[FFmpeg-devel] [PATCH]Avoid overflows when reading pgm with maxval != 255 and 65535

[Michael Niedermayer]

On Sun, Jul 28, 2013 at 10:17:45PM +0000, Carl Eugen Hoyos wrote:
> Michael Niedermayer <michaelni <at> gmx.at> writes:
> 
> > > libjpeg v6b can write 12bit pgm files that show 
> > > white noise with ImageMagick and gimp, FFmpeg 
> > > shows recognizable but very broken images.
> 
> > > The reason is that the bits above maxval (4095) 
> > > are not initialized.
> 
> The actual reason is that libjpeg v6b writes little 
> endian 16bit pgm files, the specification requires 
> big endian values.

i have realized this as well once i saw the commit fixing it in
libjpeg
this btw is easy to detect if we want to do it ...


> 
> > > Attached patch is only tested for the specific 
> > > 12bit gray16 case.
> 
> > please upload the pmn file this patch is about
> 
> Uploaded to ticket #2812.

thanks

[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Complexity theory is the science of finding the exact solution to an
approximation. Benchmarking OTOH is finding an approximation of the exact
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20130729/dc8c0a26/attachment.asc>