[patch] mm, slab: initialize object alignment on cache creation
From: David Rientjes
Date: Tue Sep 23 2014 - 17:43:27 EST
Since 4590685546a3 ("mm/sl[aou]b: Common alignment code"), the "ralign"
automatic variable in __kmem_cache_create() may be used as uninitialized.
The proper alignment defaults to BYTES_PER_WORD and can be overridden by
SLAB_RED_ZONE or the alignment specified by the caller.
This fixes https://bugzilla.kernel.org/show_bug.cgi?id=85031
Reported-by: Andrei Elovikov <a.elovikov@xxxxxxxxx>
Signed-off-by: David Rientjes <rientjes@xxxxxxxxxx>
---
a.elovikov: If you respond to this email with your full name, Andrew can
give proper credit for reporting this issue.
mm/slab.c | 11 ++---------
1 file changed, 2 insertions(+), 9 deletions(-)
diff --git a/mm/slab.c b/mm/slab.c
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -2128,7 +2128,8 @@ static int __init_refok setup_cpu_cache(struct kmem_cache *cachep, gfp_t gfp)
int
__kmem_cache_create (struct kmem_cache *cachep, unsigned long flags)
{
- size_t left_over, freelist_size, ralign;
+ size_t left_over, freelist_size;
+ size_t ralign = BYTES_PER_WORD;
gfp_t gfp;
int err;
size_t size = cachep->size;
@@ -2161,14 +2162,6 @@ __kmem_cache_create (struct kmem_cache *cachep, unsigned long flags)
size &= ~(BYTES_PER_WORD - 1);
}
- /*
- * Redzoning and user store require word alignment or possibly larger.
- * Note this will be overridden by architecture or caller mandated
- * alignment if either is greater than BYTES_PER_WORD.
- */
- if (flags & SLAB_STORE_USER)
- ralign = BYTES_PER_WORD;
-
if (flags & SLAB_RED_ZONE) {
ralign = REDZONE_ALIGN;
/* If redzoning, ensure that the second redzone is suitably
--
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/