Re: [git pull] scheduler fixes

[Ingo Molnar]

* Alexey Zaytsev <alexey.zaytsev@xxxxxxxxx> wrote:

> On Mon, Feb 2, 2009 at 12:52, Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote:
> > On Sat, 2009-01-31 at 20:49 +0300, Alexey Zaytsev wrote:
> >> And answering an earlier question, this happens only on i386 and only
> >> with 4K stacks because x86_64 dosn't have a separate softirq stack,
> >> so the preempt count diring the soft irq is at least IRQ_EXIT_OFFSET.
> >
> > Which suggests that an initial preempt_count offset of 1 should suffice
> > to fix this issue and not confuse the softirq tracing, hmm?
> 
> Can't we just copy the preempt count when switching stacks?

that always has a dynamic cost and the thing is that after-switch we are 
fine with a constant count.

	Ingo
--
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/