From: Sergei Shtylyov Subject: Re: [PATCH] ide/libata: fix ata_id_is_cfa() (take 2) Date: Mon, 26 Jan 2009 23:34:17 +0300 Message-ID: <497E1E49.1090409@ru.mvista.com> References: <200901262309.31771.sshtylyov@ru.mvista.com> <20090126201905.43e57378@lxorguk.ukuu.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from h155.mvista.com ([63.81.120.155]:4823 "EHLO imap.sh.mvista.com" rhost-flags-OK-FAIL-OK-FAIL) by vger.kernel.org with ESMTP id S1753053AbZAZUdr (ORCPT ); Mon, 26 Jan 2009 15:33:47 -0500 In-Reply-To: <20090126201905.43e57378@lxorguk.ukuu.org.uk> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Alan Cox Cc: bzolnier@gmail.com, jgarzik@pobox.com, linux-ide@vger.kernel.org, gdu@mns.spb.ru Alan Cox wrote: >>When checking for the CFA feature set support, ata_id_is_cfa() tests bit 2 in >>word 82 of the identify data instead the word 83; it also checks the ATA/PI >>standard support in the word 80 -- which the CompactFlash specification has >>as reserved, this having no slightest chance to work on CF drives that don't >>have 0x848A in the word 0. Use instead the usual validity check: word 83 bit >>14 must be set, bit 15 cleared. >>Signed-off-by: Sergei Shtylyov > NAK please stop mixing corrections with bugs that have been explained. Huh? You haven't explained anything. Your ATA base explanation is simply not valid in the CF world. > You can't exercise logic in the id_is_cfa check that relies upon the > device being CFA, which is what your comment says you are doing. Otherwise it has no chance to detect a CF drive *following the CF specs* and not having 0x848A in word 0. Please suggest a better solution. Perhaps we could check for word 80 being 0? MBR, Sergei