Re: [FFmpeg-devel] Fix for aviobuf.c::fill_buffer

2016-07-21 Thread code bythepound
x); /* note: the internal buffer could have changed, and be != avio_ctx_buffer */ On Thu, Jul 21, 2016 at 11:20 AM code bythepound wrote: > I will see if I can put together a test case that exposes the problem. > The comment for this block says "make buffer smaller in case it

Re: [FFmpeg-devel] Fix for aviobuf.c::fill_buffer

2016-07-21 Thread code bythepound
e assert doesn't seem like the right way to handle this here. On Wed, Jul 20, 2016 at 4:41 PM Michael Niedermayer wrote: > On Wed, Jul 20, 2016 at 03:59:39PM +, code bythepound wrote: > > My app seems to hit the av_assert0(len >= s->orig_buffer_size) (line 535 > in &g

[FFmpeg-devel] Fix for aviobuf.c::fill_buffer

2016-07-20 Thread code bythepound
My app seems to hit the av_assert0(len >= s->orig_buffer_size) (line 535 in my sources) in aviobuf.c::fill_buffer. FWIW I have registered a custom AVIOContext that will consistently not fill the entire buffer when the read callback is called. The outcome is that 'len' in fill_buffer is decremente