[Bug tree-optimization/31966] Miscompiles valid code with -ftree-vectorize and -march=nocona
--- Comment #3 from dorit at gcc dot gnu dot org 2007-07-01 08:47 --- I was able to reproduce the behavior on a Pentinum4 with a recent 4.3 snapshot. It doesn't look like it's the vectorizer's fault - nothing gets vectorized. However, if I disable the tree-level if-conversion (which is automatically enabled by -ftree-vectorize) - the result of '-ftree-vectorize -O2 -march=nocona divop.c -o divop' becomes '3CBA83'. -- dorit at gcc dot gnu dot org changed: What|Removed |Added Keywords||wrong-code http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31966
[Bug tree-optimization/31966] Miscompiles valid code with -ftree-vectorize and -march=nocona
--- Comment #4 from dorit at gcc dot gnu dot org 2007-07-01 08:52 --- This may be related to PR32533 (also wrong-code due to tree-if-conversion). -- dorit at gcc dot gnu dot org changed: What|Removed |Added Status|UNCONFIRMED |NEW Ever Confirmed|0 |1 Last reconfirmed|-00-00 00:00:00 |2007-07-01 08:52:43 date|| http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31966
[Bug tree-optimization/31966] Miscompiles valid code with -ftree-vectorize and -march=nocona
--- Comment #2 from lloyd at randombit dot net 2007-06-30 22:11 --- The behavior still exists in the 4.3 20070622 snapshot. It does not occur using -march=core2 (the actual CPU in question). The bad value results when using -ftree-vectorize and -march or -mtune =nocona. -O, -O2, or -O3 is also required; a binary compiled with -Os produces the correct value. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31966
[Bug tree-optimization/31966] Miscompiles valid code with -ftree-vectorize and -march=nocona
--- Comment #1 from lloyd at randombit dot net 2007-05-17 07:36 --- Created an attachment (id=13567) -- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=13567action=view) A short testcase for bug 31966 -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31966