- Apr 07, 2017
-
-
Rich Salz authored
(cherry picked from commit dfc63ccebd8262648640fc6e35fffa132102b967) Reviewed-by:
Richard Levitte <levitte@openssl.org> (Merged from https://github.com/openssl/openssl/pull/3117)
-
Rich Salz authored
(cherry picked from commit 7ea8ceab4b21d93eb272a89bff73958010d5c2c4) Reviewed-by:
-
Richard Levitte authored
When configured no-engine, we still refered to rand_engine_lock. Rework the lock init code to avoid that. Reviewed-by:
Matt Caswell <matt@openssl.org> (Merged from https://github.com/openssl/openssl/pull/3145) (cherry picked from commit 2f881d2d)
-
Richard Levitte authored
This is especially harmful since OPENSSL_cleanup() has already called the RAND cleanup function Reviewed-by:
-
Richard Levitte authored
Reviewed-by:
-
- Apr 06, 2017
-
-
Qin Long authored
Under UEFI build environment, we may encounter the OSSL_SSIZE macro re-definition error in e_os2.h if any module call OpenSSL API directly by including "openssl/xxxx.h" (caused by the predefined _WIN32/_WIN64 macro, which should have been un-defined under OPENSSL_SYS_UEFI). Though it's not one recommended usage, this patch could still eliminate the possible build issue by refining the OSSL_SSIZE definition under OPENSSL_SYS_UEFI. Reviewed-by:
Rich Salz <rsalz@openssl.org> Reviewed-by:
Andy Polyakov <appro@openssl.org> (Merged from https://github.com/openssl/openssl/pull/3121) (cherry picked from commit 005f6766)
-
Richard Levitte authored
If no default method was yet given, RAND_get_rand_method() will set it up. Doing so just to clean it away seems pretty silly, so instead, use the default_RAND_meth variable directly. This also clears a possible race condition where this will try to init things, such as ERR or ENGINE when in the middle of a OPENSSL_cleanup. Fixes #3128 Reviewed-by:
-
- Apr 04, 2017
-
-
Andy Polyakov authored
Fixes GH#3116. Reviewed-by:
-
Richard Levitte authored
Bug uncovered by test [extended tests] Reviewed-by:
-
Matt Caswell authored
Credit to OSS-Fuzz for finding this. Reviewed-by:
-
Richard Levitte authored
This increases portability of SSL_SESSION files between architectures where the size of |long| may vary. Before this, SSL_SESSION files produced on a 64-bit long architecture may break on a 32-bit long architecture. Reviewed-by:
-
Richard Levitte authored
Reviewed-by:
-
Richard Levitte authored
Also Z varieties. Reviewed-by:
-
- Apr 03, 2017
-
-
Matt Caswell authored
dhparams correctly handles X9.42 params in PEM format. However it failed to correctly processes them when reading/writing DER format. Fixes #3102 Reviewed-by:
-
Matt Caswell authored
DHparams has d2i_DHparams_fp, d2i_DHxparams_bio etc, but the equivalent macros for DHxparams were omitted. Reviewed-by:
-
- Apr 02, 2017
-
-
Andy Polyakov authored
It seems to be problematic to probe processor capabilities with SIGILL on MacOS X. The problem should be limited to cases when application code is debugged, but crashes were reported even during normal execution... Reviewed-by:
Kurt Roeckx <kurt@openssl.org> Reviewed-by:
-
- Apr 01, 2017
-
-
Andy Polyakov authored
Configure started with 'require 5.10.0', but if executed by older perl it failed with "might be runaway multi-line // string" instead of naturally expected "Perl v5.10.0 required--this is only v5.x.y". Reviewed-by:
-
Andy Polyakov authored
Reviewed-by:
-
- Mar 31, 2017
-
-
Sebastian Andrzej Siewior authored
The man pages for ciphers, s_client and s_server mention the tls1_3 option but it is not implemented in the 1.1.0 branch. Thus remove it to avoid confusion. Signed-off-by:
Sebastian Andrzej Siewior <sebastian@breakpoint.cc> Reviewed-by:
-
Jon Spillett authored
Reviewed-by:
-
- Mar 30, 2017
-
-
FdaSilvaYY authored
Backport of 69687aa8 (Merged from #3069) Reviewed-by:
-
- Mar 29, 2017
-
-
Andy Polyakov authored
Reviewed-by:
-
Andy Polyakov authored
Reviewed-by:
-
Andy Polyakov authored
Reviewed-by:
-
Andy Polyakov authored
Reviewed-by:
-
Steven Collison authored
These ciphers don't appear to be documented anywhere. Given the performance[1] benefits I think it makes sense to expose them. [1] https://software.intel.com/sites/default/files/open-ssl-performance-paper.pdf Reviewed-by:
-
Steven Collison authored
These were added to the help in ad775e04 but not the pods. Reviewed-by:
-
Jon Spillett authored
Reviewed-by:
-
Rich Salz authored
Reviewed-by:
-
- Mar 28, 2017
-
-
FdaSilvaYY authored
Reviewed-by:
-
FdaSilvaYY authored
Reviewed-by:
-
FdaSilvaYY authored
Reviewed-by:
-
FdaSilvaYY authored
Reviewed-by:
-
Jon Spillett authored
[skip ci] Reviewed-by:
-
Rich Salz authored
Reviewed-by:
-
Matt Caswell authored
Reviewed-by:
-
Emilia Kasper authored
Remove unnecessary include of apps.h. Tests shouldn't take a dependency on apps. In this case, there is no dependency, the include is unnecessary. Reviewed-by:
-
Hannes Magnusson authored
a47bc283 accidentally adds another define for X509_STORE_set_flags It is already defined 5lines prior CLA: trivial Reviewed-by:
Tim Hudson <tjh@openssl.org> Reviewed-by:
-
- Mar 27, 2017
-
-
Richard Levitte authored
Fixes #3033 Reviewed-by:
-
- Mar 26, 2017
-
-
Andy Polyakov authored
Even though Apple refers to Procedure Call Standard for ARM Architecture (AAPCS), they apparently adhere to custom version that doesn't follow stack alignment constraints in the said standard. [Why or why? If it's vendor lock-in thing, then it would be like worst spot ever.] And since bsaes-armv7 relied on standard alignment, it became problematic to execute the code on iOS. Reviewed-by:
-
