| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
| |
from http://cvs.pld-linux.org/cgi-bin/viewvc.cgi/cvs/packages/openssl/openssl-c_rehash.sh
fixes #2266
(cherry picked from commit 3fc0e1d57b15a8c6c652f215450c810a3c43601b)
Conflicts:
main/openssl/APKBUILD
|
|
|
|
|
| |
Apply patch from openssl rt.
(cherry picked from commit 095daa82b97cc5dd0de237e470e155476a3bc256)
|
|
|
|
|
|
|
|
|
| |
Add EVP_MD_FLAG_PKEY_METHOD_SIGNATURE to padlock_sha1_md to fix
DSA/SHA1 verification in certain cases. Seems that NID_sha1 instead
of NID_dss is used sometimes incorrectly, and this seems to be the
workaround regular SHA1 code does too.
Suggested-by: Daniel Mansfield <daniel.mansfield@unsw.edu.au>
|
| |
|
|
|
|
| |
http://marc.info/?t=136018837600003&r=1&w=2
|
| |
|
|
|
|
|
|
| |
(CVE-2013-0169,CVE-2012-2686,CVE-2013-0166)
fixes #1591
|
|
|
|
| |
when building on eglibc we need versioned symbols
|
| |
|
|
|
|
|
| |
* fixed hmac oneshot flag to work as expected
* renamed the patch series, and rebased against 1.0.1c
|
|
|
|
|
|
| |
We want to handle bytes upto next block boundary, to work with
hardware from block boundary. The code incorrectly fed just the
amount of bytes in the block.
|
|
|
|
|
| |
fixes #1151
(cherry picked from commit 1831053bb87f432f0d45ccd9f7a368fc885a1d64)
|
| |
|
|
|
|
| |
fixes #1107
|
|
|
|
|
|
|
| |
Remove the unneeded 0002-apps-speed-fix; the speed utility can now
measure evp speeds with -evp flag.
Padlock autoloading patch is rebased.
|
| |
|
|
|
|
| |
fixes #935
|
|
|
|
|
|
|
|
|
|
|
| |
CVE-2011-4108
CVE-2011-4109
CVE-2011-4576
CVE-2011-4577
CVE-2011-4619
CVE-2012-0027
fixes #893
|
| |
|
|
|
|
| |
* contains security fix to CVE-2011-0014
|
| |
|
| |
|
|
|
|
|
|
|
|
| |
- Fix for security issue CVE-2010-4180
- Fix for CVE-2010-4252
- Fix mishandling of absent EC point format extension.
- Fix various platform compilation issues.
- Corrected fix for security issue CVE-2010-3864.
|
|
|
|
| |
so we avoid /usr/lib64 on x86_64
|
|
|
|
|
|
| |
* upgrade to 1.0.0b which has security fixes
* update patch which did not apply anymore
* delete patch merged upstream
|
| |
|
| |
|
|
|
|
| |
The new feature is support for VIA Nano Padlock in 64-bit mode.
|
|
|
|
|
|
|
|
|
|
|
|
| |
Add new version of padlock patches which enable:
- limited support of VIA C7 SHA acceleration
- full support for VIA Nano SHA acceleration
Openssl HMAC core is also patched to take full performance out of
padlock. Speed application is updated for measuring hmac(sha1).
Padlock was moved to be dynamic engine in openssl-1.0.0. So add some
code that losfd automatically that engine.
|
| |
|
|
|
|
| |
and introduce libssl1.0 and libcrypto1.0 packages
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
| |
* CVE-2009-1377
* CVE-2009-1378
* CVE-2009-1379
* CVE-2009-1387
* CVE-2009-2409
* CVE-2009-3555
|
| |
|
|
|
|
| |
gives about 2x performance improvement on sse enabled chips.
|
| |
|
|
|
|
|
|
|
| |
fix copying of oneshot sha1 context. also decrease the size
of "small buffer", to make context smaller (and context copy
faster in most cases); it should be still enough to catch
most hmac operations.
|
| |
|
|
|
|
|
|
|
|
| |
there was a bug sha context copying, that caused all hmac users
(and possibly others) to crash. also implemented a third
intermediate hashing mode with small buffer: it'll speed up
hashing in most cases considerably (packets / certificates to
be hashed are not usually too long).
|
|
added maintainer to several packages as well
|