Re: BUG_ON(!cpus_equal(cpumask, tmp));

[Srivatsa Vaddagiri <vatsa@in.ibm.com>]

On Thu, Apr 01, 2004 at 01:31:15AM +0000, Andy Whitcroft wrote:
> 	spin_lock(&tlbstate_lock);
> +
> +	/* Subtle, mask the request mask with the currently online cpu's.
> +	 * Sample this under the lock; cpus in the the middle of going
> +	 * offline will wait until there is noone in this critical section
> +	 * before disabling IPI handling. */
> +	cpus_and(tmp, cpumask, cpu_online_map);
> +	if(cpus_empty(tmp))
> +		return;

Hmm ..Doesn't it need to drop tlbstate_lock before returning?

> +	/* Subtle, IPI users assume that they will be able to get IPI's
> +	 * though to the cpus listed in cpu_online_map.  To ensure this
> +	 * we add the requirement that they check cpu_online_map within
> +	 * the IPI critical sections.  Here we remove ourselves from the
> +	 * map, then ensure that all other cpus have left the relevant
> +	 * critical sections since the change.  We do this by aquiring
> +	 * the relevant section locks, if we have them none else is in
> +	 * them.  Once this is done we can go offline. */
> +	spin_lock(&tlbstate_lock);
> +	spin_unlock(&tlbstate_lock);
> +	spin_lock(&tlbstate_lock);
> +	spin_unlock(&tlbstate_lock);

The second lock should be call_lock?


-- 


Thanks and Regards,
Srivatsa Vaddagiri,
Linux Technology Center,
IBM Software Labs,
Bangalore, INDIA - 560017