Re: [PATCH] shrink_dcache_parent() races against shrink_dcache_memory()

[Jan Blunck <jblunck@suse.de>]

On Mon, Jan 30, Kirill Korotaev wrote:

> >>We can think about optimizing this to
> >>  if (!sb->sprunes)
> >>	wake_up(&sb->s_wait_prunes);
> >>
> >
> >
> >Hardly. This is only the case when two or more shrinkers are active in
> >parallel. If that was the case often, we would have seen this much more
> >frequent IMHO.
> But this avoids taking 2nd lock on fast path.
> 

No, the fast path (more frequent) is s_prunes == 0.

sb->s_prunes--;
if (likely(!sb->s_prunes))
   wake_up(&sb->s_wait_prunes);

This is only optimizing a rare case ... and unmounting isn't very time
critical.

Regards,
	Jan

-- 
Jan Blunck                                               jblunck@suse.de
SuSE LINUX AG - A Novell company
Maxfeldstr. 5                                          +49-911-74053-608
D-90409 Nürnberg                                      http://www.suse.de