From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff Garzik Subject: Re: [Q] PCI Express and ide (native) leads to irq storm? Date: Wed, 15 Nov 2006 13:49:21 -0500 Message-ID: <455B6131.1@pobox.com> References: <453DC2A9.8000507@sw.ru> <453DC65C.8000408@sw.ru> <454206EE.9080206@sw.ru> <1161958862.16839.26.camel@localhost.localdomain> <4559879D.8090105@sw.ru> <455AF01C.5090307@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=KOI8-R; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from srv5.dvmed.net ([207.36.208.214]:32144 "EHLO mail.dvmed.net") by vger.kernel.org with ESMTP id S1030863AbWKOStd (ORCPT ); Wed, 15 Nov 2006 13:49:33 -0500 In-Reply-To: <455AF01C.5090307@gmail.com> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Tejun Heo Cc: Vasily Averin , Alan Cox , Jens Axboe , linux-kernel@vger.kernel.org, Bartlomiej Zolnierkiewicz , linux-ide@vger.kernel.org, devel@openvz.org Tejun Heo wrote: > In short, some piix controllers including ICH7, when put into enhanced > mode (PCI native mode), uses BMDMA Interrupt bit as interrupt > pending/clear bit for *all* commands. ie. Reading STATUS does NOT clear Yep. I thought I had mentioned this, ages ago. > Fortunately, libata is immune to the problem because it does > ap->ops->irq_clear(ap) in ata_host_intr() regardless of command type in > flight. So, not loading IDE piix and using libata to drive all piix > ports solves the problem. Yep, that's intentional :) Jeff