[FFmpeg-devel] [PATCH] Faster CABAC H.264 residual decoding

[Alexander Strange]

On May 3, 2008, at 5:21 PM, Michael Niedermayer wrote:

> On Sat, May 03, 2008 at 01:30:22PM -0700, Jason Garrett-Glaser wrote:
>> Is this patch ready to be applied?  I haven't seen any objections to
>> the most recent version.
>
> I prefer uint8_t arrays, the speed loss they could cause is purely
> hypothetical while the int arrays will need 100 bytes more precious  
> L1 cache.
>
> Besides this, the patch is ok if its faster on average.
>
> [...]

Here it is with uint8_t.

On Core 2:
old: avg 9.113 max 9.135 min 9.09
new: avg 9.068 max 9.077 min 9.061

It should be faster than with int on any RISC. I'll apply it tomorrow  
unless someone thinks it's still worse on Athlon.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: cabac-transition-uint8.diff
Type: text/x-diff
Size: 2589 bytes
Desc: not available
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20080503/8ec6956f/attachment.diff>
-------------- next part --------------