On 08 Nov 1998 10:48:16 +0100, Andi Kleen <ak@muc.de> said:
> - The scheduler is not fair enough. When I have a non-niced cpu hog
> running on both CPUs all my xterms become very unresponsible. This
> didn't happen in 2.0, it is a 2.1 regression. When I nice the hogs it
> gets better, but I think the kernel should be fairer to interactive
> processes even without user help.
If a wake_up_process() is delivered to a process which last ran on a
different CPU, there's a good chance that the scheduler on the current
CPU will just ignore the reschedule request due to the processor
affinity penalty. The wake_up is essentially lost until the other CPU
finishes it's current scheduling timeslice or finds another excuse to
run the scheduler.
At one point Ingo was talking about new code to send a cross-CPU
reschedule request at times like this: Ingo, has there been any update
on this?
--Stephen
-
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/