* [PATCH] mm: slab: Verify the nodeid passed to ____cache_alloc_node [not found] <1285971995.6305502.1361465165007.JavaMail.root@redhat.com> @ 2013-02-21 22:07 ` Aaron Tomlin 2013-02-22 13:42 ` Glauber Costa 0 siblings, 1 reply; 3+ messages in thread From: Aaron Tomlin @ 2013-02-21 22:07 UTC (permalink / raw) To: linux-mm; +Cc: linux-kernel, Rik Hi, The addition of this BUG_ON should make debugging easier. While I understand that this code path is "hot", surely it is better to assert the condition than to wait until some random NULL pointer dereference or page fault. If the caller passes an invalid nodeid, at this stage in my opinion it's already a BUG. Cheers, Aaron ---8<--- mm: slab: Verify the nodeid passed to ____cache_alloc_node If the nodeid is > num_online_nodes() this can cause an Oops and a panic(). The purpose of this patch is to assert if this condition is true to aid debugging efforts rather than some random NULL pointer dereference or page fault. Signed-off-by: Aaron Tomlin <atomlin@redhat.com> slab.c | 1 + 1 file changed, 1 insertion(+) diff --git a/mm/slab.c b/mm/slab.c index e7667a3..735e8bd 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -3412,6 +3412,7 @@ static void *____cache_alloc_node(struct kmem_cache *cachep, gfp_t flags, void *obj; int x; + BUG_ON(nodeid > num_online_nodes()); l3 = cachep->nodelists[nodeid]; BUG_ON(!l3); -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a> ^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH] mm: slab: Verify the nodeid passed to ____cache_alloc_node 2013-02-21 22:07 ` [PATCH] mm: slab: Verify the nodeid passed to ____cache_alloc_node Aaron Tomlin @ 2013-02-22 13:42 ` Glauber Costa 2013-02-25 9:11 ` Aaron Tomlin 0 siblings, 1 reply; 3+ messages in thread From: Glauber Costa @ 2013-02-22 13:42 UTC (permalink / raw) To: Aaron Tomlin; +Cc: linux-mm, linux-kernel, Rik On 02/22/2013 02:07 AM, Aaron Tomlin wrote: > The addition of this BUG_ON should make debugging easier. > While I understand that this code path is "hot", surely > it is better to assert the condition than to wait until > some random NULL pointer dereference or page fault. If the > caller passes an invalid nodeid, at this stage in my opinion > it's already a BUG. If you assert with VM_BUG_ON, it will be active on debugging kernels only, which I believe is better suited for a hotpath. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a> ^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] mm: slab: Verify the nodeid passed to ____cache_alloc_node 2013-02-22 13:42 ` Glauber Costa @ 2013-02-25 9:11 ` Aaron Tomlin 0 siblings, 0 replies; 3+ messages in thread From: Aaron Tomlin @ 2013-02-25 9:11 UTC (permalink / raw) To: Glauber Costa; +Cc: linux-mm, linux-kernel, Rik > If you assert with VM_BUG_ON, it will be active on debugging kernels > only, which I believe is better suited for a hotpath. Agreed. Regards, Aaron -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a> ^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2013-02-25 9:11 UTC | newest] Thread overview: 3+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- [not found] <1285971995.6305502.1361465165007.JavaMail.root@redhat.com> 2013-02-21 22:07 ` [PATCH] mm: slab: Verify the nodeid passed to ____cache_alloc_node Aaron Tomlin 2013-02-22 13:42 ` Glauber Costa 2013-02-25 9:11 ` Aaron Tomlin
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).