> Hi, > > On 04/06/06 00:51, Alexander Kolbasov wrote: > > Hello, > > > > here is you chance to bite a piece of the scheduler code. > > > > The purpose of the fix is to make idle thread less less aggressive in > > stealin > > work from other CPUs when the work was only placed on the run-queue > > recently. > > > > The diffs are available at > > > > http://cr.grommit.com/~akolb/sticky/index.html > > The changes look solid. > > I have one question. It looks like disp_getwork will traverse the > other cpus (relative to the argument) in a deterministic order > for a given cpu and lgroup configuration (I may be wrong there - > it's not obvious!). So is there a chance that cpu A in the idle > loop might keep failing to find work to do if it keeps getting > "don't steal" back from disp_getbest for some cpu which keeps > turning up in the disp_getwork loop as having the highest > unbound pri of all. This may, I think, happen if some cpu > has say an FX class thread of high fixed priority that > keeps running and then blocking for a very short time. > Maybe a pathalogical worry!
I updated the webrev. In the new code disp_getwork() tries to find another stealable thread and don't just stop if it finds the first one that can't be stolen. I think that thi sshould address your concern. The new webrev is at http://cr.grommit.com/~akolb/sticky/index.html The original webrev is available at http://cr.grommit.com/~akolb/sticky.save - Alexander Kolbasov _______________________________________________ perf-discuss mailing list perf-discuss@opensolaris.org
