From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from nz-out-0102.google.com (nz-out-0506.google.com [64.233.162.226]) by ozlabs.org (Postfix) with ESMTP id 3768867C94 for ; Tue, 12 Dec 2006 01:23:47 +1100 (EST) Received: by nz-out-0102.google.com with SMTP id l1so833610nzf for ; Mon, 11 Dec 2006 06:23:44 -0800 (PST) Message-ID: <457D69E7.9040708@gmail.com> Date: Mon, 11 Dec 2006 23:23:35 +0900 From: Tejun Heo MIME-Version: 1.0 To: Arnd Bergmann Subject: Re: [PATCH] libata: don't initialize sg in ata_exec_internal() if DMA_NONE References: <200612081914.41810.arnd.bergmann@de.ibm.com> <20061211140258.GB18947@htj.dyndns.org> In-Reply-To: <20061211140258.GB18947@htj.dyndns.org> Content-Type: text/plain; charset=ISO-8859-1 Cc: linux-ide@vger.kernel.org, jgarzik@pobox.com, linux-kernel@vger.kernel.org, linuxppc-dev@ozlabs.org List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Tejun Heo wrote: > I'll follow up with conversion to ata_do_simple_cmd(). The current situation is... ata_exec_internal_sg() : no user except for ata_exec_internal() yet ata_exec_internal() : one data transferring user. other are non-data ata_do_simple_cmd() : three users So, adding another exec_internal variant doesn't look so hot. It seems we already have enough variants considering the small number of users. I think it's best to leave it alone for now. Thanks. -- tejun