Re: [FFmpeg-devel] Add Mediacodec audio decoders support

[Anton Khirnov <anton@khirnov.net>]

Quoting Cosmin Stejerean via ffmpeg-devel (2024-07-16 22:14:19)
> 
> 
> > On Jul 16, 2024, at 8:24 PM, Rémi Denis-Courmont <remi@remlab.net> wrote:
> > 
> > Le tiistaina 16. heinäkuuta 2024, 18.48.06 EEST Cosmin Stejerean via ffmpeg-
> > devel a écrit :
> >> To add another data point, the platform decoders might also be more secure
> >> due to sandboxing. I believe as of Android Q the software decoders provided
> >> by MediaCodec have been moved to run within a constrained sandbox.
> > 
> > Platform decoders are in all likelihood strictly less secure than software 
> > decoders. Software decoders will run in a user-space sandboxed within their 
> > respective application. Platform decoders will run in a more privileged system 
> > service, with direct access to a kernel driver in EL1, through that to the 
> > firmware running on the video DSP.
> > 
> > More performant and energy-efficient. But also way way less secure.
> > 
> > The only viewpoint whence this is more secure, is the content publisher's: 
> > this model enables DRM with hardware pass-through (but that does not even 
> > apply if you use FFmpeg as the front end).
> > 
> 
> Platform provided *software* decoders should be more secure than bundled software decoders due to the sandboxing of software decoders in recent versions of Android.

If that is such an important feature to someone then it is not
inconceivable to implement some sort of sandboxing inside avcodec.

I'm not a big fan of the argument "we should provide passthrough to
proprietary decoders because they are more secure".

-- 
Anton Khirnov
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".