From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758288Ab1LGT6W (ORCPT ); Wed, 7 Dec 2011 14:58:22 -0500 Received: from tx2ehsobe004.messaging.microsoft.com ([65.55.88.14]:48550 "EHLO TX2EHSOBE007.bigfish.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756215Ab1LGT6U (ORCPT ); Wed, 7 Dec 2011 14:58:20 -0500 X-SpamScore: -15 X-BigFish: VPS-15(zzbb2dK9371K1432N98dKzz1202hzzz2dh668h839h944h61h) X-Spam-TCS-SCL: 0:0 X-Forefront-Antispam-Report: CIP:163.181.249.108;KIP:(null);UIP:(null);IPV:NLI;H:ausb3twp01.amd.com;RD:none;EFVD:NLI X-WSS-ID: 0LVUNH2-01-PVZ-02 X-M-MSG: Date: Wed, 7 Dec 2011 20:57:53 +0100 From: Hans Rosenfeld To: "H. Peter Anvin" CC: Andi Kleen , , , , , , , , , Subject: Re: [PATCH 0/9] rework of extended state handling, LWP support Message-ID: <20111207195753.GD665@escobedo.osrc.amd.com> References: <1322570488-21798-1-git-send-email-hans.rosenfeld@amd.com> <20111130173745.GT878@escobedo.osrc.amd.com> <20111130215200.GS24062@one.firstfloor.org> <20111201203605.GB411@escobedo.osrc.amd.com> <4ED83177.7070706@zytor.com> <20111202112050.GC411@escobedo.osrc.amd.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Disposition: inline In-Reply-To: <20111202112050.GC411@escobedo.osrc.amd.com> Organization: Advanced Micro Devices GmbH, Einsteinring 24, 85609 Dornach b. Muenchen; Geschaeftsfuehrer: Andrew Bowd, Alberto Bozzo; Sitz: Dornach, Gemeinde Aschheim, Landkreis Muenchen; Registergericht Muenchen, HRB Nr. 43632 User-Agent: Mutt/1.5.21 (2010-09-15) X-OriginatorOrg: amd.com Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi, On Fri, Dec 02, 2011 at 12:20:50PM +0100, Hans Rosenfeld wrote: > On Thu, Dec 01, 2011 at 06:01:27PM -0800, H. Peter Anvin wrote: > > On 12/01/2011 12:36 PM, Hans Rosenfeld wrote: > > > > > > So I assume, if you have 10000s of processes on a legacy 32bit system > > > that never do any FPU stuff or SSE optimizations, you might indeed waste > > > a couple of megabytes. I don't think thats very realistic, but that's > > > just my opinion. > > > > > > > A couple of megabytes of *lowmem*... > > Ok, I'll rework that part, so that preallocation only happens on systems > that support non-lazy states. That means patch #7 is going away and > patch #8 is getting slightly bigger. This may take a few days as I have > to test that again. The reworked LWP patches are ready and will follow shortly. I also added a Kconfig option to completely disable support for non-lazy states, allowing to completely avoid preallocation of the xstate area if required. Did you look at the other patches already? Hans -- %SYSTEM-F-ANARCHISM, The operating system has been overthrown