public inbox for gcc-bugs@sourceware.org help / color / mirror / Atom feed
From: "evstupac at gmail dot com" <gcc-bugzilla@gcc.gnu.org> To: gcc-bugs@gcc.gnu.org Subject: [Bug target/53967] GCC produces slow code for convolution algorithm with -mfpmath=sse (the AMD_64 default) Date: Wed, 18 Jul 2012 09:45:00 -0000 [thread overview] Message-ID: <bug-53967-4-oOohZZVND1@http.gcc.gnu.org/bugzilla/> (raw) In-Reply-To: <bug-53967-4@http.gcc.gnu.org/bugzilla/> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53967 Stupachenko Evgeny <evstupac at gmail dot com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |evstupac at gmail dot com --- Comment #12 from Stupachenko Evgeny <evstupac at gmail dot com> 2012-07-18 09:45:15 UTC --- I tried it at "-O2" and got low performance with -mfpmath=sse. It looks like it is caused by register dependency (%xmm0) between: addss %xmm0, %xmm1 cvtsi2ss %eax, %xmm0 Renaming %xmm0 in cvtsi2ss to another free register in all such cases resolves the issue. Also you can try "-O2 -funroll-loops", which made "sse" code even faster and and "-O2 -fschedule-insns" which significantly reduced performance loses in "sse" case.
next prev parent reply other threads:[~2012-07-18 9:45 UTC|newest] Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top 2012-07-14 20:52 [Bug c/53967] New: " bfriesen at simple dot dallas.tx.us 2012-07-14 20:56 ` [Bug c/53967] " bfriesen at simple dot dallas.tx.us 2012-07-14 20:57 ` bfriesen at simple dot dallas.tx.us 2012-07-14 20:58 ` bfriesen at simple dot dallas.tx.us 2012-07-14 20:59 ` bfriesen at simple dot dallas.tx.us 2012-07-14 21:06 ` [Bug target/53967] " bfriesen at simple dot dallas.tx.us 2012-07-14 21:42 ` bfriesen at simple dot dallas.tx.us 2012-07-16 12:42 ` rguenth at gcc dot gnu.org 2012-07-16 14:17 ` bfriesen at simple dot dallas.tx.us 2012-07-16 14:57 ` rguenth at gcc dot gnu.org 2012-07-16 15:35 ` bfriesen at simple dot dallas.tx.us 2012-07-16 15:41 ` bfriesen at simple dot dallas.tx.us 2012-07-17 9:20 ` rguenth at gcc dot gnu.org 2012-07-18 9:45 ` evstupac at gmail dot com [this message] 2012-07-18 10:50 ` rguenth at gcc dot gnu.org 2012-07-18 14:28 ` bfriesen at simple dot dallas.tx.us 2012-07-18 20:42 ` bfriesen at simple dot dallas.tx.us 2012-07-19 14:29 ` bfriesen at simple dot dallas.tx.us 2012-07-21 1:05 ` bfriesen at simple dot dallas.tx.us 2012-08-12 15:41 ` xunxun1982 at gmail dot com
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-53967-4-oOohZZVND1@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).