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

Loren Merritt lorenm
Mon Apr 28 07:21:19 CEST 2008

On Sun, 27 Apr 2008, Nicolas George wrote:
> Le nonidi 9 flor?al, an CCXVI, matthieu castet a ?crit?:
>>> On the advice of #ffmpeg-devel I have made a version with uint8_t
>>> arrays instead of int.
>> Don't forget that some cpu (arm for example) don't have native 8 bits
>> operation. Everything is done in 32 bits, and 8 bits behavior is
>> emulated with extra operation.
> I think it may be worth reminding the existence if uint_fast8_t, wihci is
> the fastest integer type of size at least 8.

Fastest for what? e.g. int_fast{16,32}_t are 64bit on x86_64, which makes 
them faster for signed pointer math, but slower for non-trivial arithmetic 
(mul, div, etc). Afaict, this could be remedied only if int_fastX_t were 
variable attributes rather than types.

--Loren Merritt

More information about the ffmpeg-devel mailing list