On Mar 20, 2011 7:15 PM, "Colin Guthrie" <<a href="mailto:gmane@colin.guthr.ie">gmane@colin.guthr.ie</a>> wrote: > > 'Twas brillig, and Daniel Mack at 20/03/11 17:39 did gyre and gimble: > > From: Daniel Mack <<a href="mailto:daniel@caiaq.de">daniel@caiaq.de</a>> > > > > This would have crashed when unloading the module. > > --- > > src/modules/coreaudio/module-coreaudio-detect.c | 5 +---- > > 1 files changed, 1 insertions(+), 4 deletions(-) > > > > diff --git a/src/modules/coreaudio/module-coreaudio-detect.c b/src/modules/coreaudio/module-coreaudio-detect.c > > index 1d26b84..fda1f17 100644 > > --- a/src/modules/coreaudio/module-coreaudio-detect.c > > +++ b/src/modules/coreaudio/module-coreaudio-detect.c > > @@ -231,13 +231,10 @@ fail: > > } > > > > void pa__done(pa_module *m) { > > - struct userdata *u; > > + struct userdata *u = m->userdata; > > struct ca_device *dev = u->devices; > > AudioObjectPropertyAddress property_address; > > > > - pa_assert(m); > > - pa_assert_se(u = m->userdata); > > - > > property_address.mSelector = kAudioHardwarePropertyDevices; > > property_address.mScope = kAudioObjectPropertyScopeGlobal; > > property_address.mElement = kAudioObjectPropertyElementMaster; > > From what I can see, most modules are still calling pa_assert(m) so it's > probably better to keep the assert in there and just move the dev = > u->devices line lower. > > Other than that the other patches are all fine. Ok, can you add it back and amend the patch? I think that would be easier than resending :-) Thanks, Daniel