Re: [patch] voluntary-preempt-2.6.8.1-P0

[Ingo Molnar]

* Lee Revell <rlrevell@xxxxxxxxxxx> wrote:

> > > What do you think of Ingo's solution of trying to move the problematic
> > > call to mdio_read out of the spinlocked section?  It does seem that the
> > 
> > Can't comment on that, I missed it. I am aware that locking in via-rhine
> > needs work, though, it's one of the things I haven't touched.
> > 
> > > awfully long time.  In a live audio setting you would actually get lots
> > > of media events.
> > 
> > Don't trip over the network cables. Duh.
> > 
> 
> You might want to intentionally plug or unplug them.  Live music != a
> server room.  Think laptop DJs.  It would be bad if plugging into the
> network caused a click in your sound output - this could be VERY loud
> depnding on the setting!

right now the mdio_read() latency causes a latency of 140 usecs which
isnt that bad. But looking at it in isolation, mdio_read() is 70 usecs a
pop which is excessive for a kernel function.

	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/