Re: [BUGFIX] cpufreq: dont BUG_ON uninitialized policy cpu

[Dave Jones]

On Wed, Jan 13, 2010 at 01:14:38PM +0800, Wu Fengguang wrote:
 > Remove a BUG_ON which is always triggered here after commit 0b19a310eebb93:
 > 
 > --- linux-mm.orig/drivers/cpufreq/cpufreq.c	2010-01-13 13:06:03.000000000 +0800
 > +++ linux-mm/drivers/cpufreq/cpufreq.c	2010-01-13 13:06:57.000000000 +0800
 > @@ -72,7 +72,8 @@ int lock_policy_rwsem_##mode						\
 >  (int cpu)								\
 >  {									\
 >  	int policy_cpu = per_cpu(cpufreq_policy_cpu, cpu);		\
 > -	BUG_ON(policy_cpu == -1);					\
 > +	if (policy_cpu == -1)						\
 > +		return -1;						\
 >  	down_##mode(&per_cpu(cpu_policy_rwsem, policy_cpu));		\
 >  	if (unlikely(!cpu_online(cpu))) {				\
 >  		up_##mode(&per_cpu(cpu_policy_rwsem, policy_cpu));	\

So instead of always BUG'ing, we'll always fail to get the lock ?
How can this be correct ?

This makes no sense. cpufreq_update_policy shouldn't be called without
an existing policy.   Can someone send me the full trace ?

Andrew cc'd me on a revert of Thomas's last patch to this file,
but again, no trace, and I can't find the mails he refers to.

	Dave

--
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/