From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tejun Heo Subject: Re: [PATCH 16/22] libata: implement new SCR handling and port on/offline functions Date: Sun, 14 May 2006 08:18:51 +0900 Message-ID: <4466695B.4050204@gmail.com> References: <1147348791300-git-send-email-htejun@gmail.com> <44665314.4040805@pobox.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from wr-out-0506.google.com ([64.233.184.239]:21048 "EHLO wr-out-0506.google.com") by vger.kernel.org with ESMTP id S964797AbWEMXS6 (ORCPT ); Sat, 13 May 2006 19:18:58 -0400 Received: by wr-out-0506.google.com with SMTP id i23so585447wra for ; Sat, 13 May 2006 16:18:58 -0700 (PDT) In-Reply-To: <44665314.4040805@pobox.com> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Jeff Garzik Cc: alan@lxorguk.ukuu.org.uk, axboe@suse.de, albertcc@tw.ibm.com, forrest.zhao@intel.com, efalk@google.com, linux-ide@vger.kernel.org Jeff Garzik wrote: > Tejun Heo wrote: >> Implement ata_scr_{valid|read|write|write_flush}() and >> ata_port_{online|offline}(). These functions replace >> scr_{read|write}() and sata_dev_present(). >> >> Major difference between between the new SCR functions and the old >> ones is that the new ones have a way to signal error to the caller. >> This makes handling SCR-available and SCR-unavailable cases in the >> same path easier. Also, it eases later PM implementation where SCR >> access can fail due to various reasons. >> >> ata_port_{online|offline}() functions return 1 only when they are >> affirmitive of the condition. e.g. if SCR is unaccessible or >> presence cannot be determined for other reasons, these functions >> return 0. So, ata_port_online() != !ata_port_offline(). This >> distinction is useful in many exception handling cases. > > If its SATA-specific, it should have a "sata_" not "ata_" prefix. > I thought about it but the 'S' in SCR stands for Serial ATA and we also need to rename all PM functions to sata_pmp_xxx() - SCR and PMP are already SATA specific. Do you think that's the way to go? For ata_port_on/offline(), I think it's better to leave them alone for hotpluggable IDE hotbays with presence detection. -- tejun