In slab_alloc_node(), after the cpu_slab is assigned, if the task is
preempted and moves to another CPU, there's nothing keeping the page and
object in sync. The -rt kernel crashed because page was NULL and object
was not, and the node_match() dereferences page. Even though the crash
happened on -rt, there's nothing that's keeping this from happening on
mainline.
The easiest fix is to disable preemption for the entire time from
acquiring the current CPU cpu_slab and assigning the object and page.
After that, it's fine to allow preemption.
Also add a check if page is NULL in node_match().
Signed-off-by: Steven Rostedt <rostedt@xxxxxxxxxxx>
diff --git a/mm/slub.c b/mm/slub.c
index ba2ca53..10714ee 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -2041,7 +2041,7 @@ static void flush_all(struct kmem_cache *s)
static inline int node_match(struct page *page, int node)
{
#ifdef CONFIG_NUMA
- if (node != NUMA_NO_NODE && page_to_nid(page) != node)
+ if (!page || (node != NUMA_NO_NODE && page_to_nid(page) != node))
return 0;
#endif
return 1;
@@ -2337,7 +2337,10 @@ redo:
* enabled. We may switch back and forth between cpus while
* reading from one cpu area. That does not matter as long
* as we end up on the original cpu again when doing the cmpxchg.
+ *
+ * But we need to sync the setting of page and object.
*/
+ preempt_disable();
c = __this_cpu_ptr(s->cpu_slab);
/*
@@ -2347,10 +2350,14 @@ redo:
* linked list in between.
*/
tid = c->tid;
+
+ /* Must have tid first in case an interrupt comes in */
barrier();
object = c->freelist;
page = c->page;
+ preempt_enable();
+
if (unlikely(!object || !node_match(page, node)))
object = __slab_alloc(s, gfpflags, node, addr, c);
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>