Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1377002
| From | Christoph Lameter <cl@linux.com> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | Re: [PATCH v2 01/11] mm/slab: fix the theoretical race by holding proper lock |
| Date | 2016-04-12 18:40 +0200 |
| Message-ID | <rn9w5-1Xf-3@gated-at.bofh.it> (permalink) |
| References | <rmYAF-1fb-3@gated-at.bofh.it> <rmYAG-1fb-17@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
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?
Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
[PATCH v2 01/11] mm/slab: fix the theoretical race by holding proper lock js1304@gmail.com - 2016-04-12 07:00 +0200
Re: [PATCH v2 01/11] mm/slab: fix the theoretical race by holding proper lock Christoph Lameter <cl@linux.com> - 2016-04-12 18:40 +0200
Re: [PATCH v2 01/11] mm/slab: fix the theoretical race by holding proper lock Joonsoo Kim <iamjoonsoo.kim@lge.com> - 2016-04-14 04:00 +0200
csiph-web