public inbox for gcc-bugs@sourceware.org help / color / mirror / Atom feed
From: "rguenth at gcc dot gnu dot org" <gcc-bugzilla@gcc.gnu.org> To: gcc-bugs@gcc.gnu.org Subject: [Bug tree-optimization/32921] [4.3 Regression] Revision 126326 causes 12% slowdown Date: Wed, 24 Oct 2007 12:23:00 -0000 [thread overview] Message-ID: <20071024122323.27354.qmail@sourceware.org> (raw) In-Reply-To: <bug-32921-682@http.gcc.gnu.org/bugzilla/> ------- Comment #31 from rguenth at gcc dot gnu dot org 2007-10-24 12:23 ------- This is a partitioning issue. We allocate the same MPT to loads and stores in a loop. Fixed with for example --param max-aliased-vops=10000. Diego, this is yours. Avoiding false aliasing of loads and stores should be top priority in the partitioning heuristics. Though even with no paritioning we are not able to move the invariant load of dtvol: .L68: movsd (%rsi), %xmm0 addl $1, %ecx addq %rdi, %rsi subsd (%rdx), %xmm0 addq %rdi, %rdx mulsd __les3d_data_MOD_dtvol(%rip), %xmm0 xorpd %xmm1, %xmm0 movsd %xmm0, (%rax) addq %r8, %rax cmpl $6, %ecx jne .L68 because: # VUSE <dtvol_613> dtvol.23_424 = dtvol; D.1276_425 = D.1274_423 * dtvol.23_424; D.1277_426 = -D.1276_425; # dtvol_1865 = VDEF <dtvol_613> # HEAP.185_1160 = VDEF <HEAP.185_609> # HEAP.186_1142 = VDEF <HEAP.186_1194> # HEAP.202_1108 = VDEF <HEAP.202_1177> (*pretmp.248_955)[D.1259_397] = D.1277_426; as we think that pretmp.248_955 might point to dtvol. The pointers just have a constraint like pretmp.246_921 = du and du = &ANYTHING du.ubound = &ANYTHING du.lbound = &ANYTHING du.stride = &ANYTHING du.ubound = &ANYTHING du.lbound = &ANYTHING du.stride = &ANYTHING du.ubound = &ANYTHING du.lbound = &ANYTHING du.stride = &ANYTHING du.ubound = &ANYTHING du.lbound = &ANYTHING du.stride = &ANYTHING du.dtype = &ANYTHING du.offset = &ANYTHING and dtvol and DU are global variables. So we are possibly out of luck for this one. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32921
next prev parent reply other threads:[~2007-10-24 12:23 UTC|newest] Thread overview: 64+ messages / expand[flat|nested] mbox.gz Atom feed top 2007-07-28 0:23 [Bug tree-optimization/32921] New: " hjl at lucon dot org 2007-07-28 0:32 ` [Bug tree-optimization/32921] " pinskia at gcc dot gnu dot org 2007-07-28 0:32 ` pinskia at gcc dot gnu dot org 2007-07-28 0:51 ` hjl at lucon dot org 2007-07-28 1:02 ` pinskia at gcc dot gnu dot org 2007-07-28 10:02 ` rguenth at gcc dot gnu dot org 2007-07-28 20:51 ` hjl at lucon dot org 2007-07-28 20:59 ` rguenther at suse dot de 2007-07-28 23:25 ` hjl at lucon dot org 2007-09-11 11:48 ` jsm28 at gcc dot gnu dot org 2007-10-07 15:17 ` hjl at lucon dot org 2007-10-07 15:17 ` hjl at lucon dot org 2007-10-07 15:19 ` hjl at lucon dot org 2007-10-10 17:48 ` mmitchel at gcc dot gnu dot org 2007-10-17 16:14 ` pthaugen at gcc dot gnu dot org 2007-10-17 16:19 ` pthaugen at gcc dot gnu dot org 2007-10-17 16:59 ` pinskia at gcc dot gnu dot org 2007-10-17 17:41 ` dberlin at dberlin dot org 2007-10-17 21:22 ` rguenth at gcc dot gnu dot org 2007-10-18 11:45 ` rguenth at gcc dot gnu dot org 2007-10-18 20:55 ` hjl at lucon dot org 2007-10-19 11:26 ` rguenth at gcc dot gnu dot org 2007-10-19 12:28 ` rguenth at gcc dot gnu dot org 2007-10-19 12:40 ` rguenth at gcc dot gnu dot org 2007-10-19 15:36 ` rguenth at gcc dot gnu dot org 2007-10-19 16:28 ` rguenth at gcc dot gnu dot org 2007-10-19 22:13 ` hjl at lucon dot org 2007-10-20 4:21 ` patchapp at dberlin dot org 2007-10-20 9:39 ` rguenth at gcc dot gnu dot org 2007-10-20 16:05 ` hjl at lucon dot org 2007-10-22 21:11 ` pthaugen at gcc dot gnu dot org 2007-10-23 2:23 ` Joey dot ye at intel dot com 2007-10-23 20:29 ` pthaugen at gcc dot gnu dot org 2007-10-23 20:31 ` pthaugen at gcc dot gnu dot org 2007-10-24 12:23 ` rguenth at gcc dot gnu dot org [this message] 2007-10-24 15:02 ` rguenth at gcc dot gnu dot org 2007-10-24 15:18 ` rguenth at gcc dot gnu dot org 2007-10-24 15:19 ` rguenth at gcc dot gnu dot org 2007-10-24 23:15 ` pthaugen at gcc dot gnu dot org 2007-10-25 8:55 ` rguenth at gcc dot gnu dot org 2007-10-25 10:06 ` pinskia at gcc dot gnu dot org 2007-10-25 11:02 ` rguenth at gcc dot gnu dot org 2007-10-25 11:07 ` rguenth at gcc dot gnu dot org 2008-03-14 16:53 ` [Bug tree-optimization/32921] [4.3/4.4 " rguenth at gcc dot gnu dot org 2008-04-23 14:10 ` rguenth at gcc dot gnu dot org 2008-04-23 14:10 ` rguenth at gcc dot gnu dot org 2008-04-30 18:50 ` pthaugen at gcc dot gnu dot org 2008-04-30 20:30 ` rguenth at gcc dot gnu dot org 2008-04-30 21:43 ` rguenth at gcc dot gnu dot org 2008-04-30 22:36 ` pthaugen at gcc dot gnu dot org 2008-05-01 19:27 ` pthaugen at gcc dot gnu dot org 2008-05-02 10:29 ` rguenth at gcc dot gnu dot org 2008-05-02 12:17 ` rguenth at gcc dot gnu dot org 2008-05-02 12:33 ` dnovillo at google dot com 2008-05-02 12:56 ` rguenther at suse dot de 2008-05-02 14:16 ` mmitchel at gcc dot gnu dot org 2008-05-02 15:06 ` rguenther at suse dot de 2008-06-06 14:59 ` rguenth at gcc dot gnu dot org 2008-08-27 22:04 ` jsm28 at gcc dot gnu dot org 2008-12-08 15:41 ` rguenth at gcc dot gnu dot org 2009-01-24 10:22 ` rguenth at gcc dot gnu dot org 2009-01-31 14:27 ` [Bug tree-optimization/32921] [4.3 " bonzini at gnu dot org 2009-01-31 14:36 ` [Bug tree-optimization/32921] [4.3/4.4 " rguenth at gcc dot gnu dot org 2009-02-02 19:17 ` pthaugen at gcc dot gnu dot 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=20071024122323.27354.qmail@sourceware.org \ --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).