From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from e39.co.us.ibm.com (e39.co.us.ibm.com [32.97.110.160]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client CN "e39.co.us.ibm.com", Issuer "Equifax" (verified OK)) by ozlabs.org (Postfix) with ESMTPS id CA01CB70EA for ; Sat, 5 Mar 2011 10:07:37 +1100 (EST) Received: from d03relay03.boulder.ibm.com (d03relay03.boulder.ibm.com [9.17.195.228]) by e39.co.us.ibm.com (8.14.4/8.13.1) with ESMTP id p24MsmPR014393 for ; Fri, 4 Mar 2011 15:54:48 -0700 Received: from d03av05.boulder.ibm.com (d03av05.boulder.ibm.com [9.17.195.85]) by d03relay03.boulder.ibm.com (8.13.8/8.13.8/NCO v10.0) with ESMTP id p24N7YdP101036 for ; Fri, 4 Mar 2011 16:07:34 -0700 Received: from d03av05.boulder.ibm.com (loopback [127.0.0.1]) by d03av05.boulder.ibm.com (8.14.4/8.13.1/NCO v10.0 AVout) with ESMTP id p24N7YwV001267 for ; Fri, 4 Mar 2011 16:07:34 -0700 Subject: Re: [PATCH v4 2/2] add icswx support From: "Tseng-Hui (Frank) Lin" To: Benjamin Herrenschmidt In-Reply-To: <1299278441.8833.915.camel@pasglop> References: <1299086454.28840.10.camel@flin.austin.ibm.com> <1299200560.8833.869.camel@pasglop> <1299259794.28840.57.camel@flin.austin.ibm.com> <1299270417.8833.897.camel@pasglop> <1299277344.28840.79.camel@flin.austin.ibm.com> <1299278441.8833.915.camel@pasglop> Content-Type: text/plain; charset="UTF-8" Date: Fri, 04 Mar 2011 17:07:33 -0600 Message-ID: <1299280053.28840.96.camel@flin.austin.ibm.com> Mime-Version: 1.0 Cc: tsenglin@us.ibm.com, linuxppc-dev@ozlabs.org List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Sat, 2011-03-05 at 09:40 +1100, Benjamin Herrenschmidt wrote: > > The lazy switching checks the shadow variable first before setting ACOP > > register. This saves mtspr() only if the new value is the same as > > current. If there are several coprocessors on the system, the ACOP > > register may have to be changed frequently. In that case, the lazy > > switching will not save time. In extreme case when the ACOP register > > needs to be changed every time, it actually slows down the execution by > > the additional shadow variable checking. > > By how much ? Is it even measurable ? > I don't have any measurable numbers. That's why I made it an option in case people wants to disable it. I do agree that the kernel has so many options that we should refrain from adding more. If the chance that the lazy switching slows down the execution is really low, we should just take out the option. Is there any one has an idea how much the numbers are? > Cheers, > Ben. > > > _______________________________________________ > Linuxppc-dev mailing list > Linuxppc-dev@lists.ozlabs.org > https://lists.ozlabs.org/listinfo/linuxppc-dev