Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753471Ab0KCHZa (ORCPT ); Wed, 3 Nov 2010 03:25:30 -0400 Received: from mail-ww0-f44.google.com ([74.125.82.44]:61955 "EHLO mail-ww0-f44.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753206Ab0KCHZ3 (ORCPT ); Wed, 3 Nov 2010 03:25:29 -0400 DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=subject:from:to:cc:in-reply-to:references:content-type:date :message-id:mime-version:x-mailer:content-transfer-encoding; b=Lqh0Ph1Wbx4+F+wktdZ6W3V4AyBGBe4TMCFgOryUbcapA+322M13gVi2/C2ds/PZrW V3pDQcOiKIBY970xij0G+PXTG3INb+2lmg45RbPIt4/7W+A6aQgJR4KNNyTBzpFdMJo9 pSH3S1zeZlxUZIrI7vklgRIv8jwFzsiZGn3kY= Subject: Re: [RFC 4/4]x86: avoid tlbstate lock if no enough cpus From: Eric Dumazet To: Shaohua Li Cc: lkml , Ingo Molnar , Andi Kleen , "hpa@zytor.com" In-Reply-To: <1288768795.23014.123.camel@sli10-conroe> References: <1288766668.23014.117.camel@sli10-conroe> <1288767580.2467.636.camel@edumazet-laptop> <1288767995.23014.120.camel@sli10-conroe> <1288768330.2467.660.camel@edumazet-laptop> <1288768795.23014.123.camel@sli10-conroe> Content-Type: text/plain; charset="UTF-8" Date: Wed, 03 Nov 2010 08:25:23 +0100 Message-ID: <1288769123.2467.681.camel@edumazet-laptop> Mime-Version: 1.0 X-Mailer: Evolution 2.30.3 Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1264 Lines: 35 Le mercredi 03 novembre 2010 à 15:19 +0800, Shaohua Li a écrit : > On Wed, 2010-11-03 at 15:12 +0800, Eric Dumazet wrote: > > Le mercredi 03 novembre 2010 à 15:06 +0800, Shaohua Li a écrit : > > > just don't want to include the non-present cpus here. I wonder why we > > > haven't a variable to record online cpu number. > > > > What prevents a 256 cpus machine, to have 8 online cpus that all use the > > same TLB vector ? > > > > (Max 32 vectors, so 8 cpus share each vector, settled at boot time) > > > > Forget about 'online', and think 'possible' ;) > Hmm, the spread vectors to node already merged, how could the 8 cpus > share a vector? > You boot a machine with 256 cpu. They are online and very well. Each vector is shared by at least 8 cpus, because 256/32 = 8. OK ? Now you off-line 256-8 cpus, because you have HOTPLUG capability in your kernel and you have some policy to bring them up later if needed. What happens ? Do you rebalance TLB vectors to make sure each cpu has its own vector ? -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/