--=_courier-11405-1058204587-0001-2
Content-Type: text/plain; charset=iso-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Nicolas wrote:
>kernel BUG at mm/slab.c:1631!
>
That's
BUG_ON(GET_PAGE_CACHE(page) != cachep);
Someone called kmem_cache_free(cachep, obj), but cachep is for a
different object type.
>Call Trace:
>[sys_open+120/133] sys_open+0x78/0x85
>
Within sys_open - probably putname().
I have no idea how the bug could be triggered. If you can easily
reproduce it: can you try the attached patch? It prints additional data.
And please add more details: Which gcc compiler, which filesystems, etc.
--
Manfred
--=_courier-11405-1058204587-0001-2
Content-Type: text/plain; name=patch-slab-cachedebug; charset=iso-8859-1
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
filename="patch-slab-cachedebug"
--- 2.5/mm/slab.c 2003-07-10 23:27:00.000000000 +0200
+++ build-2.5/mm/slab.c 2003-07-14 19:36:26.000000000 +0200
@@ -1628,7 +1628,13 @@
kfree_debugcheck(objp);
page = virt_to_page(objp);
- BUG_ON(GET_PAGE_CACHE(page) != cachep);
+ if (GET_PAGE_CACHE(page) != cachep) {
+ printk(KERN_ERR "mismatch in kmem_cache_free: expected cache %p, got %p\n",
+ GET_PAGE_CACHE(page),cachep);
+ printk(KERN_ERR "%p is %s.\n", cachep, cachep->name);
+ printk(KERN_ERR "%p is %s.\n", GET_PAGE_CACHE(page), GET_PAGE_CACHE(page)->name);
+ WARN_ON(1);
+ }
slabp = GET_PAGE_SLAB(page);
if (cachep->flags & SLAB_STORE_USER) {
--=_courier-11405-1058204587-0001-2--