This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug middle-end/52209] [4.7 Regression] wrong code at -O0
- 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: Sat, 11 Feb 2012 09:15:41 +0000
- Subject: [Bug middle-end/52209] [4.7 Regression] wrong code at -O0
- Auto-submitted: auto-generated
- References: <bug-52209-4@http.gcc.gnu.org/bugzilla/>
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52209
Jakub Jelinek <jakub at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |ASSIGNED
AssignedTo|unassigned at gcc dot |jakub at gcc dot gnu.org
|gnu.org |
--- Comment #4 from Jakub Jelinek <jakub at gcc dot gnu.org> 2012-02-11 09:15:41 UTC ---
Created attachment 26640
--> http://gcc.gnu.org/bugzilla/attachment.cgi?id=26640
gcc47-pr52209.patch
I don't think we need REDUCE_BIT_FIELD here, IMHO the old code was just fine
for the signed bitfields. Because then op0 will be either 0 or -1, NOT on it
turns it into -1 or 0 and REDUCE_BIT_FIELD would do nothing on that. Richard's
change was IMHO only needed for unsigned bitfields.