public inbox for gcc-bugs@sourceware.org help / color / mirror / Atom feed
From: "wschmidt at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org> To: gcc-bugs@gcc.gnu.org Subject: [Bug tree-optimization/49749] Reassociation rank algorithm does not include all non-NULL operands Date: Wed, 20 Jul 2011 16:30:00 -0000 [thread overview] Message-ID: <bug-49749-4-anzoDinpTB@http.gcc.gnu.org/bugzilla/> (raw) In-Reply-To: <bug-49749-4@http.gcc.gnu.org/bugzilla/> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49749 William J. Schmidt <wschmidt at gcc dot gnu.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |ASSIGNED AssignedTo|unassigned at gcc dot |wschmidt at gcc dot gnu.org |gnu.org | --- Comment #9 from William J. Schmidt <wschmidt at gcc dot gnu.org> 2011-07-20 16:28:49 UTC --- Created attachment 24798 --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=24798 Proposed patch I'm attaching a patch that solves this issue. The patch was produced against the ibm/gcc-4_6-branch, but should apply OK to trunk -- I'll verify that later if the direction of this patch is acceptable. This regains the 20% performance loss we had experienced in 410.bwaves, and also gives a 5% boost to 444.namd. No significant degradations were observed on SPEC cpu2006. In addition to fixing the operand access problems, the patch looks for loop-carried dependencies in innermost loops, and biases the reassociation so that the phi target is summed last. The purpose of this is to identify accumulator variables in inner loops and make each iteration of their accumulations independent. When these loops are unrolled, multiple independent iterations can be interleaved for improved performance. The optimization is restricted to innermost loops to avoid unnecessarily raising register pressure. There may be a better way to achieve the bias than what I've chosen here, so please comment. If the general approach is acceptable, I'll apply comments and submit the patch for approval.
next prev parent reply other threads:[~2011-07-20 16:30 UTC|newest] Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top 2011-07-14 16:51 [Bug tree-optimization/49749] New: " wschmidt at gcc dot gnu.org 2011-07-14 19:23 ` [Bug tree-optimization/49749] " wschmidt at gcc dot gnu.org 2011-07-14 19:23 ` wschmidt at gcc dot gnu.org 2011-07-14 19:27 ` wschmidt at gcc dot gnu.org 2011-07-14 19:28 ` wschmidt at gcc dot gnu.org 2011-07-14 19:30 ` wschmidt at gcc dot gnu.org 2011-07-15 20:16 ` wschmidt at gcc dot gnu.org 2011-07-15 20:21 ` wschmidt at gcc dot gnu.org 2011-07-18 9:11 ` rguenth at gcc dot gnu.org 2011-07-20 16:30 ` wschmidt at gcc dot gnu.org [this message] 2011-07-20 16:39 ` wschmidt at gcc dot gnu.org 2011-07-20 19:02 ` wschmidt at gcc dot gnu.org 2011-07-21 12:07 ` rguenth at gcc dot gnu.org 2011-07-21 18:09 ` wschmidt at gcc dot gnu.org 2011-07-21 20:28 ` wschmidt at gcc dot gnu.org 2011-07-29 18:41 ` wschmidt at gcc dot gnu.org 2011-07-31 19:00 ` wschmidt at gcc dot gnu.org 2011-07-31 19:01 ` wschmidt at gcc dot gnu.org 2021-09-28 12:11 ` cvs-commit 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-49749-4-anzoDinpTB@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).