On Thu, Feb 02, 2006 at 01:19:15PM +0100, Jordi Mallach wrote: > On Thu, Feb 02, 2006 at 12:05:53PM +0100, Alexander Sack - Debian Bugmail > wrote: >> On Wed, Feb 01, 2006 at 05:45:37PM -0500, Ryan Richter wrote:
>> Loic, you a time-slot NMU this? I am currently under 120% load, so >> I probably won't have time for it in the near future! > What's the correct patch to apply here? Build with -O instead of > -O2, or to try the bugzilla patch provided by Lionel? -O instead of -O2 is a work-around that will work. My (backported) bugzilla patch seems to work for me (Galeon still crashes occasionally and non-reproducibly, but not more than on an i386 machine...), but is not guaranteed to catch all occurrences of the problem: see http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11376#c38 . Note: The upstream log mentions gcc bugs thrown is *as* *well*: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22591 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23192 These are fixed in gcc 4.0.2, which we have. Ubuntu compiles with -fno-strict-aliasing. On all platforms, I presume. And my amd64-using neighbour is not having particular problems. This actually makes a lot of sense since the bug in Mozilla that triggers all this is an aliasing violation... -- Lionel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
