From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262946AbUBZTQo (ORCPT ); Thu, 26 Feb 2004 14:16:44 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262948AbUBZTQm (ORCPT ); Thu, 26 Feb 2004 14:16:42 -0500 Received: from chaos.analogic.com ([204.178.40.224]:12161 "EHLO chaos.analogic.com") by vger.kernel.org with ESMTP id S262946AbUBZTQB (ORCPT ); Thu, 26 Feb 2004 14:16:01 -0500 Date: Thu, 26 Feb 2004 14:18:31 -0500 (EST) From: "Richard B. Johnson" X-X-Sender: root@chaos Reply-To: root@chaos.analogic.com To: Jim Deas cc: linux-kernel@vger.kernel.org Subject: Re: kernel disables interrupts In-Reply-To: <200402261025.AA3240886544@jadsystems.com> Message-ID: References: <200402261025.AA3240886544@jadsystems.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 26 Feb 2004, Jim Deas wrote: > I am trouble shooting a new driver and have found a new > kernel item that makes trouble shooting a bit harder. > When I unload my test driver and before I can reload it > (reseting the interrut controls)the Kernerl disables > the chattering interrupt. > Once the kernel has disable a spurious interrupt is there > a way to get it back? > > JD Linux version 2.4.24 doesn't "disable" a spurious interrupt! When you unload the driver, you deallocate the interrupt you are using (I sure hope so). When you load the driver again, you reallocate the interrupt. If you FAILED to release the IRQ when you unloaded the module, guess what? It's still allocated and you can't have it anymore. That's probably what's happening. Cheers, Dick Johnson Penguin : Linux version 2.4.24 on an i686 machine (797.90 BogoMips). Note 96.31% of all statistics are fiction.