[Bug tree-optimization/60042] vectorizer still does too many dependence tests for himeno:jacobi

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=60042 Bug 60042 depends on bug 23855, which changed state. Bug 23855 Summary: loop header should also be pulled out of the inner loop too https://gcc.gnu.org/bugzilla/show_bug.cgi?id=23855 What|Removed |Added

[Bug tree-optimization/60042] vectorizer still does too many dependence tests for himeno:jacobi

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=60042 Richard Biener changed: What|Removed |Added CC||amker.cheng at gmail dot com --- Commen

[Bug tree-optimization/60042] vectorizer still does too many dependence tests for himeno:jacobi

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60042 --- Comment #10 from Richard Biener --- Author: rguenth Date: Mon Apr 14 13:57:00 2014 New Revision: 209374 URL: http://gcc.gnu.org/viewcvs?rev=209374&root=gcc&view=rev Log: 2014-04-14 Richard Biener PR tree-optimization/60042 * tree-

[Bug tree-optimization/60042] vectorizer still does too many dependence tests for himeno:jacobi

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60042 --- Comment #9 from Richard Biener --- PR23855 fixed we'd get for the analysis of the remaining DRs in the outermost loop for example for matrix A: Creating dr for *_290 analyze_innermost: success. base_address: pretmp_1792 + (sizetype) (

[Bug tree-optimization/60042] vectorizer still does too many dependence tests for himeno:jacobi

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60042 Richard Biener changed: What|Removed |Added Depends on||23855 --- Comment #8 from Richard Biener

[Bug tree-optimization/60042] vectorizer still does too many dependence tests for himeno:jacobi

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60042 Jakub Jelinek changed: What|Removed |Added CC||jakub at gcc dot gnu.org --- Comment #7 f

[Bug tree-optimization/60042] vectorizer still does too many dependence tests for himeno:jacobi

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60042 Richard Biener changed: What|Removed |Added Attachment #32038|0 |1 is obsolete|

[Bug tree-optimization/60042] vectorizer still does too many dependence tests for himeno:jacobi

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60042 Richard Biener changed: What|Removed |Added Status|UNCONFIRMED |ASSIGNED Last reconfirmed|

[Bug tree-optimization/60042] vectorizer still does too many dependence tests for himeno:jacobi

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60042 --- Comment #4 from Richard Biener --- (In reply to Richard Biener from comment #3) > Bah, and with -fipa-pta -Ofast -fwhole-program we now _do_ see that there > isn't any aliasing but PRE messes up the loop and creates loop carried > dependencies

[Bug tree-optimization/60042] vectorizer still does too many dependence tests for himeno:jacobi

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60042 --- Comment #3 from Richard Biener --- Bah, and with -fipa-pta -Ofast -fwhole-program we now _do_ see that there isn't any aliasing but PRE messes up the loop and creates loop carried dependencies ... :/

[Bug tree-optimization/60042] vectorizer still does too many dependence tests for himeno:jacobi

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60042 --- Comment #2 from Richard Biener --- With some more dumping I seee himenobmtxpa.c:296:9: note: === vect_prune_runtime_alias_test_list === himenobmtxpa.c:296:9: note: merging ranges for *_205, *_324 and *_49, *_324 himenobmtxpa.c:296:9: note: me

[Bug tree-optimization/60042] vectorizer still does too many dependence tests for himeno:jacobi

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60042 --- Comment #1 from Richard Biener --- Created attachment 32027 --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=32027&action=edit patch to prune deps to scalar globals My patch to prune dependences to scalar global vars.