From: Noah Goldstein via Libc-alpha <libc-alpha@sourceware.org>
To: "H.J. Lu" <hjl.tools@gmail.com>
Cc: GNU C Library <libc-alpha@sourceware.org>
Subject: Re: [PATCH v2 1/5] x86_64: Add support for bcmp using sse2, sse4_1, avx2, and evex
Date: Tue, 14 Sep 2021 14:23:50 -0500 [thread overview]
Message-ID: <CAFUsyfLRWFtJ4F1=JMJ828PtvjcGvn94stgaDFuoXfv09yUo9w@mail.gmail.com> (raw)
In-Reply-To: <CAMe9rOqrL+KCxezBWSzd6aNBOSmfyEO=QCOC0o5HbinzweCK5g@mail.gmail.com>
On Tue, Sep 14, 2021 at 9:40 AM H.J. Lu <hjl.tools@gmail.com> wrote:
> On Mon, Sep 13, 2021 at 11:30 PM Noah Goldstein <goldstein.w.n@gmail.com>
> wrote:
> >
> > No bug. This commit adds support for an optimized bcmp implementation.
> > Support is for sse2, sse4_1, avx2, and evex.
> >
> > All string tests passing and build succeeding.
>
> memcmp can be a little slower than bcmp. But bcmp isn't a standard C
> function.
> All new codes should use memcmp. Can you improve memcmp instead?
> Thanks.
>
There are some small improvements to memcmp I could imagine.
Use vptestm{b|d} instead of vpcmp for zero tests.
Aligning to 64 bytes so that target alignments/placements can be better
optimized.
But for the most part the biggest optimization is just reducing all the
work to compute
1/-1 for the result.
I think, however, that since GLIBC supports bcmp it makes sense to offer
the best
version we can. Especially since compilers (Clang at least) will use it
when possible
to optimize memcmp usage.
I don't fully understand the concern with adding it. AFAICT if GLIBC
decides it no
longer wants to support bcmp we can remove it, but essentially the same work
would need to be done regardless. Can you elaborate on why?
>
>
> --
> H.J.
>
next prev parent reply other threads:[~2021-09-14 19:24 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-09-13 23:05 [PATCH 1/5] x86_64: Add support for bcmp using sse2, sse4_1, avx2, and evex Noah Goldstein via Libc-alpha
2021-09-13 23:05 ` [PATCH 2/5] x86_64: Add sse2 optimized bcmp implementation in memcmp.S Noah Goldstein via Libc-alpha
2021-09-13 23:05 ` [PATCH 3/5] x86_64: Add sse4_1 optimized bcmp implementation in memcmp-sse4.S Noah Goldstein via Libc-alpha
2021-09-13 23:05 ` [PATCH 4/5] x86_64: Add avx2 optimized bcmp implementation in bcmp-avx2.S Noah Goldstein via Libc-alpha
2021-09-13 23:05 ` [PATCH 5/5] x86_64: Add evex optimized bcmp implementation in bcmp-evex.S Noah Goldstein via Libc-alpha
2021-09-14 1:18 ` Carlos O'Donell via Libc-alpha
2021-09-14 2:05 ` Noah Goldstein via Libc-alpha
2021-09-14 2:35 ` Carlos O'Donell via Libc-alpha
2021-09-14 2:55 ` DJ Delorie via Libc-alpha
2021-09-14 3:24 ` Noah Goldstein via Libc-alpha
2021-09-14 3:40 ` Noah Goldstein via Libc-alpha
2021-09-14 4:21 ` DJ Delorie via Libc-alpha
2021-09-14 5:29 ` Noah Goldstein via Libc-alpha
2021-09-14 5:42 ` DJ Delorie via Libc-alpha
2021-09-14 5:55 ` Noah Goldstein via Libc-alpha
2021-09-13 23:22 ` [PATCH 1/5] x86_64: Add support for bcmp using sse2, sse4_1, avx2, and evex Noah Goldstein via Libc-alpha
2021-09-14 6:30 ` [PATCH v2 " Noah Goldstein via Libc-alpha
2021-09-14 6:30 ` [PATCH v2 2/5] x86_64: Add sse2 optimized bcmp implementation in memcmp.S Noah Goldstein via Libc-alpha
2021-09-14 6:30 ` [PATCH v2 3/5] x86_64: Add sse4_1 optimized bcmp implementation in memcmp-sse4.S Noah Goldstein via Libc-alpha
2021-09-14 6:30 ` [PATCH v2 4/5] x86_64: Add avx2 optimized bcmp implementation in bcmp-avx2.S Noah Goldstein via Libc-alpha
2021-09-14 6:30 ` [PATCH v2 5/5] x86_64: Add evex optimized bcmp implementation in bcmp-evex.S Noah Goldstein via Libc-alpha
2021-09-14 14:40 ` [PATCH v2 1/5] x86_64: Add support for bcmp using sse2, sse4_1, avx2, and evex H.J. Lu via Libc-alpha
2021-09-14 19:23 ` Noah Goldstein via Libc-alpha [this message]
2021-09-14 20:30 ` Florian Weimer via Libc-alpha
2021-09-15 0:00 ` [PATCH " Joseph Myers
2021-09-15 13:37 ` Zack Weinberg via Libc-alpha
2021-09-15 14:01 ` Re: [PATCH 1/5] x86_64: Add support for bcmp using sse2, sse 4_1, " Florian Weimer via Libc-alpha
2021-09-15 18:06 ` Noah Goldstein via Libc-alpha
2021-09-15 18:30 ` Joseph Myers
2021-09-27 1:35 ` Noah Goldstein via Libc-alpha
2021-09-27 7:29 ` Florian Weimer via Libc-alpha
2021-09-27 16:49 ` Noah Goldstein via Libc-alpha
2021-09-27 16:54 ` Florian Weimer via Libc-alpha
2021-09-27 17:54 ` Noah Goldstein via Libc-alpha
2021-09-27 17:56 ` Florian Weimer via Libc-alpha
2021-09-27 18:05 ` Noah Goldstein via Libc-alpha
2021-09-27 18:10 ` Florian Weimer via Libc-alpha
2021-09-27 18:15 ` Noah Goldstein via Libc-alpha
2021-09-27 18:22 ` Florian Weimer via Libc-alpha
2021-09-27 18:34 ` Noah Goldstein via Libc-alpha
2021-09-27 18:56 ` Florian Weimer via Libc-alpha
2021-09-27 19:20 ` Noah Goldstein via Libc-alpha
2021-09-27 19:34 ` Florian Weimer via Libc-alpha
2021-09-27 19:43 ` Noah Goldstein via Libc-alpha
2021-09-27 19:59 ` Florian Weimer via Libc-alpha
2021-09-27 20:22 ` Noah Goldstein via Libc-alpha
2021-09-27 20:24 ` Florian Weimer via Libc-alpha
2021-09-27 20:38 ` Noah Goldstein via Libc-alpha
2021-09-28 0:07 ` Noah Goldstein via Libc-alpha
2021-09-27 17:42 ` Joseph Myers
2021-09-27 17:48 ` Noah Goldstein via Libc-alpha
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
List information: https://www.gnu.org/software/libc/involved.html
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='CAFUsyfLRWFtJ4F1=JMJ828PtvjcGvn94stgaDFuoXfv09yUo9w@mail.gmail.com' \
--to=libc-alpha@sourceware.org \
--cc=goldstein.w.n@gmail.com \
--cc=hjl.tools@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).