On Wed, Apr 13, 2005 at 10:13:46PM -0500, Richard Rauch wrote:
> The short answer is: Yes, I have been seeing the same thing.
 [...]
> The best way that I can summarize this bug is to say: Once in a
> while (about once every 15 seconds in the case I have), a small
> number of randomly-chosen processes (perhaps just 1) will be refused
> access to the CPU for a "long" period of time (in computer's terms).

Oh, BTW, it's interesting that you observe ~30 seconds and ~2 seconds
where I see ~15 and ~1.  I timed my system in bzflag by a slow-count.
This was in bzflag (no disk I/O, some network I/O; lots of OpenGL and
X work).  The computer is an AMD64 at approx. 2GH (a "3200" Athlon 64)
with 512MB of RAM.  I was playing in a very low-resolution mode, though
subjectively I don't think that that affects this issue; I played at
the low resolution (about 320x240?) so that the software rendering of
Mesa could give a decent frame rate.  (^&

My kernel is fairly close to GENERIC.  The motherboard uses an nVidia
nForce chipset.

I've tried fiddling with buffer controls and saw little or no change
to the problem.


-- 
  "I probably don't know what I'm talking about."  http://www.olib.org/~rkr/