public inbox for gcc-bugs@sourceware.org help / color / mirror / Atom feed
From: "jschoen4 at gmail dot com" <gcc-bugzilla@gcc.gnu.org> To: gcc-bugs@gcc.gnu.org Subject: [Bug target/103393] [12 Regression] Generating 256bit register usage with -mprefer-avx128 -mprefer-vector-width=128 Date: Wed, 24 Nov 2021 20:38:59 +0000 [thread overview] Message-ID: <bug-103393-4-uOXsE6vOmY@http.gcc.gnu.org/bugzilla/> (raw) In-Reply-To: <bug-103393-4@http.gcc.gnu.org/bugzilla/> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103393 --- Comment #4 from John S <jschoen4 at gmail dot com> --- I can Confirm from my side that it does appear to be the memmove inline expansion and not the auto vectorizer. It also occurs with builtin_memset/builtin_memcpy as well. For some context, this is an issue would prevent the usage of gcc in my production environment. It will certainly impact other use cases outside of my own as well. For example, it becomes impossible to use "-mno-vzeroupper -mavx -mpreferred-vector-width=128" and use _mm256_xxx + _mm256_zeroupper() intrinsics to properly manage the ymm state (clear or not) since the compiler is now able to insert ymm's almost anywhere via the memmove inlining. Up until now the prefer-width has always behaved as in a way that all auto generated vector uses will not exceed the preferred width. Only explicit use of the _mm256/_mm512_ .. intrinsics or the "vector types" i.e. `__m256 var; __m512 var;` would result in wider register usage. I do believe Clang/icc behave this way as well and there are dependencies on this behavior. The same also applies w/ avx-512 enabled with ZMM usage + prefer=128/256 where the downclocking issues can be even more pronounced.
next prev parent reply other threads:[~2021-11-24 20:38 UTC|newest] Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top 2021-11-23 19:20 [Bug tree-optimization/103393] New: [ 12 Regression ] Auto vectorizer generating " jschoen4 at gmail dot com 2021-11-24 8:57 ` [Bug tree-optimization/103393] [12 Regression] Generating " rguenth at gcc dot gnu.org 2021-11-24 13:45 ` [Bug target/103393] " hjl.tools at gmail dot com 2021-11-24 13:53 ` rguenth at gcc dot gnu.org 2021-11-24 20:38 ` jschoen4 at gmail dot com [this message] 2021-11-25 1:15 ` crazylht at gmail dot com 2021-11-25 1:25 ` crazylht at gmail dot com 2021-11-25 7:16 ` rguenther at suse dot de 2021-11-25 7:28 ` [Bug middle-end/103393] " rguenth at gcc dot gnu.org 2021-11-25 7:40 ` rguenth at gcc dot gnu.org 2021-11-25 17:57 ` jakub at gcc dot gnu.org 2021-11-25 18:09 ` jakub at gcc dot gnu.org 2021-11-25 20:54 ` rearnsha at gcc dot gnu.org 2021-11-25 20:57 ` rearnsha at gcc dot gnu.org 2021-11-25 22:49 ` hjl.tools at gmail dot com 2021-11-26 11:31 ` rearnsha at gcc dot gnu.org 2021-11-26 11:37 ` jakub at gcc dot gnu.org 2021-11-26 11:44 ` rearnsha at gcc dot gnu.org 2021-11-26 11:48 ` jakub at gcc dot gnu.org 2021-11-26 11:51 ` rearnsha at gcc dot gnu.org 2021-11-26 11:58 ` jakub at gcc dot gnu.org 2021-11-26 12:26 ` rguenth at gcc dot gnu.org 2021-11-26 12:51 ` rearnsha at gcc dot gnu.org 2022-03-01 22:41 ` hjl.tools at gmail dot com 2022-03-02 14:52 ` hjl.tools at gmail dot com 2022-03-31 7:31 ` 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-103393-4-uOXsE6vOmY@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).