From: Aurelien Jarno <aurelien@aurel32.net>
To: libc-alpha@sourceware.org
Cc: Noah Goldstein <goldstein.w.n@gmail.com>,
"H . J . Lu" <hjl.tools@gmail.com>,
Sunil K Pandey <skpgkp2@gmail.com>,
Aurelien Jarno <aurelien@aurel32.net>
Subject: [PATCH v3 0/8] x86: Fix AVX2 string functions requiring BMI1, BMI2 or LZCNT (BZ #29611)
Date: Mon, 3 Oct 2022 21:59:36 +0200 [thread overview]
Message-ID: <20221003195944.3274548-1-aurelien@aurel32.net> (raw)
Some early Intel Haswell CPU have AVX2 instructions, but do not have
BMI1 and BMI2 instructions. Some AVX2 string functions only check for
AVX2, but use BMI1, BMI2 or LZCNT instructions. This patchset tries to
fix that.
While most fixes only change ifunc-impl-list.c, and thus only concerns
the testsuite, the changes to str*cmp and wcs(n)cmp are a real issue
affecting early Intel Haswell CPU, reported to affect Debian Sid and
Fedora Rawhide.
On the other hand, the check for LZCNT in memrchr is purely for
correctness, I am not aware of a CPU implementing AVX2 without LZCNT.
This has been tested by remplacing all BMI1 and BMI2 instructions in the
source code by the "ud2" instruction and disabling the BMI1, BMI2
feature detection, and running the testsuite.
Resolves: BZ #29611
Changes v2 -> v3:
- Change the way patches are split. No code change.
Change v1 -> v2:
- Better scan for BMI2 instructions (shlx and shrx) and BMI1
instructions (blsmsk) instructions following the feedback from Noah
Goldstein
Aurelien Jarno (8):
x86: include BMI1 and BMI2 in x86-64-v3 level
x86-64: Require BMI2 for AVX2 str(n)casecmp implementations
x86-64: Require BMI2 for AVX2 strcmp implementation
x86-64: Require BMI2 for AVX2 strncmp implementation
x86-64: Require BMI2 for AVX2 wcs(n)cmp implementations
x86-64: Require BMI2 for AVX2 (raw|w)memchr implementations
x86-64: Require BMI2 and LZCNT for AVX2 memrchr implementation
x86-64: Require BMI1/BMI2 for AVX2 strrchr and wcsrchr implementations
sysdeps/x86/get-isa-level.h | 2 +
sysdeps/x86/isa-level.h | 2 +
sysdeps/x86_64/multiarch/ifunc-avx2.h | 2 +
sysdeps/x86_64/multiarch/ifunc-impl-list.c | 86 ++++++++++++++++-----
sysdeps/x86_64/multiarch/ifunc-strcasecmp.h | 1 +
sysdeps/x86_64/multiarch/strcmp.c | 4 +-
sysdeps/x86_64/multiarch/strncmp.c | 4 +-
7 files changed, 76 insertions(+), 25 deletions(-)
--
2.35.1
next reply other threads:[~2022-10-03 19:59 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-10-03 19:59 Aurelien Jarno [this message]
2022-10-03 19:59 ` [PATCH v3 1/8] x86: include BMI1 and BMI2 in x86-64-v3 level Aurelien Jarno
2022-10-03 21:11 ` Noah Goldstein
2022-10-03 19:59 ` [PATCH v3 2/8] x86-64: Require BMI2 for AVX2 str(n)casecmp implementations Aurelien Jarno
2022-10-03 21:11 ` Noah Goldstein
2022-10-03 19:59 ` [PATCH v3 3/8] x86-64: Require BMI2 for AVX2 strcmp implementation Aurelien Jarno
2022-10-03 21:12 ` Noah Goldstein
2022-10-03 19:59 ` [PATCH v3 4/8] x86-64: Require BMI2 for AVX2 strncmp implementation Aurelien Jarno
2022-10-03 21:11 ` Noah Goldstein
2022-10-03 19:59 ` [PATCH v3 5/8] x86-64: Require BMI2 for AVX2 wcs(n)cmp implementations Aurelien Jarno
2022-10-03 21:11 ` Noah Goldstein
2022-10-03 19:59 ` [PATCH v3 6/8] x86-64: Require BMI2 for AVX2 (raw|w)memchr implementations Aurelien Jarno
2022-10-03 21:12 ` Noah Goldstein
2022-10-03 19:59 ` [PATCH v3 7/8] x86-64: Require BMI2 and LZCNT for AVX2 memrchr implementation Aurelien Jarno
2022-10-03 21:12 ` Noah Goldstein
2022-10-03 19:59 ` [PATCH v3 8/8] x86-64: Require BMI1/BMI2 for AVX2 strrchr and wcsrchr implementations Aurelien Jarno
2022-10-03 21:12 ` Noah Goldstein
2022-10-03 20:47 ` [PATCH v3 0/8] x86: Fix AVX2 string functions requiring BMI1, BMI2 or LZCNT (BZ #29611) Sunil Pandey
2022-10-03 21:11 ` Noah Goldstein
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20221003195944.3274548-1-aurelien@aurel32.net \
--to=aurelien@aurel32.net \
--cc=goldstein.w.n@gmail.com \
--cc=hjl.tools@gmail.com \
--cc=libc-alpha@sourceware.org \
--cc=skpgkp2@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).