Messages in this thread | | | Date | Fri, 12 Jun 2009 11:13:04 +0200 | From | Nick Piggin <> | Subject | Re: slab: setup allocators earlier in the boot sequence |
| |
On Fri, Jun 12, 2009 at 11:49:31AM +0300, Pekka Enberg wrote: > On Fri, Jun 12, 2009 at 11:44 AM, Benjamin > Herrenschmidt<benh@kernel.crashing.org> wrote: > > On Fri, 2009-06-12 at 11:04 +0300, Pekka Enberg wrote: > >> Hi Nick, > >> > >> On Fri, 2009-06-12 at 10:02 +0200, Nick Piggin wrote: > >> > Fair enough, but this can be done right down in the synchronous > >> > reclaim path in the page allocator. This will catch more cases > >> > of code using the page allocator directly, and should be not > >> > as hot as the slab allocator. > >> > >> So you want to push the local_irq_enable() to the page allocator too? We > >> can certainly do that but I think we ought to wait for Andrew to merge > >> Mel's patches to mainline first, OK? > > > > Doesn't my patch take care of all the cases in a much more simple way ? > > Nick, the patch Ben is talking about is here: > > http://patchwork.kernel.org/patch/29700/
It's OK. I'd make it gfp_notsmellybits, and avoid the ~. And read_mostly.
> The biggest problem with the patch is that the gfp_smellybits is wide > open for abuse. Hmm.
Probably would be better to hide it in mm/ and then just allow it to be modified with a couple of calls. OTOH if it is only modified in a couple of places then maybe that's overkill.
The whole problem comes about because we don't just restore our previously saved flags here... I guess it probably adds even more overhead to do that and make everything just work :(
| |