Am Montag, 30. September 2002 01:45 schrieb Dieter N�tzel:
> Am Montag, 30. September 2002 00:41 schrieb Dieter N�tzel:
> > Am Sonntag, 29. September 2002 23:48 schrieb Felix K�hling:
> > > On Sun, 29 Sep 2002 22:37:36 +0100
> > >
> > > Keith Whitwell <[EMAIL PROTECTED]> wrote:
> > > > Felix K�hling wrote:
> > > > > On Sun, 29 Sep 2002 23:25:03 +0200
> > > > > Dieter N�tzel <[EMAIL PROTECTED]> wrote:
> > >
> > > [snip]
> > >
> > > > >>Is r100/r200 a completely different thing?
> > > > >>If not why not a patch against both?
> > > > >>Then the testing audience should be much "wider".
> > > > >
> > > > > Sure. As far as I could see the code is very similar. However,
> > > > > this: rmesa->do_irqs = (0 &&
> > > > > rmesa->dri.drmMinor >= 6 &&
> > > > > !getenv("R200_NO_IRQS") &&
> > > > > rmesa->r200Screen->irq);
> > > > > looks like IRQs are turned off by default on R200. So my code
> > > > > wouldn't be used. Is the reason for IRQs being disabled that the
> > > > > frame throttling is not implemented properly or are there lower
> > > > > level problems with IRQs?
> > > >
> > > > No, this is a hangover from the bugs last week. It can be removed
> > > > now.
>
> GREAT.
> Even without Felix new stuff coming soon for the r200, CPU load drops from
> 100% (gears took 99%, the other CPU was 100% idle) down to 25% for gears on
> my dual Athlon MP 1900+.
>
> 1:28am up 10 min, 1 user, load average: 0.26, 0.28, 0.18
> 108 processes: 105 sleeping, 3 running, 0 zombie, 0 stopped
> CPU0 states: 8.0% user, 3.0% system, 0.0% nice, 88.3% idle
> CPU1 states: 11.0% user, 3.0% system, 0.0% nice, 85.3% idle
> Mem: 1032728K av, 594820K used, 437908K free, 0K shrd, 311180K
> buff Swap: 1028120K av, 0K used, 1028120K free
> 78272K cached
>
> PID USER PRI NI SIZE RSS SHARE WCHAN STAT %CPU %MEM TIME
> COMMAND
> 3422 nuetzel 15 0 77448 4356 1708 R 24.6 0.4 1:21
> gears 3442 nuetzel 15 0 1448 1448 1212 R 0.5 0.1
> 0:02 top 1 root 15 0 212 212 176 schedule_ S 0.0 0.0
> 0:00 init 2 root 0K 0 0 0 0 migration SW 0.0 0.0
> 0:00 migration_CPU0
> 3 root 0K 0 0 0 0 migration SW 0.0 0.0 0:00
> migration_CPU1
> 4 root 15 0 0 0 0 context_t SW 0.0 0.0 0:00
> keventd
>
>
> gears is a little bit slower
>
> Mesa/demos> ./gears
> r200CreateScreen
> 4000 frames in 5.001 seconds = 799.840 FPS
> 11608 frames in 5.000 seconds = 2321.600 FPS
> 11642 frames in 5.000 seconds = 2328.400 FPS
> 11612 frames in 5.001 seconds = 2321.936 FPS
> 11630 frames in 5.000 seconds = 2326.000 FPS
>
> then with "setenv R200_NO_USLEEPS 1" before
>
> Mesa/demos> ./gears
> r200CreateScreen
> 6465 frames in 5.000 seconds = 1293.000 FPS
> 11955 frames in 5.000 seconds = 2391.000 FPS
> 11954 frames in 5.000 seconds = 2390.800 FPS
> 11955 frames in 5.000 seconds = 2391.000 FPS
> 11954 frames in 5.000 seconds = 2390.800 FPS
Addition:
Q3A only run at full speed (135.6 fps @ 640x480x32) with "setenv
R200_NO_USLEEPS 1".
Without it Q3A is running at 66-76 fps (very shakily).
-Dieter
-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
Dri-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/dri-devel
