From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tejun Heo Subject: Re: Possibly SATA related freeze killed networking and RAID Date: Fri, 30 Nov 2007 09:02:21 +0900 Message-ID: <474F530D.8090302@gmail.com> References: <20071120220512.46b9e975@the-village.bc.nu> <20071126120649.GC4701@ucw.cz> <474CCA82.7030000@gmail.com> <474F3A4B.3080304@cfl.rr.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: Received: from wa-out-1112.google.com ([209.85.146.176]:50160 "EHLO wa-out-1112.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1763598AbXK3ACa (ORCPT ); Thu, 29 Nov 2007 19:02:30 -0500 Received: by wa-out-1112.google.com with SMTP id v27so2453309wah for ; Thu, 29 Nov 2007 16:02:28 -0800 (PST) In-Reply-To: <474F3A4B.3080304@cfl.rr.com> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Phillip Susi Cc: Pavel Machek , Alan Cox , noah , Linux Kernel Mailing List , linux-ide@vger.kernel.org Phillip Susi wrote: > Tejun Heo wrote: >> Agreed. Nobody cared on ATA controllers is usually very effective at >> taking the whole machine down. Is there any reason why we don't turn on >> irqpoll on turned off IRQs automatically? > > Why does a single spurious interrupt cause it to be shut down? I can > see if the interrupt is stuck on and keeps interrupting constantly, but > if it's just the occasional spurious interrupt, why not just ignore it > and move on? Because SFF ATA controller don't have IRQ pending bit. You don't know whether IRQ is raised or not. Plus, accessing the status register which clears pending IRQ can be very slow on PATA machines. It has to go through the PCI and ATA bus and come back. So, unconditionally trying to clear IRQ by accessing Status can incur noticeable overhead if the IRQ is shared with devices which raise a lot of IRQs. -- tejun