[Bug optimization/11846] New: [3.3 regression] mozilla/amd64 miscompilation

[Bug optimization/11846] New: [3.3 regression] mozilla/amd64 miscompilation

[gbeauchesne at mandrakesoft dot com]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11846

           Summary: [3.3 regression] mozilla/amd64 miscompilation
           Product: gcc
           Version: 3.3.1
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: optimization
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: gbeauchesne at mandrakesoft dot com
                CC: aj at suse dot de,gcc-bugs at gcc dot gnu dot org,jakub
                    at redhat dot com
  GCC host triplet: x86_64-unknown-linux-gnu

Hi,

mozilla 1.4 is miscompiled on AMD64 with gcc 3.3.1 at -O2. Building
modules/libpr0n/decoders/gif/imgContainerGIF.cpp without GCSE solves the
problem. This is a regression from 3.2.3 available on all 3.3 branches.

I failed to reduce to a simple testcase yet. But I am pretty sure it lies in
imgContainerGIF.cpp (imgContainerGIF::Notify) at first call to inlinedGetFrameAt().

[Bug optimization/11846] [3.3 regression] mozilla/amd64 miscompilation

[pinskia at physics dot uc dot edu]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11846


pinskia at physics dot uc dot edu changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |WAITING


------- Additional Comments From pinskia at physics dot uc dot edu  2003-08-07 17:05 -------
Is by chance any related to bug 11376?
Can you retry with -fno-strict-aliasing?

[Bug optimization/11846] [3.3 regression] mozilla/amd64 miscompilation

[gbeauchesne at mandrakesoft dot com]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11846



------- Additional Comments From gbeauchesne at mandrakesoft dot com  2003-08-07 18:36 -------
Subject: Re:  [3.3 regression] mozilla/amd64 miscompilation

> ------- Additional Comments From pinskia at physics dot uc dot edu  
> 2003-08-07 17:05 -------
> Is by chance any related to bug 11376?

tcase5 & tcase6 indeed segfault on amd64 too. The fact that moz worked 
with -fno-gcse on my machine was probably pure luck.

> Can you retry with -fno-strict-aliasing?

I will check tomorrow. Thanks for linking those two bugs.

[Bug optimization/11846] [3.3 regression] mozilla/amd64 miscompilation

[gbeauchesne at mandrakesoft dot com]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11846


gbeauchesne at mandrakesoft dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|WAITING                     |RESOLVED
         Resolution|                            |DUPLICATE


------- Additional Comments From gbeauchesne at mandrakesoft dot com  2003-08-08 08:29 -------
-fno-strict-aliasing indeed fixes the bug. Replying with that version of mozilla
on amd64.

*** This bug has been marked as a duplicate of 11376 ***