Re: [sched-devel/latest] WARNING: at mm/slub.c:2443

[Pekka J Enberg]

On Tue, 22 Apr 2008, Ingo Molnar wrote:
> * Frans Pop <elendil@xxxxxxxxx> wrote:
> 
> > I got the following warning while running sched-devel/latest.
> > 
> > I was not doing any tracing at that point (for the glibc compilation 
> > issue). I suspect it was while I was using a KDE application (amarok) 
> > as the desktop froze for a few seconds at one point.
> 
> i think this message is harmless to system stability.
> 
> Cc:-ing USB folks as the kmem_cache_destroy() comes from 
> drivers/usb/mon/mon_text.c.
> 
> Context: SLUB recently started warning about caches that get destroyed 
> with still outstanding allocations.
> 
> Pekka, it might make sense to turn this SLUB warning into something more 
> specific? It does seem to trigger in practice.

Oh absolutely. Something like this?

[PATCH] slub: improve kmem_cache_destroy() error message

As pointed out by Ingo, the SLUB warning of calling kmem_cache_destroy() 
with cache that still has objects triggers in practice. So turn this 
WARN_ON() into a nice SLUB specific error message to avoid people 
confusing it to a SLUB bug.

Cc: Ingo Molnar <mingo@xxxxxxx>
Cc: Christoph Lameter <clameter@xxxxxxx>
Signed-off-by: Pekka Enberg <penberg@xxxxxxxxxxxxxx>
---

diff --git a/mm/slub.c b/mm/slub.c
index 7f8aaa2..c372e88 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -2426,8 +2426,11 @@ void kmem_cache_destroy(struct kmem_cache *s)
 	if (!s->refcount) {
 		list_del(&s->list);
 		up_write(&slub_lock);
-		if (kmem_cache_close(s))
-			WARN_ON(1);
+		if (kmem_cache_close(s)) {
+			printk(KERN_ERR "SLUB %s: %s called for cache that "
+				"still has objects.\n", s->name, __func__);
+			dump_stack();
+		}
 		sysfs_slab_remove(s);
 	} else
 		up_write(&slub_lock);
--
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/