From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: cmd64x: irq 14: nobody cared - system is dreadfully slow Date: Tue, 23 Jun 2009 16:04:40 -0700 (PDT) Message-ID: <20090623.160440.243068553.davem@davemloft.net> References: <20090623.031506.214747802.davem@davemloft.net> <200906231658.56489.elendil@planet.nl> <200906231813.09570.bzolnier@gmail.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Return-path: Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:57856 "EHLO sunset.davemloft.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755247AbZFWXEf (ORCPT ); Tue, 23 Jun 2009 19:04:35 -0400 In-Reply-To: <200906231813.09570.bzolnier@gmail.com> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: bzolnier@gmail.com Cc: elendil@planet.nl, sparclinux@vger.kernel.org, linux-ide@vger.kernel.org, linux-kernel@vger.kernel.org From: Bartlomiej Zolnierkiewicz Date: Tue, 23 Jun 2009 18:13:08 +0200 > On Tuesday 23 June 2009 16:58:54 Frans Pop wrote: >> So it seems to me that in 2.6.26 something was broken in the way these ID >> fields were handled, at least in this check. This is now fixed (possibly > > Great debugging work, thanks! Indeed, thanks for all of that work Frans! >> by the changes around 5b90e990..48fb2688) and *that* causes the >> regression. Note that the hard disks are also affected. > > I see it now -- in commit c419993 ("ide-iops: only clear DMA words > on setting DMA mode") we fixed a bug in ide_config_drive_speed().. > > [ It was a real bug resulting in incorrect data being passed to > the user-space through HDIO_GET_IDENTITY ioctl ('hdparm -i'). ] Given all of this I also agree that we should apply Bart's patch to remove the debugging check altogether, and I will thus do that right now.