Messages in this thread | | | Date | Tue, 22 Nov 2011 08:50:03 -0600 (CST) | From | Christoph Lameter <> | Subject | Re: WARNING: at mm/slub.c:3357, kernel BUG at mm/slub.c:3413 |
| |
On Tue, 22 Nov 2011, Markus Trippelsdorf wrote:
> > Given slub is now lockless, validate_slab_slab() is probably very wrong > > these days. > > OK "slabinfo -v" is useless then. But that doesn't invalidate the BUGs > that I saw during boot. They happend before I could even run slabinfo > for the first time.
Correct. Also the use of debugging disables the use of cmpxchg_double() but not this_cpu_cmpxchg() use. See cmpxchg_double_slab() and kmem_cache_open()s determination of the __CMPXCHG_DOUBLE flag.
| |