[FFmpeg-devel] [PATCH] address speed regression in h264 deblocking filter after PAFF

[Uoti Urpala]

On Wed, 2007-11-07 at 11:47 +0100, Andreas ?man wrote:
> The speed increase is a bit more than I initially expected (2%),
> and decoding of the SVT trailer is almost 6% faster !? (this is
> on a core2)
> I've double checked the numbers and md5sums matches.
> 
> If someone else can perform some checks on various content I'd be happy.

In my tests your patch made decoding consistently slower. I tested a
CABAC sample both with and without the patches I posted earlier. In
every case adding your patch made decoding about 1.5% slower.

Additionally I tested a CAVLC sample only in addition to my patches
(including the one disabling interlace support). Adding your patch made
decoding about 8% slower.

The tests were run on athlon-xp with a recent gcc 4.2 version from
Debian unstable.