public inbox for gcc-bugs@sourceware.org help / color / mirror / Atom feed
From: "rguenth at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org> To: gcc-bugs@gcc.gnu.org Subject: [Bug tree-optimization/99395] s116 benchmark of TSVC is vectorized by clang and not by gcc Date: Tue, 18 Oct 2022 10:28:04 +0000 [thread overview] Message-ID: <bug-99395-4-ar7GfXuh5A@http.gcc.gnu.org/bugzilla/> (raw) In-Reply-To: <bug-99395-4@http.gcc.gnu.org/bugzilla/> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99395 --- Comment #4 from Richard Biener <rguenth at gcc dot gnu.org> --- So in the BB SLP attempt from loop vectorization (or in the BB SLP pass with -fno-predictive-commoning) we get confused during DR group building because of a duplicate access and fixup splitting the candidates at odd points. For the reduced testcase we see <bb 3> [local count: 1063004409]: # i_16 = PHI <_5(5), 0(2)> # ivtmp_18 = PHI <ivtmp_15(5), 511(2)> _1 = i_16 + 1; _2 = a[_1]; _3 = a[i_16]; _4 = _2 * _3; a[i_16] = _4; _5 = i_16 + 2; _6 = a[_5]; _7 = a[_1]; _8 = _6 * _7; a[_1] = _8; ivtmp_15 = ivtmp_18 - 1; if (ivtmp_15 != 0) goto <bb 5>; [99.00%] else goto <bb 4>; [1.00%] so a[_1] is loaded twice because CSE doesn't figure that a[i_16] cannot alias it. That causes us to split the load group.
next prev parent reply other threads:[~2022-10-18 10:28 UTC|newest] Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top 2021-03-04 23:01 [Bug middle-end/99395] New: " hubicka at gcc dot gnu.org 2021-03-04 23:24 ` [Bug middle-end/99395] " hubicka at gcc dot gnu.org 2021-03-05 8:27 ` [Bug tree-optimization/99395] " rguenth at gcc dot gnu.org 2021-03-05 15:42 ` hubicka at gcc dot gnu.org 2022-10-18 10:28 ` rguenth at gcc dot gnu.org [this message] 2022-10-18 10:37 ` rguenth at gcc dot gnu.org 2023-10-07 9:10 ` juzhe.zhong at rivai dot ai 2023-10-09 6:30 ` rguenth at gcc dot gnu.org 2024-01-30 11:37 ` juzhe.zhong at rivai dot ai 2024-01-30 12:26 ` rguenth at gcc dot gnu.org 2024-01-31 1:32 ` juzhe.zhong at rivai dot ai 2024-01-31 3:50 ` juzhe.zhong at rivai dot ai 2024-01-31 7:00 ` juzhe.zhong at rivai dot ai 2024-01-31 7:59 ` rguenth at gcc dot gnu.org 2024-01-31 8:09 ` juzhe.zhong at rivai dot ai 2024-01-31 8:18 ` rguenther at suse dot de 2024-01-31 8:27 ` juzhe.zhong at rivai dot ai 2024-01-31 9:08 ` rguenther at suse dot de 2024-01-31 9:19 ` juzhe.zhong at rivai dot ai 2024-01-31 9:56 ` rguenther at suse dot de 2024-04-07 21:33 ` pinskia 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-99395-4-ar7GfXuh5A@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: linkBe 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).