From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932184AbWGKWHZ (ORCPT ); Tue, 11 Jul 2006 18:07:25 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932185AbWGKWHZ (ORCPT ); Tue, 11 Jul 2006 18:07:25 -0400 Received: from dsl027-180-168.sfo1.dsl.speakeasy.net ([216.27.180.168]:57316 "EHLO sunset.davemloft.net") by vger.kernel.org with ESMTP id S932184AbWGKWHY (ORCPT ); Tue, 11 Jul 2006 18:07:24 -0400 Date: Tue, 11 Jul 2006 15:08:10 -0700 (PDT) Message-Id: <20060711.150810.13051231.davem@davemloft.net> To: bos@serpentine.com Cc: linux-kernel@vger.kernel.org, arjan@infradead.org Subject: Re: [PATCH] Add memcpy_cachebypass, a copy routine that tries to keep cache pressure down From: David Miller In-Reply-To: <1152655509.16499.49.camel@chalcedony.pathscale.com> References: <1152653401.16499.35.camel@chalcedony.pathscale.com> <20060711.145751.77136364.davem@davemloft.net> <1152655509.16499.49.camel@chalcedony.pathscale.com> X-Mailer: Mew version 4.2 on Emacs 21.4 / Mule 5.0 (SAKAKI) Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org From: Bryan O'Sullivan Date: Tue, 11 Jul 2006 15:05:08 -0700 > Well, exactly this scheme seems to work for __iowrite_copy*. There's a > weak generic version and a strong version in arch/x86_64/lib that > overrides it, and it gets picked up at kernel link time. It is linked in as an object, not into the library archive, that's why that one works like that. That is why io.o is added to the "obj-y" variable instead of the "lib-y" variable. It is also necessary to link these things in as objects when module exports are present, because if there is no in-kernel reference to the function, you won't get the function nor it's module export :)