[FFmpeg-devel] [PATCH] Huffyuv speed tweaks

[Vitor Sessak]

Jason Garrett-Glaser wrote:
>> When compiling add_median_predition() as a non-inline function with gcc
>> 4.2.3 this optimization gives in the same code as the original function.
>>  Are you sure it is not still the case as an inline function?
> 
> That's rather odd.  Have you tried forcing the inline to see if it changes?

Indeed, forcing the inlining makes GCC produce different code. It is 
really sad that gcc can't optimize that, since block like

for (i=0; i < len; i++) { do something to buf1[i], buf2[i], ..., etc }

are really common.

-Vitor