Re: [PATCH] mm/dmapool.c: Do not create/destroy sysfs file whileholding pools_lock

[Shaun Ruffell]

On Tue, Mar 01, 2011 at 08:35:53PM -0800, Eric W. Biederman wrote:
> Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> writes:
> > One way of fixing this would be to create another singleton lock:
> >
> >
> > 	{
> > 		static DEFINE_MUTEX(pools_sysfs_lock);
> > 		static bool pools_sysfs_done;
> >
> > 		mutex_lock(&pools_sysfs_lock);
> > 		if (pools_sysfs_done == false) {
> > 			create_sysfs_stuff();
> > 			pools_sysfs_done = true;
> > 		}
> > 		mutex_unlock(&pools_sysfs_lock);
> > 	}
> >
> > That's not terribly pretty.
> 
> Or possibly use module_init style magic.  Where use module
> initialization and remove to trigger creation and deletion of the sysfs.
> 

I'm not following how module initialization can help here. Are you suggesting
that all devices get a 'pools' attribute regardless of whether any dma pools
are actually created?

Shaun
--
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/