[theora] Optimised ARM Ogg/Theora/Vorbis decoder

[mail at stefan-guertler.de]

Zitat von Robin Watts <theora at wss.co.uk>:

> In message <4AB79AB0.7080500 at mozilla.com>
>           Christopher Blizzard <blizzard at mozilla.com> wrote:
> > On 9/19/2009 6:19 AM, Robin Watts wrote:
> > > I've started to revisit the hotspots, rewriting the code to use
> > > ARM v6 SIMD extensions, but I am unable to test it due to not
> > > having an appropriate device. If anyone out there has a suitable
> > > device and would be prepared to test my code on it, please get in
> > > touch!
> > >

May be you should contact some of the hackers of the ffmpeg project, which have
ARM-based systems. Maybe they will grant you a ssh shell on those machines, if
the heare about your optimization efforts on the ARM platform. Especial Måns
Rullgard (http://hardwarebug.org/category/arm/) is very active
(http://x264dev.multimedia.cx/?p=142). Another one is Kostya Shishkov
(http://codecs.multimedia.cx/?p=216), but maybe he will have dificulties with
the internet connection.

> > > I hope to look at using v7 or NEON instructions as appropriate
> > > too.
> > >
> > > Thanks,
> > Would a beagleboard help with this?  I'm told it has video out and
> > I  think it runs with newer ARM cores.
>
> Yes, a beagleboard would be perfect - but I've been putting off
> buying one, as I need to buy a monitor to go with it (none of mine
> are DVI compatible). I think I might bite the bullet and spend the
> cash in the morning though :)
>
> I've got ARMv6 SIMD working for the fragment reconstruction (running
> under an emulation environment), so hopefully progress won't be
> completely blocked until it arrives.
>
Stefan Gürtler