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

Corey Hickey bugfood-ml at fatooh.org
Mon Sep 18 07:18:39 CEST 2006


Carl Eugen Hoyos wrote:
> Hi!
> 
> Attached patch improves H264 decoding speed on my Athlon XP 2600+ by ~3%.
> Tested with gcc 4.0.2 and 4.0.3.

I can't quite figure out what file everybody else is using for the
benchmarks, so I used my current "can't-quite-play-it" file:

http://images.apple.com/movies/us/hd_gallery/gl1800/1080p/bbc-japan_1080p.mov

All the times listed are user times for:
$ time mplayer bbc-japan_1080p.mov -vo null -nosound -benchmark -quiet

Athlon64 3400+ at 2556 MHz
Linux 2.6.17-beyond1 (64-bit)
Debian (also 64-bit)

Each mplayer is compiled with -O4--the labels below refer only to
dsputil_mmx.c as per the original post.

------------------------------------------------------------------------------
gcc-3.3 (GCC) 3.3.6 (Debian 1:3.3.6-13)

---  -O4   ---
user    3m8.006s
user    3m8.001s
user    3m8.271s

---  -O2   ---
user    3m7.687s
user    3m7.075s
user    3m6.753s


------------------------------------------------------------------------------
gcc-3.4 (GCC) 3.4.6 (Debian 3.4.6-4)

---  -O4   ---
user    2m59.716s
user    2m59.123s
user    2m58.791s

---  -O2   ---
user    2m57.868s
user    2m59.303s
user    2m56.339s


------------------------------------------------------------------------------
gcc-4.0 (GCC) 4.0.4 20060904 (prerelease) (Debian 4.0.3-7)

---  -O4   ---
user    3m3.537s
user    3m1.586s
user    3m0.826s

---  -O2   ---
user    2m57.618s
user    2m58.628s
user    2m58.142s

------------------------------------------------------------------------------
gcc-4.1 (GCC) 4.1.2 20060901 (prerelease) (Debian 4.1.1-13)

---  -O4   ---
user    3m20.476s
user    3m21.011s
user    3m24.656s

---  -Os   ---
user    3m3.894s
user    3m4.803s
user    3m3.673s

---  -O2   ---
user    2m57.453s
user    2m57.820s
user    2m58.419s



------------------------------------------------------------------------------
I also tried gcc-4.1 using -O2 for the whole compile to see if there was
any detriment overall. There doesn't seem to be.

user    2m56.588s
user    2m56.852s
user    2m57.408s


-Corey



More information about the MPlayer-dev-eng mailing list