[Bug rtl-optimization/49390] [4.6/4.7 Regression] GCSE miscompilation

2011-06-14 Thread jakub at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49390 Jakub Jelinek changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution|

[Bug rtl-optimization/49390] [4.6/4.7 Regression] GCSE miscompilation

2011-06-14 Thread jakub at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49390 --- Comment #6 from Jakub Jelinek 2011-06-14 15:01:15 UTC --- Author: jakub Date: Tue Jun 14 15:01:10 2011 New Revision: 175024 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=175024 Log: PR rtl-optimization/49390 Revert: 2010-

[Bug rtl-optimization/49390] [4.6/4.7 Regression] GCSE miscompilation

2011-06-14 Thread jakub at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49390 --- Comment #5 from Jakub Jelinek 2011-06-14 14:59:55 UTC --- Author: jakub Date: Tue Jun 14 14:59:52 2011 New Revision: 175023 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=175023 Log: PR rtl-optimization/49390 Revert: 2010-

[Bug rtl-optimization/49390] [4.6/4.7 Regression] GCSE miscompilation

2011-06-14 Thread jakub at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49390 Jakub Jelinek changed: What|Removed |Added Status|UNCONFIRMED |ASSIGNED Last reconfirmed|

[Bug rtl-optimization/49390] [4.6/4.7 Regression] GCSE miscompilation

2011-06-13 Thread jakub at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49390 --- Comment #4 from Jakub Jelinek 2011-06-13 15:35:13 UTC --- Created attachment 24510 --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=24510 gcc46-pr49390.patch Untested patch. Richard, what do you think about it? Bernd, do you still have s

[Bug rtl-optimization/49390] [4.6/4.7 Regression] GCSE miscompilation

2011-06-13 Thread jakub at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49390 --- Comment #3 from Jakub Jelinek 2011-06-13 14:28:31 UTC --- Perhaps, if the tests are more expensive, case MEM: if (for_gcse) could first do the cheap tests, then if (!exp_equiv_p (XEXP (x, 0), XEXP (y, 0), validate, 1)) return 0; then do

[Bug rtl-optimization/49390] [4.6/4.7 Regression] GCSE miscompilation

2011-06-13 Thread jakub at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49390 --- Comment #2 from Jakub Jelinek 2011-06-13 13:53:23 UTC --- Perhaps we should have some exceptions where we allow different MEM_ATTRS, but they need to be carefully chosen. E.g. if both refs are indirect refs and are similar, with the same poi

[Bug rtl-optimization/49390] [4.6/4.7 Regression] GCSE miscompilation

2011-06-13 Thread jakub at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49390 --- Comment #1 from Jakub Jelinek 2011-06-13 12:36:58 UTC --- Blindly ignoring MEM_EXPR or other attributes looks very wrong to me. Guess in some cases it could return true even when MEM_ATTRS aren't identical, but they'd need to have the same be

[Bug rtl-optimization/49390] [4.6/4.7 Regression] GCSE miscompilation

2011-06-13 Thread jakub at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49390 Jakub Jelinek changed: What|Removed |Added Known to work||4.4.6, 4.5.2 Target Milestone|---