Re: [patch 3/3 -mmotm] oom: invoke oom killer for __GFP_NOFAIL

From: Pekka Enberg
Date: Tue Jun 02 2009 - 02:42:09 EST


Hi Nick,

On Mon, Jun 01, 2009 at 10:56:02PM -0700, Andrew Morton wrote:
>> Do we know of any callsites which do greater-than-order-0 allocations
>> with GFP_NOFAIL?  If so, we should fix them.

On Tue, Jun 2, 2009 at 9:27 AM, Nick Piggin <npiggin@xxxxxxx> wrote:
> Yeah, we have GFP_NOFAIL going to the slab allocator, so all bets are
> off in regard to the order.

SLUB is supports variable order allocations so we can certainly do
GFP_NOFAIL special casing there for < PAGE_SIZE allocations to enforce
zero order allocation for those. I suspect we could do something
similar in SLQB too?

Pekka
--
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/