This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug tree-optimization/51466] [4.7 Regression] ICE in gimple_rhs_has_side_effects,
- From: "jakub at gcc dot gnu.org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Thu, 08 Dec 2011 14:45:44 +0000
- Subject: [Bug tree-optimization/51466] [4.7 Regression] ICE in gimple_rhs_has_side_effects,
- Auto-submitted: auto-generated
- References: <bug-51466-4@http.gcc.gnu.org/bugzilla/>
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51466
--- Comment #1 from Jakub Jelinek <jakub at gcc dot gnu.org> 2011-12-08 14:45:44 UTC ---
Created attachment 26021
--> http://gcc.gnu.org/bugzilla/attachment.cgi?id=26021
gcc47-pr51466.patch
Possible fix. Though, I wonder if this isn't unnecessarily pessimistic.
The reason why we ICE is because when clearing TREE_THIS_VOLATILE on the
new_rhs, we don't also clear TREE_SIDE_EFFECTS.
So another alternative would be perhaps to also copy TREE_SIDE_EFFECTS in all
the 4 places where we currently copy TREE_THIS_VOLATILE. That fixes this too.