From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754559Ab1KTBzR (ORCPT ); Sat, 19 Nov 2011 20:55:17 -0500 Received: from mail-vx0-f174.google.com ([209.85.220.174]:35981 "EHLO mail-vx0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752963Ab1KTBzQ (ORCPT ); Sat, 19 Nov 2011 20:55:16 -0500 From: Edward Donovan To: tglx@linutronix.de Cc: linux-kernel@vger.kernel.org, maciej.rutecki@gmail.com Subject: [PATCH] genirq: fix second 2.6.39 regression in irqfixup, irqpoll Date: Sat, 19 Nov 2011 20:54:22 -0500 Message-Id: <1321754063-2887-1-git-send-email-edward.donovan@numble.net> X-Mailer: git-send-email 1.7.5.4 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi Thomas - I did isolate another regression in the bad-irq handling, and this patch sets things right on my machines. After bisecting it, I've just been troubleshooting to find which code was making the difference. This is my attempt to restore it, in a reasonable place. I don't know if this is the way you'd like to do it in the reworked IRQ code, since I'm not very familiar with the system yet. This is the most C I've dealt with, and partly I'm just moving puzzle pieces around. I have tested it a good bit, though, and functionally it has been fine. So, let me know if you have a preferred patch, and I'll test it on these machines. Thanks! Edward