Re: rt20 scheduling latency testcase and failure data

[Steven Rostedt]

On Sat, 13 May 2006, Mike Galbraith wrote:

> On Sat, 2006-05-13 at 11:20 +0200, Florian Paul Schmidt wrote:
>
> > P.S.: I ran the test a few [20 or so] times and didn't get any failures
> > of the sort you see. Even with a 1ms period:
>
> Something odd happened here... the first time I booted rt21, I could
> reproduce the problem quite regularly.  Since reboot though, poof.
>
> Elves and Gremlins.
>

Careful, rt21 has a bug slipped in that might have funny results on SMP
machines:

+		if (!cpus_equal(current->cpus_allowed, irq_affinity[irq]));
+			set_cpus_allowed(current, irq_affinity[irq]);

John (although he later fixed it) added a ; after the if.  But the fix is
not yet in Ingo's patch.

-- Steve
-
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/