On Sun, 2006-02-05 at 10:41 +0200, Pekka Enberg wrote:

Ah, sorry about that, I forgot to verify the NUMA case. The problem is
that to kmalloc_node() is calling cache_alloc() now which is forced
inline. I am wondering, would it be ok to make __cache_alloc()
non-inline for NUMA? The relevant numbers are:

[snip]

Btw, we can also change kmalloc_node() to use kmem_cache_alloc_node()
again but for that, we have a minor correctness issue, namely, the
__builtin_return_address(0) won't work for kmalloc_node(). Hmm.

			Pekka

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>