From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Fri, 2 Nov 2001 08:34:23 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Fri, 2 Nov 2001 08:34:13 -0500 Received: from berta.E-Technik.Uni-Dortmund.DE ([129.217.182.12]:52229 "HELO kt.e-technik.uni-dortmund.de") by vger.kernel.org with SMTP id ; Fri, 2 Nov 2001 08:33:56 -0500 Date: Fri, 2 Nov 2001 14:33:54 +0100 From: Wolfgang Wegner To: Alan Cox Cc: linux-kernel@vger.kernel.org Subject: Re: PCI IRQ (routing) problem with mediaGX Message-ID: <20011102143354.E30033@bigmac.e-technik.uni-dortmund.de> In-Reply-To: <20011026170143.B15834@bigmac.e-technik.uni-dortmund.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.3.16i In-Reply-To: ; from alan@lxorguk.ukuu.org.uk on Fri, Oct 26, 2001 at 05:04:27PM +0100 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hi again, sorry for making such a fuzz about such a simple problem... :) I dropped natsemi.c from 2.4.4 in 2.4.13-ac5, and now network works flawlessly, although the strange error message still remains: (dmesg output): natsemi.c:v1.07 1/9/2001 Written by Donald Becker http://www.scyld.com/network/natsemi.html (unofficial 2.4.x kernel port, version 1.0.5, April 17, 2001 Jeff Garzik, Tjeerd Mulder) PCI: Found IRQ 10 for device 00:0e.0 IRQ routing conflict for 00:0e.0, have irq 11, want irq 10 eth0: NatSemi DP83815 at 0xc281f000, fb:ff:fb:ff:fb:ff, IRQ 11. eth0: Transceiver status 0x7869 advertising 05e1. The non-working natsemi.c dmesg output looks like this: natsemi.c:v1.07 1/9/2001 Written by Donald Becker http://www.scyld.com/network/natsemi.html (unofficial 2.4.x kernel port, version 1.07+LK1.0.11, Oct 19, 2001 Jeff Garzi k, Tjeerd Mulder) PCI: Found IRQ 10 for device 00:0e.0 IRQ routing conflict for 00:0e.0, have irq 11, want irq 10 eth%d: EEPROM did not reload in 2000 usec. eth0: NatSemi DP8381[56] at 0xc2820000, fb:ff:fb:ff:fb:ff, IRQ 11. eth0: Transceiver status 0x7869 advertising 05e1. eth0: link is back. Enabling watchdog. eth0: Link changed: Autonegotiation advertising 05e1 partner 0000. eth0: no link. Disabling watchdog. eth0: Link wake-up event 00400617 eth0: Link changed: Autonegotiation advertising 05e1 partner 0000. Note the additional transceiver messages! The only difference of the first one to original 2.4.4 is the strange occurence of IRQ10, but at least it works now... :) (2.4.9 and 2.4.13 without any patches contain the same natsemi.c and thus do _not_ work either) In case there is anything I can do to help solve this strange IRQ mapping inconsistency, let me know! Thanks to everybody! Wolfgang