@bojo42, Ah, ok according to those links they just expired it invalid, so there's no patch we should be looking for. I'll drop the bug link, so this doesn't show up as a "bug fixed upstream".
Also you earlier asked a question about your Xorg.0.log. I think you're right that xrandr causes the rescanning; however I think this only occurs when inputs are activated/deactivated. I've not experimented with this myself so don't know for certain, however. Looking at the full backtrace you posted earlier, the crash is occurring in xf86_load_cursor_argb(), which is a fairly short routine. Segfaults typically come from invalid memory access, such as a) dereferencing a null pointer, b) using an uninitialized pointer, or b) accessing array elements beyond their allocated length. Looking at your full backtrace, all the pointer values are printed and none are NULL, so that sort of rules out (a). I see that it's crashing while c = 0, which rules out (b) - unless the array was unallocated. So (b) seems to be the most likely situation. I think the next step might be to build the xserver with optimizations turned off (-O0) and recreate the crash; if it's an uninitialized pointer problem, this should hopefully help give more info on it. Other than that, I might suggest trying to step through it in the debugger. It might also be worthwhile to build the xserver from upstream's git HEAD and see if the problem still exists there; if not, then there might be a patch we could pull. I compared the xf86Cursors.c files from Ubuntu and upstream and there were a few changes but nothing that jumped out as obviously related to this problem, so I'm guessing it still remains unfixed, but it could be worth testing. -- [gutsy] X crashes randomly - something with cursors? (945GM) https://bugs.launchpad.net/bugs/153466 You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
