From mboxrd@z Thu Jan 1 00:00:00 1970 From: "H. Peter Anvin" Subject: Re: [PATCHSET] percpu: generalize first chunk allocators and improve lpage NUMA support Date: Wed, 24 Jun 2009 17:13:37 -0700 Message-ID: <4A42C131.6040007@zytor.com> References: <1245850216-31653-1-git-send-email-tj@kernel.org> <20090624165508.30b88343.akpm@linux-foundation.org> <20090625000202.GR6760@one.firstfloor.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: Received: from terminus.zytor.com ([198.137.202.10]:57738 "EHLO terminus.zytor.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752147AbZFYAOu (ORCPT ); Wed, 24 Jun 2009 20:14:50 -0400 In-Reply-To: <20090625000202.GR6760@one.firstfloor.org> Sender: linux-arch-owner@vger.kernel.org List-ID: To: Andi Kleen Cc: Andrew Morton , Tejun Heo , linux-kernel@vger.kernel.org, x86@kernel.org, linux-arch@vger.kernel.org, mingo@elte.hu, tglx@linutronix.de, cl@linux-foundation.org Andi Kleen wrote: > > Haven't read the new patches, but per cpu data always was sized > for all possible CPUs. > >> and N is large, what did it cost? > >> And what are reasonable values of N? > > N should normally not be large anymore, since num_possible_cpus() > is sized based on firmware information now. > *Ahem* virtual machines *ahem*... I have discussed this with Tejun, and the plan is to allocate the percpu information when a processor is first brought online (but not removed when it is offlined again.) It's a real problem for 32-bit VMs, so it's more important than you'd think. -hpa