Re: [patch 1/6] x86, ftrace, hw-branch-tracer: support hotplug cpus

[Nathan Lynch]

On Tue, 24 Feb 2009 20:36:22 +0100
Ingo Molnar <mingo@xxxxxxx> wrote:

> 
> * Markus Metzger <markus.t.metzger@xxxxxxxxx> wrote:
> 
> > Support hotplug cpus.
> 
> > +static int __cpuinit bts_hotcpu_handler(struct notifier_block *nfb,
> > +				     unsigned long action, void *hcpu)
> > +{
> > +	unsigned int cpu = (unsigned long)hcpu;
> > +
> > +	mutex_lock(&bts_tracer_mutex);
> 
> This is buggy: CPU hotplug handlers must not sleep (they are 
> called with irqs disabled)

I don't think that is the case; the cpu notifier chains are "raw", not
atomic, and the cpu hotplug core doesn't disable irqs while processing
the chains afaict.  And there are several current examples of cpu
hotplug callbacks performing sleeping operations (slab and slub,
page_alloc, probably many others).

--
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/