From mboxrd@z Thu Jan 1 00:00:00 1970 From: Bartlomiej Zolnierkiewicz Subject: Re: PIIX4: DMA timeout issue Date: Wed, 30 Jul 2008 21:43:37 +0200 Message-ID: <200807302143.37693.bzolnier@gmail.com> References: <488ED837.2060509@alcatel-lucent.com> <200807291713.29580.bzolnier@gmail.com> <488FBC6F.2080800@alcatel-lucent.com> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Return-path: Received: from fk-out-0910.google.com ([209.85.128.189]:4426 "EHLO fk-out-0910.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754731AbYG3TqK (ORCPT ); Wed, 30 Jul 2008 15:46:10 -0400 Received: by fk-out-0910.google.com with SMTP id 18so131605fkq.5 for ; Wed, 30 Jul 2008 12:46:09 -0700 (PDT) In-Reply-To: <488FBC6F.2080800@alcatel-lucent.com> Content-Disposition: inline Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: gshan Cc: Jeff Garzik , linux-kernel@vger.kernel.org, Linux IDE mailing list , "Maciej W. Rozycki" On Wednesday 30 July 2008, gshan wrote: [...] > The IRQ routing issue has been fixed. IRQ routing system > on my board is somewhat complicated: i8259->mlba (FPGA based > PCI device)->MV64460(system controller). I forgot to set MV64460 > IRQ pin to level sensitive and edge sensitive was used, which caused > interrupt lost finally. After changing this to level sensitive, everything > seems fine. Now, the problem isn't interrupt lost, but suspending DMA > operations. Thanks for explaining this. > > Also is the problem still present in 2.6.26 or 2.6.27-rc1? > > > > [ I don't see an information about the kernel version that is > > being used in either of threads? ] > > > Sorry, I'm using 2.6.21 kernel. Would be great to try with some recent one, the issue may have been fixed already (which is quite likely given a number of bugfixes that went in recent kernels). Thanks, Bart