public inbox for gcc-bugs@sourceware.org help / color / mirror / Atom feed
From: "tnfchris at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org> To: gcc-bugs@gcc.gnu.org Subject: [Bug c/113134] Middle end early break vectorization: Fail to vectorize a simple early break code Date: Mon, 25 Dec 2023 12:35:46 +0000 [thread overview] Message-ID: <bug-113134-4-8VVSFKUQtD@http.gcc.gnu.org/bugzilla/> (raw) In-Reply-To: <bug-113134-4@http.gcc.gnu.org/bugzilla/> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113134 Tamar Christina <tnfchris at gcc dot gnu.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Last reconfirmed| |2023-12-25 Keywords| |missed-optimization --- Comment #1 from Tamar Christina <tnfchris at gcc dot gnu.org> --- Yeah, the patch wasn't intended to handle that case. As the patch says it requires the sizes of the memories being accessed to be constant. i.e. https://compiler-explorer.com/z/87d61h3x5 works and that already works today. my patch would allow this if we relax vect_analyze_early_break_dependences to not worry about memory accesses if the condition itself does not access memory. However I don't think this should get to my patch at all. this loop void add(int N, int *__restrict a, int *__restrict b, int *__restrict c) { for (int i = 0; i < N; i++) { c[i] = a[i] + b[i]; if (i > 1000) { break; } } } is essentially void add2(int N, int *__restrict a, int *__restrict b, int *__restrict c) { for (int i = 0; i < 1001; i++) { c[i] = a[i] + b[i]; } } and is what clang has rewritten it to. GCC does the same in ivcanon but then doesn't realize the additional exit is unreachable and should be removed. So this case needs to be handled before it reaches the vectorizer.
next prev parent reply other threads:[~2023-12-25 12:35 UTC|newest] Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top 2023-12-25 11:29 [Bug c/113134] New: " juzhe.zhong at rivai dot ai 2023-12-25 12:35 ` tnfchris at gcc dot gnu.org [this message] 2023-12-27 15:21 ` [Bug c/113134] gcc does not version loops with side-effect early breaks tnfchris at gcc dot gnu.org 2023-12-28 1:21 ` [Bug c/113134] gcc does not version loops with early break conditions that don't have side-effects juzhe.zhong at rivai dot ai 2023-12-28 3:48 ` tnfchris at gcc dot gnu.org 2023-12-28 3:55 ` juzhe.zhong at rivai dot ai 2023-12-28 4:02 ` tnfchris at gcc dot gnu.org 2023-12-28 4:05 ` tnfchris at gcc dot gnu.org 2023-12-28 4:23 ` juzhe.zhong at rivai dot ai 2023-12-28 4:30 ` tnfchris at gcc dot gnu.org 2023-12-28 4:35 ` juzhe.zhong at rivai dot ai 2023-12-28 4:45 ` tnfchris at gcc dot gnu.org 2023-12-28 4:46 ` juzhe.zhong at rivai dot ai 2023-12-28 4:49 ` tnfchris at gcc dot gnu.org 2023-12-28 4:51 ` juzhe.zhong at rivai dot ai 2023-12-28 4:53 ` tnfchris at gcc dot gnu.org 2023-12-28 5:08 ` tnfchris at gcc dot gnu.org 2023-12-28 9:11 ` juzhe.zhong at rivai dot ai 2023-12-28 21:20 ` [Bug tree-optimization/113134] " pinskia at gcc dot gnu.org 2024-01-08 8:11 ` rguenth at gcc dot gnu.org 2024-01-31 11:48 ` juzhe.zhong at rivai dot ai 2024-01-31 12:04 ` rguenth at gcc dot gnu.org 2024-02-02 3:38 ` juzhe.zhong at rivai dot ai 2024-02-02 8:49 ` juzhe.zhong at rivai dot ai
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-113134-4-8VVSFKUQtD@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).