From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756813AbZETXDV (ORCPT ); Wed, 20 May 2009 19:03:21 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753614AbZETXDN (ORCPT ); Wed, 20 May 2009 19:03:13 -0400 Received: from mga09.intel.com ([134.134.136.24]:59163 "EHLO mga09.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753229AbZETXDM (ORCPT ); Wed, 20 May 2009 19:03:12 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.41,224,1241420400"; d="scan'208";a="414336723" Subject: Re: [GIT PATCH] x86,percpu: fix pageattr handling with remap allocator From: Suresh Siddha Reply-To: suresh.b.siddha@intel.com To: Tejun Heo Cc: "H. Peter Anvin" , "JBeulich@novell.com" , "andi@firstfloor.org" , "mingo@elte.hu" , "linux-kernel-owner@vger.kernel.org" , "tglx@linutronix.de" , "linux-kernel@vger.kernel.org" In-Reply-To: <4A120B47.8060200@kernel.org> References: <1242305390-21958-1-git-send-email-tj@kernel.org> <1242436626.27006.8623.camel@localhost.localdomain> <4A0ED8D8.2010303@kernel.org> <1242500964.27006.8636.camel@localhost.localdomain> <4A0F672A.3000309@kernel.org> <1242674444.27006.8691.camel@localhost.localdomain> <4A11B9E7.8010707@zytor.com> <1242680835.27006.8734.camel@localhost.localdomain> <4A120B47.8060200@kernel.org> Content-Type: text/plain Organization: Intel Corp Date: Wed, 20 May 2009 16:01:10 -0700 Message-Id: <1242860470.27006.10106.camel@localhost.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.24.1 (2.24.1-2.fc10) Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, 2009-05-18 at 18:28 -0700, Tejun Heo wrote: > Hello, Suresh. > > Suresh Siddha wrote: > > This %gs:absolute type accesses are for static percpu data. > > > > But what I was referring to is the dynamic percpu data(accessed through > > per_cpu_ptr()). Instead of combining some part of the dynamic percpu > > data into the static percpu data(first percpu chunk), we can use > > different chunks for dynamic percpu data and governed by a different > > per_cpu_dynamic_offset[NR_CPUS] array. > > > > Then we can use large-page kernel direct mappings for static percpu data > > (or %gs:offset) and small-page vmalloc mappings for dynamic percpu data. > > Hmmm... I can't really follow what you're suggesting. Can you please > explain it in more detailed way? Ok. Before I make another attempt walking that hill :) I was talking to Peter and it seems there are some requests to change the first percpu unit allocation for each possible cpu using bootmem allocator, to allocating the corresponding unit at the cpu online time. Do you have plans to change this? If we do this allocation during the corresponding cpu online time and don't end up using big pages, then also we avoid all these aliasing issues... thanks, suresh