Messages in this thread |  | | Date | Thu, 8 Nov 2001 18:09:29 -0800 (PST) | From | Davide Libenzi <> | Subject | Re: [patch] scheduler cache affinity improvement for 2.4 kernels |
| |
On Thu, 8 Nov 2001, Mike Fedyk wrote:
> > The MQ scheduler has the same roots of the proposed one but has a longest > > fast path due the try to make global scheduling decisions at every > > schedule. > > Ahh, so that's why it hasn't been adopted...
Changing the scheduler is not easy ( not to code patches but to make everyone agree on the need of changing it ) and as i already said, it's easier to force my cat to have a bath instead of Linus to change the scheduler :)
> > I'm in contact ( close contact coz we're both in Beaverton :) ) with IBM > > guys to have the two scheduler tested on bigger machines if the proposed > > scheduler will give some fruit. > > > > >From what I've seen, it probably will... > > I hope something like this will go into 2.5... > > What do other unixes do in this case? Are there any commercial Unixes that > have loose affinity like linux currently does? What about NT?
I can't say about NT. I've tried a "cvs checkout" from cvs.microsoft.com but the running server ( Nimda-CVS running on port 2401 ) said me that, although I've full read/write access on the repository, the server is busy scanning port 80s :)
- Davide
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
|  |