Bug#778529: lame: fill_buffer_resample segmentation fault

2015-02-18 Thread Fabian Greffrath
Am Mittwoch, den 18.02.2015, 12:11 +0100 schrieb Fabian Greffrath: > This time, it was a simple logical error in the lame sources: The "fake" > sample rate of the fuzzed input file is 1631 kHz which lame tries to > sample down to 48 kHz in the process of encoding. The ratio between > input and

Bug#778529: lame: fill_buffer_resample segmentation fault

2015-02-18 Thread Henri Salo
On Wed, Feb 18, 2015 at 12:11:35PM +0100, Fabian Greffrath wrote: > Phew, got it. Thank you for your comprehensive analysis. I have verified that the patch fixes this issue. Should I report this to upstream bug tracker or does package maintainer handle that? Bug tracker in sourceforge.net does not

Processed: Re: Bug#778529: lame: fill_buffer_resample segmentation fault

2015-02-18 Thread Debian Bug Tracking System
Processing control commands: > tags -1 + patch Bug #778529 [lame] lame: fill_buffer_resample segmentation fault Added tag(s) patch. -- 778529: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=778529 Debian Bug Tracking System Contact ow...@bugs.debian.org with problems _

Bug#778529: lame: fill_buffer_resample segmentation fault

2015-02-18 Thread Fabian Greffrath
Control: tags -1 + patch Am Mittwoch, den 18.02.2015, 10:53 +0100 schrieb Fabian Greffrath: > But this is still not the cause of the crash, sigh! Patching the sample > to report 1 channel, it still crashes at the same location. Phew, got it. This time, it was a simple logical error in the lame

Bug#778529: lame: fill_buffer_resample segmentation fault

2015-02-18 Thread Fabian Greffrath
Am Dienstag, den 17.02.2015, 11:19 +0100 schrieb Fabian Greffrath: > But, the sample at hand reports -251 channels. Adding "... || > gfp->num_channels < 0)" to Maks' patch actually fixes the crash. But this is still not the cause of the crash, sigh! Patching the sample to report 1 channel, it sti

Bug#778529: lame: fill_buffer_resample segmentation fault

2015-02-17 Thread Fabian Greffrath
Am Dienstag, den 17.02.2015, 09:11 +0100 schrieb Fabian Greffrath: > Does anyone know the highest reasonable input samplerate supported by > lame? Na, this was a red herring. I just patched a (valid) WAV file to report a samplerate of some 10^6kHz and lame still "successfully" processed on it. B

Bug#778529: lame: fill_buffer_resample segmentation fault

2015-02-17 Thread Fabian Greffrath
Hi Henri, thank you for submitting this bug report! Am Montag, den 16.02.2015, 12:45 +0200 schrieb Henri Salo: > Error reading input file > lame: util.c:595: fill_buffer_resample: Assertion `fabs(offset) <= .501' > failed. This seems to be the inverse problem this time: The samplerate in the f