From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tejun Heo Subject: Re: [PATCH 2/4] libata: implement ata_dev_init() Date: Mon, 13 Mar 2006 18:25:21 +0900 Message-ID: <20060313092521.GA2091@htj.dyndns.org> References: <11421469551948-git-send-email-htejun@gmail.com> <11421469552851-git-send-email-htejun@gmail.com> <20060313061926.GA30429@htj.dyndns.org> <44152CBD.3010405@pobox.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from zproxy.gmail.com ([64.233.162.203]:64570 "EHLO zproxy.gmail.com") by vger.kernel.org with ESMTP id S932389AbWCMJZ0 (ORCPT ); Mon, 13 Mar 2006 04:25:26 -0500 Received: by zproxy.gmail.com with SMTP id 13so1280024nzn for ; Mon, 13 Mar 2006 01:25:25 -0800 (PST) Content-Disposition: inline In-Reply-To: <44152CBD.3010405@pobox.com> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Jeff Garzik Cc: albertcc@tw.ibm.com, alan@lxorguk.ukuu.org.uk, linux-ide@vger.kernel.org On Mon, Mar 13, 2006 at 03:26:37AM -0500, Jeff Garzik wrote: > NAK: > > 1) the devices are embedded with ata_port, and so all static > initialization should be done in ata_host_init(). It's only called once > for each device, as you say, anyway... no need to separate it from the > rest of the struct ata_port initialization AFAICS. I was mainly thinking of hotplug cases. dev liftime becomes different form ap lifetime in hotplug cases, so it needs reinitialization before recycling. I guess it can wait till we actually do hotplug. I'll drop this patch. > 2) there is no need to memset() [unless you are re-initializing, I suppose] ditto. -- tejun