Andy Polyakov
0b48a24ce9
s390x assembly pack: cache capability query results.
...
IBM argues that in certain scenarios capability query is really
expensive. At the same time it's asserted that query results can
be safely cached, because disabling CPACF is incompatible with
reboot-free operation.
Reviewed-by: Tim Hudson <tjh@openssl.org>
(cherry picked from commit 670ad0fbf6ebcf113e278d8174081a7e2d2fa44c)
2016-04-25 11:54:38 +02:00
Andy Polyakov
9676402c3a
PPC assembly pack: remove branch hints.
...
As it turns out branch hints grew as kind of a misconception. In
addition their interpretation by GNU assembler is affected by
assembler flags and can end up with opposite meaning on different
processors. As we have to loose quite a lot on misinterprerations,
especially on newer processors, we just omit them altogether.
Reviewed-by: Tim Hudson <tjh@openssl.org>
(cherry picked from commit 20b88bb160865b0c2998056fe841b1cbacd6d4c7)
2016-04-07 21:29:57 +02:00
Andy Polyakov
e780ed0336
x86_64 assembly pack: tune clang version detection even further.
...
RT#4171
Reviewed-by: Kurt Roeckx <kurt@openssl.org>
(cherry picked from commit b9749432346f69b29d82070041e71b237d718ce7)
2015-12-13 22:19:32 +01:00
Andy Polyakov
301a6dcd45
x86_64 assembly pack: tune clang version detection.
...
RT#4142
Reviewed-by: Richard Levitte <levitte@openssl.org>
(cherry picked from commit 76eba0d94bb418325be6409b272eac5e2bd4a0a9)
Resolved conflicts:
crypto/bn/asm/x86_64-mont.pl
crypto/bn/asm/x86_64-mont5.pl
Reviewed-by: Kurt Roeckx <kurt@openssl.org>
2015-11-30 13:37:39 +01:00
Andy Polyakov
bc2ab6ed2f
aes/asm/vpaes-ppc.pl: eliminate overhung stores in misaligned cases.
...
Reviewed-by: Richard Levitte <levitte@openssl.org>
(cherry picked from commit ce24d2ed232a095349886aecc8cb959d5dda8342)
2015-11-18 11:57:45 +01:00
Andy Polyakov
817ddb9fb0
aesni-sha256-x86_64.pl: fix crash on AMD Jaguar.
...
It was also found that stich performs suboptimally on AMD Jaguar, hence
execution is limited to XOP-capable and Intel processors.
Reviewed-by: Kurt Roeckx <kurt@openssl.org>
(cherry picked from commit a5fd24d19bbb586b1c6d235c2021e9bead22c9f5)
2015-11-16 13:07:33 +01:00
Alessandro Ghedini
ec3a7c9b37
Fix typos
...
Reviewed-by: Rich Salz <rsalz@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(cherry picked from commit d900a015b5a89ea2018e6122dd2738925e4d68ab)
2015-10-23 20:32:58 +02:00
Richard Levitte
9f0b86c68b
Conversion to UTF-8 where needed
...
This leaves behind files with names ending with '.iso-8859-1'. These
should be safe to remove. If something went wrong when re-encoding,
there will be some files with names ending with '.utf8' left behind.
Reviewed-by: Rich Salz <rsalz@openssl.org>
2015-07-14 01:17:45 +02:00
Andy Polyakov
d3bb25e30a
aes/asm/aesni-sha256-x86_64.pl: fix Windows compilation failure with old assembler.
...
Reviewed-by: Matt Caswell <matt@openssl.org>
(cherry picked from commit 82c4a079395223c0063100854a0981976db9848f)
2015-05-13 17:00:06 +02:00
Andy Polyakov
73824ba8fe
aes/asm/aesni-x86.pl: fix typo affecting Windows build.
...
Reviewed-by: Matt Caswell <matt@openssl.org>
(cherry picked from commit 7be6bc68c6baef87d4d730c2505a05810a5a1684)
2015-04-20 18:40:46 +02:00
Andy Polyakov
e95e22af50
aes/asm/aesni-x86[_64].pl update.
...
This addresses
- request for improvement for faster key setup in RT#3576;
- clearing registers and stack in RT#3554 (this is more of a gesture to
see if there will be some traction from compiler side);
- more commentary around input parameters handling and stack layout
(desired when RT#3553 was reviewed);
- minor size and single block performance optimization (was lying around);
Reviewed-by: Matt Caswell <matt@openssl.org>
(cherry picked from commit 23f6eec71dbd472044db7dc854599f1de14a1f48)
2015-04-20 15:44:36 +02:00
Andy Polyakov
5a27a20be3
aes/asm/aesv8-armx.pl: optimize for Cortex-A5x.
...
ARM has optimized Cortex-A5x pipeline to favour pairs of complementary
AES instructions. While modified code improves performance of post-r0p0
Cortex-A53 performance by >40% (for CBC decrypt and CTR), it hurts
original r0p0. We favour later revisions, because one can't prevent
future from coming. Improvement on post-r0p0 Cortex-A57 exceeds 50%,
while new code is not slower on r0p0, or Apple A7 for that matter.
[Update even SHA results for latest Cortex-A53.]
Reviewed-by: Richard Levitte <levitte@openssl.org>
(cherry picked from commit 94376cccb4ed5b376220bffe0739140ea9dad8c8)
2015-04-02 09:51:24 +02:00
Andy Polyakov
2fc2649173
Fix macosx-ppc build (and typos in unwind info).
...
Reviewed-by: Tim Hudson <tjh@openssl.org>
(cherry picked from commit c462a6817bb05a4c8dded1aa9fa3aa8fd7e176bf)
2015-01-22 12:15:34 +01:00
Andy Polyakov
2d63d0c84a
Fix irix-cc build.
...
Reviewed-by: Matt Caswell <matt@openssl.org>
(cherry picked from commit e464403d0bda2f1f74eb68582e4988e591c32433)
2015-01-07 18:41:17 +01:00
Andy Polyakov
f4868c9921
Remove inconsistency in ARM support.
...
This facilitates "universal" builds, ones that target multiple
architectures, e.g. ARMv5 through ARMv7. See commentary in
Configure for details.
Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(cherry picked from commit c1669e1c205dc8e695fb0c10a655f434e758b9f7)
2015-01-06 11:14:23 +01:00
Andy Polyakov
3e3cc471c2
aesni-x86_64.pl: make ECB subroutine Windows ABI compliant.
...
RT: 3553
Reviewed-by: Emilia Kasper <emilia@openssl.org>
(cherry picked from commit 69d5747f90136aa026a96204f26ab39549dfc69b)
2014-10-15 11:12:24 +02:00
Andy Polyakov
9dd6240201
x86[_64] assembly pack: add Silvermont performance data.
...
Reviewed-by: Rich Salz <rsalz@openssl.org>
(cherry picked from commit b59f92e75d334c9281082a02faa6c68afb614fd2)
2014-08-30 19:14:49 +02:00
Andy Polyakov
c991d8ae8b
Initial POWER8 support from development branch.
...
Reviewed-by: Kurt Roeckx <kurt@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
2014-07-20 14:36:49 +02:00
Andy Polyakov
371feee876
x86_64 assembly pack: improve masm support.
...
(cherry picked from commit 1b0fe79f3ee27ebd20510da3af9ec04c6ee0f800)
2014-07-09 22:46:13 +02:00
Andy Polyakov
377551b9c4
x86_64 assembly pack: refine clang detection.
...
(cherry picked from commit a356e488ad50ca9de3fc4955839441c5aff11ed3)
Resolved conflicts:
crypto/bn/asm/rsaz-avx2.pl
2014-06-28 17:26:03 +02:00
Andy Polyakov
52f856526c
x86_64 assembly pack: addendum to last clang commit.
...
(cherry picked from commit 7eb048828008f195fb6edceb8f767622694e7426)
2014-06-27 22:55:22 +02:00
Andy Polyakov
912f08dd5e
x86_64 assembly pack: allow clang to compile AVX code.
...
(cherry picked from commit ac171925ab527a55fbb27872ff69af94f7ec995b)
2014-06-27 22:55:07 +02:00
Andy Polyakov
1536bcfd56
aesv8-armx.pl: rigid input verification in key setup.
...
(cherry picked from commit 7b8c8c4d7973f17bbbee45d4718224c5853d9dc7)
2014-06-25 22:12:08 +02:00
Andy Polyakov
a073ceeff4
aesv8-armx.pl: inclrease interleave factor.
...
This is to compensate for higher aes* instruction latency on Cortex-A57.
(cherry picked from commit 015364baf3328b93dbed2613e59170715a2a11a6)
2014-06-24 08:10:37 +02:00
Andy Polyakov
d940b3b9c7
aesni-sha[1|256]-x86_64.pl: fix logical error and MacOS X build.
...
(cherry picked from commit 9024b84b7c9a36f75bffc5c77cc86bbe2911653e)
2014-06-16 10:12:56 +02:00
Andy Polyakov
8301245a5e
aesni-sha256-x86_64.pl: add missing rex in shaext.
...
PR: 3405
(cherry picked from commit 91a6bf80f8af5fcafececc443e7fbc7981af127b)
2014-06-14 16:04:04 +02:00
Andy Polyakov
56ba280ccd
Facilitate back-porting of AESNI and SHA modules.
...
Fix SEH and stack handling in Win64 build.
(cherry picked from commit 977f32e85241cba8be53e44dade32231e8a91718)
2014-06-12 21:51:35 +02:00
Andy Polyakov
70fddbe32a
Add support for Intel SHA extension.
...
(cherry picked from commit 619b94667cc7a097f6d1e2123c4f4c2c85afb8f7)
2014-06-11 10:30:31 +02:00
Andy Polyakov
9af4cb3d3b
Add AES module for ARMv8 Crypto Extension [from HEAD].
2014-06-11 00:06:27 +02:00
Andy Polyakov
56d973709c
aesni-mb-x86_64.pl: add Win64 SEH.
...
(cherry picked from commit e2eabed110b6d88fd15ac3063cab23dcd5090b44)
2014-06-10 23:09:04 +02:00
Andy Polyakov
3a97ebb16b
ARM assembly pack: get ARMv7 instruction endianness right.
...
Pointer out and suggested by: Ard Biesheuvel.
(cherry picked from commit 5dcf70a1c57c2019bfad640fe14fd4a73212860a)
2014-06-10 22:51:15 +02:00
Andy Polyakov
c90c694bc4
vpaes-ppc.pl: comply with ABI.
...
(cherry picked from commit b83d09f5528990364e0f7ed498eb8a9b56d95b58)
2014-05-23 20:16:21 +02:00
Andy Polyakov
ffdff9f12f
aes/asm/bsaes-x86_64.pl: Atom-specific optimization.
...
(cherry picked from commit 558ff0f0c1d87d74e70b0a70ddd67c0ff7f596ad)
2014-04-24 10:14:46 +02:00
Andy Polyakov
dacb698ada
vpaes-[x86_64|ppc].pl: fix typo, which for some reason triggers rkhunter.
...
(cherry picked from commit 6eebcf345933694e08aba400faf6f639fb4db196)
2014-04-06 12:53:17 +02:00
Andy Polyakov
aa1bb606f3
aes/asm/vpaes-ppc.pl: fix traceback info.
...
(cherry picked from commit e704741bf3ce02c7dbcd9b81c4df257b3d919f7e)
2014-02-25 20:13:41 +01:00
Andy Polyakov
2d4d9623da
aes/asm/aesni-x86[_64].pl: minor Atom-specific performance tweak.
...
(cherry picked from commit 214368ffee5736836e2dbb80a16a4fbd85f0eaf9)
2014-02-21 12:15:07 +01:00
Andy Polyakov
b347341c75
aes/asm/aesni-x86_64.pl: further optimization for Atom Silvermont.
...
Improve CBC decrypt and CTR by ~13/16%, which adds up to ~25/33%
improvement over "pre-Silvermont" version. [Add performance table to
aesni-x86.pl].
(cherry picked from commit 5599c7331b90d9d29c9914c2a95c16d91485415a)
2014-02-14 17:17:39 +01:00
Andy Polyakov
41c373fa3e
[aesni|sha*]-mb-x86_64.pl: add multi-block assembly modules [from master].
2014-02-05 14:33:44 +01:00
Andy Polyakov
50f1b47c7f
PPC assembly pack: jumbo update from master.
...
Add Vector Permutation AES and little-endian support.
2014-02-01 21:48:31 +01:00
Andy Polyakov
5572bc4e2f
crypto/aes/asm/aesni-x86[_64].pl: jumbo update from master.
2014-02-01 21:27:46 +01:00
Andy Polyakov
729d334106
crypto/sha/asm/sha1-x86_64.pl: jumbo update from master.
2014-02-01 21:24:55 +01:00
Andy Polyakov
acd9121085
aesni-sha1-x86_64.pl: harmonize [Atom-specific optimizations] with master branch.
2014-01-04 17:42:13 +01:00
Andy Polyakov
b76310ba74
ARM assembly pack: AES update from master (including bit-sliced module).
2013-12-09 23:44:45 +01:00
Andy Polyakov
e41a49c625
PPC assembly pack: make new .size directives profiler-friendly.
...
Suggested by: Anton Blanchard
(cherry picked from commit 76c15d790e07f6cc098be2d7b7f6ddc8acd11ca6)
2013-10-15 23:42:18 +02:00
Andy Polyakov
43ce9cdde9
PPC assembly pack: update from master branch.
...
Includes multiple updates: AES module to comply with more ABI
flavors, SHA512 for PPC32, .size directives.
2013-10-15 00:31:45 +02:00
Andy Polyakov
9ed6fba2b4
aes/asm/bsaes-x86_64.pl: update from master.
...
Performance improvement and Windows-specific bugfix (PR#3139).
2013-10-12 21:47:54 +02:00
Andy Polyakov
fccec89084
aesni-sha256-x86_64.pl: fix typo in Windows SEH.
...
(cherry picked from commit 42386fdb62176ed3af237dd3e802ba9b8c0d4def)
2013-06-30 23:14:04 +02:00
Andy Polyakov
6c7fa27404
PA-RISC assembler pack: switch to bve in 64-bit builds.
...
PR: 3074
(cherry picked from commit 02450ec69dda7815ba1e7bd74eb30f0ae1eb3042)
2013-06-30 23:13:23 +02:00
Andy Polyakov
f00962aaf3
aesni-sha1-x86_64.pl: update performance data.
...
(cherry picked from commit 3b848d3401de1892415e46561ae75f06ff65e823)
2013-06-10 22:52:06 +02:00
Andy Polyakov
3eccd2e53a
aesni-sha256-x86_64.pl: harmonize with latest sha512-x86_64.pl.
...
(cherry picked from commit 42b9a4177bf9d465ff9ac8d4e1526301d809a87e)
2013-06-10 22:51:55 +02:00