From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753078Ab0CHJBp (ORCPT ); Mon, 8 Mar 2010 04:01:45 -0500 Received: from metis.ext.pengutronix.de ([92.198.50.35]:52648 "EHLO metis.ext.pengutronix.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753006Ab0CHJBm (ORCPT ); Mon, 8 Mar 2010 04:01:42 -0500 Date: Mon, 8 Mar 2010 10:01:39 +0100 From: Uwe =?iso-8859-1?Q?Kleine-K=F6nig?= To: Thomas Gleixner Cc: linux-kernel@vger.kernel.org, Ingo Molnar Subject: Re: [PATCH v2] genirq: spurious irq detection for threaded irqs Message-ID: <20100308090139.GA27502@pengutronix.de> References: <1267802291-5197-1-git-send-email-u.kleine-koenig@pengutronix.de> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: User-Agent: Mutt/1.5.18 (2008-05-17) X-SA-Exim-Connect-IP: 2001:6f8:1178:2:215:17ff:fe12:23b0 X-SA-Exim-Mail-From: ukl@pengutronix.de X-SA-Exim-Scanned: No (on metis.ext.pengutronix.de); SAEximRunCond expanded to false X-PTX-Original-Recipient: linux-kernel@vger.kernel.org Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hey Thomas, On Fri, Mar 05, 2010 at 09:16:53PM +0100, Thomas Gleixner wrote: > On Fri, 5 Mar 2010, Uwe Kleine-König wrote: > > > For threaded irqs the top half returns IRQ_WAKE_THREAD. Don't treat > > that value like IRQ_HANDLED for the spurious check. Instead check the > > return value of the threaded handler to be able to detect stuck irqs > > that only have threaded handlers and no top half that can detect the > > problem. > > > > Signed-off-by: Uwe Kleine-König > > Cc: Thomas Gleixner > > Cc: Ingo Molnar > > --- > > Hello, > > > > changed since v1: > > > > - use note_interrupt for the threaded handler's return value, getting rid of > > note_threaded_interrupt. The only downside is that a threaded handler > > returning IRQ_WAKE_THREAD remains uncatched now. > > Well that's easy to fix. > > > + if (!noirqdebug) > > + note_interrupt(action->irq, desc, action_ret); > > note_interrupt(action->irq, desc, > action_ret & ~IRQ_WAKE_THREAD); Hmmm, this makes IRQ_WAKE_THREAD be noted as IRQ_NONE. If this is intended it IMHO deserves a comment. Something like: /* * The threaded handler must return IRQ_NONE or IRQ_HANDLED. * As IRQ_WAKE_THREAD is handled special by note_interrupt * report it as IRQ_NONE. */ Well and with that approach (IRQ_WAKE_THREAD | IRQ_HANDLED) is handled as IRQ_HANDLED, but I think I start wasting time. Actually I don't care what we do here, should I resend with action_ret & ~IRQ_WAKE_THREAD and the comment? Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-König | Industrial Linux Solutions | http://www.pengutronix.de/ |