Re: SMP Panic caused by [PATCH] sched: consolidate sched domains

[Nathan Lynch]

On Sun, 2004-08-29 at 12:24, James Bottomley wrote:
> On Sun, 2004-08-29 at 13:07, Jesse Barnes wrote:
> > I've up and downed a few CPUs on an Altix, and it seems to work ok, but that's 
> > a pretty basic test.  How about this?
> 
> Incidentally, down and up tests won't pick up these initialisation
> problems because the SMP paths will already have created the start of
> day data structures for these CPUs.  You need to boot with the CPU down
> and bring it up after boot to see the issues.

I've got a patch which reinitializes sched domains at cpu hotplug time. 
We need something like this on ppc64 for partitioned systems (we run
into the same issue when adding a cpu which wasn't present at boot).  I
had been waiting to post it until some cpu hotplug issues with preempt
were solved, but it seems it would help the case of hotplugging
secondary cpus at boot, so I'll submit that soon.

Nathan


-
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/