Re: scheduler interactivity: timeslice calculation seem wrong

[Mike Fedyk]

On Tue, Aug 19, 2003 at 12:07:13AM -0400, Eric St-Laurent wrote:
> - a /proc tunable might be needed to let the user choose the
> interactivity vs efficiency compromise. for example, with 100 background
> tasks, does the user want the most efficiency or prefer wasting some
> cycles to get a smoother progress between jobs?
> 

No, this should be avoided.  Then you get a bunch of web pages talking about
how this tunable should be set this way, and the problem cases will not be
reported to the people who can fix them.  This is good.

> - nanoseconds, or better, cycle accurate (rdtsc) timeslice accouting, it
> may help heuristics.
> 

Already done.  See the scheduler patch Ingo submitted a few weeks ago.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/