From mboxrd@z Thu Jan 1 00:00:00 1970 From: Glauber Costa Subject: Re: [PATCH 2/4] Add a __GFP_SLABMEMCG flag Date: Sat, 9 Jun 2012 12:19:44 +0400 Message-ID: <4FD30720.6040908@parallels.com> References: <1339148601-20096-1-git-send-email-glommer@parallels.com> <1339148601-20096-3-git-send-email-glommer@parallels.com> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: Sender: cgroups-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-ID: Content-Type: text/plain; charset="us-ascii"; format="flowed" To: Christoph Lameter Cc: cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org, Tejun Heo , Frederic Weisbecker , devel-GEFAQzZX7r8dnm+yROfE0A@public.gmane.org, kamezawa.hiroyu-+CUm20s59erQFUHtdCDX3A@public.gmane.org, Pekka Enberg , Michal Hocko , Johannes Weiner , Suleiman Souhlal On 06/08/2012 11:31 PM, Christoph Lameter wrote: > Please make this conditional on CONFIG_MEMCG or so. The bit can be useful > in particular on 32 bit architectures. Looking at how __GFP_NOTRACK works - which is also ifdef'd, the bit it uses is skipped if that is not defined, which I believe is a sane thing to do. Given that, I don't see the point of conditionally defining the memcg bit, It basically means that the only way we can reuse the bit saved is by making a future feature fundamentally incompatible with memcg.