kernel 2.6.9-rc2-mm2 vs glxgears

Gene Heskett gene.heskett at
Tue Sep 28 20:56:15 PDT 2004

On Tuesday 28 September 2004 18:21, Felix Kühling wrote:
>On Tue, 28 Sep 2004 15:34:36 -0400
>Gene Heskett <gene.heskett at> wrote:
>> On Thursday 23 September 2004 23:39, Gene Heskett wrote:
>> >Greetings;
>> >
>> >Also add a wish that AGP 8x was supported on nforce2 equipt
>> > mobo's, but thats not why glxgears, which in previous kernels
>> > would start out at 350fps but drop to about 100fps in about a
>> > minutes running, is now running at a consistent 10fps effective
>> > the 2.6.9-rc2-mm2 kernel.
>> >
>> >Anyone have any ideas why?
>> And this continues to be the case for the mm3 and mm4 kernels.
>> Some testing here indicates that if I replace *all* the glx stuff
>> with that from MesaLibs, I get around 550 fps, but with 100% cpu
>> going to glxgears.  glxinfo says drm = no
>Software rendering on a fast CPU.
>> With the 6.8.1 glx stuff installed, glxinfo says drm - yes, and I
>> get 10 fps at 0% cpu, even at full screen, something the MesaLibs
>> stuff can't handle at all.  Other screen actions do seem to be
>> quite snappy even if glxgears is a crippled dog.
>Sounds like it could be a problem with frame throttling. My guess is
>that interrupts get delayed somehow. The driver uses them to wait
> for previous frames to finish rendering. You can try different
> settings that don't use interrupts using the environment variable
> fthrottle_mode, e.g.
>fthrottle_mode=0 glxgears (busy waiting)
>fthrottle_mode=1 glxgears (usleeps)
>fthrottle_mode=2 glxgears (interrupts, the default)

Humm, interesting results:
[root at coyote linux-2.6.9-rc2]# fthrottle_mode=0 glxgears
ATTENTION: default value of option fthrottle_mode overridden by 
IRQ's not enabled, falling back to busy waits: 11 0 5
826 frames in 5.0 seconds = 165.200 FPS
992 frames in 5.0 seconds = 198.400 FPS
991 frames in 5.0 seconds = 198.200 FPS
821 frames in 5.0 seconds = 164.200 FPS

[root at coyote linux-2.6.9-rc2]# fthrottle_mode=1 glxgears
ATTENTION: default value of option fthrottle_mode overridden by 
IRQ's not enabled, falling back to usleeps: 11 1 5
823 frames in 5.0 seconds = 164.600 FPS
984 frames in 5.0 seconds = 196.800 FPS

[root at coyote linux-2.6.9-rc2]# fthrottle_mode=2 glxgears
ATTENTION: default value of option fthrottle_mode overridden by 
834 frames in 5.0 seconds = 166.800 FPS
995 frames in 5.0 seconds = 199.000 FPS

But a cat /proc/interrupts:
  0:    1824349          XT-PIC  timer
  1:       6423          XT-PIC  i8042
  2:          0          XT-PIC  cascade
  3:         20          XT-PIC  serial
  4:       8137          XT-PIC  serial
  5:     111753          XT-PIC  ehci_hcd, radeon at pci:0000:02:00.0
  7:          0          XT-PIC  parport0
  8:          2          XT-PIC  rtc
 11:     181971          XT-PIC  ohci_hcd, eth0, Bt87x audio
 12:      14652          XT-PIC  ohci_hcd, NVidia nForce2
 14:      14736          XT-PIC  ide0
 15:       2670          XT-PIC  ide1
NMI:          0
ERR:          0

>The first one is the best one, performance wise. But it'll result in
>100% CPU usage by glxgears.

It did not. Very low cpu usage.

>The second one used to throttle the 
> frame rate to 100 or 50 on Linux 2.4. Linux 2.6 has a higher timer
> resolution so it may not be that bad. Interrupts guarantee
> (usually) good performance while keeping the CPU usage down.
>If it's really related to frame throttling with interrupts, then it
>could have different reasons. Either interrupts really get delayed
> and the interrupt handler is executed too late. It could also be
> that the interrupt handler receives the interrupts promptly but
> waking up the waiting applications takes some time, maybe a flaw in
> the scheduler of those experimental kernels you're using.
>> Noting John Smirls announce on lkml just now, I wonder if I should
>> be a tester of his new drm?
>  Felix
>| Felix Kühling <fxkuehl at>                    
>| | PGP Fingerprint: 6A3C 9566 5B30 DDED 73C3  B152
>| 151C 5CC1 D888 E595 |

Cheers, Gene
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
99.26% setiathome rank, not too shabby for a WV hillbilly attorneys please note, additions to this message
by Gene Heskett are:
Copyright 2004 by Maurice Eugene Heskett, all rights reserved.

More information about the xorg mailing list