René J.V. Bertin wrote: In addition to px->pixels==NULL (which it isn't always when the crash occurs), I also notice that px->bytesPerRow has a suspicious value. When I see the line `srow = sptr + (y * (sbpr / 4));`, I think that sbpr (px->bytesPerRow) should be equal or a known multiple of the mask width `sw`, correct?
The question is, why would px (pixmap.data.data()) seem to be corrupt if pixmap itself is allocated one level up in the backtrace? Is this a case of missing initialisation, perhaps? R. > Is there anything that can be done with a backtrace like the one below > that would make it worthwhile to file a bug report? Some prodding after > liberal use of `volatile` in qt_mac_create_imagemask suggests that > px->pixels==NULL > > R. > > (lldb) bt all > * thread #1: tid = 0x6280a, 0x00007fff8914fe20 > libsystem_kernel.dylib`__wait4 + 8, queue = 'com.apple.main-thread', stop > reason = signal SIGSTOP > * frame #0: 0x00007fff8914fe20 libsystem_kernel.dylib`__wait4 + 8 > frame #1: 0x000000010dab494e > libkdeui.5.dylib`KCrash::startProcess(int, char const**, bool) > [inlined] startProcessInternal(argc=<unavailable>, > directly=<unavailable>) + 265 at kcrash.cpp:556 frame #2: > 0x000000010dab4845 > libkdeui.5.dylib`KCrash::startProcess(argc=<unavailable>, > argv=<unavailable>, waitAndExit=<unavailable>) + 21 at kcrash.cpp:538 > frame #3: 0x000000010dab3a79 > libkdeui.5.dylib`KCrash::defaultCrashHandler(sig=<unavailable>) + 1209 > at kcrash.cpp:441 frame #4: 0x00007fff886fd5aa > libsystem_platform.dylib`_sigtramp + 26 frame #5: 0x000000010e681c21 > QtGui`qt_mac_create_imagemask(pixmap=<unavailable>, > sr=0x00007fff5360b620) + 1057 at qpixmap_mac.cpp:1090 frame #6: > 0x000000010e78f9ff QtGui`qt_mac_draw_pattern(info=0x00007fef4c72fa80, > c=0x00007fef4c70fab0) + 1167 at qpaintengine_mac.cpp:494 frame #7: _______________________________________________ Development mailing list [email protected] http://lists.qt-project.org/mailman/listinfo/development
