[Bug tree-optimization/25621] Missed optimization when unrolling the loop (splitting up the sum) (only with -ffast-math)

[iliyapalachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25621

Ilya Palachev  changed:

   What|Removed |Added

 CC||iliyapalachev at gmail dot com

--- Comment #14 from Ilya Palachev  ---
(In reply to Joost VandeVondele from comment #13)

At page http://gcc.gnu.org/wiki/VectorizationTasks

it is written that the generalization of reduction support
(http://gcc.gnu.org/ml/gcc-patches/2006-04/msg00172.html) can help to fix this
bug.

Is this information still correct for gcc-4.9?

[Bug tree-optimization/57169] fully unrolled matrix multiplication not vectorized

[iliyapalachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57169

Ilya Palachev  changed:

   What|Removed |Added

 CC||iliyapalachev at gmail dot com

--- Comment #2 from Ilya Palachev  ---
(In reply to Richard Biener from comment #1)
> This is because basic-block SLP does not support vectorizing reductions.


At page http://gcc.gnu.org/wiki/VectorizationTasks

it is written that the generalization of reduction support
(http://gcc.gnu.org/ml/gcc-patches/2006-04/msg00172.html) can help to fix this
PR25621

Has this bug the same reason as the bug discussed in PR25621 ?

[Bug bootstrap/61164] GCC 4.9.0 fails to build libitm when fortification enabled

[iliyapalachev at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61164

Ilya Palachev  changed:

   What|Removed |Added

 CC||iliyapalachev at gmail dot com

--- Comment #1 from Ilya Palachev  ---
I have the same build error.

The build error is reproduced with Linaro GCC 4.9 2014.06 release.

Also switching off -D_FORTIFY_SOURCE=2 helps.