Re: An ICE with gfortran 4.6.1, optimization -O3, not with -O2.

[Tobias Burnus <burnus at net-b dot de>]

Am 23.08.2011 23:05, schrieb Steve Kargl:
> Here's the reduce test case that Toon meant to send.

Can someone fill a PR - and tell the bug number? I think it is a 
"middle-end" PR.

Some more data: No ICE with 4.7. While 4.6.2-20110818 (Rev. 177878)  
crashes at:

==27008== Invalid read of size 8
==27008==    at 0x89A58F: vect_is_simple_use_1 (tree-vect-stmts.c:5346)
==27008==    by 0x89B4E5: vectorizable_operation (tree-vect-stmts.c:2440)
==27008==    by 0x89F452: vect_transform_stmt (tree-vect-stmts.c:4841)
==27008==    by 0x8ACB23: vect_transform_loop (tree-vect-loop.c:4872)
==27008==    by 0x8B66E9: vectorize_loops (tree-vectorizer.c:205)

Working: 2010-04-07-r158027*
Failing: 2010-04-13-r158277* (= GCC 4.6 experimental)
(* build might contain extra Fortran patches, not checked but probably 
not relevent.)

FIXED: Must be fixed after 4.6 branching. GCC 4.7 2011-05-28-r174379 
still failed, while 2011-06-21-r175272 doesn't.

Tobias