From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tejun Heo Subject: Re: [PATCH 4/5] libata: support the ata host which implements a queue depth less than 32 Date: Thu, 3 Jul 2014 09:50:58 -0400 Message-ID: <20140703135058.GB16191@htj.dyndns.org> References: <1404389382-27298-1-git-send-email-haokexin@gmail.com> <1404389382-27298-5-git-send-email-haokexin@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from mail-qc0-f172.google.com ([209.85.216.172]:49086 "EHLO mail-qc0-f172.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756760AbaGCNvC (ORCPT ); Thu, 3 Jul 2014 09:51:02 -0400 Received: by mail-qc0-f172.google.com with SMTP id o8so229925qcw.31 for ; Thu, 03 Jul 2014 06:51:01 -0700 (PDT) Content-Disposition: inline In-Reply-To: <1404389382-27298-5-git-send-email-haokexin@gmail.com> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Kevin Hao Cc: linux-ide@vger.kernel.org, Dan Williams On Thu, Jul 03, 2014 at 08:09:41PM +0800, Kevin Hao wrote: > The sata on fsl mpc8315e is broken after the commit 8a4aeec8d2d6 > ("libata/ahci: accommodate tag ordered controllers"). The reason is > that the ata controller on this SoC only implement a queue depth of > 16. When issuing the commands in tag order, all the commands in tag > 16 ~ 17 are mapped to tag 0 unconditionally and then causes the sata Shouldn't the function at least trigger WARN_ON_ONCE() if it's forcing the tag to be 0 when it shouldn't be? Silently mapping things to zero doesn't look like a particularly good idea. Thanks. -- tejun