public inbox for gcc-bugs@sourceware.org help / color / mirror / Atom feed
From: "jakub at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org> To: gcc-bugs@gcc.gnu.org Subject: [Bug tree-optimization/54073] [4.7/4.8 Regression] SciMark Monte Carlo test performance has seriously decreased in recent GCC releases Date: Tue, 13 Nov 2012 13:05:00 -0000 [thread overview] Message-ID: <bug-54073-4-YkGCrzbFye@http.gcc.gnu.org/bugzilla/> (raw) In-Reply-To: <bug-54073-4@http.gcc.gnu.org/bugzilla/> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54073 --- Comment #8 from Jakub Jelinek <jakub at gcc dot gnu.org> 2012-11-13 13:04:28 UTC --- Created attachment 28674 --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=28674 gcc48-pr54073.patch On x86_64-linux on SandyBridge CPU with -O3 -march=corei7-avx I've tracked it down to the http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=171341 change, in particular emit_conditional_move part of the changes. Before the change emit_conditional_move would completely ignore the predicate on the comparison operand (operands[1]), starting with r171341 it honors it. And the movsicc's ordered_comparison_operator would give up on the UNLT comparison in the MonteCarlo testcase, while ix86_expand_int_movcc expands it just fine and at least on that loop it is beneficial to use vucomisd %xmm0, %xmm1 cmovae %eax, %ebp instead of: .L4: addl $1, %ebx ... vucomisd %xmm0, %xmm2 jb .L4 The attached proof of concept patch attempts to just restore the 4.6 and earlier behavior by allowing in all comparisons. Of course perhaps it might be possible it needs better tuning than that, I meant it just as a start for discussions. vanilla trunk: ** ** ** SciMark2 Numeric Benchmark, see http://math.nist.gov/scimark ** ** for details. (Results can be submitted to pozo@nist.gov) ** ** ** Using 2.00 seconds min time per kenel. Composite Score: 1886.79 FFT Mflops: 1726.97 (N=1024) SOR Mflops: 1239.20 (100 x 100) MonteCarlo: Mflops: 374.13 Sparse matmult Mflops: 1956.30 (N=1000, nz=5000) LU Mflops: 4137.37 (M=100, N=100) patched trunk: ** ** ** SciMark2 Numeric Benchmark, see http://math.nist.gov/scimark ** ** for details. (Results can be submitted to pozo@nist.gov) ** ** ** Using 2.00 seconds min time per kenel. Composite Score: 1910.08 FFT Mflops: 1726.97 (N=1024) SOR Mflops: 1239.20 (100 x 100) MonteCarlo: Mflops: 528.94 Sparse matmult Mflops: 1949.03 (N=1000, nz=5000) LU Mflops: 4106.27 (M=100, N=100)
next prev parent reply other threads:[~2012-11-13 13:05 UTC|newest] Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top 2012-07-23 15:25 [Bug tree-optimization/54073] New: " t.artem at mailcity dot com 2012-07-23 15:44 ` [Bug tree-optimization/54073] " t.artem at mailcity dot com 2012-07-24 9:23 ` [Bug tree-optimization/54073] [4.7/4.8 Regression] " rguenth at gcc dot gnu.org 2012-07-24 11:29 ` markus at trippelsdorf dot de 2012-07-24 13:21 ` rguenth at gcc dot gnu.org 2012-07-26 15:41 ` venkataramanan.kumar at amd dot com 2012-07-26 16:13 ` markus at trippelsdorf dot de 2012-08-16 11:06 ` rguenth at gcc dot gnu.org 2012-09-07 10:09 ` rguenth at gcc dot gnu.org 2012-09-20 10:28 ` jakub at gcc dot gnu.org 2012-11-13 13:05 ` jakub at gcc dot gnu.org [this message] 2012-11-13 15:07 ` t.artem at mailcity dot com 2012-11-13 15:14 ` ubizjak at gmail dot com 2012-11-13 15:24 ` jakub at gcc dot gnu.org 2012-11-13 15:55 ` hubicka at gcc dot gnu.org 2012-11-16 11:41 ` jakub at gcc dot gnu.org 2012-11-16 14:50 ` [Bug tree-optimization/54073] [4.7 " jakub at gcc dot gnu.org 2012-12-31 9:41 ` pinskia at gcc dot gnu.org 2013-02-16 19:12 ` jake.stine at gmail dot com 2013-02-17 8:41 ` ubizjak at gmail dot com 2013-04-11 7:59 ` rguenth at gcc dot gnu.org 2014-06-12 13:16 ` 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-54073-4-YkGCrzbFye@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).