Messages in this thread | | | Date | Fri, 10 Oct 2008 09:05:30 -0600 | From | "Chris Friesen" <> | Subject | Re: cgroup task groups appears sensitive to absolute magnitude of shares |
| |
Peter Zijlstra wrote: > On Fri, 2008-10-10 at 08:03 +0200, Peter Zijlstra wrote: >>On Thu, 2008-10-09 at 17:00 -0600, Chris Friesen wrote:
>>>Going to 4000/2000/1000/200 doesn't seem to give noticeable >>>improvements, and going to 40000/20000/10000/2000 causes the test to >>>behave unpredictably, either taking abnormally long to complete or else >>>not completing at all. >> >>Hmm, I would have expected it to work for at least the normal nice range >>of weight values.. >> >>Will have to look into that I suppose..
> One thought, are you running 32 or 64 bit? If you're on 32, could you > try 64? If that fixes it its probably an easy fix, otherwise I fear it > might take a bit more..
No such luck, this is 64-bit (powerpc though, not x86). I tried reproducing it again today and was unable to reproduce the hang, but saw delays of up to 7 seconds (where the expected delay is 10 sec).
Chris
| |