@ -332,7 +332,7 @@ static void usbatm_extract_one_cell(stru + if (printk_ratelimit()) + atm_warn(instance, "%s: OAM not supported (vpi %d, vci %d)!\n", + __func__, vpi, vci); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + return; + } + +@@ -360,7 +360,7 @@ static void usbatm_extract_one_cell(stru + if (length > ATM_MAX_AAL5_PDU) { + atm_rldbg(instance, "%s: bogus length %u (vcc: 0x%p)!\n", + __func__, length, vcc); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + goto out; + } + +@@ -369,14 +369,14 @@ static void usbatm_extract_one_cell(stru + if (sarb->len < pdu_length) { + atm_rldbg(instance, "%s: bogus pdu_length %u (sarb->len: %u, vcc: 0x%p)!\n", + __func__, pdu_length, sarb->len, vcc); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + goto out; + } + + if (crc32_be(~0, skb_tail_pointer(sarb) - pdu_length, pdu_length) != 0xc704dd7b) { + atm_rldbg(instance, "%s: packet failed crc check (vcc: 0x%p)!\n", + __func__, vcc); +- atomic_inc(&vcc->stats->rx_err); ++ atomic_inc_unchecked(&vcc->stats->rx_err); + goto out; + } + +@@ -386,7 +386,7 @@ static void usbatm_extract_one_cell(stru + if (printk_ratelimit()) + atm_err(instance, "%s: no memory for skb (length: %u)!\n", + __func__, length); +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + goto out; + } + +@@ -411,7 +411,7 @@ static void usbatm_extract_one_cell(stru + + vcc->push(vcc, skb); + +- atomic_inc(&vcc->stats->rx); ++ atomic_inc_unchecked(&vcc->stats->rx); + out: + skb_trim(sarb, 0); + } +@@ -614,7 +614,7 @@ static void usbatm_tx_process(unsigned l + struct atm_vcc *vcc = UDSL_SKB(skb)->atm.vcc; + + usbatm_pop(vcc, skb); +- atomic_inc(&vcc->stats->tx); ++ atomic_inc_unchecked(&vcc->stats->tx); + + skb = skb_dequeue(&instance->sndqueue); + } +@@ -773,11 +773,11 @@ static int usbatm_atm_proc_read(struct a + if (!left--) + return sprintf(page, + "AAL5: tx %d ( %d err ), rx %d ( %d err, %d drop )\n", +- atomic_read(&atm_dev->stats.aal5.tx), +- atomic_read(&atm_dev->stats.aal5.tx_err), +- atomic_read(&atm_dev->stats.aal5.rx), +- atomic_read(&atm_dev->stats.aal5.rx_err), +- atomic_read(&atm_dev->stats.aal5.rx_drop)); ++ atomic_read_unchecked(&atm_dev->stats.aal5.tx), ++ atomic_read_unchecked(&atm_dev->stats.aal5.tx_err), ++ atomic_read_unchecked(&atm_dev->stats.aal5.rx), ++ atomic_read_unchecked(&atm_dev->stats.aal5.rx_err), ++ atomic_read_unchecked(&atm_dev->stats.aal5.rx_drop)); + + if (!left--) { + if (instance->disconnected) +diff -urNp linux-2.6.38.3/drivers/usb/class/cdc-acm.c linux-2.6.38.3/drivers/usb/class/cdc-acm.c +--- linux-2.6.38.3/drivers/usb/class/cdc-acm.c 2011-03-28 17:42:40.000000000 -0400 ++++ linux-2.6.38.3/drivers/usb/class/cdc-acm.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1640,7 +1640,7 @@ static const struct usb_device_id acm_id + { USB_INTERFACE_INFO(USB_CLASS_COMM, USB_CDC_SUBCLASS_ACM, + USB_CDC_ACM_PROTO_AT_CDMA) }, + +- { } ++ { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(usb, acm_ids); +diff -urNp linux-2.6.38.3/drivers/usb/class/usblp.c linux-2.6.38.3/drivers/usb/class/usblp.c +--- linux-2.6.38.3/drivers/usb/class/usblp.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/usb/class/usblp.c 2011-04-17 15:57:32.000000000 -0400 +@@ -227,7 +227,7 @@ static const struct quirk_printer_struct + { 0x0482, 0x0010, USBLP_QUIRK_BIDIR }, /* Kyocera Mita FS 820, by zut <kernel@zut.de> */ + { 0x04f9, 0x000d, USBLP_QUIRK_BIDIR }, /* Brother Industries, Ltd HL-1440 Laser Printer */ + { 0x04b8, 0x0202, USBLP_QUIRK_BAD_CLASS }, /* Seiko Epson Receipt Printer M129C */ +- { 0, 0 } ++ { 0, 0, 0 } + }; + + static int usblp_wwait(struct usblp *usblp, int nonblock); +@@ -1398,7 +1398,7 @@ static const struct usb_device_id usblp_ + { USB_INTERFACE_INFO(7, 1, 2) }, + { USB_INTERFACE_INFO(7, 1, 3) }, + { USB_DEVICE(0x04b8, 0x0202) }, /* Seiko Epson Receipt Printer M129C */ +- { } /* Terminating entry */ ++ { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 } /* Terminating entry */ + }; + + MODULE_DEVICE_TABLE(usb, usblp_ids); +diff -urNp linux-2.6.38.3/drivers/usb/core/hcd.c linux-2.6.38.3/drivers/usb/core/hcd.c +--- linux-2.6.38.3/drivers/usb/core/hcd.c 2011-03-23 17:20:07.000000000 -0400 ++++ linux-2.6.38.3/drivers/usb/core/hcd.c 2011-04-17 15:57:32.000000000 -0400 +@@ -2457,7 +2457,7 @@ EXPORT_SYMBOL_GPL(usb_hcd_platform_shutd + + #if defined(CONFIG_USB_MON) || defined(CONFIG_USB_MON_MODULE) + +-struct usb_mon_operations *mon_ops; ++const struct usb_mon_operations *mon_ops; + + /* + * The registration is unlocked. +@@ -2467,7 +2467,7 @@ struct usb_mon_operations *mon_ops; + * symbols from usbcore, usbcore gets referenced and cannot be unloaded first. + */ + +-int usb_mon_register (struct usb_mon_operations *ops) ++int usb_mon_register (const struct usb_mon_operations *ops) + { + + if (mon_ops) +diff -urNp linux-2.6.38.3/drivers/usb/core/hub.c linux-2.6.38.3/drivers/usb/core/hub.c +--- linux-2.6.38.3/drivers/usb/core/hub.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/usb/core/hub.c 2011-04-17 15:57:32.000000000 -0400 +@@ -3492,7 +3492,7 @@ static const struct usb_device_id hub_id + .bDeviceClass = USB_CLASS_HUB}, + { .match_flags = USB_DEVICE_ID_MATCH_INT_CLASS, + .bInterfaceClass = USB_CLASS_HUB}, +- { } /* Terminating entry */ ++ { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 } /* Terminating entry */ + }; + + MODULE_DEVICE_TABLE (usb, hub_id_table); +diff -urNp linux-2.6.38.3/drivers/usb/core/message.c linux-2.6.38.3/drivers/usb/core/message.c +--- linux-2.6.38.3/drivers/usb/core/message.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/usb/core/message.c 2011-04-17 15:57:32.000000000 -0400 +@@ -869,8 +869,8 @@ char *usb_cache_string(struct usb_device + buf = kmalloc(MAX_USB_STRING_SIZE, GFP_NOIO); + if (buf) { + len = usb_string(udev, index, buf, MAX_USB_STRING_SIZE); +- if (len > 0) { +- smallbuf = kmalloc(++len, GFP_NOIO); ++ if (len++ > 0) { ++ smallbuf = kmalloc(len, GFP_NOIO); + if (!smallbuf) + return buf; + memcpy(smallbuf, buf, len); +diff -urNp linux-2.6.38.3/drivers/usb/early/ehci-dbgp.c linux-2.6.38.3/drivers/usb/early/ehci-dbgp.c +--- linux-2.6.38.3/drivers/usb/early/ehci-dbgp.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/usb/early/ehci-dbgp.c 2011-04-17 15:57:32.000000000 -0400 +@@ -96,7 +96,7 @@ static inline u32 dbgp_len_update(u32 x, + } + + #ifdef CONFIG_KGDB +-static struct kgdb_io kgdbdbgp_io_ops; ++static struct kgdb_io kgdbdbgp_io_ops; /* cannot be const, see kgdbdbgp_parse_config */ + #define dbgp_kgdb_mode (dbg_io_ops == &kgdbdbgp_io_ops) + #else + #define dbgp_kgdb_mode (0) +@@ -1026,7 +1026,7 @@ static void kgdbdbgp_write_char(u8 chr) + early_dbgp_write(NULL, &chr, 1); + } + +-static struct kgdb_io kgdbdbgp_io_ops = { ++static struct kgdb_io kgdbdbgp_io_ops = { /* cannot be const, see kgdbdbgp_parse_config() */ + .name = "kgdbdbgp", + .read_char = kgdbdbgp_read_char, + .write_char = kgdbdbgp_write_char, +diff -urNp linux-2.6.38.3/drivers/usb/host/ehci-pci.c linux-2.6.38.3/drivers/usb/host/ehci-pci.c +--- linux-2.6.38.3/drivers/usb/host/ehci-pci.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/usb/host/ehci-pci.c 2011-04-17 15:57:32.000000000 -0400 +@@ -516,7 +516,7 @@ static const struct pci_device_id pci_id + PCI_DEVICE_CLASS(PCI_CLASS_SERIAL_USB_EHCI, ~0), + .driver_data = (unsigned long) &ehci_pci_hc_driver, + }, +- { /* end: all zeroes */ } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + MODULE_DEVICE_TABLE(pci, pci_ids); + +diff -urNp linux-2.6.38.3/drivers/usb/host/uhci-hcd.c linux-2.6.38.3/drivers/usb/host/uhci-hcd.c +--- linux-2.6.38.3/drivers/usb/host/uhci-hcd.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/usb/host/uhci-hcd.c 2011-04-17 15:57:32.000000000 -0400 +@@ -948,7 +948,7 @@ static const struct pci_device_id uhci_p + /* handle any USB UHCI controller */ + PCI_DEVICE_CLASS(PCI_CLASS_SERIAL_USB_UHCI, ~0), + .driver_data = (unsigned long) &uhci_driver, +- }, { /* end: all zeroes */ } ++ }, { 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(pci, uhci_pci_ids); +diff -urNp linux-2.6.38.3/drivers/usb/mon/mon_main.c linux-2.6.38.3/drivers/usb/mon/mon_main.c +--- linux-2.6.38.3/drivers/usb/mon/mon_main.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/usb/mon/mon_main.c 2011-04-17 15:57:32.000000000 -0400 +@@ -238,7 +238,7 @@ static struct notifier_block mon_nb = { + /* + * Ops + */ +-static struct usb_mon_operations mon_ops_0 = { ++static const struct usb_mon_operations mon_ops_0 = { + .urb_submit = mon_submit, + .urb_submit_error = mon_submit_error, + .urb_complete = mon_complete, +diff -urNp linux-2.6.38.3/drivers/usb/storage/debug.h linux-2.6.38.3/drivers/usb/storage/debug.h +--- linux-2.6.38.3/drivers/usb/storage/debug.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/usb/storage/debug.h 2011-04-17 15:57:32.000000000 -0400 +@@ -54,9 +54,9 @@ void usb_stor_show_sense( unsigned char + #define US_DEBUGPX(x...) printk( x ) + #define US_DEBUG(x) x + #else +-#define US_DEBUGP(x...) +-#define US_DEBUGPX(x...) +-#define US_DEBUG(x) ++#define US_DEBUGP(x...) do {} while (0) ++#define US_DEBUGPX(x...) do {} while (0) ++#define US_DEBUG(x) do {} while (0) + #endif + + #endif +diff -urNp linux-2.6.38.3/drivers/usb/storage/usb.c linux-2.6.38.3/drivers/usb/storage/usb.c +--- linux-2.6.38.3/drivers/usb/storage/usb.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/usb/storage/usb.c 2011-04-17 15:57:32.000000000 -0400 +@@ -122,7 +122,7 @@ MODULE_PARM_DESC(quirks, "supplemental l + + static struct us_unusual_dev us_unusual_dev_list[] = { + # include "unusual_devs.h" +- { } /* Terminating entry */ ++ { NULL, NULL, 0, 0, NULL } /* Terminating entry */ + }; + + #undef UNUSUAL_DEV +diff -urNp linux-2.6.38.3/drivers/usb/storage/usual-tables.c linux-2.6.38.3/drivers/usb/storage/usual-tables.c +--- linux-2.6.38.3/drivers/usb/storage/usual-tables.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/usb/storage/usual-tables.c 2011-04-17 15:57:32.000000000 -0400 +@@ -48,7 +48,7 @@ + + struct usb_device_id usb_storage_usb_ids[] = { + # include "unusual_devs.h" +- { } /* Terminating entry */ ++ { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 } /* Terminating entry */ + }; + EXPORT_SYMBOL_GPL(usb_storage_usb_ids); + +diff -urNp linux-2.6.38.3/drivers/vhost/vhost.c linux-2.6.38.3/drivers/vhost/vhost.c +--- linux-2.6.38.3/drivers/vhost/vhost.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/vhost/vhost.c 2011-04-17 15:57:32.000000000 -0400 +@@ -565,7 +565,7 @@ static int init_used(struct vhost_virtqu + return get_user(vq->last_used_idx, &used->idx); + } + +-static long vhost_set_vring(struct vhost_dev *d, int ioctl, void __user *argp) ++static long vhost_set_vring(struct vhost_dev *d, unsigned int ioctl, void __user *argp) + { + struct file *eventfp, *filep = NULL, + *pollstart = NULL, *pollstop = NULL; +diff -urNp linux-2.6.38.3/drivers/video/fbcmap.c linux-2.6.38.3/drivers/video/fbcmap.c +--- linux-2.6.38.3/drivers/video/fbcmap.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/video/fbcmap.c 2011-04-17 15:57:32.000000000 -0400 +@@ -285,8 +285,7 @@ int fb_set_user_cmap(struct fb_cmap_user + rc = -ENODEV; + goto out; + } +- if (cmap->start < 0 || (!info->fbops->fb_setcolreg && +- !info->fbops->fb_setcmap)) { ++ if (!info->fbops->fb_setcolreg && !info->fbops->fb_setcmap) { + rc = -EINVAL; + goto out1; + } +diff -urNp linux-2.6.38.3/drivers/video/fbmem.c linux-2.6.38.3/drivers/video/fbmem.c +--- linux-2.6.38.3/drivers/video/fbmem.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/video/fbmem.c 2011-04-17 15:57:32.000000000 -0400 +@@ -403,7 +403,7 @@ static void fb_do_show_logo(struct fb_in + image->dx += image->width + 8; + } + } else if (rotate == FB_ROTATE_UD) { +- for (x = 0; x < num && image->dx >= 0; x++) { ++ for (x = 0; x < num && (__s32)image->dx >= 0; x++) { + info->fbops->fb_imageblit(info, image); + image->dx -= image->width + 8; + } +@@ -415,7 +415,7 @@ static void fb_do_show_logo(struct fb_in + image->dy += image->height + 8; + } + } else if (rotate == FB_ROTATE_CCW) { +- for (x = 0; x < num && image->dy >= 0; x++) { ++ for (x = 0; x < num && (__s32)image->dy >= 0; x++) { + info->fbops->fb_imageblit(info, image); + image->dy -= image->height + 8; + } +@@ -1101,7 +1101,7 @@ static long do_fb_ioctl(struct fb_info * + return -EFAULT; + if (con2fb.console < 1 || con2fb.console > MAX_NR_CONSOLES) + return -EINVAL; +- if (con2fb.framebuffer < 0 || con2fb.framebuffer >= FB_MAX) ++ if (con2fb.framebuffer >= FB_MAX) + return -EINVAL; + if (!registered_fb[con2fb.framebuffer]) + request_module("fb%d", con2fb.framebuffer); +diff -urNp linux-2.6.38.3/drivers/video/fbmon.c linux-2.6.38.3/drivers/video/fbmon.c +--- linux-2.6.38.3/drivers/video/fbmon.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/video/fbmon.c 2011-04-17 15:57:32.000000000 -0400 +@@ -46,7 +46,7 @@ + #ifdef DEBUG + #define DPRINTK(fmt, args...) printk(fmt,## args) + #else +-#define DPRINTK(fmt, args...) ++#define DPRINTK(fmt, args...) do {} while (0) + #endif + + #define FBMON_FIX_HEADER 1 +diff -urNp linux-2.6.38.3/drivers/video/i810/i810_accel.c linux-2.6.38.3/drivers/video/i810/i810_accel.c +--- linux-2.6.38.3/drivers/video/i810/i810_accel.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/video/i810/i810_accel.c 2011-04-17 15:57:32.000000000 -0400 +@@ -73,6 +73,7 @@ static inline int wait_for_space(struct + } + } + printk("ringbuffer lockup!!!\n"); ++ printk("head:%u tail:%u iring.size:%u space:%u\n", head, tail, par->iring.size, space); + i810_report_error(mmio); + par->dev_flags |= LOCKUP; + info->pixmap.scan_align = 1; +diff -urNp linux-2.6.38.3/drivers/video/i810/i810_main.c linux-2.6.38.3/drivers/video/i810/i810_main.c +--- linux-2.6.38.3/drivers/video/i810/i810_main.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/video/i810/i810_main.c 2011-04-17 15:57:32.000000000 -0400 +@@ -120,7 +120,7 @@ static struct pci_device_id i810fb_pci_t + PCI_ANY_ID, PCI_ANY_ID, 0, 0, 4 }, + { PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82815_CGC, + PCI_ANY_ID, PCI_ANY_ID, 0, 0, 5 }, +- { 0 }, ++ { 0, 0, 0, 0, 0, 0, 0 }, + }; + + static struct pci_driver i810fb_driver = { +diff -urNp linux-2.6.38.3/drivers/video/modedb.c linux-2.6.38.3/drivers/video/modedb.c +--- linux-2.6.38.3/drivers/video/modedb.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/video/modedb.c 2011-04-17 15:57:32.000000000 -0400 +@@ -40,255 +40,255 @@ static const struct fb_videomode modedb[ + + /* 640x400 @ 70 Hz, 31.5 kHz hsync */ + { NULL, 70, 640, 400, 39721, 40, 24, 39, 9, 96, 2, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 640x480 @ 60 Hz, 31.5 kHz hsync */ + { NULL, 60, 640, 480, 39721, 40, 24, 32, 11, 96, 2, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 800x600 @ 56 Hz, 35.15 kHz hsync */ + { NULL, 56, 800, 600, 27777, 128, 24, 22, 1, 72, 2, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1024x768 @ 87 Hz interlaced, 35.5 kHz hsync */ + { NULL, 87, 1024, 768, 22271, 56, 24, 33, 8, 160, 8, 0, +- FB_VMODE_INTERLACED }, ++ FB_VMODE_INTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 640x400 @ 85 Hz, 37.86 kHz hsync */ + { NULL, 85, 640, 400, 31746, 96, 32, 41, 1, 64, 3, +- FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED }, ++ FB_SYNC_VERT_HIGH_ACT, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 640x480 @ 72 Hz, 36.5 kHz hsync */ + { NULL, 72, 640, 480, 31746, 144, 40, 30, 8, 40, 3, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 640x480 @ 75 Hz, 37.50 kHz hsync */ + { NULL, 75, 640, 480, 31746, 120, 16, 16, 1, 64, 3, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 800x600 @ 60 Hz, 37.8 kHz hsync */ + { NULL, 60, 800, 600, 25000, 88, 40, 23, 1, 128, 4, + FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 640x480 @ 85 Hz, 43.27 kHz hsync */ + { NULL, 85, 640, 480, 27777, 80, 56, 25, 1, 56, 3, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1152x864 @ 89 Hz interlaced, 44 kHz hsync */ + { NULL, 89, 1152, 864, 15384, 96, 16, 110, 1, 216, 10, 0, +- FB_VMODE_INTERLACED }, ++ FB_VMODE_INTERLACED, FB_MODE_IS_UNKNOWN }, + /* 800x600 @ 72 Hz, 48.0 kHz hsync */ + { NULL, 72, 800, 600, 20000, 64, 56, 23, 37, 120, 6, + FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1024x768 @ 60 Hz, 48.4 kHz hsync */ + { NULL, 60, 1024, 768, 15384, 168, 8, 29, 3, 144, 6, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 640x480 @ 100 Hz, 53.01 kHz hsync */ + { NULL, 100, 640, 480, 21834, 96, 32, 36, 8, 96, 6, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1152x864 @ 60 Hz, 53.5 kHz hsync */ + { NULL, 60, 1152, 864, 11123, 208, 64, 16, 4, 256, 8, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 800x600 @ 85 Hz, 55.84 kHz hsync */ + { NULL, 85, 800, 600, 16460, 160, 64, 36, 16, 64, 5, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1024x768 @ 70 Hz, 56.5 kHz hsync */ + { NULL, 70, 1024, 768, 13333, 144, 24, 29, 3, 136, 6, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1280x1024 @ 87 Hz interlaced, 51 kHz hsync */ + { NULL, 87, 1280, 1024, 12500, 56, 16, 128, 1, 216, 12, 0, +- FB_VMODE_INTERLACED }, ++ FB_VMODE_INTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 800x600 @ 100 Hz, 64.02 kHz hsync */ + { NULL, 100, 800, 600, 14357, 160, 64, 30, 4, 64, 6, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1024x768 @ 76 Hz, 62.5 kHz hsync */ + { NULL, 76, 1024, 768, 11764, 208, 8, 36, 16, 120, 3, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1152x864 @ 70 Hz, 62.4 kHz hsync */ + { NULL, 70, 1152, 864, 10869, 106, 56, 20, 1, 160, 10, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1280x1024 @ 61 Hz, 64.2 kHz hsync */ + { NULL, 61, 1280, 1024, 9090, 200, 48, 26, 1, 184, 3, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1400x1050 @ 60Hz, 63.9 kHz hsync */ + { NULL, 60, 1400, 1050, 9259, 136, 40, 13, 1, 112, 3, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1400x1050 @ 75,107 Hz, 82,392 kHz +hsync +vsync*/ + { NULL, 75, 1400, 1050, 7190, 120, 56, 23, 10, 112, 13, + FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1400x1050 @ 60 Hz, ? kHz +hsync +vsync*/ + { NULL, 60, 1400, 1050, 9259, 128, 40, 12, 0, 112, 3, + FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1024x768 @ 85 Hz, 70.24 kHz hsync */ + { NULL, 85, 1024, 768, 10111, 192, 32, 34, 14, 160, 6, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1152x864 @ 78 Hz, 70.8 kHz hsync */ + { NULL, 78, 1152, 864, 9090, 228, 88, 32, 0, 84, 12, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1280x1024 @ 70 Hz, 74.59 kHz hsync */ + { NULL, 70, 1280, 1024, 7905, 224, 32, 28, 8, 160, 8, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1600x1200 @ 60Hz, 75.00 kHz hsync */ + { NULL, 60, 1600, 1200, 6172, 304, 64, 46, 1, 192, 3, + FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1152x864 @ 84 Hz, 76.0 kHz hsync */ + { NULL, 84, 1152, 864, 7407, 184, 312, 32, 0, 128, 12, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1280x1024 @ 74 Hz, 78.85 kHz hsync */ + { NULL, 74, 1280, 1024, 7407, 256, 32, 34, 3, 144, 3, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1024x768 @ 100Hz, 80.21 kHz hsync */ + { NULL, 100, 1024, 768, 8658, 192, 32, 21, 3, 192, 10, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1280x1024 @ 76 Hz, 81.13 kHz hsync */ + { NULL, 76, 1280, 1024, 7407, 248, 32, 34, 3, 104, 3, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1600x1200 @ 70 Hz, 87.50 kHz hsync */ + { NULL, 70, 1600, 1200, 5291, 304, 64, 46, 1, 192, 3, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1152x864 @ 100 Hz, 89.62 kHz hsync */ + { NULL, 100, 1152, 864, 7264, 224, 32, 17, 2, 128, 19, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1280x1024 @ 85 Hz, 91.15 kHz hsync */ + { NULL, 85, 1280, 1024, 6349, 224, 64, 44, 1, 160, 3, + FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1600x1200 @ 75 Hz, 93.75 kHz hsync */ + { NULL, 75, 1600, 1200, 4938, 304, 64, 46, 1, 192, 3, + FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1680x1050 @ 60 Hz, 65.191 kHz hsync */ + { NULL, 60, 1680, 1050, 6848, 280, 104, 30, 3, 176, 6, + FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1600x1200 @ 85 Hz, 105.77 kHz hsync */ + { NULL, 85, 1600, 1200, 4545, 272, 16, 37, 4, 192, 3, + FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1280x1024 @ 100 Hz, 107.16 kHz hsync */ + { NULL, 100, 1280, 1024, 5502, 256, 32, 26, 7, 128, 15, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1800x1440 @ 64Hz, 96.15 kHz hsync */ + { NULL, 64, 1800, 1440, 4347, 304, 96, 46, 1, 192, 3, + FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1800x1440 @ 70Hz, 104.52 kHz hsync */ + { NULL, 70, 1800, 1440, 4000, 304, 96, 46, 1, 192, 3, + FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 512x384 @ 78 Hz, 31.50 kHz hsync */ + { NULL, 78, 512, 384, 49603, 48, 16, 16, 1, 64, 3, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 512x384 @ 85 Hz, 34.38 kHz hsync */ + { NULL, 85, 512, 384, 45454, 48, 16, 16, 1, 64, 3, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 320x200 @ 70 Hz, 31.5 kHz hsync, 8:5 aspect ratio */ + { NULL, 70, 320, 200, 79440, 16, 16, 20, 4, 48, 1, 0, +- FB_VMODE_DOUBLE }, ++ FB_VMODE_DOUBLE, FB_MODE_IS_UNKNOWN }, + + /* 320x240 @ 60 Hz, 31.5 kHz hsync, 4:3 aspect ratio */ + { NULL, 60, 320, 240, 79440, 16, 16, 16, 5, 48, 1, 0, +- FB_VMODE_DOUBLE }, ++ FB_VMODE_DOUBLE, FB_MODE_IS_UNKNOWN }, + + /* 320x240 @ 72 Hz, 36.5 kHz hsync */ + { NULL, 72, 320, 240, 63492, 16, 16, 16, 4, 48, 2, 0, +- FB_VMODE_DOUBLE }, ++ FB_VMODE_DOUBLE, FB_MODE_IS_UNKNOWN }, + + /* 400x300 @ 56 Hz, 35.2 kHz hsync, 4:3 aspect ratio */ + { NULL, 56, 400, 300, 55555, 64, 16, 10, 1, 32, 1, 0, +- FB_VMODE_DOUBLE }, ++ FB_VMODE_DOUBLE, FB_MODE_IS_UNKNOWN }, + + /* 400x300 @ 60 Hz, 37.8 kHz hsync */ + { NULL, 60, 400, 300, 50000, 48, 16, 11, 1, 64, 2, 0, +- FB_VMODE_DOUBLE }, ++ FB_VMODE_DOUBLE, FB_MODE_IS_UNKNOWN }, + + /* 400x300 @ 72 Hz, 48.0 kHz hsync */ + { NULL, 72, 400, 300, 40000, 32, 24, 11, 19, 64, 3, 0, +- FB_VMODE_DOUBLE }, ++ FB_VMODE_DOUBLE, FB_MODE_IS_UNKNOWN }, + + /* 480x300 @ 56 Hz, 35.2 kHz hsync, 8:5 aspect ratio */ + { NULL, 56, 480, 300, 46176, 80, 16, 10, 1, 40, 1, 0, +- FB_VMODE_DOUBLE }, ++ FB_VMODE_DOUBLE, FB_MODE_IS_UNKNOWN }, + + /* 480x300 @ 60 Hz, 37.8 kHz hsync */ + { NULL, 60, 480, 300, 41858, 56, 16, 11, 1, 80, 2, 0, +- FB_VMODE_DOUBLE }, ++ FB_VMODE_DOUBLE, FB_MODE_IS_UNKNOWN }, + + /* 480x300 @ 63 Hz, 39.6 kHz hsync */ + { NULL, 63, 480, 300, 40000, 56, 16, 11, 1, 80, 2, 0, +- FB_VMODE_DOUBLE }, ++ FB_VMODE_DOUBLE, FB_MODE_IS_UNKNOWN }, + + /* 480x300 @ 72 Hz, 48.0 kHz hsync */ + { NULL, 72, 480, 300, 33386, 40, 24, 11, 19, 80, 3, 0, +- FB_VMODE_DOUBLE }, ++ FB_VMODE_DOUBLE, FB_MODE_IS_UNKNOWN }, + + /* 1920x1200 @ 60 Hz, 74.5 Khz hsync */ + { NULL, 60, 1920, 1200, 5177, 128, 336, 1, 38, 208, 3, + FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1152x768, 60 Hz, PowerBook G4 Titanium I and II */ + { NULL, 60, 1152, 768, 14047, 158, 26, 29, 3, 136, 6, + FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1366x768, 60 Hz, 47.403 kHz hsync, WXGA 16:9 aspect ratio */ + { NULL, 60, 1366, 768, 13806, 120, 10, 14, 3, 32, 5, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 1280x800, 60 Hz, 47.403 kHz hsync, WXGA 16:10 aspect ratio */ + { NULL, 60, 1280, 800, 12048, 200, 64, 24, 1, 136, 3, 0, +- FB_VMODE_NONINTERLACED }, ++ FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 720x576i @ 50 Hz, 15.625 kHz hsync (PAL RGB) */ + { NULL, 50, 720, 576, 74074, 64, 16, 39, 5, 64, 5, 0, +- FB_VMODE_INTERLACED }, ++ FB_VMODE_INTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 800x520i @ 50 Hz, 15.625 kHz hsync (PAL RGB) */ + { NULL, 50, 800, 520, 58823, 144, 64, 72, 28, 80, 5, 0, +- FB_VMODE_INTERLACED }, ++ FB_VMODE_INTERLACED, FB_MODE_IS_UNKNOWN }, + + /* 864x480 @ 60 Hz, 35.15 kHz hsync */ + { NULL, 60, 864, 480, 27777, 1, 1, 1, 1, 0, 0, +- 0, FB_VMODE_NONINTERLACED }, ++ 0, FB_VMODE_NONINTERLACED, FB_MODE_IS_UNKNOWN }, + }; + + #ifdef CONFIG_FB_MODE_HELPERS +diff -urNp linux-2.6.38.3/drivers/video/pxa3xx-gcu.c linux-2.6.38.3/drivers/video/pxa3xx-gcu.c +--- linux-2.6.38.3/drivers/video/pxa3xx-gcu.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/video/pxa3xx-gcu.c 2011-04-17 15:57:32.000000000 -0400 +@@ -103,7 +103,7 @@ struct pxa3xx_gcu_priv { + dma_addr_t shared_phys; + struct resource *resource_mem; + struct miscdevice misc_dev; +- struct file_operations misc_fops; ++ const struct file_operations misc_fops; + wait_queue_head_t wait_idle; + wait_queue_head_t wait_free; + spinlock_t spinlock; +diff -urNp linux-2.6.38.3/drivers/video/uvesafb.c linux-2.6.38.3/drivers/video/uvesafb.c +--- linux-2.6.38.3/drivers/video/uvesafb.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/video/uvesafb.c 2011-04-17 15:57:32.000000000 -0400 +@@ -19,6 +19,7 @@ + #include <linux/io.h> + #include <linux/mutex.h> + #include <linux/slab.h> ++#include <linux/moduleloader.h> + #include <video/edid.h> + #include <video/uvesafb.h> + #ifdef CONFIG_X86 +@@ -121,7 +122,7 @@ static int uvesafb_helper_start(void) + NULL, + }; + +- return call_usermodehelper(v86d_path, argv, envp, 1); ++ return call_usermodehelper(v86d_path, argv, envp, UMH_WAIT_PROC); + } + + /* +@@ -569,10 +570,32 @@ static int __devinit uvesafb_vbe_getpmi( + if ((task->t.regs.eax & 0xffff) != 0x4f || task->t.regs.es < 0xc000) { + par->pmi_setpal = par->ypan = 0; + } else { ++ ++#ifdef CONFIG_PAX_KERNEXEC ++#ifdef CONFIG_MODULES ++ par->pmi_code = module_alloc_exec((u16)task->t.regs.ecx); ++#endif ++ if (!par->pmi_code) { ++ par->pmi_setpal = par->ypan = 0; ++ return 0; ++ } ++#endif ++ + par->pmi_base = (u16 *)phys_to_virt(((u32)task->t.regs.es << 4) + + task->t.regs.edi); ++ ++#if defined(CONFIG_MODULES) && defined(CONFIG_PAX_KERNEXEC) ++ pax_open_kernel(); ++ memcpy(par->pmi_code, par->pmi_base, (u16)task->t.regs.ecx); ++ pax_close_kernel(); ++ ++ par->pmi_start = ktva_ktla(par->pmi_code + par->pmi_base[1]); ++ par->pmi_pal = ktva_ktla(par->pmi_code + par->pmi_base[2]); ++#else + par->pmi_start = (u8 *)par->pmi_base + par->pmi_base[1]; + par->pmi_pal = (u8 *)par->pmi_base + par->pmi_base[2]; ++#endif ++ + printk(KERN_INFO "uvesafb: protected mode interface info at " + "%04x:%04x\n", + (u16)task->t.regs.es, (u16)task->t.regs.edi); +@@ -1800,6 +1823,11 @@ out: + if (par->vbe_modes) + kfree(par->vbe_modes); + ++#if defined(CONFIG_MODULES) && defined(CONFIG_PAX_KERNEXEC) ++ if (par->pmi_code) ++ module_free_exec(NULL, par->pmi_code); ++#endif ++ + framebuffer_release(info); + return err; + } +@@ -1826,6 +1854,12 @@ static int uvesafb_remove(struct platfor + kfree(par->vbe_state_orig); + if (par->vbe_state_saved) + kfree(par->vbe_state_saved); ++ ++#if defined(CONFIG_MODULES) && defined(CONFIG_PAX_KERNEXEC) ++ if (par->pmi_code) ++ module_free_exec(NULL, par->pmi_code); ++#endif ++ + } + + framebuffer_release(info); +diff -urNp linux-2.6.38.3/drivers/video/vesafb.c linux-2.6.38.3/drivers/video/vesafb.c +--- linux-2.6.38.3/drivers/video/vesafb.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/drivers/video/vesafb.c 2011-04-17 15:57:32.000000000 -0400 +@@ -9,6 +9,7 @@ + */ + + #include <linux/module.h> ++#include <linux/moduleloader.h> + #include <linux/kernel.h> + #include <linux/errno.h> + #include <linux/string.h> +@@ -52,8 +53,8 @@ static int vram_remap __initdata; /* + static int vram_total __initdata; /* Set total amount of memory */ + static int pmi_setpal __read_mostly = 1; /* pmi for palette changes ??? */ + static int ypan __read_mostly; /* 0..nothing, 1..ypan, 2..ywrap */ +-static void (*pmi_start)(void) __read_mostly; +-static void (*pmi_pal) (void) __read_mostly; ++static void (*pmi_start)(void) __read_only; ++static void (*pmi_pal) (void) __read_only; + static int depth __read_mostly; + static int vga_compat __read_mostly; + /* --------------------------------------------------------------------- */ +@@ -232,6 +233,7 @@ static int __init vesafb_probe(struct pl + unsigned int size_vmode; + unsigned int size_remap; + unsigned int size_total; ++ void *pmi_code = NULL; + + if (screen_info.orig_video_isVGA != VIDEO_TYPE_VLFB) + return -ENODEV; +@@ -274,10 +276,6 @@ static int __init vesafb_probe(struct pl + size_remap = size_total; + vesafb_fix.smem_len = size_remap; + +-#ifndef __i386__ +- screen_info.vesapm_seg = 0; +-#endif +- + if (!request_mem_region(vesafb_fix.smem_start, size_total, "vesafb")) { + printk(KERN_WARNING + "vesafb: cannot reserve video memory at 0x%lx\n", +@@ -319,9 +317,21 @@ static int __init vesafb_probe(struct pl + printk(KERN_INFO "vesafb: mode is %dx%dx%d, linelength=%d, pages=%d\n", + vesafb_defined.xres, vesafb_defined.yres, vesafb_defined.bits_per_pixel, vesafb_fix.line_length, screen_info.pages); + ++#ifdef __i386__ ++ ++#if defined(CONFIG_MODULES) && defined(CONFIG_PAX_KERNEXEC) ++ pmi_code = module_alloc_exec(screen_info.vesapm_size); ++ if (!pmi_code) ++#elif !defined(CONFIG_PAX_KERNEXEC) ++ if (0) ++#endif ++ ++#endif ++ screen_info.vesapm_seg = 0; ++ + if (screen_info.vesapm_seg) { +- printk(KERN_INFO "vesafb: protected mode interface info at %04x:%04x\n", +- screen_info.vesapm_seg,screen_info.vesapm_off); ++ printk(KERN_INFO "vesafb: protected mode interface info at %04x:%04x %04x bytes\n", ++ screen_info.vesapm_seg,screen_info.vesapm_off,screen_info.vesapm_size); + } + + if (screen_info.vesapm_seg < 0xc000) +@@ -329,9 +339,25 @@ static int __init vesafb_probe(struct pl + + if (ypan || pmi_setpal) { + unsigned short *pmi_base; +- pmi_base = (unsigned short*)phys_to_virt(((unsigned long)screen_info.vesapm_seg << 4) + screen_info.vesapm_off); +- pmi_start = (void*)((char*)pmi_base + pmi_base[1]); +- pmi_pal = (void*)((char*)pmi_base + pmi_base[2]); ++ ++ pmi_base = (unsigned short*)phys_to_virt(((unsigned long)screen_info.vesapm_seg << 4) + screen_info.vesapm_off); ++ ++#if defined(CONFIG_MODULES) && defined(CONFIG_PAX_KERNEXEC) ++ pax_open_kernel(); ++ memcpy(pmi_code, pmi_base, screen_info.vesapm_size); ++#else ++ pmi_code = pmi_base; ++#endif ++ ++ pmi_start = (void*)((char*)pmi_code + pmi_base[1]); ++ pmi_pal = (void*)((char*)pmi_code + pmi_base[2]); ++ ++#if defined(CONFIG_MODULES) && defined(CONFIG_PAX_KERNEXEC) ++ pmi_start = ktva_ktla(pmi_start); ++ pmi_pal = ktva_ktla(pmi_pal); ++ pax_close_kernel(); ++#endif ++ + printk(KERN_INFO "vesafb: pmi: set display start = %p, set palette = %p\n",pmi_start,pmi_pal); + if (pmi_base[3]) { + printk(KERN_INFO "vesafb: pmi: ports = "); +@@ -473,6 +499,11 @@ static int __init vesafb_probe(struct pl + info->node, info->fix.id); + return 0; + err: ++ ++#if defined(__i386__) && defined(CONFIG_MODULES) && defined(CONFIG_PAX_KERNEXEC) ++ module_free_exec(NULL, pmi_code); ++#endif ++ + if (info->screen_base) + iounmap(info->screen_base); + framebuffer_release(info); +diff -urNp linux-2.6.38.3/fs/9p/vfs_inode.c linux-2.6.38.3/fs/9p/vfs_inode.c +--- linux-2.6.38.3/fs/9p/vfs_inode.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/9p/vfs_inode.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1094,7 +1094,7 @@ static void *v9fs_vfs_follow_link(struct + void + v9fs_vfs_put_link(struct dentry *dentry, struct nameidata *nd, void *p) + { +- char *s = nd_get_link(nd); ++ const char *s = nd_get_link(nd); + + P9_DPRINTK(P9_DEBUG_VFS, " %s %s\n", dentry->d_name.name, + IS_ERR(s) ? "<error>" : s); +diff -urNp linux-2.6.38.3/fs/aio.c linux-2.6.38.3/fs/aio.c +--- linux-2.6.38.3/fs/aio.c 2011-03-28 17:42:40.000000000 -0400 ++++ linux-2.6.38.3/fs/aio.c 2011-04-17 15:57:32.000000000 -0400 +@@ -130,7 +130,7 @@ static int aio_setup_ring(struct kioctx + size += sizeof(struct io_event) * nr_events; + nr_pages = (size + PAGE_SIZE-1) >> PAGE_SHIFT; + +- if (nr_pages < 0) ++ if (nr_pages <= 0) + return -EINVAL; + + nr_events = (PAGE_SIZE * nr_pages - sizeof(struct aio_ring)) / sizeof(struct io_event); +diff -urNp linux-2.6.38.3/fs/attr.c linux-2.6.38.3/fs/attr.c +--- linux-2.6.38.3/fs/attr.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/attr.c 2011-04-17 15:57:32.000000000 -0400 +@@ -98,6 +98,7 @@ int inode_newsize_ok(const struct inode + unsigned long limit; + + limit = rlimit(RLIMIT_FSIZE); ++ gr_learn_resource(current, RLIMIT_FSIZE, (unsigned long)offset, 1); + if (limit != RLIM_INFINITY && offset > limit) + goto out_sig; + if (offset > inode->i_sb->s_maxbytes) +diff -urNp linux-2.6.38.3/fs/befs/linuxvfs.c linux-2.6.38.3/fs/befs/linuxvfs.c +--- linux-2.6.38.3/fs/befs/linuxvfs.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/befs/linuxvfs.c 2011-04-17 15:57:32.000000000 -0400 +@@ -499,7 +499,7 @@ static void befs_put_link(struct dentry + { + befs_inode_info *befs_ino = BEFS_I(dentry->d_inode); + if (befs_ino->i_flags & BEFS_LONG_SYMLINK) { +- char *link = nd_get_link(nd); ++ const char *link = nd_get_link(nd); + if (!IS_ERR(link)) + kfree(link); + } +diff -urNp linux-2.6.38.3/fs/binfmt_aout.c linux-2.6.38.3/fs/binfmt_aout.c +--- linux-2.6.38.3/fs/binfmt_aout.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/binfmt_aout.c 2011-04-17 15:57:32.000000000 -0400 +@@ -16,6 +16,7 @@ + #include <linux/string.h> + #include <linux/fs.h> + #include <linux/file.h> ++#include <linux/security.h> + #include <linux/stat.h> + #include <linux/fcntl.h> + #include <linux/ptrace.h> +@@ -86,6 +87,8 @@ static int aout_core_dump(struct coredum + #endif + # define START_STACK(u) ((void __user *)u.start_stack) + ++ memset(&dump, 0, sizeof(dump)); ++ + fs = get_fs(); + set_fs(KERNEL_DS); + has_dumped = 1; +@@ -97,10 +100,12 @@ static int aout_core_dump(struct coredum + + /* If the size of the dump file exceeds the rlimit, then see what would happen + if we wrote the stack, but not the data area. */ ++ gr_learn_resource(current, RLIMIT_CORE, (dump.u_dsize + dump.u_ssize+1) * PAGE_SIZE, 1); + if ((dump.u_dsize + dump.u_ssize+1) * PAGE_SIZE > cprm->limit) + dump.u_dsize = 0; + + /* Make sure we have enough room to write the stack and data areas. */ ++ gr_learn_resource(current, RLIMIT_CORE, (dump.u_ssize + 1) * PAGE_SIZE, 1); + if ((dump.u_ssize + 1) * PAGE_SIZE > cprm->limit) + dump.u_ssize = 0; + +@@ -234,6 +239,8 @@ static int load_aout_binary(struct linux + rlim = rlimit(RLIMIT_DATA); + if (rlim >= RLIM_INFINITY) + rlim = ~0; ++ ++ gr_learn_resource(current, RLIMIT_DATA, ex.a_data + ex.a_bss, 1); + if (ex.a_data + ex.a_bss > rlim) + return -ENOMEM; + +@@ -262,6 +269,27 @@ static int load_aout_binary(struct linux + install_exec_creds(bprm); + current->flags &= ~PF_FORKNOEXEC; + ++#if defined(CONFIG_PAX_NOEXEC) || defined(CONFIG_PAX_ASLR) ++ current->mm->pax_flags = 0UL; ++#endif ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if (!(N_FLAGS(ex) & F_PAX_PAGEEXEC)) { ++ current->mm->pax_flags |= MF_PAX_PAGEEXEC; ++ ++#ifdef CONFIG_PAX_EMUTRAMP ++ if (N_FLAGS(ex) & F_PAX_EMUTRAMP) ++ current->mm->pax_flags |= MF_PAX_EMUTRAMP; ++#endif ++ ++#ifdef CONFIG_PAX_MPROTECT ++ if (!(N_FLAGS(ex) & F_PAX_MPROTECT)) ++ current->mm->pax_flags |= MF_PAX_MPROTECT; ++#endif ++ ++ } ++#endif ++ + if (N_MAGIC(ex) == OMAGIC) { + unsigned long text_addr, map_size; + loff_t pos; +@@ -334,7 +362,7 @@ static int load_aout_binary(struct linux + + down_write(¤t->mm->mmap_sem); + error = do_mmap(bprm->file, N_DATADDR(ex), ex.a_data, +- PROT_READ | PROT_WRITE | PROT_EXEC, ++ PROT_READ | PROT_WRITE, + MAP_FIXED | MAP_PRIVATE | MAP_DENYWRITE | MAP_EXECUTABLE, + fd_offset + ex.a_text); + up_write(¤t->mm->mmap_sem); +diff -urNp linux-2.6.38.3/fs/binfmt_elf.c linux-2.6.38.3/fs/binfmt_elf.c +--- linux-2.6.38.3/fs/binfmt_elf.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/binfmt_elf.c 2011-04-17 15:57:32.000000000 -0400 +@@ -51,6 +51,10 @@ static int elf_core_dump(struct coredump + #define elf_core_dump NULL + #endif + ++#ifdef CONFIG_PAX_MPROTECT ++static void elf_handle_mprotect(struct vm_area_struct *vma, unsigned long newflags); ++#endif ++ + #if ELF_EXEC_PAGESIZE > PAGE_SIZE + #define ELF_MIN_ALIGN ELF_EXEC_PAGESIZE + #else +@@ -70,6 +74,11 @@ static struct linux_binfmt elf_format = + .load_binary = load_elf_binary, + .load_shlib = load_elf_library, + .core_dump = elf_core_dump, ++ ++#ifdef CONFIG_PAX_MPROTECT ++ .handle_mprotect= elf_handle_mprotect, ++#endif ++ + .min_coredump = ELF_EXEC_PAGESIZE, + }; + +@@ -77,6 +86,8 @@ static struct linux_binfmt elf_format = + + static int set_brk(unsigned long start, unsigned long end) + { ++ unsigned long e = end; ++ + start = ELF_PAGEALIGN(start); + end = ELF_PAGEALIGN(end); + if (end > start) { +@@ -87,7 +98,7 @@ static int set_brk(unsigned long start, + if (BAD_ADDR(addr)) + return addr; + } +- current->mm->start_brk = current->mm->brk = end; ++ current->mm->start_brk = current->mm->brk = e; + return 0; + } + +@@ -148,12 +159,13 @@ create_elf_tables(struct linux_binprm *b + elf_addr_t __user *u_rand_bytes; + const char *k_platform = ELF_PLATFORM; + const char *k_base_platform = ELF_BASE_PLATFORM; +- unsigned char k_rand_bytes[16]; ++ u32 k_rand_bytes[4]; + int items; + elf_addr_t *elf_info; + int ei_index = 0; + const struct cred *cred = current_cred(); + struct vm_area_struct *vma; ++ unsigned long saved_auxv[AT_VECTOR_SIZE]; + + /* + * In some cases (e.g. Hyper-Threading), we want to avoid L1 +@@ -195,8 +207,12 @@ create_elf_tables(struct linux_binprm *b + * Generate 16 random bytes for userspace PRNG seeding. + */ + get_random_bytes(k_rand_bytes, sizeof(k_rand_bytes)); +- u_rand_bytes = (elf_addr_t __user *) +- STACK_ALLOC(p, sizeof(k_rand_bytes)); ++ srandom32(k_rand_bytes[0] ^ random32()); ++ srandom32(k_rand_bytes[1] ^ random32()); ++ srandom32(k_rand_bytes[2] ^ random32()); ++ srandom32(k_rand_bytes[3] ^ random32()); ++ p = STACK_ROUND(p, sizeof(k_rand_bytes)); ++ u_rand_bytes = (elf_addr_t __user *) p; + if (__copy_to_user(u_rand_bytes, k_rand_bytes, sizeof(k_rand_bytes))) + return -EFAULT; + +@@ -308,9 +324,11 @@ create_elf_tables(struct linux_binprm *b + return -EFAULT; + current->mm->env_end = p; + ++ memcpy(saved_auxv, elf_info, ei_index * sizeof(elf_addr_t)); ++ + /* Put the elf_info on the stack in the right place. */ + sp = (elf_addr_t __user *)envp + 1; +- if (copy_to_user(sp, elf_info, ei_index * sizeof(elf_addr_t))) ++ if (copy_to_user(sp, saved_auxv, ei_index * sizeof(elf_addr_t))) + return -EFAULT; + return 0; + } +@@ -381,10 +399,10 @@ static unsigned long load_elf_interp(str + { + struct elf_phdr *elf_phdata; + struct elf_phdr *eppnt; +- unsigned long load_addr = 0; ++ unsigned long load_addr = 0, pax_task_size = TASK_SIZE; + int load_addr_set = 0; + unsigned long last_bss = 0, elf_bss = 0; +- unsigned long error = ~0UL; ++ unsigned long error = -EINVAL; + unsigned long total_size; + int retval, i, size; + +@@ -430,6 +448,11 @@ static unsigned long load_elf_interp(str + goto out_close; + } + ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (current->mm->pax_flags & MF_PAX_SEGMEXEC) ++ pax_task_size = SEGMEXEC_TASK_SIZE; ++#endif ++ + eppnt = elf_phdata; + for (i = 0; i < interp_elf_ex->e_phnum; i++, eppnt++) { + if (eppnt->p_type == PT_LOAD) { +@@ -473,8 +496,8 @@ static unsigned long load_elf_interp(str + k = load_addr + eppnt->p_vaddr; + if (BAD_ADDR(k) || + eppnt->p_filesz > eppnt->p_memsz || +- eppnt->p_memsz > TASK_SIZE || +- TASK_SIZE - eppnt->p_memsz < k) { ++ eppnt->p_memsz > pax_task_size || ++ pax_task_size - eppnt->p_memsz < k) { + error = -ENOMEM; + goto out_close; + } +@@ -528,6 +551,193 @@ out: + return error; + } + ++#if (defined(CONFIG_PAX_EI_PAX) || defined(CONFIG_PAX_PT_PAX_FLAGS)) && defined(CONFIG_PAX_SOFTMODE) ++static unsigned long pax_parse_softmode(const struct elf_phdr * const elf_phdata) ++{ ++ unsigned long pax_flags = 0UL; ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if (elf_phdata->p_flags & PF_PAGEEXEC) ++ pax_flags |= MF_PAX_PAGEEXEC; ++#endif ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (elf_phdata->p_flags & PF_SEGMEXEC) ++ pax_flags |= MF_PAX_SEGMEXEC; ++#endif ++ ++#if defined(CONFIG_PAX_PAGEEXEC) && defined(CONFIG_PAX_SEGMEXEC) ++ if ((pax_flags & (MF_PAX_PAGEEXEC | MF_PAX_SEGMEXEC)) == (MF_PAX_PAGEEXEC | MF_PAX_SEGMEXEC)) { ++ if ((__supported_pte_mask & _PAGE_NX)) ++ pax_flags &= ~MF_PAX_SEGMEXEC; ++ else ++ pax_flags &= ~MF_PAX_PAGEEXEC; ++ } ++#endif ++ ++#ifdef CONFIG_PAX_EMUTRAMP ++ if (elf_phdata->p_flags & PF_EMUTRAMP) ++ pax_flags |= MF_PAX_EMUTRAMP; ++#endif ++ ++#ifdef CONFIG_PAX_MPROTECT ++ if (elf_phdata->p_flags & PF_MPROTECT) ++ pax_flags |= MF_PAX_MPROTECT; ++#endif ++ ++#if defined(CONFIG_PAX_RANDMMAP) || defined(CONFIG_PAX_RANDUSTACK) ++ if (randomize_va_space && (elf_phdata->p_flags & PF_RANDMMAP)) ++ pax_flags |= MF_PAX_RANDMMAP; ++#endif ++ ++ return pax_flags; ++} ++#endif ++ ++#ifdef CONFIG_PAX_PT_PAX_FLAGS ++static unsigned long pax_parse_hardmode(const struct elf_phdr * const elf_phdata) ++{ ++ unsigned long pax_flags = 0UL; ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if (!(elf_phdata->p_flags & PF_NOPAGEEXEC)) ++ pax_flags |= MF_PAX_PAGEEXEC; ++#endif ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (!(elf_phdata->p_flags & PF_NOSEGMEXEC)) ++ pax_flags |= MF_PAX_SEGMEXEC; ++#endif ++ ++#if defined(CONFIG_PAX_PAGEEXEC) && defined(CONFIG_PAX_SEGMEXEC) ++ if ((pax_flags & (MF_PAX_PAGEEXEC | MF_PAX_SEGMEXEC)) == (MF_PAX_PAGEEXEC | MF_PAX_SEGMEXEC)) { ++ if ((__supported_pte_mask & _PAGE_NX)) ++ pax_flags &= ~MF_PAX_SEGMEXEC; ++ else ++ pax_flags &= ~MF_PAX_PAGEEXEC; ++ } ++#endif ++ ++#ifdef CONFIG_PAX_EMUTRAMP ++ if (!(elf_phdata->p_flags & PF_NOEMUTRAMP)) ++ pax_flags |= MF_PAX_EMUTRAMP; ++#endif ++ ++#ifdef CONFIG_PAX_MPROTECT ++ if (!(elf_phdata->p_flags & PF_NOMPROTECT)) ++ pax_flags |= MF_PAX_MPROTECT; ++#endif ++ ++#if defined(CONFIG_PAX_RANDMMAP) || defined(CONFIG_PAX_RANDUSTACK) ++ if (randomize_va_space && !(elf_phdata->p_flags & PF_NORANDMMAP)) ++ pax_flags |= MF_PAX_RANDMMAP; ++#endif ++ ++ return pax_flags; ++} ++#endif ++ ++#ifdef CONFIG_PAX_EI_PAX ++static unsigned long pax_parse_ei_pax(const struct elfhdr * const elf_ex) ++{ ++ unsigned long pax_flags = 0UL; ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ if (!(elf_ex->e_ident[EI_PAX] & EF_PAX_PAGEEXEC)) ++ pax_flags |= MF_PAX_PAGEEXEC; ++#endif ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (!(elf_ex->e_ident[EI_PAX] & EF_PAX_SEGMEXEC)) ++ pax_flags |= MF_PAX_SEGMEXEC; ++#endif ++ ++#if defined(CONFIG_PAX_PAGEEXEC) && defined(CONFIG_PAX_SEGMEXEC) ++ if ((pax_flags & (MF_PAX_PAGEEXEC | MF_PAX_SEGMEXEC)) == (MF_PAX_PAGEEXEC | MF_PAX_SEGMEXEC)) { ++ if ((__supported_pte_mask & _PAGE_NX)) ++ pax_flags &= ~MF_PAX_SEGMEXEC; ++ else ++ pax_flags &= ~MF_PAX_PAGEEXEC; ++ } ++#endif ++ ++#ifdef CONFIG_PAX_EMUTRAMP ++ if ((pax_flags & (MF_PAX_PAGEEXEC | MF_PAX_SEGMEXEC)) && (elf_ex->e_ident[EI_PAX] & EF_PAX_EMUTRAMP)) ++ pax_flags |= MF_PAX_EMUTRAMP; ++#endif ++ ++#ifdef CONFIG_PAX_MPROTECT ++ if ((pax_flags & (MF_PAX_PAGEEXEC | MF_PAX_SEGMEXEC)) && !(elf_ex->e_ident[EI_PAX] & EF_PAX_MPROTECT)) ++ pax_flags |= MF_PAX_MPROTECT; ++#endif ++ ++#ifdef CONFIG_PAX_ASLR ++ if (randomize_va_space && !(elf_ex->e_ident[EI_PAX] & EF_PAX_RANDMMAP)) ++ pax_flags |= MF_PAX_RANDMMAP; ++#endif ++ ++ return pax_flags; ++} ++#endif ++ ++#if defined(CONFIG_PAX_EI_PAX) || defined(CONFIG_PAX_PT_PAX_FLAGS) ++static long pax_parse_elf_flags(const struct elfhdr * const elf_ex, const struct elf_phdr * const elf_phdata) ++{ ++ unsigned long pax_flags = 0UL; ++ ++#ifdef CONFIG_PAX_PT_PAX_FLAGS ++ unsigned long i; ++ int found_flags = 0; ++#endif ++ ++#ifdef CONFIG_PAX_EI_PAX ++ pax_flags = pax_parse_ei_pax(elf_ex); ++#endif ++ ++#ifdef CONFIG_PAX_PT_PAX_FLAGS ++ for (i = 0UL; i < elf_ex->e_phnum; i++) ++ if (elf_phdata[i].p_type == PT_PAX_FLAGS) { ++ if (((elf_phdata[i].p_flags & PF_PAGEEXEC) && (elf_phdata[i].p_flags & PF_NOPAGEEXEC)) || ++ ((elf_phdata[i].p_flags & PF_SEGMEXEC) && (elf_phdata[i].p_flags & PF_NOSEGMEXEC)) || ++ ((elf_phdata[i].p_flags & PF_EMUTRAMP) && (elf_phdata[i].p_flags & PF_NOEMUTRAMP)) || ++ ((elf_phdata[i].p_flags & PF_MPROTECT) && (elf_phdata[i].p_flags & PF_NOMPROTECT)) || ++ ((elf_phdata[i].p_flags & PF_RANDMMAP) && (elf_phdata[i].p_flags & PF_NORANDMMAP))) ++ return -EINVAL; ++ ++#ifdef CONFIG_PAX_SOFTMODE ++ if (pax_softmode) ++ pax_flags = pax_parse_softmode(&elf_phdata[i]); ++ else ++#endif ++ ++ pax_flags = pax_parse_hardmode(&elf_phdata[i]); ++ found_flags = 1; ++ break; ++ } ++#endif ++ ++#if !defined(CONFIG_PAX_EI_PAX) && defined(CONFIG_PAX_PT_PAX_FLAGS) ++ if (found_flags == 0) { ++ struct elf_phdr phdr; ++ memset(&phdr, 0, sizeof(phdr)); ++ phdr.p_flags = PF_NOEMUTRAMP; ++#ifdef CONFIG_PAX_SOFTMODE ++ if (pax_softmode) ++ pax_flags = pax_parse_softmode(&phdr); ++ else ++#endif ++ pax_flags = pax_parse_hardmode(&phdr); ++ } ++#endif ++ ++ if (0 > pax_check_flags(&pax_flags)) ++ return -EINVAL; ++ ++ current->mm->pax_flags = pax_flags; ++ return 0; ++} ++#endif ++ + /* + * These are the functions used to load ELF style executables and shared + * libraries. There is no binary dependent code anywhere else. +@@ -544,6 +754,11 @@ static unsigned long randomize_stack_top + { + unsigned int random_variable = 0; + ++#ifdef CONFIG_PAX_RANDUSTACK ++ if (randomize_va_space) ++ return stack_top - current->mm->delta_stack; ++#endif ++ + if ((current->flags & PF_RANDOMIZE) && + !(current->personality & ADDR_NO_RANDOMIZE)) { + random_variable = get_random_int() & STACK_RND_MASK; +@@ -562,7 +777,7 @@ static int load_elf_binary(struct linux_ + unsigned long load_addr = 0, load_bias = 0; + int load_addr_set = 0; + char * elf_interpreter = NULL; +- unsigned long error; ++ unsigned long error = 0; + struct elf_phdr *elf_ppnt, *elf_phdata; + unsigned long elf_bss, elf_brk; + int retval, i; +@@ -572,11 +787,11 @@ static int load_elf_binary(struct linux_ + unsigned long start_code, end_code, start_data, end_data; + unsigned long reloc_func_desc = 0; + int executable_stack = EXSTACK_DEFAULT; +- unsigned long def_flags = 0; + struct { + struct elfhdr elf_ex; + struct elfhdr interp_elf_ex; + } *loc; ++ unsigned long pax_task_size = TASK_SIZE; + + loc = kmalloc(sizeof(*loc), GFP_KERNEL); + if (!loc) { +@@ -714,11 +929,80 @@ static int load_elf_binary(struct linux_ + + /* OK, This is the point of no return */ + current->flags &= ~PF_FORKNOEXEC; +- current->mm->def_flags = def_flags; ++ ++#if defined(CONFIG_PAX_NOEXEC) || defined(CONFIG_PAX_ASLR) ++ current->mm->pax_flags = 0UL; ++#endif ++ ++#ifdef CONFIG_PAX_DLRESOLVE ++ current->mm->call_dl_resolve = 0UL; ++#endif ++ ++#if defined(CONFIG_PPC32) && defined(CONFIG_PAX_EMUSIGRT) ++ current->mm->call_syscall = 0UL; ++#endif ++ ++#ifdef CONFIG_PAX_ASLR ++ current->mm->delta_mmap = 0UL; ++ current->mm->delta_stack = 0UL; ++#endif ++ ++ current->mm->def_flags = 0; ++ ++#if defined(CONFIG_PAX_EI_PAX) || defined(CONFIG_PAX_PT_PAX_FLAGS) ++ if (0 > pax_parse_elf_flags(&loc->elf_ex, elf_phdata)) { ++ send_sig(SIGKILL, current, 0); ++ goto out_free_dentry; ++ } ++#endif ++ ++#ifdef CONFIG_PAX_HAVE_ACL_FLAGS ++ pax_set_initial_flags(bprm); ++#elif defined(CONFIG_PAX_HOOK_ACL_FLAGS) ++ if (pax_set_initial_flags_func) ++ (pax_set_initial_flags_func)(bprm); ++#endif ++ ++#ifdef CONFIG_ARCH_TRACK_EXEC_LIMIT ++ if ((current->mm->pax_flags & MF_PAX_PAGEEXEC) && !(__supported_pte_mask & _PAGE_NX)) { ++ current->mm->context.user_cs_limit = PAGE_SIZE; ++ current->mm->def_flags |= VM_PAGEEXEC; ++ } ++#endif ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (current->mm->pax_flags & MF_PAX_SEGMEXEC) { ++ current->mm->context.user_cs_base = SEGMEXEC_TASK_SIZE; ++ current->mm->context.user_cs_limit = TASK_SIZE-SEGMEXEC_TASK_SIZE; ++ pax_task_size = SEGMEXEC_TASK_SIZE; ++ } ++#endif ++ ++#if defined(CONFIG_ARCH_TRACK_EXEC_LIMIT) || defined(CONFIG_PAX_SEGMEXEC) ++ if (current->mm->pax_flags & (MF_PAX_PAGEEXEC | MF_PAX_SEGMEXEC)) { ++ set_user_cs(current->mm->context.user_cs_base, current->mm->context.user_cs_limit, get_cpu()); ++ put_cpu(); ++ } ++#endif + + /* Do this immediately, since STACK_TOP as used in setup_arg_pages + may depend on the personality. */ + SET_PERSONALITY(loc->elf_ex); ++ ++#ifdef CONFIG_PAX_ASLR ++ if (current->mm->pax_flags & MF_PAX_RANDMMAP) { ++ current->mm->delta_mmap = (pax_get_random_long() & ((1UL << PAX_DELTA_MMAP_LEN)-1)) << PAGE_SHIFT; ++ current->mm->delta_stack = (pax_get_random_long() & ((1UL << PAX_DELTA_STACK_LEN)-1)) << PAGE_SHIFT; ++ } ++#endif ++ ++#if defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) ++ if (current->mm->pax_flags & (MF_PAX_PAGEEXEC | MF_PAX_SEGMEXEC)) { ++ executable_stack = EXSTACK_DISABLE_X; ++ current->personality &= ~READ_IMPLIES_EXEC; ++ } else ++#endif ++ + if (elf_read_implies_exec(loc->elf_ex, executable_stack)) + current->personality |= READ_IMPLIES_EXEC; + +@@ -800,6 +1084,20 @@ static int load_elf_binary(struct linux_ + #else + load_bias = ELF_PAGESTART(ELF_ET_DYN_BASE - vaddr); + #endif ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ /* PaX: randomize base address at the default exe base if requested */ ++ if ((current->mm->pax_flags & MF_PAX_RANDMMAP) && elf_interpreter) { ++#ifdef CONFIG_SPARC64 ++ load_bias = (pax_get_random_long() & ((1UL << PAX_DELTA_MMAP_LEN) - 1)) << (PAGE_SHIFT+1); ++#else ++ load_bias = (pax_get_random_long() & ((1UL << PAX_DELTA_MMAP_LEN) - 1)) << PAGE_SHIFT; ++#endif ++ load_bias = ELF_PAGESTART(PAX_ELF_ET_DYN_BASE - vaddr + load_bias); ++ elf_flags |= MAP_FIXED; ++ } ++#endif ++ + } + + error = elf_map(bprm->file, load_bias + vaddr, elf_ppnt, +@@ -832,9 +1130,9 @@ static int load_elf_binary(struct linux_ + * allowed task size. Note that p_filesz must always be + * <= p_memsz so it is only necessary to check p_memsz. + */ +- if (BAD_ADDR(k) || elf_ppnt->p_filesz > elf_ppnt->p_memsz || +- elf_ppnt->p_memsz > TASK_SIZE || +- TASK_SIZE - elf_ppnt->p_memsz < k) { ++ if (k >= pax_task_size || elf_ppnt->p_filesz > elf_ppnt->p_memsz || ++ elf_ppnt->p_memsz > pax_task_size || ++ pax_task_size - elf_ppnt->p_memsz < k) { + /* set_brk can never work. Avoid overflows. */ + send_sig(SIGKILL, current, 0); + retval = -EINVAL; +@@ -862,6 +1160,11 @@ static int load_elf_binary(struct linux_ + start_data += load_bias; + end_data += load_bias; + ++#ifdef CONFIG_PAX_RANDMMAP ++ if (current->mm->pax_flags & MF_PAX_RANDMMAP) ++ elf_brk += PAGE_SIZE + ((pax_get_random_long() & ~PAGE_MASK) << 4); ++#endif ++ + /* Calling set_brk effectively mmaps the pages that we need + * for the bss and break sections. We must do this before + * mapping in the interpreter, to make sure it doesn't wind +@@ -873,9 +1176,11 @@ static int load_elf_binary(struct linux_ + goto out_free_dentry; + } + if (likely(elf_bss != elf_brk) && unlikely(padzero(elf_bss))) { +- send_sig(SIGSEGV, current, 0); +- retval = -EFAULT; /* Nobody gets to see this, but.. */ +- goto out_free_dentry; ++ /* ++ * This bss-zeroing can fail if the ELF ++ * file specifies odd protections. So ++ * we don't check the return value ++ */ + } + + if (elf_interpreter) { +@@ -1086,7 +1391,7 @@ out: + * Decide what to dump of a segment, part, all or none. + */ + static unsigned long vma_dump_size(struct vm_area_struct *vma, +- unsigned long mm_flags) ++ unsigned long mm_flags, long signr) + { + #define FILTER(type) (mm_flags & (1UL << MMF_DUMP_##type)) + +@@ -1120,7 +1425,7 @@ static unsigned long vma_dump_size(struc + if (vma->vm_file == NULL) + return 0; + +- if (FILTER(MAPPED_PRIVATE)) ++ if (signr == SIGKILL || FILTER(MAPPED_PRIVATE)) + goto whole; + + /* +@@ -1342,9 +1647,9 @@ static void fill_auxv_note(struct memelf + { + elf_addr_t *auxv = (elf_addr_t *) mm->saved_auxv; + int i = 0; +- do ++ do { + i += 2; +- while (auxv[i - 2] != AT_NULL); ++ } while (auxv[i - 2] != AT_NULL); + fill_note(note, "CORE", NT_AUXV, i * sizeof(elf_addr_t), auxv); + } + +@@ -1850,14 +2155,14 @@ static void fill_extnum_info(struct elfh + } + + static size_t elf_core_vma_data_size(struct vm_area_struct *gate_vma, +- unsigned long mm_flags) ++ struct coredump_params *cprm) + { + struct vm_area_struct *vma; + size_t size = 0; + + for (vma = first_vma(current, gate_vma); vma != NULL; + vma = next_vma(vma, gate_vma)) +- size += vma_dump_size(vma, mm_flags); ++ size += vma_dump_size(vma, cprm->mm_flags, cprm->signr); + return size; + } + +@@ -1951,7 +2256,7 @@ static int elf_core_dump(struct coredump + + dataoff = offset = roundup(offset, ELF_EXEC_PAGESIZE); + +- offset += elf_core_vma_data_size(gate_vma, cprm->mm_flags); ++ offset += elf_core_vma_data_size(gate_vma, cprm); + offset += elf_core_extra_data_size(); + e_shoff = offset; + +@@ -1965,10 +2270,12 @@ static int elf_core_dump(struct coredump + offset = dataoff; + + size += sizeof(*elf); ++ gr_learn_resource(current, RLIMIT_CORE, size, 1); + if (size > cprm->limit || !dump_write(cprm->file, elf, sizeof(*elf))) + goto end_coredump; + + size += sizeof(*phdr4note); ++ gr_learn_resource(current, RLIMIT_CORE, size, 1); + if (size > cprm->limit + || !dump_write(cprm->file, phdr4note, sizeof(*phdr4note))) + goto end_coredump; +@@ -1982,7 +2289,7 @@ static int elf_core_dump(struct coredump + phdr.p_offset = offset; + phdr.p_vaddr = vma->vm_start; + phdr.p_paddr = 0; +- phdr.p_filesz = vma_dump_size(vma, cprm->mm_flags); ++ phdr.p_filesz = vma_dump_size(vma, cprm->mm_flags, cprm->signr); + phdr.p_memsz = vma->vm_end - vma->vm_start; + offset += phdr.p_filesz; + phdr.p_flags = vma->vm_flags & VM_READ ? PF_R : 0; +@@ -1993,6 +2300,7 @@ static int elf_core_dump(struct coredump + phdr.p_align = ELF_EXEC_PAGESIZE; + + size += sizeof(phdr); ++ gr_learn_resource(current, RLIMIT_CORE, size, 1); + if (size > cprm->limit + || !dump_write(cprm->file, &phdr, sizeof(phdr))) + goto end_coredump; +@@ -2017,7 +2325,7 @@ static int elf_core_dump(struct coredump + unsigned long addr; + unsigned long end; + +- end = vma->vm_start + vma_dump_size(vma, cprm->mm_flags); ++ end = vma->vm_start + vma_dump_size(vma, cprm->mm_flags, cprm->signr); + + for (addr = vma->vm_start; addr < end; addr += PAGE_SIZE) { + struct page *page; +@@ -2026,6 +2334,7 @@ static int elf_core_dump(struct coredump + page = get_dump_page(addr); + if (page) { + void *kaddr = kmap(page); ++ gr_learn_resource(current, RLIMIT_CORE, size + PAGE_SIZE, 1); + stop = ((size += PAGE_SIZE) > cprm->limit) || + !dump_write(cprm->file, kaddr, + PAGE_SIZE); +@@ -2043,6 +2352,7 @@ static int elf_core_dump(struct coredump + + if (e_phnum == PN_XNUM) { + size += sizeof(*shdr4extnum); ++ gr_learn_resource(current, RLIMIT_CORE, size, 1); + if (size > cprm->limit + || !dump_write(cprm->file, shdr4extnum, + sizeof(*shdr4extnum))) +@@ -2063,6 +2373,97 @@ out: + + #endif /* CONFIG_ELF_CORE */ + ++#ifdef CONFIG_PAX_MPROTECT ++/* PaX: non-PIC ELF libraries need relocations on their executable segments ++ * therefore we'll grant them VM_MAYWRITE once during their life. Similarly ++ * we'll remove VM_MAYWRITE for good on RELRO segments. ++ * ++ * The checks favour ld-linux.so behaviour which operates on a per ELF segment ++ * basis because we want to allow the common case and not the special ones. ++ */ ++static void elf_handle_mprotect(struct vm_area_struct *vma, unsigned long newflags) ++{ ++ struct elfhdr elf_h; ++ struct elf_phdr elf_p; ++ unsigned long i; ++ unsigned long oldflags; ++ bool is_textrel_rw, is_textrel_rx, is_relro; ++ ++ if (!(vma->vm_mm->pax_flags & MF_PAX_MPROTECT)) ++ return; ++ ++ oldflags = vma->vm_flags & (VM_MAYEXEC | VM_MAYWRITE | VM_MAYREAD | VM_EXEC | VM_WRITE | VM_READ); ++ newflags &= VM_MAYEXEC | VM_MAYWRITE | VM_MAYREAD | VM_EXEC | VM_WRITE | VM_READ; ++ ++#ifdef CONFIG_PAX_ELFRELOCS ++ /* possible TEXTREL */ ++ is_textrel_rw = vma->vm_file && !vma->anon_vma && oldflags == (VM_MAYEXEC | VM_MAYREAD | VM_EXEC | VM_READ) && newflags == (VM_WRITE | VM_READ); ++ is_textrel_rx = vma->vm_file && vma->anon_vma && oldflags == (VM_MAYEXEC | VM_MAYWRITE | VM_MAYREAD | VM_WRITE | VM_READ) && newflags == (VM_EXEC | VM_READ); ++#else ++ is_textrel_rw = false; ++ is_textrel_rx = false; ++#endif ++ ++ /* possible RELRO */ ++ is_relro = vma->vm_file && vma->anon_vma && oldflags == (VM_MAYWRITE | VM_MAYREAD | VM_READ) && newflags == (VM_MAYWRITE | VM_MAYREAD | VM_READ); ++ ++ if (!is_textrel_rw && !is_textrel_rx && !is_relro) ++ return; ++ ++ if (sizeof(elf_h) != kernel_read(vma->vm_file, 0UL, (char *)&elf_h, sizeof(elf_h)) || ++ memcmp(elf_h.e_ident, ELFMAG, SELFMAG) || ++ ++#ifdef CONFIG_PAX_ETEXECRELOCS ++ ((is_textrel_rw || is_textrel_rx) && (elf_h.e_type != ET_DYN && elf_h.e_type != ET_EXEC)) || ++#else ++ ((is_textrel_rw || is_textrel_rx) && elf_h.e_type != ET_DYN) || ++#endif ++ ++ (is_relro && (elf_h.e_type != ET_DYN && elf_h.e_type != ET_EXEC)) || ++ !elf_check_arch(&elf_h) || ++ elf_h.e_phentsize != sizeof(struct elf_phdr) || ++ elf_h.e_phnum > 65536UL / sizeof(struct elf_phdr)) ++ return; ++ ++ for (i = 0UL; i < elf_h.e_phnum; i++) { ++ if (sizeof(elf_p) != kernel_read(vma->vm_file, elf_h.e_phoff + i*sizeof(elf_p), (char *)&elf_p, sizeof(elf_p))) ++ return; ++ switch (elf_p.p_type) { ++ case PT_DYNAMIC: ++ if (!is_textrel_rw && !is_textrel_rx) ++ continue; ++ i = 0UL; ++ while ((i+1) * sizeof(elf_dyn) <= elf_p.p_filesz) { ++ elf_dyn dyn; ++ ++ if (sizeof(dyn) != kernel_read(vma->vm_file, elf_p.p_offset + i*sizeof(dyn), (char *)&dyn, sizeof(dyn))) ++ return; ++ if (dyn.d_tag == DT_NULL) ++ return; ++ if (dyn.d_tag == DT_TEXTREL || (dyn.d_tag == DT_FLAGS && (dyn.d_un.d_val & DF_TEXTREL))) { ++ gr_log_textrel(vma); ++ if (is_textrel_rw) ++ vma->vm_flags |= VM_MAYWRITE; ++ else ++ /* PaX: disallow write access after relocs are done, hopefully noone else needs it... */ ++ vma->vm_flags &= ~VM_MAYWRITE; ++ return; ++ } ++ i++; ++ } ++ return; ++ ++ case PT_GNU_RELRO: ++ if (!is_relro) ++ continue; ++ if ((elf_p.p_offset >> PAGE_SHIFT) == vma->vm_pgoff && ELF_PAGEALIGN(elf_p.p_memsz) == vma->vm_end - vma->vm_start) ++ vma->vm_flags &= ~VM_MAYWRITE; ++ return; ++ } ++ } ++} ++#endif ++ + static int __init init_elf_binfmt(void) + { + return register_binfmt(&elf_format); +diff -urNp linux-2.6.38.3/fs/binfmt_flat.c linux-2.6.38.3/fs/binfmt_flat.c +--- linux-2.6.38.3/fs/binfmt_flat.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/binfmt_flat.c 2011-04-17 15:57:32.000000000 -0400 +@@ -567,7 +567,9 @@ static int load_flat_file(struct linux_b + realdatastart = (unsigned long) -ENOMEM; + printk("Unable to allocate RAM for process data, errno %d\n", + (int)-realdatastart); ++ down_write(¤t->mm->mmap_sem); + do_munmap(current->mm, textpos, text_len); ++ up_write(¤t->mm->mmap_sem); + ret = realdatastart; + goto err; + } +@@ -591,8 +593,10 @@ static int load_flat_file(struct linux_b + } + if (IS_ERR_VALUE(result)) { + printk("Unable to read data+bss, errno %d\n", (int)-result); ++ down_write(¤t->mm->mmap_sem); + do_munmap(current->mm, textpos, text_len); + do_munmap(current->mm, realdatastart, len); ++ up_write(¤t->mm->mmap_sem); + ret = result; + goto err; + } +@@ -661,8 +665,10 @@ static int load_flat_file(struct linux_b + } + if (IS_ERR_VALUE(result)) { + printk("Unable to read code+data+bss, errno %d\n",(int)-result); ++ down_write(¤t->mm->mmap_sem); + do_munmap(current->mm, textpos, text_len + data_len + extra + + MAX_SHARED_LIBS * sizeof(unsigned long)); ++ up_write(¤t->mm->mmap_sem); + ret = result; + goto err; + } +diff -urNp linux-2.6.38.3/fs/binfmt_misc.c linux-2.6.38.3/fs/binfmt_misc.c +--- linux-2.6.38.3/fs/binfmt_misc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/binfmt_misc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -698,7 +698,7 @@ static int bm_fill_super(struct super_bl + static struct tree_descr bm_files[] = { + [2] = {"status", &bm_status_operations, S_IWUSR|S_IRUGO}, + [3] = {"register", &bm_register_operations, S_IWUSR}, +- /* last one */ {""} ++ /* last one */ {"", NULL, 0} + }; + int err = simple_fill_super(sb, 0x42494e4d, bm_files); + if (!err) +diff -urNp linux-2.6.38.3/fs/bio.c linux-2.6.38.3/fs/bio.c +--- linux-2.6.38.3/fs/bio.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/bio.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1233,7 +1233,7 @@ static void bio_copy_kern_endio(struct b + const int read = bio_data_dir(bio) == READ; + struct bio_map_data *bmd = bio->bi_private; + int i; +- char *p = bmd->sgvecs[0].iov_base; ++ char *p = (__force char *)bmd->sgvecs[0].iov_base; + + __bio_for_each_segment(bvec, bio, i, 0) { + char *addr = page_address(bvec->bv_page); +diff -urNp linux-2.6.38.3/fs/block_dev.c linux-2.6.38.3/fs/block_dev.c +--- linux-2.6.38.3/fs/block_dev.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/block_dev.c 2011-04-17 15:57:32.000000000 -0400 +@@ -669,7 +669,7 @@ static bool bd_may_claim(struct block_de + else if (bdev->bd_contains == bdev) + return true; /* is a whole device which isn't held */ + +- else if (whole->bd_holder == bd_may_claim) ++ else if (whole->bd_holder == (void *)bd_may_claim) + return true; /* is a partition of a device that is being partitioned */ + else if (whole->bd_holder != NULL) + return false; /* is a partition of a held device */ +diff -urNp linux-2.6.38.3/fs/btrfs/ctree.c linux-2.6.38.3/fs/btrfs/ctree.c +--- linux-2.6.38.3/fs/btrfs/ctree.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/btrfs/ctree.c 2011-04-17 15:57:32.000000000 -0400 +@@ -468,9 +468,12 @@ static noinline int __btrfs_cow_block(st + free_extent_buffer(buf); + add_root_to_dirty_list(root); + } else { +- if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) +- parent_start = parent->start; +- else ++ if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) { ++ if (parent) ++ parent_start = parent->start; ++ else ++ parent_start = 0; ++ } else + parent_start = 0; + + WARN_ON(trans->transid != btrfs_header_generation(parent)); +@@ -3776,7 +3779,6 @@ setup_items_for_insert(struct btrfs_tran + + ret = 0; + if (slot == 0) { +- struct btrfs_disk_key disk_key; + btrfs_cpu_key_to_disk(&disk_key, cpu_key); + ret = fixup_low_keys(trans, root, path, &disk_key, 1); + } +diff -urNp linux-2.6.38.3/fs/btrfs/disk-io.c linux-2.6.38.3/fs/btrfs/disk-io.c +--- linux-2.6.38.3/fs/btrfs/disk-io.c 2011-04-17 16:53:01.000000000 -0400 ++++ linux-2.6.38.3/fs/btrfs/disk-io.c 2011-04-17 16:53:48.000000000 -0400 +@@ -41,7 +41,7 @@ + #include "tree-log.h" + #include "free-space-cache.h" + +-static struct extent_io_ops btree_extent_io_ops; ++static const struct extent_io_ops btree_extent_io_ops; + static void end_workqueue_fn(struct btrfs_work *work); + static void free_fs_root(struct btrfs_root *root); + static void btrfs_check_super_valid(struct btrfs_fs_info *fs_info, +@@ -3030,7 +3030,7 @@ static int btrfs_cleanup_transaction(str + return 0; + } + +-static struct extent_io_ops btree_extent_io_ops = { ++static const struct extent_io_ops btree_extent_io_ops = { + .write_cache_pages_lock_hook = btree_lock_page_hook, + .readpage_end_io_hook = btree_readpage_end_io_hook, + .submit_bio_hook = btree_submit_bio_hook, +diff -urNp linux-2.6.38.3/fs/btrfs/extent_io.h linux-2.6.38.3/fs/btrfs/extent_io.h +--- linux-2.6.38.3/fs/btrfs/extent_io.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/btrfs/extent_io.h 2011-04-17 15:57:32.000000000 -0400 +@@ -55,36 +55,36 @@ typedef int (extent_submit_bio_hook_t)(s + struct bio *bio, int mirror_num, + unsigned long bio_flags, u64 bio_offset); + struct extent_io_ops { +- int (*fill_delalloc)(struct inode *inode, struct page *locked_page, ++ int (* const fill_delalloc)(struct inode *inode, struct page *locked_page, + u64 start, u64 end, int *page_started, + unsigned long *nr_written); +- int (*writepage_start_hook)(struct page *page, u64 start, u64 end); +- int (*writepage_io_hook)(struct page *page, u64 start, u64 end); ++ int (* const writepage_start_hook)(struct page *page, u64 start, u64 end); ++ int (* const writepage_io_hook)(struct page *page, u64 start, u64 end); + extent_submit_bio_hook_t *submit_bio_hook; +- int (*merge_bio_hook)(struct page *page, unsigned long offset, ++ int (* const merge_bio_hook)(struct page *page, unsigned long offset, + size_t size, struct bio *bio, + unsigned long bio_flags); +- int (*readpage_io_hook)(struct page *page, u64 start, u64 end); +- int (*readpage_io_failed_hook)(struct bio *bio, struct page *page, ++ int (* const readpage_io_hook)(struct page *page, u64 start, u64 end); ++ int (* const readpage_io_failed_hook)(struct bio *bio, struct page *page, + u64 start, u64 end, + struct extent_state *state); +- int (*writepage_io_failed_hook)(struct bio *bio, struct page *page, ++ int (* const writepage_io_failed_hook)(struct bio *bio, struct page *page, + u64 start, u64 end, + struct extent_state *state); +- int (*readpage_end_io_hook)(struct page *page, u64 start, u64 end, ++ int (* const readpage_end_io_hook)(struct page *page, u64 start, u64 end, + struct extent_state *state); +- int (*writepage_end_io_hook)(struct page *page, u64 start, u64 end, ++ int (* const writepage_end_io_hook)(struct page *page, u64 start, u64 end, + struct extent_state *state, int uptodate); +- int (*set_bit_hook)(struct inode *inode, struct extent_state *state, ++ int (* const set_bit_hook)(struct inode *inode, struct extent_state *state, + int *bits); +- int (*clear_bit_hook)(struct inode *inode, struct extent_state *state, ++ int (* const clear_bit_hook)(struct inode *inode, struct extent_state *state, + int *bits); +- int (*merge_extent_hook)(struct inode *inode, ++ int (* const merge_extent_hook)(struct inode *inode, + struct extent_state *new, + struct extent_state *other); +- int (*split_extent_hook)(struct inode *inode, ++ int (* const split_extent_hook)(struct inode *inode, + struct extent_state *orig, u64 split); +- int (*write_cache_pages_lock_hook)(struct page *page); ++ int (* const write_cache_pages_lock_hook)(struct page *page); + }; + + struct extent_io_tree { +@@ -94,7 +94,7 @@ struct extent_io_tree { + u64 dirty_bytes; + spinlock_t lock; + spinlock_t buffer_lock; +- struct extent_io_ops *ops; ++ const struct extent_io_ops *ops; + }; + + struct extent_state { +diff -urNp linux-2.6.38.3/fs/btrfs/free-space-cache.c linux-2.6.38.3/fs/btrfs/free-space-cache.c +--- linux-2.6.38.3/fs/btrfs/free-space-cache.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/btrfs/free-space-cache.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1855,8 +1855,6 @@ u64 btrfs_alloc_from_cluster(struct btrf + + while(1) { + if (entry->bytes < bytes || entry->offset < min_start) { +- struct rb_node *node; +- + node = rb_next(&entry->offset_index); + if (!node) + break; +@@ -2018,7 +2016,7 @@ again: + */ + while (entry->bitmap || found_bitmap || + (!entry->bitmap && entry->bytes < min_bytes)) { +- struct rb_node *node = rb_next(&entry->offset_index); ++ node = rb_next(&entry->offset_index); + + if (entry->bitmap && entry->bytes > bytes + empty_size) { + ret = btrfs_bitmap_cluster(block_group, entry, cluster, +diff -urNp linux-2.6.38.3/fs/btrfs/inode.c linux-2.6.38.3/fs/btrfs/inode.c +--- linux-2.6.38.3/fs/btrfs/inode.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/btrfs/inode.c 2011-04-17 15:57:32.000000000 -0400 +@@ -64,7 +64,7 @@ static const struct inode_operations btr + static const struct address_space_operations btrfs_aops; + static const struct address_space_operations btrfs_symlink_aops; + static const struct file_operations btrfs_dir_file_operations; +-static struct extent_io_ops btrfs_extent_io_ops; ++static const struct extent_io_ops btrfs_extent_io_ops; + + static struct kmem_cache *btrfs_inode_cachep; + struct kmem_cache *btrfs_trans_handle_cachep; +@@ -6796,7 +6796,7 @@ fail: + return -ENOMEM; + } + +-static int btrfs_getattr(struct vfsmount *mnt, ++int btrfs_getattr(struct vfsmount *mnt, + struct dentry *dentry, struct kstat *stat) + { + struct inode *inode = dentry->d_inode; +@@ -6808,6 +6808,14 @@ static int btrfs_getattr(struct vfsmount + return 0; + } + ++EXPORT_SYMBOL(btrfs_getattr); ++ ++dev_t get_btrfs_dev_from_inode(struct inode *inode) ++{ ++ return BTRFS_I(inode)->root->anon_super.s_dev; ++} ++EXPORT_SYMBOL(get_btrfs_dev_from_inode); ++ + static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, + struct inode *new_dir, struct dentry *new_dentry) + { +@@ -7311,7 +7319,7 @@ static const struct file_operations btrf + .fsync = btrfs_sync_file, + }; + +-static struct extent_io_ops btrfs_extent_io_ops = { ++static const struct extent_io_ops btrfs_extent_io_ops = { + .fill_delalloc = run_delalloc_range, + .submit_bio_hook = btrfs_submit_bio_hook, + .merge_bio_hook = btrfs_merge_bio_hook, +diff -urNp linux-2.6.38.3/fs/btrfs/ioctl.c linux-2.6.38.3/fs/btrfs/ioctl.c +--- linux-2.6.38.3/fs/btrfs/ioctl.c 2011-04-17 16:53:01.000000000 -0400 ++++ linux-2.6.38.3/fs/btrfs/ioctl.c 2011-04-17 16:53:48.000000000 -0400 +@@ -2274,9 +2274,12 @@ long btrfs_ioctl_space_info(struct btrfs + for (i = 0; i < num_types; i++) { + struct btrfs_space_info *tmp; + ++ /* Don't copy in more than we allocated */ + if (!slot_count) + break; + ++ slot_count--; ++ + info = NULL; + rcu_read_lock(); + list_for_each_entry_rcu(tmp, &root->fs_info->space_info, +@@ -2298,10 +2301,7 @@ long btrfs_ioctl_space_info(struct btrfs + memcpy(dest, &space, sizeof(space)); + dest++; + space_args.total_spaces++; +- slot_count--; + } +- if (!slot_count) +- break; + } + up_read(&info->groups_sem); + } +diff -urNp linux-2.6.38.3/fs/btrfs/relocation.c linux-2.6.38.3/fs/btrfs/relocation.c +--- linux-2.6.38.3/fs/btrfs/relocation.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/btrfs/relocation.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1239,7 +1239,7 @@ static int __update_reloc_root(struct bt + } + spin_unlock(&rc->reloc_root_tree.lock); + +- BUG_ON((struct btrfs_root *)node->data != root); ++ BUG_ON(!node || (struct btrfs_root *)node->data != root); + + if (!del) { + spin_lock(&rc->reloc_root_tree.lock); +diff -urNp linux-2.6.38.3/fs/cachefiles/bind.c linux-2.6.38.3/fs/cachefiles/bind.c +--- linux-2.6.38.3/fs/cachefiles/bind.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/cachefiles/bind.c 2011-04-17 15:57:32.000000000 -0400 +@@ -39,13 +39,11 @@ int cachefiles_daemon_bind(struct cachef + args); + + /* start by checking things over */ +- ASSERT(cache->fstop_percent >= 0 && +- cache->fstop_percent < cache->fcull_percent && ++ ASSERT(cache->fstop_percent < cache->fcull_percent && + cache->fcull_percent < cache->frun_percent && + cache->frun_percent < 100); + +- ASSERT(cache->bstop_percent >= 0 && +- cache->bstop_percent < cache->bcull_percent && ++ ASSERT(cache->bstop_percent < cache->bcull_percent && + cache->bcull_percent < cache->brun_percent && + cache->brun_percent < 100); + +diff -urNp linux-2.6.38.3/fs/cachefiles/daemon.c linux-2.6.38.3/fs/cachefiles/daemon.c +--- linux-2.6.38.3/fs/cachefiles/daemon.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/cachefiles/daemon.c 2011-04-17 15:57:32.000000000 -0400 +@@ -196,7 +196,7 @@ static ssize_t cachefiles_daemon_read(st + if (n > buflen) + return -EMSGSIZE; + +- if (copy_to_user(_buffer, buffer, n) != 0) ++ if (n > sizeof(buffer) || copy_to_user(_buffer, buffer, n) != 0) + return -EFAULT; + + return n; +@@ -222,7 +222,7 @@ static ssize_t cachefiles_daemon_write(s + if (test_bit(CACHEFILES_DEAD, &cache->flags)) + return -EIO; + +- if (datalen < 0 || datalen > PAGE_SIZE - 1) ++ if (datalen > PAGE_SIZE - 1) + return -EOPNOTSUPP; + + /* drag the command string into the kernel so we can parse it */ +@@ -386,7 +386,7 @@ static int cachefiles_daemon_fstop(struc + if (args[0] != '%' || args[1] != '\0') + return -EINVAL; + +- if (fstop < 0 || fstop >= cache->fcull_percent) ++ if (fstop >= cache->fcull_percent) + return cachefiles_daemon_range_error(cache, args); + + cache->fstop_percent = fstop; +@@ -458,7 +458,7 @@ static int cachefiles_daemon_bstop(struc + if (args[0] != '%' || args[1] != '\0') + return -EINVAL; + +- if (bstop < 0 || bstop >= cache->bcull_percent) ++ if (bstop >= cache->bcull_percent) + return cachefiles_daemon_range_error(cache, args); + + cache->bstop_percent = bstop; +diff -urNp linux-2.6.38.3/fs/cachefiles/rdwr.c linux-2.6.38.3/fs/cachefiles/rdwr.c +--- linux-2.6.38.3/fs/cachefiles/rdwr.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/cachefiles/rdwr.c 2011-04-17 15:57:32.000000000 -0400 +@@ -945,7 +945,7 @@ int cachefiles_write_page(struct fscache + old_fs = get_fs(); + set_fs(KERNEL_DS); + ret = file->f_op->write( +- file, (const void __user *) data, len, &pos); ++ file, (__force const void __user *) data, len, &pos); + set_fs(old_fs); + kunmap(page); + if (ret != len) +diff -urNp linux-2.6.38.3/fs/ceph/dir.c linux-2.6.38.3/fs/ceph/dir.c +--- linux-2.6.38.3/fs/ceph/dir.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/ceph/dir.c 2011-04-17 15:57:32.000000000 -0400 +@@ -226,7 +226,7 @@ static int ceph_readdir(struct file *fil + struct ceph_fs_client *fsc = ceph_inode_to_client(inode); + struct ceph_mds_client *mdsc = fsc->mdsc; + unsigned frag = fpos_frag(filp->f_pos); +- int off = fpos_off(filp->f_pos); ++ unsigned int off = fpos_off(filp->f_pos); + int err; + u32 ftype; + struct ceph_mds_reply_info_parsed *rinfo; +@@ -358,7 +358,7 @@ more: + rinfo = &fi->last_readdir->r_reply_info; + dout("readdir frag %x num %d off %d chunkoff %d\n", frag, + rinfo->dir_nr, off, fi->offset); +- while (off - fi->offset >= 0 && off - fi->offset < rinfo->dir_nr) { ++ while (off >= fi->offset && off - fi->offset < rinfo->dir_nr) { + u64 pos = ceph_make_fpos(frag, off); + struct ceph_mds_reply_inode *in = + rinfo->dir_in[off - fi->offset].in; +diff -urNp linux-2.6.38.3/fs/cifs/cifs_uniupr.h linux-2.6.38.3/fs/cifs/cifs_uniupr.h +--- linux-2.6.38.3/fs/cifs/cifs_uniupr.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/cifs/cifs_uniupr.h 2011-04-17 15:57:32.000000000 -0400 +@@ -132,7 +132,7 @@ const struct UniCaseRange CifsUniUpperRa + {0x0490, 0x04cc, UniCaseRangeU0490}, + {0x1e00, 0x1ffc, UniCaseRangeU1e00}, + {0xff40, 0xff5a, UniCaseRangeUff40}, +- {0} ++ {0, 0, NULL} + }; + #endif + +diff -urNp linux-2.6.38.3/fs/cifs/link.c linux-2.6.38.3/fs/cifs/link.c +--- linux-2.6.38.3/fs/cifs/link.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/cifs/link.c 2011-04-17 15:57:32.000000000 -0400 +@@ -577,7 +577,7 @@ symlink_exit: + + void cifs_put_link(struct dentry *direntry, struct nameidata *nd, void *cookie) + { +- char *p = nd_get_link(nd); ++ const char *p = nd_get_link(nd); + if (!IS_ERR(p)) + kfree(p); + } +diff -urNp linux-2.6.38.3/fs/compat_binfmt_elf.c linux-2.6.38.3/fs/compat_binfmt_elf.c +--- linux-2.6.38.3/fs/compat_binfmt_elf.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/compat_binfmt_elf.c 2011-04-17 15:57:32.000000000 -0400 +@@ -30,11 +30,13 @@ + #undef elf_phdr + #undef elf_shdr + #undef elf_note ++#undef elf_dyn + #undef elf_addr_t + #define elfhdr elf32_hdr + #define elf_phdr elf32_phdr + #define elf_shdr elf32_shdr + #define elf_note elf32_note ++#define elf_dyn Elf32_Dyn + #define elf_addr_t Elf32_Addr + + /* +diff -urNp linux-2.6.38.3/fs/compat.c linux-2.6.38.3/fs/compat.c +--- linux-2.6.38.3/fs/compat.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/compat.c 2011-04-17 15:57:32.000000000 -0400 +@@ -594,7 +594,7 @@ ssize_t compat_rw_copy_check_uvector(int + goto out; + + ret = -EINVAL; +- if (nr_segs > UIO_MAXIOV || nr_segs < 0) ++ if (nr_segs > UIO_MAXIOV) + goto out; + if (nr_segs > fast_segs) { + ret = -ENOMEM; +@@ -876,6 +876,7 @@ struct compat_old_linux_dirent { + + struct compat_readdir_callback { + struct compat_old_linux_dirent __user *dirent; ++ struct file * file; + int result; + }; + +@@ -893,6 +894,10 @@ static int compat_fillonedir(void *__buf + buf->result = -EOVERFLOW; + return -EOVERFLOW; + } ++ ++ if (!gr_acl_handle_filldir(buf->file, name, namlen, ino)) ++ return 0; ++ + buf->result++; + dirent = buf->dirent; + if (!access_ok(VERIFY_WRITE, dirent, +@@ -925,6 +930,7 @@ asmlinkage long compat_sys_old_readdir(u + + buf.result = 0; + buf.dirent = dirent; ++ buf.file = file; + + error = vfs_readdir(file, compat_fillonedir, &buf); + if (buf.result) +@@ -945,6 +951,7 @@ struct compat_linux_dirent { + struct compat_getdents_callback { + struct compat_linux_dirent __user *current_dir; + struct compat_linux_dirent __user *previous; ++ struct file * file; + int count; + int error; + }; +@@ -966,6 +973,10 @@ static int compat_filldir(void *__buf, c + buf->error = -EOVERFLOW; + return -EOVERFLOW; + } ++ ++ if (!gr_acl_handle_filldir(buf->file, name, namlen, ino)) ++ return 0; ++ + dirent = buf->previous; + if (dirent) { + if (__put_user(offset, &dirent->d_off)) +@@ -1013,6 +1024,7 @@ asmlinkage long compat_sys_getdents(unsi + buf.previous = NULL; + buf.count = count; + buf.error = 0; ++ buf.file = file; + + error = vfs_readdir(file, compat_filldir, &buf); + if (error >= 0) +@@ -1034,6 +1046,7 @@ out: + struct compat_getdents_callback64 { + struct linux_dirent64 __user *current_dir; + struct linux_dirent64 __user *previous; ++ struct file * file; + int count; + int error; + }; +@@ -1050,6 +1063,10 @@ static int compat_filldir64(void * __buf + buf->error = -EINVAL; /* only used if we fail.. */ + if (reclen > buf->count) + return -EINVAL; ++ ++ if (!gr_acl_handle_filldir(buf->file, name, namlen, ino)) ++ return 0; ++ + dirent = buf->previous; + + if (dirent) { +@@ -1101,6 +1118,7 @@ asmlinkage long compat_sys_getdents64(un + buf.previous = NULL; + buf.count = count; + buf.error = 0; ++ buf.file = file; + + error = vfs_readdir(file, compat_filldir64, &buf); + if (error >= 0) +@@ -1423,6 +1441,7 @@ static int compat_copy_strings(int argc, + + page = get_arg_page(bprm, pos, 1); + if (!page) { ++ /* We've exceed the stack rlimit. */ + ret = -E2BIG; + goto out; + } +@@ -1464,6 +1483,11 @@ int compat_do_execve(char * filename, + compat_uptr_t __user *envp, + struct pt_regs * regs) + { ++#ifdef CONFIG_GRKERNSEC ++ struct file *old_exec_file; ++ struct acl_subject_label *old_acl; ++ struct rlimit old_rlim[RLIM_NLIMITS]; ++#endif + struct linux_binprm *bprm; + struct file *file; + struct files_struct *displaced; +@@ -1500,6 +1524,19 @@ int compat_do_execve(char * filename, + bprm->filename = filename; + bprm->interp = filename; + ++ if (gr_process_user_ban()) { ++ retval = -EPERM; ++ goto out_file; ++ } ++ ++ gr_learn_resource(current, RLIMIT_NPROC, atomic_read(¤t->cred->user->processes), 1); ++ retval = -EAGAIN; ++ if (gr_handle_nproc()) ++ goto out_file; ++ retval = -EACCES; ++ if (!gr_acl_handle_execve(file->f_dentry, file->f_vfsmnt)) ++ goto out_file; ++ + retval = bprm_mm_init(bprm); + if (retval) + goto out_file; +@@ -1529,9 +1566,40 @@ int compat_do_execve(char * filename, + if (retval < 0) + goto out; + ++ if (!gr_tpe_allow(file)) { ++ retval = -EACCES; ++ goto out; ++ } ++ ++ if (gr_check_crash_exec(file)) { ++ retval = -EACCES; ++ goto out; ++ } ++ ++ gr_log_chroot_exec(file->f_dentry, file->f_vfsmnt); ++ ++ gr_handle_exec_args_compat(bprm, argv); ++ ++#ifdef CONFIG_GRKERNSEC ++ old_acl = current->acl; ++ memcpy(old_rlim, current->signal->rlim, sizeof(old_rlim)); ++ old_exec_file = current->exec_file; ++ get_file(file); ++ current->exec_file = file; ++#endif ++ ++ retval = gr_set_proc_label(file->f_dentry, file->f_vfsmnt, ++ bprm->unsafe & LSM_UNSAFE_SHARE); ++ if (retval < 0) ++ goto out_fail; ++ + retval = search_binary_handler(bprm, regs); + if (retval < 0) +- goto out; ++ goto out_fail; ++#ifdef CONFIG_GRKERNSEC ++ if (old_exec_file) ++ fput(old_exec_file); ++#endif + + /* execve succeeded */ + current->fs->in_exec = 0; +@@ -1542,6 +1610,14 @@ int compat_do_execve(char * filename, + put_files_struct(displaced); + return retval; + ++out_fail: ++#ifdef CONFIG_GRKERNSEC ++ current->acl = old_acl; ++ memcpy(current->signal->rlim, old_rlim, sizeof(old_rlim)); ++ fput(current->exec_file); ++ current->exec_file = old_exec_file; ++#endif ++ + out: + if (bprm->mm) { + acct_arg_size(bprm, 0); +diff -urNp linux-2.6.38.3/fs/compat_ioctl.c linux-2.6.38.3/fs/compat_ioctl.c +--- linux-2.6.38.3/fs/compat_ioctl.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/compat_ioctl.c 2011-04-17 15:57:32.000000000 -0400 +@@ -208,6 +208,8 @@ static int do_video_set_spu_palette(unsi + + err = get_user(palp, &up->palette); + err |= get_user(length, &up->length); ++ if (err) ++ return -EFAULT; + + up_native = compat_alloc_user_space(sizeof(struct video_spu_palette)); + err = put_user(compat_ptr(palp), &up_native->palette); +@@ -1638,8 +1640,8 @@ asmlinkage long compat_sys_ioctl(unsigne + static int __init init_sys32_ioctl_cmp(const void *p, const void *q) + { + unsigned int a, b; +- a = *(unsigned int *)p; +- b = *(unsigned int *)q; ++ a = *(const unsigned int *)p; ++ b = *(const unsigned int *)q; + if (a > b) + return 1; + if (a < b) +diff -urNp linux-2.6.38.3/fs/dcache.c linux-2.6.38.3/fs/dcache.c +--- linux-2.6.38.3/fs/dcache.c 2011-03-28 17:42:40.000000000 -0400 ++++ linux-2.6.38.3/fs/dcache.c 2011-04-17 15:57:32.000000000 -0400 +@@ -3092,7 +3092,7 @@ void __init vfs_caches_init(unsigned lon + mempages -= reserve; + + names_cachep = kmem_cache_create("names_cache", PATH_MAX, 0, +- SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL); ++ SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_USERCOPY, NULL); + + dcache_init(); + inode_init(); +diff -urNp linux-2.6.38.3/fs/debugfs/inode.c linux-2.6.38.3/fs/debugfs/inode.c +--- linux-2.6.38.3/fs/debugfs/inode.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/debugfs/inode.c 2011-04-17 15:57:32.000000000 -0400 +@@ -130,7 +130,7 @@ static inline int debugfs_positive(struc + + static int debug_fill_super(struct super_block *sb, void *data, int silent) + { +- static struct tree_descr debug_files[] = {{""}}; ++ static struct tree_descr debug_files[] = {{"", NULL, 0}}; + + return simple_fill_super(sb, DEBUGFS_MAGIC, debug_files); + } +diff -urNp linux-2.6.38.3/fs/dlm/lockspace.c linux-2.6.38.3/fs/dlm/lockspace.c +--- linux-2.6.38.3/fs/dlm/lockspace.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/dlm/lockspace.c 2011-04-17 15:57:32.000000000 -0400 +@@ -200,7 +200,7 @@ static int dlm_uevent(struct kset *kset, + return 0; + } + +-static struct kset_uevent_ops dlm_uevent_ops = { ++static const struct kset_uevent_ops dlm_uevent_ops = { + .uevent = dlm_uevent, + }; + +diff -urNp linux-2.6.38.3/fs/ecryptfs/inode.c linux-2.6.38.3/fs/ecryptfs/inode.c +--- linux-2.6.38.3/fs/ecryptfs/inode.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/ecryptfs/inode.c 2011-04-17 15:57:32.000000000 -0400 +@@ -658,7 +658,7 @@ static int ecryptfs_readlink_lower(struc + old_fs = get_fs(); + set_fs(get_ds()); + rc = lower_dentry->d_inode->i_op->readlink(lower_dentry, +- (char __user *)lower_buf, ++ (__force char __user *)lower_buf, + lower_bufsiz); + set_fs(old_fs); + if (rc < 0) +@@ -704,7 +704,7 @@ static void *ecryptfs_follow_link(struct + } + old_fs = get_fs(); + set_fs(get_ds()); +- rc = dentry->d_inode->i_op->readlink(dentry, (char __user *)buf, len); ++ rc = dentry->d_inode->i_op->readlink(dentry, (__force char __user *)buf, len); + set_fs(old_fs); + if (rc < 0) { + kfree(buf); +@@ -719,7 +719,7 @@ out: + static void + ecryptfs_put_link(struct dentry *dentry, struct nameidata *nd, void *ptr) + { +- char *buf = nd_get_link(nd); ++ const char *buf = nd_get_link(nd); + if (!IS_ERR(buf)) { + /* Free the char* */ + kfree(buf); +diff -urNp linux-2.6.38.3/fs/ecryptfs/miscdev.c linux-2.6.38.3/fs/ecryptfs/miscdev.c +--- linux-2.6.38.3/fs/ecryptfs/miscdev.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/ecryptfs/miscdev.c 2011-04-17 15:57:32.000000000 -0400 +@@ -328,7 +328,7 @@ check_list: + goto out_unlock_msg_ctx; + i = 5; + if (msg_ctx->msg) { +- if (copy_to_user(&buf[i], packet_length, packet_length_size)) ++ if (packet_length_size > sizeof(packet_length) || copy_to_user(&buf[i], packet_length, packet_length_size)) + goto out_unlock_msg_ctx; + i += packet_length_size; + if (copy_to_user(&buf[i], msg_ctx->msg, msg_ctx->msg_size)) +diff -urNp linux-2.6.38.3/fs/exec.c linux-2.6.38.3/fs/exec.c +--- linux-2.6.38.3/fs/exec.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/exec.c 2011-04-17 15:57:32.000000000 -0400 +@@ -55,12 +55,24 @@ + #include <linux/fs_struct.h> + #include <linux/pipe_fs_i.h> + #include <linux/oom.h> ++#include <linux/random.h> ++#include <linux/seq_file.h> ++ ++#ifdef CONFIG_PAX_REFCOUNT ++#include <linux/kallsyms.h> ++#include <linux/kdebug.h> ++#endif + + #include <asm/uaccess.h> + #include <asm/mmu_context.h> + #include <asm/tlb.h> + #include "internal.h" + ++#ifdef CONFIG_PAX_HOOK_ACL_FLAGS ++void (*pax_set_initial_flags_func)(struct linux_binprm *bprm); ++EXPORT_SYMBOL(pax_set_initial_flags_func); ++#endif ++ + int core_uses_pid; + char core_pattern[CORENAME_MAX_SIZE] = "core"; + unsigned int core_pipe_limit; +@@ -120,7 +132,7 @@ SYSCALL_DEFINE1(uselib, const char __use + goto out; + + file = do_filp_open(AT_FDCWD, tmp, +- O_LARGEFILE | O_RDONLY | __FMODE_EXEC, 0, ++ O_LARGEFILE | O_RDONLY | __FMODE_EXEC | FMODE_GREXEC, 0, + MAY_READ | MAY_EXEC | MAY_OPEN); + putname(tmp); + error = PTR_ERR(file); +@@ -187,18 +199,10 @@ struct page *get_arg_page(struct linux_b + int write) + { + struct page *page; +- int ret; + +-#ifdef CONFIG_STACK_GROWSUP +- if (write) { +- ret = expand_stack_downwards(bprm->vma, pos); +- if (ret < 0) +- return NULL; +- } +-#endif +- ret = get_user_pages(current, bprm->mm, pos, +- 1, write, 1, &page, NULL); +- if (ret <= 0) ++ if (0 > expand_stack_downwards(bprm->vma, pos)) ++ return NULL; ++ if (0 >= get_user_pages(current, bprm->mm, pos, 1, write, 1, &page, NULL)) + return NULL; + + if (write) { +@@ -273,6 +277,11 @@ static int __bprm_mm_init(struct linux_b + vma->vm_end = STACK_TOP_MAX; + vma->vm_start = vma->vm_end - PAGE_SIZE; + vma->vm_flags = VM_STACK_FLAGS | VM_STACK_INCOMPLETE_SETUP; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ vma->vm_flags &= ~(VM_EXEC | VM_MAYEXEC); ++#endif ++ + vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); + INIT_LIST_HEAD(&vma->anon_vma_chain); + +@@ -287,6 +296,12 @@ static int __bprm_mm_init(struct linux_b + mm->stack_vm = mm->total_vm = 1; + up_write(&mm->mmap_sem); + bprm->p = vma->vm_end - sizeof(void *); ++ ++#ifdef CONFIG_PAX_RANDUSTACK ++ if (randomize_va_space) ++ bprm->p ^= (pax_get_random_long() & ~15) & ~PAGE_MASK; ++#endif ++ + return 0; + err: + up_write(&mm->mmap_sem); +@@ -522,7 +537,7 @@ int copy_strings_kernel(int argc, const + int r; + mm_segment_t oldfs = get_fs(); + set_fs(KERNEL_DS); +- r = copy_strings(argc, (const char __user *const __user *)argv, bprm); ++ r = copy_strings(argc, (__force const char __user *const __user *)argv, bprm); + set_fs(oldfs); + return r; + } +@@ -552,7 +567,8 @@ static int shift_arg_pages(struct vm_are + unsigned long new_end = old_end - shift; + struct mmu_gather *tlb; + +- BUG_ON(new_start > new_end); ++ if (new_start >= new_end || new_start < mmap_min_addr) ++ return -ENOMEM; + + /* + * ensure there are no vmas between where we want to go +@@ -561,6 +577,10 @@ static int shift_arg_pages(struct vm_are + if (vma != find_vma(mm, new_start)) + return -EFAULT; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ BUG_ON(pax_find_mirror_vma(vma)); ++#endif ++ + /* + * cover the whole range: [new_start, old_end) + */ +@@ -641,10 +661,6 @@ int setup_arg_pages(struct linux_binprm + stack_top = arch_align_stack(stack_top); + stack_top = PAGE_ALIGN(stack_top); + +- if (unlikely(stack_top < mmap_min_addr) || +- unlikely(vma->vm_end - vma->vm_start >= stack_top - mmap_min_addr)) +- return -ENOMEM; +- + stack_shift = vma->vm_end - stack_top; + + bprm->p -= stack_shift; +@@ -656,8 +672,28 @@ int setup_arg_pages(struct linux_binprm + bprm->exec -= stack_shift; + + down_write(&mm->mmap_sem); ++ ++ /* Move stack pages down in memory. */ ++ if (stack_shift) { ++ ret = shift_arg_pages(vma, stack_shift); ++ if (ret) ++ goto out_unlock; ++ } ++ + vm_flags = VM_STACK_FLAGS; + ++#if defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) ++ if (mm->pax_flags & (MF_PAX_PAGEEXEC | MF_PAX_SEGMEXEC)) { ++ vm_flags &= ~VM_EXEC; ++ ++#ifdef CONFIG_PAX_MPROTECT ++ if (mm->pax_flags & MF_PAX_MPROTECT) ++ vm_flags &= ~VM_MAYEXEC; ++#endif ++ ++ } ++#endif ++ + /* + * Adjust stack execute permissions; explicitly enable for + * EXSTACK_ENABLE_X, disable for EXSTACK_DISABLE_X and leave alone +@@ -676,13 +712,6 @@ int setup_arg_pages(struct linux_binprm + goto out_unlock; + BUG_ON(prev != vma); + +- /* Move stack pages down in memory. */ +- if (stack_shift) { +- ret = shift_arg_pages(vma, stack_shift); +- if (ret) +- goto out_unlock; +- } +- + /* mprotect_fixup is overkill to remove the temporary stack flags */ + vma->vm_flags &= ~VM_STACK_INCOMPLETE_SETUP; + +@@ -723,7 +752,7 @@ struct file *open_exec(const char *name) + int err; + + file = do_filp_open(AT_FDCWD, name, +- O_LARGEFILE | O_RDONLY | __FMODE_EXEC, 0, ++ O_LARGEFILE | O_RDONLY | __FMODE_EXEC | FMODE_GREXEC, 0, + MAY_EXEC | MAY_OPEN); + if (IS_ERR(file)) + goto out; +@@ -760,7 +789,7 @@ int kernel_read(struct file *file, loff_ + old_fs = get_fs(); + set_fs(get_ds()); + /* The cast to a user pointer is valid due to the set_fs() */ +- result = vfs_read(file, (void __user *)addr, count, &pos); ++ result = vfs_read(file, (__force void __user *)addr, count, &pos); + set_fs(old_fs); + return result; + } +@@ -1182,7 +1211,7 @@ int check_unsafe_exec(struct linux_binpr + } + rcu_read_unlock(); + +- if (p->fs->users > n_fs) { ++ if (atomic_read(&p->fs->users) > n_fs) { + bprm->unsafe |= LSM_UNSAFE_SHARE; + } else { + res = -EAGAIN; +@@ -1378,6 +1407,11 @@ int do_execve(const char * filename, + const char __user *const __user *envp, + struct pt_regs * regs) + { ++#ifdef CONFIG_GRKERNSEC ++ struct file *old_exec_file; ++ struct acl_subject_label *old_acl; ++ struct rlimit old_rlim[RLIM_NLIMITS]; ++#endif + struct linux_binprm *bprm; + struct file *file; + struct files_struct *displaced; +@@ -1414,6 +1448,23 @@ int do_execve(const char * filename, + bprm->filename = filename; + bprm->interp = filename; + ++ if (gr_process_user_ban()) { ++ retval = -EPERM; ++ goto out_file; ++ } ++ ++ gr_learn_resource(current, RLIMIT_NPROC, atomic_read(¤t->cred->user->processes), 1); ++ ++ if (gr_handle_nproc()) { ++ retval = -EAGAIN; ++ goto out_file; ++ } ++ ++ if (!gr_acl_handle_execve(file->f_dentry, file->f_vfsmnt)) { ++ retval = -EACCES; ++ goto out_file; ++ } ++ + retval = bprm_mm_init(bprm); + if (retval) + goto out_file; +@@ -1443,9 +1494,40 @@ int do_execve(const char * filename, + if (retval < 0) + goto out; + ++ if (!gr_tpe_allow(file)) { ++ retval = -EACCES; ++ goto out; ++ } ++ ++ if (gr_check_crash_exec(file)) { ++ retval = -EACCES; ++ goto out; ++ } ++ ++ gr_log_chroot_exec(file->f_dentry, file->f_vfsmnt); ++ ++ gr_handle_exec_args(bprm, argv); ++ ++#ifdef CONFIG_GRKERNSEC ++ old_acl = current->acl; ++ memcpy(old_rlim, current->signal->rlim, sizeof(old_rlim)); ++ old_exec_file = current->exec_file; ++ get_file(file); ++ current->exec_file = file; ++#endif ++ ++ retval = gr_set_proc_label(file->f_dentry, file->f_vfsmnt, ++ bprm->unsafe & LSM_UNSAFE_SHARE); ++ if (retval < 0) ++ goto out_fail; ++ + retval = search_binary_handler(bprm,regs); + if (retval < 0) +- goto out; ++ goto out_fail; ++#ifdef CONFIG_GRKERNSEC ++ if (old_exec_file) ++ fput(old_exec_file); ++#endif + + /* execve succeeded */ + current->fs->in_exec = 0; +@@ -1456,6 +1538,14 @@ int do_execve(const char * filename, + put_files_struct(displaced); + return retval; + ++out_fail: ++#ifdef CONFIG_GRKERNSEC ++ current->acl = old_acl; ++ memcpy(current->signal->rlim, old_rlim, sizeof(old_rlim)); ++ fput(current->exec_file); ++ current->exec_file = old_exec_file; ++#endif ++ + out: + if (bprm->mm) { + acct_arg_size(bprm, 0); +@@ -1642,6 +1732,208 @@ out: + return ispipe; + } + ++int pax_check_flags(unsigned long *flags) ++{ ++ int retval = 0; ++ ++#if !defined(CONFIG_X86_32) || !defined(CONFIG_PAX_SEGMEXEC) ++ if (*flags & MF_PAX_SEGMEXEC) ++ { ++ *flags &= ~MF_PAX_SEGMEXEC; ++ retval = -EINVAL; ++ } ++#endif ++ ++ if ((*flags & MF_PAX_PAGEEXEC) ++ ++#ifdef CONFIG_PAX_PAGEEXEC ++ && (*flags & MF_PAX_SEGMEXEC) ++#endif ++ ++ ) ++ { ++ *flags &= ~MF_PAX_PAGEEXEC; ++ retval = -EINVAL; ++ } ++ ++ if ((*flags & MF_PAX_MPROTECT) ++ ++#ifdef CONFIG_PAX_MPROTECT ++ && !(*flags & (MF_PAX_PAGEEXEC | MF_PAX_SEGMEXEC)) ++#endif ++ ++ ) ++ { ++ *flags &= ~MF_PAX_MPROTECT; ++ retval = -EINVAL; ++ } ++ ++ if ((*flags & MF_PAX_EMUTRAMP) ++ ++#ifdef CONFIG_PAX_EMUTRAMP ++ && !(*flags & (MF_PAX_PAGEEXEC | MF_PAX_SEGMEXEC)) ++#endif ++ ++ ) ++ { ++ *flags &= ~MF_PAX_EMUTRAMP; ++ retval = -EINVAL; ++ } ++ ++ return retval; ++} ++ ++EXPORT_SYMBOL(pax_check_flags); ++ ++#if defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) ++void pax_report_fault(struct pt_regs *regs, void *pc, void *sp) ++{ ++ struct task_struct *tsk = current; ++ struct mm_struct *mm = current->mm; ++ char *buffer_exec = (char *)__get_free_page(GFP_KERNEL); ++ char *buffer_fault = (char *)__get_free_page(GFP_KERNEL); ++ char *path_exec = NULL; ++ char *path_fault = NULL; ++ unsigned long start = 0UL, end = 0UL, offset = 0UL; ++ ++ if (buffer_exec && buffer_fault) { ++ struct vm_area_struct *vma, *vma_exec = NULL, *vma_fault = NULL; ++ ++ down_read(&mm->mmap_sem); ++ vma = mm->mmap; ++ while (vma && (!vma_exec || !vma_fault)) { ++ if ((vma->vm_flags & VM_EXECUTABLE) && vma->vm_file) ++ vma_exec = vma; ++ if (vma->vm_start <= (unsigned long)pc && (unsigned long)pc < vma->vm_end) ++ vma_fault = vma; ++ vma = vma->vm_next; ++ } ++ if (vma_exec) { ++ path_exec = d_path(&vma_exec->vm_file->f_path, buffer_exec, PAGE_SIZE); ++ if (IS_ERR(path_exec)) ++ path_exec = "<path too long>"; ++ else { ++ path_exec = mangle_path(buffer_exec, path_exec, "\t\n\\"); ++ if (path_exec) { ++ *path_exec = 0; ++ path_exec = buffer_exec; ++ } else ++ path_exec = "<path too long>"; ++ } ++ } ++ if (vma_fault) { ++ start = vma_fault->vm_start; ++ end = vma_fault->vm_end; ++ offset = vma_fault->vm_pgoff << PAGE_SHIFT; ++ if (vma_fault->vm_file) { ++ path_fault = d_path(&vma_fault->vm_file->f_path, buffer_fault, PAGE_SIZE); ++ if (IS_ERR(path_fault)) ++ path_fault = "<path too long>"; ++ else { ++ path_fault = mangle_path(buffer_fault, path_fault, "\t\n\\"); ++ if (path_fault) { ++ *path_fault = 0; ++ path_fault = buffer_fault; ++ } else ++ path_fault = "<path too long>"; ++ } ++ } else ++ path_fault = "<anonymous mapping>"; ++ } ++ up_read(&mm->mmap_sem); ++ } ++ if (tsk->signal->curr_ip) ++ printk(KERN_ERR "PAX: From %pI4: execution attempt in: %s, %08lx-%08lx %08lx\n", &tsk->signal->curr_ip, path_fault, start, end, offset); ++ else ++ printk(KERN_ERR "PAX: execution attempt in: %s, %08lx-%08lx %08lx\n", path_fault, start, end, offset); ++ printk(KERN_ERR "PAX: terminating task: %s(%s):%d, uid/euid: %u/%u, " ++ "PC: %p, SP: %p\n", path_exec, tsk->comm, task_pid_nr(tsk), ++ task_uid(tsk), task_euid(tsk), pc, sp); ++ free_page((unsigned long)buffer_exec); ++ free_page((unsigned long)buffer_fault); ++ pax_report_insns(pc, sp); ++ do_coredump(SIGKILL, SIGKILL, regs); ++} ++#endif ++ ++#ifdef CONFIG_PAX_REFCOUNT ++void pax_report_refcount_overflow(struct pt_regs *regs) ++{ ++ if (current->signal->curr_ip) ++ printk(KERN_ERR "PAX: From %pI4: refcount overflow detected in: %s:%d, uid/euid: %u/%u\n", ++ ¤t->signal->curr_ip, current->comm, task_pid_nr(current), current_uid(), current_euid()); ++ else ++ printk(KERN_ERR "PAX: refcount overflow detected in: %s:%d, uid/euid: %u/%u\n", ++ current->comm, task_pid_nr(current), current_uid(), current_euid()); ++ print_symbol(KERN_ERR "PAX: refcount overflow occured at: %s\n", instruction_pointer(regs)); ++ show_regs(regs); ++ force_sig_info(SIGKILL, SEND_SIG_FORCED, current); ++} ++#endif ++ ++#ifdef CONFIG_PAX_USERCOPY ++/* 0: not at all, 1: fully, 2: fully inside frame, -1: partially (implies an error) */ ++int object_is_on_stack(const void *obj, unsigned long len) ++{ ++ const void * const stack = task_stack_page(current); ++ const void * const stackend = stack + THREAD_SIZE; ++ ++#if defined(CONFIG_FRAME_POINTER) && defined(CONFIG_X86) ++ const void *frame = NULL; ++ const void *oldframe; ++#endif ++ ++ if (obj + len < obj) ++ return -1; ++ ++ if (obj + len <= stack || stackend <= obj) ++ return 0; ++ ++ if (obj < stack || stackend < obj + len) ++ return -1; ++ ++#if defined(CONFIG_FRAME_POINTER) && defined(CONFIG_X86) ++ oldframe = __builtin_frame_address(1); ++ if (oldframe) ++ frame = __builtin_frame_address(2); ++ /* ++ low ----------------------------------------------> high ++ [saved bp][saved ip][args][local vars][saved bp][saved ip] ++ ^----------------^ ++ allow copies only within here ++ */ ++ while (stack <= frame && frame < stackend) { ++ /* if obj + len extends past the last frame, this ++ check won't pass and the next frame will be 0, ++ causing us to bail out and correctly report ++ the copy as invalid ++ */ ++ if (obj + len <= frame) ++ return obj >= oldframe + 2 * sizeof(void *) ? 2 : -1; ++ oldframe = frame; ++ frame = *(const void * const *)frame; ++ } ++ return -1; ++#else ++ return 1; ++#endif ++} ++ ++ ++void pax_report_usercopy(const void *ptr, unsigned long len, bool to, const char *type) ++{ ++ if (current->signal->curr_ip) ++ printk(KERN_ERR "PAX: From %pI4: kernel memory %s attempt detected %s %p (%s) (%lu bytes)\n", ++ ¤t->signal->curr_ip, to ? "leak" : "overwrite", to ? "from" : "to", ptr, type ? : "unknown", len); ++ else ++ printk(KERN_ERR "PAX: kernel memory %s attempt detected %s %p (%s) (%lu bytes)\n", ++ to ? "leak" : "overwrite", to ? "from" : "to", ptr, type ? : "unknown", len); ++ dump_stack(); ++ gr_handle_kernel_exploit(); ++ do_group_exit(SIGKILL); ++} ++#endif ++ + static int zap_process(struct task_struct *start, int exit_code) + { + struct task_struct *t; +@@ -1852,17 +2144,17 @@ static void wait_for_dump_helpers(struct + pipe = file->f_path.dentry->d_inode->i_pipe; + + pipe_lock(pipe); +- pipe->readers++; +- pipe->writers--; ++ atomic_inc(&pipe->readers); ++ atomic_dec(&pipe->writers); + +- while ((pipe->readers > 1) && (!signal_pending(current))) { ++ while ((atomic_read(&pipe->readers) > 1) && (!signal_pending(current))) { + wake_up_interruptible_sync(&pipe->wait); + kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN); + pipe_wait(pipe); + } + +- pipe->readers--; +- pipe->writers++; ++ atomic_dec(&pipe->readers); ++ atomic_inc(&pipe->writers); + pipe_unlock(pipe); + + } +@@ -1938,6 +2230,9 @@ void do_coredump(long signr, int exit_co + + audit_core_dumps(signr); + ++ if (signr == SIGSEGV || signr == SIGBUS || signr == SIGKILL || signr == SIGILL) ++ gr_handle_brute_attach(current, cprm.mm_flags); ++ + binfmt = mm->binfmt; + if (!binfmt || !binfmt->core_dump) + goto fail; +@@ -1978,6 +2273,8 @@ void do_coredump(long signr, int exit_co + goto fail_corename; + } + ++ gr_learn_resource(current, RLIMIT_CORE, binfmt->min_coredump, 1); ++ + if (ispipe) { + int dump_count; + char **helper_argv; +diff -urNp linux-2.6.38.3/fs/ext2/balloc.c linux-2.6.38.3/fs/ext2/balloc.c +--- linux-2.6.38.3/fs/ext2/balloc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/ext2/balloc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1192,7 +1192,7 @@ static int ext2_has_free_blocks(struct e + + free_blocks = percpu_counter_read_positive(&sbi->s_freeblocks_counter); + root_blocks = le32_to_cpu(sbi->s_es->s_r_blocks_count); +- if (free_blocks < root_blocks + 1 && !capable(CAP_SYS_RESOURCE) && ++ if (free_blocks < root_blocks + 1 && !capable_nolog(CAP_SYS_RESOURCE) && + sbi->s_resuid != current_fsuid() && + (sbi->s_resgid == 0 || !in_group_p (sbi->s_resgid))) { + return 0; +diff -urNp linux-2.6.38.3/fs/ext2/xattr.c linux-2.6.38.3/fs/ext2/xattr.c +--- linux-2.6.38.3/fs/ext2/xattr.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/ext2/xattr.c 2011-04-17 15:57:32.000000000 -0400 +@@ -86,8 +86,8 @@ + printk("\n"); \ + } while (0) + #else +-# define ea_idebug(f...) +-# define ea_bdebug(f...) ++# define ea_idebug(inode, f...) do {} while (0) ++# define ea_bdebug(bh, f...) do {} while (0) + #endif + + static int ext2_xattr_set2(struct inode *, struct buffer_head *, +diff -urNp linux-2.6.38.3/fs/ext3/balloc.c linux-2.6.38.3/fs/ext3/balloc.c +--- linux-2.6.38.3/fs/ext3/balloc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/ext3/balloc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1441,7 +1441,7 @@ static int ext3_has_free_blocks(struct e + + free_blocks = percpu_counter_read_positive(&sbi->s_freeblocks_counter); + root_blocks = le32_to_cpu(sbi->s_es->s_r_blocks_count); +- if (free_blocks < root_blocks + 1 && !capable(CAP_SYS_RESOURCE) && ++ if (free_blocks < root_blocks + 1 && !capable_nolog(CAP_SYS_RESOURCE) && + sbi->s_resuid != current_fsuid() && + (sbi->s_resgid == 0 || !in_group_p (sbi->s_resgid))) { + return 0; +diff -urNp linux-2.6.38.3/fs/ext3/namei.c linux-2.6.38.3/fs/ext3/namei.c +--- linux-2.6.38.3/fs/ext3/namei.c 2011-03-23 17:20:08.000000000 -0400 ++++ linux-2.6.38.3/fs/ext3/namei.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1159,7 +1159,7 @@ static struct ext3_dir_entry_2 *do_split + char *data1 = (*bh)->b_data, *data2; + unsigned split, move, size; + struct ext3_dir_entry_2 *de = NULL, *de2; +- int err = 0, i; ++ int i, err = 0; + + bh2 = ext3_append (handle, dir, &newblock, &err); + if (!(bh2)) { +diff -urNp linux-2.6.38.3/fs/ext3/xattr.c linux-2.6.38.3/fs/ext3/xattr.c +--- linux-2.6.38.3/fs/ext3/xattr.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/ext3/xattr.c 2011-04-17 15:57:32.000000000 -0400 +@@ -89,8 +89,8 @@ + printk("\n"); \ + } while (0) + #else +-# define ea_idebug(f...) +-# define ea_bdebug(f...) ++# define ea_idebug(f...) do {} while (0) ++# define ea_bdebug(f...) do {} while (0) + #endif + + static void ext3_xattr_cache_insert(struct buffer_head *); +diff -urNp linux-2.6.38.3/fs/ext4/balloc.c linux-2.6.38.3/fs/ext4/balloc.c +--- linux-2.6.38.3/fs/ext4/balloc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/ext4/balloc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -519,7 +519,7 @@ static int ext4_has_free_blocks(struct e + /* Hm, nope. Are (enough) root reserved blocks available? */ + if (sbi->s_resuid == current_fsuid() || + ((sbi->s_resgid != 0) && in_group_p(sbi->s_resgid)) || +- capable(CAP_SYS_RESOURCE)) { ++ capable_nolog(CAP_SYS_RESOURCE)) { + if (free_blocks >= (nblocks + dirty_blocks)) + return 1; + } +diff -urNp linux-2.6.38.3/fs/ext4/ext4.h linux-2.6.38.3/fs/ext4/ext4.h +--- linux-2.6.38.3/fs/ext4/ext4.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/ext4/ext4.h 2011-04-17 15:57:32.000000000 -0400 +@@ -1166,19 +1166,19 @@ struct ext4_sb_info { + unsigned long s_mb_last_start; + + /* stats for buddy allocator */ +- atomic_t s_bal_reqs; /* number of reqs with len > 1 */ +- atomic_t s_bal_success; /* we found long enough chunks */ +- atomic_t s_bal_allocated; /* in blocks */ +- atomic_t s_bal_ex_scanned; /* total extents scanned */ +- atomic_t s_bal_goals; /* goal hits */ +- atomic_t s_bal_breaks; /* too long searches */ +- atomic_t s_bal_2orders; /* 2^order hits */ ++ atomic_unchecked_t s_bal_reqs; /* number of reqs with len > 1 */ ++ atomic_unchecked_t s_bal_success; /* we found long enough chunks */ ++ atomic_unchecked_t s_bal_allocated; /* in blocks */ ++ atomic_unchecked_t s_bal_ex_scanned; /* total extents scanned */ ++ atomic_unchecked_t s_bal_goals; /* goal hits */ ++ atomic_unchecked_t s_bal_breaks; /* too long searches */ ++ atomic_unchecked_t s_bal_2orders; /* 2^order hits */ + spinlock_t s_bal_lock; + unsigned long s_mb_buddies_generated; + unsigned long long s_mb_generation_time; +- atomic_t s_mb_lost_chunks; +- atomic_t s_mb_preallocated; +- atomic_t s_mb_discarded; ++ atomic_unchecked_t s_mb_lost_chunks; ++ atomic_unchecked_t s_mb_preallocated; ++ atomic_unchecked_t s_mb_discarded; + atomic_t s_lock_busy; + + /* locality groups */ +diff -urNp linux-2.6.38.3/fs/ext4/mballoc.c linux-2.6.38.3/fs/ext4/mballoc.c +--- linux-2.6.38.3/fs/ext4/mballoc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/ext4/mballoc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1846,7 +1846,7 @@ void ext4_mb_simple_scan_group(struct ex + BUG_ON(ac->ac_b_ex.fe_len != ac->ac_g_ex.fe_len); + + if (EXT4_SB(sb)->s_mb_stats) +- atomic_inc(&EXT4_SB(sb)->s_bal_2orders); ++ atomic_inc_unchecked(&EXT4_SB(sb)->s_bal_2orders); + + break; + } +@@ -2140,7 +2140,7 @@ repeat: + ac->ac_status = AC_STATUS_CONTINUE; + ac->ac_flags |= EXT4_MB_HINT_FIRST; + cr = 3; +- atomic_inc(&sbi->s_mb_lost_chunks); ++ atomic_inc_unchecked(&sbi->s_mb_lost_chunks); + goto repeat; + } + } +@@ -2606,25 +2606,25 @@ int ext4_mb_release(struct super_block * + if (sbi->s_mb_stats) { + printk(KERN_INFO + "EXT4-fs: mballoc: %u blocks %u reqs (%u success)\n", +- atomic_read(&sbi->s_bal_allocated), +- atomic_read(&sbi->s_bal_reqs), +- atomic_read(&sbi->s_bal_success)); ++ atomic_read_unchecked(&sbi->s_bal_allocated), ++ atomic_read_unchecked(&sbi->s_bal_reqs), ++ atomic_read_unchecked(&sbi->s_bal_success)); + printk(KERN_INFO + "EXT4-fs: mballoc: %u extents scanned, %u goal hits, " + "%u 2^N hits, %u breaks, %u lost\n", +- atomic_read(&sbi->s_bal_ex_scanned), +- atomic_read(&sbi->s_bal_goals), +- atomic_read(&sbi->s_bal_2orders), +- atomic_read(&sbi->s_bal_breaks), +- atomic_read(&sbi->s_mb_lost_chunks)); ++ atomic_read_unchecked(&sbi->s_bal_ex_scanned), ++ atomic_read_unchecked(&sbi->s_bal_goals), ++ atomic_read_unchecked(&sbi->s_bal_2orders), ++ atomic_read_unchecked(&sbi->s_bal_breaks), ++ atomic_read_unchecked(&sbi->s_mb_lost_chunks)); + printk(KERN_INFO + "EXT4-fs: mballoc: %lu generated and it took %Lu\n", + sbi->s_mb_buddies_generated++, + sbi->s_mb_generation_time); + printk(KERN_INFO + "EXT4-fs: mballoc: %u preallocated, %u discarded\n", +- atomic_read(&sbi->s_mb_preallocated), +- atomic_read(&sbi->s_mb_discarded)); ++ atomic_read_unchecked(&sbi->s_mb_preallocated), ++ atomic_read_unchecked(&sbi->s_mb_discarded)); + } + + free_percpu(sbi->s_locality_groups); +@@ -3100,16 +3100,16 @@ static void ext4_mb_collect_stats(struct + struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb); + + if (sbi->s_mb_stats && ac->ac_g_ex.fe_len > 1) { +- atomic_inc(&sbi->s_bal_reqs); +- atomic_add(ac->ac_b_ex.fe_len, &sbi->s_bal_allocated); ++ atomic_inc_unchecked(&sbi->s_bal_reqs); ++ atomic_add_unchecked(ac->ac_b_ex.fe_len, &sbi->s_bal_allocated); + if (ac->ac_b_ex.fe_len >= ac->ac_o_ex.fe_len) +- atomic_inc(&sbi->s_bal_success); +- atomic_add(ac->ac_found, &sbi->s_bal_ex_scanned); ++ atomic_inc_unchecked(&sbi->s_bal_success); ++ atomic_add_unchecked(ac->ac_found, &sbi->s_bal_ex_scanned); + if (ac->ac_g_ex.fe_start == ac->ac_b_ex.fe_start && + ac->ac_g_ex.fe_group == ac->ac_b_ex.fe_group) +- atomic_inc(&sbi->s_bal_goals); ++ atomic_inc_unchecked(&sbi->s_bal_goals); + if (ac->ac_found > sbi->s_mb_max_to_scan) +- atomic_inc(&sbi->s_bal_breaks); ++ atomic_inc_unchecked(&sbi->s_bal_breaks); + } + + if (ac->ac_op == EXT4_MB_HISTORY_ALLOC) +@@ -3507,7 +3507,7 @@ ext4_mb_new_inode_pa(struct ext4_allocat + trace_ext4_mb_new_inode_pa(ac, pa); + + ext4_mb_use_inode_pa(ac, pa); +- atomic_add(pa->pa_free, &EXT4_SB(sb)->s_mb_preallocated); ++ atomic_add_unchecked(pa->pa_free, &EXT4_SB(sb)->s_mb_preallocated); + + ei = EXT4_I(ac->ac_inode); + grp = ext4_get_group_info(sb, ac->ac_b_ex.fe_group); +@@ -3567,7 +3567,7 @@ ext4_mb_new_group_pa(struct ext4_allocat + trace_ext4_mb_new_group_pa(ac, pa); + + ext4_mb_use_group_pa(ac, pa); +- atomic_add(pa->pa_free, &EXT4_SB(sb)->s_mb_preallocated); ++ atomic_add_unchecked(pa->pa_free, &EXT4_SB(sb)->s_mb_preallocated); + + grp = ext4_get_group_info(sb, ac->ac_b_ex.fe_group); + lg = ac->ac_lg; +@@ -3654,7 +3654,7 @@ ext4_mb_release_inode_pa(struct ext4_bud + * from the bitmap and continue. + */ + } +- atomic_add(free, &sbi->s_mb_discarded); ++ atomic_add_unchecked(free, &sbi->s_mb_discarded); + + return err; + } +@@ -3672,7 +3672,7 @@ ext4_mb_release_group_pa(struct ext4_bud + ext4_get_group_no_and_offset(sb, pa->pa_pstart, &group, &bit); + BUG_ON(group != e4b->bd_group && pa->pa_len != 0); + mb_free_blocks(pa->pa_inode, e4b, bit, pa->pa_len); +- atomic_add(pa->pa_len, &EXT4_SB(sb)->s_mb_discarded); ++ atomic_add_unchecked(pa->pa_len, &EXT4_SB(sb)->s_mb_discarded); + trace_ext4_mballoc_discard(sb, NULL, group, bit, pa->pa_len); + + return 0; +diff -urNp linux-2.6.38.3/fs/ext4/namei.c linux-2.6.38.3/fs/ext4/namei.c +--- linux-2.6.38.3/fs/ext4/namei.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/ext4/namei.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1161,7 +1161,7 @@ static struct ext4_dir_entry_2 *do_split + char *data1 = (*bh)->b_data, *data2; + unsigned split, move, size; + struct ext4_dir_entry_2 *de = NULL, *de2; +- int err = 0, i; ++ int i, err = 0; + + bh2 = ext4_append (handle, dir, &newblock, &err); + if (!(bh2)) { +diff -urNp linux-2.6.38.3/fs/ext4/xattr.c linux-2.6.38.3/fs/ext4/xattr.c +--- linux-2.6.38.3/fs/ext4/xattr.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/ext4/xattr.c 2011-04-17 15:57:32.000000000 -0400 +@@ -82,8 +82,8 @@ + printk("\n"); \ + } while (0) + #else +-# define ea_idebug(f...) +-# define ea_bdebug(f...) ++# define ea_idebug(inode, f...) do {} while (0) ++# define ea_bdebug(bh, f...) do {} while (0) + #endif + + static void ext4_xattr_cache_insert(struct buffer_head *); +diff -urNp linux-2.6.38.3/fs/fcntl.c linux-2.6.38.3/fs/fcntl.c +--- linux-2.6.38.3/fs/fcntl.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/fcntl.c 2011-04-17 15:57:32.000000000 -0400 +@@ -224,6 +224,11 @@ int __f_setown(struct file *filp, struct + if (err) + return err; + ++ if (gr_handle_chroot_fowner(pid, type)) ++ return -ENOENT; ++ if (gr_check_protected_task_fowner(pid, type)) ++ return -EACCES; ++ + f_modown(filp, pid, type, force); + return 0; + } +@@ -348,6 +353,7 @@ static long do_fcntl(int fd, unsigned in + switch (cmd) { + case F_DUPFD: + case F_DUPFD_CLOEXEC: ++ gr_learn_resource(current, RLIMIT_NOFILE, arg, 0); + if (arg >= rlimit(RLIMIT_NOFILE)) + break; + err = alloc_fd(arg, cmd == F_DUPFD_CLOEXEC ? O_CLOEXEC : 0); +@@ -808,14 +814,14 @@ static int __init fcntl_init(void) + * Exceptions: O_NONBLOCK is a two bit define on parisc; O_NDELAY + * is defined as O_NONBLOCK on some platforms and not on others. + */ +- BUILD_BUG_ON(18 - 1 /* for O_RDONLY being 0 */ != HWEIGHT32( ++ BUILD_BUG_ON(19 - 1 /* for O_RDONLY being 0 */ != HWEIGHT32( + O_RDONLY | O_WRONLY | O_RDWR | + O_CREAT | O_EXCL | O_NOCTTY | + O_TRUNC | O_APPEND | /* O_NONBLOCK | */ + __O_SYNC | O_DSYNC | FASYNC | + O_DIRECT | O_LARGEFILE | O_DIRECTORY | + O_NOFOLLOW | O_NOATIME | O_CLOEXEC | +- __FMODE_EXEC ++ __FMODE_EXEC | FMODE_GREXEC + )); + + fasync_cache = kmem_cache_create("fasync_cache", +diff -urNp linux-2.6.38.3/fs/fifo.c linux-2.6.38.3/fs/fifo.c +--- linux-2.6.38.3/fs/fifo.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/fifo.c 2011-04-17 15:57:32.000000000 -0400 +@@ -58,10 +58,10 @@ static int fifo_open(struct inode *inode + */ + filp->f_op = &read_pipefifo_fops; + pipe->r_counter++; +- if (pipe->readers++ == 0) ++ if (atomic_inc_return(&pipe->readers) == 1) + wake_up_partner(inode); + +- if (!pipe->writers) { ++ if (!atomic_read(&pipe->writers)) { + if ((filp->f_flags & O_NONBLOCK)) { + /* suppress POLLHUP until we have + * seen a writer */ +@@ -82,15 +82,15 @@ static int fifo_open(struct inode *inode + * errno=ENXIO when there is no process reading the FIFO. + */ + ret = -ENXIO; +- if ((filp->f_flags & O_NONBLOCK) && !pipe->readers) ++ if ((filp->f_flags & O_NONBLOCK) && !atomic_read(&pipe->readers)) + goto err; + + filp->f_op = &write_pipefifo_fops; + pipe->w_counter++; +- if (!pipe->writers++) ++ if (atomic_inc_return(&pipe->writers) == 1) + wake_up_partner(inode); + +- if (!pipe->readers) { ++ if (!atomic_read(&pipe->readers)) { + wait_for_partner(inode, &pipe->r_counter); + if (signal_pending(current)) + goto err_wr; +@@ -106,11 +106,11 @@ static int fifo_open(struct inode *inode + */ + filp->f_op = &rdwr_pipefifo_fops; + +- pipe->readers++; +- pipe->writers++; ++ atomic_inc(&pipe->readers); ++ atomic_inc(&pipe->writers); + pipe->r_counter++; + pipe->w_counter++; +- if (pipe->readers == 1 || pipe->writers == 1) ++ if (atomic_read(&pipe->readers) == 1 || atomic_read(&pipe->writers) == 1) + wake_up_partner(inode); + break; + +@@ -124,19 +124,19 @@ static int fifo_open(struct inode *inode + return 0; + + err_rd: +- if (!--pipe->readers) ++ if (atomic_dec_and_test(&pipe->readers)) + wake_up_interruptible(&pipe->wait); + ret = -ERESTARTSYS; + goto err; + + err_wr: +- if (!--pipe->writers) ++ if (atomic_dec_and_test(&pipe->writers)) + wake_up_interruptible(&pipe->wait); + ret = -ERESTARTSYS; + goto err; + + err: +- if (!pipe->readers && !pipe->writers) ++ if (!atomic_read(&pipe->readers) && !atomic_read(&pipe->writers)) + free_pipe_info(inode); + + err_nocleanup: +diff -urNp linux-2.6.38.3/fs/file.c linux-2.6.38.3/fs/file.c +--- linux-2.6.38.3/fs/file.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/file.c 2011-04-17 15:57:32.000000000 -0400 +@@ -14,6 +14,7 @@ + #include <linux/slab.h> + #include <linux/vmalloc.h> + #include <linux/file.h> ++#include <linux/security.h> + #include <linux/fdtable.h> + #include <linux/bitops.h> + #include <linux/interrupt.h> +@@ -250,6 +251,7 @@ int expand_files(struct files_struct *fi + * N.B. For clone tasks sharing a files structure, this test + * will limit the total number of files that can be opened. + */ ++ gr_learn_resource(current, RLIMIT_NOFILE, nr, 0); + if (nr >= rlimit(RLIMIT_NOFILE)) + return -EMFILE; + +diff -urNp linux-2.6.38.3/fs/filesystems.c linux-2.6.38.3/fs/filesystems.c +--- linux-2.6.38.3/fs/filesystems.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/filesystems.c 2011-04-17 15:57:32.000000000 -0400 +@@ -275,7 +275,12 @@ struct file_system_type *get_fs_type(con + int len = dot ? dot - name : strlen(name); + + fs = __get_fs_type(name, len); ++ ++#ifdef CONFIG_GRKERNSEC_MODHARDEN ++ if (!fs && (___request_module(true, "grsec_modharden_fs", "%.*s", len, name) == 0)) ++#else + if (!fs && (request_module("%.*s", len, name) == 0)) ++#endif + fs = __get_fs_type(name, len); + + if (dot && fs && !(fs->fs_flags & FS_HAS_SUBTYPE)) { +diff -urNp linux-2.6.38.3/fs/fs_struct.c linux-2.6.38.3/fs/fs_struct.c +--- linux-2.6.38.3/fs/fs_struct.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/fs_struct.c 2011-04-17 15:57:32.000000000 -0400 +@@ -4,6 +4,7 @@ + #include <linux/path.h> + #include <linux/slab.h> + #include <linux/fs_struct.h> ++#include <linux/grsecurity.h> + #include "internal.h" + + static inline void path_get_longterm(struct path *path) +@@ -31,6 +32,7 @@ void set_fs_root(struct fs_struct *fs, s + old_root = fs->root; + fs->root = *path; + path_get_longterm(path); ++ gr_set_chroot_entries(current, path); + write_seqcount_end(&fs->seq); + spin_unlock(&fs->lock); + if (old_root.dentry) +@@ -74,6 +76,7 @@ void chroot_fs_refs(struct path *old_roo + && fs->root.mnt == old_root->mnt) { + path_get_longterm(new_root); + fs->root = *new_root; ++ gr_set_chroot_entries(p, new_root); + count++; + } + if (fs->pwd.dentry == old_root->dentry +@@ -109,7 +112,8 @@ void exit_fs(struct task_struct *tsk) + spin_lock(&fs->lock); + write_seqcount_begin(&fs->seq); + tsk->fs = NULL; +- kill = !--fs->users; ++ gr_clear_chroot_entries(tsk); ++ kill = !atomic_dec_return(&fs->users); + write_seqcount_end(&fs->seq); + spin_unlock(&fs->lock); + task_unlock(tsk); +@@ -123,7 +127,7 @@ struct fs_struct *copy_fs_struct(struct + struct fs_struct *fs = kmem_cache_alloc(fs_cachep, GFP_KERNEL); + /* We don't need to lock fs - think why ;-) */ + if (fs) { +- fs->users = 1; ++ atomic_set(&fs->users, 1); + fs->in_exec = 0; + spin_lock_init(&fs->lock); + seqcount_init(&fs->seq); +@@ -132,6 +136,9 @@ struct fs_struct *copy_fs_struct(struct + spin_lock(&old->lock); + fs->root = old->root; + path_get_longterm(&fs->root); ++ /* instead of calling gr_set_chroot_entries here, ++ we call it from every caller of this function ++ */ + fs->pwd = old->pwd; + path_get_longterm(&fs->pwd); + spin_unlock(&old->lock); +@@ -150,8 +157,9 @@ int unshare_fs_struct(void) + + task_lock(current); + spin_lock(&fs->lock); +- kill = !--fs->users; ++ kill = !atomic_dec_return(&fs->users); + current->fs = new_fs; ++ gr_set_chroot_entries(current, &new_fs->root); + spin_unlock(&fs->lock); + task_unlock(current); + +@@ -170,7 +178,7 @@ EXPORT_SYMBOL(current_umask); + + /* to be mentioned only in INIT_TASK */ + struct fs_struct init_fs = { +- .users = 1, ++ .users = ATOMIC_INIT(1), + .lock = __SPIN_LOCK_UNLOCKED(init_fs.lock), + .seq = SEQCNT_ZERO, + .umask = 0022, +@@ -186,12 +194,13 @@ void daemonize_fs_struct(void) + task_lock(current); + + spin_lock(&init_fs.lock); +- init_fs.users++; ++ atomic_inc(&init_fs.users); + spin_unlock(&init_fs.lock); + + spin_lock(&fs->lock); + current->fs = &init_fs; +- kill = !--fs->users; ++ gr_set_chroot_entries(current, ¤t->fs->root); ++ kill = !atomic_dec_return(&fs->users); + spin_unlock(&fs->lock); + + task_unlock(current); +diff -urNp linux-2.6.38.3/fs/fuse/control.c linux-2.6.38.3/fs/fuse/control.c +--- linux-2.6.38.3/fs/fuse/control.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/fuse/control.c 2011-04-17 15:57:32.000000000 -0400 +@@ -298,7 +298,7 @@ void fuse_ctl_remove_conn(struct fuse_co + + static int fuse_ctl_fill_super(struct super_block *sb, void *data, int silent) + { +- struct tree_descr empty_descr = {""}; ++ struct tree_descr empty_descr = {"", NULL, 0}; + struct fuse_conn *fc; + int err; + +diff -urNp linux-2.6.38.3/fs/fuse/cuse.c linux-2.6.38.3/fs/fuse/cuse.c +--- linux-2.6.38.3/fs/fuse/cuse.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/fuse/cuse.c 2011-04-17 15:57:32.000000000 -0400 +@@ -530,8 +530,18 @@ static int cuse_channel_release(struct i + return rc; + } + +-static struct file_operations cuse_channel_fops; /* initialized during init */ +- ++static const struct file_operations cuse_channel_fops = { /* initialized during init */ ++ .owner = THIS_MODULE, ++ .llseek = no_llseek, ++ .read = do_sync_read, ++ .aio_read = fuse_dev_read, ++ .write = do_sync_write, ++ .aio_write = fuse_dev_write, ++ .poll = fuse_dev_poll, ++ .open = cuse_channel_open, ++ .release = cuse_channel_release, ++ .fasync = fuse_dev_fasync, ++}; + + /************************************************************************** + * Misc stuff and module initializatiion +@@ -577,12 +587,6 @@ static int __init cuse_init(void) + for (i = 0; i < CUSE_CONNTBL_LEN; i++) + INIT_LIST_HEAD(&cuse_conntbl[i]); + +- /* inherit and extend fuse_dev_operations */ +- cuse_channel_fops = fuse_dev_operations; +- cuse_channel_fops.owner = THIS_MODULE; +- cuse_channel_fops.open = cuse_channel_open; +- cuse_channel_fops.release = cuse_channel_release; +- + cuse_class = class_create(THIS_MODULE, "cuse"); + if (IS_ERR(cuse_class)) + return PTR_ERR(cuse_class); +diff -urNp linux-2.6.38.3/fs/fuse/dev.c linux-2.6.38.3/fs/fuse/dev.c +--- linux-2.6.38.3/fs/fuse/dev.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/fuse/dev.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1183,7 +1183,7 @@ static ssize_t fuse_dev_do_read(struct f + return err; + } + +-static ssize_t fuse_dev_read(struct kiocb *iocb, const struct iovec *iov, ++ssize_t fuse_dev_read(struct kiocb *iocb, const struct iovec *iov, + unsigned long nr_segs, loff_t pos) + { + struct fuse_copy_state cs; +@@ -1197,6 +1197,8 @@ static ssize_t fuse_dev_read(struct kioc + return fuse_dev_do_read(fc, file, &cs, iov_length(iov, nr_segs)); + } + ++EXPORT_SYMBOL_GPL(fuse_dev_read); ++ + static int fuse_dev_pipe_buf_steal(struct pipe_inode_info *pipe, + struct pipe_buffer *buf) + { +@@ -1240,7 +1242,7 @@ static ssize_t fuse_dev_splice_read(stru + ret = 0; + pipe_lock(pipe); + +- if (!pipe->readers) { ++ if (!atomic_read(&pipe->readers)) { + send_sig(SIGPIPE, current, 0); + if (!ret) + ret = -EPIPE; +@@ -1733,7 +1735,7 @@ static ssize_t fuse_dev_do_write(struct + return err; + } + +-static ssize_t fuse_dev_write(struct kiocb *iocb, const struct iovec *iov, ++ssize_t fuse_dev_write(struct kiocb *iocb, const struct iovec *iov, + unsigned long nr_segs, loff_t pos) + { + struct fuse_copy_state cs; +@@ -1746,6 +1748,8 @@ static ssize_t fuse_dev_write(struct kio + return fuse_dev_do_write(fc, &cs, iov_length(iov, nr_segs)); + } + ++EXPORT_SYMBOL_GPL(fuse_dev_write); ++ + static ssize_t fuse_dev_splice_write(struct pipe_inode_info *pipe, + struct file *out, loff_t *ppos, + size_t len, unsigned int flags) +@@ -1824,7 +1828,7 @@ out: + return ret; + } + +-static unsigned fuse_dev_poll(struct file *file, poll_table *wait) ++unsigned fuse_dev_poll(struct file *file, poll_table *wait) + { + unsigned mask = POLLOUT | POLLWRNORM; + struct fuse_conn *fc = fuse_get_conn(file); +@@ -1843,6 +1847,8 @@ static unsigned fuse_dev_poll(struct fil + return mask; + } + ++EXPORT_SYMBOL_GPL(fuse_dev_poll); ++ + /* + * Abort all requests on the given list (pending or processing) + * +@@ -1962,7 +1968,7 @@ int fuse_dev_release(struct inode *inode + } + EXPORT_SYMBOL_GPL(fuse_dev_release); + +-static int fuse_dev_fasync(int fd, struct file *file, int on) ++int fuse_dev_fasync(int fd, struct file *file, int on) + { + struct fuse_conn *fc = fuse_get_conn(file); + if (!fc) +@@ -1972,6 +1978,8 @@ static int fuse_dev_fasync(int fd, struc + return fasync_helper(fd, file, on, &fc->fasync); + } + ++EXPORT_SYMBOL_GPL(fuse_dev_fasync); ++ + const struct file_operations fuse_dev_operations = { + .owner = THIS_MODULE, + .llseek = no_llseek, +diff -urNp linux-2.6.38.3/fs/fuse/dir.c linux-2.6.38.3/fs/fuse/dir.c +--- linux-2.6.38.3/fs/fuse/dir.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/fuse/dir.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1133,7 +1133,7 @@ static char *read_link(struct dentry *de + return link; + } + +-static void free_link(char *link) ++static void free_link(const char *link) + { + if (!IS_ERR(link)) + free_page((unsigned long) link); +diff -urNp linux-2.6.38.3/fs/fuse/fuse_i.h linux-2.6.38.3/fs/fuse/fuse_i.h +--- linux-2.6.38.3/fs/fuse/fuse_i.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/fuse/fuse_i.h 2011-04-17 15:57:32.000000000 -0400 +@@ -541,6 +541,16 @@ extern const struct file_operations fuse + + extern const struct dentry_operations fuse_dentry_operations; + ++extern ssize_t fuse_dev_read(struct kiocb *iocb, const struct iovec *iov, ++ unsigned long nr_segs, loff_t pos); ++ ++extern ssize_t fuse_dev_write(struct kiocb *iocb, const struct iovec *iov, ++ unsigned long nr_segs, loff_t pos); ++ ++extern unsigned fuse_dev_poll(struct file *file, poll_table *wait); ++ ++extern int fuse_dev_fasync(int fd, struct file *file, int on); ++ + /** + * Inode to nodeid comparison. + */ +diff -urNp linux-2.6.38.3/fs/hfs/inode.c linux-2.6.38.3/fs/hfs/inode.c +--- linux-2.6.38.3/fs/hfs/inode.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/hfs/inode.c 2011-04-17 15:57:32.000000000 -0400 +@@ -447,7 +447,7 @@ int hfs_write_inode(struct inode *inode, + + if (S_ISDIR(main_inode->i_mode)) { + if (fd.entrylength < sizeof(struct hfs_cat_dir)) +- /* panic? */; ++ {/* panic? */} + hfs_bnode_read(fd.bnode, &rec, fd.entryoffset, + sizeof(struct hfs_cat_dir)); + if (rec.type != HFS_CDR_DIR || +@@ -468,7 +468,7 @@ int hfs_write_inode(struct inode *inode, + sizeof(struct hfs_cat_file)); + } else { + if (fd.entrylength < sizeof(struct hfs_cat_file)) +- /* panic? */; ++ {/* panic? */} + hfs_bnode_read(fd.bnode, &rec, fd.entryoffset, + sizeof(struct hfs_cat_file)); + if (rec.type != HFS_CDR_FIL || +diff -urNp linux-2.6.38.3/fs/hfsplus/inode.c linux-2.6.38.3/fs/hfsplus/inode.c +--- linux-2.6.38.3/fs/hfsplus/inode.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/hfsplus/inode.c 2011-04-17 15:57:32.000000000 -0400 +@@ -498,7 +498,7 @@ int hfsplus_cat_read_inode(struct inode + struct hfsplus_cat_folder *folder = &entry.folder; + + if (fd->entrylength < sizeof(struct hfsplus_cat_folder)) +- /* panic? */; ++ {/* panic? */} + hfs_bnode_read(fd->bnode, &entry, fd->entryoffset, + sizeof(struct hfsplus_cat_folder)); + hfsplus_get_perms(inode, &folder->permissions, 1); +@@ -515,7 +515,7 @@ int hfsplus_cat_read_inode(struct inode + struct hfsplus_cat_file *file = &entry.file; + + if (fd->entrylength < sizeof(struct hfsplus_cat_file)) +- /* panic? */; ++ {/* panic? */} + hfs_bnode_read(fd->bnode, &entry, fd->entryoffset, + sizeof(struct hfsplus_cat_file)); + +@@ -572,7 +572,7 @@ int hfsplus_cat_write_inode(struct inode + struct hfsplus_cat_folder *folder = &entry.folder; + + if (fd.entrylength < sizeof(struct hfsplus_cat_folder)) +- /* panic? */; ++ {/* panic? */} + hfs_bnode_read(fd.bnode, &entry, fd.entryoffset, + sizeof(struct hfsplus_cat_folder)); + /* simple node checks? */ +@@ -594,7 +594,7 @@ int hfsplus_cat_write_inode(struct inode + struct hfsplus_cat_file *file = &entry.file; + + if (fd.entrylength < sizeof(struct hfsplus_cat_file)) +- /* panic? */; ++ {/* panic? */} + hfs_bnode_read(fd.bnode, &entry, fd.entryoffset, + sizeof(struct hfsplus_cat_file)); + hfsplus_inode_write_fork(inode, &file->data_fork); +diff -urNp linux-2.6.38.3/fs/hugetlbfs/inode.c linux-2.6.38.3/fs/hugetlbfs/inode.c +--- linux-2.6.38.3/fs/hugetlbfs/inode.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/hugetlbfs/inode.c 2011-04-17 15:57:32.000000000 -0400 +@@ -915,7 +915,7 @@ static struct file_system_type hugetlbfs + .kill_sb = kill_litter_super, + }; + +-static struct vfsmount *hugetlbfs_vfsmount; ++struct vfsmount *hugetlbfs_vfsmount; + + static int can_do_hugetlb_shm(void) + { +diff -urNp linux-2.6.38.3/fs/jffs2/debug.h linux-2.6.38.3/fs/jffs2/debug.h +--- linux-2.6.38.3/fs/jffs2/debug.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/jffs2/debug.h 2011-04-17 15:57:32.000000000 -0400 +@@ -53,13 +53,13 @@ + #if CONFIG_JFFS2_FS_DEBUG > 0 + #define D1(x) x + #else +-#define D1(x) ++#define D1(x) do {} while (0); + #endif + + #if CONFIG_JFFS2_FS_DEBUG > 1 + #define D2(x) x + #else +-#define D2(x) ++#define D2(x) do {} while (0); + #endif + + /* The prefixes of JFFS2 messages */ +@@ -115,73 +115,73 @@ + #ifdef JFFS2_DBG_READINODE_MESSAGES + #define dbg_readinode(fmt, ...) JFFS2_DEBUG(fmt, ##__VA_ARGS__) + #else +-#define dbg_readinode(fmt, ...) ++#define dbg_readinode(fmt, ...) do {} while (0) + #endif + #ifdef JFFS2_DBG_READINODE2_MESSAGES + #define dbg_readinode2(fmt, ...) JFFS2_DEBUG(fmt, ##__VA_ARGS__) + #else +-#define dbg_readinode2(fmt, ...) ++#define dbg_readinode2(fmt, ...) do {} while (0) + #endif + + /* Fragtree build debugging messages */ + #ifdef JFFS2_DBG_FRAGTREE_MESSAGES + #define dbg_fragtree(fmt, ...) JFFS2_DEBUG(fmt, ##__VA_ARGS__) + #else +-#define dbg_fragtree(fmt, ...) ++#define dbg_fragtree(fmt, ...) do {} while (0) + #endif + #ifdef JFFS2_DBG_FRAGTREE2_MESSAGES + #define dbg_fragtree2(fmt, ...) JFFS2_DEBUG(fmt, ##__VA_ARGS__) + #else +-#define dbg_fragtree2(fmt, ...) ++#define dbg_fragtree2(fmt, ...) do {} while (0) + #endif + + /* Directory entry list manilulation debugging messages */ + #ifdef JFFS2_DBG_DENTLIST_MESSAGES + #define dbg_dentlist(fmt, ...) JFFS2_DEBUG(fmt, ##__VA_ARGS__) + #else +-#define dbg_dentlist(fmt, ...) ++#define dbg_dentlist(fmt, ...) do {} while (0) + #endif + + /* Print the messages about manipulating node_refs */ + #ifdef JFFS2_DBG_NODEREF_MESSAGES + #define dbg_noderef(fmt, ...) JFFS2_DEBUG(fmt, ##__VA_ARGS__) + #else +-#define dbg_noderef(fmt, ...) ++#define dbg_noderef(fmt, ...) do {} while (0) + #endif + + /* Manipulations with the list of inodes (JFFS2 inocache) */ + #ifdef JFFS2_DBG_INOCACHE_MESSAGES + #define dbg_inocache(fmt, ...) JFFS2_DEBUG(fmt, ##__VA_ARGS__) + #else +-#define dbg_inocache(fmt, ...) ++#define dbg_inocache(fmt, ...) do {} while (0) + #endif + + /* Summary debugging messages */ + #ifdef JFFS2_DBG_SUMMARY_MESSAGES + #define dbg_summary(fmt, ...) JFFS2_DEBUG(fmt, ##__VA_ARGS__) + #else +-#define dbg_summary(fmt, ...) ++#define dbg_summary(fmt, ...) do {} while (0) + #endif + + /* File system build messages */ + #ifdef JFFS2_DBG_FSBUILD_MESSAGES + #define dbg_fsbuild(fmt, ...) JFFS2_DEBUG(fmt, ##__VA_ARGS__) + #else +-#define dbg_fsbuild(fmt, ...) ++#define dbg_fsbuild(fmt, ...) do {} while (0) + #endif + + /* Watch the object allocations */ + #ifdef JFFS2_DBG_MEMALLOC_MESSAGES + #define dbg_memalloc(fmt, ...) JFFS2_DEBUG(fmt, ##__VA_ARGS__) + #else +-#define dbg_memalloc(fmt, ...) ++#define dbg_memalloc(fmt, ...) do {} while (0) + #endif + + /* Watch the XATTR subsystem */ + #ifdef JFFS2_DBG_XATTR_MESSAGES + #define dbg_xattr(fmt, ...) JFFS2_DEBUG(fmt, ##__VA_ARGS__) + #else +-#define dbg_xattr(fmt, ...) ++#define dbg_xattr(fmt, ...) do {} while (0) + #endif + + /* "Sanity" checks */ +diff -urNp linux-2.6.38.3/fs/jffs2/erase.c linux-2.6.38.3/fs/jffs2/erase.c +--- linux-2.6.38.3/fs/jffs2/erase.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/jffs2/erase.c 2011-04-17 15:57:32.000000000 -0400 +@@ -439,7 +439,8 @@ static void jffs2_mark_erased_block(stru + struct jffs2_unknown_node marker = { + .magic = cpu_to_je16(JFFS2_MAGIC_BITMASK), + .nodetype = cpu_to_je16(JFFS2_NODETYPE_CLEANMARKER), +- .totlen = cpu_to_je32(c->cleanmarker_size) ++ .totlen = cpu_to_je32(c->cleanmarker_size), ++ .hdr_crc = cpu_to_je32(0) + }; + + jffs2_prealloc_raw_node_refs(c, jeb, 1); +diff -urNp linux-2.6.38.3/fs/jffs2/summary.h linux-2.6.38.3/fs/jffs2/summary.h +--- linux-2.6.38.3/fs/jffs2/summary.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/jffs2/summary.h 2011-04-17 15:57:32.000000000 -0400 +@@ -194,18 +194,18 @@ int jffs2_sum_scan_sumnode(struct jffs2_ + + #define jffs2_sum_active() (0) + #define jffs2_sum_init(a) (0) +-#define jffs2_sum_exit(a) +-#define jffs2_sum_disable_collecting(a) ++#define jffs2_sum_exit(a) do {} while (0) ++#define jffs2_sum_disable_collecting(a) do {} while (0) + #define jffs2_sum_is_disabled(a) (0) +-#define jffs2_sum_reset_collected(a) ++#define jffs2_sum_reset_collected(a) do {} while (0) + #define jffs2_sum_add_kvec(a,b,c,d) (0) +-#define jffs2_sum_move_collected(a,b) ++#define jffs2_sum_move_collected(a,b) do {} while (0) + #define jffs2_sum_write_sumnode(a) (0) +-#define jffs2_sum_add_padding_mem(a,b) +-#define jffs2_sum_add_inode_mem(a,b,c) +-#define jffs2_sum_add_dirent_mem(a,b,c) +-#define jffs2_sum_add_xattr_mem(a,b,c) +-#define jffs2_sum_add_xref_mem(a,b,c) ++#define jffs2_sum_add_padding_mem(a,b) do {} while (0) ++#define jffs2_sum_add_inode_mem(a,b,c) do {} while (0) ++#define jffs2_sum_add_dirent_mem(a,b,c) do {} while (0) ++#define jffs2_sum_add_xattr_mem(a,b,c) do {} while (0) ++#define jffs2_sum_add_xref_mem(a,b,c) do {} while (0) + #define jffs2_sum_scan_sumnode(a,b,c,d,e) (0) + + #endif /* CONFIG_JFFS2_SUMMARY */ +diff -urNp linux-2.6.38.3/fs/jffs2/wbuf.c linux-2.6.38.3/fs/jffs2/wbuf.c +--- linux-2.6.38.3/fs/jffs2/wbuf.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/jffs2/wbuf.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1012,7 +1012,8 @@ static const struct jffs2_unknown_node o + { + .magic = constant_cpu_to_je16(JFFS2_MAGIC_BITMASK), + .nodetype = constant_cpu_to_je16(JFFS2_NODETYPE_CLEANMARKER), +- .totlen = constant_cpu_to_je32(8) ++ .totlen = constant_cpu_to_je32(8), ++ .hdr_crc = constant_cpu_to_je32(0) + }; + + /* +diff -urNp linux-2.6.38.3/fs/Kconfig.binfmt linux-2.6.38.3/fs/Kconfig.binfmt +--- linux-2.6.38.3/fs/Kconfig.binfmt 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/Kconfig.binfmt 2011-04-17 15:57:32.000000000 -0400 +@@ -86,7 +86,7 @@ config HAVE_AOUT + + config BINFMT_AOUT + tristate "Kernel support for a.out and ECOFF binaries" +- depends on HAVE_AOUT ++ depends on HAVE_AOUT && BROKEN + ---help--- + A.out (Assembler.OUTput) is a set of formats for libraries and + executables used in the earliest versions of UNIX. Linux used +diff -urNp linux-2.6.38.3/fs/libfs.c linux-2.6.38.3/fs/libfs.c +--- linux-2.6.38.3/fs/libfs.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/libfs.c 2011-04-17 15:57:32.000000000 -0400 +@@ -138,6 +138,8 @@ int dcache_readdir(struct file * filp, v + struct dentry *dentry = filp->f_path.dentry; + struct dentry *cursor = filp->private_data; + struct list_head *p, *q = &cursor->d_u.d_child; ++ char d_name[DNAME_INLINE_LEN]; ++ const char *name; + ino_t ino; + int i = filp->f_pos; + +@@ -172,7 +174,12 @@ int dcache_readdir(struct file * filp, v + + spin_unlock(&next->d_lock); + spin_unlock(&dentry->d_lock); +- if (filldir(dirent, next->d_name.name, ++ if (next->d_name.len < DNAME_INLINE_LEN) { ++ memcpy(d_name, next->d_name.name, next->d_name.len); ++ name = d_name; ++ } else ++ name = next->d_name.name; ++ if (filldir(dirent, name, + next->d_name.len, filp->f_pos, + next->d_inode->i_ino, + dt_type(next->d_inode)) < 0) +diff -urNp linux-2.6.38.3/fs/lockd/svc.c linux-2.6.38.3/fs/lockd/svc.c +--- linux-2.6.38.3/fs/lockd/svc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/lockd/svc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -41,7 +41,7 @@ + + static struct svc_program nlmsvc_program; + +-struct nlmsvc_binding * nlmsvc_ops; ++const struct nlmsvc_binding * nlmsvc_ops; + EXPORT_SYMBOL_GPL(nlmsvc_ops); + + static DEFINE_MUTEX(nlmsvc_mutex); +diff -urNp linux-2.6.38.3/fs/locks.c linux-2.6.38.3/fs/locks.c +--- linux-2.6.38.3/fs/locks.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/locks.c 2011-04-17 15:57:32.000000000 -0400 +@@ -2044,16 +2044,16 @@ void locks_remove_flock(struct file *fil + return; + + if (filp->f_op && filp->f_op->flock) { +- struct file_lock fl = { ++ struct file_lock flock = { + .fl_pid = current->tgid, + .fl_file = filp, + .fl_flags = FL_FLOCK, + .fl_type = F_UNLCK, + .fl_end = OFFSET_MAX, + }; +- filp->f_op->flock(filp, F_SETLKW, &fl); +- if (fl.fl_ops && fl.fl_ops->fl_release_private) +- fl.fl_ops->fl_release_private(&fl); ++ filp->f_op->flock(filp, F_SETLKW, &flock); ++ if (flock.fl_ops && flock.fl_ops->fl_release_private) ++ flock.fl_ops->fl_release_private(&flock); + } + + lock_flocks(); +diff -urNp linux-2.6.38.3/fs/namei.c linux-2.6.38.3/fs/namei.c +--- linux-2.6.38.3/fs/namei.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/namei.c 2011-04-17 15:57:32.000000000 -0400 +@@ -226,14 +226,6 @@ int generic_permission(struct inode *ino + return ret; + + /* +- * Read/write DACs are always overridable. +- * Executable DACs are overridable if at least one exec bit is set. +- */ +- if (!(mask & MAY_EXEC) || execute_ok(inode)) +- if (capable(CAP_DAC_OVERRIDE)) +- return 0; +- +- /* + * Searching includes executable on directories, else just read. + */ + mask &= MAY_READ | MAY_WRITE | MAY_EXEC; +@@ -241,6 +233,14 @@ int generic_permission(struct inode *ino + if (capable(CAP_DAC_READ_SEARCH)) + return 0; + ++ /* ++ * Read/write DACs are always overridable. ++ * Executable DACs are overridable if at least one exec bit is set. ++ */ ++ if (!(mask & MAY_EXEC) || execute_ok(inode)) ++ if (capable(CAP_DAC_OVERRIDE)) ++ return 0; ++ + return -EACCES; + } + +@@ -687,7 +687,8 @@ static inline int exec_permission(struct + if (ret == -ECHILD) + return ret; + +- if (capable(CAP_DAC_OVERRIDE) || capable(CAP_DAC_READ_SEARCH)) ++ if (capable_nolog(CAP_DAC_OVERRIDE) || capable(CAP_DAC_READ_SEARCH) || ++ capable(CAP_DAC_OVERRIDE)) + goto ok; + + return ret; +@@ -775,7 +776,7 @@ __do_follow_link(const struct path *link + *p = dentry->d_inode->i_op->follow_link(dentry, nd); + error = PTR_ERR(*p); + if (!IS_ERR(*p)) { +- char *s = nd_get_link(nd); ++ const char *s = nd_get_link(nd); + error = 0; + if (s) + error = __vfs_follow_link(nd, s); +@@ -814,6 +815,13 @@ static inline int do_follow_link(struct + err = security_inode_follow_link(path->dentry, nd); + if (err) + goto loop; ++ ++ if (gr_handle_follow_link(path->dentry->d_parent->d_inode, ++ path->dentry->d_inode, path->dentry, nd->path.mnt)) { ++ err = -EACCES; ++ goto loop; ++ } ++ + current->link_count++; + current->total_link_count++; + nd->depth++; +@@ -1505,13 +1513,36 @@ return_reval: + return_base: + if (nameidata_drop_rcu_last_maybe(nd)) + return -ECHILD; ++ ++ if (!(nd->flags & LOOKUP_PARENT) && !gr_acl_handle_hidden_file(nd->path.dentry, nd->path.mnt)) { ++ err = -ENOENT; ++ goto err_and_ret; ++ } ++ + return 0; + out_dput: + if (!(nd->flags & LOOKUP_RCU)) + path_put_conditional(&next, nd); + break; + } ++#ifdef CONFIG_GRKERNSEC ++ /* we do this because we can't operate here on an rcu'd dentry, ++ acquire a properly-referenced copy ++ */ ++ if (nameidata_drop_rcu_last_maybe(nd)) ++ return -ECHILD; ++#endif ++ ++ if (!(nd->flags & LOOKUP_PARENT) && !gr_acl_handle_hidden_file(nd->path.dentry, nd->path.mnt)) ++ err = -ENOENT; ++ ++err_and_ret: ++#ifndef CONFIG_GRKERNSEC ++ /* since we convert to ref-walk above, always put the path if we reach ++ here ++ */ + if (!(nd->flags & LOOKUP_RCU)) ++#endif + path_put(&nd->path); + return_err: + return err; +@@ -1738,6 +1769,9 @@ static int do_path_lookup(int dfd, const + } + + if (likely(!retval)) { ++ if (*name != '/' && nd->path.dentry && nd->inode && !gr_chroot_fchdir(nd->path.dentry, nd->path.mnt)) ++ return -ENOENT; ++ + if (unlikely(!audit_dummy_context())) { + if (nd->path.dentry && nd->inode) + audit_inode(name, nd->path.dentry); +@@ -2078,6 +2112,30 @@ int vfs_create(struct inode *dir, struct + return error; + } + ++/* ++ * Note that while the flag value (low two bits) for sys_open means: ++ * 00 - read-only ++ * 01 - write-only ++ * 10 - read-write ++ * 11 - special ++ * it is changed into ++ * 00 - no permissions needed ++ * 01 - read-permission ++ * 10 - write-permission ++ * 11 - read-write ++ * for the internal routines (ie open_namei()/follow_link() etc) ++ * This is more logical, and also allows the 00 "no perm needed" ++ * to be used for symlinks (where the permissions are checked ++ * later). ++ * ++*/ ++static inline int open_to_namei_flags(int flag) ++{ ++ if ((flag+1) & O_ACCMODE) ++ flag++; ++ return flag; ++} ++ + int may_open(struct path *path, int acc_mode, int flag) + { + struct dentry *dentry = path->dentry; +@@ -2126,7 +2184,27 @@ int may_open(struct path *path, int acc_ + /* + * Ensure there are no outstanding leases on the file. + */ +- return break_lease(inode, flag); ++ error = break_lease(inode, flag); ++ ++ if (error) ++ return error; ++ ++ if (gr_handle_rofs_blockwrite(dentry, path->mnt, acc_mode)) { ++ error = -EPERM; ++ goto exit; ++ } ++ ++ if (gr_handle_rawio(inode)) { ++ error = -EPERM; ++ goto exit; ++ } ++ ++ if (!gr_acl_handle_open(dentry, path->mnt, open_to_namei_flags(flag))) { ++ error = -EACCES; ++ goto exit; ++ } ++exit: ++ return error; + } + + static int handle_truncate(struct file *filp) +@@ -2161,6 +2239,12 @@ static int __open_namei_create(struct na + { + int error; + struct dentry *dir = nd->path.dentry; ++ int flag = open_to_namei_flags(open_flag); ++ ++ if (!gr_acl_handle_creat(path->dentry, nd->path.dentry, nd->path.mnt, flag, mode)) { ++ error = -EACCES; ++ goto out_unlock; ++ } + + if (!IS_POSIXACL(dir->d_inode)) + mode &= ~current_umask(); +@@ -2168,6 +2252,8 @@ static int __open_namei_create(struct na + if (error) + goto out_unlock; + error = vfs_create(dir->d_inode, path->dentry, mode, nd); ++ if (!error) ++ gr_handle_create(path->dentry, nd->path.mnt); + out_unlock: + mutex_unlock(&dir->d_inode->i_mutex); + dput(nd->path.dentry); +@@ -2179,30 +2265,6 @@ out_unlock: + return may_open(&nd->path, 0, open_flag & ~O_TRUNC); + } + +-/* +- * Note that while the flag value (low two bits) for sys_open means: +- * 00 - read-only +- * 01 - write-only +- * 10 - read-write +- * 11 - special +- * it is changed into +- * 00 - no permissions needed +- * 01 - read-permission +- * 10 - write-permission +- * 11 - read-write +- * for the internal routines (ie open_namei()/follow_link() etc) +- * This is more logical, and also allows the 00 "no perm needed" +- * to be used for symlinks (where the permissions are checked +- * later). +- * +-*/ +-static inline int open_to_namei_flags(int flag) +-{ +- if ((flag+1) & O_ACCMODE) +- flag++; +- return flag; +-} +- + static int open_will_truncate(int flag, struct inode *inode) + { + /* +@@ -2273,6 +2335,7 @@ static struct file *do_last(struct namei + int mode, const char *pathname) + { + struct dentry *dir = nd->path.dentry; ++ int flag = open_to_namei_flags(open_flag); + struct file *filp; + int error = -EISDIR; + +@@ -2351,6 +2414,14 @@ static struct file *do_last(struct namei + /* + * It already exists. + */ ++ ++ /* only check if O_CREAT is specified, all other checks need to go ++ into may_open */ ++ if (gr_handle_fifo(path->dentry, path->mnt, dir, flag, acc_mode)) { ++ error = -EACCES; ++ goto exit_mutex_unlock; ++ } ++ + mutex_unlock(&dir->d_inode->i_mutex); + audit_inode(pathname, path->dentry); + +@@ -2534,6 +2605,11 @@ reval: + error = security_inode_follow_link(link.dentry, &nd); + if (error) + goto exit_dput; ++ if (gr_handle_follow_link(link.dentry->d_parent->d_inode, ++ link.dentry->d_inode, link.dentry, nd.path.mnt)) { ++ error = -EACCES; ++ goto exit_dput; ++ } + error = __do_follow_link(&link, &nd, &cookie); + if (unlikely(error)) { + if (!IS_ERR(cookie) && linki->i_op->put_link) +@@ -2704,6 +2780,17 @@ SYSCALL_DEFINE4(mknodat, int, dfd, const + error = may_mknod(mode); + if (error) + goto out_dput; ++ ++ if (gr_handle_chroot_mknod(dentry, nd.path.mnt, mode)) { ++ error = -EPERM; ++ goto out_dput; ++ } ++ ++ if (!gr_acl_handle_mknod(dentry, nd.path.dentry, nd.path.mnt, mode)) { ++ error = -EACCES; ++ goto out_dput; ++ } ++ + error = mnt_want_write(nd.path.mnt); + if (error) + goto out_dput; +@@ -2724,6 +2811,9 @@ SYSCALL_DEFINE4(mknodat, int, dfd, const + } + out_drop_write: + mnt_drop_write(nd.path.mnt); ++ ++ if (!error) ++ gr_handle_create(dentry, nd.path.mnt); + out_dput: + dput(dentry); + out_unlock: +@@ -2776,6 +2866,11 @@ SYSCALL_DEFINE3(mkdirat, int, dfd, const + if (IS_ERR(dentry)) + goto out_unlock; + ++ if (!gr_acl_handle_mkdir(dentry, nd.path.dentry, nd.path.mnt)) { ++ error = -EACCES; ++ goto out_dput; ++ } ++ + if (!IS_POSIXACL(nd.path.dentry->d_inode)) + mode &= ~current_umask(); + error = mnt_want_write(nd.path.mnt); +@@ -2787,6 +2882,10 @@ SYSCALL_DEFINE3(mkdirat, int, dfd, const + error = vfs_mkdir(nd.path.dentry->d_inode, dentry, mode); + out_drop_write: + mnt_drop_write(nd.path.mnt); ++ ++ if (!error) ++ gr_handle_create(dentry, nd.path.mnt); ++ + out_dput: + dput(dentry); + out_unlock: +@@ -2866,6 +2965,8 @@ static long do_rmdir(int dfd, const char + char * name; + struct dentry *dentry; + struct nameidata nd; ++ ino_t saved_ino = 0; ++ dev_t saved_dev = 0; + + error = user_path_parent(dfd, pathname, &nd, &name); + if (error) +@@ -2890,6 +2991,19 @@ static long do_rmdir(int dfd, const char + error = PTR_ERR(dentry); + if (IS_ERR(dentry)) + goto exit2; ++ ++ if (dentry->d_inode != NULL) { ++ if (dentry->d_inode->i_nlink <= 1) { ++ saved_ino = dentry->d_inode->i_ino; ++ saved_dev = gr_get_dev_from_dentry(dentry); ++ } ++ ++ if (!gr_acl_handle_rmdir(dentry, nd.path.mnt)) { ++ error = -EACCES; ++ goto exit3; ++ } ++ } ++ + error = mnt_want_write(nd.path.mnt); + if (error) + goto exit3; +@@ -2897,6 +3011,8 @@ static long do_rmdir(int dfd, const char + if (error) + goto exit4; + error = vfs_rmdir(nd.path.dentry->d_inode, dentry); ++ if (!error && (saved_dev || saved_ino)) ++ gr_handle_delete(saved_ino, saved_dev); + exit4: + mnt_drop_write(nd.path.mnt); + exit3: +@@ -2959,6 +3075,8 @@ static long do_unlinkat(int dfd, const c + struct dentry *dentry; + struct nameidata nd; + struct inode *inode = NULL; ++ ino_t saved_ino = 0; ++ dev_t saved_dev = 0; + + error = user_path_parent(dfd, pathname, &nd, &name); + if (error) +@@ -2978,8 +3096,17 @@ static long do_unlinkat(int dfd, const c + if (nd.last.name[nd.last.len]) + goto slashes; + inode = dentry->d_inode; +- if (inode) ++ if (inode) { + ihold(inode); ++ if (inode->i_nlink <= 1) { ++ saved_ino = inode->i_ino; ++ saved_dev = gr_get_dev_from_dentry(dentry); ++ } ++ if (!gr_acl_handle_unlink(dentry, nd.path.mnt)) { ++ error = -EACCES; ++ goto exit2; ++ } ++ } + error = mnt_want_write(nd.path.mnt); + if (error) + goto exit2; +@@ -2987,6 +3114,8 @@ static long do_unlinkat(int dfd, const c + if (error) + goto exit3; + error = vfs_unlink(nd.path.dentry->d_inode, dentry); ++ if (!error && (saved_ino || saved_dev)) ++ gr_handle_delete(saved_ino, saved_dev); + exit3: + mnt_drop_write(nd.path.mnt); + exit2: +@@ -3064,6 +3193,11 @@ SYSCALL_DEFINE3(symlinkat, const char __ + if (IS_ERR(dentry)) + goto out_unlock; + ++ if (!gr_acl_handle_symlink(dentry, nd.path.dentry, nd.path.mnt, from)) { ++ error = -EACCES; ++ goto out_dput; ++ } ++ + error = mnt_want_write(nd.path.mnt); + if (error) + goto out_dput; +@@ -3071,6 +3205,8 @@ SYSCALL_DEFINE3(symlinkat, const char __ + if (error) + goto out_drop_write; + error = vfs_symlink(nd.path.dentry->d_inode, dentry, from); ++ if (!error) ++ gr_handle_create(dentry, nd.path.mnt); + out_drop_write: + mnt_drop_write(nd.path.mnt); + out_dput: +@@ -3163,6 +3299,20 @@ SYSCALL_DEFINE5(linkat, int, olddfd, con + error = PTR_ERR(new_dentry); + if (IS_ERR(new_dentry)) + goto out_unlock; ++ ++ if (gr_handle_hardlink(old_path.dentry, old_path.mnt, ++ old_path.dentry->d_inode, ++ old_path.dentry->d_inode->i_mode, to)) { ++ error = -EACCES; ++ goto out_dput; ++ } ++ ++ if (!gr_acl_handle_link(new_dentry, nd.path.dentry, nd.path.mnt, ++ old_path.dentry, old_path.mnt, to)) { ++ error = -EACCES; ++ goto out_dput; ++ } ++ + error = mnt_want_write(nd.path.mnt); + if (error) + goto out_dput; +@@ -3170,6 +3320,8 @@ SYSCALL_DEFINE5(linkat, int, olddfd, con + if (error) + goto out_drop_write; + error = vfs_link(old_path.dentry, nd.path.dentry->d_inode, new_dentry); ++ if (!error) ++ gr_handle_create(new_dentry, nd.path.mnt); + out_drop_write: + mnt_drop_write(nd.path.mnt); + out_dput: +@@ -3403,6 +3555,12 @@ SYSCALL_DEFINE4(renameat, int, olddfd, c + if (new_dentry == trap) + goto exit5; + ++ error = gr_acl_handle_rename(new_dentry, new_dir, newnd.path.mnt, ++ old_dentry, old_dir->d_inode, oldnd.path.mnt, ++ to); ++ if (error) ++ goto exit5; ++ + error = mnt_want_write(oldnd.path.mnt); + if (error) + goto exit5; +@@ -3412,6 +3570,9 @@ SYSCALL_DEFINE4(renameat, int, olddfd, c + goto exit6; + error = vfs_rename(old_dir->d_inode, old_dentry, + new_dir->d_inode, new_dentry); ++ if (!error) ++ gr_handle_rename(old_dir->d_inode, new_dir->d_inode, old_dentry, ++ new_dentry, oldnd.path.mnt, new_dentry->d_inode ? 1 : 0); + exit6: + mnt_drop_write(oldnd.path.mnt); + exit5: +@@ -3437,6 +3598,8 @@ SYSCALL_DEFINE2(rename, const char __use + + int vfs_readlink(struct dentry *dentry, char __user *buffer, int buflen, const char *link) + { ++ char tmpbuf[64]; ++ const char *newlink; + int len; + + len = PTR_ERR(link); +@@ -3446,7 +3609,14 @@ int vfs_readlink(struct dentry *dentry, + len = strlen(link); + if (len > (unsigned) buflen) + len = buflen; +- if (copy_to_user(buffer, link, len)) ++ ++ if (len < sizeof(tmpbuf)) { ++ memcpy(tmpbuf, link, len); ++ newlink = tmpbuf; ++ } else ++ newlink = link; ++ ++ if (copy_to_user(buffer, newlink, len)) + len = -EFAULT; + out: + return len; +diff -urNp linux-2.6.38.3/fs/namespace.c linux-2.6.38.3/fs/namespace.c +--- linux-2.6.38.3/fs/namespace.c 2011-03-28 17:42:40.000000000 -0400 ++++ linux-2.6.38.3/fs/namespace.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1285,6 +1285,9 @@ static int do_umount(struct vfsmount *mn + if (!(sb->s_flags & MS_RDONLY)) + retval = do_remount_sb(sb, MS_RDONLY, NULL, 0); + up_write(&sb->s_umount); ++ ++ gr_log_remount(mnt->mnt_devname, retval); ++ + return retval; + } + +@@ -1304,6 +1307,9 @@ static int do_umount(struct vfsmount *mn + br_write_unlock(vfsmount_lock); + up_write(&namespace_sem); + release_mounts(&umount_list); ++ ++ gr_log_unmount(mnt->mnt_devname, retval); ++ + return retval; + } + +@@ -2241,6 +2247,16 @@ long do_mount(char *dev_name, char *dir_ + MS_NOATIME | MS_NODIRATIME | MS_RELATIME| MS_KERNMOUNT | + MS_STRICTATIME); + ++ if (gr_handle_rofs_mount(path.dentry, path.mnt, mnt_flags)) { ++ retval = -EPERM; ++ goto dput_out; ++ } ++ ++ if (gr_handle_chroot_mount(path.dentry, path.mnt, dev_name)) { ++ retval = -EPERM; ++ goto dput_out; ++ } ++ + if (flags & MS_REMOUNT) + retval = do_remount(&path, flags & ~MS_REMOUNT, mnt_flags, + data_page); +@@ -2255,6 +2271,9 @@ long do_mount(char *dev_name, char *dir_ + dev_name, data_page); + dput_out: + path_put(&path); ++ ++ gr_log_mount(dev_name, dir_name, retval); ++ + return retval; + } + +@@ -2480,6 +2499,12 @@ SYSCALL_DEFINE2(pivot_root, const char _ + goto out1; + } + ++ if (gr_handle_chroot_pivot()) { ++ error = -EPERM; ++ path_put(&old); ++ goto out1; ++ } ++ + get_fs_root(current->fs, &root); + down_write(&namespace_sem); + mutex_lock(&old.dentry->d_inode->i_mutex); +diff -urNp linux-2.6.38.3/fs/nfs/inode.c linux-2.6.38.3/fs/nfs/inode.c +--- linux-2.6.38.3/fs/nfs/inode.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/nfs/inode.c 2011-04-17 15:57:32.000000000 -0400 +@@ -998,16 +998,16 @@ static int nfs_size_need_update(const st + return nfs_size_to_loff_t(fattr->size) > i_size_read(inode); + } + +-static atomic_long_t nfs_attr_generation_counter; ++static atomic_long_unchecked_t nfs_attr_generation_counter; + + static unsigned long nfs_read_attr_generation_counter(void) + { +- return atomic_long_read(&nfs_attr_generation_counter); ++ return atomic_long_read_unchecked(&nfs_attr_generation_counter); + } + + unsigned long nfs_inc_attr_generation_counter(void) + { +- return atomic_long_inc_return(&nfs_attr_generation_counter); ++ return atomic_long_inc_return_unchecked(&nfs_attr_generation_counter); + } + + void nfs_fattr_init(struct nfs_fattr *fattr) +diff -urNp linux-2.6.38.3/fs/nfs/nfs4proc.c linux-2.6.38.3/fs/nfs/nfs4proc.c +--- linux-2.6.38.3/fs/nfs/nfs4proc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/nfs/nfs4proc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1198,7 +1198,7 @@ static int _nfs4_do_open_reclaim(struct + static int nfs4_do_open_reclaim(struct nfs_open_context *ctx, struct nfs4_state *state) + { + struct nfs_server *server = NFS_SERVER(state->inode); +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + do { + err = _nfs4_do_open_reclaim(ctx, state); +@@ -1240,7 +1240,7 @@ static int _nfs4_open_delegation_recall( + + int nfs4_open_delegation_recall(struct nfs_open_context *ctx, struct nfs4_state *state, const nfs4_stateid *stateid) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + struct nfs_server *server = NFS_SERVER(state->inode); + int err; + do { +@@ -1615,7 +1615,7 @@ static int _nfs4_open_expired(struct nfs + static int nfs4_do_open_expired(struct nfs_open_context *ctx, struct nfs4_state *state) + { + struct nfs_server *server = NFS_SERVER(state->inode); +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + + do { +@@ -1730,7 +1730,7 @@ out_err: + + static struct nfs4_state *nfs4_do_open(struct inode *dir, struct path *path, fmode_t fmode, int flags, struct iattr *sattr, struct rpc_cred *cred) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + struct nfs4_state *res; + int status; + +@@ -1821,7 +1821,7 @@ static int nfs4_do_setattr(struct inode + struct nfs4_state *state) + { + struct nfs_server *server = NFS_SERVER(inode); +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + do { + err = nfs4_handle_exception(server, +@@ -2111,7 +2111,7 @@ static int _nfs4_server_capabilities(str + + int nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + do { + err = nfs4_handle_exception(server, +@@ -2145,7 +2145,7 @@ static int _nfs4_lookup_root(struct nfs_ + static int nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_fsinfo *info) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + do { + err = nfs4_handle_exception(server, +@@ -2233,7 +2233,7 @@ static int _nfs4_proc_getattr(struct nfs + + static int nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fattr *fattr) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + do { + err = nfs4_handle_exception(server, +@@ -2321,7 +2321,7 @@ static int nfs4_proc_lookupfh(struct nfs + struct qstr *name, struct nfs_fh *fhandle, + struct nfs_fattr *fattr) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + do { + err = _nfs4_proc_lookupfh(server, dirfh, name, fhandle, fattr); +@@ -2350,7 +2350,7 @@ static int _nfs4_proc_lookup(struct inod + + static int nfs4_proc_lookup(struct inode *dir, struct qstr *name, struct nfs_fh *fhandle, struct nfs_fattr *fattr) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + do { + err = nfs4_handle_exception(NFS_SERVER(dir), +@@ -2417,7 +2417,7 @@ static int _nfs4_proc_access(struct inod + + static int nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + do { + err = nfs4_handle_exception(NFS_SERVER(inode), +@@ -2473,7 +2473,7 @@ static int _nfs4_proc_readlink(struct in + static int nfs4_proc_readlink(struct inode *inode, struct page *page, + unsigned int pgbase, unsigned int pglen) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + do { + err = nfs4_handle_exception(NFS_SERVER(inode), +@@ -2568,7 +2568,7 @@ out: + + static int nfs4_proc_remove(struct inode *dir, struct qstr *name) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + do { + err = nfs4_handle_exception(NFS_SERVER(dir), +@@ -2673,7 +2673,7 @@ out: + static int nfs4_proc_rename(struct inode *old_dir, struct qstr *old_name, + struct inode *new_dir, struct qstr *new_name) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + do { + err = nfs4_handle_exception(NFS_SERVER(old_dir), +@@ -2722,7 +2722,7 @@ out: + + static int nfs4_proc_link(struct inode *inode, struct inode *dir, struct qstr *name) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + do { + err = nfs4_handle_exception(NFS_SERVER(inode), +@@ -2814,7 +2814,7 @@ out: + static int nfs4_proc_symlink(struct inode *dir, struct dentry *dentry, + struct page *page, unsigned int len, struct iattr *sattr) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + do { + err = nfs4_handle_exception(NFS_SERVER(dir), +@@ -2845,7 +2845,7 @@ out: + static int nfs4_proc_mkdir(struct inode *dir, struct dentry *dentry, + struct iattr *sattr) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + + sattr->ia_mode &= ~current_umask(); +@@ -2899,7 +2899,7 @@ static int _nfs4_proc_readdir(struct den + static int nfs4_proc_readdir(struct dentry *dentry, struct rpc_cred *cred, + u64 cookie, struct page **pages, unsigned int count, int plus) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + do { + err = nfs4_handle_exception(NFS_SERVER(dentry->d_inode), +@@ -2947,7 +2947,7 @@ out: + static int nfs4_proc_mknod(struct inode *dir, struct dentry *dentry, + struct iattr *sattr, dev_t rdev) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + + sattr->ia_mode &= ~current_umask(); +@@ -2981,7 +2981,7 @@ static int _nfs4_proc_statfs(struct nfs_ + + static int nfs4_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsstat *fsstat) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + do { + err = nfs4_handle_exception(server, +@@ -3012,7 +3012,7 @@ static int _nfs4_do_fsinfo(struct nfs_se + + static int nfs4_do_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsinfo *fsinfo) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + + do { +@@ -3058,7 +3058,7 @@ static int _nfs4_proc_pathconf(struct nf + static int nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_pathconf *pathconf) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + + do { +@@ -3404,7 +3404,7 @@ out_free: + + static ssize_t nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t buflen) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + ssize_t ret; + do { + ret = __nfs4_get_acl_uncached(inode, buf, buflen); +@@ -3479,7 +3479,7 @@ static int __nfs4_proc_set_acl(struct in + + static int nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + do { + err = nfs4_handle_exception(NFS_SERVER(inode), +@@ -3760,7 +3760,7 @@ out: + int nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, const nfs4_stateid *stateid, int issync) + { + struct nfs_server *server = NFS_SERVER(inode); +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + do { + err = _nfs4_proc_delegreturn(inode, cred, stateid, issync); +@@ -3834,7 +3834,7 @@ out: + + static int nfs4_proc_getlk(struct nfs4_state *state, int cmd, struct file_lock *request) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + + do { +@@ -4239,7 +4239,7 @@ static int _nfs4_do_setlk(struct nfs4_st + static int nfs4_lock_reclaim(struct nfs4_state *state, struct file_lock *request) + { + struct nfs_server *server = NFS_SERVER(state->inode); +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + + do { +@@ -4257,7 +4257,7 @@ static int nfs4_lock_reclaim(struct nfs4 + static int nfs4_lock_expired(struct nfs4_state *state, struct file_lock *request) + { + struct nfs_server *server = NFS_SERVER(state->inode); +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + + err = nfs4_set_lock_state(state, request); +@@ -4321,7 +4321,7 @@ out: + + static int nfs4_proc_setlk(struct nfs4_state *state, int cmd, struct file_lock *request) + { +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + + do { +@@ -4381,7 +4381,7 @@ nfs4_proc_lock(struct file *filp, int cm + int nfs4_lock_delegation_recall(struct nfs4_state *state, struct file_lock *fl) + { + struct nfs_server *server = NFS_SERVER(state->inode); +- struct nfs4_exception exception = { }; ++ struct nfs4_exception exception = {0, 0}; + int err; + + err = nfs4_set_lock_state(state, fl); +diff -urNp linux-2.6.38.3/fs/nfsd/lockd.c linux-2.6.38.3/fs/nfsd/lockd.c +--- linux-2.6.38.3/fs/nfsd/lockd.c 2011-04-17 16:53:01.000000000 -0400 ++++ linux-2.6.38.3/fs/nfsd/lockd.c 2011-04-17 16:53:48.000000000 -0400 +@@ -60,7 +60,7 @@ nlm_fclose(struct file *filp) + fput(filp); + } + +-static struct nlmsvc_binding nfsd_nlm_ops = { ++static const struct nlmsvc_binding nfsd_nlm_ops = { + .fopen = nlm_fopen, /* open file for locking */ + .fclose = nlm_fclose, /* close file */ + }; +diff -urNp linux-2.6.38.3/fs/nfsd/nfsctl.c linux-2.6.38.3/fs/nfsd/nfsctl.c +--- linux-2.6.38.3/fs/nfsd/nfsctl.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/nfsd/nfsctl.c 2011-04-17 15:57:32.000000000 -0400 +@@ -180,7 +180,7 @@ static int export_features_open(struct i + return single_open(file, export_features_show, NULL); + } + +-static struct file_operations export_features_operations = { ++static const struct file_operations export_features_operations = { + .open = export_features_open, + .read = seq_read, + .llseek = seq_lseek, +diff -urNp linux-2.6.38.3/fs/nfsd/vfs.c linux-2.6.38.3/fs/nfsd/vfs.c +--- linux-2.6.38.3/fs/nfsd/vfs.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/nfsd/vfs.c 2011-04-17 15:57:32.000000000 -0400 +@@ -898,7 +898,7 @@ nfsd_vfs_read(struct svc_rqst *rqstp, st + } else { + oldfs = get_fs(); + set_fs(KERNEL_DS); +- host_err = vfs_readv(file, (struct iovec __user *)vec, vlen, &offset); ++ host_err = vfs_readv(file, (__force struct iovec __user *)vec, vlen, &offset); + set_fs(oldfs); + } + +@@ -1002,7 +1002,7 @@ nfsd_vfs_write(struct svc_rqst *rqstp, s + + /* Write the data. */ + oldfs = get_fs(); set_fs(KERNEL_DS); +- host_err = vfs_writev(file, (struct iovec __user *)vec, vlen, &offset); ++ host_err = vfs_writev(file, (__force struct iovec __user *)vec, vlen, &offset); + set_fs(oldfs); + if (host_err < 0) + goto out_nfserr; +@@ -1518,7 +1518,7 @@ nfsd_readlink(struct svc_rqst *rqstp, st + */ + + oldfs = get_fs(); set_fs(KERNEL_DS); +- host_err = inode->i_op->readlink(dentry, buf, *lenp); ++ host_err = inode->i_op->readlink(dentry, (__force char __user *)buf, *lenp); + set_fs(oldfs); + + if (host_err < 0) +diff -urNp linux-2.6.38.3/fs/nls/nls_base.c linux-2.6.38.3/fs/nls/nls_base.c +--- linux-2.6.38.3/fs/nls/nls_base.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/nls/nls_base.c 2011-04-17 15:57:32.000000000 -0400 +@@ -41,7 +41,7 @@ static const struct utf8_table utf8_tabl + {0xF8, 0xF0, 3*6, 0x1FFFFF, 0x10000, /* 4 byte sequence */}, + {0xFC, 0xF8, 4*6, 0x3FFFFFF, 0x200000, /* 5 byte sequence */}, + {0xFE, 0xFC, 5*6, 0x7FFFFFFF, 0x4000000, /* 6 byte sequence */}, +- {0, /* end of table */} ++ {0, 0, 0, 0, 0, /* end of table */} + }; + + #define UNICODE_MAX 0x0010ffff +diff -urNp linux-2.6.38.3/fs/notify/dnotify/dnotify.c linux-2.6.38.3/fs/notify/dnotify/dnotify.c +--- linux-2.6.38.3/fs/notify/dnotify/dnotify.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/notify/dnotify/dnotify.c 2011-04-17 15:57:32.000000000 -0400 +@@ -151,7 +151,7 @@ static void dnotify_free_mark(struct fsn + kmem_cache_free(dnotify_mark_cache, dn_mark); + } + +-static struct fsnotify_ops dnotify_fsnotify_ops = { ++static const struct fsnotify_ops dnotify_fsnotify_ops = { + .handle_event = dnotify_handle_event, + .should_send_event = dnotify_should_send_event, + .free_group_priv = NULL, +diff -urNp linux-2.6.38.3/fs/ntfs/dir.c linux-2.6.38.3/fs/ntfs/dir.c +--- linux-2.6.38.3/fs/ntfs/dir.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/ntfs/dir.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1329,7 +1329,7 @@ find_next_index_buffer: + ia = (INDEX_ALLOCATION*)(kaddr + (ia_pos & ~PAGE_CACHE_MASK & + ~(s64)(ndir->itype.index.block_size - 1))); + /* Bounds checks. */ +- if (unlikely((u8*)ia < kaddr || (u8*)ia > kaddr + PAGE_CACHE_SIZE)) { ++ if (unlikely(!kaddr || (u8*)ia < kaddr || (u8*)ia > kaddr + PAGE_CACHE_SIZE)) { + ntfs_error(sb, "Out of bounds check failed. Corrupt directory " + "inode 0x%lx or driver bug.", vdir->i_ino); + goto err_out; +diff -urNp linux-2.6.38.3/fs/ntfs/file.c linux-2.6.38.3/fs/ntfs/file.c +--- linux-2.6.38.3/fs/ntfs/file.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/ntfs/file.c 2011-04-17 15:57:32.000000000 -0400 +@@ -2222,6 +2222,6 @@ const struct inode_operations ntfs_file_ + #endif /* NTFS_RW */ + }; + +-const struct file_operations ntfs_empty_file_ops = {}; ++const struct file_operations ntfs_empty_file_ops __read_only; + +-const struct inode_operations ntfs_empty_inode_ops = {}; ++const struct inode_operations ntfs_empty_inode_ops __read_only; +diff -urNp linux-2.6.38.3/fs/ocfs2/localalloc.c linux-2.6.38.3/fs/ocfs2/localalloc.c +--- linux-2.6.38.3/fs/ocfs2/localalloc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/ocfs2/localalloc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1307,7 +1307,7 @@ static int ocfs2_local_alloc_slide_windo + goto bail; + } + +- atomic_inc(&osb->alloc_stats.moves); ++ atomic_inc_unchecked(&osb->alloc_stats.moves); + + bail: + if (handle) +diff -urNp linux-2.6.38.3/fs/ocfs2/ocfs2.h linux-2.6.38.3/fs/ocfs2/ocfs2.h +--- linux-2.6.38.3/fs/ocfs2/ocfs2.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/ocfs2/ocfs2.h 2011-04-17 15:57:32.000000000 -0400 +@@ -230,11 +230,11 @@ enum ocfs2_vol_state + + struct ocfs2_alloc_stats + { +- atomic_t moves; +- atomic_t local_data; +- atomic_t bitmap_data; +- atomic_t bg_allocs; +- atomic_t bg_extends; ++ atomic_unchecked_t moves; ++ atomic_unchecked_t local_data; ++ atomic_unchecked_t bitmap_data; ++ atomic_unchecked_t bg_allocs; ++ atomic_unchecked_t bg_extends; + }; + + enum ocfs2_local_alloc_state +diff -urNp linux-2.6.38.3/fs/ocfs2/suballoc.c linux-2.6.38.3/fs/ocfs2/suballoc.c +--- linux-2.6.38.3/fs/ocfs2/suballoc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/ocfs2/suballoc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -877,7 +877,7 @@ static int ocfs2_reserve_suballoc_bits(s + mlog_errno(status); + goto bail; + } +- atomic_inc(&osb->alloc_stats.bg_extends); ++ atomic_inc_unchecked(&osb->alloc_stats.bg_extends); + + /* You should never ask for this much metadata */ + BUG_ON(bits_wanted > +@@ -2012,7 +2012,7 @@ int ocfs2_claim_metadata(handle_t *handl + mlog_errno(status); + goto bail; + } +- atomic_inc(&OCFS2_SB(ac->ac_inode->i_sb)->alloc_stats.bg_allocs); ++ atomic_inc_unchecked(&OCFS2_SB(ac->ac_inode->i_sb)->alloc_stats.bg_allocs); + + *suballoc_loc = res.sr_bg_blkno; + *suballoc_bit_start = res.sr_bit_offset; +@@ -2219,7 +2219,7 @@ int ocfs2_claim_new_inode(handle_t *hand + mlog_errno(status); + goto bail; + } +- atomic_inc(&OCFS2_SB(ac->ac_inode->i_sb)->alloc_stats.bg_allocs); ++ atomic_inc_unchecked(&OCFS2_SB(ac->ac_inode->i_sb)->alloc_stats.bg_allocs); + + BUG_ON(res.sr_bits != 1); + +@@ -2324,7 +2324,7 @@ int __ocfs2_claim_clusters(handle_t *han + cluster_start, + num_clusters); + if (!status) +- atomic_inc(&osb->alloc_stats.local_data); ++ atomic_inc_unchecked(&osb->alloc_stats.local_data); + } else { + if (min_clusters > (osb->bitmap_cpg - 1)) { + /* The only paths asking for contiguousness +@@ -2350,7 +2350,7 @@ int __ocfs2_claim_clusters(handle_t *han + ocfs2_desc_bitmap_to_cluster_off(ac->ac_inode, + res.sr_bg_blkno, + res.sr_bit_offset); +- atomic_inc(&osb->alloc_stats.bitmap_data); ++ atomic_inc_unchecked(&osb->alloc_stats.bitmap_data); + *num_clusters = res.sr_bits; + } + } +diff -urNp linux-2.6.38.3/fs/ocfs2/super.c linux-2.6.38.3/fs/ocfs2/super.c +--- linux-2.6.38.3/fs/ocfs2/super.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/ocfs2/super.c 2011-04-17 15:57:32.000000000 -0400 +@@ -297,11 +297,11 @@ static int ocfs2_osb_dump(struct ocfs2_s + "%10s => GlobalAllocs: %d LocalAllocs: %d " + "SubAllocs: %d LAWinMoves: %d SAExtends: %d\n", + "Stats", +- atomic_read(&osb->alloc_stats.bitmap_data), +- atomic_read(&osb->alloc_stats.local_data), +- atomic_read(&osb->alloc_stats.bg_allocs), +- atomic_read(&osb->alloc_stats.moves), +- atomic_read(&osb->alloc_stats.bg_extends)); ++ atomic_read_unchecked(&osb->alloc_stats.bitmap_data), ++ atomic_read_unchecked(&osb->alloc_stats.local_data), ++ atomic_read_unchecked(&osb->alloc_stats.bg_allocs), ++ atomic_read_unchecked(&osb->alloc_stats.moves), ++ atomic_read_unchecked(&osb->alloc_stats.bg_extends)); + + out += snprintf(buf + out, len - out, + "%10s => State: %u Descriptor: %llu Size: %u bits " +@@ -2141,11 +2141,11 @@ static int ocfs2_initialize_super(struct + spin_lock_init(&osb->osb_xattr_lock); + ocfs2_init_steal_slots(osb); + +- atomic_set(&osb->alloc_stats.moves, 0); +- atomic_set(&osb->alloc_stats.local_data, 0); +- atomic_set(&osb->alloc_stats.bitmap_data, 0); +- atomic_set(&osb->alloc_stats.bg_allocs, 0); +- atomic_set(&osb->alloc_stats.bg_extends, 0); ++ atomic_set_unchecked(&osb->alloc_stats.moves, 0); ++ atomic_set_unchecked(&osb->alloc_stats.local_data, 0); ++ atomic_set_unchecked(&osb->alloc_stats.bitmap_data, 0); ++ atomic_set_unchecked(&osb->alloc_stats.bg_allocs, 0); ++ atomic_set_unchecked(&osb->alloc_stats.bg_extends, 0); + + /* Copy the blockcheck stats from the superblock probe */ + osb->osb_ecc_stats = *stats; +diff -urNp linux-2.6.38.3/fs/ocfs2/symlink.c linux-2.6.38.3/fs/ocfs2/symlink.c +--- linux-2.6.38.3/fs/ocfs2/symlink.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/ocfs2/symlink.c 2011-04-17 15:57:32.000000000 -0400 +@@ -148,7 +148,7 @@ bail: + + static void ocfs2_fast_put_link(struct dentry *dentry, struct nameidata *nd, void *cookie) + { +- char *link = nd_get_link(nd); ++ const char *link = nd_get_link(nd); + if (!IS_ERR(link)) + kfree(link); + } +diff -urNp linux-2.6.38.3/fs/open.c linux-2.6.38.3/fs/open.c +--- linux-2.6.38.3/fs/open.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/open.c 2011-04-17 15:57:32.000000000 -0400 +@@ -112,6 +112,10 @@ static long do_sys_truncate(const char _ + error = locks_verify_truncate(inode, NULL, length); + if (!error) + error = security_path_truncate(&path); ++ ++ if (!error && !gr_acl_handle_truncate(path.dentry, path.mnt)) ++ error = -EACCES; ++ + if (!error) + error = do_truncate(path.dentry, length, 0, NULL); + +@@ -358,6 +362,9 @@ SYSCALL_DEFINE3(faccessat, int, dfd, con + if (__mnt_is_readonly(path.mnt)) + res = -EROFS; + ++ if (!res && !gr_acl_handle_access(path.dentry, path.mnt, mode)) ++ res = -EACCES; ++ + out_path_release: + path_put(&path); + out: +@@ -384,6 +391,8 @@ SYSCALL_DEFINE1(chdir, const char __user + if (error) + goto dput_and_out; + ++ gr_log_chdir(path.dentry, path.mnt); ++ + set_fs_pwd(current->fs, &path); + + dput_and_out: +@@ -410,6 +419,13 @@ SYSCALL_DEFINE1(fchdir, unsigned int, fd + goto out_putf; + + error = inode_permission(inode, MAY_EXEC | MAY_CHDIR); ++ ++ if (!error && !gr_chroot_fchdir(file->f_path.dentry, file->f_path.mnt)) ++ error = -EPERM; ++ ++ if (!error) ++ gr_log_chdir(file->f_path.dentry, file->f_path.mnt); ++ + if (!error) + set_fs_pwd(current->fs, &file->f_path); + out_putf: +@@ -438,7 +454,18 @@ SYSCALL_DEFINE1(chroot, const char __use + if (error) + goto dput_and_out; + ++ if (gr_handle_chroot_chroot(path.dentry, path.mnt)) ++ goto dput_and_out; ++ ++ if (gr_handle_chroot_caps(&path)) { ++ error = -ENOMEM; ++ goto dput_and_out; ++ } ++ + set_fs_root(current->fs, &path); ++ ++ gr_handle_chroot_chdir(&path); ++ + error = 0; + dput_and_out: + path_put(&path); +@@ -466,12 +493,25 @@ SYSCALL_DEFINE2(fchmod, unsigned int, fd + err = mnt_want_write_file(file); + if (err) + goto out_putf; ++ + mutex_lock(&inode->i_mutex); ++ ++ if (!gr_acl_handle_fchmod(dentry, file->f_vfsmnt, mode)) { ++ err = -EACCES; ++ goto out_unlock; ++ } ++ + err = security_path_chmod(dentry, file->f_vfsmnt, mode); + if (err) + goto out_unlock; + if (mode == (mode_t) -1) + mode = inode->i_mode; ++ ++ if (gr_handle_chroot_chmod(dentry, file->f_vfsmnt, mode)) { ++ err = -EACCES; ++ goto out_unlock; ++ } ++ + newattrs.ia_mode = (mode & S_IALLUGO) | (inode->i_mode & ~S_IALLUGO); + newattrs.ia_valid = ATTR_MODE | ATTR_CTIME; + err = notify_change(dentry, &newattrs); +@@ -499,12 +539,25 @@ SYSCALL_DEFINE3(fchmodat, int, dfd, cons + error = mnt_want_write(path.mnt); + if (error) + goto dput_and_out; ++ + mutex_lock(&inode->i_mutex); ++ ++ if (!gr_acl_handle_chmod(path.dentry, path.mnt, mode)) { ++ error = -EACCES; ++ goto out_unlock; ++ } ++ + error = security_path_chmod(path.dentry, path.mnt, mode); + if (error) + goto out_unlock; + if (mode == (mode_t) -1) + mode = inode->i_mode; ++ ++ if (gr_handle_chroot_chmod(path.dentry, path.mnt, mode)) { ++ error = -EACCES; ++ goto out_unlock; ++ } ++ + newattrs.ia_mode = (mode & S_IALLUGO) | (inode->i_mode & ~S_IALLUGO); + newattrs.ia_valid = ATTR_MODE | ATTR_CTIME; + error = notify_change(path.dentry, &newattrs); +@@ -528,6 +581,9 @@ static int chown_common(struct path *pat + int error; + struct iattr newattrs; + ++ if (!gr_acl_handle_chown(path->dentry, path->mnt)) ++ return -EACCES; ++ + newattrs.ia_valid = ATTR_CTIME; + if (user != (uid_t) -1) { + newattrs.ia_valid |= ATTR_UID; +@@ -898,7 +954,10 @@ long do_sys_open(int dfd, const char __u + if (!IS_ERR(tmp)) { + fd = get_unused_fd_flags(flags); + if (fd >= 0) { +- struct file *f = do_filp_open(dfd, tmp, flags, mode, 0); ++ struct file *f; ++ /* don't allow to be set by userland */ ++ flags &= ~FMODE_GREXEC; ++ f = do_filp_open(dfd, tmp, flags, mode, 0); + if (IS_ERR(f)) { + put_unused_fd(fd); + fd = PTR_ERR(f); +diff -urNp linux-2.6.38.3/fs/partitions/ldm.c linux-2.6.38.3/fs/partitions/ldm.c +--- linux-2.6.38.3/fs/partitions/ldm.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/partitions/ldm.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1313,7 +1313,7 @@ static bool ldm_frag_add (const u8 *data + goto found; + } + +- f = kmalloc (sizeof (*f) + size*num, GFP_KERNEL); ++ f = kmalloc (size*num + sizeof (*f), GFP_KERNEL); + if (!f) { + ldm_crit ("Out of memory."); + return false; +diff -urNp linux-2.6.38.3/fs/pipe.c linux-2.6.38.3/fs/pipe.c +--- linux-2.6.38.3/fs/pipe.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/pipe.c 2011-04-17 15:57:32.000000000 -0400 +@@ -420,9 +420,9 @@ redo: + } + if (bufs) /* More to do? */ + continue; +- if (!pipe->writers) ++ if (!atomic_read(&pipe->writers)) + break; +- if (!pipe->waiting_writers) { ++ if (!atomic_read(&pipe->waiting_writers)) { + /* syscall merging: Usually we must not sleep + * if O_NONBLOCK is set, or if we got some data. + * But if a writer sleeps in kernel space, then +@@ -481,7 +481,7 @@ pipe_write(struct kiocb *iocb, const str + mutex_lock(&inode->i_mutex); + pipe = inode->i_pipe; + +- if (!pipe->readers) { ++ if (!atomic_read(&pipe->readers)) { + send_sig(SIGPIPE, current, 0); + ret = -EPIPE; + goto out; +@@ -530,7 +530,7 @@ redo1: + for (;;) { + int bufs; + +- if (!pipe->readers) { ++ if (!atomic_read(&pipe->readers)) { + send_sig(SIGPIPE, current, 0); + if (!ret) + ret = -EPIPE; +@@ -616,9 +616,9 @@ redo2: + kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN); + do_wakeup = 0; + } +- pipe->waiting_writers++; ++ atomic_inc(&pipe->waiting_writers); + pipe_wait(pipe); +- pipe->waiting_writers--; ++ atomic_dec(&pipe->waiting_writers); + } + out: + mutex_unlock(&inode->i_mutex); +@@ -685,7 +685,7 @@ pipe_poll(struct file *filp, poll_table + mask = 0; + if (filp->f_mode & FMODE_READ) { + mask = (nrbufs > 0) ? POLLIN | POLLRDNORM : 0; +- if (!pipe->writers && filp->f_version != pipe->w_counter) ++ if (!atomic_read(&pipe->writers) && filp->f_version != pipe->w_counter) + mask |= POLLHUP; + } + +@@ -695,7 +695,7 @@ pipe_poll(struct file *filp, poll_table + * Most Unices do not set POLLERR for FIFOs but on Linux they + * behave exactly like pipes for poll(). + */ +- if (!pipe->readers) ++ if (!atomic_read(&pipe->readers)) + mask |= POLLERR; + } + +@@ -709,10 +709,10 @@ pipe_release(struct inode *inode, int de + + mutex_lock(&inode->i_mutex); + pipe = inode->i_pipe; +- pipe->readers -= decr; +- pipe->writers -= decw; ++ atomic_sub(decr, &pipe->readers); ++ atomic_sub(decw, &pipe->writers); + +- if (!pipe->readers && !pipe->writers) { ++ if (!atomic_read(&pipe->readers) && !atomic_read(&pipe->writers)) { + free_pipe_info(inode); + } else { + wake_up_interruptible_sync_poll(&pipe->wait, POLLIN | POLLOUT | POLLRDNORM | POLLWRNORM | POLLERR | POLLHUP); +@@ -802,7 +802,7 @@ pipe_read_open(struct inode *inode, stru + + if (inode->i_pipe) { + ret = 0; +- inode->i_pipe->readers++; ++ atomic_inc(&inode->i_pipe->readers); + } + + mutex_unlock(&inode->i_mutex); +@@ -819,7 +819,7 @@ pipe_write_open(struct inode *inode, str + + if (inode->i_pipe) { + ret = 0; +- inode->i_pipe->writers++; ++ atomic_inc(&inode->i_pipe->writers); + } + + mutex_unlock(&inode->i_mutex); +@@ -837,9 +837,9 @@ pipe_rdwr_open(struct inode *inode, stru + if (inode->i_pipe) { + ret = 0; + if (filp->f_mode & FMODE_READ) +- inode->i_pipe->readers++; ++ atomic_inc(&inode->i_pipe->readers); + if (filp->f_mode & FMODE_WRITE) +- inode->i_pipe->writers++; ++ atomic_inc(&inode->i_pipe->writers); + } + + mutex_unlock(&inode->i_mutex); +@@ -931,7 +931,7 @@ void free_pipe_info(struct inode *inode) + inode->i_pipe = NULL; + } + +-static struct vfsmount *pipe_mnt __read_mostly; ++struct vfsmount *pipe_mnt __read_mostly; + + /* + * pipefs_dname() is called from d_path(). +@@ -961,7 +961,8 @@ static struct inode * get_pipe_inode(voi + goto fail_iput; + inode->i_pipe = pipe; + +- pipe->readers = pipe->writers = 1; ++ atomic_set(&pipe->readers, 1); ++ atomic_set(&pipe->writers, 1); + inode->i_fop = &rdwr_pipefifo_fops; + + /* +diff -urNp linux-2.6.38.3/fs/proc/array.c linux-2.6.38.3/fs/proc/array.c +--- linux-2.6.38.3/fs/proc/array.c 2011-03-28 17:42:40.000000000 -0400 ++++ linux-2.6.38.3/fs/proc/array.c 2011-04-17 15:57:32.000000000 -0400 +@@ -60,6 +60,7 @@ + #include <linux/tty.h> + #include <linux/string.h> + #include <linux/mman.h> ++#include <linux/grsecurity.h> + #include <linux/proc_fs.h> + #include <linux/ioport.h> + #include <linux/uaccess.h> +@@ -337,6 +338,21 @@ static void task_cpus_allowed(struct seq + seq_putc(m, '\n'); + } + ++#if defined(CONFIG_PAX_NOEXEC) || defined(CONFIG_PAX_ASLR) ++static inline void task_pax(struct seq_file *m, struct task_struct *p) ++{ ++ if (p->mm) ++ seq_printf(m, "PaX:\t%c%c%c%c%c\n", ++ p->mm->pax_flags & MF_PAX_PAGEEXEC ? 'P' : 'p', ++ p->mm->pax_flags & MF_PAX_EMUTRAMP ? 'E' : 'e', ++ p->mm->pax_flags & MF_PAX_MPROTECT ? 'M' : 'm', ++ p->mm->pax_flags & MF_PAX_RANDMMAP ? 'R' : 'r', ++ p->mm->pax_flags & MF_PAX_SEGMEXEC ? 'S' : 's'); ++ else ++ seq_printf(m, "PaX:\t-----\n"); ++} ++#endif ++ + int proc_pid_status(struct seq_file *m, struct pid_namespace *ns, + struct pid *pid, struct task_struct *task) + { +@@ -354,9 +370,24 @@ int proc_pid_status(struct seq_file *m, + task_cpus_allowed(m, task); + cpuset_task_status_allowed(m, task); + task_context_switch_counts(m, task); ++ ++#if defined(CONFIG_PAX_NOEXEC) || defined(CONFIG_PAX_ASLR) ++ task_pax(m, task); ++#endif ++ ++#if defined(CONFIG_GRKERNSEC) && !defined(CONFIG_GRKERNSEC_NO_RBAC) ++ task_grsec_rbac(m, task); ++#endif ++ + return 0; + } + ++#ifdef CONFIG_GRKERNSEC_PROC_MEMMAP ++#define PAX_RAND_FLAGS(_mm) (_mm != NULL && _mm != current->mm && \ ++ (_mm->pax_flags & MF_PAX_RANDMMAP || \ ++ _mm->pax_flags & MF_PAX_SEGMEXEC)) ++#endif ++ + static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, + struct pid *pid, struct task_struct *task, int whole) + { +@@ -449,6 +480,19 @@ static int do_task_stat(struct seq_file + gtime = task->gtime; + } + ++#ifdef CONFIG_GRKERNSEC_PROC_MEMMAP ++ if (PAX_RAND_FLAGS(mm)) { ++ eip = 0; ++ esp = 0; ++ wchan = 0; ++ } ++#endif ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ wchan = 0; ++ eip =0; ++ esp =0; ++#endif ++ + /* scale priority and nice values from timeslices to -20..20 */ + /* to make it look like a "normal" Unix priority/nice value */ + priority = task_prio(task); +@@ -489,9 +533,15 @@ static int do_task_stat(struct seq_file + vsize, + mm ? get_mm_rss(mm) : 0, + rsslim, ++#ifdef CONFIG_GRKERNSEC_PROC_MEMMAP ++ PAX_RAND_FLAGS(mm) ? 1 : (mm ? (permitted ? mm->start_code : 1) : 0), ++ PAX_RAND_FLAGS(mm) ? 1 : (mm ? (permitted ? mm->end_code : 1) : 0), ++ PAX_RAND_FLAGS(mm) ? 0 : ((permitted && mm) ? mm->start_stack : 0), ++#else + mm ? (permitted ? mm->start_code : 1) : 0, + mm ? (permitted ? mm->end_code : 1) : 0, + (permitted && mm) ? mm->start_stack : 0, ++#endif + esp, + eip, + /* The signal information here is obsolete. +@@ -544,3 +594,10 @@ int proc_pid_statm(struct seq_file *m, s + + return 0; + } ++ ++#ifdef CONFIG_GRKERNSEC_PROC_IPADDR ++int proc_pid_ipaddr(struct task_struct *task, char *buffer) ++{ ++ return sprintf(buffer, "%pI4\n", &task->signal->curr_ip); ++} ++#endif +diff -urNp linux-2.6.38.3/fs/proc/base.c linux-2.6.38.3/fs/proc/base.c +--- linux-2.6.38.3/fs/proc/base.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/proc/base.c 2011-04-17 15:57:32.000000000 -0400 +@@ -104,6 +104,22 @@ struct pid_entry { + union proc_op op; + }; + ++struct getdents_callback { ++ struct linux_dirent __user * current_dir; ++ struct linux_dirent __user * previous; ++ struct file * file; ++ int count; ++ int error; ++}; ++ ++static int gr_fake_filldir(void * __buf, const char *name, int namlen, ++ loff_t offset, u64 ino, unsigned int d_type) ++{ ++ struct getdents_callback * buf = (struct getdents_callback *) __buf; ++ buf->error = -EINVAL; ++ return 0; ++} ++ + #define NOD(NAME, MODE, IOP, FOP, OP) { \ + .name = (NAME), \ + .len = sizeof(NAME) - 1, \ +@@ -203,6 +219,9 @@ static int check_mem_permission(struct t + if (task == current) + return 0; + ++ if (gr_handle_proc_ptrace(task) || gr_acl_handle_procpidmem(task)) ++ return -EPERM; ++ + /* + * If current is actively ptrace'ing, and would also be + * permitted to freshly attach with ptrace now, permit it. +@@ -250,6 +269,9 @@ static int proc_pid_cmdline(struct task_ + if (!mm->arg_end) + goto out_mm; /* Shh! No looking before we're done */ + ++ if (gr_acl_handle_procpidmem(task)) ++ goto out_mm; ++ + len = mm->arg_end - mm->arg_start; + + if (len > PAGE_SIZE) +@@ -277,12 +299,28 @@ out: + return res; + } + ++#ifdef CONFIG_GRKERNSEC_PROC_MEMMAP ++#define PAX_RAND_FLAGS(_mm) (_mm != NULL && _mm != current->mm && \ ++ (_mm->pax_flags & MF_PAX_RANDMMAP || \ ++ _mm->pax_flags & MF_PAX_SEGMEXEC)) ++#endif ++ + static int proc_pid_auxv(struct task_struct *task, char *buffer) + { + int res = 0; + struct mm_struct *mm = get_task_mm(task); + if (mm) { + unsigned int nwords = 0; ++ ++#ifdef CONFIG_GRKERNSEC_PROC_MEMMAP ++ /* allow if we're currently ptracing this task */ ++ if (PAX_RAND_FLAGS(mm) && ++ (!(task->ptrace & PT_PTRACED) || (task->parent != current))) { ++ mmput(mm); ++ return res; ++ } ++#endif ++ + do { + nwords += 2; + } while (mm->saved_auxv[nwords - 2] != 0); /* AT_NULL */ +@@ -296,7 +334,7 @@ static int proc_pid_auxv(struct task_str + } + + +-#ifdef CONFIG_KALLSYMS ++#if defined(CONFIG_KALLSYMS) && !defined(CONFIG_GRKERNSEC_HIDESYM) + /* + * Provides a wchan file via kallsyms in a proper one-value-per-file format. + * Returns the resolved symbol. If that fails, simply return the address. +@@ -318,7 +356,7 @@ static int proc_pid_wchan(struct task_st + } + #endif /* CONFIG_KALLSYMS */ + +-#ifdef CONFIG_STACKTRACE ++#if defined(CONFIG_STACKTRACE) && !defined(CONFIG_GRKERNSEC_HIDESYM) + + #define MAX_STACK_TRACE_DEPTH 64 + +@@ -503,7 +541,7 @@ static int proc_pid_limits(struct task_s + return count; + } + +-#ifdef CONFIG_HAVE_ARCH_TRACEHOOK ++#if defined(CONFIG_HAVE_ARCH_TRACEHOOK) && !defined(CONFIG_GRKERNSEC_PROC_MEMMAP) + static int proc_pid_syscall(struct task_struct *task, char *buffer) + { + long nr; +@@ -528,7 +566,7 @@ static int proc_pid_syscall(struct task_ + /************************************************************************/ + + /* permission checks */ +-static int proc_fd_access_allowed(struct inode *inode) ++static int proc_fd_access_allowed(struct inode *inode, unsigned int log) + { + struct task_struct *task; + int allowed = 0; +@@ -538,7 +576,10 @@ static int proc_fd_access_allowed(struct + */ + task = get_proc_task(inode); + if (task) { +- allowed = ptrace_may_access(task, PTRACE_MODE_READ); ++ if (log) ++ allowed = ptrace_may_access_log(task, PTRACE_MODE_READ); ++ else ++ allowed = ptrace_may_access(task, PTRACE_MODE_READ); + put_task_struct(task); + } + return allowed; +@@ -917,6 +958,9 @@ static ssize_t environ_read(struct file + if (!task) + goto out_no_task; + ++ if (gr_acl_handle_procpidmem(task)) ++ goto out; ++ + if (!ptrace_may_access(task, PTRACE_MODE_READ)) + goto out; + +@@ -1606,7 +1650,7 @@ static void *proc_pid_follow_link(struct + path_put(&nd->path); + + /* Are we allowed to snoop on the tasks file descriptors? */ +- if (!proc_fd_access_allowed(inode)) ++ if (!proc_fd_access_allowed(inode,0)) + goto out; + + error = PROC_I(inode)->op.proc_get_link(inode, &nd->path); +@@ -1645,8 +1689,18 @@ static int proc_pid_readlink(struct dent + struct path path; + + /* Are we allowed to snoop on the tasks file descriptors? */ +- if (!proc_fd_access_allowed(inode)) +- goto out; ++ /* logging this is needed for learning on chromium to work properly, ++ but we don't want to flood the logs from 'ps' which does a readlink ++ on /proc/fd/2 of tasks in the listing, nor do we want 'ps' to learn ++ CAP_SYS_PTRACE as it's not necessary for its basic functionality ++ */ ++ if (dentry->d_name.name[0] == '2' && dentry->d_name.name[1] == '\0') { ++ if (!proc_fd_access_allowed(inode,0)) ++ goto out; ++ } else { ++ if (!proc_fd_access_allowed(inode,1)) ++ goto out; ++ } + + error = PROC_I(inode)->op.proc_get_link(inode, &path); + if (error) +@@ -1712,7 +1766,11 @@ static struct inode *proc_pid_make_inode + rcu_read_lock(); + cred = __task_cred(task); + inode->i_uid = cred->euid; ++#ifdef CONFIG_GRKERNSEC_PROC_USERGROUP ++ inode->i_gid = CONFIG_GRKERNSEC_PROC_GID; ++#else + inode->i_gid = cred->egid; ++#endif + rcu_read_unlock(); + } + security_task_to_inode(task, inode); +@@ -1730,6 +1788,9 @@ static int pid_getattr(struct vfsmount * + struct inode *inode = dentry->d_inode; + struct task_struct *task; + const struct cred *cred; ++#if defined(CONFIG_GRKERNSEC_PROC_USER) || defined(CONFIG_GRKERNSEC_PROC_USERGROUP) ++ const struct cred *tmpcred = current_cred(); ++#endif + + generic_fillattr(inode, stat); + +@@ -1737,12 +1798,34 @@ static int pid_getattr(struct vfsmount * + stat->uid = 0; + stat->gid = 0; + task = pid_task(proc_pid(inode), PIDTYPE_PID); ++ ++ if (task && (gr_pid_is_chrooted(task) || gr_check_hidden_task(task))) { ++ rcu_read_unlock(); ++ return -ENOENT; ++ } ++ + if (task) { ++ cred = __task_cred(task); ++#if defined(CONFIG_GRKERNSEC_PROC_USER) || defined(CONFIG_GRKERNSEC_PROC_USERGROUP) ++ if (!tmpcred->uid || (tmpcred->uid == cred->uid) ++#ifdef CONFIG_GRKERNSEC_PROC_USERGROUP ++ || in_group_p(CONFIG_GRKERNSEC_PROC_GID) ++#endif ++ ) ++#endif + if ((inode->i_mode == (S_IFDIR|S_IRUGO|S_IXUGO)) || ++#ifdef CONFIG_GRKERNSEC_PROC_USER ++ (inode->i_mode == (S_IFDIR|S_IRUSR|S_IXUSR)) || ++#elif defined(CONFIG_GRKERNSEC_PROC_USERGROUP) ++ (inode->i_mode == (S_IFDIR|S_IRUSR|S_IRGRP|S_IXUSR|S_IXGRP)) || ++#endif + task_dumpable(task)) { +- cred = __task_cred(task); + stat->uid = cred->euid; ++#ifdef CONFIG_GRKERNSEC_PROC_USERGROUP ++ stat->gid = CONFIG_GRKERNSEC_PROC_GID; ++#else + stat->gid = cred->egid; ++#endif + } + } + rcu_read_unlock(); +@@ -1780,11 +1863,20 @@ static int pid_revalidate(struct dentry + + if (task) { + if ((inode->i_mode == (S_IFDIR|S_IRUGO|S_IXUGO)) || ++#ifdef CONFIG_GRKERNSEC_PROC_USER ++ (inode->i_mode == (S_IFDIR|S_IRUSR|S_IXUSR)) || ++#elif defined(CONFIG_GRKERNSEC_PROC_USERGROUP) ++ (inode->i_mode == (S_IFDIR|S_IRUSR|S_IRGRP|S_IXUSR|S_IXGRP)) || ++#endif + task_dumpable(task)) { + rcu_read_lock(); + cred = __task_cred(task); + inode->i_uid = cred->euid; ++#ifdef CONFIG_GRKERNSEC_PROC_USERGROUP ++ inode->i_gid = CONFIG_GRKERNSEC_PROC_GID; ++#else + inode->i_gid = cred->egid; ++#endif + rcu_read_unlock(); + } else { + inode->i_uid = 0; +@@ -1905,7 +1997,8 @@ static int proc_fd_info(struct inode *in + int fd = proc_fd(inode); + + if (task) { +- files = get_files_struct(task); ++ if (!gr_acl_handle_procpidmem(task)) ++ files = get_files_struct(task); + put_task_struct(task); + } + if (files) { +@@ -2165,15 +2258,25 @@ static const struct file_operations proc + */ + static int proc_fd_permission(struct inode *inode, int mask, unsigned int flags) + { ++ struct task_struct *task; + int rv; + + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + rv = generic_permission(inode, mask, flags, NULL); +- if (rv == 0) +- return 0; ++ + if (task_pid(current) == proc_pid(inode)) + rv = 0; ++ ++ task = get_proc_task(inode); ++ if (task == NULL) ++ return rv; ++ ++ if (gr_acl_handle_procpidmem(task)) ++ rv = -EACCES; ++ ++ put_task_struct(task); ++ + return rv; + } + +@@ -2283,6 +2386,9 @@ static struct dentry *proc_pident_lookup + if (!task) + goto out_no_task; + ++ if (gr_pid_is_chrooted(task) || gr_check_hidden_task(task)) ++ goto out; ++ + /* + * Yes, it does not scale. And it should not. Don't add + * new entries into /proc/<tgid>/ without very good reasons. +@@ -2327,6 +2433,9 @@ static int proc_pident_readdir(struct fi + if (!task) + goto out_no_task; + ++ if (gr_pid_is_chrooted(task) || gr_check_hidden_task(task)) ++ goto out; ++ + ret = 0; + i = filp->f_pos; + switch (i) { +@@ -2597,7 +2706,7 @@ static void *proc_self_follow_link(struc + static void proc_self_put_link(struct dentry *dentry, struct nameidata *nd, + void *cookie) + { +- char *s = nd_get_link(nd); ++ const char *s = nd_get_link(nd); + if (!IS_ERR(s)) + __putname(s); + } +@@ -2777,7 +2886,7 @@ static const struct pid_entry tgid_base_ + REG("autogroup", S_IRUGO|S_IWUSR, proc_pid_sched_autogroup_operations), + #endif + REG("comm", S_IRUGO|S_IWUSR, proc_pid_set_comm_operations), +-#ifdef CONFIG_HAVE_ARCH_TRACEHOOK ++#if defined(CONFIG_HAVE_ARCH_TRACEHOOK) && !defined(CONFIG_GRKERNSEC_PROC_MEMMAP) + INF("syscall", S_IRUSR, proc_pid_syscall), + #endif + INF("cmdline", S_IRUGO, proc_pid_cmdline), +@@ -2802,10 +2911,10 @@ static const struct pid_entry tgid_base_ + #ifdef CONFIG_SECURITY + DIR("attr", S_IRUGO|S_IXUGO, proc_attr_dir_inode_operations, proc_attr_dir_operations), + #endif +-#ifdef CONFIG_KALLSYMS ++#if defined(CONFIG_KALLSYMS) && !defined(CONFIG_GRKERNSEC_HIDESYM) + INF("wchan", S_IRUGO, proc_pid_wchan), + #endif +-#ifdef CONFIG_STACKTRACE ++#if defined(CONFIG_STACKTRACE) && !defined(CONFIG_GRKERNSEC_HIDESYM) + ONE("stack", S_IRUSR, proc_pid_stack), + #endif + #ifdef CONFIG_SCHEDSTATS +@@ -2836,6 +2945,9 @@ static const struct pid_entry tgid_base_ + #ifdef CONFIG_TASK_IO_ACCOUNTING + INF("io", S_IRUGO, proc_tgid_io_accounting), + #endif ++#ifdef CONFIG_GRKERNSEC_PROC_IPADDR ++ INF("ipaddr", S_IRUSR, proc_pid_ipaddr), ++#endif + }; + + static int proc_tgid_base_readdir(struct file * filp, +@@ -2961,7 +3073,14 @@ static struct dentry *proc_pid_instantia + if (!inode) + goto out; + ++#ifdef CONFIG_GRKERNSEC_PROC_USER ++ inode->i_mode = S_IFDIR|S_IRUSR|S_IXUSR; ++#elif defined(CONFIG_GRKERNSEC_PROC_USERGROUP) ++ inode->i_gid = CONFIG_GRKERNSEC_PROC_GID; ++ inode->i_mode = S_IFDIR|S_IRUSR|S_IRGRP|S_IXUSR|S_IXGRP; ++#else + inode->i_mode = S_IFDIR|S_IRUGO|S_IXUGO; ++#endif + inode->i_op = &proc_tgid_base_inode_operations; + inode->i_fop = &proc_tgid_base_operations; + inode->i_flags|=S_IMMUTABLE; +@@ -3003,7 +3122,11 @@ struct dentry *proc_pid_lookup(struct in + if (!task) + goto out; + ++ if (gr_pid_is_chrooted(task) || gr_check_hidden_task(task)) ++ goto out_put_task; ++ + result = proc_pid_instantiate(dir, dentry, task, NULL); ++out_put_task: + put_task_struct(task); + out: + return result; +@@ -3068,6 +3191,11 @@ int proc_pid_readdir(struct file * filp, + { + unsigned int nr = filp->f_pos - FIRST_PROCESS_ENTRY; + struct task_struct *reaper = get_proc_task(filp->f_path.dentry->d_inode); ++#if defined(CONFIG_GRKERNSEC_PROC_USER) || defined(CONFIG_GRKERNSEC_PROC_USERGROUP) ++ const struct cred *tmpcred = current_cred(); ++ const struct cred *itercred; ++#endif ++ filldir_t __filldir = filldir; + struct tgid_iter iter; + struct pid_namespace *ns; + +@@ -3086,8 +3214,27 @@ int proc_pid_readdir(struct file * filp, + for (iter = next_tgid(ns, iter); + iter.task; + iter.tgid += 1, iter = next_tgid(ns, iter)) { ++#if defined(CONFIG_GRKERNSEC_PROC_USER) || defined(CONFIG_GRKERNSEC_PROC_USERGROUP) ++ rcu_read_lock(); ++ itercred = __task_cred(iter.task); ++#endif ++ if (gr_pid_is_chrooted(iter.task) || gr_check_hidden_task(iter.task) ++#if defined(CONFIG_GRKERNSEC_PROC_USER) || defined(CONFIG_GRKERNSEC_PROC_USERGROUP) ++ || (tmpcred->uid && (itercred->uid != tmpcred->uid) ++#ifdef CONFIG_GRKERNSEC_PROC_USERGROUP ++ && !in_group_p(CONFIG_GRKERNSEC_PROC_GID) ++#endif ++ ) ++#endif ++ ) ++ __filldir = &gr_fake_filldir; ++ else ++ __filldir = filldir; ++#if defined(CONFIG_GRKERNSEC_PROC_USER) || defined(CONFIG_GRKERNSEC_PROC_USERGROUP) ++ rcu_read_unlock(); ++#endif + filp->f_pos = iter.tgid + TGID_OFFSET; +- if (proc_pid_fill_cache(filp, dirent, filldir, iter) < 0) { ++ if (proc_pid_fill_cache(filp, dirent, __filldir, iter) < 0) { + put_task_struct(iter.task); + goto out; + } +@@ -3114,7 +3261,7 @@ static const struct pid_entry tid_base_s + REG("sched", S_IRUGO|S_IWUSR, proc_pid_sched_operations), + #endif + REG("comm", S_IRUGO|S_IWUSR, proc_pid_set_comm_operations), +-#ifdef CONFIG_HAVE_ARCH_TRACEHOOK ++#if defined(CONFIG_HAVE_ARCH_TRACEHOOK) && !defined(CONFIG_GRKERNSEC_PROC_MEMMAP) + INF("syscall", S_IRUSR, proc_pid_syscall), + #endif + INF("cmdline", S_IRUGO, proc_pid_cmdline), +@@ -3138,10 +3285,10 @@ static const struct pid_entry tid_base_s + #ifdef CONFIG_SECURITY + DIR("attr", S_IRUGO|S_IXUGO, proc_attr_dir_inode_operations, proc_attr_dir_operations), + #endif +-#ifdef CONFIG_KALLSYMS ++#if defined(CONFIG_KALLSYMS) && !defined(CONFIG_GRKERNSEC_HIDESYM) + INF("wchan", S_IRUGO, proc_pid_wchan), + #endif +-#ifdef CONFIG_STACKTRACE ++#if defined(CONFIG_STACKTRACE) && !defined(CONFIG_GRKERNSEC_HIDESYM) + ONE("stack", S_IRUSR, proc_pid_stack), + #endif + #ifdef CONFIG_SCHEDSTATS +diff -urNp linux-2.6.38.3/fs/proc/cmdline.c linux-2.6.38.3/fs/proc/cmdline.c +--- linux-2.6.38.3/fs/proc/cmdline.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/proc/cmdline.c 2011-04-17 15:57:32.000000000 -0400 +@@ -23,7 +23,11 @@ static const struct file_operations cmdl + + static int __init proc_cmdline_init(void) + { ++#ifdef CONFIG_GRKERNSEC_PROC_ADD ++ proc_create_grsec("cmdline", 0, NULL, &cmdline_proc_fops); ++#else + proc_create("cmdline", 0, NULL, &cmdline_proc_fops); ++#endif + return 0; + } + module_init(proc_cmdline_init); +diff -urNp linux-2.6.38.3/fs/proc/devices.c linux-2.6.38.3/fs/proc/devices.c +--- linux-2.6.38.3/fs/proc/devices.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/proc/devices.c 2011-04-17 15:57:32.000000000 -0400 +@@ -64,7 +64,11 @@ static const struct file_operations proc + + static int __init proc_devices_init(void) + { ++#ifdef CONFIG_GRKERNSEC_PROC_ADD ++ proc_create_grsec("devices", 0, NULL, &proc_devinfo_operations); ++#else + proc_create("devices", 0, NULL, &proc_devinfo_operations); ++#endif + return 0; + } + module_init(proc_devices_init); +diff -urNp linux-2.6.38.3/fs/proc/inode.c linux-2.6.38.3/fs/proc/inode.c +--- linux-2.6.38.3/fs/proc/inode.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/proc/inode.c 2011-04-17 15:57:32.000000000 -0400 +@@ -435,7 +435,11 @@ struct inode *proc_get_inode(struct supe + if (de->mode) { + inode->i_mode = de->mode; + inode->i_uid = de->uid; ++#ifdef CONFIG_GRKERNSEC_PROC_USERGROUP ++ inode->i_gid = CONFIG_GRKERNSEC_PROC_GID; ++#else + inode->i_gid = de->gid; ++#endif + } + if (de->size) + inode->i_size = de->size; +diff -urNp linux-2.6.38.3/fs/proc/internal.h linux-2.6.38.3/fs/proc/internal.h +--- linux-2.6.38.3/fs/proc/internal.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/proc/internal.h 2011-04-17 15:57:32.000000000 -0400 +@@ -51,6 +51,9 @@ extern int proc_pid_status(struct seq_fi + struct pid *pid, struct task_struct *task); + extern int proc_pid_statm(struct seq_file *m, struct pid_namespace *ns, + struct pid *pid, struct task_struct *task); ++#ifdef CONFIG_GRKERNSEC_PROC_IPADDR ++extern int proc_pid_ipaddr(struct task_struct *task, char *buffer); ++#endif + extern loff_t mem_lseek(struct file *file, loff_t offset, int orig); + + extern const struct file_operations proc_maps_operations; +diff -urNp linux-2.6.38.3/fs/proc/Kconfig linux-2.6.38.3/fs/proc/Kconfig +--- linux-2.6.38.3/fs/proc/Kconfig 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/proc/Kconfig 2011-04-17 15:57:32.000000000 -0400 +@@ -30,12 +30,12 @@ config PROC_FS + + config PROC_KCORE + bool "/proc/kcore support" if !ARM +- depends on PROC_FS && MMU ++ depends on PROC_FS && MMU && !GRKERNSEC_PROC_ADD + + config PROC_VMCORE + bool "/proc/vmcore support" +- depends on PROC_FS && CRASH_DUMP +- default y ++ depends on PROC_FS && CRASH_DUMP && !GRKERNSEC ++ default n + help + Exports the dump image of crashed kernel in ELF format. + +@@ -59,8 +59,8 @@ config PROC_SYSCTL + limited in memory. + + config PROC_PAGE_MONITOR +- default y +- depends on PROC_FS && MMU ++ default n ++ depends on PROC_FS && MMU && !GRKERNSEC + bool "Enable /proc page monitoring" if EXPERT + help + Various /proc files exist to monitor process memory utilization: +diff -urNp linux-2.6.38.3/fs/proc/kcore.c linux-2.6.38.3/fs/proc/kcore.c +--- linux-2.6.38.3/fs/proc/kcore.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/proc/kcore.c 2011-04-17 15:57:32.000000000 -0400 +@@ -478,9 +478,10 @@ read_kcore(struct file *file, char __use + * the addresses in the elf_phdr on our list. + */ + start = kc_offset_to_vaddr(*fpos - elf_buflen); +- if ((tsz = (PAGE_SIZE - (start & ~PAGE_MASK))) > buflen) ++ tsz = PAGE_SIZE - (start & ~PAGE_MASK); ++ if (tsz > buflen) + tsz = buflen; +- ++ + while (buflen) { + struct kcore_list *m; + +@@ -509,20 +510,23 @@ read_kcore(struct file *file, char __use + kfree(elf_buf); + } else { + if (kern_addr_valid(start)) { +- unsigned long n; ++ char *elf_buf; ++ mm_segment_t oldfs; + +- n = copy_to_user(buffer, (char *)start, tsz); +- /* +- * We cannot distingush between fault on source +- * and fault on destination. When this happens +- * we clear too and hope it will trigger the +- * EFAULT again. +- */ +- if (n) { +- if (clear_user(buffer + tsz - n, +- n)) ++ elf_buf = kmalloc(tsz, GFP_KERNEL); ++ if (!elf_buf) ++ return -ENOMEM; ++ oldfs = get_fs(); ++ set_fs(KERNEL_DS); ++ if (!__copy_from_user(elf_buf, (const void __user *)start, tsz)) { ++ set_fs(oldfs); ++ if (copy_to_user(buffer, elf_buf, tsz)) { ++ kfree(elf_buf); + return -EFAULT; ++ } + } ++ set_fs(oldfs); ++ kfree(elf_buf); + } else { + if (clear_user(buffer, tsz)) + return -EFAULT; +@@ -542,6 +546,9 @@ read_kcore(struct file *file, char __use + + static int open_kcore(struct inode *inode, struct file *filp) + { ++#if defined(CONFIG_GRKERNSEC_PROC_ADD) || defined(CONFIG_GRKERNSEC_HIDESYM) ++ return -EPERM; ++#endif + if (!capable(CAP_SYS_RAWIO)) + return -EPERM; + if (kcore_need_update) +diff -urNp linux-2.6.38.3/fs/proc/meminfo.c linux-2.6.38.3/fs/proc/meminfo.c +--- linux-2.6.38.3/fs/proc/meminfo.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/proc/meminfo.c 2011-04-17 15:57:32.000000000 -0400 +@@ -157,7 +157,7 @@ static int meminfo_proc_show(struct seq_ + vmi.used >> 10, + vmi.largest_chunk >> 10 + #ifdef CONFIG_MEMORY_FAILURE +- ,atomic_long_read(&mce_bad_pages) << (PAGE_SHIFT - 10) ++ ,atomic_long_read_unchecked(&mce_bad_pages) << (PAGE_SHIFT - 10) + #endif + #ifdef CONFIG_TRANSPARENT_HUGEPAGE + ,K(global_page_state(NR_ANON_TRANSPARENT_HUGEPAGES) * +diff -urNp linux-2.6.38.3/fs/proc/nommu.c linux-2.6.38.3/fs/proc/nommu.c +--- linux-2.6.38.3/fs/proc/nommu.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/proc/nommu.c 2011-04-17 15:57:32.000000000 -0400 +@@ -66,7 +66,7 @@ static int nommu_region_show(struct seq_ + if (len < 1) + len = 1; + seq_printf(m, "%*c", len, ' '); +- seq_path(m, &file->f_path, ""); ++ seq_path(m, &file->f_path, "\n\\"); + } + + seq_putc(m, '\n'); +diff -urNp linux-2.6.38.3/fs/proc/proc_net.c linux-2.6.38.3/fs/proc/proc_net.c +--- linux-2.6.38.3/fs/proc/proc_net.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/proc/proc_net.c 2011-04-17 15:57:32.000000000 -0400 +@@ -105,6 +105,17 @@ static struct net *get_proc_task_net(str + struct task_struct *task; + struct nsproxy *ns; + struct net *net = NULL; ++#if defined(CONFIG_GRKERNSEC_PROC_USER) || defined(CONFIG_GRKERNSEC_PROC_USERGROUP) ++ const struct cred *cred = current_cred(); ++#endif ++ ++#ifdef CONFIG_GRKERNSEC_PROC_USER ++ if (cred->fsuid) ++ return net; ++#elif defined(CONFIG_GRKERNSEC_PROC_USERGROUP) ++ if (cred->fsuid && !in_group_p(CONFIG_GRKERNSEC_PROC_GID)) ++ return net; ++#endif + + rcu_read_lock(); + task = pid_task(proc_pid(dir), PIDTYPE_PID); +diff -urNp linux-2.6.38.3/fs/proc/proc_sysctl.c linux-2.6.38.3/fs/proc/proc_sysctl.c +--- linux-2.6.38.3/fs/proc/proc_sysctl.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/proc/proc_sysctl.c 2011-04-17 15:57:32.000000000 -0400 +@@ -8,6 +8,8 @@ + #include <linux/namei.h> + #include "internal.h" + ++extern __u32 gr_handle_sysctl(const struct ctl_table *table, const int op); ++ + static const struct dentry_operations proc_sys_dentry_operations; + static const struct file_operations proc_sys_file_operations; + static const struct inode_operations proc_sys_inode_operations; +@@ -112,6 +114,9 @@ static struct dentry *proc_sys_lookup(st + if (!p) + goto out; + ++ if (gr_handle_sysctl(p, MAY_EXEC)) ++ goto out; ++ + err = ERR_PTR(-ENOMEM); + inode = proc_sys_make_inode(dir->i_sb, h ? h : head, p); + if (h) +@@ -231,6 +236,9 @@ static int scan(struct ctl_table_header + if (*pos < file->f_pos) + continue; + ++ if (gr_handle_sysctl(table, 0)) ++ continue; ++ + res = proc_sys_fill_cache(file, dirent, filldir, head, table); + if (res) + return res; +@@ -359,6 +367,9 @@ static int proc_sys_getattr(struct vfsmo + if (IS_ERR(head)) + return PTR_ERR(head); + ++ if (table && gr_handle_sysctl(table, MAY_EXEC)) ++ return -ENOENT; ++ + generic_fillattr(inode, stat); + if (table) + stat->mode = (stat->mode & S_IFMT) | table->mode; +diff -urNp linux-2.6.38.3/fs/proc/root.c linux-2.6.38.3/fs/proc/root.c +--- linux-2.6.38.3/fs/proc/root.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/proc/root.c 2011-04-17 15:57:32.000000000 -0400 +@@ -132,7 +132,15 @@ void __init proc_root_init(void) + #ifdef CONFIG_PROC_DEVICETREE + proc_device_tree_init(); + #endif ++#ifdef CONFIG_GRKERNSEC_PROC_ADD ++#ifdef CONFIG_GRKERNSEC_PROC_USER ++ proc_mkdir_mode("bus", S_IRUSR | S_IXUSR, NULL); ++#elif defined(CONFIG_GRKERNSEC_PROC_USERGROUP) ++ proc_mkdir_mode("bus", S_IRUSR | S_IXUSR | S_IRGRP | S_IXGRP, NULL); ++#endif ++#else + proc_mkdir("bus", NULL); ++#endif + proc_sys_init(); + } + +diff -urNp linux-2.6.38.3/fs/proc/task_mmu.c linux-2.6.38.3/fs/proc/task_mmu.c +--- linux-2.6.38.3/fs/proc/task_mmu.c 2011-03-28 17:42:40.000000000 -0400 ++++ linux-2.6.38.3/fs/proc/task_mmu.c 2011-04-17 15:57:32.000000000 -0400 +@@ -49,8 +49,13 @@ void task_mem(struct seq_file *m, struct + "VmExe:\t%8lu kB\n" + "VmLib:\t%8lu kB\n" + "VmPTE:\t%8lu kB\n" +- "VmSwap:\t%8lu kB\n", +- hiwater_vm << (PAGE_SHIFT-10), ++ "VmSwap:\t%8lu kB\n" ++ ++#ifdef CONFIG_ARCH_TRACK_EXEC_LIMIT ++ "CsBase:\t%8lx\nCsLim:\t%8lx\n" ++#endif ++ ++ ,hiwater_vm << (PAGE_SHIFT-10), + (total_vm - mm->reserved_vm) << (PAGE_SHIFT-10), + mm->locked_vm << (PAGE_SHIFT-10), + hiwater_rss << (PAGE_SHIFT-10), +@@ -58,7 +63,13 @@ void task_mem(struct seq_file *m, struct + data << (PAGE_SHIFT-10), + mm->stack_vm << (PAGE_SHIFT-10), text, lib, + (PTRS_PER_PTE*sizeof(pte_t)*mm->nr_ptes) >> 10, +- swap << (PAGE_SHIFT-10)); ++ swap << (PAGE_SHIFT-10) ++ ++#ifdef CONFIG_ARCH_TRACK_EXEC_LIMIT ++ , mm->context.user_cs_base, mm->context.user_cs_limit ++#endif ++ ++ ); + } + + unsigned long task_vsize(struct mm_struct *mm) +@@ -204,6 +215,12 @@ static int do_maps_open(struct inode *in + return ret; + } + ++#ifdef CONFIG_GRKERNSEC_PROC_MEMMAP ++#define PAX_RAND_FLAGS(_mm) (_mm != NULL && _mm != current->mm && \ ++ (_mm->pax_flags & MF_PAX_RANDMMAP || \ ++ _mm->pax_flags & MF_PAX_SEGMEXEC)) ++#endif ++ + static void show_map_vma(struct seq_file *m, struct vm_area_struct *vma) + { + struct mm_struct *mm = vma->vm_mm; +@@ -211,7 +228,6 @@ static void show_map_vma(struct seq_file + int flags = vma->vm_flags; + unsigned long ino = 0; + unsigned long long pgoff = 0; +- unsigned long start; + dev_t dev = 0; + int len; + +@@ -222,20 +238,23 @@ static void show_map_vma(struct seq_file + pgoff = ((loff_t)vma->vm_pgoff) << PAGE_SHIFT; + } + +- /* We don't show the stack guard page in /proc/maps */ +- start = vma->vm_start; +- if (vma->vm_flags & VM_GROWSDOWN) +- if (!vma_stack_continue(vma->vm_prev, vma->vm_start)) +- start += PAGE_SIZE; +- + seq_printf(m, "%08lx-%08lx %c%c%c%c %08llx %02x:%02x %lu %n", +- start, ++#ifdef CONFIG_GRKERNSEC_PROC_MEMMAP ++ PAX_RAND_FLAGS(mm) ? 0UL : vma->vm_start, ++ PAX_RAND_FLAGS(mm) ? 0UL : vma->vm_end, ++#else ++ vma->vm_start, + vma->vm_end, ++#endif + flags & VM_READ ? 'r' : '-', + flags & VM_WRITE ? 'w' : '-', + flags & VM_EXEC ? 'x' : '-', + flags & VM_MAYSHARE ? 's' : 'p', ++#ifdef CONFIG_GRKERNSEC_PROC_MEMMAP ++ PAX_RAND_FLAGS(mm) ? 0UL : pgoff, ++#else + pgoff, ++#endif + MAJOR(dev), MINOR(dev), ino, &len); + + /* +@@ -244,16 +263,16 @@ static void show_map_vma(struct seq_file + */ + if (file) { + pad_len_spaces(m, len); +- seq_path(m, &file->f_path, "\n"); ++ seq_path(m, &file->f_path, "\n\\"); + } else { + const char *name = arch_vma_name(vma); + if (!name) { + if (mm) { +- if (vma->vm_start <= mm->brk && +- vma->vm_end >= mm->start_brk) { ++ if (vma->vm_start <= mm->brk && vma->vm_end >= mm->start_brk) { + name = "[heap]"; +- } else if (vma->vm_start <= mm->start_stack && +- vma->vm_end >= mm->start_stack) { ++ } else if ((vma->vm_flags & (VM_GROWSDOWN | VM_GROWSUP)) || ++ (vma->vm_start <= mm->start_stack && ++ vma->vm_end >= mm->start_stack)) { + name = "[stack]"; + } + } else { +@@ -399,11 +418,16 @@ static int show_smap(struct seq_file *m, + }; + + memset(&mss, 0, sizeof mss); +- mss.vma = vma; +- /* mmap_sem is held in m_start */ +- if (vma->vm_mm && !is_vm_hugetlb_page(vma)) +- walk_page_range(vma->vm_start, vma->vm_end, &smaps_walk); +- ++#ifdef CONFIG_GRKERNSEC_PROC_MEMMAP ++ if (!PAX_RAND_FLAGS(vma->vm_mm)) { ++#endif ++ mss.vma = vma; ++ /* mmap_sem is held in m_start */ ++ if (vma->vm_mm && !is_vm_hugetlb_page(vma)) ++ walk_page_range(vma->vm_start, vma->vm_end, &smaps_walk); ++#ifdef CONFIG_GRKERNSEC_PROC_MEMMAP ++ } ++#endif + show_map_vma(m, vma); + + seq_printf(m, +@@ -420,7 +444,11 @@ static int show_smap(struct seq_file *m, + "KernelPageSize: %8lu kB\n" + "MMUPageSize: %8lu kB\n" + "Locked: %8lu kB\n", ++#ifdef CONFIG_GRKERNSEC_PROC_MEMMAP ++ PAX_RAND_FLAGS(vma->vm_mm) ? 0UL : (vma->vm_end - vma->vm_start) >> 10, ++#else + (vma->vm_end - vma->vm_start) >> 10, ++#endif + mss.resident >> 10, + (unsigned long)(mss.pss >> (10 + PSS_SHIFT)), + mss.shared_clean >> 10, +diff -urNp linux-2.6.38.3/fs/proc/task_nommu.c linux-2.6.38.3/fs/proc/task_nommu.c +--- linux-2.6.38.3/fs/proc/task_nommu.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/proc/task_nommu.c 2011-04-17 15:57:32.000000000 -0400 +@@ -51,7 +51,7 @@ void task_mem(struct seq_file *m, struct + else + bytes += kobjsize(mm); + +- if (current->fs && current->fs->users > 1) ++ if (current->fs && atomic_read(¤t->fs->users) > 1) + sbytes += kobjsize(current->fs); + else + bytes += kobjsize(current->fs); +@@ -166,7 +166,7 @@ static int nommu_vma_show(struct seq_fil + + if (file) { + pad_len_spaces(m, len); +- seq_path(m, &file->f_path, ""); ++ seq_path(m, &file->f_path, "\n\\"); + } else if (mm) { + if (vma->vm_start <= mm->start_stack && + vma->vm_end >= mm->start_stack) { +diff -urNp linux-2.6.38.3/fs/readdir.c linux-2.6.38.3/fs/readdir.c +--- linux-2.6.38.3/fs/readdir.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/readdir.c 2011-04-17 15:57:32.000000000 -0400 +@@ -17,6 +17,7 @@ + #include <linux/security.h> + #include <linux/syscalls.h> + #include <linux/unistd.h> ++#include <linux/namei.h> + + #include <asm/uaccess.h> + +@@ -67,6 +68,7 @@ struct old_linux_dirent { + + struct readdir_callback { + struct old_linux_dirent __user * dirent; ++ struct file * file; + int result; + }; + +@@ -84,6 +86,10 @@ static int fillonedir(void * __buf, cons + buf->result = -EOVERFLOW; + return -EOVERFLOW; + } ++ ++ if (!gr_acl_handle_filldir(buf->file, name, namlen, ino)) ++ return 0; ++ + buf->result++; + dirent = buf->dirent; + if (!access_ok(VERIFY_WRITE, dirent, +@@ -116,6 +122,7 @@ SYSCALL_DEFINE3(old_readdir, unsigned in + + buf.result = 0; + buf.dirent = dirent; ++ buf.file = file; + + error = vfs_readdir(file, fillonedir, &buf); + if (buf.result) +@@ -142,6 +149,7 @@ struct linux_dirent { + struct getdents_callback { + struct linux_dirent __user * current_dir; + struct linux_dirent __user * previous; ++ struct file * file; + int count; + int error; + }; +@@ -163,6 +171,10 @@ static int filldir(void * __buf, const c + buf->error = -EOVERFLOW; + return -EOVERFLOW; + } ++ ++ if (!gr_acl_handle_filldir(buf->file, name, namlen, ino)) ++ return 0; ++ + dirent = buf->previous; + if (dirent) { + if (__put_user(offset, &dirent->d_off)) +@@ -210,6 +222,7 @@ SYSCALL_DEFINE3(getdents, unsigned int, + buf.previous = NULL; + buf.count = count; + buf.error = 0; ++ buf.file = file; + + error = vfs_readdir(file, filldir, &buf); + if (error >= 0) +@@ -229,6 +242,7 @@ out: + struct getdents_callback64 { + struct linux_dirent64 __user * current_dir; + struct linux_dirent64 __user * previous; ++ struct file *file; + int count; + int error; + }; +@@ -244,6 +258,10 @@ static int filldir64(void * __buf, const + buf->error = -EINVAL; /* only used if we fail.. */ + if (reclen > buf->count) + return -EINVAL; ++ ++ if (!gr_acl_handle_filldir(buf->file, name, namlen, ino)) ++ return 0; ++ + dirent = buf->previous; + if (dirent) { + if (__put_user(offset, &dirent->d_off)) +@@ -291,6 +309,7 @@ SYSCALL_DEFINE3(getdents64, unsigned int + + buf.current_dir = dirent; + buf.previous = NULL; ++ buf.file = file; + buf.count = count; + buf.error = 0; + +diff -urNp linux-2.6.38.3/fs/reiserfs/do_balan.c linux-2.6.38.3/fs/reiserfs/do_balan.c +--- linux-2.6.38.3/fs/reiserfs/do_balan.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/reiserfs/do_balan.c 2011-04-17 15:57:32.000000000 -0400 +@@ -2051,7 +2051,7 @@ void do_balance(struct tree_balance *tb, + return; + } + +- atomic_inc(&(fs_generation(tb->tb_sb))); ++ atomic_inc_unchecked(&(fs_generation(tb->tb_sb))); + do_balance_starts(tb); + + /* balance leaf returns 0 except if combining L R and S into +diff -urNp linux-2.6.38.3/fs/reiserfs/item_ops.c linux-2.6.38.3/fs/reiserfs/item_ops.c +--- linux-2.6.38.3/fs/reiserfs/item_ops.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/reiserfs/item_ops.c 2011-04-17 15:57:32.000000000 -0400 +@@ -102,7 +102,7 @@ static void sd_print_vi(struct virtual_i + vi->vi_index, vi->vi_type, vi->vi_ih); + } + +-static struct item_operations stat_data_ops = { ++static const struct item_operations stat_data_ops = { + .bytes_number = sd_bytes_number, + .decrement_key = sd_decrement_key, + .is_left_mergeable = sd_is_left_mergeable, +@@ -196,7 +196,7 @@ static void direct_print_vi(struct virtu + vi->vi_index, vi->vi_type, vi->vi_ih); + } + +-static struct item_operations direct_ops = { ++static const struct item_operations direct_ops = { + .bytes_number = direct_bytes_number, + .decrement_key = direct_decrement_key, + .is_left_mergeable = direct_is_left_mergeable, +@@ -341,7 +341,7 @@ static void indirect_print_vi(struct vir + vi->vi_index, vi->vi_type, vi->vi_ih); + } + +-static struct item_operations indirect_ops = { ++static const struct item_operations indirect_ops = { + .bytes_number = indirect_bytes_number, + .decrement_key = indirect_decrement_key, + .is_left_mergeable = indirect_is_left_mergeable, +@@ -628,7 +628,7 @@ static void direntry_print_vi(struct vir + printk("\n"); + } + +-static struct item_operations direntry_ops = { ++static const struct item_operations direntry_ops = { + .bytes_number = direntry_bytes_number, + .decrement_key = direntry_decrement_key, + .is_left_mergeable = direntry_is_left_mergeable, +@@ -724,7 +724,7 @@ static void errcatch_print_vi(struct vir + "Invalid item type observed, run fsck ASAP"); + } + +-static struct item_operations errcatch_ops = { ++static const struct item_operations errcatch_ops = { + errcatch_bytes_number, + errcatch_decrement_key, + errcatch_is_left_mergeable, +@@ -746,7 +746,7 @@ static struct item_operations errcatch_o + #error Item types must use disk-format assigned values. + #endif + +-struct item_operations *item_ops[TYPE_ANY + 1] = { ++const struct item_operations * const item_ops[TYPE_ANY + 1] = { + &stat_data_ops, + &indirect_ops, + &direct_ops, +diff -urNp linux-2.6.38.3/fs/reiserfs/procfs.c linux-2.6.38.3/fs/reiserfs/procfs.c +--- linux-2.6.38.3/fs/reiserfs/procfs.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/reiserfs/procfs.c 2011-04-17 15:57:32.000000000 -0400 +@@ -113,7 +113,7 @@ static int show_super(struct seq_file *m + "SMALL_TAILS " : "NO_TAILS ", + replay_only(sb) ? "REPLAY_ONLY " : "", + convert_reiserfs(sb) ? "CONV " : "", +- atomic_read(&r->s_generation_counter), ++ atomic_read_unchecked(&r->s_generation_counter), + SF(s_disk_reads), SF(s_disk_writes), SF(s_fix_nodes), + SF(s_do_balance), SF(s_unneeded_left_neighbor), + SF(s_good_search_by_key_reada), SF(s_bmaps), +diff -urNp linux-2.6.38.3/fs/select.c linux-2.6.38.3/fs/select.c +--- linux-2.6.38.3/fs/select.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/select.c 2011-04-17 15:57:32.000000000 -0400 +@@ -20,6 +20,7 @@ + #include <linux/module.h> + #include <linux/slab.h> + #include <linux/poll.h> ++#include <linux/security.h> + #include <linux/personality.h> /* for STICKY_TIMEOUTS */ + #include <linux/file.h> + #include <linux/fdtable.h> +@@ -840,6 +841,7 @@ int do_sys_poll(struct pollfd __user *uf + struct poll_list *walk = head; + unsigned long todo = nfds; + ++ gr_learn_resource(current, RLIMIT_NOFILE, nfds, 1); + if (nfds > rlimit(RLIMIT_NOFILE)) + return -EINVAL; + +diff -urNp linux-2.6.38.3/fs/seq_file.c linux-2.6.38.3/fs/seq_file.c +--- linux-2.6.38.3/fs/seq_file.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/seq_file.c 2011-04-17 15:57:32.000000000 -0400 +@@ -76,7 +76,8 @@ static int traverse(struct seq_file *m, + return 0; + } + if (!m->buf) { +- m->buf = kmalloc(m->size = PAGE_SIZE, GFP_KERNEL); ++ m->size = PAGE_SIZE; ++ m->buf = kmalloc(PAGE_SIZE, GFP_KERNEL); + if (!m->buf) + return -ENOMEM; + } +@@ -116,7 +117,8 @@ static int traverse(struct seq_file *m, + Eoverflow: + m->op->stop(m, p); + kfree(m->buf); +- m->buf = kmalloc(m->size <<= 1, GFP_KERNEL); ++ m->size <<= 1; ++ m->buf = kmalloc(m->size, GFP_KERNEL); + return !m->buf ? -ENOMEM : -EAGAIN; + } + +@@ -169,7 +171,8 @@ ssize_t seq_read(struct file *file, char + m->version = file->f_version; + /* grab buffer if we didn't have one */ + if (!m->buf) { +- m->buf = kmalloc(m->size = PAGE_SIZE, GFP_KERNEL); ++ m->size = PAGE_SIZE; ++ m->buf = kmalloc(PAGE_SIZE, GFP_KERNEL); + if (!m->buf) + goto Enomem; + } +@@ -210,7 +213,8 @@ ssize_t seq_read(struct file *file, char + goto Fill; + m->op->stop(m, p); + kfree(m->buf); +- m->buf = kmalloc(m->size <<= 1, GFP_KERNEL); ++ m->size <<= 1; ++ m->buf = kmalloc(m->size, GFP_KERNEL); + if (!m->buf) + goto Enomem; + m->count = 0; +diff -urNp linux-2.6.38.3/fs/splice.c linux-2.6.38.3/fs/splice.c +--- linux-2.6.38.3/fs/splice.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/splice.c 2011-04-17 15:57:32.000000000 -0400 +@@ -186,7 +186,7 @@ ssize_t splice_to_pipe(struct pipe_inode + pipe_lock(pipe); + + for (;;) { +- if (!pipe->readers) { ++ if (!atomic_read(&pipe->readers)) { + send_sig(SIGPIPE, current, 0); + if (!ret) + ret = -EPIPE; +@@ -240,9 +240,9 @@ ssize_t splice_to_pipe(struct pipe_inode + do_wakeup = 0; + } + +- pipe->waiting_writers++; ++ atomic_inc(&pipe->waiting_writers); + pipe_wait(pipe); +- pipe->waiting_writers--; ++ atomic_dec(&pipe->waiting_writers); + } + + pipe_unlock(pipe); +@@ -556,7 +556,7 @@ static ssize_t kernel_readv(struct file + old_fs = get_fs(); + set_fs(get_ds()); + /* The cast to a user pointer is valid due to the set_fs() */ +- res = vfs_readv(file, (const struct iovec __user *)vec, vlen, &pos); ++ res = vfs_readv(file, (__force const struct iovec __user *)vec, vlen, &pos); + set_fs(old_fs); + + return res; +@@ -571,7 +571,7 @@ static ssize_t kernel_write(struct file + old_fs = get_fs(); + set_fs(get_ds()); + /* The cast to a user pointer is valid due to the set_fs() */ +- res = vfs_write(file, (const char __user *)buf, count, &pos); ++ res = vfs_write(file, (__force const char __user *)buf, count, &pos); + set_fs(old_fs); + + return res; +@@ -622,7 +622,7 @@ ssize_t default_file_splice_read(struct + goto err; + + this_len = min_t(size_t, len, PAGE_CACHE_SIZE - offset); +- vec[i].iov_base = (void __user *) page_address(page); ++ vec[i].iov_base = (__force void __user *) page_address(page); + vec[i].iov_len = this_len; + spd.pages[i] = page; + spd.nr_pages++; +@@ -842,10 +842,10 @@ EXPORT_SYMBOL(splice_from_pipe_feed); + int splice_from_pipe_next(struct pipe_inode_info *pipe, struct splice_desc *sd) + { + while (!pipe->nrbufs) { +- if (!pipe->writers) ++ if (!atomic_read(&pipe->writers)) + return 0; + +- if (!pipe->waiting_writers && sd->num_spliced) ++ if (!atomic_read(&pipe->waiting_writers) && sd->num_spliced) + return 0; + + if (sd->flags & SPLICE_F_NONBLOCK) +@@ -1178,7 +1178,7 @@ ssize_t splice_direct_to_actor(struct fi + * out of the pipe right after the splice_to_pipe(). So set + * PIPE_READERS appropriately. + */ +- pipe->readers = 1; ++ atomic_set(&pipe->readers, 1); + + current->splice_pipe = pipe; + } +@@ -1730,9 +1730,9 @@ static int ipipe_prep(struct pipe_inode_ + ret = -ERESTARTSYS; + break; + } +- if (!pipe->writers) ++ if (!atomic_read(&pipe->writers)) + break; +- if (!pipe->waiting_writers) { ++ if (!atomic_read(&pipe->waiting_writers)) { + if (flags & SPLICE_F_NONBLOCK) { + ret = -EAGAIN; + break; +@@ -1764,7 +1764,7 @@ static int opipe_prep(struct pipe_inode_ + pipe_lock(pipe); + + while (pipe->nrbufs >= pipe->buffers) { +- if (!pipe->readers) { ++ if (!atomic_read(&pipe->readers)) { + send_sig(SIGPIPE, current, 0); + ret = -EPIPE; + break; +@@ -1777,9 +1777,9 @@ static int opipe_prep(struct pipe_inode_ + ret = -ERESTARTSYS; + break; + } +- pipe->waiting_writers++; ++ atomic_inc(&pipe->waiting_writers); + pipe_wait(pipe); +- pipe->waiting_writers--; ++ atomic_dec(&pipe->waiting_writers); + } + + pipe_unlock(pipe); +@@ -1815,14 +1815,14 @@ retry: + pipe_double_lock(ipipe, opipe); + + do { +- if (!opipe->readers) { ++ if (!atomic_read(&opipe->readers)) { + send_sig(SIGPIPE, current, 0); + if (!ret) + ret = -EPIPE; + break; + } + +- if (!ipipe->nrbufs && !ipipe->writers) ++ if (!ipipe->nrbufs && !atomic_read(&ipipe->writers)) + break; + + /* +@@ -1922,7 +1922,7 @@ static int link_pipe(struct pipe_inode_i + pipe_double_lock(ipipe, opipe); + + do { +- if (!opipe->readers) { ++ if (!atomic_read(&opipe->readers)) { + send_sig(SIGPIPE, current, 0); + if (!ret) + ret = -EPIPE; +@@ -1967,7 +1967,7 @@ static int link_pipe(struct pipe_inode_i + * return EAGAIN if we have the potential of some data in the + * future, otherwise just return 0 + */ +- if (!ret && ipipe->waiting_writers && (flags & SPLICE_F_NONBLOCK)) ++ if (!ret && atomic_read(&ipipe->waiting_writers) && (flags & SPLICE_F_NONBLOCK)) + ret = -EAGAIN; + + pipe_unlock(ipipe); +diff -urNp linux-2.6.38.3/fs/sysfs/mount.c linux-2.6.38.3/fs/sysfs/mount.c +--- linux-2.6.38.3/fs/sysfs/mount.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/sysfs/mount.c 2011-04-17 15:57:32.000000000 -0400 +@@ -36,7 +36,11 @@ struct sysfs_dirent sysfs_root = { + .s_name = "", + .s_count = ATOMIC_INIT(1), + .s_flags = SYSFS_DIR | (KOBJ_NS_TYPE_NONE << SYSFS_NS_TYPE_SHIFT), ++#ifdef CONFIG_GRKERNSEC_SYSFS_RESTRICT ++ .s_mode = S_IFDIR | S_IRWXU, ++#else + .s_mode = S_IFDIR | S_IRWXU | S_IRUGO | S_IXUGO, ++#endif + .s_ino = 1, + }; + +diff -urNp linux-2.6.38.3/fs/sysfs/symlink.c linux-2.6.38.3/fs/sysfs/symlink.c +--- linux-2.6.38.3/fs/sysfs/symlink.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/sysfs/symlink.c 2011-04-17 15:57:32.000000000 -0400 +@@ -286,7 +286,7 @@ static void *sysfs_follow_link(struct de + + static void sysfs_put_link(struct dentry *dentry, struct nameidata *nd, void *cookie) + { +- char *page = nd_get_link(nd); ++ const char *page = nd_get_link(nd); + if (!IS_ERR(page)) + free_page((unsigned long)page); + } +diff -urNp linux-2.6.38.3/fs/ubifs/debug.c linux-2.6.38.3/fs/ubifs/debug.c +--- linux-2.6.38.3/fs/ubifs/debug.c 2011-04-17 16:53:01.000000000 -0400 ++++ linux-2.6.38.3/fs/ubifs/debug.c 2011-04-17 16:53:48.000000000 -0400 +@@ -2844,19 +2844,19 @@ int dbg_debugfs_init_fs(struct ubifs_inf + } + + fname = "dump_lprops"; +- dent = debugfs_create_file(fname, S_IWUGO, d->dfs_dir, c, &dfs_fops); ++ dent = debugfs_create_file(fname, S_IWUSR, d->dfs_dir, c, &dfs_fops); + if (IS_ERR(dent)) + goto out_remove; + d->dfs_dump_lprops = dent; + + fname = "dump_budg"; +- dent = debugfs_create_file(fname, S_IWUGO, d->dfs_dir, c, &dfs_fops); ++ dent = debugfs_create_file(fname, S_IWUSR, d->dfs_dir, c, &dfs_fops); + if (IS_ERR(dent)) + goto out_remove; + d->dfs_dump_budg = dent; + + fname = "dump_tnc"; +- dent = debugfs_create_file(fname, S_IWUGO, d->dfs_dir, c, &dfs_fops); ++ dent = debugfs_create_file(fname, S_IWUSR, d->dfs_dir, c, &dfs_fops); + if (IS_ERR(dent)) + goto out_remove; + d->dfs_dump_tnc = dent; +diff -urNp linux-2.6.38.3/fs/udf/misc.c linux-2.6.38.3/fs/udf/misc.c +--- linux-2.6.38.3/fs/udf/misc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/udf/misc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -142,8 +142,8 @@ struct genericFormat *udf_add_extendedat + iinfo->i_lenEAttr += size; + return (struct genericFormat *)&ea[offset]; + } +- if (loc & 0x02) +- ; ++ if (loc & 0x02) { ++ } + + return NULL; + } +@@ -286,7 +286,7 @@ void udf_new_tag(char *data, uint16_t id + + u8 udf_tag_checksum(const struct tag *t) + { +- u8 *data = (u8 *)t; ++ const u8 *data = (const u8 *)t; + u8 checksum = 0; + int i; + for (i = 0; i < sizeof(struct tag); ++i) +diff -urNp linux-2.6.38.3/fs/udf/udfdecl.h linux-2.6.38.3/fs/udf/udfdecl.h +--- linux-2.6.38.3/fs/udf/udfdecl.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/udf/udfdecl.h 2011-04-17 15:57:32.000000000 -0400 +@@ -26,7 +26,7 @@ do { \ + printk(f, ##a); \ + } while (0) + #else +-#define udf_debug(f, a...) /**/ ++#define udf_debug(f, a...) do {} while (0) + #endif + + #define udf_info(f, a...) \ +diff -urNp linux-2.6.38.3/fs/utimes.c linux-2.6.38.3/fs/utimes.c +--- linux-2.6.38.3/fs/utimes.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/utimes.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1,6 +1,7 @@ + #include <linux/compiler.h> + #include <linux/file.h> + #include <linux/fs.h> ++#include <linux/security.h> + #include <linux/linkage.h> + #include <linux/mount.h> + #include <linux/namei.h> +@@ -101,6 +102,12 @@ static int utimes_common(struct path *pa + goto mnt_drop_write_and_out; + } + } ++ ++ if (!gr_acl_handle_utime(path->dentry, path->mnt)) { ++ error = -EACCES; ++ goto mnt_drop_write_and_out; ++ } ++ + mutex_lock(&inode->i_mutex); + error = notify_change(path->dentry, &newattrs); + mutex_unlock(&inode->i_mutex); +diff -urNp linux-2.6.38.3/fs/xattr_acl.c linux-2.6.38.3/fs/xattr_acl.c +--- linux-2.6.38.3/fs/xattr_acl.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/xattr_acl.c 2011-04-17 15:57:32.000000000 -0400 +@@ -17,8 +17,8 @@ + struct posix_acl * + posix_acl_from_xattr(const void *value, size_t size) + { +- posix_acl_xattr_header *header = (posix_acl_xattr_header *)value; +- posix_acl_xattr_entry *entry = (posix_acl_xattr_entry *)(header+1), *end; ++ const posix_acl_xattr_header *header = (const posix_acl_xattr_header *)value; ++ const posix_acl_xattr_entry *entry = (const posix_acl_xattr_entry *)(header+1), *end; + int count; + struct posix_acl *acl; + struct posix_acl_entry *acl_e; +diff -urNp linux-2.6.38.3/fs/xattr.c linux-2.6.38.3/fs/xattr.c +--- linux-2.6.38.3/fs/xattr.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/xattr.c 2011-04-17 15:57:32.000000000 -0400 +@@ -247,7 +247,7 @@ EXPORT_SYMBOL_GPL(vfs_removexattr); + * Extended attribute SET operations + */ + static long +-setxattr(struct dentry *d, const char __user *name, const void __user *value, ++setxattr(struct path *path, const char __user *name, const void __user *value, + size_t size, int flags) + { + int error; +@@ -271,7 +271,13 @@ setxattr(struct dentry *d, const char __ + return PTR_ERR(kvalue); + } + +- error = vfs_setxattr(d, kname, kvalue, size, flags); ++ if (!gr_acl_handle_setxattr(path->dentry, path->mnt)) { ++ error = -EACCES; ++ goto out; ++ } ++ ++ error = vfs_setxattr(path->dentry, kname, kvalue, size, flags); ++out: + kfree(kvalue); + return error; + } +@@ -288,7 +294,7 @@ SYSCALL_DEFINE5(setxattr, const char __u + return error; + error = mnt_want_write(path.mnt); + if (!error) { +- error = setxattr(path.dentry, name, value, size, flags); ++ error = setxattr(&path, name, value, size, flags); + mnt_drop_write(path.mnt); + } + path_put(&path); +@@ -307,7 +313,7 @@ SYSCALL_DEFINE5(lsetxattr, const char __ + return error; + error = mnt_want_write(path.mnt); + if (!error) { +- error = setxattr(path.dentry, name, value, size, flags); ++ error = setxattr(&path, name, value, size, flags); + mnt_drop_write(path.mnt); + } + path_put(&path); +@@ -318,17 +324,15 @@ SYSCALL_DEFINE5(fsetxattr, int, fd, cons + const void __user *,value, size_t, size, int, flags) + { + struct file *f; +- struct dentry *dentry; + int error = -EBADF; + + f = fget(fd); + if (!f) + return error; +- dentry = f->f_path.dentry; +- audit_inode(NULL, dentry); ++ audit_inode(NULL, f->f_path.dentry); + error = mnt_want_write_file(f); + if (!error) { +- error = setxattr(dentry, name, value, size, flags); ++ error = setxattr(&f->f_path, name, value, size, flags); + mnt_drop_write(f->f_path.mnt); + } + fput(f); +diff -urNp linux-2.6.38.3/fs/xfs/linux-2.6/xfs_ioctl32.c linux-2.6.38.3/fs/xfs/linux-2.6/xfs_ioctl32.c +--- linux-2.6.38.3/fs/xfs/linux-2.6/xfs_ioctl32.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/xfs/linux-2.6/xfs_ioctl32.c 2011-04-17 15:57:32.000000000 -0400 +@@ -73,6 +73,7 @@ xfs_compat_ioc_fsgeometry_v1( + xfs_fsop_geom_t fsgeo; + int error; + ++ memset(&fsgeo, 0, sizeof(fsgeo)); + error = xfs_fs_geometry(mp, &fsgeo, 3); + if (error) + return -error; +diff -urNp linux-2.6.38.3/fs/xfs/linux-2.6/xfs_ioctl.c linux-2.6.38.3/fs/xfs/linux-2.6/xfs_ioctl.c +--- linux-2.6.38.3/fs/xfs/linux-2.6/xfs_ioctl.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/xfs/linux-2.6/xfs_ioctl.c 2011-04-17 15:57:32.000000000 -0400 +@@ -128,7 +128,7 @@ xfs_find_handle( + } + + error = -EFAULT; +- if (copy_to_user(hreq->ohandle, &handle, hsize) || ++ if (hsize > sizeof handle || copy_to_user(hreq->ohandle, &handle, hsize) || + copy_to_user(hreq->ohandlen, &hsize, sizeof(__s32))) + goto out_put; + +@@ -720,6 +720,7 @@ xfs_ioc_fsgeometry( + xfs_fsop_geom_t fsgeo; + int error; + ++ memset(&fsgeo, 0, sizeof(fsgeo)); + error = xfs_fs_geometry(mp, &fsgeo, 4); + if (error) + return -error; +diff -urNp linux-2.6.38.3/fs/xfs/linux-2.6/xfs_iops.c linux-2.6.38.3/fs/xfs/linux-2.6/xfs_iops.c +--- linux-2.6.38.3/fs/xfs/linux-2.6/xfs_iops.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/xfs/linux-2.6/xfs_iops.c 2011-04-17 15:57:32.000000000 -0400 +@@ -436,7 +436,7 @@ xfs_vn_put_link( + struct nameidata *nd, + void *p) + { +- char *s = nd_get_link(nd); ++ const char *s = nd_get_link(nd); + + if (!IS_ERR(s)) + kfree(s); +diff -urNp linux-2.6.38.3/fs/xfs/xfs_bmap.c linux-2.6.38.3/fs/xfs/xfs_bmap.c +--- linux-2.6.38.3/fs/xfs/xfs_bmap.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/fs/xfs/xfs_bmap.c 2011-04-17 15:57:32.000000000 -0400 +@@ -287,7 +287,7 @@ xfs_bmap_validate_ret( + int nmap, + int ret_nmap); + #else +-#define xfs_bmap_validate_ret(bno,len,flags,mval,onmap,nmap) ++#define xfs_bmap_validate_ret(bno,len,flags,mval,onmap,nmap) do {} while (0) + #endif /* DEBUG */ + + STATIC int +diff -urNp linux-2.6.38.3/grsecurity/gracl_alloc.c linux-2.6.38.3/grsecurity/gracl_alloc.c +--- linux-2.6.38.3/grsecurity/gracl_alloc.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/gracl_alloc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,105 @@ ++#include <linux/kernel.h> ++#include <linux/mm.h> ++#include <linux/slab.h> ++#include <linux/vmalloc.h> ++#include <linux/gracl.h> ++#include <linux/grsecurity.h> ++ ++static unsigned long alloc_stack_next = 1; ++static unsigned long alloc_stack_size = 1; ++static void **alloc_stack; ++ ++static __inline__ int ++alloc_pop(void) ++{ ++ if (alloc_stack_next == 1) ++ return 0; ++ ++ kfree(alloc_stack[alloc_stack_next - 2]); ++ ++ alloc_stack_next--; ++ ++ return 1; ++} ++ ++static __inline__ int ++alloc_push(void *buf) ++{ ++ if (alloc_stack_next >= alloc_stack_size) ++ return 1; ++ ++ alloc_stack[alloc_stack_next - 1] = buf; ++ ++ alloc_stack_next++; ++ ++ return 0; ++} ++ ++void * ++acl_alloc(unsigned long len) ++{ ++ void *ret = NULL; ++ ++ if (!len || len > PAGE_SIZE) ++ goto out; ++ ++ ret = kmalloc(len, GFP_KERNEL); ++ ++ if (ret) { ++ if (alloc_push(ret)) { ++ kfree(ret); ++ ret = NULL; ++ } ++ } ++ ++out: ++ return ret; ++} ++ ++void * ++acl_alloc_num(unsigned long num, unsigned long len) ++{ ++ if (!len || (num > (PAGE_SIZE / len))) ++ return NULL; ++ ++ return acl_alloc(num * len); ++} ++ ++void ++acl_free_all(void) ++{ ++ if (gr_acl_is_enabled() || !alloc_stack) ++ return; ++ ++ while (alloc_pop()) ; ++ ++ if (alloc_stack) { ++ if ((alloc_stack_size * sizeof (void *)) <= PAGE_SIZE) ++ kfree(alloc_stack); ++ else ++ vfree(alloc_stack); ++ } ++ ++ alloc_stack = NULL; ++ alloc_stack_size = 1; ++ alloc_stack_next = 1; ++ ++ return; ++} ++ ++int ++acl_alloc_stack_init(unsigned long size) ++{ ++ if ((size * sizeof (void *)) <= PAGE_SIZE) ++ alloc_stack = ++ (void **) kmalloc(size * sizeof (void *), GFP_KERNEL); ++ else ++ alloc_stack = (void **) vmalloc(size * sizeof (void *)); ++ ++ alloc_stack_size = size; ++ ++ if (!alloc_stack) ++ return 0; ++ else ++ return 1; ++} +diff -urNp linux-2.6.38.3/grsecurity/gracl.c linux-2.6.38.3/grsecurity/gracl.c +--- linux-2.6.38.3/grsecurity/gracl.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/gracl.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,4074 @@ ++#include <linux/kernel.h> ++#include <linux/module.h> ++#include <linux/sched.h> ++#include <linux/mm.h> ++#include <linux/file.h> ++#include <linux/fs.h> ++#include <linux/namei.h> ++#include <linux/mount.h> ++#include <linux/tty.h> ++#include <linux/proc_fs.h> ++#include <linux/smp_lock.h> ++#include <linux/lglock.h> ++#include <linux/slab.h> ++#include <linux/vmalloc.h> ++#include <linux/types.h> ++#include <linux/sysctl.h> ++#include <linux/netdevice.h> ++#include <linux/ptrace.h> ++#include <linux/gracl.h> ++#include <linux/gralloc.h> ++#include <linux/grsecurity.h> ++#include <linux/grinternal.h> ++#include <linux/pid_namespace.h> ++#include <linux/fdtable.h> ++#include <linux/percpu.h> ++ ++#include <asm/uaccess.h> ++#include <asm/errno.h> ++#include <asm/mman.h> ++ ++static struct acl_role_db acl_role_set; ++static struct name_db name_set; ++static struct inodev_db inodev_set; ++ ++/* for keeping track of userspace pointers used for subjects, so we ++ can share references in the kernel as well ++*/ ++ ++static struct path real_root; ++ ++static struct acl_subj_map_db subj_map_set; ++ ++static struct acl_role_label *default_role; ++ ++static struct acl_role_label *role_list; ++ ++static u16 acl_sp_role_value; ++ ++extern char *gr_shared_page[4]; ++static DEFINE_MUTEX(gr_dev_mutex); ++DEFINE_RWLOCK(gr_inode_lock); ++ ++struct gr_arg *gr_usermode; ++ ++static unsigned int gr_status __read_only = GR_STATUS_INIT; ++ ++extern int chkpw(struct gr_arg *entry, unsigned char *salt, unsigned char *sum); ++extern void gr_clear_learn_entries(void); ++ ++#ifdef CONFIG_GRKERNSEC_RESLOG ++extern void gr_log_resource(const struct task_struct *task, ++ const int res, const unsigned long wanted, const int gt); ++#endif ++ ++unsigned char *gr_system_salt; ++unsigned char *gr_system_sum; ++ ++static struct sprole_pw **acl_special_roles = NULL; ++static __u16 num_sprole_pws = 0; ++ ++static struct acl_role_label *kernel_role = NULL; ++ ++static unsigned int gr_auth_attempts = 0; ++static unsigned long gr_auth_expires = 0UL; ++ ++extern struct vfsmount *sock_mnt; ++extern struct vfsmount *pipe_mnt; ++extern struct vfsmount *shm_mnt; ++#ifdef CONFIG_HUGETLBFS ++extern struct vfsmount *hugetlbfs_vfsmount; ++#endif ++ ++static struct acl_object_label *fakefs_obj; ++ ++extern int gr_init_uidset(void); ++extern void gr_free_uidset(void); ++extern void gr_remove_uid(uid_t uid); ++extern int gr_find_uid(uid_t uid); ++ ++DECLARE_BRLOCK(vfsmount_lock); ++ ++__inline__ int ++gr_acl_is_enabled(void) ++{ ++ return (gr_status & GR_READY); ++} ++ ++#ifdef CONFIG_BTRFS_FS ++extern dev_t get_btrfs_dev_from_inode(struct inode *inode); ++extern int btrfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat); ++#endif ++ ++static inline dev_t __get_dev(const struct dentry *dentry) ++{ ++#ifdef CONFIG_BTRFS_FS ++ if (dentry->d_inode->i_op && dentry->d_inode->i_op->getattr == &btrfs_getattr) ++ return get_btrfs_dev_from_inode(dentry->d_inode); ++ else ++#endif ++ return dentry->d_inode->i_sb->s_dev; ++} ++ ++dev_t gr_get_dev_from_dentry(struct dentry *dentry) ++{ ++ return __get_dev(dentry); ++} ++ ++static char gr_task_roletype_to_char(struct task_struct *task) ++{ ++ switch (task->role->roletype & ++ (GR_ROLE_DEFAULT | GR_ROLE_USER | GR_ROLE_GROUP | ++ GR_ROLE_SPECIAL)) { ++ case GR_ROLE_DEFAULT: ++ return 'D'; ++ case GR_ROLE_USER: ++ return 'U'; ++ case GR_ROLE_GROUP: ++ return 'G'; ++ case GR_ROLE_SPECIAL: ++ return 'S'; ++ } ++ ++ return 'X'; ++} ++ ++char gr_roletype_to_char(void) ++{ ++ return gr_task_roletype_to_char(current); ++} ++ ++__inline__ int ++gr_acl_tpe_check(void) ++{ ++ if (unlikely(!(gr_status & GR_READY))) ++ return 0; ++ if (current->role->roletype & GR_ROLE_TPE) ++ return 1; ++ else ++ return 0; ++} ++ ++int ++gr_handle_rawio(const struct inode *inode) ++{ ++#ifdef CONFIG_GRKERNSEC_CHROOT_CAPS ++ if (inode && S_ISBLK(inode->i_mode) && ++ grsec_enable_chroot_caps && proc_is_chrooted(current) && ++ !capable(CAP_SYS_RAWIO)) ++ return 1; ++#endif ++ return 0; ++} ++ ++static int ++gr_streq(const char *a, const char *b, const unsigned int lena, const unsigned int lenb) ++{ ++ if (likely(lena != lenb)) ++ return 0; ++ ++ return !memcmp(a, b, lena); ++} ++ ++static int prepend(char **buffer, int *buflen, const char *str, int namelen) ++{ ++ *buflen -= namelen; ++ if (*buflen < 0) ++ return -ENAMETOOLONG; ++ *buffer -= namelen; ++ memcpy(*buffer, str, namelen); ++ return 0; ++} ++ ++static int prepend_name(char **buffer, int *buflen, struct qstr *name) ++{ ++ return prepend(buffer, buflen, name->name, name->len); ++} ++ ++static int prepend_path(const struct path *path, struct path *root, ++ char **buffer, int *buflen) ++{ ++ struct dentry *dentry = path->dentry; ++ struct vfsmount *vfsmnt = path->mnt; ++ bool slash = false; ++ int error = 0; ++ ++ while (dentry != root->dentry || vfsmnt != root->mnt) { ++ struct dentry * parent; ++ ++ if (dentry == vfsmnt->mnt_root || IS_ROOT(dentry)) { ++ /* Global root? */ ++ if (vfsmnt->mnt_parent == vfsmnt) { ++ goto out; ++ } ++ dentry = vfsmnt->mnt_mountpoint; ++ vfsmnt = vfsmnt->mnt_parent; ++ continue; ++ } ++ parent = dentry->d_parent; ++ prefetch(parent); ++ spin_lock(&dentry->d_lock); ++ error = prepend_name(buffer, buflen, &dentry->d_name); ++ spin_unlock(&dentry->d_lock); ++ if (!error) ++ error = prepend(buffer, buflen, "/", 1); ++ if (error) ++ break; ++ ++ slash = true; ++ dentry = parent; ++ } ++ ++out: ++ if (!error && !slash) ++ error = prepend(buffer, buflen, "/", 1); ++ ++ return error; ++} ++ ++/* this must be called with vfsmount_lock and rename_lock held */ ++ ++static char *__our_d_path(const struct path *path, struct path *root, ++ char *buf, int buflen) ++{ ++ char *res = buf + buflen; ++ int error; ++ ++ prepend(&res, &buflen, "\0", 1); ++ error = prepend_path(path, root, &res, &buflen); ++ if (error) ++ return ERR_PTR(error); ++ ++ return res; ++} ++ ++static char * ++gen_full_path(struct path *path, struct path *root, char *buf, int buflen) ++{ ++ char *retval; ++ ++ retval = __our_d_path(path, root, buf, buflen); ++ if (unlikely(IS_ERR(retval))) ++ retval = strcpy(buf, "<path too long>"); ++ else if (unlikely(retval[1] == '/' && retval[2] == '\0')) ++ retval[1] = '\0'; ++ ++ return retval; ++} ++ ++static char * ++__d_real_path(const struct dentry *dentry, const struct vfsmount *vfsmnt, ++ char *buf, int buflen) ++{ ++ struct path path; ++ char *res; ++ ++ path.dentry = (struct dentry *)dentry; ++ path.mnt = (struct vfsmount *)vfsmnt; ++ ++ /* we can use real_root.dentry, real_root.mnt, because this is only called ++ by the RBAC system */ ++ res = gen_full_path(&path, &real_root, buf, buflen); ++ ++ return res; ++} ++ ++static char * ++d_real_path(const struct dentry *dentry, const struct vfsmount *vfsmnt, ++ char *buf, int buflen) ++{ ++ char *res; ++ struct path path; ++ struct path root; ++ struct task_struct *reaper = &init_task; ++ ++ path.dentry = (struct dentry *)dentry; ++ path.mnt = (struct vfsmount *)vfsmnt; ++ ++ /* we can't use real_root.dentry, real_root.mnt, because they belong only to the RBAC system */ ++ get_fs_root(reaper->fs, &root); ++ ++ write_seqlock(&rename_lock); ++ br_read_lock(vfsmount_lock); ++ res = gen_full_path(&path, &root, buf, buflen); ++ br_read_unlock(vfsmount_lock); ++ write_sequnlock(&rename_lock); ++ ++ path_put(&root); ++ return res; ++} ++ ++static char * ++gr_to_filename_rbac(const struct dentry *dentry, const struct vfsmount *mnt) ++{ ++ char *ret; ++ write_seqlock(&rename_lock); ++ br_read_lock(vfsmount_lock); ++ ret = __d_real_path(dentry, mnt, per_cpu_ptr(gr_shared_page[0],smp_processor_id()), ++ PAGE_SIZE); ++ br_read_unlock(vfsmount_lock); ++ write_sequnlock(&rename_lock); ++ return ret; ++} ++ ++char * ++gr_to_filename_nolock(const struct dentry *dentry, const struct vfsmount *mnt) ++{ ++ return __d_real_path(dentry, mnt, per_cpu_ptr(gr_shared_page[0],smp_processor_id()), ++ PAGE_SIZE); ++} ++ ++char * ++gr_to_filename(const struct dentry *dentry, const struct vfsmount *mnt) ++{ ++ return d_real_path(dentry, mnt, per_cpu_ptr(gr_shared_page[0], smp_processor_id()), ++ PAGE_SIZE); ++} ++ ++char * ++gr_to_filename1(const struct dentry *dentry, const struct vfsmount *mnt) ++{ ++ return d_real_path(dentry, mnt, per_cpu_ptr(gr_shared_page[1], smp_processor_id()), ++ PAGE_SIZE); ++} ++ ++char * ++gr_to_filename2(const struct dentry *dentry, const struct vfsmount *mnt) ++{ ++ return d_real_path(dentry, mnt, per_cpu_ptr(gr_shared_page[2], smp_processor_id()), ++ PAGE_SIZE); ++} ++ ++char * ++gr_to_filename3(const struct dentry *dentry, const struct vfsmount *mnt) ++{ ++ return d_real_path(dentry, mnt, per_cpu_ptr(gr_shared_page[3], smp_processor_id()), ++ PAGE_SIZE); ++} ++ ++__inline__ __u32 ++to_gr_audit(const __u32 reqmode) ++{ ++ /* masks off auditable permission flags, then shifts them to create ++ auditing flags, and adds the special case of append auditing if ++ we're requesting write */ ++ return (((reqmode & ~GR_AUDITS) << 10) | ((reqmode & GR_WRITE) ? GR_AUDIT_APPEND : 0)); ++} ++ ++struct acl_subject_label * ++lookup_subject_map(const struct acl_subject_label *userp) ++{ ++ unsigned int index = shash(userp, subj_map_set.s_size); ++ struct subject_map *match; ++ ++ match = subj_map_set.s_hash[index]; ++ ++ while (match && match->user != userp) ++ match = match->next; ++ ++ if (match != NULL) ++ return match->kernel; ++ else ++ return NULL; ++} ++ ++static void ++insert_subj_map_entry(struct subject_map *subjmap) ++{ ++ unsigned int index = shash(subjmap->user, subj_map_set.s_size); ++ struct subject_map **curr; ++ ++ subjmap->prev = NULL; ++ ++ curr = &subj_map_set.s_hash[index]; ++ if (*curr != NULL) ++ (*curr)->prev = subjmap; ++ ++ subjmap->next = *curr; ++ *curr = subjmap; ++ ++ return; ++} ++ ++static struct acl_role_label * ++lookup_acl_role_label(const struct task_struct *task, const uid_t uid, ++ const gid_t gid) ++{ ++ unsigned int index = rhash(uid, GR_ROLE_USER, acl_role_set.r_size); ++ struct acl_role_label *match; ++ struct role_allowed_ip *ipp; ++ unsigned int x; ++ u32 curr_ip = task->signal->curr_ip; ++ ++ task->signal->saved_ip = curr_ip; ++ ++ match = acl_role_set.r_hash[index]; ++ ++ while (match) { ++ if ((match->roletype & (GR_ROLE_DOMAIN | GR_ROLE_USER)) == (GR_ROLE_DOMAIN | GR_ROLE_USER)) { ++ for (x = 0; x < match->domain_child_num; x++) { ++ if (match->domain_children[x] == uid) ++ goto found; ++ } ++ } else if (match->uidgid == uid && match->roletype & GR_ROLE_USER) ++ break; ++ match = match->next; ++ } ++found: ++ if (match == NULL) { ++ try_group: ++ index = rhash(gid, GR_ROLE_GROUP, acl_role_set.r_size); ++ match = acl_role_set.r_hash[index]; ++ ++ while (match) { ++ if ((match->roletype & (GR_ROLE_DOMAIN | GR_ROLE_GROUP)) == (GR_ROLE_DOMAIN | GR_ROLE_GROUP)) { ++ for (x = 0; x < match->domain_child_num; x++) { ++ if (match->domain_children[x] == gid) ++ goto found2; ++ } ++ } else if (match->uidgid == gid && match->roletype & GR_ROLE_GROUP) ++ break; ++ match = match->next; ++ } ++found2: ++ if (match == NULL) ++ match = default_role; ++ if (match->allowed_ips == NULL) ++ return match; ++ else { ++ for (ipp = match->allowed_ips; ipp; ipp = ipp->next) { ++ if (likely ++ ((ntohl(curr_ip) & ipp->netmask) == ++ (ntohl(ipp->addr) & ipp->netmask))) ++ return match; ++ } ++ match = default_role; ++ } ++ } else if (match->allowed_ips == NULL) { ++ return match; ++ } else { ++ for (ipp = match->allowed_ips; ipp; ipp = ipp->next) { ++ if (likely ++ ((ntohl(curr_ip) & ipp->netmask) == ++ (ntohl(ipp->addr) & ipp->netmask))) ++ return match; ++ } ++ goto try_group; ++ } ++ ++ return match; ++} ++ ++struct acl_subject_label * ++lookup_acl_subj_label(const ino_t ino, const dev_t dev, ++ const struct acl_role_label *role) ++{ ++ unsigned int index = fhash(ino, dev, role->subj_hash_size); ++ struct acl_subject_label *match; ++ ++ match = role->subj_hash[index]; ++ ++ while (match && (match->inode != ino || match->device != dev || ++ (match->mode & GR_DELETED))) { ++ match = match->next; ++ } ++ ++ if (match && !(match->mode & GR_DELETED)) ++ return match; ++ else ++ return NULL; ++} ++ ++struct acl_subject_label * ++lookup_acl_subj_label_deleted(const ino_t ino, const dev_t dev, ++ const struct acl_role_label *role) ++{ ++ unsigned int index = fhash(ino, dev, role->subj_hash_size); ++ struct acl_subject_label *match; ++ ++ match = role->subj_hash[index]; ++ ++ while (match && (match->inode != ino || match->device != dev || ++ !(match->mode & GR_DELETED))) { ++ match = match->next; ++ } ++ ++ if (match && (match->mode & GR_DELETED)) ++ return match; ++ else ++ return NULL; ++} ++ ++static struct acl_object_label * ++lookup_acl_obj_label(const ino_t ino, const dev_t dev, ++ const struct acl_subject_label *subj) ++{ ++ unsigned int index = fhash(ino, dev, subj->obj_hash_size); ++ struct acl_object_label *match; ++ ++ match = subj->obj_hash[index]; ++ ++ while (match && (match->inode != ino || match->device != dev || ++ (match->mode & GR_DELETED))) { ++ match = match->next; ++ } ++ ++ if (match && !(match->mode & GR_DELETED)) ++ return match; ++ else ++ return NULL; ++} ++ ++static struct acl_object_label * ++lookup_acl_obj_label_create(const ino_t ino, const dev_t dev, ++ const struct acl_subject_label *subj) ++{ ++ unsigned int index = fhash(ino, dev, subj->obj_hash_size); ++ struct acl_object_label *match; ++ ++ match = subj->obj_hash[index]; ++ ++ while (match && (match->inode != ino || match->device != dev || ++ !(match->mode & GR_DELETED))) { ++ match = match->next; ++ } ++ ++ if (match && (match->mode & GR_DELETED)) ++ return match; ++ ++ match = subj->obj_hash[index]; ++ ++ while (match && (match->inode != ino || match->device != dev || ++ (match->mode & GR_DELETED))) { ++ match = match->next; ++ } ++ ++ if (match && !(match->mode & GR_DELETED)) ++ return match; ++ else ++ return NULL; ++} ++ ++static struct name_entry * ++lookup_name_entry(const char *name) ++{ ++ unsigned int len = strlen(name); ++ unsigned int key = full_name_hash(name, len); ++ unsigned int index = key % name_set.n_size; ++ struct name_entry *match; ++ ++ match = name_set.n_hash[index]; ++ ++ while (match && (match->key != key || !gr_streq(match->name, name, match->len, len))) ++ match = match->next; ++ ++ return match; ++} ++ ++static struct name_entry * ++lookup_name_entry_create(const char *name) ++{ ++ unsigned int len = strlen(name); ++ unsigned int key = full_name_hash(name, len); ++ unsigned int index = key % name_set.n_size; ++ struct name_entry *match; ++ ++ match = name_set.n_hash[index]; ++ ++ while (match && (match->key != key || !gr_streq(match->name, name, match->len, len) || ++ !match->deleted)) ++ match = match->next; ++ ++ if (match && match->deleted) ++ return match; ++ ++ match = name_set.n_hash[index]; ++ ++ while (match && (match->key != key || !gr_streq(match->name, name, match->len, len) || ++ match->deleted)) ++ match = match->next; ++ ++ if (match && !match->deleted) ++ return match; ++ else ++ return NULL; ++} ++ ++static struct inodev_entry * ++lookup_inodev_entry(const ino_t ino, const dev_t dev) ++{ ++ unsigned int index = fhash(ino, dev, inodev_set.i_size); ++ struct inodev_entry *match; ++ ++ match = inodev_set.i_hash[index]; ++ ++ while (match && (match->nentry->inode != ino || match->nentry->device != dev)) ++ match = match->next; ++ ++ return match; ++} ++ ++static void ++insert_inodev_entry(struct inodev_entry *entry) ++{ ++ unsigned int index = fhash(entry->nentry->inode, entry->nentry->device, ++ inodev_set.i_size); ++ struct inodev_entry **curr; ++ ++ entry->prev = NULL; ++ ++ curr = &inodev_set.i_hash[index]; ++ if (*curr != NULL) ++ (*curr)->prev = entry; ++ ++ entry->next = *curr; ++ *curr = entry; ++ ++ return; ++} ++ ++static void ++__insert_acl_role_label(struct acl_role_label *role, uid_t uidgid) ++{ ++ unsigned int index = ++ rhash(uidgid, role->roletype & (GR_ROLE_USER | GR_ROLE_GROUP), acl_role_set.r_size); ++ struct acl_role_label **curr; ++ struct acl_role_label *tmp; ++ ++ curr = &acl_role_set.r_hash[index]; ++ ++ /* if role was already inserted due to domains and already has ++ a role in the same bucket as it attached, then we need to ++ combine these two buckets ++ */ ++ if (role->next) { ++ tmp = role->next; ++ while (tmp->next) ++ tmp = tmp->next; ++ tmp->next = *curr; ++ } else ++ role->next = *curr; ++ *curr = role; ++ ++ return; ++} ++ ++static void ++insert_acl_role_label(struct acl_role_label *role) ++{ ++ int i; ++ ++ if (role_list == NULL) { ++ role_list = role; ++ role->prev = NULL; ++ } else { ++ role->prev = role_list; ++ role_list = role; ++ } ++ ++ /* used for hash chains */ ++ role->next = NULL; ++ ++ if (role->roletype & GR_ROLE_DOMAIN) { ++ for (i = 0; i < role->domain_child_num; i++) ++ __insert_acl_role_label(role, role->domain_children[i]); ++ } else ++ __insert_acl_role_label(role, role->uidgid); ++} ++ ++static int ++insert_name_entry(char *name, const ino_t inode, const dev_t device, __u8 deleted) ++{ ++ struct name_entry **curr, *nentry; ++ struct inodev_entry *ientry; ++ unsigned int len = strlen(name); ++ unsigned int key = full_name_hash(name, len); ++ unsigned int index = key % name_set.n_size; ++ ++ curr = &name_set.n_hash[index]; ++ ++ while (*curr && ((*curr)->key != key || !gr_streq((*curr)->name, name, (*curr)->len, len))) ++ curr = &((*curr)->next); ++ ++ if (*curr != NULL) ++ return 1; ++ ++ nentry = acl_alloc(sizeof (struct name_entry)); ++ if (nentry == NULL) ++ return 0; ++ ientry = acl_alloc(sizeof (struct inodev_entry)); ++ if (ientry == NULL) ++ return 0; ++ ientry->nentry = nentry; ++ ++ nentry->key = key; ++ nentry->name = name; ++ nentry->inode = inode; ++ nentry->device = device; ++ nentry->len = len; ++ nentry->deleted = deleted; ++ ++ nentry->prev = NULL; ++ curr = &name_set.n_hash[index]; ++ if (*curr != NULL) ++ (*curr)->prev = nentry; ++ nentry->next = *curr; ++ *curr = nentry; ++ ++ /* insert us into the table searchable by inode/dev */ ++ insert_inodev_entry(ientry); ++ ++ return 1; ++} ++ ++static void ++insert_acl_obj_label(struct acl_object_label *obj, ++ struct acl_subject_label *subj) ++{ ++ unsigned int index = ++ fhash(obj->inode, obj->device, subj->obj_hash_size); ++ struct acl_object_label **curr; ++ ++ ++ obj->prev = NULL; ++ ++ curr = &subj->obj_hash[index]; ++ if (*curr != NULL) ++ (*curr)->prev = obj; ++ ++ obj->next = *curr; ++ *curr = obj; ++ ++ return; ++} ++ ++static void ++insert_acl_subj_label(struct acl_subject_label *obj, ++ struct acl_role_label *role) ++{ ++ unsigned int index = fhash(obj->inode, obj->device, role->subj_hash_size); ++ struct acl_subject_label **curr; ++ ++ obj->prev = NULL; ++ ++ curr = &role->subj_hash[index]; ++ if (*curr != NULL) ++ (*curr)->prev = obj; ++ ++ obj->next = *curr; ++ *curr = obj; ++ ++ return; ++} ++ ++/* allocating chained hash tables, so optimal size is where lambda ~ 1 */ ++ ++static void * ++create_table(__u32 * len, int elementsize) ++{ ++ unsigned int table_sizes[] = { ++ 7, 13, 31, 61, 127, 251, 509, 1021, 2039, 4093, 8191, 16381, ++ 32749, 65521, 131071, 262139, 524287, 1048573, 2097143, ++ 4194301, 8388593, 16777213, 33554393, 67108859 ++ }; ++ void *newtable = NULL; ++ unsigned int pwr = 0; ++ ++ while ((pwr < ((sizeof (table_sizes) / sizeof (table_sizes[0])) - 1)) && ++ table_sizes[pwr] <= *len) ++ pwr++; ++ ++ if (table_sizes[pwr] <= *len || (table_sizes[pwr] > ULONG_MAX / elementsize)) ++ return newtable; ++ ++ if ((table_sizes[pwr] * elementsize) <= PAGE_SIZE) ++ newtable = ++ kmalloc(table_sizes[pwr] * elementsize, GFP_KERNEL); ++ else ++ newtable = vmalloc(table_sizes[pwr] * elementsize); ++ ++ *len = table_sizes[pwr]; ++ ++ return newtable; ++} ++ ++static int ++init_variables(const struct gr_arg *arg) ++{ ++ struct task_struct *reaper = &init_task; ++ unsigned int stacksize; ++ ++ subj_map_set.s_size = arg->role_db.num_subjects; ++ acl_role_set.r_size = arg->role_db.num_roles + arg->role_db.num_domain_children; ++ name_set.n_size = arg->role_db.num_objects; ++ inodev_set.i_size = arg->role_db.num_objects; ++ ++ if (!subj_map_set.s_size || !acl_role_set.r_size || ++ !name_set.n_size || !inodev_set.i_size) ++ return 1; ++ ++ if (!gr_init_uidset()) ++ return 1; ++ ++ /* set up the stack that holds allocation info */ ++ ++ stacksize = arg->role_db.num_pointers + 5; ++ ++ if (!acl_alloc_stack_init(stacksize)) ++ return 1; ++ ++ /* grab reference for the real root dentry and vfsmount */ ++ get_fs_root(reaper->fs, &real_root); ++ ++#ifdef CONFIG_GRKERNSEC_RBAC_DEBUG ++ printk(KERN_ALERT "Obtained real root device=%d, inode=%lu\n", __get_dev(real_root.dentry), real_root.dentry->d_inode->i_ino); ++#endif ++ ++ fakefs_obj = acl_alloc(sizeof(struct acl_object_label)); ++ if (fakefs_obj == NULL) ++ return 1; ++ fakefs_obj->mode = GR_FIND | GR_READ | GR_WRITE | GR_EXEC; ++ ++ subj_map_set.s_hash = ++ (struct subject_map **) create_table(&subj_map_set.s_size, sizeof(void *)); ++ acl_role_set.r_hash = ++ (struct acl_role_label **) create_table(&acl_role_set.r_size, sizeof(void *)); ++ name_set.n_hash = (struct name_entry **) create_table(&name_set.n_size, sizeof(void *)); ++ inodev_set.i_hash = ++ (struct inodev_entry **) create_table(&inodev_set.i_size, sizeof(void *)); ++ ++ if (!subj_map_set.s_hash || !acl_role_set.r_hash || ++ !name_set.n_hash || !inodev_set.i_hash) ++ return 1; ++ ++ memset(subj_map_set.s_hash, 0, ++ sizeof(struct subject_map *) * subj_map_set.s_size); ++ memset(acl_role_set.r_hash, 0, ++ sizeof (struct acl_role_label *) * acl_role_set.r_size); ++ memset(name_set.n_hash, 0, ++ sizeof (struct name_entry *) * name_set.n_size); ++ memset(inodev_set.i_hash, 0, ++ sizeof (struct inodev_entry *) * inodev_set.i_size); ++ ++ return 0; ++} ++ ++/* free information not needed after startup ++ currently contains user->kernel pointer mappings for subjects ++*/ ++ ++static void ++free_init_variables(void) ++{ ++ __u32 i; ++ ++ if (subj_map_set.s_hash) { ++ for (i = 0; i < subj_map_set.s_size; i++) { ++ if (subj_map_set.s_hash[i]) { ++ kfree(subj_map_set.s_hash[i]); ++ subj_map_set.s_hash[i] = NULL; ++ } ++ } ++ ++ if ((subj_map_set.s_size * sizeof (struct subject_map *)) <= ++ PAGE_SIZE) ++ kfree(subj_map_set.s_hash); ++ else ++ vfree(subj_map_set.s_hash); ++ } ++ ++ return; ++} ++ ++static void ++free_variables(void) ++{ ++ struct acl_subject_label *s; ++ struct acl_role_label *r; ++ struct task_struct *task, *task2; ++ unsigned int x; ++ ++ gr_clear_learn_entries(); ++ ++ read_lock(&tasklist_lock); ++ do_each_thread(task2, task) { ++ task->acl_sp_role = 0; ++ task->acl_role_id = 0; ++ task->acl = NULL; ++ task->role = NULL; ++ } while_each_thread(task2, task); ++ read_unlock(&tasklist_lock); ++ ++ /* release the reference to the real root dentry and vfsmount */ ++ path_put(&real_root); ++ ++ /* free all object hash tables */ ++ ++ FOR_EACH_ROLE_START(r) ++ if (r->subj_hash == NULL) ++ goto next_role; ++ FOR_EACH_SUBJECT_START(r, s, x) ++ if (s->obj_hash == NULL) ++ break; ++ if ((s->obj_hash_size * sizeof (struct acl_object_label *)) <= PAGE_SIZE) ++ kfree(s->obj_hash); ++ else ++ vfree(s->obj_hash); ++ FOR_EACH_SUBJECT_END(s, x) ++ FOR_EACH_NESTED_SUBJECT_START(r, s) ++ if (s->obj_hash == NULL) ++ break; ++ if ((s->obj_hash_size * sizeof (struct acl_object_label *)) <= PAGE_SIZE) ++ kfree(s->obj_hash); ++ else ++ vfree(s->obj_hash); ++ FOR_EACH_NESTED_SUBJECT_END(s) ++ if ((r->subj_hash_size * sizeof (struct acl_subject_label *)) <= PAGE_SIZE) ++ kfree(r->subj_hash); ++ else ++ vfree(r->subj_hash); ++ r->subj_hash = NULL; ++next_role: ++ FOR_EACH_ROLE_END(r) ++ ++ acl_free_all(); ++ ++ if (acl_role_set.r_hash) { ++ if ((acl_role_set.r_size * sizeof (struct acl_role_label *)) <= ++ PAGE_SIZE) ++ kfree(acl_role_set.r_hash); ++ else ++ vfree(acl_role_set.r_hash); ++ } ++ if (name_set.n_hash) { ++ if ((name_set.n_size * sizeof (struct name_entry *)) <= ++ PAGE_SIZE) ++ kfree(name_set.n_hash); ++ else ++ vfree(name_set.n_hash); ++ } ++ ++ if (inodev_set.i_hash) { ++ if ((inodev_set.i_size * sizeof (struct inodev_entry *)) <= ++ PAGE_SIZE) ++ kfree(inodev_set.i_hash); ++ else ++ vfree(inodev_set.i_hash); ++ } ++ ++ gr_free_uidset(); ++ ++ memset(&name_set, 0, sizeof (struct name_db)); ++ memset(&inodev_set, 0, sizeof (struct inodev_db)); ++ memset(&acl_role_set, 0, sizeof (struct acl_role_db)); ++ memset(&subj_map_set, 0, sizeof (struct acl_subj_map_db)); ++ ++ default_role = NULL; ++ role_list = NULL; ++ ++ return; ++} ++ ++static __u32 ++count_user_objs(struct acl_object_label *userp) ++{ ++ struct acl_object_label o_tmp; ++ __u32 num = 0; ++ ++ while (userp) { ++ if (copy_from_user(&o_tmp, userp, ++ sizeof (struct acl_object_label))) ++ break; ++ ++ userp = o_tmp.prev; ++ num++; ++ } ++ ++ return num; ++} ++ ++static struct acl_subject_label * ++do_copy_user_subj(struct acl_subject_label *userp, struct acl_role_label *role); ++ ++static int ++copy_user_glob(struct acl_object_label *obj) ++{ ++ struct acl_object_label *g_tmp, **guser; ++ unsigned int len; ++ char *tmp; ++ ++ if (obj->globbed == NULL) ++ return 0; ++ ++ guser = &obj->globbed; ++ while (*guser) { ++ g_tmp = (struct acl_object_label *) ++ acl_alloc(sizeof (struct acl_object_label)); ++ if (g_tmp == NULL) ++ return -ENOMEM; ++ ++ if (copy_from_user(g_tmp, *guser, ++ sizeof (struct acl_object_label))) ++ return -EFAULT; ++ ++ len = strnlen_user(g_tmp->filename, PATH_MAX); ++ ++ if (!len || len >= PATH_MAX) ++ return -EINVAL; ++ ++ if ((tmp = (char *) acl_alloc(len)) == NULL) ++ return -ENOMEM; ++ ++ if (copy_from_user(tmp, g_tmp->filename, len)) ++ return -EFAULT; ++ tmp[len-1] = '\0'; ++ g_tmp->filename = tmp; ++ ++ *guser = g_tmp; ++ guser = &(g_tmp->next); ++ } ++ ++ return 0; ++} ++ ++static int ++copy_user_objs(struct acl_object_label *userp, struct acl_subject_label *subj, ++ struct acl_role_label *role) ++{ ++ struct acl_object_label *o_tmp; ++ unsigned int len; ++ int ret; ++ char *tmp; ++ ++ while (userp) { ++ if ((o_tmp = (struct acl_object_label *) ++ acl_alloc(sizeof (struct acl_object_label))) == NULL) ++ return -ENOMEM; ++ ++ if (copy_from_user(o_tmp, userp, ++ sizeof (struct acl_object_label))) ++ return -EFAULT; ++ ++ userp = o_tmp->prev; ++ ++ len = strnlen_user(o_tmp->filename, PATH_MAX); ++ ++ if (!len || len >= PATH_MAX) ++ return -EINVAL; ++ ++ if ((tmp = (char *) acl_alloc(len)) == NULL) ++ return -ENOMEM; ++ ++ if (copy_from_user(tmp, o_tmp->filename, len)) ++ return -EFAULT; ++ tmp[len-1] = '\0'; ++ o_tmp->filename = tmp; ++ ++ insert_acl_obj_label(o_tmp, subj); ++ if (!insert_name_entry(o_tmp->filename, o_tmp->inode, ++ o_tmp->device, (o_tmp->mode & GR_DELETED) ? 1 : 0)) ++ return -ENOMEM; ++ ++ ret = copy_user_glob(o_tmp); ++ if (ret) ++ return ret; ++ ++ if (o_tmp->nested) { ++ o_tmp->nested = do_copy_user_subj(o_tmp->nested, role); ++ if (IS_ERR(o_tmp->nested)) ++ return PTR_ERR(o_tmp->nested); ++ ++ /* insert into nested subject list */ ++ o_tmp->nested->next = role->hash->first; ++ role->hash->first = o_tmp->nested; ++ } ++ } ++ ++ return 0; ++} ++ ++static __u32 ++count_user_subjs(struct acl_subject_label *userp) ++{ ++ struct acl_subject_label s_tmp; ++ __u32 num = 0; ++ ++ while (userp) { ++ if (copy_from_user(&s_tmp, userp, ++ sizeof (struct acl_subject_label))) ++ break; ++ ++ userp = s_tmp.prev; ++ /* do not count nested subjects against this count, since ++ they are not included in the hash table, but are ++ attached to objects. We have already counted ++ the subjects in userspace for the allocation ++ stack ++ */ ++ if (!(s_tmp.mode & GR_NESTED)) ++ num++; ++ } ++ ++ return num; ++} ++ ++static int ++copy_user_allowedips(struct acl_role_label *rolep) ++{ ++ struct role_allowed_ip *ruserip, *rtmp = NULL, *rlast; ++ ++ ruserip = rolep->allowed_ips; ++ ++ while (ruserip) { ++ rlast = rtmp; ++ ++ if ((rtmp = (struct role_allowed_ip *) ++ acl_alloc(sizeof (struct role_allowed_ip))) == NULL) ++ return -ENOMEM; ++ ++ if (copy_from_user(rtmp, ruserip, ++ sizeof (struct role_allowed_ip))) ++ return -EFAULT; ++ ++ ruserip = rtmp->prev; ++ ++ if (!rlast) { ++ rtmp->prev = NULL; ++ rolep->allowed_ips = rtmp; ++ } else { ++ rlast->next = rtmp; ++ rtmp->prev = rlast; ++ } ++ ++ if (!ruserip) ++ rtmp->next = NULL; ++ } ++ ++ return 0; ++} ++ ++static int ++copy_user_transitions(struct acl_role_label *rolep) ++{ ++ struct role_transition *rusertp, *rtmp = NULL, *rlast; ++ ++ unsigned int len; ++ char *tmp; ++ ++ rusertp = rolep->transitions; ++ ++ while (rusertp) { ++ rlast = rtmp; ++ ++ if ((rtmp = (struct role_transition *) ++ acl_alloc(sizeof (struct role_transition))) == NULL) ++ return -ENOMEM; ++ ++ if (copy_from_user(rtmp, rusertp, ++ sizeof (struct role_transition))) ++ return -EFAULT; ++ ++ rusertp = rtmp->prev; ++ ++ len = strnlen_user(rtmp->rolename, GR_SPROLE_LEN); ++ ++ if (!len || len >= GR_SPROLE_LEN) ++ return -EINVAL; ++ ++ if ((tmp = (char *) acl_alloc(len)) == NULL) ++ return -ENOMEM; ++ ++ if (copy_from_user(tmp, rtmp->rolename, len)) ++ return -EFAULT; ++ tmp[len-1] = '\0'; ++ rtmp->rolename = tmp; ++ ++ if (!rlast) { ++ rtmp->prev = NULL; ++ rolep->transitions = rtmp; ++ } else { ++ rlast->next = rtmp; ++ rtmp->prev = rlast; ++ } ++ ++ if (!rusertp) ++ rtmp->next = NULL; ++ } ++ ++ return 0; ++} ++ ++static struct acl_subject_label * ++do_copy_user_subj(struct acl_subject_label *userp, struct acl_role_label *role) ++{ ++ struct acl_subject_label *s_tmp = NULL, *s_tmp2; ++ unsigned int len; ++ char *tmp; ++ __u32 num_objs; ++ struct acl_ip_label **i_tmp, *i_utmp2; ++ struct gr_hash_struct ghash; ++ struct subject_map *subjmap; ++ unsigned int i_num; ++ int err; ++ ++ s_tmp = lookup_subject_map(userp); ++ ++ /* we've already copied this subject into the kernel, just return ++ the reference to it, and don't copy it over again ++ */ ++ if (s_tmp) ++ return(s_tmp); ++ ++ if ((s_tmp = (struct acl_subject_label *) ++ acl_alloc(sizeof (struct acl_subject_label))) == NULL) ++ return ERR_PTR(-ENOMEM); ++ ++ subjmap = (struct subject_map *)kmalloc(sizeof (struct subject_map), GFP_KERNEL); ++ if (subjmap == NULL) ++ return ERR_PTR(-ENOMEM); ++ ++ subjmap->user = userp; ++ subjmap->kernel = s_tmp; ++ insert_subj_map_entry(subjmap); ++ ++ if (copy_from_user(s_tmp, userp, ++ sizeof (struct acl_subject_label))) ++ return ERR_PTR(-EFAULT); ++ ++ len = strnlen_user(s_tmp->filename, PATH_MAX); ++ ++ if (!len || len >= PATH_MAX) ++ return ERR_PTR(-EINVAL); ++ ++ if ((tmp = (char *) acl_alloc(len)) == NULL) ++ return ERR_PTR(-ENOMEM); ++ ++ if (copy_from_user(tmp, s_tmp->filename, len)) ++ return ERR_PTR(-EFAULT); ++ tmp[len-1] = '\0'; ++ s_tmp->filename = tmp; ++ ++ if (!strcmp(s_tmp->filename, "/")) ++ role->root_label = s_tmp; ++ ++ if (copy_from_user(&ghash, s_tmp->hash, sizeof(struct gr_hash_struct))) ++ return ERR_PTR(-EFAULT); ++ ++ /* copy user and group transition tables */ ++ ++ if (s_tmp->user_trans_num) { ++ uid_t *uidlist; ++ ++ uidlist = (uid_t *)acl_alloc_num(s_tmp->user_trans_num, sizeof(uid_t)); ++ if (uidlist == NULL) ++ return ERR_PTR(-ENOMEM); ++ if (copy_from_user(uidlist, s_tmp->user_transitions, s_tmp->user_trans_num * sizeof(uid_t))) ++ return ERR_PTR(-EFAULT); ++ ++ s_tmp->user_transitions = uidlist; ++ } ++ ++ if (s_tmp->group_trans_num) { ++ gid_t *gidlist; ++ ++ gidlist = (gid_t *)acl_alloc_num(s_tmp->group_trans_num, sizeof(gid_t)); ++ if (gidlist == NULL) ++ return ERR_PTR(-ENOMEM); ++ if (copy_from_user(gidlist, s_tmp->group_transitions, s_tmp->group_trans_num * sizeof(gid_t))) ++ return ERR_PTR(-EFAULT); ++ ++ s_tmp->group_transitions = gidlist; ++ } ++ ++ /* set up object hash table */ ++ num_objs = count_user_objs(ghash.first); ++ ++ s_tmp->obj_hash_size = num_objs; ++ s_tmp->obj_hash = ++ (struct acl_object_label **) ++ create_table(&(s_tmp->obj_hash_size), sizeof(void *)); ++ ++ if (!s_tmp->obj_hash) ++ return ERR_PTR(-ENOMEM); ++ ++ memset(s_tmp->obj_hash, 0, ++ s_tmp->obj_hash_size * ++ sizeof (struct acl_object_label *)); ++ ++ /* add in objects */ ++ err = copy_user_objs(ghash.first, s_tmp, role); ++ ++ if (err) ++ return ERR_PTR(err); ++ ++ /* set pointer for parent subject */ ++ if (s_tmp->parent_subject) { ++ s_tmp2 = do_copy_user_subj(s_tmp->parent_subject, role); ++ ++ if (IS_ERR(s_tmp2)) ++ return s_tmp2; ++ ++ s_tmp->parent_subject = s_tmp2; ++ } ++ ++ /* add in ip acls */ ++ ++ if (!s_tmp->ip_num) { ++ s_tmp->ips = NULL; ++ goto insert; ++ } ++ ++ i_tmp = ++ (struct acl_ip_label **) acl_alloc_num(s_tmp->ip_num, ++ sizeof (struct acl_ip_label *)); ++ ++ if (!i_tmp) ++ return ERR_PTR(-ENOMEM); ++ ++ for (i_num = 0; i_num < s_tmp->ip_num; i_num++) { ++ *(i_tmp + i_num) = ++ (struct acl_ip_label *) ++ acl_alloc(sizeof (struct acl_ip_label)); ++ if (!*(i_tmp + i_num)) ++ return ERR_PTR(-ENOMEM); ++ ++ if (copy_from_user ++ (&i_utmp2, s_tmp->ips + i_num, ++ sizeof (struct acl_ip_label *))) ++ return ERR_PTR(-EFAULT); ++ ++ if (copy_from_user ++ (*(i_tmp + i_num), i_utmp2, ++ sizeof (struct acl_ip_label))) ++ return ERR_PTR(-EFAULT); ++ ++ if ((*(i_tmp + i_num))->iface == NULL) ++ continue; ++ ++ len = strnlen_user((*(i_tmp + i_num))->iface, IFNAMSIZ); ++ if (!len || len >= IFNAMSIZ) ++ return ERR_PTR(-EINVAL); ++ tmp = acl_alloc(len); ++ if (tmp == NULL) ++ return ERR_PTR(-ENOMEM); ++ if (copy_from_user(tmp, (*(i_tmp + i_num))->iface, len)) ++ return ERR_PTR(-EFAULT); ++ (*(i_tmp + i_num))->iface = tmp; ++ } ++ ++ s_tmp->ips = i_tmp; ++ ++insert: ++ if (!insert_name_entry(s_tmp->filename, s_tmp->inode, ++ s_tmp->device, (s_tmp->mode & GR_DELETED) ? 1 : 0)) ++ return ERR_PTR(-ENOMEM); ++ ++ return s_tmp; ++} ++ ++static int ++copy_user_subjs(struct acl_subject_label *userp, struct acl_role_label *role) ++{ ++ struct acl_subject_label s_pre; ++ struct acl_subject_label * ret; ++ int err; ++ ++ while (userp) { ++ if (copy_from_user(&s_pre, userp, ++ sizeof (struct acl_subject_label))) ++ return -EFAULT; ++ ++ /* do not add nested subjects here, add ++ while parsing objects ++ */ ++ ++ if (s_pre.mode & GR_NESTED) { ++ userp = s_pre.prev; ++ continue; ++ } ++ ++ ret = do_copy_user_subj(userp, role); ++ ++ err = PTR_ERR(ret); ++ if (IS_ERR(ret)) ++ return err; ++ ++ insert_acl_subj_label(ret, role); ++ ++ userp = s_pre.prev; ++ } ++ ++ return 0; ++} ++ ++static int ++copy_user_acl(struct gr_arg *arg) ++{ ++ struct acl_role_label *r_tmp = NULL, **r_utmp, *r_utmp2; ++ struct sprole_pw *sptmp; ++ struct gr_hash_struct *ghash; ++ uid_t *domainlist; ++ unsigned int r_num; ++ unsigned int len; ++ char *tmp; ++ int err = 0; ++ __u16 i; ++ __u32 num_subjs; ++ ++ /* we need a default and kernel role */ ++ if (arg->role_db.num_roles < 2) ++ return -EINVAL; ++ ++ /* copy special role authentication info from userspace */ ++ ++ num_sprole_pws = arg->num_sprole_pws; ++ acl_special_roles = (struct sprole_pw **) acl_alloc_num(num_sprole_pws, sizeof(struct sprole_pw *)); ++ ++ if (!acl_special_roles) { ++ err = -ENOMEM; ++ goto cleanup; ++ } ++ ++ for (i = 0; i < num_sprole_pws; i++) { ++ sptmp = (struct sprole_pw *) acl_alloc(sizeof(struct sprole_pw)); ++ if (!sptmp) { ++ err = -ENOMEM; ++ goto cleanup; ++ } ++ if (copy_from_user(sptmp, arg->sprole_pws + i, ++ sizeof (struct sprole_pw))) { ++ err = -EFAULT; ++ goto cleanup; ++ } ++ ++ len = ++ strnlen_user(sptmp->rolename, GR_SPROLE_LEN); ++ ++ if (!len || len >= GR_SPROLE_LEN) { ++ err = -EINVAL; ++ goto cleanup; ++ } ++ ++ if ((tmp = (char *) acl_alloc(len)) == NULL) { ++ err = -ENOMEM; ++ goto cleanup; ++ } ++ ++ if (copy_from_user(tmp, sptmp->rolename, len)) { ++ err = -EFAULT; ++ goto cleanup; ++ } ++ tmp[len-1] = '\0'; ++#ifdef CONFIG_GRKERNSEC_RBAC_DEBUG ++ printk(KERN_ALERT "Copying special role %s\n", tmp); ++#endif ++ sptmp->rolename = tmp; ++ acl_special_roles[i] = sptmp; ++ } ++ ++ r_utmp = (struct acl_role_label **) arg->role_db.r_table; ++ ++ for (r_num = 0; r_num < arg->role_db.num_roles; r_num++) { ++ r_tmp = acl_alloc(sizeof (struct acl_role_label)); ++ ++ if (!r_tmp) { ++ err = -ENOMEM; ++ goto cleanup; ++ } ++ ++ if (copy_from_user(&r_utmp2, r_utmp + r_num, ++ sizeof (struct acl_role_label *))) { ++ err = -EFAULT; ++ goto cleanup; ++ } ++ ++ if (copy_from_user(r_tmp, r_utmp2, ++ sizeof (struct acl_role_label))) { ++ err = -EFAULT; ++ goto cleanup; ++ } ++ ++ len = strnlen_user(r_tmp->rolename, GR_SPROLE_LEN); ++ ++ if (!len || len >= PATH_MAX) { ++ err = -EINVAL; ++ goto cleanup; ++ } ++ ++ if ((tmp = (char *) acl_alloc(len)) == NULL) { ++ err = -ENOMEM; ++ goto cleanup; ++ } ++ if (copy_from_user(tmp, r_tmp->rolename, len)) { ++ err = -EFAULT; ++ goto cleanup; ++ } ++ tmp[len-1] = '\0'; ++ r_tmp->rolename = tmp; ++ ++ if (!strcmp(r_tmp->rolename, "default") ++ && (r_tmp->roletype & GR_ROLE_DEFAULT)) { ++ default_role = r_tmp; ++ } else if (!strcmp(r_tmp->rolename, ":::kernel:::")) { ++ kernel_role = r_tmp; ++ } ++ ++ if ((ghash = (struct gr_hash_struct *) acl_alloc(sizeof(struct gr_hash_struct))) == NULL) { ++ err = -ENOMEM; ++ goto cleanup; ++ } ++ if (copy_from_user(ghash, r_tmp->hash, sizeof(struct gr_hash_struct))) { ++ err = -EFAULT; ++ goto cleanup; ++ } ++ ++ r_tmp->hash = ghash; ++ ++ num_subjs = count_user_subjs(r_tmp->hash->first); ++ ++ r_tmp->subj_hash_size = num_subjs; ++ r_tmp->subj_hash = ++ (struct acl_subject_label **) ++ create_table(&(r_tmp->subj_hash_size), sizeof(void *)); ++ ++ if (!r_tmp->subj_hash) { ++ err = -ENOMEM; ++ goto cleanup; ++ } ++ ++ err = copy_user_allowedips(r_tmp); ++ if (err) ++ goto cleanup; ++ ++ /* copy domain info */ ++ if (r_tmp->domain_children != NULL) { ++ domainlist = acl_alloc_num(r_tmp->domain_child_num, sizeof(uid_t)); ++ if (domainlist == NULL) { ++ err = -ENOMEM; ++ goto cleanup; ++ } ++ if (copy_from_user(domainlist, r_tmp->domain_children, r_tmp->domain_child_num * sizeof(uid_t))) { ++ err = -EFAULT; ++ goto cleanup; ++ } ++ r_tmp->domain_children = domainlist; ++ } ++ ++ err = copy_user_transitions(r_tmp); ++ if (err) ++ goto cleanup; ++ ++ memset(r_tmp->subj_hash, 0, ++ r_tmp->subj_hash_size * ++ sizeof (struct acl_subject_label *)); ++ ++ err = copy_user_subjs(r_tmp->hash->first, r_tmp); ++ ++ if (err) ++ goto cleanup; ++ ++ /* set nested subject list to null */ ++ r_tmp->hash->first = NULL; ++ ++ insert_acl_role_label(r_tmp); ++ } ++ ++ goto return_err; ++ cleanup: ++ free_variables(); ++ return_err: ++ return err; ++ ++} ++ ++static int ++gracl_init(struct gr_arg *args) ++{ ++ int error = 0; ++ ++ memcpy(gr_system_salt, args->salt, GR_SALT_LEN); ++ memcpy(gr_system_sum, args->sum, GR_SHA_LEN); ++ ++ if (init_variables(args)) { ++ gr_log_str(GR_DONT_AUDIT_GOOD, GR_INITF_ACL_MSG, GR_VERSION); ++ error = -ENOMEM; ++ free_variables(); ++ goto out; ++ } ++ ++ error = copy_user_acl(args); ++ free_init_variables(); ++ if (error) { ++ free_variables(); ++ goto out; ++ } ++ ++ if ((error = gr_set_acls(0))) { ++ free_variables(); ++ goto out; ++ } ++ ++ pax_open_kernel(); ++ gr_status |= GR_READY; ++ pax_close_kernel(); ++ ++ out: ++ return error; ++} ++ ++/* derived from glibc fnmatch() 0: match, 1: no match*/ ++ ++static int ++glob_match(const char *p, const char *n) ++{ ++ char c; ++ ++ while ((c = *p++) != '\0') { ++ switch (c) { ++ case '?': ++ if (*n == '\0') ++ return 1; ++ else if (*n == '/') ++ return 1; ++ break; ++ case '\\': ++ if (*n != c) ++ return 1; ++ break; ++ case '*': ++ for (c = *p++; c == '?' || c == '*'; c = *p++) { ++ if (*n == '/') ++ return 1; ++ else if (c == '?') { ++ if (*n == '\0') ++ return 1; ++ else ++ ++n; ++ } ++ } ++ if (c == '\0') { ++ return 0; ++ } else { ++ const char *endp; ++ ++ if ((endp = strchr(n, '/')) == NULL) ++ endp = n + strlen(n); ++ ++ if (c == '[') { ++ for (--p; n < endp; ++n) ++ if (!glob_match(p, n)) ++ return 0; ++ } else if (c == '/') { ++ while (*n != '\0' && *n != '/') ++ ++n; ++ if (*n == '/' && !glob_match(p, n + 1)) ++ return 0; ++ } else { ++ for (--p; n < endp; ++n) ++ if (*n == c && !glob_match(p, n)) ++ return 0; ++ } ++ ++ return 1; ++ } ++ case '[': ++ { ++ int not; ++ char cold; ++ ++ if (*n == '\0' || *n == '/') ++ return 1; ++ ++ not = (*p == '!' || *p == '^'); ++ if (not) ++ ++p; ++ ++ c = *p++; ++ for (;;) { ++ unsigned char fn = (unsigned char)*n; ++ ++ if (c == '\0') ++ return 1; ++ else { ++ if (c == fn) ++ goto matched; ++ cold = c; ++ c = *p++; ++ ++ if (c == '-' && *p != ']') { ++ unsigned char cend = *p++; ++ ++ if (cend == '\0') ++ return 1; ++ ++ if (cold <= fn && fn <= cend) ++ goto matched; ++ ++ c = *p++; ++ } ++ } ++ ++ if (c == ']') ++ break; ++ } ++ if (!not) ++ return 1; ++ break; ++ matched: ++ while (c != ']') { ++ if (c == '\0') ++ return 1; ++ ++ c = *p++; ++ } ++ if (not) ++ return 1; ++ } ++ break; ++ default: ++ if (c != *n) ++ return 1; ++ } ++ ++ ++n; ++ } ++ ++ if (*n == '\0') ++ return 0; ++ ++ if (*n == '/') ++ return 0; ++ ++ return 1; ++} ++ ++static struct acl_object_label * ++chk_glob_label(struct acl_object_label *globbed, ++ struct dentry *dentry, struct vfsmount *mnt, char **path) ++{ ++ struct acl_object_label *tmp; ++ ++ if (*path == NULL) ++ *path = gr_to_filename_nolock(dentry, mnt); ++ ++ tmp = globbed; ++ ++ while (tmp) { ++ if (!glob_match(tmp->filename, *path)) ++ return tmp; ++ tmp = tmp->next; ++ } ++ ++ return NULL; ++} ++ ++static struct acl_object_label * ++__full_lookup(const struct dentry *orig_dentry, const struct vfsmount *orig_mnt, ++ const ino_t curr_ino, const dev_t curr_dev, ++ const struct acl_subject_label *subj, char **path, const int checkglob) ++{ ++ struct acl_subject_label *tmpsubj; ++ struct acl_object_label *retval; ++ struct acl_object_label *retval2; ++ ++ tmpsubj = (struct acl_subject_label *) subj; ++ read_lock(&gr_inode_lock); ++ do { ++ retval = lookup_acl_obj_label(curr_ino, curr_dev, tmpsubj); ++ if (retval) { ++ if (checkglob && retval->globbed) { ++ retval2 = chk_glob_label(retval->globbed, (struct dentry *)orig_dentry, ++ (struct vfsmount *)orig_mnt, path); ++ if (retval2) ++ retval = retval2; ++ } ++ break; ++ } ++ } while ((tmpsubj = tmpsubj->parent_subject)); ++ read_unlock(&gr_inode_lock); ++ ++ return retval; ++} ++ ++static __inline__ struct acl_object_label * ++full_lookup(const struct dentry *orig_dentry, const struct vfsmount *orig_mnt, ++ struct dentry *curr_dentry, ++ const struct acl_subject_label *subj, char **path, const int checkglob) ++{ ++ int newglob = checkglob; ++ ino_t inode; ++ dev_t device; ++ ++ /* if we aren't checking a subdirectory of the original path yet, don't do glob checking ++ as we don't want a / * rule to match instead of the / object ++ don't do this for create lookups that call this function though, since they're looking up ++ on the parent and thus need globbing checks on all paths ++ */ ++ if (orig_dentry == curr_dentry && newglob != GR_CREATE_GLOB) ++ newglob = GR_NO_GLOB; ++ ++ spin_lock(&curr_dentry->d_lock); ++ inode = curr_dentry->d_inode->i_ino; ++ device = __get_dev(curr_dentry); ++ spin_unlock(&curr_dentry->d_lock); ++ ++ return __full_lookup(orig_dentry, orig_mnt, inode, device, subj, path, newglob); ++} ++ ++static struct acl_object_label * ++__chk_obj_label(const struct dentry *l_dentry, const struct vfsmount *l_mnt, ++ const struct acl_subject_label *subj, char *path, const int checkglob) ++{ ++ struct dentry *dentry = (struct dentry *) l_dentry; ++ struct vfsmount *mnt = (struct vfsmount *) l_mnt; ++ struct acl_object_label *retval; ++ struct dentry *parent; ++ ++ write_seqlock(&rename_lock); ++ br_read_lock(vfsmount_lock); ++ ++ if (unlikely((mnt == shm_mnt && dentry->d_inode->i_nlink == 0) || mnt == pipe_mnt || mnt == sock_mnt || ++#ifdef CONFIG_HUGETLBFS ++ (mnt == hugetlbfs_vfsmount && dentry->d_inode->i_nlink == 0) || ++#endif ++ /* ignore Eric Biederman */ ++ IS_PRIVATE(l_dentry->d_inode))) { ++ retval = fakefs_obj; ++ goto out; ++ } ++ ++ for (;;) { ++ if (dentry == real_root.dentry && mnt == real_root.mnt) ++ break; ++ ++ if (dentry == mnt->mnt_root || IS_ROOT(dentry)) { ++ if (mnt->mnt_parent == mnt) ++ break; ++ ++ retval = full_lookup(l_dentry, l_mnt, dentry, subj, &path, checkglob); ++ if (retval != NULL) ++ goto out; ++ ++ dentry = mnt->mnt_mountpoint; ++ mnt = mnt->mnt_parent; ++ continue; ++ } ++ ++ parent = dentry->d_parent; ++ retval = full_lookup(l_dentry, l_mnt, dentry, subj, &path, checkglob); ++ if (retval != NULL) ++ goto out; ++ ++ dentry = parent; ++ } ++ ++ retval = full_lookup(l_dentry, l_mnt, dentry, subj, &path, checkglob); ++ ++ /* real_root is pinned so we don't have to hold a reference */ ++ if (retval == NULL) ++ retval = full_lookup(l_dentry, l_mnt, real_root.dentry, subj, &path, checkglob); ++out: ++ br_read_unlock(vfsmount_lock); ++ write_sequnlock(&rename_lock); ++ ++ BUG_ON(retval == NULL); ++ ++ return retval; ++} ++ ++static __inline__ struct acl_object_label * ++chk_obj_label(const struct dentry *l_dentry, const struct vfsmount *l_mnt, ++ const struct acl_subject_label *subj) ++{ ++ char *path = NULL; ++ return __chk_obj_label(l_dentry, l_mnt, subj, path, GR_REG_GLOB); ++} ++ ++static __inline__ struct acl_object_label * ++chk_obj_label_noglob(const struct dentry *l_dentry, const struct vfsmount *l_mnt, ++ const struct acl_subject_label *subj) ++{ ++ char *path = NULL; ++ return __chk_obj_label(l_dentry, l_mnt, subj, path, GR_NO_GLOB); ++} ++ ++static __inline__ struct acl_object_label * ++chk_obj_create_label(const struct dentry *l_dentry, const struct vfsmount *l_mnt, ++ const struct acl_subject_label *subj, char *path) ++{ ++ return __chk_obj_label(l_dentry, l_mnt, subj, path, GR_CREATE_GLOB); ++} ++ ++static struct acl_subject_label * ++chk_subj_label(const struct dentry *l_dentry, const struct vfsmount *l_mnt, ++ const struct acl_role_label *role) ++{ ++ struct dentry *dentry = (struct dentry *) l_dentry; ++ struct vfsmount *mnt = (struct vfsmount *) l_mnt; ++ struct acl_subject_label *retval; ++ struct dentry *parent; ++ ++ write_seqlock(&rename_lock); ++ br_read_lock(vfsmount_lock); ++ ++ for (;;) { ++ if (dentry == real_root.dentry && mnt == real_root.mnt) ++ break; ++ if (dentry == mnt->mnt_root || IS_ROOT(dentry)) { ++ if (mnt->mnt_parent == mnt) ++ break; ++ ++ spin_lock(&dentry->d_lock); ++ read_lock(&gr_inode_lock); ++ retval = ++ lookup_acl_subj_label(dentry->d_inode->i_ino, ++ __get_dev(dentry), role); ++ read_unlock(&gr_inode_lock); ++ spin_unlock(&dentry->d_lock); ++ if (retval != NULL) ++ goto out; ++ ++ dentry = mnt->mnt_mountpoint; ++ mnt = mnt->mnt_parent; ++ continue; ++ } ++ ++ spin_lock(&dentry->d_lock); ++ read_lock(&gr_inode_lock); ++ retval = lookup_acl_subj_label(dentry->d_inode->i_ino, ++ __get_dev(dentry), role); ++ read_unlock(&gr_inode_lock); ++ parent = dentry->d_parent; ++ spin_unlock(&dentry->d_lock); ++ ++ if (retval != NULL) ++ goto out; ++ ++ dentry = parent; ++ } ++ ++ spin_lock(&dentry->d_lock); ++ read_lock(&gr_inode_lock); ++ retval = lookup_acl_subj_label(dentry->d_inode->i_ino, ++ __get_dev(dentry), role); ++ read_unlock(&gr_inode_lock); ++ spin_unlock(&dentry->d_lock); ++ ++ if (unlikely(retval == NULL)) { ++ /* real_root is pinned, we don't need to hold a reference */ ++ read_lock(&gr_inode_lock); ++ retval = lookup_acl_subj_label(real_root.dentry->d_inode->i_ino, ++ __get_dev(real_root.dentry), role); ++ read_unlock(&gr_inode_lock); ++ } ++out: ++ br_read_unlock(vfsmount_lock); ++ write_sequnlock(&rename_lock); ++ ++ BUG_ON(retval == NULL); ++ ++ return retval; ++} ++ ++static void ++gr_log_learn(const struct dentry *dentry, const struct vfsmount *mnt, const __u32 mode) ++{ ++ struct task_struct *task = current; ++ const struct cred *cred = current_cred(); ++ ++ security_learn(GR_LEARN_AUDIT_MSG, task->role->rolename, task->role->roletype, ++ cred->uid, cred->gid, task->exec_file ? gr_to_filename1(task->exec_file->f_path.dentry, ++ task->exec_file->f_path.mnt) : task->acl->filename, task->acl->filename, ++ 1UL, 1UL, gr_to_filename(dentry, mnt), (unsigned long) mode, &task->signal->saved_ip); ++ ++ return; ++} ++ ++static void ++gr_log_learn_sysctl(const char *path, const __u32 mode) ++{ ++ struct task_struct *task = current; ++ const struct cred *cred = current_cred(); ++ ++ security_learn(GR_LEARN_AUDIT_MSG, task->role->rolename, task->role->roletype, ++ cred->uid, cred->gid, task->exec_file ? gr_to_filename1(task->exec_file->f_path.dentry, ++ task->exec_file->f_path.mnt) : task->acl->filename, task->acl->filename, ++ 1UL, 1UL, path, (unsigned long) mode, &task->signal->saved_ip); ++ ++ return; ++} ++ ++static void ++gr_log_learn_id_change(const char type, const unsigned int real, ++ const unsigned int effective, const unsigned int fs) ++{ ++ struct task_struct *task = current; ++ const struct cred *cred = current_cred(); ++ ++ security_learn(GR_ID_LEARN_MSG, task->role->rolename, task->role->roletype, ++ cred->uid, cred->gid, task->exec_file ? gr_to_filename1(task->exec_file->f_path.dentry, ++ task->exec_file->f_path.mnt) : task->acl->filename, task->acl->filename, ++ type, real, effective, fs, &task->signal->saved_ip); ++ ++ return; ++} ++ ++__u32 ++gr_check_link(const struct dentry * new_dentry, ++ const struct dentry * parent_dentry, ++ const struct vfsmount * parent_mnt, ++ const struct dentry * old_dentry, const struct vfsmount * old_mnt) ++{ ++ struct acl_object_label *obj; ++ __u32 oldmode, newmode; ++ __u32 needmode; ++ ++ if (unlikely(!(gr_status & GR_READY))) ++ return (GR_CREATE | GR_LINK); ++ ++ obj = chk_obj_label(old_dentry, old_mnt, current->acl); ++ oldmode = obj->mode; ++ ++ if (current->acl->mode & (GR_LEARN | GR_INHERITLEARN)) ++ oldmode |= (GR_CREATE | GR_LINK); ++ ++ needmode = GR_CREATE | GR_AUDIT_CREATE | GR_SUPPRESS; ++ if (old_dentry->d_inode->i_mode & (S_ISUID | S_ISGID)) ++ needmode |= GR_SETID | GR_AUDIT_SETID; ++ ++ newmode = ++ gr_check_create(new_dentry, parent_dentry, parent_mnt, ++ oldmode | needmode); ++ ++ needmode = newmode & (GR_FIND | GR_APPEND | GR_WRITE | GR_EXEC | ++ GR_SETID | GR_READ | GR_FIND | GR_DELETE | ++ GR_INHERIT | GR_AUDIT_INHERIT); ++ ++ if (old_dentry->d_inode->i_mode & (S_ISUID | S_ISGID) && !(newmode & GR_SETID)) ++ goto bad; ++ ++ if ((oldmode & needmode) != needmode) ++ goto bad; ++ ++ needmode = oldmode & (GR_NOPTRACE | GR_PTRACERD | GR_INHERIT | GR_AUDITS); ++ if ((newmode & needmode) != needmode) ++ goto bad; ++ ++ if ((newmode & (GR_CREATE | GR_LINK)) == (GR_CREATE | GR_LINK)) ++ return newmode; ++bad: ++ needmode = oldmode; ++ if (old_dentry->d_inode->i_mode & (S_ISUID | S_ISGID)) ++ needmode |= GR_SETID; ++ ++ if (current->acl->mode & (GR_LEARN | GR_INHERITLEARN)) { ++ gr_log_learn(old_dentry, old_mnt, needmode); ++ return (GR_CREATE | GR_LINK); ++ } else if (newmode & GR_SUPPRESS) ++ return GR_SUPPRESS; ++ else ++ return 0; ++} ++ ++__u32 ++gr_search_file(const struct dentry * dentry, const __u32 mode, ++ const struct vfsmount * mnt) ++{ ++ __u32 retval = mode; ++ struct acl_subject_label *curracl; ++ struct acl_object_label *currobj; ++ ++ if (unlikely(!(gr_status & GR_READY))) ++ return (mode & ~GR_AUDITS); ++ ++ curracl = current->acl; ++ ++ currobj = chk_obj_label(dentry, mnt, curracl); ++ retval = currobj->mode & mode; ++ ++ /* if we're opening a specified transfer file for writing ++ (e.g. /dev/initctl), then transfer our role to init ++ */ ++ if (unlikely(currobj->mode & GR_INIT_TRANSFER && retval & GR_WRITE && ++ current->role->roletype & GR_ROLE_PERSIST)) { ++ struct task_struct *task = init_pid_ns.child_reaper; ++ ++ if (task->role != current->role) { ++ task->acl_sp_role = 0; ++ task->acl_role_id = current->acl_role_id; ++ task->role = current->role; ++ rcu_read_lock(); ++ read_lock(&grsec_exec_file_lock); ++ gr_apply_subject_to_task(task); ++ read_unlock(&grsec_exec_file_lock); ++ rcu_read_unlock(); ++ gr_log_noargs(GR_DONT_AUDIT_GOOD, GR_INIT_TRANSFER_MSG); ++ } ++ } ++ ++ if (unlikely ++ ((curracl->mode & (GR_LEARN | GR_INHERITLEARN)) && !(mode & GR_NOPTRACE) ++ && (retval != (mode & ~(GR_AUDITS | GR_SUPPRESS))))) { ++ __u32 new_mode = mode; ++ ++ new_mode &= ~(GR_AUDITS | GR_SUPPRESS); ++ ++ retval = new_mode; ++ ++ if (new_mode & GR_EXEC && curracl->mode & GR_INHERITLEARN) ++ new_mode |= GR_INHERIT; ++ ++ if (!(mode & GR_NOLEARN)) ++ gr_log_learn(dentry, mnt, new_mode); ++ } ++ ++ return retval; ++} ++ ++__u32 ++gr_check_create(const struct dentry * new_dentry, const struct dentry * parent, ++ const struct vfsmount * mnt, const __u32 mode) ++{ ++ struct name_entry *match; ++ struct acl_object_label *matchpo; ++ struct acl_subject_label *curracl; ++ char *path; ++ __u32 retval; ++ ++ if (unlikely(!(gr_status & GR_READY))) ++ return (mode & ~GR_AUDITS); ++ ++ preempt_disable(); ++ path = gr_to_filename_rbac(new_dentry, mnt); ++ match = lookup_name_entry_create(path); ++ ++ if (!match) ++ goto check_parent; ++ ++ curracl = current->acl; ++ ++ read_lock(&gr_inode_lock); ++ matchpo = lookup_acl_obj_label_create(match->inode, match->device, curracl); ++ read_unlock(&gr_inode_lock); ++ ++ if (matchpo) { ++ if ((matchpo->mode & mode) != ++ (mode & ~(GR_AUDITS | GR_SUPPRESS)) ++ && curracl->mode & (GR_LEARN | GR_INHERITLEARN)) { ++ __u32 new_mode = mode; ++ ++ new_mode &= ~(GR_AUDITS | GR_SUPPRESS); ++ ++ gr_log_learn(new_dentry, mnt, new_mode); ++ ++ preempt_enable(); ++ return new_mode; ++ } ++ preempt_enable(); ++ return (matchpo->mode & mode); ++ } ++ ++ check_parent: ++ curracl = current->acl; ++ ++ matchpo = chk_obj_create_label(parent, mnt, curracl, path); ++ retval = matchpo->mode & mode; ++ ++ if ((retval != (mode & ~(GR_AUDITS | GR_SUPPRESS))) ++ && (curracl->mode & (GR_LEARN | GR_INHERITLEARN))) { ++ __u32 new_mode = mode; ++ ++ new_mode &= ~(GR_AUDITS | GR_SUPPRESS); ++ ++ gr_log_learn(new_dentry, mnt, new_mode); ++ preempt_enable(); ++ return new_mode; ++ } ++ ++ preempt_enable(); ++ return retval; ++} ++ ++int ++gr_check_hidden_task(const struct task_struct *task) ++{ ++ if (unlikely(!(gr_status & GR_READY))) ++ return 0; ++ ++ if (!(task->acl->mode & GR_PROCFIND) && !(current->acl->mode & GR_VIEW)) ++ return 1; ++ ++ return 0; ++} ++ ++int ++gr_check_protected_task(const struct task_struct *task) ++{ ++ if (unlikely(!(gr_status & GR_READY) || !task)) ++ return 0; ++ ++ if ((task->acl->mode & GR_PROTECTED) && !(current->acl->mode & GR_KILL) && ++ task->acl != current->acl) ++ return 1; ++ ++ return 0; ++} ++ ++int ++gr_check_protected_task_fowner(struct pid *pid, enum pid_type type) ++{ ++ struct task_struct *p; ++ int ret = 0; ++ ++ if (unlikely(!(gr_status & GR_READY) || !pid)) ++ return ret; ++ ++ read_lock(&tasklist_lock); ++ do_each_pid_task(pid, type, p) { ++ if ((p->acl->mode & GR_PROTECTED) && !(current->acl->mode & GR_KILL) && ++ p->acl != current->acl) { ++ ret = 1; ++ goto out; ++ } ++ } while_each_pid_task(pid, type, p); ++out: ++ read_unlock(&tasklist_lock); ++ ++ return ret; ++} ++ ++void ++gr_copy_label(struct task_struct *tsk) ++{ ++ tsk->signal->used_accept = 0; ++ tsk->acl_sp_role = 0; ++ tsk->acl_role_id = current->acl_role_id; ++ tsk->acl = current->acl; ++ tsk->role = current->role; ++ tsk->signal->curr_ip = current->signal->curr_ip; ++ tsk->signal->saved_ip = current->signal->saved_ip; ++ if (current->exec_file) ++ get_file(current->exec_file); ++ tsk->exec_file = current->exec_file; ++ tsk->is_writable = current->is_writable; ++ if (unlikely(current->signal->used_accept)) { ++ current->signal->curr_ip = 0; ++ current->signal->saved_ip = 0; ++ } ++ ++ return; ++} ++ ++static void ++gr_set_proc_res(struct task_struct *task) ++{ ++ struct acl_subject_label *proc; ++ unsigned short i; ++ ++ proc = task->acl; ++ ++ if (proc->mode & (GR_LEARN | GR_INHERITLEARN)) ++ return; ++ ++ for (i = 0; i < RLIM_NLIMITS; i++) { ++ if (!(proc->resmask & (1 << i))) ++ continue; ++ ++ task->signal->rlim[i].rlim_cur = proc->res[i].rlim_cur; ++ task->signal->rlim[i].rlim_max = proc->res[i].rlim_max; ++ } ++ ++ return; ++} ++ ++int ++gr_check_user_change(int real, int effective, int fs) ++{ ++ unsigned int i; ++ __u16 num; ++ uid_t *uidlist; ++ int curuid; ++ int realok = 0; ++ int effectiveok = 0; ++ int fsok = 0; ++ ++ if (unlikely(!(gr_status & GR_READY))) ++ return 0; ++ ++ if (current->acl->mode & (GR_LEARN | GR_INHERITLEARN)) ++ gr_log_learn_id_change('u', real, effective, fs); ++ ++ num = current->acl->user_trans_num; ++ uidlist = current->acl->user_transitions; ++ ++ if (uidlist == NULL) ++ return 0; ++ ++ if (real == -1) ++ realok = 1; ++ if (effective == -1) ++ effectiveok = 1; ++ if (fs == -1) ++ fsok = 1; ++ ++ if (current->acl->user_trans_type & GR_ID_ALLOW) { ++ for (i = 0; i < num; i++) { ++ curuid = (int)uidlist[i]; ++ if (real == curuid) ++ realok = 1; ++ if (effective == curuid) ++ effectiveok = 1; ++ if (fs == curuid) ++ fsok = 1; ++ } ++ } else if (current->acl->user_trans_type & GR_ID_DENY) { ++ for (i = 0; i < num; i++) { ++ curuid = (int)uidlist[i]; ++ if (real == curuid) ++ break; ++ if (effective == curuid) ++ break; ++ if (fs == curuid) ++ break; ++ } ++ /* not in deny list */ ++ if (i == num) { ++ realok = 1; ++ effectiveok = 1; ++ fsok = 1; ++ } ++ } ++ ++ if (realok && effectiveok && fsok) ++ return 0; ++ else { ++ gr_log_int(GR_DONT_AUDIT, GR_USRCHANGE_ACL_MSG, realok ? (effectiveok ? (fsok ? 0 : fs) : effective) : real); ++ return 1; ++ } ++} ++ ++int ++gr_check_group_change(int real, int effective, int fs) ++{ ++ unsigned int i; ++ __u16 num; ++ gid_t *gidlist; ++ int curgid; ++ int realok = 0; ++ int effectiveok = 0; ++ int fsok = 0; ++ ++ if (unlikely(!(gr_status & GR_READY))) ++ return 0; ++ ++ if (current->acl->mode & (GR_LEARN | GR_INHERITLEARN)) ++ gr_log_learn_id_change('g', real, effective, fs); ++ ++ num = current->acl->group_trans_num; ++ gidlist = current->acl->group_transitions; ++ ++ if (gidlist == NULL) ++ return 0; ++ ++ if (real == -1) ++ realok = 1; ++ if (effective == -1) ++ effectiveok = 1; ++ if (fs == -1) ++ fsok = 1; ++ ++ if (current->acl->group_trans_type & GR_ID_ALLOW) { ++ for (i = 0; i < num; i++) { ++ curgid = (int)gidlist[i]; ++ if (real == curgid) ++ realok = 1; ++ if (effective == curgid) ++ effectiveok = 1; ++ if (fs == curgid) ++ fsok = 1; ++ } ++ } else if (current->acl->group_trans_type & GR_ID_DENY) { ++ for (i = 0; i < num; i++) { ++ curgid = (int)gidlist[i]; ++ if (real == curgid) ++ break; ++ if (effective == curgid) ++ break; ++ if (fs == curgid) ++ break; ++ } ++ /* not in deny list */ ++ if (i == num) { ++ realok = 1; ++ effectiveok = 1; ++ fsok = 1; ++ } ++ } ++ ++ if (realok && effectiveok && fsok) ++ return 0; ++ else { ++ gr_log_int(GR_DONT_AUDIT, GR_GRPCHANGE_ACL_MSG, realok ? (effectiveok ? (fsok ? 0 : fs) : effective) : real); ++ return 1; ++ } ++} ++ ++void ++gr_set_role_label(struct task_struct *task, const uid_t uid, const uid_t gid) ++{ ++ struct acl_role_label *role = task->role; ++ struct acl_subject_label *subj = NULL; ++ struct acl_object_label *obj; ++ struct file *filp; ++ ++ if (unlikely(!(gr_status & GR_READY))) ++ return; ++ ++ filp = task->exec_file; ++ ++ /* kernel process, we'll give them the kernel role */ ++ if (unlikely(!filp)) { ++ task->role = kernel_role; ++ task->acl = kernel_role->root_label; ++ return; ++ } else if (!task->role || !(task->role->roletype & GR_ROLE_SPECIAL)) ++ role = lookup_acl_role_label(task, uid, gid); ++ ++ /* perform subject lookup in possibly new role ++ we can use this result below in the case where role == task->role ++ */ ++ subj = chk_subj_label(filp->f_path.dentry, filp->f_path.mnt, role); ++ ++ /* if we changed uid/gid, but result in the same role ++ and are using inheritance, don't lose the inherited subject ++ if current subject is other than what normal lookup ++ would result in, we arrived via inheritance, don't ++ lose subject ++ */ ++ if (role != task->role || (!(task->acl->mode & GR_INHERITLEARN) && ++ (subj == task->acl))) ++ task->acl = subj; ++ ++ task->role = role; ++ ++ task->is_writable = 0; ++ ++ /* ignore additional mmap checks for processes that are writable ++ by the default ACL */ ++ obj = chk_obj_label(filp->f_path.dentry, filp->f_path.mnt, default_role->root_label); ++ if (unlikely(obj->mode & GR_WRITE)) ++ task->is_writable = 1; ++ obj = chk_obj_label(filp->f_path.dentry, filp->f_path.mnt, task->role->root_label); ++ if (unlikely(obj->mode & GR_WRITE)) ++ task->is_writable = 1; ++ ++#ifdef CONFIG_GRKERNSEC_RBAC_DEBUG ++ printk(KERN_ALERT "Set role label for (%s:%d): role:%s, subject:%s\n", task->comm, task->pid, task->role->rolename, task->acl->filename); ++#endif ++ ++ gr_set_proc_res(task); ++ ++ return; ++} ++ ++int ++gr_set_proc_label(const struct dentry *dentry, const struct vfsmount *mnt, ++ const int unsafe_share) ++{ ++ struct task_struct *task = current; ++ struct acl_subject_label *newacl; ++ struct acl_object_label *obj; ++ __u32 retmode; ++ ++ if (unlikely(!(gr_status & GR_READY))) ++ return 0; ++ ++ newacl = chk_subj_label(dentry, mnt, task->role); ++ ++ task_lock(task); ++ if ((((task->ptrace & PT_PTRACED) || unsafe_share) && ++ !(task->acl->mode & GR_POVERRIDE) && (task->acl != newacl) && ++ !(task->role->roletype & GR_ROLE_GOD) && ++ !gr_search_file(dentry, GR_PTRACERD, mnt) && ++ !(task->acl->mode & (GR_LEARN | GR_INHERITLEARN)))) { ++ task_unlock(task); ++ if (unsafe_share) ++ gr_log_fs_generic(GR_DONT_AUDIT, GR_UNSAFESHARE_EXEC_ACL_MSG, dentry, mnt); ++ else ++ gr_log_fs_generic(GR_DONT_AUDIT, GR_PTRACE_EXEC_ACL_MSG, dentry, mnt); ++ return -EACCES; ++ } ++ task_unlock(task); ++ ++ obj = chk_obj_label(dentry, mnt, task->acl); ++ retmode = obj->mode & (GR_INHERIT | GR_AUDIT_INHERIT); ++ ++ if (!(task->acl->mode & GR_INHERITLEARN) && ++ ((newacl->mode & GR_LEARN) || !(retmode & GR_INHERIT))) { ++ if (obj->nested) ++ task->acl = obj->nested; ++ else ++ task->acl = newacl; ++ } else if (retmode & GR_INHERIT && retmode & GR_AUDIT_INHERIT) ++ gr_log_str_fs(GR_DO_AUDIT, GR_INHERIT_ACL_MSG, task->acl->filename, dentry, mnt); ++ ++ task->is_writable = 0; ++ ++ /* ignore additional mmap checks for processes that are writable ++ by the default ACL */ ++ obj = chk_obj_label(dentry, mnt, default_role->root_label); ++ if (unlikely(obj->mode & GR_WRITE)) ++ task->is_writable = 1; ++ obj = chk_obj_label(dentry, mnt, task->role->root_label); ++ if (unlikely(obj->mode & GR_WRITE)) ++ task->is_writable = 1; ++ ++ gr_set_proc_res(task); ++ ++#ifdef CONFIG_GRKERNSEC_RBAC_DEBUG ++ printk(KERN_ALERT "Set subject label for (%s:%d): role:%s, subject:%s\n", task->comm, task->pid, task->role->rolename, task->acl->filename); ++#endif ++ return 0; ++} ++ ++/* always called with valid inodev ptr */ ++static void ++do_handle_delete(struct inodev_entry *inodev, const ino_t ino, const dev_t dev) ++{ ++ struct acl_object_label *matchpo; ++ struct acl_subject_label *matchps; ++ struct acl_subject_label *subj; ++ struct acl_role_label *role; ++ unsigned int x; ++ ++ FOR_EACH_ROLE_START(role) ++ FOR_EACH_SUBJECT_START(role, subj, x) ++ if ((matchpo = lookup_acl_obj_label(ino, dev, subj)) != NULL) ++ matchpo->mode |= GR_DELETED; ++ FOR_EACH_SUBJECT_END(subj,x) ++ FOR_EACH_NESTED_SUBJECT_START(role, subj) ++ if (subj->inode == ino && subj->device == dev) ++ subj->mode |= GR_DELETED; ++ FOR_EACH_NESTED_SUBJECT_END(subj) ++ if ((matchps = lookup_acl_subj_label(ino, dev, role)) != NULL) ++ matchps->mode |= GR_DELETED; ++ FOR_EACH_ROLE_END(role) ++ ++ inodev->nentry->deleted = 1; ++ ++ return; ++} ++ ++void ++gr_handle_delete(const ino_t ino, const dev_t dev) ++{ ++ struct inodev_entry *inodev; ++ ++ if (unlikely(!(gr_status & GR_READY))) ++ return; ++ ++ write_lock(&gr_inode_lock); ++ inodev = lookup_inodev_entry(ino, dev); ++ if (inodev != NULL) ++ do_handle_delete(inodev, ino, dev); ++ write_unlock(&gr_inode_lock); ++ ++ return; ++} ++ ++static void ++update_acl_obj_label(const ino_t oldinode, const dev_t olddevice, ++ const ino_t newinode, const dev_t newdevice, ++ struct acl_subject_label *subj) ++{ ++ unsigned int index = fhash(oldinode, olddevice, subj->obj_hash_size); ++ struct acl_object_label *match; ++ ++ match = subj->obj_hash[index]; ++ ++ while (match && (match->inode != oldinode || ++ match->device != olddevice || ++ !(match->mode & GR_DELETED))) ++ match = match->next; ++ ++ if (match && (match->inode == oldinode) ++ && (match->device == olddevice) ++ && (match->mode & GR_DELETED)) { ++ if (match->prev == NULL) { ++ subj->obj_hash[index] = match->next; ++ if (match->next != NULL) ++ match->next->prev = NULL; ++ } else { ++ match->prev->next = match->next; ++ if (match->next != NULL) ++ match->next->prev = match->prev; ++ } ++ match->prev = NULL; ++ match->next = NULL; ++ match->inode = newinode; ++ match->device = newdevice; ++ match->mode &= ~GR_DELETED; ++ ++ insert_acl_obj_label(match, subj); ++ } ++ ++ return; ++} ++ ++static void ++update_acl_subj_label(const ino_t oldinode, const dev_t olddevice, ++ const ino_t newinode, const dev_t newdevice, ++ struct acl_role_label *role) ++{ ++ unsigned int index = fhash(oldinode, olddevice, role->subj_hash_size); ++ struct acl_subject_label *match; ++ ++ match = role->subj_hash[index]; ++ ++ while (match && (match->inode != oldinode || ++ match->device != olddevice || ++ !(match->mode & GR_DELETED))) ++ match = match->next; ++ ++ if (match && (match->inode == oldinode) ++ && (match->device == olddevice) ++ && (match->mode & GR_DELETED)) { ++ if (match->prev == NULL) { ++ role->subj_hash[index] = match->next; ++ if (match->next != NULL) ++ match->next->prev = NULL; ++ } else { ++ match->prev->next = match->next; ++ if (match->next != NULL) ++ match->next->prev = match->prev; ++ } ++ match->prev = NULL; ++ match->next = NULL; ++ match->inode = newinode; ++ match->device = newdevice; ++ match->mode &= ~GR_DELETED; ++ ++ insert_acl_subj_label(match, role); ++ } ++ ++ return; ++} ++ ++static void ++update_inodev_entry(const ino_t oldinode, const dev_t olddevice, ++ const ino_t newinode, const dev_t newdevice) ++{ ++ unsigned int index = fhash(oldinode, olddevice, inodev_set.i_size); ++ struct inodev_entry *match; ++ ++ match = inodev_set.i_hash[index]; ++ ++ while (match && (match->nentry->inode != oldinode || ++ match->nentry->device != olddevice || !match->nentry->deleted)) ++ match = match->next; ++ ++ if (match && (match->nentry->inode == oldinode) ++ && (match->nentry->device == olddevice) && ++ match->nentry->deleted) { ++ if (match->prev == NULL) { ++ inodev_set.i_hash[index] = match->next; ++ if (match->next != NULL) ++ match->next->prev = NULL; ++ } else { ++ match->prev->next = match->next; ++ if (match->next != NULL) ++ match->next->prev = match->prev; ++ } ++ match->prev = NULL; ++ match->next = NULL; ++ match->nentry->inode = newinode; ++ match->nentry->device = newdevice; ++ match->nentry->deleted = 0; ++ ++ insert_inodev_entry(match); ++ } ++ ++ return; ++} ++ ++static void ++do_handle_create(const struct name_entry *matchn, const struct dentry *dentry, ++ const struct vfsmount *mnt) ++{ ++ struct acl_subject_label *subj; ++ struct acl_role_label *role; ++ unsigned int x; ++ ino_t ino = dentry->d_inode->i_ino; ++ dev_t dev = __get_dev(dentry); ++ ++ FOR_EACH_ROLE_START(role) ++ update_acl_subj_label(matchn->inode, matchn->device, ino, dev, role); ++ ++ FOR_EACH_NESTED_SUBJECT_START(role, subj) ++ if ((subj->inode == ino) && (subj->device == dev)) { ++ subj->inode = ino; ++ subj->device = dev; ++ } ++ FOR_EACH_NESTED_SUBJECT_END(subj) ++ FOR_EACH_SUBJECT_START(role, subj, x) ++ update_acl_obj_label(matchn->inode, matchn->device, ++ ino, dev, subj); ++ FOR_EACH_SUBJECT_END(subj,x) ++ FOR_EACH_ROLE_END(role) ++ ++ update_inodev_entry(matchn->inode, matchn->device, ino, dev); ++ ++ return; ++} ++ ++void ++gr_handle_create(const struct dentry *dentry, const struct vfsmount *mnt) ++{ ++ struct name_entry *matchn; ++ ++ if (unlikely(!(gr_status & GR_READY))) ++ return; ++ ++ preempt_disable(); ++ matchn = lookup_name_entry(gr_to_filename_rbac(dentry, mnt)); ++ ++ if (unlikely((unsigned long)matchn)) { ++ write_lock(&gr_inode_lock); ++ do_handle_create(matchn, dentry, mnt); ++ write_unlock(&gr_inode_lock); ++ } ++ preempt_enable(); ++ ++ return; ++} ++ ++void ++gr_handle_rename(struct inode *old_dir, struct inode *new_dir, ++ struct dentry *old_dentry, ++ struct dentry *new_dentry, ++ struct vfsmount *mnt, const __u8 replace) ++{ ++ struct name_entry *matchn; ++ struct inodev_entry *inodev; ++ ino_t old_ino = old_dentry->d_inode->i_ino; ++ dev_t old_dev = __get_dev(old_dentry); ++ ++ /* vfs_rename swaps the name and parent link for old_dentry and ++ new_dentry ++ at this point, old_dentry has the new name, parent link, and inode ++ for the renamed file ++ if a file is being replaced by a rename, new_dentry has the inode ++ and name for the replaced file ++ */ ++ ++ if (unlikely(!(gr_status & GR_READY))) ++ return; ++ ++ preempt_disable(); ++ matchn = lookup_name_entry(gr_to_filename_rbac(old_dentry, mnt)); ++ ++ /* we wouldn't have to check d_inode if it weren't for ++ NFS silly-renaming ++ */ ++ ++ write_lock(&gr_inode_lock); ++ if (unlikely(replace && new_dentry->d_inode)) { ++ ino_t new_ino = new_dentry->d_inode->i_ino; ++ dev_t new_dev = __get_dev(new_dentry); ++ ++ inodev = lookup_inodev_entry(new_ino, new_dev); ++ if (inodev != NULL && (new_dentry->d_inode->i_nlink <= 1)) ++ do_handle_delete(inodev, new_ino, new_dev); ++ } ++ ++ inodev = lookup_inodev_entry(old_ino, old_dev); ++ if (inodev != NULL && (old_dentry->d_inode->i_nlink <= 1)) ++ do_handle_delete(inodev, old_ino, old_dev); ++ ++ if (unlikely((unsigned long)matchn)) ++ do_handle_create(matchn, old_dentry, mnt); ++ ++ write_unlock(&gr_inode_lock); ++ preempt_enable(); ++ ++ return; ++} ++ ++static int ++lookup_special_role_auth(__u16 mode, const char *rolename, unsigned char **salt, ++ unsigned char **sum) ++{ ++ struct acl_role_label *r; ++ struct role_allowed_ip *ipp; ++ struct role_transition *trans; ++ unsigned int i; ++ int found = 0; ++ u32 curr_ip = current->signal->curr_ip; ++ ++ current->signal->saved_ip = curr_ip; ++ ++ /* check transition table */ ++ ++ for (trans = current->role->transitions; trans; trans = trans->next) { ++ if (!strcmp(rolename, trans->rolename)) { ++ found = 1; ++ break; ++ } ++ } ++ ++ if (!found) ++ return 0; ++ ++ /* handle special roles that do not require authentication ++ and check ip */ ++ ++ FOR_EACH_ROLE_START(r) ++ if (!strcmp(rolename, r->rolename) && ++ (r->roletype & GR_ROLE_SPECIAL)) { ++ found = 0; ++ if (r->allowed_ips != NULL) { ++ for (ipp = r->allowed_ips; ipp; ipp = ipp->next) { ++ if ((ntohl(curr_ip) & ipp->netmask) == ++ (ntohl(ipp->addr) & ipp->netmask)) ++ found = 1; ++ } ++ } else ++ found = 2; ++ if (!found) ++ return 0; ++ ++ if (((mode == GR_SPROLE) && (r->roletype & GR_ROLE_NOPW)) || ++ ((mode == GR_SPROLEPAM) && (r->roletype & GR_ROLE_PAM))) { ++ *salt = NULL; ++ *sum = NULL; ++ return 1; ++ } ++ } ++ FOR_EACH_ROLE_END(r) ++ ++ for (i = 0; i < num_sprole_pws; i++) { ++ if (!strcmp(rolename, acl_special_roles[i]->rolename)) { ++ *salt = acl_special_roles[i]->salt; ++ *sum = acl_special_roles[i]->sum; ++ return 1; ++ } ++ } ++ ++ return 0; ++} ++ ++static void ++assign_special_role(char *rolename) ++{ ++ struct acl_object_label *obj; ++ struct acl_role_label *r; ++ struct acl_role_label *assigned = NULL; ++ struct task_struct *tsk; ++ struct file *filp; ++ ++ FOR_EACH_ROLE_START(r) ++ if (!strcmp(rolename, r->rolename) && ++ (r->roletype & GR_ROLE_SPECIAL)) { ++ assigned = r; ++ break; ++ } ++ FOR_EACH_ROLE_END(r) ++ ++ if (!assigned) ++ return; ++ ++ read_lock(&tasklist_lock); ++ read_lock(&grsec_exec_file_lock); ++ ++ tsk = current->real_parent; ++ if (tsk == NULL) ++ goto out_unlock; ++ ++ filp = tsk->exec_file; ++ if (filp == NULL) ++ goto out_unlock; ++ ++ tsk->is_writable = 0; ++ ++ tsk->acl_sp_role = 1; ++ tsk->acl_role_id = ++acl_sp_role_value; ++ tsk->role = assigned; ++ tsk->acl = chk_subj_label(filp->f_path.dentry, filp->f_path.mnt, tsk->role); ++ ++ /* ignore additional mmap checks for processes that are writable ++ by the default ACL */ ++ obj = chk_obj_label(filp->f_path.dentry, filp->f_path.mnt, default_role->root_label); ++ if (unlikely(obj->mode & GR_WRITE)) ++ tsk->is_writable = 1; ++ obj = chk_obj_label(filp->f_path.dentry, filp->f_path.mnt, tsk->role->root_label); ++ if (unlikely(obj->mode & GR_WRITE)) ++ tsk->is_writable = 1; ++ ++#ifdef CONFIG_GRKERNSEC_RBAC_DEBUG ++ printk(KERN_ALERT "Assigning special role:%s subject:%s to process (%s:%d)\n", tsk->role->rolename, tsk->acl->filename, tsk->comm, tsk->pid); ++#endif ++ ++out_unlock: ++ read_unlock(&grsec_exec_file_lock); ++ read_unlock(&tasklist_lock); ++ return; ++} ++ ++int gr_check_secure_terminal(struct task_struct *task) ++{ ++ struct task_struct *p, *p2, *p3; ++ struct files_struct *files; ++ struct fdtable *fdt; ++ struct file *our_file = NULL, *file; ++ int i; ++ ++ if (task->signal->tty == NULL) ++ return 1; ++ ++ files = get_files_struct(task); ++ if (files != NULL) { ++ rcu_read_lock(); ++ fdt = files_fdtable(files); ++ for (i=0; i < fdt->max_fds; i++) { ++ file = fcheck_files(files, i); ++ if (file && (our_file == NULL) && (file->private_data == task->signal->tty)) { ++ get_file(file); ++ our_file = file; ++ } ++ } ++ rcu_read_unlock(); ++ put_files_struct(files); ++ } ++ ++ if (our_file == NULL) ++ return 1; ++ ++ read_lock(&tasklist_lock); ++ do_each_thread(p2, p) { ++ files = get_files_struct(p); ++ if (files == NULL || ++ (p->signal && p->signal->tty == task->signal->tty)) { ++ if (files != NULL) ++ put_files_struct(files); ++ continue; ++ } ++ rcu_read_lock(); ++ fdt = files_fdtable(files); ++ for (i=0; i < fdt->max_fds; i++) { ++ file = fcheck_files(files, i); ++ if (file && S_ISCHR(file->f_path.dentry->d_inode->i_mode) && ++ file->f_path.dentry->d_inode->i_rdev == our_file->f_path.dentry->d_inode->i_rdev) { ++ p3 = task; ++ while (p3->pid > 0) { ++ if (p3 == p) ++ break; ++ p3 = p3->real_parent; ++ } ++ if (p3 == p) ++ break; ++ gr_log_ttysniff(GR_DONT_AUDIT_GOOD, GR_TTYSNIFF_ACL_MSG, p); ++ gr_handle_alertkill(p); ++ rcu_read_unlock(); ++ put_files_struct(files); ++ read_unlock(&tasklist_lock); ++ fput(our_file); ++ return 0; ++ } ++ } ++ rcu_read_unlock(); ++ put_files_struct(files); ++ } while_each_thread(p2, p); ++ read_unlock(&tasklist_lock); ++ ++ fput(our_file); ++ return 1; ++} ++ ++ssize_t ++write_grsec_handler(struct file *file, const char * buf, size_t count, loff_t *ppos) ++{ ++ struct gr_arg_wrapper uwrap; ++ unsigned char *sprole_salt = NULL; ++ unsigned char *sprole_sum = NULL; ++ int error = sizeof (struct gr_arg_wrapper); ++ int error2 = 0; ++ ++ mutex_lock(&gr_dev_mutex); ++ ++ if ((gr_status & GR_READY) && !(current->acl->mode & GR_KERNELAUTH)) { ++ error = -EPERM; ++ goto out; ++ } ++ ++ if (count != sizeof (struct gr_arg_wrapper)) { ++ gr_log_int_int(GR_DONT_AUDIT_GOOD, GR_DEV_ACL_MSG, (int)count, (int)sizeof(struct gr_arg_wrapper)); ++ error = -EINVAL; ++ goto out; ++ } ++ ++ ++ if (gr_auth_expires && time_after_eq(get_seconds(), gr_auth_expires)) { ++ gr_auth_expires = 0; ++ gr_auth_attempts = 0; ++ } ++ ++ if (copy_from_user(&uwrap, buf, sizeof (struct gr_arg_wrapper))) { ++ error = -EFAULT; ++ goto out; ++ } ++ ++ if ((uwrap.version != GRSECURITY_VERSION) || (uwrap.size != sizeof(struct gr_arg))) { ++ error = -EINVAL; ++ goto out; ++ } ++ ++ if (copy_from_user(gr_usermode, uwrap.arg, sizeof (struct gr_arg))) { ++ error = -EFAULT; ++ goto out; ++ } ++ ++ if (gr_usermode->mode != GR_SPROLE && gr_usermode->mode != GR_SPROLEPAM && ++ gr_auth_attempts >= CONFIG_GRKERNSEC_ACL_MAXTRIES && ++ time_after(gr_auth_expires, get_seconds())) { ++ error = -EBUSY; ++ goto out; ++ } ++ ++ /* if non-root trying to do anything other than use a special role, ++ do not attempt authentication, do not count towards authentication ++ locking ++ */ ++ ++ if (gr_usermode->mode != GR_SPROLE && gr_usermode->mode != GR_STATUS && ++ gr_usermode->mode != GR_UNSPROLE && gr_usermode->mode != GR_SPROLEPAM && ++ current_uid()) { ++ error = -EPERM; ++ goto out; ++ } ++ ++ /* ensure pw and special role name are null terminated */ ++ ++ gr_usermode->pw[GR_PW_LEN - 1] = '\0'; ++ gr_usermode->sp_role[GR_SPROLE_LEN - 1] = '\0'; ++ ++ /* Okay. ++ * We have our enough of the argument structure..(we have yet ++ * to copy_from_user the tables themselves) . Copy the tables ++ * only if we need them, i.e. for loading operations. */ ++ ++ switch (gr_usermode->mode) { ++ case GR_STATUS: ++ if (gr_status & GR_READY) { ++ error = 1; ++ if (!gr_check_secure_terminal(current)) ++ error = 3; ++ } else ++ error = 2; ++ goto out; ++ case GR_SHUTDOWN: ++ if ((gr_status & GR_READY) ++ && !(chkpw(gr_usermode, gr_system_salt, gr_system_sum))) { ++ pax_open_kernel(); ++ gr_status &= ~GR_READY; ++ pax_close_kernel(); ++ ++ gr_log_noargs(GR_DONT_AUDIT_GOOD, GR_SHUTS_ACL_MSG); ++ free_variables(); ++ memset(gr_usermode, 0, sizeof (struct gr_arg)); ++ memset(gr_system_salt, 0, GR_SALT_LEN); ++ memset(gr_system_sum, 0, GR_SHA_LEN); ++ } else if (gr_status & GR_READY) { ++ gr_log_noargs(GR_DONT_AUDIT, GR_SHUTF_ACL_MSG); ++ error = -EPERM; ++ } else { ++ gr_log_noargs(GR_DONT_AUDIT_GOOD, GR_SHUTI_ACL_MSG); ++ error = -EAGAIN; ++ } ++ break; ++ case GR_ENABLE: ++ if (!(gr_status & GR_READY) && !(error2 = gracl_init(gr_usermode))) ++ gr_log_str(GR_DONT_AUDIT_GOOD, GR_ENABLE_ACL_MSG, GR_VERSION); ++ else { ++ if (gr_status & GR_READY) ++ error = -EAGAIN; ++ else ++ error = error2; ++ gr_log_str(GR_DONT_AUDIT, GR_ENABLEF_ACL_MSG, GR_VERSION); ++ } ++ break; ++ case GR_RELOAD: ++ if (!(gr_status & GR_READY)) { ++ gr_log_str(GR_DONT_AUDIT_GOOD, GR_RELOADI_ACL_MSG, GR_VERSION); ++ error = -EAGAIN; ++ } else if (!(chkpw(gr_usermode, gr_system_salt, gr_system_sum))) { ++ preempt_disable(); ++ ++ pax_open_kernel(); ++ gr_status &= ~GR_READY; ++ pax_close_kernel(); ++ ++ free_variables(); ++ if (!(error2 = gracl_init(gr_usermode))) { ++ preempt_enable(); ++ gr_log_str(GR_DONT_AUDIT_GOOD, GR_RELOAD_ACL_MSG, GR_VERSION); ++ } else { ++ preempt_enable(); ++ error = error2; ++ gr_log_str(GR_DONT_AUDIT, GR_RELOADF_ACL_MSG, GR_VERSION); ++ } ++ } else { ++ gr_log_str(GR_DONT_AUDIT, GR_RELOADF_ACL_MSG, GR_VERSION); ++ error = -EPERM; ++ } ++ break; ++ case GR_SEGVMOD: ++ if (unlikely(!(gr_status & GR_READY))) { ++ gr_log_noargs(GR_DONT_AUDIT_GOOD, GR_SEGVMODI_ACL_MSG); ++ error = -EAGAIN; ++ break; ++ } ++ ++ if (!(chkpw(gr_usermode, gr_system_salt, gr_system_sum))) { ++ gr_log_noargs(GR_DONT_AUDIT_GOOD, GR_SEGVMODS_ACL_MSG); ++ if (gr_usermode->segv_device && gr_usermode->segv_inode) { ++ struct acl_subject_label *segvacl; ++ segvacl = ++ lookup_acl_subj_label(gr_usermode->segv_inode, ++ gr_usermode->segv_device, ++ current->role); ++ if (segvacl) { ++ segvacl->crashes = 0; ++ segvacl->expires = 0; ++ } ++ } else if (gr_find_uid(gr_usermode->segv_uid) >= 0) { ++ gr_remove_uid(gr_usermode->segv_uid); ++ } ++ } else { ++ gr_log_noargs(GR_DONT_AUDIT, GR_SEGVMODF_ACL_MSG); ++ error = -EPERM; ++ } ++ break; ++ case GR_SPROLE: ++ case GR_SPROLEPAM: ++ if (unlikely(!(gr_status & GR_READY))) { ++ gr_log_noargs(GR_DONT_AUDIT_GOOD, GR_SPROLEI_ACL_MSG); ++ error = -EAGAIN; ++ break; ++ } ++ ++ if (current->role->expires && time_after_eq(get_seconds(), current->role->expires)) { ++ current->role->expires = 0; ++ current->role->auth_attempts = 0; ++ } ++ ++ if (current->role->auth_attempts >= CONFIG_GRKERNSEC_ACL_MAXTRIES && ++ time_after(current->role->expires, get_seconds())) { ++ error = -EBUSY; ++ goto out; ++ } ++ ++ if (lookup_special_role_auth ++ (gr_usermode->mode, gr_usermode->sp_role, &sprole_salt, &sprole_sum) ++ && ((!sprole_salt && !sprole_sum) ++ || !(chkpw(gr_usermode, sprole_salt, sprole_sum)))) { ++ char *p = ""; ++ assign_special_role(gr_usermode->sp_role); ++ read_lock(&tasklist_lock); ++ if (current->real_parent) ++ p = current->real_parent->role->rolename; ++ read_unlock(&tasklist_lock); ++ gr_log_str_int(GR_DONT_AUDIT_GOOD, GR_SPROLES_ACL_MSG, ++ p, acl_sp_role_value); ++ } else { ++ gr_log_str(GR_DONT_AUDIT, GR_SPROLEF_ACL_MSG, gr_usermode->sp_role); ++ error = -EPERM; ++ if(!(current->role->auth_attempts++)) ++ current->role->expires = get_seconds() + CONFIG_GRKERNSEC_ACL_TIMEOUT; ++ ++ goto out; ++ } ++ break; ++ case GR_UNSPROLE: ++ if (unlikely(!(gr_status & GR_READY))) { ++ gr_log_noargs(GR_DONT_AUDIT_GOOD, GR_UNSPROLEI_ACL_MSG); ++ error = -EAGAIN; ++ break; ++ } ++ ++ if (current->role->roletype & GR_ROLE_SPECIAL) { ++ char *p = ""; ++ int i = 0; ++ ++ read_lock(&tasklist_lock); ++ if (current->real_parent) { ++ p = current->real_parent->role->rolename; ++ i = current->real_parent->acl_role_id; ++ } ++ read_unlock(&tasklist_lock); ++ ++ gr_log_str_int(GR_DONT_AUDIT_GOOD, GR_UNSPROLES_ACL_MSG, p, i); ++ gr_set_acls(1); ++ } else { ++ error = -EPERM; ++ goto out; ++ } ++ break; ++ default: ++ gr_log_int(GR_DONT_AUDIT, GR_INVMODE_ACL_MSG, gr_usermode->mode); ++ error = -EINVAL; ++ break; ++ } ++ ++ if (error != -EPERM) ++ goto out; ++ ++ if(!(gr_auth_attempts++)) ++ gr_auth_expires = get_seconds() + CONFIG_GRKERNSEC_ACL_TIMEOUT; ++ ++ out: ++ mutex_unlock(&gr_dev_mutex); ++ return error; ++} ++ ++/* must be called with ++ rcu_read_lock(); ++ read_lock(&tasklist_lock); ++ read_lock(&grsec_exec_file_lock); ++*/ ++int gr_apply_subject_to_task(struct task_struct *task) ++{ ++ struct acl_object_label *obj; ++ char *tmpname; ++ struct acl_subject_label *tmpsubj; ++ struct file *filp; ++ struct name_entry *nmatch; ++ ++ filp = task->exec_file; ++ if (filp == NULL) ++ return 0; ++ ++ /* the following is to apply the correct subject ++ on binaries running when the RBAC system ++ is enabled, when the binaries have been ++ replaced or deleted since their execution ++ ----- ++ when the RBAC system starts, the inode/dev ++ from exec_file will be one the RBAC system ++ is unaware of. It only knows the inode/dev ++ of the present file on disk, or the absence ++ of it. ++ */ ++ preempt_disable(); ++ tmpname = gr_to_filename_rbac(filp->f_path.dentry, filp->f_path.mnt); ++ ++ nmatch = lookup_name_entry(tmpname); ++ preempt_enable(); ++ tmpsubj = NULL; ++ if (nmatch) { ++ if (nmatch->deleted) ++ tmpsubj = lookup_acl_subj_label_deleted(nmatch->inode, nmatch->device, task->role); ++ else ++ tmpsubj = lookup_acl_subj_label(nmatch->inode, nmatch->device, task->role); ++ if (tmpsubj != NULL) ++ task->acl = tmpsubj; ++ } ++ if (tmpsubj == NULL) ++ task->acl = chk_subj_label(filp->f_path.dentry, filp->f_path.mnt, ++ task->role); ++ if (task->acl) { ++ struct acl_subject_label *curr; ++ curr = task->acl; ++ ++ task->is_writable = 0; ++ /* ignore additional mmap checks for processes that are writable ++ by the default ACL */ ++ obj = chk_obj_label(filp->f_path.dentry, filp->f_path.mnt, default_role->root_label); ++ if (unlikely(obj->mode & GR_WRITE)) ++ task->is_writable = 1; ++ obj = chk_obj_label(filp->f_path.dentry, filp->f_path.mnt, task->role->root_label); ++ if (unlikely(obj->mode & GR_WRITE)) ++ task->is_writable = 1; ++ ++ gr_set_proc_res(task); ++ ++#ifdef CONFIG_GRKERNSEC_RBAC_DEBUG ++ printk(KERN_ALERT "gr_set_acls for (%s:%d): role:%s, subject:%s\n", task->comm, task->pid, task->role->rolename, task->acl->filename); ++#endif ++ } else { ++ return 1; ++ } ++ ++ return 0; ++} ++ ++int ++gr_set_acls(const int type) ++{ ++ struct task_struct *task, *task2; ++ struct acl_role_label *role = current->role; ++ __u16 acl_role_id = current->acl_role_id; ++ const struct cred *cred; ++ int ret; ++ ++ rcu_read_lock(); ++ read_lock(&tasklist_lock); ++ read_lock(&grsec_exec_file_lock); ++ do_each_thread(task2, task) { ++ /* check to see if we're called from the exit handler, ++ if so, only replace ACLs that have inherited the admin ++ ACL */ ++ ++ if (type && (task->role != role || ++ task->acl_role_id != acl_role_id)) ++ continue; ++ ++ task->acl_role_id = 0; ++ task->acl_sp_role = 0; ++ ++ if (task->exec_file) { ++ cred = __task_cred(task); ++ task->role = lookup_acl_role_label(task, cred->uid, cred->gid); ++ ret = gr_apply_subject_to_task(task); ++ if (ret) { ++ read_unlock(&grsec_exec_file_lock); ++ read_unlock(&tasklist_lock); ++ rcu_read_unlock(); ++ gr_log_str_int(GR_DONT_AUDIT_GOOD, GR_DEFACL_MSG, task->comm, task->pid); ++ return ret; ++ } ++ } else { ++ // it's a kernel process ++ task->role = kernel_role; ++ task->acl = kernel_role->root_label; ++#ifdef CONFIG_GRKERNSEC_ACL_HIDEKERN ++ task->acl->mode &= ~GR_PROCFIND; ++#endif ++ } ++ } while_each_thread(task2, task); ++ read_unlock(&grsec_exec_file_lock); ++ read_unlock(&tasklist_lock); ++ rcu_read_unlock(); ++ ++ return 0; ++} ++ ++void ++gr_learn_resource(const struct task_struct *task, ++ const int res, const unsigned long wanted, const int gt) ++{ ++ struct acl_subject_label *acl; ++ const struct cred *cred; ++ ++ if (unlikely((gr_status & GR_READY) && ++ task->acl && (task->acl->mode & (GR_LEARN | GR_INHERITLEARN)))) ++ goto skip_reslog; ++ ++#ifdef CONFIG_GRKERNSEC_RESLOG ++ gr_log_resource(task, res, wanted, gt); ++#endif ++ skip_reslog: ++ ++ if (unlikely(!(gr_status & GR_READY) || !wanted || res >= GR_NLIMITS)) ++ return; ++ ++ acl = task->acl; ++ ++ if (likely(!acl || !(acl->mode & (GR_LEARN | GR_INHERITLEARN)) || ++ !(acl->resmask & (1 << (unsigned short) res)))) ++ return; ++ ++ if (wanted >= acl->res[res].rlim_cur) { ++ unsigned long res_add; ++ ++ res_add = wanted; ++ switch (res) { ++ case RLIMIT_CPU: ++ res_add += GR_RLIM_CPU_BUMP; ++ break; ++ case RLIMIT_FSIZE: ++ res_add += GR_RLIM_FSIZE_BUMP; ++ break; ++ case RLIMIT_DATA: ++ res_add += GR_RLIM_DATA_BUMP; ++ break; ++ case RLIMIT_STACK: ++ res_add += GR_RLIM_STACK_BUMP; ++ break; ++ case RLIMIT_CORE: ++ res_add += GR_RLIM_CORE_BUMP; ++ break; ++ case RLIMIT_RSS: ++ res_add += GR_RLIM_RSS_BUMP; ++ break; ++ case RLIMIT_NPROC: ++ res_add += GR_RLIM_NPROC_BUMP; ++ break; ++ case RLIMIT_NOFILE: ++ res_add += GR_RLIM_NOFILE_BUMP; ++ break; ++ case RLIMIT_MEMLOCK: ++ res_add += GR_RLIM_MEMLOCK_BUMP; ++ break; ++ case RLIMIT_AS: ++ res_add += GR_RLIM_AS_BUMP; ++ break; ++ case RLIMIT_LOCKS: ++ res_add += GR_RLIM_LOCKS_BUMP; ++ break; ++ case RLIMIT_SIGPENDING: ++ res_add += GR_RLIM_SIGPENDING_BUMP; ++ break; ++ case RLIMIT_MSGQUEUE: ++ res_add += GR_RLIM_MSGQUEUE_BUMP; ++ break; ++ case RLIMIT_NICE: ++ res_add += GR_RLIM_NICE_BUMP; ++ break; ++ case RLIMIT_RTPRIO: ++ res_add += GR_RLIM_RTPRIO_BUMP; ++ break; ++ case RLIMIT_RTTIME: ++ res_add += GR_RLIM_RTTIME_BUMP; ++ break; ++ } ++ ++ acl->res[res].rlim_cur = res_add; ++ ++ if (wanted > acl->res[res].rlim_max) ++ acl->res[res].rlim_max = res_add; ++ ++ /* only log the subject filename, since resource logging is supported for ++ single-subject learning only */ ++ rcu_read_lock(); ++ cred = __task_cred(task); ++ security_learn(GR_LEARN_AUDIT_MSG, task->role->rolename, ++ task->role->roletype, cred->uid, cred->gid, acl->filename, ++ acl->filename, acl->res[res].rlim_cur, acl->res[res].rlim_max, ++ "", (unsigned long) res, &task->signal->saved_ip); ++ rcu_read_unlock(); ++ } ++ ++ return; ++} ++ ++#if defined(CONFIG_PAX_HAVE_ACL_FLAGS) && (defined(CONFIG_PAX_NOEXEC) || defined(CONFIG_PAX_ASLR)) ++void ++pax_set_initial_flags(struct linux_binprm *bprm) ++{ ++ struct task_struct *task = current; ++ struct acl_subject_label *proc; ++ unsigned long flags; ++ ++ if (unlikely(!(gr_status & GR_READY))) ++ return; ++ ++ flags = pax_get_flags(task); ++ ++ proc = task->acl; ++ ++ if (proc->pax_flags & GR_PAX_DISABLE_PAGEEXEC) ++ flags &= ~MF_PAX_PAGEEXEC; ++ if (proc->pax_flags & GR_PAX_DISABLE_SEGMEXEC) ++ flags &= ~MF_PAX_SEGMEXEC; ++ if (proc->pax_flags & GR_PAX_DISABLE_RANDMMAP) ++ flags &= ~MF_PAX_RANDMMAP; ++ if (proc->pax_flags & GR_PAX_DISABLE_EMUTRAMP) ++ flags &= ~MF_PAX_EMUTRAMP; ++ if (proc->pax_flags & GR_PAX_DISABLE_MPROTECT) ++ flags &= ~MF_PAX_MPROTECT; ++ ++ if (proc->pax_flags & GR_PAX_ENABLE_PAGEEXEC) ++ flags |= MF_PAX_PAGEEXEC; ++ if (proc->pax_flags & GR_PAX_ENABLE_SEGMEXEC) ++ flags |= MF_PAX_SEGMEXEC; ++ if (proc->pax_flags & GR_PAX_ENABLE_RANDMMAP) ++ flags |= MF_PAX_RANDMMAP; ++ if (proc->pax_flags & GR_PAX_ENABLE_EMUTRAMP) ++ flags |= MF_PAX_EMUTRAMP; ++ if (proc->pax_flags & GR_PAX_ENABLE_MPROTECT) ++ flags |= MF_PAX_MPROTECT; ++ ++ pax_set_flags(task, flags); ++ ++ return; ++} ++#endif ++ ++#ifdef CONFIG_SYSCTL ++/* Eric Biederman likes breaking userland ABI and every inode-based security ++ system to save 35kb of memory */ ++ ++/* we modify the passed in filename, but adjust it back before returning */ ++static struct acl_object_label *gr_lookup_by_name(char *name, unsigned int len) ++{ ++ struct name_entry *nmatch; ++ char *p, *lastp = NULL; ++ struct acl_object_label *obj = NULL, *tmp; ++ struct acl_subject_label *tmpsubj; ++ char c = '\0'; ++ ++ read_lock(&gr_inode_lock); ++ ++ p = name + len - 1; ++ do { ++ nmatch = lookup_name_entry(name); ++ if (lastp != NULL) ++ *lastp = c; ++ ++ if (nmatch == NULL) ++ goto next_component; ++ tmpsubj = current->acl; ++ do { ++ obj = lookup_acl_obj_label(nmatch->inode, nmatch->device, tmpsubj); ++ if (obj != NULL) { ++ tmp = obj->globbed; ++ while (tmp) { ++ if (!glob_match(tmp->filename, name)) { ++ obj = tmp; ++ goto found_obj; ++ } ++ tmp = tmp->next; ++ } ++ goto found_obj; ++ } ++ } while ((tmpsubj = tmpsubj->parent_subject)); ++next_component: ++ /* end case */ ++ if (p == name) ++ break; ++ ++ while (*p != '/') ++ p--; ++ if (p == name) ++ lastp = p + 1; ++ else { ++ lastp = p; ++ p--; ++ } ++ c = *lastp; ++ *lastp = '\0'; ++ } while (1); ++found_obj: ++ read_unlock(&gr_inode_lock); ++ /* obj returned will always be non-null */ ++ return obj; ++} ++ ++/* returns 0 when allowing, non-zero on error ++ op of 0 is used for readdir, so we don't log the names of hidden files ++*/ ++__u32 ++gr_handle_sysctl(const struct ctl_table *table, const int op) ++{ ++ struct ctl_table *tmp; ++ const char *proc_sys = "/proc/sys"; ++ char *path; ++ struct acl_object_label *obj; ++ unsigned short len = 0, pos = 0, depth = 0, i; ++ __u32 err = 0; ++ __u32 mode = 0; ++ ++ if (unlikely(!(gr_status & GR_READY))) ++ return 0; ++ ++ /* for now, ignore operations on non-sysctl entries if it's not a ++ readdir*/ ++ if (table->child != NULL && op != 0) ++ return 0; ++ ++ mode |= GR_FIND; ++ /* it's only a read if it's an entry, read on dirs is for readdir */ ++ if (op & MAY_READ) ++ mode |= GR_READ; ++ if (op & MAY_WRITE) ++ mode |= GR_WRITE; ++ ++ preempt_disable(); ++ ++ path = per_cpu_ptr(gr_shared_page[0], smp_processor_id()); ++ ++ /* it's only a read/write if it's an actual entry, not a dir ++ (which are opened for readdir) ++ */ ++ ++ /* convert the requested sysctl entry into a pathname */ ++ ++ for (tmp = (struct ctl_table *)table; tmp != NULL; tmp = tmp->parent) { ++ len += strlen(tmp->procname); ++ len++; ++ depth++; ++ } ++ ++ if ((len + depth + strlen(proc_sys) + 1) > PAGE_SIZE) { ++ /* deny */ ++ goto out; ++ } ++ ++ memset(path, 0, PAGE_SIZE); ++ ++ memcpy(path, proc_sys, strlen(proc_sys)); ++ ++ pos += strlen(proc_sys); ++ ++ for (; depth > 0; depth--) { ++ path[pos] = '/'; ++ pos++; ++ for (i = 1, tmp = (struct ctl_table *)table; tmp != NULL; tmp = tmp->parent) { ++ if (depth == i) { ++ memcpy(path + pos, tmp->procname, ++ strlen(tmp->procname)); ++ pos += strlen(tmp->procname); ++ } ++ i++; ++ } ++ } ++ ++ obj = gr_lookup_by_name(path, pos); ++ err = obj->mode & (mode | to_gr_audit(mode) | GR_SUPPRESS); ++ ++ if (unlikely((current->acl->mode & (GR_LEARN | GR_INHERITLEARN)) && ++ ((err & mode) != mode))) { ++ __u32 new_mode = mode; ++ ++ new_mode &= ~(GR_AUDITS | GR_SUPPRESS); ++ ++ err = 0; ++ gr_log_learn_sysctl(path, new_mode); ++ } else if (!(err & GR_FIND) && !(err & GR_SUPPRESS) && op != 0) { ++ gr_log_hidden_sysctl(GR_DONT_AUDIT, GR_HIDDEN_ACL_MSG, path); ++ err = -ENOENT; ++ } else if (!(err & GR_FIND)) { ++ err = -ENOENT; ++ } else if (((err & mode) & ~GR_FIND) != (mode & ~GR_FIND) && !(err & GR_SUPPRESS)) { ++ gr_log_str4(GR_DONT_AUDIT, GR_SYSCTL_ACL_MSG, "denied", ++ path, (mode & GR_READ) ? " reading" : "", ++ (mode & GR_WRITE) ? " writing" : ""); ++ err = -EACCES; ++ } else if ((err & mode) != mode) { ++ err = -EACCES; ++ } else if ((((err & mode) & ~GR_FIND) == (mode & ~GR_FIND)) && (err & GR_AUDITS)) { ++ gr_log_str4(GR_DO_AUDIT, GR_SYSCTL_ACL_MSG, "successful", ++ path, (mode & GR_READ) ? " reading" : "", ++ (mode & GR_WRITE) ? " writing" : ""); ++ err = 0; ++ } else ++ err = 0; ++ ++ out: ++ preempt_enable(); ++ ++ return err; ++} ++#endif ++ ++int ++gr_handle_proc_ptrace(struct task_struct *task) ++{ ++ struct file *filp; ++ struct task_struct *tmp = task; ++ struct task_struct *curtemp = current; ++ __u32 retmode; ++ ++#ifndef CONFIG_GRKERNSEC_HARDEN_PTRACE ++ if (unlikely(!(gr_status & GR_READY))) ++ return 0; ++#endif ++ ++ read_lock(&tasklist_lock); ++ read_lock(&grsec_exec_file_lock); ++ filp = task->exec_file; ++ ++ while (tmp->pid > 0) { ++ if (tmp == curtemp) ++ break; ++ tmp = tmp->real_parent; ++ } ++ ++ if (!filp || (tmp->pid == 0 && ((grsec_enable_harden_ptrace && current_uid() && !(gr_status & GR_READY)) || ++ ((gr_status & GR_READY) && !(current->acl->mode & GR_RELAXPTRACE))))) { ++ read_unlock(&grsec_exec_file_lock); ++ read_unlock(&tasklist_lock); ++ return 1; ++ } ++ ++#ifdef CONFIG_GRKERNSEC_HARDEN_PTRACE ++ if (!(gr_status & GR_READY)) { ++ read_unlock(&grsec_exec_file_lock); ++ read_unlock(&tasklist_lock); ++ return 0; ++ } ++#endif ++ ++ retmode = gr_search_file(filp->f_path.dentry, GR_NOPTRACE, filp->f_path.mnt); ++ read_unlock(&grsec_exec_file_lock); ++ read_unlock(&tasklist_lock); ++ ++ if (retmode & GR_NOPTRACE) ++ return 1; ++ ++ if (!(current->acl->mode & GR_POVERRIDE) && !(current->role->roletype & GR_ROLE_GOD) ++ && (current->acl != task->acl || (current->acl != current->role->root_label ++ && current->pid != task->pid))) ++ return 1; ++ ++ return 0; ++} ++ ++void task_grsec_rbac(struct seq_file *m, struct task_struct *p) ++{ ++ if (unlikely(!(gr_status & GR_READY))) ++ return; ++ ++ if (!(current->role->roletype & GR_ROLE_GOD)) ++ return; ++ ++ seq_printf(m, "RBAC:\t%.64s:%c:%.950s\n", ++ p->role->rolename, gr_task_roletype_to_char(p), ++ p->acl->filename); ++} ++ ++int ++gr_handle_ptrace(struct task_struct *task, const long request) ++{ ++ struct task_struct *tmp = task; ++ struct task_struct *curtemp = current; ++ __u32 retmode; ++ ++#ifndef CONFIG_GRKERNSEC_HARDEN_PTRACE ++ if (unlikely(!(gr_status & GR_READY))) ++ return 0; ++#endif ++ ++ read_lock(&tasklist_lock); ++ while (tmp->pid > 0) { ++ if (tmp == curtemp) ++ break; ++ tmp = tmp->real_parent; ++ } ++ ++ if (tmp->pid == 0 && ((grsec_enable_harden_ptrace && current_uid() && !(gr_status & GR_READY)) || ++ ((gr_status & GR_READY) && !(current->acl->mode & GR_RELAXPTRACE)))) { ++ read_unlock(&tasklist_lock); ++ gr_log_ptrace(GR_DONT_AUDIT, GR_PTRACE_ACL_MSG, task); ++ return 1; ++ } ++ read_unlock(&tasklist_lock); ++ ++#ifdef CONFIG_GRKERNSEC_HARDEN_PTRACE ++ if (!(gr_status & GR_READY)) ++ return 0; ++#endif ++ ++ read_lock(&grsec_exec_file_lock); ++ if (unlikely(!task->exec_file)) { ++ read_unlock(&grsec_exec_file_lock); ++ return 0; ++ } ++ ++ retmode = gr_search_file(task->exec_file->f_path.dentry, GR_PTRACERD | GR_NOPTRACE, task->exec_file->f_path.mnt); ++ read_unlock(&grsec_exec_file_lock); ++ ++ if (retmode & GR_NOPTRACE) { ++ gr_log_ptrace(GR_DONT_AUDIT, GR_PTRACE_ACL_MSG, task); ++ return 1; ++ } ++ ++ if (retmode & GR_PTRACERD) { ++ switch (request) { ++ case PTRACE_POKETEXT: ++ case PTRACE_POKEDATA: ++ case PTRACE_POKEUSR: ++#if !defined(CONFIG_PPC32) && !defined(CONFIG_PPC64) && !defined(CONFIG_PARISC) && !defined(CONFIG_ALPHA) && !defined(CONFIG_IA64) ++ case PTRACE_SETREGS: ++ case PTRACE_SETFPREGS: ++#endif ++#ifdef CONFIG_X86 ++ case PTRACE_SETFPXREGS: ++#endif ++#ifdef CONFIG_ALTIVEC ++ case PTRACE_SETVRREGS: ++#endif ++ return 1; ++ default: ++ return 0; ++ } ++ } else if (!(current->acl->mode & GR_POVERRIDE) && ++ !(current->role->roletype & GR_ROLE_GOD) && ++ (current->acl != task->acl)) { ++ gr_log_ptrace(GR_DONT_AUDIT, GR_PTRACE_ACL_MSG, task); ++ return 1; ++ } ++ ++ return 0; ++} ++ ++static int is_writable_mmap(const struct file *filp) ++{ ++ struct task_struct *task = current; ++ struct acl_object_label *obj, *obj2; ++ ++ if (gr_status & GR_READY && !(task->acl->mode & GR_OVERRIDE) && ++ !task->is_writable && S_ISREG(filp->f_path.dentry->d_inode->i_mode) && (filp->f_path.mnt != shm_mnt || (filp->f_path.dentry->d_inode->i_nlink > 0))) { ++ obj = chk_obj_label(filp->f_path.dentry, filp->f_path.mnt, default_role->root_label); ++ obj2 = chk_obj_label(filp->f_path.dentry, filp->f_path.mnt, ++ task->role->root_label); ++ if (unlikely((obj->mode & GR_WRITE) || (obj2->mode & GR_WRITE))) { ++ gr_log_fs_generic(GR_DONT_AUDIT, GR_WRITLIB_ACL_MSG, filp->f_path.dentry, filp->f_path.mnt); ++ return 1; ++ } ++ } ++ return 0; ++} ++ ++int ++gr_acl_handle_mmap(const struct file *file, const unsigned long prot) ++{ ++ __u32 mode; ++ ++ if (unlikely(!file || !(prot & PROT_EXEC))) ++ return 1; ++ ++ if (is_writable_mmap(file)) ++ return 0; ++ ++ mode = ++ gr_search_file(file->f_path.dentry, ++ GR_EXEC | GR_AUDIT_EXEC | GR_SUPPRESS, ++ file->f_path.mnt); ++ ++ if (!gr_tpe_allow(file)) ++ return 0; ++ ++ if (unlikely(!(mode & GR_EXEC) && !(mode & GR_SUPPRESS))) { ++ gr_log_fs_rbac_generic(GR_DONT_AUDIT, GR_MMAP_ACL_MSG, file->f_path.dentry, file->f_path.mnt); ++ return 0; ++ } else if (unlikely(!(mode & GR_EXEC))) { ++ return 0; ++ } else if (unlikely(mode & GR_EXEC && mode & GR_AUDIT_EXEC)) { ++ gr_log_fs_rbac_generic(GR_DO_AUDIT, GR_MMAP_ACL_MSG, file->f_path.dentry, file->f_path.mnt); ++ return 1; ++ } ++ ++ return 1; ++} ++ ++int ++gr_acl_handle_mprotect(const struct file *file, const unsigned long prot) ++{ ++ __u32 mode; ++ ++ if (unlikely(!file || !(prot & PROT_EXEC))) ++ return 1; ++ ++ if (is_writable_mmap(file)) ++ return 0; ++ ++ mode = ++ gr_search_file(file->f_path.dentry, ++ GR_EXEC | GR_AUDIT_EXEC | GR_SUPPRESS, ++ file->f_path.mnt); ++ ++ if (!gr_tpe_allow(file)) ++ return 0; ++ ++ if (unlikely(!(mode & GR_EXEC) && !(mode & GR_SUPPRESS))) { ++ gr_log_fs_rbac_generic(GR_DONT_AUDIT, GR_MPROTECT_ACL_MSG, file->f_path.dentry, file->f_path.mnt); ++ return 0; ++ } else if (unlikely(!(mode & GR_EXEC))) { ++ return 0; ++ } else if (unlikely(mode & GR_EXEC && mode & GR_AUDIT_EXEC)) { ++ gr_log_fs_rbac_generic(GR_DO_AUDIT, GR_MPROTECT_ACL_MSG, file->f_path.dentry, file->f_path.mnt); ++ return 1; ++ } ++ ++ return 1; ++} ++ ++void ++gr_acl_handle_psacct(struct task_struct *task, const long code) ++{ ++ unsigned long runtime; ++ unsigned long cputime; ++ unsigned int wday, cday; ++ __u8 whr, chr; ++ __u8 wmin, cmin; ++ __u8 wsec, csec; ++ struct timespec timeval; ++ ++ if (unlikely(!(gr_status & GR_READY) || !task->acl || ++ !(task->acl->mode & GR_PROCACCT))) ++ return; ++ ++ do_posix_clock_monotonic_gettime(&timeval); ++ runtime = timeval.tv_sec - task->start_time.tv_sec; ++ wday = runtime / (3600 * 24); ++ runtime -= wday * (3600 * 24); ++ whr = runtime / 3600; ++ runtime -= whr * 3600; ++ wmin = runtime / 60; ++ runtime -= wmin * 60; ++ wsec = runtime; ++ ++ cputime = (task->utime + task->stime) / HZ; ++ cday = cputime / (3600 * 24); ++ cputime -= cday * (3600 * 24); ++ chr = cputime / 3600; ++ cputime -= chr * 3600; ++ cmin = cputime / 60; ++ cputime -= cmin * 60; ++ csec = cputime; ++ ++ gr_log_procacct(GR_DO_AUDIT, GR_ACL_PROCACCT_MSG, task, wday, whr, wmin, wsec, cday, chr, cmin, csec, code); ++ ++ return; ++} ++ ++void gr_set_kernel_label(struct task_struct *task) ++{ ++ if (gr_status & GR_READY) { ++ task->role = kernel_role; ++ task->acl = kernel_role->root_label; ++ } ++ return; ++} ++ ++#ifdef CONFIG_TASKSTATS ++int gr_is_taskstats_denied(int pid) ++{ ++ struct task_struct *task; ++#if defined(CONFIG_GRKERNSEC_PROC_USER) || defined(CONFIG_GRKERNSEC_PROC_USERGROUP) ++ const struct cred *cred; ++#endif ++ int ret = 0; ++ ++ /* restrict taskstats viewing to un-chrooted root users ++ who have the 'view' subject flag if the RBAC system is enabled ++ */ ++ ++ rcu_read_lock(); ++ read_lock(&tasklist_lock); ++ task = find_task_by_vpid(pid); ++ if (task) { ++#ifdef CONFIG_GRKERNSEC_CHROOT ++ if (proc_is_chrooted(task)) ++ ret = -EACCES; ++#endif ++#if defined(CONFIG_GRKERNSEC_PROC_USER) || defined(CONFIG_GRKERNSEC_PROC_USERGROUP) ++ cred = __task_cred(task); ++#ifdef CONFIG_GRKERNSEC_PROC_USER ++ if (cred->uid != 0) ++ ret = -EACCES; ++#elif defined(CONFIG_GRKERNSEC_PROC_USERGROUP) ++ if (cred->uid != 0 && !groups_search(cred->group_info, CONFIG_GRKERNSEC_PROC_GID)) ++ ret = -EACCES; ++#endif ++#endif ++ if (gr_status & GR_READY) { ++ if (!(task->acl->mode & GR_VIEW)) ++ ret = -EACCES; ++ } ++ } else ++ ret = -ENOENT; ++ ++ read_unlock(&tasklist_lock); ++ rcu_read_unlock(); ++ ++ return ret; ++} ++#endif ++ ++/* AUXV entries are filled via a descendant of search_binary_handler ++ after we've already applied the subject for the target ++*/ ++int gr_acl_enable_at_secure(void) ++{ ++ if (unlikely(!(gr_status & GR_READY))) ++ return 0; ++ ++ if (current->acl->mode & GR_ATSECURE) ++ return 1; ++ ++ return 0; ++} ++ ++int gr_acl_handle_filldir(const struct file *file, const char *name, const unsigned int namelen, const ino_t ino) ++{ ++ struct task_struct *task = current; ++ struct dentry *dentry = file->f_path.dentry; ++ struct vfsmount *mnt = file->f_path.mnt; ++ struct acl_object_label *obj, *tmp; ++ struct acl_subject_label *subj; ++ unsigned int bufsize; ++ int is_not_root; ++ char *path; ++ dev_t dev = __get_dev(dentry); ++ ++ if (unlikely(!(gr_status & GR_READY))) ++ return 1; ++ ++ if (task->acl->mode & (GR_LEARN | GR_INHERITLEARN)) ++ return 1; ++ ++ /* ignore Eric Biederman */ ++ if (IS_PRIVATE(dentry->d_inode)) ++ return 1; ++ ++ subj = task->acl; ++ do { ++ obj = lookup_acl_obj_label(ino, dev, subj); ++ if (obj != NULL) ++ return (obj->mode & GR_FIND) ? 1 : 0; ++ } while ((subj = subj->parent_subject)); ++ ++ /* this is purely an optimization since we're looking for an object ++ for the directory we're doing a readdir on ++ if it's possible for any globbed object to match the entry we're ++ filling into the directory, then the object we find here will be ++ an anchor point with attached globbed objects ++ */ ++ obj = chk_obj_label_noglob(dentry, mnt, task->acl); ++ if (obj->globbed == NULL) ++ return (obj->mode & GR_FIND) ? 1 : 0; ++ ++ is_not_root = ((obj->filename[0] == '/') && ++ (obj->filename[1] == '\0')) ? 0 : 1; ++ bufsize = PAGE_SIZE - namelen - is_not_root; ++ ++ /* check bufsize > PAGE_SIZE || bufsize == 0 */ ++ if (unlikely((bufsize - 1) > (PAGE_SIZE - 1))) ++ return 1; ++ ++ preempt_disable(); ++ path = d_real_path(dentry, mnt, per_cpu_ptr(gr_shared_page[0], smp_processor_id()), ++ bufsize); ++ ++ bufsize = strlen(path); ++ ++ /* if base is "/", don't append an additional slash */ ++ if (is_not_root) ++ *(path + bufsize) = '/'; ++ memcpy(path + bufsize + is_not_root, name, namelen); ++ *(path + bufsize + namelen + is_not_root) = '\0'; ++ ++ tmp = obj->globbed; ++ while (tmp) { ++ if (!glob_match(tmp->filename, path)) { ++ preempt_enable(); ++ return (tmp->mode & GR_FIND) ? 1 : 0; ++ } ++ tmp = tmp->next; ++ } ++ preempt_enable(); ++ return (obj->mode & GR_FIND) ? 1 : 0; ++} ++ ++#ifdef CONFIG_NETFILTER_XT_MATCH_GRADM_MODULE ++EXPORT_SYMBOL(gr_acl_is_enabled); ++#endif ++EXPORT_SYMBOL(gr_learn_resource); ++EXPORT_SYMBOL(gr_set_kernel_label); ++#ifdef CONFIG_SECURITY ++EXPORT_SYMBOL(gr_check_user_change); ++EXPORT_SYMBOL(gr_check_group_change); ++#endif ++ +diff -urNp linux-2.6.38.3/grsecurity/gracl_cap.c linux-2.6.38.3/grsecurity/gracl_cap.c +--- linux-2.6.38.3/grsecurity/gracl_cap.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/gracl_cap.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,139 @@ ++#include <linux/kernel.h> ++#include <linux/module.h> ++#include <linux/sched.h> ++#include <linux/gracl.h> ++#include <linux/grsecurity.h> ++#include <linux/grinternal.h> ++ ++static const char *captab_log[] = { ++ "CAP_CHOWN", ++ "CAP_DAC_OVERRIDE", ++ "CAP_DAC_READ_SEARCH", ++ "CAP_FOWNER", ++ "CAP_FSETID", ++ "CAP_KILL", ++ "CAP_SETGID", ++ "CAP_SETUID", ++ "CAP_SETPCAP", ++ "CAP_LINUX_IMMUTABLE", ++ "CAP_NET_BIND_SERVICE", ++ "CAP_NET_BROADCAST", ++ "CAP_NET_ADMIN", ++ "CAP_NET_RAW", ++ "CAP_IPC_LOCK", ++ "CAP_IPC_OWNER", ++ "CAP_SYS_MODULE", ++ "CAP_SYS_RAWIO", ++ "CAP_SYS_CHROOT", ++ "CAP_SYS_PTRACE", ++ "CAP_SYS_PACCT", ++ "CAP_SYS_ADMIN", ++ "CAP_SYS_BOOT", ++ "CAP_SYS_NICE", ++ "CAP_SYS_RESOURCE", ++ "CAP_SYS_TIME", ++ "CAP_SYS_TTY_CONFIG", ++ "CAP_MKNOD", ++ "CAP_LEASE", ++ "CAP_AUDIT_WRITE", ++ "CAP_AUDIT_CONTROL", ++ "CAP_SETFCAP", ++ "CAP_MAC_OVERRIDE", ++ "CAP_MAC_ADMIN", ++ "CAP_SYSLOG" ++}; ++ ++EXPORT_SYMBOL(gr_is_capable); ++EXPORT_SYMBOL(gr_is_capable_nolog); ++ ++int ++gr_is_capable(const int cap) ++{ ++ struct task_struct *task = current; ++ const struct cred *cred = current_cred(); ++ struct acl_subject_label *curracl; ++ kernel_cap_t cap_drop = __cap_empty_set, cap_mask = __cap_empty_set; ++ kernel_cap_t cap_audit = __cap_empty_set; ++ ++ if (!gr_acl_is_enabled()) ++ return 1; ++ ++ curracl = task->acl; ++ ++ cap_drop = curracl->cap_lower; ++ cap_mask = curracl->cap_mask; ++ cap_audit = curracl->cap_invert_audit; ++ ++ while ((curracl = curracl->parent_subject)) { ++ /* if the cap isn't specified in the current computed mask but is specified in the ++ current level subject, and is lowered in the current level subject, then add ++ it to the set of dropped capabilities ++ otherwise, add the current level subject's mask to the current computed mask ++ */ ++ if (!cap_raised(cap_mask, cap) && cap_raised(curracl->cap_mask, cap)) { ++ cap_raise(cap_mask, cap); ++ if (cap_raised(curracl->cap_lower, cap)) ++ cap_raise(cap_drop, cap); ++ if (cap_raised(curracl->cap_invert_audit, cap)) ++ cap_raise(cap_audit, cap); ++ } ++ } ++ ++ if (!cap_raised(cap_drop, cap)) { ++ if (cap_raised(cap_audit, cap)) ++ gr_log_cap(GR_DO_AUDIT, GR_CAP_ACL_MSG2, task, captab_log[cap]); ++ return 1; ++ } ++ ++ curracl = task->acl; ++ ++ if ((curracl->mode & (GR_LEARN | GR_INHERITLEARN)) ++ && cap_raised(cred->cap_effective, cap)) { ++ security_learn(GR_LEARN_AUDIT_MSG, task->role->rolename, ++ task->role->roletype, cred->uid, ++ cred->gid, task->exec_file ? ++ gr_to_filename(task->exec_file->f_path.dentry, ++ task->exec_file->f_path.mnt) : curracl->filename, ++ curracl->filename, 0UL, ++ 0UL, "", (unsigned long) cap, &task->signal->saved_ip); ++ return 1; ++ } ++ ++ if ((cap >= 0) && (cap < (sizeof(captab_log)/sizeof(captab_log[0]))) && cap_raised(cred->cap_effective, cap) && !cap_raised(cap_audit, cap)) ++ gr_log_cap(GR_DONT_AUDIT, GR_CAP_ACL_MSG, task, captab_log[cap]); ++ return 0; ++} ++ ++int ++gr_is_capable_nolog(const int cap) ++{ ++ struct acl_subject_label *curracl; ++ kernel_cap_t cap_drop = __cap_empty_set, cap_mask = __cap_empty_set; ++ ++ if (!gr_acl_is_enabled()) ++ return 1; ++ ++ curracl = current->acl; ++ ++ cap_drop = curracl->cap_lower; ++ cap_mask = curracl->cap_mask; ++ ++ while ((curracl = curracl->parent_subject)) { ++ /* if the cap isn't specified in the current computed mask but is specified in the ++ current level subject, and is lowered in the current level subject, then add ++ it to the set of dropped capabilities ++ otherwise, add the current level subject's mask to the current computed mask ++ */ ++ if (!cap_raised(cap_mask, cap) && cap_raised(curracl->cap_mask, cap)) { ++ cap_raise(cap_mask, cap); ++ if (cap_raised(curracl->cap_lower, cap)) ++ cap_raise(cap_drop, cap); ++ } ++ } ++ ++ if (!cap_raised(cap_drop, cap)) ++ return 1; ++ ++ return 0; ++} ++ +diff -urNp linux-2.6.38.3/grsecurity/gracl_fs.c linux-2.6.38.3/grsecurity/gracl_fs.c +--- linux-2.6.38.3/grsecurity/gracl_fs.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/gracl_fs.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,431 @@ ++#include <linux/kernel.h> ++#include <linux/sched.h> ++#include <linux/types.h> ++#include <linux/fs.h> ++#include <linux/file.h> ++#include <linux/stat.h> ++#include <linux/grsecurity.h> ++#include <linux/grinternal.h> ++#include <linux/gracl.h> ++ ++__u32 ++gr_acl_handle_hidden_file(const struct dentry * dentry, ++ const struct vfsmount * mnt) ++{ ++ __u32 mode; ++ ++ if (unlikely(!dentry->d_inode)) ++ return GR_FIND; ++ ++ mode = ++ gr_search_file(dentry, GR_FIND | GR_AUDIT_FIND | GR_SUPPRESS, mnt); ++ ++ if (unlikely(mode & GR_FIND && mode & GR_AUDIT_FIND)) { ++ gr_log_fs_rbac_generic(GR_DO_AUDIT, GR_HIDDEN_ACL_MSG, dentry, mnt); ++ return mode; ++ } else if (unlikely(!(mode & GR_FIND) && !(mode & GR_SUPPRESS))) { ++ gr_log_fs_rbac_generic(GR_DONT_AUDIT, GR_HIDDEN_ACL_MSG, dentry, mnt); ++ return 0; ++ } else if (unlikely(!(mode & GR_FIND))) ++ return 0; ++ ++ return GR_FIND; ++} ++ ++__u32 ++gr_acl_handle_open(const struct dentry * dentry, const struct vfsmount * mnt, ++ const int fmode) ++{ ++ __u32 reqmode = GR_FIND; ++ __u32 mode; ++ ++ if (unlikely(!dentry->d_inode)) ++ return reqmode; ++ ++ if (unlikely(fmode & O_APPEND)) ++ reqmode |= GR_APPEND; ++ else if (unlikely(fmode & FMODE_WRITE)) ++ reqmode |= GR_WRITE; ++ if (likely((fmode & FMODE_READ) && !(fmode & O_DIRECTORY))) ++ reqmode |= GR_READ; ++ if ((fmode & FMODE_GREXEC) && (fmode & __FMODE_EXEC)) ++ reqmode &= ~GR_READ; ++ mode = ++ gr_search_file(dentry, reqmode | to_gr_audit(reqmode) | GR_SUPPRESS, ++ mnt); ++ ++ if (unlikely(((mode & reqmode) == reqmode) && mode & GR_AUDITS)) { ++ gr_log_fs_rbac_mode2(GR_DO_AUDIT, GR_OPEN_ACL_MSG, dentry, mnt, ++ reqmode & GR_READ ? " reading" : "", ++ reqmode & GR_WRITE ? " writing" : reqmode & ++ GR_APPEND ? " appending" : ""); ++ return reqmode; ++ } else ++ if (unlikely((mode & reqmode) != reqmode && !(mode & GR_SUPPRESS))) ++ { ++ gr_log_fs_rbac_mode2(GR_DONT_AUDIT, GR_OPEN_ACL_MSG, dentry, mnt, ++ reqmode & GR_READ ? " reading" : "", ++ reqmode & GR_WRITE ? " writing" : reqmode & ++ GR_APPEND ? " appending" : ""); ++ return 0; ++ } else if (unlikely((mode & reqmode) != reqmode)) ++ return 0; ++ ++ return reqmode; ++} ++ ++__u32 ++gr_acl_handle_creat(const struct dentry * dentry, ++ const struct dentry * p_dentry, ++ const struct vfsmount * p_mnt, const int fmode, ++ const int imode) ++{ ++ __u32 reqmode = GR_WRITE | GR_CREATE; ++ __u32 mode; ++ ++ if (unlikely(fmode & O_APPEND)) ++ reqmode |= GR_APPEND; ++ if (unlikely((fmode & FMODE_READ) && !(fmode & O_DIRECTORY))) ++ reqmode |= GR_READ; ++ if (unlikely((fmode & O_CREAT) && (imode & (S_ISUID | S_ISGID)))) ++ reqmode |= GR_SETID; ++ ++ mode = ++ gr_check_create(dentry, p_dentry, p_mnt, ++ reqmode | to_gr_audit(reqmode) | GR_SUPPRESS); ++ ++ if (unlikely(((mode & reqmode) == reqmode) && mode & GR_AUDITS)) { ++ gr_log_fs_rbac_mode2(GR_DO_AUDIT, GR_CREATE_ACL_MSG, dentry, p_mnt, ++ reqmode & GR_READ ? " reading" : "", ++ reqmode & GR_WRITE ? " writing" : reqmode & ++ GR_APPEND ? " appending" : ""); ++ return reqmode; ++ } else ++ if (unlikely((mode & reqmode) != reqmode && !(mode & GR_SUPPRESS))) ++ { ++ gr_log_fs_rbac_mode2(GR_DONT_AUDIT, GR_CREATE_ACL_MSG, dentry, p_mnt, ++ reqmode & GR_READ ? " reading" : "", ++ reqmode & GR_WRITE ? " writing" : reqmode & ++ GR_APPEND ? " appending" : ""); ++ return 0; ++ } else if (unlikely((mode & reqmode) != reqmode)) ++ return 0; ++ ++ return reqmode; ++} ++ ++__u32 ++gr_acl_handle_access(const struct dentry * dentry, const struct vfsmount * mnt, ++ const int fmode) ++{ ++ __u32 mode, reqmode = GR_FIND; ++ ++ if ((fmode & S_IXOTH) && !S_ISDIR(dentry->d_inode->i_mode)) ++ reqmode |= GR_EXEC; ++ if (fmode & S_IWOTH) ++ reqmode |= GR_WRITE; ++ if (fmode & S_IROTH) ++ reqmode |= GR_READ; ++ ++ mode = ++ gr_search_file(dentry, reqmode | to_gr_audit(reqmode) | GR_SUPPRESS, ++ mnt); ++ ++ if (unlikely(((mode & reqmode) == reqmode) && mode & GR_AUDITS)) { ++ gr_log_fs_rbac_mode3(GR_DO_AUDIT, GR_ACCESS_ACL_MSG, dentry, mnt, ++ reqmode & GR_READ ? " reading" : "", ++ reqmode & GR_WRITE ? " writing" : "", ++ reqmode & GR_EXEC ? " executing" : ""); ++ return reqmode; ++ } else ++ if (unlikely((mode & reqmode) != reqmode && !(mode & GR_SUPPRESS))) ++ { ++ gr_log_fs_rbac_mode3(GR_DONT_AUDIT, GR_ACCESS_ACL_MSG, dentry, mnt, ++ reqmode & GR_READ ? " reading" : "", ++ reqmode & GR_WRITE ? " writing" : "", ++ reqmode & GR_EXEC ? " executing" : ""); ++ return 0; ++ } else if (unlikely((mode & reqmode) != reqmode)) ++ return 0; ++ ++ return reqmode; ++} ++ ++static __u32 generic_fs_handler(const struct dentry *dentry, const struct vfsmount *mnt, __u32 reqmode, const char *fmt) ++{ ++ __u32 mode; ++ ++ mode = gr_search_file(dentry, reqmode | to_gr_audit(reqmode) | GR_SUPPRESS, mnt); ++ ++ if (unlikely(((mode & (reqmode)) == (reqmode)) && mode & GR_AUDITS)) { ++ gr_log_fs_rbac_generic(GR_DO_AUDIT, fmt, dentry, mnt); ++ return mode; ++ } else if (unlikely((mode & (reqmode)) != (reqmode) && !(mode & GR_SUPPRESS))) { ++ gr_log_fs_rbac_generic(GR_DONT_AUDIT, fmt, dentry, mnt); ++ return 0; ++ } else if (unlikely((mode & (reqmode)) != (reqmode))) ++ return 0; ++ ++ return (reqmode); ++} ++ ++__u32 ++gr_acl_handle_rmdir(const struct dentry * dentry, const struct vfsmount * mnt) ++{ ++ return generic_fs_handler(dentry, mnt, GR_WRITE | GR_DELETE , GR_RMDIR_ACL_MSG); ++} ++ ++__u32 ++gr_acl_handle_unlink(const struct dentry *dentry, const struct vfsmount *mnt) ++{ ++ return generic_fs_handler(dentry, mnt, GR_WRITE | GR_DELETE , GR_UNLINK_ACL_MSG); ++} ++ ++__u32 ++gr_acl_handle_truncate(const struct dentry *dentry, const struct vfsmount *mnt) ++{ ++ return generic_fs_handler(dentry, mnt, GR_WRITE, GR_TRUNCATE_ACL_MSG); ++} ++ ++__u32 ++gr_acl_handle_utime(const struct dentry *dentry, const struct vfsmount *mnt) ++{ ++ return generic_fs_handler(dentry, mnt, GR_WRITE, GR_ATIME_ACL_MSG); ++} ++ ++__u32 ++gr_acl_handle_fchmod(const struct dentry *dentry, const struct vfsmount *mnt, ++ mode_t mode) ++{ ++ if (unlikely(dentry->d_inode && S_ISSOCK(dentry->d_inode->i_mode))) ++ return 1; ++ ++ if (unlikely((mode != (mode_t)-1) && (mode & (S_ISUID | S_ISGID)))) { ++ return generic_fs_handler(dentry, mnt, GR_WRITE | GR_SETID, ++ GR_FCHMOD_ACL_MSG); ++ } else { ++ return generic_fs_handler(dentry, mnt, GR_WRITE, GR_FCHMOD_ACL_MSG); ++ } ++} ++ ++__u32 ++gr_acl_handle_chmod(const struct dentry *dentry, const struct vfsmount *mnt, ++ mode_t mode) ++{ ++ if (unlikely((mode != (mode_t)-1) && (mode & (S_ISUID | S_ISGID)))) { ++ return generic_fs_handler(dentry, mnt, GR_WRITE | GR_SETID, ++ GR_CHMOD_ACL_MSG); ++ } else { ++ return generic_fs_handler(dentry, mnt, GR_WRITE, GR_CHMOD_ACL_MSG); ++ } ++} ++ ++__u32 ++gr_acl_handle_chown(const struct dentry *dentry, const struct vfsmount *mnt) ++{ ++ return generic_fs_handler(dentry, mnt, GR_WRITE, GR_CHOWN_ACL_MSG); ++} ++ ++__u32 ++gr_acl_handle_setxattr(const struct dentry *dentry, const struct vfsmount *mnt) ++{ ++ return generic_fs_handler(dentry, mnt, GR_WRITE, GR_SETXATTR_ACL_MSG); ++} ++ ++__u32 ++gr_acl_handle_execve(const struct dentry *dentry, const struct vfsmount *mnt) ++{ ++ return generic_fs_handler(dentry, mnt, GR_EXEC, GR_EXEC_ACL_MSG); ++} ++ ++__u32 ++gr_acl_handle_unix(const struct dentry *dentry, const struct vfsmount *mnt) ++{ ++ return generic_fs_handler(dentry, mnt, GR_READ | GR_WRITE, ++ GR_UNIXCONNECT_ACL_MSG); ++} ++ ++/* hardlinks require at minimum create permission, ++ any additional privilege required is based on the ++ privilege of the file being linked to ++*/ ++__u32 ++gr_acl_handle_link(const struct dentry * new_dentry, ++ const struct dentry * parent_dentry, ++ const struct vfsmount * parent_mnt, ++ const struct dentry * old_dentry, ++ const struct vfsmount * old_mnt, const char *to) ++{ ++ __u32 mode; ++ __u32 needmode = GR_CREATE | GR_LINK; ++ __u32 needaudit = GR_AUDIT_CREATE | GR_AUDIT_LINK; ++ ++ mode = ++ gr_check_link(new_dentry, parent_dentry, parent_mnt, old_dentry, ++ old_mnt); ++ ++ if (unlikely(((mode & needmode) == needmode) && (mode & needaudit))) { ++ gr_log_fs_rbac_str(GR_DO_AUDIT, GR_LINK_ACL_MSG, old_dentry, old_mnt, to); ++ return mode; ++ } else if (unlikely(((mode & needmode) != needmode) && !(mode & GR_SUPPRESS))) { ++ gr_log_fs_rbac_str(GR_DONT_AUDIT, GR_LINK_ACL_MSG, old_dentry, old_mnt, to); ++ return 0; ++ } else if (unlikely((mode & needmode) != needmode)) ++ return 0; ++ ++ return 1; ++} ++ ++__u32 ++gr_acl_handle_symlink(const struct dentry * new_dentry, ++ const struct dentry * parent_dentry, ++ const struct vfsmount * parent_mnt, const char *from) ++{ ++ __u32 needmode = GR_WRITE | GR_CREATE; ++ __u32 mode; ++ ++ mode = ++ gr_check_create(new_dentry, parent_dentry, parent_mnt, ++ GR_CREATE | GR_AUDIT_CREATE | ++ GR_WRITE | GR_AUDIT_WRITE | GR_SUPPRESS); ++ ++ if (unlikely(mode & GR_WRITE && mode & GR_AUDITS)) { ++ gr_log_fs_str_rbac(GR_DO_AUDIT, GR_SYMLINK_ACL_MSG, from, new_dentry, parent_mnt); ++ return mode; ++ } else if (unlikely(((mode & needmode) != needmode) && !(mode & GR_SUPPRESS))) { ++ gr_log_fs_str_rbac(GR_DONT_AUDIT, GR_SYMLINK_ACL_MSG, from, new_dentry, parent_mnt); ++ return 0; ++ } else if (unlikely((mode & needmode) != needmode)) ++ return 0; ++ ++ return (GR_WRITE | GR_CREATE); ++} ++ ++static __u32 generic_fs_create_handler(const struct dentry *new_dentry, const struct dentry *parent_dentry, const struct vfsmount *parent_mnt, __u32 reqmode, const char *fmt) ++{ ++ __u32 mode; ++ ++ mode = gr_check_create(new_dentry, parent_dentry, parent_mnt, reqmode | to_gr_audit(reqmode) | GR_SUPPRESS); ++ ++ if (unlikely(((mode & (reqmode)) == (reqmode)) && mode & GR_AUDITS)) { ++ gr_log_fs_rbac_generic(GR_DO_AUDIT, fmt, new_dentry, parent_mnt); ++ return mode; ++ } else if (unlikely((mode & (reqmode)) != (reqmode) && !(mode & GR_SUPPRESS))) { ++ gr_log_fs_rbac_generic(GR_DONT_AUDIT, fmt, new_dentry, parent_mnt); ++ return 0; ++ } else if (unlikely((mode & (reqmode)) != (reqmode))) ++ return 0; ++ ++ return (reqmode); ++} ++ ++__u32 ++gr_acl_handle_mknod(const struct dentry * new_dentry, ++ const struct dentry * parent_dentry, ++ const struct vfsmount * parent_mnt, ++ const int mode) ++{ ++ __u32 reqmode = GR_WRITE | GR_CREATE; ++ if (unlikely(mode & (S_ISUID | S_ISGID))) ++ reqmode |= GR_SETID; ++ ++ return generic_fs_create_handler(new_dentry, parent_dentry, parent_mnt, ++ reqmode, GR_MKNOD_ACL_MSG); ++} ++ ++__u32 ++gr_acl_handle_mkdir(const struct dentry *new_dentry, ++ const struct dentry *parent_dentry, ++ const struct vfsmount *parent_mnt) ++{ ++ return generic_fs_create_handler(new_dentry, parent_dentry, parent_mnt, ++ GR_WRITE | GR_CREATE, GR_MKDIR_ACL_MSG); ++} ++ ++#define RENAME_CHECK_SUCCESS(old, new) \ ++ (((old & (GR_WRITE | GR_READ)) == (GR_WRITE | GR_READ)) && \ ++ ((new & (GR_WRITE | GR_READ)) == (GR_WRITE | GR_READ))) ++ ++int ++gr_acl_handle_rename(struct dentry *new_dentry, ++ struct dentry *parent_dentry, ++ const struct vfsmount *parent_mnt, ++ struct dentry *old_dentry, ++ struct inode *old_parent_inode, ++ struct vfsmount *old_mnt, const char *newname) ++{ ++ __u32 comp1, comp2; ++ int error = 0; ++ ++ if (unlikely(!gr_acl_is_enabled())) ++ return 0; ++ ++ if (!new_dentry->d_inode) { ++ comp1 = gr_check_create(new_dentry, parent_dentry, parent_mnt, ++ GR_READ | GR_WRITE | GR_CREATE | GR_AUDIT_READ | ++ GR_AUDIT_WRITE | GR_AUDIT_CREATE | GR_SUPPRESS); ++ comp2 = gr_search_file(old_dentry, GR_READ | GR_WRITE | ++ GR_DELETE | GR_AUDIT_DELETE | ++ GR_AUDIT_READ | GR_AUDIT_WRITE | ++ GR_SUPPRESS, old_mnt); ++ } else { ++ comp1 = gr_search_file(new_dentry, GR_READ | GR_WRITE | ++ GR_CREATE | GR_DELETE | ++ GR_AUDIT_CREATE | GR_AUDIT_DELETE | ++ GR_AUDIT_READ | GR_AUDIT_WRITE | ++ GR_SUPPRESS, parent_mnt); ++ comp2 = ++ gr_search_file(old_dentry, ++ GR_READ | GR_WRITE | GR_AUDIT_READ | ++ GR_DELETE | GR_AUDIT_DELETE | ++ GR_AUDIT_WRITE | GR_SUPPRESS, old_mnt); ++ } ++ ++ if (RENAME_CHECK_SUCCESS(comp1, comp2) && ++ ((comp1 & GR_AUDITS) || (comp2 & GR_AUDITS))) ++ gr_log_fs_rbac_str(GR_DO_AUDIT, GR_RENAME_ACL_MSG, old_dentry, old_mnt, newname); ++ else if (!RENAME_CHECK_SUCCESS(comp1, comp2) && !(comp1 & GR_SUPPRESS) ++ && !(comp2 & GR_SUPPRESS)) { ++ gr_log_fs_rbac_str(GR_DONT_AUDIT, GR_RENAME_ACL_MSG, old_dentry, old_mnt, newname); ++ error = -EACCES; ++ } else if (unlikely(!RENAME_CHECK_SUCCESS(comp1, comp2))) ++ error = -EACCES; ++ ++ return error; ++} ++ ++void ++gr_acl_handle_exit(void) ++{ ++ u16 id; ++ char *rolename; ++ struct file *exec_file; ++ ++ if (unlikely(current->acl_sp_role && gr_acl_is_enabled() && ++ !(current->role->roletype & GR_ROLE_PERSIST))) { ++ id = current->acl_role_id; ++ rolename = current->role->rolename; ++ gr_set_acls(1); ++ gr_log_str_int(GR_DONT_AUDIT_GOOD, GR_SPROLEL_ACL_MSG, rolename, id); ++ } ++ ++ write_lock(&grsec_exec_file_lock); ++ exec_file = current->exec_file; ++ current->exec_file = NULL; ++ write_unlock(&grsec_exec_file_lock); ++ ++ if (exec_file) ++ fput(exec_file); ++} ++ ++int ++gr_acl_handle_procpidmem(const struct task_struct *task) ++{ ++ if (unlikely(!gr_acl_is_enabled())) ++ return 0; ++ ++ if (task != current && task->acl->mode & GR_PROTPROCFD) ++ return -EACCES; ++ ++ return 0; ++} +diff -urNp linux-2.6.38.3/grsecurity/gracl_ip.c linux-2.6.38.3/grsecurity/gracl_ip.c +--- linux-2.6.38.3/grsecurity/gracl_ip.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/gracl_ip.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,382 @@ ++#include <linux/kernel.h> ++#include <asm/uaccess.h> ++#include <asm/errno.h> ++#include <net/sock.h> ++#include <linux/file.h> ++#include <linux/fs.h> ++#include <linux/net.h> ++#include <linux/in.h> ++#include <linux/skbuff.h> ++#include <linux/ip.h> ++#include <linux/udp.h> ++#include <linux/smp_lock.h> ++#include <linux/types.h> ++#include <linux/sched.h> ++#include <linux/netdevice.h> ++#include <linux/inetdevice.h> ++#include <linux/gracl.h> ++#include <linux/grsecurity.h> ++#include <linux/grinternal.h> ++ ++#define GR_BIND 0x01 ++#define GR_CONNECT 0x02 ++#define GR_INVERT 0x04 ++#define GR_BINDOVERRIDE 0x08 ++#define GR_CONNECTOVERRIDE 0x10 ++#define GR_SOCK_FAMILY 0x20 ++ ++static const char * gr_protocols[IPPROTO_MAX] = { ++ "ip", "icmp", "igmp", "ggp", "ipencap", "st", "tcp", "cbt", ++ "egp", "igp", "bbn-rcc", "nvp", "pup", "argus", "emcon", "xnet", ++ "chaos", "udp", "mux", "dcn", "hmp", "prm", "xns-idp", "trunk-1", ++ "trunk-2", "leaf-1", "leaf-2", "rdp", "irtp", "iso-tp4", "netblt", "mfe-nsp", ++ "merit-inp", "sep", "3pc", "idpr", "xtp", "ddp", "idpr-cmtp", "tp++", ++ "il", "ipv6", "sdrp", "ipv6-route", "ipv6-frag", "idrp", "rsvp", "gre", ++ "mhrp", "bna", "ipv6-crypt", "ipv6-auth", "i-nlsp", "swipe", "narp", "mobile", ++ "tlsp", "skip", "ipv6-icmp", "ipv6-nonxt", "ipv6-opts", "unknown:61", "cftp", "unknown:63", ++ "sat-expak", "kryptolan", "rvd", "ippc", "unknown:68", "sat-mon", "visa", "ipcv", ++ "cpnx", "cphb", "wsn", "pvp", "br-sat-mon", "sun-nd", "wb-mon", "wb-expak", ++ "iso-ip", "vmtp", "secure-vmtp", "vines", "ttp", "nfsnet-igp", "dgp", "tcf", ++ "eigrp", "ospf", "sprite-rpc", "larp", "mtp", "ax.25", "ipip", "micp", ++ "scc-sp", "etherip", "encap", "unknown:99", "gmtp", "ifmp", "pnni", "pim", ++ "aris", "scps", "qnx", "a/n", "ipcomp", "snp", "compaq-peer", "ipx-in-ip", ++ "vrrp", "pgm", "unknown:114", "l2tp", "ddx", "iatp", "stp", "srp", ++ "uti", "smp", "sm", "ptp", "isis", "fire", "crtp", "crdup", ++ "sscopmce", "iplt", "sps", "pipe", "sctp", "fc", "unkown:134", "unknown:135", ++ "unknown:136", "unknown:137", "unknown:138", "unknown:139", "unknown:140", "unknown:141", "unknown:142", "unknown:143", ++ "unknown:144", "unknown:145", "unknown:146", "unknown:147", "unknown:148", "unknown:149", "unknown:150", "unknown:151", ++ "unknown:152", "unknown:153", "unknown:154", "unknown:155", "unknown:156", "unknown:157", "unknown:158", "unknown:159", ++ "unknown:160", "unknown:161", "unknown:162", "unknown:163", "unknown:164", "unknown:165", "unknown:166", "unknown:167", ++ "unknown:168", "unknown:169", "unknown:170", "unknown:171", "unknown:172", "unknown:173", "unknown:174", "unknown:175", ++ "unknown:176", "unknown:177", "unknown:178", "unknown:179", "unknown:180", "unknown:181", "unknown:182", "unknown:183", ++ "unknown:184", "unknown:185", "unknown:186", "unknown:187", "unknown:188", "unknown:189", "unknown:190", "unknown:191", ++ "unknown:192", "unknown:193", "unknown:194", "unknown:195", "unknown:196", "unknown:197", "unknown:198", "unknown:199", ++ "unknown:200", "unknown:201", "unknown:202", "unknown:203", "unknown:204", "unknown:205", "unknown:206", "unknown:207", ++ "unknown:208", "unknown:209", "unknown:210", "unknown:211", "unknown:212", "unknown:213", "unknown:214", "unknown:215", ++ "unknown:216", "unknown:217", "unknown:218", "unknown:219", "unknown:220", "unknown:221", "unknown:222", "unknown:223", ++ "unknown:224", "unknown:225", "unknown:226", "unknown:227", "unknown:228", "unknown:229", "unknown:230", "unknown:231", ++ "unknown:232", "unknown:233", "unknown:234", "unknown:235", "unknown:236", "unknown:237", "unknown:238", "unknown:239", ++ "unknown:240", "unknown:241", "unknown:242", "unknown:243", "unknown:244", "unknown:245", "unknown:246", "unknown:247", ++ "unknown:248", "unknown:249", "unknown:250", "unknown:251", "unknown:252", "unknown:253", "unknown:254", "unknown:255", ++ }; ++ ++static const char * gr_socktypes[SOCK_MAX] = { ++ "unknown:0", "stream", "dgram", "raw", "rdm", "seqpacket", "unknown:6", ++ "unknown:7", "unknown:8", "unknown:9", "packet" ++ }; ++ ++static const char * gr_sockfamilies[AF_MAX+1] = { ++ "unspec", "unix", "inet", "ax25", "ipx", "appletalk", "netrom", "bridge", "atmpvc", "x25", ++ "inet6", "rose", "decnet", "netbeui", "security", "key", "netlink", "packet", "ash", ++ "econet", "atmsvc", "rds", "sna", "irda", "ppox", "wanpipe", "llc", "fam_27", "fam_28", ++ "tipc", "bluetooth", "iucv", "rxrpc", "isdn", "phonet", "ieee802154", "ciaf" ++ }; ++ ++const char * ++gr_proto_to_name(unsigned char proto) ++{ ++ return gr_protocols[proto]; ++} ++ ++const char * ++gr_socktype_to_name(unsigned char type) ++{ ++ return gr_socktypes[type]; ++} ++ ++const char * ++gr_sockfamily_to_name(unsigned char family) ++{ ++ return gr_sockfamilies[family]; ++} ++ ++int ++gr_search_socket(const int domain, const int type, const int protocol) ++{ ++ struct acl_subject_label *curr; ++ const struct cred *cred = current_cred(); ++ ++ if (unlikely(!gr_acl_is_enabled())) ++ goto exit; ++ ++ if ((domain < 0) || (type < 0) || (protocol < 0) || ++ (domain >= AF_MAX) || (type >= SOCK_MAX) || (protocol >= IPPROTO_MAX)) ++ goto exit; // let the kernel handle it ++ ++ curr = current->acl; ++ ++ if (curr->sock_families[domain / 32] & (1 << (domain % 32))) { ++ /* the family is allowed, if this is PF_INET allow it only if ++ the extra sock type/protocol checks pass */ ++ if (domain == PF_INET) ++ goto inet_check; ++ goto exit; ++ } else { ++ if (curr->mode & (GR_LEARN | GR_INHERITLEARN)) { ++ __u32 fakeip = 0; ++ security_learn(GR_IP_LEARN_MSG, current->role->rolename, ++ current->role->roletype, cred->uid, ++ cred->gid, current->exec_file ? ++ gr_to_filename(current->exec_file->f_path.dentry, ++ current->exec_file->f_path.mnt) : ++ curr->filename, curr->filename, ++ &fakeip, domain, 0, 0, GR_SOCK_FAMILY, ++ ¤t->signal->saved_ip); ++ goto exit; ++ } ++ goto exit_fail; ++ } ++ ++inet_check: ++ /* the rest of this checking is for IPv4 only */ ++ if (!curr->ips) ++ goto exit; ++ ++ if ((curr->ip_type & (1 << type)) && ++ (curr->ip_proto[protocol / 32] & (1 << (protocol % 32)))) ++ goto exit; ++ ++ if (curr->mode & (GR_LEARN | GR_INHERITLEARN)) { ++ /* we don't place acls on raw sockets , and sometimes ++ dgram/ip sockets are opened for ioctl and not ++ bind/connect, so we'll fake a bind learn log */ ++ if (type == SOCK_RAW || type == SOCK_PACKET) { ++ __u32 fakeip = 0; ++ security_learn(GR_IP_LEARN_MSG, current->role->rolename, ++ current->role->roletype, cred->uid, ++ cred->gid, current->exec_file ? ++ gr_to_filename(current->exec_file->f_path.dentry, ++ current->exec_file->f_path.mnt) : ++ curr->filename, curr->filename, ++ &fakeip, 0, type, ++ protocol, GR_CONNECT, ¤t->signal->saved_ip); ++ } else if ((type == SOCK_DGRAM) && (protocol == IPPROTO_IP)) { ++ __u32 fakeip = 0; ++ security_learn(GR_IP_LEARN_MSG, current->role->rolename, ++ current->role->roletype, cred->uid, ++ cred->gid, current->exec_file ? ++ gr_to_filename(current->exec_file->f_path.dentry, ++ current->exec_file->f_path.mnt) : ++ curr->filename, curr->filename, ++ &fakeip, 0, type, ++ protocol, GR_BIND, ¤t->signal->saved_ip); ++ } ++ /* we'll log when they use connect or bind */ ++ goto exit; ++ } ++ ++exit_fail: ++ if (domain == PF_INET) ++ gr_log_str3(GR_DONT_AUDIT, GR_SOCK_MSG, gr_sockfamily_to_name(domain), ++ gr_socktype_to_name(type), gr_proto_to_name(protocol)); ++ else ++ gr_log_str2_int(GR_DONT_AUDIT, GR_SOCK_NOINET_MSG, gr_sockfamily_to_name(domain), ++ gr_socktype_to_name(type), protocol); ++ ++ return 0; ++exit: ++ return 1; ++} ++ ++int check_ip_policy(struct acl_ip_label *ip, __u32 ip_addr, __u16 ip_port, __u8 protocol, const int mode, const int type, __u32 our_addr, __u32 our_netmask) ++{ ++ if ((ip->mode & mode) && ++ (ip_port >= ip->low) && ++ (ip_port <= ip->high) && ++ ((ntohl(ip_addr) & our_netmask) == ++ (ntohl(our_addr) & our_netmask)) ++ && (ip->proto[protocol / 32] & (1 << (protocol % 32))) ++ && (ip->type & (1 << type))) { ++ if (ip->mode & GR_INVERT) ++ return 2; // specifically denied ++ else ++ return 1; // allowed ++ } ++ ++ return 0; // not specifically allowed, may continue parsing ++} ++ ++static int ++gr_search_connectbind(const int full_mode, struct sock *sk, ++ struct sockaddr_in *addr, const int type) ++{ ++ char iface[IFNAMSIZ] = {0}; ++ struct acl_subject_label *curr; ++ struct acl_ip_label *ip; ++ struct inet_sock *isk; ++ struct net_device *dev; ++ struct in_device *idev; ++ unsigned long i; ++ int ret; ++ int mode = full_mode & (GR_BIND | GR_CONNECT); ++ __u32 ip_addr = 0; ++ __u32 our_addr; ++ __u32 our_netmask; ++ char *p; ++ __u16 ip_port = 0; ++ const struct cred *cred = current_cred(); ++ ++ if (unlikely(!gr_acl_is_enabled() || sk->sk_family != PF_INET)) ++ return 0; ++ ++ curr = current->acl; ++ isk = inet_sk(sk); ++ ++ /* INADDR_ANY overriding for binds, inaddr_any_override is already in network order */ ++ if ((full_mode & GR_BINDOVERRIDE) && addr->sin_addr.s_addr == htonl(INADDR_ANY) && curr->inaddr_any_override != 0) ++ addr->sin_addr.s_addr = curr->inaddr_any_override; ++ if ((full_mode & GR_CONNECT) && isk->inet_saddr == htonl(INADDR_ANY) && curr->inaddr_any_override != 0) { ++ struct sockaddr_in saddr; ++ int err; ++ ++ saddr.sin_family = AF_INET; ++ saddr.sin_addr.s_addr = curr->inaddr_any_override; ++ saddr.sin_port = isk->inet_sport; ++ ++ err = security_socket_bind(sk->sk_socket, (struct sockaddr *)&saddr, sizeof(struct sockaddr_in)); ++ if (err) ++ return err; ++ ++ err = sk->sk_socket->ops->bind(sk->sk_socket, (struct sockaddr *)&saddr, sizeof(struct sockaddr_in)); ++ if (err) ++ return err; ++ } ++ ++ if (!curr->ips) ++ return 0; ++ ++ ip_addr = addr->sin_addr.s_addr; ++ ip_port = ntohs(addr->sin_port); ++ ++ if (curr->mode & (GR_LEARN | GR_INHERITLEARN)) { ++ security_learn(GR_IP_LEARN_MSG, current->role->rolename, ++ current->role->roletype, cred->uid, ++ cred->gid, current->exec_file ? ++ gr_to_filename(current->exec_file->f_path.dentry, ++ current->exec_file->f_path.mnt) : ++ curr->filename, curr->filename, ++ &ip_addr, ip_port, type, ++ sk->sk_protocol, mode, ¤t->signal->saved_ip); ++ return 0; ++ } ++ ++ for (i = 0; i < curr->ip_num; i++) { ++ ip = *(curr->ips + i); ++ if (ip->iface != NULL) { ++ strncpy(iface, ip->iface, IFNAMSIZ - 1); ++ p = strchr(iface, ':'); ++ if (p != NULL) ++ *p = '\0'; ++ dev = dev_get_by_name(sock_net(sk), iface); ++ if (dev == NULL) ++ continue; ++ idev = in_dev_get(dev); ++ if (idev == NULL) { ++ dev_put(dev); ++ continue; ++ } ++ rcu_read_lock(); ++ for_ifa(idev) { ++ if (!strcmp(ip->iface, ifa->ifa_label)) { ++ our_addr = ifa->ifa_address; ++ our_netmask = 0xffffffff; ++ ret = check_ip_policy(ip, ip_addr, ip_port, sk->sk_protocol, mode, type, our_addr, our_netmask); ++ if (ret == 1) { ++ rcu_read_unlock(); ++ in_dev_put(idev); ++ dev_put(dev); ++ return 0; ++ } else if (ret == 2) { ++ rcu_read_unlock(); ++ in_dev_put(idev); ++ dev_put(dev); ++ goto denied; ++ } ++ } ++ } endfor_ifa(idev); ++ rcu_read_unlock(); ++ in_dev_put(idev); ++ dev_put(dev); ++ } else { ++ our_addr = ip->addr; ++ our_netmask = ip->netmask; ++ ret = check_ip_policy(ip, ip_addr, ip_port, sk->sk_protocol, mode, type, our_addr, our_netmask); ++ if (ret == 1) ++ return 0; ++ else if (ret == 2) ++ goto denied; ++ } ++ } ++ ++denied: ++ if (mode == GR_BIND) ++ gr_log_int5_str2(GR_DONT_AUDIT, GR_BIND_ACL_MSG, &ip_addr, ip_port, gr_socktype_to_name(type), gr_proto_to_name(sk->sk_protocol)); ++ else if (mode == GR_CONNECT) ++ gr_log_int5_str2(GR_DONT_AUDIT, GR_CONNECT_ACL_MSG, &ip_addr, ip_port, gr_socktype_to_name(type), gr_proto_to_name(sk->sk_protocol)); ++ ++ return -EACCES; ++} ++ ++int ++gr_search_connect(struct socket *sock, struct sockaddr_in *addr) ++{ ++ return gr_search_connectbind(GR_CONNECT | GR_CONNECTOVERRIDE, sock->sk, addr, sock->type); ++} ++ ++int ++gr_search_bind(struct socket *sock, struct sockaddr_in *addr) ++{ ++ return gr_search_connectbind(GR_BIND | GR_BINDOVERRIDE, sock->sk, addr, sock->type); ++} ++ ++int gr_search_listen(struct socket *sock) ++{ ++ struct sock *sk = sock->sk; ++ struct sockaddr_in addr; ++ ++ addr.sin_addr.s_addr = inet_sk(sk)->inet_saddr; ++ addr.sin_port = inet_sk(sk)->inet_sport; ++ ++ return gr_search_connectbind(GR_BIND | GR_CONNECTOVERRIDE, sock->sk, &addr, sock->type); ++} ++ ++int gr_search_accept(struct socket *sock) ++{ ++ struct sock *sk = sock->sk; ++ struct sockaddr_in addr; ++ ++ addr.sin_addr.s_addr = inet_sk(sk)->inet_saddr; ++ addr.sin_port = inet_sk(sk)->inet_sport; ++ ++ return gr_search_connectbind(GR_BIND | GR_CONNECTOVERRIDE, sock->sk, &addr, sock->type); ++} ++ ++int ++gr_search_udp_sendmsg(struct sock *sk, struct sockaddr_in *addr) ++{ ++ if (addr) ++ return gr_search_connectbind(GR_CONNECT, sk, addr, SOCK_DGRAM); ++ else { ++ struct sockaddr_in sin; ++ const struct inet_sock *inet = inet_sk(sk); ++ ++ sin.sin_addr.s_addr = inet->inet_daddr; ++ sin.sin_port = inet->inet_dport; ++ ++ return gr_search_connectbind(GR_CONNECT | GR_CONNECTOVERRIDE, sk, &sin, SOCK_DGRAM); ++ } ++} ++ ++int ++gr_search_udp_recvmsg(struct sock *sk, const struct sk_buff *skb) ++{ ++ struct sockaddr_in sin; ++ ++ if (unlikely(skb->len < sizeof (struct udphdr))) ++ return 0; // skip this packet ++ ++ sin.sin_addr.s_addr = ip_hdr(skb)->saddr; ++ sin.sin_port = udp_hdr(skb)->source; ++ ++ return gr_search_connectbind(GR_CONNECT | GR_CONNECTOVERRIDE, sk, &sin, SOCK_DGRAM); ++} +diff -urNp linux-2.6.38.3/grsecurity/gracl_learn.c linux-2.6.38.3/grsecurity/gracl_learn.c +--- linux-2.6.38.3/grsecurity/gracl_learn.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/gracl_learn.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,211 @@ ++#include <linux/kernel.h> ++#include <linux/mm.h> ++#include <linux/sched.h> ++#include <linux/poll.h> ++#include <linux/smp_lock.h> ++#include <linux/string.h> ++#include <linux/file.h> ++#include <linux/types.h> ++#include <linux/vmalloc.h> ++#include <linux/grinternal.h> ++ ++extern ssize_t write_grsec_handler(struct file * file, const char __user * buf, ++ size_t count, loff_t *ppos); ++extern int gr_acl_is_enabled(void); ++ ++static DECLARE_WAIT_QUEUE_HEAD(learn_wait); ++static int gr_learn_attached; ++ ++/* use a 512k buffer */ ++#define LEARN_BUFFER_SIZE (512 * 1024) ++ ++static DEFINE_SPINLOCK(gr_learn_lock); ++static DEFINE_MUTEX(gr_learn_user_mutex); ++ ++/* we need to maintain two buffers, so that the kernel context of grlearn ++ uses a semaphore around the userspace copying, and the other kernel contexts ++ use a spinlock when copying into the buffer, since they cannot sleep ++*/ ++static char *learn_buffer; ++static char *learn_buffer_user; ++static int learn_buffer_len; ++static int learn_buffer_user_len; ++ ++static ssize_t ++read_learn(struct file *file, char __user * buf, size_t count, loff_t * ppos) ++{ ++ DECLARE_WAITQUEUE(wait, current); ++ ssize_t retval = 0; ++ ++ add_wait_queue(&learn_wait, &wait); ++ set_current_state(TASK_INTERRUPTIBLE); ++ do { ++ mutex_lock(&gr_learn_user_mutex); ++ spin_lock(&gr_learn_lock); ++ if (learn_buffer_len) ++ break; ++ spin_unlock(&gr_learn_lock); ++ mutex_unlock(&gr_learn_user_mutex); ++ if (file->f_flags & O_NONBLOCK) { ++ retval = -EAGAIN; ++ goto out; ++ } ++ if (signal_pending(current)) { ++ retval = -ERESTARTSYS; ++ goto out; ++ } ++ ++ schedule(); ++ } while (1); ++ ++ memcpy(learn_buffer_user, learn_buffer, learn_buffer_len); ++ learn_buffer_user_len = learn_buffer_len; ++ retval = learn_buffer_len; ++ learn_buffer_len = 0; ++ ++ spin_unlock(&gr_learn_lock); ++ ++ if (copy_to_user(buf, learn_buffer_user, learn_buffer_user_len)) ++ retval = -EFAULT; ++ ++ mutex_unlock(&gr_learn_user_mutex); ++out: ++ set_current_state(TASK_RUNNING); ++ remove_wait_queue(&learn_wait, &wait); ++ return retval; ++} ++ ++static unsigned int ++poll_learn(struct file * file, poll_table * wait) ++{ ++ poll_wait(file, &learn_wait, wait); ++ ++ if (learn_buffer_len) ++ return (POLLIN | POLLRDNORM); ++ ++ return 0; ++} ++ ++void ++gr_clear_learn_entries(void) ++{ ++ char *tmp; ++ ++ mutex_lock(&gr_learn_user_mutex); ++ if (learn_buffer != NULL) { ++ spin_lock(&gr_learn_lock); ++ tmp = learn_buffer; ++ learn_buffer = NULL; ++ spin_unlock(&gr_learn_lock); ++ vfree(learn_buffer); ++ } ++ if (learn_buffer_user != NULL) { ++ vfree(learn_buffer_user); ++ learn_buffer_user = NULL; ++ } ++ learn_buffer_len = 0; ++ mutex_unlock(&gr_learn_user_mutex); ++ ++ return; ++} ++ ++void ++gr_add_learn_entry(const char *fmt, ...) ++{ ++ va_list args; ++ unsigned int len; ++ ++ if (!gr_learn_attached) ++ return; ++ ++ spin_lock(&gr_learn_lock); ++ ++ /* leave a gap at the end so we know when it's "full" but don't have to ++ compute the exact length of the string we're trying to append ++ */ ++ if (learn_buffer_len > LEARN_BUFFER_SIZE - 16384) { ++ spin_unlock(&gr_learn_lock); ++ wake_up_interruptible(&learn_wait); ++ return; ++ } ++ if (learn_buffer == NULL) { ++ spin_unlock(&gr_learn_lock); ++ return; ++ } ++ ++ va_start(args, fmt); ++ len = vsnprintf(learn_buffer + learn_buffer_len, LEARN_BUFFER_SIZE - learn_buffer_len, fmt, args); ++ va_end(args); ++ ++ learn_buffer_len += len + 1; ++ ++ spin_unlock(&gr_learn_lock); ++ wake_up_interruptible(&learn_wait); ++ ++ return; ++} ++ ++static int ++open_learn(struct inode *inode, struct file *file) ++{ ++ if (file->f_mode & FMODE_READ && gr_learn_attached) ++ return -EBUSY; ++ if (file->f_mode & FMODE_READ) { ++ int retval = 0; ++ mutex_lock(&gr_learn_user_mutex); ++ if (learn_buffer == NULL) ++ learn_buffer = vmalloc(LEARN_BUFFER_SIZE); ++ if (learn_buffer_user == NULL) ++ learn_buffer_user = vmalloc(LEARN_BUFFER_SIZE); ++ if (learn_buffer == NULL) { ++ retval = -ENOMEM; ++ goto out_error; ++ } ++ if (learn_buffer_user == NULL) { ++ retval = -ENOMEM; ++ goto out_error; ++ } ++ learn_buffer_len = 0; ++ learn_buffer_user_len = 0; ++ gr_learn_attached = 1; ++out_error: ++ mutex_unlock(&gr_learn_user_mutex); ++ return retval; ++ } ++ return 0; ++} ++ ++static int ++close_learn(struct inode *inode, struct file *file) ++{ ++ char *tmp; ++ ++ if (file->f_mode & FMODE_READ) { ++ mutex_lock(&gr_learn_user_mutex); ++ if (learn_buffer != NULL) { ++ spin_lock(&gr_learn_lock); ++ tmp = learn_buffer; ++ learn_buffer = NULL; ++ spin_unlock(&gr_learn_lock); ++ vfree(tmp); ++ } ++ if (learn_buffer_user != NULL) { ++ vfree(learn_buffer_user); ++ learn_buffer_user = NULL; ++ } ++ learn_buffer_len = 0; ++ learn_buffer_user_len = 0; ++ gr_learn_attached = 0; ++ mutex_unlock(&gr_learn_user_mutex); ++ } ++ ++ return 0; ++} ++ ++const struct file_operations grsec_fops = { ++ .read = read_learn, ++ .write = write_grsec_handler, ++ .open = open_learn, ++ .release = close_learn, ++ .poll = poll_learn, ++}; +diff -urNp linux-2.6.38.3/grsecurity/gracl_res.c linux-2.6.38.3/grsecurity/gracl_res.c +--- linux-2.6.38.3/grsecurity/gracl_res.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/gracl_res.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,68 @@ ++#include <linux/kernel.h> ++#include <linux/sched.h> ++#include <linux/gracl.h> ++#include <linux/grinternal.h> ++ ++static const char *restab_log[] = { ++ [RLIMIT_CPU] = "RLIMIT_CPU", ++ [RLIMIT_FSIZE] = "RLIMIT_FSIZE", ++ [RLIMIT_DATA] = "RLIMIT_DATA", ++ [RLIMIT_STACK] = "RLIMIT_STACK", ++ [RLIMIT_CORE] = "RLIMIT_CORE", ++ [RLIMIT_RSS] = "RLIMIT_RSS", ++ [RLIMIT_NPROC] = "RLIMIT_NPROC", ++ [RLIMIT_NOFILE] = "RLIMIT_NOFILE", ++ [RLIMIT_MEMLOCK] = "RLIMIT_MEMLOCK", ++ [RLIMIT_AS] = "RLIMIT_AS", ++ [RLIMIT_LOCKS] = "RLIMIT_LOCKS", ++ [RLIMIT_SIGPENDING] = "RLIMIT_SIGPENDING", ++ [RLIMIT_MSGQUEUE] = "RLIMIT_MSGQUEUE", ++ [RLIMIT_NICE] = "RLIMIT_NICE", ++ [RLIMIT_RTPRIO] = "RLIMIT_RTPRIO", ++ [RLIMIT_RTTIME] = "RLIMIT_RTTIME", ++ [GR_CRASH_RES] = "RLIMIT_CRASH" ++}; ++ ++void ++gr_log_resource(const struct task_struct *task, ++ const int res, const unsigned long wanted, const int gt) ++{ ++ const struct cred *cred; ++ unsigned long rlim; ++ ++ if (!gr_acl_is_enabled() && !grsec_resource_logging) ++ return; ++ ++ // not yet supported resource ++ if (unlikely(!restab_log[res])) ++ return; ++ ++ if (res == RLIMIT_CPU || res == RLIMIT_RTTIME) ++ rlim = task_rlimit_max(task, res); ++ else ++ rlim = task_rlimit(task, res); ++ ++ if (likely((rlim == RLIM_INFINITY) || (gt && wanted <= rlim) || (!gt && wanted < rlim))) ++ return; ++ ++ rcu_read_lock(); ++ cred = __task_cred(task); ++ ++ if (res == RLIMIT_NPROC && ++ (cap_raised(cred->cap_effective, CAP_SYS_ADMIN) || ++ cap_raised(cred->cap_effective, CAP_SYS_RESOURCE))) ++ goto out_rcu_unlock; ++ else if (res == RLIMIT_MEMLOCK && ++ cap_raised(cred->cap_effective, CAP_IPC_LOCK)) ++ goto out_rcu_unlock; ++ else if (res == RLIMIT_NICE && cap_raised(cred->cap_effective, CAP_SYS_NICE)) ++ goto out_rcu_unlock; ++ rcu_read_unlock(); ++ ++ gr_log_res_ulong2_str(GR_DONT_AUDIT, GR_RESOURCE_MSG, task, wanted, restab_log[res], rlim); ++ ++ return; ++out_rcu_unlock: ++ rcu_read_unlock(); ++ return; ++} +diff -urNp linux-2.6.38.3/grsecurity/gracl_segv.c linux-2.6.38.3/grsecurity/gracl_segv.c +--- linux-2.6.38.3/grsecurity/gracl_segv.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/gracl_segv.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,300 @@ ++#include <linux/kernel.h> ++#include <linux/mm.h> ++#include <asm/uaccess.h> ++#include <asm/errno.h> ++#include <asm/mman.h> ++#include <net/sock.h> ++#include <linux/file.h> ++#include <linux/fs.h> ++#include <linux/net.h> ++#include <linux/in.h> ++#include <linux/smp_lock.h> ++#include <linux/slab.h> ++#include <linux/types.h> ++#include <linux/sched.h> ++#include <linux/timer.h> ++#include <linux/gracl.h> ++#include <linux/grsecurity.h> ++#include <linux/grinternal.h> ++ ++static struct crash_uid *uid_set; ++static unsigned short uid_used; ++static DEFINE_SPINLOCK(gr_uid_lock); ++extern rwlock_t gr_inode_lock; ++extern struct acl_subject_label * ++ lookup_acl_subj_label(const ino_t inode, const dev_t dev, ++ struct acl_role_label *role); ++ ++#ifdef CONFIG_BTRFS_FS ++extern dev_t get_btrfs_dev_from_inode(struct inode *inode); ++extern int btrfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat); ++#endif ++ ++static inline dev_t __get_dev(const struct dentry *dentry) ++{ ++#ifdef CONFIG_BTRFS_FS ++ if (dentry->d_inode->i_op && dentry->d_inode->i_op->getattr == &btrfs_getattr) ++ return get_btrfs_dev_from_inode(dentry->d_inode); ++ else ++#endif ++ return dentry->d_inode->i_sb->s_dev; ++} ++ ++int ++gr_init_uidset(void) ++{ ++ uid_set = ++ kmalloc(GR_UIDTABLE_MAX * sizeof (struct crash_uid), GFP_KERNEL); ++ uid_used = 0; ++ ++ return uid_set ? 1 : 0; ++} ++ ++void ++gr_free_uidset(void) ++{ ++ if (uid_set) ++ kfree(uid_set); ++ ++ return; ++} ++ ++int ++gr_find_uid(const uid_t uid) ++{ ++ struct crash_uid *tmp = uid_set; ++ uid_t buid; ++ int low = 0, high = uid_used - 1, mid; ++ ++ while (high >= low) { ++ mid = (low + high) >> 1; ++ buid = tmp[mid].uid; ++ if (buid == uid) ++ return mid; ++ if (buid > uid) ++ high = mid - 1; ++ if (buid < uid) ++ low = mid + 1; ++ } ++ ++ return -1; ++} ++ ++static __inline__ void ++gr_insertsort(void) ++{ ++ unsigned short i, j; ++ struct crash_uid index; ++ ++ for (i = 1; i < uid_used; i++) { ++ index = uid_set[i]; ++ j = i; ++ while ((j > 0) && uid_set[j - 1].uid > index.uid) { ++ uid_set[j] = uid_set[j - 1]; ++ j--; ++ } ++ uid_set[j] = index; ++ } ++ ++ return; ++} ++ ++static __inline__ void ++gr_insert_uid(const uid_t uid, const unsigned long expires) ++{ ++ int loc; ++ ++ if (uid_used == GR_UIDTABLE_MAX) ++ return; ++ ++ loc = gr_find_uid(uid); ++ ++ if (loc >= 0) { ++ uid_set[loc].expires = expires; ++ return; ++ } ++ ++ uid_set[uid_used].uid = uid; ++ uid_set[uid_used].expires = expires; ++ uid_used++; ++ ++ gr_insertsort(); ++ ++ return; ++} ++ ++void ++gr_remove_uid(const unsigned short loc) ++{ ++ unsigned short i; ++ ++ for (i = loc + 1; i < uid_used; i++) ++ uid_set[i - 1] = uid_set[i]; ++ ++ uid_used--; ++ ++ return; ++} ++ ++int ++gr_check_crash_uid(const uid_t uid) ++{ ++ int loc; ++ int ret = 0; ++ ++ if (unlikely(!gr_acl_is_enabled())) ++ return 0; ++ ++ spin_lock(&gr_uid_lock); ++ loc = gr_find_uid(uid); ++ ++ if (loc < 0) ++ goto out_unlock; ++ ++ if (time_before_eq(uid_set[loc].expires, get_seconds())) ++ gr_remove_uid(loc); ++ else ++ ret = 1; ++ ++out_unlock: ++ spin_unlock(&gr_uid_lock); ++ return ret; ++} ++ ++static __inline__ int ++proc_is_setxid(const struct cred *cred) ++{ ++ if (cred->uid != cred->euid || cred->uid != cred->suid || ++ cred->uid != cred->fsuid) ++ return 1; ++ if (cred->gid != cred->egid || cred->gid != cred->sgid || ++ cred->gid != cred->fsgid) ++ return 1; ++ ++ return 0; ++} ++ ++extern int gr_fake_force_sig(int sig, struct task_struct *t); ++ ++void ++gr_handle_crash(struct task_struct *task, const int sig) ++{ ++ struct acl_subject_label *curr; ++ struct acl_subject_label *curr2; ++ struct task_struct *tsk, *tsk2; ++ const struct cred *cred; ++ const struct cred *cred2; ++ ++ if (sig != SIGSEGV && sig != SIGKILL && sig != SIGBUS && sig != SIGILL) ++ return; ++ ++ if (unlikely(!gr_acl_is_enabled())) ++ return; ++ ++ curr = task->acl; ++ ++ if (!(curr->resmask & (1 << GR_CRASH_RES))) ++ return; ++ ++ if (time_before_eq(curr->expires, get_seconds())) { ++ curr->expires = 0; ++ curr->crashes = 0; ++ } ++ ++ curr->crashes++; ++ ++ if (!curr->expires) ++ curr->expires = get_seconds() + curr->res[GR_CRASH_RES].rlim_max; ++ ++ if ((curr->crashes >= curr->res[GR_CRASH_RES].rlim_cur) && ++ time_after(curr->expires, get_seconds())) { ++ rcu_read_lock(); ++ cred = __task_cred(task); ++ if (cred->uid && proc_is_setxid(cred)) { ++ gr_log_crash1(GR_DONT_AUDIT, GR_SEGVSTART_ACL_MSG, task, curr->res[GR_CRASH_RES].rlim_max); ++ spin_lock(&gr_uid_lock); ++ gr_insert_uid(cred->uid, curr->expires); ++ spin_unlock(&gr_uid_lock); ++ curr->expires = 0; ++ curr->crashes = 0; ++ read_lock(&tasklist_lock); ++ do_each_thread(tsk2, tsk) { ++ cred2 = __task_cred(tsk); ++ if (tsk != task && cred2->uid == cred->uid) ++ gr_fake_force_sig(SIGKILL, tsk); ++ } while_each_thread(tsk2, tsk); ++ read_unlock(&tasklist_lock); ++ } else { ++ gr_log_crash2(GR_DONT_AUDIT, GR_SEGVNOSUID_ACL_MSG, task, curr->res[GR_CRASH_RES].rlim_max); ++ read_lock(&tasklist_lock); ++ do_each_thread(tsk2, tsk) { ++ if (likely(tsk != task)) { ++ curr2 = tsk->acl; ++ ++ if (curr2->device == curr->device && ++ curr2->inode == curr->inode) ++ gr_fake_force_sig(SIGKILL, tsk); ++ } ++ } while_each_thread(tsk2, tsk); ++ read_unlock(&tasklist_lock); ++ } ++ rcu_read_unlock(); ++ } ++ ++ return; ++} ++ ++int ++gr_check_crash_exec(const struct file *filp) ++{ ++ struct acl_subject_label *curr; ++ ++ if (unlikely(!gr_acl_is_enabled())) ++ return 0; ++ ++ read_lock(&gr_inode_lock); ++ curr = lookup_acl_subj_label(filp->f_path.dentry->d_inode->i_ino, ++ __get_dev(filp->f_path.dentry), ++ current->role); ++ read_unlock(&gr_inode_lock); ++ ++ if (!curr || !(curr->resmask & (1 << GR_CRASH_RES)) || ++ (!curr->crashes && !curr->expires)) ++ return 0; ++ ++ if ((curr->crashes >= curr->res[GR_CRASH_RES].rlim_cur) && ++ time_after(curr->expires, get_seconds())) ++ return 1; ++ else if (time_before_eq(curr->expires, get_seconds())) { ++ curr->crashes = 0; ++ curr->expires = 0; ++ } ++ ++ return 0; ++} ++ ++void ++gr_handle_alertkill(struct task_struct *task) ++{ ++ struct acl_subject_label *curracl; ++ __u32 curr_ip; ++ struct task_struct *p, *p2; ++ ++ if (unlikely(!gr_acl_is_enabled())) ++ return; ++ ++ curracl = task->acl; ++ curr_ip = task->signal->curr_ip; ++ ++ if ((curracl->mode & GR_KILLIPPROC) && curr_ip) { ++ read_lock(&tasklist_lock); ++ do_each_thread(p2, p) { ++ if (p->signal->curr_ip == curr_ip) ++ gr_fake_force_sig(SIGKILL, p); ++ } while_each_thread(p2, p); ++ read_unlock(&tasklist_lock); ++ } else if (curracl->mode & GR_KILLPROC) ++ gr_fake_force_sig(SIGKILL, task); ++ ++ return; ++} +diff -urNp linux-2.6.38.3/grsecurity/gracl_shm.c linux-2.6.38.3/grsecurity/gracl_shm.c +--- linux-2.6.38.3/grsecurity/gracl_shm.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/gracl_shm.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,40 @@ ++#include <linux/kernel.h> ++#include <linux/mm.h> ++#include <linux/sched.h> ++#include <linux/file.h> ++#include <linux/ipc.h> ++#include <linux/gracl.h> ++#include <linux/grsecurity.h> ++#include <linux/grinternal.h> ++ ++int ++gr_handle_shmat(const pid_t shm_cprid, const pid_t shm_lapid, ++ const time_t shm_createtime, const uid_t cuid, const int shmid) ++{ ++ struct task_struct *task; ++ ++ if (!gr_acl_is_enabled()) ++ return 1; ++ ++ rcu_read_lock(); ++ read_lock(&tasklist_lock); ++ ++ task = find_task_by_vpid(shm_cprid); ++ ++ if (unlikely(!task)) ++ task = find_task_by_vpid(shm_lapid); ++ ++ if (unlikely(task && (time_before_eq((unsigned long)task->start_time.tv_sec, (unsigned long)shm_createtime) || ++ (task->pid == shm_lapid)) && ++ (task->acl->mode & GR_PROTSHM) && ++ (task->acl != current->acl))) { ++ read_unlock(&tasklist_lock); ++ rcu_read_unlock(); ++ gr_log_int3(GR_DONT_AUDIT, GR_SHMAT_ACL_MSG, cuid, shm_cprid, shmid); ++ return 0; ++ } ++ read_unlock(&tasklist_lock); ++ rcu_read_unlock(); ++ ++ return 1; ++} +diff -urNp linux-2.6.38.3/grsecurity/grsec_chdir.c linux-2.6.38.3/grsecurity/grsec_chdir.c +--- linux-2.6.38.3/grsecurity/grsec_chdir.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/grsec_chdir.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,19 @@ ++#include <linux/kernel.h> ++#include <linux/sched.h> ++#include <linux/fs.h> ++#include <linux/file.h> ++#include <linux/grsecurity.h> ++#include <linux/grinternal.h> ++ ++void ++gr_log_chdir(const struct dentry *dentry, const struct vfsmount *mnt) ++{ ++#ifdef CONFIG_GRKERNSEC_AUDIT_CHDIR ++ if ((grsec_enable_chdir && grsec_enable_group && ++ in_group_p(grsec_audit_gid)) || (grsec_enable_chdir && ++ !grsec_enable_group)) { ++ gr_log_fs_generic(GR_DO_AUDIT, GR_CHDIR_AUDIT_MSG, dentry, mnt); ++ } ++#endif ++ return; ++} +diff -urNp linux-2.6.38.3/grsecurity/grsec_chroot.c linux-2.6.38.3/grsecurity/grsec_chroot.c +--- linux-2.6.38.3/grsecurity/grsec_chroot.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/grsec_chroot.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,355 @@ ++#include <linux/kernel.h> ++#include <linux/module.h> ++#include <linux/sched.h> ++#include <linux/file.h> ++#include <linux/fs.h> ++#include <linux/mount.h> ++#include <linux/types.h> ++#include <linux/pid_namespace.h> ++#include <linux/grsecurity.h> ++#include <linux/grinternal.h> ++ ++void gr_set_chroot_entries(struct task_struct *task, struct path *path) ++{ ++#ifdef CONFIG_GRKERNSEC ++ if (task->pid > 1 && path->dentry != init_task.fs->root.dentry && ++ path->dentry != task->nsproxy->mnt_ns->root->mnt_root) ++ task->gr_is_chrooted = 1; ++ else ++ task->gr_is_chrooted = 0; ++ ++ task->gr_chroot_dentry = path->dentry; ++#endif ++ return; ++} ++ ++void gr_clear_chroot_entries(struct task_struct *task) ++{ ++#ifdef CONFIG_GRKERNSEC ++ task->gr_is_chrooted = 0; ++ task->gr_chroot_dentry = NULL; ++#endif ++ return; ++} ++ ++int ++gr_handle_chroot_unix(struct pid *pid) ++{ ++#ifdef CONFIG_GRKERNSEC_CHROOT_UNIX ++ struct task_struct *p; ++ ++ if (unlikely(!grsec_enable_chroot_unix)) ++ return 1; ++ ++ if (likely(!proc_is_chrooted(current))) ++ return 1; ++ ++ rcu_read_lock(); ++ read_lock(&tasklist_lock); ++ p = pid_task(pid, PIDTYPE_PID); ++ if (unlikely(p && !have_same_root(current, p))) { ++ read_unlock(&tasklist_lock); ++ rcu_read_unlock(); ++ gr_log_noargs(GR_DONT_AUDIT, GR_UNIX_CHROOT_MSG); ++ return 0; ++ } ++ read_unlock(&tasklist_lock); ++ rcu_read_unlock(); ++#endif ++ return 1; ++} ++ ++int ++gr_handle_chroot_nice(void) ++{ ++#ifdef CONFIG_GRKERNSEC_CHROOT_NICE ++ if (grsec_enable_chroot_nice && proc_is_chrooted(current)) { ++ gr_log_noargs(GR_DONT_AUDIT, GR_NICE_CHROOT_MSG); ++ return -EPERM; ++ } ++#endif ++ return 0; ++} ++ ++int ++gr_handle_chroot_setpriority(struct task_struct *p, const int niceval) ++{ ++#ifdef CONFIG_GRKERNSEC_CHROOT_NICE ++ if (grsec_enable_chroot_nice && (niceval < task_nice(p)) ++ && proc_is_chrooted(current)) { ++ gr_log_str_int(GR_DONT_AUDIT, GR_PRIORITY_CHROOT_MSG, p->comm, p->pid); ++ return -EACCES; ++ } ++#endif ++ return 0; ++} ++ ++int ++gr_handle_chroot_rawio(const struct inode *inode) ++{ ++#ifdef CONFIG_GRKERNSEC_CHROOT_CAPS ++ if (grsec_enable_chroot_caps && proc_is_chrooted(current) && ++ inode && S_ISBLK(inode->i_mode) && !capable(CAP_SYS_RAWIO)) ++ return 1; ++#endif ++ return 0; ++} ++ ++int ++gr_handle_chroot_fowner(struct pid *pid, enum pid_type type) ++{ ++#ifdef CONFIG_GRKERNSEC_CHROOT_FINDTASK ++ struct task_struct *p; ++ int ret = 0; ++ if (!grsec_enable_chroot_findtask || !proc_is_chrooted(current) || !pid) ++ return ret; ++ ++ read_lock(&tasklist_lock); ++ do_each_pid_task(pid, type, p) { ++ if (!have_same_root(current, p)) { ++ ret = 1; ++ goto out; ++ } ++ } while_each_pid_task(pid, type, p); ++out: ++ read_unlock(&tasklist_lock); ++ return ret; ++#endif ++ return 0; ++} ++ ++int ++gr_pid_is_chrooted(struct task_struct *p) ++{ ++#ifdef CONFIG_GRKERNSEC_CHROOT_FINDTASK ++ if (!grsec_enable_chroot_findtask || !proc_is_chrooted(current) || p == NULL) ++ return 0; ++ ++ if ((p->exit_state & (EXIT_ZOMBIE | EXIT_DEAD)) || ++ !have_same_root(current, p)) { ++ return 1; ++ } ++#endif ++ return 0; ++} ++ ++EXPORT_SYMBOL(gr_pid_is_chrooted); ++ ++#if defined(CONFIG_GRKERNSEC_CHROOT_DOUBLE) || defined(CONFIG_GRKERNSEC_CHROOT_FCHDIR) ++int gr_is_outside_chroot(const struct dentry *u_dentry, const struct vfsmount *u_mnt) ++{ ++ struct path path, currentroot; ++ int ret = 0; ++ ++ path.dentry = (struct dentry *)u_dentry; ++ path.mnt = (struct vfsmount *)u_mnt; ++ get_fs_root(current->fs, ¤troot); ++ if (path_is_under(&path, ¤troot)) ++ ret = 1; ++ path_put(¤troot); ++ ++ return ret; ++} ++#endif ++ ++int ++gr_chroot_fchdir(struct dentry *u_dentry, struct vfsmount *u_mnt) ++{ ++#ifdef CONFIG_GRKERNSEC_CHROOT_FCHDIR ++ if (!grsec_enable_chroot_fchdir) ++ return 1; ++ ++ if (!proc_is_chrooted(current)) ++ return 1; ++ else if (!gr_is_outside_chroot(u_dentry, u_mnt)) { ++ gr_log_fs_generic(GR_DONT_AUDIT, GR_CHROOT_FCHDIR_MSG, u_dentry, u_mnt); ++ return 0; ++ } ++#endif ++ return 1; ++} ++ ++int ++gr_chroot_shmat(const pid_t shm_cprid, const pid_t shm_lapid, ++ const time_t shm_createtime) ++{ ++#ifdef CONFIG_GRKERNSEC_CHROOT_SHMAT ++ struct pid *pid = NULL; ++ time_t starttime; ++ ++ if (unlikely(!grsec_enable_chroot_shmat)) ++ return 1; ++ ++ if (likely(!proc_is_chrooted(current))) ++ return 1; ++ ++ rcu_read_lock(); ++ read_lock(&tasklist_lock); ++ ++ pid = find_vpid(shm_cprid); ++ if (pid) { ++ struct task_struct *p; ++ p = pid_task(pid, PIDTYPE_PID); ++ if (p == NULL) ++ goto unlock; ++ starttime = p->start_time.tv_sec; ++ if (unlikely(!have_same_root(current, p) && ++ time_before_eq((unsigned long)starttime, (unsigned long)shm_createtime))) { ++ read_unlock(&tasklist_lock); ++ rcu_read_unlock(); ++ gr_log_noargs(GR_DONT_AUDIT, GR_SHMAT_CHROOT_MSG); ++ return 0; ++ } ++ } else { ++ pid = find_vpid(shm_lapid); ++ if (pid) { ++ struct task_struct *p; ++ p = pid_task(pid, PIDTYPE_PID); ++ if (p == NULL) ++ goto unlock; ++ if (unlikely(!have_same_root(current, p))) { ++ read_unlock(&tasklist_lock); ++ rcu_read_unlock(); ++ gr_log_noargs(GR_DONT_AUDIT, GR_SHMAT_CHROOT_MSG); ++ return 0; ++ } ++ } ++ } ++unlock: ++ read_unlock(&tasklist_lock); ++ rcu_read_unlock(); ++#endif ++ return 1; ++} ++ ++void ++gr_log_chroot_exec(const struct dentry *dentry, const struct vfsmount *mnt) ++{ ++#ifdef CONFIG_GRKERNSEC_CHROOT_EXECLOG ++ if (grsec_enable_chroot_execlog && proc_is_chrooted(current)) ++ gr_log_fs_generic(GR_DO_AUDIT, GR_EXEC_CHROOT_MSG, dentry, mnt); ++#endif ++ return; ++} ++ ++int ++gr_handle_chroot_mknod(const struct dentry *dentry, ++ const struct vfsmount *mnt, const int mode) ++{ ++#ifdef CONFIG_GRKERNSEC_CHROOT_MKNOD ++ if (grsec_enable_chroot_mknod && !S_ISFIFO(mode) && !S_ISREG(mode) && ++ proc_is_chrooted(current)) { ++ gr_log_fs_generic(GR_DONT_AUDIT, GR_MKNOD_CHROOT_MSG, dentry, mnt); ++ return -EPERM; ++ } ++#endif ++ return 0; ++} ++ ++int ++gr_handle_chroot_mount(const struct dentry *dentry, ++ const struct vfsmount *mnt, const char *dev_name) ++{ ++#ifdef CONFIG_GRKERNSEC_CHROOT_MOUNT ++ if (grsec_enable_chroot_mount && proc_is_chrooted(current)) { ++ gr_log_str_fs(GR_DONT_AUDIT, GR_MOUNT_CHROOT_MSG, dev_name, dentry, mnt); ++ return -EPERM; ++ } ++#endif ++ return 0; ++} ++ ++int ++gr_handle_chroot_pivot(void) ++{ ++#ifdef CONFIG_GRKERNSEC_CHROOT_PIVOT ++ if (grsec_enable_chroot_pivot && proc_is_chrooted(current)) { ++ gr_log_noargs(GR_DONT_AUDIT, GR_PIVOT_CHROOT_MSG); ++ return -EPERM; ++ } ++#endif ++ return 0; ++} ++ ++int ++gr_handle_chroot_chroot(const struct dentry *dentry, const struct vfsmount *mnt) ++{ ++#ifdef CONFIG_GRKERNSEC_CHROOT_DOUBLE ++ if (grsec_enable_chroot_double && proc_is_chrooted(current) && ++ !gr_is_outside_chroot(dentry, mnt)) { ++ gr_log_fs_generic(GR_DONT_AUDIT, GR_CHROOT_CHROOT_MSG, dentry, mnt); ++ return -EPERM; ++ } ++#endif ++ return 0; ++} ++ ++int ++gr_handle_chroot_caps(struct path *path) ++{ ++#ifdef CONFIG_GRKERNSEC_CHROOT_CAPS ++ if (grsec_enable_chroot_caps && current->pid > 1 && current->fs != NULL && ++ (init_task.fs->root.dentry != path->dentry) && ++ (current->nsproxy->mnt_ns->root->mnt_root != path->dentry)) { ++ ++ kernel_cap_t chroot_caps = GR_CHROOT_CAPS; ++ const struct cred *old = current_cred(); ++ struct cred *new = prepare_creds(); ++ if (new == NULL) ++ return 1; ++ ++ new->cap_permitted = cap_drop(old->cap_permitted, ++ chroot_caps); ++ new->cap_inheritable = cap_drop(old->cap_inheritable, ++ chroot_caps); ++ new->cap_effective = cap_drop(old->cap_effective, ++ chroot_caps); ++ ++ commit_creds(new); ++ ++ return 0; ++ } ++#endif ++ return 0; ++} ++ ++int ++gr_handle_chroot_sysctl(const int op) ++{ ++#ifdef CONFIG_GRKERNSEC_CHROOT_SYSCTL ++ if (grsec_enable_chroot_sysctl && (op & MAY_WRITE) && ++ proc_is_chrooted(current)) ++ return -EACCES; ++#endif ++ return 0; ++} ++ ++void ++gr_handle_chroot_chdir(struct path *path) ++{ ++#ifdef CONFIG_GRKERNSEC_CHROOT_CHDIR ++ if (grsec_enable_chroot_chdir) ++ set_fs_pwd(current->fs, path); ++#endif ++ return; ++} ++ ++int ++gr_handle_chroot_chmod(const struct dentry *dentry, ++ const struct vfsmount *mnt, const int mode) ++{ ++#ifdef CONFIG_GRKERNSEC_CHROOT_CHMOD ++ /* allow chmod +s on directories, but not files */ ++ if (grsec_enable_chroot_chmod && !S_ISDIR(dentry->d_inode->i_mode) && ++ ((mode & S_ISUID) || ((mode & (S_ISGID | S_IXGRP)) == (S_ISGID | S_IXGRP))) && ++ proc_is_chrooted(current)) { ++ gr_log_fs_generic(GR_DONT_AUDIT, GR_CHMOD_CHROOT_MSG, dentry, mnt); ++ return -EPERM; ++ } ++#endif ++ return 0; ++} ++ ++#ifdef CONFIG_SECURITY ++EXPORT_SYMBOL(gr_handle_chroot_caps); ++#endif +diff -urNp linux-2.6.38.3/grsecurity/grsec_disabled.c linux-2.6.38.3/grsecurity/grsec_disabled.c +--- linux-2.6.38.3/grsecurity/grsec_disabled.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/grsec_disabled.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,447 @@ ++#include <linux/kernel.h> ++#include <linux/module.h> ++#include <linux/sched.h> ++#include <linux/file.h> ++#include <linux/fs.h> ++#include <linux/kdev_t.h> ++#include <linux/net.h> ++#include <linux/in.h> ++#include <linux/ip.h> ++#include <linux/skbuff.h> ++#include <linux/sysctl.h> ++ ++#ifdef CONFIG_PAX_HAVE_ACL_FLAGS ++void ++pax_set_initial_flags(struct linux_binprm *bprm) ++{ ++ return; ++} ++#endif ++ ++#ifdef CONFIG_SYSCTL ++__u32 ++gr_handle_sysctl(const struct ctl_table * table, const int op) ++{ ++ return 0; ++} ++#endif ++ ++#ifdef CONFIG_TASKSTATS ++int gr_is_taskstats_denied(int pid) ++{ ++ return 0; ++} ++#endif ++ ++int ++gr_acl_is_enabled(void) ++{ ++ return 0; ++} ++ ++int ++gr_handle_rawio(const struct inode *inode) ++{ ++ return 0; ++} ++ ++void ++gr_acl_handle_psacct(struct task_struct *task, const long code) ++{ ++ return; ++} ++ ++int ++gr_handle_ptrace(struct task_struct *task, const long request) ++{ ++ return 0; ++} ++ ++int ++gr_handle_proc_ptrace(struct task_struct *task) ++{ ++ return 0; ++} ++ ++void ++gr_learn_resource(const struct task_struct *task, ++ const int res, const unsigned long wanted, const int gt) ++{ ++ return; ++} ++ ++int ++gr_set_acls(const int type) ++{ ++ return 0; ++} ++ ++int ++gr_check_hidden_task(const struct task_struct *tsk) ++{ ++ return 0; ++} ++ ++int ++gr_check_protected_task(const struct task_struct *task) ++{ ++ return 0; ++} ++ ++int ++gr_check_protected_task_fowner(struct pid *pid, enum pid_type type) ++{ ++ return 0; ++} ++ ++void ++gr_copy_label(struct task_struct *tsk) ++{ ++ return; ++} ++ ++void ++gr_set_pax_flags(struct task_struct *task) ++{ ++ return; ++} ++ ++int ++gr_set_proc_label(const struct dentry *dentry, const struct vfsmount *mnt, ++ const int unsafe_share) ++{ ++ return 0; ++} ++ ++void ++gr_handle_delete(const ino_t ino, const dev_t dev) ++{ ++ return; ++} ++ ++void ++gr_handle_create(const struct dentry *dentry, const struct vfsmount *mnt) ++{ ++ return; ++} ++ ++void ++gr_handle_crash(struct task_struct *task, const int sig) ++{ ++ return; ++} ++ ++int ++gr_check_crash_exec(const struct file *filp) ++{ ++ return 0; ++} ++ ++int ++gr_check_crash_uid(const uid_t uid) ++{ ++ return 0; ++} ++ ++void ++gr_handle_rename(struct inode *old_dir, struct inode *new_dir, ++ struct dentry *old_dentry, ++ struct dentry *new_dentry, ++ struct vfsmount *mnt, const __u8 replace) ++{ ++ return; ++} ++ ++int ++gr_search_socket(const int family, const int type, const int protocol) ++{ ++ return 1; ++} ++ ++int ++gr_search_connectbind(const int mode, const struct socket *sock, ++ const struct sockaddr_in *addr) ++{ ++ return 0; ++} ++ ++int ++gr_is_capable(const int cap) ++{ ++ return 1; ++} ++ ++int ++gr_is_capable_nolog(const int cap) ++{ ++ return 1; ++} ++ ++void ++gr_handle_alertkill(struct task_struct *task) ++{ ++ return; ++} ++ ++__u32 ++gr_acl_handle_execve(const struct dentry * dentry, const struct vfsmount * mnt) ++{ ++ return 1; ++} ++ ++__u32 ++gr_acl_handle_hidden_file(const struct dentry * dentry, ++ const struct vfsmount * mnt) ++{ ++ return 1; ++} ++ ++__u32 ++gr_acl_handle_open(const struct dentry * dentry, const struct vfsmount * mnt, ++ const int fmode) ++{ ++ return 1; ++} ++ ++__u32 ++gr_acl_handle_rmdir(const struct dentry * dentry, const struct vfsmount * mnt) ++{ ++ return 1; ++} ++ ++__u32 ++gr_acl_handle_unlink(const struct dentry * dentry, const struct vfsmount * mnt) ++{ ++ return 1; ++} ++ ++int ++gr_acl_handle_mmap(const struct file *file, const unsigned long prot, ++ unsigned int *vm_flags) ++{ ++ return 1; ++} ++ ++__u32 ++gr_acl_handle_truncate(const struct dentry * dentry, ++ const struct vfsmount * mnt) ++{ ++ return 1; ++} ++ ++__u32 ++gr_acl_handle_utime(const struct dentry * dentry, const struct vfsmount * mnt) ++{ ++ return 1; ++} ++ ++__u32 ++gr_acl_handle_access(const struct dentry * dentry, ++ const struct vfsmount * mnt, const int fmode) ++{ ++ return 1; ++} ++ ++__u32 ++gr_acl_handle_fchmod(const struct dentry * dentry, const struct vfsmount * mnt, ++ mode_t mode) ++{ ++ return 1; ++} ++ ++__u32 ++gr_acl_handle_chmod(const struct dentry * dentry, const struct vfsmount * mnt, ++ mode_t mode) ++{ ++ return 1; ++} ++ ++__u32 ++gr_acl_handle_chown(const struct dentry * dentry, const struct vfsmount * mnt) ++{ ++ return 1; ++} ++ ++__u32 ++gr_acl_handle_setxattr(const struct dentry * dentry, const struct vfsmount * mnt) ++{ ++ return 1; ++} ++ ++void ++grsecurity_init(void) ++{ ++ return; ++} ++ ++__u32 ++gr_acl_handle_mknod(const struct dentry * new_dentry, ++ const struct dentry * parent_dentry, ++ const struct vfsmount * parent_mnt, ++ const int mode) ++{ ++ return 1; ++} ++ ++__u32 ++gr_acl_handle_mkdir(const struct dentry * new_dentry, ++ const struct dentry * parent_dentry, ++ const struct vfsmount * parent_mnt) ++{ ++ return 1; ++} ++ ++__u32 ++gr_acl_handle_symlink(const struct dentry * new_dentry, ++ const struct dentry * parent_dentry, ++ const struct vfsmount * parent_mnt, const char *from) ++{ ++ return 1; ++} ++ ++__u32 ++gr_acl_handle_link(const struct dentry * new_dentry, ++ const struct dentry * parent_dentry, ++ const struct vfsmount * parent_mnt, ++ const struct dentry * old_dentry, ++ const struct vfsmount * old_mnt, const char *to) ++{ ++ return 1; ++} ++ ++int ++gr_acl_handle_rename(const struct dentry *new_dentry, ++ const struct dentry *parent_dentry, ++ const struct vfsmount *parent_mnt, ++ const struct dentry *old_dentry, ++ const struct inode *old_parent_inode, ++ const struct vfsmount *old_mnt, const char *newname) ++{ ++ return 0; ++} ++ ++int ++gr_acl_handle_filldir(const struct file *file, const char *name, ++ const int namelen, const ino_t ino) ++{ ++ return 1; ++} ++ ++int ++gr_handle_shmat(const pid_t shm_cprid, const pid_t shm_lapid, ++ const time_t shm_createtime, const uid_t cuid, const int shmid) ++{ ++ return 1; ++} ++ ++int ++gr_search_bind(const struct socket *sock, const struct sockaddr_in *addr) ++{ ++ return 0; ++} ++ ++int ++gr_search_accept(const struct socket *sock) ++{ ++ return 0; ++} ++ ++int ++gr_search_listen(const struct socket *sock) ++{ ++ return 0; ++} ++ ++int ++gr_search_connect(const struct socket *sock, const struct sockaddr_in *addr) ++{ ++ return 0; ++} ++ ++__u32 ++gr_acl_handle_unix(const struct dentry * dentry, const struct vfsmount * mnt) ++{ ++ return 1; ++} ++ ++__u32 ++gr_acl_handle_creat(const struct dentry * dentry, ++ const struct dentry * p_dentry, ++ const struct vfsmount * p_mnt, const int fmode, ++ const int imode) ++{ ++ return 1; ++} ++ ++void ++gr_acl_handle_exit(void) ++{ ++ return; ++} ++ ++int ++gr_acl_handle_mprotect(const struct file *file, const unsigned long prot) ++{ ++ return 1; ++} ++ ++void ++gr_set_role_label(const uid_t uid, const gid_t gid) ++{ ++ return; ++} ++ ++int ++gr_acl_handle_procpidmem(const struct task_struct *task) ++{ ++ return 0; ++} ++ ++int ++gr_search_udp_recvmsg(const struct sock *sk, const struct sk_buff *skb) ++{ ++ return 0; ++} ++ ++int ++gr_search_udp_sendmsg(const struct sock *sk, const struct sockaddr_in *addr) ++{ ++ return 0; ++} ++ ++void ++gr_set_kernel_label(struct task_struct *task) ++{ ++ return; ++} ++ ++int ++gr_check_user_change(int real, int effective, int fs) ++{ ++ return 0; ++} ++ ++int ++gr_check_group_change(int real, int effective, int fs) ++{ ++ return 0; ++} ++ ++int gr_acl_enable_at_secure(void) ++{ ++ return 0; ++} ++ ++dev_t gr_get_dev_from_dentry(struct dentry *dentry) ++{ ++ return dentry->d_inode->i_sb->s_dev; ++} ++ ++EXPORT_SYMBOL(gr_is_capable); ++EXPORT_SYMBOL(gr_is_capable_nolog); ++EXPORT_SYMBOL(gr_learn_resource); ++EXPORT_SYMBOL(gr_set_kernel_label); ++#ifdef CONFIG_SECURITY ++EXPORT_SYMBOL(gr_check_user_change); ++EXPORT_SYMBOL(gr_check_group_change); ++#endif +diff -urNp linux-2.6.38.3/grsecurity/grsec_exec.c linux-2.6.38.3/grsecurity/grsec_exec.c +--- linux-2.6.38.3/grsecurity/grsec_exec.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/grsec_exec.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,147 @@ ++#include <linux/kernel.h> ++#include <linux/sched.h> ++#include <linux/file.h> ++#include <linux/binfmts.h> ++#include <linux/smp_lock.h> ++#include <linux/fs.h> ++#include <linux/types.h> ++#include <linux/grdefs.h> ++#include <linux/grinternal.h> ++#include <linux/capability.h> ++#include <linux/compat.h> ++ ++#include <asm/uaccess.h> ++ ++#ifdef CONFIG_GRKERNSEC_EXECLOG ++static char gr_exec_arg_buf[132]; ++static DEFINE_MUTEX(gr_exec_arg_mutex); ++#endif ++ ++int ++gr_handle_nproc(void) ++{ ++#ifdef CONFIG_GRKERNSEC_EXECVE ++ const struct cred *cred = current_cred(); ++ if (grsec_enable_execve && cred->user && ++ (atomic_read(&cred->user->processes) > rlimit(RLIMIT_NPROC)) && ++ !capable(CAP_SYS_ADMIN) && !capable(CAP_SYS_RESOURCE)) { ++ gr_log_noargs(GR_DONT_AUDIT, GR_NPROC_MSG); ++ return -EAGAIN; ++ } ++#endif ++ return 0; ++} ++ ++void ++gr_handle_exec_args(struct linux_binprm *bprm, const char __user *const __user *argv) ++{ ++#ifdef CONFIG_GRKERNSEC_EXECLOG ++ char *grarg = gr_exec_arg_buf; ++ unsigned int i, x, execlen = 0; ++ char c; ++ ++ if (!((grsec_enable_execlog && grsec_enable_group && ++ in_group_p(grsec_audit_gid)) ++ || (grsec_enable_execlog && !grsec_enable_group))) ++ return; ++ ++ mutex_lock(&gr_exec_arg_mutex); ++ memset(grarg, 0, sizeof(gr_exec_arg_buf)); ++ ++ if (unlikely(argv == NULL)) ++ goto log; ++ ++ for (i = 0; i < bprm->argc && execlen < 128; i++) { ++ const char __user *p; ++ unsigned int len; ++ ++ if (copy_from_user(&p, argv + i, sizeof(p))) ++ goto log; ++ if (!p) ++ goto log; ++ len = strnlen_user(p, 128 - execlen); ++ if (len > 128 - execlen) ++ len = 128 - execlen; ++ else if (len > 0) ++ len--; ++ if (copy_from_user(grarg + execlen, p, len)) ++ goto log; ++ ++ /* rewrite unprintable characters */ ++ for (x = 0; x < len; x++) { ++ c = *(grarg + execlen + x); ++ if (c < 32 || c > 126) ++ *(grarg + execlen + x) = ' '; ++ } ++ ++ execlen += len; ++ *(grarg + execlen) = ' '; ++ *(grarg + execlen + 1) = '\0'; ++ execlen++; ++ } ++ ++ log: ++ gr_log_fs_str(GR_DO_AUDIT, GR_EXEC_AUDIT_MSG, bprm->file->f_path.dentry, ++ bprm->file->f_path.mnt, grarg); ++ mutex_unlock(&gr_exec_arg_mutex); ++#endif ++ return; ++} ++ ++#ifdef CONFIG_COMPAT ++void ++gr_handle_exec_args_compat(struct linux_binprm *bprm, compat_uptr_t __user *argv) ++{ ++#ifdef CONFIG_GRKERNSEC_EXECLOG ++ char *grarg = gr_exec_arg_buf; ++ unsigned int i, x, execlen = 0; ++ char c; ++ ++ if (!((grsec_enable_execlog && grsec_enable_group && ++ in_group_p(grsec_audit_gid)) ++ || (grsec_enable_execlog && !grsec_enable_group))) ++ return; ++ ++ mutex_lock(&gr_exec_arg_mutex); ++ memset(grarg, 0, sizeof(gr_exec_arg_buf)); ++ ++ if (unlikely(argv == NULL)) ++ goto log; ++ ++ for (i = 0; i < bprm->argc && execlen < 128; i++) { ++ compat_uptr_t p; ++ unsigned int len; ++ ++ if (get_user(p, argv + i)) ++ goto log; ++ len = strnlen_user(compat_ptr(p), 128 - execlen); ++ if (len > 128 - execlen) ++ len = 128 - execlen; ++ else if (len > 0) ++ len--; ++ else ++ goto log; ++ if (copy_from_user(grarg + execlen, compat_ptr(p), len)) ++ goto log; ++ ++ /* rewrite unprintable characters */ ++ for (x = 0; x < len; x++) { ++ c = *(grarg + execlen + x); ++ if (c < 32 || c > 126) ++ *(grarg + execlen + x) = ' '; ++ } ++ ++ execlen += len; ++ *(grarg + execlen) = ' '; ++ *(grarg + execlen + 1) = '\0'; ++ execlen++; ++ } ++ ++ log: ++ gr_log_fs_str(GR_DO_AUDIT, GR_EXEC_AUDIT_MSG, bprm->file->f_path.dentry, ++ bprm->file->f_path.mnt, grarg); ++ mutex_unlock(&gr_exec_arg_mutex); ++#endif ++ return; ++} ++#endif +diff -urNp linux-2.6.38.3/grsecurity/grsec_fifo.c linux-2.6.38.3/grsecurity/grsec_fifo.c +--- linux-2.6.38.3/grsecurity/grsec_fifo.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/grsec_fifo.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,24 @@ ++#include <linux/kernel.h> ++#include <linux/sched.h> ++#include <linux/fs.h> ++#include <linux/file.h> ++#include <linux/grinternal.h> ++ ++int ++gr_handle_fifo(const struct dentry *dentry, const struct vfsmount *mnt, ++ const struct dentry *dir, const int flag, const int acc_mode) ++{ ++#ifdef CONFIG_GRKERNSEC_FIFO ++ const struct cred *cred = current_cred(); ++ ++ if (grsec_enable_fifo && S_ISFIFO(dentry->d_inode->i_mode) && ++ !(flag & O_EXCL) && (dir->d_inode->i_mode & S_ISVTX) && ++ (dentry->d_inode->i_uid != dir->d_inode->i_uid) && ++ (cred->fsuid != dentry->d_inode->i_uid)) { ++ if (!inode_permission(dentry->d_inode, acc_mode)) ++ gr_log_fs_int2(GR_DONT_AUDIT, GR_FIFO_MSG, dentry, mnt, dentry->d_inode->i_uid, dentry->d_inode->i_gid); ++ return -EACCES; ++ } ++#endif ++ return 0; ++} +diff -urNp linux-2.6.38.3/grsecurity/grsec_fork.c linux-2.6.38.3/grsecurity/grsec_fork.c +--- linux-2.6.38.3/grsecurity/grsec_fork.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/grsec_fork.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,23 @@ ++#include <linux/kernel.h> ++#include <linux/sched.h> ++#include <linux/grsecurity.h> ++#include <linux/grinternal.h> ++#include <linux/errno.h> ++ ++void ++gr_log_forkfail(const int retval) ++{ ++#ifdef CONFIG_GRKERNSEC_FORKFAIL ++ if (grsec_enable_forkfail && (retval == -EAGAIN || retval == -ENOMEM)) { ++ switch (retval) { ++ case -EAGAIN: ++ gr_log_str(GR_DONT_AUDIT, GR_FAILFORK_MSG, "EAGAIN"); ++ break; ++ case -ENOMEM: ++ gr_log_str(GR_DONT_AUDIT, GR_FAILFORK_MSG, "ENOMEM"); ++ break; ++ } ++ } ++#endif ++ return; ++} +diff -urNp linux-2.6.38.3/grsecurity/grsec_init.c linux-2.6.38.3/grsecurity/grsec_init.c +--- linux-2.6.38.3/grsecurity/grsec_init.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/grsec_init.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,270 @@ ++#include <linux/kernel.h> ++#include <linux/sched.h> ++#include <linux/mm.h> ++#include <linux/smp_lock.h> ++#include <linux/gracl.h> ++#include <linux/slab.h> ++#include <linux/vmalloc.h> ++#include <linux/percpu.h> ++#include <linux/module.h> ++ ++int grsec_enable_link; ++int grsec_enable_dmesg; ++int grsec_enable_harden_ptrace; ++int grsec_enable_fifo; ++int grsec_enable_execve; ++int grsec_enable_execlog; ++int grsec_enable_signal; ++int grsec_enable_forkfail; ++int grsec_enable_audit_ptrace; ++int grsec_enable_time; ++int grsec_enable_audit_textrel; ++int grsec_enable_group; ++int grsec_audit_gid; ++int grsec_enable_chdir; ++int grsec_enable_mount; ++int grsec_enable_rofs; ++int grsec_enable_chroot_findtask; ++int grsec_enable_chroot_mount; ++int grsec_enable_chroot_shmat; ++int grsec_enable_chroot_fchdir; ++int grsec_enable_chroot_double; ++int grsec_enable_chroot_pivot; ++int grsec_enable_chroot_chdir; ++int grsec_enable_chroot_chmod; ++int grsec_enable_chroot_mknod; ++int grsec_enable_chroot_nice; ++int grsec_enable_chroot_execlog; ++int grsec_enable_chroot_caps; ++int grsec_enable_chroot_sysctl; ++int grsec_enable_chroot_unix; ++int grsec_enable_tpe; ++int grsec_tpe_gid; ++int grsec_enable_blackhole; ++#ifdef CONFIG_IPV6_MODULE ++EXPORT_SYMBOL(grsec_enable_blackhole); ++#endif ++int grsec_lastack_retries; ++int grsec_enable_tpe_all; ++int grsec_enable_tpe_invert; ++int grsec_enable_socket_all; ++int grsec_socket_all_gid; ++int grsec_enable_socket_client; ++int grsec_socket_client_gid; ++int grsec_enable_socket_server; ++int grsec_socket_server_gid; ++int grsec_resource_logging; ++int grsec_disable_privio; ++int grsec_enable_log_rwxmaps; ++int grsec_lock; ++ ++DEFINE_SPINLOCK(grsec_alert_lock); ++unsigned long grsec_alert_wtime = 0; ++unsigned long grsec_alert_fyet = 0; ++ ++DEFINE_SPINLOCK(grsec_audit_lock); ++ ++DEFINE_RWLOCK(grsec_exec_file_lock); ++ ++char *gr_shared_page[4]; ++ ++char *gr_alert_log_fmt; ++char *gr_audit_log_fmt; ++char *gr_alert_log_buf; ++char *gr_audit_log_buf; ++ ++extern struct gr_arg *gr_usermode; ++extern unsigned char *gr_system_salt; ++extern unsigned char *gr_system_sum; ++ ++void __init ++grsecurity_init(void) ++{ ++ int j; ++ /* create the per-cpu shared pages */ ++ ++#ifdef CONFIG_X86 ++ memset((char *)(0x41a + PAGE_OFFSET), 0, 36); ++#endif ++ ++ for (j = 0; j < 4; j++) { ++ gr_shared_page[j] = (char *)__alloc_percpu(PAGE_SIZE, __alignof__(unsigned long long)); ++ if (gr_shared_page[j] == NULL) { ++ panic("Unable to allocate grsecurity shared page"); ++ return; ++ } ++ } ++ ++ /* allocate log buffers */ ++ gr_alert_log_fmt = kmalloc(512, GFP_KERNEL); ++ if (!gr_alert_log_fmt) { ++ panic("Unable to allocate grsecurity alert log format buffer"); ++ return; ++ } ++ gr_audit_log_fmt = kmalloc(512, GFP_KERNEL); ++ if (!gr_audit_log_fmt) { ++ panic("Unable to allocate grsecurity audit log format buffer"); ++ return; ++ } ++ gr_alert_log_buf = (char *) get_zeroed_page(GFP_KERNEL); ++ if (!gr_alert_log_buf) { ++ panic("Unable to allocate grsecurity alert log buffer"); ++ return; ++ } ++ gr_audit_log_buf = (char *) get_zeroed_page(GFP_KERNEL); ++ if (!gr_audit_log_buf) { ++ panic("Unable to allocate grsecurity audit log buffer"); ++ return; ++ } ++ ++ /* allocate memory for authentication structure */ ++ gr_usermode = kmalloc(sizeof(struct gr_arg), GFP_KERNEL); ++ gr_system_salt = kmalloc(GR_SALT_LEN, GFP_KERNEL); ++ gr_system_sum = kmalloc(GR_SHA_LEN, GFP_KERNEL); ++ ++ if (!gr_usermode || !gr_system_salt || !gr_system_sum) { ++ panic("Unable to allocate grsecurity authentication structure"); ++ return; ++ } ++ ++ ++#ifdef CONFIG_GRKERNSEC_IO ++#if !defined(CONFIG_GRKERNSEC_SYSCTL_DISTRO) ++ grsec_disable_privio = 1; ++#elif defined(CONFIG_GRKERNSEC_SYSCTL_ON) ++ grsec_disable_privio = 1; ++#else ++ grsec_disable_privio = 0; ++#endif ++#endif ++ ++#ifdef CONFIG_GRKERNSEC_TPE_INVERT ++ /* for backward compatibility, tpe_invert always defaults to on if ++ enabled in the kernel ++ */ ++ grsec_enable_tpe_invert = 1; ++#endif ++ ++#if !defined(CONFIG_GRKERNSEC_SYSCTL) || defined(CONFIG_GRKERNSEC_SYSCTL_ON) ++#ifndef CONFIG_GRKERNSEC_SYSCTL ++ grsec_lock = 1; ++#endif ++ ++#ifdef CONFIG_GRKERNSEC_AUDIT_TEXTREL ++ grsec_enable_audit_textrel = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_RWXMAP_LOG ++ grsec_enable_log_rwxmaps = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_AUDIT_GROUP ++ grsec_enable_group = 1; ++ grsec_audit_gid = CONFIG_GRKERNSEC_AUDIT_GID; ++#endif ++#ifdef CONFIG_GRKERNSEC_AUDIT_CHDIR ++ grsec_enable_chdir = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_HARDEN_PTRACE ++ grsec_enable_harden_ptrace = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_AUDIT_MOUNT ++ grsec_enable_mount = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_LINK ++ grsec_enable_link = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_DMESG ++ grsec_enable_dmesg = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_BLACKHOLE ++ grsec_enable_blackhole = 1; ++ grsec_lastack_retries = 4; ++#endif ++#ifdef CONFIG_GRKERNSEC_FIFO ++ grsec_enable_fifo = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_EXECVE ++ grsec_enable_execve = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_EXECLOG ++ grsec_enable_execlog = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_SIGNAL ++ grsec_enable_signal = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_FORKFAIL ++ grsec_enable_forkfail = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_TIME ++ grsec_enable_time = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_RESLOG ++ grsec_resource_logging = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_FINDTASK ++ grsec_enable_chroot_findtask = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_UNIX ++ grsec_enable_chroot_unix = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_MOUNT ++ grsec_enable_chroot_mount = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_FCHDIR ++ grsec_enable_chroot_fchdir = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_SHMAT ++ grsec_enable_chroot_shmat = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_AUDIT_PTRACE ++ grsec_enable_audit_ptrace = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_DOUBLE ++ grsec_enable_chroot_double = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_PIVOT ++ grsec_enable_chroot_pivot = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_CHDIR ++ grsec_enable_chroot_chdir = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_CHMOD ++ grsec_enable_chroot_chmod = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_MKNOD ++ grsec_enable_chroot_mknod = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_NICE ++ grsec_enable_chroot_nice = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_EXECLOG ++ grsec_enable_chroot_execlog = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_CAPS ++ grsec_enable_chroot_caps = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_SYSCTL ++ grsec_enable_chroot_sysctl = 1; ++#endif ++#ifdef CONFIG_GRKERNSEC_TPE ++ grsec_enable_tpe = 1; ++ grsec_tpe_gid = CONFIG_GRKERNSEC_TPE_GID; ++#ifdef CONFIG_GRKERNSEC_TPE_ALL ++ grsec_enable_tpe_all = 1; ++#endif ++#endif ++#ifdef CONFIG_GRKERNSEC_SOCKET_ALL ++ grsec_enable_socket_all = 1; ++ grsec_socket_all_gid = CONFIG_GRKERNSEC_SOCKET_ALL_GID; ++#endif ++#ifdef CONFIG_GRKERNSEC_SOCKET_CLIENT ++ grsec_enable_socket_client = 1; ++ grsec_socket_client_gid = CONFIG_GRKERNSEC_SOCKET_CLIENT_GID; ++#endif ++#ifdef CONFIG_GRKERNSEC_SOCKET_SERVER ++ grsec_enable_socket_server = 1; ++ grsec_socket_server_gid = CONFIG_GRKERNSEC_SOCKET_SERVER_GID; ++#endif ++#endif ++ ++ return; ++} +diff -urNp linux-2.6.38.3/grsecurity/grsec_link.c linux-2.6.38.3/grsecurity/grsec_link.c +--- linux-2.6.38.3/grsecurity/grsec_link.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/grsec_link.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,43 @@ ++#include <linux/kernel.h> ++#include <linux/sched.h> ++#include <linux/fs.h> ++#include <linux/file.h> ++#include <linux/grinternal.h> ++ ++int ++gr_handle_follow_link(const struct inode *parent, ++ const struct inode *inode, ++ const struct dentry *dentry, const struct vfsmount *mnt) ++{ ++#ifdef CONFIG_GRKERNSEC_LINK ++ const struct cred *cred = current_cred(); ++ ++ if (grsec_enable_link && S_ISLNK(inode->i_mode) && ++ (parent->i_mode & S_ISVTX) && (parent->i_uid != inode->i_uid) && ++ (parent->i_mode & S_IWOTH) && (cred->fsuid != inode->i_uid)) { ++ gr_log_fs_int2(GR_DONT_AUDIT, GR_SYMLINK_MSG, dentry, mnt, inode->i_uid, inode->i_gid); ++ return -EACCES; ++ } ++#endif ++ return 0; ++} ++ ++int ++gr_handle_hardlink(const struct dentry *dentry, ++ const struct vfsmount *mnt, ++ struct inode *inode, const int mode, const char *to) ++{ ++#ifdef CONFIG_GRKERNSEC_LINK ++ const struct cred *cred = current_cred(); ++ ++ if (grsec_enable_link && cred->fsuid != inode->i_uid && ++ (!S_ISREG(mode) || (mode & S_ISUID) || ++ ((mode & (S_ISGID | S_IXGRP)) == (S_ISGID | S_IXGRP)) || ++ (inode_permission(inode, MAY_READ | MAY_WRITE))) && ++ !capable(CAP_FOWNER) && cred->uid) { ++ gr_log_fs_int2_str(GR_DONT_AUDIT, GR_HARDLINK_MSG, dentry, mnt, inode->i_uid, inode->i_gid, to); ++ return -EPERM; ++ } ++#endif ++ return 0; ++} +diff -urNp linux-2.6.38.3/grsecurity/grsec_log.c linux-2.6.38.3/grsecurity/grsec_log.c +--- linux-2.6.38.3/grsecurity/grsec_log.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/grsec_log.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,310 @@ ++#include <linux/kernel.h> ++#include <linux/sched.h> ++#include <linux/file.h> ++#include <linux/tty.h> ++#include <linux/fs.h> ++#include <linux/grinternal.h> ++ ++#ifdef CONFIG_TREE_PREEMPT_RCU ++#define DISABLE_PREEMPT() preempt_disable() ++#define ENABLE_PREEMPT() preempt_enable() ++#else ++#define DISABLE_PREEMPT() ++#define ENABLE_PREEMPT() ++#endif ++ ++#define BEGIN_LOCKS(x) \ ++ DISABLE_PREEMPT(); \ ++ rcu_read_lock(); \ ++ read_lock(&tasklist_lock); \ ++ read_lock(&grsec_exec_file_lock); \ ++ if (x != GR_DO_AUDIT) \ ++ spin_lock(&grsec_alert_lock); \ ++ else \ ++ spin_lock(&grsec_audit_lock) ++ ++#define END_LOCKS(x) \ ++ if (x != GR_DO_AUDIT) \ ++ spin_unlock(&grsec_alert_lock); \ ++ else \ ++ spin_unlock(&grsec_audit_lock); \ ++ read_unlock(&grsec_exec_file_lock); \ ++ read_unlock(&tasklist_lock); \ ++ rcu_read_unlock(); \ ++ ENABLE_PREEMPT(); \ ++ if (x == GR_DONT_AUDIT) \ ++ gr_handle_alertkill(current) ++ ++enum { ++ FLOODING, ++ NO_FLOODING ++}; ++ ++extern char *gr_alert_log_fmt; ++extern char *gr_audit_log_fmt; ++extern char *gr_alert_log_buf; ++extern char *gr_audit_log_buf; ++ ++static int gr_log_start(int audit) ++{ ++ char *loglevel = (audit == GR_DO_AUDIT) ? KERN_INFO : KERN_ALERT; ++ char *fmt = (audit == GR_DO_AUDIT) ? gr_audit_log_fmt : gr_alert_log_fmt; ++ char *buf = (audit == GR_DO_AUDIT) ? gr_audit_log_buf : gr_alert_log_buf; ++ ++ if (audit == GR_DO_AUDIT) ++ goto set_fmt; ++ ++ if (!grsec_alert_wtime || jiffies - grsec_alert_wtime > CONFIG_GRKERNSEC_FLOODTIME * HZ) { ++ grsec_alert_wtime = jiffies; ++ grsec_alert_fyet = 0; ++ } else if ((jiffies - grsec_alert_wtime < CONFIG_GRKERNSEC_FLOODTIME * HZ) && (grsec_alert_fyet < CONFIG_GRKERNSEC_FLOODBURST)) { ++ grsec_alert_fyet++; ++ } else if (grsec_alert_fyet == CONFIG_GRKERNSEC_FLOODBURST) { ++ grsec_alert_wtime = jiffies; ++ grsec_alert_fyet++; ++ printk(KERN_ALERT "grsec: more alerts, logging disabled for %d seconds\n", CONFIG_GRKERNSEC_FLOODTIME); ++ return FLOODING; ++ } else return FLOODING; ++ ++set_fmt: ++ memset(buf, 0, PAGE_SIZE); ++ if (current->signal->curr_ip && gr_acl_is_enabled()) { ++ sprintf(fmt, "%s%s", loglevel, "grsec: From %pI4: (%.64s:%c:%.950s) "); ++ snprintf(buf, PAGE_SIZE - 1, fmt, ¤t->signal->curr_ip, current->role->rolename, gr_roletype_to_char(), current->acl->filename); ++ } else if (current->signal->curr_ip) { ++ sprintf(fmt, "%s%s", loglevel, "grsec: From %pI4: "); ++ snprintf(buf, PAGE_SIZE - 1, fmt, ¤t->signal->curr_ip); ++ } else if (gr_acl_is_enabled()) { ++ sprintf(fmt, "%s%s", loglevel, "grsec: (%.64s:%c:%.950s) "); ++ snprintf(buf, PAGE_SIZE - 1, fmt, current->role->rolename, gr_roletype_to_char(), current->acl->filename); ++ } else { ++ sprintf(fmt, "%s%s", loglevel, "grsec: "); ++ strcpy(buf, fmt); ++ } ++ ++ return NO_FLOODING; ++} ++ ++static void gr_log_middle(int audit, const char *msg, va_list ap) ++ __attribute__ ((format (printf, 2, 0))); ++ ++static void gr_log_middle(int audit, const char *msg, va_list ap) ++{ ++ char *buf = (audit == GR_DO_AUDIT) ? gr_audit_log_buf : gr_alert_log_buf; ++ unsigned int len = strlen(buf); ++ ++ vsnprintf(buf + len, PAGE_SIZE - len - 1, msg, ap); ++ ++ return; ++} ++ ++static void gr_log_middle_varargs(int audit, const char *msg, ...) ++ __attribute__ ((format (printf, 2, 3))); ++ ++static void gr_log_middle_varargs(int audit, const char *msg, ...) ++{ ++ char *buf = (audit == GR_DO_AUDIT) ? gr_audit_log_buf : gr_alert_log_buf; ++ unsigned int len = strlen(buf); ++ va_list ap; ++ ++ va_start(ap, msg); ++ vsnprintf(buf + len, PAGE_SIZE - len - 1, msg, ap); ++ va_end(ap); ++ ++ return; ++} ++ ++static void gr_log_end(int audit) ++{ ++ char *buf = (audit == GR_DO_AUDIT) ? gr_audit_log_buf : gr_alert_log_buf; ++ unsigned int len = strlen(buf); ++ ++ snprintf(buf + len, PAGE_SIZE - len - 1, DEFAULTSECMSG, DEFAULTSECARGS(current, current_cred(), __task_cred(current->real_parent))); ++ printk("%s\n", buf); ++ ++ return; ++} ++ ++void gr_log_varargs(int audit, const char *msg, int argtypes, ...) ++{ ++ int logtype; ++ char *result = (audit == GR_DO_AUDIT) ? "successful" : "denied"; ++ char *str1, *str2, *str3; ++ void *voidptr; ++ int num1, num2; ++ unsigned long ulong1, ulong2; ++ struct dentry *dentry; ++ struct vfsmount *mnt; ++ struct file *file; ++ struct task_struct *task; ++ const struct cred *cred, *pcred; ++ va_list ap; ++ ++ BEGIN_LOCKS(audit); ++ logtype = gr_log_start(audit); ++ if (logtype == FLOODING) { ++ END_LOCKS(audit); ++ return; ++ } ++ va_start(ap, argtypes); ++ switch (argtypes) { ++ case GR_TTYSNIFF: ++ task = va_arg(ap, struct task_struct *); ++ gr_log_middle_varargs(audit, msg, &task->signal->curr_ip, gr_task_fullpath0(task), task->comm, task->pid, gr_parent_task_fullpath0(task), task->real_parent->comm, task->real_parent->pid); ++ break; ++ case GR_SYSCTL_HIDDEN: ++ str1 = va_arg(ap, char *); ++ gr_log_middle_varargs(audit, msg, result, str1); ++ break; ++ case GR_RBAC: ++ dentry = va_arg(ap, struct dentry *); ++ mnt = va_arg(ap, struct vfsmount *); ++ gr_log_middle_varargs(audit, msg, result, gr_to_filename(dentry, mnt)); ++ break; ++ case GR_RBAC_STR: ++ dentry = va_arg(ap, struct dentry *); ++ mnt = va_arg(ap, struct vfsmount *); ++ str1 = va_arg(ap, char *); ++ gr_log_middle_varargs(audit, msg, result, gr_to_filename(dentry, mnt), str1); ++ break; ++ case GR_STR_RBAC: ++ str1 = va_arg(ap, char *); ++ dentry = va_arg(ap, struct dentry *); ++ mnt = va_arg(ap, struct vfsmount *); ++ gr_log_middle_varargs(audit, msg, result, str1, gr_to_filename(dentry, mnt)); ++ break; ++ case GR_RBAC_MODE2: ++ dentry = va_arg(ap, struct dentry *); ++ mnt = va_arg(ap, struct vfsmount *); ++ str1 = va_arg(ap, char *); ++ str2 = va_arg(ap, char *); ++ gr_log_middle_varargs(audit, msg, result, gr_to_filename(dentry, mnt), str1, str2); ++ break; ++ case GR_RBAC_MODE3: ++ dentry = va_arg(ap, struct dentry *); ++ mnt = va_arg(ap, struct vfsmount *); ++ str1 = va_arg(ap, char *); ++ str2 = va_arg(ap, char *); ++ str3 = va_arg(ap, char *); ++ gr_log_middle_varargs(audit, msg, result, gr_to_filename(dentry, mnt), str1, str2, str3); ++ break; ++ case GR_FILENAME: ++ dentry = va_arg(ap, struct dentry *); ++ mnt = va_arg(ap, struct vfsmount *); ++ gr_log_middle_varargs(audit, msg, gr_to_filename(dentry, mnt)); ++ break; ++ case GR_STR_FILENAME: ++ str1 = va_arg(ap, char *); ++ dentry = va_arg(ap, struct dentry *); ++ mnt = va_arg(ap, struct vfsmount *); ++ gr_log_middle_varargs(audit, msg, str1, gr_to_filename(dentry, mnt)); ++ break; ++ case GR_FILENAME_STR: ++ dentry = va_arg(ap, struct dentry *); ++ mnt = va_arg(ap, struct vfsmount *); ++ str1 = va_arg(ap, char *); ++ gr_log_middle_varargs(audit, msg, gr_to_filename(dentry, mnt), str1); ++ break; ++ case GR_FILENAME_TWO_INT: ++ dentry = va_arg(ap, struct dentry *); ++ mnt = va_arg(ap, struct vfsmount *); ++ num1 = va_arg(ap, int); ++ num2 = va_arg(ap, int); ++ gr_log_middle_varargs(audit, msg, gr_to_filename(dentry, mnt), num1, num2); ++ break; ++ case GR_FILENAME_TWO_INT_STR: ++ dentry = va_arg(ap, struct dentry *); ++ mnt = va_arg(ap, struct vfsmount *); ++ num1 = va_arg(ap, int); ++ num2 = va_arg(ap, int); ++ str1 = va_arg(ap, char *); ++ gr_log_middle_varargs(audit, msg, gr_to_filename(dentry, mnt), num1, num2, str1); ++ break; ++ case GR_TEXTREL: ++ file = va_arg(ap, struct file *); ++ ulong1 = va_arg(ap, unsigned long); ++ ulong2 = va_arg(ap, unsigned long); ++ gr_log_middle_varargs(audit, msg, file ? gr_to_filename(file->f_path.dentry, file->f_path.mnt) : "<anonymous mapping>", ulong1, ulong2); ++ break; ++ case GR_PTRACE: ++ task = va_arg(ap, struct task_struct *); ++ gr_log_middle_varargs(audit, msg, task->exec_file ? gr_to_filename(task->exec_file->f_path.dentry, task->exec_file->f_path.mnt) : "(none)", task->comm, task->pid); ++ break; ++ case GR_RESOURCE: ++ task = va_arg(ap, struct task_struct *); ++ cred = __task_cred(task); ++ pcred = __task_cred(task->real_parent); ++ ulong1 = va_arg(ap, unsigned long); ++ str1 = va_arg(ap, char *); ++ ulong2 = va_arg(ap, unsigned long); ++ gr_log_middle_varargs(audit, msg, ulong1, str1, ulong2, gr_task_fullpath(task), task->comm, task->pid, cred->uid, cred->euid, cred->gid, cred->egid, gr_parent_task_fullpath(task), task->real_parent->comm, task->real_parent->pid, pcred->uid, pcred->euid, pcred->gid, pcred->egid); ++ break; ++ case GR_CAP: ++ task = va_arg(ap, struct task_struct *); ++ cred = __task_cred(task); ++ pcred = __task_cred(task->real_parent); ++ str1 = va_arg(ap, char *); ++ gr_log_middle_varargs(audit, msg, str1, gr_task_fullpath(task), task->comm, task->pid, cred->uid, cred->euid, cred->gid, cred->egid, gr_parent_task_fullpath(task), task->real_parent->comm, task->real_parent->pid, pcred->uid, pcred->euid, pcred->gid, pcred->egid); ++ break; ++ case GR_SIG: ++ str1 = va_arg(ap, char *); ++ voidptr = va_arg(ap, void *); ++ gr_log_middle_varargs(audit, msg, str1, voidptr); ++ break; ++ case GR_SIG2: ++ task = va_arg(ap, struct task_struct *); ++ cred = __task_cred(task); ++ pcred = __task_cred(task->real_parent); ++ num1 = va_arg(ap, int); ++ gr_log_middle_varargs(audit, msg, num1, gr_task_fullpath0(task), task->comm, task->pid, cred->uid, cred->euid, cred->gid, cred->egid, gr_parent_task_fullpath0(task), task->real_parent->comm, task->real_parent->pid, pcred->uid, pcred->euid, pcred->gid, pcred->egid); ++ break; ++ case GR_CRASH1: ++ task = va_arg(ap, struct task_struct *); ++ cred = __task_cred(task); ++ pcred = __task_cred(task->real_parent); ++ ulong1 = va_arg(ap, unsigned long); ++ gr_log_middle_varargs(audit, msg, gr_task_fullpath(task), task->comm, task->pid, cred->uid, cred->euid, cred->gid, cred->egid, gr_parent_task_fullpath(task), task->real_parent->comm, task->real_parent->pid, pcred->uid, pcred->euid, pcred->gid, pcred->egid, cred->uid, ulong1); ++ break; ++ case GR_CRASH2: ++ task = va_arg(ap, struct task_struct *); ++ cred = __task_cred(task); ++ pcred = __task_cred(task->real_parent); ++ ulong1 = va_arg(ap, unsigned long); ++ gr_log_middle_varargs(audit, msg, gr_task_fullpath(task), task->comm, task->pid, cred->uid, cred->euid, cred->gid, cred->egid, gr_parent_task_fullpath(task), task->real_parent->comm, task->real_parent->pid, pcred->uid, pcred->euid, pcred->gid, pcred->egid, ulong1); ++ break; ++ case GR_RWXMAP: ++ file = va_arg(ap, struct file *); ++ gr_log_middle_varargs(audit, msg, file ? gr_to_filename(file->f_path.dentry, file->f_path.mnt) : "<anonymous mapping>"); ++ break; ++ case GR_PSACCT: ++ { ++ unsigned int wday, cday; ++ __u8 whr, chr; ++ __u8 wmin, cmin; ++ __u8 wsec, csec; ++ char cur_tty[64] = { 0 }; ++ char parent_tty[64] = { 0 }; ++ ++ task = va_arg(ap, struct task_struct *); ++ wday = va_arg(ap, unsigned int); ++ cday = va_arg(ap, unsigned int); ++ whr = va_arg(ap, int); ++ chr = va_arg(ap, int); ++ wmin = va_arg(ap, int); ++ cmin = va_arg(ap, int); ++ wsec = va_arg(ap, int); ++ csec = va_arg(ap, int); ++ ulong1 = va_arg(ap, unsigned long); ++ cred = __task_cred(task); ++ pcred = __task_cred(task->real_parent); ++ ++ gr_log_middle_varargs(audit, msg, gr_task_fullpath(task), task->comm, task->pid, &task->signal->curr_ip, tty_name(task->signal->tty, cur_tty), cred->uid, cred->euid, cred->gid, cred->egid, wday, whr, wmin, wsec, cday, chr, cmin, csec, (task->flags & PF_SIGNALED) ? "killed by signal" : "exited", ulong1, gr_parent_task_fullpath(task), task->real_parent->comm, task->real_parent->pid, &task->real_parent->signal->curr_ip, tty_name(task->real_parent->signal->tty, parent_tty), pcred->uid, pcred->euid, pcred->gid, pcred->egid); ++ } ++ break; ++ default: ++ gr_log_middle(audit, msg, ap); ++ } ++ va_end(ap); ++ gr_log_end(audit); ++ END_LOCKS(audit); ++} +diff -urNp linux-2.6.38.3/grsecurity/grsec_mem.c linux-2.6.38.3/grsecurity/grsec_mem.c +--- linux-2.6.38.3/grsecurity/grsec_mem.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/grsec_mem.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,33 @@ ++#include <linux/kernel.h> ++#include <linux/sched.h> ++#include <linux/mm.h> ++#include <linux/mman.h> ++#include <linux/grinternal.h> ++ ++void ++gr_handle_ioperm(void) ++{ ++ gr_log_noargs(GR_DONT_AUDIT, GR_IOPERM_MSG); ++ return; ++} ++ ++void ++gr_handle_iopl(void) ++{ ++ gr_log_noargs(GR_DONT_AUDIT, GR_IOPL_MSG); ++ return; ++} ++ ++void ++gr_handle_mem_readwrite(u64 from, u64 to) ++{ ++ gr_log_two_u64(GR_DONT_AUDIT, GR_MEM_READWRITE_MSG, from, to); ++ return; ++} ++ ++void ++gr_handle_vm86(void) ++{ ++ gr_log_noargs(GR_DONT_AUDIT, GR_VM86_MSG); ++ return; ++} +diff -urNp linux-2.6.38.3/grsecurity/grsec_mount.c linux-2.6.38.3/grsecurity/grsec_mount.c +--- linux-2.6.38.3/grsecurity/grsec_mount.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/grsec_mount.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,62 @@ ++#include <linux/kernel.h> ++#include <linux/sched.h> ++#include <linux/mount.h> ++#include <linux/grsecurity.h> ++#include <linux/grinternal.h> ++ ++void ++gr_log_remount(const char *devname, const int retval) ++{ ++#ifdef CONFIG_GRKERNSEC_AUDIT_MOUNT ++ if (grsec_enable_mount && (retval >= 0)) ++ gr_log_str(GR_DO_AUDIT, GR_REMOUNT_AUDIT_MSG, devname ? devname : "none"); ++#endif ++ return; ++} ++ ++void ++gr_log_unmount(const char *devname, const int retval) ++{ ++#ifdef CONFIG_GRKERNSEC_AUDIT_MOUNT ++ if (grsec_enable_mount && (retval >= 0)) ++ gr_log_str(GR_DO_AUDIT, GR_UNMOUNT_AUDIT_MSG, devname ? devname : "none"); ++#endif ++ return; ++} ++ ++void ++gr_log_mount(const char *from, const char *to, const int retval) ++{ ++#ifdef CONFIG_GRKERNSEC_AUDIT_MOUNT ++ if (grsec_enable_mount && (retval >= 0)) ++ gr_log_str_str(GR_DO_AUDIT, GR_MOUNT_AUDIT_MSG, from, to); ++#endif ++ return; ++} ++ ++int ++gr_handle_rofs_mount(struct dentry *dentry, struct vfsmount *mnt, int mnt_flags) ++{ ++#ifdef CONFIG_GRKERNSEC_ROFS ++ if (grsec_enable_rofs && !(mnt_flags & MNT_READONLY)) { ++ gr_log_fs_generic(GR_DO_AUDIT, GR_ROFS_MOUNT_MSG, dentry, mnt); ++ return -EPERM; ++ } else ++ return 0; ++#endif ++ return 0; ++} ++ ++int ++gr_handle_rofs_blockwrite(struct dentry *dentry, struct vfsmount *mnt, int acc_mode) ++{ ++#ifdef CONFIG_GRKERNSEC_ROFS ++ if (grsec_enable_rofs && (acc_mode & MAY_WRITE) && ++ dentry->d_inode && S_ISBLK(dentry->d_inode->i_mode)) { ++ gr_log_fs_generic(GR_DO_AUDIT, GR_ROFS_BLOCKWRITE_MSG, dentry, mnt); ++ return -EPERM; ++ } else ++ return 0; ++#endif ++ return 0; ++} +diff -urNp linux-2.6.38.3/grsecurity/grsec_pax.c linux-2.6.38.3/grsecurity/grsec_pax.c +--- linux-2.6.38.3/grsecurity/grsec_pax.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/grsec_pax.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,36 @@ ++#include <linux/kernel.h> ++#include <linux/sched.h> ++#include <linux/mm.h> ++#include <linux/file.h> ++#include <linux/grinternal.h> ++#include <linux/grsecurity.h> ++ ++void ++gr_log_textrel(struct vm_area_struct * vma) ++{ ++#ifdef CONFIG_GRKERNSEC_AUDIT_TEXTREL ++ if (grsec_enable_audit_textrel) ++ gr_log_textrel_ulong_ulong(GR_DO_AUDIT, GR_TEXTREL_AUDIT_MSG, vma->vm_file, vma->vm_start, vma->vm_pgoff); ++#endif ++ return; ++} ++ ++void ++gr_log_rwxmmap(struct file *file) ++{ ++#ifdef CONFIG_GRKERNSEC_RWXMAP_LOG ++ if (grsec_enable_log_rwxmaps) ++ gr_log_rwxmap(GR_DONT_AUDIT, GR_RWXMMAP_MSG, file); ++#endif ++ return; ++} ++ ++void ++gr_log_rwxmprotect(struct file *file) ++{ ++#ifdef CONFIG_GRKERNSEC_RWXMAP_LOG ++ if (grsec_enable_log_rwxmaps) ++ gr_log_rwxmap(GR_DONT_AUDIT, GR_RWXMPROTECT_MSG, file); ++#endif ++ return; ++} +diff -urNp linux-2.6.38.3/grsecurity/grsec_ptrace.c linux-2.6.38.3/grsecurity/grsec_ptrace.c +--- linux-2.6.38.3/grsecurity/grsec_ptrace.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/grsec_ptrace.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,14 @@ ++#include <linux/kernel.h> ++#include <linux/sched.h> ++#include <linux/grinternal.h> ++#include <linux/grsecurity.h> ++ ++void ++gr_audit_ptrace(struct task_struct *task) ++{ ++#ifdef CONFIG_GRKERNSEC_AUDIT_PTRACE ++ if (grsec_enable_audit_ptrace) ++ gr_log_ptrace(GR_DO_AUDIT, GR_PTRACE_AUDIT_MSG, task); ++#endif ++ return; ++} +diff -urNp linux-2.6.38.3/grsecurity/grsec_sig.c linux-2.6.38.3/grsecurity/grsec_sig.c +--- linux-2.6.38.3/grsecurity/grsec_sig.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/grsec_sig.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,197 @@ ++#include <linux/kernel.h> ++#include <linux/sched.h> ++#include <linux/delay.h> ++#include <linux/grsecurity.h> ++#include <linux/grinternal.h> ++#include <linux/hardirq.h> ++ ++char *signames[] = { ++ [SIGSEGV] = "Segmentation fault", ++ [SIGILL] = "Illegal instruction", ++ [SIGABRT] = "Abort", ++ [SIGBUS] = "Invalid alignment/Bus error" ++}; ++ ++void ++gr_log_signal(const int sig, const void *addr, const struct task_struct *t) ++{ ++#ifdef CONFIG_GRKERNSEC_SIGNAL ++ if (grsec_enable_signal && ((sig == SIGSEGV) || (sig == SIGILL) || ++ (sig == SIGABRT) || (sig == SIGBUS))) { ++ if (t->pid == current->pid) { ++ gr_log_sig_addr(GR_DONT_AUDIT_GOOD, GR_UNISIGLOG_MSG, signames[sig], addr); ++ } else { ++ gr_log_sig_task(GR_DONT_AUDIT_GOOD, GR_DUALSIGLOG_MSG, t, sig); ++ } ++ } ++#endif ++ return; ++} ++ ++int ++gr_handle_signal(const struct task_struct *p, const int sig) ++{ ++#ifdef CONFIG_GRKERNSEC ++ if (current->pid > 1 && gr_check_protected_task(p)) { ++ gr_log_sig_task(GR_DONT_AUDIT, GR_SIG_ACL_MSG, p, sig); ++ return -EPERM; ++ } else if (gr_pid_is_chrooted((struct task_struct *)p)) { ++ return -EPERM; ++ } ++#endif ++ return 0; ++} ++ ++#ifdef CONFIG_GRKERNSEC ++extern int specific_send_sig_info(int sig, struct siginfo *info, struct task_struct *t); ++ ++int gr_fake_force_sig(int sig, struct task_struct *t) ++{ ++ unsigned long int flags; ++ int ret, blocked, ignored; ++ struct k_sigaction *action; ++ ++ spin_lock_irqsave(&t->sighand->siglock, flags); ++ action = &t->sighand->action[sig-1]; ++ ignored = action->sa.sa_handler == SIG_IGN; ++ blocked = sigismember(&t->blocked, sig); ++ if (blocked || ignored) { ++ action->sa.sa_handler = SIG_DFL; ++ if (blocked) { ++ sigdelset(&t->blocked, sig); ++ recalc_sigpending_and_wake(t); ++ } ++ } ++ if (action->sa.sa_handler == SIG_DFL) ++ t->signal->flags &= ~SIGNAL_UNKILLABLE; ++ ret = specific_send_sig_info(sig, SEND_SIG_PRIV, t); ++ ++ spin_unlock_irqrestore(&t->sighand->siglock, flags); ++ ++ return ret; ++} ++#endif ++ ++#ifdef CONFIG_GRKERNSEC_BRUTE ++#define GR_USER_BAN_TIME (15 * 60) ++ ++static int __get_dumpable(unsigned long mm_flags) ++{ ++ int ret; ++ ++ ret = mm_flags & MMF_DUMPABLE_MASK; ++ return (ret >= 2) ? 2 : ret; ++} ++#endif ++ ++void gr_handle_brute_attach(struct task_struct *p, unsigned long mm_flags) ++{ ++#ifdef CONFIG_GRKERNSEC_BRUTE ++ uid_t uid = 0; ++ ++ rcu_read_lock(); ++ read_lock(&tasklist_lock); ++ read_lock(&grsec_exec_file_lock); ++ if (p->real_parent && p->real_parent->exec_file == p->exec_file) ++ p->real_parent->brute = 1; ++ else { ++ const struct cred *cred = __task_cred(p), *cred2; ++ struct task_struct *tsk, *tsk2; ++ ++ if (!__get_dumpable(mm_flags) && cred->uid) { ++ struct user_struct *user; ++ ++ uid = cred->uid; ++ ++ /* this is put upon execution past expiration */ ++ user = find_user(uid); ++ if (user == NULL) ++ goto unlock; ++ user->banned = 1; ++ user->ban_expires = get_seconds() + GR_USER_BAN_TIME; ++ if (user->ban_expires == ~0UL) ++ user->ban_expires--; ++ ++ do_each_thread(tsk2, tsk) { ++ cred2 = __task_cred(tsk); ++ if (tsk != p && cred2->uid == uid) ++ gr_fake_force_sig(SIGKILL, tsk); ++ } while_each_thread(tsk2, tsk); ++ } ++ } ++unlock: ++ read_unlock(&grsec_exec_file_lock); ++ read_unlock(&tasklist_lock); ++ rcu_read_unlock(); ++ ++ if (uid) ++ printk(KERN_ALERT "grsec: bruteforce prevention initiated against uid %u, banning for %d minutes\n", uid, GR_USER_BAN_TIME / 60); ++ ++#endif ++ return; ++} ++ ++void gr_handle_brute_check(void) ++{ ++#ifdef CONFIG_GRKERNSEC_BRUTE ++ if (current->brute) ++ msleep(30 * 1000); ++#endif ++ return; ++} ++ ++void gr_handle_kernel_exploit(void) ++{ ++#ifdef CONFIG_GRKERNSEC_KERN_LOCKOUT ++ const struct cred *cred; ++ struct task_struct *tsk, *tsk2; ++ struct user_struct *user; ++ uid_t uid; ++ ++ if (in_irq() || in_serving_softirq() || in_nmi()) ++ panic("grsec: halting the system due to suspicious kernel crash caused in interrupt context"); ++ ++ uid = current_uid(); ++ ++ if (uid == 0) ++ panic("grsec: halting the system due to suspicious kernel crash caused by root"); ++ else { ++ /* kill all the processes of this user, hold a reference ++ to their creds struct, and prevent them from creating ++ another process until system reset ++ */ ++ printk(KERN_ALERT "grsec: banning user with uid %u until system restart for suspicious kernel crash\n", uid); ++ /* we intentionally leak this ref */ ++ user = get_uid(current->cred->user); ++ if (user) { ++ user->banned = 1; ++ user->ban_expires = ~0UL; ++ } ++ ++ read_lock(&tasklist_lock); ++ do_each_thread(tsk2, tsk) { ++ cred = __task_cred(tsk); ++ if (cred->uid == uid) ++ gr_fake_force_sig(SIGKILL, tsk); ++ } while_each_thread(tsk2, tsk); ++ read_unlock(&tasklist_lock); ++ } ++#endif ++} ++ ++int gr_process_user_ban(void) ++{ ++#if defined(CONFIG_GRKERNSEC_KERN_LOCKOUT) || defined(CONFIG_GRKERNSEC_BRUTE) ++ if (unlikely(current->cred->user->banned)) { ++ struct user_struct *user = current->cred->user; ++ if (user->ban_expires != ~0UL && time_after_eq(get_seconds(), user->ban_expires)) { ++ user->banned = 0; ++ user->ban_expires = 0; ++ free_uid(user); ++ } else ++ return -EPERM; ++ } ++#endif ++ return 0; ++ ++} +diff -urNp linux-2.6.38.3/grsecurity/grsec_sock.c linux-2.6.38.3/grsecurity/grsec_sock.c +--- linux-2.6.38.3/grsecurity/grsec_sock.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/grsec_sock.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,275 @@ ++#include <linux/kernel.h> ++#include <linux/module.h> ++#include <linux/sched.h> ++#include <linux/file.h> ++#include <linux/net.h> ++#include <linux/in.h> ++#include <linux/ip.h> ++#include <net/sock.h> ++#include <net/inet_sock.h> ++#include <linux/grsecurity.h> ++#include <linux/grinternal.h> ++#include <linux/gracl.h> ++ ++kernel_cap_t gr_cap_rtnetlink(struct sock *sock); ++EXPORT_SYMBOL(gr_cap_rtnetlink); ++ ++extern int gr_search_udp_recvmsg(const struct sock *sk, const struct sk_buff *skb); ++extern int gr_search_udp_sendmsg(const struct sock *sk, const struct sockaddr_in *addr); ++ ++EXPORT_SYMBOL(gr_search_udp_recvmsg); ++EXPORT_SYMBOL(gr_search_udp_sendmsg); ++ ++#ifdef CONFIG_UNIX_MODULE ++EXPORT_SYMBOL(gr_acl_handle_unix); ++EXPORT_SYMBOL(gr_acl_handle_mknod); ++EXPORT_SYMBOL(gr_handle_chroot_unix); ++EXPORT_SYMBOL(gr_handle_create); ++#endif ++ ++#ifdef CONFIG_GRKERNSEC ++#define gr_conn_table_size 32749 ++struct conn_table_entry { ++ struct conn_table_entry *next; ++ struct signal_struct *sig; ++}; ++ ++struct conn_table_entry *gr_conn_table[gr_conn_table_size]; ++DEFINE_SPINLOCK(gr_conn_table_lock); ++ ++extern const char * gr_socktype_to_name(unsigned char type); ++extern const char * gr_proto_to_name(unsigned char proto); ++extern const char * gr_sockfamily_to_name(unsigned char family); ++ ++static __inline__ int ++conn_hash(__u32 saddr, __u32 daddr, __u16 sport, __u16 dport, unsigned int size) ++{ ++ return ((daddr + saddr + (sport << 8) + (dport << 16)) % size); ++} ++ ++static __inline__ int ++conn_match(const struct signal_struct *sig, __u32 saddr, __u32 daddr, ++ __u16 sport, __u16 dport) ++{ ++ if (unlikely(sig->gr_saddr == saddr && sig->gr_daddr == daddr && ++ sig->gr_sport == sport && sig->gr_dport == dport)) ++ return 1; ++ else ++ return 0; ++} ++ ++static void gr_add_to_task_ip_table_nolock(struct signal_struct *sig, struct conn_table_entry *newent) ++{ ++ struct conn_table_entry **match; ++ unsigned int index; ++ ++ index = conn_hash(sig->gr_saddr, sig->gr_daddr, ++ sig->gr_sport, sig->gr_dport, ++ gr_conn_table_size); ++ ++ newent->sig = sig; ++ ++ match = &gr_conn_table[index]; ++ newent->next = *match; ++ *match = newent; ++ ++ return; ++} ++ ++static void gr_del_task_from_ip_table_nolock(struct signal_struct *sig) ++{ ++ struct conn_table_entry *match, *last = NULL; ++ unsigned int index; ++ ++ index = conn_hash(sig->gr_saddr, sig->gr_daddr, ++ sig->gr_sport, sig->gr_dport, ++ gr_conn_table_size); ++ ++ match = gr_conn_table[index]; ++ while (match && !conn_match(match->sig, ++ sig->gr_saddr, sig->gr_daddr, sig->gr_sport, ++ sig->gr_dport)) { ++ last = match; ++ match = match->next; ++ } ++ ++ if (match) { ++ if (last) ++ last->next = match->next; ++ else ++ gr_conn_table[index] = NULL; ++ kfree(match); ++ } ++ ++ return; ++} ++ ++static struct signal_struct * gr_lookup_task_ip_table(__u32 saddr, __u32 daddr, ++ __u16 sport, __u16 dport) ++{ ++ struct conn_table_entry *match; ++ unsigned int index; ++ ++ index = conn_hash(saddr, daddr, sport, dport, gr_conn_table_size); ++ ++ match = gr_conn_table[index]; ++ while (match && !conn_match(match->sig, saddr, daddr, sport, dport)) ++ match = match->next; ++ ++ if (match) ++ return match->sig; ++ else ++ return NULL; ++} ++ ++#endif ++ ++void gr_update_task_in_ip_table(struct task_struct *task, const struct inet_sock *inet) ++{ ++#ifdef CONFIG_GRKERNSEC ++ struct signal_struct *sig = task->signal; ++ struct conn_table_entry *newent; ++ ++ newent = kmalloc(sizeof(struct conn_table_entry), GFP_ATOMIC); ++ if (newent == NULL) ++ return; ++ /* no bh lock needed since we are called with bh disabled */ ++ spin_lock(&gr_conn_table_lock); ++ gr_del_task_from_ip_table_nolock(sig); ++ sig->gr_saddr = inet->inet_rcv_saddr; ++ sig->gr_daddr = inet->inet_daddr; ++ sig->gr_sport = inet->inet_sport; ++ sig->gr_dport = inet->inet_dport; ++ gr_add_to_task_ip_table_nolock(sig, newent); ++ spin_unlock(&gr_conn_table_lock); ++#endif ++ return; ++} ++ ++void gr_del_task_from_ip_table(struct task_struct *task) ++{ ++#ifdef CONFIG_GRKERNSEC ++ spin_lock_bh(&gr_conn_table_lock); ++ gr_del_task_from_ip_table_nolock(task->signal); ++ spin_unlock_bh(&gr_conn_table_lock); ++#endif ++ return; ++} ++ ++void ++gr_attach_curr_ip(const struct sock *sk) ++{ ++#ifdef CONFIG_GRKERNSEC ++ struct signal_struct *p, *set; ++ const struct inet_sock *inet = inet_sk(sk); ++ ++ if (unlikely(sk->sk_protocol != IPPROTO_TCP)) ++ return; ++ ++ set = current->signal; ++ ++ spin_lock_bh(&gr_conn_table_lock); ++ p = gr_lookup_task_ip_table(inet->inet_daddr, inet->inet_rcv_saddr, ++ inet->inet_dport, inet->inet_sport); ++ if (unlikely(p != NULL)) { ++ set->curr_ip = p->curr_ip; ++ set->used_accept = 1; ++ gr_del_task_from_ip_table_nolock(p); ++ spin_unlock_bh(&gr_conn_table_lock); ++ return; ++ } ++ spin_unlock_bh(&gr_conn_table_lock); ++ ++ set->curr_ip = inet->inet_daddr; ++ set->used_accept = 1; ++#endif ++ return; ++} ++ ++int ++gr_handle_sock_all(const int family, const int type, const int protocol) ++{ ++#ifdef CONFIG_GRKERNSEC_SOCKET_ALL ++ if (grsec_enable_socket_all && in_group_p(grsec_socket_all_gid) && ++ (family != AF_UNIX)) { ++ if (family == AF_INET) ++ gr_log_str3(GR_DONT_AUDIT, GR_SOCK_MSG, gr_sockfamily_to_name(family), gr_socktype_to_name(type), gr_proto_to_name(protocol)); ++ else ++ gr_log_str2_int(GR_DONT_AUDIT, GR_SOCK_NOINET_MSG, gr_sockfamily_to_name(family), gr_socktype_to_name(type), protocol); ++ return -EACCES; ++ } ++#endif ++ return 0; ++} ++ ++int ++gr_handle_sock_server(const struct sockaddr *sck) ++{ ++#ifdef CONFIG_GRKERNSEC_SOCKET_SERVER ++ if (grsec_enable_socket_server && ++ in_group_p(grsec_socket_server_gid) && ++ sck && (sck->sa_family != AF_UNIX) && ++ (sck->sa_family != AF_LOCAL)) { ++ gr_log_noargs(GR_DONT_AUDIT, GR_BIND_MSG); ++ return -EACCES; ++ } ++#endif ++ return 0; ++} ++ ++int ++gr_handle_sock_server_other(const struct sock *sck) ++{ ++#ifdef CONFIG_GRKERNSEC_SOCKET_SERVER ++ if (grsec_enable_socket_server && ++ in_group_p(grsec_socket_server_gid) && ++ sck && (sck->sk_family != AF_UNIX) && ++ (sck->sk_family != AF_LOCAL)) { ++ gr_log_noargs(GR_DONT_AUDIT, GR_BIND_MSG); ++ return -EACCES; ++ } ++#endif ++ return 0; ++} ++ ++int ++gr_handle_sock_client(const struct sockaddr *sck) ++{ ++#ifdef CONFIG_GRKERNSEC_SOCKET_CLIENT ++ if (grsec_enable_socket_client && in_group_p(grsec_socket_client_gid) && ++ sck && (sck->sa_family != AF_UNIX) && ++ (sck->sa_family != AF_LOCAL)) { ++ gr_log_noargs(GR_DONT_AUDIT, GR_CONNECT_MSG); ++ return -EACCES; ++ } ++#endif ++ return 0; ++} ++ ++kernel_cap_t ++gr_cap_rtnetlink(struct sock *sock) ++{ ++#ifdef CONFIG_GRKERNSEC ++ if (!gr_acl_is_enabled()) ++ return current_cap(); ++ else if (sock->sk_protocol == NETLINK_ISCSI && ++ cap_raised(current_cap(), CAP_SYS_ADMIN) && ++ gr_is_capable(CAP_SYS_ADMIN)) ++ return current_cap(); ++ else if (sock->sk_protocol == NETLINK_AUDIT && ++ cap_raised(current_cap(), CAP_AUDIT_WRITE) && ++ gr_is_capable(CAP_AUDIT_WRITE) && ++ cap_raised(current_cap(), CAP_AUDIT_CONTROL) && ++ gr_is_capable(CAP_AUDIT_CONTROL)) ++ return current_cap(); ++ else if (cap_raised(current_cap(), CAP_NET_ADMIN) && ++ ((sock->sk_protocol == NETLINK_ROUTE) ? ++ gr_is_capable_nolog(CAP_NET_ADMIN) : ++ gr_is_capable(CAP_NET_ADMIN))) ++ return current_cap(); ++ else ++ return __cap_empty_set; ++#else ++ return current_cap(); ++#endif ++} +diff -urNp linux-2.6.38.3/grsecurity/grsec_sysctl.c linux-2.6.38.3/grsecurity/grsec_sysctl.c +--- linux-2.6.38.3/grsecurity/grsec_sysctl.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/grsec_sysctl.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,433 @@ ++#include <linux/kernel.h> ++#include <linux/sched.h> ++#include <linux/sysctl.h> ++#include <linux/grsecurity.h> ++#include <linux/grinternal.h> ++ ++int ++gr_handle_sysctl_mod(const char *dirname, const char *name, const int op) ++{ ++#ifdef CONFIG_GRKERNSEC_SYSCTL ++ if (!strcmp(dirname, "grsecurity") && grsec_lock && (op & MAY_WRITE)) { ++ gr_log_str(GR_DONT_AUDIT, GR_SYSCTL_MSG, name); ++ return -EACCES; ++ } ++#endif ++ return 0; ++} ++ ++#ifdef CONFIG_GRKERNSEC_ROFS ++static int __maybe_unused one = 1; ++#endif ++ ++#if defined(CONFIG_GRKERNSEC_SYSCTL) || defined(CONFIG_GRKERNSEC_ROFS) ++struct ctl_table grsecurity_table[] = { ++#ifdef CONFIG_GRKERNSEC_SYSCTL ++#ifdef CONFIG_GRKERNSEC_SYSCTL_DISTRO ++#ifdef CONFIG_GRKERNSEC_IO ++ { ++ .procname = "disable_priv_io", ++ .data = &grsec_disable_privio, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#endif ++#ifdef CONFIG_GRKERNSEC_LINK ++ { ++ .procname = "linking_restrictions", ++ .data = &grsec_enable_link, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_FIFO ++ { ++ .procname = "fifo_restrictions", ++ .data = &grsec_enable_fifo, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_EXECVE ++ { ++ .procname = "execve_limiting", ++ .data = &grsec_enable_execve, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_BLACKHOLE ++ { ++ .procname = "ip_blackhole", ++ .data = &grsec_enable_blackhole, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++ { ++ .procname = "lastack_retries", ++ .data = &grsec_lastack_retries, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_EXECLOG ++ { ++ .procname = "exec_logging", ++ .data = &grsec_enable_execlog, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_RWXMAP_LOG ++ { ++ .procname = "rwxmap_logging", ++ .data = &grsec_enable_log_rwxmaps, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_SIGNAL ++ { ++ .procname = "signal_logging", ++ .data = &grsec_enable_signal, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_FORKFAIL ++ { ++ .procname = "forkfail_logging", ++ .data = &grsec_enable_forkfail, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_TIME ++ { ++ .procname = "timechange_logging", ++ .data = &grsec_enable_time, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_SHMAT ++ { ++ .procname = "chroot_deny_shmat", ++ .data = &grsec_enable_chroot_shmat, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_UNIX ++ { ++ .procname = "chroot_deny_unix", ++ .data = &grsec_enable_chroot_unix, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_MOUNT ++ { ++ .procname = "chroot_deny_mount", ++ .data = &grsec_enable_chroot_mount, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_FCHDIR ++ { ++ .procname = "chroot_deny_fchdir", ++ .data = &grsec_enable_chroot_fchdir, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_DOUBLE ++ { ++ .procname = "chroot_deny_chroot", ++ .data = &grsec_enable_chroot_double, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_PIVOT ++ { ++ .procname = "chroot_deny_pivot", ++ .data = &grsec_enable_chroot_pivot, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_CHDIR ++ { ++ .procname = "chroot_enforce_chdir", ++ .data = &grsec_enable_chroot_chdir, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_CHMOD ++ { ++ .procname = "chroot_deny_chmod", ++ .data = &grsec_enable_chroot_chmod, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_MKNOD ++ { ++ .procname = "chroot_deny_mknod", ++ .data = &grsec_enable_chroot_mknod, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_NICE ++ { ++ .procname = "chroot_restrict_nice", ++ .data = &grsec_enable_chroot_nice, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_EXECLOG ++ { ++ .procname = "chroot_execlog", ++ .data = &grsec_enable_chroot_execlog, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_CAPS ++ { ++ .procname = "chroot_caps", ++ .data = &grsec_enable_chroot_caps, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_SYSCTL ++ { ++ .procname = "chroot_deny_sysctl", ++ .data = &grsec_enable_chroot_sysctl, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_TPE ++ { ++ .procname = "tpe", ++ .data = &grsec_enable_tpe, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++ { ++ .procname = "tpe_gid", ++ .data = &grsec_tpe_gid, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_TPE_INVERT ++ { ++ .procname = "tpe_invert", ++ .data = &grsec_enable_tpe_invert, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_TPE_ALL ++ { ++ .procname = "tpe_restrict_all", ++ .data = &grsec_enable_tpe_all, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_SOCKET_ALL ++ { ++ .procname = "socket_all", ++ .data = &grsec_enable_socket_all, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++ { ++ .procname = "socket_all_gid", ++ .data = &grsec_socket_all_gid, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_SOCKET_CLIENT ++ { ++ .procname = "socket_client", ++ .data = &grsec_enable_socket_client, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++ { ++ .procname = "socket_client_gid", ++ .data = &grsec_socket_client_gid, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_SOCKET_SERVER ++ { ++ .procname = "socket_server", ++ .data = &grsec_enable_socket_server, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++ { ++ .procname = "socket_server_gid", ++ .data = &grsec_socket_server_gid, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_AUDIT_GROUP ++ { ++ .procname = "audit_group", ++ .data = &grsec_enable_group, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++ { ++ .procname = "audit_gid", ++ .data = &grsec_audit_gid, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_AUDIT_CHDIR ++ { ++ .procname = "audit_chdir", ++ .data = &grsec_enable_chdir, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_AUDIT_MOUNT ++ { ++ .procname = "audit_mount", ++ .data = &grsec_enable_mount, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_AUDIT_TEXTREL ++ { ++ .procname = "audit_textrel", ++ .data = &grsec_enable_audit_textrel, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_DMESG ++ { ++ .procname = "dmesg", ++ .data = &grsec_enable_dmesg, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_CHROOT_FINDTASK ++ { ++ .procname = "chroot_findtask", ++ .data = &grsec_enable_chroot_findtask, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_RESLOG ++ { ++ .procname = "resource_logging", ++ .data = &grsec_resource_logging, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_AUDIT_PTRACE ++ { ++ .procname = "audit_ptrace", ++ .data = &grsec_enable_audit_ptrace, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_HARDEN_PTRACE ++ { ++ .procname = "harden_ptrace", ++ .data = &grsec_enable_harden_ptrace, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++ { ++ .procname = "grsec_lock", ++ .data = &grsec_lock, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++#endif ++#ifdef CONFIG_GRKERNSEC_ROFS ++ { ++ .procname = "romount_protect", ++ .data = &grsec_enable_rofs, ++ .maxlen = sizeof(int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec_minmax, ++ .extra1 = &one, ++ .extra2 = &one, ++ }, ++#endif ++ { } ++}; ++#endif +diff -urNp linux-2.6.38.3/grsecurity/grsec_time.c linux-2.6.38.3/grsecurity/grsec_time.c +--- linux-2.6.38.3/grsecurity/grsec_time.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/grsec_time.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,16 @@ ++#include <linux/kernel.h> ++#include <linux/sched.h> ++#include <linux/grinternal.h> ++#include <linux/module.h> ++ ++void ++gr_log_timechange(void) ++{ ++#ifdef CONFIG_GRKERNSEC_TIME ++ if (grsec_enable_time) ++ gr_log_noargs(GR_DONT_AUDIT_GOOD, GR_TIME_MSG); ++#endif ++ return; ++} ++ ++EXPORT_SYMBOL(gr_log_timechange); +diff -urNp linux-2.6.38.3/grsecurity/grsec_tpe.c linux-2.6.38.3/grsecurity/grsec_tpe.c +--- linux-2.6.38.3/grsecurity/grsec_tpe.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/grsec_tpe.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,39 @@ ++#include <linux/kernel.h> ++#include <linux/sched.h> ++#include <linux/file.h> ++#include <linux/fs.h> ++#include <linux/grinternal.h> ++ ++extern int gr_acl_tpe_check(void); ++ ++int ++gr_tpe_allow(const struct file *file) ++{ ++#ifdef CONFIG_GRKERNSEC ++ struct inode *inode = file->f_path.dentry->d_parent->d_inode; ++ const struct cred *cred = current_cred(); ++ ++ if (cred->uid && ((grsec_enable_tpe && ++#ifdef CONFIG_GRKERNSEC_TPE_INVERT ++ ((grsec_enable_tpe_invert && !in_group_p(grsec_tpe_gid)) || ++ (!grsec_enable_tpe_invert && in_group_p(grsec_tpe_gid))) ++#else ++ in_group_p(grsec_tpe_gid) ++#endif ++ ) || gr_acl_tpe_check()) && ++ (inode->i_uid || (!inode->i_uid && ((inode->i_mode & S_IWGRP) || ++ (inode->i_mode & S_IWOTH))))) { ++ gr_log_fs_generic(GR_DONT_AUDIT, GR_EXEC_TPE_MSG, file->f_path.dentry, file->f_path.mnt); ++ return 0; ++ } ++#ifdef CONFIG_GRKERNSEC_TPE_ALL ++ if (cred->uid && grsec_enable_tpe && grsec_enable_tpe_all && ++ ((inode->i_uid && (inode->i_uid != cred->uid)) || ++ (inode->i_mode & S_IWGRP) || (inode->i_mode & S_IWOTH))) { ++ gr_log_fs_generic(GR_DONT_AUDIT, GR_EXEC_TPE_MSG, file->f_path.dentry, file->f_path.mnt); ++ return 0; ++ } ++#endif ++#endif ++ return 1; ++} +diff -urNp linux-2.6.38.3/grsecurity/grsum.c linux-2.6.38.3/grsecurity/grsum.c +--- linux-2.6.38.3/grsecurity/grsum.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/grsum.c 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,61 @@ ++#include <linux/err.h> ++#include <linux/kernel.h> ++#include <linux/sched.h> ++#include <linux/mm.h> ++#include <linux/scatterlist.h> ++#include <linux/crypto.h> ++#include <linux/gracl.h> ++ ++ ++#if !defined(CONFIG_CRYPTO) || defined(CONFIG_CRYPTO_MODULE) || !defined(CONFIG_CRYPTO_SHA256) || defined(CONFIG_CRYPTO_SHA256_MODULE) ++#error "crypto and sha256 must be built into the kernel" ++#endif ++ ++int ++chkpw(struct gr_arg *entry, unsigned char *salt, unsigned char *sum) ++{ ++ char *p; ++ struct crypto_hash *tfm; ++ struct hash_desc desc; ++ struct scatterlist sg; ++ unsigned char temp_sum[GR_SHA_LEN]; ++ volatile int retval = 0; ++ volatile int dummy = 0; ++ unsigned int i; ++ ++ sg_init_table(&sg, 1); ++ ++ tfm = crypto_alloc_hash("sha256", 0, CRYPTO_ALG_ASYNC); ++ if (IS_ERR(tfm)) { ++ /* should never happen, since sha256 should be built in */ ++ return 1; ++ } ++ ++ desc.tfm = tfm; ++ desc.flags = 0; ++ ++ crypto_hash_init(&desc); ++ ++ p = salt; ++ sg_set_buf(&sg, p, GR_SALT_LEN); ++ crypto_hash_update(&desc, &sg, sg.length); ++ ++ p = entry->pw; ++ sg_set_buf(&sg, p, strlen(p)); ++ ++ crypto_hash_update(&desc, &sg, sg.length); ++ ++ crypto_hash_final(&desc, temp_sum); ++ ++ memset(entry->pw, 0, GR_PW_LEN); ++ ++ for (i = 0; i < GR_SHA_LEN; i++) ++ if (sum[i] != temp_sum[i]) ++ retval = 1; ++ else ++ dummy = 1; // waste a cycle ++ ++ crypto_free_hash(tfm); ++ ++ return retval; ++} +diff -urNp linux-2.6.38.3/grsecurity/Kconfig linux-2.6.38.3/grsecurity/Kconfig +--- linux-2.6.38.3/grsecurity/Kconfig 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/Kconfig 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,1045 @@ ++# ++# grecurity configuration ++# ++ ++menu "Grsecurity" ++ ++config GRKERNSEC ++ bool "Grsecurity" ++ select CRYPTO ++ select CRYPTO_SHA256 ++ help ++ If you say Y here, you will be able to configure many features ++ that will enhance the security of your system. It is highly ++ recommended that you say Y here and read through the help ++ for each option so that you fully understand the features and ++ can evaluate their usefulness for your machine. ++ ++choice ++ prompt "Security Level" ++ depends on GRKERNSEC ++ default GRKERNSEC_CUSTOM ++ ++config GRKERNSEC_LOW ++ bool "Low" ++ select GRKERNSEC_LINK ++ select GRKERNSEC_FIFO ++ select GRKERNSEC_EXECVE ++ select GRKERNSEC_RANDNET ++ select GRKERNSEC_DMESG ++ select GRKERNSEC_CHROOT ++ select GRKERNSEC_CHROOT_CHDIR ++ ++ help ++ If you choose this option, several of the grsecurity options will ++ be enabled that will give you greater protection against a number ++ of attacks, while assuring that none of your software will have any ++ conflicts with the additional security measures. If you run a lot ++ of unusual software, or you are having problems with the higher ++ security levels, you should say Y here. With this option, the ++ following features are enabled: ++ ++ - Linking restrictions ++ - FIFO restrictions ++ - Enforcing RLIMIT_NPROC on execve ++ - Restricted dmesg ++ - Enforced chdir("/") on chroot ++ - Runtime module disabling ++ ++config GRKERNSEC_MEDIUM ++ bool "Medium" ++ select PAX ++ select PAX_EI_PAX ++ select PAX_PT_PAX_FLAGS ++ select PAX_HAVE_ACL_FLAGS ++ select GRKERNSEC_PROC_MEMMAP if (PAX_NOEXEC || PAX_ASLR) ++ select GRKERNSEC_CHROOT ++ select GRKERNSEC_CHROOT_SYSCTL ++ select GRKERNSEC_LINK ++ select GRKERNSEC_FIFO ++ select GRKERNSEC_EXECVE ++ select GRKERNSEC_DMESG ++ select GRKERNSEC_RANDNET ++ select GRKERNSEC_FORKFAIL ++ select GRKERNSEC_TIME ++ select GRKERNSEC_SIGNAL ++ select GRKERNSEC_CHROOT ++ select GRKERNSEC_CHROOT_UNIX ++ select GRKERNSEC_CHROOT_MOUNT ++ select GRKERNSEC_CHROOT_PIVOT ++ select GRKERNSEC_CHROOT_DOUBLE ++ select GRKERNSEC_CHROOT_CHDIR ++ select GRKERNSEC_CHROOT_MKNOD ++ select GRKERNSEC_PROC ++ select GRKERNSEC_PROC_USERGROUP ++ select PAX_RANDUSTACK ++ select PAX_ASLR ++ select PAX_RANDMMAP ++ select PAX_REFCOUNT if (X86 || SPARC64) ++ select PAX_USERCOPY if ((X86 || SPARC32 || SPARC64 || PPC) && (SLAB || SLUB || SLOB)) ++ ++ help ++ If you say Y here, several features in addition to those included ++ in the low additional security level will be enabled. These ++ features provide even more security to your system, though in rare ++ cases they may be incompatible with very old or poorly written ++ software. If you enable this option, make sure that your auth ++ service (identd) is running as gid 1001. With this option, ++ the following features (in addition to those provided in the ++ low additional security level) will be enabled: ++ ++ - Failed fork logging ++ - Time change logging ++ - Signal logging ++ - Deny mounts in chroot ++ - Deny double chrooting ++ - Deny sysctl writes in chroot ++ - Deny mknod in chroot ++ - Deny access to abstract AF_UNIX sockets out of chroot ++ - Deny pivot_root in chroot ++ - Denied writes of /dev/kmem, /dev/mem, and /dev/port ++ - /proc restrictions with special GID set to 10 (usually wheel) ++ - Address Space Layout Randomization (ASLR) ++ - Prevent exploitation of most refcount overflows ++ - Bounds checking of copying between the kernel and userland ++ ++config GRKERNSEC_HIGH ++ bool "High" ++ select GRKERNSEC_LINK ++ select GRKERNSEC_FIFO ++ select GRKERNSEC_EXECVE ++ select GRKERNSEC_DMESG ++ select GRKERNSEC_FORKFAIL ++ select GRKERNSEC_TIME ++ select GRKERNSEC_SIGNAL ++ select GRKERNSEC_CHROOT ++ select GRKERNSEC_CHROOT_SHMAT ++ select GRKERNSEC_CHROOT_UNIX ++ select GRKERNSEC_CHROOT_MOUNT ++ select GRKERNSEC_CHROOT_FCHDIR ++ select GRKERNSEC_CHROOT_PIVOT ++ select GRKERNSEC_CHROOT_DOUBLE ++ select GRKERNSEC_CHROOT_CHDIR ++ select GRKERNSEC_CHROOT_MKNOD ++ select GRKERNSEC_CHROOT_CAPS ++ select GRKERNSEC_CHROOT_SYSCTL ++ select GRKERNSEC_CHROOT_FINDTASK ++ select GRKERNSEC_SYSFS_RESTRICT ++ select GRKERNSEC_PROC ++ select GRKERNSEC_PROC_MEMMAP if (PAX_NOEXEC || PAX_ASLR) ++ select GRKERNSEC_HIDESYM ++ select GRKERNSEC_BRUTE ++ select GRKERNSEC_PROC_USERGROUP ++ select GRKERNSEC_KMEM ++ select GRKERNSEC_RESLOG ++ select GRKERNSEC_RANDNET ++ select GRKERNSEC_PROC_ADD ++ select GRKERNSEC_CHROOT_CHMOD ++ select GRKERNSEC_CHROOT_NICE ++ select GRKERNSEC_AUDIT_MOUNT ++ select GRKERNSEC_MODHARDEN if (MODULES) ++ select GRKERNSEC_HARDEN_PTRACE ++ select GRKERNSEC_VM86 if (X86_32) ++ select GRKERNSEC_KERN_LOCKOUT if (X86) ++ select PAX ++ select PAX_RANDUSTACK ++ select PAX_ASLR ++ select PAX_RANDMMAP ++ select PAX_NOEXEC ++ select PAX_MPROTECT ++ select PAX_EI_PAX ++ select PAX_PT_PAX_FLAGS ++ select PAX_HAVE_ACL_FLAGS ++ select PAX_KERNEXEC if ((PPC || X86) && (!X86_32 || X86_WP_WORKS_OK) && !XEN) ++ select PAX_MEMORY_UDEREF if (X86 && !XEN) ++ select PAX_RANDKSTACK if (X86_TSC && X86) ++ select PAX_SEGMEXEC if (X86_32) ++ select PAX_PAGEEXEC ++ select PAX_EMUPLT if (ALPHA || PARISC || SPARC32 || SPARC64) ++ select PAX_EMUTRAMP if (PARISC) ++ select PAX_EMUSIGRT if (PARISC) ++ select PAX_ETEXECRELOCS if (ALPHA || IA64 || PARISC) ++ select PAX_ELFRELOCS if (PAX_ETEXECRELOCS || (IA64 || PPC || X86)) ++ select PAX_REFCOUNT if (X86 || SPARC64) ++ select PAX_USERCOPY if ((X86 || PPC || SPARC32 || SPARC64) && (SLAB || SLUB || SLOB)) ++ help ++ If you say Y here, many of the features of grsecurity will be ++ enabled, which will protect you against many kinds of attacks ++ against your system. The heightened security comes at a cost ++ of an increased chance of incompatibilities with rare software ++ on your machine. Since this security level enables PaX, you should ++ view <http://pax.grsecurity.net> and read about the PaX ++ project. While you are there, download chpax and run it on ++ binaries that cause problems with PaX. Also remember that ++ since the /proc restrictions are enabled, you must run your ++ identd as gid 1001. This security level enables the following ++ features in addition to those listed in the low and medium ++ security levels: ++ ++ - Additional /proc restrictions ++ - Chmod restrictions in chroot ++ - No signals, ptrace, or viewing of processes outside of chroot ++ - Capability restrictions in chroot ++ - Deny fchdir out of chroot ++ - Priority restrictions in chroot ++ - Segmentation-based implementation of PaX ++ - Mprotect restrictions ++ - Removal of addresses from /proc/<pid>/[smaps|maps|stat] ++ - Kernel stack randomization ++ - Mount/unmount/remount logging ++ - Kernel symbol hiding ++ - Prevention of memory exhaustion-based exploits ++ - Hardening of module auto-loading ++ - Ptrace restrictions ++ - Restricted vm86 mode ++ - Restricted sysfs/debugfs ++ - Active kernel exploit response ++ ++config GRKERNSEC_CUSTOM ++ bool "Custom" ++ help ++ If you say Y here, you will be able to configure every grsecurity ++ option, which allows you to enable many more features that aren't ++ covered in the basic security levels. These additional features ++ include TPE, socket restrictions, and the sysctl system for ++ grsecurity. It is advised that you read through the help for ++ each option to determine its usefulness in your situation. ++ ++endchoice ++ ++menu "Address Space Protection" ++depends on GRKERNSEC ++ ++config GRKERNSEC_KMEM ++ bool "Deny writing to /dev/kmem, /dev/mem, and /dev/port" ++ select STRICT_DEVMEM if (X86 || ARM || TILE || S390) ++ help ++ If you say Y here, /dev/kmem and /dev/mem won't be allowed to ++ be written to via mmap or otherwise to modify the running kernel. ++ /dev/port will also not be allowed to be opened. If you have module ++ support disabled, enabling this will close up four ways that are ++ currently used to insert malicious code into the running kernel. ++ Even with all these features enabled, we still highly recommend that ++ you use the RBAC system, as it is still possible for an attacker to ++ modify the running kernel through privileged I/O granted by ioperm/iopl. ++ If you are not using XFree86, you may be able to stop this additional ++ case by enabling the 'Disable privileged I/O' option. Though nothing ++ legitimately writes to /dev/kmem, XFree86 does need to write to /dev/mem, ++ but only to video memory, which is the only writing we allow in this ++ case. If /dev/kmem or /dev/mem are mmaped without PROT_WRITE, they will ++ not be allowed to mprotect it with PROT_WRITE later. ++ It is highly recommended that you say Y here if you meet all the ++ conditions above. ++ ++config GRKERNSEC_VM86 ++ bool "Restrict VM86 mode" ++ depends on X86_32 ++ ++ help ++ If you say Y here, only processes with CAP_SYS_RAWIO will be able to ++ make use of a special execution mode on 32bit x86 processors called ++ Virtual 8086 (VM86) mode. XFree86 may need vm86 mode for certain ++ video cards and will still work with this option enabled. The purpose ++ of the option is to prevent exploitation of emulation errors in ++ virtualization of vm86 mode like the one discovered in VMWare in 2009. ++ Nearly all users should be able to enable this option. ++ ++config GRKERNSEC_IO ++ bool "Disable privileged I/O" ++ depends on X86 ++ select RTC_CLASS ++ select RTC_INTF_DEV ++ select RTC_DRV_CMOS ++ ++ help ++ If you say Y here, all ioperm and iopl calls will return an error. ++ Ioperm and iopl can be used to modify the running kernel. ++ Unfortunately, some programs need this access to operate properly, ++ the most notable of which are XFree86 and hwclock. hwclock can be ++ remedied by having RTC support in the kernel, so real-time ++ clock support is enabled if this option is enabled, to ensure ++ that hwclock operates correctly. XFree86 still will not ++ operate correctly with this option enabled, so DO NOT CHOOSE Y ++ IF YOU USE XFree86. If you use XFree86 and you still want to ++ protect your kernel against modification, use the RBAC system. ++ ++config GRKERNSEC_PROC_MEMMAP ++ bool "Remove addresses from /proc/<pid>/[smaps|maps|stat]" ++ default y if (PAX_NOEXEC || PAX_ASLR) ++ depends on PAX_NOEXEC || PAX_ASLR ++ help ++ If you say Y here, the /proc/<pid>/maps and /proc/<pid>/stat files will ++ give no information about the addresses of its mappings if ++ PaX features that rely on random addresses are enabled on the task. ++ If you use PaX it is greatly recommended that you say Y here as it ++ closes up a hole that makes the full ASLR useless for suid ++ binaries. ++ ++config GRKERNSEC_BRUTE ++ bool "Deter exploit bruteforcing" ++ help ++ If you say Y here, attempts to bruteforce exploits against forking ++ daemons such as apache or sshd, as well as against suid/sgid binaries ++ will be deterred. When a child of a forking daemon is killed by PaX ++ or crashes due to an illegal instruction or other suspicious signal, ++ the parent process will be delayed 30 seconds upon every subsequent ++ fork until the administrator is able to assess the situation and ++ restart the daemon. ++ In the suid/sgid case, the attempt is logged, the user has all their ++ processes terminated, and they are prevented from executing any further ++ processes for 15 minutes. ++ It is recommended that you also enable signal logging in the auditing ++ section so that logs are generated when a process triggers a suspicious ++ signal. ++ ++config GRKERNSEC_MODHARDEN ++ bool "Harden module auto-loading" ++ depends on MODULES ++ help ++ If you say Y here, module auto-loading in response to use of some ++ feature implemented by an unloaded module will be restricted to ++ root users. Enabling this option helps defend against attacks ++ by unprivileged users who abuse the auto-loading behavior to ++ cause a vulnerable module to load that is then exploited. ++ ++ If this option prevents a legitimate use of auto-loading for a ++ non-root user, the administrator can execute modprobe manually ++ with the exact name of the module mentioned in the alert log. ++ Alternatively, the administrator can add the module to the list ++ of modules loaded at boot by modifying init scripts. ++ ++ Modification of init scripts will most likely be needed on ++ Ubuntu servers with encrypted home directory support enabled, ++ as the first non-root user logging in will cause the ecb(aes), ++ ecb(aes)-all, cbc(aes), and cbc(aes)-all modules to be loaded. ++ ++config GRKERNSEC_HIDESYM ++ bool "Hide kernel symbols" ++ help ++ If you say Y here, getting information on loaded modules, and ++ displaying all kernel symbols through a syscall will be restricted ++ to users with CAP_SYS_MODULE. For software compatibility reasons, ++ /proc/kallsyms will be restricted to the root user. The RBAC ++ system can hide that entry even from root. ++ ++ This option also prevents leaking of kernel addresses through ++ several /proc entries. ++ ++ Note that this option is only effective provided the following ++ conditions are met: ++ 1) The kernel using grsecurity is not precompiled by some distribution ++ 2) You have also enabled GRKERNSEC_DMESG ++ 3) You are using the RBAC system and hiding other files such as your ++ kernel image and System.map. Alternatively, enabling this option ++ causes the permissions on /boot, /lib/modules, and the kernel ++ source directory to change at compile time to prevent ++ reading by non-root users. ++ If the above conditions are met, this option will aid in providing a ++ useful protection against local kernel exploitation of overflows ++ and arbitrary read/write vulnerabilities. ++ ++config GRKERNSEC_KERN_LOCKOUT ++ bool "Active kernel exploit response" ++ depends on X86 ++ help ++ If you say Y here, when a PaX alert is triggered due to suspicious ++ activity in the kernel (from KERNEXEC/UDEREF/USERCOPY) ++ or an OOPs occurs due to bad memory accesses, instead of just ++ terminating the offending process (and potentially allowing ++ a subsequent exploit from the same user), we will take one of two ++ actions: ++ If the user was root, we will panic the system ++ If the user was non-root, we will log the attempt, terminate ++ all processes owned by the user, then prevent them from creating ++ any new processes until the system is restarted ++ This deters repeated kernel exploitation/bruteforcing attempts ++ and is useful for later forensics. ++ ++endmenu ++menu "Role Based Access Control Options" ++depends on GRKERNSEC ++ ++config GRKERNSEC_RBAC_DEBUG ++ bool ++ ++config GRKERNSEC_NO_RBAC ++ bool "Disable RBAC system" ++ help ++ If you say Y here, the /dev/grsec device will be removed from the kernel, ++ preventing the RBAC system from being enabled. You should only say Y ++ here if you have no intention of using the RBAC system, so as to prevent ++ an attacker with root access from misusing the RBAC system to hide files ++ and processes when loadable module support and /dev/[k]mem have been ++ locked down. ++ ++config GRKERNSEC_ACL_HIDEKERN ++ bool "Hide kernel processes" ++ help ++ If you say Y here, all kernel threads will be hidden to all ++ processes but those whose subject has the "view hidden processes" ++ flag. ++ ++config GRKERNSEC_ACL_MAXTRIES ++ int "Maximum tries before password lockout" ++ default 3 ++ help ++ This option enforces the maximum number of times a user can attempt ++ to authorize themselves with the grsecurity RBAC system before being ++ denied the ability to attempt authorization again for a specified time. ++ The lower the number, the harder it will be to brute-force a password. ++ ++config GRKERNSEC_ACL_TIMEOUT ++ int "Time to wait after max password tries, in seconds" ++ default 30 ++ help ++ This option specifies the time the user must wait after attempting to ++ authorize to the RBAC system with the maximum number of invalid ++ passwords. The higher the number, the harder it will be to brute-force ++ a password. ++ ++endmenu ++menu "Filesystem Protections" ++depends on GRKERNSEC ++ ++config GRKERNSEC_PROC ++ bool "Proc restrictions" ++ help ++ If you say Y here, the permissions of the /proc filesystem ++ will be altered to enhance system security and privacy. You MUST ++ choose either a user only restriction or a user and group restriction. ++ Depending upon the option you choose, you can either restrict users to ++ see only the processes they themselves run, or choose a group that can ++ view all processes and files normally restricted to root if you choose ++ the "restrict to user only" option. NOTE: If you're running identd as ++ a non-root user, you will have to run it as the group you specify here. ++ ++config GRKERNSEC_PROC_USER ++ bool "Restrict /proc to user only" ++ depends on GRKERNSEC_PROC ++ help ++ If you say Y here, non-root users will only be able to view their own ++ processes, and restricts them from viewing network-related information, ++ and viewing kernel symbol and module information. ++ ++config GRKERNSEC_PROC_USERGROUP ++ bool "Allow special group" ++ depends on GRKERNSEC_PROC && !GRKERNSEC_PROC_USER ++ help ++ If you say Y here, you will be able to select a group that will be ++ able to view all processes and network-related information. If you've ++ enabled GRKERNSEC_HIDESYM, kernel and symbol information may still ++ remain hidden. This option is useful if you want to run identd as ++ a non-root user. ++ ++config GRKERNSEC_PROC_GID ++ int "GID for special group" ++ depends on GRKERNSEC_PROC_USERGROUP ++ default 1001 ++ ++config GRKERNSEC_PROC_ADD ++ bool "Additional restrictions" ++ depends on GRKERNSEC_PROC_USER || GRKERNSEC_PROC_USERGROUP ++ help ++ If you say Y here, additional restrictions will be placed on ++ /proc that keep normal users from viewing device information and ++ slabinfo information that could be useful for exploits. ++ ++config GRKERNSEC_LINK ++ bool "Linking restrictions" ++ help ++ If you say Y here, /tmp race exploits will be prevented, since users ++ will no longer be able to follow symlinks owned by other users in ++ world-writable +t directories (e.g. /tmp), unless the owner of the ++ symlink is the owner of the directory. users will also not be ++ able to hardlink to files they do not own. If the sysctl option is ++ enabled, a sysctl option with name "linking_restrictions" is created. ++ ++config GRKERNSEC_FIFO ++ bool "FIFO restrictions" ++ help ++ If you say Y here, users will not be able to write to FIFOs they don't ++ own in world-writable +t directories (e.g. /tmp), unless the owner of ++ the FIFO is the same owner of the directory it's held in. If the sysctl ++ option is enabled, a sysctl option with name "fifo_restrictions" is ++ created. ++ ++config GRKERNSEC_SYSFS_RESTRICT ++ bool "Sysfs/debugfs restriction" ++ depends on SYSFS ++ help ++ If you say Y here, sysfs (the pseudo-filesystem mounted at /sys) and ++ any filesystem normally mounted under it (e.g. debugfs) will only ++ be accessible by root. These filesystems generally provide access ++ to hardware and debug information that isn't appropriate for unprivileged ++ users of the system. Sysfs and debugfs have also become a large source ++ of new vulnerabilities, ranging from infoleaks to local compromise. ++ There has been very little oversight with an eye toward security involved ++ in adding new exporters of information to these filesystems, so their ++ use is discouraged. ++ This option is equivalent to a chmod 0700 of the mount paths. ++ ++config GRKERNSEC_ROFS ++ bool "Runtime read-only mount protection" ++ help ++ If you say Y here, a sysctl option with name "romount_protect" will ++ be created. By setting this option to 1 at runtime, filesystems ++ will be protected in the following ways: ++ * No new writable mounts will be allowed ++ * Existing read-only mounts won't be able to be remounted read/write ++ * Write operations will be denied on all block devices ++ This option acts independently of grsec_lock: once it is set to 1, ++ it cannot be turned off. Therefore, please be mindful of the resulting ++ behavior if this option is enabled in an init script on a read-only ++ filesystem. This feature is mainly intended for secure embedded systems. ++ ++config GRKERNSEC_CHROOT ++ bool "Chroot jail restrictions" ++ help ++ If you say Y here, you will be able to choose several options that will ++ make breaking out of a chrooted jail much more difficult. If you ++ encounter no software incompatibilities with the following options, it ++ is recommended that you enable each one. ++ ++config GRKERNSEC_CHROOT_MOUNT ++ bool "Deny mounts" ++ depends on GRKERNSEC_CHROOT ++ help ++ If you say Y here, processes inside a chroot will not be able to ++ mount or remount filesystems. If the sysctl option is enabled, a ++ sysctl option with name "chroot_deny_mount" is created. ++ ++config GRKERNSEC_CHROOT_DOUBLE ++ bool "Deny double-chroots" ++ depends on GRKERNSEC_CHROOT ++ help ++ If you say Y here, processes inside a chroot will not be able to chroot ++ again outside the chroot. This is a widely used method of breaking ++ out of a chroot jail and should not be allowed. If the sysctl ++ option is enabled, a sysctl option with name ++ "chroot_deny_chroot" is created. ++ ++config GRKERNSEC_CHROOT_PIVOT ++ bool "Deny pivot_root in chroot" ++ depends on GRKERNSEC_CHROOT ++ help ++ If you say Y here, processes inside a chroot will not be able to use ++ a function called pivot_root() that was introduced in Linux 2.3.41. It ++ works similar to chroot in that it changes the root filesystem. This ++ function could be misused in a chrooted process to attempt to break out ++ of the chroot, and therefore should not be allowed. If the sysctl ++ option is enabled, a sysctl option with name "chroot_deny_pivot" is ++ created. ++ ++config GRKERNSEC_CHROOT_CHDIR ++ bool "Enforce chdir(\"/\") on all chroots" ++ depends on GRKERNSEC_CHROOT ++ help ++ If you say Y here, the current working directory of all newly-chrooted ++ applications will be set to the the root directory of the chroot. ++ The man page on chroot(2) states: ++ Note that this call does not change the current working ++ directory, so that `.' can be outside the tree rooted at ++ `/'. In particular, the super-user can escape from a ++ `chroot jail' by doing `mkdir foo; chroot foo; cd ..'. ++ ++ It is recommended that you say Y here, since it's not known to break ++ any software. If the sysctl option is enabled, a sysctl option with ++ name "chroot_enforce_chdir" is created. ++ ++config GRKERNSEC_CHROOT_CHMOD ++ bool "Deny (f)chmod +s" ++ depends on GRKERNSEC_CHROOT ++ help ++ If you say Y here, processes inside a chroot will not be able to chmod ++ or fchmod files to make them have suid or sgid bits. This protects ++ against another published method of breaking a chroot. If the sysctl ++ option is enabled, a sysctl option with name "chroot_deny_chmod" is ++ created. ++ ++config GRKERNSEC_CHROOT_FCHDIR ++ bool "Deny fchdir out of chroot" ++ depends on GRKERNSEC_CHROOT ++ help ++ If you say Y here, a well-known method of breaking chroots by fchdir'ing ++ to a file descriptor of the chrooting process that points to a directory ++ outside the filesystem will be stopped. If the sysctl option ++ is enabled, a sysctl option with name "chroot_deny_fchdir" is created. ++ ++config GRKERNSEC_CHROOT_MKNOD ++ bool "Deny mknod" ++ depends on GRKERNSEC_CHROOT ++ help ++ If you say Y here, processes inside a chroot will not be allowed to ++ mknod. The problem with using mknod inside a chroot is that it ++ would allow an attacker to create a device entry that is the same ++ as one on the physical root of your system, which could range from ++ anything from the console device to a device for your harddrive (which ++ they could then use to wipe the drive or steal data). It is recommended ++ that you say Y here, unless you run into software incompatibilities. ++ If the sysctl option is enabled, a sysctl option with name ++ "chroot_deny_mknod" is created. ++ ++config GRKERNSEC_CHROOT_SHMAT ++ bool "Deny shmat() out of chroot" ++ depends on GRKERNSEC_CHROOT ++ help ++ If you say Y here, processes inside a chroot will not be able to attach ++ to shared memory segments that were created outside of the chroot jail. ++ It is recommended that you say Y here. If the sysctl option is enabled, ++ a sysctl option with name "chroot_deny_shmat" is created. ++ ++config GRKERNSEC_CHROOT_UNIX ++ bool "Deny access to abstract AF_UNIX sockets out of chroot" ++ depends on GRKERNSEC_CHROOT ++ help ++ If you say Y here, processes inside a chroot will not be able to ++ connect to abstract (meaning not belonging to a filesystem) Unix ++ domain sockets that were bound outside of a chroot. It is recommended ++ that you say Y here. If the sysctl option is enabled, a sysctl option ++ with name "chroot_deny_unix" is created. ++ ++config GRKERNSEC_CHROOT_FINDTASK ++ bool "Protect outside processes" ++ depends on GRKERNSEC_CHROOT ++ help ++ If you say Y here, processes inside a chroot will not be able to ++ kill, send signals with fcntl, ptrace, capget, getpgid, setpgid, ++ getsid, or view any process outside of the chroot. If the sysctl ++ option is enabled, a sysctl option with name "chroot_findtask" is ++ created. ++ ++config GRKERNSEC_CHROOT_NICE ++ bool "Restrict priority changes" ++ depends on GRKERNSEC_CHROOT ++ help ++ If you say Y here, processes inside a chroot will not be able to raise ++ the priority of processes in the chroot, or alter the priority of ++ processes outside the chroot. This provides more security than simply ++ removing CAP_SYS_NICE from the process' capability set. If the ++ sysctl option is enabled, a sysctl option with name "chroot_restrict_nice" ++ is created. ++ ++config GRKERNSEC_CHROOT_SYSCTL ++ bool "Deny sysctl writes" ++ depends on GRKERNSEC_CHROOT ++ help ++ If you say Y here, an attacker in a chroot will not be able to ++ write to sysctl entries, either by sysctl(2) or through a /proc ++ interface. It is strongly recommended that you say Y here. If the ++ sysctl option is enabled, a sysctl option with name ++ "chroot_deny_sysctl" is created. ++ ++config GRKERNSEC_CHROOT_CAPS ++ bool "Capability restrictions" ++ depends on GRKERNSEC_CHROOT ++ help ++ If you say Y here, the capabilities on all root processes within a ++ chroot jail will be lowered to stop module insertion, raw i/o, ++ system and net admin tasks, rebooting the system, modifying immutable ++ files, modifying IPC owned by another, and changing the system time. ++ This is left an option because it can break some apps. Disable this ++ if your chrooted apps are having problems performing those kinds of ++ tasks. If the sysctl option is enabled, a sysctl option with ++ name "chroot_caps" is created. ++ ++endmenu ++menu "Kernel Auditing" ++depends on GRKERNSEC ++ ++config GRKERNSEC_AUDIT_GROUP ++ bool "Single group for auditing" ++ help ++ If you say Y here, the exec, chdir, and (un)mount logging features ++ will only operate on a group you specify. This option is recommended ++ if you only want to watch certain users instead of having a large ++ amount of logs from the entire system. If the sysctl option is enabled, ++ a sysctl option with name "audit_group" is created. ++ ++config GRKERNSEC_AUDIT_GID ++ int "GID for auditing" ++ depends on GRKERNSEC_AUDIT_GROUP ++ default 1007 ++ ++config GRKERNSEC_EXECLOG ++ bool "Exec logging" ++ help ++ If you say Y here, all execve() calls will be logged (since the ++ other exec*() calls are frontends to execve(), all execution ++ will be logged). Useful for shell-servers that like to keep track ++ of their users. If the sysctl option is enabled, a sysctl option with ++ name "exec_logging" is created. ++ WARNING: This option when enabled will produce a LOT of logs, especially ++ on an active system. ++ ++config GRKERNSEC_RESLOG ++ bool "Resource logging" ++ help ++ If you say Y here, all attempts to overstep resource limits will ++ be logged with the resource name, the requested size, and the current ++ limit. It is highly recommended that you say Y here. If the sysctl ++ option is enabled, a sysctl option with name "resource_logging" is ++ created. If the RBAC system is enabled, the sysctl value is ignored. ++ ++config GRKERNSEC_CHROOT_EXECLOG ++ bool "Log execs within chroot" ++ help ++ If you say Y here, all executions inside a chroot jail will be logged ++ to syslog. This can cause a large amount of logs if certain ++ applications (eg. djb's daemontools) are installed on the system, and ++ is therefore left as an option. If the sysctl option is enabled, a ++ sysctl option with name "chroot_execlog" is created. ++ ++config GRKERNSEC_AUDIT_PTRACE ++ bool "Ptrace logging" ++ help ++ If you say Y here, all attempts to attach to a process via ptrace ++ will be logged. If the sysctl option is enabled, a sysctl option ++ with name "audit_ptrace" is created. ++ ++config GRKERNSEC_AUDIT_CHDIR ++ bool "Chdir logging" ++ help ++ If you say Y here, all chdir() calls will be logged. If the sysctl ++ option is enabled, a sysctl option with name "audit_chdir" is created. ++ ++config GRKERNSEC_AUDIT_MOUNT ++ bool "(Un)Mount logging" ++ help ++ If you say Y here, all mounts and unmounts will be logged. If the ++ sysctl option is enabled, a sysctl option with name "audit_mount" is ++ created. ++ ++config GRKERNSEC_SIGNAL ++ bool "Signal logging" ++ help ++ If you say Y here, certain important signals will be logged, such as ++ SIGSEGV, which will as a result inform you of when a error in a program ++ occurred, which in some cases could mean a possible exploit attempt. ++ If the sysctl option is enabled, a sysctl option with name ++ "signal_logging" is created. ++ ++config GRKERNSEC_FORKFAIL ++ bool "Fork failure logging" ++ help ++ If you say Y here, all failed fork() attempts will be logged. ++ This could suggest a fork bomb, or someone attempting to overstep ++ their process limit. If the sysctl option is enabled, a sysctl option ++ with name "forkfail_logging" is created. ++ ++config GRKERNSEC_TIME ++ bool "Time change logging" ++ help ++ If you say Y here, any changes of the system clock will be logged. ++ If the sysctl option is enabled, a sysctl option with name ++ "timechange_logging" is created. ++ ++config GRKERNSEC_PROC_IPADDR ++ bool "/proc/<pid>/ipaddr support" ++ help ++ If you say Y here, a new entry will be added to each /proc/<pid> ++ directory that contains the IP address of the person using the task. ++ The IP is carried across local TCP and AF_UNIX stream sockets. ++ This information can be useful for IDS/IPSes to perform remote response ++ to a local attack. The entry is readable by only the owner of the ++ process (and root if he has CAP_DAC_OVERRIDE, which can be removed via ++ the RBAC system), and thus does not create privacy concerns. ++ ++config GRKERNSEC_RWXMAP_LOG ++ bool 'Denied RWX mmap/mprotect logging' ++ depends on PAX_MPROTECT && !PAX_EMUPLT && !PAX_EMUSIGRT ++ help ++ If you say Y here, calls to mmap() and mprotect() with explicit ++ usage of PROT_WRITE and PROT_EXEC together will be logged when ++ denied by the PAX_MPROTECT feature. If the sysctl option is ++ enabled, a sysctl option with name "rwxmap_logging" is created. ++ ++config GRKERNSEC_AUDIT_TEXTREL ++ bool 'ELF text relocations logging (READ HELP)' ++ depends on PAX_MPROTECT ++ help ++ If you say Y here, text relocations will be logged with the filename ++ of the offending library or binary. The purpose of the feature is ++ to help Linux distribution developers get rid of libraries and ++ binaries that need text relocations which hinder the future progress ++ of PaX. Only Linux distribution developers should say Y here, and ++ never on a production machine, as this option creates an information ++ leak that could aid an attacker in defeating the randomization of ++ a single memory region. If the sysctl option is enabled, a sysctl ++ option with name "audit_textrel" is created. ++ ++endmenu ++ ++menu "Executable Protections" ++depends on GRKERNSEC ++ ++config GRKERNSEC_EXECVE ++ bool "Enforce RLIMIT_NPROC on execs" ++ help ++ If you say Y here, users with a resource limit on processes will ++ have the value checked during execve() calls. The current system ++ only checks the system limit during fork() calls. If the sysctl option ++ is enabled, a sysctl option with name "execve_limiting" is created. ++ ++config GRKERNSEC_DMESG ++ bool "Dmesg(8) restriction" ++ help ++ If you say Y here, non-root users will not be able to use dmesg(8) ++ to view up to the last 4kb of messages in the kernel's log buffer. ++ The kernel's log buffer often contains kernel addresses and other ++ identifying information useful to an attacker in fingerprinting a ++ system for a targeted exploit. ++ If the sysctl option is enabled, a sysctl option with name "dmesg" is ++ created. ++ ++config GRKERNSEC_HARDEN_PTRACE ++ bool "Deter ptrace-based process snooping" ++ help ++ If you say Y here, TTY sniffers and other malicious monitoring ++ programs implemented through ptrace will be defeated. If you ++ have been using the RBAC system, this option has already been ++ enabled for several years for all users, with the ability to make ++ fine-grained exceptions. ++ ++ This option only affects the ability of non-root users to ptrace ++ processes that are not a descendent of the ptracing process. ++ This means that strace ./binary and gdb ./binary will still work, ++ but attaching to arbitrary processes will not. If the sysctl ++ option is enabled, a sysctl option with name "harden_ptrace" is ++ created. ++ ++config GRKERNSEC_TPE ++ bool "Trusted Path Execution (TPE)" ++ help ++ If you say Y here, you will be able to choose a gid to add to the ++ supplementary groups of users you want to mark as "untrusted." ++ These users will not be able to execute any files that are not in ++ root-owned directories writable only by root. If the sysctl option ++ is enabled, a sysctl option with name "tpe" is created. ++ ++config GRKERNSEC_TPE_ALL ++ bool "Partially restrict all non-root users" ++ depends on GRKERNSEC_TPE ++ help ++ If you say Y here, all non-root users will be covered under ++ a weaker TPE restriction. This is separate from, and in addition to, ++ the main TPE options that you have selected elsewhere. Thus, if a ++ "trusted" GID is chosen, this restriction applies to even that GID. ++ Under this restriction, all non-root users will only be allowed to ++ execute files in directories they own that are not group or ++ world-writable, or in directories owned by root and writable only by ++ root. If the sysctl option is enabled, a sysctl option with name ++ "tpe_restrict_all" is created. ++ ++config GRKERNSEC_TPE_INVERT ++ bool "Invert GID option" ++ depends on GRKERNSEC_TPE ++ help ++ If you say Y here, the group you specify in the TPE configuration will ++ decide what group TPE restrictions will be *disabled* for. This ++ option is useful if you want TPE restrictions to be applied to most ++ users on the system. If the sysctl option is enabled, a sysctl option ++ with name "tpe_invert" is created. Unlike other sysctl options, this ++ entry will default to on for backward-compatibility. ++ ++config GRKERNSEC_TPE_GID ++ int "GID for untrusted users" ++ depends on GRKERNSEC_TPE && !GRKERNSEC_TPE_INVERT ++ default 1005 ++ help ++ Setting this GID determines what group TPE restrictions will be ++ *enabled* for. If the sysctl option is enabled, a sysctl option ++ with name "tpe_gid" is created. ++ ++config GRKERNSEC_TPE_GID ++ int "GID for trusted users" ++ depends on GRKERNSEC_TPE && GRKERNSEC_TPE_INVERT ++ default 1005 ++ help ++ Setting this GID determines what group TPE restrictions will be ++ *disabled* for. If the sysctl option is enabled, a sysctl option ++ with name "tpe_gid" is created. ++ ++endmenu ++menu "Network Protections" ++depends on GRKERNSEC ++ ++config GRKERNSEC_RANDNET ++ bool "Larger entropy pools" ++ help ++ If you say Y here, the entropy pools used for many features of Linux ++ and grsecurity will be doubled in size. Since several grsecurity ++ features use additional randomness, it is recommended that you say Y ++ here. Saying Y here has a similar effect as modifying ++ /proc/sys/kernel/random/poolsize. ++ ++config GRKERNSEC_BLACKHOLE ++ bool "TCP/UDP blackhole and LAST_ACK DoS prevention" ++ help ++ If you say Y here, neither TCP resets nor ICMP ++ destination-unreachable packets will be sent in response to packets ++ sent to ports for which no associated listening process exists. ++ This feature supports both IPV4 and IPV6 and exempts the ++ loopback interface from blackholing. Enabling this feature ++ makes a host more resilient to DoS attacks and reduces network ++ visibility against scanners. ++ ++ The blackhole feature as-implemented is equivalent to the FreeBSD ++ blackhole feature, as it prevents RST responses to all packets, not ++ just SYNs. Under most application behavior this causes no ++ problems, but applications (like haproxy) may not close certain ++ connections in a way that cleanly terminates them on the remote ++ end, leaving the remote host in LAST_ACK state. Because of this ++ side-effect and to prevent intentional LAST_ACK DoSes, this ++ feature also adds automatic mitigation against such attacks. ++ The mitigation drastically reduces the amount of time a socket ++ can spend in LAST_ACK state. If you're using haproxy and not ++ all servers it connects to have this option enabled, consider ++ disabling this feature on the haproxy host. ++ ++ If the sysctl option is enabled, two sysctl options with names ++ "ip_blackhole" and "lastack_retries" will be created. ++ While "ip_blackhole" takes the standard zero/non-zero on/off ++ toggle, "lastack_retries" uses the same kinds of values as ++ "tcp_retries1" and "tcp_retries2". The default value of 4 ++ prevents a socket from lasting more than 45 seconds in LAST_ACK ++ state. ++ ++config GRKERNSEC_SOCKET ++ bool "Socket restrictions" ++ help ++ If you say Y here, you will be able to choose from several options. ++ If you assign a GID on your system and add it to the supplementary ++ groups of users you want to restrict socket access to, this patch ++ will perform up to three things, based on the option(s) you choose. ++ ++config GRKERNSEC_SOCKET_ALL ++ bool "Deny any sockets to group" ++ depends on GRKERNSEC_SOCKET ++ help ++ If you say Y here, you will be able to choose a GID of whose users will ++ be unable to connect to other hosts from your machine or run server ++ applications from your machine. If the sysctl option is enabled, a ++ sysctl option with name "socket_all" is created. ++ ++config GRKERNSEC_SOCKET_ALL_GID ++ int "GID to deny all sockets for" ++ depends on GRKERNSEC_SOCKET_ALL ++ default 1004 ++ help ++ Here you can choose the GID to disable socket access for. Remember to ++ add the users you want socket access disabled for to the GID ++ specified here. If the sysctl option is enabled, a sysctl option ++ with name "socket_all_gid" is created. ++ ++config GRKERNSEC_SOCKET_CLIENT ++ bool "Deny client sockets to group" ++ depends on GRKERNSEC_SOCKET ++ help ++ If you say Y here, you will be able to choose a GID of whose users will ++ be unable to connect to other hosts from your machine, but will be ++ able to run servers. If this option is enabled, all users in the group ++ you specify will have to use passive mode when initiating ftp transfers ++ from the shell on your machine. If the sysctl option is enabled, a ++ sysctl option with name "socket_client" is created. ++ ++config GRKERNSEC_SOCKET_CLIENT_GID ++ int "GID to deny client sockets for" ++ depends on GRKERNSEC_SOCKET_CLIENT ++ default 1003 ++ help ++ Here you can choose the GID to disable client socket access for. ++ Remember to add the users you want client socket access disabled for to ++ the GID specified here. If the sysctl option is enabled, a sysctl ++ option with name "socket_client_gid" is created. ++ ++config GRKERNSEC_SOCKET_SERVER ++ bool "Deny server sockets to group" ++ depends on GRKERNSEC_SOCKET ++ help ++ If you say Y here, you will be able to choose a GID of whose users will ++ be unable to run server applications from your machine. If the sysctl ++ option is enabled, a sysctl option with name "socket_server" is created. ++ ++config GRKERNSEC_SOCKET_SERVER_GID ++ int "GID to deny server sockets for" ++ depends on GRKERNSEC_SOCKET_SERVER ++ default 1002 ++ help ++ Here you can choose the GID to disable server socket access for. ++ Remember to add the users you want server socket access disabled for to ++ the GID specified here. If the sysctl option is enabled, a sysctl ++ option with name "socket_server_gid" is created. ++ ++endmenu ++menu "Sysctl support" ++depends on GRKERNSEC && SYSCTL ++ ++config GRKERNSEC_SYSCTL ++ bool "Sysctl support" ++ help ++ If you say Y here, you will be able to change the options that ++ grsecurity runs with at bootup, without having to recompile your ++ kernel. You can echo values to files in /proc/sys/kernel/grsecurity ++ to enable (1) or disable (0) various features. All the sysctl entries ++ are mutable until the "grsec_lock" entry is set to a non-zero value. ++ All features enabled in the kernel configuration are disabled at boot ++ if you do not say Y to the "Turn on features by default" option. ++ All options should be set at startup, and the grsec_lock entry should ++ be set to a non-zero value after all the options are set. ++ *THIS IS EXTREMELY IMPORTANT* ++ ++config GRKERNSEC_SYSCTL_DISTRO ++ bool "Extra sysctl support for distro makers (READ HELP)" ++ depends on GRKERNSEC_SYSCTL && GRKERNSEC_IO ++ help ++ If you say Y here, additional sysctl options will be created ++ for features that affect processes running as root. Therefore, ++ it is critical when using this option that the grsec_lock entry be ++ enabled after boot. Only distros with prebuilt kernel packages ++ with this option enabled that can ensure grsec_lock is enabled ++ after boot should use this option. ++ *Failure to set grsec_lock after boot makes all grsec features ++ this option covers useless* ++ ++ Currently this option creates the following sysctl entries: ++ "Disable Privileged I/O": "disable_priv_io" ++ ++config GRKERNSEC_SYSCTL_ON ++ bool "Turn on features by default" ++ depends on GRKERNSEC_SYSCTL ++ help ++ If you say Y here, instead of having all features enabled in the ++ kernel configuration disabled at boot time, the features will be ++ enabled at boot time. It is recommended you say Y here unless ++ there is some reason you would want all sysctl-tunable features to ++ be disabled by default. As mentioned elsewhere, it is important ++ to enable the grsec_lock entry once you have finished modifying ++ the sysctl entries. ++ ++endmenu ++menu "Logging Options" ++depends on GRKERNSEC ++ ++config GRKERNSEC_FLOODTIME ++ int "Seconds in between log messages (minimum)" ++ default 10 ++ help ++ This option allows you to enforce the number of seconds between ++ grsecurity log messages. The default should be suitable for most ++ people, however, if you choose to change it, choose a value small enough ++ to allow informative logs to be produced, but large enough to ++ prevent flooding. ++ ++config GRKERNSEC_FLOODBURST ++ int "Number of messages in a burst (maximum)" ++ default 4 ++ help ++ This option allows you to choose the maximum number of messages allowed ++ within the flood time interval you chose in a separate option. The ++ default should be suitable for most people, however if you find that ++ many of your logs are being interpreted as flooding, you may want to ++ raise this value. ++ ++endmenu ++ ++endmenu +diff -urNp linux-2.6.38.3/grsecurity/Makefile linux-2.6.38.3/grsecurity/Makefile +--- linux-2.6.38.3/grsecurity/Makefile 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/grsecurity/Makefile 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,29 @@ ++# grsecurity's ACL system was originally written in 2001 by Michael Dalton ++# during 2001-2009 it has been completely redesigned by Brad Spengler ++# into an RBAC system ++# ++# All code in this directory and various hooks inserted throughout the kernel ++# are copyright Brad Spengler - Open Source Security, Inc., and released ++# under the GPL v2 or higher ++ ++obj-y = grsec_chdir.o grsec_chroot.o grsec_exec.o grsec_fifo.o grsec_fork.o \ ++ grsec_mount.o grsec_sig.o grsec_sock.o grsec_sysctl.o \ ++ grsec_time.o grsec_tpe.o grsec_link.o grsec_pax.o grsec_ptrace.o ++ ++obj-$(CONFIG_GRKERNSEC) += grsec_init.o grsum.o gracl.o gracl_ip.o gracl_segv.o \ ++ gracl_cap.o gracl_alloc.o gracl_shm.o grsec_mem.o gracl_fs.o \ ++ gracl_learn.o grsec_log.o ++obj-$(CONFIG_GRKERNSEC_RESLOG) += gracl_res.o ++ ++ifndef CONFIG_GRKERNSEC ++obj-y += grsec_disabled.o ++endif ++ ++ifdef CONFIG_GRKERNSEC_HIDESYM ++extra-y := grsec_hidesym.o ++$(obj)/grsec_hidesym.o: ++ @-chmod -f 500 /boot ++ @-chmod -f 500 /lib/modules ++ @-chmod -f 700 . ++ @echo ' grsec: protected kernel image paths' ++endif +diff -urNp linux-2.6.38.3/include/acpi/acoutput.h linux-2.6.38.3/include/acpi/acoutput.h +--- linux-2.6.38.3/include/acpi/acoutput.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/acpi/acoutput.h 2011-04-17 15:57:32.000000000 -0400 +@@ -269,8 +269,8 @@ + * leaving no executable debug code! + */ + #define ACPI_FUNCTION_NAME(a) +-#define ACPI_DEBUG_PRINT(pl) +-#define ACPI_DEBUG_PRINT_RAW(pl) ++#define ACPI_DEBUG_PRINT(pl) do {} while (0) ++#define ACPI_DEBUG_PRINT_RAW(pl) do {} while (0) + + #endif /* ACPI_DEBUG_OUTPUT */ + +diff -urNp linux-2.6.38.3/include/acpi/acpi_drivers.h linux-2.6.38.3/include/acpi/acpi_drivers.h +--- linux-2.6.38.3/include/acpi/acpi_drivers.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/acpi/acpi_drivers.h 2011-04-17 15:57:32.000000000 -0400 +@@ -119,8 +119,8 @@ void pci_acpi_crs_quirks(void); + Dock Station + -------------------------------------------------------------------------- */ + struct acpi_dock_ops { +- acpi_notify_handler handler; +- acpi_notify_handler uevent; ++ const acpi_notify_handler handler; ++ const acpi_notify_handler uevent; + }; + + #if defined(CONFIG_ACPI_DOCK) || defined(CONFIG_ACPI_DOCK_MODULE) +@@ -128,7 +128,7 @@ extern int is_dock_device(acpi_handle ha + extern int register_dock_notifier(struct notifier_block *nb); + extern void unregister_dock_notifier(struct notifier_block *nb); + extern int register_hotplug_dock_device(acpi_handle handle, +- struct acpi_dock_ops *ops, ++ const struct acpi_dock_ops *ops, + void *context); + extern void unregister_hotplug_dock_device(acpi_handle handle); + #else +@@ -144,7 +144,7 @@ static inline void unregister_dock_notif + { + } + static inline int register_hotplug_dock_device(acpi_handle handle, +- struct acpi_dock_ops *ops, ++ const struct acpi_dock_ops *ops, + void *context) + { + return -ENODEV; +diff -urNp linux-2.6.38.3/include/asm-generic/atomic-long.h linux-2.6.38.3/include/asm-generic/atomic-long.h +--- linux-2.6.38.3/include/asm-generic/atomic-long.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/asm-generic/atomic-long.h 2011-04-17 15:57:32.000000000 -0400 +@@ -22,6 +22,12 @@ + + typedef atomic64_t atomic_long_t; + ++#ifdef CONFIG_PAX_REFCOUNT ++typedef atomic64_unchecked_t atomic_long_unchecked_t; ++#else ++typedef atomic64_t atomic_long_unchecked_t; ++#endif ++ + #define ATOMIC_LONG_INIT(i) ATOMIC64_INIT(i) + + static inline long atomic_long_read(atomic_long_t *l) +@@ -31,6 +37,15 @@ static inline long atomic_long_read(atom + return (long)atomic64_read(v); + } + ++#ifdef CONFIG_PAX_REFCOUNT ++static inline long atomic_long_read_unchecked(atomic_long_unchecked_t *l) ++{ ++ atomic64_unchecked_t *v = (atomic64_unchecked_t *)l; ++ ++ return (long)atomic64_read_unchecked(v); ++} ++#endif ++ + static inline void atomic_long_set(atomic_long_t *l, long i) + { + atomic64_t *v = (atomic64_t *)l; +@@ -38,6 +53,15 @@ static inline void atomic_long_set(atomi + atomic64_set(v, i); + } + ++#ifdef CONFIG_PAX_REFCOUNT ++static inline void atomic_long_set_unchecked(atomic_long_unchecked_t *l, long i) ++{ ++ atomic64_unchecked_t *v = (atomic64_unchecked_t *)l; ++ ++ atomic64_set_unchecked(v, i); ++} ++#endif ++ + static inline void atomic_long_inc(atomic_long_t *l) + { + atomic64_t *v = (atomic64_t *)l; +@@ -45,6 +69,15 @@ static inline void atomic_long_inc(atomi + atomic64_inc(v); + } + ++#ifdef CONFIG_PAX_REFCOUNT ++static inline void atomic_long_inc_unchecked(atomic_long_unchecked_t *l) ++{ ++ atomic64_unchecked_t *v = (atomic64_unchecked_t *)l; ++ ++ atomic64_inc_unchecked(v); ++} ++#endif ++ + static inline void atomic_long_dec(atomic_long_t *l) + { + atomic64_t *v = (atomic64_t *)l; +@@ -52,6 +85,15 @@ static inline void atomic_long_dec(atomi + atomic64_dec(v); + } + ++#ifdef CONFIG_PAX_REFCOUNT ++static inline void atomic_long_dec_unchecked(atomic_long_unchecked_t *l) ++{ ++ atomic64_unchecked_t *v = (atomic64_unchecked_t *)l; ++ ++ atomic64_dec_unchecked(v); ++} ++#endif ++ + static inline void atomic_long_add(long i, atomic_long_t *l) + { + atomic64_t *v = (atomic64_t *)l; +@@ -59,6 +101,15 @@ static inline void atomic_long_add(long + atomic64_add(i, v); + } + ++#ifdef CONFIG_PAX_REFCOUNT ++static inline void atomic_long_add_unchecked(long i, atomic_long_unchecked_t *l) ++{ ++ atomic64_unchecked_t *v = (atomic64_unchecked_t *)l; ++ ++ atomic64_add_unchecked(i, v); ++} ++#endif ++ + static inline void atomic_long_sub(long i, atomic_long_t *l) + { + atomic64_t *v = (atomic64_t *)l; +@@ -66,6 +117,15 @@ static inline void atomic_long_sub(long + atomic64_sub(i, v); + } + ++#ifdef CONFIG_PAX_REFCOUNT ++static inline void atomic_long_sub_unchecked(long i, atomic_long_unchecked_t *l) ++{ ++ atomic64_unchecked_t *v = (atomic64_unchecked_t *)l; ++ ++ atomic64_sub_unchecked(i, v); ++} ++#endif ++ + static inline int atomic_long_sub_and_test(long i, atomic_long_t *l) + { + atomic64_t *v = (atomic64_t *)l; +@@ -115,6 +175,15 @@ static inline long atomic_long_inc_retur + return (long)atomic64_inc_return(v); + } + ++#ifdef CONFIG_PAX_REFCOUNT ++static inline long atomic_long_inc_return_unchecked(atomic_long_unchecked_t *l) ++{ ++ atomic64_unchecked_t *v = (atomic64_unchecked_t *)l; ++ ++ return (long)atomic64_inc_return_unchecked(v); ++} ++#endif ++ + static inline long atomic_long_dec_return(atomic_long_t *l) + { + atomic64_t *v = (atomic64_t *)l; +@@ -140,6 +209,12 @@ static inline long atomic_long_add_unles + + typedef atomic_t atomic_long_t; + ++#ifdef CONFIG_PAX_REFCOUNT ++typedef atomic_unchecked_t atomic_long_unchecked_t; ++#else ++typedef atomic_t atomic_long_unchecked_t; ++#endif ++ + #define ATOMIC_LONG_INIT(i) ATOMIC_INIT(i) + static inline long atomic_long_read(atomic_long_t *l) + { +@@ -148,6 +223,15 @@ static inline long atomic_long_read(atom + return (long)atomic_read(v); + } + ++#ifdef CONFIG_PAX_REFCOUNT ++static inline long atomic_long_read_unchecked(atomic_long_unchecked_t *l) ++{ ++ atomic_unchecked_t *v = (atomic_unchecked_t *)l; ++ ++ return (long)atomic_read_unchecked(v); ++} ++#endif ++ + static inline void atomic_long_set(atomic_long_t *l, long i) + { + atomic_t *v = (atomic_t *)l; +@@ -155,6 +239,15 @@ static inline void atomic_long_set(atomi + atomic_set(v, i); + } + ++#ifdef CONFIG_PAX_REFCOUNT ++static inline void atomic_long_set_unchecked(atomic_long_unchecked_t *l, long i) ++{ ++ atomic_unchecked_t *v = (atomic_unchecked_t *)l; ++ ++ atomic_set_unchecked(v, i); ++} ++#endif ++ + static inline void atomic_long_inc(atomic_long_t *l) + { + atomic_t *v = (atomic_t *)l; +@@ -162,6 +255,15 @@ static inline void atomic_long_inc(atomi + atomic_inc(v); + } + ++#ifdef CONFIG_PAX_REFCOUNT ++static inline void atomic_long_inc_unchecked(atomic_long_unchecked_t *l) ++{ ++ atomic_unchecked_t *v = (atomic_unchecked_t *)l; ++ ++ atomic_inc_unchecked(v); ++} ++#endif ++ + static inline void atomic_long_dec(atomic_long_t *l) + { + atomic_t *v = (atomic_t *)l; +@@ -169,6 +271,15 @@ static inline void atomic_long_dec(atomi + atomic_dec(v); + } + ++#ifdef CONFIG_PAX_REFCOUNT ++static inline void atomic_long_dec_unchecked(atomic_long_unchecked_t *l) ++{ ++ atomic_unchecked_t *v = (atomic_unchecked_t *)l; ++ ++ atomic_dec_unchecked(v); ++} ++#endif ++ + static inline void atomic_long_add(long i, atomic_long_t *l) + { + atomic_t *v = (atomic_t *)l; +@@ -176,6 +287,15 @@ static inline void atomic_long_add(long + atomic_add(i, v); + } + ++#ifdef CONFIG_PAX_REFCOUNT ++static inline void atomic_long_add_unchecked(long i, atomic_long_unchecked_t *l) ++{ ++ atomic_unchecked_t *v = (atomic_unchecked_t *)l; ++ ++ atomic_add_unchecked(i, v); ++} ++#endif ++ + static inline void atomic_long_sub(long i, atomic_long_t *l) + { + atomic_t *v = (atomic_t *)l; +@@ -183,6 +303,15 @@ static inline void atomic_long_sub(long + atomic_sub(i, v); + } + ++#ifdef CONFIG_PAX_REFCOUNT ++static inline void atomic_long_sub_unchecked(long i, atomic_long_unchecked_t *l) ++{ ++ atomic_unchecked_t *v = (atomic_unchecked_t *)l; ++ ++ atomic_sub_unchecked(i, v); ++} ++#endif ++ + static inline int atomic_long_sub_and_test(long i, atomic_long_t *l) + { + atomic_t *v = (atomic_t *)l; +@@ -232,6 +361,15 @@ static inline long atomic_long_inc_retur + return (long)atomic_inc_return(v); + } + ++#ifdef CONFIG_PAX_REFCOUNT ++static inline long atomic_long_inc_return_unchecked(atomic_long_unchecked_t *l) ++{ ++ atomic_unchecked_t *v = (atomic_unchecked_t *)l; ++ ++ return (long)atomic_inc_return_unchecked(v); ++} ++#endif ++ + static inline long atomic_long_dec_return(atomic_long_t *l) + { + atomic_t *v = (atomic_t *)l; +@@ -255,4 +393,41 @@ static inline long atomic_long_add_unles + + #endif /* BITS_PER_LONG == 64 */ + ++#ifdef CONFIG_PAX_REFCOUNT ++static inline void pax_refcount_needs_these_functions(void) ++{ ++ atomic_read_unchecked((atomic_unchecked_t *)NULL); ++ atomic_set_unchecked((atomic_unchecked_t *)NULL, 0); ++ atomic_add_unchecked(0, (atomic_unchecked_t *)NULL); ++ atomic_sub_unchecked(0, (atomic_unchecked_t *)NULL); ++ atomic_inc_unchecked((atomic_unchecked_t *)NULL); ++ atomic_inc_return_unchecked((atomic_unchecked_t *)NULL); ++ atomic_add_return_unchecked(0, (atomic_unchecked_t *)NULL); ++ ++ atomic_long_read_unchecked((atomic_long_unchecked_t *)NULL); ++ atomic_long_set_unchecked((atomic_long_unchecked_t *)NULL, 0); ++ atomic_long_add_unchecked(0, (atomic_long_unchecked_t *)NULL); ++ atomic_long_sub_unchecked(0, (atomic_long_unchecked_t *)NULL); ++ atomic_long_inc_unchecked((atomic_long_unchecked_t *)NULL); ++ atomic_long_inc_return_unchecked((atomic_long_unchecked_t *)NULL); ++ atomic_long_dec_unchecked((atomic_long_unchecked_t *)NULL); ++} ++#else ++#define atomic_read_unchecked(v) atomic_read(v) ++#define atomic_set_unchecked(v, i) atomic_set((v), (i)) ++#define atomic_add_unchecked(i, v) atomic_add((i), (v)) ++#define atomic_sub_unchecked(i, v) atomic_sub((i), (v)) ++#define atomic_inc_unchecked(v) atomic_inc(v) ++#define atomic_inc_return_unchecked(v) atomic_inc_return(v) ++#define atomic_add_return_unchecked(i, v) atomic_add_return((i), (v)) ++ ++#define atomic_long_read_unchecked(v) atomic_long_read(v) ++#define atomic_long_set_unchecked(v, i) atomic_long_set((v), (i)) ++#define atomic_long_add_unchecked(i, v) atomic_long_add((i), (v)) ++#define atomic_long_sub_unchecked(i, v) atomic_long_sub((i), (v)) ++#define atomic_long_inc_unchecked(v) atomic_long_inc(v) ++#define atomic_long_inc_return_unchecked(v) atomic_long_inc_return(v) ++#define atomic_long_dec_unchecked(v) atomic_long_dec(v) ++#endif ++ + #endif /* _ASM_GENERIC_ATOMIC_LONG_H */ +diff -urNp linux-2.6.38.3/include/asm-generic/dma-mapping-common.h linux-2.6.38.3/include/asm-generic/dma-mapping-common.h +--- linux-2.6.38.3/include/asm-generic/dma-mapping-common.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/asm-generic/dma-mapping-common.h 2011-04-17 15:57:32.000000000 -0400 +@@ -11,7 +11,7 @@ static inline dma_addr_t dma_map_single_ + enum dma_data_direction dir, + struct dma_attrs *attrs) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + dma_addr_t addr; + + kmemcheck_mark_initialized(ptr, size); +@@ -30,7 +30,7 @@ static inline void dma_unmap_single_attr + enum dma_data_direction dir, + struct dma_attrs *attrs) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + + BUG_ON(!valid_dma_direction(dir)); + if (ops->unmap_page) +@@ -42,7 +42,7 @@ static inline int dma_map_sg_attrs(struc + int nents, enum dma_data_direction dir, + struct dma_attrs *attrs) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + int i, ents; + struct scatterlist *s; + +@@ -59,7 +59,7 @@ static inline void dma_unmap_sg_attrs(st + int nents, enum dma_data_direction dir, + struct dma_attrs *attrs) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + + BUG_ON(!valid_dma_direction(dir)); + debug_dma_unmap_sg(dev, sg, nents, dir); +@@ -71,7 +71,7 @@ static inline dma_addr_t dma_map_page(st + size_t offset, size_t size, + enum dma_data_direction dir) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + dma_addr_t addr; + + kmemcheck_mark_initialized(page_address(page) + offset, size); +@@ -85,7 +85,7 @@ static inline dma_addr_t dma_map_page(st + static inline void dma_unmap_page(struct device *dev, dma_addr_t addr, + size_t size, enum dma_data_direction dir) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + + BUG_ON(!valid_dma_direction(dir)); + if (ops->unmap_page) +@@ -97,7 +97,7 @@ static inline void dma_sync_single_for_c + size_t size, + enum dma_data_direction dir) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + + BUG_ON(!valid_dma_direction(dir)); + if (ops->sync_single_for_cpu) +@@ -109,7 +109,7 @@ static inline void dma_sync_single_for_d + dma_addr_t addr, size_t size, + enum dma_data_direction dir) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + + BUG_ON(!valid_dma_direction(dir)); + if (ops->sync_single_for_device) +@@ -139,7 +139,7 @@ static inline void + dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, + int nelems, enum dma_data_direction dir) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + + BUG_ON(!valid_dma_direction(dir)); + if (ops->sync_sg_for_cpu) +@@ -151,7 +151,7 @@ static inline void + dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, + int nelems, enum dma_data_direction dir) + { +- struct dma_map_ops *ops = get_dma_ops(dev); ++ const struct dma_map_ops *ops = get_dma_ops(dev); + + BUG_ON(!valid_dma_direction(dir)); + if (ops->sync_sg_for_device) +diff -urNp linux-2.6.38.3/include/asm-generic/futex.h linux-2.6.38.3/include/asm-generic/futex.h +--- linux-2.6.38.3/include/asm-generic/futex.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/asm-generic/futex.h 2011-04-17 15:57:32.000000000 -0400 +@@ -6,7 +6,7 @@ + #include <asm/errno.h> + + static inline int +-futex_atomic_op_inuser (int encoded_op, int __user *uaddr) ++futex_atomic_op_inuser (int encoded_op, u32 __user *uaddr) + { + int op = (encoded_op >> 28) & 7; + int cmp = (encoded_op >> 24) & 15; +@@ -48,7 +48,7 @@ futex_atomic_op_inuser (int encoded_op, + } + + static inline int +-futex_atomic_cmpxchg_inatomic(int __user *uaddr, int oldval, int newval) ++futex_atomic_cmpxchg_inatomic(u32 __user *uaddr, int oldval, int newval) + { + return -ENOSYS; + } +diff -urNp linux-2.6.38.3/include/asm-generic/int-l64.h linux-2.6.38.3/include/asm-generic/int-l64.h +--- linux-2.6.38.3/include/asm-generic/int-l64.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/asm-generic/int-l64.h 2011-04-17 15:57:32.000000000 -0400 +@@ -46,6 +46,8 @@ typedef unsigned int u32; + typedef signed long s64; + typedef unsigned long u64; + ++typedef unsigned int intoverflow_t __attribute__ ((mode(TI))); ++ + #define S8_C(x) x + #define U8_C(x) x ## U + #define S16_C(x) x +diff -urNp linux-2.6.38.3/include/asm-generic/int-ll64.h linux-2.6.38.3/include/asm-generic/int-ll64.h +--- linux-2.6.38.3/include/asm-generic/int-ll64.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/asm-generic/int-ll64.h 2011-04-17 15:57:32.000000000 -0400 +@@ -51,6 +51,8 @@ typedef unsigned int u32; + typedef signed long long s64; + typedef unsigned long long u64; + ++typedef unsigned long long intoverflow_t; ++ + #define S8_C(x) x + #define U8_C(x) x ## U + #define S16_C(x) x +diff -urNp linux-2.6.38.3/include/asm-generic/kmap_types.h linux-2.6.38.3/include/asm-generic/kmap_types.h +--- linux-2.6.38.3/include/asm-generic/kmap_types.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/asm-generic/kmap_types.h 2011-04-17 15:57:32.000000000 -0400 +@@ -29,10 +29,11 @@ KMAP_D(16) KM_IRQ_PTE, + KMAP_D(17) KM_NMI, + KMAP_D(18) KM_NMI_PTE, + KMAP_D(19) KM_KDB, ++KMAP_D(20) KM_CLEARPAGE, + /* + * Remember to update debug_kmap_atomic() when adding new kmap types! + */ +-KMAP_D(20) KM_TYPE_NR ++KMAP_D(21) KM_TYPE_NR + }; + + #undef KMAP_D +diff -urNp linux-2.6.38.3/include/asm-generic/pgtable.h linux-2.6.38.3/include/asm-generic/pgtable.h +--- linux-2.6.38.3/include/asm-generic/pgtable.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/asm-generic/pgtable.h 2011-04-17 15:57:32.000000000 -0400 +@@ -447,6 +447,14 @@ static inline int pmd_write(pmd_t pmd) + #endif /* __HAVE_ARCH_PMD_WRITE */ + #endif + ++#ifndef __HAVE_ARCH_PAX_OPEN_KERNEL ++static inline unsigned long pax_open_kernel(void) { return 0; } ++#endif ++ ++#ifndef __HAVE_ARCH_PAX_CLOSE_KERNEL ++static inline unsigned long pax_close_kernel(void) { return 0; } ++#endif ++ + #endif /* !__ASSEMBLY__ */ + + #endif /* _ASM_GENERIC_PGTABLE_H */ +diff -urNp linux-2.6.38.3/include/asm-generic/pgtable-nopmd.h linux-2.6.38.3/include/asm-generic/pgtable-nopmd.h +--- linux-2.6.38.3/include/asm-generic/pgtable-nopmd.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/asm-generic/pgtable-nopmd.h 2011-04-17 15:57:32.000000000 -0400 +@@ -1,14 +1,19 @@ + #ifndef _PGTABLE_NOPMD_H + #define _PGTABLE_NOPMD_H + +-#ifndef __ASSEMBLY__ +- + #include <asm-generic/pgtable-nopud.h> + +-struct mm_struct; +- + #define __PAGETABLE_PMD_FOLDED + ++#define PMD_SHIFT PUD_SHIFT ++#define PTRS_PER_PMD 1 ++#define PMD_SIZE (_AC(1,UL) << PMD_SHIFT) ++#define PMD_MASK (~(PMD_SIZE-1)) ++ ++#ifndef __ASSEMBLY__ ++ ++struct mm_struct; ++ + /* + * Having the pmd type consist of a pud gets the size right, and allows + * us to conceptually access the pud entry that this pmd is folded into +@@ -16,11 +21,6 @@ struct mm_struct; + */ + typedef struct { pud_t pud; } pmd_t; + +-#define PMD_SHIFT PUD_SHIFT +-#define PTRS_PER_PMD 1 +-#define PMD_SIZE (1UL << PMD_SHIFT) +-#define PMD_MASK (~(PMD_SIZE-1)) +- + /* + * The "pud_xxx()" functions here are trivial for a folded two-level + * setup: the pmd is never bad, and a pmd always exists (as it's folded +diff -urNp linux-2.6.38.3/include/asm-generic/pgtable-nopud.h linux-2.6.38.3/include/asm-generic/pgtable-nopud.h +--- linux-2.6.38.3/include/asm-generic/pgtable-nopud.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/asm-generic/pgtable-nopud.h 2011-04-17 15:57:32.000000000 -0400 +@@ -1,10 +1,15 @@ + #ifndef _PGTABLE_NOPUD_H + #define _PGTABLE_NOPUD_H + +-#ifndef __ASSEMBLY__ +- + #define __PAGETABLE_PUD_FOLDED + ++#define PUD_SHIFT PGDIR_SHIFT ++#define PTRS_PER_PUD 1 ++#define PUD_SIZE (_AC(1,UL) << PUD_SHIFT) ++#define PUD_MASK (~(PUD_SIZE-1)) ++ ++#ifndef __ASSEMBLY__ ++ + /* + * Having the pud type consist of a pgd gets the size right, and allows + * us to conceptually access the pgd entry that this pud is folded into +@@ -12,11 +17,6 @@ + */ + typedef struct { pgd_t pgd; } pud_t; + +-#define PUD_SHIFT PGDIR_SHIFT +-#define PTRS_PER_PUD 1 +-#define PUD_SIZE (1UL << PUD_SHIFT) +-#define PUD_MASK (~(PUD_SIZE-1)) +- + /* + * The "pgd_xxx()" functions here are trivial for a folded two-level + * setup: the pud is never bad, and a pud always exists (as it's folded +diff -urNp linux-2.6.38.3/include/asm-generic/vmlinux.lds.h linux-2.6.38.3/include/asm-generic/vmlinux.lds.h +--- linux-2.6.38.3/include/asm-generic/vmlinux.lds.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/asm-generic/vmlinux.lds.h 2011-04-17 15:57:32.000000000 -0400 +@@ -213,6 +213,7 @@ + .rodata : AT(ADDR(.rodata) - LOAD_OFFSET) { \ + VMLINUX_SYMBOL(__start_rodata) = .; \ + *(.rodata) *(.rodata.*) \ ++ *(.data..read_only) \ + *(__vermagic) /* Kernel version magic */ \ + . = ALIGN(8); \ + VMLINUX_SYMBOL(__start___tracepoints_ptrs) = .; \ +@@ -696,14 +697,15 @@ + * section in the linker script will go there too. @phdr should have + * a leading colon. + * +- * Note that this macros defines __per_cpu_load as an absolute symbol. ++ * Note that this macros defines per_cpu_load as an absolute symbol. + * If there is no need to put the percpu section at a predetermined + * address, use PERCPU(). + */ + #define PERCPU_VADDR(vaddr, phdr) \ +- VMLINUX_SYMBOL(__per_cpu_load) = .; \ +- .data..percpu vaddr : AT(VMLINUX_SYMBOL(__per_cpu_load) \ ++ per_cpu_load = .; \ ++ .data..percpu vaddr : AT(VMLINUX_SYMBOL(per_cpu_load) \ + - LOAD_OFFSET) { \ ++ VMLINUX_SYMBOL(__per_cpu_load) = . + per_cpu_load; \ + VMLINUX_SYMBOL(__per_cpu_start) = .; \ + *(.data..percpu..first) \ + . = ALIGN(PAGE_SIZE); \ +@@ -713,7 +715,7 @@ + *(.data..percpu..shared_aligned) \ + VMLINUX_SYMBOL(__per_cpu_end) = .; \ + } phdr \ +- . = VMLINUX_SYMBOL(__per_cpu_load) + SIZEOF(.data..percpu); ++ . = VMLINUX_SYMBOL(per_cpu_load) + SIZEOF(.data..percpu); + + /** + * PERCPU - define output section for percpu area, simple version +diff -urNp linux-2.6.38.3/include/drm/drm_pciids.h linux-2.6.38.3/include/drm/drm_pciids.h +--- linux-2.6.38.3/include/drm/drm_pciids.h 2011-04-17 16:53:01.000000000 -0400 ++++ linux-2.6.38.3/include/drm/drm_pciids.h 2011-04-17 16:58:34.000000000 -0400 +@@ -460,7 +460,7 @@ + {0x1002, 0x9805, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PALM|RADEON_NEW_MEMMAP|RADEON_IS_IGP}, \ + {0x1002, 0x9806, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PALM|RADEON_NEW_MEMMAP|RADEON_IS_IGP}, \ + {0x1002, 0x9807, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PALM|RADEON_NEW_MEMMAP|RADEON_IS_IGP}, \ +- {0, 0, 0} ++ {0, 0, 0, 0, 0, 0} + + #define r128_PCI_IDS \ + {0x1002, 0x4c45, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ +@@ -500,14 +500,14 @@ + {0x1002, 0x5446, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ + {0x1002, 0x544C, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ + {0x1002, 0x5452, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ +- {0, 0, 0} ++ {0, 0, 0, 0, 0, 0} + + #define mga_PCI_IDS \ + {0x102b, 0x0520, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MGA_CARD_TYPE_G200}, \ + {0x102b, 0x0521, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MGA_CARD_TYPE_G200}, \ + {0x102b, 0x0525, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MGA_CARD_TYPE_G400}, \ + {0x102b, 0x2527, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MGA_CARD_TYPE_G550}, \ +- {0, 0, 0} ++ {0, 0, 0, 0, 0, 0} + + #define mach64_PCI_IDS \ + {0x1002, 0x4749, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ +@@ -530,7 +530,7 @@ + {0x1002, 0x4c53, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ + {0x1002, 0x4c4d, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ + {0x1002, 0x4c4e, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ +- {0, 0, 0} ++ {0, 0, 0, 0, 0, 0} + + #define sisdrv_PCI_IDS \ + {0x1039, 0x0300, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ +@@ -541,7 +541,7 @@ + {0x1039, 0x7300, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ + {0x18CA, 0x0040, PCI_ANY_ID, PCI_ANY_ID, 0, 0, SIS_CHIP_315}, \ + {0x18CA, 0x0042, PCI_ANY_ID, PCI_ANY_ID, 0, 0, SIS_CHIP_315}, \ +- {0, 0, 0} ++ {0, 0, 0, 0, 0, 0} + + #define tdfx_PCI_IDS \ + {0x121a, 0x0003, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ +@@ -550,7 +550,7 @@ + {0x121a, 0x0007, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ + {0x121a, 0x0009, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ + {0x121a, 0x000b, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ +- {0, 0, 0} ++ {0, 0, 0, 0, 0, 0} + + #define viadrv_PCI_IDS \ + {0x1106, 0x3022, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ +@@ -562,14 +562,14 @@ + {0x1106, 0x3343, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ + {0x1106, 0x3230, PCI_ANY_ID, PCI_ANY_ID, 0, 0, VIA_DX9_0}, \ + {0x1106, 0x3157, PCI_ANY_ID, PCI_ANY_ID, 0, 0, VIA_PRO_GROUP_A}, \ +- {0, 0, 0} ++ {0, 0, 0, 0, 0, 0} + + #define i810_PCI_IDS \ + {0x8086, 0x7121, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ + {0x8086, 0x7123, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ + {0x8086, 0x7125, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ + {0x8086, 0x1132, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ +- {0, 0, 0} ++ {0, 0, 0, 0, 0, 0} + + #define i830_PCI_IDS \ + {0x8086, 0x3577, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ +@@ -577,11 +577,11 @@ + {0x8086, 0x3582, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ + {0x8086, 0x2572, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ + {0x8086, 0x358e, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ +- {0, 0, 0} ++ {0, 0, 0, 0, 0, 0} + + #define gamma_PCI_IDS \ + {0x3d3d, 0x0008, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \ +- {0, 0, 0} ++ {0, 0, 0, 0, 0, 0} + + #define savage_PCI_IDS \ + {0x5333, 0x8a20, PCI_ANY_ID, PCI_ANY_ID, 0, 0, S3_SAVAGE3D}, \ +@@ -607,10 +607,10 @@ + {0x5333, 0x8d02, PCI_ANY_ID, PCI_ANY_ID, 0, 0, S3_TWISTER}, \ + {0x5333, 0x8d03, PCI_ANY_ID, PCI_ANY_ID, 0, 0, S3_PROSAVAGEDDR}, \ + {0x5333, 0x8d04, PCI_ANY_ID, PCI_ANY_ID, 0, 0, S3_PROSAVAGEDDR}, \ +- {0, 0, 0} ++ {0, 0, 0, 0, 0, 0} + + #define ffb_PCI_IDS \ +- {0, 0, 0} ++ {0, 0, 0, 0, 0, 0} + + #define i915_PCI_IDS \ + {0x8086, 0x3577, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \ +@@ -644,4 +644,4 @@ + {0x8086, 0x0042, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \ + {0x8086, 0x0046, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \ + {0x8086, 0x0102, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \ +- {0, 0, 0} ++ {0, 0, 0, 0, 0, 0} +diff -urNp linux-2.6.38.3/include/drm/drmP.h linux-2.6.38.3/include/drm/drmP.h +--- linux-2.6.38.3/include/drm/drmP.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/drm/drmP.h 2011-04-17 15:57:32.000000000 -0400 +@@ -73,6 +73,7 @@ + #include <linux/workqueue.h> + #include <linux/poll.h> + #include <asm/pgalloc.h> ++#include <asm/local.h> + #include "drm.h" + + #include <linux/idr.h> +@@ -881,7 +882,7 @@ struct drm_driver { + void (*vgaarb_irq)(struct drm_device *dev, bool state); + + /* Driver private ops for this object */ +- struct vm_operations_struct *gem_vm_ops; ++ const struct vm_operations_struct *gem_vm_ops; + + int major; + int minor; +@@ -894,7 +895,7 @@ struct drm_driver { + int dev_priv_size; + struct drm_ioctl_desc *ioctls; + int num_ioctls; +- struct file_operations fops; ++ const struct file_operations fops; + struct pci_driver pci_driver; + struct platform_device *platform_device; + /* List of devices hanging off this driver */ +@@ -991,7 +992,7 @@ struct drm_device { + + /** \name Usage Counters */ + /*@{ */ +- int open_count; /**< Outstanding files open */ ++ local_t open_count; /**< Outstanding files open */ + atomic_t ioctl_count; /**< Outstanding IOCTLs pending */ + atomic_t vma_count; /**< Outstanding vma areas open */ + int buf_use; /**< Buffers in use -- cannot alloc */ +@@ -1002,7 +1003,7 @@ struct drm_device { + /*@{ */ + unsigned long counters; + enum drm_stat_type types[15]; +- atomic_t counts[15]; ++ atomic_unchecked_t counts[15]; + /*@} */ + + struct list_head filelist; +@@ -1101,7 +1102,7 @@ struct drm_device { + struct platform_device *platformdev; /**< Platform device struture */ + + struct drm_sg_mem *sg; /**< Scatter gather memory */ +- unsigned int num_crtcs; /**< Number of CRTCs on this device */ ++ unsigned int num_crtcs; /**< Number of CRTCs on this device */ + void *dev_private; /**< device private data */ + void *mm_private; + struct address_space *dev_mapping; +diff -urNp linux-2.6.38.3/include/linux/a.out.h linux-2.6.38.3/include/linux/a.out.h +--- linux-2.6.38.3/include/linux/a.out.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/a.out.h 2011-04-17 15:57:32.000000000 -0400 +@@ -39,6 +39,14 @@ enum machine_type { + M_MIPS2 = 152 /* MIPS R6000/R4000 binary */ + }; + ++/* Constants for the N_FLAGS field */ ++#define F_PAX_PAGEEXEC 1 /* Paging based non-executable pages */ ++#define F_PAX_EMUTRAMP 2 /* Emulate trampolines */ ++#define F_PAX_MPROTECT 4 /* Restrict mprotect() */ ++#define F_PAX_RANDMMAP 8 /* Randomize mmap() base */ ++/*#define F_PAX_RANDEXEC 16*/ /* Randomize ET_EXEC base */ ++#define F_PAX_SEGMEXEC 32 /* Segmentation based non-executable pages */ ++ + #if !defined (N_MAGIC) + #define N_MAGIC(exec) ((exec).a_info & 0xffff) + #endif +diff -urNp linux-2.6.38.3/include/linux/atmdev.h linux-2.6.38.3/include/linux/atmdev.h +--- linux-2.6.38.3/include/linux/atmdev.h 2011-04-17 16:53:01.000000000 -0400 ++++ linux-2.6.38.3/include/linux/atmdev.h 2011-04-17 16:53:48.000000000 -0400 +@@ -237,7 +237,7 @@ struct compat_atm_iobuf { + #endif + + struct k_atm_aal_stats { +-#define __HANDLE_ITEM(i) atomic_t i ++#define __HANDLE_ITEM(i) atomic_unchecked_t i + __AAL_STAT_ITEMS + #undef __HANDLE_ITEM + }; +diff -urNp linux-2.6.38.3/include/linux/binfmts.h linux-2.6.38.3/include/linux/binfmts.h +--- linux-2.6.38.3/include/linux/binfmts.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/binfmts.h 2011-04-17 15:57:32.000000000 -0400 +@@ -92,6 +92,7 @@ struct linux_binfmt { + int (*load_binary)(struct linux_binprm *, struct pt_regs * regs); + int (*load_shlib)(struct file *); + int (*core_dump)(struct coredump_params *cprm); ++ void (*handle_mprotect)(struct vm_area_struct *vma, unsigned long newflags); + unsigned long min_coredump; /* minimal dump size */ + }; + +diff -urNp linux-2.6.38.3/include/linux/blkdev.h linux-2.6.38.3/include/linux/blkdev.h +--- linux-2.6.38.3/include/linux/blkdev.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/blkdev.h 2011-04-17 15:57:32.000000000 -0400 +@@ -1247,22 +1247,22 @@ queue_max_integrity_segments(struct requ + #endif /* CONFIG_BLK_DEV_INTEGRITY */ + + struct block_device_operations { +- int (*open) (struct block_device *, fmode_t); +- int (*release) (struct gendisk *, fmode_t); +- int (*ioctl) (struct block_device *, fmode_t, unsigned, unsigned long); +- int (*compat_ioctl) (struct block_device *, fmode_t, unsigned, unsigned long); +- int (*direct_access) (struct block_device *, sector_t, ++ int (* const open) (struct block_device *, fmode_t); ++ int (* const release) (struct gendisk *, fmode_t); ++ int (* const ioctl) (struct block_device *, fmode_t, unsigned, unsigned long); ++ int (* const compat_ioctl) (struct block_device *, fmode_t, unsigned, unsigned long); ++ int (* const direct_access) (struct block_device *, sector_t, + void **, unsigned long *); +- unsigned int (*check_events) (struct gendisk *disk, ++ unsigned int (* const check_events) (struct gendisk *disk, + unsigned int clearing); + /* ->media_changed() is DEPRECATED, use ->check_events() instead */ +- int (*media_changed) (struct gendisk *); +- void (*unlock_native_capacity) (struct gendisk *); +- int (*revalidate_disk) (struct gendisk *); +- int (*getgeo)(struct block_device *, struct hd_geometry *); ++ int (* const media_changed) (struct gendisk *); ++ void (* const unlock_native_capacity) (struct gendisk *); ++ int (* const revalidate_disk) (struct gendisk *); ++ int (* const getgeo)(struct block_device *, struct hd_geometry *); + /* this callback is with swap_lock and sometimes page table lock held */ +- void (*swap_slot_free_notify) (struct block_device *, unsigned long); +- struct module *owner; ++ void (* const swap_slot_free_notify) (struct block_device *, unsigned long); ++ struct module * const owner; + }; + + extern int __blkdev_driver_ioctl(struct block_device *, fmode_t, unsigned int, +diff -urNp linux-2.6.38.3/include/linux/byteorder/little_endian.h linux-2.6.38.3/include/linux/byteorder/little_endian.h +--- linux-2.6.38.3/include/linux/byteorder/little_endian.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/byteorder/little_endian.h 2011-04-17 15:57:32.000000000 -0400 +@@ -42,51 +42,51 @@ + + static inline __le64 __cpu_to_le64p(const __u64 *p) + { +- return (__force __le64)*p; ++ return (__force const __le64)*p; + } + static inline __u64 __le64_to_cpup(const __le64 *p) + { +- return (__force __u64)*p; ++ return (__force const __u64)*p; + } + static inline __le32 __cpu_to_le32p(const __u32 *p) + { +- return (__force __le32)*p; ++ return (__force const __le32)*p; + } + static inline __u32 __le32_to_cpup(const __le32 *p) + { +- return (__force __u32)*p; ++ return (__force const __u32)*p; + } + static inline __le16 __cpu_to_le16p(const __u16 *p) + { +- return (__force __le16)*p; ++ return (__force const __le16)*p; + } + static inline __u16 __le16_to_cpup(const __le16 *p) + { +- return (__force __u16)*p; ++ return (__force const __u16)*p; + } + static inline __be64 __cpu_to_be64p(const __u64 *p) + { +- return (__force __be64)__swab64p(p); ++ return (__force const __be64)__swab64p(p); + } + static inline __u64 __be64_to_cpup(const __be64 *p) + { +- return __swab64p((__u64 *)p); ++ return __swab64p((const __u64 *)p); + } + static inline __be32 __cpu_to_be32p(const __u32 *p) + { +- return (__force __be32)__swab32p(p); ++ return (__force const __be32)__swab32p(p); + } + static inline __u32 __be32_to_cpup(const __be32 *p) + { +- return __swab32p((__u32 *)p); ++ return __swab32p((const __u32 *)p); + } + static inline __be16 __cpu_to_be16p(const __u16 *p) + { +- return (__force __be16)__swab16p(p); ++ return (__force const __be16)__swab16p(p); + } + static inline __u16 __be16_to_cpup(const __be16 *p) + { +- return __swab16p((__u16 *)p); ++ return __swab16p((const __u16 *)p); + } + #define __cpu_to_le64s(x) do { (void)(x); } while (0) + #define __le64_to_cpus(x) do { (void)(x); } while (0) +diff -urNp linux-2.6.38.3/include/linux/cache.h linux-2.6.38.3/include/linux/cache.h +--- linux-2.6.38.3/include/linux/cache.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/cache.h 2011-04-17 15:57:32.000000000 -0400 +@@ -16,6 +16,10 @@ + #define __read_mostly + #endif + ++#ifndef __read_only ++#define __read_only __read_mostly ++#endif ++ + #ifndef ____cacheline_aligned + #define ____cacheline_aligned __attribute__((__aligned__(SMP_CACHE_BYTES))) + #endif +diff -urNp linux-2.6.38.3/include/linux/capability.h linux-2.6.38.3/include/linux/capability.h +--- linux-2.6.38.3/include/linux/capability.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/capability.h 2011-04-17 15:57:32.000000000 -0400 +@@ -561,6 +561,7 @@ extern const kernel_cap_t __cap_init_eff + (security_real_capable_noaudit((t), (cap)) == 0) + + extern int capable(int cap); ++int capable_nolog(int cap); + + /* audit system wants to get cap info from files as well */ + struct dentry; +diff -urNp linux-2.6.38.3/include/linux/compiler-gcc4.h linux-2.6.38.3/include/linux/compiler-gcc4.h +--- linux-2.6.38.3/include/linux/compiler-gcc4.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/compiler-gcc4.h 2011-04-17 15:57:32.000000000 -0400 +@@ -54,6 +54,10 @@ + + #endif + ++#define __alloc_size(...) __attribute((alloc_size(__VA_ARGS__))) ++#define __bos(ptr, arg) __builtin_object_size((ptr), (arg)) ++#define __bos0(ptr) __bos((ptr), 0) ++#define __bos1(ptr) __bos((ptr), 1) + #endif + + #if __GNUC_MINOR__ > 0 +diff -urNp linux-2.6.38.3/include/linux/compiler.h linux-2.6.38.3/include/linux/compiler.h +--- linux-2.6.38.3/include/linux/compiler.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/compiler.h 2011-04-17 15:57:32.000000000 -0400 +@@ -273,6 +273,22 @@ void ftrace_likely_update(struct ftrace_ + #define __cold + #endif + ++#ifndef __alloc_size ++#define __alloc_size ++#endif ++ ++#ifndef __bos ++#define __bos ++#endif ++ ++#ifndef __bos0 ++#define __bos0 ++#endif ++ ++#ifndef __bos1 ++#define __bos1 ++#endif ++ + /* Simple shorthand for a section definition */ + #ifndef __section + # define __section(S) __attribute__ ((__section__(#S))) +@@ -306,6 +322,7 @@ void ftrace_likely_update(struct ftrace_ + * use is to mediate communication between process-level code and irq/NMI + * handlers, all running on the same CPU. + */ +-#define ACCESS_ONCE(x) (*(volatile typeof(x) *)&(x)) ++#define ACCESS_ONCE(x) (*(volatile const typeof(x) *)&(x)) ++#define ACCESS_ONCE_RW(x) (*(volatile typeof(x) *)&(x)) + + #endif /* __LINUX_COMPILER_H */ +diff -urNp linux-2.6.38.3/include/linux/cpuset.h linux-2.6.38.3/include/linux/cpuset.h +--- linux-2.6.38.3/include/linux/cpuset.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/cpuset.h 2011-04-17 15:57:32.000000000 -0400 +@@ -118,7 +118,7 @@ static inline void put_mems_allowed(void + * nodemask. + */ + smp_mb(); +- --ACCESS_ONCE(current->mems_allowed_change_disable); ++ --ACCESS_ONCE_RW(current->mems_allowed_change_disable); + } + + static inline void set_mems_allowed(nodemask_t nodemask) +diff -urNp linux-2.6.38.3/include/linux/decompress/mm.h linux-2.6.38.3/include/linux/decompress/mm.h +--- linux-2.6.38.3/include/linux/decompress/mm.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/decompress/mm.h 2011-04-17 15:57:32.000000000 -0400 +@@ -77,7 +77,7 @@ static void free(void *where) + * warnings when not needed (indeed large_malloc / large_free are not + * needed by inflate */ + +-#define malloc(a) kmalloc(a, GFP_KERNEL) ++#define malloc(a) kmalloc((a), GFP_KERNEL) + #define free(a) kfree(a) + + #define large_malloc(a) vmalloc(a) +diff -urNp linux-2.6.38.3/include/linux/dma-mapping.h linux-2.6.38.3/include/linux/dma-mapping.h +--- linux-2.6.38.3/include/linux/dma-mapping.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/dma-mapping.h 2011-04-17 15:57:32.000000000 -0400 +@@ -16,40 +16,40 @@ enum dma_data_direction { + }; + + struct dma_map_ops { +- void* (*alloc_coherent)(struct device *dev, size_t size, ++ void* (* const alloc_coherent)(struct device *dev, size_t size, + dma_addr_t *dma_handle, gfp_t gfp); +- void (*free_coherent)(struct device *dev, size_t size, ++ void (* const free_coherent)(struct device *dev, size_t size, + void *vaddr, dma_addr_t dma_handle); +- dma_addr_t (*map_page)(struct device *dev, struct page *page, ++ dma_addr_t (* const map_page)(struct device *dev, struct page *page, + unsigned long offset, size_t size, + enum dma_data_direction dir, + struct dma_attrs *attrs); +- void (*unmap_page)(struct device *dev, dma_addr_t dma_handle, ++ void (* const unmap_page)(struct device *dev, dma_addr_t dma_handle, + size_t size, enum dma_data_direction dir, + struct dma_attrs *attrs); +- int (*map_sg)(struct device *dev, struct scatterlist *sg, ++ int (* const map_sg)(struct device *dev, struct scatterlist *sg, + int nents, enum dma_data_direction dir, + struct dma_attrs *attrs); +- void (*unmap_sg)(struct device *dev, ++ void (* const unmap_sg)(struct device *dev, + struct scatterlist *sg, int nents, + enum dma_data_direction dir, + struct dma_attrs *attrs); +- void (*sync_single_for_cpu)(struct device *dev, ++ void (* const sync_single_for_cpu)(struct device *dev, + dma_addr_t dma_handle, size_t size, + enum dma_data_direction dir); +- void (*sync_single_for_device)(struct device *dev, ++ void (* const sync_single_for_device)(struct device *dev, + dma_addr_t dma_handle, size_t size, + enum dma_data_direction dir); +- void (*sync_sg_for_cpu)(struct device *dev, ++ void (* const sync_sg_for_cpu)(struct device *dev, + struct scatterlist *sg, int nents, + enum dma_data_direction dir); +- void (*sync_sg_for_device)(struct device *dev, ++ void (* const sync_sg_for_device)(struct device *dev, + struct scatterlist *sg, int nents, + enum dma_data_direction dir); +- int (*mapping_error)(struct device *dev, dma_addr_t dma_addr); +- int (*dma_supported)(struct device *dev, u64 mask); +- int (*set_dma_mask)(struct device *dev, u64 mask); +- int is_phys; ++ int (* const mapping_error)(struct device *dev, dma_addr_t dma_addr); ++ int (* const dma_supported)(struct device *dev, u64 mask); ++ int (* set_dma_mask)(struct device *dev, u64 mask); ++ const int is_phys; + }; + + #define DMA_BIT_MASK(n) (((n) == 64) ? ~0ULL : ((1ULL<<(n))-1)) +diff -urNp linux-2.6.38.3/include/linux/elf.h linux-2.6.38.3/include/linux/elf.h +--- linux-2.6.38.3/include/linux/elf.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/elf.h 2011-04-17 15:57:32.000000000 -0400 +@@ -49,6 +49,17 @@ typedef __s64 Elf64_Sxword; + #define PT_GNU_EH_FRAME 0x6474e550 + + #define PT_GNU_STACK (PT_LOOS + 0x474e551) ++#define PT_GNU_RELRO (PT_LOOS + 0x474e552) ++ ++#define PT_PAX_FLAGS (PT_LOOS + 0x5041580) ++ ++/* Constants for the e_flags field */ ++#define EF_PAX_PAGEEXEC 1 /* Paging based non-executable pages */ ++#define EF_PAX_EMUTRAMP 2 /* Emulate trampolines */ ++#define EF_PAX_MPROTECT 4 /* Restrict mprotect() */ ++#define EF_PAX_RANDMMAP 8 /* Randomize mmap() base */ ++/*#define EF_PAX_RANDEXEC 16*/ /* Randomize ET_EXEC base */ ++#define EF_PAX_SEGMEXEC 32 /* Segmentation based non-executable pages */ + + /* + * Extended Numbering +@@ -106,6 +117,8 @@ typedef __s64 Elf64_Sxword; + #define DT_DEBUG 21 + #define DT_TEXTREL 22 + #define DT_JMPREL 23 ++#define DT_FLAGS 30 ++ #define DF_TEXTREL 0x00000004 + #define DT_ENCODING 32 + #define OLD_DT_LOOS 0x60000000 + #define DT_LOOS 0x6000000d +@@ -252,6 +265,19 @@ typedef struct elf64_hdr { + #define PF_W 0x2 + #define PF_X 0x1 + ++#define PF_PAGEEXEC (1U << 4) /* Enable PAGEEXEC */ ++#define PF_NOPAGEEXEC (1U << 5) /* Disable PAGEEXEC */ ++#define PF_SEGMEXEC (1U << 6) /* Enable SEGMEXEC */ ++#define PF_NOSEGMEXEC (1U << 7) /* Disable SEGMEXEC */ ++#define PF_MPROTECT (1U << 8) /* Enable MPROTECT */ ++#define PF_NOMPROTECT (1U << 9) /* Disable MPROTECT */ ++/*#define PF_RANDEXEC (1U << 10)*/ /* Enable RANDEXEC */ ++/*#define PF_NORANDEXEC (1U << 11)*/ /* Disable RANDEXEC */ ++#define PF_EMUTRAMP (1U << 12) /* Enable EMUTRAMP */ ++#define PF_NOEMUTRAMP (1U << 13) /* Disable EMUTRAMP */ ++#define PF_RANDMMAP (1U << 14) /* Enable RANDMMAP */ ++#define PF_NORANDMMAP (1U << 15) /* Disable RANDMMAP */ ++ + typedef struct elf32_phdr{ + Elf32_Word p_type; + Elf32_Off p_offset; +@@ -344,6 +370,8 @@ typedef struct elf64_shdr { + #define EI_OSABI 7 + #define EI_PAD 8 + ++#define EI_PAX 14 ++ + #define ELFMAG0 0x7f /* EI_MAG */ + #define ELFMAG1 'E' + #define ELFMAG2 'L' +@@ -421,6 +449,7 @@ extern Elf32_Dyn _DYNAMIC []; + #define elf_note elf32_note + #define elf_addr_t Elf32_Off + #define Elf_Half Elf32_Half ++#define elf_dyn Elf32_Dyn + + #else + +@@ -431,6 +460,7 @@ extern Elf64_Dyn _DYNAMIC []; + #define elf_note elf64_note + #define elf_addr_t Elf64_Off + #define Elf_Half Elf64_Half ++#define elf_dyn Elf64_Dyn + + #endif + +diff -urNp linux-2.6.38.3/include/linux/fs.h linux-2.6.38.3/include/linux/fs.h +--- linux-2.6.38.3/include/linux/fs.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/fs.h 2011-04-17 15:57:32.000000000 -0400 +@@ -105,6 +105,11 @@ struct inodes_stat_t { + /* File was opened by fanotify and shouldn't generate fanotify events */ + #define FMODE_NONOTIFY ((__force fmode_t)0x1000000) + ++/* Hack for grsec so as not to require read permission simply to execute ++ * a binary ++ */ ++#define FMODE_GREXEC ((__force fmode_t)0x2000000) ++ + /* + * The below are the various read and write types that we support. Some of + * them include behavioral modifiers that send information down to the +@@ -581,42 +586,42 @@ typedef int (*read_actor_t)(read_descrip + unsigned long, unsigned long); + + struct address_space_operations { +- int (*writepage)(struct page *page, struct writeback_control *wbc); +- int (*readpage)(struct file *, struct page *); +- void (*sync_page)(struct page *); ++ int (* const writepage)(struct page *page, struct writeback_control *wbc); ++ int (* const readpage)(struct file *, struct page *); ++ void (* const sync_page)(struct page *); + + /* Write back some dirty pages from this mapping. */ +- int (*writepages)(struct address_space *, struct writeback_control *); ++ int (* const writepages)(struct address_space *, struct writeback_control *); + + /* Set a page dirty. Return true if this dirtied it */ +- int (*set_page_dirty)(struct page *page); ++ int (* const set_page_dirty)(struct page *page); + +- int (*readpages)(struct file *filp, struct address_space *mapping, ++ int (* const readpages)(struct file *filp, struct address_space *mapping, + struct list_head *pages, unsigned nr_pages); + +- int (*write_begin)(struct file *, struct address_space *mapping, ++ int (* const write_begin)(struct file *, struct address_space *mapping, + loff_t pos, unsigned len, unsigned flags, + struct page **pagep, void **fsdata); +- int (*write_end)(struct file *, struct address_space *mapping, ++ int (* const write_end)(struct file *, struct address_space *mapping, + loff_t pos, unsigned len, unsigned copied, + struct page *page, void *fsdata); + + /* Unfortunately this kludge is needed for FIBMAP. Don't use it */ +- sector_t (*bmap)(struct address_space *, sector_t); +- void (*invalidatepage) (struct page *, unsigned long); +- int (*releasepage) (struct page *, gfp_t); +- void (*freepage)(struct page *); +- ssize_t (*direct_IO)(int, struct kiocb *, const struct iovec *iov, ++ sector_t (* const bmap)(struct address_space *, sector_t); ++ void (* const invalidatepage) (struct page *, unsigned long); ++ int (* const releasepage) (struct page *, gfp_t); ++ void (* const freepage)(struct page *); ++ ssize_t (* const direct_IO)(int, struct kiocb *, const struct iovec *iov, + loff_t offset, unsigned long nr_segs); +- int (*get_xip_mem)(struct address_space *, pgoff_t, int, ++ int (* const get_xip_mem)(struct address_space *, pgoff_t, int, + void **, unsigned long *); + /* migrate the contents of a page to the specified target */ +- int (*migratepage) (struct address_space *, ++ int (* const migratepage) (struct address_space *, + struct page *, struct page *); +- int (*launder_page) (struct page *); +- int (*is_partially_uptodate) (struct page *, read_descriptor_t *, ++ int (* const launder_page) (struct page *); ++ int (* const is_partially_uptodate) (struct page *, read_descriptor_t *, + unsigned long); +- int (*error_remove_page)(struct address_space *, struct page *); ++ int (* const error_remove_page)(struct address_space *, struct page *); + }; + + /* +@@ -1059,17 +1064,17 @@ static inline int file_check_writeable(s + typedef struct files_struct *fl_owner_t; + + struct file_lock_operations { +- void (*fl_copy_lock)(struct file_lock *, struct file_lock *); +- void (*fl_release_private)(struct file_lock *); ++ void (* const fl_copy_lock)(struct file_lock *, struct file_lock *); ++ void (* const fl_release_private)(struct file_lock *); + }; + + struct lock_manager_operations { +- int (*fl_compare_owner)(struct file_lock *, struct file_lock *); +- void (*fl_notify)(struct file_lock *); /* unblock callback */ +- int (*fl_grant)(struct file_lock *, struct file_lock *, int); +- void (*fl_release_private)(struct file_lock *); +- void (*fl_break)(struct file_lock *); +- int (*fl_change)(struct file_lock **, int); ++ int (* const fl_compare_owner)(struct file_lock *, struct file_lock *); ++ void (* const fl_notify)(struct file_lock *); /* unblock callback */ ++ int (* const fl_grant)(struct file_lock *, struct file_lock *, int); ++ void (* const fl_release_private)(struct file_lock *); ++ void (* const fl_break)(struct file_lock *); ++ int (* const fl_change)(struct file_lock **, int); + }; + + struct lock_manager { +@@ -1604,29 +1609,29 @@ extern ssize_t vfs_writev(struct file *, + unsigned long, loff_t *); + + struct super_operations { +- struct inode *(*alloc_inode)(struct super_block *sb); +- void (*destroy_inode)(struct inode *); ++ struct inode *(* const alloc_inode)(struct super_block *sb); ++ void (* const destroy_inode)(struct inode *); + +- void (*dirty_inode) (struct inode *); +- int (*write_inode) (struct inode *, struct writeback_control *wbc); +- int (*drop_inode) (struct inode *); +- void (*evict_inode) (struct inode *); +- void (*put_super) (struct super_block *); +- void (*write_super) (struct super_block *); +- int (*sync_fs)(struct super_block *sb, int wait); +- int (*freeze_fs) (struct super_block *); +- int (*unfreeze_fs) (struct super_block *); +- int (*statfs) (struct dentry *, struct kstatfs *); +- int (*remount_fs) (struct super_block *, int *, char *); +- void (*umount_begin) (struct super_block *); ++ void (* const dirty_inode) (struct inode *); ++ int (* const write_inode) (struct inode *, struct writeback_control *wbc); ++ int (* const drop_inode) (struct inode *); ++ void (* const evict_inode) (struct inode *); ++ void (* const put_super) (struct super_block *); ++ void (* const write_super) (struct super_block *); ++ int (* const sync_fs)(struct super_block *sb, int wait); ++ int (* const freeze_fs) (struct super_block *); ++ int (* const unfreeze_fs) (struct super_block *); ++ int (* const statfs) (struct dentry *, struct kstatfs *); ++ int (* const remount_fs) (struct super_block *, int *, char *); ++ void (* const umount_begin) (struct super_block *); + +- int (*show_options)(struct seq_file *, struct vfsmount *); +- int (*show_stats)(struct seq_file *, struct vfsmount *); ++ int (* const show_options)(struct seq_file *, struct vfsmount *); ++ int (* const show_stats)(struct seq_file *, struct vfsmount *); + #ifdef CONFIG_QUOTA +- ssize_t (*quota_read)(struct super_block *, int, char *, size_t, loff_t); +- ssize_t (*quota_write)(struct super_block *, int, const char *, size_t, loff_t); ++ ssize_t (* const quota_read)(struct super_block *, int, char *, size_t, loff_t); ++ ssize_t (* const quota_write)(struct super_block *, int, const char *, size_t, loff_t); + #endif +- int (*bdev_try_to_free_page)(struct super_block*, struct page*, gfp_t); ++ int (* const bdev_try_to_free_page)(struct super_block*, struct page*, gfp_t); + }; + + /* +diff -urNp linux-2.6.38.3/include/linux/fs_struct.h linux-2.6.38.3/include/linux/fs_struct.h +--- linux-2.6.38.3/include/linux/fs_struct.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/fs_struct.h 2011-04-17 15:57:32.000000000 -0400 +@@ -6,7 +6,7 @@ + #include <linux/seqlock.h> + + struct fs_struct { +- int users; ++ atomic_t users; + spinlock_t lock; + seqcount_t seq; + int umask; +diff -urNp linux-2.6.38.3/include/linux/genhd.h linux-2.6.38.3/include/linux/genhd.h +--- linux-2.6.38.3/include/linux/genhd.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/genhd.h 2011-04-17 15:57:32.000000000 -0400 +@@ -183,7 +183,7 @@ struct gendisk { + struct kobject *slave_dir; + + struct timer_rand_state *random; +- atomic_t sync_io; /* RAID */ ++ atomic_unchecked_t sync_io; /* RAID */ + struct disk_events *ev; + #ifdef CONFIG_BLK_DEV_INTEGRITY + struct blk_integrity *integrity; +diff -urNp linux-2.6.38.3/include/linux/gracl.h linux-2.6.38.3/include/linux/gracl.h +--- linux-2.6.38.3/include/linux/gracl.h 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/include/linux/gracl.h 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,317 @@ ++#ifndef GR_ACL_H ++#define GR_ACL_H ++ ++#include <linux/grdefs.h> ++#include <linux/resource.h> ++#include <linux/capability.h> ++#include <linux/dcache.h> ++#include <asm/resource.h> ++ ++/* Major status information */ ++ ++#define GR_VERSION "grsecurity 2.2.2" ++#define GRSECURITY_VERSION 0x2202 ++ ++enum { ++ GR_SHUTDOWN = 0, ++ GR_ENABLE = 1, ++ GR_SPROLE = 2, ++ GR_RELOAD = 3, ++ GR_SEGVMOD = 4, ++ GR_STATUS = 5, ++ GR_UNSPROLE = 6, ++ GR_PASSSET = 7, ++ GR_SPROLEPAM = 8, ++}; ++ ++/* Password setup definitions ++ * kernel/grhash.c */ ++enum { ++ GR_PW_LEN = 128, ++ GR_SALT_LEN = 16, ++ GR_SHA_LEN = 32, ++}; ++ ++enum { ++ GR_SPROLE_LEN = 64, ++}; ++ ++enum { ++ GR_NO_GLOB = 0, ++ GR_REG_GLOB, ++ GR_CREATE_GLOB ++}; ++ ++#define GR_NLIMITS 32 ++ ++/* Begin Data Structures */ ++ ++struct sprole_pw { ++ unsigned char *rolename; ++ unsigned char salt[GR_SALT_LEN]; ++ unsigned char sum[GR_SHA_LEN]; /* 256-bit SHA hash of the password */ ++}; ++ ++struct name_entry { ++ __u32 key; ++ ino_t inode; ++ dev_t device; ++ char *name; ++ __u16 len; ++ __u8 deleted; ++ struct name_entry *prev; ++ struct name_entry *next; ++}; ++ ++struct inodev_entry { ++ struct name_entry *nentry; ++ struct inodev_entry *prev; ++ struct inodev_entry *next; ++}; ++ ++struct acl_role_db { ++ struct acl_role_label **r_hash; ++ __u32 r_size; ++}; ++ ++struct inodev_db { ++ struct inodev_entry **i_hash; ++ __u32 i_size; ++}; ++ ++struct name_db { ++ struct name_entry **n_hash; ++ __u32 n_size; ++}; ++ ++struct crash_uid { ++ uid_t uid; ++ unsigned long expires; ++}; ++ ++struct gr_hash_struct { ++ void **table; ++ void **nametable; ++ void *first; ++ __u32 table_size; ++ __u32 used_size; ++ int type; ++}; ++ ++/* Userspace Grsecurity ACL data structures */ ++ ++struct acl_subject_label { ++ char *filename; ++ ino_t inode; ++ dev_t device; ++ __u32 mode; ++ kernel_cap_t cap_mask; ++ kernel_cap_t cap_lower; ++ kernel_cap_t cap_invert_audit; ++ ++ struct rlimit res[GR_NLIMITS]; ++ __u32 resmask; ++ ++ __u8 user_trans_type; ++ __u8 group_trans_type; ++ uid_t *user_transitions; ++ gid_t *group_transitions; ++ __u16 user_trans_num; ++ __u16 group_trans_num; ++ ++ __u32 sock_families[2]; ++ __u32 ip_proto[8]; ++ __u32 ip_type; ++ struct acl_ip_label **ips; ++ __u32 ip_num; ++ __u32 inaddr_any_override; ++ ++ __u32 crashes; ++ unsigned long expires; ++ ++ struct acl_subject_label *parent_subject; ++ struct gr_hash_struct *hash; ++ struct acl_subject_label *prev; ++ struct acl_subject_label *next; ++ ++ struct acl_object_label **obj_hash; ++ __u32 obj_hash_size; ++ __u16 pax_flags; ++}; ++ ++struct role_allowed_ip { ++ __u32 addr; ++ __u32 netmask; ++ ++ struct role_allowed_ip *prev; ++ struct role_allowed_ip *next; ++}; ++ ++struct role_transition { ++ char *rolename; ++ ++ struct role_transition *prev; ++ struct role_transition *next; ++}; ++ ++struct acl_role_label { ++ char *rolename; ++ uid_t uidgid; ++ __u16 roletype; ++ ++ __u16 auth_attempts; ++ unsigned long expires; ++ ++ struct acl_subject_label *root_label; ++ struct gr_hash_struct *hash; ++ ++ struct acl_role_label *prev; ++ struct acl_role_label *next; ++ ++ struct role_transition *transitions; ++ struct role_allowed_ip *allowed_ips; ++ uid_t *domain_children; ++ __u16 domain_child_num; ++ ++ struct acl_subject_label **subj_hash; ++ __u32 subj_hash_size; ++}; ++ ++struct user_acl_role_db { ++ struct acl_role_label **r_table; ++ __u32 num_pointers; /* Number of allocations to track */ ++ __u32 num_roles; /* Number of roles */ ++ __u32 num_domain_children; /* Number of domain children */ ++ __u32 num_subjects; /* Number of subjects */ ++ __u32 num_objects; /* Number of objects */ ++}; ++ ++struct acl_object_label { ++ char *filename; ++ ino_t inode; ++ dev_t device; ++ __u32 mode; ++ ++ struct acl_subject_label *nested; ++ struct acl_object_label *globbed; ++ ++ /* next two structures not used */ ++ ++ struct acl_object_label *prev; ++ struct acl_object_label *next; ++}; ++ ++struct acl_ip_label { ++ char *iface; ++ __u32 addr; ++ __u32 netmask; ++ __u16 low, high; ++ __u8 mode; ++ __u32 type; ++ __u32 proto[8]; ++ ++ /* next two structures not used */ ++ ++ struct acl_ip_label *prev; ++ struct acl_ip_label *next; ++}; ++ ++struct gr_arg { ++ struct user_acl_role_db role_db; ++ unsigned char pw[GR_PW_LEN]; ++ unsigned char salt[GR_SALT_LEN]; ++ unsigned char sum[GR_SHA_LEN]; ++ unsigned char sp_role[GR_SPROLE_LEN]; ++ struct sprole_pw *sprole_pws; ++ dev_t segv_device; ++ ino_t segv_inode; ++ uid_t segv_uid; ++ __u16 num_sprole_pws; ++ __u16 mode; ++}; ++ ++struct gr_arg_wrapper { ++ struct gr_arg *arg; ++ __u32 version; ++ __u32 size; ++}; ++ ++struct subject_map { ++ struct acl_subject_label *user; ++ struct acl_subject_label *kernel; ++ struct subject_map *prev; ++ struct subject_map *next; ++}; ++ ++struct acl_subj_map_db { ++ struct subject_map **s_hash; ++ __u32 s_size; ++}; ++ ++/* End Data Structures Section */ ++ ++/* Hash functions generated by empirical testing by Brad Spengler ++ Makes good use of the low bits of the inode. Generally 0-1 times ++ in loop for successful match. 0-3 for unsuccessful match. ++ Shift/add algorithm with modulus of table size and an XOR*/ ++ ++static __inline__ unsigned int ++rhash(const uid_t uid, const __u16 type, const unsigned int sz) ++{ ++ return ((((uid + type) << (16 + type)) ^ uid) % sz); ++} ++ ++ static __inline__ unsigned int ++shash(const struct acl_subject_label *userp, const unsigned int sz) ++{ ++ return ((const unsigned long)userp % sz); ++} ++ ++static __inline__ unsigned int ++fhash(const ino_t ino, const dev_t dev, const unsigned int sz) ++{ ++ return (((ino + dev) ^ ((ino << 13) + (ino << 23) + (dev << 9))) % sz); ++} ++ ++static __inline__ unsigned int ++nhash(const char *name, const __u16 len, const unsigned int sz) ++{ ++ return full_name_hash((const unsigned char *)name, len) % sz; ++} ++ ++#define FOR_EACH_ROLE_START(role) \ ++ role = role_list; \ ++ while (role) { ++ ++#define FOR_EACH_ROLE_END(role) \ ++ role = role->prev; \ ++ } ++ ++#define FOR_EACH_SUBJECT_START(role,subj,iter) \ ++ subj = NULL; \ ++ iter = 0; \ ++ while (iter < role->subj_hash_size) { \ ++ if (subj == NULL) \ ++ subj = role->subj_hash[iter]; \ ++ if (subj == NULL) { \ ++ iter++; \ ++ continue; \ ++ } ++ ++#define FOR_EACH_SUBJECT_END(subj,iter) \ ++ subj = subj->next; \ ++ if (subj == NULL) \ ++ iter++; \ ++ } ++ ++ ++#define FOR_EACH_NESTED_SUBJECT_START(role,subj) \ ++ subj = role->hash->first; \ ++ while (subj != NULL) { ++ ++#define FOR_EACH_NESTED_SUBJECT_END(subj) \ ++ subj = subj->next; \ ++ } ++ ++#endif ++ +diff -urNp linux-2.6.38.3/include/linux/gralloc.h linux-2.6.38.3/include/linux/gralloc.h +--- linux-2.6.38.3/include/linux/gralloc.h 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/include/linux/gralloc.h 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,9 @@ ++#ifndef __GRALLOC_H ++#define __GRALLOC_H ++ ++void acl_free_all(void); ++int acl_alloc_stack_init(unsigned long size); ++void *acl_alloc(unsigned long len); ++void *acl_alloc_num(unsigned long num, unsigned long len); ++ ++#endif +diff -urNp linux-2.6.38.3/include/linux/grdefs.h linux-2.6.38.3/include/linux/grdefs.h +--- linux-2.6.38.3/include/linux/grdefs.h 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/include/linux/grdefs.h 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,139 @@ ++#ifndef GRDEFS_H ++#define GRDEFS_H ++ ++/* Begin grsecurity status declarations */ ++ ++enum { ++ GR_READY = 0x01, ++ GR_STATUS_INIT = 0x00 // disabled state ++}; ++ ++/* Begin ACL declarations */ ++ ++/* Role flags */ ++ ++enum { ++ GR_ROLE_USER = 0x0001, ++ GR_ROLE_GROUP = 0x0002, ++ GR_ROLE_DEFAULT = 0x0004, ++ GR_ROLE_SPECIAL = 0x0008, ++ GR_ROLE_AUTH = 0x0010, ++ GR_ROLE_NOPW = 0x0020, ++ GR_ROLE_GOD = 0x0040, ++ GR_ROLE_LEARN = 0x0080, ++ GR_ROLE_TPE = 0x0100, ++ GR_ROLE_DOMAIN = 0x0200, ++ GR_ROLE_PAM = 0x0400, ++ GR_ROLE_PERSIST = 0x0800 ++}; ++ ++/* ACL Subject and Object mode flags */ ++enum { ++ GR_DELETED = 0x80000000 ++}; ++ ++/* ACL Object-only mode flags */ ++enum { ++ GR_READ = 0x00000001, ++ GR_APPEND = 0x00000002, ++ GR_WRITE = 0x00000004, ++ GR_EXEC = 0x00000008, ++ GR_FIND = 0x00000010, ++ GR_INHERIT = 0x00000020, ++ GR_SETID = 0x00000040, ++ GR_CREATE = 0x00000080, ++ GR_DELETE = 0x00000100, ++ GR_LINK = 0x00000200, ++ GR_AUDIT_READ = 0x00000400, ++ GR_AUDIT_APPEND = 0x00000800, ++ GR_AUDIT_WRITE = 0x00001000, ++ GR_AUDIT_EXEC = 0x00002000, ++ GR_AUDIT_FIND = 0x00004000, ++ GR_AUDIT_INHERIT= 0x00008000, ++ GR_AUDIT_SETID = 0x00010000, ++ GR_AUDIT_CREATE = 0x00020000, ++ GR_AUDIT_DELETE = 0x00040000, ++ GR_AUDIT_LINK = 0x00080000, ++ GR_PTRACERD = 0x00100000, ++ GR_NOPTRACE = 0x00200000, ++ GR_SUPPRESS = 0x00400000, ++ GR_NOLEARN = 0x00800000, ++ GR_INIT_TRANSFER= 0x01000000 ++}; ++ ++#define GR_AUDITS (GR_AUDIT_READ | GR_AUDIT_WRITE | GR_AUDIT_APPEND | GR_AUDIT_EXEC | \ ++ GR_AUDIT_FIND | GR_AUDIT_INHERIT | GR_AUDIT_SETID | \ ++ GR_AUDIT_CREATE | GR_AUDIT_DELETE | GR_AUDIT_LINK) ++ ++/* ACL subject-only mode flags */ ++enum { ++ GR_KILL = 0x00000001, ++ GR_VIEW = 0x00000002, ++ GR_PROTECTED = 0x00000004, ++ GR_LEARN = 0x00000008, ++ GR_OVERRIDE = 0x00000010, ++ /* just a placeholder, this mode is only used in userspace */ ++ GR_DUMMY = 0x00000020, ++ GR_PROTSHM = 0x00000040, ++ GR_KILLPROC = 0x00000080, ++ GR_KILLIPPROC = 0x00000100, ++ /* just a placeholder, this mode is only used in userspace */ ++ GR_NOTROJAN = 0x00000200, ++ GR_PROTPROCFD = 0x00000400, ++ GR_PROCACCT = 0x00000800, ++ GR_RELAXPTRACE = 0x00001000, ++ GR_NESTED = 0x00002000, ++ GR_INHERITLEARN = 0x00004000, ++ GR_PROCFIND = 0x00008000, ++ GR_POVERRIDE = 0x00010000, ++ GR_KERNELAUTH = 0x00020000, ++ GR_ATSECURE = 0x00040000 ++}; ++ ++enum { ++ GR_PAX_ENABLE_SEGMEXEC = 0x0001, ++ GR_PAX_ENABLE_PAGEEXEC = 0x0002, ++ GR_PAX_ENABLE_MPROTECT = 0x0004, ++ GR_PAX_ENABLE_RANDMMAP = 0x0008, ++ GR_PAX_ENABLE_EMUTRAMP = 0x0010, ++ GR_PAX_DISABLE_SEGMEXEC = 0x0100, ++ GR_PAX_DISABLE_PAGEEXEC = 0x0200, ++ GR_PAX_DISABLE_MPROTECT = 0x0400, ++ GR_PAX_DISABLE_RANDMMAP = 0x0800, ++ GR_PAX_DISABLE_EMUTRAMP = 0x1000, ++}; ++ ++enum { ++ GR_ID_USER = 0x01, ++ GR_ID_GROUP = 0x02, ++}; ++ ++enum { ++ GR_ID_ALLOW = 0x01, ++ GR_ID_DENY = 0x02, ++}; ++ ++#define GR_CRASH_RES 31 ++#define GR_UIDTABLE_MAX 500 ++ ++/* begin resource learning section */ ++enum { ++ GR_RLIM_CPU_BUMP = 60, ++ GR_RLIM_FSIZE_BUMP = 50000, ++ GR_RLIM_DATA_BUMP = 10000, ++ GR_RLIM_STACK_BUMP = 1000, ++ GR_RLIM_CORE_BUMP = 10000, ++ GR_RLIM_RSS_BUMP = 500000, ++ GR_RLIM_NPROC_BUMP = 1, ++ GR_RLIM_NOFILE_BUMP = 5, ++ GR_RLIM_MEMLOCK_BUMP = 50000, ++ GR_RLIM_AS_BUMP = 500000, ++ GR_RLIM_LOCKS_BUMP = 2, ++ GR_RLIM_SIGPENDING_BUMP = 5, ++ GR_RLIM_MSGQUEUE_BUMP = 10000, ++ GR_RLIM_NICE_BUMP = 1, ++ GR_RLIM_RTPRIO_BUMP = 1, ++ GR_RLIM_RTTIME_BUMP = 1000000 ++}; ++ ++#endif +diff -urNp linux-2.6.38.3/include/linux/grinternal.h linux-2.6.38.3/include/linux/grinternal.h +--- linux-2.6.38.3/include/linux/grinternal.h 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/include/linux/grinternal.h 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,219 @@ ++#ifndef __GRINTERNAL_H ++#define __GRINTERNAL_H ++ ++#ifdef CONFIG_GRKERNSEC ++ ++#include <linux/fs.h> ++#include <linux/mnt_namespace.h> ++#include <linux/nsproxy.h> ++#include <linux/gracl.h> ++#include <linux/grdefs.h> ++#include <linux/grmsg.h> ++ ++void gr_add_learn_entry(const char *fmt, ...) ++ __attribute__ ((format (printf, 1, 2))); ++__u32 gr_search_file(const struct dentry *dentry, const __u32 mode, ++ const struct vfsmount *mnt); ++__u32 gr_check_create(const struct dentry *new_dentry, ++ const struct dentry *parent, ++ const struct vfsmount *mnt, const __u32 mode); ++int gr_check_protected_task(const struct task_struct *task); ++__u32 to_gr_audit(const __u32 reqmode); ++int gr_set_acls(const int type); ++int gr_apply_subject_to_task(struct task_struct *task); ++int gr_acl_is_enabled(void); ++char gr_roletype_to_char(void); ++ ++void gr_handle_alertkill(struct task_struct *task); ++char *gr_to_filename(const struct dentry *dentry, ++ const struct vfsmount *mnt); ++char *gr_to_filename1(const struct dentry *dentry, ++ const struct vfsmount *mnt); ++char *gr_to_filename2(const struct dentry *dentry, ++ const struct vfsmount *mnt); ++char *gr_to_filename3(const struct dentry *dentry, ++ const struct vfsmount *mnt); ++ ++extern int grsec_enable_harden_ptrace; ++extern int grsec_enable_link; ++extern int grsec_enable_fifo; ++extern int grsec_enable_execve; ++extern int grsec_enable_shm; ++extern int grsec_enable_execlog; ++extern int grsec_enable_signal; ++extern int grsec_enable_audit_ptrace; ++extern int grsec_enable_forkfail; ++extern int grsec_enable_time; ++extern int grsec_enable_rofs; ++extern int grsec_enable_chroot_shmat; ++extern int grsec_enable_chroot_findtask; ++extern int grsec_enable_chroot_mount; ++extern int grsec_enable_chroot_double; ++extern int grsec_enable_chroot_pivot; ++extern int grsec_enable_chroot_chdir; ++extern int grsec_enable_chroot_chmod; ++extern int grsec_enable_chroot_mknod; ++extern int grsec_enable_chroot_fchdir; ++extern int grsec_enable_chroot_nice; ++extern int grsec_enable_chroot_execlog; ++extern int grsec_enable_chroot_caps; ++extern int grsec_enable_chroot_sysctl; ++extern int grsec_enable_chroot_unix; ++extern int grsec_enable_tpe; ++extern int grsec_tpe_gid; ++extern int grsec_enable_tpe_all; ++extern int grsec_enable_tpe_invert; ++extern int grsec_enable_socket_all; ++extern int grsec_socket_all_gid; ++extern int grsec_enable_socket_client; ++extern int grsec_socket_client_gid; ++extern int grsec_enable_socket_server; ++extern int grsec_socket_server_gid; ++extern int grsec_audit_gid; ++extern int grsec_enable_group; ++extern int grsec_enable_audit_textrel; ++extern int grsec_enable_log_rwxmaps; ++extern int grsec_enable_mount; ++extern int grsec_enable_chdir; ++extern int grsec_resource_logging; ++extern int grsec_enable_blackhole; ++extern int grsec_lastack_retries; ++extern int grsec_lock; ++ ++extern spinlock_t grsec_alert_lock; ++extern unsigned long grsec_alert_wtime; ++extern unsigned long grsec_alert_fyet; ++ ++extern spinlock_t grsec_audit_lock; ++ ++extern rwlock_t grsec_exec_file_lock; ++ ++#define gr_task_fullpath(tsk) ((tsk)->exec_file ? \ ++ gr_to_filename2((tsk)->exec_file->f_path.dentry, \ ++ (tsk)->exec_file->f_vfsmnt) : "/") ++ ++#define gr_parent_task_fullpath(tsk) ((tsk)->real_parent->exec_file ? \ ++ gr_to_filename3((tsk)->real_parent->exec_file->f_path.dentry, \ ++ (tsk)->real_parent->exec_file->f_vfsmnt) : "/") ++ ++#define gr_task_fullpath0(tsk) ((tsk)->exec_file ? \ ++ gr_to_filename((tsk)->exec_file->f_path.dentry, \ ++ (tsk)->exec_file->f_vfsmnt) : "/") ++ ++#define gr_parent_task_fullpath0(tsk) ((tsk)->real_parent->exec_file ? \ ++ gr_to_filename1((tsk)->real_parent->exec_file->f_path.dentry, \ ++ (tsk)->real_parent->exec_file->f_vfsmnt) : "/") ++ ++#define proc_is_chrooted(tsk_a) ((tsk_a)->gr_is_chrooted) ++ ++#define have_same_root(tsk_a,tsk_b) ((tsk_a)->gr_chroot_dentry == (tsk_b)->gr_chroot_dentry) ++ ++#define DEFAULTSECARGS(task, cred, pcred) gr_task_fullpath(task), (task)->comm, \ ++ (task)->pid, (cred)->uid, \ ++ (cred)->euid, (cred)->gid, (cred)->egid, \ ++ gr_parent_task_fullpath(task), \ ++ (task)->real_parent->comm, (task)->real_parent->pid, \ ++ (pcred)->uid, (pcred)->euid, \ ++ (pcred)->gid, (pcred)->egid ++ ++#define GR_CHROOT_CAPS {{ \ ++ CAP_TO_MASK(CAP_LINUX_IMMUTABLE) | CAP_TO_MASK(CAP_NET_ADMIN) | \ ++ CAP_TO_MASK(CAP_SYS_MODULE) | CAP_TO_MASK(CAP_SYS_RAWIO) | \ ++ CAP_TO_MASK(CAP_SYS_PACCT) | CAP_TO_MASK(CAP_SYS_ADMIN) | \ ++ CAP_TO_MASK(CAP_SYS_BOOT) | CAP_TO_MASK(CAP_SYS_TIME) | \ ++ CAP_TO_MASK(CAP_NET_RAW) | CAP_TO_MASK(CAP_SYS_TTY_CONFIG) | \ ++ CAP_TO_MASK(CAP_IPC_OWNER) , 0 }} ++ ++#define security_learn(normal_msg,args...) \ ++({ \ ++ read_lock(&grsec_exec_file_lock); \ ++ gr_add_learn_entry(normal_msg "\n", ## args); \ ++ read_unlock(&grsec_exec_file_lock); \ ++}) ++ ++enum { ++ GR_DO_AUDIT, ++ GR_DONT_AUDIT, ++ /* used for non-audit messages that we shouldn't kill the task on */ ++ GR_DONT_AUDIT_GOOD ++}; ++ ++enum { ++ GR_TTYSNIFF, ++ GR_RBAC, ++ GR_RBAC_STR, ++ GR_STR_RBAC, ++ GR_RBAC_MODE2, ++ GR_RBAC_MODE3, ++ GR_FILENAME, ++ GR_SYSCTL_HIDDEN, ++ GR_NOARGS, ++ GR_ONE_INT, ++ GR_ONE_INT_TWO_STR, ++ GR_ONE_STR, ++ GR_STR_INT, ++ GR_TWO_STR_INT, ++ GR_TWO_INT, ++ GR_TWO_U64, ++ GR_THREE_INT, ++ GR_FIVE_INT_TWO_STR, ++ GR_TWO_STR, ++ GR_THREE_STR, ++ GR_FOUR_STR, ++ GR_STR_FILENAME, ++ GR_FILENAME_STR, ++ GR_FILENAME_TWO_INT, ++ GR_FILENAME_TWO_INT_STR, ++ GR_TEXTREL, ++ GR_PTRACE, ++ GR_RESOURCE, ++ GR_CAP, ++ GR_SIG, ++ GR_SIG2, ++ GR_CRASH1, ++ GR_CRASH2, ++ GR_PSACCT, ++ GR_RWXMAP ++}; ++ ++#define gr_log_hidden_sysctl(audit, msg, str) gr_log_varargs(audit, msg, GR_SYSCTL_HIDDEN, str) ++#define gr_log_ttysniff(audit, msg, task) gr_log_varargs(audit, msg, GR_TTYSNIFF, task) ++#define gr_log_fs_rbac_generic(audit, msg, dentry, mnt) gr_log_varargs(audit, msg, GR_RBAC, dentry, mnt) ++#define gr_log_fs_rbac_str(audit, msg, dentry, mnt, str) gr_log_varargs(audit, msg, GR_RBAC_STR, dentry, mnt, str) ++#define gr_log_fs_str_rbac(audit, msg, str, dentry, mnt) gr_log_varargs(audit, msg, GR_STR_RBAC, str, dentry, mnt) ++#define gr_log_fs_rbac_mode2(audit, msg, dentry, mnt, str1, str2) gr_log_varargs(audit, msg, GR_RBAC_MODE2, dentry, mnt, str1, str2) ++#define gr_log_fs_rbac_mode3(audit, msg, dentry, mnt, str1, str2, str3) gr_log_varargs(audit, msg, GR_RBAC_MODE3, dentry, mnt, str1, str2, str3) ++#define gr_log_fs_generic(audit, msg, dentry, mnt) gr_log_varargs(audit, msg, GR_FILENAME, dentry, mnt) ++#define gr_log_noargs(audit, msg) gr_log_varargs(audit, msg, GR_NOARGS) ++#define gr_log_int(audit, msg, num) gr_log_varargs(audit, msg, GR_ONE_INT, num) ++#define gr_log_int_str2(audit, msg, num, str1, str2) gr_log_varargs(audit, msg, GR_ONE_INT_TWO_STR, num, str1, str2) ++#define gr_log_str(audit, msg, str) gr_log_varargs(audit, msg, GR_ONE_STR, str) ++#define gr_log_str_int(audit, msg, str, num) gr_log_varargs(audit, msg, GR_STR_INT, str, num) ++#define gr_log_int_int(audit, msg, num1, num2) gr_log_varargs(audit, msg, GR_TWO_INT, num1, num2) ++#define gr_log_two_u64(audit, msg, num1, num2) gr_log_varargs(audit, msg, GR_TWO_U64, num1, num2) ++#define gr_log_int3(audit, msg, num1, num2, num3) gr_log_varargs(audit, msg, GR_THREE_INT, num1, num2, num3) ++#define gr_log_int5_str2(audit, msg, num1, num2, str1, str2) gr_log_varargs(audit, msg, GR_FIVE_INT_TWO_STR, num1, num2, str1, str2) ++#define gr_log_str_str(audit, msg, str1, str2) gr_log_varargs(audit, msg, GR_TWO_STR, str1, str2) ++#define gr_log_str2_int(audit, msg, str1, str2, num) gr_log_varargs(audit, msg, GR_TWO_STR_INT, str1, str2, num) ++#define gr_log_str3(audit, msg, str1, str2, str3) gr_log_varargs(audit, msg, GR_THREE_STR, str1, str2, str3) ++#define gr_log_str4(audit, msg, str1, str2, str3, str4) gr_log_varargs(audit, msg, GR_FOUR_STR, str1, str2, str3, str4) ++#define gr_log_str_fs(audit, msg, str, dentry, mnt) gr_log_varargs(audit, msg, GR_STR_FILENAME, str, dentry, mnt) ++#define gr_log_fs_str(audit, msg, dentry, mnt, str) gr_log_varargs(audit, msg, GR_FILENAME_STR, dentry, mnt, str) ++#define gr_log_fs_int2(audit, msg, dentry, mnt, num1, num2) gr_log_varargs(audit, msg, GR_FILENAME_TWO_INT, dentry, mnt, num1, num2) ++#define gr_log_fs_int2_str(audit, msg, dentry, mnt, num1, num2, str) gr_log_varargs(audit, msg, GR_FILENAME_TWO_INT_STR, dentry, mnt, num1, num2, str) ++#define gr_log_textrel_ulong_ulong(audit, msg, file, ulong1, ulong2) gr_log_varargs(audit, msg, GR_TEXTREL, file, ulong1, ulong2) ++#define gr_log_ptrace(audit, msg, task) gr_log_varargs(audit, msg, GR_PTRACE, task) ++#define gr_log_res_ulong2_str(audit, msg, task, ulong1, str, ulong2) gr_log_varargs(audit, msg, GR_RESOURCE, task, ulong1, str, ulong2) ++#define gr_log_cap(audit, msg, task, str) gr_log_varargs(audit, msg, GR_CAP, task, str) ++#define gr_log_sig_addr(audit, msg, str, addr) gr_log_varargs(audit, msg, GR_SIG, str, addr) ++#define gr_log_sig_task(audit, msg, task, num) gr_log_varargs(audit, msg, GR_SIG2, task, num) ++#define gr_log_crash1(audit, msg, task, ulong) gr_log_varargs(audit, msg, GR_CRASH1, task, ulong) ++#define gr_log_crash2(audit, msg, task, ulong1) gr_log_varargs(audit, msg, GR_CRASH2, task, ulong1) ++#define gr_log_procacct(audit, msg, task, num1, num2, num3, num4, num5, num6, num7, num8, num9) gr_log_varargs(audit, msg, GR_PSACCT, task, num1, num2, num3, num4, num5, num6, num7, num8, num9) ++#define gr_log_rwxmap(audit, msg, str) gr_log_varargs(audit, msg, GR_RWXMAP, str) ++ ++void gr_log_varargs(int audit, const char *msg, int argtypes, ...); ++ ++#endif ++ ++#endif +diff -urNp linux-2.6.38.3/include/linux/grmsg.h linux-2.6.38.3/include/linux/grmsg.h +--- linux-2.6.38.3/include/linux/grmsg.h 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/include/linux/grmsg.h 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,108 @@ ++#define DEFAULTSECMSG "%.256s[%.16s:%d] uid/euid:%u/%u gid/egid:%u/%u, parent %.256s[%.16s:%d] uid/euid:%u/%u gid/egid:%u/%u" ++#define GR_ACL_PROCACCT_MSG "%.256s[%.16s:%d] IP:%pI4 TTY:%.64s uid/euid:%u/%u gid/egid:%u/%u run time:[%ud %uh %um %us] cpu time:[%ud %uh %um %us] %s with exit code %ld, parent %.256s[%.16s:%d] IP:%pI4 TTY:%.64s uid/euid:%u/%u gid/egid:%u/%u" ++#define GR_PTRACE_ACL_MSG "denied ptrace of %.950s(%.16s:%d) by " ++#define GR_STOPMOD_MSG "denied modification of module state by " ++#define GR_ROFS_BLOCKWRITE_MSG "denied write to block device %.950s by " ++#define GR_ROFS_MOUNT_MSG "denied writable mount of %.950s by " ++#define GR_IOPERM_MSG "denied use of ioperm() by " ++#define GR_IOPL_MSG "denied use of iopl() by " ++#define GR_SHMAT_ACL_MSG "denied attach of shared memory of UID %u, PID %d, ID %u by " ++#define GR_UNIX_CHROOT_MSG "denied connect() to abstract AF_UNIX socket outside of chroot by " ++#define GR_SHMAT_CHROOT_MSG "denied attach of shared memory outside of chroot by " ++#define GR_MEM_READWRITE_MSG "denied access of range %Lx -> %Lx in /dev/mem by " ++#define GR_SYMLINK_MSG "not following symlink %.950s owned by %d.%d by " ++#define GR_LEARN_AUDIT_MSG "%s\t%u\t%u\t%u\t%.4095s\t%.4095s\t%lu\t%lu\t%.4095s\t%lu\t%pI4" ++#define GR_ID_LEARN_MSG "%s\t%u\t%u\t%u\t%.4095s\t%.4095s\t%c\t%d\t%d\t%d\t%pI4" ++#define GR_HIDDEN_ACL_MSG "%s access to hidden file %.950s by " ++#define GR_OPEN_ACL_MSG "%s open of %.950s for%s%s by " ++#define GR_CREATE_ACL_MSG "%s create of %.950s for%s%s by " ++#define GR_FIFO_MSG "denied writing FIFO %.950s of %d.%d by " ++#define GR_MKNOD_CHROOT_MSG "denied mknod of %.950s from chroot by " ++#define GR_MKNOD_ACL_MSG "%s mknod of %.950s by " ++#define GR_UNIXCONNECT_ACL_MSG "%s connect() to the unix domain socket %.950s by " ++#define GR_TTYSNIFF_ACL_MSG "terminal being sniffed by IP:%pI4 %.480s[%.16s:%d], parent %.480s[%.16s:%d] against " ++#define GR_MKDIR_ACL_MSG "%s mkdir of %.950s by " ++#define GR_RMDIR_ACL_MSG "%s rmdir of %.950s by " ++#define GR_UNLINK_ACL_MSG "%s unlink of %.950s by " ++#define GR_SYMLINK_ACL_MSG "%s symlink from %.480s to %.480s by " ++#define GR_HARDLINK_MSG "denied hardlink of %.930s (owned by %d.%d) to %.30s for " ++#define GR_LINK_ACL_MSG "%s link of %.480s to %.480s by " ++#define GR_INHERIT_ACL_MSG "successful inherit of %.480s's ACL for %.480s by " ++#define GR_RENAME_ACL_MSG "%s rename of %.480s to %.480s by " ++#define GR_UNSAFESHARE_EXEC_ACL_MSG "denied exec with cloned fs of %.950s by " ++#define GR_PTRACE_EXEC_ACL_MSG "denied ptrace of %.950s by " ++#define GR_NPROC_MSG "denied overstep of process limit by " ++#define GR_EXEC_ACL_MSG "%s execution of %.950s by " ++#define GR_EXEC_TPE_MSG "denied untrusted exec of %.950s by " ++#define GR_SEGVSTART_ACL_MSG "possible exploit bruteforcing on " DEFAULTSECMSG " banning uid %u from login for %lu seconds" ++#define GR_SEGVNOSUID_ACL_MSG "possible exploit bruteforcing on " DEFAULTSECMSG " banning execution for %lu seconds" ++#define GR_MOUNT_CHROOT_MSG "denied mount of %.256s as %.930s from chroot by " ++#define GR_PIVOT_CHROOT_MSG "denied pivot_root from chroot by " ++#define GR_TRUNCATE_ACL_MSG "%s truncate of %.950s by " ++#define GR_ATIME_ACL_MSG "%s access time change of %.950s by " ++#define GR_ACCESS_ACL_MSG "%s access of %.950s for%s%s%s by " ++#define GR_CHROOT_CHROOT_MSG "denied double chroot to %.950s by " ++#define GR_FCHMOD_ACL_MSG "%s fchmod of %.950s by " ++#define GR_CHMOD_CHROOT_MSG "denied chmod +s of %.950s by " ++#define GR_CHMOD_ACL_MSG "%s chmod of %.950s by " ++#define GR_CHROOT_FCHDIR_MSG "denied fchdir outside of chroot to %.950s by " ++#define GR_CHOWN_ACL_MSG "%s chown of %.950s by " ++#define GR_SETXATTR_ACL_MSG "%s setting extended attributes of %.950s by " ++#define GR_WRITLIB_ACL_MSG "denied load of writable library %.950s by " ++#define GR_INITF_ACL_MSG "init_variables() failed %s by " ++#define GR_DISABLED_ACL_MSG "Error loading %s, trying to run kernel with acls disabled. To disable acls at startup use <kernel image name> gracl=off from your boot loader" ++#define GR_DEV_ACL_MSG "/dev/grsec: %d bytes sent %d required, being fed garbaged by " ++#define GR_SHUTS_ACL_MSG "shutdown auth success for " ++#define GR_SHUTF_ACL_MSG "shutdown auth failure for " ++#define GR_SHUTI_ACL_MSG "ignoring shutdown for disabled RBAC system for " ++#define GR_SEGVMODS_ACL_MSG "segvmod auth success for " ++#define GR_SEGVMODF_ACL_MSG "segvmod auth failure for " ++#define GR_SEGVMODI_ACL_MSG "ignoring segvmod for disabled RBAC system for " ++#define GR_ENABLE_ACL_MSG "%s RBAC system loaded by " ++#define GR_ENABLEF_ACL_MSG "unable to load %s for " ++#define GR_RELOADI_ACL_MSG "ignoring reload request for disabled RBAC system" ++#define GR_RELOAD_ACL_MSG "%s RBAC system reloaded by " ++#define GR_RELOADF_ACL_MSG "failed reload of %s for " ++#define GR_SPROLEI_ACL_MSG "ignoring change to special role for disabled RBAC system for " ++#define GR_SPROLES_ACL_MSG "successful change to special role %s (id %d) by " ++#define GR_SPROLEL_ACL_MSG "special role %s (id %d) exited by " ++#define GR_SPROLEF_ACL_MSG "special role %s failure for " ++#define GR_UNSPROLEI_ACL_MSG "ignoring unauth of special role for disabled RBAC system for " ++#define GR_UNSPROLES_ACL_MSG "successful unauth of special role %s (id %d) by " ++#define GR_INVMODE_ACL_MSG "invalid mode %d by " ++#define GR_PRIORITY_CHROOT_MSG "denied priority change of process (%.16s:%d) by " ++#define GR_FAILFORK_MSG "failed fork with errno %s by " ++#define GR_NICE_CHROOT_MSG "denied priority change by " ++#define GR_UNISIGLOG_MSG "%.32s occurred at %p in " ++#define GR_DUALSIGLOG_MSG "signal %d sent to " DEFAULTSECMSG " by " ++#define GR_SIG_ACL_MSG "denied send of signal %d to protected task " DEFAULTSECMSG " by " ++#define GR_SYSCTL_MSG "denied modification of grsecurity sysctl value : %.32s by " ++#define GR_SYSCTL_ACL_MSG "%s sysctl of %.950s for%s%s by " ++#define GR_TIME_MSG "time set by " ++#define GR_DEFACL_MSG "fatal: unable to find subject for (%.16s:%d), loaded by " ++#define GR_MMAP_ACL_MSG "%s executable mmap of %.950s by " ++#define GR_MPROTECT_ACL_MSG "%s executable mprotect of %.950s by " ++#define GR_SOCK_MSG "denied socket(%.16s,%.16s,%.16s) by " ++#define GR_SOCK_NOINET_MSG "denied socket(%.16s,%.16s,%d) by " ++#define GR_BIND_MSG "denied bind() by " ++#define GR_CONNECT_MSG "denied connect() by " ++#define GR_BIND_ACL_MSG "denied bind() to %pI4 port %u sock type %.16s protocol %.16s by " ++#define GR_CONNECT_ACL_MSG "denied connect() to %pI4 port %u sock type %.16s protocol %.16s by " ++#define GR_IP_LEARN_MSG "%s\t%u\t%u\t%u\t%.4095s\t%.4095s\t%pI4\t%u\t%u\t%u\t%u\t%pI4" ++#define GR_EXEC_CHROOT_MSG "exec of %.980s within chroot by process " ++#define GR_CAP_ACL_MSG "use of %s denied for " ++#define GR_CAP_ACL_MSG2 "use of %s permitted for " ++#define GR_USRCHANGE_ACL_MSG "change to uid %u denied for " ++#define GR_GRPCHANGE_ACL_MSG "change to gid %u denied for " ++#define GR_REMOUNT_AUDIT_MSG "remount of %.256s by " ++#define GR_UNMOUNT_AUDIT_MSG "unmount of %.256s by " ++#define GR_MOUNT_AUDIT_MSG "mount of %.256s to %.256s by " ++#define GR_CHDIR_AUDIT_MSG "chdir to %.980s by " ++#define GR_EXEC_AUDIT_MSG "exec of %.930s (%.128s) by " ++#define GR_RESOURCE_MSG "denied resource overstep by requesting %lu for %.16s against limit %lu for " ++#define GR_RWXMMAP_MSG "denied RWX mmap of %.950s by " ++#define GR_RWXMPROTECT_MSG "denied RWX mprotect of %.950s by " ++#define GR_TEXTREL_AUDIT_MSG "text relocation in %s, VMA:0x%08lx 0x%08lx by " ++#define GR_VM86_MSG "denied use of vm86 by " ++#define GR_PTRACE_AUDIT_MSG "process %.950s(%.16s:%d) attached to via ptrace by " ++#define GR_INIT_TRANSFER_MSG "persistent special role transferred privilege to init by " +diff -urNp linux-2.6.38.3/include/linux/grsecurity.h linux-2.6.38.3/include/linux/grsecurity.h +--- linux-2.6.38.3/include/linux/grsecurity.h 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/include/linux/grsecurity.h 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,212 @@ ++#ifndef GR_SECURITY_H ++#define GR_SECURITY_H ++#include <linux/fs.h> ++#include <linux/fs_struct.h> ++#include <linux/binfmts.h> ++#include <linux/gracl.h> ++#include <linux/compat.h> ++ ++/* notify of brain-dead configs */ ++#if defined(CONFIG_PAX_NOEXEC) && !defined(CONFIG_PAX_PAGEEXEC) && !defined(CONFIG_PAX_SEGMEXEC) && !defined(CONFIG_PAX_KERNEXEC) ++#error "CONFIG_PAX_NOEXEC enabled, but PAGEEXEC, SEGMEXEC, and KERNEXEC are disabled." ++#endif ++#if defined(CONFIG_PAX_NOEXEC) && !defined(CONFIG_PAX_EI_PAX) && !defined(CONFIG_PAX_PT_PAX_FLAGS) ++#error "CONFIG_PAX_NOEXEC enabled, but neither CONFIG_PAX_EI_PAX nor CONFIG_PAX_PT_PAX_FLAGS are enabled." ++#endif ++#if defined(CONFIG_PAX_ASLR) && (defined(CONFIG_PAX_RANDMMAP) || defined(CONFIG_PAX_RANDUSTACK)) && !defined(CONFIG_PAX_EI_PAX) && !defined(CONFIG_PAX_PT_PAX_FLAGS) ++#error "CONFIG_PAX_ASLR enabled, but neither CONFIG_PAX_EI_PAX nor CONFIG_PAX_PT_PAX_FLAGS are enabled." ++#endif ++#if defined(CONFIG_PAX_ASLR) && !defined(CONFIG_PAX_RANDKSTACK) && !defined(CONFIG_PAX_RANDUSTACK) && !defined(CONFIG_PAX_RANDMMAP) ++#error "CONFIG_PAX_ASLR enabled, but RANDKSTACK, RANDUSTACK, and RANDMMAP are disabled." ++#endif ++#if defined(CONFIG_PAX) && !defined(CONFIG_PAX_NOEXEC) && !defined(CONFIG_PAX_ASLR) ++#error "CONFIG_PAX enabled, but no PaX options are enabled." ++#endif ++ ++void gr_handle_brute_attach(struct task_struct *p, unsigned long mm_flags); ++void gr_handle_brute_check(void); ++void gr_handle_kernel_exploit(void); ++int gr_process_user_ban(void); ++ ++char gr_roletype_to_char(void); ++ ++int gr_acl_enable_at_secure(void); ++ ++int gr_check_user_change(int real, int effective, int fs); ++int gr_check_group_change(int real, int effective, int fs); ++ ++void gr_del_task_from_ip_table(struct task_struct *p); ++ ++int gr_pid_is_chrooted(struct task_struct *p); ++int gr_handle_chroot_fowner(struct pid *pid, enum pid_type type); ++int gr_handle_chroot_nice(void); ++int gr_handle_chroot_sysctl(const int op); ++int gr_handle_chroot_setpriority(struct task_struct *p, ++ const int niceval); ++int gr_chroot_fchdir(struct dentry *u_dentry, struct vfsmount *u_mnt); ++int gr_handle_chroot_chroot(const struct dentry *dentry, ++ const struct vfsmount *mnt); ++int gr_handle_chroot_caps(struct path *path); ++void gr_handle_chroot_chdir(struct path *path); ++int gr_handle_chroot_chmod(const struct dentry *dentry, ++ const struct vfsmount *mnt, const int mode); ++int gr_handle_chroot_mknod(const struct dentry *dentry, ++ const struct vfsmount *mnt, const int mode); ++int gr_handle_chroot_mount(const struct dentry *dentry, ++ const struct vfsmount *mnt, ++ const char *dev_name); ++int gr_handle_chroot_pivot(void); ++int gr_handle_chroot_unix(struct pid *pid); ++ ++int gr_handle_rawio(const struct inode *inode); ++int gr_handle_nproc(void); ++ ++void gr_handle_ioperm(void); ++void gr_handle_iopl(void); ++ ++int gr_tpe_allow(const struct file *file); ++ ++void gr_set_chroot_entries(struct task_struct *task, struct path *path); ++void gr_clear_chroot_entries(struct task_struct *task); ++ ++void gr_log_forkfail(const int retval); ++void gr_log_timechange(void); ++void gr_log_signal(const int sig, const void *addr, const struct task_struct *t); ++void gr_log_chdir(const struct dentry *dentry, ++ const struct vfsmount *mnt); ++void gr_log_chroot_exec(const struct dentry *dentry, ++ const struct vfsmount *mnt); ++void gr_handle_exec_args(struct linux_binprm *bprm, const char __user *const __user *argv); ++#ifdef CONFIG_COMPAT ++void gr_handle_exec_args_compat(struct linux_binprm *bprm, compat_uptr_t __user *argv); ++#endif ++void gr_log_remount(const char *devname, const int retval); ++void gr_log_unmount(const char *devname, const int retval); ++void gr_log_mount(const char *from, const char *to, const int retval); ++void gr_log_textrel(struct vm_area_struct *vma); ++void gr_log_rwxmmap(struct file *file); ++void gr_log_rwxmprotect(struct file *file); ++ ++int gr_handle_follow_link(const struct inode *parent, ++ const struct inode *inode, ++ const struct dentry *dentry, ++ const struct vfsmount *mnt); ++int gr_handle_fifo(const struct dentry *dentry, ++ const struct vfsmount *mnt, ++ const struct dentry *dir, const int flag, ++ const int acc_mode); ++int gr_handle_hardlink(const struct dentry *dentry, ++ const struct vfsmount *mnt, ++ struct inode *inode, ++ const int mode, const char *to); ++ ++int gr_is_capable(const int cap); ++int gr_is_capable_nolog(const int cap); ++void gr_learn_resource(const struct task_struct *task, const int limit, ++ const unsigned long wanted, const int gt); ++void gr_copy_label(struct task_struct *tsk); ++void gr_handle_crash(struct task_struct *task, const int sig); ++int gr_handle_signal(const struct task_struct *p, const int sig); ++int gr_check_crash_uid(const uid_t uid); ++int gr_check_protected_task(const struct task_struct *task); ++int gr_check_protected_task_fowner(struct pid *pid, enum pid_type type); ++int gr_acl_handle_mmap(const struct file *file, ++ const unsigned long prot); ++int gr_acl_handle_mprotect(const struct file *file, ++ const unsigned long prot); ++int gr_check_hidden_task(const struct task_struct *tsk); ++__u32 gr_acl_handle_truncate(const struct dentry *dentry, ++ const struct vfsmount *mnt); ++__u32 gr_acl_handle_utime(const struct dentry *dentry, ++ const struct vfsmount *mnt); ++__u32 gr_acl_handle_access(const struct dentry *dentry, ++ const struct vfsmount *mnt, const int fmode); ++__u32 gr_acl_handle_fchmod(const struct dentry *dentry, ++ const struct vfsmount *mnt, mode_t mode); ++__u32 gr_acl_handle_chmod(const struct dentry *dentry, ++ const struct vfsmount *mnt, mode_t mode); ++__u32 gr_acl_handle_chown(const struct dentry *dentry, ++ const struct vfsmount *mnt); ++__u32 gr_acl_handle_setxattr(const struct dentry *dentry, ++ const struct vfsmount *mnt); ++int gr_handle_ptrace(struct task_struct *task, const long request); ++int gr_handle_proc_ptrace(struct task_struct *task); ++__u32 gr_acl_handle_execve(const struct dentry *dentry, ++ const struct vfsmount *mnt); ++int gr_check_crash_exec(const struct file *filp); ++int gr_acl_is_enabled(void); ++void gr_set_kernel_label(struct task_struct *task); ++void gr_set_role_label(struct task_struct *task, const uid_t uid, ++ const gid_t gid); ++int gr_set_proc_label(const struct dentry *dentry, ++ const struct vfsmount *mnt, ++ const int unsafe_share); ++__u32 gr_acl_handle_hidden_file(const struct dentry *dentry, ++ const struct vfsmount *mnt); ++__u32 gr_acl_handle_open(const struct dentry *dentry, ++ const struct vfsmount *mnt, const int fmode); ++__u32 gr_acl_handle_creat(const struct dentry *dentry, ++ const struct dentry *p_dentry, ++ const struct vfsmount *p_mnt, const int fmode, ++ const int imode); ++void gr_handle_create(const struct dentry *dentry, ++ const struct vfsmount *mnt); ++__u32 gr_acl_handle_mknod(const struct dentry *new_dentry, ++ const struct dentry *parent_dentry, ++ const struct vfsmount *parent_mnt, ++ const int mode); ++__u32 gr_acl_handle_mkdir(const struct dentry *new_dentry, ++ const struct dentry *parent_dentry, ++ const struct vfsmount *parent_mnt); ++__u32 gr_acl_handle_rmdir(const struct dentry *dentry, ++ const struct vfsmount *mnt); ++void gr_handle_delete(const ino_t ino, const dev_t dev); ++__u32 gr_acl_handle_unlink(const struct dentry *dentry, ++ const struct vfsmount *mnt); ++__u32 gr_acl_handle_symlink(const struct dentry *new_dentry, ++ const struct dentry *parent_dentry, ++ const struct vfsmount *parent_mnt, ++ const char *from); ++__u32 gr_acl_handle_link(const struct dentry *new_dentry, ++ const struct dentry *parent_dentry, ++ const struct vfsmount *parent_mnt, ++ const struct dentry *old_dentry, ++ const struct vfsmount *old_mnt, const char *to); ++int gr_acl_handle_rename(struct dentry *new_dentry, ++ struct dentry *parent_dentry, ++ const struct vfsmount *parent_mnt, ++ struct dentry *old_dentry, ++ struct inode *old_parent_inode, ++ struct vfsmount *old_mnt, const char *newname); ++void gr_handle_rename(struct inode *old_dir, struct inode *new_dir, ++ struct dentry *old_dentry, ++ struct dentry *new_dentry, ++ struct vfsmount *mnt, const __u8 replace); ++__u32 gr_check_link(const struct dentry *new_dentry, ++ const struct dentry *parent_dentry, ++ const struct vfsmount *parent_mnt, ++ const struct dentry *old_dentry, ++ const struct vfsmount *old_mnt); ++int gr_acl_handle_filldir(const struct file *file, const char *name, ++ const unsigned int namelen, const ino_t ino); ++ ++__u32 gr_acl_handle_unix(const struct dentry *dentry, ++ const struct vfsmount *mnt); ++void gr_acl_handle_exit(void); ++void gr_acl_handle_psacct(struct task_struct *task, const long code); ++int gr_acl_handle_procpidmem(const struct task_struct *task); ++int gr_handle_rofs_mount(struct dentry *dentry, struct vfsmount *mnt, int mnt_flags); ++int gr_handle_rofs_blockwrite(struct dentry *dentry, struct vfsmount *mnt, int acc_mode); ++void gr_audit_ptrace(struct task_struct *task); ++dev_t gr_get_dev_from_dentry(struct dentry *dentry); ++ ++#ifdef CONFIG_GRKERNSEC ++void task_grsec_rbac(struct seq_file *m, struct task_struct *p); ++void gr_handle_vm86(void); ++void gr_handle_mem_readwrite(u64 from, u64 to); ++ ++extern int grsec_enable_dmesg; ++extern int grsec_disable_privio; ++#endif ++ ++#endif +diff -urNp linux-2.6.38.3/include/linux/grsock.h linux-2.6.38.3/include/linux/grsock.h +--- linux-2.6.38.3/include/linux/grsock.h 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/include/linux/grsock.h 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,19 @@ ++#ifndef __GRSOCK_H ++#define __GRSOCK_H ++ ++extern void gr_attach_curr_ip(const struct sock *sk); ++extern int gr_handle_sock_all(const int family, const int type, ++ const int protocol); ++extern int gr_handle_sock_server(const struct sockaddr *sck); ++extern int gr_handle_sock_server_other(const struct sock *sck); ++extern int gr_handle_sock_client(const struct sockaddr *sck); ++extern int gr_search_connect(struct socket * sock, ++ struct sockaddr_in * addr); ++extern int gr_search_bind(struct socket * sock, ++ struct sockaddr_in * addr); ++extern int gr_search_listen(struct socket * sock); ++extern int gr_search_accept(struct socket * sock); ++extern int gr_search_socket(const int domain, const int type, ++ const int protocol); ++ ++#endif +diff -urNp linux-2.6.38.3/include/linux/highmem.h linux-2.6.38.3/include/linux/highmem.h +--- linux-2.6.38.3/include/linux/highmem.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/highmem.h 2011-04-17 15:57:32.000000000 -0400 +@@ -185,6 +185,18 @@ static inline void clear_highpage(struct + kunmap_atomic(kaddr, KM_USER0); + } + ++static inline void sanitize_highpage(struct page *page) ++{ ++ void *kaddr; ++ unsigned long flags; ++ ++ local_irq_save(flags); ++ kaddr = kmap_atomic(page, KM_CLEARPAGE); ++ clear_page(kaddr); ++ kunmap_atomic(kaddr, KM_CLEARPAGE); ++ local_irq_restore(flags); ++} ++ + static inline void zero_user_segments(struct page *page, + unsigned start1, unsigned end1, + unsigned start2, unsigned end2) +diff -urNp linux-2.6.38.3/include/linux/init.h linux-2.6.38.3/include/linux/init.h +--- linux-2.6.38.3/include/linux/init.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/init.h 2011-04-17 15:57:32.000000000 -0400 +@@ -293,13 +293,13 @@ void __init parse_early_options(char *cm + + /* Each module must use one module_init(). */ + #define module_init(initfn) \ +- static inline initcall_t __inittest(void) \ ++ static inline __used initcall_t __inittest(void) \ + { return initfn; } \ + int init_module(void) __attribute__((alias(#initfn))); + + /* This is only required if you want to be unloadable. */ + #define module_exit(exitfn) \ +- static inline exitcall_t __exittest(void) \ ++ static inline __used exitcall_t __exittest(void) \ + { return exitfn; } \ + void cleanup_module(void) __attribute__((alias(#exitfn))); + +diff -urNp linux-2.6.38.3/include/linux/init_task.h linux-2.6.38.3/include/linux/init_task.h +--- linux-2.6.38.3/include/linux/init_task.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/init_task.h 2011-04-17 15:57:32.000000000 -0400 +@@ -83,6 +83,12 @@ extern struct group_info init_groups; + #define INIT_IDS + #endif + ++#ifdef CONFIG_X86 ++#define INIT_TASK_THREAD_INFO INIT_THREAD_INFO ++#else ++#define INIT_TASK_THREAD_INFO ++#endif ++ + /* + * Because of the reduced scope of CAP_SETPCAP when filesystem + * capabilities are in effect, it is safe to allow CAP_SETPCAP to +@@ -163,6 +169,7 @@ extern struct cred init_cred; + RCU_INIT_POINTER(.cred, &init_cred), \ + .comm = "swapper", \ + .thread = INIT_THREAD, \ ++ .tinfo = INIT_TASK_THREAD_INFO, \ + .fs = &init_fs, \ + .files = &init_files, \ + .signal = &init_signals, \ +diff -urNp linux-2.6.38.3/include/linux/interrupt.h linux-2.6.38.3/include/linux/interrupt.h +--- linux-2.6.38.3/include/linux/interrupt.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/interrupt.h 2011-04-17 15:57:32.000000000 -0400 +@@ -393,7 +393,7 @@ enum + /* map softirq index to softirq name. update 'softirq_to_name' in + * kernel/softirq.c when adding a new softirq. + */ +-extern char *softirq_to_name[NR_SOFTIRQS]; ++extern const char * const softirq_to_name[NR_SOFTIRQS]; + + /* softirq mask and active fields moved to irq_cpustat_t in + * asm/hardirq.h to get better cache usage. KAO +@@ -401,12 +401,12 @@ extern char *softirq_to_name[NR_SOFTIRQS + + struct softirq_action + { +- void (*action)(struct softirq_action *); ++ void (*action)(void); + }; + + asmlinkage void do_softirq(void); + asmlinkage void __do_softirq(void); +-extern void open_softirq(int nr, void (*action)(struct softirq_action *)); ++extern void open_softirq(int nr, void (*action)(void)); + extern void softirq_init(void); + static inline void __raise_softirq_irqoff(unsigned int nr) + { +diff -urNp linux-2.6.38.3/include/linux/jbd2.h linux-2.6.38.3/include/linux/jbd2.h +--- linux-2.6.38.3/include/linux/jbd2.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/jbd2.h 2011-04-17 15:57:32.000000000 -0400 +@@ -67,7 +67,7 @@ extern u8 jbd2_journal_enable_debug; + } \ + } while (0) + #else +-#define jbd_debug(f, a...) /**/ ++#define jbd_debug(f, a...) do {} while (0) + #endif + + extern void *jbd2_alloc(size_t size, gfp_t flags); +diff -urNp linux-2.6.38.3/include/linux/jbd.h linux-2.6.38.3/include/linux/jbd.h +--- linux-2.6.38.3/include/linux/jbd.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/jbd.h 2011-04-17 15:57:32.000000000 -0400 +@@ -67,7 +67,7 @@ extern u8 journal_enable_debug; + } \ + } while (0) + #else +-#define jbd_debug(f, a...) /**/ ++#define jbd_debug(f, a...) do {} while (0) + #endif + + static inline void *jbd_alloc(size_t size, gfp_t flags) +diff -urNp linux-2.6.38.3/include/linux/kallsyms.h linux-2.6.38.3/include/linux/kallsyms.h +--- linux-2.6.38.3/include/linux/kallsyms.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/kallsyms.h 2011-04-17 15:57:32.000000000 -0400 +@@ -15,7 +15,8 @@ + + struct module; + +-#ifdef CONFIG_KALLSYMS ++#if !defined(__INCLUDED_BY_HIDESYM) || !defined(CONFIG_KALLSYMS) ++#if defined(CONFIG_KALLSYMS) && !defined(CONFIG_GRKERNSEC_HIDESYM) + /* Lookup the address for a symbol. Returns 0 if not found. */ + unsigned long kallsyms_lookup_name(const char *name); + +@@ -92,6 +93,15 @@ static inline int lookup_symbol_attrs(un + /* Stupid that this does nothing, but I didn't create this mess. */ + #define __print_symbol(fmt, addr) + #endif /*CONFIG_KALLSYMS*/ ++#else /* when included by kallsyms.c, vsnprintf.c, or ++ arch/x86/kernel/dumpstack.c, with HIDESYM enabled */ ++extern void __print_symbol(const char *fmt, unsigned long address); ++extern int sprint_symbol(char *buffer, unsigned long address); ++const char *kallsyms_lookup(unsigned long addr, ++ unsigned long *symbolsize, ++ unsigned long *offset, ++ char **modname, char *namebuf); ++#endif + + /* This macro allows us to keep printk typechecking */ + static void __check_printsym_format(const char *fmt, ...) +diff -urNp linux-2.6.38.3/include/linux/kgdb.h linux-2.6.38.3/include/linux/kgdb.h +--- linux-2.6.38.3/include/linux/kgdb.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/kgdb.h 2011-04-17 15:57:32.000000000 -0400 +@@ -269,22 +269,22 @@ struct kgdb_arch { + */ + struct kgdb_io { + const char *name; +- int (*read_char) (void); +- void (*write_char) (u8); +- void (*flush) (void); +- int (*init) (void); +- void (*pre_exception) (void); +- void (*post_exception) (void); ++ int (* const read_char) (void); ++ void (* const write_char) (u8); ++ void (* const flush) (void); ++ int (* const init) (void); ++ void (* const pre_exception) (void); ++ void (* const post_exception) (void); + int is_console; + }; + +-extern struct kgdb_arch arch_kgdb_ops; ++extern const struct kgdb_arch arch_kgdb_ops; + + extern unsigned long __weak kgdb_arch_pc(int exception, struct pt_regs *regs); + +-extern int kgdb_register_io_module(struct kgdb_io *local_kgdb_io_ops); +-extern void kgdb_unregister_io_module(struct kgdb_io *local_kgdb_io_ops); +-extern struct kgdb_io *dbg_io_ops; ++extern int kgdb_register_io_module(const struct kgdb_io *local_kgdb_io_ops); ++extern void kgdb_unregister_io_module(const struct kgdb_io *local_kgdb_io_ops); ++extern const struct kgdb_io *dbg_io_ops; + + extern int kgdb_hex2long(char **ptr, unsigned long *long_val); + extern char *kgdb_mem2hex(char *mem, char *buf, int count); +diff -urNp linux-2.6.38.3/include/linux/kmod.h linux-2.6.38.3/include/linux/kmod.h +--- linux-2.6.38.3/include/linux/kmod.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/kmod.h 2011-04-17 15:57:32.000000000 -0400 +@@ -33,6 +33,8 @@ extern char modprobe_path[]; /* for sysc + * usually useless though. */ + extern int __request_module(bool wait, const char *name, ...) \ + __attribute__((format(printf, 2, 3))); ++extern int ___request_module(bool wait, char *param_name, const char *name, ...) \ ++ __attribute__((format(printf, 3, 4))); + #define request_module(mod...) __request_module(true, mod) + #define request_module_nowait(mod...) __request_module(false, mod) + #define try_then_request_module(x, mod...) \ +diff -urNp linux-2.6.38.3/include/linux/kvm_host.h linux-2.6.38.3/include/linux/kvm_host.h +--- linux-2.6.38.3/include/linux/kvm_host.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/kvm_host.h 2011-04-17 15:57:32.000000000 -0400 +@@ -288,7 +288,7 @@ void kvm_vcpu_uninit(struct kvm_vcpu *vc + void vcpu_load(struct kvm_vcpu *vcpu); + void vcpu_put(struct kvm_vcpu *vcpu); + +-int kvm_init(void *opaque, unsigned vcpu_size, unsigned vcpu_align, ++int kvm_init(const void *opaque, unsigned vcpu_size, unsigned vcpu_align, + struct module *module); + void kvm_exit(void); + +@@ -428,7 +428,7 @@ int kvm_arch_vcpu_ioctl_set_guest_debug( + struct kvm_guest_debug *dbg); + int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run); + +-int kvm_arch_init(void *opaque); ++int kvm_arch_init(const void *opaque); + void kvm_arch_exit(void); + + int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu); +diff -urNp linux-2.6.38.3/include/linux/libata.h linux-2.6.38.3/include/linux/libata.h +--- linux-2.6.38.3/include/linux/libata.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/libata.h 2011-04-17 15:57:32.000000000 -0400 +@@ -65,11 +65,11 @@ + #ifdef ATA_VERBOSE_DEBUG + #define VPRINTK(fmt, args...) printk(KERN_ERR "%s: " fmt, __func__, ## args) + #else +-#define VPRINTK(fmt, args...) ++#define VPRINTK(fmt, args...) do {} while (0) + #endif /* ATA_VERBOSE_DEBUG */ + #else +-#define DPRINTK(fmt, args...) +-#define VPRINTK(fmt, args...) ++#define DPRINTK(fmt, args...) do {} while (0) ++#define VPRINTK(fmt, args...) do {} while (0) + #endif /* ATA_DEBUG */ + + #define BPRINTK(fmt, args...) if (ap->flags & ATA_FLAG_DEBUGMSG) printk(KERN_ERR "%s: " fmt, __func__, ## args) +@@ -530,11 +530,11 @@ struct ata_ioports { + + struct ata_host { + spinlock_t lock; +- struct device *dev; ++ struct device *dev; + void __iomem * const *iomap; + unsigned int n_ports; + void *private_data; +- struct ata_port_operations *ops; ++ const struct ata_port_operations *ops; + unsigned long flags; + + struct mutex eh_mutex; +@@ -725,7 +725,7 @@ struct ata_link { + + struct ata_port { + struct Scsi_Host *scsi_host; /* our co-allocated scsi host */ +- struct ata_port_operations *ops; ++ const struct ata_port_operations *ops; + spinlock_t *lock; + /* Flags owned by the EH context. Only EH should touch these once the + port is active */ +@@ -913,7 +913,7 @@ struct ata_port_info { + unsigned long pio_mask; + unsigned long mwdma_mask; + unsigned long udma_mask; +- struct ata_port_operations *port_ops; ++ const struct ata_port_operations *port_ops; + void *private_data; + }; + +@@ -937,7 +937,7 @@ extern const unsigned long sata_deb_timi + extern const unsigned long sata_deb_timing_hotplug[]; + extern const unsigned long sata_deb_timing_long[]; + +-extern struct ata_port_operations ata_dummy_port_ops; ++extern const struct ata_port_operations ata_dummy_port_ops; + extern const struct ata_port_info ata_dummy_port_info; + + static inline const unsigned long * +@@ -983,7 +983,7 @@ extern int ata_host_activate(struct ata_ + struct scsi_host_template *sht); + extern void ata_host_detach(struct ata_host *host); + extern void ata_host_init(struct ata_host *, struct device *, +- unsigned long, struct ata_port_operations *); ++ unsigned long, const struct ata_port_operations *); + extern int ata_scsi_detect(struct scsi_host_template *sht); + extern int ata_scsi_ioctl(struct scsi_device *dev, int cmd, void __user *arg); + extern int ata_scsi_queuecmd(struct Scsi_Host *h, struct scsi_cmnd *cmd); +diff -urNp linux-2.6.38.3/include/linux/lockd/bind.h linux-2.6.38.3/include/linux/lockd/bind.h +--- linux-2.6.38.3/include/linux/lockd/bind.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/lockd/bind.h 2011-04-17 15:57:32.000000000 -0400 +@@ -23,13 +23,13 @@ struct svc_rqst; + * This is the set of functions for lockd->nfsd communication + */ + struct nlmsvc_binding { +- __be32 (*fopen)(struct svc_rqst *, ++ __be32 (* const fopen)(struct svc_rqst *, + struct nfs_fh *, + struct file **); +- void (*fclose)(struct file *); ++ void (* const fclose)(struct file *); + }; + +-extern struct nlmsvc_binding * nlmsvc_ops; ++extern const struct nlmsvc_binding * nlmsvc_ops; + + /* + * Similar to nfs_client_initdata, but without the NFS-specific +diff -urNp linux-2.6.38.3/include/linux/mm.h linux-2.6.38.3/include/linux/mm.h +--- linux-2.6.38.3/include/linux/mm.h 2011-03-28 17:42:40.000000000 -0400 ++++ linux-2.6.38.3/include/linux/mm.h 2011-04-17 15:57:32.000000000 -0400 +@@ -113,7 +113,14 @@ extern unsigned int kobjsize(const void + + #define VM_CAN_NONLINEAR 0x08000000 /* Has ->fault & does nonlinear pages */ + #define VM_MIXEDMAP 0x10000000 /* Can contain "struct page" and pure PFN pages */ ++ ++#if defined(CONFIG_PAX_PAGEEXEC) && defined(CONFIG_X86_32) ++#define VM_SAO 0x00000000 /* Strong Access Ordering (powerpc) */ ++#define VM_PAGEEXEC 0x20000000 /* vma->vm_page_prot needs special handling */ ++#else + #define VM_SAO 0x20000000 /* Strong Access Ordering (powerpc) */ ++#endif ++ + #define VM_PFN_AT_MMAP 0x40000000 /* PFNMAP vma that is fully mapped at mmap time */ + #define VM_MERGEABLE 0x80000000 /* KSM may merge identical pages */ + +@@ -992,12 +999,6 @@ int set_page_dirty(struct page *page); + int set_page_dirty_lock(struct page *page); + int clear_page_dirty_for_io(struct page *page); + +-/* Is the vma a continuation of the stack vma above it? */ +-static inline int vma_stack_continue(struct vm_area_struct *vma, unsigned long addr) +-{ +- return vma && (vma->vm_end == addr) && (vma->vm_flags & VM_GROWSDOWN); +-} +- + extern unsigned long move_page_tables(struct vm_area_struct *vma, + unsigned long old_addr, struct vm_area_struct *new_vma, + unsigned long new_addr, unsigned long len); +@@ -1149,6 +1150,15 @@ struct shrinker { + extern void register_shrinker(struct shrinker *); + extern void unregister_shrinker(struct shrinker *); + ++#ifdef CONFIG_MMU ++pgprot_t vm_get_page_prot(unsigned long vm_flags); ++#else ++static inline pgprot_t vm_get_page_prot(unsigned long vm_flags) ++{ ++ return __pgprot(0); ++} ++#endif ++ + int vma_wants_writenotify(struct vm_area_struct *vma); + + extern pte_t *__get_locked_pte(struct mm_struct *mm, unsigned long addr, +@@ -1438,6 +1448,7 @@ out: + } + + extern int do_munmap(struct mm_struct *, unsigned long, size_t); ++extern int __do_munmap(struct mm_struct *, unsigned long, size_t); + + extern unsigned long do_brk(unsigned long, unsigned long); + +@@ -1494,6 +1505,10 @@ extern struct vm_area_struct * find_vma( + extern struct vm_area_struct * find_vma_prev(struct mm_struct * mm, unsigned long addr, + struct vm_area_struct **pprev); + ++extern struct vm_area_struct *pax_find_mirror_vma(struct vm_area_struct *vma); ++extern __must_check long pax_mirror_vma(struct vm_area_struct *vma_m, struct vm_area_struct *vma); ++extern void pax_mirror_file_pte(struct vm_area_struct *vma, unsigned long address, struct page *page_m, spinlock_t *ptl); ++ + /* Look up the first VMA which intersects the interval start_addr..end_addr-1, + NULL if none. Assume start_addr < end_addr. */ + static inline struct vm_area_struct * find_vma_intersection(struct mm_struct * mm, unsigned long start_addr, unsigned long end_addr) +@@ -1510,15 +1525,6 @@ static inline unsigned long vma_pages(st + return (vma->vm_end - vma->vm_start) >> PAGE_SHIFT; + } + +-#ifdef CONFIG_MMU +-pgprot_t vm_get_page_prot(unsigned long vm_flags); +-#else +-static inline pgprot_t vm_get_page_prot(unsigned long vm_flags) +-{ +- return __pgprot(0); +-} +-#endif +- + struct vm_area_struct *find_extend_vma(struct mm_struct *, unsigned long addr); + int remap_pfn_range(struct vm_area_struct *, unsigned long addr, + unsigned long pfn, unsigned long size, pgprot_t); +@@ -1627,7 +1633,7 @@ extern int unpoison_memory(unsigned long + extern int sysctl_memory_failure_early_kill; + extern int sysctl_memory_failure_recovery; + extern void shake_page(struct page *p, int access); +-extern atomic_long_t mce_bad_pages; ++extern atomic_long_unchecked_t mce_bad_pages; + extern int soft_offline_page(struct page *page, int flags); + #ifdef CONFIG_MEMORY_FAILURE + int is_hwpoison_address(unsigned long addr); +@@ -1649,5 +1655,11 @@ extern void copy_user_huge_page(struct p + unsigned int pages_per_huge_page); + #endif /* CONFIG_TRANSPARENT_HUGEPAGE || CONFIG_HUGETLBFS */ + ++#ifdef CONFIG_ARCH_TRACK_EXEC_LIMIT ++extern void track_exec_limit(struct mm_struct *mm, unsigned long start, unsigned long end, unsigned long prot); ++#else ++static inline void track_exec_limit(struct mm_struct *mm, unsigned long start, unsigned long end, unsigned long prot) {} ++#endif ++ + #endif /* __KERNEL__ */ + #endif /* _LINUX_MM_H */ +diff -urNp linux-2.6.38.3/include/linux/mm_types.h linux-2.6.38.3/include/linux/mm_types.h +--- linux-2.6.38.3/include/linux/mm_types.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/mm_types.h 2011-04-17 15:57:32.000000000 -0400 +@@ -183,6 +183,8 @@ struct vm_area_struct { + #ifdef CONFIG_NUMA + struct mempolicy *vm_policy; /* NUMA policy for the VMA */ + #endif ++ ++ struct vm_area_struct *vm_mirror;/* PaX: mirror vma or NULL */ + }; + + struct core_thread { +@@ -315,6 +317,24 @@ struct mm_struct { + #endif + /* How many tasks sharing this mm are OOM_DISABLE */ + atomic_t oom_disable_count; ++ ++#if defined(CONFIG_PAX_EI_PAX) || defined(CONFIG_PAX_PT_PAX_FLAGS) || defined(CONFIG_PAX_NOEXEC) || defined(CONFIG_PAX_ASLR) ++ unsigned long pax_flags; ++#endif ++ ++#ifdef CONFIG_PAX_DLRESOLVE ++ unsigned long call_dl_resolve; ++#endif ++ ++#if defined(CONFIG_PPC32) && defined(CONFIG_PAX_EMUSIGRT) ++ unsigned long call_syscall; ++#endif ++ ++#ifdef CONFIG_PAX_ASLR ++ unsigned long delta_mmap; /* randomized offset */ ++ unsigned long delta_stack; /* randomized offset */ ++#endif ++ + }; + + /* Future-safe accessor for struct mm_struct's cpu_vm_mask. */ +diff -urNp linux-2.6.38.3/include/linux/mmu_notifier.h linux-2.6.38.3/include/linux/mmu_notifier.h +--- linux-2.6.38.3/include/linux/mmu_notifier.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/mmu_notifier.h 2011-04-17 15:57:32.000000000 -0400 +@@ -255,12 +255,12 @@ static inline void mmu_notifier_mm_destr + */ + #define ptep_clear_flush_notify(__vma, __address, __ptep) \ + ({ \ +- pte_t __pte; \ ++ pte_t ___pte; \ + struct vm_area_struct *___vma = __vma; \ + unsigned long ___address = __address; \ +- __pte = ptep_clear_flush(___vma, ___address, __ptep); \ ++ ___pte = ptep_clear_flush(___vma, ___address, __ptep); \ + mmu_notifier_invalidate_page(___vma->vm_mm, ___address); \ +- __pte; \ ++ ___pte; \ + }) + + #define pmdp_clear_flush_notify(__vma, __address, __pmdp) \ +diff -urNp linux-2.6.38.3/include/linux/mmzone.h linux-2.6.38.3/include/linux/mmzone.h +--- linux-2.6.38.3/include/linux/mmzone.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/mmzone.h 2011-04-17 15:57:32.000000000 -0400 +@@ -355,7 +355,7 @@ struct zone { + unsigned long flags; /* zone flags, see below */ + + /* Zone statistics */ +- atomic_long_t vm_stat[NR_VM_ZONE_STAT_ITEMS]; ++ atomic_long_unchecked_t vm_stat[NR_VM_ZONE_STAT_ITEMS]; + + /* + * The target ratio of ACTIVE_ANON to INACTIVE_ANON pages on +diff -urNp linux-2.6.38.3/include/linux/mod_devicetable.h linux-2.6.38.3/include/linux/mod_devicetable.h +--- linux-2.6.38.3/include/linux/mod_devicetable.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/mod_devicetable.h 2011-04-17 15:57:32.000000000 -0400 +@@ -12,7 +12,7 @@ + typedef unsigned long kernel_ulong_t; + #endif + +-#define PCI_ANY_ID (~0) ++#define PCI_ANY_ID ((__u16)~0) + + struct pci_device_id { + __u32 vendor, device; /* Vendor and device ID or PCI_ANY_ID*/ +@@ -131,7 +131,7 @@ struct usb_device_id { + #define USB_DEVICE_ID_MATCH_INT_SUBCLASS 0x0100 + #define USB_DEVICE_ID_MATCH_INT_PROTOCOL 0x0200 + +-#define HID_ANY_ID (~0) ++#define HID_ANY_ID (~0U) + + struct hid_device_id { + __u16 bus; +diff -urNp linux-2.6.38.3/include/linux/module.h linux-2.6.38.3/include/linux/module.h +--- linux-2.6.38.3/include/linux/module.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/module.h 2011-04-17 15:57:32.000000000 -0400 +@@ -324,19 +324,16 @@ struct module + int (*init)(void); + + /* If this is non-NULL, vfree after init() returns */ +- void *module_init; ++ void *module_init_rx, *module_init_rw; + + /* Here is the actual code + data, vfree'd on unload. */ +- void *module_core; ++ void *module_core_rx, *module_core_rw; + + /* Here are the sizes of the init and core sections */ +- unsigned int init_size, core_size; ++ unsigned int init_size_rw, core_size_rw; + + /* The size of the executable code in each section. */ +- unsigned int init_text_size, core_text_size; +- +- /* Size of RO sections of the module (text+rodata) */ +- unsigned int init_ro_size, core_ro_size; ++ unsigned int init_size_rx, core_size_rx; + + /* Arch-specific module values */ + struct mod_arch_specific arch; +@@ -441,16 +438,46 @@ bool is_module_address(unsigned long add + bool is_module_percpu_address(unsigned long addr); + bool is_module_text_address(unsigned long addr); + ++static inline int within_module_range(unsigned long addr, void *start, unsigned long size) ++{ ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ if (ktla_ktva(addr) >= (unsigned long)start && ++ ktla_ktva(addr) < (unsigned long)start + size) ++ return 1; ++#endif ++ ++ return ((void *)addr >= start && (void *)addr < start + size); ++} ++ ++static inline int within_module_core_rx(unsigned long addr, struct module *mod) ++{ ++ return within_module_range(addr, mod->module_core_rx, mod->core_size_rx); ++} ++ ++static inline int within_module_core_rw(unsigned long addr, struct module *mod) ++{ ++ return within_module_range(addr, mod->module_core_rw, mod->core_size_rw); ++} ++ ++static inline int within_module_init_rx(unsigned long addr, struct module *mod) ++{ ++ return within_module_range(addr, mod->module_init_rx, mod->init_size_rx); ++} ++ ++static inline int within_module_init_rw(unsigned long addr, struct module *mod) ++{ ++ return within_module_range(addr, mod->module_init_rw, mod->init_size_rw); ++} ++ + static inline int within_module_core(unsigned long addr, struct module *mod) + { +- return (unsigned long)mod->module_core <= addr && +- addr < (unsigned long)mod->module_core + mod->core_size; ++ return within_module_core_rx(addr, mod) || within_module_core_rw(addr, mod); + } + + static inline int within_module_init(unsigned long addr, struct module *mod) + { +- return (unsigned long)mod->module_init <= addr && +- addr < (unsigned long)mod->module_init + mod->init_size; ++ return within_module_init_rx(addr, mod) || within_module_init_rw(addr, mod); + } + + /* Search for module by name: must hold module_mutex. */ +diff -urNp linux-2.6.38.3/include/linux/moduleloader.h linux-2.6.38.3/include/linux/moduleloader.h +--- linux-2.6.38.3/include/linux/moduleloader.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/moduleloader.h 2011-04-17 15:57:32.000000000 -0400 +@@ -20,9 +20,21 @@ unsigned int arch_mod_section_prepend(st + sections. Returns NULL on failure. */ + void *module_alloc(unsigned long size); + ++#ifdef CONFIG_PAX_KERNEXEC ++void *module_alloc_exec(unsigned long size); ++#else ++#define module_alloc_exec(x) module_alloc(x) ++#endif ++ + /* Free memory returned from module_alloc. */ + void module_free(struct module *mod, void *module_region); + ++#ifdef CONFIG_PAX_KERNEXEC ++void module_free_exec(struct module *mod, void *module_region); ++#else ++#define module_free_exec(x, y) module_free((x), (y)) ++#endif ++ + /* Apply the given relocation to the (simplified) ELF. Return -error + or 0. */ + int apply_relocate(Elf_Shdr *sechdrs, +diff -urNp linux-2.6.38.3/include/linux/moduleparam.h linux-2.6.38.3/include/linux/moduleparam.h +--- linux-2.6.38.3/include/linux/moduleparam.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/moduleparam.h 2011-04-17 15:57:32.000000000 -0400 +@@ -255,7 +255,7 @@ static inline void __kernel_param_unlock + * @len is usually just sizeof(string). + */ + #define module_param_string(name, string, len, perm) \ +- static const struct kparam_string __param_string_##name \ ++ static const struct kparam_string __param_string_##name __used \ + = { len, string }; \ + __module_param_call(MODULE_PARAM_PREFIX, name, \ + ¶m_ops_string, \ +@@ -370,7 +370,7 @@ extern int param_get_invbool(char *buffe + * module_param_named() for why this might be necessary. + */ + #define module_param_array_named(name, array, type, nump, perm) \ +- static const struct kparam_array __param_arr_##name \ ++ static const struct kparam_array __param_arr_##name __used \ + = { ARRAY_SIZE(array), nump, ¶m_ops_##type, \ + sizeof(array[0]), array }; \ + __module_param_call(MODULE_PARAM_PREFIX, name, \ +diff -urNp linux-2.6.38.3/include/linux/mutex.h linux-2.6.38.3/include/linux/mutex.h +--- linux-2.6.38.3/include/linux/mutex.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/mutex.h 2011-04-17 17:44:37.000000000 -0400 +@@ -51,7 +51,7 @@ struct mutex { + spinlock_t wait_lock; + struct list_head wait_list; + #if defined(CONFIG_DEBUG_MUTEXES) || defined(CONFIG_SMP) +- struct thread_info *owner; ++ struct task_struct *owner; + #endif + #ifdef CONFIG_DEBUG_MUTEXES + const char *name; +diff -urNp linux-2.6.38.3/include/linux/namei.h linux-2.6.38.3/include/linux/namei.h +--- linux-2.6.38.3/include/linux/namei.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/namei.h 2011-04-17 15:57:32.000000000 -0400 +@@ -25,7 +25,7 @@ struct nameidata { + unsigned seq; + int last_type; + unsigned depth; +- char *saved_names[MAX_NESTED_LINKS + 1]; ++ const char *saved_names[MAX_NESTED_LINKS + 1]; + + /* Intent data */ + union { +@@ -88,12 +88,12 @@ extern int follow_up(struct path *); + extern struct dentry *lock_rename(struct dentry *, struct dentry *); + extern void unlock_rename(struct dentry *, struct dentry *); + +-static inline void nd_set_link(struct nameidata *nd, char *path) ++static inline void nd_set_link(struct nameidata *nd, const char *path) + { + nd->saved_names[nd->depth] = path; + } + +-static inline char *nd_get_link(struct nameidata *nd) ++static inline const char *nd_get_link(const struct nameidata *nd) + { + return nd->saved_names[nd->depth]; + } +diff -urNp linux-2.6.38.3/include/linux/netfilter/xt_gradm.h linux-2.6.38.3/include/linux/netfilter/xt_gradm.h +--- linux-2.6.38.3/include/linux/netfilter/xt_gradm.h 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/include/linux/netfilter/xt_gradm.h 2011-04-17 15:57:32.000000000 -0400 +@@ -0,0 +1,9 @@ ++#ifndef _LINUX_NETFILTER_XT_GRADM_H ++#define _LINUX_NETFILTER_XT_GRADM_H 1 ++ ++struct xt_gradm_mtinfo { ++ __u16 flags; ++ __u16 invflags; ++}; ++ ++#endif +diff -urNp linux-2.6.38.3/include/linux/oprofile.h linux-2.6.38.3/include/linux/oprofile.h +--- linux-2.6.38.3/include/linux/oprofile.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/oprofile.h 2011-04-17 15:57:32.000000000 -0400 +@@ -132,9 +132,9 @@ int oprofilefs_create_ulong(struct super + int oprofilefs_create_ro_ulong(struct super_block * sb, struct dentry * root, + char const * name, ulong * val); + +-/** Create a file for read-only access to an atomic_t. */ ++/** Create a file for read-only access to an atomic_unchecked_t. */ + int oprofilefs_create_ro_atomic(struct super_block * sb, struct dentry * root, +- char const * name, atomic_t * val); ++ char const * name, atomic_unchecked_t * val); + + /** create a directory */ + struct dentry * oprofilefs_mkdir(struct super_block * sb, struct dentry * root, +diff -urNp linux-2.6.38.3/include/linux/pipe_fs_i.h linux-2.6.38.3/include/linux/pipe_fs_i.h +--- linux-2.6.38.3/include/linux/pipe_fs_i.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/pipe_fs_i.h 2011-04-17 15:57:32.000000000 -0400 +@@ -46,9 +46,9 @@ struct pipe_buffer { + struct pipe_inode_info { + wait_queue_head_t wait; + unsigned int nrbufs, curbuf, buffers; +- unsigned int readers; +- unsigned int writers; +- unsigned int waiting_writers; ++ atomic_t readers; ++ atomic_t writers; ++ atomic_t waiting_writers; + unsigned int r_counter; + unsigned int w_counter; + struct page *tmp_page; +diff -urNp linux-2.6.38.3/include/linux/pm_runtime.h linux-2.6.38.3/include/linux/pm_runtime.h +--- linux-2.6.38.3/include/linux/pm_runtime.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/pm_runtime.h 2011-04-17 15:57:32.000000000 -0400 +@@ -89,7 +89,7 @@ static inline bool pm_runtime_enabled(st + + static inline void pm_runtime_mark_last_busy(struct device *dev) + { +- ACCESS_ONCE(dev->power.last_busy) = jiffies; ++ ACCESS_ONCE_RW(dev->power.last_busy) = jiffies; + } + + #else /* !CONFIG_PM_RUNTIME */ +diff -urNp linux-2.6.38.3/include/linux/poison.h linux-2.6.38.3/include/linux/poison.h +--- linux-2.6.38.3/include/linux/poison.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/poison.h 2011-04-17 15:57:32.000000000 -0400 +@@ -19,8 +19,8 @@ + * under normal circumstances, used to verify that nobody uses + * non-initialized list entries. + */ +-#define LIST_POISON1 ((void *) 0x00100100 + POISON_POINTER_DELTA) +-#define LIST_POISON2 ((void *) 0x00200200 + POISON_POINTER_DELTA) ++#define LIST_POISON1 ((void *) (long)0xFFFFFF01) ++#define LIST_POISON2 ((void *) (long)0xFFFFFF02) + + /********** include/linux/timer.h **********/ + /* +diff -urNp linux-2.6.38.3/include/linux/proc_fs.h linux-2.6.38.3/include/linux/proc_fs.h +--- linux-2.6.38.3/include/linux/proc_fs.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/proc_fs.h 2011-04-17 15:57:32.000000000 -0400 +@@ -155,6 +155,19 @@ static inline struct proc_dir_entry *pro + return proc_create_data(name, mode, parent, proc_fops, NULL); + } + ++static inline struct proc_dir_entry *proc_create_grsec(const char *name, mode_t mode, ++ struct proc_dir_entry *parent, const struct file_operations *proc_fops) ++{ ++#ifdef CONFIG_GRKERNSEC_PROC_USER ++ return proc_create_data(name, S_IRUSR, parent, proc_fops, NULL); ++#elif defined(CONFIG_GRKERNSEC_PROC_USERGROUP) ++ return proc_create_data(name, S_IRUSR | S_IRGRP, parent, proc_fops, NULL); ++#else ++ return proc_create_data(name, mode, parent, proc_fops, NULL); ++#endif ++} ++ ++ + static inline struct proc_dir_entry *create_proc_read_entry(const char *name, + mode_t mode, struct proc_dir_entry *base, + read_proc_t *read_proc, void * data) +diff -urNp linux-2.6.38.3/include/linux/ptrace.h linux-2.6.38.3/include/linux/ptrace.h +--- linux-2.6.38.3/include/linux/ptrace.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/ptrace.h 2011-04-17 15:57:32.000000000 -0400 +@@ -115,10 +115,10 @@ extern void __ptrace_unlink(struct task_ + extern void exit_ptrace(struct task_struct *tracer); + #define PTRACE_MODE_READ 1 + #define PTRACE_MODE_ATTACH 2 +-/* Returns 0 on success, -errno on denial. */ +-extern int __ptrace_may_access(struct task_struct *task, unsigned int mode); + /* Returns true on success, false on denial. */ + extern bool ptrace_may_access(struct task_struct *task, unsigned int mode); ++/* Returns true on success, false on denial. */ ++extern bool ptrace_may_access_log(struct task_struct *task, unsigned int mode); + + static inline int ptrace_reparented(struct task_struct *child) + { +diff -urNp linux-2.6.38.3/include/linux/random.h linux-2.6.38.3/include/linux/random.h +--- linux-2.6.38.3/include/linux/random.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/random.h 2011-04-17 15:57:32.000000000 -0400 +@@ -80,12 +80,17 @@ void srandom32(u32 seed); + + u32 prandom32(struct rnd_state *); + ++static inline unsigned long pax_get_random_long(void) ++{ ++ return random32() + (sizeof(long) > 4 ? (unsigned long)random32() << 32 : 0); ++} ++ + /* + * Handle minimum values for seeds + */ + static inline u32 __seed(u32 x, u32 m) + { +- return (x < m) ? x + m : x; ++ return (x <= m) ? x + m + 1 : x; + } + + /** +diff -urNp linux-2.6.38.3/include/linux/reiserfs_fs.h linux-2.6.38.3/include/linux/reiserfs_fs.h +--- linux-2.6.38.3/include/linux/reiserfs_fs.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/reiserfs_fs.h 2011-04-17 15:57:32.000000000 -0400 +@@ -1403,7 +1403,7 @@ static inline loff_t max_reiserfs_offset + #define REISERFS_USER_MEM 1 /* reiserfs user memory mode */ + + #define fs_generation(s) (REISERFS_SB(s)->s_generation_counter) +-#define get_generation(s) atomic_read (&fs_generation(s)) ++#define get_generation(s) atomic_read_unchecked (&fs_generation(s)) + #define FILESYSTEM_CHANGED_TB(tb) (get_generation((tb)->tb_sb) != (tb)->fs_gen) + #define __fs_changed(gen,s) (gen != get_generation (s)) + #define fs_changed(gen,s) \ +@@ -1615,24 +1615,24 @@ static inline struct super_block *sb_fro + */ + + struct item_operations { +- int (*bytes_number) (struct item_head * ih, int block_size); +- void (*decrement_key) (struct cpu_key *); +- int (*is_left_mergeable) (struct reiserfs_key * ih, ++ int (* const bytes_number) (struct item_head * ih, int block_size); ++ void (* const decrement_key) (struct cpu_key *); ++ int (* const is_left_mergeable) (struct reiserfs_key * ih, + unsigned long bsize); +- void (*print_item) (struct item_head *, char *item); +- void (*check_item) (struct item_head *, char *item); ++ void (* const print_item) (struct item_head *, char *item); ++ void (* const check_item) (struct item_head *, char *item); + +- int (*create_vi) (struct virtual_node * vn, struct virtual_item * vi, ++ int (* const create_vi) (struct virtual_node * vn, struct virtual_item * vi, + int is_affected, int insert_size); +- int (*check_left) (struct virtual_item * vi, int free, ++ int (* const check_left) (struct virtual_item * vi, int free, + int start_skip, int end_skip); +- int (*check_right) (struct virtual_item * vi, int free); +- int (*part_size) (struct virtual_item * vi, int from, int to); +- int (*unit_num) (struct virtual_item * vi); +- void (*print_vi) (struct virtual_item * vi); ++ int (* const check_right) (struct virtual_item * vi, int free); ++ int (* const part_size) (struct virtual_item * vi, int from, int to); ++ int (* const unit_num) (struct virtual_item * vi); ++ void (* const print_vi) (struct virtual_item * vi); + }; + +-extern struct item_operations *item_ops[TYPE_ANY + 1]; ++extern const struct item_operations * const item_ops[TYPE_ANY + 1]; + + #define op_bytes_number(ih,bsize) item_ops[le_ih_k_type (ih)]->bytes_number (ih, bsize) + #define op_is_left_mergeable(key,bsize) item_ops[le_key_k_type (le_key_version (key), key)]->is_left_mergeable (key, bsize) +diff -urNp linux-2.6.38.3/include/linux/reiserfs_fs_sb.h linux-2.6.38.3/include/linux/reiserfs_fs_sb.h +--- linux-2.6.38.3/include/linux/reiserfs_fs_sb.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/reiserfs_fs_sb.h 2011-04-17 15:57:32.000000000 -0400 +@@ -386,7 +386,7 @@ struct reiserfs_sb_info { + /* Comment? -Hans */ + wait_queue_head_t s_wait; + /* To be obsoleted soon by per buffer seals.. -Hans */ +- atomic_t s_generation_counter; // increased by one every time the ++ atomic_unchecked_t s_generation_counter; // increased by one every time the + // tree gets re-balanced + unsigned long s_properties; /* File system properties. Currently holds + on-disk FS format */ +diff -urNp linux-2.6.38.3/include/linux/rmap.h linux-2.6.38.3/include/linux/rmap.h +--- linux-2.6.38.3/include/linux/rmap.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/rmap.h 2011-04-17 15:57:32.000000000 -0400 +@@ -145,8 +145,8 @@ static inline void anon_vma_unlock(struc + void anon_vma_init(void); /* create anon_vma_cachep */ + int anon_vma_prepare(struct vm_area_struct *); + void unlink_anon_vmas(struct vm_area_struct *); +-int anon_vma_clone(struct vm_area_struct *, struct vm_area_struct *); +-int anon_vma_fork(struct vm_area_struct *, struct vm_area_struct *); ++int anon_vma_clone(struct vm_area_struct *, const struct vm_area_struct *); ++int anon_vma_fork(struct vm_area_struct *, const struct vm_area_struct *); + void __anon_vma_link(struct vm_area_struct *); + void anon_vma_free(struct anon_vma *); + +diff -urNp linux-2.6.38.3/include/linux/sched.h linux-2.6.38.3/include/linux/sched.h +--- linux-2.6.38.3/include/linux/sched.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/sched.h 2011-04-17 17:23:53.000000000 -0400 +@@ -99,6 +99,7 @@ struct robust_list_head; + struct bio_list; + struct fs_struct; + struct perf_event_context; ++struct linux_binprm; + + /* + * List of flags we want to share for kernel threads, +@@ -359,7 +360,7 @@ extern signed long schedule_timeout_inte + extern signed long schedule_timeout_killable(signed long timeout); + extern signed long schedule_timeout_uninterruptible(signed long timeout); + asmlinkage void schedule(void); +-extern int mutex_spin_on_owner(struct mutex *lock, struct thread_info *owner); ++extern int mutex_spin_on_owner(struct mutex *lock, struct task_struct *owner); + + struct nsproxy; + struct user_namespace; +@@ -380,10 +381,13 @@ struct user_namespace; + #define DEFAULT_MAX_MAP_COUNT (USHRT_MAX - MAPCOUNT_ELF_CORE_MARGIN) + + extern int sysctl_max_map_count; ++extern unsigned long sysctl_heap_stack_gap; + + #include <linux/aio.h> + + #ifdef CONFIG_MMU ++extern bool check_heap_stack_gap(const struct vm_area_struct *vma, unsigned long addr, unsigned long len); ++extern unsigned long skip_heap_stack_gap(const struct vm_area_struct *vma, unsigned long len); + extern void arch_pick_mmap_layout(struct mm_struct *mm); + extern unsigned long + arch_get_unmapped_area(struct file *, unsigned long, unsigned long, +@@ -628,6 +632,17 @@ struct signal_struct { + #ifdef CONFIG_TASKSTATS + struct taskstats *stats; + #endif ++ ++#ifdef CONFIG_GRKERNSEC ++ u32 curr_ip; ++ u32 saved_ip; ++ u32 gr_saddr; ++ u32 gr_daddr; ++ u16 gr_sport; ++ u16 gr_dport; ++ u8 used_accept:1; ++#endif ++ + #ifdef CONFIG_AUDIT + unsigned audit_tty; + struct tty_audit_buf *tty_audit_buf; +@@ -700,6 +715,11 @@ struct user_struct { + struct key *session_keyring; /* UID's default session keyring */ + #endif + ++#if defined(CONFIG_GRKERNSEC_KERN_LOCKOUT) || defined(CONFIG_GRKERNSEC_BRUTE) ++ unsigned int banned; ++ unsigned long ban_expires; ++#endif ++ + /* Hash table maintenance information */ + struct hlist_node uidhash_node; + uid_t uid; +@@ -1307,8 +1327,8 @@ struct task_struct { + struct list_head thread_group; + + struct completion *vfork_done; /* for vfork() */ +- int __user *set_child_tid; /* CLONE_CHILD_SETTID */ +- int __user *clear_child_tid; /* CLONE_CHILD_CLEARTID */ ++ pid_t __user *set_child_tid; /* CLONE_CHILD_SETTID */ ++ pid_t __user *clear_child_tid; /* CLONE_CHILD_CLEARTID */ + + cputime_t utime, stime, utimescaled, stimescaled; + cputime_t gtime; +@@ -1324,13 +1344,6 @@ struct task_struct { + struct task_cputime cputime_expires; + struct list_head cpu_timers[3]; + +-/* process credentials */ +- const struct cred __rcu *real_cred; /* objective and real subjective task +- * credentials (COW) */ +- const struct cred __rcu *cred; /* effective (overridable) subjective task +- * credentials (COW) */ +- struct cred *replacement_session_keyring; /* for KEYCTL_SESSION_TO_PARENT */ +- + char comm[TASK_COMM_LEN]; /* executable name excluding path + - access with [gs]et_task_comm (which lock + it with task_lock()) +@@ -1347,8 +1360,16 @@ struct task_struct { + #endif + /* CPU-specific state of this task */ + struct thread_struct thread; ++/* thread_info moved to task_struct */ ++#ifdef CONFIG_X86 ++ struct thread_info tinfo; ++#endif + /* filesystem information */ + struct fs_struct *fs; ++ ++ const struct cred __rcu *cred; /* effective (overridable) subjective task ++ * credentials (COW) */ ++ + /* open file information */ + struct files_struct *files; + /* namespaces */ +@@ -1395,6 +1416,11 @@ struct task_struct { + struct rt_mutex_waiter *pi_blocked_on; + #endif + ++/* process credentials */ ++ const struct cred __rcu *real_cred; /* objective and real subjective task ++ * credentials (COW) */ ++ struct cred *replacement_session_keyring; /* for KEYCTL_SESSION_TO_PARENT */ ++ + #ifdef CONFIG_DEBUG_MUTEXES + /* mutex deadlock detection */ + struct mutex_waiter *blocked_on; +@@ -1499,6 +1525,21 @@ struct task_struct { + unsigned long default_timer_slack_ns; + + struct list_head *scm_work_list; ++ ++#ifdef CONFIG_GRKERNSEC ++ /* grsecurity */ ++ struct dentry *gr_chroot_dentry; ++ struct acl_subject_label *acl; ++ struct acl_role_label *role; ++ struct file *exec_file; ++ u16 acl_role_id; ++ /* is this the task that authenticated to the special role */ ++ u8 acl_sp_role; ++ u8 is_writable; ++ u8 brute; ++ u8 gr_is_chrooted; ++#endif ++ + #ifdef CONFIG_FUNCTION_GRAPH_TRACER + /* Index of current stored address in ret_stack */ + int curr_ret_stack; +@@ -1530,6 +1571,51 @@ struct task_struct { + #endif + }; + ++#define MF_PAX_PAGEEXEC 0x01000000 /* Paging based non-executable pages */ ++#define MF_PAX_EMUTRAMP 0x02000000 /* Emulate trampolines */ ++#define MF_PAX_MPROTECT 0x04000000 /* Restrict mprotect() */ ++#define MF_PAX_RANDMMAP 0x08000000 /* Randomize mmap() base */ ++/*#define MF_PAX_RANDEXEC 0x10000000*/ /* Randomize ET_EXEC base */ ++#define MF_PAX_SEGMEXEC 0x20000000 /* Segmentation based non-executable pages */ ++ ++#ifdef CONFIG_PAX_SOFTMODE ++extern unsigned int pax_softmode; ++#endif ++ ++extern int pax_check_flags(unsigned long *); ++ ++/* if tsk != current then task_lock must be held on it */ ++#if defined(CONFIG_PAX_NOEXEC) || defined(CONFIG_PAX_ASLR) ++static inline unsigned long pax_get_flags(struct task_struct *tsk) ++{ ++ if (likely(tsk->mm)) ++ return tsk->mm->pax_flags; ++ else ++ return 0UL; ++} ++ ++/* if tsk != current then task_lock must be held on it */ ++static inline long pax_set_flags(struct task_struct *tsk, unsigned long flags) ++{ ++ if (likely(tsk->mm)) { ++ tsk->mm->pax_flags = flags; ++ return 0; ++ } ++ return -EINVAL; ++} ++#endif ++ ++#ifdef CONFIG_PAX_HAVE_ACL_FLAGS ++extern void pax_set_initial_flags(struct linux_binprm *bprm); ++#elif defined(CONFIG_PAX_HOOK_ACL_FLAGS) ++extern void (*pax_set_initial_flags_func)(struct linux_binprm *bprm); ++#endif ++ ++void pax_report_fault(struct pt_regs *regs, void *pc, void *sp); ++void pax_report_insns(void *pc, void *sp); ++void pax_report_refcount_overflow(struct pt_regs *regs); ++void pax_report_usercopy(const void *ptr, unsigned long len, bool to, const char *type); ++ + /* Future-safe accessor for struct task_struct's cpus_allowed. */ + #define tsk_cpus_allowed(tsk) (&(tsk)->cpus_allowed) + +@@ -2169,7 +2255,7 @@ extern void __cleanup_sighand(struct sig + extern void exit_itimers(struct signal_struct *); + extern void flush_itimer_signals(void); + +-extern NORET_TYPE void do_group_exit(int); ++extern NORET_TYPE void do_group_exit(int) ATTRIB_NORET; + + extern void daemonize(const char *, ...); + extern int allow_signal(int); +@@ -2310,13 +2396,17 @@ static inline unsigned long *end_of_stac + + #endif + +-static inline int object_is_on_stack(void *obj) ++static inline int object_starts_on_stack(void *obj) + { +- void *stack = task_stack_page(current); ++ const void *stack = task_stack_page(current); + + return (obj >= stack) && (obj < (stack + THREAD_SIZE)); + } + ++#ifdef CONFIG_PAX_USERCOPY ++extern int object_is_on_stack(const void *obj, unsigned long len); ++#endif ++ + extern void thread_info_cache_init(void); + + #ifdef CONFIG_DEBUG_STACK_USAGE +diff -urNp linux-2.6.38.3/include/linux/screen_info.h linux-2.6.38.3/include/linux/screen_info.h +--- linux-2.6.38.3/include/linux/screen_info.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/screen_info.h 2011-04-17 15:57:32.000000000 -0400 +@@ -43,7 +43,8 @@ struct screen_info { + __u16 pages; /* 0x32 */ + __u16 vesa_attributes; /* 0x34 */ + __u32 capabilities; /* 0x36 */ +- __u8 _reserved[6]; /* 0x3a */ ++ __u16 vesapm_size; /* 0x3a */ ++ __u8 _reserved[4]; /* 0x3c */ + } __attribute__((packed)); + + #define VIDEO_TYPE_MDA 0x10 /* Monochrome Text Display */ +diff -urNp linux-2.6.38.3/include/linux/security.h linux-2.6.38.3/include/linux/security.h +--- linux-2.6.38.3/include/linux/security.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/security.h 2011-04-17 15:57:32.000000000 -0400 +@@ -35,6 +35,7 @@ + #include <linux/key.h> + #include <linux/xfrm.h> + #include <linux/slab.h> ++#include <linux/grsecurity.h> + #include <net/flow.h> + + /* Maximum number of letters for an LSM name string */ +diff -urNp linux-2.6.38.3/include/linux/shm.h linux-2.6.38.3/include/linux/shm.h +--- linux-2.6.38.3/include/linux/shm.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/shm.h 2011-04-17 15:57:32.000000000 -0400 +@@ -95,6 +95,10 @@ struct shmid_kernel /* private to the ke + pid_t shm_cprid; + pid_t shm_lprid; + struct user_struct *mlock_user; ++#ifdef CONFIG_GRKERNSEC ++ time_t shm_createtime; ++ pid_t shm_lapid; ++#endif + }; + + /* shm_mode upper byte flags */ +diff -urNp linux-2.6.38.3/include/linux/skbuff.h linux-2.6.38.3/include/linux/skbuff.h +--- linux-2.6.38.3/include/linux/skbuff.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/skbuff.h 2011-04-17 15:57:32.000000000 -0400 +@@ -589,7 +589,7 @@ static inline struct skb_shared_hwtstamp + */ + static inline int skb_queue_empty(const struct sk_buff_head *list) + { +- return list->next == (struct sk_buff *)list; ++ return list->next == (const struct sk_buff *)list; + } + + /** +@@ -602,7 +602,7 @@ static inline int skb_queue_empty(const + static inline bool skb_queue_is_last(const struct sk_buff_head *list, + const struct sk_buff *skb) + { +- return skb->next == (struct sk_buff *)list; ++ return skb->next == (const struct sk_buff *)list; + } + + /** +@@ -615,7 +615,7 @@ static inline bool skb_queue_is_last(con + static inline bool skb_queue_is_first(const struct sk_buff_head *list, + const struct sk_buff *skb) + { +- return skb->prev == (struct sk_buff *)list; ++ return skb->prev == (const struct sk_buff *)list; + } + + /** +diff -urNp linux-2.6.38.3/include/linux/slab.h linux-2.6.38.3/include/linux/slab.h +--- linux-2.6.38.3/include/linux/slab.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/slab.h 2011-04-17 15:57:32.000000000 -0400 +@@ -11,12 +11,20 @@ + + #include <linux/gfp.h> + #include <linux/types.h> ++#include <linux/err.h> + + /* + * Flags to pass to kmem_cache_create(). + * The ones marked DEBUG are only valid if CONFIG_SLAB_DEBUG is set. + */ + #define SLAB_DEBUG_FREE 0x00000100UL /* DEBUG: Perform (expensive) checks on free */ ++ ++#ifdef CONFIG_PAX_USERCOPY ++#define SLAB_USERCOPY 0x00000200UL /* PaX: Allow copying objs to/from userland */ ++#else ++#define SLAB_USERCOPY 0x00000000UL ++#endif ++ + #define SLAB_RED_ZONE 0x00000400UL /* DEBUG: Red zone objs in a cache */ + #define SLAB_POISON 0x00000800UL /* DEBUG: Poison objects */ + #define SLAB_HWCACHE_ALIGN 0x00002000UL /* Align objs on cache lines */ +@@ -87,10 +95,13 @@ + * ZERO_SIZE_PTR can be passed to kfree though in the same way that NULL can. + * Both make kfree a no-op. + */ +-#define ZERO_SIZE_PTR ((void *)16) ++#define ZERO_SIZE_PTR \ ++({ \ ++ BUILD_BUG_ON(!(MAX_ERRNO & ~PAGE_MASK));\ ++ (void *)(-MAX_ERRNO-1L); \ ++}) + +-#define ZERO_OR_NULL_PTR(x) ((unsigned long)(x) <= \ +- (unsigned long)ZERO_SIZE_PTR) ++#define ZERO_OR_NULL_PTR(x) ((unsigned long)(x) - 1 >= (unsigned long)ZERO_SIZE_PTR - 1) + + /* + * struct kmem_cache related prototypes +@@ -142,6 +153,7 @@ void * __must_check krealloc(const void + void kfree(const void *); + void kzfree(const void *); + size_t ksize(const void *); ++void check_object_size(const void *ptr, unsigned long n, bool to); + + /* + * Allocator specific definitions. These are mainly used to establish optimized +@@ -334,4 +346,37 @@ static inline void *kzalloc_node(size_t + + void __init kmem_cache_init_late(void); + ++#define kmalloc(x, y) \ ++({ \ ++ void *___retval; \ ++ intoverflow_t ___x = (intoverflow_t)x; \ ++ if (WARN(___x > ULONG_MAX, "kmalloc size overflow\n"))\ ++ ___retval = NULL; \ ++ else \ ++ ___retval = kmalloc((size_t)___x, (y)); \ ++ ___retval; \ ++}) ++ ++#define kmalloc_node(x, y, z) \ ++({ \ ++ void *___retval; \ ++ intoverflow_t ___x = (intoverflow_t)x; \ ++ if (WARN(___x > ULONG_MAX, "kmalloc_node size overflow\n"))\ ++ ___retval = NULL; \ ++ else \ ++ ___retval = kmalloc_node((size_t)___x, (y), (z));\ ++ ___retval; \ ++}) ++ ++#define kzalloc(x, y) \ ++({ \ ++ void *___retval; \ ++ intoverflow_t ___x = (intoverflow_t)x; \ ++ if (WARN(___x > ULONG_MAX, "kzalloc size overflow\n"))\ ++ ___retval = NULL; \ ++ else \ ++ ___retval = kzalloc((size_t)___x, (y)); \ ++ ___retval; \ ++}) ++ + #endif /* _LINUX_SLAB_H */ +diff -urNp linux-2.6.38.3/include/linux/slub_def.h linux-2.6.38.3/include/linux/slub_def.h +--- linux-2.6.38.3/include/linux/slub_def.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/slub_def.h 2011-04-17 15:57:32.000000000 -0400 +@@ -79,7 +79,7 @@ struct kmem_cache { + struct kmem_cache_order_objects max; + struct kmem_cache_order_objects min; + gfp_t allocflags; /* gfp flags to use on each alloc */ +- int refcount; /* Refcount for slab cache destroy */ ++ atomic_t refcount; /* Refcount for slab cache destroy */ + void (*ctor)(void *); + int inuse; /* Offset to metadata */ + int align; /* Alignment */ +diff -urNp linux-2.6.38.3/include/linux/sonet.h linux-2.6.38.3/include/linux/sonet.h +--- linux-2.6.38.3/include/linux/sonet.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/sonet.h 2011-04-17 15:57:32.000000000 -0400 +@@ -61,7 +61,7 @@ struct sonet_stats { + #include <asm/atomic.h> + + struct k_sonet_stats { +-#define __HANDLE_ITEM(i) atomic_t i ++#define __HANDLE_ITEM(i) atomic_unchecked_t i + __SONET_ITEMS + #undef __HANDLE_ITEM + }; +diff -urNp linux-2.6.38.3/include/linux/sunrpc/clnt.h linux-2.6.38.3/include/linux/sunrpc/clnt.h +--- linux-2.6.38.3/include/linux/sunrpc/clnt.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/sunrpc/clnt.h 2011-04-17 15:57:32.000000000 -0400 +@@ -168,9 +168,9 @@ static inline unsigned short rpc_get_por + { + switch (sap->sa_family) { + case AF_INET: +- return ntohs(((struct sockaddr_in *)sap)->sin_port); ++ return ntohs(((const struct sockaddr_in *)sap)->sin_port); + case AF_INET6: +- return ntohs(((struct sockaddr_in6 *)sap)->sin6_port); ++ return ntohs(((const struct sockaddr_in6 *)sap)->sin6_port); + } + return 0; + } +@@ -203,7 +203,7 @@ static inline bool __rpc_cmp_addr4(const + static inline bool __rpc_copy_addr4(struct sockaddr *dst, + const struct sockaddr *src) + { +- const struct sockaddr_in *ssin = (struct sockaddr_in *) src; ++ const struct sockaddr_in *ssin = (const struct sockaddr_in *) src; + struct sockaddr_in *dsin = (struct sockaddr_in *) dst; + + dsin->sin_family = ssin->sin_family; +@@ -300,7 +300,7 @@ static inline u32 rpc_get_scope_id(const + if (sa->sa_family != AF_INET6) + return 0; + +- return ((struct sockaddr_in6 *) sa)->sin6_scope_id; ++ return ((const struct sockaddr_in6 *) sa)->sin6_scope_id; + } + + #endif /* __KERNEL__ */ +diff -urNp linux-2.6.38.3/include/linux/suspend.h linux-2.6.38.3/include/linux/suspend.h +--- linux-2.6.38.3/include/linux/suspend.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/suspend.h 2011-04-17 15:57:32.000000000 -0400 +@@ -106,15 +106,15 @@ typedef int __bitwise suspend_state_t; + * which require special recovery actions in that situation. + */ + struct platform_suspend_ops { +- int (*valid)(suspend_state_t state); +- int (*begin)(suspend_state_t state); +- int (*prepare)(void); +- int (*prepare_late)(void); +- int (*enter)(suspend_state_t state); +- void (*wake)(void); +- void (*finish)(void); +- void (*end)(void); +- void (*recover)(void); ++ int (* const valid)(suspend_state_t state); ++ int (* const begin)(suspend_state_t state); ++ int (* const prepare)(void); ++ int (* const prepare_late)(void); ++ int (* const enter)(suspend_state_t state); ++ void (* const wake)(void); ++ void (* const finish)(void); ++ void (* const end)(void); ++ void (* const recover)(void); + }; + + #ifdef CONFIG_SUSPEND +@@ -217,16 +217,16 @@ extern void mark_free_pages(struct zone + * platforms which require special recovery actions in that situation. + */ + struct platform_hibernation_ops { +- int (*begin)(void); +- void (*end)(void); +- int (*pre_snapshot)(void); +- void (*finish)(void); +- int (*prepare)(void); +- int (*enter)(void); +- void (*leave)(void); +- int (*pre_restore)(void); +- void (*restore_cleanup)(void); +- void (*recover)(void); ++ int (* const begin)(void); ++ void (* const end)(void); ++ int (* const pre_snapshot)(void); ++ void (* const finish)(void); ++ int (* const prepare)(void); ++ int (* const enter)(void); ++ void (* const leave)(void); ++ int (* const pre_restore)(void); ++ void (* const restore_cleanup)(void); ++ void (* const recover)(void); + }; + + #ifdef CONFIG_HIBERNATION +diff -urNp linux-2.6.38.3/include/linux/sysctl.h linux-2.6.38.3/include/linux/sysctl.h +--- linux-2.6.38.3/include/linux/sysctl.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/sysctl.h 2011-04-17 15:57:32.000000000 -0400 +@@ -155,7 +155,11 @@ enum + KERN_PANIC_ON_NMI=76, /* int: whether we will panic on an unrecovered */ + }; + +- ++#ifdef CONFIG_PAX_SOFTMODE ++enum { ++ PAX_SOFTMODE=1 /* PaX: disable/enable soft mode */ ++}; ++#endif + + /* CTL_VM names: */ + enum +@@ -967,6 +971,8 @@ typedef int proc_handler (struct ctl_tab + + extern int proc_dostring(struct ctl_table *, int, + void __user *, size_t *, loff_t *); ++extern int proc_dostring_modpriv(struct ctl_table *, int, ++ void __user *, size_t *, loff_t *); + extern int proc_dointvec(struct ctl_table *, int, + void __user *, size_t *, loff_t *); + extern int proc_dointvec_minmax(struct ctl_table *, int, +diff -urNp linux-2.6.38.3/include/linux/sysfs.h linux-2.6.38.3/include/linux/sysfs.h +--- linux-2.6.38.3/include/linux/sysfs.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/sysfs.h 2011-04-17 15:57:32.000000000 -0400 +@@ -110,8 +110,8 @@ struct bin_attribute { + #define sysfs_bin_attr_init(bin_attr) sysfs_attr_init(&(bin_attr)->attr) + + struct sysfs_ops { +- ssize_t (*show)(struct kobject *, struct attribute *,char *); +- ssize_t (*store)(struct kobject *,struct attribute *,const char *, size_t); ++ ssize_t (* const show)(struct kobject *, struct attribute *,char *); ++ ssize_t (* const store)(struct kobject *,struct attribute *,const char *, size_t); + }; + + struct sysfs_dirent; +diff -urNp linux-2.6.38.3/include/linux/tty.h linux-2.6.38.3/include/linux/tty.h +--- linux-2.6.38.3/include/linux/tty.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/tty.h 2011-04-17 15:57:32.000000000 -0400 +@@ -13,6 +13,8 @@ + #include <linux/tty_driver.h> + #include <linux/tty_ldisc.h> + #include <linux/mutex.h> ++#include <linux/poll.h> ++#include <linux/smp_lock.h> + + #include <asm/system.h> + +@@ -465,7 +467,6 @@ extern int tty_perform_flush(struct tty_ + extern dev_t tty_devnum(struct tty_struct *tty); + extern void proc_clear_tty(struct task_struct *p); + extern struct tty_struct *get_current_tty(void); +-extern void tty_default_fops(struct file_operations *fops); + extern struct tty_struct *alloc_tty_struct(void); + extern int tty_add_file(struct tty_struct *tty, struct file *file); + extern void free_tty_struct(struct tty_struct *tty); +@@ -528,6 +529,18 @@ extern void tty_ldisc_begin(void); + /* This last one is just for the tty layer internals and shouldn't be used elsewhere */ + extern void tty_ldisc_enable(struct tty_struct *tty); + ++/* tty_io.c */ ++extern ssize_t tty_read(struct file *, char __user *, size_t, loff_t *); ++extern ssize_t tty_write(struct file *, const char __user *, size_t, loff_t *); ++extern unsigned int tty_poll(struct file *, poll_table *); ++#ifdef CONFIG_COMPAT ++extern long tty_compat_ioctl(struct file *file, unsigned int cmd, ++ unsigned long arg); ++#else ++#define tty_compat_ioctl NULL ++#endif ++extern int tty_release(struct inode *, struct file *); ++extern int tty_fasync(int fd, struct file *filp, int on); + + /* n_tty.c */ + extern struct tty_ldisc_ops tty_ldisc_N_TTY; +diff -urNp linux-2.6.38.3/include/linux/tty_ldisc.h linux-2.6.38.3/include/linux/tty_ldisc.h +--- linux-2.6.38.3/include/linux/tty_ldisc.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/tty_ldisc.h 2011-04-17 15:57:32.000000000 -0400 +@@ -148,7 +148,7 @@ struct tty_ldisc_ops { + + struct module *owner; + +- int refcount; ++ atomic_t refcount; + }; + + struct tty_ldisc { +diff -urNp linux-2.6.38.3/include/linux/types.h linux-2.6.38.3/include/linux/types.h +--- linux-2.6.38.3/include/linux/types.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/types.h 2011-04-17 15:57:32.000000000 -0400 +@@ -207,10 +207,26 @@ typedef struct { + int counter; + } atomic_t; + ++#ifdef CONFIG_PAX_REFCOUNT ++typedef struct { ++ int counter; ++} atomic_unchecked_t; ++#else ++typedef atomic_t atomic_unchecked_t; ++#endif ++ + #ifdef CONFIG_64BIT + typedef struct { + long counter; + } atomic64_t; ++ ++#ifdef CONFIG_PAX_REFCOUNT ++typedef struct { ++ long counter; ++} atomic64_unchecked_t; ++#else ++typedef atomic64_t atomic64_unchecked_t; ++#endif + #endif + + struct list_head { +diff -urNp linux-2.6.38.3/include/linux/uaccess.h linux-2.6.38.3/include/linux/uaccess.h +--- linux-2.6.38.3/include/linux/uaccess.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/uaccess.h 2011-04-17 15:57:32.000000000 -0400 +@@ -76,11 +76,11 @@ static inline unsigned long __copy_from_ + long ret; \ + mm_segment_t old_fs = get_fs(); \ + \ +- set_fs(KERNEL_DS); \ + pagefault_disable(); \ ++ set_fs(KERNEL_DS); \ + ret = __copy_from_user_inatomic(&(retval), (__force typeof(retval) __user *)(addr), sizeof(retval)); \ +- pagefault_enable(); \ + set_fs(old_fs); \ ++ pagefault_enable(); \ + ret; \ + }) + +@@ -93,8 +93,8 @@ static inline unsigned long __copy_from_ + * Safely read from address @src to the buffer at @dst. If a kernel fault + * happens, handle that and return -EFAULT. + */ +-extern long probe_kernel_read(void *dst, void *src, size_t size); +-extern long __probe_kernel_read(void *dst, void *src, size_t size); ++extern long probe_kernel_read(void *dst, const void *src, size_t size); ++extern long __probe_kernel_read(void *dst, const void *src, size_t size); + + /* + * probe_kernel_write(): safely attempt to write to a location +@@ -105,7 +105,7 @@ extern long __probe_kernel_read(void *ds + * Safely write to address @dst from the buffer at @src. If a kernel fault + * happens, handle that and return -EFAULT. + */ +-extern long notrace probe_kernel_write(void *dst, void *src, size_t size); +-extern long notrace __probe_kernel_write(void *dst, void *src, size_t size); ++extern long notrace probe_kernel_write(void *dst, const void *src, size_t size); ++extern long notrace __probe_kernel_write(void *dst, const void *src, size_t size); + + #endif /* __LINUX_UACCESS_H__ */ +diff -urNp linux-2.6.38.3/include/linux/unaligned/access_ok.h linux-2.6.38.3/include/linux/unaligned/access_ok.h +--- linux-2.6.38.3/include/linux/unaligned/access_ok.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/unaligned/access_ok.h 2011-04-17 15:57:32.000000000 -0400 +@@ -6,32 +6,32 @@ + + static inline u16 get_unaligned_le16(const void *p) + { +- return le16_to_cpup((__le16 *)p); ++ return le16_to_cpup((const __le16 *)p); + } + + static inline u32 get_unaligned_le32(const void *p) + { +- return le32_to_cpup((__le32 *)p); ++ return le32_to_cpup((const __le32 *)p); + } + + static inline u64 get_unaligned_le64(const void *p) + { +- return le64_to_cpup((__le64 *)p); ++ return le64_to_cpup((const __le64 *)p); + } + + static inline u16 get_unaligned_be16(const void *p) + { +- return be16_to_cpup((__be16 *)p); ++ return be16_to_cpup((const __be16 *)p); + } + + static inline u32 get_unaligned_be32(const void *p) + { +- return be32_to_cpup((__be32 *)p); ++ return be32_to_cpup((const __be32 *)p); + } + + static inline u64 get_unaligned_be64(const void *p) + { +- return be64_to_cpup((__be64 *)p); ++ return be64_to_cpup((const __be64 *)p); + } + + static inline void put_unaligned_le16(u16 val, void *p) +diff -urNp linux-2.6.38.3/include/linux/usb/hcd.h linux-2.6.38.3/include/linux/usb/hcd.h +--- linux-2.6.38.3/include/linux/usb/hcd.h 2011-03-23 17:20:08.000000000 -0400 ++++ linux-2.6.38.3/include/linux/usb/hcd.h 2011-04-17 15:57:32.000000000 -0400 +@@ -589,7 +589,7 @@ struct usb_mon_operations { + /* void (*urb_unlink)(struct usb_bus *bus, struct urb *urb); */ + }; + +-extern struct usb_mon_operations *mon_ops; ++extern const struct usb_mon_operations *mon_ops; + + static inline void usbmon_urb_submit(struct usb_bus *bus, struct urb *urb) + { +@@ -611,7 +611,7 @@ static inline void usbmon_urb_complete(s + (*mon_ops->urb_complete)(bus, urb, status); + } + +-int usb_mon_register(struct usb_mon_operations *ops); ++int usb_mon_register(const struct usb_mon_operations *ops); + void usb_mon_deregister(void); + + #else +diff -urNp linux-2.6.38.3/include/linux/vmalloc.h linux-2.6.38.3/include/linux/vmalloc.h +--- linux-2.6.38.3/include/linux/vmalloc.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/vmalloc.h 2011-04-17 15:57:32.000000000 -0400 +@@ -13,6 +13,11 @@ struct vm_area_struct; /* vma defining + #define VM_MAP 0x00000004 /* vmap()ed pages */ + #define VM_USERMAP 0x00000008 /* suitable for remap_vmalloc_range */ + #define VM_VPAGES 0x00000010 /* buffer for pages was vmalloc'ed */ ++ ++#if defined(CONFIG_MODULES) && defined(CONFIG_X86) && defined(CONFIG_PAX_KERNEXEC) ++#define VM_KERNEXEC 0x00000020 /* allocate from executable kernel memory range */ ++#endif ++ + /* bits [20..32] reserved for arch specific ioremap internals */ + + /* +@@ -123,4 +128,103 @@ struct vm_struct **pcpu_get_vm_areas(con + void pcpu_free_vm_areas(struct vm_struct **vms, int nr_vms); + #endif + ++#define vmalloc(x) \ ++({ \ ++ void *___retval; \ ++ intoverflow_t ___x = (intoverflow_t)x; \ ++ if (WARN(___x > ULONG_MAX, "vmalloc size overflow\n")) \ ++ ___retval = NULL; \ ++ else \ ++ ___retval = vmalloc((unsigned long)___x); \ ++ ___retval; \ ++}) ++ ++#define vzalloc(x) \ ++({ \ ++ void *___retval; \ ++ intoverflow_t ___x = (intoverflow_t)x; \ ++ if (WARN(___x > ULONG_MAX, "vzalloc size overflow\n")) \ ++ ___retval = NULL; \ ++ else \ ++ ___retval = vzalloc((unsigned long)___x); \ ++ ___retval; \ ++}) ++ ++#define __vmalloc(x, y, z) \ ++({ \ ++ void *___retval; \ ++ intoverflow_t ___x = (intoverflow_t)x; \ ++ if (WARN(___x > ULONG_MAX, "__vmalloc size overflow\n"))\ ++ ___retval = NULL; \ ++ else \ ++ ___retval = __vmalloc((unsigned long)___x, (y), (z));\ ++ ___retval; \ ++}) ++ ++#define vmalloc_user(x) \ ++({ \ ++ void *___retval; \ ++ intoverflow_t ___x = (intoverflow_t)x; \ ++ if (WARN(___x > ULONG_MAX, "vmalloc_user size overflow\n"))\ ++ ___retval = NULL; \ ++ else \ ++ ___retval = vmalloc_user((unsigned long)___x); \ ++ ___retval; \ ++}) ++ ++#define vmalloc_exec(x) \ ++({ \ ++ void *___retval; \ ++ intoverflow_t ___x = (intoverflow_t)x; \ ++ if (WARN(___x > ULONG_MAX, "vmalloc_exec size overflow\n"))\ ++ ___retval = NULL; \ ++ else \ ++ ___retval = vmalloc_exec((unsigned long)___x); \ ++ ___retval; \ ++}) ++ ++#define vmalloc_node(x, y) \ ++({ \ ++ void *___retval; \ ++ intoverflow_t ___x = (intoverflow_t)x; \ ++ if (WARN(___x > ULONG_MAX, "vmalloc_node size overflow\n"))\ ++ ___retval = NULL; \ ++ else \ ++ ___retval = vmalloc_node((unsigned long)___x, (y));\ ++ ___retval; \ ++}) ++ ++#define vzalloc_node(x, y) \ ++({ \ ++ void *___retval; \ ++ intoverflow_t ___x = (intoverflow_t)x; \ ++ if (WARN(___x > ULONG_MAX, "vzalloc_node size overflow\n"))\ ++ ___retval = NULL; \ ++ else \ ++ ___retval = vzalloc_node((unsigned long)___x, (y));\ ++ ___retval; \ ++}) ++ ++#define vmalloc_32(x) \ ++({ \ ++ void *___retval; \ ++ intoverflow_t ___x = (intoverflow_t)x; \ ++ if (WARN(___x > ULONG_MAX, "vmalloc_32 size overflow\n"))\ ++ ___retval = NULL; \ ++ else \ ++ ___retval = vmalloc_32((unsigned long)___x); \ ++ ___retval; \ ++}) ++ ++#define vmalloc_32_user(x) \ ++({ \ ++void *___retval; \ ++ intoverflow_t ___x = (intoverflow_t)x; \ ++ if (WARN(___x > ULONG_MAX, "vmalloc_32_user size overflow\n"))\ ++ ___retval = NULL; \ ++ else \ ++ ___retval = vmalloc_32_user((unsigned long)___x);\ ++ ___retval; \ ++}) ++ + #endif /* _LINUX_VMALLOC_H */ +diff -urNp linux-2.6.38.3/include/linux/vmstat.h linux-2.6.38.3/include/linux/vmstat.h +--- linux-2.6.38.3/include/linux/vmstat.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/linux/vmstat.h 2011-04-17 15:57:32.000000000 -0400 +@@ -140,18 +140,18 @@ static inline void vm_events_fold_cpu(in + /* + * Zone based page accounting with per cpu differentials. + */ +-extern atomic_long_t vm_stat[NR_VM_ZONE_STAT_ITEMS]; ++extern atomic_long_unchecked_t vm_stat[NR_VM_ZONE_STAT_ITEMS]; + + static inline void zone_page_state_add(long x, struct zone *zone, + enum zone_stat_item item) + { +- atomic_long_add(x, &zone->vm_stat[item]); +- atomic_long_add(x, &vm_stat[item]); ++ atomic_long_add_unchecked(x, &zone->vm_stat[item]); ++ atomic_long_add_unchecked(x, &vm_stat[item]); + } + + static inline unsigned long global_page_state(enum zone_stat_item item) + { +- long x = atomic_long_read(&vm_stat[item]); ++ long x = atomic_long_read_unchecked(&vm_stat[item]); + #ifdef CONFIG_SMP + if (x < 0) + x = 0; +@@ -162,7 +162,7 @@ static inline unsigned long global_page_ + static inline unsigned long zone_page_state(struct zone *zone, + enum zone_stat_item item) + { +- long x = atomic_long_read(&zone->vm_stat[item]); ++ long x = atomic_long_read_unchecked(&zone->vm_stat[item]); + #ifdef CONFIG_SMP + if (x < 0) + x = 0; +@@ -179,7 +179,7 @@ static inline unsigned long zone_page_st + static inline unsigned long zone_page_state_snapshot(struct zone *zone, + enum zone_stat_item item) + { +- long x = atomic_long_read(&zone->vm_stat[item]); ++ long x = atomic_long_read_unchecked(&zone->vm_stat[item]); + + #ifdef CONFIG_SMP + int cpu; +@@ -273,8 +273,8 @@ static inline void __mod_zone_page_state + + static inline void __inc_zone_state(struct zone *zone, enum zone_stat_item item) + { +- atomic_long_inc(&zone->vm_stat[item]); +- atomic_long_inc(&vm_stat[item]); ++ atomic_long_inc_unchecked(&zone->vm_stat[item]); ++ atomic_long_inc_unchecked(&vm_stat[item]); + } + + static inline void __inc_zone_page_state(struct page *page, +@@ -285,8 +285,8 @@ static inline void __inc_zone_page_state + + static inline void __dec_zone_state(struct zone *zone, enum zone_stat_item item) + { +- atomic_long_dec(&zone->vm_stat[item]); +- atomic_long_dec(&vm_stat[item]); ++ atomic_long_dec_unchecked(&zone->vm_stat[item]); ++ atomic_long_dec_unchecked(&vm_stat[item]); + } + + static inline void __dec_zone_page_state(struct page *page, +diff -urNp linux-2.6.38.3/include/net/inetpeer.h linux-2.6.38.3/include/net/inetpeer.h +--- linux-2.6.38.3/include/net/inetpeer.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/net/inetpeer.h 2011-04-17 15:57:32.000000000 -0400 +@@ -38,8 +38,8 @@ struct inet_peer { + */ + union { + struct { +- atomic_t rid; /* Frag reception counter */ +- atomic_t ip_id_count; /* IP ID for the next packet */ ++ atomic_unchecked_t rid; /* Frag reception counter */ ++ atomic_unchecked_t ip_id_count; /* IP ID for the next packet */ + __u32 tcp_ts; + __u32 tcp_ts_stamp; + }; +@@ -88,7 +88,7 @@ static inline __u16 inet_getid(struct in + { + more++; + inet_peer_refcheck(p); +- return atomic_add_return(more, &p->ip_id_count) - more; ++ return atomic_add_return_unchecked(more, &p->ip_id_count) - more; + } + + #endif /* _NET_INETPEER_H */ +diff -urNp linux-2.6.38.3/include/net/irda/ircomm_tty.h linux-2.6.38.3/include/net/irda/ircomm_tty.h +--- linux-2.6.38.3/include/net/irda/ircomm_tty.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/net/irda/ircomm_tty.h 2011-04-17 15:57:32.000000000 -0400 +@@ -35,6 +35,7 @@ + #include <linux/termios.h> + #include <linux/timer.h> + #include <linux/tty.h> /* struct tty_struct */ ++#include <asm/local.h> + + #include <net/irda/irias_object.h> + #include <net/irda/ircomm_core.h> +@@ -105,8 +106,8 @@ struct ircomm_tty_cb { + unsigned short close_delay; + unsigned short closing_wait; /* time to wait before closing */ + +- int open_count; +- int blocked_open; /* # of blocked opens */ ++ local_t open_count; ++ local_t blocked_open; /* # of blocked opens */ + + /* Protect concurent access to : + * o self->open_count +diff -urNp linux-2.6.38.3/include/net/neighbour.h linux-2.6.38.3/include/net/neighbour.h +--- linux-2.6.38.3/include/net/neighbour.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/net/neighbour.h 2011-04-17 15:57:32.000000000 -0400 +@@ -118,12 +118,12 @@ struct neighbour { + + struct neigh_ops { + int family; +- void (*solicit)(struct neighbour *, struct sk_buff*); +- void (*error_report)(struct neighbour *, struct sk_buff*); +- int (*output)(struct sk_buff*); +- int (*connected_output)(struct sk_buff*); +- int (*hh_output)(struct sk_buff*); +- int (*queue_xmit)(struct sk_buff*); ++ void (* const solicit)(struct neighbour *, struct sk_buff*); ++ void (* const error_report)(struct neighbour *, struct sk_buff*); ++ int (* const output)(struct sk_buff*); ++ int (* const connected_output)(struct sk_buff*); ++ int (* const hh_output)(struct sk_buff*); ++ int (* const queue_xmit)(struct sk_buff*); + }; + + struct pneigh_entry { +diff -urNp linux-2.6.38.3/include/net/netlink.h linux-2.6.38.3/include/net/netlink.h +--- linux-2.6.38.3/include/net/netlink.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/net/netlink.h 2011-04-17 15:57:32.000000000 -0400 +@@ -562,7 +562,7 @@ static inline void *nlmsg_get_pos(struct + static inline void nlmsg_trim(struct sk_buff *skb, const void *mark) + { + if (mark) +- skb_trim(skb, (unsigned char *) mark - skb->data); ++ skb_trim(skb, (const unsigned char *) mark - skb->data); + } + + /** +diff -urNp linux-2.6.38.3/include/net/sctp/sctp.h linux-2.6.38.3/include/net/sctp/sctp.h +--- linux-2.6.38.3/include/net/sctp/sctp.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/net/sctp/sctp.h 2011-04-17 15:57:32.000000000 -0400 +@@ -316,9 +316,9 @@ do { \ + + #else /* SCTP_DEBUG */ + +-#define SCTP_DEBUG_PRINTK(whatever...) +-#define SCTP_DEBUG_PRINTK_CONT(fmt, args...) +-#define SCTP_DEBUG_PRINTK_IPADDR(whatever...) ++#define SCTP_DEBUG_PRINTK(whatever...) do {} while (0) ++#define SCTP_DEBUG_PRINTK_CONT(fmt, args...) do {} while (0) ++#define SCTP_DEBUG_PRINTK_IPADDR(whatever...) do {} while (0) + #define SCTP_ENABLE_DEBUG + #define SCTP_DISABLE_DEBUG + #define SCTP_ASSERT(expr, str, func) +diff -urNp linux-2.6.38.3/include/net/tcp.h linux-2.6.38.3/include/net/tcp.h +--- linux-2.6.38.3/include/net/tcp.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/net/tcp.h 2011-04-17 15:57:32.000000000 -0400 +@@ -1382,7 +1382,7 @@ enum tcp_seq_states { + struct tcp_seq_afinfo { + char *name; + sa_family_t family; +- struct file_operations seq_fops; ++ struct file_operations seq_fops; /* cannot be const */ + struct seq_operations seq_ops; + }; + +diff -urNp linux-2.6.38.3/include/net/udp.h linux-2.6.38.3/include/net/udp.h +--- linux-2.6.38.3/include/net/udp.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/net/udp.h 2011-04-17 15:57:32.000000000 -0400 +@@ -223,7 +223,7 @@ struct udp_seq_afinfo { + char *name; + sa_family_t family; + struct udp_table *udp_table; +- struct file_operations seq_fops; ++ struct file_operations seq_fops; /* cannot be const */ + struct seq_operations seq_ops; + }; + +diff -urNp linux-2.6.38.3/include/sound/ac97_codec.h linux-2.6.38.3/include/sound/ac97_codec.h +--- linux-2.6.38.3/include/sound/ac97_codec.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/sound/ac97_codec.h 2011-04-17 15:57:32.000000000 -0400 +@@ -419,15 +419,15 @@ + struct snd_ac97; + + struct snd_ac97_build_ops { +- int (*build_3d) (struct snd_ac97 *ac97); +- int (*build_specific) (struct snd_ac97 *ac97); +- int (*build_spdif) (struct snd_ac97 *ac97); +- int (*build_post_spdif) (struct snd_ac97 *ac97); ++ int (* const build_3d) (struct snd_ac97 *ac97); ++ int (* const build_specific) (struct snd_ac97 *ac97); ++ int (* const build_spdif) (struct snd_ac97 *ac97); ++ int (* const build_post_spdif) (struct snd_ac97 *ac97); + #ifdef CONFIG_PM +- void (*suspend) (struct snd_ac97 *ac97); +- void (*resume) (struct snd_ac97 *ac97); ++ void (* const suspend) (struct snd_ac97 *ac97); ++ void (* const resume) (struct snd_ac97 *ac97); + #endif +- void (*update_jacks) (struct snd_ac97 *ac97); /* for jack-sharing */ ++ void (* const update_jacks) (struct snd_ac97 *ac97); /* for jack-sharing */ + }; + + struct snd_ac97_bus_ops { +diff -urNp linux-2.6.38.3/include/trace/events/irq.h linux-2.6.38.3/include/trace/events/irq.h +--- linux-2.6.38.3/include/trace/events/irq.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/trace/events/irq.h 2011-04-17 15:57:32.000000000 -0400 +@@ -36,7 +36,7 @@ struct softirq_action; + */ + TRACE_EVENT(irq_handler_entry, + +- TP_PROTO(int irq, struct irqaction *action), ++ TP_PROTO(int irq, const struct irqaction *action), + + TP_ARGS(irq, action), + +@@ -66,7 +66,7 @@ TRACE_EVENT(irq_handler_entry, + */ + TRACE_EVENT(irq_handler_exit, + +- TP_PROTO(int irq, struct irqaction *action, int ret), ++ TP_PROTO(int irq, const struct irqaction *action, int ret), + + TP_ARGS(irq, action, ret), + +diff -urNp linux-2.6.38.3/include/video/uvesafb.h linux-2.6.38.3/include/video/uvesafb.h +--- linux-2.6.38.3/include/video/uvesafb.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/include/video/uvesafb.h 2011-04-17 15:57:32.000000000 -0400 +@@ -177,6 +177,7 @@ struct uvesafb_par { + u8 ypan; /* 0 - nothing, 1 - ypan, 2 - ywrap */ + u8 pmi_setpal; /* PMI for palette changes */ + u16 *pmi_base; /* protected mode interface location */ ++ u8 *pmi_code; /* protected mode code location */ + void *pmi_start; + void *pmi_pal; + u8 *vbe_state_orig; /* +diff -urNp linux-2.6.38.3/init/do_mounts.c linux-2.6.38.3/init/do_mounts.c +--- linux-2.6.38.3/init/do_mounts.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/init/do_mounts.c 2011-04-17 15:57:32.000000000 -0400 +@@ -287,7 +287,7 @@ static void __init get_fs_names(char *pa + + static int __init do_mount_root(char *name, char *fs, int flags, void *data) + { +- int err = sys_mount(name, "/root", fs, flags, data); ++ int err = sys_mount((__force char __user *)name, (__force char __user *)"/root", (__force char __user *)fs, flags, (__force void __user *)data); + if (err) + return err; + +@@ -382,18 +382,18 @@ void __init change_floppy(char *fmt, ... + va_start(args, fmt); + vsprintf(buf, fmt, args); + va_end(args); +- fd = sys_open("/dev/root", O_RDWR | O_NDELAY, 0); ++ fd = sys_open((char __user *)"/dev/root", O_RDWR | O_NDELAY, 0); + if (fd >= 0) { + sys_ioctl(fd, FDEJECT, 0); + sys_close(fd); + } + printk(KERN_NOTICE "VFS: Insert %s and press ENTER\n", buf); +- fd = sys_open("/dev/console", O_RDWR, 0); ++ fd = sys_open((__force const char __user *)"/dev/console", O_RDWR, 0); + if (fd >= 0) { + sys_ioctl(fd, TCGETS, (long)&termios); + termios.c_lflag &= ~ICANON; + sys_ioctl(fd, TCSETSF, (long)&termios); +- sys_read(fd, &c, 1); ++ sys_read(fd, (char __user *)&c, 1); + termios.c_lflag |= ICANON; + sys_ioctl(fd, TCSETSF, (long)&termios); + sys_close(fd); +@@ -487,6 +487,6 @@ void __init prepare_namespace(void) + mount_root(); + out: + devtmpfs_mount("dev"); +- sys_mount(".", "/", NULL, MS_MOVE, NULL); ++ sys_mount((__force char __user *)".", (__force char __user *)"/", NULL, MS_MOVE, NULL); + sys_chroot((const char __user __force *)"."); + } +diff -urNp linux-2.6.38.3/init/do_mounts.h linux-2.6.38.3/init/do_mounts.h +--- linux-2.6.38.3/init/do_mounts.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/init/do_mounts.h 2011-04-17 15:57:32.000000000 -0400 +@@ -15,15 +15,15 @@ extern int root_mountflags; + + static inline int create_dev(char *name, dev_t dev) + { +- sys_unlink(name); +- return sys_mknod(name, S_IFBLK|0600, new_encode_dev(dev)); ++ sys_unlink((__force char __user *)name); ++ return sys_mknod((__force char __user *)name, S_IFBLK|0600, new_encode_dev(dev)); + } + + #if BITS_PER_LONG == 32 + static inline u32 bstat(char *name) + { + struct stat64 stat; +- if (sys_stat64(name, &stat) != 0) ++ if (sys_stat64((__force char __user *)name, (__force struct stat64 __user *)&stat) != 0) + return 0; + if (!S_ISBLK(stat.st_mode)) + return 0; +diff -urNp linux-2.6.38.3/init/do_mounts_initrd.c linux-2.6.38.3/init/do_mounts_initrd.c +--- linux-2.6.38.3/init/do_mounts_initrd.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/init/do_mounts_initrd.c 2011-04-17 15:57:32.000000000 -0400 +@@ -44,13 +44,13 @@ static void __init handle_initrd(void) + create_dev("/dev/root.old", Root_RAM0); + /* mount initrd on rootfs' /root */ + mount_block_root("/dev/root.old", root_mountflags & ~MS_RDONLY); +- sys_mkdir("/old", 0700); +- root_fd = sys_open("/", 0, 0); +- old_fd = sys_open("/old", 0, 0); ++ sys_mkdir((__force const char __user *)"/old", 0700); ++ root_fd = sys_open((__force const char __user *)"/", 0, 0); ++ old_fd = sys_open((__force const char __user *)"/old", 0, 0); + /* move initrd over / and chdir/chroot in initrd root */ +- sys_chdir("/root"); +- sys_mount(".", "/", NULL, MS_MOVE, NULL); +- sys_chroot("."); ++ sys_chdir((__force const char __user *)"/root"); ++ sys_mount((__force char __user *)".", (__force char __user *)"/", NULL, MS_MOVE, NULL); ++ sys_chroot((__force const char __user *)"."); + + /* + * In case that a resume from disk is carried out by linuxrc or one of +@@ -67,15 +67,15 @@ static void __init handle_initrd(void) + + /* move initrd to rootfs' /old */ + sys_fchdir(old_fd); +- sys_mount("/", ".", NULL, MS_MOVE, NULL); ++ sys_mount((__force char __user *)"/", (__force char __user *)".", NULL, MS_MOVE, NULL); + /* switch root and cwd back to / of rootfs */ + sys_fchdir(root_fd); +- sys_chroot("."); ++ sys_chroot((__force const char __user *)"."); + sys_close(old_fd); + sys_close(root_fd); + + if (new_decode_dev(real_root_dev) == Root_RAM0) { +- sys_chdir("/old"); ++ sys_chdir((__force const char __user *)"/old"); + return; + } + +@@ -83,17 +83,17 @@ static void __init handle_initrd(void) + mount_root(); + + printk(KERN_NOTICE "Trying to move old root to /initrd ... "); +- error = sys_mount("/old", "/root/initrd", NULL, MS_MOVE, NULL); ++ error = sys_mount((__force char __user *)"/old", (__force char __user *)"/root/initrd", NULL, MS_MOVE, NULL); + if (!error) + printk("okay\n"); + else { +- int fd = sys_open("/dev/root.old", O_RDWR, 0); ++ int fd = sys_open((__force const char __user *)"/dev/root.old", O_RDWR, 0); + if (error == -ENOENT) + printk("/initrd does not exist. Ignored.\n"); + else + printk("failed\n"); + printk(KERN_NOTICE "Unmounting old root\n"); +- sys_umount("/old", MNT_DETACH); ++ sys_umount((__force char __user *)"/old", MNT_DETACH); + printk(KERN_NOTICE "Trying to free ramdisk memory ... "); + if (fd < 0) { + error = fd; +@@ -116,11 +116,11 @@ int __init initrd_load(void) + * mounted in the normal path. + */ + if (rd_load_image("/initrd.image") && ROOT_DEV != Root_RAM0) { +- sys_unlink("/initrd.image"); ++ sys_unlink((__force const char __user *)"/initrd.image"); + handle_initrd(); + return 1; + } + } +- sys_unlink("/initrd.image"); ++ sys_unlink((__force const char __user *)"/initrd.image"); + return 0; + } +diff -urNp linux-2.6.38.3/init/do_mounts_md.c linux-2.6.38.3/init/do_mounts_md.c +--- linux-2.6.38.3/init/do_mounts_md.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/init/do_mounts_md.c 2011-04-17 15:57:32.000000000 -0400 +@@ -170,7 +170,7 @@ static void __init md_setup_drive(void) + partitioned ? "_d" : "", minor, + md_setup_args[ent].device_names); + +- fd = sys_open(name, 0, 0); ++ fd = sys_open((__force char __user *)name, 0, 0); + if (fd < 0) { + printk(KERN_ERR "md: open failed - cannot start " + "array %s\n", name); +@@ -233,7 +233,7 @@ static void __init md_setup_drive(void) + * array without it + */ + sys_close(fd); +- fd = sys_open(name, 0, 0); ++ fd = sys_open((__force char __user *)name, 0, 0); + sys_ioctl(fd, BLKRRPART, 0); + } + sys_close(fd); +diff -urNp linux-2.6.38.3/init/initramfs.c linux-2.6.38.3/init/initramfs.c +--- linux-2.6.38.3/init/initramfs.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/init/initramfs.c 2011-04-17 15:57:32.000000000 -0400 +@@ -74,7 +74,7 @@ static void __init free_hash(void) + } + } + +-static long __init do_utime(char __user *filename, time_t mtime) ++static long __init do_utime(__force char __user *filename, time_t mtime) + { + struct timespec t[2]; + +@@ -109,7 +109,7 @@ static void __init dir_utime(void) + struct dir_entry *de, *tmp; + list_for_each_entry_safe(de, tmp, &dir_list, list) { + list_del(&de->list); +- do_utime(de->name, de->mtime); ++ do_utime((__force char __user *)de->name, de->mtime); + kfree(de->name); + kfree(de); + } +@@ -271,7 +271,7 @@ static int __init maybe_link(void) + if (nlink >= 2) { + char *old = find_link(major, minor, ino, mode, collected); + if (old) +- return (sys_link(old, collected) < 0) ? -1 : 1; ++ return (sys_link((__force char __user *)old, (__force char __user *)collected) < 0) ? -1 : 1; + } + return 0; + } +@@ -280,11 +280,11 @@ static void __init clean_path(char *path + { + struct stat st; + +- if (!sys_newlstat(path, &st) && (st.st_mode^mode) & S_IFMT) { ++ if (!sys_newlstat((__force char __user *)path, (__force struct stat __user *)&st) && (st.st_mode^mode) & S_IFMT) { + if (S_ISDIR(st.st_mode)) +- sys_rmdir(path); ++ sys_rmdir((__force char __user *)path); + else +- sys_unlink(path); ++ sys_unlink((__force char __user *)path); + } + } + +@@ -305,7 +305,7 @@ static int __init do_name(void) + int openflags = O_WRONLY|O_CREAT; + if (ml != 1) + openflags |= O_TRUNC; +- wfd = sys_open(collected, openflags, mode); ++ wfd = sys_open((__force char __user *)collected, openflags, mode); + + if (wfd >= 0) { + sys_fchown(wfd, uid, gid); +@@ -317,17 +317,17 @@ static int __init do_name(void) + } + } + } else if (S_ISDIR(mode)) { +- sys_mkdir(collected, mode); +- sys_chown(collected, uid, gid); +- sys_chmod(collected, mode); ++ sys_mkdir((__force char __user *)collected, mode); ++ sys_chown((__force char __user *)collected, uid, gid); ++ sys_chmod((__force char __user *)collected, mode); + dir_add(collected, mtime); + } else if (S_ISBLK(mode) || S_ISCHR(mode) || + S_ISFIFO(mode) || S_ISSOCK(mode)) { + if (maybe_link() == 0) { +- sys_mknod(collected, mode, rdev); +- sys_chown(collected, uid, gid); +- sys_chmod(collected, mode); +- do_utime(collected, mtime); ++ sys_mknod((__force char __user *)collected, mode, rdev); ++ sys_chown((__force char __user *)collected, uid, gid); ++ sys_chmod((__force char __user *)collected, mode); ++ do_utime((__force char __user *)collected, mtime); + } + } + return 0; +@@ -336,15 +336,15 @@ static int __init do_name(void) + static int __init do_copy(void) + { + if (count >= body_len) { +- sys_write(wfd, victim, body_len); ++ sys_write(wfd, (__force char __user *)victim, body_len); + sys_close(wfd); +- do_utime(vcollected, mtime); ++ do_utime((__force char __user *)vcollected, mtime); + kfree(vcollected); + eat(body_len); + state = SkipIt; + return 0; + } else { +- sys_write(wfd, victim, count); ++ sys_write(wfd, (__force char __user *)victim, count); + body_len -= count; + eat(count); + return 1; +@@ -355,9 +355,9 @@ static int __init do_symlink(void) + { + collected[N_ALIGN(name_len) + body_len] = '\0'; + clean_path(collected, 0); +- sys_symlink(collected + N_ALIGN(name_len), collected); +- sys_lchown(collected, uid, gid); +- do_utime(collected, mtime); ++ sys_symlink((__force char __user *)collected + N_ALIGN(name_len), (__force char __user *)collected); ++ sys_lchown((__force char __user *)collected, uid, gid); ++ do_utime((__force char __user *)collected, mtime); + state = SkipIt; + next_state = Reset; + return 0; +diff -urNp linux-2.6.38.3/init/Kconfig linux-2.6.38.3/init/Kconfig +--- linux-2.6.38.3/init/Kconfig 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/init/Kconfig 2011-04-17 15:57:32.000000000 -0400 +@@ -1185,7 +1185,7 @@ config SLUB_DEBUG + + config COMPAT_BRK + bool "Disable heap randomization" +- default y ++ default n + help + Randomizing heap placement makes heap exploits harder, but it + also breaks ancient binaries (including anything libc5 based). +diff -urNp linux-2.6.38.3/init/main.c linux-2.6.38.3/init/main.c +--- linux-2.6.38.3/init/main.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/init/main.c 2011-04-17 15:57:32.000000000 -0400 +@@ -96,6 +96,8 @@ static inline void mark_rodata_ro(void) + extern void tc_init(void); + #endif + ++extern void grsecurity_init(void); ++ + /* + * Debug helper: via this flag we know that we are in 'early bootup code' + * where only the boot processor is running with IRQ disabled. This means +@@ -206,6 +208,47 @@ static int __init set_reset_devices(char + + __setup("reset_devices", set_reset_devices); + ++#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) ++extern char pax_enter_kernel_user[]; ++extern char pax_exit_kernel_user[]; ++extern pgdval_t clone_pgd_mask; ++#endif ++ ++#if defined(CONFIG_X86) && defined(CONFIG_PAX_MEMORY_UDEREF) ++static int __init setup_pax_nouderef(char *str) ++{ ++#ifdef CONFIG_X86_32 ++ unsigned int cpu; ++ ++ for (cpu = 0; cpu < NR_CPUS; cpu++) { ++ get_cpu_gdt_table(cpu)[GDT_ENTRY_KERNEL_DS].type = 3; ++ get_cpu_gdt_table(cpu)[GDT_ENTRY_KERNEL_DS].limit = 0xf; ++ get_cpu_gdt_table(cpu)[GDT_ENTRY_DEFAULT_USER_CS].limit = 0xf; ++ get_cpu_gdt_table(cpu)[GDT_ENTRY_DEFAULT_USER_DS].limit = 0xf; ++ } ++ asm("mov %0, %%ds; mov %0, %%es; mov %0, %%ss" : : "r" (__KERNEL_DS) : "memory"); ++#else ++ memcpy(pax_enter_kernel_user, (unsigned char []){0xc3}, 1); ++ memcpy(pax_exit_kernel_user, (unsigned char []){0xc3}, 1); ++ clone_pgd_mask = ~(pgdval_t)0UL; ++#endif ++ ++ return 0; ++} ++early_param("pax_nouderef", setup_pax_nouderef); ++#endif ++ ++#ifdef CONFIG_PAX_SOFTMODE ++unsigned int pax_softmode; ++ ++static int __init setup_pax_softmode(char *str) ++{ ++ get_option(&str, &pax_softmode); ++ return 1; ++} ++__setup("pax_softmode=", setup_pax_softmode); ++#endif ++ + static const char * argv_init[MAX_INIT_ARGS+2] = { "init", NULL, }; + const char * envp_init[MAX_INIT_ENVS+2] = { "HOME=/", "TERM=linux", NULL, }; + static const char *panic_later, *panic_param; +@@ -751,6 +794,7 @@ int __init_or_module do_one_initcall(ini + { + int count = preempt_count(); + int ret; ++ const char *msg1 = "", *msg2 = ""; + + if (initcall_debug) + ret = do_one_initcall_debug(fn); +@@ -763,15 +807,15 @@ int __init_or_module do_one_initcall(ini + sprintf(msgbuf, "error code %d ", ret); + + if (preempt_count() != count) { +- strlcat(msgbuf, "preemption imbalance ", sizeof(msgbuf)); ++ msg1 = " preemption imbalance"; + preempt_count() = count; + } + if (irqs_disabled()) { +- strlcat(msgbuf, "disabled interrupts ", sizeof(msgbuf)); ++ msg2 = " disabled interrupts"; + local_irq_enable(); + } +- if (msgbuf[0]) { +- printk("initcall %pF returned with %s\n", fn, msgbuf); ++ if (msgbuf[0] || *msg1 || *msg2) { ++ printk("initcall %pF returned with %s%s%s\n", fn, msgbuf, msg1, msg2); + } + + return ret; +@@ -898,7 +942,7 @@ static int __init kernel_init(void * unu + do_basic_setup(); + + /* Open the /dev/console on the rootfs, this should never fail */ +- if (sys_open((const char __user *) "/dev/console", O_RDWR, 0) < 0) ++ if (sys_open((__force const char __user *) "/dev/console", O_RDWR, 0) < 0) + printk(KERN_WARNING "Warning: unable to open an initial console.\n"); + + (void) sys_dup(0); +@@ -911,11 +955,13 @@ static int __init kernel_init(void * unu + if (!ramdisk_execute_command) + ramdisk_execute_command = "/init"; + +- if (sys_access((const char __user *) ramdisk_execute_command, 0) != 0) { ++ if (sys_access((__force const char __user *) ramdisk_execute_command, 0) != 0) { + ramdisk_execute_command = NULL; + prepare_namespace(); + } + ++ grsecurity_init(); ++ + /* + * Ok, we have completed the initial bootup, and + * we're essentially up and running. Get rid of the +diff -urNp linux-2.6.38.3/ipc/mqueue.c linux-2.6.38.3/ipc/mqueue.c +--- linux-2.6.38.3/ipc/mqueue.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/ipc/mqueue.c 2011-04-17 15:57:32.000000000 -0400 +@@ -154,6 +154,7 @@ static struct inode *mqueue_get_inode(st + mq_bytes = (mq_msg_tblsz + + (info->attr.mq_maxmsg * info->attr.mq_msgsize)); + ++ gr_learn_resource(current, RLIMIT_MSGQUEUE, u->mq_bytes + mq_bytes, 1); + spin_lock(&mq_lock); + if (u->mq_bytes + mq_bytes < u->mq_bytes || + u->mq_bytes + mq_bytes > +diff -urNp linux-2.6.38.3/ipc/shm.c linux-2.6.38.3/ipc/shm.c +--- linux-2.6.38.3/ipc/shm.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/ipc/shm.c 2011-04-17 15:57:32.000000000 -0400 +@@ -69,6 +69,14 @@ static void shm_destroy (struct ipc_name + static int sysvipc_shm_proc_show(struct seq_file *s, void *it); + #endif + ++#ifdef CONFIG_GRKERNSEC ++extern int gr_handle_shmat(const pid_t shm_cprid, const pid_t shm_lapid, ++ const time_t shm_createtime, const uid_t cuid, ++ const int shmid); ++extern int gr_chroot_shmat(const pid_t shm_cprid, const pid_t shm_lapid, ++ const time_t shm_createtime); ++#endif ++ + void shm_init_ns(struct ipc_namespace *ns) + { + ns->shm_ctlmax = SHMMAX; +@@ -401,6 +409,14 @@ static int newseg(struct ipc_namespace * + shp->shm_lprid = 0; + shp->shm_atim = shp->shm_dtim = 0; + shp->shm_ctim = get_seconds(); ++#ifdef CONFIG_GRKERNSEC ++ { ++ struct timespec timeval; ++ do_posix_clock_monotonic_gettime(&timeval); ++ ++ shp->shm_createtime = timeval.tv_sec; ++ } ++#endif + shp->shm_segsz = size; + shp->shm_nattch = 0; + shp->shm_file = file; +@@ -761,8 +777,6 @@ SYSCALL_DEFINE3(shmctl, int, shmid, int, + case SHM_LOCK: + case SHM_UNLOCK: + { +- struct file *uninitialized_var(shm_file); +- + lru_add_drain_all(); /* drain pagevecs to lru lists */ + + shp = shm_lock_check(ns, shmid); +@@ -895,9 +909,21 @@ long do_shmat(int shmid, char __user *sh + if (err) + goto out_unlock; + ++#ifdef CONFIG_GRKERNSEC ++ if (!gr_handle_shmat(shp->shm_cprid, shp->shm_lapid, shp->shm_createtime, ++ shp->shm_perm.cuid, shmid) || ++ !gr_chroot_shmat(shp->shm_cprid, shp->shm_lapid, shp->shm_createtime)) { ++ err = -EACCES; ++ goto out_unlock; ++ } ++#endif ++ + path = shp->shm_file->f_path; + path_get(&path); + shp->shm_nattch++; ++#ifdef CONFIG_GRKERNSEC ++ shp->shm_lapid = current->pid; ++#endif + size = i_size_read(path.dentry->d_inode); + shm_unlock(shp); + +diff -urNp linux-2.6.38.3/kernel/acct.c linux-2.6.38.3/kernel/acct.c +--- linux-2.6.38.3/kernel/acct.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/acct.c 2011-04-17 15:57:32.000000000 -0400 +@@ -570,7 +570,7 @@ static void do_acct_process(struct bsd_a + */ + flim = current->signal->rlim[RLIMIT_FSIZE].rlim_cur; + current->signal->rlim[RLIMIT_FSIZE].rlim_cur = RLIM_INFINITY; +- file->f_op->write(file, (char *)&ac, ++ file->f_op->write(file, (__force char __user *)&ac, + sizeof(acct_t), &file->f_pos); + current->signal->rlim[RLIMIT_FSIZE].rlim_cur = flim; + set_fs(fs); +diff -urNp linux-2.6.38.3/kernel/capability.c linux-2.6.38.3/kernel/capability.c +--- linux-2.6.38.3/kernel/capability.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/capability.c 2011-04-17 15:57:32.000000000 -0400 +@@ -205,6 +205,9 @@ SYSCALL_DEFINE2(capget, cap_user_header_ + * before modification is attempted and the application + * fails. + */ ++ if (tocopy > ARRAY_SIZE(kdata)) ++ return -EFAULT; ++ + if (copy_to_user(dataptr, kdata, tocopy + * sizeof(struct __user_cap_data_struct))) { + return -EFAULT; +@@ -306,10 +309,26 @@ int capable(int cap) + BUG(); + } + +- if (security_capable(current_cred(), cap) == 0) { ++ if (security_capable(current_cred(), cap) == 0 && gr_is_capable(cap)) { + current->flags |= PF_SUPERPRIV; + return 1; + } + return 0; + } ++ ++int capable_nolog(int cap) ++{ ++ if (unlikely(!cap_valid(cap))) { ++ printk(KERN_CRIT "capable() called with invalid cap=%u\n", cap); ++ BUG(); ++ } ++ ++ if (security_capable(current_cred(), cap) == 0 && gr_is_capable_nolog(cap)) { ++ current->flags |= PF_SUPERPRIV; ++ return 1; ++ } ++ return 0; ++} ++ + EXPORT_SYMBOL(capable); ++EXPORT_SYMBOL(capable_nolog); +diff -urNp linux-2.6.38.3/kernel/compat.c linux-2.6.38.3/kernel/compat.c +--- linux-2.6.38.3/kernel/compat.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/compat.c 2011-04-17 15:57:32.000000000 -0400 +@@ -13,6 +13,7 @@ + + #include <linux/linkage.h> + #include <linux/compat.h> ++#include <linux/module.h> + #include <linux/errno.h> + #include <linux/time.h> + #include <linux/signal.h> +diff -urNp linux-2.6.38.3/kernel/configs.c linux-2.6.38.3/kernel/configs.c +--- linux-2.6.38.3/kernel/configs.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/configs.c 2011-04-17 15:57:32.000000000 -0400 +@@ -74,8 +74,19 @@ static int __init ikconfig_init(void) + struct proc_dir_entry *entry; + + /* create the current config file */ ++#if defined(CONFIG_GRKERNSEC_PROC_ADD) || defined(CONFIG_GRKERNSEC_HIDESYM) ++#if defined(CONFIG_GRKERNSEC_PROC_USER) || defined(CONFIG_GRKERNSEC_HIDESYM) ++ entry = proc_create("config.gz", S_IFREG | S_IRUSR, NULL, ++ &ikconfig_file_ops); ++#elif defined(CONFIG_GRKERNSEC_PROC_USERGROUP) ++ entry = proc_create("config.gz", S_IFREG | S_IRUSR | S_IRGRP, NULL, ++ &ikconfig_file_ops); ++#endif ++#else + entry = proc_create("config.gz", S_IFREG | S_IRUGO, NULL, + &ikconfig_file_ops); ++#endif ++ + if (!entry) + return -ENOMEM; + +diff -urNp linux-2.6.38.3/kernel/cred.c linux-2.6.38.3/kernel/cred.c +--- linux-2.6.38.3/kernel/cred.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/cred.c 2011-04-17 15:57:32.000000000 -0400 +@@ -483,6 +483,8 @@ int commit_creds(struct cred *new) + + get_cred(new); /* we will require a ref for the subj creds too */ + ++ gr_set_role_label(task, new->uid, new->gid); ++ + /* dumpability changes */ + if (old->euid != new->euid || + old->egid != new->egid || +diff -urNp linux-2.6.38.3/kernel/debug/debug_core.c linux-2.6.38.3/kernel/debug/debug_core.c +--- linux-2.6.38.3/kernel/debug/debug_core.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/debug/debug_core.c 2011-04-17 15:57:32.000000000 -0400 +@@ -72,7 +72,7 @@ int kgdb_io_module_registered; + /* Guard for recursive entry */ + static int exception_level; + +-struct kgdb_io *dbg_io_ops; ++const struct kgdb_io *dbg_io_ops; + static DEFINE_SPINLOCK(kgdb_registration_lock); + + /* kgdb console driver is loaded */ +@@ -864,7 +864,7 @@ static void kgdb_initial_breakpoint(void + * + * Register it with the KGDB core. + */ +-int kgdb_register_io_module(struct kgdb_io *new_dbg_io_ops) ++int kgdb_register_io_module(const struct kgdb_io *new_dbg_io_ops) + { + int err; + +@@ -909,7 +909,7 @@ EXPORT_SYMBOL_GPL(kgdb_register_io_modul + * + * Unregister it with the KGDB core. + */ +-void kgdb_unregister_io_module(struct kgdb_io *old_dbg_io_ops) ++void kgdb_unregister_io_module(const struct kgdb_io *old_dbg_io_ops) + { + BUG_ON(kgdb_connected); + +diff -urNp linux-2.6.38.3/kernel/debug/kdb/kdb_main.c linux-2.6.38.3/kernel/debug/kdb/kdb_main.c +--- linux-2.6.38.3/kernel/debug/kdb/kdb_main.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/debug/kdb/kdb_main.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1980,7 +1980,7 @@ static int kdb_lsmod(int argc, const cha + list_for_each_entry(mod, kdb_modules, list) { + + kdb_printf("%-20s%8u 0x%p ", mod->name, +- mod->core_size, (void *)mod); ++ mod->core_size_rx + mod->core_size_rw, (void *)mod); + #ifdef CONFIG_MODULE_UNLOAD + kdb_printf("%4d ", module_refcount(mod)); + #endif +@@ -1990,7 +1990,7 @@ static int kdb_lsmod(int argc, const cha + kdb_printf(" (Loading)"); + else + kdb_printf(" (Live)"); +- kdb_printf(" 0x%p", mod->module_core); ++ kdb_printf(" 0x%p 0x%p", mod->module_core_rx, mod->module_core_rw); + + #ifdef CONFIG_MODULE_UNLOAD + { +diff -urNp linux-2.6.38.3/kernel/exit.c linux-2.6.38.3/kernel/exit.c +--- linux-2.6.38.3/kernel/exit.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/exit.c 2011-04-17 15:57:32.000000000 -0400 +@@ -57,6 +57,10 @@ + #include <asm/pgtable.h> + #include <asm/mmu_context.h> + ++#ifdef CONFIG_GRKERNSEC ++extern rwlock_t grsec_exec_file_lock; ++#endif ++ + static void exit_mm(struct task_struct * tsk); + + static void __unhash_process(struct task_struct *p, bool group_dead) +@@ -169,6 +173,8 @@ void release_task(struct task_struct * p + struct task_struct *leader; + int zap_leader; + repeat: ++ gr_del_task_from_ip_table(p); ++ + tracehook_prepare_release_task(p); + /* don't need to get the RCU readlock here - the process is dead and + * can't be modifying its own credentials. But shut RCU-lockdep up */ +@@ -338,11 +344,22 @@ static void reparent_to_kthreadd(void) + { + write_lock_irq(&tasklist_lock); + ++#ifdef CONFIG_GRKERNSEC ++ write_lock(&grsec_exec_file_lock); ++ if (current->exec_file) { ++ fput(current->exec_file); ++ current->exec_file = NULL; ++ } ++ write_unlock(&grsec_exec_file_lock); ++#endif ++ + ptrace_unlink(current); + /* Reparent to init */ + current->real_parent = current->parent = kthreadd_task; + list_move_tail(¤t->sibling, ¤t->real_parent->children); + ++ gr_set_kernel_label(current); ++ + /* Set the exit signal to SIGCHLD so we signal init on exit */ + current->exit_signal = SIGCHLD; + +@@ -394,7 +411,7 @@ int allow_signal(int sig) + * know it'll be handled, so that they don't get converted to + * SIGKILL or just silently dropped. + */ +- current->sighand->action[(sig)-1].sa.sa_handler = (void __user *)2; ++ current->sighand->action[(sig)-1].sa.sa_handler = (__force void __user *)2; + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); + return 0; +@@ -430,6 +447,17 @@ void daemonize(const char *name, ...) + vsnprintf(current->comm, sizeof(current->comm), name, args); + va_end(args); + ++#ifdef CONFIG_GRKERNSEC ++ write_lock(&grsec_exec_file_lock); ++ if (current->exec_file) { ++ fput(current->exec_file); ++ current->exec_file = NULL; ++ } ++ write_unlock(&grsec_exec_file_lock); ++#endif ++ ++ gr_set_kernel_label(current); ++ + /* + * If we were started as result of loading a module, close all of the + * user space pages. We don't need them, and if we didn't close them +@@ -905,17 +933,17 @@ NORET_TYPE void do_exit(long code) + struct task_struct *tsk = current; + int group_dead; + +- profile_task_exit(tsk); +- +- WARN_ON(atomic_read(&tsk->fs_excl)); +- ++ /* ++ * Check this first since set_fs() below depends on ++ * current_thread_info(), which we better not access when we're in ++ * interrupt context. Other than that, we want to do the set_fs() ++ * as early as possible. ++ */ + if (unlikely(in_interrupt())) + panic("Aiee, killing interrupt handler!"); +- if (unlikely(!tsk->pid)) +- panic("Attempted to kill the idle task!"); + + /* +- * If do_exit is called because this processes oopsed, it's possible ++ * If do_exit is called because this processes Oops'ed, it's possible + * that get_fs() was left as KERNEL_DS, so reset it to USER_DS before + * continuing. Amongst other possible reasons, this is to prevent + * mm_release()->clear_child_tid() from writing to a user-controlled +@@ -923,6 +951,13 @@ NORET_TYPE void do_exit(long code) + */ + set_fs(USER_DS); + ++ profile_task_exit(tsk); ++ ++ WARN_ON(atomic_read(&tsk->fs_excl)); ++ ++ if (unlikely(!tsk->pid)) ++ panic("Attempted to kill the idle task!"); ++ + tracehook_report_exit(&code); + + validate_creds_for_do_exit(tsk); +@@ -983,6 +1018,9 @@ NORET_TYPE void do_exit(long code) + tsk->exit_code = code; + taskstats_exit(tsk, group_dead); + ++ gr_acl_handle_psacct(tsk, code); ++ gr_acl_handle_exit(); ++ + exit_mm(tsk); + + if (group_dead) +diff -urNp linux-2.6.38.3/kernel/fork.c linux-2.6.38.3/kernel/fork.c +--- linux-2.6.38.3/kernel/fork.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/fork.c 2011-04-17 15:57:32.000000000 -0400 +@@ -280,7 +280,7 @@ static struct task_struct *dup_task_stru + *stackend = STACK_END_MAGIC; /* for overflow detection */ + + #ifdef CONFIG_CC_STACKPROTECTOR +- tsk->stack_canary = get_random_int(); ++ tsk->stack_canary = pax_get_random_long(); + #endif + + /* One for us, one for whoever does the "release_task()" (usually parent) */ +@@ -302,13 +302,78 @@ out: + } + + #ifdef CONFIG_MMU ++static struct vm_area_struct *dup_vma(struct mm_struct *mm, struct vm_area_struct *mpnt) ++{ ++ struct vm_area_struct *tmp; ++ unsigned long charge; ++ struct mempolicy *pol; ++ struct file *file; ++ ++ charge = 0; ++ if (mpnt->vm_flags & VM_ACCOUNT) { ++ unsigned int len = (mpnt->vm_end - mpnt->vm_start) >> PAGE_SHIFT; ++ if (security_vm_enough_memory(len)) ++ goto fail_nomem; ++ charge = len; ++ } ++ tmp = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); ++ if (!tmp) ++ goto fail_nomem; ++ *tmp = *mpnt; ++ tmp->vm_mm = mm; ++ INIT_LIST_HEAD(&tmp->anon_vma_chain); ++ pol = mpol_dup(vma_policy(mpnt)); ++ if (IS_ERR(pol)) ++ goto fail_nomem_policy; ++ vma_set_policy(tmp, pol); ++ if (anon_vma_fork(tmp, mpnt)) ++ goto fail_nomem_anon_vma_fork; ++ tmp->vm_flags &= ~VM_LOCKED; ++ tmp->vm_next = tmp->vm_prev = NULL; ++ tmp->vm_mirror = NULL; ++ file = tmp->vm_file; ++ if (file) { ++ struct inode *inode = file->f_path.dentry->d_inode; ++ struct address_space *mapping = file->f_mapping; ++ ++ get_file(file); ++ if (tmp->vm_flags & VM_DENYWRITE) ++ atomic_dec(&inode->i_writecount); ++ spin_lock(&mapping->i_mmap_lock); ++ if (tmp->vm_flags & VM_SHARED) ++ mapping->i_mmap_writable++; ++ tmp->vm_truncate_count = mpnt->vm_truncate_count; ++ flush_dcache_mmap_lock(mapping); ++ /* insert tmp into the share list, just after mpnt */ ++ vma_prio_tree_add(tmp, mpnt); ++ flush_dcache_mmap_unlock(mapping); ++ spin_unlock(&mapping->i_mmap_lock); ++ } ++ ++ /* ++ * Clear hugetlb-related page reserves for children. This only ++ * affects MAP_PRIVATE mappings. Faults generated by the child ++ * are not guaranteed to succeed, even if read-only ++ */ ++ if (is_vm_hugetlb_page(tmp)) ++ reset_vma_resv_huge_pages(tmp); ++ ++ return tmp; ++ ++fail_nomem_anon_vma_fork: ++ mpol_put(pol); ++fail_nomem_policy: ++ kmem_cache_free(vm_area_cachep, tmp); ++fail_nomem: ++ vm_unacct_memory(charge); ++ return NULL; ++} ++ + static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm) + { + struct vm_area_struct *mpnt, *tmp, *prev, **pprev; + struct rb_node **rb_link, *rb_parent; + int retval; +- unsigned long charge; +- struct mempolicy *pol; + + down_write(&oldmm->mmap_sem); + flush_cache_dup_mm(oldmm); +@@ -320,8 +385,8 @@ static int dup_mmap(struct mm_struct *mm + mm->locked_vm = 0; + mm->mmap = NULL; + mm->mmap_cache = NULL; +- mm->free_area_cache = oldmm->mmap_base; +- mm->cached_hole_size = ~0UL; ++ mm->free_area_cache = oldmm->free_area_cache; ++ mm->cached_hole_size = oldmm->cached_hole_size; + mm->map_count = 0; + cpumask_clear(mm_cpumask(mm)); + mm->mm_rb = RB_ROOT; +@@ -337,8 +402,6 @@ static int dup_mmap(struct mm_struct *mm + + prev = NULL; + for (mpnt = oldmm->mmap; mpnt; mpnt = mpnt->vm_next) { +- struct file *file; +- + if (mpnt->vm_flags & VM_DONTCOPY) { + long pages = vma_pages(mpnt); + mm->total_vm -= pages; +@@ -346,56 +409,13 @@ static int dup_mmap(struct mm_struct *mm + -pages); + continue; + } +- charge = 0; +- if (mpnt->vm_flags & VM_ACCOUNT) { +- unsigned int len = (mpnt->vm_end - mpnt->vm_start) >> PAGE_SHIFT; +- if (security_vm_enough_memory(len)) +- goto fail_nomem; +- charge = len; +- } +- tmp = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); +- if (!tmp) +- goto fail_nomem; +- *tmp = *mpnt; +- INIT_LIST_HEAD(&tmp->anon_vma_chain); +- pol = mpol_dup(vma_policy(mpnt)); +- retval = PTR_ERR(pol); +- if (IS_ERR(pol)) +- goto fail_nomem_policy; +- vma_set_policy(tmp, pol); +- tmp->vm_mm = mm; +- if (anon_vma_fork(tmp, mpnt)) +- goto fail_nomem_anon_vma_fork; +- tmp->vm_flags &= ~VM_LOCKED; +- tmp->vm_next = tmp->vm_prev = NULL; +- file = tmp->vm_file; +- if (file) { +- struct inode *inode = file->f_path.dentry->d_inode; +- struct address_space *mapping = file->f_mapping; +- +- get_file(file); +- if (tmp->vm_flags & VM_DENYWRITE) +- atomic_dec(&inode->i_writecount); +- spin_lock(&mapping->i_mmap_lock); +- if (tmp->vm_flags & VM_SHARED) +- mapping->i_mmap_writable++; +- tmp->vm_truncate_count = mpnt->vm_truncate_count; +- flush_dcache_mmap_lock(mapping); +- /* insert tmp into the share list, just after mpnt */ +- vma_prio_tree_add(tmp, mpnt); +- flush_dcache_mmap_unlock(mapping); +- spin_unlock(&mapping->i_mmap_lock); ++ tmp = dup_vma(mm, mpnt); ++ if (!tmp) { ++ retval = -ENOMEM; ++ goto out; + } + + /* +- * Clear hugetlb-related page reserves for children. This only +- * affects MAP_PRIVATE mappings. Faults generated by the child +- * are not guaranteed to succeed, even if read-only +- */ +- if (is_vm_hugetlb_page(tmp)) +- reset_vma_resv_huge_pages(tmp); +- +- /* + * Link in the new vma and copy the page table entries. + */ + *pprev = tmp; +@@ -416,6 +436,31 @@ static int dup_mmap(struct mm_struct *mm + if (retval) + goto out; + } ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (oldmm->pax_flags & MF_PAX_SEGMEXEC) { ++ struct vm_area_struct *mpnt_m; ++ ++ for (mpnt = oldmm->mmap, mpnt_m = mm->mmap; mpnt; mpnt = mpnt->vm_next, mpnt_m = mpnt_m->vm_next) { ++ BUG_ON(!mpnt_m || mpnt_m->vm_mirror || mpnt->vm_mm != oldmm || mpnt_m->vm_mm != mm); ++ ++ if (!mpnt->vm_mirror) ++ continue; ++ ++ if (mpnt->vm_end <= SEGMEXEC_TASK_SIZE) { ++ BUG_ON(mpnt->vm_mirror->vm_mirror != mpnt); ++ mpnt->vm_mirror = mpnt_m; ++ } else { ++ BUG_ON(mpnt->vm_mirror->vm_mirror == mpnt || mpnt->vm_mirror->vm_mirror->vm_mm != mm); ++ mpnt_m->vm_mirror = mpnt->vm_mirror->vm_mirror; ++ mpnt_m->vm_mirror->vm_mirror = mpnt_m; ++ mpnt->vm_mirror->vm_mirror = mpnt; ++ } ++ } ++ BUG_ON(mpnt_m); ++ } ++#endif ++ + /* a new mm has just been created */ + arch_dup_mmap(oldmm, mm); + retval = 0; +@@ -424,14 +469,6 @@ out: + flush_tlb_mm(oldmm); + up_write(&oldmm->mmap_sem); + return retval; +-fail_nomem_anon_vma_fork: +- mpol_put(pol); +-fail_nomem_policy: +- kmem_cache_free(vm_area_cachep, tmp); +-fail_nomem: +- retval = -ENOMEM; +- vm_unacct_memory(charge); +- goto out; + } + + static inline int mm_alloc_pgd(struct mm_struct * mm) +@@ -778,13 +815,14 @@ static int copy_fs(unsigned long clone_f + spin_unlock(&fs->lock); + return -EAGAIN; + } +- fs->users++; ++ atomic_inc(&fs->users); + spin_unlock(&fs->lock); + return 0; + } + tsk->fs = copy_fs_struct(fs); + if (!tsk->fs) + return -ENOMEM; ++ gr_set_chroot_entries(tsk, &tsk->fs->root); + return 0; + } + +@@ -1042,10 +1080,13 @@ static struct task_struct *copy_process( + DEBUG_LOCKS_WARN_ON(!p->softirqs_enabled); + #endif + retval = -EAGAIN; ++ ++ gr_learn_resource(p, RLIMIT_NPROC, atomic_read(&p->real_cred->user->processes), 0); ++ + if (atomic_read(&p->real_cred->user->processes) >= + task_rlimit(p, RLIMIT_NPROC)) { +- if (!capable(CAP_SYS_ADMIN) && !capable(CAP_SYS_RESOURCE) && +- p->real_cred->user != INIT_USER) ++ if (p->real_cred->user != INIT_USER && ++ !capable(CAP_SYS_ADMIN) && !capable(CAP_SYS_RESOURCE)) + goto bad_fork_free; + } + +@@ -1199,6 +1240,8 @@ static struct task_struct *copy_process( + goto bad_fork_free_pid; + } + ++ gr_copy_label(p); ++ + p->set_child_tid = (clone_flags & CLONE_CHILD_SETTID) ? child_tidptr : NULL; + /* + * Clear TID on mm_release()? +@@ -1356,6 +1399,8 @@ bad_fork_cleanup_count: + bad_fork_free: + free_task(p); + fork_out: ++ gr_log_forkfail(retval); ++ + return ERR_PTR(retval); + } + +@@ -1444,6 +1489,8 @@ long do_fork(unsigned long clone_flags, + if (clone_flags & CLONE_PARENT_SETTID) + put_user(nr, parent_tidptr); + ++ gr_handle_brute_check(); ++ + if (clone_flags & CLONE_VFORK) { + p->vfork_done = &vfork; + init_completion(&vfork); +@@ -1559,7 +1606,7 @@ static int unshare_fs(unsigned long unsh + return 0; + + /* don't need lock here; in the worst case we'll do useless copy */ +- if (fs->users == 1) ++ if (atomic_read(&fs->users) == 1) + return 0; + + *new_fsp = copy_fs_struct(fs); +@@ -1682,7 +1729,8 @@ SYSCALL_DEFINE1(unshare, unsigned long, + fs = current->fs; + spin_lock(&fs->lock); + current->fs = new_fs; +- if (--fs->users) ++ gr_set_chroot_entries(current, ¤t->fs->root); ++ if (atomic_dec_return(&fs->users)) + new_fs = NULL; + else + new_fs = fs; +diff -urNp linux-2.6.38.3/kernel/futex.c linux-2.6.38.3/kernel/futex.c +--- linux-2.6.38.3/kernel/futex.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/futex.c 2011-04-17 15:57:32.000000000 -0400 +@@ -54,6 +54,7 @@ + #include <linux/mount.h> + #include <linux/pagemap.h> + #include <linux/syscalls.h> ++#include <linux/ptrace.h> + #include <linux/signal.h> + #include <linux/module.h> + #include <linux/magic.h> +@@ -236,6 +237,11 @@ get_futex_key(u32 __user *uaddr, int fsh + struct page *page, *page_head; + int err; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ if ((mm->pax_flags & MF_PAX_SEGMEXEC) && address >= SEGMEXEC_TASK_SIZE) ++ return -EFAULT; ++#endif ++ + /* + * The futex address must be "naturally" aligned. + */ +@@ -2404,7 +2410,9 @@ SYSCALL_DEFINE3(get_robust_list, int, pi + { + struct robust_list_head __user *head; + unsigned long ret; ++#ifndef CONFIG_GRKERNSEC_PROC_MEMMAP + const struct cred *cred = current_cred(), *pcred; ++#endif + + if (!futex_cmpxchg_enabled) + return -ENOSYS; +@@ -2420,11 +2428,16 @@ SYSCALL_DEFINE3(get_robust_list, int, pi + if (!p) + goto err_unlock; + ret = -EPERM; ++#ifdef CONFIG_GRKERNSEC_PROC_MEMMAP ++ if (!ptrace_may_access(p, PTRACE_MODE_READ)) ++ goto err_unlock; ++#else + pcred = __task_cred(p); + if (cred->euid != pcred->euid && + cred->euid != pcred->uid && + !capable(CAP_SYS_PTRACE)) + goto err_unlock; ++#endif + head = p->robust_list; + rcu_read_unlock(); + } +@@ -2667,6 +2680,7 @@ static int __init futex_init(void) + { + u32 curval; + int i; ++ mm_segment_t oldfs; + + /* + * This will fail and we want it. Some arch implementations do +@@ -2678,7 +2692,10 @@ static int __init futex_init(void) + * implementation, the non-functional ones will return + * -ENOSYS. + */ ++ oldfs = get_fs(); ++ set_fs(USER_DS); + curval = cmpxchg_futex_value_locked(NULL, 0, 0); ++ set_fs(oldfs); + if (curval == -EFAULT) + futex_cmpxchg_enabled = 1; + +diff -urNp linux-2.6.38.3/kernel/futex_compat.c linux-2.6.38.3/kernel/futex_compat.c +--- linux-2.6.38.3/kernel/futex_compat.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/futex_compat.c 2011-04-17 15:57:32.000000000 -0400 +@@ -10,6 +10,7 @@ + #include <linux/compat.h> + #include <linux/nsproxy.h> + #include <linux/futex.h> ++#include <linux/ptrace.h> + + #include <asm/uaccess.h> + +@@ -136,7 +137,10 @@ compat_sys_get_robust_list(int pid, comp + { + struct compat_robust_list_head __user *head; + unsigned long ret; +- const struct cred *cred = current_cred(), *pcred; ++#ifndef CONFIG_GRKERNSEC_PROC_MEMMAP ++ const struct cred *cred = current_cred(); ++ const struct cred *pcred; ++#endif + + if (!futex_cmpxchg_enabled) + return -ENOSYS; +@@ -152,11 +156,16 @@ compat_sys_get_robust_list(int pid, comp + if (!p) + goto err_unlock; + ret = -EPERM; ++#ifdef CONFIG_GRKERNSEC_PROC_MEMMAP ++ if (!ptrace_may_access(p, PTRACE_MODE_READ)) ++ goto err_unlock; ++#else + pcred = __task_cred(p); + if (cred->euid != pcred->euid && + cred->euid != pcred->uid && + !capable(CAP_SYS_PTRACE)) + goto err_unlock; ++#endif + head = p->compat_robust_list; + rcu_read_unlock(); + } +diff -urNp linux-2.6.38.3/kernel/gcov/base.c linux-2.6.38.3/kernel/gcov/base.c +--- linux-2.6.38.3/kernel/gcov/base.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/gcov/base.c 2011-04-17 15:57:32.000000000 -0400 +@@ -102,11 +102,6 @@ void gcov_enable_events(void) + } + + #ifdef CONFIG_MODULES +-static inline int within(void *addr, void *start, unsigned long size) +-{ +- return ((addr >= start) && (addr < start + size)); +-} +- + /* Update list and generate events when modules are unloaded. */ + static int gcov_module_notifier(struct notifier_block *nb, unsigned long event, + void *data) +@@ -121,7 +116,7 @@ static int gcov_module_notifier(struct n + prev = NULL; + /* Remove entries located in module from linked list. */ + for (info = gcov_info_head; info; info = info->next) { +- if (within(info, mod->module_core, mod->core_size)) { ++ if (within_module_core_rw((unsigned long)info, mod)) { + if (prev) + prev->next = info->next; + else +diff -urNp linux-2.6.38.3/kernel/hrtimer.c linux-2.6.38.3/kernel/hrtimer.c +--- linux-2.6.38.3/kernel/hrtimer.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/hrtimer.c 2011-04-17 15:57:32.000000000 -0400 +@@ -1371,7 +1371,7 @@ void hrtimer_peek_ahead_timers(void) + local_irq_restore(flags); + } + +-static void run_hrtimer_softirq(struct softirq_action *h) ++static void run_hrtimer_softirq(void) + { + hrtimer_peek_ahead_timers(); + } +diff -urNp linux-2.6.38.3/kernel/jump_label.c linux-2.6.38.3/kernel/jump_label.c +--- linux-2.6.38.3/kernel/jump_label.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/jump_label.c 2011-04-17 15:57:32.000000000 -0400 +@@ -49,6 +49,17 @@ void jump_label_unlock(void) + mutex_unlock(&jump_label_mutex); + } + ++static void jump_label_swap(void *a, void *b, int size) ++{ ++ struct jump_entry t; ++ ++ t = *(struct jump_entry *)a; ++ pax_open_kernel(); ++ *(struct jump_entry *)a = *(struct jump_entry *)b; ++ *(struct jump_entry *)b = t; ++ pax_close_kernel(); ++} ++ + static int jump_label_cmp(const void *a, const void *b) + { + const struct jump_entry *jea = a; +@@ -70,7 +81,7 @@ sort_jump_label_entries(struct jump_entr + + size = (((unsigned long)stop - (unsigned long)start) + / sizeof(struct jump_entry)); +- sort(start, size, sizeof(struct jump_entry), jump_label_cmp, NULL); ++ sort(start, size, sizeof(struct jump_entry), jump_label_cmp, jump_label_swap); + } + + static struct jump_label_entry *get_jump_label_entry(jump_label_t key) +@@ -407,8 +418,11 @@ static void remove_jump_label_module_ini + count = e_module->nr_entries; + iter = e_module->table; + while (count--) { +- if (within_module_init(iter->code, mod)) ++ if (within_module_init(iter->code, mod)) { ++ pax_open_kernel(); + iter->key = 0; ++ pax_close_kernel(); ++ } + iter++; + } + } +diff -urNp linux-2.6.38.3/kernel/kallsyms.c linux-2.6.38.3/kernel/kallsyms.c +--- linux-2.6.38.3/kernel/kallsyms.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/kallsyms.c 2011-04-17 15:57:32.000000000 -0400 +@@ -11,6 +11,9 @@ + * Changed the compression method from stem compression to "table lookup" + * compression (see scripts/kallsyms.c for a more complete description) + */ ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++#define __INCLUDED_BY_HIDESYM 1 ++#endif + #include <linux/kallsyms.h> + #include <linux/module.h> + #include <linux/init.h> +@@ -53,12 +56,33 @@ extern const unsigned long kallsyms_mark + + static inline int is_kernel_inittext(unsigned long addr) + { ++ if (system_state != SYSTEM_BOOTING) ++ return 0; ++ + if (addr >= (unsigned long)_sinittext + && addr <= (unsigned long)_einittext) + return 1; + return 0; + } + ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) ++#ifdef CONFIG_MODULES ++static inline int is_module_text(unsigned long addr) ++{ ++ if ((unsigned long)MODULES_EXEC_VADDR <= addr && addr <= (unsigned long)MODULES_EXEC_END) ++ return 1; ++ ++ addr = ktla_ktva(addr); ++ return (unsigned long)MODULES_EXEC_VADDR <= addr && addr <= (unsigned long)MODULES_EXEC_END; ++} ++#else ++static inline int is_module_text(unsigned long addr) ++{ ++ return 0; ++} ++#endif ++#endif ++ + static inline int is_kernel_text(unsigned long addr) + { + if ((addr >= (unsigned long)_stext && addr <= (unsigned long)_etext) || +@@ -69,13 +93,28 @@ static inline int is_kernel_text(unsigne + + static inline int is_kernel(unsigned long addr) + { ++ ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) ++ if (is_kernel_text(addr) || is_kernel_inittext(addr)) ++ return 1; ++ ++ if (ktla_ktva((unsigned long)_text) <= addr && addr < (unsigned long)_end) ++#else + if (addr >= (unsigned long)_stext && addr <= (unsigned long)_end) ++#endif ++ + return 1; + return in_gate_area_no_task(addr); + } + + static int is_ksym_addr(unsigned long addr) + { ++ ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) ++ if (is_module_text(addr)) ++ return 0; ++#endif ++ + if (all_var) + return is_kernel(addr); + +@@ -416,7 +455,6 @@ static unsigned long get_ksymbol_core(st + + static void reset_iter(struct kallsym_iter *iter, loff_t new_pos) + { +- iter->name[0] = '\0'; + iter->nameoff = get_symbol_offset(new_pos); + iter->pos = new_pos; + } +@@ -464,6 +502,11 @@ static int s_show(struct seq_file *m, vo + { + struct kallsym_iter *iter = m->private; + ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ if (current_uid()) ++ return 0; ++#endif ++ + /* Some debugging symbols have no name. Ignore them. */ + if (!iter->name[0]) + return 0; +@@ -504,7 +547,7 @@ static int kallsyms_open(struct inode *i + struct kallsym_iter *iter; + int ret; + +- iter = kmalloc(sizeof(*iter), GFP_KERNEL); ++ iter = kzalloc(sizeof(*iter), GFP_KERNEL); + if (!iter) + return -ENOMEM; + reset_iter(iter, 0); +diff -urNp linux-2.6.38.3/kernel/kmod.c linux-2.6.38.3/kernel/kmod.c +--- linux-2.6.38.3/kernel/kmod.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/kmod.c 2011-04-17 15:57:32.000000000 -0400 +@@ -65,13 +65,12 @@ char modprobe_path[KMOD_PATH_LEN] = "/sb + * If module auto-loading support is disabled then this function + * becomes a no-operation. + */ +-int __request_module(bool wait, const char *fmt, ...) ++static int ____request_module(bool wait, char *module_param, const char *fmt, va_list ap) + { +- va_list args; + char module_name[MODULE_NAME_LEN]; + unsigned int max_modprobes; + int ret; +- char *argv[] = { modprobe_path, "-q", "--", module_name, NULL }; ++ char *argv[] = { modprobe_path, "-q", "--", module_name, module_param, NULL }; + static char *envp[] = { "HOME=/", + "TERM=linux", + "PATH=/sbin:/usr/sbin:/bin:/usr/bin", +@@ -80,9 +79,7 @@ int __request_module(bool wait, const ch + #define MAX_KMOD_CONCURRENT 50 /* Completely arbitrary value - KAO */ + static int kmod_loop_msg; + +- va_start(args, fmt); +- ret = vsnprintf(module_name, MODULE_NAME_LEN, fmt, args); +- va_end(args); ++ ret = vsnprintf(module_name, MODULE_NAME_LEN, fmt, ap); + if (ret >= MODULE_NAME_LEN) + return -ENAMETOOLONG; + +@@ -90,6 +87,20 @@ int __request_module(bool wait, const ch + if (ret) + return ret; + ++#ifdef CONFIG_GRKERNSEC_MODHARDEN ++ if (!current_uid()) { ++ /* hack to workaround consolekit/udisks stupidity */ ++ read_lock(&tasklist_lock); ++ if (!strcmp(current->comm, "mount") && ++ current->real_parent && !strncmp(current->real_parent->comm, "udisk", 5)) { ++ read_unlock(&tasklist_lock); ++ printk(KERN_ALERT "grsec: denied attempt to auto-load fs module %.64s by udisks\n", module_name); ++ return -EPERM; ++ } ++ read_unlock(&tasklist_lock); ++ } ++#endif ++ + /* If modprobe needs a service that is in a module, we get a recursive + * loop. Limit the number of running kmod threads to max_threads/2 or + * MAX_KMOD_CONCURRENT, whichever is the smaller. A cleaner method +@@ -123,6 +134,47 @@ int __request_module(bool wait, const ch + atomic_dec(&kmod_concurrent); + return ret; + } ++ ++int ___request_module(bool wait, char *module_param, const char *fmt, ...) ++{ ++ va_list args; ++ int ret; ++ ++ va_start(args, fmt); ++ ret = ____request_module(wait, module_param, fmt, args); ++ va_end(args); ++ ++ return ret; ++} ++ ++int __request_module(bool wait, const char *fmt, ...) ++{ ++ va_list args; ++ int ret; ++ ++#ifdef CONFIG_GRKERNSEC_MODHARDEN ++ if (current_uid()) { ++ char module_param[MODULE_NAME_LEN]; ++ ++ memset(module_param, 0, sizeof(module_param)); ++ ++ snprintf(module_param, sizeof(module_param) - 1, "grsec_modharden_normal%u_", current_uid()); ++ ++ va_start(args, fmt); ++ ret = ____request_module(wait, module_param, fmt, args); ++ va_end(args); ++ ++ return ret; ++ } ++#endif ++ ++ va_start(args, fmt); ++ ret = ____request_module(wait, NULL, fmt, args); ++ va_end(args); ++ ++ return ret; ++} ++ + EXPORT_SYMBOL(__request_module); + #endif /* CONFIG_MODULES */ + +diff -urNp linux-2.6.38.3/kernel/kprobes.c linux-2.6.38.3/kernel/kprobes.c +--- linux-2.6.38.3/kernel/kprobes.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/kprobes.c 2011-04-17 15:57:32.000000000 -0400 +@@ -185,7 +185,7 @@ static kprobe_opcode_t __kprobes *__get_ + * kernel image and loaded module images reside. This is required + * so x86_64 can correctly handle the %rip-relative fixups. + */ +- kip->insns = module_alloc(PAGE_SIZE); ++ kip->insns = module_alloc_exec(PAGE_SIZE); + if (!kip->insns) { + kfree(kip); + return NULL; +@@ -225,7 +225,7 @@ static int __kprobes collect_one_slot(st + */ + if (!list_is_singular(&kip->list)) { + list_del(&kip->list); +- module_free(NULL, kip->insns); ++ module_free_exec(NULL, kip->insns); + kfree(kip); + } + return 1; +@@ -1936,7 +1936,7 @@ static int __init init_kprobes(void) + { + int i, err = 0; + unsigned long offset = 0, size = 0; +- char *modname, namebuf[128]; ++ char *modname, namebuf[KSYM_NAME_LEN]; + const char *symbol_name; + void *addr; + struct kprobe_blackpoint *kb; +@@ -2062,7 +2062,7 @@ static int __kprobes show_kprobe_addr(st + const char *sym = NULL; + unsigned int i = *(loff_t *) v; + unsigned long offset = 0; +- char *modname, namebuf[128]; ++ char *modname, namebuf[KSYM_NAME_LEN]; + + head = &kprobe_table[i]; + preempt_disable(); +diff -urNp linux-2.6.38.3/kernel/lockdep.c linux-2.6.38.3/kernel/lockdep.c +--- linux-2.6.38.3/kernel/lockdep.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/lockdep.c 2011-04-17 15:57:32.000000000 -0400 +@@ -571,6 +571,10 @@ static int static_obj(void *obj) + end = (unsigned long) &_end, + addr = (unsigned long) obj; + ++#ifdef CONFIG_PAX_KERNEXEC ++ start = ktla_ktva(start); ++#endif ++ + /* + * static variable? + */ +@@ -706,6 +710,7 @@ register_lock_class(struct lockdep_map * + if (!static_obj(lock->key)) { + debug_locks_off(); + printk("INFO: trying to register non-static key.\n"); ++ printk("lock:%pS key:%pS.\n", lock, lock->key); + printk("the code is fine but needs lockdep annotation.\n"); + printk("turning off the locking correctness validator.\n"); + dump_stack(); +@@ -2752,7 +2757,7 @@ static int __lock_acquire(struct lockdep + if (!class) + return 0; + } +- atomic_inc((atomic_t *)&class->ops); ++ atomic_inc_unchecked((atomic_unchecked_t *)&class->ops); + if (very_verbose(class)) { + printk("\nacquire class [%p] %s", class->key, class->name); + if (class->name_version > 1) +diff -urNp linux-2.6.38.3/kernel/lockdep_proc.c linux-2.6.38.3/kernel/lockdep_proc.c +--- linux-2.6.38.3/kernel/lockdep_proc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/lockdep_proc.c 2011-04-17 15:57:32.000000000 -0400 +@@ -39,7 +39,7 @@ static void l_stop(struct seq_file *m, v + + static void print_name(struct seq_file *m, struct lock_class *class) + { +- char str[128]; ++ char str[KSYM_NAME_LEN]; + const char *name = class->name; + + if (!name) { +diff -urNp linux-2.6.38.3/kernel/module.c linux-2.6.38.3/kernel/module.c +--- linux-2.6.38.3/kernel/module.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/module.c 2011-04-17 16:05:04.000000000 -0400 +@@ -57,6 +57,7 @@ + #include <linux/kmemleak.h> + #include <linux/jump_label.h> + #include <linux/pfn.h> ++#include <linux/grsecurity.h> + + #define CREATE_TRACE_POINTS + #include <trace/events/module.h> +@@ -118,7 +119,8 @@ static BLOCKING_NOTIFIER_HEAD(module_not + + /* Bounds of module allocation, for speeding __module_address. + * Protected by module_mutex. */ +-static unsigned long module_addr_min = -1UL, module_addr_max = 0; ++static unsigned long module_addr_min_rw = -1UL, module_addr_max_rw = 0; ++static unsigned long module_addr_min_rx = -1UL, module_addr_max_rx = 0; + + int register_module_notifier(struct notifier_block * nb) + { +@@ -282,7 +284,7 @@ bool each_symbol(bool (*fn)(const struct + return true; + + list_for_each_entry_rcu(mod, &modules, list) { +- struct symsearch arr[] = { ++ struct symsearch modarr[] = { + { mod->syms, mod->syms + mod->num_syms, mod->crcs, + NOT_GPL_ONLY, false }, + { mod->gpl_syms, mod->gpl_syms + mod->num_gpl_syms, +@@ -304,7 +306,7 @@ bool each_symbol(bool (*fn)(const struct + #endif + }; + +- if (each_symbol_in_section(arr, ARRAY_SIZE(arr), mod, fn, data)) ++ if (each_symbol_in_section(modarr, ARRAY_SIZE(modarr), mod, fn, data)) + return true; + } + return false; +@@ -415,7 +417,7 @@ static inline void __percpu *mod_percpu( + static int percpu_modalloc(struct module *mod, + unsigned long size, unsigned long align) + { +- if (align > PAGE_SIZE) { ++ if (align-1 >= PAGE_SIZE) { + printk(KERN_WARNING "%s: per-cpu alignment %li > %li\n", + mod->name, align, PAGE_SIZE); + align = PAGE_SIZE; +@@ -1143,7 +1145,7 @@ resolve_symbol_wait(struct module *mod, + */ + #ifdef CONFIG_SYSFS + +-#ifdef CONFIG_KALLSYMS ++#if defined(CONFIG_KALLSYMS) && !defined(CONFIG_GRKERNSEC_HIDESYM) + static inline bool sect_empty(const Elf_Shdr *sect) + { + return !(sect->sh_flags & SHF_ALLOC) || sect->sh_size == 0; +@@ -1612,17 +1614,17 @@ void unset_section_ro_nx(struct module * + { + unsigned long total_pages; + +- if (mod->module_core == module_region) { ++ if (mod->module_core_rx == module_region) { + /* Set core as NX+RW */ +- total_pages = MOD_NUMBER_OF_PAGES(mod->module_core, mod->core_size); +- set_memory_nx((unsigned long)mod->module_core, total_pages); +- set_memory_rw((unsigned long)mod->module_core, total_pages); ++ total_pages = MOD_NUMBER_OF_PAGES(mod->module_core_rx, mod->core_size_rx); ++ set_memory_nx((unsigned long)mod->module_core_rx, total_pages); ++ set_memory_rw((unsigned long)mod->module_core_rx, total_pages); + +- } else if (mod->module_init == module_region) { ++ } else if (mod->module_init_rx == module_region) { + /* Set init as NX+RW */ +- total_pages = MOD_NUMBER_OF_PAGES(mod->module_init, mod->init_size); +- set_memory_nx((unsigned long)mod->module_init, total_pages); +- set_memory_rw((unsigned long)mod->module_init, total_pages); ++ total_pages = MOD_NUMBER_OF_PAGES(mod->module_init_rx, mod->init_size_rx); ++ set_memory_nx((unsigned long)mod->module_init_rx, total_pages); ++ set_memory_rw((unsigned long)mod->module_init_rx, total_pages); + } + } + +@@ -1633,14 +1635,14 @@ void set_all_modules_text_rw() + + mutex_lock(&module_mutex); + list_for_each_entry_rcu(mod, &modules, list) { +- if ((mod->module_core) && (mod->core_text_size)) { +- set_page_attributes(mod->module_core, +- mod->module_core + mod->core_text_size, ++ if ((mod->module_core_rx) && (mod->core_size_rx)) { ++ set_page_attributes(mod->module_core_rx, ++ mod->module_core_rx + mod->core_size_rx, + set_memory_rw); + } +- if ((mod->module_init) && (mod->init_text_size)) { +- set_page_attributes(mod->module_init, +- mod->module_init + mod->init_text_size, ++ if ((mod->module_init_rx) && (mod->init_size_rx)) { ++ set_page_attributes(mod->module_init_rx, ++ mod->module_init_rx + mod->init_size_rx, + set_memory_rw); + } + } +@@ -1654,14 +1656,14 @@ void set_all_modules_text_ro() + + mutex_lock(&module_mutex); + list_for_each_entry_rcu(mod, &modules, list) { +- if ((mod->module_core) && (mod->core_text_size)) { +- set_page_attributes(mod->module_core, +- mod->module_core + mod->core_text_size, ++ if ((mod->module_core_rx) && (mod->core_size_rx)) { ++ set_page_attributes(mod->module_core_rx, ++ mod->module_core_rx + mod->core_size_rx, + set_memory_ro); + } +- if ((mod->module_init) && (mod->init_text_size)) { +- set_page_attributes(mod->module_init, +- mod->module_init + mod->init_text_size, ++ if ((mod->module_init_rx) && (mod->init_size_rx)) { ++ set_page_attributes(mod->module_init_rx, ++ mod->module_init_rx + mod->init_size_rx, + set_memory_ro); + } + } +@@ -1696,17 +1698,20 @@ static void free_module(struct module *m + destroy_params(mod->kp, mod->num_kp); + + /* This may be NULL, but that's OK */ +- unset_section_ro_nx(mod, mod->module_init); +- module_free(mod, mod->module_init); ++ unset_section_ro_nx(mod, mod->module_init_rx); ++ module_free(mod, mod->module_init_rw); ++ module_free_exec(mod, mod->module_init_rx); + kfree(mod->args); + percpu_modfree(mod); + + /* Free lock-classes: */ +- lockdep_free_key_range(mod->module_core, mod->core_size); ++ lockdep_free_key_range(mod->module_core_rx, mod->core_size_rx); ++ lockdep_free_key_range(mod->module_core_rw, mod->core_size_rw); + + /* Finally, free the core (containing the module structure) */ +- unset_section_ro_nx(mod, mod->module_core); +- module_free(mod, mod->module_core); ++ unset_section_ro_nx(mod, mod->module_core_rx); ++ module_free_exec(mod, mod->module_core_rx); ++ module_free(mod, mod->module_core_rw); + + #ifdef CONFIG_MPU + update_protections(current->mm); +@@ -1775,10 +1780,25 @@ static int simplify_symbols(struct modul + unsigned int i; + int ret = 0; + const struct kernel_symbol *ksym; ++#ifdef CONFIG_GRKERNSEC_MODHARDEN ++ int is_fs_load = 0; ++ int register_filesystem_found = 0; ++ ++ if (strstr(mod->args, "grsec_modharden_fs")) ++ is_fs_load = 1; ++#endif + + for (i = 1; i < symsec->sh_size / sizeof(Elf_Sym); i++) { + const char *name = info->strtab + sym[i].st_name; + ++#ifdef CONFIG_GRKERNSEC_MODHARDEN ++ /* it's a real shame this will never get ripped and copied ++ upstream! ;( ++ */ ++ if (is_fs_load && !strcmp(name, "register_filesystem")) ++ register_filesystem_found = 1; ++#endif ++ + switch (sym[i].st_shndx) { + case SHN_COMMON: + /* We compiled with -fno-common. These are not +@@ -1799,7 +1819,9 @@ static int simplify_symbols(struct modul + ksym = resolve_symbol_wait(mod, info, name); + /* Ok if resolved. */ + if (ksym && !IS_ERR(ksym)) { ++ pax_open_kernel(); + sym[i].st_value = ksym->value; ++ pax_close_kernel(); + break; + } + +@@ -1818,11 +1840,20 @@ static int simplify_symbols(struct modul + secbase = (unsigned long)mod_percpu(mod); + else + secbase = info->sechdrs[sym[i].st_shndx].sh_addr; ++ pax_open_kernel(); + sym[i].st_value += secbase; ++ pax_close_kernel(); + break; + } + } + ++#ifdef CONFIG_GRKERNSEC_MODHARDEN ++ if (is_fs_load && !register_filesystem_found) { ++ printk(KERN_ALERT "grsec: Denied attempt to load non-fs module %.64s through mount\n", mod->name); ++ ret = -EPERM; ++ } ++#endif ++ + return ret; + } + +@@ -1906,22 +1937,12 @@ static void layout_sections(struct modul + || s->sh_entsize != ~0UL + || strstarts(sname, ".init")) + continue; +- s->sh_entsize = get_offset(mod, &mod->core_size, s, i); ++ if ((s->sh_flags & SHF_WRITE) || !(s->sh_flags & SHF_ALLOC)) ++ s->sh_entsize = get_offset(mod, &mod->core_size_rw, s, i); ++ else ++ s->sh_entsize = get_offset(mod, &mod->core_size_rx, s, i); + DEBUGP("\t%s\n", name); + } +- switch (m) { +- case 0: /* executable */ +- mod->core_size = debug_align(mod->core_size); +- mod->core_text_size = mod->core_size; +- break; +- case 1: /* RO: text and ro-data */ +- mod->core_size = debug_align(mod->core_size); +- mod->core_ro_size = mod->core_size; +- break; +- case 3: /* whole core */ +- mod->core_size = debug_align(mod->core_size); +- break; +- } + } + + DEBUGP("Init section allocation order:\n"); +@@ -1935,23 +1956,13 @@ static void layout_sections(struct modul + || s->sh_entsize != ~0UL + || !strstarts(sname, ".init")) + continue; +- s->sh_entsize = (get_offset(mod, &mod->init_size, s, i) +- | INIT_OFFSET_MASK); ++ if ((s->sh_flags & SHF_WRITE) || !(s->sh_flags & SHF_ALLOC)) ++ s->sh_entsize = get_offset(mod, &mod->init_size_rw, s, i); ++ else ++ s->sh_entsize = get_offset(mod, &mod->init_size_rx, s, i); ++ s->sh_entsize |= INIT_OFFSET_MASK; + DEBUGP("\t%s\n", sname); + } +- switch (m) { +- case 0: /* executable */ +- mod->init_size = debug_align(mod->init_size); +- mod->init_text_size = mod->init_size; +- break; +- case 1: /* RO: text and ro-data */ +- mod->init_size = debug_align(mod->init_size); +- mod->init_ro_size = mod->init_size; +- break; +- case 3: /* whole init */ +- mod->init_size = debug_align(mod->init_size); +- break; +- } + } + } + +@@ -2119,7 +2130,7 @@ static void layout_symtab(struct module + + /* Put symbol section at end of init part of module. */ + symsect->sh_flags |= SHF_ALLOC; +- symsect->sh_entsize = get_offset(mod, &mod->init_size, symsect, ++ symsect->sh_entsize = get_offset(mod, &mod->init_size_rx, symsect, + info->index.sym) | INIT_OFFSET_MASK; + DEBUGP("\t%s\n", info->secstrings + symsect->sh_name); + +@@ -2136,19 +2147,19 @@ static void layout_symtab(struct module + } + + /* Append room for core symbols at end of core part. */ +- info->symoffs = ALIGN(mod->core_size, symsect->sh_addralign ?: 1); +- mod->core_size = info->symoffs + ndst * sizeof(Elf_Sym); ++ info->symoffs = ALIGN(mod->core_size_rx, symsect->sh_addralign ?: 1); ++ mod->core_size_rx = info->symoffs + ndst * sizeof(Elf_Sym); + + /* Put string table section at end of init part of module. */ + strsect->sh_flags |= SHF_ALLOC; +- strsect->sh_entsize = get_offset(mod, &mod->init_size, strsect, ++ strsect->sh_entsize = get_offset(mod, &mod->init_size_rx, strsect, + info->index.str) | INIT_OFFSET_MASK; + DEBUGP("\t%s\n", info->secstrings + strsect->sh_name); + + /* Append room for core symbols' strings at end of core part. */ +- info->stroffs = mod->core_size; ++ info->stroffs = mod->core_size_rx; + __set_bit(0, info->strmap); +- mod->core_size += bitmap_weight(info->strmap, strsect->sh_size); ++ mod->core_size_rx += bitmap_weight(info->strmap, strsect->sh_size); + } + + static void add_kallsyms(struct module *mod, const struct load_info *info) +@@ -2164,11 +2175,13 @@ static void add_kallsyms(struct module * + /* Make sure we get permanent strtab: don't use info->strtab. */ + mod->strtab = (void *)info->sechdrs[info->index.str].sh_addr; + ++ pax_open_kernel(); ++ + /* Set types up while we still have access to sections. */ + for (i = 0; i < mod->num_symtab; i++) + mod->symtab[i].st_info = elf_type(&mod->symtab[i], info); + +- mod->core_symtab = dst = mod->module_core + info->symoffs; ++ mod->core_symtab = dst = mod->module_core_rx + info->symoffs; + src = mod->symtab; + *dst = *src; + for (ndst = i = 1; i < mod->num_symtab; ++i, ++src) { +@@ -2181,10 +2194,12 @@ static void add_kallsyms(struct module * + } + mod->core_num_syms = ndst; + +- mod->core_strtab = s = mod->module_core + info->stroffs; ++ mod->core_strtab = s = mod->module_core_rx + info->stroffs; + for (*s = 0, i = 1; i < info->sechdrs[info->index.str].sh_size; ++i) + if (test_bit(i, info->strmap)) + *++s = mod->strtab[i]; ++ ++ pax_close_kernel(); + } + #else + static inline void layout_symtab(struct module *mod, struct load_info *info) +@@ -2213,17 +2228,33 @@ static void dynamic_debug_remove(struct + ddebug_remove_module(debug->modname); + } + +-static void *module_alloc_update_bounds(unsigned long size) ++static void *module_alloc_update_bounds_rw(unsigned long size) + { + void *ret = module_alloc(size); + + if (ret) { + mutex_lock(&module_mutex); + /* Update module bounds. */ +- if ((unsigned long)ret < module_addr_min) +- module_addr_min = (unsigned long)ret; +- if ((unsigned long)ret + size > module_addr_max) +- module_addr_max = (unsigned long)ret + size; ++ if ((unsigned long)ret < module_addr_min_rw) ++ module_addr_min_rw = (unsigned long)ret; ++ if ((unsigned long)ret + size > module_addr_max_rw) ++ module_addr_max_rw = (unsigned long)ret + size; ++ mutex_unlock(&module_mutex); ++ } ++ return ret; ++} ++ ++static void *module_alloc_update_bounds_rx(unsigned long size) ++{ ++ void *ret = module_alloc_exec(size); ++ ++ if (ret) { ++ mutex_lock(&module_mutex); ++ /* Update module bounds. */ ++ if ((unsigned long)ret < module_addr_min_rx) ++ module_addr_min_rx = (unsigned long)ret; ++ if ((unsigned long)ret + size > module_addr_max_rx) ++ module_addr_max_rx = (unsigned long)ret + size; + mutex_unlock(&module_mutex); + } + return ret; +@@ -2516,7 +2547,7 @@ static int move_module(struct module *mo + void *ptr; + + /* Do the allocs. */ +- ptr = module_alloc_update_bounds(mod->core_size); ++ ptr = module_alloc_update_bounds_rw(mod->core_size_rw); + /* + * The pointer to this block is stored in the module structure + * which is inside the block. Just mark it as not being a +@@ -2526,23 +2557,50 @@ static int move_module(struct module *mo + if (!ptr) + return -ENOMEM; + +- memset(ptr, 0, mod->core_size); +- mod->module_core = ptr; ++ memset(ptr, 0, mod->core_size_rw); ++ mod->module_core_rw = ptr; + +- ptr = module_alloc_update_bounds(mod->init_size); ++ ptr = module_alloc_update_bounds_rw(mod->init_size_rw); + /* + * The pointer to this block is stored in the module structure + * which is inside the block. This block doesn't need to be + * scanned as it contains data and code that will be freed + * after the module is initialized. + */ +- kmemleak_ignore(ptr); +- if (!ptr && mod->init_size) { +- module_free(mod, mod->module_core); ++ kmemleak_not_leak(ptr); ++ if (!ptr && mod->init_size_rw) { ++ module_free(mod, mod->module_core_rw); + return -ENOMEM; + } +- memset(ptr, 0, mod->init_size); +- mod->module_init = ptr; ++ memset(ptr, 0, mod->init_size_rw); ++ mod->module_init_rw = ptr; ++ ++ ptr = module_alloc_update_bounds_rx(mod->core_size_rx); ++ kmemleak_not_leak(ptr); ++ if (!ptr) { ++ module_free(mod, mod->module_init_rw); ++ module_free(mod, mod->module_core_rw); ++ return -ENOMEM; ++ } ++ ++ pax_open_kernel(); ++ memset(ptr, 0, mod->core_size_rx); ++ pax_close_kernel(); ++ mod->module_core_rx = ptr; ++ ++ ptr = module_alloc_update_bounds_rx(mod->init_size_rx); ++ kmemleak_not_leak(ptr); ++ if (!ptr && mod->init_size_rx) { ++ module_free_exec(mod, mod->module_core_rx); ++ module_free(mod, mod->module_init_rw); ++ module_free(mod, mod->module_core_rw); ++ return -ENOMEM; ++ } ++ ++ pax_open_kernel(); ++ memset(ptr, 0, mod->init_size_rx); ++ pax_close_kernel(); ++ mod->module_init_rx = ptr; + + /* Transfer each section which specifies SHF_ALLOC */ + DEBUGP("final section addresses:\n"); +@@ -2553,16 +2611,45 @@ static int move_module(struct module *mo + if (!(shdr->sh_flags & SHF_ALLOC)) + continue; + +- if (shdr->sh_entsize & INIT_OFFSET_MASK) +- dest = mod->module_init +- + (shdr->sh_entsize & ~INIT_OFFSET_MASK); +- else +- dest = mod->module_core + shdr->sh_entsize; ++ if (shdr->sh_entsize & INIT_OFFSET_MASK) { ++ if ((shdr->sh_flags & SHF_WRITE) || !(shdr->sh_flags & SHF_ALLOC)) ++ dest = mod->module_init_rw ++ + (shdr->sh_entsize & ~INIT_OFFSET_MASK); ++ else ++ dest = mod->module_init_rx ++ + (shdr->sh_entsize & ~INIT_OFFSET_MASK); ++ } else { ++ if ((shdr->sh_flags & SHF_WRITE) || !(shdr->sh_flags & SHF_ALLOC)) ++ dest = mod->module_core_rw + shdr->sh_entsize; ++ else ++ dest = mod->module_core_rx + shdr->sh_entsize; ++ } ++ ++ if (shdr->sh_type != SHT_NOBITS) { ++ ++#ifdef CONFIG_PAX_KERNEXEC ++#ifdef CONFIG_X86_64 ++ if ((shdr->sh_flags & SHF_WRITE) && (shdr->sh_flags & SHF_EXECINSTR)) ++ set_memory_x((unsigned long)dest, (shdr->sh_size + PAGE_SIZE) >> PAGE_SHIFT); ++#endif ++ if (!(shdr->sh_flags & SHF_WRITE) && (shdr->sh_flags & SHF_ALLOC)) { ++ pax_open_kernel(); ++ memcpy(dest, (void *)shdr->sh_addr, shdr->sh_size); ++ pax_close_kernel(); ++ } else ++#endif + +- if (shdr->sh_type != SHT_NOBITS) + memcpy(dest, (void *)shdr->sh_addr, shdr->sh_size); ++ } + /* Update sh_addr to point to copy in image. */ +- shdr->sh_addr = (unsigned long)dest; ++ ++#ifdef CONFIG_PAX_KERNEXEC ++ if (shdr->sh_flags & SHF_EXECINSTR) ++ shdr->sh_addr = ktva_ktla((unsigned long)dest); ++ else ++#endif ++ ++ shdr->sh_addr = (unsigned long)dest; + DEBUGP("\t0x%lx %s\n", + shdr->sh_addr, info->secstrings + shdr->sh_name); + } +@@ -2613,12 +2700,12 @@ static void flush_module_icache(const st + * Do it before processing of module parameters, so the module + * can provide parameter accessor functions of its own. + */ +- if (mod->module_init) +- flush_icache_range((unsigned long)mod->module_init, +- (unsigned long)mod->module_init +- + mod->init_size); +- flush_icache_range((unsigned long)mod->module_core, +- (unsigned long)mod->module_core + mod->core_size); ++ if (mod->module_init_rx) ++ flush_icache_range((unsigned long)mod->module_init_rx, ++ (unsigned long)mod->module_init_rx ++ + mod->init_size_rx); ++ flush_icache_range((unsigned long)mod->module_core_rx, ++ (unsigned long)mod->module_core_rx + mod->core_size_rx); + + set_fs(old_fs); + } +@@ -2690,8 +2777,10 @@ static void module_deallocate(struct mod + { + kfree(info->strmap); + percpu_modfree(mod); +- module_free(mod, mod->module_init); +- module_free(mod, mod->module_core); ++ module_free_exec(mod, mod->module_init_rx); ++ module_free_exec(mod, mod->module_core_rx); ++ module_free(mod, mod->module_init_rw); ++ module_free(mod, mod->module_core_rw); + } + + static int post_relocation(struct module *mod, const struct load_info *info) +@@ -2748,9 +2837,38 @@ static struct module *load_module(void _ + if (err) + goto free_unload; + ++ /* Now copy in args */ ++ mod->args = strndup_user(uargs, ~0UL >> 1); ++ if (IS_ERR(mod->args)) { ++ err = PTR_ERR(mod->args); ++ goto free_unload; ++ } ++ + /* Set up MODINFO_ATTR fields */ + setup_modinfo(mod, &info); + ++#ifdef CONFIG_GRKERNSEC_MODHARDEN ++ { ++ char *p, *p2; ++ ++ if (strstr(mod->args, "grsec_modharden_netdev")) { ++ printk(KERN_ALERT "grsec: denied auto-loading kernel module for a network device with CAP_SYS_MODULE (deprecated). Use CAP_NET_ADMIN and alias netdev-%.64s instead.", mod->name); ++ err = -EPERM; ++ goto free_modinfo; ++ } else if ((p = strstr(mod->args, "grsec_modharden_normal"))) { ++ p += strlen("grsec_modharden_normal"); ++ p2 = strstr(p, "_"); ++ if (p2) { ++ *p2 = '\0'; ++ printk(KERN_ALERT "grsec: denied kernel module auto-load of %.64s by uid %.9s\n", mod->name, p); ++ *p2 = '_'; ++ } ++ err = -EPERM; ++ goto free_modinfo; ++ } ++ } ++#endif ++ + /* Fix up syms, so that st_value is a pointer to location. */ + err = simplify_symbols(mod, &info); + if (err < 0) +@@ -2766,13 +2884,6 @@ static struct module *load_module(void _ + + flush_module_icache(mod); + +- /* Now copy in args */ +- mod->args = strndup_user(uargs, ~0UL >> 1); +- if (IS_ERR(mod->args)) { +- err = PTR_ERR(mod->args); +- goto free_arch_cleanup; +- } +- + /* Mark state as coming so strong_try_module_get() ignores us. */ + mod->state = MODULE_STATE_COMING; + +@@ -2832,11 +2943,10 @@ static struct module *load_module(void _ + unlock: + mutex_unlock(&module_mutex); + synchronize_sched(); +- kfree(mod->args); +- free_arch_cleanup: + module_arch_cleanup(mod); + free_modinfo: + free_modinfo(mod); ++ kfree(mod->args); + free_unload: + module_unload_free(mod); + free_module: +@@ -2877,16 +2987,16 @@ SYSCALL_DEFINE3(init_module, void __user + MODULE_STATE_COMING, mod); + + /* Set RO and NX regions for core */ +- set_section_ro_nx(mod->module_core, +- mod->core_text_size, +- mod->core_ro_size, +- mod->core_size); ++ set_section_ro_nx(mod->module_core_rx, ++ mod->core_size_rx, ++ mod->core_size_rx, ++ mod->core_size_rx); + + /* Set RO and NX regions for init */ +- set_section_ro_nx(mod->module_init, +- mod->init_text_size, +- mod->init_ro_size, +- mod->init_size); ++ set_section_ro_nx(mod->module_init_rx, ++ mod->init_size_rx, ++ mod->init_size_rx, ++ mod->init_size_rx); + + do_mod_ctors(mod); + /* Start the module */ +@@ -2931,11 +3041,13 @@ SYSCALL_DEFINE3(init_module, void __user + mod->symtab = mod->core_symtab; + mod->strtab = mod->core_strtab; + #endif +- unset_section_ro_nx(mod, mod->module_init); +- module_free(mod, mod->module_init); +- mod->module_init = NULL; +- mod->init_size = 0; +- mod->init_text_size = 0; ++ unset_section_ro_nx(mod, mod->module_init_rx); ++ module_free(mod, mod->module_init_rw); ++ module_free_exec(mod, mod->module_init_rx); ++ mod->module_init_rw = NULL; ++ mod->module_init_rx = NULL; ++ mod->init_size_rw = 0; ++ mod->init_size_rx = 0; + mutex_unlock(&module_mutex); + + return 0; +@@ -2966,10 +3078,16 @@ static const char *get_ksymbol(struct mo + unsigned long nextval; + + /* At worse, next value is at end of module */ +- if (within_module_init(addr, mod)) +- nextval = (unsigned long)mod->module_init+mod->init_text_size; ++ if (within_module_init_rx(addr, mod)) ++ nextval = (unsigned long)mod->module_init_rx+mod->init_size_rx; ++ else if (within_module_init_rw(addr, mod)) ++ nextval = (unsigned long)mod->module_init_rw+mod->init_size_rw; ++ else if (within_module_core_rx(addr, mod)) ++ nextval = (unsigned long)mod->module_core_rx+mod->core_size_rx; ++ else if (within_module_core_rw(addr, mod)) ++ nextval = (unsigned long)mod->module_core_rw+mod->core_size_rw; + else +- nextval = (unsigned long)mod->module_core+mod->core_text_size; ++ return NULL; + + /* Scan for closest preceeding symbol, and next symbol. (ELF + starts real symbols at 1). */ +@@ -3215,7 +3333,7 @@ static int m_show(struct seq_file *m, vo + char buf[8]; + + seq_printf(m, "%s %u", +- mod->name, mod->init_size + mod->core_size); ++ mod->name, mod->init_size_rx + mod->init_size_rw + mod->core_size_rx + mod->core_size_rw); + print_unload_info(m, mod); + + /* Informative for users. */ +@@ -3224,7 +3342,7 @@ static int m_show(struct seq_file *m, vo + mod->state == MODULE_STATE_COMING ? "Loading": + "Live"); + /* Used by oprofile and other similar tools. */ +- seq_printf(m, " 0x%p", mod->module_core); ++ seq_printf(m, " 0x%p 0x%p", mod->module_core_rx, mod->module_core_rw); + + /* Taints info */ + if (mod->taints) +@@ -3260,7 +3378,17 @@ static const struct file_operations proc + + static int __init proc_modules_init(void) + { ++#ifndef CONFIG_GRKERNSEC_HIDESYM ++#ifdef CONFIG_GRKERNSEC_PROC_USER ++ proc_create("modules", S_IRUSR, NULL, &proc_modules_operations); ++#elif defined(CONFIG_GRKERNSEC_PROC_USERGROUP) ++ proc_create("modules", S_IRUSR | S_IRGRP, NULL, &proc_modules_operations); ++#else + proc_create("modules", 0, NULL, &proc_modules_operations); ++#endif ++#else ++ proc_create("modules", S_IRUSR, NULL, &proc_modules_operations); ++#endif + return 0; + } + module_init(proc_modules_init); +@@ -3319,12 +3447,12 @@ struct module *__module_address(unsigned + { + struct module *mod; + +- if (addr < module_addr_min || addr > module_addr_max) ++ if ((addr < module_addr_min_rx || addr > module_addr_max_rx) && ++ (addr < module_addr_min_rw || addr > module_addr_max_rw)) + return NULL; + + list_for_each_entry_rcu(mod, &modules, list) +- if (within_module_core(addr, mod) +- || within_module_init(addr, mod)) ++ if (within_module_init(addr, mod) || within_module_core(addr, mod)) + return mod; + return NULL; + } +@@ -3358,11 +3486,20 @@ bool is_module_text_address(unsigned lon + */ + struct module *__module_text_address(unsigned long addr) + { +- struct module *mod = __module_address(addr); ++ struct module *mod; ++ ++#ifdef CONFIG_X86_32 ++ addr = ktla_ktva(addr); ++#endif ++ ++ if (addr < module_addr_min_rx || addr > module_addr_max_rx) ++ return NULL; ++ ++ mod = __module_address(addr); ++ + if (mod) { + /* Make sure it's within the text section. */ +- if (!within(addr, mod->module_init, mod->init_text_size) +- && !within(addr, mod->module_core, mod->core_text_size)) ++ if (!within_module_init_rx(addr, mod) && !within_module_core_rx(addr, mod)) + mod = NULL; + } + return mod; +diff -urNp linux-2.6.38.3/kernel/mutex.c linux-2.6.38.3/kernel/mutex.c +--- linux-2.6.38.3/kernel/mutex.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/mutex.c 2011-04-17 16:24:38.000000000 -0400 +@@ -160,7 +160,7 @@ __mutex_lock_common(struct mutex *lock, + */ + + for (;;) { +- struct thread_info *owner; ++ struct task_struct *owner; + + /* + * If we own the BKL, then don't spin. The owner of +@@ -205,7 +205,7 @@ __mutex_lock_common(struct mutex *lock, + spin_lock_mutex(&lock->wait_lock, flags); + + debug_mutex_lock_common(lock, &waiter); +- debug_mutex_add_waiter(lock, &waiter, task_thread_info(task)); ++ debug_mutex_add_waiter(lock, &waiter, task); + + /* add waiting tasks to the end of the waitqueue (FIFO): */ + list_add_tail(&waiter.list, &lock->wait_list); +@@ -234,8 +234,7 @@ __mutex_lock_common(struct mutex *lock, + * TASK_UNINTERRUPTIBLE case.) + */ + if (unlikely(signal_pending_state(state, task))) { +- mutex_remove_waiter(lock, &waiter, +- task_thread_info(task)); ++ mutex_remove_waiter(lock, &waiter, task); + mutex_release(&lock->dep_map, 1, ip); + spin_unlock_mutex(&lock->wait_lock, flags); + +@@ -256,7 +255,7 @@ __mutex_lock_common(struct mutex *lock, + done: + lock_acquired(&lock->dep_map, ip); + /* got the lock - rejoice! */ +- mutex_remove_waiter(lock, &waiter, current_thread_info()); ++ mutex_remove_waiter(lock, &waiter, current); + mutex_set_owner(lock); + + /* set it to 0 if there are no waiters left: */ +diff -urNp linux-2.6.38.3/kernel/mutex-debug.c linux-2.6.38.3/kernel/mutex-debug.c +--- linux-2.6.38.3/kernel/mutex-debug.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/mutex-debug.c 2011-04-17 16:23:07.000000000 -0400 +@@ -49,21 +49,21 @@ void debug_mutex_free_waiter(struct mute + } + + void debug_mutex_add_waiter(struct mutex *lock, struct mutex_waiter *waiter, +- struct thread_info *ti) ++ struct task_struct *task) + { + SMP_DEBUG_LOCKS_WARN_ON(!spin_is_locked(&lock->wait_lock)); + + /* Mark the current thread as blocked on the lock: */ +- ti->task->blocked_on = waiter; ++ task->blocked_on = waiter; + } + + void mutex_remove_waiter(struct mutex *lock, struct mutex_waiter *waiter, +- struct thread_info *ti) ++ struct task_struct *task) + { + DEBUG_LOCKS_WARN_ON(list_empty(&waiter->list)); +- DEBUG_LOCKS_WARN_ON(waiter->task != ti->task); +- DEBUG_LOCKS_WARN_ON(ti->task->blocked_on != waiter); +- ti->task->blocked_on = NULL; ++ DEBUG_LOCKS_WARN_ON(waiter->task != task); ++ DEBUG_LOCKS_WARN_ON(task->blocked_on != waiter->task); ++ task->blocked_on = NULL; + + list_del_init(&waiter->list); + waiter->task = NULL; +@@ -75,7 +75,7 @@ void debug_mutex_unlock(struct mutex *lo + return; + + DEBUG_LOCKS_WARN_ON(lock->magic != lock); +- DEBUG_LOCKS_WARN_ON(lock->owner != current_thread_info()); ++ DEBUG_LOCKS_WARN_ON(lock->owner != current); + DEBUG_LOCKS_WARN_ON(!lock->wait_list.prev && !lock->wait_list.next); + mutex_clear_owner(lock); + } +diff -urNp linux-2.6.38.3/kernel/mutex-debug.h linux-2.6.38.3/kernel/mutex-debug.h +--- linux-2.6.38.3/kernel/mutex-debug.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/mutex-debug.h 2011-04-17 16:26:49.000000000 -0400 +@@ -20,16 +20,16 @@ extern void debug_mutex_wake_waiter(stru + extern void debug_mutex_free_waiter(struct mutex_waiter *waiter); + extern void debug_mutex_add_waiter(struct mutex *lock, + struct mutex_waiter *waiter, +- struct thread_info *ti); ++ struct task_struct *task); + extern void mutex_remove_waiter(struct mutex *lock, struct mutex_waiter *waiter, +- struct thread_info *ti); ++ struct task_struct *task); + extern void debug_mutex_unlock(struct mutex *lock); + extern void debug_mutex_init(struct mutex *lock, const char *name, + struct lock_class_key *key); + + static inline void mutex_set_owner(struct mutex *lock) + { +- lock->owner = current_thread_info(); ++ lock->owner = current; + } + + static inline void mutex_clear_owner(struct mutex *lock) +diff -urNp linux-2.6.38.3/kernel/mutex.h linux-2.6.38.3/kernel/mutex.h +--- linux-2.6.38.3/kernel/mutex.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/mutex.h 2011-04-17 16:24:51.000000000 -0400 +@@ -19,7 +19,7 @@ + #ifdef CONFIG_SMP + static inline void mutex_set_owner(struct mutex *lock) + { +- lock->owner = current_thread_info(); ++ lock->owner = current; + } + + static inline void mutex_clear_owner(struct mutex *lock) +diff -urNp linux-2.6.38.3/kernel/panic.c linux-2.6.38.3/kernel/panic.c +--- linux-2.6.38.3/kernel/panic.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/panic.c 2011-04-17 15:57:32.000000000 -0400 +@@ -369,7 +369,7 @@ static void warn_slowpath_common(const c + const char *board; + + printk(KERN_WARNING "------------[ cut here ]------------\n"); +- printk(KERN_WARNING "WARNING: at %s:%d %pS()\n", file, line, caller); ++ printk(KERN_WARNING "WARNING: at %s:%d %pA()\n", file, line, caller); + board = dmi_get_system_info(DMI_PRODUCT_NAME); + if (board) + printk(KERN_WARNING "Hardware name: %s\n", board); +@@ -424,7 +424,8 @@ EXPORT_SYMBOL(warn_slowpath_null); + */ + void __stack_chk_fail(void) + { +- panic("stack-protector: Kernel stack is corrupted in: %p\n", ++ dump_stack(); ++ panic("stack-protector: Kernel stack is corrupted in: %pA\n", + __builtin_return_address(0)); + } + EXPORT_SYMBOL(__stack_chk_fail); +diff -urNp linux-2.6.38.3/kernel/pid.c linux-2.6.38.3/kernel/pid.c +--- linux-2.6.38.3/kernel/pid.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/pid.c 2011-04-17 15:57:32.000000000 -0400 +@@ -33,6 +33,7 @@ + #include <linux/rculist.h> + #include <linux/bootmem.h> + #include <linux/hash.h> ++#include <linux/security.h> + #include <linux/pid_namespace.h> + #include <linux/init_task.h> + #include <linux/syscalls.h> +@@ -45,7 +46,7 @@ struct pid init_struct_pid = INIT_STRUCT + + int pid_max = PID_MAX_DEFAULT; + +-#define RESERVED_PIDS 300 ++#define RESERVED_PIDS 500 + + int pid_max_min = RESERVED_PIDS + 1; + int pid_max_max = PID_MAX_LIMIT; +@@ -416,8 +417,15 @@ EXPORT_SYMBOL(pid_task); + */ + struct task_struct *find_task_by_pid_ns(pid_t nr, struct pid_namespace *ns) + { ++ struct task_struct *task; ++ + rcu_lockdep_assert(rcu_read_lock_held()); +- return pid_task(find_pid_ns(nr, ns), PIDTYPE_PID); ++ task = pid_task(find_pid_ns(nr, ns), PIDTYPE_PID); ++ ++ if (gr_pid_is_chrooted(task)) ++ return NULL; ++ ++ return task; + } + + struct task_struct *find_task_by_vpid(pid_t vnr) +diff -urNp linux-2.6.38.3/kernel/posix-cpu-timers.c linux-2.6.38.3/kernel/posix-cpu-timers.c +--- linux-2.6.38.3/kernel/posix-cpu-timers.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/posix-cpu-timers.c 2011-04-17 15:57:32.000000000 -0400 +@@ -6,6 +6,7 @@ + #include <linux/posix-timers.h> + #include <linux/errno.h> + #include <linux/math64.h> ++#include <linux/security.h> + #include <asm/uaccess.h> + #include <linux/kernel_stat.h> + #include <trace/events/timer.h> +diff -urNp linux-2.6.38.3/kernel/posix-timers.c linux-2.6.38.3/kernel/posix-timers.c +--- linux-2.6.38.3/kernel/posix-timers.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/posix-timers.c 2011-04-17 15:57:33.000000000 -0400 +@@ -42,6 +42,7 @@ + #include <linux/compiler.h> + #include <linux/idr.h> + #include <linux/posix-timers.h> ++#include <linux/grsecurity.h> + #include <linux/syscalls.h> + #include <linux/wait.h> + #include <linux/workqueue.h> +@@ -955,6 +956,13 @@ SYSCALL_DEFINE2(clock_settime, const clo + if (copy_from_user(&new_tp, tp, sizeof (*tp))) + return -EFAULT; + ++ /* only the CLOCK_REALTIME clock can be set, all other clocks ++ have their clock_set fptr set to a nosettime dummy function ++ CLOCK_REALTIME has a NULL clock_set fptr which causes it to ++ call common_clock_set, which calls do_sys_settimeofday, which ++ we hook ++ */ ++ + return CLOCK_DISPATCH(which_clock, clock_set, (which_clock, &new_tp)); + } + +diff -urNp linux-2.6.38.3/kernel/power/poweroff.c linux-2.6.38.3/kernel/power/poweroff.c +--- linux-2.6.38.3/kernel/power/poweroff.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/power/poweroff.c 2011-04-17 15:57:33.000000000 -0400 +@@ -37,7 +37,7 @@ static struct sysrq_key_op sysrq_powerof + .enable_mask = SYSRQ_ENABLE_BOOT, + }; + +-static int pm_sysrq_init(void) ++static int __init pm_sysrq_init(void) + { + register_sysrq_key('o', &sysrq_poweroff_op); + return 0; +diff -urNp linux-2.6.38.3/kernel/power/process.c linux-2.6.38.3/kernel/power/process.c +--- linux-2.6.38.3/kernel/power/process.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/power/process.c 2011-04-17 15:57:33.000000000 -0400 +@@ -41,6 +41,7 @@ static int try_to_freeze_tasks(bool sig_ + u64 elapsed_csecs64; + unsigned int elapsed_csecs; + bool wakeup = false; ++ bool timedout = false; + + do_gettimeofday(&start); + +@@ -51,6 +52,8 @@ static int try_to_freeze_tasks(bool sig_ + + while (true) { + todo = 0; ++ if (time_after(jiffies, end_time)) ++ timedout = true; + read_lock(&tasklist_lock); + do_each_thread(g, p) { + if (frozen(p) || !freezable(p)) +@@ -71,9 +74,13 @@ static int try_to_freeze_tasks(bool sig_ + * try_to_stop() after schedule() in ptrace/signal + * stop sees TIF_FREEZE. + */ +- if (!task_is_stopped_or_traced(p) && +- !freezer_should_skip(p)) ++ if (!task_is_stopped_or_traced(p) && !freezer_should_skip(p)) { + todo++; ++ if (timedout) { ++ printk(KERN_ERR "Task refusing to freeze:\n"); ++ sched_show_task(p); ++ } ++ } + } while_each_thread(g, p); + read_unlock(&tasklist_lock); + +@@ -82,7 +89,7 @@ static int try_to_freeze_tasks(bool sig_ + todo += wq_busy; + } + +- if (!todo || time_after(jiffies, end_time)) ++ if (!todo || timedout) + break; + + if (pm_wakeup_pending()) { +diff -urNp linux-2.6.38.3/kernel/printk.c linux-2.6.38.3/kernel/printk.c +--- linux-2.6.38.3/kernel/printk.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/printk.c 2011-04-17 15:57:33.000000000 -0400 +@@ -279,12 +279,17 @@ static int check_syslog_permissions(int + if (from_file && type != SYSLOG_ACTION_OPEN) + return 0; + ++#ifdef CONFIG_GRKERNSEC_DMESG ++ if (grsec_enable_dmesg && !capable(CAP_SYSLOG) && !capable_nolog(CAP_SYS_ADMIN)) ++ return -EPERM; ++#endif ++ + if (syslog_action_restricted(type)) { + if (capable(CAP_SYSLOG)) + return 0; + /* For historical reasons, accept CAP_SYS_ADMIN too, with a warning */ + if (capable(CAP_SYS_ADMIN)) { +- WARN_ONCE(1, "Attempt to access syslog with CAP_SYS_ADMIN " ++ printk_once(KERN_WARNING "Attempt to access syslog with CAP_SYS_ADMIN " + "but no CAP_SYSLOG (deprecated).\n"); + return 0; + } +diff -urNp linux-2.6.38.3/kernel/ptrace.c linux-2.6.38.3/kernel/ptrace.c +--- linux-2.6.38.3/kernel/ptrace.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/ptrace.c 2011-04-17 15:57:33.000000000 -0400 +@@ -116,7 +116,8 @@ int ptrace_check_attach(struct task_stru + return ret; + } + +-int __ptrace_may_access(struct task_struct *task, unsigned int mode) ++static int __ptrace_may_access(struct task_struct *task, unsigned int mode, ++ unsigned int log) + { + const struct cred *cred = current_cred(), *tcred; + +@@ -140,7 +141,9 @@ int __ptrace_may_access(struct task_stru + cred->gid != tcred->egid || + cred->gid != tcred->sgid || + cred->gid != tcred->gid) && +- !capable(CAP_SYS_PTRACE)) { ++ ((!log && !capable_nolog(CAP_SYS_PTRACE)) || ++ (log && !capable(CAP_SYS_PTRACE))) ++ ) { + rcu_read_unlock(); + return -EPERM; + } +@@ -148,7 +151,9 @@ int __ptrace_may_access(struct task_stru + smp_rmb(); + if (task->mm) + dumpable = get_dumpable(task->mm); +- if (!dumpable && !capable(CAP_SYS_PTRACE)) ++ if (!dumpable && ++ ((!log && !capable_nolog(CAP_SYS_PTRACE)) || ++ (log && !capable(CAP_SYS_PTRACE)))) + return -EPERM; + + return security_ptrace_access_check(task, mode); +@@ -158,7 +163,16 @@ bool ptrace_may_access(struct task_struc + { + int err; + task_lock(task); +- err = __ptrace_may_access(task, mode); ++ err = __ptrace_may_access(task, mode, 0); ++ task_unlock(task); ++ return !err; ++} ++ ++bool ptrace_may_access_log(struct task_struct *task, unsigned int mode) ++{ ++ int err; ++ task_lock(task); ++ err = __ptrace_may_access(task, mode, 1); + task_unlock(task); + return !err; + } +@@ -185,7 +199,7 @@ static int ptrace_attach(struct task_str + goto out; + + task_lock(task); +- retval = __ptrace_may_access(task, PTRACE_MODE_ATTACH); ++ retval = __ptrace_may_access(task, PTRACE_MODE_ATTACH, 1); + task_unlock(task); + if (retval) + goto unlock_creds; +@@ -198,7 +212,7 @@ static int ptrace_attach(struct task_str + goto unlock_tasklist; + + task->ptrace = PT_PTRACED; +- if (capable(CAP_SYS_PTRACE)) ++ if (capable_nolog(CAP_SYS_PTRACE)) + task->ptrace |= PT_PTRACE_CAP; + + __ptrace_link(task, current); +@@ -369,7 +383,7 @@ int ptrace_readdata(struct task_struct * + break; + return -EIO; + } +- if (copy_to_user(dst, buf, retval)) ++ if (retval > sizeof(buf) || copy_to_user(dst, buf, retval)) + return -EFAULT; + copied += retval; + src += retval; +@@ -565,7 +579,7 @@ int ptrace_request(struct task_struct *c + { + int ret = -EIO; + siginfo_t siginfo; +- void __user *datavp = (void __user *) data; ++ void __user *datavp = (__force void __user *) data; + unsigned long __user *datalp = datavp; + + switch (request) { +@@ -713,14 +727,21 @@ SYSCALL_DEFINE4(ptrace, long, request, l + goto out; + } + ++ if (gr_handle_ptrace(child, request)) { ++ ret = -EPERM; ++ goto out_put_task_struct; ++ } ++ + if (request == PTRACE_ATTACH) { + ret = ptrace_attach(child); + /* + * Some architectures need to do book-keeping after + * a ptrace attach. + */ +- if (!ret) ++ if (!ret) { + arch_ptrace_attach(child); ++ gr_audit_ptrace(child); ++ } + goto out_put_task_struct; + } + +@@ -745,7 +766,7 @@ int generic_ptrace_peekdata(struct task_ + copied = access_process_vm(tsk, addr, &tmp, sizeof(tmp), 0); + if (copied != sizeof(tmp)) + return -EIO; +- return put_user(tmp, (unsigned long __user *)data); ++ return put_user(tmp, (__force unsigned long __user *)data); + } + + int generic_ptrace_pokedata(struct task_struct *tsk, unsigned long addr, +@@ -855,14 +876,21 @@ asmlinkage long compat_sys_ptrace(compat + goto out; + } + ++ if (gr_handle_ptrace(child, request)) { ++ ret = -EPERM; ++ goto out_put_task_struct; ++ } ++ + if (request == PTRACE_ATTACH) { + ret = ptrace_attach(child); + /* + * Some architectures need to do book-keeping after + * a ptrace attach. + */ +- if (!ret) ++ if (!ret) { + arch_ptrace_attach(child); ++ gr_audit_ptrace(child); ++ } + goto out_put_task_struct; + } + +diff -urNp linux-2.6.38.3/kernel/rcutree.c linux-2.6.38.3/kernel/rcutree.c +--- linux-2.6.38.3/kernel/rcutree.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/rcutree.c 2011-04-17 15:57:33.000000000 -0400 +@@ -1389,7 +1389,7 @@ __rcu_process_callbacks(struct rcu_state + /* + * Do softirq processing for the current CPU. + */ +-static void rcu_process_callbacks(struct softirq_action *unused) ++static void rcu_process_callbacks(void) + { + /* + * Memory references from any prior RCU read-side critical sections +diff -urNp linux-2.6.38.3/kernel/rcutree_plugin.h linux-2.6.38.3/kernel/rcutree_plugin.h +--- linux-2.6.38.3/kernel/rcutree_plugin.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/rcutree_plugin.h 2011-04-17 15:57:33.000000000 -0400 +@@ -730,7 +730,7 @@ void synchronize_rcu_expedited(void) + + /* Clean up and exit. */ + smp_mb(); /* ensure expedited GP seen before counter increment. */ +- ACCESS_ONCE(sync_rcu_preempt_exp_count)++; ++ ACCESS_ONCE_RW(sync_rcu_preempt_exp_count)++; + unlock_mb_ret: + mutex_unlock(&sync_rcu_preempt_exp_mutex); + mb_ret: +diff -urNp linux-2.6.38.3/kernel/resource.c linux-2.6.38.3/kernel/resource.c +--- linux-2.6.38.3/kernel/resource.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/resource.c 2011-04-17 15:57:33.000000000 -0400 +@@ -133,8 +133,18 @@ static const struct file_operations proc + + static int __init ioresources_init(void) + { ++#ifdef CONFIG_GRKERNSEC_PROC_ADD ++#ifdef CONFIG_GRKERNSEC_PROC_USER ++ proc_create("ioports", S_IRUSR, NULL, &proc_ioports_operations); ++ proc_create("iomem", S_IRUSR, NULL, &proc_iomem_operations); ++#elif defined(CONFIG_GRKERNSEC_PROC_USERGROUP) ++ proc_create("ioports", S_IRUSR | S_IRGRP, NULL, &proc_ioports_operations); ++ proc_create("iomem", S_IRUSR | S_IRGRP, NULL, &proc_iomem_operations); ++#endif ++#else + proc_create("ioports", 0, NULL, &proc_ioports_operations); + proc_create("iomem", 0, NULL, &proc_iomem_operations); ++#endif + return 0; + } + __initcall(ioresources_init); +diff -urNp linux-2.6.38.3/kernel/rtmutex.c linux-2.6.38.3/kernel/rtmutex.c +--- linux-2.6.38.3/kernel/rtmutex.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/rtmutex.c 2011-04-17 15:57:33.000000000 -0400 +@@ -511,7 +511,7 @@ static void wakeup_next_waiter(struct rt + */ + raw_spin_lock_irqsave(&pendowner->pi_lock, flags); + +- WARN_ON(!pendowner->pi_blocked_on); ++ BUG_ON(!pendowner->pi_blocked_on); + WARN_ON(pendowner->pi_blocked_on != waiter); + WARN_ON(pendowner->pi_blocked_on->lock != lock); + +diff -urNp linux-2.6.38.3/kernel/sched.c linux-2.6.38.3/kernel/sched.c +--- linux-2.6.38.3/kernel/sched.c 2011-03-23 17:20:08.000000000 -0400 ++++ linux-2.6.38.3/kernel/sched.c 2011-04-17 16:29:21.000000000 -0400 +@@ -4024,7 +4024,7 @@ EXPORT_SYMBOL(schedule); + * Look out! "owner" is an entirely speculative pointer + * access and not reliable. + */ +-int mutex_spin_on_owner(struct mutex *lock, struct thread_info *owner) ++int mutex_spin_on_owner(struct mutex *lock, struct task_struct *owner) + { + unsigned int cpu; + struct rq *rq; +@@ -4038,10 +4038,10 @@ int mutex_spin_on_owner(struct mutex *lo + * DEBUG_PAGEALLOC could have unmapped it if + * the mutex owner just released it and exited. + */ +- if (probe_kernel_address(&owner->cpu, cpu)) ++ if (probe_kernel_address(&task_thread_info(owner)->cpu, cpu)) + return 0; + #else +- cpu = owner->cpu; ++ cpu = task_thread_info(owner)->cpu; + #endif + + /* +@@ -4078,7 +4078,7 @@ int mutex_spin_on_owner(struct mutex *lo + /* + * Is that owner really running on that cpu? + */ +- if (task_thread_info(rq->curr) != owner || need_resched()) ++ if (rq->curr != owner || need_resched()) + return 0; + + arch_mutex_cpu_relax(); +@@ -4638,6 +4638,8 @@ int can_nice(const struct task_struct *p + /* convert nice value [19,-20] to rlimit style value [1,40] */ + int nice_rlim = 20 - nice; + ++ gr_learn_resource(p, RLIMIT_NICE, nice_rlim, 1); ++ + return (nice_rlim <= task_rlimit(p, RLIMIT_NICE) || + capable(CAP_SYS_NICE)); + } +@@ -4671,7 +4673,8 @@ SYSCALL_DEFINE1(nice, int, increment) + if (nice > 19) + nice = 19; + +- if (increment < 0 && !can_nice(current, nice)) ++ if (increment < 0 && (!can_nice(current, nice) || ++ gr_handle_chroot_nice())) + return -EPERM; + + retval = security_task_setnice(current, nice); +@@ -4814,6 +4817,7 @@ recheck: + unsigned long rlim_rtprio = + task_rlimit(p, RLIMIT_RTPRIO); + ++ gr_learn_resource(p, RLIMIT_RTPRIO, param->sched_priority, 1); + /* can't set/change the rt policy */ + if (policy != p->policy && !rlim_rtprio) + return -EPERM; +@@ -6942,7 +6946,7 @@ static void init_sched_groups_power(int + long power; + int weight; + +- WARN_ON(!sd || !sd->groups); ++ BUG_ON(!sd || !sd->groups); + + if (cpu != group_first_cpu(sd->groups)) + return; +diff -urNp linux-2.6.38.3/kernel/sched_fair.c linux-2.6.38.3/kernel/sched_fair.c +--- linux-2.6.38.3/kernel/sched_fair.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/sched_fair.c 2011-04-17 15:57:33.000000000 -0400 +@@ -3960,7 +3960,7 @@ static void nohz_idle_balance(int this_c + * run_rebalance_domains is triggered when needed from the scheduler tick. + * Also triggered for nohz idle balancing (with nohz_balancing_kick set). + */ +-static void run_rebalance_domains(struct softirq_action *h) ++static void run_rebalance_domains(void) + { + int this_cpu = smp_processor_id(); + struct rq *this_rq = cpu_rq(this_cpu); +diff -urNp linux-2.6.38.3/kernel/signal.c linux-2.6.38.3/kernel/signal.c +--- linux-2.6.38.3/kernel/signal.c 2011-04-17 16:53:01.000000000 -0400 ++++ linux-2.6.38.3/kernel/signal.c 2011-04-17 16:53:48.000000000 -0400 +@@ -45,12 +45,12 @@ static struct kmem_cache *sigqueue_cache + + int print_fatal_signals __read_mostly; + +-static void __user *sig_handler(struct task_struct *t, int sig) ++static __sighandler_t sig_handler(struct task_struct *t, int sig) + { + return t->sighand->action[sig - 1].sa.sa_handler; + } + +-static int sig_handler_ignored(void __user *handler, int sig) ++static int sig_handler_ignored(__sighandler_t handler, int sig) + { + /* Is it explicitly or implicitly ignored? */ + return handler == SIG_IGN || +@@ -60,7 +60,7 @@ static int sig_handler_ignored(void __us + static int sig_task_ignored(struct task_struct *t, int sig, + int from_ancestor_ns) + { +- void __user *handler; ++ __sighandler_t handler; + + handler = sig_handler(t, sig); + +@@ -243,6 +243,9 @@ __sigqueue_alloc(int sig, struct task_st + atomic_inc(&user->sigpending); + rcu_read_unlock(); + ++ if (!override_rlimit) ++ gr_learn_resource(t, RLIMIT_SIGPENDING, atomic_read(&user->sigpending), 1); ++ + if (override_rlimit || + atomic_read(&user->sigpending) <= + task_rlimit(t, RLIMIT_SIGPENDING)) { +@@ -367,7 +370,7 @@ flush_signal_handlers(struct task_struct + + int unhandled_signal(struct task_struct *tsk, int sig) + { +- void __user *handler = tsk->sighand->action[sig-1].sa.sa_handler; ++ __sighandler_t handler = tsk->sighand->action[sig-1].sa.sa_handler; + if (is_global_init(tsk)) + return 1; + if (handler != SIG_IGN && handler != SIG_DFL) +@@ -678,6 +681,9 @@ static int check_kill_permission(int sig + } + } + ++ if (gr_handle_signal(t, sig)) ++ return -EPERM; ++ + return security_task_kill(t, info, sig, 0); + } + +@@ -1025,7 +1031,7 @@ __group_send_sig_info(int sig, struct si + return send_signal(sig, info, p, 1); + } + +-static int ++int + specific_send_sig_info(int sig, struct siginfo *info, struct task_struct *t) + { + return send_signal(sig, info, t, 0); +@@ -1062,6 +1068,7 @@ force_sig_info(int sig, struct siginfo * + unsigned long int flags; + int ret, blocked, ignored; + struct k_sigaction *action; ++ int is_unhandled = 0; + + spin_lock_irqsave(&t->sighand->siglock, flags); + action = &t->sighand->action[sig-1]; +@@ -1076,9 +1083,18 @@ force_sig_info(int sig, struct siginfo * + } + if (action->sa.sa_handler == SIG_DFL) + t->signal->flags &= ~SIGNAL_UNKILLABLE; ++ if (action->sa.sa_handler == SIG_IGN || action->sa.sa_handler == SIG_DFL) ++ is_unhandled = 1; + ret = specific_send_sig_info(sig, info, t); + spin_unlock_irqrestore(&t->sighand->siglock, flags); + ++ /* only deal with unhandled signals, java etc trigger SIGSEGV during ++ normal operation */ ++ if (is_unhandled) { ++ gr_log_signal(sig, !is_si_special(info) ? info->si_addr : NULL, t); ++ gr_handle_crash(t, sig); ++ } ++ + return ret; + } + +@@ -1137,8 +1153,11 @@ int group_send_sig_info(int sig, struct + ret = check_kill_permission(sig, info, p); + rcu_read_unlock(); + +- if (!ret && sig) ++ if (!ret && sig) { + ret = do_send_sig_info(sig, info, p, true); ++ if (!ret) ++ gr_log_signal(sig, !is_si_special(info) ? info->si_addr : NULL, p); ++ } + + return ret; + } +diff -urNp linux-2.6.38.3/kernel/smp.c linux-2.6.38.3/kernel/smp.c +--- linux-2.6.38.3/kernel/smp.c 2011-03-23 17:20:08.000000000 -0400 ++++ linux-2.6.38.3/kernel/smp.c 2011-04-17 15:57:33.000000000 -0400 +@@ -583,22 +583,22 @@ int smp_call_function(smp_call_func_t fu + } + EXPORT_SYMBOL(smp_call_function); + +-void ipi_call_lock(void) ++void ipi_call_lock(void) __acquires(call_function.lock) + { + raw_spin_lock(&call_function.lock); + } + +-void ipi_call_unlock(void) ++void ipi_call_unlock(void) __releases(call_function.lock) + { + raw_spin_unlock(&call_function.lock); + } + +-void ipi_call_lock_irq(void) ++void ipi_call_lock_irq(void) __acquires(call_function.lock) + { + raw_spin_lock_irq(&call_function.lock); + } + +-void ipi_call_unlock_irq(void) ++void ipi_call_unlock_irq(void) __releases(call_function.lock) + { + raw_spin_unlock_irq(&call_function.lock); + } +diff -urNp linux-2.6.38.3/kernel/softirq.c linux-2.6.38.3/kernel/softirq.c +--- linux-2.6.38.3/kernel/softirq.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/softirq.c 2011-04-17 15:57:33.000000000 -0400 +@@ -56,7 +56,7 @@ static struct softirq_action softirq_vec + + static DEFINE_PER_CPU(struct task_struct *, ksoftirqd); + +-char *softirq_to_name[NR_SOFTIRQS] = { ++const char * const softirq_to_name[NR_SOFTIRQS] = { + "HI", "TIMER", "NET_TX", "NET_RX", "BLOCK", "BLOCK_IOPOLL", + "TASKLET", "SCHED", "HRTIMER", "RCU" + }; +@@ -206,7 +206,7 @@ EXPORT_SYMBOL(local_bh_enable_ip); + + asmlinkage void __do_softirq(void) + { +- struct softirq_action *h; ++ const struct softirq_action *h; + __u32 pending; + int max_restart = MAX_SOFTIRQ_RESTART; + int cpu; +@@ -235,7 +235,7 @@ restart: + kstat_incr_softirqs_this_cpu(vec_nr); + + trace_softirq_entry(vec_nr); +- h->action(h); ++ h->action(); + trace_softirq_exit(vec_nr); + if (unlikely(prev_count != preempt_count())) { + printk(KERN_ERR "huh, entered softirq %u %s %p" +@@ -365,7 +365,7 @@ void raise_softirq(unsigned int nr) + local_irq_restore(flags); + } + +-void open_softirq(int nr, void (*action)(struct softirq_action *)) ++void open_softirq(int nr, void (*action)(void)) + { + softirq_vec[nr].action = action; + } +@@ -421,7 +421,7 @@ void __tasklet_hi_schedule_first(struct + + EXPORT_SYMBOL(__tasklet_hi_schedule_first); + +-static void tasklet_action(struct softirq_action *a) ++static void tasklet_action(void) + { + struct tasklet_struct *list; + +@@ -456,7 +456,7 @@ static void tasklet_action(struct softir + } + } + +-static void tasklet_hi_action(struct softirq_action *a) ++static void tasklet_hi_action(void) + { + struct tasklet_struct *list; + +diff -urNp linux-2.6.38.3/kernel/sys.c linux-2.6.38.3/kernel/sys.c +--- linux-2.6.38.3/kernel/sys.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/sys.c 2011-04-17 15:57:33.000000000 -0400 +@@ -136,6 +136,12 @@ static int set_one_prio(struct task_stru + error = -EACCES; + goto out; + } ++ ++ if (gr_handle_chroot_setpriority(p, niceval)) { ++ error = -EACCES; ++ goto out; ++ } ++ + no_nice = security_task_setnice(p, niceval); + if (no_nice) { + error = no_nice; +@@ -517,6 +523,9 @@ SYSCALL_DEFINE2(setregid, gid_t, rgid, g + goto error; + } + ++ if (gr_check_group_change(new->gid, new->egid, -1)) ++ goto error; ++ + if (rgid != (gid_t) -1 || + (egid != (gid_t) -1 && egid != old->gid)) + new->sgid = new->egid; +@@ -546,6 +555,10 @@ SYSCALL_DEFINE1(setgid, gid_t, gid) + old = current_cred(); + + retval = -EPERM; ++ ++ if (gr_check_group_change(gid, gid, gid)) ++ goto error; ++ + if (capable(CAP_SETGID)) + new->gid = new->egid = new->sgid = new->fsgid = gid; + else if (gid == old->gid || gid == old->sgid) +@@ -626,6 +639,9 @@ SYSCALL_DEFINE2(setreuid, uid_t, ruid, u + goto error; + } + ++ if (gr_check_user_change(new->uid, new->euid, -1)) ++ goto error; ++ + if (new->uid != old->uid) { + retval = set_user(new); + if (retval < 0) +@@ -670,6 +686,12 @@ SYSCALL_DEFINE1(setuid, uid_t, uid) + old = current_cred(); + + retval = -EPERM; ++ ++ if (gr_check_crash_uid(uid)) ++ goto error; ++ if (gr_check_user_change(uid, uid, uid)) ++ goto error; ++ + if (capable(CAP_SETUID)) { + new->suid = new->uid = uid; + if (uid != old->uid) { +@@ -724,6 +746,9 @@ SYSCALL_DEFINE3(setresuid, uid_t, ruid, + goto error; + } + ++ if (gr_check_user_change(ruid, euid, -1)) ++ goto error; ++ + if (ruid != (uid_t) -1) { + new->uid = ruid; + if (ruid != old->uid) { +@@ -788,6 +813,9 @@ SYSCALL_DEFINE3(setresgid, gid_t, rgid, + goto error; + } + ++ if (gr_check_group_change(rgid, egid, -1)) ++ goto error; ++ + if (rgid != (gid_t) -1) + new->gid = rgid; + if (egid != (gid_t) -1) +@@ -834,6 +862,9 @@ SYSCALL_DEFINE1(setfsuid, uid_t, uid) + old = current_cred(); + old_fsuid = old->fsuid; + ++ if (gr_check_user_change(-1, -1, uid)) ++ goto error; ++ + if (uid == old->uid || uid == old->euid || + uid == old->suid || uid == old->fsuid || + capable(CAP_SETUID)) { +@@ -844,6 +875,7 @@ SYSCALL_DEFINE1(setfsuid, uid_t, uid) + } + } + ++error: + abort_creds(new); + return old_fsuid; + +@@ -870,12 +902,16 @@ SYSCALL_DEFINE1(setfsgid, gid_t, gid) + if (gid == old->gid || gid == old->egid || + gid == old->sgid || gid == old->fsgid || + capable(CAP_SETGID)) { ++ if (gr_check_group_change(-1, -1, gid)) ++ goto error; ++ + if (gid != old_fsgid) { + new->fsgid = gid; + goto change_okay; + } + } + ++error: + abort_creds(new); + return old_fsgid; + +@@ -1616,7 +1652,7 @@ SYSCALL_DEFINE5(prctl, int, option, unsi + error = get_dumpable(me->mm); + break; + case PR_SET_DUMPABLE: +- if (arg2 < 0 || arg2 > 1) { ++ if (arg2 > 1) { + error = -EINVAL; + break; + } +diff -urNp linux-2.6.38.3/kernel/sysctl.c linux-2.6.38.3/kernel/sysctl.c +--- linux-2.6.38.3/kernel/sysctl.c 2011-03-28 17:42:40.000000000 -0400 ++++ linux-2.6.38.3/kernel/sysctl.c 2011-04-17 15:57:33.000000000 -0400 +@@ -84,6 +84,13 @@ + + + #if defined(CONFIG_SYSCTL) ++#include <linux/grsecurity.h> ++#include <linux/grinternal.h> ++ ++extern __u32 gr_handle_sysctl(const ctl_table *table, const int op); ++extern int gr_handle_sysctl_mod(const char *dirname, const char *name, ++ const int op); ++extern int gr_handle_chroot_sysctl(const int op); + + /* External variables not in a header file. */ + extern int sysctl_overcommit_memory; +@@ -195,6 +202,7 @@ static int sysrq_sysctl_handler(ctl_tabl + } + + #endif ++extern struct ctl_table grsecurity_table[]; + + static struct ctl_table root_table[]; + static struct ctl_table_root sysctl_table_root; +@@ -224,6 +232,20 @@ extern struct ctl_table epoll_table[]; + int sysctl_legacy_va_layout; + #endif + ++#ifdef CONFIG_PAX_SOFTMODE ++static ctl_table pax_table[] = { ++ { ++ .procname = "softmode", ++ .data = &pax_softmode, ++ .maxlen = sizeof(unsigned int), ++ .mode = 0600, ++ .proc_handler = &proc_dointvec, ++ }, ++ ++ { } ++}; ++#endif ++ + /* The default sysctl tables: */ + + static struct ctl_table root_table[] = { +@@ -270,6 +292,22 @@ static int max_extfrag_threshold = 1000; + #endif + + static struct ctl_table kern_table[] = { ++#if defined(CONFIG_GRKERNSEC_SYSCTL) || defined(CONFIG_GRKERNSEC_ROFS) ++ { ++ .procname = "grsecurity", ++ .mode = 0500, ++ .child = grsecurity_table, ++ }, ++#endif ++ ++#ifdef CONFIG_PAX_SOFTMODE ++ { ++ .procname = "pax", ++ .mode = 0500, ++ .child = pax_table, ++ }, ++#endif ++ + { + .procname = "sched_child_runs_first", + .data = &sysctl_sched_child_runs_first, +@@ -551,7 +589,7 @@ static struct ctl_table kern_table[] = { + .data = &modprobe_path, + .maxlen = KMOD_PATH_LEN, + .mode = 0644, +- .proc_handler = proc_dostring, ++ .proc_handler = proc_dostring_modpriv, + }, + { + .procname = "modules_disabled", +@@ -713,16 +751,20 @@ static struct ctl_table kern_table[] = { + .extra1 = &zero, + .extra2 = &one, + }, ++#endif + { + .procname = "kptr_restrict", + .data = &kptr_restrict, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dmesg_restrict, ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ .extra1 = &two, ++#else + .extra1 = &zero, ++#endif + .extra2 = &two, + }, +-#endif + { + .procname = "ngroups_max", + .data = &ngroups_max, +@@ -1187,6 +1229,13 @@ static struct ctl_table vm_table[] = { + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + }, ++ { ++ .procname = "heap_stack_gap", ++ .data = &sysctl_heap_stack_gap, ++ .maxlen = sizeof(sysctl_heap_stack_gap), ++ .mode = 0644, ++ .proc_handler = proc_doulongvec_minmax, ++ }, + #else + { + .procname = "nr_trim_pages", +@@ -1698,6 +1747,16 @@ int sysctl_perm(struct ctl_table_root *r + int error; + int mode; + ++ if (table->parent != NULL && table->parent->procname != NULL && ++ table->procname != NULL && ++ gr_handle_sysctl_mod(table->parent->procname, table->procname, op)) ++ return -EACCES; ++ if (gr_handle_chroot_sysctl(op)) ++ return -EACCES; ++ error = gr_handle_sysctl(table, op); ++ if (error) ++ return error; ++ + error = security_sysctl(table, op & (MAY_READ | MAY_WRITE | MAY_EXEC)); + if (error) + return error; +@@ -2105,6 +2164,16 @@ int proc_dostring(struct ctl_table *tabl + buffer, lenp, ppos); + } + ++int proc_dostring_modpriv(struct ctl_table *table, int write, ++ void __user *buffer, size_t *lenp, loff_t *ppos) ++{ ++ if (write && !capable(CAP_SYS_MODULE)) ++ return -EPERM; ++ ++ return _proc_do_string(table->data, table->maxlen, write, ++ buffer, lenp, ppos); ++} ++ + static size_t proc_skip_spaces(char **buf) + { + size_t ret; +@@ -2210,6 +2279,8 @@ static int proc_put_long(void __user **b + len = strlen(tmp); + if (len > *size) + len = *size; ++ if (len > sizeof(tmp)) ++ len = sizeof(tmp); + if (copy_to_user(*buf, tmp, len)) + return -EFAULT; + *size -= len; +@@ -2526,8 +2597,11 @@ static int __do_proc_doulongvec_minmax(v + *i = val; + } else { + val = convdiv * (*i) / convmul; +- if (!first) ++ if (!first) { + err = proc_put_char(&buffer, &left, '\t'); ++ if (err) ++ break; ++ } + err = proc_put_long(&buffer, &left, val, false); + if (err) + break; +@@ -2922,6 +2996,12 @@ int proc_dostring(struct ctl_table *tabl + return -ENOSYS; + } + ++int proc_dostring_modpriv(struct ctl_table *table, int write, ++ void __user *buffer, size_t *lenp, loff_t *ppos) ++{ ++ return -ENOSYS; ++} ++ + int proc_dointvec(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) + { +@@ -2978,6 +3058,7 @@ EXPORT_SYMBOL(proc_dointvec_minmax); + EXPORT_SYMBOL(proc_dointvec_userhz_jiffies); + EXPORT_SYMBOL(proc_dointvec_ms_jiffies); + EXPORT_SYMBOL(proc_dostring); ++EXPORT_SYMBOL(proc_dostring_modpriv); + EXPORT_SYMBOL(proc_doulongvec_minmax); + EXPORT_SYMBOL(proc_doulongvec_ms_jiffies_minmax); + EXPORT_SYMBOL(register_sysctl_table); +diff -urNp linux-2.6.38.3/kernel/sysctl_check.c linux-2.6.38.3/kernel/sysctl_check.c +--- linux-2.6.38.3/kernel/sysctl_check.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/sysctl_check.c 2011-04-17 15:57:33.000000000 -0400 +@@ -131,6 +131,7 @@ int sysctl_check_table(struct nsproxy *n + set_fail(&fail, table, "Directory with extra2"); + } else { + if ((table->proc_handler == proc_dostring) || ++ (table->proc_handler == proc_dostring_modpriv) || + (table->proc_handler == proc_dointvec) || + (table->proc_handler == proc_dointvec_minmax) || + (table->proc_handler == proc_dointvec_jiffies) || +diff -urNp linux-2.6.38.3/kernel/taskstats.c linux-2.6.38.3/kernel/taskstats.c +--- linux-2.6.38.3/kernel/taskstats.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/taskstats.c 2011-04-17 15:57:33.000000000 -0400 +@@ -27,9 +27,12 @@ + #include <linux/cgroup.h> + #include <linux/fs.h> + #include <linux/file.h> ++#include <linux/grsecurity.h> + #include <net/genetlink.h> + #include <asm/atomic.h> + ++extern int gr_is_taskstats_denied(int pid); ++ + /* + * Maximum length of a cpumask that can be specified in + * the TASKSTATS_CMD_ATTR_REGISTER/DEREGISTER_CPUMASK attribute +@@ -549,6 +552,9 @@ err: + + static int taskstats_user_cmd(struct sk_buff *skb, struct genl_info *info) + { ++ if (gr_is_taskstats_denied(current->pid)) ++ return -EACCES; ++ + if (info->attrs[TASKSTATS_CMD_ATTR_REGISTER_CPUMASK]) + return cmd_attr_register_cpumask(info); + else if (info->attrs[TASKSTATS_CMD_ATTR_DEREGISTER_CPUMASK]) +diff -urNp linux-2.6.38.3/kernel/time/tick-broadcast.c linux-2.6.38.3/kernel/time/tick-broadcast.c +--- linux-2.6.38.3/kernel/time/tick-broadcast.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/time/tick-broadcast.c 2011-04-17 15:57:33.000000000 -0400 +@@ -116,7 +116,7 @@ int tick_device_uses_broadcast(struct cl + * then clear the broadcast bit. + */ + if (!(dev->features & CLOCK_EVT_FEAT_C3STOP)) { +- int cpu = smp_processor_id(); ++ cpu = smp_processor_id(); + + cpumask_clear_cpu(cpu, tick_get_broadcast_mask()); + tick_broadcast_clear_oneshot(cpu); +diff -urNp linux-2.6.38.3/kernel/time/timekeeping.c linux-2.6.38.3/kernel/time/timekeeping.c +--- linux-2.6.38.3/kernel/time/timekeeping.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/time/timekeeping.c 2011-04-17 15:57:33.000000000 -0400 +@@ -14,6 +14,7 @@ + #include <linux/init.h> + #include <linux/mm.h> + #include <linux/sched.h> ++#include <linux/grsecurity.h> + #include <linux/sysdev.h> + #include <linux/clocksource.h> + #include <linux/jiffies.h> +@@ -361,6 +362,8 @@ int do_settimeofday(struct timespec *tv) + if ((unsigned long)tv->tv_nsec >= NSEC_PER_SEC) + return -EINVAL; + ++ gr_log_timechange(); ++ + write_seqlock_irqsave(&xtime_lock, flags); + + timekeeping_forward_now(); +diff -urNp linux-2.6.38.3/kernel/time/timer_list.c linux-2.6.38.3/kernel/time/timer_list.c +--- linux-2.6.38.3/kernel/time/timer_list.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/time/timer_list.c 2011-04-17 15:57:33.000000000 -0400 +@@ -38,12 +38,16 @@ DECLARE_PER_CPU(struct hrtimer_cpu_base, + + static void print_name_offset(struct seq_file *m, void *sym) + { ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ SEQ_printf(m, "<%p>", NULL); ++#else + char symname[KSYM_NAME_LEN]; + + if (lookup_symbol_name((unsigned long)sym, symname) < 0) + SEQ_printf(m, "<%pK>", sym); + else + SEQ_printf(m, "%s", symname); ++#endif + } + + static void +@@ -112,7 +116,11 @@ next_one: + static void + print_base(struct seq_file *m, struct hrtimer_clock_base *base, u64 now) + { ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ SEQ_printf(m, " .base: %p\n", NULL); ++#else + SEQ_printf(m, " .base: %pK\n", base); ++#endif + SEQ_printf(m, " .index: %d\n", + base->index); + SEQ_printf(m, " .resolution: %Lu nsecs\n", +@@ -293,7 +301,11 @@ static int __init init_timer_list_procfs + { + struct proc_dir_entry *pe; + ++#ifdef CONFIG_GRKERNSEC_PROC_ADD ++ pe = proc_create("timer_list", 0400, NULL, &timer_list_fops); ++#else + pe = proc_create("timer_list", 0444, NULL, &timer_list_fops); ++#endif + if (!pe) + return -ENOMEM; + return 0; +diff -urNp linux-2.6.38.3/kernel/time/timer_stats.c linux-2.6.38.3/kernel/time/timer_stats.c +--- linux-2.6.38.3/kernel/time/timer_stats.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/time/timer_stats.c 2011-04-17 15:57:33.000000000 -0400 +@@ -269,12 +269,16 @@ void timer_stats_update_stats(void *time + + static void print_name_offset(struct seq_file *m, unsigned long addr) + { ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ seq_printf(m, "<%p>", NULL); ++#else + char symname[KSYM_NAME_LEN]; + + if (lookup_symbol_name(addr, symname) < 0) + seq_printf(m, "<%p>", (void *)addr); + else + seq_printf(m, "%s", symname); ++#endif + } + + static int tstats_show(struct seq_file *m, void *v) +@@ -417,7 +421,11 @@ static int __init init_tstats_procfs(voi + { + struct proc_dir_entry *pe; + ++#ifdef CONFIG_GRKERNSEC_PROC_ADD ++ pe = proc_create("timer_stats", 0600, NULL, &tstats_fops); ++#else + pe = proc_create("timer_stats", 0644, NULL, &tstats_fops); ++#endif + if (!pe) + return -ENOMEM; + return 0; +diff -urNp linux-2.6.38.3/kernel/time.c linux-2.6.38.3/kernel/time.c +--- linux-2.6.38.3/kernel/time.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/time.c 2011-04-17 15:57:33.000000000 -0400 +@@ -163,6 +163,11 @@ int do_sys_settimeofday(struct timespec + return error; + + if (tz) { ++ /* we log in do_settimeofday called below, so don't log twice ++ */ ++ if (!tv) ++ gr_log_timechange(); ++ + /* SMP safe, global irq locking makes it work. */ + sys_tz = *tz; + update_vsyscall_tz(); +diff -urNp linux-2.6.38.3/kernel/timer.c linux-2.6.38.3/kernel/timer.c +--- linux-2.6.38.3/kernel/timer.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/timer.c 2011-04-17 15:57:33.000000000 -0400 +@@ -1276,7 +1276,7 @@ void update_process_times(int user_tick) + /* + * This function runs timers and the timer-tq in bottom half context. + */ +-static void run_timer_softirq(struct softirq_action *h) ++static void run_timer_softirq(void) + { + struct tvec_base *base = __this_cpu_read(tvec_bases); + +diff -urNp linux-2.6.38.3/kernel/trace/ftrace.c linux-2.6.38.3/kernel/trace/ftrace.c +--- linux-2.6.38.3/kernel/trace/ftrace.c 2011-03-23 17:20:08.000000000 -0400 ++++ linux-2.6.38.3/kernel/trace/ftrace.c 2011-04-17 15:57:33.000000000 -0400 +@@ -1107,13 +1107,18 @@ ftrace_code_disable(struct module *mod, + + ip = rec->ip; + ++ ret = ftrace_arch_code_modify_prepare(); ++ FTRACE_WARN_ON(ret); ++ if (ret) ++ return 0; ++ + ret = ftrace_make_nop(mod, rec, MCOUNT_ADDR); ++ FTRACE_WARN_ON(ftrace_arch_code_modify_post_process()); + if (ret) { + ftrace_bug(ret, ip); + rec->flags |= FTRACE_FL_FAILED; +- return 0; + } +- return 1; ++ return ret ? 0 : 1; + } + + /* +diff -urNp linux-2.6.38.3/kernel/trace/ring_buffer.c linux-2.6.38.3/kernel/trace/ring_buffer.c +--- linux-2.6.38.3/kernel/trace/ring_buffer.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/trace/ring_buffer.c 2011-04-17 15:57:33.000000000 -0400 +@@ -669,7 +669,7 @@ static struct list_head *rb_list_head(st + * the reader page). But if the next page is a header page, + * its flags will be non zero. + */ +-static int inline ++static inline int + rb_is_head_page(struct ring_buffer_per_cpu *cpu_buffer, + struct buffer_page *page, struct list_head *list) + { +diff -urNp linux-2.6.38.3/kernel/trace/trace.c linux-2.6.38.3/kernel/trace/trace.c +--- linux-2.6.38.3/kernel/trace/trace.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/trace/trace.c 2011-04-17 15:57:33.000000000 -0400 +@@ -3967,10 +3967,9 @@ static const struct file_operations trac + }; + #endif + +-static struct dentry *d_tracer; +- + struct dentry *tracing_init_dentry(void) + { ++ static struct dentry *d_tracer; + static int once; + + if (d_tracer) +@@ -3990,10 +3989,9 @@ struct dentry *tracing_init_dentry(void) + return d_tracer; + } + +-static struct dentry *d_percpu; +- + struct dentry *tracing_dentry_percpu(void) + { ++ static struct dentry *d_percpu; + static int once; + struct dentry *d_tracer; + +diff -urNp linux-2.6.38.3/kernel/trace/trace_events.c linux-2.6.38.3/kernel/trace/trace_events.c +--- linux-2.6.38.3/kernel/trace/trace_events.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/trace/trace_events.c 2011-04-17 15:57:33.000000000 -0400 +@@ -1240,10 +1240,10 @@ static LIST_HEAD(ftrace_module_file_list + struct ftrace_module_file_ops { + struct list_head list; + struct module *mod; +- struct file_operations id; +- struct file_operations enable; +- struct file_operations format; +- struct file_operations filter; ++ struct file_operations id; /* cannot be const, see trace_create_file_ops() */ ++ struct file_operations enable; /* cannot be const, see trace_create_file_ops() */ ++ struct file_operations format; /* cannot be const, see trace_create_file_ops() */ ++ struct file_operations filter; /* cannot be const, see trace_create_file_ops() */ + }; + + static struct ftrace_module_file_ops * +diff -urNp linux-2.6.38.3/kernel/trace/trace_output.c linux-2.6.38.3/kernel/trace/trace_output.c +--- linux-2.6.38.3/kernel/trace/trace_output.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/trace/trace_output.c 2011-04-17 15:57:33.000000000 -0400 +@@ -278,7 +278,7 @@ int trace_seq_path(struct trace_seq *s, + + p = d_path(path, s->buffer + s->len, PAGE_SIZE - s->len); + if (!IS_ERR(p)) { +- p = mangle_path(s->buffer + s->len, p, "\n"); ++ p = mangle_path(s->buffer + s->len, p, "\n\\"); + if (p) { + s->len = p - s->buffer; + return 1; +diff -urNp linux-2.6.38.3/kernel/trace/trace_stack.c linux-2.6.38.3/kernel/trace/trace_stack.c +--- linux-2.6.38.3/kernel/trace/trace_stack.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/trace/trace_stack.c 2011-04-17 15:57:33.000000000 -0400 +@@ -50,7 +50,7 @@ static inline void check_stack(void) + return; + + /* we do not handle interrupt stacks yet */ +- if (!object_is_on_stack(&this_size)) ++ if (!object_starts_on_stack(&this_size)) + return; + + local_irq_save(flags); +diff -urNp linux-2.6.38.3/kernel/trace/trace_workqueue.c linux-2.6.38.3/kernel/trace/trace_workqueue.c +--- linux-2.6.38.3/kernel/trace/trace_workqueue.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/kernel/trace/trace_workqueue.c 2011-04-17 15:57:33.000000000 -0400 +@@ -22,7 +22,7 @@ struct cpu_workqueue_stats { + int cpu; + pid_t pid; + /* Can be inserted from interrupt or user context, need to be atomic */ +- atomic_t inserted; ++ atomic_unchecked_t inserted; + /* + * Don't need to be atomic, works are serialized in a single workqueue thread + * on a single CPU. +@@ -60,7 +60,7 @@ probe_workqueue_insertion(void *ignore, + spin_lock_irqsave(&workqueue_cpu_stat(cpu)->lock, flags); + list_for_each_entry(node, &workqueue_cpu_stat(cpu)->list, list) { + if (node->pid == wq_thread->pid) { +- atomic_inc(&node->inserted); ++ atomic_inc_unchecked(&node->inserted); + goto found; + } + } +@@ -210,7 +210,7 @@ static int workqueue_stat_show(struct se + tsk = get_pid_task(pid, PIDTYPE_PID); + if (tsk) { + seq_printf(s, "%3d %6d %6u %s\n", cws->cpu, +- atomic_read(&cws->inserted), cws->executed, ++ atomic_read_unchecked(&cws->inserted), cws->executed, + tsk->comm); + put_task_struct(tsk); + } +diff -urNp linux-2.6.38.3/lib/bug.c linux-2.6.38.3/lib/bug.c +--- linux-2.6.38.3/lib/bug.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/lib/bug.c 2011-04-17 15:57:33.000000000 -0400 +@@ -133,6 +133,8 @@ enum bug_trap_type report_bug(unsigned l + return BUG_TRAP_TYPE_NONE; + + bug = find_bug(bugaddr); ++ if (!bug) ++ return BUG_TRAP_TYPE_NONE; + + file = NULL; + line = 0; +diff -urNp linux-2.6.38.3/lib/debugobjects.c linux-2.6.38.3/lib/debugobjects.c +--- linux-2.6.38.3/lib/debugobjects.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/lib/debugobjects.c 2011-04-17 15:57:33.000000000 -0400 +@@ -281,7 +281,7 @@ static void debug_object_is_on_stack(voi + if (limit > 4) + return; + +- is_on_stack = object_is_on_stack(addr); ++ is_on_stack = object_starts_on_stack(addr); + if (is_on_stack == onstack) + return; + +diff -urNp linux-2.6.38.3/lib/dma-debug.c linux-2.6.38.3/lib/dma-debug.c +--- linux-2.6.38.3/lib/dma-debug.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/lib/dma-debug.c 2011-04-17 15:57:33.000000000 -0400 +@@ -862,7 +862,7 @@ out: + + static void check_for_stack(struct device *dev, void *addr) + { +- if (object_is_on_stack(addr)) ++ if (object_starts_on_stack(addr)) + err_printk(dev, NULL, "DMA-API: device driver maps memory from" + "stack [addr=%p]\n", addr); + } +diff -urNp linux-2.6.38.3/lib/inflate.c linux-2.6.38.3/lib/inflate.c +--- linux-2.6.38.3/lib/inflate.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/lib/inflate.c 2011-04-17 15:57:33.000000000 -0400 +@@ -269,7 +269,7 @@ static void free(void *where) + malloc_ptr = free_mem_ptr; + } + #else +-#define malloc(a) kmalloc(a, GFP_KERNEL) ++#define malloc(a) kmalloc((a), GFP_KERNEL) + #define free(a) kfree(a) + #endif + +diff -urNp linux-2.6.38.3/lib/Kconfig.debug linux-2.6.38.3/lib/Kconfig.debug +--- linux-2.6.38.3/lib/Kconfig.debug 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/lib/Kconfig.debug 2011-04-17 15:57:33.000000000 -0400 +@@ -1066,6 +1066,7 @@ config LATENCYTOP + depends on DEBUG_KERNEL + depends on STACKTRACE_SUPPORT + depends on PROC_FS ++ depends on !GRKERNSEC_HIDESYM + select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE + select KALLSYMS + select KALLSYMS_ALL +diff -urNp linux-2.6.38.3/lib/kref.c linux-2.6.38.3/lib/kref.c +--- linux-2.6.38.3/lib/kref.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/lib/kref.c 2011-04-17 15:57:33.000000000 -0400 +@@ -52,7 +52,7 @@ void kref_get(struct kref *kref) + */ + int kref_put(struct kref *kref, void (*release)(struct kref *kref)) + { +- WARN_ON(release == NULL); ++ BUG_ON(release == NULL); + WARN_ON(release == (void (*)(struct kref *))kfree); + + if (atomic_dec_and_test(&kref->refcount)) { +diff -urNp linux-2.6.38.3/lib/radix-tree.c linux-2.6.38.3/lib/radix-tree.c +--- linux-2.6.38.3/lib/radix-tree.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/lib/radix-tree.c 2011-04-17 15:57:33.000000000 -0400 +@@ -80,7 +80,7 @@ struct radix_tree_preload { + int nr; + struct radix_tree_node *nodes[RADIX_TREE_MAX_PATH]; + }; +-static DEFINE_PER_CPU(struct radix_tree_preload, radix_tree_preloads) = { 0, }; ++static DEFINE_PER_CPU(struct radix_tree_preload, radix_tree_preloads); + + static inline void *ptr_to_indirect(void *ptr) + { +diff -urNp linux-2.6.38.3/lib/vsprintf.c linux-2.6.38.3/lib/vsprintf.c +--- linux-2.6.38.3/lib/vsprintf.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/lib/vsprintf.c 2011-04-17 15:57:33.000000000 -0400 +@@ -16,6 +16,9 @@ + * - scnprintf and vscnprintf + */ + ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++#define __INCLUDED_BY_HIDESYM 1 ++#endif + #include <stdarg.h> + #include <linux/module.h> + #include <linux/types.h> +@@ -574,7 +577,7 @@ char *symbol_string(char *buf, char *end + unsigned long value = (unsigned long) ptr; + #ifdef CONFIG_KALLSYMS + char sym[KSYM_SYMBOL_LEN]; +- if (ext != 'f' && ext != 's') ++ if (ext != 'f' && ext != 's' && ext != 'a') + sprint_symbol(sym, value); + else + kallsyms_lookup(value, NULL, NULL, NULL, sym); +@@ -936,7 +939,11 @@ char *uuid_string(char *buf, char *end, + return string(buf, end, uuid, spec); + } + ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++int kptr_restrict = 2; ++#else + int kptr_restrict = 1; ++#endif + + /* + * Show a '%p' thing. A kernel extension is that the '%p' is followed +@@ -949,6 +956,8 @@ int kptr_restrict = 1; + * - 'f' For simple symbolic function names without offset + * - 'S' For symbolic direct pointers with offset + * - 's' For symbolic direct pointers without offset ++ * - 'A' For symbolic direct pointers with offset approved for use with GRKERNSEC_HIDESYM ++ * - 'a' For symbolic direct pointers without offset approved for use with GRKERNSEC_HIDESYM + * - 'R' For decoded struct resource, e.g., [mem 0x0-0x1f 64bit pref] + * - 'r' For raw struct resource, e.g., [mem 0x0-0x1f flags 0x201] + * - 'M' For a 6-byte MAC address, it prints the address in the +@@ -993,12 +1002,12 @@ char *pointer(const char *fmt, char *buf + { + if (!ptr) { + /* +- * Print (null) with the same width as a pointer so it makes ++ * Print (nil) with the same width as a pointer so it makes + * tabular output look nice. + */ + if (spec.field_width == -1) + spec.field_width = 2 * sizeof(void *); +- return string(buf, end, "(null)", spec); ++ return string(buf, end, "(nil)", spec); + } + + switch (*fmt) { +@@ -1008,6 +1017,13 @@ char *pointer(const char *fmt, char *buf + /* Fallthrough */ + case 'S': + case 's': ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ break; ++#else ++ return symbol_string(buf, end, ptr, spec, *fmt); ++#endif ++ case 'A': ++ case 'a': + return symbol_string(buf, end, ptr, spec, *fmt); + case 'R': + case 'r': +@@ -1772,11 +1788,11 @@ int bstr_printf(char *buf, size_t size, + typeof(type) value; \ + if (sizeof(type) == 8) { \ + args = PTR_ALIGN(args, sizeof(u32)); \ +- *(u32 *)&value = *(u32 *)args; \ +- *((u32 *)&value + 1) = *(u32 *)(args + 4); \ ++ *(u32 *)&value = *(const u32 *)args; \ ++ *((u32 *)&value + 1) = *(const u32 *)(args + 4); \ + } else { \ + args = PTR_ALIGN(args, sizeof(type)); \ +- value = *(typeof(type) *)args; \ ++ value = *(const typeof(type) *)args; \ + } \ + args += sizeof(type); \ + value; \ +@@ -1839,7 +1855,7 @@ int bstr_printf(char *buf, size_t size, + case FORMAT_TYPE_STR: { + const char *str_arg = args; + args += strlen(str_arg) + 1; +- str = string(str, end, (char *)str_arg, spec); ++ str = string(str, end, str_arg, spec); + break; + } + +diff -urNp linux-2.6.38.3/localversion-grsec linux-2.6.38.3/localversion-grsec +--- linux-2.6.38.3/localversion-grsec 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/localversion-grsec 2011-04-17 15:57:33.000000000 -0400 +@@ -0,0 +1 @@ ++-grsec +diff -urNp linux-2.6.38.3/Makefile linux-2.6.38.3/Makefile +--- linux-2.6.38.3/Makefile 2011-04-17 16:53:01.000000000 -0400 ++++ linux-2.6.38.3/Makefile 2011-04-17 16:53:16.000000000 -0400 +@@ -233,8 +233,8 @@ CONFIG_SHELL := $(shell if [ -x "$$BASH" + + HOSTCC = gcc + HOSTCXX = g++ +-HOSTCFLAGS = -Wall -Wmissing-prototypes -Wstrict-prototypes -O2 -fomit-frame-pointer +-HOSTCXXFLAGS = -O2 ++HOSTCFLAGS = -Wall -W -Wmissing-prototypes -Wstrict-prototypes -O2 -fomit-frame-pointer -fno-delete-null-pointer-checks ++HOSTCXXFLAGS = -O2 -fno-delete-null-pointer-checks + + # Decide whether to build built-in, modular, or both. + # Normally, just do built-in. +@@ -681,7 +681,7 @@ export mod_strip_cmd + + + ifeq ($(KBUILD_EXTMOD),) +-core-y += kernel/ mm/ fs/ ipc/ security/ crypto/ block/ ++core-y += kernel/ mm/ fs/ ipc/ security/ crypto/ block/ grsecurity/ + + vmlinux-dirs := $(patsubst %/,%,$(filter %/, $(init-y) $(init-m) \ + $(core-y) $(core-m) $(drivers-y) $(drivers-m) \ +diff -urNp linux-2.6.38.3/mm/bootmem.c linux-2.6.38.3/mm/bootmem.c +--- linux-2.6.38.3/mm/bootmem.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/bootmem.c 2011-04-17 15:57:33.000000000 -0400 +@@ -201,19 +201,30 @@ static void __init __free_pages_memory(u + unsigned long __init free_all_memory_core_early(int nodeid) + { + int i; +- u64 start, end; ++ u64 start, end, startrange, endrange; + unsigned long count = 0; +- struct range *range = NULL; ++ struct range *range = NULL, rangerange = { 0, 0 }; + int nr_range; + + nr_range = get_free_all_memory_range(&range, nodeid); ++ startrange = __pa(range) >> PAGE_SHIFT; ++ endrange = (__pa(range + nr_range) - 1) >> PAGE_SHIFT; + + for (i = 0; i < nr_range; i++) { + start = range[i].start; + end = range[i].end; ++ if (start <= endrange && startrange < end) { ++ BUG_ON(rangerange.start | rangerange.end); ++ rangerange = range[i]; ++ continue; ++ } + count += end - start; + __free_pages_memory(start, end); + } ++ start = rangerange.start; ++ end = rangerange.end; ++ count += end - start; ++ __free_pages_memory(start, end); + + return count; + } +diff -urNp linux-2.6.38.3/mm/filemap.c linux-2.6.38.3/mm/filemap.c +--- linux-2.6.38.3/mm/filemap.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/filemap.c 2011-04-17 15:57:33.000000000 -0400 +@@ -1664,7 +1664,7 @@ int generic_file_mmap(struct file * file + struct address_space *mapping = file->f_mapping; + + if (!mapping->a_ops->readpage) +- return -ENOEXEC; ++ return -ENODEV; + file_accessed(file); + vma->vm_ops = &generic_file_vm_ops; + vma->vm_flags |= VM_CAN_NONLINEAR; +@@ -2060,6 +2060,7 @@ inline int generic_write_checks(struct f + *pos = i_size_read(inode); + + if (limit != RLIM_INFINITY) { ++ gr_learn_resource(current, RLIMIT_FSIZE,*pos, 0); + if (*pos >= limit) { + send_sig(SIGXFSZ, current, 0); + return -EFBIG; +diff -urNp linux-2.6.38.3/mm/fremap.c linux-2.6.38.3/mm/fremap.c +--- linux-2.6.38.3/mm/fremap.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/fremap.c 2011-04-17 15:57:33.000000000 -0400 +@@ -156,6 +156,11 @@ SYSCALL_DEFINE5(remap_file_pages, unsign + retry: + vma = find_vma(mm, start); + ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (vma && (mm->pax_flags & MF_PAX_SEGMEXEC) && (vma->vm_flags & VM_MAYEXEC)) ++ goto out; ++#endif ++ + /* + * Make sure the vma is shared, that it supports prefaulting, + * and that the remapped range is valid and fully within +@@ -224,7 +229,7 @@ SYSCALL_DEFINE5(remap_file_pages, unsign + /* + * drop PG_Mlocked flag for over-mapped range + */ +- unsigned int saved_flags = vma->vm_flags; ++ unsigned long saved_flags = vma->vm_flags; + munlock_vma_pages_range(vma, start, start + size); + vma->vm_flags = saved_flags; + } +diff -urNp linux-2.6.38.3/mm/highmem.c linux-2.6.38.3/mm/highmem.c +--- linux-2.6.38.3/mm/highmem.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/highmem.c 2011-04-17 15:57:33.000000000 -0400 +@@ -125,9 +125,10 @@ static void flush_all_zero_pkmaps(void) + * So no dangers, even with speculative execution. + */ + page = pte_page(pkmap_page_table[i]); ++ pax_open_kernel(); + pte_clear(&init_mm, (unsigned long)page_address(page), + &pkmap_page_table[i]); +- ++ pax_close_kernel(); + set_page_address(page, NULL); + need_flush = 1; + } +@@ -186,9 +187,11 @@ start: + } + } + vaddr = PKMAP_ADDR(last_pkmap_nr); ++ ++ pax_open_kernel(); + set_pte_at(&init_mm, vaddr, + &(pkmap_page_table[last_pkmap_nr]), mk_pte(page, kmap_prot)); +- ++ pax_close_kernel(); + pkmap_count[last_pkmap_nr] = 1; + set_page_address(page, (void *)vaddr); + +diff -urNp linux-2.6.38.3/mm/hugetlb.c linux-2.6.38.3/mm/hugetlb.c +--- linux-2.6.38.3/mm/hugetlb.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/hugetlb.c 2011-04-17 15:57:33.000000000 -0400 +@@ -2333,6 +2333,27 @@ static int unmap_ref_private(struct mm_s + return 1; + } + ++#ifdef CONFIG_PAX_SEGMEXEC ++static void pax_mirror_huge_pte(struct vm_area_struct *vma, unsigned long address, struct page *page_m) ++{ ++ struct mm_struct *mm = vma->vm_mm; ++ struct vm_area_struct *vma_m; ++ unsigned long address_m; ++ pte_t *ptep_m; ++ ++ vma_m = pax_find_mirror_vma(vma); ++ if (!vma_m) ++ return; ++ ++ BUG_ON(address >= SEGMEXEC_TASK_SIZE); ++ address_m = address + SEGMEXEC_TASK_SIZE; ++ ptep_m = huge_pte_offset(mm, address_m & HPAGE_MASK); ++ get_page(page_m); ++ hugepage_add_anon_rmap(page_m, vma_m, address_m); ++ set_huge_pte_at(mm, address_m, ptep_m, make_huge_pte(vma_m, page_m, 0)); ++} ++#endif ++ + /* + * Hugetlb_cow() should be called with page lock of the original hugepage held. + */ +@@ -2434,6 +2455,11 @@ retry_avoidcopy: + make_huge_pte(vma, new_page, 1)); + page_remove_rmap(old_page); + hugepage_add_new_anon_rmap(new_page, vma, address); ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ pax_mirror_huge_pte(vma, address, new_page); ++#endif ++ + /* Make the old page be freed below */ + new_page = old_page; + mmu_notifier_invalidate_range_end(mm, +@@ -2585,6 +2611,10 @@ retry: + && (vma->vm_flags & VM_SHARED))); + set_huge_pte_at(mm, address, ptep, new_pte); + ++#ifdef CONFIG_PAX_SEGMEXEC ++ pax_mirror_huge_pte(vma, address, page); ++#endif ++ + if ((flags & FAULT_FLAG_WRITE) && !(vma->vm_flags & VM_SHARED)) { + /* Optimization, do the COW without a second fault */ + ret = hugetlb_cow(mm, vma, address, ptep, new_pte, page); +@@ -2614,6 +2644,10 @@ int hugetlb_fault(struct mm_struct *mm, + static DEFINE_MUTEX(hugetlb_instantiation_mutex); + struct hstate *h = hstate_vma(vma); + ++#ifdef CONFIG_PAX_SEGMEXEC ++ struct vm_area_struct *vma_m; ++#endif ++ + ptep = huge_pte_offset(mm, address); + if (ptep) { + entry = huge_ptep_get(ptep); +@@ -2625,6 +2659,26 @@ int hugetlb_fault(struct mm_struct *mm, + VM_FAULT_SET_HINDEX(h - hstates); + } + ++#ifdef CONFIG_PAX_SEGMEXEC ++ vma_m = pax_find_mirror_vma(vma); ++ if (vma_m) { ++ unsigned long address_m; ++ ++ if (vma->vm_start > vma_m->vm_start) { ++ address_m = address; ++ address -= SEGMEXEC_TASK_SIZE; ++ vma = vma_m; ++ h = hstate_vma(vma); ++ } else ++ address_m = address + SEGMEXEC_TASK_SIZE; ++ ++ if (!huge_pte_alloc(mm, address_m, huge_page_size(h))) ++ return VM_FAULT_OOM; ++ address_m &= HPAGE_MASK; ++ unmap_hugepage_range(vma, address_m, address_m + HPAGE_SIZE, NULL); ++ } ++#endif ++ + ptep = huge_pte_alloc(mm, address, huge_page_size(h)); + if (!ptep) + return VM_FAULT_OOM; +diff -urNp linux-2.6.38.3/mm/Kconfig linux-2.6.38.3/mm/Kconfig +--- linux-2.6.38.3/mm/Kconfig 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/Kconfig 2011-04-17 15:57:33.000000000 -0400 +@@ -240,7 +240,7 @@ config KSM + config DEFAULT_MMAP_MIN_ADDR + int "Low address space to protect from user allocation" + depends on MMU +- default 4096 ++ default 65536 + help + This is the portion of low virtual memory which should be protected + from userspace allocation. Keeping a user from writing to low pages +diff -urNp linux-2.6.38.3/mm/kmemleak.c linux-2.6.38.3/mm/kmemleak.c +--- linux-2.6.38.3/mm/kmemleak.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/kmemleak.c 2011-04-17 15:57:33.000000000 -0400 +@@ -357,7 +357,7 @@ static void print_unreferenced(struct se + + for (i = 0; i < object->trace_len; i++) { + void *ptr = (void *)object->trace[i]; +- seq_printf(seq, " [<%p>] %pS\n", ptr, ptr); ++ seq_printf(seq, " [<%p>] %pA\n", ptr, ptr); + } + } + +diff -urNp linux-2.6.38.3/mm/maccess.c linux-2.6.38.3/mm/maccess.c +--- linux-2.6.38.3/mm/maccess.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/maccess.c 2011-04-17 15:57:33.000000000 -0400 +@@ -15,10 +15,10 @@ + * happens, handle that and return -EFAULT. + */ + +-long __weak probe_kernel_read(void *dst, void *src, size_t size) ++long __weak probe_kernel_read(void *dst, const void *src, size_t size) + __attribute__((alias("__probe_kernel_read"))); + +-long __probe_kernel_read(void *dst, void *src, size_t size) ++long __probe_kernel_read(void *dst, const void *src, size_t size) + { + long ret; + mm_segment_t old_fs = get_fs(); +@@ -43,10 +43,10 @@ EXPORT_SYMBOL_GPL(probe_kernel_read); + * Safely write to address @dst from the buffer at @src. If a kernel fault + * happens, handle that and return -EFAULT. + */ +-long __weak probe_kernel_write(void *dst, void *src, size_t size) ++long __weak probe_kernel_write(void *dst, const void *src, size_t size) + __attribute__((alias("__probe_kernel_write"))); + +-long __probe_kernel_write(void *dst, void *src, size_t size) ++long __probe_kernel_write(void *dst, const void *src, size_t size) + { + long ret; + mm_segment_t old_fs = get_fs(); +diff -urNp linux-2.6.38.3/mm/madvise.c linux-2.6.38.3/mm/madvise.c +--- linux-2.6.38.3/mm/madvise.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/madvise.c 2011-04-17 15:57:33.000000000 -0400 +@@ -45,6 +45,10 @@ static long madvise_behavior(struct vm_a + pgoff_t pgoff; + unsigned long new_flags = vma->vm_flags; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ struct vm_area_struct *vma_m; ++#endif ++ + switch (behavior) { + case MADV_NORMAL: + new_flags = new_flags & ~VM_RAND_READ & ~VM_SEQ_READ; +@@ -110,6 +114,13 @@ success: + /* + * vm_flags is protected by the mmap_sem held in write mode. + */ ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ vma_m = pax_find_mirror_vma(vma); ++ if (vma_m) ++ vma_m->vm_flags = new_flags & ~(VM_WRITE | VM_MAYWRITE | VM_ACCOUNT); ++#endif ++ + vma->vm_flags = new_flags; + + out: +@@ -168,6 +179,11 @@ static long madvise_dontneed(struct vm_a + struct vm_area_struct ** prev, + unsigned long start, unsigned long end) + { ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ struct vm_area_struct *vma_m; ++#endif ++ + *prev = vma; + if (vma->vm_flags & (VM_LOCKED|VM_HUGETLB|VM_PFNMAP)) + return -EINVAL; +@@ -180,6 +196,21 @@ static long madvise_dontneed(struct vm_a + zap_page_range(vma, start, end - start, &details); + } else + zap_page_range(vma, start, end - start, NULL); ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ vma_m = pax_find_mirror_vma(vma); ++ if (vma_m) { ++ if (unlikely(vma->vm_flags & VM_NONLINEAR)) { ++ struct zap_details details = { ++ .nonlinear_vma = vma_m, ++ .last_index = ULONG_MAX, ++ }; ++ zap_page_range(vma, start + SEGMEXEC_TASK_SIZE, end - start, &details); ++ } else ++ zap_page_range(vma, start + SEGMEXEC_TASK_SIZE, end - start, NULL); ++ } ++#endif ++ + return 0; + } + +@@ -376,6 +407,16 @@ SYSCALL_DEFINE3(madvise, unsigned long, + if (end < start) + goto out; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (current->mm->pax_flags & MF_PAX_SEGMEXEC) { ++ if (end > SEGMEXEC_TASK_SIZE) ++ goto out; ++ } else ++#endif ++ ++ if (end > TASK_SIZE) ++ goto out; ++ + error = 0; + if (end == start) + goto out; +diff -urNp linux-2.6.38.3/mm/memory.c linux-2.6.38.3/mm/memory.c +--- linux-2.6.38.3/mm/memory.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/memory.c 2011-04-17 15:57:33.000000000 -0400 +@@ -259,8 +259,12 @@ static inline void free_pmd_range(struct + return; + + pmd = pmd_offset(pud, start); ++ ++#if !defined(CONFIG_X86_32) || !defined(CONFIG_PAX_PER_CPU_PGD) + pud_clear(pud); + pmd_free_tlb(tlb, pmd, start); ++#endif ++ + } + + static inline void free_pud_range(struct mmu_gather *tlb, pgd_t *pgd, +@@ -291,9 +295,12 @@ static inline void free_pud_range(struct + if (end - 1 > ceiling - 1) + return; + ++#if !defined(CONFIG_X86_64) || !defined(CONFIG_PAX_PER_CPU_PGD) + pud = pud_offset(pgd, start); + pgd_clear(pgd); + pud_free_tlb(tlb, pud, start); ++#endif ++ + } + + /* +@@ -1433,10 +1440,10 @@ int __get_user_pages(struct task_struct + (VM_MAYREAD | VM_MAYWRITE) : (VM_READ | VM_WRITE); + i = 0; + +- do { ++ while (nr_pages) { + struct vm_area_struct *vma; + +- vma = find_extend_vma(mm, start); ++ vma = find_vma(mm, start); + if (!vma && in_gate_area(tsk, start)) { + unsigned long pg = start & PAGE_MASK; + struct vm_area_struct *gate_vma = get_gate_vma(tsk); +@@ -1489,7 +1496,7 @@ int __get_user_pages(struct task_struct + continue; + } + +- if (!vma || ++ if (!vma || start < vma->vm_start || + (vma->vm_flags & (VM_IO | VM_PFNMAP)) || + !(vm_flags & vma->vm_flags)) + return i ? : -EFAULT; +@@ -1575,7 +1582,7 @@ int __get_user_pages(struct task_struct + start += PAGE_SIZE; + nr_pages--; + } while (nr_pages && start < vma->vm_end); +- } while (nr_pages); ++ } + return i; + } + +@@ -1724,6 +1731,10 @@ static int insert_page(struct vm_area_st + page_add_file_rmap(page); + set_pte_at(mm, addr, pte, mk_pte(page, prot)); + ++#ifdef CONFIG_PAX_SEGMEXEC ++ pax_mirror_file_pte(vma, addr, page, ptl); ++#endif ++ + retval = 0; + pte_unmap_unlock(pte, ptl); + return retval; +@@ -1758,10 +1769,22 @@ out: + int vm_insert_page(struct vm_area_struct *vma, unsigned long addr, + struct page *page) + { ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ struct vm_area_struct *vma_m; ++#endif ++ + if (addr < vma->vm_start || addr >= vma->vm_end) + return -EFAULT; + if (!page_count(page)) + return -EINVAL; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ vma_m = pax_find_mirror_vma(vma); ++ if (vma_m) ++ vma_m->vm_flags |= VM_INSERTPAGE; ++#endif ++ + vma->vm_flags |= VM_INSERTPAGE; + return insert_page(vma, addr, page, vma->vm_page_prot); + } +@@ -1847,6 +1870,7 @@ int vm_insert_mixed(struct vm_area_struc + unsigned long pfn) + { + BUG_ON(!(vma->vm_flags & VM_MIXEDMAP)); ++ BUG_ON(vma->vm_mirror); + + if (addr < vma->vm_start || addr >= vma->vm_end) + return -EFAULT; +@@ -2162,6 +2186,186 @@ static inline void cow_user_page(struct + copy_user_highpage(dst, src, va, vma); + } + ++#ifdef CONFIG_PAX_SEGMEXEC ++static void pax_unmap_mirror_pte(struct vm_area_struct *vma, unsigned long address, pmd_t *pmd) ++{ ++ struct mm_struct *mm = vma->vm_mm; ++ spinlock_t *ptl; ++ pte_t *pte, entry; ++ ++ pte = pte_offset_map_lock(mm, pmd, address, &ptl); ++ entry = *pte; ++ if (!pte_present(entry)) { ++ if (!pte_none(entry)) { ++ BUG_ON(pte_file(entry)); ++ free_swap_and_cache(pte_to_swp_entry(entry)); ++ pte_clear_not_present_full(mm, address, pte, 0); ++ } ++ } else { ++ struct page *page; ++ ++ flush_cache_page(vma, address, pte_pfn(entry)); ++ entry = ptep_clear_flush(vma, address, pte); ++ BUG_ON(pte_dirty(entry)); ++ page = vm_normal_page(vma, address, entry); ++ if (page) { ++ update_hiwater_rss(mm); ++ if (PageAnon(page)) ++ dec_mm_counter_fast(mm, MM_ANONPAGES); ++ else ++ dec_mm_counter_fast(mm, MM_FILEPAGES); ++ page_remove_rmap(page); ++ page_cache_release(page); ++ } ++ } ++ pte_unmap_unlock(pte, ptl); ++} ++ ++/* PaX: if vma is mirrored, synchronize the mirror's PTE ++ * ++ * the ptl of the lower mapped page is held on entry and is not released on exit ++ * or inside to ensure atomic changes to the PTE states (swapout, mremap, munmap, etc) ++ */ ++static void pax_mirror_anon_pte(struct vm_area_struct *vma, unsigned long address, struct page *page_m, spinlock_t *ptl) ++{ ++ struct mm_struct *mm = vma->vm_mm; ++ unsigned long address_m; ++ spinlock_t *ptl_m; ++ struct vm_area_struct *vma_m; ++ pmd_t *pmd_m; ++ pte_t *pte_m, entry_m; ++ ++ BUG_ON(!page_m || !PageAnon(page_m)); ++ ++ vma_m = pax_find_mirror_vma(vma); ++ if (!vma_m) ++ return; ++ ++ BUG_ON(!PageLocked(page_m)); ++ BUG_ON(address >= SEGMEXEC_TASK_SIZE); ++ address_m = address + SEGMEXEC_TASK_SIZE; ++ pmd_m = pmd_offset(pud_offset(pgd_offset(mm, address_m), address_m), address_m); ++ pte_m = pte_offset_map(pmd_m, address_m); ++ ptl_m = pte_lockptr(mm, pmd_m); ++ if (ptl != ptl_m) { ++ spin_lock_nested(ptl_m, SINGLE_DEPTH_NESTING); ++ if (!pte_none(*pte_m)) ++ goto out; ++ } ++ ++ entry_m = pfn_pte(page_to_pfn(page_m), vma_m->vm_page_prot); ++ page_cache_get(page_m); ++ page_add_anon_rmap(page_m, vma_m, address_m); ++ inc_mm_counter_fast(mm, MM_ANONPAGES); ++ set_pte_at(mm, address_m, pte_m, entry_m); ++ update_mmu_cache(vma_m, address_m, entry_m); ++out: ++ if (ptl != ptl_m) ++ spin_unlock(ptl_m); ++ pte_unmap(pte_m); ++ unlock_page(page_m); ++} ++ ++void pax_mirror_file_pte(struct vm_area_struct *vma, unsigned long address, struct page *page_m, spinlock_t *ptl) ++{ ++ struct mm_struct *mm = vma->vm_mm; ++ unsigned long address_m; ++ spinlock_t *ptl_m; ++ struct vm_area_struct *vma_m; ++ pmd_t *pmd_m; ++ pte_t *pte_m, entry_m; ++ ++ BUG_ON(!page_m || PageAnon(page_m)); ++ ++ vma_m = pax_find_mirror_vma(vma); ++ if (!vma_m) ++ return; ++ ++ BUG_ON(address >= SEGMEXEC_TASK_SIZE); ++ address_m = address + SEGMEXEC_TASK_SIZE; ++ pmd_m = pmd_offset(pud_offset(pgd_offset(mm, address_m), address_m), address_m); ++ pte_m = pte_offset_map(pmd_m, address_m); ++ ptl_m = pte_lockptr(mm, pmd_m); ++ if (ptl != ptl_m) { ++ spin_lock_nested(ptl_m, SINGLE_DEPTH_NESTING); ++ if (!pte_none(*pte_m)) ++ goto out; ++ } ++ ++ entry_m = pfn_pte(page_to_pfn(page_m), vma_m->vm_page_prot); ++ page_cache_get(page_m); ++ page_add_file_rmap(page_m); ++ inc_mm_counter_fast(mm, MM_FILEPAGES); ++ set_pte_at(mm, address_m, pte_m, entry_m); ++ update_mmu_cache(vma_m, address_m, entry_m); ++out: ++ if (ptl != ptl_m) ++ spin_unlock(ptl_m); ++ pte_unmap(pte_m); ++} ++ ++static void pax_mirror_pfn_pte(struct vm_area_struct *vma, unsigned long address, unsigned long pfn_m, spinlock_t *ptl) ++{ ++ struct mm_struct *mm = vma->vm_mm; ++ unsigned long address_m; ++ spinlock_t *ptl_m; ++ struct vm_area_struct *vma_m; ++ pmd_t *pmd_m; ++ pte_t *pte_m, entry_m; ++ ++ vma_m = pax_find_mirror_vma(vma); ++ if (!vma_m) ++ return; ++ ++ BUG_ON(address >= SEGMEXEC_TASK_SIZE); ++ address_m = address + SEGMEXEC_TASK_SIZE; ++ pmd_m = pmd_offset(pud_offset(pgd_offset(mm, address_m), address_m), address_m); ++ pte_m = pte_offset_map(pmd_m, address_m); ++ ptl_m = pte_lockptr(mm, pmd_m); ++ if (ptl != ptl_m) { ++ spin_lock_nested(ptl_m, SINGLE_DEPTH_NESTING); ++ if (!pte_none(*pte_m)) ++ goto out; ++ } ++ ++ entry_m = pfn_pte(pfn_m, vma_m->vm_page_prot); ++ set_pte_at(mm, address_m, pte_m, entry_m); ++out: ++ if (ptl != ptl_m) ++ spin_unlock(ptl_m); ++ pte_unmap(pte_m); ++} ++ ++static void pax_mirror_pte(struct vm_area_struct *vma, unsigned long address, pte_t *pte, pmd_t *pmd, spinlock_t *ptl) ++{ ++ struct page *page_m; ++ pte_t entry; ++ ++ if (!(vma->vm_mm->pax_flags & MF_PAX_SEGMEXEC)) ++ goto out; ++ ++ entry = *pte; ++ page_m = vm_normal_page(vma, address, entry); ++ if (!page_m) ++ pax_mirror_pfn_pte(vma, address, pte_pfn(entry), ptl); ++ else if (PageAnon(page_m)) { ++ if (pax_find_mirror_vma(vma)) { ++ pte_unmap_unlock(pte, ptl); ++ lock_page(page_m); ++ pte = pte_offset_map_lock(vma->vm_mm, pmd, address, &ptl); ++ if (pte_same(entry, *pte)) ++ pax_mirror_anon_pte(vma, address, page_m, ptl); ++ else ++ unlock_page(page_m); ++ } ++ } else ++ pax_mirror_file_pte(vma, address, page_m, ptl); ++ ++out: ++ pte_unmap_unlock(pte, ptl); ++} ++#endif ++ + /* + * This routine handles present pages, when users try to write + * to a shared page. It is done by copying the page to a new address +@@ -2373,6 +2577,12 @@ gotten: + */ + page_table = pte_offset_map_lock(mm, pmd, address, &ptl); + if (likely(pte_same(*page_table, orig_pte))) { ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (pax_find_mirror_vma(vma)) ++ BUG_ON(!trylock_page(new_page)); ++#endif ++ + if (old_page) { + if (!PageAnon(old_page)) { + dec_mm_counter_fast(mm, MM_FILEPAGES); +@@ -2424,6 +2634,10 @@ gotten: + page_remove_rmap(old_page); + } + ++#ifdef CONFIG_PAX_SEGMEXEC ++ pax_mirror_anon_pte(vma, address, new_page, ptl); ++#endif ++ + /* Free the old page.. */ + new_page = old_page; + ret |= VM_FAULT_WRITE; +@@ -2834,6 +3048,11 @@ static int do_swap_page(struct mm_struct + swap_free(entry); + if (vm_swap_full() || (vma->vm_flags & VM_LOCKED) || PageMlocked(page)) + try_to_free_swap(page); ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if ((flags & FAULT_FLAG_WRITE) || !pax_find_mirror_vma(vma)) ++#endif ++ + unlock_page(page); + if (swapcache) { + /* +@@ -2857,6 +3076,11 @@ static int do_swap_page(struct mm_struct + + /* No need to invalidate - it was non-present before */ + update_mmu_cache(vma, address, page_table); ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ pax_mirror_anon_pte(vma, address, page, ptl); ++#endif ++ + unlock: + pte_unmap_unlock(page_table, ptl); + out: +@@ -2876,40 +3100,6 @@ out_release: + } + + /* +- * This is like a special single-page "expand_{down|up}wards()", +- * except we must first make sure that 'address{-|+}PAGE_SIZE' +- * doesn't hit another vma. +- */ +-static inline int check_stack_guard_page(struct vm_area_struct *vma, unsigned long address) +-{ +- address &= PAGE_MASK; +- if ((vma->vm_flags & VM_GROWSDOWN) && address == vma->vm_start) { +- struct vm_area_struct *prev = vma->vm_prev; +- +- /* +- * Is there a mapping abutting this one below? +- * +- * That's only ok if it's the same stack mapping +- * that has gotten split.. +- */ +- if (prev && prev->vm_end == address) +- return prev->vm_flags & VM_GROWSDOWN ? 0 : -ENOMEM; +- +- expand_stack(vma, address - PAGE_SIZE); +- } +- if ((vma->vm_flags & VM_GROWSUP) && address + PAGE_SIZE == vma->vm_end) { +- struct vm_area_struct *next = vma->vm_next; +- +- /* As VM_GROWSDOWN but s/below/above/ */ +- if (next && next->vm_start == address + PAGE_SIZE) +- return next->vm_flags & VM_GROWSUP ? 0 : -ENOMEM; +- +- expand_upwards(vma, address + PAGE_SIZE); +- } +- return 0; +-} +- +-/* + * We enter with non-exclusive mmap_sem (to exclude vma changes, + * but allow concurrent faults), and pte mapped but not yet locked. + * We return with mmap_sem still held, but pte unmapped and unlocked. +@@ -2918,27 +3108,23 @@ static int do_anonymous_page(struct mm_s + unsigned long address, pte_t *page_table, pmd_t *pmd, + unsigned int flags) + { +- struct page *page; ++ struct page *page = NULL; + spinlock_t *ptl; + pte_t entry; + +- pte_unmap(page_table); +- +- /* Check if we need to add a guard page to the stack */ +- if (check_stack_guard_page(vma, address) < 0) +- return VM_FAULT_SIGBUS; +- +- /* Use the zero-page for reads */ + if (!(flags & FAULT_FLAG_WRITE)) { + entry = pte_mkspecial(pfn_pte(my_zero_pfn(address), + vma->vm_page_prot)); +- page_table = pte_offset_map_lock(mm, pmd, address, &ptl); ++ ptl = pte_lockptr(mm, pmd); ++ spin_lock(ptl); + if (!pte_none(*page_table)) + goto unlock; + goto setpte; + } + + /* Allocate our own private page. */ ++ pte_unmap(page_table); ++ + if (unlikely(anon_vma_prepare(vma))) + goto oom; + page = alloc_zeroed_user_highpage_movable(vma, address); +@@ -2957,6 +3143,11 @@ static int do_anonymous_page(struct mm_s + if (!pte_none(*page_table)) + goto release; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (pax_find_mirror_vma(vma)) ++ BUG_ON(!trylock_page(page)); ++#endif ++ + inc_mm_counter_fast(mm, MM_ANONPAGES); + page_add_new_anon_rmap(page, vma, address); + setpte: +@@ -2964,6 +3155,12 @@ setpte: + + /* No need to invalidate - it was non-present before */ + update_mmu_cache(vma, address, page_table); ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (page) ++ pax_mirror_anon_pte(vma, address, page, ptl); ++#endif ++ + unlock: + pte_unmap_unlock(page_table, ptl); + return 0; +@@ -3101,6 +3298,12 @@ static int __do_fault(struct mm_struct * + */ + /* Only go through if we didn't race with anybody else... */ + if (likely(pte_same(*page_table, orig_pte))) { ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (anon && pax_find_mirror_vma(vma)) ++ BUG_ON(!trylock_page(page)); ++#endif ++ + flush_icache_page(vma, page); + entry = mk_pte(page, vma->vm_page_prot); + if (flags & FAULT_FLAG_WRITE) +@@ -3120,6 +3323,14 @@ static int __do_fault(struct mm_struct * + + /* no need to invalidate: a not-present page won't be cached */ + update_mmu_cache(vma, address, page_table); ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (anon) ++ pax_mirror_anon_pte(vma, address, page, ptl); ++ else ++ pax_mirror_file_pte(vma, address, page, ptl); ++#endif ++ + } else { + if (charged) + mem_cgroup_uncharge_page(page); +@@ -3267,6 +3478,12 @@ int handle_pte_fault(struct mm_struct *m + if (flags & FAULT_FLAG_WRITE) + flush_tlb_fix_spurious_fault(vma, address); + } ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ pax_mirror_pte(vma, address, pte, pmd, ptl); ++ return 0; ++#endif ++ + unlock: + pte_unmap_unlock(pte, ptl); + return 0; +@@ -3283,6 +3500,10 @@ int handle_mm_fault(struct mm_struct *mm + pmd_t *pmd; + pte_t *pte; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ struct vm_area_struct *vma_m; ++#endif ++ + __set_current_state(TASK_RUNNING); + + count_vm_event(PGFAULT); +@@ -3293,6 +3514,34 @@ int handle_mm_fault(struct mm_struct *mm + if (unlikely(is_vm_hugetlb_page(vma))) + return hugetlb_fault(mm, vma, address, flags); + ++#ifdef CONFIG_PAX_SEGMEXEC ++ vma_m = pax_find_mirror_vma(vma); ++ if (vma_m) { ++ unsigned long address_m; ++ pgd_t *pgd_m; ++ pud_t *pud_m; ++ pmd_t *pmd_m; ++ ++ if (vma->vm_start > vma_m->vm_start) { ++ address_m = address; ++ address -= SEGMEXEC_TASK_SIZE; ++ vma = vma_m; ++ } else ++ address_m = address + SEGMEXEC_TASK_SIZE; ++ ++ pgd_m = pgd_offset(mm, address_m); ++ pud_m = pud_alloc(mm, pgd_m, address_m); ++ if (!pud_m) ++ return VM_FAULT_OOM; ++ pmd_m = pmd_alloc(mm, pud_m, address_m); ++ if (!pmd_m) ++ return VM_FAULT_OOM; ++ if (!pmd_present(*pmd_m) && __pte_alloc(mm, vma_m, pmd_m, address_m)) ++ return VM_FAULT_OOM; ++ pax_unmap_mirror_pte(vma_m, address_m, pmd_m); ++ } ++#endif ++ + pgd = pgd_offset(mm, address); + pud = pud_alloc(mm, pgd, address); + if (!pud) +@@ -3426,7 +3675,7 @@ static int __init gate_vma_init(void) + gate_vma.vm_start = FIXADDR_USER_START; + gate_vma.vm_end = FIXADDR_USER_END; + gate_vma.vm_flags = VM_READ | VM_MAYREAD | VM_EXEC | VM_MAYEXEC; +- gate_vma.vm_page_prot = __P101; ++ gate_vma.vm_page_prot = vm_get_page_prot(gate_vma.vm_flags); + /* + * Make sure the vDSO gets into every core dump. + * Dumping its contents makes post-mortem fully interpretable later +diff -urNp linux-2.6.38.3/mm/memory-failure.c linux-2.6.38.3/mm/memory-failure.c +--- linux-2.6.38.3/mm/memory-failure.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/memory-failure.c 2011-04-17 15:57:33.000000000 -0400 +@@ -58,7 +58,7 @@ int sysctl_memory_failure_early_kill __r + + int sysctl_memory_failure_recovery __read_mostly = 1; + +-atomic_long_t mce_bad_pages __read_mostly = ATOMIC_LONG_INIT(0); ++atomic_long_unchecked_t mce_bad_pages __read_mostly = ATOMIC_LONG_INIT(0); + + #if defined(CONFIG_HWPOISON_INJECT) || defined(CONFIG_HWPOISON_INJECT_MODULE) + +@@ -1012,7 +1012,7 @@ int __memory_failure(unsigned long pfn, + } + + nr_pages = 1 << compound_trans_order(hpage); +- atomic_long_add(nr_pages, &mce_bad_pages); ++ atomic_long_add_unchecked(nr_pages, &mce_bad_pages); + + /* + * We need/can do nothing about count=0 pages. +@@ -1042,7 +1042,7 @@ int __memory_failure(unsigned long pfn, + if (!PageHWPoison(hpage) + || (hwpoison_filter(p) && TestClearPageHWPoison(p)) + || (p != hpage && TestSetPageHWPoison(hpage))) { +- atomic_long_sub(nr_pages, &mce_bad_pages); ++ atomic_long_sub_unchecked(nr_pages, &mce_bad_pages); + return 0; + } + set_page_hwpoison_huge_page(hpage); +@@ -1100,7 +1100,7 @@ int __memory_failure(unsigned long pfn, + } + if (hwpoison_filter(p)) { + if (TestClearPageHWPoison(p)) +- atomic_long_sub(nr_pages, &mce_bad_pages); ++ atomic_long_sub_unchecked(nr_pages, &mce_bad_pages); + unlock_page(hpage); + put_page(hpage); + return 0; +@@ -1226,7 +1226,7 @@ int unpoison_memory(unsigned long pfn) + return 0; + } + if (TestClearPageHWPoison(p)) +- atomic_long_sub(nr_pages, &mce_bad_pages); ++ atomic_long_sub_unchecked(nr_pages, &mce_bad_pages); + pr_info("MCE: Software-unpoisoned free page %#lx\n", pfn); + return 0; + } +@@ -1240,7 +1240,7 @@ int unpoison_memory(unsigned long pfn) + */ + if (TestClearPageHWPoison(page)) { + pr_info("MCE: Software-unpoisoned page %#lx\n", pfn); +- atomic_long_sub(nr_pages, &mce_bad_pages); ++ atomic_long_sub_unchecked(nr_pages, &mce_bad_pages); + freeit = 1; + if (PageHuge(page)) + clear_page_hwpoison_huge_page(page); +@@ -1353,7 +1353,7 @@ static int soft_offline_huge_page(struct + } + done: + if (!PageHWPoison(hpage)) +- atomic_long_add(1 << compound_trans_order(hpage), &mce_bad_pages); ++ atomic_long_add_unchecked(1 << compound_trans_order(hpage), &mce_bad_pages); + set_page_hwpoison_huge_page(hpage); + dequeue_hwpoisoned_huge_page(hpage); + /* keep elevated page count for bad page */ +@@ -1482,7 +1482,7 @@ int soft_offline_page(struct page *page, + return ret; + + done: +- atomic_long_add(1, &mce_bad_pages); ++ atomic_long_add_unchecked(1, &mce_bad_pages); + SetPageHWPoison(page); + /* keep elevated page count for bad page */ + return ret; +diff -urNp linux-2.6.38.3/mm/mempolicy.c linux-2.6.38.3/mm/mempolicy.c +--- linux-2.6.38.3/mm/mempolicy.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/mempolicy.c 2011-04-17 15:57:33.000000000 -0400 +@@ -643,6 +643,10 @@ static int mbind_range(struct mm_struct + unsigned long vmstart; + unsigned long vmend; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ struct vm_area_struct *vma_m; ++#endif ++ + vma = find_vma_prev(mm, start, &prev); + if (!vma || vma->vm_start > start) + return -EFAULT; +@@ -673,6 +677,16 @@ static int mbind_range(struct mm_struct + err = policy_vma(vma, new_pol); + if (err) + goto out; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ vma_m = pax_find_mirror_vma(vma); ++ if (vma_m) { ++ err = policy_vma(vma_m, new_pol); ++ if (err) ++ goto out; ++ } ++#endif ++ + } + + out: +@@ -1106,6 +1120,17 @@ static long do_mbind(unsigned long start + + if (end < start) + return -EINVAL; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (mm->pax_flags & MF_PAX_SEGMEXEC) { ++ if (end > SEGMEXEC_TASK_SIZE) ++ return -EINVAL; ++ } else ++#endif ++ ++ if (end > TASK_SIZE) ++ return -EINVAL; ++ + if (end == start) + return 0; + +@@ -1324,6 +1349,14 @@ SYSCALL_DEFINE4(migrate_pages, pid_t, pi + if (!mm) + goto out; + ++#ifdef CONFIG_GRKERNSEC_PROC_MEMMAP ++ if (mm != current->mm && ++ (mm->pax_flags & MF_PAX_RANDMMAP || mm->pax_flags & MF_PAX_SEGMEXEC)) { ++ err = -EPERM; ++ goto out; ++ } ++#endif ++ + /* + * Check if this process has the right to modify the specified + * process. The right exists if the process has administrative +@@ -1333,8 +1366,7 @@ SYSCALL_DEFINE4(migrate_pages, pid_t, pi + rcu_read_lock(); + tcred = __task_cred(task); + if (cred->euid != tcred->suid && cred->euid != tcred->uid && +- cred->uid != tcred->suid && cred->uid != tcred->uid && +- !capable(CAP_SYS_NICE)) { ++ cred->uid != tcred->suid && !capable(CAP_SYS_NICE)) { + rcu_read_unlock(); + err = -EPERM; + goto out; +@@ -2635,7 +2667,7 @@ int show_numa_map(struct seq_file *m, vo + + if (file) { + seq_printf(m, " file="); +- seq_path(m, &file->f_path, "\n\t= "); ++ seq_path(m, &file->f_path, "\n\t\\= "); + } else if (vma->vm_start <= mm->brk && vma->vm_end >= mm->start_brk) { + seq_printf(m, " heap"); + } else if (vma->vm_start <= mm->start_stack && +diff -urNp linux-2.6.38.3/mm/migrate.c linux-2.6.38.3/mm/migrate.c +--- linux-2.6.38.3/mm/migrate.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/migrate.c 2011-04-17 15:57:33.000000000 -0400 +@@ -1299,6 +1299,14 @@ SYSCALL_DEFINE6(move_pages, pid_t, pid, + if (!mm) + return -EINVAL; + ++#ifdef CONFIG_GRKERNSEC_PROC_MEMMAP ++ if (mm != current->mm && ++ (mm->pax_flags & MF_PAX_RANDMMAP || mm->pax_flags & MF_PAX_SEGMEXEC)) { ++ err = -EPERM; ++ goto out; ++ } ++#endif ++ + /* + * Check if this process has the right to modify the specified + * process. The right exists if the process has administrative +@@ -1308,8 +1316,7 @@ SYSCALL_DEFINE6(move_pages, pid_t, pid, + rcu_read_lock(); + tcred = __task_cred(task); + if (cred->euid != tcred->suid && cred->euid != tcred->uid && +- cred->uid != tcred->suid && cred->uid != tcred->uid && +- !capable(CAP_SYS_NICE)) { ++ cred->uid != tcred->suid && !capable(CAP_SYS_NICE)) { + rcu_read_unlock(); + err = -EPERM; + goto out; +diff -urNp linux-2.6.38.3/mm/mlock.c linux-2.6.38.3/mm/mlock.c +--- linux-2.6.38.3/mm/mlock.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/mlock.c 2011-04-17 15:57:33.000000000 -0400 +@@ -13,6 +13,7 @@ + #include <linux/pagemap.h> + #include <linux/mempolicy.h> + #include <linux/syscalls.h> ++#include <linux/security.h> + #include <linux/sched.h> + #include <linux/module.h> + #include <linux/rmap.h> +@@ -135,13 +136,6 @@ void munlock_vma_page(struct page *page) + } + } + +-static inline int stack_guard_page(struct vm_area_struct *vma, unsigned long addr) +-{ +- return (vma->vm_flags & VM_GROWSDOWN) && +- (vma->vm_start == addr) && +- !vma_stack_continue(vma->vm_prev, addr); +-} +- + /** + * __mlock_vma_pages_range() - mlock a range of pages in the vma. + * @vma: target vma +@@ -188,12 +182,6 @@ static long __mlock_vma_pages_range(stru + if (vma->vm_flags & VM_LOCKED) + gup_flags |= FOLL_MLOCK; + +- /* We don't try to access the guard page of a stack vma */ +- if (stack_guard_page(vma, start)) { +- addr += PAGE_SIZE; +- nr_pages--; +- } +- + return __get_user_pages(current, mm, addr, nr_pages, gup_flags, + NULL, NULL, nonblocking); + } +@@ -393,6 +381,9 @@ static int do_mlock(unsigned long start, + return -EINVAL; + if (end == start) + return 0; ++ if (end > TASK_SIZE) ++ return -EINVAL; ++ + vma = find_vma_prev(current->mm, start, &prev); + if (!vma || vma->vm_start > start) + return -ENOMEM; +@@ -403,6 +394,11 @@ static int do_mlock(unsigned long start, + for (nstart = start ; ; ) { + unsigned int newflags; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ if ((current->mm->pax_flags & MF_PAX_SEGMEXEC) && (vma->vm_start >= SEGMEXEC_TASK_SIZE)) ++ break; ++#endif ++ + /* Here we know that vma->vm_start <= nstart < vma->vm_end. */ + + newflags = vma->vm_flags | VM_LOCKED; +@@ -508,6 +504,7 @@ SYSCALL_DEFINE2(mlock, unsigned long, st + lock_limit >>= PAGE_SHIFT; + + /* check against resource limits */ ++ gr_learn_resource(current, RLIMIT_MEMLOCK, (current->mm->locked_vm << PAGE_SHIFT) + len, 1); + if ((locked <= lock_limit) || capable(CAP_IPC_LOCK)) + error = do_mlock(start, len, 1); + up_write(¤t->mm->mmap_sem); +@@ -531,17 +528,23 @@ SYSCALL_DEFINE2(munlock, unsigned long, + static int do_mlockall(int flags) + { + struct vm_area_struct * vma, * prev = NULL; +- unsigned int def_flags = 0; + + if (flags & MCL_FUTURE) +- def_flags = VM_LOCKED; +- current->mm->def_flags = def_flags; ++ current->mm->def_flags |= VM_LOCKED; ++ else ++ current->mm->def_flags &= ~VM_LOCKED; + if (flags == MCL_FUTURE) + goto out; + + for (vma = current->mm->mmap; vma ; vma = prev->vm_next) { +- unsigned int newflags; ++ unsigned long newflags; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if ((current->mm->pax_flags & MF_PAX_SEGMEXEC) && (vma->vm_start >= SEGMEXEC_TASK_SIZE)) ++ break; ++#endif + ++ BUG_ON(vma->vm_end > TASK_SIZE); + newflags = vma->vm_flags | VM_LOCKED; + if (!(flags & MCL_CURRENT)) + newflags &= ~VM_LOCKED; +@@ -573,6 +576,7 @@ SYSCALL_DEFINE1(mlockall, int, flags) + lock_limit >>= PAGE_SHIFT; + + ret = -ENOMEM; ++ gr_learn_resource(current, RLIMIT_MEMLOCK, current->mm->total_vm << PAGE_SHIFT, 1); + if (!(flags & MCL_CURRENT) || (current->mm->total_vm <= lock_limit) || + capable(CAP_IPC_LOCK)) + ret = do_mlockall(flags); +diff -urNp linux-2.6.38.3/mm/mmap.c linux-2.6.38.3/mm/mmap.c +--- linux-2.6.38.3/mm/mmap.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/mmap.c 2011-04-17 15:57:33.000000000 -0400 +@@ -46,6 +46,16 @@ + #define arch_rebalance_pgtables(addr, len) (addr) + #endif + ++static inline void verify_mm_writelocked(struct mm_struct *mm) ++{ ++#if defined(CONFIG_DEBUG_VM) || defined(CONFIG_PAX) ++ if (unlikely(down_read_trylock(&mm->mmap_sem))) { ++ up_read(&mm->mmap_sem); ++ BUG(); ++ } ++#endif ++} ++ + static void unmap_region(struct mm_struct *mm, + struct vm_area_struct *vma, struct vm_area_struct *prev, + unsigned long start, unsigned long end); +@@ -71,22 +81,32 @@ static void unmap_region(struct mm_struc + * x: (no) no x: (no) yes x: (no) yes x: (yes) yes + * + */ +-pgprot_t protection_map[16] = { ++pgprot_t protection_map[16] __read_only = { + __P000, __P001, __P010, __P011, __P100, __P101, __P110, __P111, + __S000, __S001, __S010, __S011, __S100, __S101, __S110, __S111 + }; + + pgprot_t vm_get_page_prot(unsigned long vm_flags) + { +- return __pgprot(pgprot_val(protection_map[vm_flags & ++ pgprot_t prot = __pgprot(pgprot_val(protection_map[vm_flags & + (VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)]) | + pgprot_val(arch_vm_get_page_prot(vm_flags))); ++ ++#if defined(CONFIG_PAX_PAGEEXEC) && defined(CONFIG_X86_32) ++ if (!(__supported_pte_mask & _PAGE_NX) && ++ (vm_flags & (VM_PAGEEXEC | VM_EXEC)) == VM_PAGEEXEC && ++ (vm_flags & (VM_READ | VM_WRITE))) ++ prot = __pgprot(pte_val(pte_exprotect(__pte(pgprot_val(prot))))); ++#endif ++ ++ return prot; + } + EXPORT_SYMBOL(vm_get_page_prot); + + int sysctl_overcommit_memory = OVERCOMMIT_GUESS; /* heuristic overcommit */ + int sysctl_overcommit_ratio = 50; /* default is 50% */ + int sysctl_max_map_count __read_mostly = DEFAULT_MAX_MAP_COUNT; ++unsigned long sysctl_heap_stack_gap __read_mostly = 64*1024; + struct percpu_counter vm_committed_as; + + /* +@@ -232,6 +252,7 @@ static struct vm_area_struct *remove_vma + struct vm_area_struct *next = vma->vm_next; + + might_sleep(); ++ BUG_ON(vma->vm_mirror); + if (vma->vm_ops && vma->vm_ops->close) + vma->vm_ops->close(vma); + if (vma->vm_file) { +@@ -276,6 +297,7 @@ SYSCALL_DEFINE1(brk, unsigned long, brk) + * not page aligned -Ram Gupta + */ + rlim = rlimit(RLIMIT_DATA); ++ gr_learn_resource(current, RLIMIT_DATA, (brk - mm->start_brk) + (mm->end_data - mm->start_data), 1); + if (rlim < RLIM_INFINITY && (brk - mm->start_brk) + + (mm->end_data - mm->start_data) > rlim) + goto out; +@@ -719,6 +741,12 @@ static int + can_vma_merge_before(struct vm_area_struct *vma, unsigned long vm_flags, + struct anon_vma *anon_vma, struct file *file, pgoff_t vm_pgoff) + { ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if ((vma->vm_mm->pax_flags & MF_PAX_SEGMEXEC) && vma->vm_start == SEGMEXEC_TASK_SIZE) ++ return 0; ++#endif ++ + if (is_mergeable_vma(vma, file, vm_flags) && + is_mergeable_anon_vma(anon_vma, vma->anon_vma)) { + if (vma->vm_pgoff == vm_pgoff) +@@ -738,6 +766,12 @@ static int + can_vma_merge_after(struct vm_area_struct *vma, unsigned long vm_flags, + struct anon_vma *anon_vma, struct file *file, pgoff_t vm_pgoff) + { ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if ((vma->vm_mm->pax_flags & MF_PAX_SEGMEXEC) && vma->vm_end == SEGMEXEC_TASK_SIZE) ++ return 0; ++#endif ++ + if (is_mergeable_vma(vma, file, vm_flags) && + is_mergeable_anon_vma(anon_vma, vma->anon_vma)) { + pgoff_t vm_pglen; +@@ -780,13 +814,20 @@ can_vma_merge_after(struct vm_area_struc + struct vm_area_struct *vma_merge(struct mm_struct *mm, + struct vm_area_struct *prev, unsigned long addr, + unsigned long end, unsigned long vm_flags, +- struct anon_vma *anon_vma, struct file *file, ++ struct anon_vma *anon_vma, struct file *file, + pgoff_t pgoff, struct mempolicy *policy) + { + pgoff_t pglen = (end - addr) >> PAGE_SHIFT; + struct vm_area_struct *area, *next; + int err; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ unsigned long addr_m = addr + SEGMEXEC_TASK_SIZE, end_m = end + SEGMEXEC_TASK_SIZE; ++ struct vm_area_struct *area_m = NULL, *next_m = NULL, *prev_m = NULL; ++ ++ BUG_ON((mm->pax_flags & MF_PAX_SEGMEXEC) && SEGMEXEC_TASK_SIZE < end); ++#endif ++ + /* + * We later require that vma->vm_flags == vm_flags, + * so this tests vma->vm_flags & VM_SPECIAL, too. +@@ -802,6 +843,15 @@ struct vm_area_struct *vma_merge(struct + if (next && next->vm_end == end) /* cases 6, 7, 8 */ + next = next->vm_next; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (prev) ++ prev_m = pax_find_mirror_vma(prev); ++ if (area) ++ area_m = pax_find_mirror_vma(area); ++ if (next) ++ next_m = pax_find_mirror_vma(next); ++#endif ++ + /* + * Can it merge with the predecessor? + */ +@@ -821,9 +871,24 @@ struct vm_area_struct *vma_merge(struct + /* cases 1, 6 */ + err = vma_adjust(prev, prev->vm_start, + next->vm_end, prev->vm_pgoff, NULL); +- } else /* cases 2, 5, 7 */ ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (!err && prev_m) ++ err = vma_adjust(prev_m, prev_m->vm_start, ++ next_m->vm_end, prev_m->vm_pgoff, NULL); ++#endif ++ ++ } else { /* cases 2, 5, 7 */ + err = vma_adjust(prev, prev->vm_start, + end, prev->vm_pgoff, NULL); ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (!err && prev_m) ++ err = vma_adjust(prev_m, prev_m->vm_start, ++ end_m, prev_m->vm_pgoff, NULL); ++#endif ++ ++ } + if (err) + return NULL; + khugepaged_enter_vma_merge(prev); +@@ -837,12 +902,27 @@ struct vm_area_struct *vma_merge(struct + mpol_equal(policy, vma_policy(next)) && + can_vma_merge_before(next, vm_flags, + anon_vma, file, pgoff+pglen)) { +- if (prev && addr < prev->vm_end) /* case 4 */ ++ if (prev && addr < prev->vm_end) { /* case 4 */ + err = vma_adjust(prev, prev->vm_start, + addr, prev->vm_pgoff, NULL); +- else /* cases 3, 8 */ ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (!err && prev_m) ++ err = vma_adjust(prev_m, prev_m->vm_start, ++ addr_m, prev_m->vm_pgoff, NULL); ++#endif ++ ++ } else { /* cases 3, 8 */ + err = vma_adjust(area, addr, next->vm_end, + next->vm_pgoff - pglen, NULL); ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (!err && area_m) ++ err = vma_adjust(area_m, addr_m, next_m->vm_end, ++ next_m->vm_pgoff - pglen, NULL); ++#endif ++ ++ } + if (err) + return NULL; + khugepaged_enter_vma_merge(area); +@@ -958,14 +1038,11 @@ none: + void vm_stat_account(struct mm_struct *mm, unsigned long flags, + struct file *file, long pages) + { +- const unsigned long stack_flags +- = VM_STACK_FLAGS & (VM_GROWSUP|VM_GROWSDOWN); +- + if (file) { + mm->shared_vm += pages; + if ((flags & (VM_EXEC|VM_WRITE)) == VM_EXEC) + mm->exec_vm += pages; +- } else if (flags & stack_flags) ++ } else if (flags & (VM_GROWSUP|VM_GROWSDOWN)) + mm->stack_vm += pages; + if (flags & (VM_RESERVED|VM_IO)) + mm->reserved_vm += pages; +@@ -992,7 +1069,7 @@ unsigned long do_mmap_pgoff(struct file + * (the exception is when the underlying filesystem is noexec + * mounted, in which case we dont add PROT_EXEC.) + */ +- if ((prot & PROT_READ) && (current->personality & READ_IMPLIES_EXEC)) ++ if ((prot & (PROT_READ | PROT_WRITE)) && (current->personality & READ_IMPLIES_EXEC)) + if (!(file && (file->f_path.mnt->mnt_flags & MNT_NOEXEC))) + prot |= PROT_EXEC; + +@@ -1018,7 +1095,7 @@ unsigned long do_mmap_pgoff(struct file + /* Obtain the address to map to. we verify (or select) it and ensure + * that it represents a valid section of the address space. + */ +- addr = get_unmapped_area(file, addr, len, pgoff, flags); ++ addr = get_unmapped_area(file, addr, len, pgoff, flags | ((prot & PROT_EXEC) ? MAP_EXECUTABLE : 0)); + if (addr & ~PAGE_MASK) + return addr; + +@@ -1029,6 +1106,36 @@ unsigned long do_mmap_pgoff(struct file + vm_flags = calc_vm_prot_bits(prot) | calc_vm_flag_bits(flags) | + mm->def_flags | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC; + ++#ifdef CONFIG_PAX_MPROTECT ++ if (mm->pax_flags & MF_PAX_MPROTECT) { ++#ifndef CONFIG_PAX_MPROTECT_COMPAT ++ if ((vm_flags & (VM_WRITE | VM_EXEC)) == (VM_WRITE | VM_EXEC)) { ++ gr_log_rwxmmap(file); ++ ++#ifdef CONFIG_PAX_EMUPLT ++ vm_flags &= ~VM_EXEC; ++#else ++ return -EPERM; ++#endif ++ ++ } ++ ++ if (!(vm_flags & VM_EXEC)) ++ vm_flags &= ~VM_MAYEXEC; ++#else ++ if ((vm_flags & (VM_WRITE | VM_EXEC)) != VM_EXEC) ++ vm_flags &= ~(VM_EXEC | VM_MAYEXEC); ++#endif ++ else ++ vm_flags &= ~VM_MAYWRITE; ++ } ++#endif ++ ++#if defined(CONFIG_PAX_PAGEEXEC) && defined(CONFIG_X86_32) ++ if ((mm->pax_flags & MF_PAX_PAGEEXEC) && file) ++ vm_flags &= ~VM_PAGEEXEC; ++#endif ++ + if (flags & MAP_LOCKED) + if (!can_do_mlock()) + return -EPERM; +@@ -1040,6 +1147,7 @@ unsigned long do_mmap_pgoff(struct file + locked += mm->locked_vm; + lock_limit = rlimit(RLIMIT_MEMLOCK); + lock_limit >>= PAGE_SHIFT; ++ gr_learn_resource(current, RLIMIT_MEMLOCK, locked << PAGE_SHIFT, 1); + if (locked > lock_limit && !capable(CAP_IPC_LOCK)) + return -EAGAIN; + } +@@ -1110,6 +1218,9 @@ unsigned long do_mmap_pgoff(struct file + if (error) + return error; + ++ if (!gr_acl_handle_mmap(file, prot)) ++ return -EACCES; ++ + return mmap_region(file, addr, len, flags, vm_flags, pgoff); + } + EXPORT_SYMBOL(do_mmap_pgoff); +@@ -1187,10 +1298,10 @@ SYSCALL_DEFINE1(old_mmap, struct mmap_ar + */ + int vma_wants_writenotify(struct vm_area_struct *vma) + { +- unsigned int vm_flags = vma->vm_flags; ++ unsigned long vm_flags = vma->vm_flags; + + /* If it was private or non-writable, the write bit is already clear */ +- if ((vm_flags & (VM_WRITE|VM_SHARED)) != ((VM_WRITE|VM_SHARED))) ++ if ((vm_flags & (VM_WRITE|VM_SHARED)) != (VM_WRITE|VM_SHARED)) + return 0; + + /* The backer wishes to know when pages are first written to? */ +@@ -1239,14 +1350,24 @@ unsigned long mmap_region(struct file *f + unsigned long charged = 0; + struct inode *inode = file ? file->f_path.dentry->d_inode : NULL; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ struct vm_area_struct *vma_m = NULL; ++#endif ++ ++ /* ++ * mm->mmap_sem is required to protect against another thread ++ * changing the mappings in case we sleep. ++ */ ++ verify_mm_writelocked(mm); ++ + /* Clear old maps */ + error = -ENOMEM; +-munmap_back: + vma = find_vma_prepare(mm, addr, &prev, &rb_link, &rb_parent); + if (vma && vma->vm_start < addr + len) { + if (do_munmap(mm, addr, len)) + return -ENOMEM; +- goto munmap_back; ++ vma = find_vma_prepare(mm, addr, &prev, &rb_link, &rb_parent); ++ BUG_ON(vma && vma->vm_start < addr + len); + } + + /* Check against address space limit. */ +@@ -1295,6 +1416,16 @@ munmap_back: + goto unacct_error; + } + ++#ifdef CONFIG_PAX_SEGMEXEC ++ if ((mm->pax_flags & MF_PAX_SEGMEXEC) && (vm_flags & VM_EXEC)) { ++ vma_m = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); ++ if (!vma_m) { ++ error = -ENOMEM; ++ goto free_vma; ++ } ++ } ++#endif ++ + vma->vm_mm = mm; + vma->vm_start = addr; + vma->vm_end = addr + len; +@@ -1318,6 +1449,19 @@ munmap_back: + error = file->f_op->mmap(file, vma); + if (error) + goto unmap_and_free_vma; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (vma_m && (vm_flags & VM_EXECUTABLE)) ++ added_exe_file_vma(mm); ++#endif ++ ++#if defined(CONFIG_PAX_PAGEEXEC) && defined(CONFIG_X86_32) ++ if ((mm->pax_flags & MF_PAX_PAGEEXEC) && !(vma->vm_flags & VM_SPECIAL)) { ++ vma->vm_flags |= VM_PAGEEXEC; ++ vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); ++ } ++#endif ++ + if (vm_flags & VM_EXECUTABLE) + added_exe_file_vma(mm); + +@@ -1353,6 +1497,11 @@ munmap_back: + vma_link(mm, vma, prev, rb_link, rb_parent); + file = vma->vm_file; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (vma_m) ++ BUG_ON(pax_mirror_vma(vma_m, vma)); ++#endif ++ + /* Once vma denies write, undo our temporary denial count */ + if (correct_wcount) + atomic_inc(&inode->i_writecount); +@@ -1361,6 +1510,7 @@ out: + + mm->total_vm += len >> PAGE_SHIFT; + vm_stat_account(mm, vm_flags, file, len >> PAGE_SHIFT); ++ track_exec_limit(mm, addr, addr + len, vm_flags); + if (vm_flags & VM_LOCKED) { + if (!mlock_vma_pages_range(vma, addr, addr + len)) + mm->locked_vm += (len >> PAGE_SHIFT); +@@ -1378,6 +1528,12 @@ unmap_and_free_vma: + unmap_region(mm, vma, prev, vma->vm_start, vma->vm_end); + charged = 0; + free_vma: ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (vma_m) ++ kmem_cache_free(vm_area_cachep, vma_m); ++#endif ++ + kmem_cache_free(vm_area_cachep, vma); + unacct_error: + if (charged) +@@ -1385,6 +1541,44 @@ unacct_error: + return error; + } + ++bool check_heap_stack_gap(const struct vm_area_struct *vma, unsigned long addr, unsigned long len) ++{ ++ if (!vma) { ++#ifdef CONFIG_STACK_GROWSUP ++ if (addr > sysctl_heap_stack_gap) ++ vma = find_vma(current->mm, addr - sysctl_heap_stack_gap); ++ else ++ vma = find_vma(current->mm, 0); ++ if (vma && (vma->vm_flags & VM_GROWSUP)) ++ return false; ++#endif ++ return true; ++ } ++ ++ if (addr + len > vma->vm_start) ++ return false; ++ ++ if (vma->vm_flags & VM_GROWSDOWN) ++ return sysctl_heap_stack_gap <= vma->vm_start - addr - len; ++#ifdef CONFIG_STACK_GROWSUP ++ else if (vma->vm_prev && (vma->vm_prev->vm_flags & VM_GROWSUP)) ++ return addr - vma->vm_prev->vm_end <= sysctl_heap_stack_gap; ++#endif ++ ++ return true; ++} ++ ++unsigned long skip_heap_stack_gap(const struct vm_area_struct *vma, unsigned long len) ++{ ++ if (vma->vm_start < len) ++ return -ENOMEM; ++ if (!(vma->vm_flags & VM_GROWSDOWN)) ++ return vma->vm_start - len; ++ if (sysctl_heap_stack_gap <= vma->vm_start - len) ++ return vma->vm_start - len - sysctl_heap_stack_gap; ++ return -ENOMEM; ++} ++ + /* Get an address range which is currently unmapped. + * For shmat() with addr=0. + * +@@ -1411,18 +1605,23 @@ arch_get_unmapped_area(struct file *filp + if (flags & MAP_FIXED) + return addr; + ++#ifdef CONFIG_PAX_RANDMMAP ++ if (!(mm->pax_flags & MF_PAX_RANDMMAP)) ++#endif ++ + if (addr) { + addr = PAGE_ALIGN(addr); +- vma = find_vma(mm, addr); +- if (TASK_SIZE - len >= addr && +- (!vma || addr + len <= vma->vm_start)) +- return addr; ++ if (TASK_SIZE - len >= addr) { ++ vma = find_vma(mm, addr); ++ if (check_heap_stack_gap(vma, addr, len)) ++ return addr; ++ } + } + if (len > mm->cached_hole_size) { +- start_addr = addr = mm->free_area_cache; ++ start_addr = addr = mm->free_area_cache; + } else { +- start_addr = addr = TASK_UNMAPPED_BASE; +- mm->cached_hole_size = 0; ++ start_addr = addr = mm->mmap_base; ++ mm->cached_hole_size = 0; + } + + full_search: +@@ -1433,34 +1632,40 @@ full_search: + * Start a new search - just in case we missed + * some holes. + */ +- if (start_addr != TASK_UNMAPPED_BASE) { +- addr = TASK_UNMAPPED_BASE; +- start_addr = addr; ++ if (start_addr != mm->mmap_base) { ++ start_addr = addr = mm->mmap_base; + mm->cached_hole_size = 0; + goto full_search; + } + return -ENOMEM; + } +- if (!vma || addr + len <= vma->vm_start) { +- /* +- * Remember the place where we stopped the search: +- */ +- mm->free_area_cache = addr + len; +- return addr; +- } ++ if (check_heap_stack_gap(vma, addr, len)) ++ break; + if (addr + mm->cached_hole_size < vma->vm_start) + mm->cached_hole_size = vma->vm_start - addr; + addr = vma->vm_end; + } ++ ++ /* ++ * Remember the place where we stopped the search: ++ */ ++ mm->free_area_cache = addr + len; ++ return addr; + } + #endif + + void arch_unmap_area(struct mm_struct *mm, unsigned long addr) + { ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if ((mm->pax_flags & MF_PAX_SEGMEXEC) && SEGMEXEC_TASK_SIZE <= addr) ++ return; ++#endif ++ + /* + * Is this a new hole at the lowest possible address? + */ +- if (addr >= TASK_UNMAPPED_BASE && addr < mm->free_area_cache) { ++ if (addr >= mm->mmap_base && addr < mm->free_area_cache) { + mm->free_area_cache = addr; + mm->cached_hole_size = ~0UL; + } +@@ -1478,7 +1683,7 @@ arch_get_unmapped_area_topdown(struct fi + { + struct vm_area_struct *vma; + struct mm_struct *mm = current->mm; +- unsigned long addr = addr0; ++ unsigned long base = mm->mmap_base, addr = addr0; + + /* requested length too big for entire address space */ + if (len > TASK_SIZE) +@@ -1487,13 +1692,18 @@ arch_get_unmapped_area_topdown(struct fi + if (flags & MAP_FIXED) + return addr; + ++#ifdef CONFIG_PAX_RANDMMAP ++ if (!(mm->pax_flags & MF_PAX_RANDMMAP)) ++#endif ++ + /* requesting a specific address */ + if (addr) { + addr = PAGE_ALIGN(addr); +- vma = find_vma(mm, addr); +- if (TASK_SIZE - len >= addr && +- (!vma || addr + len <= vma->vm_start)) +- return addr; ++ if (TASK_SIZE - len >= addr) { ++ vma = find_vma(mm, addr); ++ if (check_heap_stack_gap(vma, addr, len)) ++ return addr; ++ } + } + + /* check if free_area_cache is useful for us */ +@@ -1508,7 +1718,7 @@ arch_get_unmapped_area_topdown(struct fi + /* make sure it can fit in the remaining address space */ + if (addr > len) { + vma = find_vma(mm, addr-len); +- if (!vma || addr <= vma->vm_start) ++ if (check_heap_stack_gap(vma, addr - len, len)) + /* remember the address as a hint for next time */ + return (mm->free_area_cache = addr-len); + } +@@ -1525,7 +1735,7 @@ arch_get_unmapped_area_topdown(struct fi + * return with success: + */ + vma = find_vma(mm, addr); +- if (!vma || addr+len <= vma->vm_start) ++ if (check_heap_stack_gap(vma, addr, len)) + /* remember the address as a hint for next time */ + return (mm->free_area_cache = addr); + +@@ -1534,8 +1744,8 @@ arch_get_unmapped_area_topdown(struct fi + mm->cached_hole_size = vma->vm_start - addr; + + /* try just below the current vma->vm_start */ +- addr = vma->vm_start-len; +- } while (len < vma->vm_start); ++ addr = skip_heap_stack_gap(vma, len); ++ } while (!IS_ERR_VALUE(addr)); + + bottomup: + /* +@@ -1544,13 +1754,21 @@ bottomup: + * can happen with large stack limits and large mmap() + * allocations. + */ ++ mm->mmap_base = TASK_UNMAPPED_BASE; ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base += mm->delta_mmap; ++#endif ++ ++ mm->free_area_cache = mm->mmap_base; + mm->cached_hole_size = ~0UL; +- mm->free_area_cache = TASK_UNMAPPED_BASE; + addr = arch_get_unmapped_area(filp, addr0, len, pgoff, flags); + /* + * Restore the topdown base: + */ +- mm->free_area_cache = mm->mmap_base; ++ mm->mmap_base = base; ++ mm->free_area_cache = base; + mm->cached_hole_size = ~0UL; + + return addr; +@@ -1559,6 +1777,12 @@ bottomup: + + void arch_unmap_area_topdown(struct mm_struct *mm, unsigned long addr) + { ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if ((mm->pax_flags & MF_PAX_SEGMEXEC) && SEGMEXEC_TASK_SIZE <= addr) ++ return; ++#endif ++ + /* + * Is this a new hole at the highest possible address? + */ +@@ -1566,8 +1790,10 @@ void arch_unmap_area_topdown(struct mm_s + mm->free_area_cache = addr; + + /* dont allow allocations above current base */ +- if (mm->free_area_cache > mm->mmap_base) ++ if (mm->free_area_cache > mm->mmap_base) { + mm->free_area_cache = mm->mmap_base; ++ mm->cached_hole_size = ~0UL; ++ } + } + + unsigned long +@@ -1675,6 +1901,28 @@ out: + return prev ? prev->vm_next : vma; + } + ++#ifdef CONFIG_PAX_SEGMEXEC ++struct vm_area_struct *pax_find_mirror_vma(struct vm_area_struct *vma) ++{ ++ struct vm_area_struct *vma_m; ++ ++ BUG_ON(!vma || vma->vm_start >= vma->vm_end); ++ if (!(vma->vm_mm->pax_flags & MF_PAX_SEGMEXEC) || !(vma->vm_flags & VM_EXEC)) { ++ BUG_ON(vma->vm_mirror); ++ return NULL; ++ } ++ BUG_ON(vma->vm_start < SEGMEXEC_TASK_SIZE && SEGMEXEC_TASK_SIZE < vma->vm_end); ++ vma_m = vma->vm_mirror; ++ BUG_ON(!vma_m || vma_m->vm_mirror != vma); ++ BUG_ON(vma->vm_file != vma_m->vm_file); ++ BUG_ON(vma->vm_end - vma->vm_start != vma_m->vm_end - vma_m->vm_start); ++ BUG_ON(vma->vm_pgoff != vma_m->vm_pgoff); ++ BUG_ON(vma->anon_vma != vma_m->anon_vma && vma->anon_vma->root != vma_m->anon_vma->root); ++ BUG_ON((vma->vm_flags ^ vma_m->vm_flags) & ~(VM_WRITE | VM_MAYWRITE | VM_ACCOUNT | VM_LOCKED | VM_RESERVED)); ++ return vma_m; ++} ++#endif ++ + /* + * Verify that the stack growth is acceptable and + * update accounting. This is shared with both the +@@ -1691,6 +1939,7 @@ static int acct_stack_growth(struct vm_a + return -ENOMEM; + + /* Stack limit test */ ++ gr_learn_resource(current, RLIMIT_STACK, size, 1); + if (size > ACCESS_ONCE(rlim[RLIMIT_STACK].rlim_cur)) + return -ENOMEM; + +@@ -1701,6 +1950,7 @@ static int acct_stack_growth(struct vm_a + locked = mm->locked_vm + grow; + limit = ACCESS_ONCE(rlim[RLIMIT_MEMLOCK].rlim_cur); + limit >>= PAGE_SHIFT; ++ gr_learn_resource(current, RLIMIT_MEMLOCK, locked << PAGE_SHIFT, 1); + if (locked > limit && !capable(CAP_IPC_LOCK)) + return -ENOMEM; + } +@@ -1731,37 +1981,48 @@ static int acct_stack_growth(struct vm_a + * PA-RISC uses this for its stack; IA64 for its Register Backing Store. + * vma is the last one with address > vma->vm_end. Have to extend vma. + */ ++#ifndef CONFIG_IA64 ++static ++#endif + int expand_upwards(struct vm_area_struct *vma, unsigned long address) + { + int error; ++ bool locknext; + + if (!(vma->vm_flags & VM_GROWSUP)) + return -EFAULT; + ++ /* Also guard against wrapping around to address 0. */ ++ if (address < PAGE_ALIGN(address+1)) ++ address = PAGE_ALIGN(address+1); ++ else ++ return -ENOMEM; ++ + /* + * We must make sure the anon_vma is allocated + * so that the anon_vma locking is not a noop. + */ + if (unlikely(anon_vma_prepare(vma))) + return -ENOMEM; ++ locknext = vma->vm_next && (vma->vm_next->vm_flags & VM_GROWSDOWN); ++ if (locknext && anon_vma_prepare(vma->vm_next)) ++ return -ENOMEM; + vma_lock_anon_vma(vma); ++ if (locknext) ++ vma_lock_anon_vma(vma->vm_next); + + /* + * vma->vm_start/vm_end cannot change under us because the caller + * is required to hold the mmap_sem in read mode. We need the +- * anon_vma lock to serialize against concurrent expand_stacks. +- * Also guard against wrapping around to address 0. ++ * anon_vma locks to serialize against concurrent expand_stacks ++ * and expand_upwards. + */ +- if (address < PAGE_ALIGN(address+4)) +- address = PAGE_ALIGN(address+4); +- else { +- vma_unlock_anon_vma(vma); +- return -ENOMEM; +- } + error = 0; + + /* Somebody else might have raced and expanded it already */ +- if (address > vma->vm_end) { ++ if (vma->vm_next && (vma->vm_next->vm_flags & (VM_READ | VM_WRITE | VM_EXEC)) && vma->vm_next->vm_start - address < sysctl_heap_stack_gap) ++ error = -ENOMEM; ++ else if (address > vma->vm_end && (!locknext || vma->vm_next->vm_start >= address)) { + unsigned long size, grow; + + size = address - vma->vm_start; +@@ -1773,6 +2034,8 @@ int expand_upwards(struct vm_area_struct + perf_event_mmap(vma); + } + } ++ if (locknext) ++ vma_unlock_anon_vma(vma->vm_next); + vma_unlock_anon_vma(vma); + khugepaged_enter_vma_merge(vma); + return error; +@@ -1786,6 +2049,8 @@ static int expand_downwards(struct vm_ar + unsigned long address) + { + int error; ++ bool lockprev = false; ++ struct vm_area_struct *prev; + + /* + * We must make sure the anon_vma is allocated +@@ -1799,6 +2064,15 @@ static int expand_downwards(struct vm_ar + if (error) + return error; + ++ prev = vma->vm_prev; ++#if defined(CONFIG_STACK_GROWSUP) || defined(CONFIG_IA64) ++ lockprev = prev && (prev->vm_flags & VM_GROWSUP); ++#endif ++ if (lockprev && anon_vma_prepare(prev)) ++ return -ENOMEM; ++ if (lockprev) ++ vma_lock_anon_vma(prev); ++ + vma_lock_anon_vma(vma); + + /* +@@ -1808,9 +2082,17 @@ static int expand_downwards(struct vm_ar + */ + + /* Somebody else might have raced and expanded it already */ +- if (address < vma->vm_start) { ++ if (prev && (prev->vm_flags & (VM_READ | VM_WRITE | VM_EXEC)) && address - prev->vm_end < sysctl_heap_stack_gap) ++ error = -ENOMEM; ++ else if (address < vma->vm_start && (!lockprev || prev->vm_end <= address)) { + unsigned long size, grow; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ struct vm_area_struct *vma_m; ++ ++ vma_m = pax_find_mirror_vma(vma); ++#endif ++ + size = vma->vm_end - address; + grow = (vma->vm_start - address) >> PAGE_SHIFT; + +@@ -1818,10 +2100,21 @@ static int expand_downwards(struct vm_ar + if (!error) { + vma->vm_start = address; + vma->vm_pgoff -= grow; ++ track_exec_limit(vma->vm_mm, vma->vm_start, vma->vm_end, vma->vm_flags); ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (vma_m) { ++ vma_m->vm_start -= grow << PAGE_SHIFT; ++ vma_m->vm_pgoff -= grow; ++ } ++#endif ++ + perf_event_mmap(vma); + } + } + vma_unlock_anon_vma(vma); ++ if (lockprev) ++ vma_unlock_anon_vma(prev); + khugepaged_enter_vma_merge(vma); + return error; + } +@@ -1896,6 +2189,13 @@ static void remove_vma_list(struct mm_st + do { + long nrpages = vma_pages(vma); + ++#ifdef CONFIG_PAX_SEGMEXEC ++ if ((mm->pax_flags & MF_PAX_SEGMEXEC) && (vma->vm_start >= SEGMEXEC_TASK_SIZE)) { ++ vma = remove_vma(vma); ++ continue; ++ } ++#endif ++ + mm->total_vm -= nrpages; + vm_stat_account(mm, vma->vm_flags, vma->vm_file, -nrpages); + vma = remove_vma(vma); +@@ -1941,6 +2241,16 @@ detach_vmas_to_be_unmapped(struct mm_str + insertion_point = (prev ? &prev->vm_next : &mm->mmap); + vma->vm_prev = NULL; + do { ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (vma->vm_mirror) { ++ BUG_ON(!vma->vm_mirror->vm_mirror || vma->vm_mirror->vm_mirror != vma); ++ vma->vm_mirror->vm_mirror = NULL; ++ vma->vm_mirror->vm_flags &= ~VM_EXEC; ++ vma->vm_mirror = NULL; ++ } ++#endif ++ + rb_erase(&vma->vm_rb, &mm->mm_rb); + mm->map_count--; + tail_vma = vma; +@@ -1969,14 +2279,33 @@ static int __split_vma(struct mm_struct + struct vm_area_struct *new; + int err = -ENOMEM; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ struct vm_area_struct *vma_m, *new_m = NULL; ++ unsigned long addr_m = addr + SEGMEXEC_TASK_SIZE; ++#endif ++ + if (is_vm_hugetlb_page(vma) && (addr & + ~(huge_page_mask(hstate_vma(vma))))) + return -EINVAL; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ vma_m = pax_find_mirror_vma(vma); ++#endif ++ + new = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); + if (!new) + goto out_err; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (vma_m) { ++ new_m = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); ++ if (!new_m) { ++ kmem_cache_free(vm_area_cachep, new); ++ goto out_err; ++ } ++ } ++#endif ++ + /* most fields are the same, copy all, and then fixup */ + *new = *vma; + +@@ -1989,6 +2318,22 @@ static int __split_vma(struct mm_struct + new->vm_pgoff += ((addr - vma->vm_start) >> PAGE_SHIFT); + } + ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (vma_m) { ++ *new_m = *vma_m; ++ INIT_LIST_HEAD(&new_m->anon_vma_chain); ++ new_m->vm_mirror = new; ++ new->vm_mirror = new_m; ++ ++ if (new_below) ++ new_m->vm_end = addr_m; ++ else { ++ new_m->vm_start = addr_m; ++ new_m->vm_pgoff += ((addr_m - vma_m->vm_start) >> PAGE_SHIFT); ++ } ++ } ++#endif ++ + pol = mpol_dup(vma_policy(vma)); + if (IS_ERR(pol)) { + err = PTR_ERR(pol); +@@ -2014,6 +2359,42 @@ static int __split_vma(struct mm_struct + else + err = vma_adjust(vma, vma->vm_start, addr, vma->vm_pgoff, new); + ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (!err && vma_m) { ++ if (anon_vma_clone(new_m, vma_m)) ++ goto out_free_mpol; ++ ++ mpol_get(pol); ++ vma_set_policy(new_m, pol); ++ ++ if (new_m->vm_file) { ++ get_file(new_m->vm_file); ++ if (vma_m->vm_flags & VM_EXECUTABLE) ++ added_exe_file_vma(mm); ++ } ++ ++ if (new_m->vm_ops && new_m->vm_ops->open) ++ new_m->vm_ops->open(new_m); ++ ++ if (new_below) ++ err = vma_adjust(vma_m, addr_m, vma_m->vm_end, vma_m->vm_pgoff + ++ ((addr_m - new_m->vm_start) >> PAGE_SHIFT), new_m); ++ else ++ err = vma_adjust(vma_m, vma_m->vm_start, addr_m, vma_m->vm_pgoff, new_m); ++ ++ if (err) { ++ if (new_m->vm_ops && new_m->vm_ops->close) ++ new_m->vm_ops->close(new_m); ++ if (new_m->vm_file) { ++ if (vma_m->vm_flags & VM_EXECUTABLE) ++ removed_exe_file_vma(mm); ++ fput(new_m->vm_file); ++ } ++ mpol_put(pol); ++ } ++ } ++#endif ++ + /* Success. */ + if (!err) + return 0; +@@ -2026,10 +2407,18 @@ static int __split_vma(struct mm_struct + removed_exe_file_vma(mm); + fput(new->vm_file); + } +- unlink_anon_vmas(new); + out_free_mpol: + mpol_put(pol); + out_free_vma: ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (new_m) { ++ unlink_anon_vmas(new_m); ++ kmem_cache_free(vm_area_cachep, new_m); ++ } ++#endif ++ ++ unlink_anon_vmas(new); + kmem_cache_free(vm_area_cachep, new); + out_err: + return err; +@@ -2042,6 +2431,15 @@ static int __split_vma(struct mm_struct + int split_vma(struct mm_struct *mm, struct vm_area_struct *vma, + unsigned long addr, int new_below) + { ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (mm->pax_flags & MF_PAX_SEGMEXEC) { ++ BUG_ON(vma->vm_end > SEGMEXEC_TASK_SIZE); ++ if (mm->map_count >= sysctl_max_map_count-1) ++ return -ENOMEM; ++ } else ++#endif ++ + if (mm->map_count >= sysctl_max_map_count) + return -ENOMEM; + +@@ -2053,11 +2451,30 @@ int split_vma(struct mm_struct *mm, stru + * work. This now handles partial unmappings. + * Jeremy Fitzhardinge <jeremy@goop.org> + */ ++#ifdef CONFIG_PAX_SEGMEXEC + int do_munmap(struct mm_struct *mm, unsigned long start, size_t len) + { ++ int ret = __do_munmap(mm, start, len); ++ if (ret || !(mm->pax_flags & MF_PAX_SEGMEXEC)) ++ return ret; ++ ++ return __do_munmap(mm, start + SEGMEXEC_TASK_SIZE, len); ++} ++ ++int __do_munmap(struct mm_struct *mm, unsigned long start, size_t len) ++#else ++int do_munmap(struct mm_struct *mm, unsigned long start, size_t len) ++#endif ++{ + unsigned long end; + struct vm_area_struct *vma, *prev, *last; + ++ /* ++ * mm->mmap_sem is required to protect against another thread ++ * changing the mappings in case we sleep. ++ */ ++ verify_mm_writelocked(mm); ++ + if ((start & ~PAGE_MASK) || start > TASK_SIZE || len > TASK_SIZE-start) + return -EINVAL; + +@@ -2131,6 +2548,8 @@ int do_munmap(struct mm_struct *mm, unsi + /* Fix up all other VM information */ + remove_vma_list(mm, vma); + ++ track_exec_limit(mm, start, end, 0UL); ++ + return 0; + } + +@@ -2143,22 +2562,18 @@ SYSCALL_DEFINE2(munmap, unsigned long, a + + profile_munmap(addr); + ++#ifdef CONFIG_PAX_SEGMEXEC ++ if ((mm->pax_flags & MF_PAX_SEGMEXEC) && ++ (len > SEGMEXEC_TASK_SIZE || addr > SEGMEXEC_TASK_SIZE-len)) ++ return -EINVAL; ++#endif ++ + down_write(&mm->mmap_sem); + ret = do_munmap(mm, addr, len); + up_write(&mm->mmap_sem); + return ret; + } + +-static inline void verify_mm_writelocked(struct mm_struct *mm) +-{ +-#ifdef CONFIG_DEBUG_VM +- if (unlikely(down_read_trylock(&mm->mmap_sem))) { +- WARN_ON(1); +- up_read(&mm->mmap_sem); +- } +-#endif +-} +- + /* + * this is really a simplified "do_mmap". it only handles + * anonymous maps. eventually we may be able to do some +@@ -2172,6 +2587,7 @@ unsigned long do_brk(unsigned long addr, + struct rb_node ** rb_link, * rb_parent; + pgoff_t pgoff = addr >> PAGE_SHIFT; + int error; ++ unsigned long charged; + + len = PAGE_ALIGN(len); + if (!len) +@@ -2183,16 +2599,30 @@ unsigned long do_brk(unsigned long addr, + + flags = VM_DATA_DEFAULT_FLAGS | VM_ACCOUNT | mm->def_flags; + ++#if defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) ++ if (mm->pax_flags & (MF_PAX_PAGEEXEC | MF_PAX_SEGMEXEC)) { ++ flags &= ~VM_EXEC; ++ ++#ifdef CONFIG_PAX_MPROTECT ++ if (mm->pax_flags & MF_PAX_MPROTECT) ++ flags &= ~VM_MAYEXEC; ++#endif ++ ++ } ++#endif ++ + error = get_unmapped_area(NULL, addr, len, 0, MAP_FIXED); + if (error & ~PAGE_MASK) + return error; + ++ charged = len >> PAGE_SHIFT; ++ + /* + * mlock MCL_FUTURE? + */ + if (mm->def_flags & VM_LOCKED) { + unsigned long locked, lock_limit; +- locked = len >> PAGE_SHIFT; ++ locked = charged; + locked += mm->locked_vm; + lock_limit = rlimit(RLIMIT_MEMLOCK); + lock_limit >>= PAGE_SHIFT; +@@ -2209,22 +2639,22 @@ unsigned long do_brk(unsigned long addr, + /* + * Clear old maps. this also does some error checking for us + */ +- munmap_back: + vma = find_vma_prepare(mm, addr, &prev, &rb_link, &rb_parent); + if (vma && vma->vm_start < addr + len) { + if (do_munmap(mm, addr, len)) + return -ENOMEM; +- goto munmap_back; ++ vma = find_vma_prepare(mm, addr, &prev, &rb_link, &rb_parent); ++ BUG_ON(vma && vma->vm_start < addr + len); + } + + /* Check against address space limits *after* clearing old maps... */ +- if (!may_expand_vm(mm, len >> PAGE_SHIFT)) ++ if (!may_expand_vm(mm, charged)) + return -ENOMEM; + + if (mm->map_count > sysctl_max_map_count) + return -ENOMEM; + +- if (security_vm_enough_memory(len >> PAGE_SHIFT)) ++ if (security_vm_enough_memory(charged)) + return -ENOMEM; + + /* Can we just expand an old private anonymous mapping? */ +@@ -2238,7 +2668,7 @@ unsigned long do_brk(unsigned long addr, + */ + vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); + if (!vma) { +- vm_unacct_memory(len >> PAGE_SHIFT); ++ vm_unacct_memory(charged); + return -ENOMEM; + } + +@@ -2252,11 +2682,12 @@ unsigned long do_brk(unsigned long addr, + vma_link(mm, vma, prev, rb_link, rb_parent); + out: + perf_event_mmap(vma); +- mm->total_vm += len >> PAGE_SHIFT; ++ mm->total_vm += charged; + if (flags & VM_LOCKED) { + if (!mlock_vma_pages_range(vma, addr, addr + len)) +- mm->locked_vm += (len >> PAGE_SHIFT); ++ mm->locked_vm += charged; + } ++ track_exec_limit(mm, addr, addr + len, flags); + return addr; + } + +@@ -2303,8 +2734,10 @@ void exit_mmap(struct mm_struct *mm) + * Walk the list again, actually closing and freeing it, + * with preemption enabled, without holding any MM locks. + */ +- while (vma) ++ while (vma) { ++ vma->vm_mirror = NULL; + vma = remove_vma(vma); ++ } + + BUG_ON(mm->nr_ptes > (FIRST_USER_ADDRESS+PMD_SIZE-1)>>PMD_SHIFT); + } +@@ -2318,6 +2751,13 @@ int insert_vm_struct(struct mm_struct * + struct vm_area_struct * __vma, * prev; + struct rb_node ** rb_link, * rb_parent; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ struct vm_area_struct *vma_m = NULL; ++#endif ++ ++ if (security_file_mmap(NULL, 0, 0, 0, vma->vm_start, 1)) ++ return -EPERM; ++ + /* + * The vm_pgoff of a purely anonymous vma should be irrelevant + * until its first write fault, when page's anon_vma and index +@@ -2340,7 +2780,22 @@ int insert_vm_struct(struct mm_struct * + if ((vma->vm_flags & VM_ACCOUNT) && + security_vm_enough_memory_mm(mm, vma_pages(vma))) + return -ENOMEM; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if ((mm->pax_flags & MF_PAX_SEGMEXEC) && (vma->vm_flags & VM_EXEC)) { ++ vma_m = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); ++ if (!vma_m) ++ return -ENOMEM; ++ } ++#endif ++ + vma_link(mm, vma, prev, rb_link, rb_parent); ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (vma_m) ++ BUG_ON(pax_mirror_vma(vma_m, vma)); ++#endif ++ + return 0; + } + +@@ -2358,6 +2813,8 @@ struct vm_area_struct *copy_vma(struct v + struct rb_node **rb_link, *rb_parent; + struct mempolicy *pol; + ++ BUG_ON(vma->vm_mirror); ++ + /* + * If anonymous vma has not yet been faulted, update new pgoff + * to match new location, to increase its chance of merging. +@@ -2407,6 +2864,39 @@ struct vm_area_struct *copy_vma(struct v + kmem_cache_free(vm_area_cachep, new_vma); + return NULL; + } ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++long pax_mirror_vma(struct vm_area_struct *vma_m, struct vm_area_struct *vma) ++{ ++ struct vm_area_struct *prev_m; ++ struct rb_node **rb_link_m, *rb_parent_m; ++ struct mempolicy *pol_m; ++ ++ BUG_ON(!(vma->vm_mm->pax_flags & MF_PAX_SEGMEXEC) || !(vma->vm_flags & VM_EXEC)); ++ BUG_ON(vma->vm_mirror || vma_m->vm_mirror); ++ BUG_ON(!mpol_equal(vma_policy(vma), vma_policy(vma_m))); ++ *vma_m = *vma; ++ INIT_LIST_HEAD(&vma_m->anon_vma_chain); ++ if (anon_vma_clone(vma_m, vma)) ++ return -ENOMEM; ++ pol_m = vma_policy(vma_m); ++ mpol_get(pol_m); ++ vma_set_policy(vma_m, pol_m); ++ vma_m->vm_start += SEGMEXEC_TASK_SIZE; ++ vma_m->vm_end += SEGMEXEC_TASK_SIZE; ++ vma_m->vm_flags &= ~(VM_WRITE | VM_MAYWRITE | VM_ACCOUNT | VM_LOCKED); ++ vma_m->vm_page_prot = vm_get_page_prot(vma_m->vm_flags); ++ if (vma_m->vm_file) ++ get_file(vma_m->vm_file); ++ if (vma_m->vm_ops && vma_m->vm_ops->open) ++ vma_m->vm_ops->open(vma_m); ++ find_vma_prepare(vma->vm_mm, vma_m->vm_start, &prev_m, &rb_link_m, &rb_parent_m); ++ vma_link(vma->vm_mm, vma_m, prev_m, rb_link_m, rb_parent_m); ++ vma_m->vm_mirror = vma; ++ vma->vm_mirror = vma_m; ++ return 0; ++} ++#endif + + /* + * Return true if the calling process may expand its vm space by the passed +@@ -2418,7 +2908,7 @@ int may_expand_vm(struct mm_struct *mm, + unsigned long lim; + + lim = rlimit(RLIMIT_AS) >> PAGE_SHIFT; +- ++ gr_learn_resource(current, RLIMIT_AS, (cur + npages) << PAGE_SHIFT, 1); + if (cur + npages > lim) + return 0; + return 1; +@@ -2489,6 +2979,22 @@ int install_special_mapping(struct mm_st + vma->vm_start = addr; + vma->vm_end = addr + len; + ++#ifdef CONFIG_PAX_MPROTECT ++ if (mm->pax_flags & MF_PAX_MPROTECT) { ++#ifndef CONFIG_PAX_MPROTECT_COMPAT ++ if ((vm_flags & (VM_WRITE | VM_EXEC)) == (VM_WRITE | VM_EXEC)) ++ return -EPERM; ++ if (!(vm_flags & VM_EXEC)) ++ vm_flags &= ~VM_MAYEXEC; ++#else ++ if ((vm_flags & (VM_WRITE | VM_EXEC)) != VM_EXEC) ++ vm_flags &= ~(VM_EXEC | VM_MAYEXEC); ++#endif ++ else ++ vm_flags &= ~VM_MAYWRITE; ++ } ++#endif ++ + vma->vm_flags = vm_flags | mm->def_flags | VM_DONTEXPAND; + vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); + +diff -urNp linux-2.6.38.3/mm/mprotect.c linux-2.6.38.3/mm/mprotect.c +--- linux-2.6.38.3/mm/mprotect.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/mprotect.c 2011-04-17 15:57:33.000000000 -0400 +@@ -23,10 +23,16 @@ + #include <linux/mmu_notifier.h> + #include <linux/migrate.h> + #include <linux/perf_event.h> ++ ++#ifdef CONFIG_PAX_MPROTECT ++#include <linux/elf.h> ++#endif ++ + #include <asm/uaccess.h> + #include <asm/pgtable.h> + #include <asm/cacheflush.h> + #include <asm/tlbflush.h> ++#include <asm/mmu_context.h> + + #ifndef pgprot_modify + static inline pgprot_t pgprot_modify(pgprot_t oldprot, pgprot_t newprot) +@@ -141,6 +147,48 @@ static void change_protection(struct vm_ + flush_tlb_range(vma, start, end); + } + ++#ifdef CONFIG_ARCH_TRACK_EXEC_LIMIT ++/* called while holding the mmap semaphor for writing except stack expansion */ ++void track_exec_limit(struct mm_struct *mm, unsigned long start, unsigned long end, unsigned long prot) ++{ ++ unsigned long oldlimit, newlimit = 0UL; ++ ++ if (!(mm->pax_flags & MF_PAX_PAGEEXEC) || (__supported_pte_mask & _PAGE_NX)) ++ return; ++ ++ spin_lock(&mm->page_table_lock); ++ oldlimit = mm->context.user_cs_limit; ++ if ((prot & VM_EXEC) && oldlimit < end) ++ /* USER_CS limit moved up */ ++ newlimit = end; ++ else if (!(prot & VM_EXEC) && start < oldlimit && oldlimit <= end) ++ /* USER_CS limit moved down */ ++ newlimit = start; ++ ++ if (newlimit) { ++ mm->context.user_cs_limit = newlimit; ++ ++#ifdef CONFIG_SMP ++ wmb(); ++ cpus_clear(mm->context.cpu_user_cs_mask); ++ cpu_set(smp_processor_id(), mm->context.cpu_user_cs_mask); ++#endif ++ ++ set_user_cs(mm->context.user_cs_base, mm->context.user_cs_limit, smp_processor_id()); ++ } ++ spin_unlock(&mm->page_table_lock); ++ if (newlimit == end) { ++ struct vm_area_struct *vma = find_vma(mm, oldlimit); ++ ++ for (; vma && vma->vm_start < end; vma = vma->vm_next) ++ if (is_vm_hugetlb_page(vma)) ++ hugetlb_change_protection(vma, vma->vm_start, vma->vm_end, vma->vm_page_prot); ++ else ++ change_protection(vma, vma->vm_start, vma->vm_end, vma->vm_page_prot, vma_wants_writenotify(vma)); ++ } ++} ++#endif ++ + int + mprotect_fixup(struct vm_area_struct *vma, struct vm_area_struct **pprev, + unsigned long start, unsigned long end, unsigned long newflags) +@@ -153,11 +201,29 @@ mprotect_fixup(struct vm_area_struct *vm + int error; + int dirty_accountable = 0; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ struct vm_area_struct *vma_m = NULL; ++ unsigned long start_m, end_m; ++ ++ start_m = start + SEGMEXEC_TASK_SIZE; ++ end_m = end + SEGMEXEC_TASK_SIZE; ++#endif ++ + if (newflags == oldflags) { + *pprev = vma; + return 0; + } + ++ if (newflags & (VM_READ | VM_WRITE | VM_EXEC)) { ++ struct vm_area_struct *prev = vma->vm_prev, *next = vma->vm_next; ++ ++ if (next && (next->vm_flags & VM_GROWSDOWN) && sysctl_heap_stack_gap > next->vm_start - end) ++ return -ENOMEM; ++ ++ if (prev && (prev->vm_flags & VM_GROWSUP) && sysctl_heap_stack_gap > start - prev->vm_end) ++ return -ENOMEM; ++ } ++ + /* + * If we make a private mapping writable we increase our commit; + * but (without finer accounting) cannot reduce our commit if we +@@ -174,6 +240,42 @@ mprotect_fixup(struct vm_area_struct *vm + } + } + ++#ifdef CONFIG_PAX_SEGMEXEC ++ if ((mm->pax_flags & MF_PAX_SEGMEXEC) && ((oldflags ^ newflags) & VM_EXEC)) { ++ if (start != vma->vm_start) { ++ error = split_vma(mm, vma, start, 1); ++ if (error) ++ goto fail; ++ BUG_ON(!*pprev || (*pprev)->vm_next == vma); ++ *pprev = (*pprev)->vm_next; ++ } ++ ++ if (end != vma->vm_end) { ++ error = split_vma(mm, vma, end, 0); ++ if (error) ++ goto fail; ++ } ++ ++ if (pax_find_mirror_vma(vma)) { ++ error = __do_munmap(mm, start_m, end_m - start_m); ++ if (error) ++ goto fail; ++ } else { ++ vma_m = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); ++ if (!vma_m) { ++ error = -ENOMEM; ++ goto fail; ++ } ++ vma->vm_flags = newflags; ++ error = pax_mirror_vma(vma_m, vma); ++ if (error) { ++ vma->vm_flags = oldflags; ++ goto fail; ++ } ++ } ++ } ++#endif ++ + /* + * First try to merge with previous and/or next vma. + */ +@@ -204,9 +306,21 @@ success: + * vm_flags and vm_page_prot are protected by the mmap_sem + * held in write mode. + */ ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if ((mm->pax_flags & MF_PAX_SEGMEXEC) && (newflags & VM_EXEC) && ((vma->vm_flags ^ newflags) & VM_READ)) ++ pax_find_mirror_vma(vma)->vm_flags ^= VM_READ; ++#endif ++ + vma->vm_flags = newflags; ++ ++#ifdef CONFIG_PAX_MPROTECT ++ if (mm->binfmt && mm->binfmt->handle_mprotect) ++ mm->binfmt->handle_mprotect(vma, newflags); ++#endif ++ + vma->vm_page_prot = pgprot_modify(vma->vm_page_prot, +- vm_get_page_prot(newflags)); ++ vm_get_page_prot(vma->vm_flags)); + + if (vma_wants_writenotify(vma)) { + vma->vm_page_prot = vm_get_page_prot(newflags & ~VM_SHARED); +@@ -248,6 +362,17 @@ SYSCALL_DEFINE3(mprotect, unsigned long, + end = start + len; + if (end <= start) + return -ENOMEM; ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (current->mm->pax_flags & MF_PAX_SEGMEXEC) { ++ if (end > SEGMEXEC_TASK_SIZE) ++ return -EINVAL; ++ } else ++#endif ++ ++ if (end > TASK_SIZE) ++ return -EINVAL; ++ + if (!arch_validate_prot(prot)) + return -EINVAL; + +@@ -255,7 +380,7 @@ SYSCALL_DEFINE3(mprotect, unsigned long, + /* + * Does the application expect PROT_READ to imply PROT_EXEC: + */ +- if ((prot & PROT_READ) && (current->personality & READ_IMPLIES_EXEC)) ++ if ((prot & (PROT_READ | PROT_WRITE)) && (current->personality & READ_IMPLIES_EXEC)) + prot |= PROT_EXEC; + + vm_flags = calc_vm_prot_bits(prot); +@@ -287,6 +412,11 @@ SYSCALL_DEFINE3(mprotect, unsigned long, + if (start > vma->vm_start) + prev = vma; + ++#ifdef CONFIG_PAX_MPROTECT ++ if (current->mm->binfmt && current->mm->binfmt->handle_mprotect) ++ current->mm->binfmt->handle_mprotect(vma, vm_flags); ++#endif ++ + for (nstart = start ; ; ) { + unsigned long newflags; + +@@ -296,6 +426,14 @@ SYSCALL_DEFINE3(mprotect, unsigned long, + + /* newflags >> 4 shift VM_MAY% in place of VM_% */ + if ((newflags & ~(newflags >> 4)) & (VM_READ | VM_WRITE | VM_EXEC)) { ++ if (prot & (PROT_WRITE | PROT_EXEC)) ++ gr_log_rwxmprotect(vma->vm_file); ++ ++ error = -EACCES; ++ goto out; ++ } ++ ++ if (!gr_acl_handle_mprotect(vma->vm_file, prot)) { + error = -EACCES; + goto out; + } +@@ -310,6 +448,9 @@ SYSCALL_DEFINE3(mprotect, unsigned long, + error = mprotect_fixup(vma, &prev, nstart, tmp, newflags); + if (error) + goto out; ++ ++ track_exec_limit(current->mm, nstart, tmp, vm_flags); ++ + nstart = tmp; + + if (nstart < prev->vm_end) +diff -urNp linux-2.6.38.3/mm/mremap.c linux-2.6.38.3/mm/mremap.c +--- linux-2.6.38.3/mm/mremap.c 2011-04-17 16:53:01.000000000 -0400 ++++ linux-2.6.38.3/mm/mremap.c 2011-04-17 16:53:48.000000000 -0400 +@@ -114,6 +114,12 @@ static void move_ptes(struct vm_area_str + continue; + pte = ptep_clear_flush(vma, old_addr, old_pte); + pte = move_pte(pte, new_vma->vm_page_prot, old_addr, new_addr); ++ ++#ifdef CONFIG_ARCH_TRACK_EXEC_LIMIT ++ if (!(__supported_pte_mask & _PAGE_NX) && (new_vma->vm_flags & (VM_PAGEEXEC | VM_EXEC)) == VM_PAGEEXEC) ++ pte = pte_exprotect(pte); ++#endif ++ + set_pte_at(mm, new_addr, new_pte, pte); + } + +@@ -273,6 +279,11 @@ static struct vm_area_struct *vma_to_res + if (is_vm_hugetlb_page(vma)) + goto Einval; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (pax_find_mirror_vma(vma)) ++ goto Einval; ++#endif ++ + /* We can't remap across vm area boundaries */ + if (old_len > vma->vm_end - addr) + goto Efault; +@@ -329,20 +340,25 @@ static unsigned long mremap_to(unsigned + unsigned long ret = -EINVAL; + unsigned long charged = 0; + unsigned long map_flags; ++ unsigned long pax_task_size = TASK_SIZE; + + if (new_addr & ~PAGE_MASK) + goto out; + +- if (new_len > TASK_SIZE || new_addr > TASK_SIZE - new_len) ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (mm->pax_flags & MF_PAX_SEGMEXEC) ++ pax_task_size = SEGMEXEC_TASK_SIZE; ++#endif ++ ++ pax_task_size -= PAGE_SIZE; ++ ++ if (new_len > TASK_SIZE || new_addr > pax_task_size - new_len) + goto out; + + /* Check if the location we're moving into overlaps the + * old location at all, and fail if it does. + */ +- if ((new_addr <= addr) && (new_addr+new_len) > addr) +- goto out; +- +- if ((addr <= new_addr) && (addr+old_len) > new_addr) ++ if (addr + old_len > new_addr && new_addr + new_len > addr) + goto out; + + ret = security_file_mmap(NULL, 0, 0, 0, new_addr, 1); +@@ -414,6 +430,7 @@ unsigned long do_mremap(unsigned long ad + struct vm_area_struct *vma; + unsigned long ret = -EINVAL; + unsigned long charged = 0; ++ unsigned long pax_task_size = TASK_SIZE; + + if (flags & ~(MREMAP_FIXED | MREMAP_MAYMOVE)) + goto out; +@@ -432,6 +449,17 @@ unsigned long do_mremap(unsigned long ad + if (!new_len) + goto out; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (mm->pax_flags & MF_PAX_SEGMEXEC) ++ pax_task_size = SEGMEXEC_TASK_SIZE; ++#endif ++ ++ pax_task_size -= PAGE_SIZE; ++ ++ if (new_len > pax_task_size || addr > pax_task_size-new_len || ++ old_len > pax_task_size || addr > pax_task_size-old_len) ++ goto out; ++ + if (flags & MREMAP_FIXED) { + if (flags & MREMAP_MAYMOVE) + ret = mremap_to(addr, old_len, new_addr, new_len); +@@ -481,6 +509,7 @@ unsigned long do_mremap(unsigned long ad + addr + new_len); + } + ret = addr; ++ track_exec_limit(vma->vm_mm, vma->vm_start, addr + new_len, vma->vm_flags); + goto out; + } + } +@@ -507,7 +536,13 @@ unsigned long do_mremap(unsigned long ad + ret = security_file_mmap(NULL, 0, 0, 0, new_addr, 1); + if (ret) + goto out; ++ ++ map_flags = vma->vm_flags; + ret = move_vma(vma, addr, old_len, new_len, new_addr); ++ if (!(ret & ~PAGE_MASK)) { ++ track_exec_limit(current->mm, addr, addr + old_len, 0UL); ++ track_exec_limit(current->mm, new_addr, new_addr + new_len, map_flags); ++ } + } + out: + if (ret & ~PAGE_MASK) +diff -urNp linux-2.6.38.3/mm/nommu.c linux-2.6.38.3/mm/nommu.c +--- linux-2.6.38.3/mm/nommu.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/nommu.c 2011-04-17 15:57:33.000000000 -0400 +@@ -63,7 +63,6 @@ int sysctl_overcommit_memory = OVERCOMMI + int sysctl_overcommit_ratio = 50; /* default is 50% */ + int sysctl_max_map_count = DEFAULT_MAX_MAP_COUNT; + int sysctl_nr_trim_pages = CONFIG_NOMMU_INITIAL_TRIM_EXCESS; +-int heap_stack_gap = 0; + + atomic_long_t mmap_pages_allocated; + +@@ -833,15 +832,6 @@ struct vm_area_struct *find_vma(struct m + EXPORT_SYMBOL(find_vma); + + /* +- * find a VMA +- * - we don't extend stack VMAs under NOMMU conditions +- */ +-struct vm_area_struct *find_extend_vma(struct mm_struct *mm, unsigned long addr) +-{ +- return find_vma(mm, addr); +-} +- +-/* + * expand a stack to a given address + * - not supported under NOMMU conditions + */ +@@ -1563,6 +1553,7 @@ int split_vma(struct mm_struct *mm, stru + + /* most fields are the same, copy all, and then fixup */ + *new = *vma; ++ INIT_LIST_HEAD(&new->anon_vma_chain); + *region = *vma->vm_region; + new->vm_region = region; + +diff -urNp linux-2.6.38.3/mm/page_alloc.c linux-2.6.38.3/mm/page_alloc.c +--- linux-2.6.38.3/mm/page_alloc.c 2011-03-28 17:42:40.000000000 -0400 ++++ linux-2.6.38.3/mm/page_alloc.c 2011-04-17 15:57:33.000000000 -0400 +@@ -644,6 +644,10 @@ static bool free_pages_prepare(struct pa + int i; + int bad = 0; + ++#ifdef CONFIG_PAX_MEMORY_SANITIZE ++ unsigned long index = 1UL << order; ++#endif ++ + trace_mm_page_free_direct(page, order); + kmemcheck_free_shadow(page, order); + +@@ -659,6 +663,12 @@ static bool free_pages_prepare(struct pa + debug_check_no_obj_freed(page_address(page), + PAGE_SIZE << order); + } ++ ++#ifdef CONFIG_PAX_MEMORY_SANITIZE ++ for (; index; --index) ++ sanitize_highpage(page + index - 1); ++#endif ++ + arch_free_page(page, order); + kernel_map_pages(page, 1 << order, 0); + +@@ -773,8 +783,10 @@ static int prep_new_page(struct page *pa + arch_alloc_page(page, order); + kernel_map_pages(page, 1 << order, 1); + ++#ifndef CONFIG_PAX_MEMORY_SANITIZE + if (gfp_flags & __GFP_ZERO) + prep_zero_page(page, order, gfp_flags); ++#endif + + if (order && (gfp_flags & __GFP_COMP)) + prep_compound_page(page, order); +diff -urNp linux-2.6.38.3/mm/percpu.c linux-2.6.38.3/mm/percpu.c +--- linux-2.6.38.3/mm/percpu.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/percpu.c 2011-04-17 15:57:33.000000000 -0400 +@@ -121,7 +121,7 @@ static unsigned int pcpu_first_unit_cpu + static unsigned int pcpu_last_unit_cpu __read_mostly; + + /* the address of the first chunk which starts with the kernel static area */ +-void *pcpu_base_addr __read_mostly; ++void *pcpu_base_addr __read_only; + EXPORT_SYMBOL_GPL(pcpu_base_addr); + + static const int *pcpu_unit_map __read_mostly; /* cpu -> unit */ +diff -urNp linux-2.6.38.3/mm/rmap.c linux-2.6.38.3/mm/rmap.c +--- linux-2.6.38.3/mm/rmap.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/rmap.c 2011-04-17 15:57:33.000000000 -0400 +@@ -117,6 +117,10 @@ int anon_vma_prepare(struct vm_area_stru + struct anon_vma *anon_vma = vma->anon_vma; + struct anon_vma_chain *avc; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ struct anon_vma_chain *avc_m = NULL; ++#endif ++ + might_sleep(); + if (unlikely(!anon_vma)) { + struct mm_struct *mm = vma->vm_mm; +@@ -126,6 +130,12 @@ int anon_vma_prepare(struct vm_area_stru + if (!avc) + goto out_enomem; + ++#ifdef CONFIG_PAX_SEGMEXEC ++ avc_m = anon_vma_chain_alloc(); ++ if (!avc_m) ++ goto out_enomem_free_avc; ++#endif ++ + anon_vma = find_mergeable_anon_vma(vma); + allocated = NULL; + if (!anon_vma) { +@@ -144,6 +154,21 @@ int anon_vma_prepare(struct vm_area_stru + /* page_table_lock to protect against threads */ + spin_lock(&mm->page_table_lock); + if (likely(!vma->anon_vma)) { ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ struct vm_area_struct *vma_m = pax_find_mirror_vma(vma); ++ ++ if (vma_m) { ++ BUG_ON(vma_m->anon_vma); ++ vma_m->anon_vma = anon_vma; ++ avc_m->anon_vma = anon_vma; ++ avc_m->vma = vma; ++ list_add(&avc_m->same_vma, &vma_m->anon_vma_chain); ++ list_add(&avc_m->same_anon_vma, &anon_vma->head); ++ avc_m = NULL; ++ } ++#endif ++ + vma->anon_vma = anon_vma; + avc->anon_vma = anon_vma; + avc->vma = vma; +@@ -157,12 +182,24 @@ int anon_vma_prepare(struct vm_area_stru + + if (unlikely(allocated)) + anon_vma_free(allocated); ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (unlikely(avc_m)) ++ anon_vma_chain_free(avc_m); ++#endif ++ + if (unlikely(avc)) + anon_vma_chain_free(avc); + } + return 0; + + out_enomem_free_avc: ++ ++#ifdef CONFIG_PAX_SEGMEXEC ++ if (avc_m) ++ anon_vma_chain_free(avc_m); ++#endif ++ + anon_vma_chain_free(avc); + out_enomem: + return -ENOMEM; +@@ -189,7 +226,7 @@ static void anon_vma_chain_link(struct v + * Attach the anon_vmas from src to dst. + * Returns 0 on success, -ENOMEM on failure. + */ +-int anon_vma_clone(struct vm_area_struct *dst, struct vm_area_struct *src) ++int anon_vma_clone(struct vm_area_struct *dst, const struct vm_area_struct *src) + { + struct anon_vma_chain *avc, *pavc; + +@@ -211,7 +248,7 @@ int anon_vma_clone(struct vm_area_struct + * the corresponding VMA in the parent process is attached to. + * Returns 0 on success, non-zero on failure. + */ +-int anon_vma_fork(struct vm_area_struct *vma, struct vm_area_struct *pvma) ++int anon_vma_fork(struct vm_area_struct *vma, const struct vm_area_struct *pvma) + { + struct anon_vma_chain *avc; + struct anon_vma *anon_vma; +diff -urNp linux-2.6.38.3/mm/shmem.c linux-2.6.38.3/mm/shmem.c +--- linux-2.6.38.3/mm/shmem.c 2011-03-28 17:42:40.000000000 -0400 ++++ linux-2.6.38.3/mm/shmem.c 2011-04-17 15:57:33.000000000 -0400 +@@ -31,7 +31,7 @@ + #include <linux/percpu_counter.h> + #include <linux/swap.h> + +-static struct vfsmount *shm_mnt; ++struct vfsmount *shm_mnt; + + #ifdef CONFIG_SHMEM + /* +@@ -1070,6 +1070,8 @@ static int shmem_writepage(struct page * + goto unlock; + } + entry = shmem_swp_entry(info, index, NULL); ++ if (!entry) ++ goto unlock; + if (entry->val) { + /* + * The more uptodate page coming down from a stacked +diff -urNp linux-2.6.38.3/mm/slab.c linux-2.6.38.3/mm/slab.c +--- linux-2.6.38.3/mm/slab.c 2011-03-28 17:42:40.000000000 -0400 ++++ linux-2.6.38.3/mm/slab.c 2011-04-17 15:57:33.000000000 -0400 +@@ -284,7 +284,7 @@ struct kmem_list3 { + * Need this for bootstrapping a per node allocator. + */ + #define NUM_INIT_LISTS (3 * MAX_NUMNODES) +-static struct kmem_list3 __initdata initkmem_list3[NUM_INIT_LISTS]; ++static struct kmem_list3 initkmem_list3[NUM_INIT_LISTS]; + #define CACHE_CACHE 0 + #define SIZE_AC MAX_NUMNODES + #define SIZE_L3 (2 * MAX_NUMNODES) +@@ -534,7 +534,7 @@ static inline void *index_to_obj(struct + * reciprocal_divide(offset, cache->reciprocal_buffer_size) + */ + static inline unsigned int obj_to_index(const struct kmem_cache *cache, +- const struct slab *slab, void *obj) ++ const struct slab *slab, const void *obj) + { + u32 offset = (obj - slab->s_mem); + return reciprocal_divide(offset, cache->reciprocal_buffer_size); +@@ -560,14 +560,14 @@ struct cache_names { + static struct cache_names __initdata cache_names[] = { + #define CACHE(x) { .name = "size-" #x, .name_dma = "size-" #x "(DMA)" }, + #include <linux/kmalloc_sizes.h> +- {NULL,} ++ {NULL, NULL} + #undef CACHE + }; + + static struct arraycache_init initarray_cache __initdata = +- { {0, BOOT_CPUCACHE_ENTRIES, 1, 0} }; ++ { {0, BOOT_CPUCACHE_ENTRIES, 1, 0}, {NULL} }; + static struct arraycache_init initarray_generic = +- { {0, BOOT_CPUCACHE_ENTRIES, 1, 0} }; ++ { {0, BOOT_CPUCACHE_ENTRIES, 1, 0}, {NULL} }; + + /* internal cache of cache description objs */ + static struct kmem_cache cache_cache = { +@@ -1552,7 +1552,7 @@ void __init kmem_cache_init(void) + sizes->cs_cachep = kmem_cache_create(names->name, + sizes->cs_size, + ARCH_KMALLOC_MINALIGN, +- ARCH_KMALLOC_FLAGS|SLAB_PANIC, ++ ARCH_KMALLOC_FLAGS|SLAB_PANIC|SLAB_USERCOPY, + NULL); + } + #ifdef CONFIG_ZONE_DMA +@@ -4535,15 +4535,66 @@ static const struct file_operations proc + + static int __init slab_proc_init(void) + { +- proc_create("slabinfo",S_IWUSR|S_IRUGO,NULL,&proc_slabinfo_operations); ++ mode_t gr_mode = S_IRUGO; ++ ++#ifdef CONFIG_GRKERNSEC_PROC_ADD ++ gr_mode = S_IRUSR; ++#endif ++ ++ proc_create("slabinfo",S_IWUSR|gr_mode,NULL,&proc_slabinfo_operations); + #ifdef CONFIG_DEBUG_SLAB_LEAK +- proc_create("slab_allocators", 0, NULL, &proc_slabstats_operations); ++ proc_create("slab_allocators", gr_mode, NULL, &proc_slabstats_operations); + #endif + return 0; + } + module_init(slab_proc_init); + #endif + ++void check_object_size(const void *ptr, unsigned long n, bool to) ++{ ++ ++#ifdef CONFIG_PAX_USERCOPY ++ struct page *page; ++ struct kmem_cache *cachep = NULL; ++ struct slab *slabp; ++ unsigned int objnr; ++ unsigned long offset; ++ ++ if (!n) ++ return; ++ ++ if (ZERO_OR_NULL_PTR(ptr)) ++ goto report; ++ ++ if (!virt_addr_valid(ptr)) ++ return; ++ ++ page = virt_to_head_page(ptr); ++ ++ if (!PageSlab(page)) { ++ if (object_is_on_stack(ptr, n) == -1) ++ goto report; ++ return; ++ } ++ ++ cachep = page_get_cache(page); ++ if (!(cachep->flags & SLAB_USERCOPY)) ++ goto report; ++ ++ slabp = page_get_slab(page); ++ objnr = obj_to_index(cachep, slabp, ptr); ++ BUG_ON(objnr >= cachep->num); ++ offset = ptr - index_to_obj(cachep, slabp, objnr) - obj_offset(cachep); ++ if (offset <= obj_size(cachep) && n <= obj_size(cachep) - offset) ++ return; ++ ++report: ++ pax_report_usercopy(ptr, n, to, cachep ? cachep->name : NULL); ++#endif ++ ++} ++EXPORT_SYMBOL(check_object_size); ++ + /** + * ksize - get the actual amount of memory allocated for a given object + * @objp: Pointer to the object +diff -urNp linux-2.6.38.3/mm/slob.c linux-2.6.38.3/mm/slob.c +--- linux-2.6.38.3/mm/slob.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/slob.c 2011-04-17 15:57:33.000000000 -0400 +@@ -29,7 +29,7 @@ + * If kmalloc is asked for objects of PAGE_SIZE or larger, it calls + * alloc_pages() directly, allocating compound pages so the page order + * does not have to be separately tracked, and also stores the exact +- * allocation size in page->private so that it can be used to accurately ++ * allocation size in slob_page->size so that it can be used to accurately + * provide ksize(). These objects are detected in kfree() because slob_page() + * is false for them. + * +@@ -58,6 +58,7 @@ + */ + + #include <linux/kernel.h> ++#include <linux/sched.h> + #include <linux/slab.h> + #include <linux/mm.h> + #include <linux/swap.h> /* struct reclaim_state */ +@@ -102,7 +103,8 @@ struct slob_page { + unsigned long flags; /* mandatory */ + atomic_t _count; /* mandatory */ + slobidx_t units; /* free units left in page */ +- unsigned long pad[2]; ++ unsigned long pad[1]; ++ unsigned long size; /* size when >=PAGE_SIZE */ + slob_t *free; /* first free slob_t in page */ + struct list_head list; /* linked list of free pages */ + }; +@@ -135,7 +137,7 @@ static LIST_HEAD(free_slob_large); + */ + static inline int is_slob_page(struct slob_page *sp) + { +- return PageSlab((struct page *)sp); ++ return PageSlab((struct page *)sp) && !sp->size; + } + + static inline void set_slob_page(struct slob_page *sp) +@@ -150,7 +152,7 @@ static inline void clear_slob_page(struc + + static inline struct slob_page *slob_page(const void *addr) + { +- return (struct slob_page *)virt_to_page(addr); ++ return (struct slob_page *)virt_to_head_page(addr); + } + + /* +@@ -210,7 +212,7 @@ static void set_slob(slob_t *s, slobidx_ + /* + * Return the size of a slob block. + */ +-static slobidx_t slob_units(slob_t *s) ++static slobidx_t slob_units(const slob_t *s) + { + if (s->units > 0) + return s->units; +@@ -220,7 +222,7 @@ static slobidx_t slob_units(slob_t *s) + /* + * Return the next free slob block pointer after this one. + */ +-static slob_t *slob_next(slob_t *s) ++static slob_t *slob_next(const slob_t *s) + { + slob_t *base = (slob_t *)((unsigned long)s & PAGE_MASK); + slobidx_t next; +@@ -235,7 +237,7 @@ static slob_t *slob_next(slob_t *s) + /* + * Returns true if s is the last free block in its page. + */ +-static int slob_last(slob_t *s) ++static int slob_last(const slob_t *s) + { + return !((unsigned long)slob_next(s) & ~PAGE_MASK); + } +@@ -254,6 +256,7 @@ static void *slob_new_pages(gfp_t gfp, i + if (!page) + return NULL; + ++ set_slob_page(page); + return page_address(page); + } + +@@ -370,11 +373,11 @@ static void *slob_alloc(size_t size, gfp + if (!b) + return NULL; + sp = slob_page(b); +- set_slob_page(sp); + + spin_lock_irqsave(&slob_lock, flags); + sp->units = SLOB_UNITS(PAGE_SIZE); + sp->free = b; ++ sp->size = 0; + INIT_LIST_HEAD(&sp->list); + set_slob(b, SLOB_UNITS(PAGE_SIZE), b + SLOB_UNITS(PAGE_SIZE)); + set_slob_page_free(sp, slob_list); +@@ -476,10 +479,9 @@ out: + * End of slob allocator proper. Begin kmem_cache_alloc and kmalloc frontend. + */ + +-void *__kmalloc_node(size_t size, gfp_t gfp, int node) ++static void *__kmalloc_node_align(size_t size, gfp_t gfp, int node, int align) + { +- unsigned int *m; +- int align = max(ARCH_KMALLOC_MINALIGN, ARCH_SLAB_MINALIGN); ++ slob_t *m; + void *ret; + + lockdep_trace_alloc(gfp); +@@ -492,7 +494,10 @@ void *__kmalloc_node(size_t size, gfp_t + + if (!m) + return NULL; +- *m = size; ++ BUILD_BUG_ON(ARCH_KMALLOC_MINALIGN < 2 * SLOB_UNIT); ++ BUILD_BUG_ON(ARCH_SLAB_MINALIGN < 2 * SLOB_UNIT); ++ m[0].units = size; ++ m[1].units = align; + ret = (void *)m + align; + + trace_kmalloc_node(_RET_IP_, ret, +@@ -504,9 +509,9 @@ void *__kmalloc_node(size_t size, gfp_t + gfp |= __GFP_COMP; + ret = slob_new_pages(gfp, order, node); + if (ret) { +- struct page *page; +- page = virt_to_page(ret); +- page->private = size; ++ struct slob_page *sp; ++ sp = slob_page(ret); ++ sp->size = size; + } + + trace_kmalloc_node(_RET_IP_, ret, +@@ -516,6 +521,13 @@ void *__kmalloc_node(size_t size, gfp_t + kmemleak_alloc(ret, size, 1, gfp); + return ret; + } ++ ++void *__kmalloc_node(size_t size, gfp_t gfp, int node) ++{ ++ int align = max(ARCH_KMALLOC_MINALIGN, ARCH_SLAB_MINALIGN); ++ ++ return __kmalloc_node_align(size, gfp, node, align); ++} + EXPORT_SYMBOL(__kmalloc_node); + + void kfree(const void *block) +@@ -531,13 +543,81 @@ void kfree(const void *block) + sp = slob_page(block); + if (is_slob_page(sp)) { + int align = max(ARCH_KMALLOC_MINALIGN, ARCH_SLAB_MINALIGN); +- unsigned int *m = (unsigned int *)(block - align); +- slob_free(m, *m + align); +- } else ++ slob_t *m = (slob_t *)(block - align); ++ slob_free(m, m[0].units + align); ++ } else { ++ clear_slob_page(sp); ++ free_slob_page(sp); ++ sp->size = 0; + put_page(&sp->page); ++ } + } + EXPORT_SYMBOL(kfree); + ++void check_object_size(const void *ptr, unsigned long n, bool to) ++{ ++ ++#ifdef CONFIG_PAX_USERCOPY ++ struct slob_page *sp; ++ const slob_t *free; ++ const void *base; ++ ++ if (!n) ++ return; ++ ++ if (ZERO_OR_NULL_PTR(ptr)) ++ goto report; ++ ++ if (!virt_addr_valid(ptr)) ++ return; ++ ++ sp = slob_page(ptr); ++ if (!PageSlab((struct page*)sp)) { ++ if (object_is_on_stack(ptr, n) == -1) ++ goto report; ++ return; ++ } ++ ++ if (sp->size) { ++ base = page_address(&sp->page); ++ if (base <= ptr && n <= sp->size - (ptr - base)) ++ return; ++ goto report; ++ } ++ ++ /* some tricky double walking to find the chunk */ ++ base = (void *)((unsigned long)ptr & PAGE_MASK); ++ free = sp->free; ++ ++ while (!slob_last(free) && (void *)free <= ptr) { ++ base = free + slob_units(free); ++ free = slob_next(free); ++ } ++ ++ while (base < (void *)free) { ++ slobidx_t m = ((slob_t *)base)[0].units, align = ((slob_t *)base)[1].units; ++ int size = SLOB_UNIT * SLOB_UNITS(m + align); ++ int offset; ++ ++ if (ptr < base + align) ++ goto report; ++ ++ offset = ptr - base - align; ++ if (offset < m) { ++ if (n <= m - offset) ++ return; ++ goto report; ++ } ++ base += size; ++ } ++ ++report: ++ pax_report_usercopy(ptr, n, to, NULL); ++#endif ++ ++} ++EXPORT_SYMBOL(check_object_size); ++ + /* can't use ksize for kmem_cache_alloc memory, only kmalloc */ + size_t ksize(const void *block) + { +@@ -550,10 +630,10 @@ size_t ksize(const void *block) + sp = slob_page(block); + if (is_slob_page(sp)) { + int align = max(ARCH_KMALLOC_MINALIGN, ARCH_SLAB_MINALIGN); +- unsigned int *m = (unsigned int *)(block - align); +- return SLOB_UNITS(*m) * SLOB_UNIT; ++ slob_t *m = (slob_t *)(block - align); ++ return SLOB_UNITS(m[0].units) * SLOB_UNIT; + } else +- return sp->page.private; ++ return sp->size; + } + EXPORT_SYMBOL(ksize); + +@@ -608,17 +688,25 @@ void *kmem_cache_alloc_node(struct kmem_ + { + void *b; + ++#ifdef CONFIG_PAX_USERCOPY ++ b = __kmalloc_node_align(c->size, flags, node, c->align); ++#else + if (c->size < PAGE_SIZE) { + b = slob_alloc(c->size, flags, c->align, node); + trace_kmem_cache_alloc_node(_RET_IP_, b, c->size, + SLOB_UNITS(c->size) * SLOB_UNIT, + flags, node); + } else { ++ struct slob_page *sp; ++ + b = slob_new_pages(flags, get_order(c->size), node); ++ sp = slob_page(b); ++ sp->size = c->size; + trace_kmem_cache_alloc_node(_RET_IP_, b, c->size, + PAGE_SIZE << get_order(c->size), + flags, node); + } ++#endif + + if (c->ctor) + c->ctor(b); +@@ -630,10 +718,16 @@ EXPORT_SYMBOL(kmem_cache_alloc_node); + + static void __kmem_cache_free(void *b, int size) + { +- if (size < PAGE_SIZE) ++ struct slob_page *sp = slob_page(b); ++ ++ if (is_slob_page(sp)) + slob_free(b, size); +- else ++ else { ++ clear_slob_page(sp); ++ free_slob_page(sp); ++ sp->size = 0; + slob_free_pages(b, get_order(size)); ++ } + } + + static void kmem_rcu_free(struct rcu_head *head) +@@ -646,14 +740,23 @@ static void kmem_rcu_free(struct rcu_hea + + void kmem_cache_free(struct kmem_cache *c, void *b) + { ++ int size = c->size; ++ ++#ifdef CONFIG_PAX_USERCOPY ++ if (size + c->align < PAGE_SIZE) { ++ size += c->align; ++ b -= c->align; ++ } ++#endif ++ + kmemleak_free_recursive(b, c->flags); + if (unlikely(c->flags & SLAB_DESTROY_BY_RCU)) { + struct slob_rcu *slob_rcu; +- slob_rcu = b + (c->size - sizeof(struct slob_rcu)); +- slob_rcu->size = c->size; ++ slob_rcu = b + (size - sizeof(struct slob_rcu)); ++ slob_rcu->size = size; + call_rcu(&slob_rcu->head, kmem_rcu_free); + } else { +- __kmem_cache_free(b, c->size); ++ __kmem_cache_free(b, size); + } + + trace_kmem_cache_free(_RET_IP_, b); +diff -urNp linux-2.6.38.3/mm/slub.c linux-2.6.38.3/mm/slub.c +--- linux-2.6.38.3/mm/slub.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/slub.c 2011-04-17 15:57:33.000000000 -0400 +@@ -390,7 +390,7 @@ static void print_track(const char *s, s + if (!t->addr) + return; + +- printk(KERN_ERR "INFO: %s in %pS age=%lu cpu=%u pid=%d\n", ++ printk(KERN_ERR "INFO: %s in %pA age=%lu cpu=%u pid=%d\n", + s, (void *)t->addr, jiffies - t->when, t->cpu, t->pid); + } + +@@ -1927,6 +1927,8 @@ void kmem_cache_free(struct kmem_cache * + + page = virt_to_head_page(x); + ++ BUG_ON(!PageSlab(page)); ++ + slab_free(s, page, x, _RET_IP_); + + trace_kmem_cache_free(_RET_IP_, x); +@@ -1960,7 +1962,7 @@ static int slub_min_objects; + * Merge control. If this is set then no merging of slab caches will occur. + * (Could be removed. This was introduced to pacify the merge skeptics.) + */ +-static int slub_nomerge; ++static int slub_nomerge = 1; + + /* + * Calculate the order of allocation given an slab object size. +@@ -2370,7 +2372,7 @@ static int kmem_cache_open(struct kmem_c + * list to avoid pounding the page allocator excessively. + */ + set_min_partial(s, ilog2(s->size)); +- s->refcount = 1; ++ atomic_set(&s->refcount, 1); + #ifdef CONFIG_NUMA + s->remote_node_defrag_ratio = 1000; + #endif +@@ -2482,8 +2484,7 @@ static inline int kmem_cache_close(struc + void kmem_cache_destroy(struct kmem_cache *s) + { + down_write(&slub_lock); +- s->refcount--; +- if (!s->refcount) { ++ if (atomic_dec_and_test(&s->refcount)) { + list_del(&s->list); + if (kmem_cache_close(s)) { + printk(KERN_ERR "SLUB %s: %s called for cache that " +@@ -2693,6 +2694,46 @@ void *__kmalloc_node(size_t size, gfp_t + EXPORT_SYMBOL(__kmalloc_node); + #endif + ++void check_object_size(const void *ptr, unsigned long n, bool to) ++{ ++ ++#ifdef CONFIG_PAX_USERCOPY ++ struct page *page; ++ struct kmem_cache *s = NULL; ++ unsigned long offset; ++ ++ if (!n) ++ return; ++ ++ if (ZERO_OR_NULL_PTR(ptr)) ++ goto report; ++ ++ if (!virt_addr_valid(ptr)) ++ return; ++ ++ page = virt_to_head_page(ptr); ++ ++ if (!PageSlab(page)) { ++ if (object_is_on_stack(ptr, n) == -1) ++ goto report; ++ return; ++ } ++ ++ s = page->slab; ++ if (!(s->flags & SLAB_USERCOPY)) ++ goto report; ++ ++ offset = (ptr - page_address(page)) % s->size; ++ if (offset <= s->objsize && n <= s->objsize - offset) ++ return; ++ ++report: ++ pax_report_usercopy(ptr, n, to, s ? s->name : NULL); ++#endif ++ ++} ++EXPORT_SYMBOL(check_object_size); ++ + size_t ksize(const void *object) + { + struct page *page; +@@ -2958,7 +2999,7 @@ static void __init kmem_cache_bootstrap_ + int node; + + list_add(&s->list, &slab_caches); +- s->refcount = -1; ++ atomic_set(&s->refcount, -1); + + for_each_node_state(node, N_NORMAL_MEMORY) { + struct kmem_cache_node *n = get_node(s, node); +@@ -3075,17 +3116,17 @@ void __init kmem_cache_init(void) + + /* Caches that are not of the two-to-the-power-of size */ + if (KMALLOC_MIN_SIZE <= 32) { +- kmalloc_caches[1] = create_kmalloc_cache("kmalloc-96", 96, 0); ++ kmalloc_caches[1] = create_kmalloc_cache("kmalloc-96", 96, SLAB_USERCOPY); + caches++; + } + + if (KMALLOC_MIN_SIZE <= 64) { +- kmalloc_caches[2] = create_kmalloc_cache("kmalloc-192", 192, 0); ++ kmalloc_caches[2] = create_kmalloc_cache("kmalloc-192", 192, SLAB_USERCOPY); + caches++; + } + + for (i = KMALLOC_SHIFT_LOW; i < SLUB_PAGE_SHIFT; i++) { +- kmalloc_caches[i] = create_kmalloc_cache("kmalloc", 1 << i, 0); ++ kmalloc_caches[i] = create_kmalloc_cache("kmalloc", 1 << i, SLAB_USERCOPY); + caches++; + } + +@@ -3153,7 +3194,7 @@ static int slab_unmergeable(struct kmem_ + /* + * We may have set a slab to be unmergeable during bootstrap. + */ +- if (s->refcount < 0) ++ if (atomic_read(&s->refcount) < 0) + return 1; + + return 0; +@@ -3212,7 +3253,7 @@ struct kmem_cache *kmem_cache_create(con + down_write(&slub_lock); + s = find_mergeable(size, align, flags, name, ctor); + if (s) { +- s->refcount++; ++ atomic_inc(&s->refcount); + /* + * Adjust the object sizes so that we clear + * the complete object on kzalloc. +@@ -3221,7 +3262,7 @@ struct kmem_cache *kmem_cache_create(con + s->inuse = max_t(int, s->inuse, ALIGN(size, sizeof(void *))); + + if (sysfs_slab_alias(s, name)) { +- s->refcount--; ++ atomic_dec(&s->refcount); + goto err; + } + up_write(&slub_lock); +@@ -3954,7 +3995,7 @@ SLAB_ATTR_RO(ctor); + + static ssize_t aliases_show(struct kmem_cache *s, char *buf) + { +- return sprintf(buf, "%d\n", s->refcount - 1); ++ return sprintf(buf, "%d\n", atomic_read(&s->refcount) - 1); + } + SLAB_ATTR_RO(aliases); + +@@ -4691,7 +4732,13 @@ static const struct file_operations proc + + static int __init slab_proc_init(void) + { +- proc_create("slabinfo", S_IRUGO, NULL, &proc_slabinfo_operations); ++ mode_t gr_mode = S_IRUGO; ++ ++#ifdef CONFIG_GRKERNSEC_PROC_ADD ++ gr_mode = S_IRUSR; ++#endif ++ ++ proc_create("slabinfo", gr_mode, NULL, &proc_slabinfo_operations); + return 0; + } + module_init(slab_proc_init); +diff -urNp linux-2.6.38.3/mm/util.c linux-2.6.38.3/mm/util.c +--- linux-2.6.38.3/mm/util.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/util.c 2011-04-17 15:57:33.000000000 -0400 +@@ -219,6 +219,12 @@ EXPORT_SYMBOL(strndup_user); + void arch_pick_mmap_layout(struct mm_struct *mm) + { + mm->mmap_base = TASK_UNMAPPED_BASE; ++ ++#ifdef CONFIG_PAX_RANDMMAP ++ if (mm->pax_flags & MF_PAX_RANDMMAP) ++ mm->mmap_base += mm->delta_mmap; ++#endif ++ + mm->get_unmapped_area = arch_get_unmapped_area; + mm->unmap_area = arch_unmap_area; + } +diff -urNp linux-2.6.38.3/mm/vmalloc.c linux-2.6.38.3/mm/vmalloc.c +--- linux-2.6.38.3/mm/vmalloc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/vmalloc.c 2011-04-17 15:57:33.000000000 -0400 +@@ -39,8 +39,19 @@ static void vunmap_pte_range(pmd_t *pmd, + + pte = pte_offset_kernel(pmd, addr); + do { +- pte_t ptent = ptep_get_and_clear(&init_mm, addr, pte); +- WARN_ON(!pte_none(ptent) && !pte_present(ptent)); ++ ++#if defined(CONFIG_MODULES) && defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) ++ if ((unsigned long)MODULES_EXEC_VADDR <= addr && addr < (unsigned long)MODULES_EXEC_END) { ++ BUG_ON(!pte_exec(*pte)); ++ set_pte_at(&init_mm, addr, pte, pfn_pte(__pa(addr) >> PAGE_SHIFT, PAGE_KERNEL_EXEC)); ++ continue; ++ } ++#endif ++ ++ { ++ pte_t ptent = ptep_get_and_clear(&init_mm, addr, pte); ++ WARN_ON(!pte_none(ptent) && !pte_present(ptent)); ++ } + } while (pte++, addr += PAGE_SIZE, addr != end); + } + +@@ -91,6 +102,7 @@ static int vmap_pte_range(pmd_t *pmd, un + unsigned long end, pgprot_t prot, struct page **pages, int *nr) + { + pte_t *pte; ++ int ret = -ENOMEM; + + /* + * nr is a running index into the array which helps higher level +@@ -100,17 +112,30 @@ static int vmap_pte_range(pmd_t *pmd, un + pte = pte_alloc_kernel(pmd, addr); + if (!pte) + return -ENOMEM; ++ ++ pax_open_kernel(); + do { + struct page *page = pages[*nr]; + +- if (WARN_ON(!pte_none(*pte))) +- return -EBUSY; +- if (WARN_ON(!page)) +- return -ENOMEM; ++#if defined(CONFIG_MODULES) && defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) ++ if (pgprot_val(prot) & _PAGE_NX) ++#endif ++ ++ if (WARN_ON(!pte_none(*pte))) { ++ ret = -EBUSY; ++ goto out; ++ } ++ if (WARN_ON(!page)) { ++ ret = -ENOMEM; ++ goto out; ++ } + set_pte_at(&init_mm, addr, pte, mk_pte(page, prot)); + (*nr)++; + } while (pte++, addr += PAGE_SIZE, addr != end); +- return 0; ++ ret = 0; ++out: ++ pax_close_kernel(); ++ return ret; + } + + static int vmap_pmd_range(pud_t *pud, unsigned long addr, +@@ -191,11 +216,20 @@ int is_vmalloc_or_module_addr(const void + * and fall back on vmalloc() if that fails. Others + * just put it in the vmalloc space. + */ +-#if defined(CONFIG_MODULES) && defined(MODULES_VADDR) ++#ifdef CONFIG_MODULES ++#ifdef MODULES_VADDR + unsigned long addr = (unsigned long)x; + if (addr >= MODULES_VADDR && addr < MODULES_END) + return 1; + #endif ++ ++#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) ++ if (x >= (const void *)MODULES_EXEC_VADDR && x < (const void *)MODULES_EXEC_END) ++ return 1; ++#endif ++ ++#endif ++ + return is_vmalloc_addr(x); + } + +@@ -216,8 +250,14 @@ struct page *vmalloc_to_page(const void + + if (!pgd_none(*pgd)) { + pud_t *pud = pud_offset(pgd, addr); ++#ifdef CONFIG_X86 ++ if (!pud_large(*pud)) ++#endif + if (!pud_none(*pud)) { + pmd_t *pmd = pmd_offset(pud, addr); ++#ifdef CONFIG_X86 ++ if (!pmd_large(*pmd)) ++#endif + if (!pmd_none(*pmd)) { + pte_t *ptep, pte; + +@@ -1244,6 +1284,16 @@ static struct vm_struct *__get_vm_area_n + struct vm_struct *area; + + BUG_ON(in_interrupt()); ++ ++#if defined(CONFIG_MODULES) && defined(CONFIG_X86) && defined(CONFIG_PAX_KERNEXEC) ++ if (flags & VM_KERNEXEC) { ++ if (start != VMALLOC_START || end != VMALLOC_END) ++ return NULL; ++ start = (unsigned long)MODULES_EXEC_VADDR; ++ end = (unsigned long)MODULES_EXEC_END; ++ } ++#endif ++ + if (flags & VM_IOREMAP) { + int bit = fls(size); + +@@ -1462,6 +1512,11 @@ void *vmap(struct page **pages, unsigned + if (count > totalram_pages) + return NULL; + ++#if defined(CONFIG_MODULES) && defined(CONFIG_X86) && defined(CONFIG_PAX_KERNEXEC) ++ if (!(pgprot_val(prot) & _PAGE_NX)) ++ flags |= VM_KERNEXEC; ++#endif ++ + area = get_vm_area_caller((count << PAGE_SHIFT), flags, + __builtin_return_address(0)); + if (!area) +@@ -1558,6 +1613,13 @@ void *__vmalloc_node_range(unsigned long + if (!size || (size >> PAGE_SHIFT) > totalram_pages) + return NULL; + ++#if defined(CONFIG_MODULES) && defined(CONFIG_X86) && defined(CONFIG_PAX_KERNEXEC) ++ if (!(pgprot_val(prot) & _PAGE_NX)) ++ area = __get_vm_area_node(size, align, VM_ALLOC | VM_KERNEXEC, VMALLOC_START, VMALLOC_END, ++ node, gfp_mask, caller); ++ else ++#endif ++ + area = __get_vm_area_node(size, align, VM_ALLOC, start, end, node, + gfp_mask, caller); + +@@ -1597,6 +1659,7 @@ static void *__vmalloc_node(unsigned lon + gfp_mask, prot, node, caller); + } + ++#undef __vmalloc + void *__vmalloc(unsigned long size, gfp_t gfp_mask, pgprot_t prot) + { + return __vmalloc_node(size, 1, gfp_mask, prot, -1, +@@ -1620,6 +1683,7 @@ static inline void *__vmalloc_node_flags + * For tight control over page level allocator and protection flags + * use __vmalloc() instead. + */ ++#undef vmalloc + void *vmalloc(unsigned long size) + { + return __vmalloc_node_flags(size, -1, GFP_KERNEL | __GFP_HIGHMEM); +@@ -1636,6 +1700,7 @@ EXPORT_SYMBOL(vmalloc); + * For tight control over page level allocator and protection flags + * use __vmalloc() instead. + */ ++#undef vzalloc + void *vzalloc(unsigned long size) + { + return __vmalloc_node_flags(size, -1, +@@ -1650,6 +1715,7 @@ EXPORT_SYMBOL(vzalloc); + * The resulting memory area is zeroed so it can be mapped to userspace + * without leaking data. + */ ++#undef vmalloc_user + void *vmalloc_user(unsigned long size) + { + struct vm_struct *area; +@@ -1677,6 +1743,7 @@ EXPORT_SYMBOL(vmalloc_user); + * For tight control over page level allocator and protection flags + * use __vmalloc() instead. + */ ++#undef vmalloc_node + void *vmalloc_node(unsigned long size, int node) + { + return __vmalloc_node(size, 1, GFP_KERNEL | __GFP_HIGHMEM, PAGE_KERNEL, +@@ -1696,6 +1763,7 @@ EXPORT_SYMBOL(vmalloc_node); + * For tight control over page level allocator and protection flags + * use __vmalloc_node() instead. + */ ++#undef vzalloc_node + void *vzalloc_node(unsigned long size, int node) + { + return __vmalloc_node_flags(size, node, +@@ -1718,10 +1786,10 @@ EXPORT_SYMBOL(vzalloc_node); + * For tight control over page level allocator and protection flags + * use __vmalloc() instead. + */ +- ++#undef vmalloc_exec + void *vmalloc_exec(unsigned long size) + { +- return __vmalloc_node(size, 1, GFP_KERNEL | __GFP_HIGHMEM, PAGE_KERNEL_EXEC, ++ return __vmalloc_node(size, 1, GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO, PAGE_KERNEL_EXEC, + -1, __builtin_return_address(0)); + } + +@@ -1740,6 +1808,7 @@ void *vmalloc_exec(unsigned long size) + * Allocate enough 32bit PA addressable pages to cover @size from the + * page level allocator and map them into contiguous kernel virtual space. + */ ++#undef vmalloc_32 + void *vmalloc_32(unsigned long size) + { + return __vmalloc_node(size, 1, GFP_VMALLOC32, PAGE_KERNEL, +@@ -1754,6 +1823,7 @@ EXPORT_SYMBOL(vmalloc_32); + * The resulting memory area is 32bit addressable and zeroed so it can be + * mapped to userspace without leaking data. + */ ++#undef vmalloc_32_user + void *vmalloc_32_user(unsigned long size) + { + struct vm_struct *area; +@@ -2018,6 +2088,8 @@ int remap_vmalloc_range(struct vm_area_s + unsigned long uaddr = vma->vm_start; + unsigned long usize = vma->vm_end - vma->vm_start; + ++ BUG_ON(vma->vm_mirror); ++ + if ((PAGE_SIZE-1) & (unsigned long)addr) + return -EINVAL; + +diff -urNp linux-2.6.38.3/mm/vmstat.c linux-2.6.38.3/mm/vmstat.c +--- linux-2.6.38.3/mm/vmstat.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/mm/vmstat.c 2011-04-17 15:57:33.000000000 -0400 +@@ -78,7 +78,7 @@ void vm_events_fold_cpu(int cpu) + * + * vm_stat contains the global counters + */ +-atomic_long_t vm_stat[NR_VM_ZONE_STAT_ITEMS]; ++atomic_long_unchecked_t vm_stat[NR_VM_ZONE_STAT_ITEMS]; + EXPORT_SYMBOL(vm_stat); + + #ifdef CONFIG_SMP +@@ -451,7 +451,7 @@ void refresh_cpu_vm_stats(int cpu) + v = p->vm_stat_diff[i]; + p->vm_stat_diff[i] = 0; + local_irq_restore(flags); +- atomic_long_add(v, &zone->vm_stat[i]); ++ atomic_long_add_unchecked(v, &zone->vm_stat[i]); + global_diff[i] += v; + #ifdef CONFIG_NUMA + /* 3 seconds idle till flush */ +@@ -489,7 +489,7 @@ void refresh_cpu_vm_stats(int cpu) + + for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++) + if (global_diff[i]) +- atomic_long_add(global_diff[i], &vm_stat[i]); ++ atomic_long_add_unchecked(global_diff[i], &vm_stat[i]); + } + + #endif +@@ -1188,10 +1188,20 @@ static int __init setup_vmstat(void) + start_cpu_timer(cpu); + #endif + #ifdef CONFIG_PROC_FS +- proc_create("buddyinfo", S_IRUGO, NULL, &fragmentation_file_operations); +- proc_create("pagetypeinfo", S_IRUGO, NULL, &pagetypeinfo_file_ops); +- proc_create("vmstat", S_IRUGO, NULL, &proc_vmstat_file_operations); +- proc_create("zoneinfo", S_IRUGO, NULL, &proc_zoneinfo_file_operations); ++ { ++ mode_t gr_mode = S_IRUGO; ++#ifdef CONFIG_GRKERNSEC_PROC_ADD ++ gr_mode = S_IRUSR; ++#endif ++ proc_create("buddyinfo", gr_mode, NULL, &fragmentation_file_operations); ++ proc_create("pagetypeinfo", gr_mode, NULL, &pagetypeinfo_file_ops); ++#ifdef CONFIG_GRKERNSEC_PROC_USERGROUP ++ proc_create("vmstat", gr_mode | S_IRGRP, NULL, &proc_vmstat_file_operations); ++#else ++ proc_create("vmstat", gr_mode, NULL, &proc_vmstat_file_operations); ++#endif ++ proc_create("zoneinfo", gr_mode, NULL, &proc_zoneinfo_file_operations); ++ } + #endif + return 0; + } +diff -urNp linux-2.6.38.3/net/8021q/vlan.c linux-2.6.38.3/net/8021q/vlan.c +--- linux-2.6.38.3/net/8021q/vlan.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/8021q/vlan.c 2011-04-17 15:57:33.000000000 -0400 +@@ -589,8 +589,7 @@ static int vlan_ioctl_handler(struct net + err = -EPERM; + if (!capable(CAP_NET_ADMIN)) + break; +- if ((args.u.name_type >= 0) && +- (args.u.name_type < VLAN_NAME_TYPE_HIGHEST)) { ++ if (args.u.name_type < VLAN_NAME_TYPE_HIGHEST) { + struct vlan_net *vn; + + vn = net_generic(net, vlan_net_id); +diff -urNp linux-2.6.38.3/net/atm/atm_misc.c linux-2.6.38.3/net/atm/atm_misc.c +--- linux-2.6.38.3/net/atm/atm_misc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/atm/atm_misc.c 2011-04-17 15:57:33.000000000 -0400 +@@ -17,7 +17,7 @@ int atm_charge(struct atm_vcc *vcc, int + if (atomic_read(&sk_atm(vcc)->sk_rmem_alloc) <= sk_atm(vcc)->sk_rcvbuf) + return 1; + atm_return(vcc, truesize); +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + return 0; + } + EXPORT_SYMBOL(atm_charge); +@@ -39,7 +39,7 @@ struct sk_buff *atm_alloc_charge(struct + } + } + atm_return(vcc, guess); +- atomic_inc(&vcc->stats->rx_drop); ++ atomic_inc_unchecked(&vcc->stats->rx_drop); + return NULL; + } + EXPORT_SYMBOL(atm_alloc_charge); +@@ -86,7 +86,7 @@ EXPORT_SYMBOL(atm_pcr_goal); + + void sonet_copy_stats(struct k_sonet_stats *from, struct sonet_stats *to) + { +-#define __HANDLE_ITEM(i) to->i = atomic_read(&from->i) ++#define __HANDLE_ITEM(i) to->i = atomic_read_unchecked(&from->i) + __SONET_ITEMS + #undef __HANDLE_ITEM + } +@@ -94,7 +94,7 @@ EXPORT_SYMBOL(sonet_copy_stats); + + void sonet_subtract_stats(struct k_sonet_stats *from, struct sonet_stats *to) + { +-#define __HANDLE_ITEM(i) atomic_sub(to->i, &from->i) ++#define __HANDLE_ITEM(i) atomic_sub_unchecked(to->i,&from->i) + __SONET_ITEMS + #undef __HANDLE_ITEM + } +diff -urNp linux-2.6.38.3/net/atm/proc.c linux-2.6.38.3/net/atm/proc.c +--- linux-2.6.38.3/net/atm/proc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/atm/proc.c 2011-04-17 15:57:33.000000000 -0400 +@@ -45,9 +45,9 @@ static void add_stats(struct seq_file *s + const struct k_atm_aal_stats *stats) + { + seq_printf(seq, "%s ( %d %d %d %d %d )", aal, +- atomic_read(&stats->tx), atomic_read(&stats->tx_err), +- atomic_read(&stats->rx), atomic_read(&stats->rx_err), +- atomic_read(&stats->rx_drop)); ++ atomic_read_unchecked(&stats->tx),atomic_read_unchecked(&stats->tx_err), ++ atomic_read_unchecked(&stats->rx),atomic_read_unchecked(&stats->rx_err), ++ atomic_read_unchecked(&stats->rx_drop)); + } + + static void atm_dev_info(struct seq_file *seq, const struct atm_dev *dev) +@@ -191,7 +191,12 @@ static void vcc_info(struct seq_file *se + { + struct sock *sk = sk_atm(vcc); + ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ seq_printf(seq, "%p ", NULL); ++#else + seq_printf(seq, "%p ", vcc); ++#endif ++ + if (!vcc->dev) + seq_printf(seq, "Unassigned "); + else +@@ -218,7 +223,11 @@ static void svc_info(struct seq_file *se + { + if (!vcc->dev) + seq_printf(seq, sizeof(void *) == 4 ? ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ "N/A@%p%10s" : "N/A@%p%2s", NULL, ""); ++#else + "N/A@%p%10s" : "N/A@%p%2s", vcc, ""); ++#endif + else + seq_printf(seq, "%3d %3d %5d ", + vcc->dev->number, vcc->vpi, vcc->vci); +diff -urNp linux-2.6.38.3/net/atm/resources.c linux-2.6.38.3/net/atm/resources.c +--- linux-2.6.38.3/net/atm/resources.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/atm/resources.c 2011-04-17 15:57:33.000000000 -0400 +@@ -160,7 +160,7 @@ EXPORT_SYMBOL(atm_dev_deregister); + static void copy_aal_stats(struct k_atm_aal_stats *from, + struct atm_aal_stats *to) + { +-#define __HANDLE_ITEM(i) to->i = atomic_read(&from->i) ++#define __HANDLE_ITEM(i) to->i = atomic_read_unchecked(&from->i) + __AAL_STAT_ITEMS + #undef __HANDLE_ITEM + } +@@ -168,7 +168,7 @@ static void copy_aal_stats(struct k_atm_ + static void subtract_aal_stats(struct k_atm_aal_stats *from, + struct atm_aal_stats *to) + { +-#define __HANDLE_ITEM(i) atomic_sub(to->i, &from->i) ++#define __HANDLE_ITEM(i) atomic_sub_unchecked(to->i, &from->i) + __AAL_STAT_ITEMS + #undef __HANDLE_ITEM + } +diff -urNp linux-2.6.38.3/net/bridge/br_multicast.c linux-2.6.38.3/net/bridge/br_multicast.c +--- linux-2.6.38.3/net/bridge/br_multicast.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/bridge/br_multicast.c 2011-04-17 15:57:33.000000000 -0400 +@@ -1482,7 +1482,7 @@ static int br_multicast_ipv6_rcv(struct + nexthdr = ip6h->nexthdr; + offset = ipv6_skip_exthdr(skb, sizeof(*ip6h), &nexthdr); + +- if (offset < 0 || nexthdr != IPPROTO_ICMPV6) ++ if (nexthdr != IPPROTO_ICMPV6) + return 0; + + /* Okay, we found ICMPv6 header */ +diff -urNp linux-2.6.38.3/net/bridge/netfilter/ebtables.c linux-2.6.38.3/net/bridge/netfilter/ebtables.c +--- linux-2.6.38.3/net/bridge/netfilter/ebtables.c 2011-04-17 16:53:01.000000000 -0400 ++++ linux-2.6.38.3/net/bridge/netfilter/ebtables.c 2011-04-17 17:00:29.000000000 -0400 +@@ -1512,7 +1512,7 @@ static int do_ebt_get_ctl(struct sock *s + tmp.valid_hooks = t->table->valid_hooks; + } + mutex_unlock(&ebt_mutex); +- if (copy_to_user(user, &tmp, *len) != 0){ ++ if (*len > sizeof(tmp) || copy_to_user(user, &tmp, *len) != 0){ + BUGPRINT("c2u Didn't work\n"); + ret = -EFAULT; + break; +diff -urNp linux-2.6.38.3/net/can/bcm.c linux-2.6.38.3/net/can/bcm.c +--- linux-2.6.38.3/net/can/bcm.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/can/bcm.c 2011-04-17 15:57:33.000000000 -0400 +@@ -165,9 +165,15 @@ static int bcm_proc_show(struct seq_file + struct bcm_sock *bo = bcm_sk(sk); + struct bcm_op *op; + ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ seq_printf(m, ">>> socket %p", NULL); ++ seq_printf(m, " / sk %p", NULL); ++ seq_printf(m, " / bo %p", NULL); ++#else + seq_printf(m, ">>> socket %p", sk->sk_socket); + seq_printf(m, " / sk %p", sk); + seq_printf(m, " / bo %p", bo); ++#endif + seq_printf(m, " / dropped %lu", bo->dropped_usr_msgs); + seq_printf(m, " / bound %s", bcm_proc_getifname(ifname, bo->ifindex)); + seq_printf(m, " <<<\n"); +diff -urNp linux-2.6.38.3/net/core/dev.c linux-2.6.38.3/net/core/dev.c +--- linux-2.6.38.3/net/core/dev.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/core/dev.c 2011-04-17 15:57:33.000000000 -0400 +@@ -1124,10 +1124,14 @@ void dev_load(struct net *net, const cha + if (no_module && capable(CAP_NET_ADMIN)) + no_module = request_module("netdev-%s", name); + if (no_module && capable(CAP_SYS_MODULE)) { ++#ifdef CONFIG_GRKERNSEC_MODHARDEN ++ ___request_module(true, "grsec_modharden_netdev", "%s", name); ++#else + if (!request_module("%s", name)) + pr_err("Loading kernel module for a network device " + "with CAP_SYS_MODULE (deprecated). Use CAP_NET_ADMIN and alias netdev-%s " + "instead\n", name); ++#endif + } + } + EXPORT_SYMBOL(dev_load); +@@ -2787,7 +2791,7 @@ int netif_rx_ni(struct sk_buff *skb) + } + EXPORT_SYMBOL(netif_rx_ni); + +-static void net_tx_action(struct softirq_action *h) ++static void net_tx_action(void) + { + struct softnet_data *sd = &__get_cpu_var(softnet_data); + +@@ -3697,7 +3701,7 @@ void netif_napi_del(struct napi_struct * + } + EXPORT_SYMBOL(netif_napi_del); + +-static void net_rx_action(struct softirq_action *h) ++static void net_rx_action(void) + { + struct softnet_data *sd = &__get_cpu_var(softnet_data); + unsigned long time_limit = jiffies + 2; +diff -urNp linux-2.6.38.3/net/core/sock.c linux-2.6.38.3/net/core/sock.c +--- linux-2.6.38.3/net/core/sock.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/core/sock.c 2011-04-17 15:57:33.000000000 -0400 +@@ -934,7 +934,7 @@ int sock_getsockopt(struct socket *sock, + return -ENOTCONN; + if (lv < len) + return -EINVAL; +- if (copy_to_user(optval, address, len)) ++ if (len > sizeof(address) || copy_to_user(optval, address, len)) + return -EFAULT; + goto lenout; + } +@@ -967,7 +967,7 @@ int sock_getsockopt(struct socket *sock, + + if (len > lv) + len = lv; +- if (copy_to_user(optval, &v, len)) ++ if (len > sizeof(v) || copy_to_user(optval, &v, len)) + return -EFAULT; + lenout: + if (put_user(len, optlen)) +diff -urNp linux-2.6.38.3/net/dccp/ccids/ccid3.c linux-2.6.38.3/net/dccp/ccids/ccid3.c +--- linux-2.6.38.3/net/dccp/ccids/ccid3.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/dccp/ccids/ccid3.c 2011-04-17 15:57:33.000000000 -0400 +@@ -41,7 +41,7 @@ + static int ccid3_debug; + #define ccid3_pr_debug(format, a...) DCCP_PR_DEBUG(ccid3_debug, format, ##a) + #else +-#define ccid3_pr_debug(format, a...) ++#define ccid3_pr_debug(format, a...) do {} while (0) + #endif + + /* +diff -urNp linux-2.6.38.3/net/dccp/dccp.h linux-2.6.38.3/net/dccp/dccp.h +--- linux-2.6.38.3/net/dccp/dccp.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/dccp/dccp.h 2011-04-17 15:57:33.000000000 -0400 +@@ -44,9 +44,9 @@ extern int dccp_debug; + #define dccp_pr_debug_cat(format, a...) DCCP_PRINTK(dccp_debug, format, ##a) + #define dccp_debug(fmt, a...) dccp_pr_debug_cat(KERN_DEBUG fmt, ##a) + #else +-#define dccp_pr_debug(format, a...) +-#define dccp_pr_debug_cat(format, a...) +-#define dccp_debug(format, a...) ++#define dccp_pr_debug(format, a...) do {} while (0) ++#define dccp_pr_debug_cat(format, a...) do {} while (0) ++#define dccp_debug(format, a...) do {} while (0) + #endif + + extern struct inet_hashinfo dccp_hashinfo; +diff -urNp linux-2.6.38.3/net/decnet/sysctl_net_decnet.c linux-2.6.38.3/net/decnet/sysctl_net_decnet.c +--- linux-2.6.38.3/net/decnet/sysctl_net_decnet.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/decnet/sysctl_net_decnet.c 2011-04-17 15:57:33.000000000 -0400 +@@ -173,7 +173,7 @@ static int dn_node_address_handler(ctl_t + + if (len > *lenp) len = *lenp; + +- if (copy_to_user(buffer, addr, len)) ++ if (len > sizeof addr || copy_to_user(buffer, addr, len)) + return -EFAULT; + + *lenp = len; +@@ -236,7 +236,7 @@ static int dn_def_dev_handler(ctl_table + + if (len > *lenp) len = *lenp; + +- if (copy_to_user(buffer, devname, len)) ++ if (len > sizeof devname || copy_to_user(buffer, devname, len)) + return -EFAULT; + + *lenp = len; +diff -urNp linux-2.6.38.3/net/econet/Kconfig linux-2.6.38.3/net/econet/Kconfig +--- linux-2.6.38.3/net/econet/Kconfig 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/econet/Kconfig 2011-04-17 15:57:33.000000000 -0400 +@@ -4,7 +4,7 @@ + + config ECONET + tristate "Acorn Econet/AUN protocols (EXPERIMENTAL)" +- depends on EXPERIMENTAL && INET ++ depends on EXPERIMENTAL && INET && BROKEN + ---help--- + Econet is a fairly old and slow networking protocol mainly used by + Acorn computers to access file and print servers. It uses native +diff -urNp linux-2.6.38.3/net/ipv4/inet_diag.c linux-2.6.38.3/net/ipv4/inet_diag.c +--- linux-2.6.38.3/net/ipv4/inet_diag.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/ipv4/inet_diag.c 2011-04-17 15:57:33.000000000 -0400 +@@ -114,8 +114,14 @@ static int inet_csk_diag_fill(struct soc + r->idiag_retrans = 0; + + r->id.idiag_if = sk->sk_bound_dev_if; ++ ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ r->id.idiag_cookie[0] = 0; ++ r->id.idiag_cookie[1] = 0; ++#else + r->id.idiag_cookie[0] = (u32)(unsigned long)sk; + r->id.idiag_cookie[1] = (u32)(((unsigned long)sk >> 31) >> 1); ++#endif + + r->id.idiag_sport = inet->inet_sport; + r->id.idiag_dport = inet->inet_dport; +@@ -201,8 +207,15 @@ static int inet_twsk_diag_fill(struct in + r->idiag_family = tw->tw_family; + r->idiag_retrans = 0; + r->id.idiag_if = tw->tw_bound_dev_if; ++ ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ r->id.idiag_cookie[0] = 0; ++ r->id.idiag_cookie[1] = 0; ++#else + r->id.idiag_cookie[0] = (u32)(unsigned long)tw; + r->id.idiag_cookie[1] = (u32)(((unsigned long)tw >> 31) >> 1); ++#endif ++ + r->id.idiag_sport = tw->tw_sport; + r->id.idiag_dport = tw->tw_dport; + r->id.idiag_src[0] = tw->tw_rcv_saddr; +@@ -285,12 +298,14 @@ static int inet_diag_get_exact(struct sk + if (sk == NULL) + goto unlock; + ++#ifndef CONFIG_GRKERNSEC_HIDESYM + err = -ESTALE; + if ((req->id.idiag_cookie[0] != INET_DIAG_NOCOOKIE || + req->id.idiag_cookie[1] != INET_DIAG_NOCOOKIE) && + ((u32)(unsigned long)sk != req->id.idiag_cookie[0] || + (u32)((((unsigned long)sk) >> 31) >> 1) != req->id.idiag_cookie[1])) + goto out; ++#endif + + err = -ENOMEM; + rep = alloc_skb(NLMSG_SPACE((sizeof(struct inet_diag_msg) + +@@ -582,8 +597,14 @@ static int inet_diag_fill_req(struct sk_ + r->idiag_retrans = req->retrans; + + r->id.idiag_if = sk->sk_bound_dev_if; ++ ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ r->id.idiag_cookie[0] = 0; ++ r->id.idiag_cookie[1] = 0; ++#else + r->id.idiag_cookie[0] = (u32)(unsigned long)req; + r->id.idiag_cookie[1] = (u32)(((unsigned long)req >> 31) >> 1); ++#endif + + tmo = req->expires - jiffies; + if (tmo < 0) +diff -urNp linux-2.6.38.3/net/ipv4/inet_hashtables.c linux-2.6.38.3/net/ipv4/inet_hashtables.c +--- linux-2.6.38.3/net/ipv4/inet_hashtables.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/ipv4/inet_hashtables.c 2011-04-17 15:57:33.000000000 -0400 +@@ -18,11 +18,14 @@ + #include <linux/sched.h> + #include <linux/slab.h> + #include <linux/wait.h> ++#include <linux/security.h> + + #include <net/inet_connection_sock.h> + #include <net/inet_hashtables.h> + #include <net/ip.h> + ++extern void gr_update_task_in_ip_table(struct task_struct *task, const struct inet_sock *inet); ++ + /* + * Allocate and initialize a new local port bind bucket. + * The bindhash mutex for snum's hash chain must be held here. +@@ -529,6 +532,8 @@ ok: + twrefcnt += inet_twsk_bind_unhash(tw, hinfo); + spin_unlock(&head->lock); + ++ gr_update_task_in_ip_table(current, inet_sk(sk)); ++ + if (tw) { + inet_twsk_deschedule(tw, death_row); + while (twrefcnt) { +diff -urNp linux-2.6.38.3/net/ipv4/inetpeer.c linux-2.6.38.3/net/ipv4/inetpeer.c +--- linux-2.6.38.3/net/ipv4/inetpeer.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/ipv4/inetpeer.c 2011-04-17 15:57:33.000000000 -0400 +@@ -509,8 +509,8 @@ struct inet_peer *inet_getpeer(struct in + if (p) { + p->daddr = *daddr; + atomic_set(&p->refcnt, 1); +- atomic_set(&p->rid, 0); +- atomic_set(&p->ip_id_count, secure_ip_id(daddr->a4)); ++ atomic_set_unchecked(&p->rid, 0); ++ atomic_set_unchecked(&p->ip_id_count, secure_ip_id(daddr->a4)); + p->tcp_ts_stamp = 0; + INIT_LIST_HEAD(&p->unused); + +diff -urNp linux-2.6.38.3/net/ipv4/ip_fragment.c linux-2.6.38.3/net/ipv4/ip_fragment.c +--- linux-2.6.38.3/net/ipv4/ip_fragment.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/ipv4/ip_fragment.c 2011-04-17 15:57:33.000000000 -0400 +@@ -298,7 +298,7 @@ static inline int ip_frag_too_far(struct + return 0; + + start = qp->rid; +- end = atomic_inc_return(&peer->rid); ++ end = atomic_inc_return_unchecked(&peer->rid); + qp->rid = end; + + rc = qp->q.fragments && (end - start) > max; +diff -urNp linux-2.6.38.3/net/ipv4/netfilter/nf_nat_snmp_basic.c linux-2.6.38.3/net/ipv4/netfilter/nf_nat_snmp_basic.c +--- linux-2.6.38.3/net/ipv4/netfilter/nf_nat_snmp_basic.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/ipv4/netfilter/nf_nat_snmp_basic.c 2011-04-17 15:57:33.000000000 -0400 +@@ -398,7 +398,7 @@ static unsigned char asn1_octets_decode( + + *len = 0; + +- *octets = kmalloc(eoc - ctx->pointer, GFP_ATOMIC); ++ *octets = kmalloc((eoc - ctx->pointer), GFP_ATOMIC); + if (*octets == NULL) { + if (net_ratelimit()) + pr_notice("OOM in bsalg (%d)\n", __LINE__); +diff -urNp linux-2.6.38.3/net/ipv4/raw.c linux-2.6.38.3/net/ipv4/raw.c +--- linux-2.6.38.3/net/ipv4/raw.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/ipv4/raw.c 2011-04-17 15:57:33.000000000 -0400 +@@ -724,15 +724,19 @@ static int raw_init(struct sock *sk) + + static int raw_seticmpfilter(struct sock *sk, char __user *optval, int optlen) + { ++ struct icmp_filter filter; ++ + if (optlen > sizeof(struct icmp_filter)) + optlen = sizeof(struct icmp_filter); +- if (copy_from_user(&raw_sk(sk)->filter, optval, optlen)) ++ if (copy_from_user(&filter, optval, optlen)) + return -EFAULT; ++ memcpy(&raw_sk(sk)->filter, &filter, sizeof(filter)); + return 0; + } + + static int raw_geticmpfilter(struct sock *sk, char __user *optval, int __user *optlen) + { ++ struct icmp_filter filter; + int len, ret = -EFAULT; + + if (get_user(len, optlen)) +@@ -743,8 +747,9 @@ static int raw_geticmpfilter(struct sock + if (len > sizeof(struct icmp_filter)) + len = sizeof(struct icmp_filter); + ret = -EFAULT; ++ memcpy(&filter, &raw_sk(sk)->filter, len); + if (put_user(len, optlen) || +- copy_to_user(optval, &raw_sk(sk)->filter, len)) ++ copy_to_user(optval, &filter, len)) + goto out; + ret = 0; + out: return ret; +diff -urNp linux-2.6.38.3/net/ipv4/route.c linux-2.6.38.3/net/ipv4/route.c +--- linux-2.6.38.3/net/ipv4/route.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/ipv4/route.c 2011-04-17 15:57:33.000000000 -0400 +@@ -2857,7 +2857,7 @@ static int rt_fill_info(struct net *net, + expires = rt->dst.expires ? rt->dst.expires - jiffies : 0; + if (rt->peer) { + inet_peer_refcheck(rt->peer); +- id = atomic_read(&rt->peer->ip_id_count) & 0xffff; ++ id = atomic_read_unchecked(&rt->peer->ip_id_count) & 0xffff; + if (rt->peer->tcp_ts_stamp) { + ts = rt->peer->tcp_ts; + tsage = get_seconds() - rt->peer->tcp_ts_stamp; +diff -urNp linux-2.6.38.3/net/ipv4/tcp_ipv4.c linux-2.6.38.3/net/ipv4/tcp_ipv4.c +--- linux-2.6.38.3/net/ipv4/tcp_ipv4.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/ipv4/tcp_ipv4.c 2011-04-17 15:57:33.000000000 -0400 +@@ -86,6 +86,9 @@ int sysctl_tcp_tw_reuse __read_mostly; + int sysctl_tcp_low_latency __read_mostly; + EXPORT_SYMBOL(sysctl_tcp_low_latency); + ++#ifdef CONFIG_GRKERNSEC_BLACKHOLE ++extern int grsec_enable_blackhole; ++#endif + + #ifdef CONFIG_TCP_MD5SIG + static struct tcp_md5sig_key *tcp_v4_md5_do_lookup(struct sock *sk, +@@ -1593,6 +1596,9 @@ int tcp_v4_do_rcv(struct sock *sk, struc + return 0; + + reset: ++#ifdef CONFIG_GRKERNSEC_BLACKHOLE ++ if (!grsec_enable_blackhole) ++#endif + tcp_v4_send_reset(rsk, skb); + discard: + kfree_skb(skb); +@@ -1655,12 +1661,19 @@ int tcp_v4_rcv(struct sk_buff *skb) + TCP_SKB_CB(skb)->sacked = 0; + + sk = __inet_lookup_skb(&tcp_hashinfo, skb, th->source, th->dest); +- if (!sk) ++ if (!sk) { ++#ifdef CONFIG_GRKERNSEC_BLACKHOLE ++ ret = 1; ++#endif + goto no_tcp_socket; +- ++ } + process: +- if (sk->sk_state == TCP_TIME_WAIT) ++ if (sk->sk_state == TCP_TIME_WAIT) { ++#ifdef CONFIG_GRKERNSEC_BLACKHOLE ++ ret = 2; ++#endif + goto do_time_wait; ++ } + + if (unlikely(iph->ttl < inet_sk(sk)->min_ttl)) { + NET_INC_STATS_BH(net, LINUX_MIB_TCPMINTTLDROP); +@@ -1710,6 +1723,10 @@ no_tcp_socket: + bad_packet: + TCP_INC_STATS_BH(net, TCP_MIB_INERRS); + } else { ++#ifdef CONFIG_GRKERNSEC_BLACKHOLE ++ if (!grsec_enable_blackhole || (ret == 1 && ++ (skb->dev->flags & IFF_LOOPBACK))) ++#endif + tcp_v4_send_reset(NULL, skb); + } + +@@ -2373,7 +2390,11 @@ static void get_openreq4(struct sock *sk + 0, /* non standard timer */ + 0, /* open_requests have no inode */ + atomic_read(&sk->sk_refcnt), ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ NULL, ++#else + req, ++#endif + len); + } + +@@ -2423,7 +2444,12 @@ static void get_tcp4_sock(struct sock *s + sock_i_uid(sk), + icsk->icsk_probes_out, + sock_i_ino(sk), +- atomic_read(&sk->sk_refcnt), sk, ++ atomic_read(&sk->sk_refcnt), ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ NULL, ++#else ++ sk, ++#endif + jiffies_to_clock_t(icsk->icsk_rto), + jiffies_to_clock_t(icsk->icsk_ack.ato), + (icsk->icsk_ack.quick << 1) | icsk->icsk_ack.pingpong, +@@ -2451,7 +2477,13 @@ static void get_timewait4_sock(struct in + " %02X %08X:%08X %02X:%08lX %08X %5d %8d %d %d %p%n", + i, src, srcp, dest, destp, tw->tw_substate, 0, 0, + 3, jiffies_to_clock_t(ttd), 0, 0, 0, 0, +- atomic_read(&tw->tw_refcnt), tw, len); ++ atomic_read(&tw->tw_refcnt), ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ NULL, ++#else ++ tw, ++#endif ++ len); + } + + #define TMPSZ 150 +diff -urNp linux-2.6.38.3/net/ipv4/tcp_minisocks.c linux-2.6.38.3/net/ipv4/tcp_minisocks.c +--- linux-2.6.38.3/net/ipv4/tcp_minisocks.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/ipv4/tcp_minisocks.c 2011-04-17 15:57:33.000000000 -0400 +@@ -27,6 +27,10 @@ + #include <net/inet_common.h> + #include <net/xfrm.h> + ++#ifdef CONFIG_GRKERNSEC_BLACKHOLE ++extern int grsec_enable_blackhole; ++#endif ++ + int sysctl_tcp_syncookies __read_mostly = 1; + EXPORT_SYMBOL(sysctl_tcp_syncookies); + +@@ -745,6 +749,10 @@ listen_overflow: + + embryonic_reset: + NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_EMBRYONICRSTS); ++ ++#ifdef CONFIG_GRKERNSEC_BLACKHOLE ++ if (!grsec_enable_blackhole) ++#endif + if (!(flg & TCP_FLAG_RST)) + req->rsk_ops->send_reset(sk, skb); + +diff -urNp linux-2.6.38.3/net/ipv4/tcp_probe.c linux-2.6.38.3/net/ipv4/tcp_probe.c +--- linux-2.6.38.3/net/ipv4/tcp_probe.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/ipv4/tcp_probe.c 2011-04-17 15:57:33.000000000 -0400 +@@ -202,7 +202,7 @@ static ssize_t tcpprobe_read(struct file + if (cnt + width >= len) + break; + +- if (copy_to_user(buf + cnt, tbuf, width)) ++ if (width > sizeof tbuf || copy_to_user(buf + cnt, tbuf, width)) + return -EFAULT; + cnt += width; + } +diff -urNp linux-2.6.38.3/net/ipv4/tcp_timer.c linux-2.6.38.3/net/ipv4/tcp_timer.c +--- linux-2.6.38.3/net/ipv4/tcp_timer.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/ipv4/tcp_timer.c 2011-04-17 15:57:33.000000000 -0400 +@@ -22,6 +22,10 @@ + #include <linux/gfp.h> + #include <net/tcp.h> + ++#ifdef CONFIG_GRKERNSEC_BLACKHOLE ++extern int grsec_lastack_retries; ++#endif ++ + int sysctl_tcp_syn_retries __read_mostly = TCP_SYN_RETRIES; + int sysctl_tcp_synack_retries __read_mostly = TCP_SYNACK_RETRIES; + int sysctl_tcp_keepalive_time __read_mostly = TCP_KEEPALIVE_TIME; +@@ -199,6 +203,13 @@ static int tcp_write_timeout(struct sock + } + } + ++#ifdef CONFIG_GRKERNSEC_BLACKHOLE ++ if ((sk->sk_state == TCP_LAST_ACK) && ++ (grsec_lastack_retries > 0) && ++ (grsec_lastack_retries < retry_until)) ++ retry_until = grsec_lastack_retries; ++#endif ++ + if (retransmits_timed_out(sk, retry_until, + syn_set ? 0 : icsk->icsk_user_timeout, syn_set)) { + /* Has it gone just too far? */ +diff -urNp linux-2.6.38.3/net/ipv4/udp.c linux-2.6.38.3/net/ipv4/udp.c +--- linux-2.6.38.3/net/ipv4/udp.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/ipv4/udp.c 2011-04-17 15:57:33.000000000 -0400 +@@ -86,6 +86,7 @@ + #include <linux/types.h> + #include <linux/fcntl.h> + #include <linux/module.h> ++#include <linux/security.h> + #include <linux/socket.h> + #include <linux/sockios.h> + #include <linux/igmp.h> +@@ -107,6 +108,10 @@ + #include <net/xfrm.h> + #include "udp_impl.h" + ++#ifdef CONFIG_GRKERNSEC_BLACKHOLE ++extern int grsec_enable_blackhole; ++#endif ++ + struct udp_table udp_table __read_mostly; + EXPORT_SYMBOL(udp_table); + +@@ -564,6 +569,9 @@ found: + return s; + } + ++extern int gr_search_udp_recvmsg(struct sock *sk, const struct sk_buff *skb); ++extern int gr_search_udp_sendmsg(struct sock *sk, struct sockaddr_in *addr); ++ + /* + * This routine is called by the ICMP module when it gets some + * sort of error condition. If err < 0 then the socket should +@@ -832,9 +840,18 @@ int udp_sendmsg(struct kiocb *iocb, stru + dport = usin->sin_port; + if (dport == 0) + return -EINVAL; ++ ++ err = gr_search_udp_sendmsg(sk, usin); ++ if (err) ++ return err; + } else { + if (sk->sk_state != TCP_ESTABLISHED) + return -EDESTADDRREQ; ++ ++ err = gr_search_udp_sendmsg(sk, NULL); ++ if (err) ++ return err; ++ + daddr = inet->inet_daddr; + dport = inet->inet_dport; + /* Open fast path for connected socket. +@@ -1139,6 +1156,10 @@ try_again: + if (!skb) + goto out; + ++ err = gr_search_udp_recvmsg(sk, skb); ++ if (err) ++ goto out_free; ++ + ulen = skb->len - sizeof(struct udphdr); + if (len > ulen) + len = ulen; +@@ -1623,6 +1644,9 @@ int __udp4_lib_rcv(struct sk_buff *skb, + goto csum_error; + + UDP_INC_STATS_BH(net, UDP_MIB_NOPORTS, proto == IPPROTO_UDPLITE); ++#ifdef CONFIG_GRKERNSEC_BLACKHOLE ++ if (!grsec_enable_blackhole || (skb->dev->flags & IFF_LOOPBACK)) ++#endif + icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0); + + /* +@@ -2050,7 +2074,12 @@ static void udp4_format_sock(struct sock + sk_wmem_alloc_get(sp), + sk_rmem_alloc_get(sp), + 0, 0L, 0, sock_i_uid(sp), 0, sock_i_ino(sp), +- atomic_read(&sp->sk_refcnt), sp, ++ atomic_read(&sp->sk_refcnt), ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ NULL, ++#else ++ sp, ++#endif + atomic_read(&sp->sk_drops), len); + } + +diff -urNp linux-2.6.38.3/net/ipv6/exthdrs.c linux-2.6.38.3/net/ipv6/exthdrs.c +--- linux-2.6.38.3/net/ipv6/exthdrs.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/ipv6/exthdrs.c 2011-04-17 15:57:33.000000000 -0400 +@@ -634,7 +634,7 @@ static struct tlvtype_proc tlvprochopopt + .type = IPV6_TLV_JUMBO, + .func = ipv6_hop_jumbo, + }, +- { -1, } ++ { -1, NULL } + }; + + int ipv6_parse_hopopts(struct sk_buff *skb) +diff -urNp linux-2.6.38.3/net/ipv6/raw.c linux-2.6.38.3/net/ipv6/raw.c +--- linux-2.6.38.3/net/ipv6/raw.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/ipv6/raw.c 2011-04-17 15:57:33.000000000 -0400 +@@ -602,7 +602,7 @@ out: + return err; + } + +-static int rawv6_send_hdrinc(struct sock *sk, void *from, int length, ++static int rawv6_send_hdrinc(struct sock *sk, void *from, unsigned int length, + struct flowi *fl, struct dst_entry **dstp, + unsigned int flags) + { +@@ -919,12 +919,15 @@ do_confirm: + static int rawv6_seticmpfilter(struct sock *sk, int level, int optname, + char __user *optval, int optlen) + { ++ struct icmp6_filter filter; ++ + switch (optname) { + case ICMPV6_FILTER: + if (optlen > sizeof(struct icmp6_filter)) + optlen = sizeof(struct icmp6_filter); +- if (copy_from_user(&raw6_sk(sk)->filter, optval, optlen)) ++ if (copy_from_user(&filter, optval, optlen)) + return -EFAULT; ++ memcpy(&raw6_sk(sk)->filter, &filter, optlen); + return 0; + default: + return -ENOPROTOOPT; +@@ -936,6 +939,7 @@ static int rawv6_seticmpfilter(struct so + static int rawv6_geticmpfilter(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen) + { ++ struct icmp6_filter filter; + int len; + + switch (optname) { +@@ -948,7 +952,8 @@ static int rawv6_geticmpfilter(struct so + len = sizeof(struct icmp6_filter); + if (put_user(len, optlen)) + return -EFAULT; +- if (copy_to_user(optval, &raw6_sk(sk)->filter, len)) ++ memcpy(&filter, &raw6_sk(sk)->filter, len); ++ if (copy_to_user(optval, &filter, len)) + return -EFAULT; + return 0; + default: +@@ -1262,7 +1267,13 @@ static void raw6_sock_seq_show(struct se + 0, 0L, 0, + sock_i_uid(sp), 0, + sock_i_ino(sp), +- atomic_read(&sp->sk_refcnt), sp, atomic_read(&sp->sk_drops)); ++ atomic_read(&sp->sk_refcnt), ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ NULL, ++#else ++ sp, ++#endif ++ atomic_read(&sp->sk_drops)); + } + + static int raw6_seq_show(struct seq_file *seq, void *v) +diff -urNp linux-2.6.38.3/net/ipv6/tcp_ipv6.c linux-2.6.38.3/net/ipv6/tcp_ipv6.c +--- linux-2.6.38.3/net/ipv6/tcp_ipv6.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/ipv6/tcp_ipv6.c 2011-04-17 15:57:33.000000000 -0400 +@@ -92,6 +92,10 @@ static struct tcp_md5sig_key *tcp_v6_md5 + } + #endif + ++#ifdef CONFIG_GRKERNSEC_BLACKHOLE ++extern int grsec_enable_blackhole; ++#endif ++ + static void tcp_v6_hash(struct sock *sk) + { + if (sk->sk_state != TCP_CLOSE) { +@@ -1676,6 +1680,9 @@ static int tcp_v6_do_rcv(struct sock *sk + return 0; + + reset: ++#ifdef CONFIG_GRKERNSEC_BLACKHOLE ++ if (!grsec_enable_blackhole) ++#endif + tcp_v6_send_reset(sk, skb); + discard: + if (opt_skb) +@@ -1755,12 +1762,20 @@ static int tcp_v6_rcv(struct sk_buff *sk + TCP_SKB_CB(skb)->sacked = 0; + + sk = __inet6_lookup_skb(&tcp_hashinfo, skb, th->source, th->dest); +- if (!sk) ++ if (!sk) { ++#ifdef CONFIG_GRKERNSEC_BLACKHOLE ++ ret = 1; ++#endif + goto no_tcp_socket; ++ } + + process: +- if (sk->sk_state == TCP_TIME_WAIT) ++ if (sk->sk_state == TCP_TIME_WAIT) { ++#ifdef CONFIG_GRKERNSEC_BLACKHOLE ++ ret = 2; ++#endif + goto do_time_wait; ++ } + + if (hdr->hop_limit < inet6_sk(sk)->min_hopcount) { + NET_INC_STATS_BH(net, LINUX_MIB_TCPMINTTLDROP); +@@ -1808,6 +1823,10 @@ no_tcp_socket: + bad_packet: + TCP_INC_STATS_BH(net, TCP_MIB_INERRS); + } else { ++#ifdef CONFIG_GRKERNSEC_BLACKHOLE ++ if (!grsec_enable_blackhole || (ret == 1 && ++ (skb->dev->flags & IFF_LOOPBACK))) ++#endif + tcp_v6_send_reset(NULL, skb); + } + +@@ -2068,7 +2087,13 @@ static void get_openreq6(struct seq_file + uid, + 0, /* non standard timer */ + 0, /* open_requests have no inode */ +- 0, req); ++ 0, ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ NULL ++#else ++ req ++#endif ++ ); + } + + static void get_tcp6_sock(struct seq_file *seq, struct sock *sp, int i) +@@ -2118,7 +2143,12 @@ static void get_tcp6_sock(struct seq_fil + sock_i_uid(sp), + icsk->icsk_probes_out, + sock_i_ino(sp), +- atomic_read(&sp->sk_refcnt), sp, ++ atomic_read(&sp->sk_refcnt), ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ NULL, ++#else ++ sp, ++#endif + jiffies_to_clock_t(icsk->icsk_rto), + jiffies_to_clock_t(icsk->icsk_ack.ato), + (icsk->icsk_ack.quick << 1 ) | icsk->icsk_ack.pingpong, +@@ -2153,7 +2183,13 @@ static void get_timewait6_sock(struct se + dest->s6_addr32[2], dest->s6_addr32[3], destp, + tw->tw_substate, 0, 0, + 3, jiffies_to_clock_t(ttd), 0, 0, 0, 0, +- atomic_read(&tw->tw_refcnt), tw); ++ atomic_read(&tw->tw_refcnt), ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ NULL ++#else ++ tw ++#endif ++ ); + } + + static int tcp6_seq_show(struct seq_file *seq, void *v) +diff -urNp linux-2.6.38.3/net/ipv6/udp.c linux-2.6.38.3/net/ipv6/udp.c +--- linux-2.6.38.3/net/ipv6/udp.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/ipv6/udp.c 2011-04-17 15:57:33.000000000 -0400 +@@ -50,6 +50,10 @@ + #include <linux/seq_file.h> + #include "udp_impl.h" + ++#ifdef CONFIG_GRKERNSEC_BLACKHOLE ++extern int grsec_enable_blackhole; ++#endif ++ + int ipv6_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2) + { + const struct in6_addr *sk_rcv_saddr6 = &inet6_sk(sk)->rcv_saddr; +@@ -773,6 +777,9 @@ int __udp6_lib_rcv(struct sk_buff *skb, + UDP6_INC_STATS_BH(net, UDP_MIB_NOPORTS, + proto == IPPROTO_UDPLITE); + ++#ifdef CONFIG_GRKERNSEC_BLACKHOLE ++ if (!grsec_enable_blackhole || (skb->dev->flags & IFF_LOOPBACK)) ++#endif + icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0); + + kfree_skb(skb); +@@ -1407,7 +1414,12 @@ static void udp6_sock_seq_show(struct se + 0, 0L, 0, + sock_i_uid(sp), 0, + sock_i_ino(sp), +- atomic_read(&sp->sk_refcnt), sp, ++ atomic_read(&sp->sk_refcnt), ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ NULL, ++#else ++ sp, ++#endif + atomic_read(&sp->sk_drops)); + } + +diff -urNp linux-2.6.38.3/net/irda/ircomm/ircomm_tty.c linux-2.6.38.3/net/irda/ircomm/ircomm_tty.c +--- linux-2.6.38.3/net/irda/ircomm/ircomm_tty.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/irda/ircomm/ircomm_tty.c 2011-04-17 15:57:33.000000000 -0400 +@@ -281,16 +281,16 @@ static int ircomm_tty_block_til_ready(st + add_wait_queue(&self->open_wait, &wait); + + IRDA_DEBUG(2, "%s(%d):block_til_ready before block on %s open_count=%d\n", +- __FILE__,__LINE__, tty->driver->name, self->open_count ); ++ __FILE__,__LINE__, tty->driver->name, local_read(&self->open_count) ); + + /* As far as I can see, we protect open_count - Jean II */ + spin_lock_irqsave(&self->spinlock, flags); + if (!tty_hung_up_p(filp)) { + extra_count = 1; +- self->open_count--; ++ local_dec(&self->open_count); + } + spin_unlock_irqrestore(&self->spinlock, flags); +- self->blocked_open++; ++ local_inc(&self->blocked_open); + + while (1) { + if (tty->termios->c_cflag & CBAUD) { +@@ -330,7 +330,7 @@ static int ircomm_tty_block_til_ready(st + } + + IRDA_DEBUG(1, "%s(%d):block_til_ready blocking on %s open_count=%d\n", +- __FILE__,__LINE__, tty->driver->name, self->open_count ); ++ __FILE__,__LINE__, tty->driver->name, local_read(&self->open_count) ); + + schedule(); + } +@@ -341,13 +341,13 @@ static int ircomm_tty_block_til_ready(st + if (extra_count) { + /* ++ is not atomic, so this should be protected - Jean II */ + spin_lock_irqsave(&self->spinlock, flags); +- self->open_count++; ++ local_inc(&self->open_count); + spin_unlock_irqrestore(&self->spinlock, flags); + } +- self->blocked_open--; ++ local_dec(&self->blocked_open); + + IRDA_DEBUG(1, "%s(%d):block_til_ready after blocking on %s open_count=%d\n", +- __FILE__,__LINE__, tty->driver->name, self->open_count); ++ __FILE__,__LINE__, tty->driver->name, local_read(&self->open_count)); + + if (!retval) + self->flags |= ASYNC_NORMAL_ACTIVE; +@@ -416,14 +416,14 @@ static int ircomm_tty_open(struct tty_st + } + /* ++ is not atomic, so this should be protected - Jean II */ + spin_lock_irqsave(&self->spinlock, flags); +- self->open_count++; ++ local_inc(&self->open_count); + + tty->driver_data = self; + self->tty = tty; + spin_unlock_irqrestore(&self->spinlock, flags); + + IRDA_DEBUG(1, "%s(), %s%d, count = %d\n", __func__ , tty->driver->name, +- self->line, self->open_count); ++ self->line, local_read(&self->open_count)); + + /* Not really used by us, but lets do it anyway */ + self->tty->low_latency = (self->flags & ASYNC_LOW_LATENCY) ? 1 : 0; +@@ -509,7 +509,7 @@ static void ircomm_tty_close(struct tty_ + return; + } + +- if ((tty->count == 1) && (self->open_count != 1)) { ++ if ((tty->count == 1) && (local_read(&self->open_count) != 1)) { + /* + * Uh, oh. tty->count is 1, which means that the tty + * structure will be freed. state->count should always +@@ -519,16 +519,16 @@ static void ircomm_tty_close(struct tty_ + */ + IRDA_DEBUG(0, "%s(), bad serial port count; " + "tty->count is 1, state->count is %d\n", __func__ , +- self->open_count); +- self->open_count = 1; ++ local_read(&self->open_count)); ++ local_set(&self->open_count, 1); + } + +- if (--self->open_count < 0) { ++ if (local_dec_return(&self->open_count) < 0) { + IRDA_ERROR("%s(), bad serial port count for ttys%d: %d\n", +- __func__, self->line, self->open_count); +- self->open_count = 0; ++ __func__, self->line, local_read(&self->open_count)); ++ local_set(&self->open_count, 0); + } +- if (self->open_count) { ++ if (local_read(&self->open_count)) { + spin_unlock_irqrestore(&self->spinlock, flags); + + IRDA_DEBUG(0, "%s(), open count > 0\n", __func__ ); +@@ -560,7 +560,7 @@ static void ircomm_tty_close(struct tty_ + tty->closing = 0; + self->tty = NULL; + +- if (self->blocked_open) { ++ if (local_read(&self->blocked_open)) { + if (self->close_delay) + schedule_timeout_interruptible(self->close_delay); + wake_up_interruptible(&self->open_wait); +@@ -1012,7 +1012,7 @@ static void ircomm_tty_hangup(struct tty + spin_lock_irqsave(&self->spinlock, flags); + self->flags &= ~ASYNC_NORMAL_ACTIVE; + self->tty = NULL; +- self->open_count = 0; ++ local_set(&self->open_count, 0); + spin_unlock_irqrestore(&self->spinlock, flags); + + wake_up_interruptible(&self->open_wait); +@@ -1364,7 +1364,7 @@ static void ircomm_tty_line_info(struct + seq_putc(m, '\n'); + + seq_printf(m, "Role: %s\n", self->client ? "client" : "server"); +- seq_printf(m, "Open count: %d\n", self->open_count); ++ seq_printf(m, "Open count: %d\n", local_read(&self->open_count)); + seq_printf(m, "Max data size: %d\n", self->max_data_size); + seq_printf(m, "Max header size: %d\n", self->max_header_size); + +diff -urNp linux-2.6.38.3/net/key/af_key.c linux-2.6.38.3/net/key/af_key.c +--- linux-2.6.38.3/net/key/af_key.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/key/af_key.c 2011-04-17 15:57:33.000000000 -0400 +@@ -3644,7 +3644,11 @@ static int pfkey_seq_show(struct seq_fil + seq_printf(f ,"sk RefCnt Rmem Wmem User Inode\n"); + else + seq_printf(f ,"%p %-6d %-6u %-6u %-6u %-6lu\n", ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ NULL, ++#else + s, ++#endif + atomic_read(&s->sk_refcnt), + sk_rmem_alloc_get(s), + sk_wmem_alloc_get(s), +diff -urNp linux-2.6.38.3/net/mac80211/cfg.c linux-2.6.38.3/net/mac80211/cfg.c +--- linux-2.6.38.3/net/mac80211/cfg.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/mac80211/cfg.c 2011-04-17 15:57:33.000000000 -0400 +@@ -1939,7 +1939,7 @@ static int ieee80211_get_antenna(struct + return drv_get_antenna(local, tx_ant, rx_ant); + } + +-struct cfg80211_ops mac80211_config_ops = { ++const struct cfg80211_ops mac80211_config_ops = { + .add_virtual_intf = ieee80211_add_iface, + .del_virtual_intf = ieee80211_del_iface, + .change_virtual_intf = ieee80211_change_iface, +diff -urNp linux-2.6.38.3/net/mac80211/cfg.h linux-2.6.38.3/net/mac80211/cfg.h +--- linux-2.6.38.3/net/mac80211/cfg.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/mac80211/cfg.h 2011-04-17 15:57:33.000000000 -0400 +@@ -4,6 +4,6 @@ + #ifndef __CFG_H + #define __CFG_H + +-extern struct cfg80211_ops mac80211_config_ops; ++extern const struct cfg80211_ops mac80211_config_ops; + + #endif /* __CFG_H */ +diff -urNp linux-2.6.38.3/net/mac80211/ieee80211_i.h linux-2.6.38.3/net/mac80211/ieee80211_i.h +--- linux-2.6.38.3/net/mac80211/ieee80211_i.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/mac80211/ieee80211_i.h 2011-04-17 15:57:33.000000000 -0400 +@@ -27,6 +27,7 @@ + #include <net/ieee80211_radiotap.h> + #include <net/cfg80211.h> + #include <net/mac80211.h> ++#include <asm/local.h> + #include "key.h" + #include "sta_info.h" + +@@ -716,7 +717,7 @@ struct ieee80211_local { + /* also used to protect ampdu_ac_queue and amdpu_ac_stop_refcnt */ + spinlock_t queue_stop_reason_lock; + +- int open_count; ++ local_t open_count; + int monitors, cooked_mntrs; + /* number of interfaces with corresponding FIF_ flags */ + int fif_fcsfail, fif_plcpfail, fif_control, fif_other_bss, fif_pspoll, +diff -urNp linux-2.6.38.3/net/mac80211/iface.c linux-2.6.38.3/net/mac80211/iface.c +--- linux-2.6.38.3/net/mac80211/iface.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/mac80211/iface.c 2011-04-17 15:57:33.000000000 -0400 +@@ -211,7 +211,7 @@ static int ieee80211_do_open(struct net_ + break; + } + +- if (local->open_count == 0) { ++ if (local_read(&local->open_count) == 0) { + res = drv_start(local); + if (res) + goto err_del_bss; +@@ -235,7 +235,7 @@ static int ieee80211_do_open(struct net_ + memcpy(dev->perm_addr, dev->dev_addr, ETH_ALEN); + + if (!is_valid_ether_addr(dev->dev_addr)) { +- if (!local->open_count) ++ if (!local_read(&local->open_count)) + drv_stop(local); + return -EADDRNOTAVAIL; + } +@@ -327,7 +327,7 @@ static int ieee80211_do_open(struct net_ + mutex_unlock(&local->mtx); + + if (coming_up) +- local->open_count++; ++ local_inc(&local->open_count); + + if (hw_reconf_flags) { + ieee80211_hw_config(local, hw_reconf_flags); +@@ -347,7 +347,7 @@ static int ieee80211_do_open(struct net_ + err_del_interface: + drv_remove_interface(local, &sdata->vif); + err_stop: +- if (!local->open_count) ++ if (!local_read(&local->open_count)) + drv_stop(local); + err_del_bss: + sdata->bss = NULL; +@@ -473,7 +473,7 @@ static void ieee80211_do_stop(struct iee + } + + if (going_down) +- local->open_count--; ++ local_dec(&local->open_count); + + switch (sdata->vif.type) { + case NL80211_IFTYPE_AP_VLAN: +@@ -532,7 +532,7 @@ static void ieee80211_do_stop(struct iee + + ieee80211_recalc_ps(local, -1); + +- if (local->open_count == 0) { ++ if (local_read(&local->open_count) == 0) { + if (local->ops->napi_poll) + napi_disable(&local->napi); + ieee80211_clear_tx_pending(local); +diff -urNp linux-2.6.38.3/net/mac80211/main.c linux-2.6.38.3/net/mac80211/main.c +--- linux-2.6.38.3/net/mac80211/main.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/mac80211/main.c 2011-04-17 15:57:33.000000000 -0400 +@@ -161,7 +161,7 @@ int ieee80211_hw_config(struct ieee80211 + local->hw.conf.power_level = power; + } + +- if (changed && local->open_count) { ++ if (changed && local_read(&local->open_count)) { + ret = drv_config(local, changed); + /* + * Goal: +diff -urNp linux-2.6.38.3/net/mac80211/pm.c linux-2.6.38.3/net/mac80211/pm.c +--- linux-2.6.38.3/net/mac80211/pm.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/mac80211/pm.c 2011-04-17 15:57:33.000000000 -0400 +@@ -95,7 +95,7 @@ int __ieee80211_suspend(struct ieee80211 + } + + /* stop hardware - this must stop RX */ +- if (local->open_count) ++ if (local_read(&local->open_count)) + ieee80211_stop_device(local); + + local->suspended = true; +diff -urNp linux-2.6.38.3/net/mac80211/rate.c linux-2.6.38.3/net/mac80211/rate.c +--- linux-2.6.38.3/net/mac80211/rate.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/mac80211/rate.c 2011-04-17 15:57:33.000000000 -0400 +@@ -371,7 +371,7 @@ int ieee80211_init_rate_ctrl_alg(struct + + ASSERT_RTNL(); + +- if (local->open_count) ++ if (local_read(&local->open_count)) + return -EBUSY; + + if (local->hw.flags & IEEE80211_HW_HAS_RATE_CONTROL) { +diff -urNp linux-2.6.38.3/net/mac80211/rc80211_pid_debugfs.c linux-2.6.38.3/net/mac80211/rc80211_pid_debugfs.c +--- linux-2.6.38.3/net/mac80211/rc80211_pid_debugfs.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/mac80211/rc80211_pid_debugfs.c 2011-04-17 15:57:33.000000000 -0400 +@@ -192,7 +192,7 @@ static ssize_t rate_control_pid_events_r + + spin_unlock_irqrestore(&events->lock, status); + +- if (copy_to_user(buf, pb, p)) ++ if (p > sizeof(pb) || copy_to_user(buf, pb, p)) + return -EFAULT; + + return p; +diff -urNp linux-2.6.38.3/net/mac80211/tx.c linux-2.6.38.3/net/mac80211/tx.c +--- linux-2.6.38.3/net/mac80211/tx.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/mac80211/tx.c 2011-04-17 15:57:33.000000000 -0400 +@@ -173,7 +173,7 @@ static __le16 ieee80211_duration(struct + return cpu_to_le16(dur); + } + +-static int inline is_ieee80211_device(struct ieee80211_local *local, ++static inline int is_ieee80211_device(struct ieee80211_local *local, + struct net_device *dev) + { + return local == wdev_priv(dev->ieee80211_ptr); +diff -urNp linux-2.6.38.3/net/mac80211/util.c linux-2.6.38.3/net/mac80211/util.c +--- linux-2.6.38.3/net/mac80211/util.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/mac80211/util.c 2011-04-17 15:57:33.000000000 -0400 +@@ -1135,7 +1135,7 @@ int ieee80211_reconfig(struct ieee80211_ + local->resuming = true; + + /* restart hardware */ +- if (local->open_count) { ++ if (local_read(&local->open_count)) { + /* + * Upon resume hardware can sometimes be goofy due to + * various platform / driver / bus issues, so restarting +diff -urNp linux-2.6.38.3/net/netfilter/Kconfig linux-2.6.38.3/net/netfilter/Kconfig +--- linux-2.6.38.3/net/netfilter/Kconfig 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/netfilter/Kconfig 2011-04-17 15:57:33.000000000 -0400 +@@ -709,6 +709,16 @@ config NETFILTER_XT_MATCH_ESP + + To compile it as a module, choose M here. If unsure, say N. + ++config NETFILTER_XT_MATCH_GRADM ++ tristate '"gradm" match support' ++ depends on NETFILTER_XTABLES && NETFILTER_ADVANCED ++ depends on GRKERNSEC && !GRKERNSEC_NO_RBAC ++ ---help--- ++ The gradm match allows to match on grsecurity RBAC being enabled. ++ It is useful when iptables rules are applied early on bootup to ++ prevent connections to the machine (except from a trusted host) ++ while the RBAC system is disabled. ++ + config NETFILTER_XT_MATCH_HASHLIMIT + tristate '"hashlimit" match support' + depends on (IP6_NF_IPTABLES || IP6_NF_IPTABLES=n) +diff -urNp linux-2.6.38.3/net/netfilter/Makefile linux-2.6.38.3/net/netfilter/Makefile +--- linux-2.6.38.3/net/netfilter/Makefile 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/netfilter/Makefile 2011-04-17 15:57:33.000000000 -0400 +@@ -74,6 +74,7 @@ obj-$(CONFIG_NETFILTER_XT_MATCH_CPU) += + obj-$(CONFIG_NETFILTER_XT_MATCH_DCCP) += xt_dccp.o + obj-$(CONFIG_NETFILTER_XT_MATCH_DSCP) += xt_dscp.o + obj-$(CONFIG_NETFILTER_XT_MATCH_ESP) += xt_esp.o ++obj-$(CONFIG_NETFILTER_XT_MATCH_GRADM) += xt_gradm.o + obj-$(CONFIG_NETFILTER_XT_MATCH_HASHLIMIT) += xt_hashlimit.o + obj-$(CONFIG_NETFILTER_XT_MATCH_HELPER) += xt_helper.o + obj-$(CONFIG_NETFILTER_XT_MATCH_HL) += xt_hl.o +diff -urNp linux-2.6.38.3/net/netfilter/nf_conntrack_netlink.c linux-2.6.38.3/net/netfilter/nf_conntrack_netlink.c +--- linux-2.6.38.3/net/netfilter/nf_conntrack_netlink.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/netfilter/nf_conntrack_netlink.c 2011-04-17 15:57:33.000000000 -0400 +@@ -761,7 +761,7 @@ static const struct nla_policy tuple_nla + static int + ctnetlink_parse_tuple(const struct nlattr * const cda[], + struct nf_conntrack_tuple *tuple, +- enum ctattr_tuple type, u_int8_t l3num) ++ enum ctattr_type type, u_int8_t l3num) + { + struct nlattr *tb[CTA_TUPLE_MAX+1]; + int err; +diff -urNp linux-2.6.38.3/net/netfilter/xt_gradm.c linux-2.6.38.3/net/netfilter/xt_gradm.c +--- linux-2.6.38.3/net/netfilter/xt_gradm.c 1969-12-31 19:00:00.000000000 -0500 ++++ linux-2.6.38.3/net/netfilter/xt_gradm.c 2011-04-17 15:57:33.000000000 -0400 +@@ -0,0 +1,51 @@ ++/* ++ * gradm match for netfilter ++ * Copyright © Zbigniew Krzystolik, 2010 ++ * ++ * This program is free software; you can redistribute it and/or modify ++ * it under the terms of the GNU General Public License; either version ++ * 2 or 3 as published by the Free Software Foundation. ++ */ ++#include <linux/module.h> ++#include <linux/moduleparam.h> ++#include <linux/skbuff.h> ++#include <linux/netfilter/x_tables.h> ++#include <linux/grsecurity.h> ++#include <linux/netfilter/xt_gradm.h> ++ ++static bool ++gradm_mt(const struct sk_buff *skb, struct xt_action_param *par) ++{ ++ const struct xt_gradm_mtinfo *info = par->matchinfo; ++ bool retval = false; ++ if (gr_acl_is_enabled()) ++ retval = true; ++ return retval ^ info->invflags; ++} ++ ++static struct xt_match gradm_mt_reg __read_mostly = { ++ .name = "gradm", ++ .revision = 0, ++ .family = NFPROTO_UNSPEC, ++ .match = gradm_mt, ++ .matchsize = XT_ALIGN(sizeof(struct xt_gradm_mtinfo)), ++ .me = THIS_MODULE, ++}; ++ ++static int __init gradm_mt_init(void) ++{ ++ return xt_register_match(&gradm_mt_reg); ++} ++ ++static void __exit gradm_mt_exit(void) ++{ ++ xt_unregister_match(&gradm_mt_reg); ++} ++ ++module_init(gradm_mt_init); ++module_exit(gradm_mt_exit); ++MODULE_AUTHOR("Zbigniew Krzystolik <zbyniu@destrukcja.pl>"); ++MODULE_DESCRIPTION("Xtables: Grsecurity RBAC match"); ++MODULE_LICENSE("GPL"); ++MODULE_ALIAS("ipt_gradm"); ++MODULE_ALIAS("ip6t_gradm"); +diff -urNp linux-2.6.38.3/net/netlink/af_netlink.c linux-2.6.38.3/net/netlink/af_netlink.c +--- linux-2.6.38.3/net/netlink/af_netlink.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/netlink/af_netlink.c 2011-04-17 15:57:33.000000000 -0400 +@@ -2001,13 +2001,21 @@ static int netlink_seq_show(struct seq_f + struct netlink_sock *nlk = nlk_sk(s); + + seq_printf(seq, "%p %-3d %-6d %08x %-8d %-8d %p %-8d %-8d %-8lu\n", ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ NULL, ++#else + s, ++#endif + s->sk_protocol, + nlk->pid, + nlk->groups ? (u32)nlk->groups[0] : 0, + sk_rmem_alloc_get(s), + sk_wmem_alloc_get(s), ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ NULL, ++#else + nlk->cb, ++#endif + atomic_read(&s->sk_refcnt), + atomic_read(&s->sk_drops), + sock_i_ino(s) +diff -urNp linux-2.6.38.3/net/netrom/af_netrom.c linux-2.6.38.3/net/netrom/af_netrom.c +--- linux-2.6.38.3/net/netrom/af_netrom.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/netrom/af_netrom.c 2011-04-17 15:57:33.000000000 -0400 +@@ -840,6 +840,7 @@ static int nr_getname(struct socket *soc + struct sock *sk = sock->sk; + struct nr_sock *nr = nr_sk(sk); + ++ memset(sax, 0, sizeof(*sax)); + lock_sock(sk); + if (peer != 0) { + if (sk->sk_state != TCP_ESTABLISHED) { +@@ -854,7 +855,6 @@ static int nr_getname(struct socket *soc + *uaddr_len = sizeof(struct full_sockaddr_ax25); + } else { + sax->fsa_ax25.sax25_family = AF_NETROM; +- sax->fsa_ax25.sax25_ndigis = 0; + sax->fsa_ax25.sax25_call = nr->source_addr; + *uaddr_len = sizeof(struct sockaddr_ax25); + } +diff -urNp linux-2.6.38.3/net/packet/af_packet.c linux-2.6.38.3/net/packet/af_packet.c +--- linux-2.6.38.3/net/packet/af_packet.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/packet/af_packet.c 2011-04-17 15:57:33.000000000 -0400 +@@ -2134,7 +2134,7 @@ static int packet_getsockopt(struct sock + case PACKET_HDRLEN: + if (len > sizeof(int)) + len = sizeof(int); +- if (copy_from_user(&val, optval, len)) ++ if (len > sizeof(val) || copy_from_user(&val, optval, len)) + return -EFAULT; + switch (val) { + case TPACKET_V1: +@@ -2172,7 +2172,7 @@ static int packet_getsockopt(struct sock + + if (put_user(len, optlen)) + return -EFAULT; +- if (copy_to_user(optval, data, len)) ++ if (len > sizeof(st) || copy_to_user(optval, data, len)) + return -EFAULT; + return 0; + } +@@ -2684,7 +2684,11 @@ static int packet_seq_show(struct seq_fi + + seq_printf(seq, + "%p %-6d %-4d %04x %-5d %1d %-6u %-6u %-6lu\n", ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ NULL, ++#else + s, ++#endif + atomic_read(&s->sk_refcnt), + s->sk_type, + ntohs(po->num), +diff -urNp linux-2.6.38.3/net/phonet/af_phonet.c linux-2.6.38.3/net/phonet/af_phonet.c +--- linux-2.6.38.3/net/phonet/af_phonet.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/phonet/af_phonet.c 2011-04-17 15:57:33.000000000 -0400 +@@ -41,7 +41,7 @@ static struct phonet_protocol *phonet_pr + { + struct phonet_protocol *pp; + +- if (protocol >= PHONET_NPROTO) ++ if (protocol < 0 || protocol >= PHONET_NPROTO) + return NULL; + + rcu_read_lock(); +@@ -463,7 +463,7 @@ int __init_or_module phonet_proto_regist + { + int err = 0; + +- if (protocol >= PHONET_NPROTO) ++ if (protocol < 0 || protocol >= PHONET_NPROTO) + return -EINVAL; + + err = proto_register(pp->prot, 1); +diff -urNp linux-2.6.38.3/net/phonet/socket.c linux-2.6.38.3/net/phonet/socket.c +--- linux-2.6.38.3/net/phonet/socket.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/phonet/socket.c 2011-04-17 15:57:33.000000000 -0400 +@@ -637,7 +637,12 @@ static int pn_sock_seq_show(struct seq_f + sk->sk_state, + sk_wmem_alloc_get(sk), sk_rmem_alloc_get(sk), + sock_i_uid(sk), sock_i_ino(sk), +- atomic_read(&sk->sk_refcnt), sk, ++ atomic_read(&sk->sk_refcnt), ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ NULL, ++#else ++ sk, ++#endif + atomic_read(&sk->sk_drops), &len); + } + seq_printf(seq, "%*s\n", 127 - len, ""); +diff -urNp linux-2.6.38.3/net/sctp/proc.c linux-2.6.38.3/net/sctp/proc.c +--- linux-2.6.38.3/net/sctp/proc.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/sctp/proc.c 2011-04-17 15:57:33.000000000 -0400 +@@ -212,7 +212,12 @@ static int sctp_eps_seq_show(struct seq_ + sctp_for_each_hentry(epb, node, &head->chain) { + ep = sctp_ep(epb); + sk = epb->sk; +- seq_printf(seq, "%8p %8p %-3d %-3d %-4d %-5d %5d %5lu ", ep, sk, ++ seq_printf(seq, "%8p %8p %-3d %-3d %-4d %-5d %5d %5lu ", ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ NULL, NULL, ++#else ++ ep, sk, ++#endif + sctp_sk(sk)->type, sk->sk_state, hash, + epb->bind_addr.port, + sock_i_uid(sk), sock_i_ino(sk)); +@@ -318,7 +323,12 @@ static int sctp_assocs_seq_show(struct s + seq_printf(seq, + "%8p %8p %-3d %-3d %-2d %-4d " + "%4d %8d %8d %7d %5lu %-5d %5d ", +- assoc, sk, sctp_sk(sk)->type, sk->sk_state, ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ NULL, NULL, ++#else ++ assoc, sk, ++#endif ++ sctp_sk(sk)->type, sk->sk_state, + assoc->state, hash, + assoc->assoc_id, + assoc->sndbuf_used, +diff -urNp linux-2.6.38.3/net/sctp/socket.c linux-2.6.38.3/net/sctp/socket.c +--- linux-2.6.38.3/net/sctp/socket.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/sctp/socket.c 2011-04-17 15:57:33.000000000 -0400 +@@ -1496,7 +1496,7 @@ SCTP_STATIC int sctp_sendmsg(struct kioc + struct sctp_sndrcvinfo *sinfo; + struct sctp_initmsg *sinit; + sctp_assoc_t associd = 0; +- sctp_cmsgs_t cmsgs = { NULL }; ++ sctp_cmsgs_t cmsgs = { NULL, NULL }; + int err; + sctp_scope_t scope; + long timeo; +@@ -4435,7 +4435,7 @@ static int sctp_getsockopt_peer_addrs(st + addrlen = sctp_get_af_specific(temp.sa.sa_family)->sockaddr_len; + if (space_left < addrlen) + return -ENOMEM; +- if (copy_to_user(to, &temp, addrlen)) ++ if (addrlen > sizeof(temp) || copy_to_user(to, &temp, addrlen)) + return -EFAULT; + to += addrlen; + cnt++; +diff -urNp linux-2.6.38.3/net/socket.c linux-2.6.38.3/net/socket.c +--- linux-2.6.38.3/net/socket.c 2011-03-28 17:42:40.000000000 -0400 ++++ linux-2.6.38.3/net/socket.c 2011-04-17 15:57:33.000000000 -0400 +@@ -88,6 +88,7 @@ + #include <linux/nsproxy.h> + #include <linux/magic.h> + #include <linux/slab.h> ++#include <linux/in.h> + + #include <asm/uaccess.h> + #include <asm/unistd.h> +@@ -105,6 +106,8 @@ + #include <linux/sockios.h> + #include <linux/atalk.h> + ++#include <linux/grsock.h> ++ + static int sock_no_open(struct inode *irrelevant, struct file *dontcare); + static ssize_t sock_aio_read(struct kiocb *iocb, const struct iovec *iov, + unsigned long nr_segs, loff_t pos); +@@ -326,7 +329,7 @@ static struct dentry *sockfs_mount(struc + &sockfs_dentry_operations, SOCKFS_MAGIC); + } + +-static struct vfsmount *sock_mnt __read_mostly; ++struct vfsmount *sock_mnt __read_mostly; + + static struct file_system_type sock_fs_type = { + .name = "sockfs", +@@ -1174,6 +1177,8 @@ int __sock_create(struct net *net, int f + return -EAFNOSUPPORT; + if (type < 0 || type >= SOCK_MAX) + return -EINVAL; ++ if (protocol < 0) ++ return -EINVAL; + + /* Compatibility. + +@@ -1306,6 +1311,16 @@ SYSCALL_DEFINE3(socket, int, family, int + if (SOCK_NONBLOCK != O_NONBLOCK && (flags & SOCK_NONBLOCK)) + flags = (flags & ~SOCK_NONBLOCK) | O_NONBLOCK; + ++ if(!gr_search_socket(family, type, protocol)) { ++ retval = -EACCES; ++ goto out; ++ } ++ ++ if (gr_handle_sock_all(family, type, protocol)) { ++ retval = -EACCES; ++ goto out; ++ } ++ + retval = sock_create(family, type, protocol, &sock); + if (retval < 0) + goto out; +@@ -1418,6 +1433,14 @@ SYSCALL_DEFINE3(bind, int, fd, struct so + if (sock) { + err = move_addr_to_kernel(umyaddr, addrlen, (struct sockaddr *)&address); + if (err >= 0) { ++ if (gr_handle_sock_server((struct sockaddr *)&address)) { ++ err = -EACCES; ++ goto error; ++ } ++ err = gr_search_bind(sock, (struct sockaddr_in *)&address); ++ if (err) ++ goto error; ++ + err = security_socket_bind(sock, + (struct sockaddr *)&address, + addrlen); +@@ -1426,6 +1449,7 @@ SYSCALL_DEFINE3(bind, int, fd, struct so + (struct sockaddr *) + &address, addrlen); + } ++error: + fput_light(sock->file, fput_needed); + } + return err; +@@ -1449,10 +1473,20 @@ SYSCALL_DEFINE2(listen, int, fd, int, ba + if ((unsigned)backlog > somaxconn) + backlog = somaxconn; + ++ if (gr_handle_sock_server_other(sock->sk)) { ++ err = -EPERM; ++ goto error; ++ } ++ ++ err = gr_search_listen(sock); ++ if (err) ++ goto error; ++ + err = security_socket_listen(sock, backlog); + if (!err) + err = sock->ops->listen(sock, backlog); + ++error: + fput_light(sock->file, fput_needed); + } + return err; +@@ -1496,6 +1530,18 @@ SYSCALL_DEFINE4(accept4, int, fd, struct + newsock->type = sock->type; + newsock->ops = sock->ops; + ++ if (gr_handle_sock_server_other(sock->sk)) { ++ err = -EPERM; ++ sock_release(newsock); ++ goto out_put; ++ } ++ ++ err = gr_search_accept(sock); ++ if (err) { ++ sock_release(newsock); ++ goto out_put; ++ } ++ + /* + * We don't need try_module_get here, as the listening socket (sock) + * has the protocol module (sock->ops->owner) held. +@@ -1534,6 +1580,8 @@ SYSCALL_DEFINE4(accept4, int, fd, struct + fd_install(newfd, newfile); + err = newfd; + ++ gr_attach_curr_ip(newsock->sk); ++ + out_put: + fput_light(sock->file, fput_needed); + out: +@@ -1566,6 +1614,7 @@ SYSCALL_DEFINE3(connect, int, fd, struct + int, addrlen) + { + struct socket *sock; ++ struct sockaddr *sck; + struct sockaddr_storage address; + int err, fput_needed; + +@@ -1576,6 +1625,17 @@ SYSCALL_DEFINE3(connect, int, fd, struct + if (err < 0) + goto out_put; + ++ sck = (struct sockaddr *)&address; ++ ++ if (gr_handle_sock_client(sck)) { ++ err = -EACCES; ++ goto out_put; ++ } ++ ++ err = gr_search_connect(sock, (struct sockaddr_in *)sck); ++ if (err) ++ goto out_put; ++ + err = + security_socket_connect(sock, (struct sockaddr *)&address, addrlen); + if (err) +diff -urNp linux-2.6.38.3/net/sunrpc/sched.c linux-2.6.38.3/net/sunrpc/sched.c +--- linux-2.6.38.3/net/sunrpc/sched.c 2011-03-23 17:20:08.000000000 -0400 ++++ linux-2.6.38.3/net/sunrpc/sched.c 2011-04-17 15:57:33.000000000 -0400 +@@ -234,9 +234,9 @@ static int rpc_wait_bit_killable(void *w + #ifdef RPC_DEBUG + static void rpc_task_set_debuginfo(struct rpc_task *task) + { +- static atomic_t rpc_pid; ++ static atomic_unchecked_t rpc_pid; + +- task->tk_pid = atomic_inc_return(&rpc_pid); ++ task->tk_pid = atomic_inc_return_unchecked(&rpc_pid); + } + #else + static inline void rpc_task_set_debuginfo(struct rpc_task *task) +diff -urNp linux-2.6.38.3/net/sunrpc/xprtrdma/svc_rdma.c linux-2.6.38.3/net/sunrpc/xprtrdma/svc_rdma.c +--- linux-2.6.38.3/net/sunrpc/xprtrdma/svc_rdma.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/sunrpc/xprtrdma/svc_rdma.c 2011-04-17 15:57:33.000000000 -0400 +@@ -109,7 +109,7 @@ static int read_reset_stat(ctl_table *ta + len -= *ppos; + if (len > *lenp) + len = *lenp; +- if (len && copy_to_user(buffer, str_buf, len)) ++ if (len > sizeof str_buf || (len && copy_to_user(buffer, str_buf, len))) + return -EFAULT; + *lenp = len; + *ppos += len; +diff -urNp linux-2.6.38.3/net/sysctl_net.c linux-2.6.38.3/net/sysctl_net.c +--- linux-2.6.38.3/net/sysctl_net.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/sysctl_net.c 2011-04-17 15:57:33.000000000 -0400 +@@ -46,7 +46,7 @@ static int net_ctl_permissions(struct ct + struct ctl_table *table) + { + /* Allow network administrator to have same access as root. */ +- if (capable(CAP_NET_ADMIN)) { ++ if (capable_nolog(CAP_NET_ADMIN)) { + int mode = (table->mode >> 6) & 7; + return (mode << 6) | (mode << 3) | mode; + } +diff -urNp linux-2.6.38.3/net/tipc/socket.c linux-2.6.38.3/net/tipc/socket.c +--- linux-2.6.38.3/net/tipc/socket.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/tipc/socket.c 2011-04-17 15:57:33.000000000 -0400 +@@ -1447,8 +1447,9 @@ static int connect(struct socket *sock, + } else { + if (res == 0) + res = -ETIMEDOUT; +- else +- ; /* leave "res" unchanged */ ++ else { ++ /* leave "res" unchanged */ ++ } + sock->state = SS_DISCONNECTING; + } + +diff -urNp linux-2.6.38.3/net/unix/af_unix.c linux-2.6.38.3/net/unix/af_unix.c +--- linux-2.6.38.3/net/unix/af_unix.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/unix/af_unix.c 2011-04-17 15:57:33.000000000 -0400 +@@ -765,6 +765,12 @@ static struct sock *unix_find_other(stru + err = -ECONNREFUSED; + if (!S_ISSOCK(inode->i_mode)) + goto put_fail; ++ ++ if (!gr_acl_handle_unix(path.dentry, path.mnt)) { ++ err = -EACCES; ++ goto put_fail; ++ } ++ + u = unix_find_socket_byinode(inode); + if (!u) + goto put_fail; +@@ -785,6 +791,13 @@ static struct sock *unix_find_other(stru + if (u) { + struct dentry *dentry; + dentry = unix_sk(u)->dentry; ++ ++ if (!gr_handle_chroot_unix(u->sk_peer_pid)) { ++ err = -EPERM; ++ sock_put(u); ++ goto fail; ++ } ++ + if (dentry) + touch_atime(unix_sk(u)->mnt, dentry); + } else +@@ -870,11 +883,18 @@ static int unix_bind(struct socket *sock + err = security_path_mknod(&nd.path, dentry, mode, 0); + if (err) + goto out_mknod_drop_write; ++ if (!gr_acl_handle_mknod(dentry, nd.path.dentry, nd.path.mnt, mode)) { ++ err = -EACCES; ++ goto out_mknod_drop_write; ++ } + err = vfs_mknod(nd.path.dentry->d_inode, dentry, mode, 0); + out_mknod_drop_write: + mnt_drop_write(nd.path.mnt); + if (err) + goto out_mknod_dput; ++ ++ gr_handle_create(dentry, nd.path.mnt); ++ + mutex_unlock(&nd.path.dentry->d_inode->i_mutex); + dput(nd.path.dentry); + nd.path.dentry = dentry; +@@ -892,6 +912,11 @@ out_mknod_drop_write: + goto out_unlock; + } + ++#ifdef CONFIG_GRKERNSEC_CHROOT_UNIX ++ put_pid(sk->sk_peer_pid); ++ sk->sk_peer_pid = get_pid(task_tgid(current)); ++#endif ++ + list = &unix_socket_table[addr->hash]; + } else { + list = &unix_socket_table[dentry->d_inode->i_ino & (UNIX_HASH_SIZE-1)]; +@@ -2235,7 +2260,11 @@ static int unix_seq_show(struct seq_file + unix_state_lock(s); + + seq_printf(seq, "%p: %08X %08X %08X %04X %02X %5lu", ++#ifdef CONFIG_GRKERNSEC_HIDESYM ++ NULL, ++#else + s, ++#endif + atomic_read(&s->sk_refcnt), + 0, + s->sk_state == TCP_LISTEN ? __SO_ACCEPTCON : 0, +diff -urNp linux-2.6.38.3/net/wireless/reg.c linux-2.6.38.3/net/wireless/reg.c +--- linux-2.6.38.3/net/wireless/reg.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/wireless/reg.c 2011-04-17 15:57:33.000000000 -0400 +@@ -54,7 +54,7 @@ + printk(KERN_DEBUG pr_fmt(format), ##args); \ + } while (0) + #else +-#define REG_DBG_PRINT(args...) ++#define REG_DBG_PRINT(args...) do {} while (0) + #endif + + /* Receipt of information from last regulatory request */ +diff -urNp linux-2.6.38.3/net/wireless/wext-core.c linux-2.6.38.3/net/wireless/wext-core.c +--- linux-2.6.38.3/net/wireless/wext-core.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/wireless/wext-core.c 2011-04-17 15:57:33.000000000 -0400 +@@ -746,8 +746,7 @@ static int ioctl_standard_iw_point(struc + */ + + /* Support for very large requests */ +- if ((descr->flags & IW_DESCR_FLAG_NOMAX) && +- (user_length > descr->max_tokens)) { ++ if (user_length > descr->max_tokens) { + /* Allow userspace to GET more than max so + * we can support any size GET requests. + * There is still a limit : -ENOMEM. +@@ -784,22 +783,6 @@ static int ioctl_standard_iw_point(struc + } + } + +- if (IW_IS_GET(cmd) && !(descr->flags & IW_DESCR_FLAG_NOMAX)) { +- /* +- * If this is a GET, but not NOMAX, it means that the extra +- * data is not bounded by userspace, but by max_tokens. Thus +- * set the length to max_tokens. This matches the extra data +- * allocation. +- * The driver should fill it with the number of tokens it +- * provided, and it may check iwp->length rather than having +- * knowledge of max_tokens. If the driver doesn't change the +- * iwp->length, this ioctl just copies back max_token tokens +- * filled with zeroes. Hopefully the driver isn't claiming +- * them to be valid data. +- */ +- iwp->length = descr->max_tokens; +- } +- + err = handler(dev, info, (union iwreq_data *) iwp, extra); + + iwp->length += essid_compat; +diff -urNp linux-2.6.38.3/net/x25/x25_facilities.c linux-2.6.38.3/net/x25/x25_facilities.c +--- linux-2.6.38.3/net/x25/x25_facilities.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/x25/x25_facilities.c 2011-04-17 15:57:33.000000000 -0400 +@@ -167,7 +167,8 @@ int x25_parse_facilities(struct sk_buff + break; + default: + printk(KERN_DEBUG "X.25: unknown facility %02X," +- "length %d\n", p[0], p[1]); ++ "length %d, values %02X, %02X\n", ++ p[0], p[1], p[2], p[3]); + break; + } + len -= p[1] + 2; +diff -urNp linux-2.6.38.3/net/xfrm/xfrm_policy.c linux-2.6.38.3/net/xfrm/xfrm_policy.c +--- linux-2.6.38.3/net/xfrm/xfrm_policy.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/net/xfrm/xfrm_policy.c 2011-04-17 15:57:33.000000000 -0400 +@@ -1507,7 +1507,7 @@ free_dst: + goto out; + } + +-static int inline ++static inline int + xfrm_dst_alloc_copy(void **target, void *src, int size) + { + if (!*target) { +@@ -1519,7 +1519,7 @@ xfrm_dst_alloc_copy(void **target, void + return 0; + } + +-static int inline ++static inline int + xfrm_dst_update_parent(struct dst_entry *dst, struct xfrm_selector *sel) + { + #ifdef CONFIG_XFRM_SUB_POLICY +@@ -1531,7 +1531,7 @@ xfrm_dst_update_parent(struct dst_entry + #endif + } + +-static int inline ++static inline int + xfrm_dst_update_origin(struct dst_entry *dst, struct flowi *fl) + { + #ifdef CONFIG_XFRM_SUB_POLICY +diff -urNp linux-2.6.38.3/scripts/basic/fixdep.c linux-2.6.38.3/scripts/basic/fixdep.c +--- linux-2.6.38.3/scripts/basic/fixdep.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/scripts/basic/fixdep.c 2011-04-17 15:57:33.000000000 -0400 +@@ -235,9 +235,9 @@ static void use_config(const char *m, in + + static void parse_config_file(const char *map, size_t len) + { +- const int *end = (const int *) (map + len); ++ const unsigned int *end = (const unsigned int *) (map + len); + /* start at +1, so that p can never be < map */ +- const int *m = (const int *) map + 1; ++ const unsigned int *m = (const unsigned int *) map + 1; + const char *p, *q; + + for (; m < end; m++) { +@@ -405,7 +405,7 @@ static void print_deps(void) + static void traps(void) + { + static char test[] __attribute__((aligned(sizeof(int)))) = "CONF"; +- int *p = (int *)test; ++ unsigned int *p = (unsigned int *)test; + + if (*p != INT_CONF) { + fprintf(stderr, "fixdep: sizeof(int) != 4 or wrong endianess? %#x\n", +diff -urNp linux-2.6.38.3/scripts/kallsyms.c linux-2.6.38.3/scripts/kallsyms.c +--- linux-2.6.38.3/scripts/kallsyms.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/scripts/kallsyms.c 2011-04-17 15:57:33.000000000 -0400 +@@ -43,10 +43,10 @@ struct text_range { + + static unsigned long long _text; + static struct text_range text_ranges[] = { +- { "_stext", "_etext" }, +- { "_sinittext", "_einittext" }, +- { "_stext_l1", "_etext_l1" }, /* Blackfin on-chip L1 inst SRAM */ +- { "_stext_l2", "_etext_l2" }, /* Blackfin on-chip L2 SRAM */ ++ { "_stext", "_etext", 0, 0 }, ++ { "_sinittext", "_einittext", 0, 0 }, ++ { "_stext_l1", "_etext_l1", 0, 0 }, /* Blackfin on-chip L1 inst SRAM */ ++ { "_stext_l2", "_etext_l2", 0, 0 }, /* Blackfin on-chip L2 SRAM */ + }; + #define text_range_text (&text_ranges[0]) + #define text_range_inittext (&text_ranges[1]) +diff -urNp linux-2.6.38.3/scripts/mod/file2alias.c linux-2.6.38.3/scripts/mod/file2alias.c +--- linux-2.6.38.3/scripts/mod/file2alias.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/scripts/mod/file2alias.c 2011-04-17 15:57:33.000000000 -0400 +@@ -72,7 +72,7 @@ static void device_id_check(const char * + unsigned long size, unsigned long id_size, + void *symval) + { +- int i; ++ unsigned int i; + + if (size % id_size || size < id_size) { + if (cross_build != 0) +@@ -102,7 +102,7 @@ static void device_id_check(const char * + /* USB is special because the bcdDevice can be matched against a numeric range */ + /* Looks like "usb:vNpNdNdcNdscNdpNicNiscNipN" */ + static void do_usb_entry(struct usb_device_id *id, +- unsigned int bcdDevice_initial, int bcdDevice_initial_digits, ++ unsigned int bcdDevice_initial, unsigned int bcdDevice_initial_digits, + unsigned char range_lo, unsigned char range_hi, + unsigned char max, struct module *mod) + { +@@ -437,7 +437,7 @@ static void do_pnp_device_entry(void *sy + for (i = 0; i < count; i++) { + const char *id = (char *)devs[i].id; + char acpi_id[sizeof(devs[0].id)]; +- int j; ++ unsigned int j; + + buf_printf(&mod->dev_table_buf, + "MODULE_ALIAS(\"pnp:d%s*\");\n", id); +@@ -467,7 +467,7 @@ static void do_pnp_card_entries(void *sy + + for (j = 0; j < PNP_MAX_DEVICES; j++) { + const char *id = (char *)card->devs[j].id; +- int i2, j2; ++ unsigned int i2, j2; + int dup = 0; + + if (!id[0]) +@@ -493,7 +493,7 @@ static void do_pnp_card_entries(void *sy + /* add an individual alias for every device entry */ + if (!dup) { + char acpi_id[sizeof(card->devs[0].id)]; +- int k; ++ unsigned int k; + + buf_printf(&mod->dev_table_buf, + "MODULE_ALIAS(\"pnp:d%s*\");\n", id); +@@ -768,7 +768,7 @@ static void dmi_ascii_filter(char *d, co + static int do_dmi_entry(const char *filename, struct dmi_system_id *id, + char *alias) + { +- int i, j; ++ unsigned int i, j; + + sprintf(alias, "dmi*"); + +diff -urNp linux-2.6.38.3/scripts/mod/modpost.c linux-2.6.38.3/scripts/mod/modpost.c +--- linux-2.6.38.3/scripts/mod/modpost.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/scripts/mod/modpost.c 2011-04-17 15:57:33.000000000 -0400 +@@ -896,6 +896,7 @@ enum mismatch { + ANY_INIT_TO_ANY_EXIT, + ANY_EXIT_TO_ANY_INIT, + EXPORT_TO_INIT_EXIT, ++ DATA_TO_TEXT + }; + + struct sectioncheck { +@@ -1004,6 +1005,12 @@ const struct sectioncheck sectioncheck[] + .tosec = { INIT_SECTIONS, EXIT_SECTIONS, NULL }, + .mismatch = EXPORT_TO_INIT_EXIT, + .symbol_white_list = { DEFAULT_SYMBOL_WHITE_LIST, NULL }, ++}, ++/* Do not reference code from writable data */ ++{ ++ .fromsec = { DATA_SECTIONS, NULL }, ++ .tosec = { TEXT_SECTIONS, NULL }, ++ .mismatch = DATA_TO_TEXT + } + }; + +@@ -1126,10 +1133,10 @@ static Elf_Sym *find_elf_symbol(struct e + continue; + if (ELF_ST_TYPE(sym->st_info) == STT_SECTION) + continue; +- if (sym->st_value == addr) +- return sym; + /* Find a symbol nearby - addr are maybe negative */ + d = sym->st_value - addr; ++ if (d == 0) ++ return sym; + if (d < 0) + d = addr - sym->st_value; + if (d < distance) { +@@ -1401,6 +1408,14 @@ static void report_sec_mismatch(const ch + tosym, prl_to, prl_to, tosym); + free(prl_to); + break; ++ case DATA_TO_TEXT: ++/* ++ fprintf(stderr, ++ "The variable %s references\n" ++ "the %s %s%s%s\n", ++ fromsym, to, sec2annotation(tosec), tosym, to_p); ++*/ ++ break; + } + fprintf(stderr, "\n"); + } +@@ -1724,7 +1739,7 @@ void __attribute__((format(printf, 2, 3) + va_end(ap); + } + +-void buf_write(struct buffer *buf, const char *s, int len) ++void buf_write(struct buffer *buf, const char *s, unsigned int len) + { + if (buf->size - buf->pos < len) { + buf->size += len + SZ; +@@ -1936,7 +1951,7 @@ static void write_if_changed(struct buff + if (fstat(fileno(file), &st) < 0) + goto close_write; + +- if (st.st_size != b->pos) ++ if (st.st_size != (off_t)b->pos) + goto close_write; + + tmp = NOFAIL(malloc(b->pos)); +diff -urNp linux-2.6.38.3/scripts/mod/modpost.h linux-2.6.38.3/scripts/mod/modpost.h +--- linux-2.6.38.3/scripts/mod/modpost.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/scripts/mod/modpost.h 2011-04-17 15:57:33.000000000 -0400 +@@ -92,15 +92,15 @@ void *do_nofail(void *ptr, const char *e + + struct buffer { + char *p; +- int pos; +- int size; ++ unsigned int pos; ++ unsigned int size; + }; + + void __attribute__((format(printf, 2, 3))) + buf_printf(struct buffer *buf, const char *fmt, ...); + + void +-buf_write(struct buffer *buf, const char *s, int len); ++buf_write(struct buffer *buf, const char *s, unsigned int len); + + struct module { + struct module *next; +diff -urNp linux-2.6.38.3/scripts/mod/sumversion.c linux-2.6.38.3/scripts/mod/sumversion.c +--- linux-2.6.38.3/scripts/mod/sumversion.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/scripts/mod/sumversion.c 2011-04-17 15:57:33.000000000 -0400 +@@ -470,7 +470,7 @@ static void write_version(const char *fi + goto out; + } + +- if (write(fd, sum, strlen(sum)+1) != strlen(sum)+1) { ++ if (write(fd, sum, strlen(sum)+1) != (ssize_t)strlen(sum)+1) { + warn("writing sum in %s failed: %s\n", + filename, strerror(errno)); + goto out; +diff -urNp linux-2.6.38.3/scripts/pnmtologo.c linux-2.6.38.3/scripts/pnmtologo.c +--- linux-2.6.38.3/scripts/pnmtologo.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/scripts/pnmtologo.c 2011-04-17 15:57:33.000000000 -0400 +@@ -237,14 +237,14 @@ static void write_header(void) + fprintf(out, " * Linux logo %s\n", logoname); + fputs(" */\n\n", out); + fputs("#include <linux/linux_logo.h>\n\n", out); +- fprintf(out, "static unsigned char %s_data[] __initdata = {\n", ++ fprintf(out, "static unsigned char %s_data[] = {\n", + logoname); + } + + static void write_footer(void) + { + fputs("\n};\n\n", out); +- fprintf(out, "const struct linux_logo %s __initconst = {\n", logoname); ++ fprintf(out, "const struct linux_logo %s = {\n", logoname); + fprintf(out, "\t.type\t\t= %s,\n", logo_types[logo_type]); + fprintf(out, "\t.width\t\t= %d,\n", logo_width); + fprintf(out, "\t.height\t\t= %d,\n", logo_height); +@@ -374,7 +374,7 @@ static void write_logo_clut224(void) + fputs("\n};\n\n", out); + + /* write logo clut */ +- fprintf(out, "static unsigned char %s_clut[] __initdata = {\n", ++ fprintf(out, "static unsigned char %s_clut[] = {\n", + logoname); + write_hex_cnt = 0; + for (i = 0; i < logo_clutsize; i++) { +diff -urNp linux-2.6.38.3/security/apparmor/lsm.c linux-2.6.38.3/security/apparmor/lsm.c +--- linux-2.6.38.3/security/apparmor/lsm.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/security/apparmor/lsm.c 2011-04-17 15:57:33.000000000 -0400 +@@ -619,7 +619,7 @@ static int apparmor_task_setrlimit(struc + return error; + } + +-static struct security_operations apparmor_ops = { ++static struct security_operations apparmor_ops __read_only = { + .name = "apparmor", + + .ptrace_access_check = apparmor_ptrace_access_check, +diff -urNp linux-2.6.38.3/security/commoncap.c linux-2.6.38.3/security/commoncap.c +--- linux-2.6.38.3/security/commoncap.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/security/commoncap.c 2011-04-17 15:57:33.000000000 -0400 +@@ -27,6 +27,7 @@ + #include <linux/sched.h> + #include <linux/prctl.h> + #include <linux/securebits.h> ++#include <net/sock.h> + + /* + * If a non-root user executes a setuid-root binary in +@@ -50,9 +51,11 @@ static void warn_setuid_and_fcaps_mixed( + } + } + ++extern kernel_cap_t gr_cap_rtnetlink(struct sock *sk); ++ + int cap_netlink_send(struct sock *sk, struct sk_buff *skb) + { +- NETLINK_CB(skb).eff_cap = current_cap(); ++ NETLINK_CB(skb).eff_cap = gr_cap_rtnetlink(sk); + return 0; + } + +@@ -534,6 +537,9 @@ int cap_bprm_secureexec(struct linux_bin + { + const struct cred *cred = current_cred(); + ++ if (gr_acl_enable_at_secure()) ++ return 1; ++ + if (cred->uid != 0) { + if (bprm->cap_effective) + return 1; +diff -urNp linux-2.6.38.3/security/integrity/ima/ima_api.c linux-2.6.38.3/security/integrity/ima/ima_api.c +--- linux-2.6.38.3/security/integrity/ima/ima_api.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/security/integrity/ima/ima_api.c 2011-04-17 15:57:33.000000000 -0400 +@@ -75,7 +75,7 @@ void ima_add_violation(struct inode *ino + int result; + + /* can overflow, only indicator */ +- atomic_long_inc(&ima_htable.violations); ++ atomic_long_inc_unchecked(&ima_htable.violations); + + entry = kmalloc(sizeof(*entry), GFP_KERNEL); + if (!entry) { +diff -urNp linux-2.6.38.3/security/integrity/ima/ima_fs.c linux-2.6.38.3/security/integrity/ima/ima_fs.c +--- linux-2.6.38.3/security/integrity/ima/ima_fs.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/security/integrity/ima/ima_fs.c 2011-04-17 15:57:33.000000000 -0400 +@@ -28,12 +28,12 @@ + static int valid_policy = 1; + #define TMPBUFLEN 12 + static ssize_t ima_show_htable_value(char __user *buf, size_t count, +- loff_t *ppos, atomic_long_t *val) ++ loff_t *ppos, atomic_long_unchecked_t *val) + { + char tmpbuf[TMPBUFLEN]; + ssize_t len; + +- len = scnprintf(tmpbuf, TMPBUFLEN, "%li\n", atomic_long_read(val)); ++ len = scnprintf(tmpbuf, TMPBUFLEN, "%li\n", atomic_long_read_unchecked(val)); + return simple_read_from_buffer(buf, count, ppos, tmpbuf, len); + } + +diff -urNp linux-2.6.38.3/security/integrity/ima/ima.h linux-2.6.38.3/security/integrity/ima/ima.h +--- linux-2.6.38.3/security/integrity/ima/ima.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/security/integrity/ima/ima.h 2011-04-17 15:57:33.000000000 -0400 +@@ -85,8 +85,8 @@ void ima_add_violation(struct inode *ino + extern spinlock_t ima_queue_lock; + + struct ima_h_table { +- atomic_long_t len; /* number of stored measurements in the list */ +- atomic_long_t violations; ++ atomic_long_unchecked_t len; /* number of stored measurements in the list */ ++ atomic_long_unchecked_t violations; + struct hlist_head queue[IMA_MEASURE_HTABLE_SIZE]; + }; + extern struct ima_h_table ima_htable; +diff -urNp linux-2.6.38.3/security/integrity/ima/ima_queue.c linux-2.6.38.3/security/integrity/ima/ima_queue.c +--- linux-2.6.38.3/security/integrity/ima/ima_queue.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/security/integrity/ima/ima_queue.c 2011-04-17 15:57:33.000000000 -0400 +@@ -79,7 +79,7 @@ static int ima_add_digest_entry(struct i + INIT_LIST_HEAD(&qe->later); + list_add_tail_rcu(&qe->later, &ima_measurements); + +- atomic_long_inc(&ima_htable.len); ++ atomic_long_inc_unchecked(&ima_htable.len); + key = ima_hash_key(entry->digest); + hlist_add_head_rcu(&qe->hnext, &ima_htable.queue[key]); + return 0; +diff -urNp linux-2.6.38.3/security/Kconfig linux-2.6.38.3/security/Kconfig +--- linux-2.6.38.3/security/Kconfig 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/security/Kconfig 2011-04-17 15:57:33.000000000 -0400 +@@ -4,6 +4,527 @@ + + menu "Security options" + ++source grsecurity/Kconfig ++ ++menu "PaX" ++ ++ config ARCH_TRACK_EXEC_LIMIT ++ bool ++ ++ config PAX_PER_CPU_PGD ++ bool ++ ++ config TASK_SIZE_MAX_SHIFT ++ int ++ depends on X86_64 ++ default 47 if !PAX_PER_CPU_PGD ++ default 42 if PAX_PER_CPU_PGD ++ ++ config PAX_ENABLE_PAE ++ bool ++ default y if (X86_32 && (MPENTIUM4 || MK8 || MPSC || MCORE2 || MATOM)) ++ ++config PAX ++ bool "Enable various PaX features" ++ depends on GRKERNSEC && (ALPHA || ARM || AVR32 || IA64 || MIPS || PARISC || PPC || SPARC || X86) ++ help ++ This allows you to enable various PaX features. PaX adds ++ intrusion prevention mechanisms to the kernel that reduce ++ the risks posed by exploitable memory corruption bugs. ++ ++menu "PaX Control" ++ depends on PAX ++ ++config PAX_SOFTMODE ++ bool 'Support soft mode' ++ select PAX_PT_PAX_FLAGS ++ help ++ Enabling this option will allow you to run PaX in soft mode, that ++ is, PaX features will not be enforced by default, only on executables ++ marked explicitly. You must also enable PT_PAX_FLAGS support as it ++ is the only way to mark executables for soft mode use. ++ ++ Soft mode can be activated by using the "pax_softmode=1" kernel command ++ line option on boot. Furthermore you can control various PaX features ++ at runtime via the entries in /proc/sys/kernel/pax. ++ ++config PAX_EI_PAX ++ bool 'Use legacy ELF header marking' ++ help ++ Enabling this option will allow you to control PaX features on ++ a per executable basis via the 'chpax' utility available at ++ http://pax.grsecurity.net/. The control flags will be read from ++ an otherwise reserved part of the ELF header. This marking has ++ numerous drawbacks (no support for soft-mode, toolchain does not ++ know about the non-standard use of the ELF header) therefore it ++ has been deprecated in favour of PT_PAX_FLAGS support. ++ ++ Note that if you enable PT_PAX_FLAGS marking support as well, ++ the PT_PAX_FLAG marks will override the legacy EI_PAX marks. ++ ++config PAX_PT_PAX_FLAGS ++ bool 'Use ELF program header marking' ++ help ++ Enabling this option will allow you to control PaX features on ++ a per executable basis via the 'paxctl' utility available at ++ http://pax.grsecurity.net/. The control flags will be read from ++ a PaX specific ELF program header (PT_PAX_FLAGS). This marking ++ has the benefits of supporting both soft mode and being fully ++ integrated into the toolchain (the binutils patch is available ++ from http://pax.grsecurity.net). ++ ++ If your toolchain does not support PT_PAX_FLAGS markings, ++ you can create one in most cases with 'paxctl -C'. ++ ++ Note that if you enable the legacy EI_PAX marking support as well, ++ the EI_PAX marks will be overridden by the PT_PAX_FLAGS marks. ++ ++choice ++ prompt 'MAC system integration' ++ default PAX_HAVE_ACL_FLAGS ++ help ++ Mandatory Access Control systems have the option of controlling ++ PaX flags on a per executable basis, choose the method supported ++ by your particular system. ++ ++ - "none": if your MAC system does not interact with PaX, ++ - "direct": if your MAC system defines pax_set_initial_flags() itself, ++ - "hook": if your MAC system uses the pax_set_initial_flags_func callback. ++ ++ NOTE: this option is for developers/integrators only. ++ ++ config PAX_NO_ACL_FLAGS ++ bool 'none' ++ ++ config PAX_HAVE_ACL_FLAGS ++ bool 'direct' ++ ++ config PAX_HOOK_ACL_FLAGS ++ bool 'hook' ++endchoice ++ ++endmenu ++ ++menu "Non-executable pages" ++ depends on PAX ++ ++config PAX_NOEXEC ++ bool "Enforce non-executable pages" ++ depends on (PAX_EI_PAX || PAX_PT_PAX_FLAGS || PAX_HAVE_ACL_FLAGS || PAX_HOOK_ACL_FLAGS) && (ALPHA || (ARM && (CPU_V6 || CPU_V7)) || IA64 || MIPS || PARISC || PPC || S390 || SPARC || X86) ++ help ++ By design some architectures do not allow for protecting memory ++ pages against execution or even if they do, Linux does not make ++ use of this feature. In practice this means that if a page is ++ readable (such as the stack or heap) it is also executable. ++ ++ There is a well known exploit technique that makes use of this ++ fact and a common programming mistake where an attacker can ++ introduce code of his choice somewhere in the attacked program's ++ memory (typically the stack or the heap) and then execute it. ++ ++ If the attacked program was running with different (typically ++ higher) privileges than that of the attacker, then he can elevate ++ his own privilege level (e.g. get a root shell, write to files for ++ which he does not have write access to, etc). ++ ++ Enabling this option will let you choose from various features ++ that prevent the injection and execution of 'foreign' code in ++ a program. ++ ++ This will also break programs that rely on the old behaviour and ++ expect that dynamically allocated memory via the malloc() family ++ of functions is executable (which it is not). Notable examples ++ are the XFree86 4.x server, the java runtime and wine. ++ ++config PAX_PAGEEXEC ++ bool "Paging based non-executable pages" ++ depends on PAX_NOEXEC && (!X86_32 || M586 || M586TSC || M586MMX || M686 || MPENTIUMII || MPENTIUMIII || MPENTIUMM || MCORE2 || MATOM || MPENTIUM4 || MPSC || MK7 || MK8 || MWINCHIPC6 || MWINCHIP2 || MWINCHIP3D || MVIAC3_2 || MVIAC7) ++ select S390_SWITCH_AMODE if S390 ++ select S390_EXEC_PROTECT if S390 ++ select ARCH_TRACK_EXEC_LIMIT if X86_32 ++ help ++ This implementation is based on the paging feature of the CPU. ++ On i386 without hardware non-executable bit support there is a ++ variable but usually low performance impact, however on Intel's ++ P4 core based CPUs it is very high so you should not enable this ++ for kernels meant to be used on such CPUs. ++ ++ On alpha, avr32, ia64, parisc, sparc, sparc64, x86_64 and i386 ++ with hardware non-executable bit support there is no performance ++ impact, on ppc the impact is negligible. ++ ++ Note that several architectures require various emulations due to ++ badly designed userland ABIs, this will cause a performance impact ++ but will disappear as soon as userland is fixed. For example, ppc ++ userland MUST have been built with secure-plt by a recent toolchain. ++ ++config PAX_SEGMEXEC ++ bool "Segmentation based non-executable pages" ++ depends on PAX_NOEXEC && X86_32 ++ help ++ This implementation is based on the segmentation feature of the ++ CPU and has a very small performance impact, however applications ++ will be limited to a 1.5 GB address space instead of the normal ++ 3 GB. ++ ++config PAX_EMUTRAMP ++ bool "Emulate trampolines" if (PAX_PAGEEXEC || PAX_SEGMEXEC) && (PARISC || X86) ++ default y if PARISC ++ help ++ There are some programs and libraries that for one reason or ++ another attempt to execute special small code snippets from ++ non-executable memory pages. Most notable examples are the ++ signal handler return code generated by the kernel itself and ++ the GCC trampolines. ++ ++ If you enabled CONFIG_PAX_PAGEEXEC or CONFIG_PAX_SEGMEXEC then ++ such programs will no longer work under your kernel. ++ ++ As a remedy you can say Y here and use the 'chpax' or 'paxctl' ++ utilities to enable trampoline emulation for the affected programs ++ yet still have the protection provided by the non-executable pages. ++ ++ On parisc you MUST enable this option and EMUSIGRT as well, otherwise ++ your system will not even boot. ++ ++ Alternatively you can say N here and use the 'chpax' or 'paxctl' ++ utilities to disable CONFIG_PAX_PAGEEXEC and CONFIG_PAX_SEGMEXEC ++ for the affected files. ++ ++ NOTE: enabling this feature *may* open up a loophole in the ++ protection provided by non-executable pages that an attacker ++ could abuse. Therefore the best solution is to not have any ++ files on your system that would require this option. This can ++ be achieved by not using libc5 (which relies on the kernel ++ signal handler return code) and not using or rewriting programs ++ that make use of the nested function implementation of GCC. ++ Skilled users can just fix GCC itself so that it implements ++ nested function calls in a way that does not interfere with PaX. ++ ++config PAX_EMUSIGRT ++ bool "Automatically emulate sigreturn trampolines" ++ depends on PAX_EMUTRAMP && PARISC ++ default y ++ help ++ Enabling this option will have the kernel automatically detect ++ and emulate signal return trampolines executing on the stack ++ that would otherwise lead to task termination. ++ ++ This solution is intended as a temporary one for users with ++ legacy versions of libc (libc5, glibc 2.0, uClibc before 0.9.17, ++ Modula-3 runtime, etc) or executables linked to such, basically ++ everything that does not specify its own SA_RESTORER function in ++ normal executable memory like glibc 2.1+ does. ++ ++ On parisc you MUST enable this option, otherwise your system will ++ not even boot. ++ ++ NOTE: this feature cannot be disabled on a per executable basis ++ and since it *does* open up a loophole in the protection provided ++ by non-executable pages, the best solution is to not have any ++ files on your system that would require this option. ++ ++config PAX_MPROTECT ++ bool "Restrict mprotect()" ++ depends on (PAX_PAGEEXEC || PAX_SEGMEXEC) ++ help ++ Enabling this option will prevent programs from ++ - changing the executable status of memory pages that were ++ not originally created as executable, ++ - making read-only executable pages writable again, ++ - creating executable pages from anonymous memory, ++ - making read-only-after-relocations (RELRO) data pages writable again. ++ ++ You should say Y here to complete the protection provided by ++ the enforcement of non-executable pages. ++ ++ NOTE: you can use the 'chpax' or 'paxctl' utilities to control ++ this feature on a per file basis. ++ ++config PAX_MPROTECT_COMPAT ++ bool "Use legacy/compat protection demoting (read help)" ++ depends on PAX_MPROTECT ++ default n ++ help ++ The current implementation of PAX_MPROTECT denies RWX allocations/mprotects ++ by sending the proper error code to the application. For some broken ++ userland, this can cause problems with Python or other applications. The ++ current implementation however allows for applications like clamav to ++ detect if JIT compilation/execution is allowed and to fall back gracefully ++ to an interpreter-based mode if it does not. While we encourage everyone ++ to use the current implementation as-is and push upstream to fix broken ++ userland (note that the RWX logging option can assist with this), in some ++ environments this may not be possible. Having to disable MPROTECT ++ completely on certain binaries reduces the security benefit of PaX, ++ so this option is provided for those environments to revert to the old ++ behavior. ++ ++config PAX_ELFRELOCS ++ bool "Allow ELF text relocations (read help)" ++ depends on PAX_MPROTECT ++ default n ++ help ++ Non-executable pages and mprotect() restrictions are effective ++ in preventing the introduction of new executable code into an ++ attacked task's address space. There remain only two venues ++ for this kind of attack: if the attacker can execute already ++ existing code in the attacked task then he can either have it ++ create and mmap() a file containing his code or have it mmap() ++ an already existing ELF library that does not have position ++ independent code in it and use mprotect() on it to make it ++ writable and copy his code there. While protecting against ++ the former approach is beyond PaX, the latter can be prevented ++ by having only PIC ELF libraries on one's system (which do not ++ need to relocate their code). If you are sure this is your case, ++ as is the case with all modern Linux distributions, then leave ++ this option disabled. You should say 'n' here. ++ ++config PAX_ETEXECRELOCS ++ bool "Allow ELF ET_EXEC text relocations" ++ depends on PAX_MPROTECT && (ALPHA || IA64 || PARISC) ++ select PAX_ELFRELOCS ++ default y ++ help ++ On some architectures there are incorrectly created applications ++ that require text relocations and would not work without enabling ++ this option. If you are an alpha, ia64 or parisc user, you should ++ enable this option and disable it once you have made sure that ++ none of your applications need it. ++ ++config PAX_EMUPLT ++ bool "Automatically emulate ELF PLT" ++ depends on PAX_MPROTECT && (ALPHA || PARISC || SPARC) ++ default y ++ help ++ Enabling this option will have the kernel automatically detect ++ and emulate the Procedure Linkage Table entries in ELF files. ++ On some architectures such entries are in writable memory, and ++ become non-executable leading to task termination. Therefore ++ it is mandatory that you enable this option on alpha, parisc, ++ sparc and sparc64, otherwise your system would not even boot. ++ ++ NOTE: this feature *does* open up a loophole in the protection ++ provided by the non-executable pages, therefore the proper ++ solution is to modify the toolchain to produce a PLT that does ++ not need to be writable. ++ ++config PAX_DLRESOLVE ++ bool 'Emulate old glibc resolver stub' ++ depends on PAX_EMUPLT && SPARC ++ default n ++ help ++ This option is needed if userland has an old glibc (before 2.4) ++ that puts a 'save' instruction into the runtime generated resolver ++ stub that needs special emulation. ++ ++config PAX_KERNEXEC ++ bool "Enforce non-executable kernel pages" ++ depends on PAX_NOEXEC && (PPC || X86) && (!X86_32 || X86_WP_WORKS_OK) && !XEN ++ select PAX_PER_CPU_PGD if X86_64 || (X86_32 && X86_PAE) ++ help ++ This is the kernel land equivalent of PAGEEXEC and MPROTECT, ++ that is, enabling this option will make it harder to inject ++ and execute 'foreign' code in kernel memory itself. ++ ++config PAX_KERNEXEC_MODULE_TEXT ++ int "Minimum amount of memory reserved for module code" ++ default "4" ++ depends on PAX_KERNEXEC && X86_32 && MODULES ++ help ++ Due to implementation details the kernel must reserve a fixed ++ amount of memory for module code at compile time that cannot be ++ changed at runtime. Here you can specify the minimum amount ++ in MB that will be reserved. Due to the same implementation ++ details this size will always be rounded up to the next 2/4 MB ++ boundary (depends on PAE) so the actually available memory for ++ module code will usually be more than this minimum. ++ ++ The default 4 MB should be enough for most users but if you have ++ an excessive number of modules (e.g., most distribution configs ++ compile many drivers as modules) or use huge modules such as ++ nvidia's kernel driver, you will need to adjust this amount. ++ A good rule of thumb is to look at your currently loaded kernel ++ modules and add up their sizes. ++ ++endmenu ++ ++menu "Address Space Layout Randomization" ++ depends on PAX ++ ++config PAX_ASLR ++ bool "Address Space Layout Randomization" ++ depends on PAX_EI_PAX || PAX_PT_PAX_FLAGS || PAX_HAVE_ACL_FLAGS || PAX_HOOK_ACL_FLAGS ++ help ++ Many if not most exploit techniques rely on the knowledge of ++ certain addresses in the attacked program. The following options ++ will allow the kernel to apply a certain amount of randomization ++ to specific parts of the program thereby forcing an attacker to ++ guess them in most cases. Any failed guess will most likely crash ++ the attacked program which allows the kernel to detect such attempts ++ and react on them. PaX itself provides no reaction mechanisms, ++ instead it is strongly encouraged that you make use of Nergal's ++ segvguard (ftp://ftp.pl.openwall.com/misc/segvguard/) or grsecurity's ++ (http://www.grsecurity.net/) built-in crash detection features or ++ develop one yourself. ++ ++ By saying Y here you can choose to randomize the following areas: ++ - top of the task's kernel stack ++ - top of the task's userland stack ++ - base address for mmap() requests that do not specify one ++ (this includes all libraries) ++ - base address of the main executable ++ ++ It is strongly recommended to say Y here as address space layout ++ randomization has negligible impact on performance yet it provides ++ a very effective protection. ++ ++ NOTE: you can use the 'chpax' or 'paxctl' utilities to control ++ this feature on a per file basis. ++ ++config PAX_RANDKSTACK ++ bool "Randomize kernel stack base" ++ depends on PAX_ASLR && X86_TSC && X86 ++ help ++ By saying Y here the kernel will randomize every task's kernel ++ stack on every system call. This will not only force an attacker ++ to guess it but also prevent him from making use of possible ++ leaked information about it. ++ ++ Since the kernel stack is a rather scarce resource, randomization ++ may cause unexpected stack overflows, therefore you should very ++ carefully test your system. Note that once enabled in the kernel ++ configuration, this feature cannot be disabled on a per file basis. ++ ++config PAX_RANDUSTACK ++ bool "Randomize user stack base" ++ depends on PAX_ASLR ++ help ++ By saying Y here the kernel will randomize every task's userland ++ stack. The randomization is done in two steps where the second ++ one may apply a big amount of shift to the top of the stack and ++ cause problems for programs that want to use lots of memory (more ++ than 2.5 GB if SEGMEXEC is not active, or 1.25 GB when it is). ++ For this reason the second step can be controlled by 'chpax' or ++ 'paxctl' on a per file basis. ++ ++config PAX_RANDMMAP ++ bool "Randomize mmap() base" ++ depends on PAX_ASLR ++ help ++ By saying Y here the kernel will use a randomized base address for ++ mmap() requests that do not specify one themselves. As a result ++ all dynamically loaded libraries will appear at random addresses ++ and therefore be harder to exploit by a technique where an attacker ++ attempts to execute library code for his purposes (e.g. spawn a ++ shell from an exploited program that is running at an elevated ++ privilege level). ++ ++ Furthermore, if a program is relinked as a dynamic ELF file, its ++ base address will be randomized as well, completing the full ++ randomization of the address space layout. Attacking such programs ++ becomes a guess game. You can find an example of doing this at ++ http://pax.grsecurity.net/et_dyn.tar.gz and practical samples at ++ http://www.grsecurity.net/grsec-gcc-specs.tar.gz . ++ ++ NOTE: you can use the 'chpax' or 'paxctl' utilities to control this ++ feature on a per file basis. ++ ++endmenu ++ ++menu "Miscellaneous hardening features" ++ ++config PAX_MEMORY_SANITIZE ++ bool "Sanitize all freed memory" ++ help ++ By saying Y here the kernel will erase memory pages as soon as they ++ are freed. This in turn reduces the lifetime of data stored in the ++ pages, making it less likely that sensitive information such as ++ passwords, cryptographic secrets, etc stay in memory for too long. ++ ++ This is especially useful for programs whose runtime is short, long ++ lived processes and the kernel itself benefit from this as long as ++ they operate on whole memory pages and ensure timely freeing of pages ++ that may hold sensitive information. ++ ++ The tradeoff is performance impact, on a single CPU system kernel ++ compilation sees a 3% slowdown, other systems and workloads may vary ++ and you are advised to test this feature on your expected workload ++ before deploying it. ++ ++ Note that this feature does not protect data stored in live pages, ++ e.g., process memory swapped to disk may stay there for a long time. ++ ++config PAX_MEMORY_UDEREF ++ bool "Prevent invalid userland pointer dereference" ++ depends on X86 && !UML_X86 && !XEN ++ select PAX_PER_CPU_PGD if X86_64 ++ help ++ By saying Y here the kernel will be prevented from dereferencing ++ userland pointers in contexts where the kernel expects only kernel ++ pointers. This is both a useful runtime debugging feature and a ++ security measure that prevents exploiting a class of kernel bugs. ++ ++ The tradeoff is that some virtualization solutions may experience ++ a huge slowdown and therefore you should not enable this feature ++ for kernels meant to run in such environments. Whether a given VM ++ solution is affected or not is best determined by simply trying it ++ out, the performance impact will be obvious right on boot as this ++ mechanism engages from very early on. A good rule of thumb is that ++ VMs running on CPUs without hardware virtualization support (i.e., ++ the majority of IA-32 CPUs) will likely experience the slowdown. ++ ++config PAX_REFCOUNT ++ bool "Prevent various kernel object reference counter overflows" ++ depends on GRKERNSEC && (X86 || SPARC64) ++ help ++ By saying Y here the kernel will detect and prevent overflowing ++ various (but not all) kinds of object reference counters. Such ++ overflows can normally occur due to bugs only and are often, if ++ not always, exploitable. ++ ++ The tradeoff is that data structures protected by an overflowed ++ refcount will never be freed and therefore will leak memory. Note ++ that this leak also happens even without this protection but in ++ that case the overflow can eventually trigger the freeing of the ++ data structure while it is still being used elsewhere, resulting ++ in the exploitable situation that this feature prevents. ++ ++ Since this has a negligible performance impact, you should enable ++ this feature. ++ ++config PAX_USERCOPY ++ bool "Harden heap object copies between kernel and userland" ++ depends on X86 || PPC || SPARC ++ depends on GRKERNSEC && (SLAB || SLUB) ++ help ++ By saying Y here the kernel will enforce the size of heap objects ++ when they are copied in either direction between the kernel and ++ userland, even if only a part of the heap object is copied. ++ ++ Specifically, this checking prevents information leaking from the ++ kernel heap during kernel to userland copies (if the kernel heap ++ object is otherwise fully initialized) and prevents kernel heap ++ overflows during userland to kernel copies. ++ ++ Note that the current implementation provides the strictest bounds ++ checks for the SLUB allocator. ++ ++ Enabling this option also enables per-slab cache protection against ++ data in a given cache being copied into/out of via userland ++ accessors. Though the whitelist of regions will be reduced over ++ time, it notably protects important data structures like task structs. ++ ++ If frame pointers are enabled on x86, this option will also restrict ++ copies into and out of the kernel stack to local variables within a ++ single frame. ++ ++ Since this has a negligible performance impact, you should enable ++ this feature. ++ ++endmenu ++ ++endmenu ++ + config KEYS + bool "Enable access key retention support" + help +@@ -167,7 +688,7 @@ config INTEL_TXT + config LSM_MMAP_MIN_ADDR + int "Low address space for LSM to protect from user allocation" + depends on SECURITY && SECURITY_SELINUX +- default 65536 ++ default 32768 + help + This is the portion of low virtual memory which should be protected + from userspace allocation. Keeping a user from writing to low pages +diff -urNp linux-2.6.38.3/security/min_addr.c linux-2.6.38.3/security/min_addr.c +--- linux-2.6.38.3/security/min_addr.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/security/min_addr.c 2011-04-17 15:57:33.000000000 -0400 +@@ -14,6 +14,7 @@ unsigned long dac_mmap_min_addr = CONFIG + */ + static void update_mmap_min_addr(void) + { ++#ifndef SPARC + #ifdef CONFIG_LSM_MMAP_MIN_ADDR + if (dac_mmap_min_addr > CONFIG_LSM_MMAP_MIN_ADDR) + mmap_min_addr = dac_mmap_min_addr; +@@ -22,6 +23,7 @@ static void update_mmap_min_addr(void) + #else + mmap_min_addr = dac_mmap_min_addr; + #endif ++#endif + } + + /* +diff -urNp linux-2.6.38.3/security/security.c linux-2.6.38.3/security/security.c +--- linux-2.6.38.3/security/security.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/security/security.c 2011-04-17 15:57:33.000000000 -0400 +@@ -25,8 +25,8 @@ static __initdata char chosen_lsm[SECURI + /* things that live in capability.c */ + extern void __init security_fixup_ops(struct security_operations *ops); + +-static struct security_operations *security_ops; +-static struct security_operations default_security_ops = { ++static struct security_operations *security_ops __read_only; ++static struct security_operations default_security_ops __read_only = { + .name = "default", + }; + +@@ -67,7 +67,9 @@ int __init security_init(void) + + void reset_security_ops(void) + { ++ pax_open_kernel(); + security_ops = &default_security_ops; ++ pax_close_kernel(); + } + + /* Save user chosen LSM */ +diff -urNp linux-2.6.38.3/security/selinux/hooks.c linux-2.6.38.3/security/selinux/hooks.c +--- linux-2.6.38.3/security/selinux/hooks.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/security/selinux/hooks.c 2011-04-17 15:57:33.000000000 -0400 +@@ -90,7 +90,6 @@ + #define NUM_SEL_MNT_OPTS 5 + + extern int selinux_nlmsg_lookup(u16 sclass, u16 nlmsg_type, u32 *perm); +-extern struct security_operations *security_ops; + + /* SECMARK reference count */ + atomic_t selinux_secmark_refcount = ATOMIC_INIT(0); +@@ -5395,7 +5394,7 @@ static int selinux_key_getsecurity(struc + + #endif + +-static struct security_operations selinux_ops = { ++static struct security_operations selinux_ops __read_only = { + .name = "selinux", + + .ptrace_access_check = selinux_ptrace_access_check, +diff -urNp linux-2.6.38.3/security/smack/smack_lsm.c linux-2.6.38.3/security/smack/smack_lsm.c +--- linux-2.6.38.3/security/smack/smack_lsm.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/security/smack/smack_lsm.c 2011-04-17 15:57:33.000000000 -0400 +@@ -3179,7 +3179,7 @@ static int smack_inode_getsecctx(struct + return 0; + } + +-struct security_operations smack_ops = { ++struct security_operations smack_ops __read_only = { + .name = "smack", + + .ptrace_access_check = smack_ptrace_access_check, +diff -urNp linux-2.6.38.3/security/tomoyo/tomoyo.c linux-2.6.38.3/security/tomoyo/tomoyo.c +--- linux-2.6.38.3/security/tomoyo/tomoyo.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/security/tomoyo/tomoyo.c 2011-04-17 15:57:33.000000000 -0400 +@@ -240,7 +240,7 @@ static int tomoyo_sb_pivotroot(struct pa + * tomoyo_security_ops is a "struct security_operations" which is used for + * registering TOMOYO. + */ +-static struct security_operations tomoyo_security_ops = { ++static struct security_operations tomoyo_security_ops __read_only = { + .name = "tomoyo", + .cred_alloc_blank = tomoyo_cred_alloc_blank, + .cred_prepare = tomoyo_cred_prepare, +diff -urNp linux-2.6.38.3/sound/aoa/codecs/onyx.c linux-2.6.38.3/sound/aoa/codecs/onyx.c +--- linux-2.6.38.3/sound/aoa/codecs/onyx.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/sound/aoa/codecs/onyx.c 2011-04-17 15:57:33.000000000 -0400 +@@ -54,7 +54,7 @@ struct onyx { + spdif_locked:1, + analog_locked:1, + original_mute:2; +- int open_count; ++ local_t open_count; + struct codec_info *codec_info; + + /* mutex serializes concurrent access to the device +@@ -753,7 +753,7 @@ static int onyx_open(struct codec_info_i + struct onyx *onyx = cii->codec_data; + + mutex_lock(&onyx->mutex); +- onyx->open_count++; ++ local_inc(&onyx->open_count); + mutex_unlock(&onyx->mutex); + + return 0; +@@ -765,8 +765,7 @@ static int onyx_close(struct codec_info_ + struct onyx *onyx = cii->codec_data; + + mutex_lock(&onyx->mutex); +- onyx->open_count--; +- if (!onyx->open_count) ++ if (local_dec_and_test(&onyx->open_count)) + onyx->spdif_locked = onyx->analog_locked = 0; + mutex_unlock(&onyx->mutex); + +diff -urNp linux-2.6.38.3/sound/aoa/codecs/onyx.h linux-2.6.38.3/sound/aoa/codecs/onyx.h +--- linux-2.6.38.3/sound/aoa/codecs/onyx.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/sound/aoa/codecs/onyx.h 2011-04-17 15:57:33.000000000 -0400 +@@ -11,6 +11,7 @@ + #include <linux/i2c.h> + #include <asm/pmac_low_i2c.h> + #include <asm/prom.h> ++#include <asm/local.h> + + /* PCM3052 register definitions */ + +diff -urNp linux-2.6.38.3/sound/core/oss/pcm_oss.c linux-2.6.38.3/sound/core/oss/pcm_oss.c +--- linux-2.6.38.3/sound/core/oss/pcm_oss.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/sound/core/oss/pcm_oss.c 2011-04-17 15:57:33.000000000 -0400 +@@ -2971,8 +2971,8 @@ static void snd_pcm_oss_proc_done(struct + } + } + #else /* !CONFIG_SND_VERBOSE_PROCFS */ +-#define snd_pcm_oss_proc_init(pcm) +-#define snd_pcm_oss_proc_done(pcm) ++#define snd_pcm_oss_proc_init(pcm) do {} while (0) ++#define snd_pcm_oss_proc_done(pcm) do {} while (0) + #endif /* CONFIG_SND_VERBOSE_PROCFS */ + + /* +diff -urNp linux-2.6.38.3/sound/core/seq/seq_lock.h linux-2.6.38.3/sound/core/seq/seq_lock.h +--- linux-2.6.38.3/sound/core/seq/seq_lock.h 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/sound/core/seq/seq_lock.h 2011-04-17 15:57:33.000000000 -0400 +@@ -23,10 +23,10 @@ void snd_use_lock_sync_helper(snd_use_lo + #else /* SMP || CONFIG_SND_DEBUG */ + + typedef spinlock_t snd_use_lock_t; /* dummy */ +-#define snd_use_lock_init(lockp) /**/ +-#define snd_use_lock_use(lockp) /**/ +-#define snd_use_lock_free(lockp) /**/ +-#define snd_use_lock_sync(lockp) /**/ ++#define snd_use_lock_init(lockp) do {} while (0) ++#define snd_use_lock_use(lockp) do {} while (0) ++#define snd_use_lock_free(lockp) do {} while (0) ++#define snd_use_lock_sync(lockp) do {} while (0) + + #endif /* SMP || CONFIG_SND_DEBUG */ + +diff -urNp linux-2.6.38.3/sound/drivers/mts64.c linux-2.6.38.3/sound/drivers/mts64.c +--- linux-2.6.38.3/sound/drivers/mts64.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/sound/drivers/mts64.c 2011-04-17 15:57:33.000000000 -0400 +@@ -28,6 +28,7 @@ + #include <sound/initval.h> + #include <sound/rawmidi.h> + #include <sound/control.h> ++#include <asm/local.h> + + #define CARD_NAME "Miditerminal 4140" + #define DRIVER_NAME "MTS64" +@@ -66,7 +67,7 @@ struct mts64 { + struct pardevice *pardev; + int pardev_claimed; + +- int open_count; ++ local_t open_count; + int current_midi_output_port; + int current_midi_input_port; + u8 mode[MTS64_NUM_INPUT_PORTS]; +@@ -696,7 +697,7 @@ static int snd_mts64_rawmidi_open(struct + { + struct mts64 *mts = substream->rmidi->private_data; + +- if (mts->open_count == 0) { ++ if (local_read(&mts->open_count) == 0) { + /* We don't need a spinlock here, because this is just called + if the device has not been opened before. + So there aren't any IRQs from the device */ +@@ -704,7 +705,7 @@ static int snd_mts64_rawmidi_open(struct + + msleep(50); + } +- ++(mts->open_count); ++ local_inc(&mts->open_count); + + return 0; + } +@@ -714,8 +715,7 @@ static int snd_mts64_rawmidi_close(struc + struct mts64 *mts = substream->rmidi->private_data; + unsigned long flags; + +- --(mts->open_count); +- if (mts->open_count == 0) { ++ if (local_dec_return(&mts->open_count) == 0) { + /* We need the spinlock_irqsave here because we can still + have IRQs at this point */ + spin_lock_irqsave(&mts->lock, flags); +@@ -724,8 +724,8 @@ static int snd_mts64_rawmidi_close(struc + + msleep(500); + +- } else if (mts->open_count < 0) +- mts->open_count = 0; ++ } else if (local_read(&mts->open_count) < 0) ++ local_set(&mts->open_count, 0); + + return 0; + } +diff -urNp linux-2.6.38.3/sound/drivers/portman2x4.c linux-2.6.38.3/sound/drivers/portman2x4.c +--- linux-2.6.38.3/sound/drivers/portman2x4.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/sound/drivers/portman2x4.c 2011-04-17 15:57:33.000000000 -0400 +@@ -47,6 +47,7 @@ + #include <sound/initval.h> + #include <sound/rawmidi.h> + #include <sound/control.h> ++#include <asm/local.h> + + #define CARD_NAME "Portman 2x4" + #define DRIVER_NAME "portman" +@@ -84,7 +85,7 @@ struct portman { + struct pardevice *pardev; + int pardev_claimed; + +- int open_count; ++ local_t open_count; + int mode[PORTMAN_NUM_INPUT_PORTS]; + struct snd_rawmidi_substream *midi_input[PORTMAN_NUM_INPUT_PORTS]; + }; +diff -urNp linux-2.6.38.3/sound/oss/sb_audio.c linux-2.6.38.3/sound/oss/sb_audio.c +--- linux-2.6.38.3/sound/oss/sb_audio.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/sound/oss/sb_audio.c 2011-04-17 15:57:33.000000000 -0400 +@@ -901,7 +901,7 @@ sb16_copy_from_user(int dev, + buf16 = (signed short *)(localbuf + localoffs); + while (c) + { +- locallen = (c >= LBUFCOPYSIZE ? LBUFCOPYSIZE : c); ++ locallen = ((unsigned)c >= LBUFCOPYSIZE ? LBUFCOPYSIZE : c); + if (copy_from_user(lbuf8, + userbuf+useroffs + p, + locallen)) +diff -urNp linux-2.6.38.3/sound/oss/swarm_cs4297a.c linux-2.6.38.3/sound/oss/swarm_cs4297a.c +--- linux-2.6.38.3/sound/oss/swarm_cs4297a.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/sound/oss/swarm_cs4297a.c 2011-04-17 15:57:33.000000000 -0400 +@@ -2606,7 +2606,6 @@ static int __init cs4297a_init(void) + { + struct cs4297a_state *s; + u32 pwr, id; +- mm_segment_t fs; + int rval; + #ifndef CONFIG_BCM_CS4297A_CSWARM + u64 cfg; +@@ -2696,22 +2695,23 @@ static int __init cs4297a_init(void) + if (!rval) { + char *sb1250_duart_present; + ++#if 0 ++ mm_segment_t fs; + fs = get_fs(); + set_fs(KERNEL_DS); +-#if 0 + val = SOUND_MASK_LINE; + mixer_ioctl(s, SOUND_MIXER_WRITE_RECSRC, (unsigned long) &val); + for (i = 0; i < ARRAY_SIZE(initvol); i++) { + val = initvol[i].vol; + mixer_ioctl(s, initvol[i].mixch, (unsigned long) &val); + } ++ set_fs(fs); + // cs4297a_write_ac97(s, 0x18, 0x0808); + #else + // cs4297a_write_ac97(s, 0x5e, 0x180); + cs4297a_write_ac97(s, 0x02, 0x0808); + cs4297a_write_ac97(s, 0x18, 0x0808); + #endif +- set_fs(fs); + + list_add(&s->list, &cs4297a_devs); + +diff -urNp linux-2.6.38.3/sound/pci/ac97/ac97_patch.c linux-2.6.38.3/sound/pci/ac97/ac97_patch.c +--- linux-2.6.38.3/sound/pci/ac97/ac97_patch.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/sound/pci/ac97/ac97_patch.c 2011-04-17 15:57:33.000000000 -0400 +@@ -1486,7 +1486,7 @@ static const struct snd_ac97_res_table a + { AC97_VIDEO, 0x9f1f }, + { AC97_AUX, 0x9f1f }, + { AC97_PCM, 0x9f1f }, +- { } /* terminator */ ++ { 0, 0 } /* terminator */ + }; + + static int patch_ad1819(struct snd_ac97 * ac97) +@@ -3864,7 +3864,7 @@ static struct snd_ac97_res_table lm4550_ + { AC97_AUX, 0x1f1f }, + { AC97_PCM, 0x1f1f }, + { AC97_REC_GAIN, 0x0f0f }, +- { } /* terminator */ ++ { 0, 0 } /* terminator */ + }; + + static int patch_lm4550(struct snd_ac97 *ac97) +diff -urNp linux-2.6.38.3/sound/pci/ens1370.c linux-2.6.38.3/sound/pci/ens1370.c +--- linux-2.6.38.3/sound/pci/ens1370.c 2011-04-17 16:53:01.000000000 -0400 ++++ linux-2.6.38.3/sound/pci/ens1370.c 2011-04-17 16:54:08.000000000 -0400 +@@ -453,7 +453,7 @@ static DEFINE_PCI_DEVICE_TABLE(snd_audio + { PCI_VDEVICE(ENSONIQ, 0x5880), 0, }, /* ES1373 - CT5880 */ + { PCI_VDEVICE(ECTIVA, 0x8938), 0, }, /* Ectiva EV1938 */ + #endif +- { 0, } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(pci, snd_audiopci_ids); +diff -urNp linux-2.6.38.3/sound/pci/hda/patch_hdmi.c linux-2.6.38.3/sound/pci/hda/patch_hdmi.c +--- linux-2.6.38.3/sound/pci/hda/patch_hdmi.c 2011-04-17 16:53:01.000000000 -0400 ++++ linux-2.6.38.3/sound/pci/hda/patch_hdmi.c 2011-04-17 16:54:08.000000000 -0400 +@@ -733,10 +733,10 @@ static void hdmi_non_intrinsic_event(str + cp_ready); + + /* TODO */ +- if (cp_state) +- ; +- if (cp_ready) +- ; ++ if (cp_state) { ++ } ++ if (cp_ready) { ++ } + } + + +diff -urNp linux-2.6.38.3/sound/pci/intel8x0.c linux-2.6.38.3/sound/pci/intel8x0.c +--- linux-2.6.38.3/sound/pci/intel8x0.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/sound/pci/intel8x0.c 2011-04-17 15:57:33.000000000 -0400 +@@ -444,7 +444,7 @@ static DEFINE_PCI_DEVICE_TABLE(snd_intel + { PCI_VDEVICE(AMD, 0x746d), DEVICE_INTEL }, /* AMD8111 */ + { PCI_VDEVICE(AMD, 0x7445), DEVICE_INTEL }, /* AMD768 */ + { PCI_VDEVICE(AL, 0x5455), DEVICE_ALI }, /* Ali5455 */ +- { 0, } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(pci, snd_intel8x0_ids); +@@ -2141,7 +2141,7 @@ static struct ac97_quirk ac97_quirks[] _ + .type = AC97_TUNE_HP_ONLY + }, + #endif +- { } /* terminator */ ++ { 0, 0, 0, 0, NULL, 0 } /* terminator */ + }; + + static int __devinit snd_intel8x0_mixer(struct intel8x0 *chip, int ac97_clock, +diff -urNp linux-2.6.38.3/sound/pci/intel8x0m.c linux-2.6.38.3/sound/pci/intel8x0m.c +--- linux-2.6.38.3/sound/pci/intel8x0m.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/sound/pci/intel8x0m.c 2011-04-17 15:57:33.000000000 -0400 +@@ -239,7 +239,7 @@ static DEFINE_PCI_DEVICE_TABLE(snd_intel + { PCI_VDEVICE(AMD, 0x746d), DEVICE_INTEL }, /* AMD8111 */ + { PCI_VDEVICE(AL, 0x5455), DEVICE_ALI }, /* Ali5455 */ + #endif +- { 0, } ++ { 0, 0, 0, 0, 0, 0, 0 } + }; + + MODULE_DEVICE_TABLE(pci, snd_intel8x0m_ids); +@@ -1264,7 +1264,7 @@ static struct shortname_table { + { 0x5455, "ALi M5455" }, + { 0x746d, "AMD AMD8111" }, + #endif +- { 0 }, ++ { 0, NULL }, + }; + + static int __devinit snd_intel8x0m_probe(struct pci_dev *pci, +diff -urNp linux-2.6.38.3/usr/gen_init_cpio.c linux-2.6.38.3/usr/gen_init_cpio.c +--- linux-2.6.38.3/usr/gen_init_cpio.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/usr/gen_init_cpio.c 2011-04-17 15:57:33.000000000 -0400 +@@ -305,7 +305,7 @@ static int cpio_mkfile(const char *name, + int retval; + int rc = -1; + int namesize; +- int i; ++ unsigned int i; + + mode |= S_IFREG; + +@@ -394,9 +394,10 @@ static char *cpio_replace_env(char *new_ + *env_var = *expanded = '\0'; + strncat(env_var, start + 2, end - start - 2); + strncat(expanded, new_location, start - new_location); +- strncat(expanded, getenv(env_var), PATH_MAX); +- strncat(expanded, end + 1, PATH_MAX); ++ strncat(expanded, getenv(env_var), PATH_MAX - strlen(expanded)); ++ strncat(expanded, end + 1, PATH_MAX - strlen(expanded)); + strncpy(new_location, expanded, PATH_MAX); ++ new_location[PATH_MAX] = 0; + } else + break; + } +diff -urNp linux-2.6.38.3/virt/kvm/kvm_main.c linux-2.6.38.3/virt/kvm/kvm_main.c +--- linux-2.6.38.3/virt/kvm/kvm_main.c 2011-03-14 21:20:32.000000000 -0400 ++++ linux-2.6.38.3/virt/kvm/kvm_main.c 2011-04-17 15:57:33.000000000 -0400 +@@ -1521,7 +1521,7 @@ static int kvm_vcpu_release(struct inode + return 0; + } + +-static struct file_operations kvm_vcpu_fops = { ++static struct file_operations kvm_vcpu_fops = { /* cannot be const */ + .release = kvm_vcpu_release, + .unlocked_ioctl = kvm_vcpu_ioctl, + .compat_ioctl = kvm_vcpu_ioctl, +@@ -1990,7 +1990,7 @@ static int kvm_vm_mmap(struct file *file + return 0; + } + +-static struct file_operations kvm_vm_fops = { ++static struct file_operations kvm_vm_fops = { /* cannot be const */ + .release = kvm_vm_release, + .unlocked_ioctl = kvm_vm_ioctl, + #ifdef CONFIG_COMPAT +@@ -2088,7 +2088,7 @@ out: + return r; + } + +-static struct file_operations kvm_chardev_ops = { ++static struct file_operations kvm_chardev_ops = { /* cannot be const */ + .unlocked_ioctl = kvm_dev_ioctl, + .compat_ioctl = kvm_dev_ioctl, + .llseek = noop_llseek, +@@ -2098,6 +2098,9 @@ static struct miscdevice kvm_dev = { + KVM_MINOR, + "kvm", + &kvm_chardev_ops, ++ {NULL, NULL}, ++ NULL, ++ NULL + }; + + static void hardware_enable_nolock(void *junk) +@@ -2443,7 +2446,7 @@ static void kvm_sched_out(struct preempt + kvm_arch_vcpu_put(vcpu); + } + +-int kvm_init(void *opaque, unsigned vcpu_size, unsigned vcpu_align, ++int kvm_init(const void *opaque, unsigned vcpu_size, unsigned vcpu_align, + struct module *module) + { + int r; |