Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754657AbYBXHek (ORCPT ); Sun, 24 Feb 2008 02:34:40 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751324AbYBXHeb (ORCPT ); Sun, 24 Feb 2008 02:34:31 -0500 Received: from mail08a.verio.de ([213.198.55.73]:17993 "HELO mail08a.verio.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1751254AbYBXHe3 (ORCPT ); Sun, 24 Feb 2008 02:34:29 -0500 X-Greylist: delayed 399 seconds by postgrey-1.27 at vger.kernel.org; Sun, 24 Feb 2008 02:34:29 EST Message-Id: <6.1.1.1.2.20080224082113.01e51e50@192.168.6.12> X-Mailer: QUALCOMM Windows Eudora Version 6.1.1.1 Date: Sun, 24 Feb 2008 08:27:30 +0100 To: linux-kernel@vger.kernel.org From: Roger While Subject: Re: [patch 1/2] x86,fpu: split FPU state from task struct Cc: suresh.b.siddha@intel.com Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed X-SIMBasis-MailScanner-Information: Please contact the ISP for more information X-SIMBasis-MailScanner: Found to be clean X-SIMBasis-MailScanner-SpamCheck: not spam (whitelisted), SpamAssassin (score=-5.899, required 5, autolearn=not spam, ALL_TRUSTED -3.30, BAYES_00 -2.60) X-SIMBasis-MailScanner-From: simrw@sim-basis.de X-Spam: [F=0.3029429977; heur=0.500(-6600); stat=0.302; spamtraq-heur=0.500(2007101727)] X-MAIL-FROM: X-SOURCE-IP: [213.198.55.139] X-SF-Loop: 1 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 997 Lines: 28 On Sat, Feb 23, 2008 at 06:34:38PM -0800, Suresh Siddha wrote: > Split the FPU save area from the task struct. This allows easy migration > of FPU context, and it's generally cleaner. It also allows the following > two optimizations: > > 1) only allocate when the application actually uses FPU, so in the first > lazy FPU trap. This could save memory for non-fpu using apps. Next patch > does this lazy allocation. > > 2) allocate the right size for the actual cpu rather than 512 bytes always. > Patches enabling xsave/xrstor support (coming shortly) will take advantage > of this. > if (next_p->fpu_counter>5) > - prefetch(&next->i387.fxsave); > + prefetch(FXSAVE(next_p)); Shouldn't that be prefetch(FXSAVE(next)); ? Roger -- 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/