openbsd/src/lib/libcrypto/sha, branch OPENBSD_7_9_BASEA mirror of https://github.com/libressl/openbsd.git
https://git.lua4.win/openbsd/atom?h=OPENBSD_7_9_BASE2026-03-28T13:11:28+00:00Include crypto_assembly.h instead of manually ensuring _CET_ENDBR exists.2026-03-28T13:11:28+00:00jsing2026-03-28T13:11:28+00:00urn:sha1:b691bbfcac89ff4cadab20821c931ce1d672b9c9
ok kenjiro@ tb@
Make SHA aarch64 assembly build with gcc.2026-01-25T08:22:17+00:00jsing2026-01-25T08:22:17+00:00urn:sha1:c6ef8fc8d3285a55c002f52df2bb2df42b7734c0
gcc is extremely fussy about register naming and insists on q and s naming
for the ARM CE SHA instructions, even though they're referring to the same
register (while LLVM just figures it out). Work around this by mapping
registers to their required variant at usage and defining a handful of
mappings between v registers and alternate names/views.
This is still somewhat ugly, but seems to be one of the cleaner options
that will allow portable to enable SHA assembly on platforms that use gcc.
ok kenjiro@ tb@
Tidy instruction separators in SHA assembly.2026-01-24T14:20:52+00:00jsing2026-01-24T14:20:52+00:00urn:sha1:1d4a03f3d650a577581aedee8cc799a3ad6668d8
Remove unnecessary separators and add a few to macros that call other
macros (instead of expecting them to exist).
Use .section before .rodata to appease gas.2026-01-17T06:31:45+00:00jsing2026-01-17T06:31:45+00:00urn:sha1:650cf9e91c9b18210aaa53983bfb7725a344b7bb
gas dislikes bare .rodata - add .section before .rodata to make it happier
(LLVM does not care and is happy with either). For consistency, do the same
with .text.
Use local label prefix for loop labels.2026-01-17T06:23:42+00:00jsing2026-01-17T06:23:42+00:00urn:sha1:f8fcc4d17be5065d3823964ee7fd62dbfb090e96Provide accelerated SHA-1 for aarch64.2025-06-28T12:51:08+00:00jsing2025-06-28T12:51:08+00:00urn:sha1:8d0be76398371402b3426575c7326b6e08b0e023
Provide an assembly implementation of SHA-1 for aarch64 using the ARM
Cryptographic Extension (CE). This results in around a 2x speed up for
larger block sizes.
ok tb@
Make OPENSSL_IA32_SSE2 the default for i386 and remove the flag.2025-06-09T14:28:34+00:00jsing2025-06-09T14:28:34+00:00urn:sha1:951cd4503ff3c4cc93c1a36cf06138b1ddd739d7
The OPENSSL_IA32_SSE2 flag controls whether a number of the perlasm
scripts generate additional implementations that use SSE2 functionality.
In all cases except ghash, the code checks OPENSSL_ia32cap_P for SSE2
support, before trying to run SSE2 code. For ghash it generates a CLMUL
based implementation in addition to different MMX version (one MMX
version hides behind OPENSSL_IA32_SSE2, the other does not), however this
does not appear to actually use SSE2. We also disable AES-NI on i386 if
OPENSSL_IA32_SSE2.
On OpenBSD, we've always defined OPENSSL_IA32_SSE2 so this is effectively
a no-op. The only change is that we now check MMX rather than SSE2 for the
ghash MMX implementation.
ok bcook@ beck@
Remove GNU assembler version check.2025-06-09T13:58:28+00:00jsing2025-06-09T13:58:28+00:00urn:sha1:d8ae322f91ae730f31b750787e0180588dab36dc
GNU assembler version 2.19 was released in 2014, so it does not seem
unreasonable to expect that we have an assembler that supports AVX.
Furthermore, the current check fails on LLVM.
ok bcook@ beck@
Use 'ctx' for sha3_ctx variables, rather than the less readable 'c'.2025-04-18T07:36:11+00:00jsing2025-04-18T07:36:11+00:00urn:sha1:40f74488b8ad99303861a7ad88abf6108df59fb0
ok tb@
Pull casts from void * to uint8_t * up to variables, rather than inline.2025-04-18T07:27:42+00:00jsing2025-04-18T07:27:42+00:00urn:sha1:b9a0523d97444dc48412549cc52bb57d6dcd7d7e
ok tb@