Messages in this thread | | | Date | Mon, 23 Nov 1998 16:29:51 +0100 (CET) | From | Rik van Riel <> | Subject | Re: Scheduling: interactive problems still there in 129 |
| |
On Fri, 20 Nov 1998, Neil Conway wrote:
> This has been mentioned by many people so far (I seem to recall even > seeing patches) but is still there in 129.
[SNIP]
> Now on a 2.1.129 SMP machine (2 cpus) I run two copies of the same > CPU hog. Now my key-holding down gets lots of pauses, some short, > some quite long (like the odd one is 100's of milliseconds) > (measured by eye :-)). > > when I run say 3 CPU hogs on a 2 CPU machine, each of which needs > say 1 minute of CPU time, and I start them maybe 2 seconds apart, I > get the weird effect that the first one to start gets CONSISTENTLY > more CPU time than the others
I guess I should fix reschedule_idle() to fix this. I'll get working on this ASAP and I even have an idea on how to fix the problem relatively inexpensively.
Note the /relatively/, since cache locality is quite important when we want to achieve scalable SMP. To properly implement things, we also need to take in account things like the time the program left the run queue, so I'll probably fix things in the context of my scheduler_bigpatch...
> I presume this implies a sorting problem in the scheduler code ?
The current version of my scheduler_bigpatch doesn't fix the reschedule_idle() thing, but it does contain some other (more generic) fixes that should help you with your problem.
cheers,
Rik -- slowly getting used to dvorak kbd layout... +-------------------------------------------------------------------+ | 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/
| |