From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1764030AbYEAQiu (ORCPT ); Thu, 1 May 2008 12:38:50 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1760187AbYEAQil (ORCPT ); Thu, 1 May 2008 12:38:41 -0400 Received: from el-out-1112.google.com ([209.85.162.183]:9371 "EHLO el-out-1112.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1759262AbYEAQij (ORCPT ); Thu, 1 May 2008 12:38:39 -0400 DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=bVQ0yudsVbYjXDVv3+m84hufRLuDP4Zmc4bxbpyI6AGGHh4f+Zb9Oqqioa7U6DuFAJAHXipi9B6mIuEraxswCbL17vMJSD+kjVXQCXatzD5LBv60d/FhuBteH0ZaGJFKbHiNR47WW/dIivYDhJMW8VgHAZuDtknMQTtcay4Wgi8= Message-ID: <86802c440805010938s1fb6c3feu67a31b281223686c@mail.gmail.com> Date: Thu, 1 May 2008 09:38:38 -0700 From: "Yinghai Lu" To: "Randy Dunlap" , "Ingo Molnar" Subject: Re: [PATCH] x86: mtrr cleanup for converting continuous to discrete - auto detect Cc: "Yinghai Lu" , "Andrew Morton" , "H. Peter Anvin" , "Thomas Gleixner" , "Gabriel C" , "Mika Fischer" , "linux-kernel@vger.kernel.org" In-Reply-To: <20080501080927.026b6243.randy.dunlap@oracle.com> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <200804272337.40130.yhlu.kernel@gmail.com> <200804290352.33543.yhlu.kernel@gmail.com> <200804292025.58268.yhlu.kernel@gmail.com> <200805010100.34751.yhlu.kernel@gmail.com> <20080501080927.026b6243.randy.dunlap@oracle.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, May 1, 2008 at 8:09 AM, Randy Dunlap wrote: > On Thu, 1 May 2008 01:00:34 -0700 Yinghai Lu wrote: > > > > > loop mtrr chunk_size and gran_size from 1M to 2G to find out optimal value. > > What size step (increment) is used for these loops? > > > > so user don't need to add mtrr_chunk_size and mtrr_gran_size, > > > > if optimal value is not found, print out all list to help select less optimal > > value. > > > > add mtrr_spare_reg_nr= so user could set 2 instead of 1, if the card need more entries. > > > > Signed-off-by: Yinghai Lu > > > > > > Index: linux-2.6/Documentation/kernel-parameters.txt > > =================================================================== > > --- linux-2.6.orig/Documentation/kernel-parameters.txt > > +++ linux-2.6/Documentation/kernel-parameters.txt > > @@ -613,9 +613,17 @@ and is between 256 and 4096 characters. > > that could hold holes aka. UC entries. > > > > mtrr_gran_size=nn[KMG] [X86] > > - used for mtrr cleanup. It is granity of mtrr block. > > + used for mtrr cleanup. It is granularity of mtrr block. > > + default is 1. > > Default > > > > Big value could prevent small alignment use up MTRRs. > > Large value could prevent small alignment from > using up MTRRs. > > > > > > + mtrr_spare_reg_nr=n [X86] > > + Format: > > + range: 0,7 : spare reg number > > + default : 1 > > + used for mtrr cleanup. It is spare mtrr entries number. > > Used > > > > + set to 2 or more if your graphical card need more. > > Set needs more. > > > > + > > disable_mtrr_trim [X86, Intel and AMD only] > > By default the kernel will trim any uncacheable > > memory out of your available memory pool based on > Ingo, can you change that directly in the patch? or need me send another updated patch? YH