Crash when re-importing from divecomputer
Dirk Hohndel
dirk at hohndel.org
Wed Jan 23 12:09:48 PST 2013
Henrik Brautaset Aronsen <subsurface at henrik.synth.no> writes:
> 1) Download dives from divecomputer, save and exit
> 2) Open Subsurface again, and delete the two latest dives
> 3) Start a download from the divecomputer again
>
> *kaboom*
>
> Thread 0:: Dispatch queue: com.apple.main-thread
> 0 libsystem_kernel.dylib 0x00007fff92f53686 mach_msg_trap + 10
> 1 libsystem_kernel.dylib 0x00007fff92f52c42 mach_msg + 70
> 2 com.apple.CoreFoundation 0x00007fff90d53803
> __CFRunLoopServiceMachPort + 195
> 3 com.apple.CoreFoundation 0x00007fff90d58ee6 __CFRunLoopRun
> + 1078
> 4 com.apple.CoreFoundation 0x00007fff90d586b2
> CFRunLoopRunSpecific + 290
> 5 com.apple.HIToolbox 0x00007fff91a6a0a4
> RunCurrentEventLoopInMode + 209
> 6 com.apple.HIToolbox 0x00007fff91a69e42
> ReceiveNextEventCommon + 356
> 7 com.apple.HIToolbox 0x00007fff91a69cd3
> BlockUntilNextEventMatchingListInMode + 62
> 8 com.apple.AppKit 0x00007fff964ad613 _DPSNextEvent + 685
> 9 com.apple.AppKit 0x00007fff964aced2 -[NSApplication
> nextEventMatchingMask:untilDate:inMode:dequeue:] + 128
> 10 libgdk-quartz-2.0.0.dylib 0x0000000105181a12 poll_func + 178
> 11 libglib-2.0.0.dylib 0x00000001056cd9b5
> g_main_context_iterate + 421
> 12 libglib-2.0.0.dylib 0x00000001056cefaf g_main_loop_run
> + 287
Everything above is the internal gtk/quartz implementation of
gtk_dialog_run - so this makes me think that this must be a memory
corruption issue.
> 13 libgtk-quartz-2.0.0.dylib 0x0000000104d27b99 gtk_dialog_run
> + 377
> 14 subsurface 0x00000001049b4a9a do_import + 234
> (libdivecomputer.c:752)
> 15 subsurface 0x00000001049ce4dc
> import_dive_computer + 156 (download-dialog.c:134)
> 16 subsurface 0x00000001049cf999 download_dialog
> + 1897 (download-dialog.c:451)
>
> If I save and exit after 2) and restart Subsurface, the reimport works fine.
Can you run this under Valgrind and try to reproduce?
/D
More information about the subsurface
mailing list