[MPlayer-dev-eng] Re: dsputil_mmx.c: -O2

[Carl Eugen Hoyos]

Bonsoir Hervé,

Hervé W. <H.O.W.aka.V+mplayer <at> gmail.com> writes:

> On 17/09/06, Carl Eugen Hoyos wrote:
> > Attached patch improves H264 decoding speed on my Athlon XP 2600+ by ~3%.
> > Tested with gcc 4.0.2 and 4.0.3.
> 
> Perhaps some tests with gcc 4.1.1, gcc 3.4.6 and 2.95.3 are in order?

Not much difference for 2.95.3 and 3.4.6.

I double checked the numbers for 4.1.1.

4.1.1:
without my patch (best results in >10 tests):
BENCHMARKs: VC:  64.651s VO:  17.885s A:   0.000s Sys:   2.112s =   84.648s
BENCHMARKs: VC:  64.659s VO:  17.850s A:   0.000s Sys:   2.100s =   84.609s
with my patch - -O2 for dsputil_mmx.c (worst results!):
BENCHMARKs: VC:  57.822s VO:  18.448s A:   0.000s Sys:   2.181s =   78.452s
BENCHMARKs: VC:  57.824s VO:  18.608s A:   0.000s Sys:   2.109s =   78.541s

3.4.6:
without my patch:
BENCHMARKs: VC:  57.439s VO:  17.993s A:   0.000s Sys:   2.517s =   77.948s
BENCHMARKs: VC:  57.138s VO:  17.943s A:   0.000s Sys:   2.160s =   77.242s
with my patch - -O2 for dsputil_mmx.c:
BENCHMARKs: VC:  56.687s VO:  17.948s A:   0.000s Sys:   2.181s =   76.816s
BENCHMARKs: VC:  56.795s VO:  18.129s A:   0.000s Sys:   2.114s =   77.038s

2.95.3:
without my patch:
BENCHMARKs: VC:  66.599s VO:  17.656s A:   0.000s Sys:   2.108s =   86.362s
BENCHMARKs: VC:  66.591s VO:  17.776s A:   0.000s Sys:   2.082s =   86.450s
with my patch - -O2 for dsputil_mmx.c:
BENCHMARKs: VC:  66.062s VO:  17.688s A:   0.000s Sys:   2.136s =   85.887s
BENCHMARKs: VC:  65.648s VO:  17.680s A:   0.000s Sys:   2.089s =   85.417s

> > Index: libavcodec/Makefile
> 
> libavcodec? shouldn't this patch go to ffmpeg-devel instead of 
> mplayer-dev-eng?

Should I send the patch again there?

Carl Eugen