Re: [PATCH 6/7] work_on_cpu: use on x86/kernel/cpu/intel_cacheinfo.c

[Rusty Russell <rusty@rustcorp.com.au>]

On Thursday 23 October 2008 20:46:21 Oleg Nesterov wrote:
> On 10/23, Rusty Russell wrote:
> > +static int __cpuinit detect_cache_attributes(unsigned int cpu)
> > +{
> > +	int			retval;
> > +
> > +	if (num_cache_leaves == 0)
> > +		return -ENOENT;
> > +
> > +	per_cpu(cpuid4_info, cpu) = kzalloc(
> > +	    sizeof(struct _cpuid4_info) * num_cache_leaves, GFP_KERNEL);
> > +	if (per_cpu(cpuid4_info, cpu) == NULL)
> > +		return -ENOMEM;
> > +
> > +	retval = work_on_cpu(cpu, get_cpu_leaves, NULL);
>
> This doesn't look right.
>
> cache_add_dev()->cpuid4_cache_sysfs_init()->detect_cache_attributes()
> is called by CPU_ONLINE under cpu_hotplug_begin(), this is deadlockable.
>
> Yes, the caller (cpu_hotplug.active_writer) can safely take
> get_online_cpus(), but if we have another pending work which needs
> get_online_cpus(), we have the deadlock.

Hi Oleg,

   Nice catch.  Can we use smp_call_function here instead?

Rusty.

>
> Oleg.