Messages in this thread | | | Date | Tue, 29 Sep 1998 00:14:47 +0200 (CEST) | From | Rik van Riel <> | Subject | Re: [PATCH] scheduler patch, faster still |
| |
On Mon, 28 Sep 1998, MOLNAR Ingo wrote: > On Mon, 28 Sep 1998 yodaiken@chelm.cs.nmt.edu wrote: > > > Is it? And, isn't Richard arguing that his change makes sense only for > > long queues? > > aggreed. 'long runqueues' are definitely a red herring. Very big (Linux) > servers and even artificial 'exploit' programs could not show any RL case > where there are long runqueues. Eg. someone at Kiva (now they are > Netscape) has asked the same questions and has run statistics on a very > big webserver and the average runqueue length was around 3, maxing out at > 15.
OK. This is a valid point against Richard's proposed changes.
> Long runqueues might be a problem on other OSs, but not on Linux. (or if > yes, please show me the case) > (barring one case, when there are alot of CPU-bound processes around, in > which case those processes waste so much cache resources anyway)
That's OK, since they still run their full timeslice before having their priority restored. Two CPU-bound tasks thrash the cache just as much ast ten.
The problem with this is that interactive processes also get their priority restored slower when there are a lot of CPU-bound processes running. It seems like that is fixed with my patch (which should be a little more efficient, but I'd really like to keep the slow-recharging for niced processes).
Knowing you, I think you might actually be interested to take a look at my scheduler changes and make it more efficient (it avoids the for_each_task() and adds the cost in another place; it could probably be a little faster).
Rik. +-------------------------------------------------------------------+ | Linux memory management tour guide. H.H.vanRiel@phys.uu.nl | | Scouting Vries cubscout leader. http://www.phys.uu.nl/~riel/ | +-------------------------------------------------------------------+
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |