summaryrefslogtreecommitdiffstats
path: root/libc/string/sh/sh4
Commit message (Collapse)AuthorAgeFilesLines
* sh4: Reorder code in memmove.cCarmelo Amoroso2009-10-101-32/+30
| | | | | | | Remove forward declaration for service routine. Reorder code and keep hidden_def right after the respective function. Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
* sh: Add new optimisation to the SH4 memcpyGiuseppe Cavallaro2009-09-271-2/+107
| | | | | | | | | | | | | | | | | | This optimization is based on prefetching and 64bit data transfer via FPU (only for the little endianess) Tests shows that: ---------------------------------------- Memory bandwidth | Gain | sh4-300 | sh4-200 ---------------------------------------- 512 bytes to 16KiB | ~20% | ~25% from 32KiB to 16MiB | ~190% | ~5% ---------------------------------------- Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com> Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
* sh: add assembly strcpy and strncpyGiuseppe Cavallaro2009-09-272-0/+71
| | | | | | | | | | | | | | | | | This patch adds the strcpy and strncpy assembly routines. Benchmarks showed the following gains: ~7% for strcpy ~30% for strncpy Note: uClibc string tests pass without any failures. These functions have been only tested on SH4, for this reason I've voluntarily added them within the sh4 sub-folder. If somebody would like to test them on other SH CPUs, these can be moved on sh common folder. Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com> Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
* sh: add a new memmove optimised for SH4Giuseppe Cavallaro2009-09-271-0/+119
| | | | | | | | | | | | | | | This patch adds the memmove fuction for SH4. By default, it used the generic implementation. This new code uses the memcpy for BWD copies and implements FWD copy when required (see comment within the code itself). The idea behind is to get advantage of using the optimised memcpy for SH4 and use the FPU for FWD copies (for big sizes) as well. LMBench bw_mem test showed a significant improvement on uClibc because bcopy invokes memmove, directly. Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com> Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
* sh: add assembly memset from Kernel and optimise it for SH4Giuseppe Cavallaro2009-09-271-0/+146
| | | | | | | | | | | This patch adds the SH memset assembly implementation currenlty included into the Kernel. It also adds, only for little endian mode, the 64bit data transfer via FPU (using single paired precision mode). Tests shows that on SH4-300 we gain ~100% for size greater than 1KiB. Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com> Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
* sh: Fix up optimized SH-4 memcpy on big endian.Giuseppe Cavallaro2009-07-141-12/+12
| | | | | | | | | | Signed-off-by: Hideo Saito <saito@densan.co.jp> Signed-off-by: Paul Mundt <lethal@linux-sh.org> Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com> See Linux Kernel commit: http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=e08b954c9a140f2062649faec72514eb505f18c3 Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
* Synch with trunk @ 24075.Carmelo Amoroso2008-11-171-54/+53
| | | | | Step 4 libc/string and asm implementation
* Added optmized memcpy implementation for SH4Carmelo Amoroso2007-11-211-0/+804
(backward memcpy algorithm) Modified libc/string/generic/Makefile.in to handle subtarget implementations. Fixed generic memmove code to handle backward memcpy by using a selectable config option __ARCH_HAS_BWD_MEMCPY__ This option is on for SH4 arch Signed-off-by: Carmelo Amoroso <carmelo.amoroso@st.com>
generated by cgit v1.2.3 (git 2.25.1) at 2025-08-09 17:44:41 +0000