public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "pinskia at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/107546] [10/11/12/13 Regression] simd, redundant pcmpeqb and pxor
Date: Mon, 07 Nov 2022 06:58:23 +0000	[thread overview]
Message-ID: <bug-107546-4-GZ09FuIoni@http.gcc.gnu.org/bugzilla/> (raw)
In-Reply-To: <bug-107546-4@http.gcc.gnu.org/bugzilla/>

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107546

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|simd, redundant pcmpeqb and |[10/11/12/13 Regression]
                   |pxor                        |simd, redundant pcmpeqb and
                   |                            |pxor
   Target Milestone|---                         |10.5

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
For GNU C++ vectors produced
GCC 4.8 until GCC 11 produced:
        movdqa  xmm0, XMMWORD PTR [rdi]
        pcmpeqd xmm1, xmm1
        pcmpgtb xmm0, XMMWORD PTR .LC0[rip]
        pandn   xmm0, xmm1
        ret

GCC 11+ produces:
        movdqa  xmm0, XMMWORD PTR [rdi]
        pxor    xmm1, xmm1
        pcmpgtb xmm0, XMMWORD PTR .LC0[rip]
        pcmpeqb xmm0, xmm1
        ret

But the intrinics produced the expected thing until GCC 9.

in GCC 8 the intrinsics produces:
  _3 = VEC_COND_EXPR <_4 < { 48, 48, 48, 48, 48, 48, 48, 48, 48, 48, 48, 48,
48, 48, 48, 48 }, { -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
-1 }, { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 }>;

even. Notice the < vs <= there.
I suspect the <= expansion part of the x86_64 backend needs to be fixed up to
produce better code.

So this is a regression for the intrinsics and marking it as such.

  parent reply	other threads:[~2022-11-07  6:58 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-11-07  6:39 [Bug rtl-optimization/107546] New: " i.nixman at autistici dot org
2022-11-07  6:51 ` [Bug target/107546] " pinskia at gcc dot gnu.org
2022-11-07  6:58 ` pinskia at gcc dot gnu.org [this message]
2022-11-07  7:07 ` [Bug target/107546] [10/11/12/13 Regression] " rguenth at gcc dot gnu.org
2022-11-07  7:07 ` crazylht at gmail dot com
2022-11-07  7:14 ` crazylht at gmail dot com
2022-11-07  7:16 ` glisse at gcc dot gnu.org
2022-11-07  7:20 ` pinskia at gcc dot gnu.org
2022-11-07  8:27 ` crazylht at gmail dot com
2022-11-07 10:22 ` jakub at gcc dot gnu.org
2022-11-07 12:02 ` jakub at gcc dot gnu.org
2022-11-07 12:05 ` i.nixman at autistici dot org
2022-11-07 13:47 ` i.nixman at autistici dot org
2022-11-08 11:24 ` cvs-commit at gcc dot gnu.org
2022-11-08 11:59 ` [Bug target/107546] [10/11/12 " jakub at gcc dot gnu.org
2022-11-10 10:48 ` cvs-commit at gcc dot gnu.org
2023-07-07 10:44 ` [Bug target/107546] [11/12 " rguenth at gcc dot gnu.org

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=bug-107546-4-GZ09FuIoni@http.gcc.gnu.org/bugzilla/ \
    --to=gcc-bugzilla@gcc.gnu.org \
    --cc=gcc-bugs@gcc.gnu.org \
    /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).