lkml.org 
[lkml]   [2016]   [Apr]   [13]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH v2 01/11] mm/slab: fix the theoretical race by holding proper lock
On Tue, Apr 12, 2016 at 11:38:39AM -0500, Christoph Lameter wrote:
> On Tue, 12 Apr 2016, js1304@gmail.com wrote:
>
> > @@ -2222,6 +2241,7 @@ static void drain_cpu_caches(struct kmem_cache *cachep)
> > {
> > struct kmem_cache_node *n;
> > int node;
> > + LIST_HEAD(list);
> >
> > on_each_cpu(do_drain, cachep, 1);
> > check_irq_on();
> > @@ -2229,8 +2249,13 @@ static void drain_cpu_caches(struct kmem_cache *cachep)
> > if (n->alien)
> > drain_alien_cache(cachep, n->alien);
> >
> > - for_each_kmem_cache_node(cachep, node, n)
> > - drain_array(cachep, n, n->shared, 1, node);
> > + for_each_kmem_cache_node(cachep, node, n) {
> > + spin_lock_irq(&n->list_lock);
> > + drain_array_locked(cachep, n->shared, node, true, &list);
> > + spin_unlock_irq(&n->list_lock);
> > +
> > + slabs_destroy(cachep, &list);
>
> Can the slabs_destroy() call be moved outside of the loop? It may be
> faster then?

Yes, it can. But, I'd prefer to call it on each node. It would be
better for cache although it would be marginal.

Thanks.

\
 
 \ /
  Last update: 2016-04-14 04:21    [W:0.100 / U:0.128 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site