From mboxrd@z Thu Jan 1 00:00:00 1970 From: Danny Feng Subject: Re: e1000e: why does pci_enable_pcie_error_reporting() fail on my hp2510p? Date: Fri, 21 Aug 2009 10:08:46 +0800 Message-ID: <4A8E01AE.5070900@redhat.com> References: <200908201714.30638.elendil@planet.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Cc: Netdev , linux-kernel@vger.kernel.org, Jeff Kirsher To: Frans Pop Return-path: Received: from mx1.redhat.com ([209.132.183.28]:61750 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751222AbZHUCIl (ORCPT ); Thu, 20 Aug 2009 22:08:41 -0400 In-Reply-To: <200908201714.30638.elendil@planet.nl> Sender: netdev-owner@vger.kernel.org List-ID: On 08/20/2009 11:14 PM, Frans Pop wrote: > With the current kernel I suddenly get the following error displayed > during boot: > e1000e: Intel(R) PRO/1000 Network Driver - 1.0.2-k2 > e1000e: Copyright (c) 1999-2008 Intel Corporation. > e1000e 0000:00:19.0: PCI INT A -> GSI 22 (level, low) -> IRQ 22 > ! e1000e 0000:00:19.0: pci_enable_pcie_error_reporting failed 0xfffffffb > e1000e 0000:00:19.0: setting latency timer to 64 > e1000e 0000:00:19.0: irq 26 for MSI/MSI-X > > This looks to be the consequence of: > commit 68eac4602b9104cdaa6c18b3edd914cececa6a1e > Author: Xiaotian Feng > Date: Fri Aug 14 14:35:52 2009 +0000 > e1000e: fix use of pci_enable_pcie_error_reporting > > Looking at the change it seems logical that this error, that I previously > only saw during resumes from suppend to RAM, now also shows up during > system boot. So I don't think it can be qualified as a regression, but it > is certainly rather annoying. > > Questions: > - is it possible that the error occurs due to a kernel bug, or is this > certain to be due to broken hardware? I think this messages comes from hardware does not support pcie aer. In fact the messages is not fatal and can be ignored. Some drivers is silence with pci_enable_pcie_error_reporting .... My patch is just to fix the logic of pcie aer usage ... > - as the error is non-fatal, should it maybe be changed from dev_err > to dev_info, so that at least it does not show up during a boot > with 'quiet'? I agree with this.... Maybe some of other drivers needs this too (igb,ixgbe)... > - if broken hardware is common, should the kernel complain at all, > or at least maybe show a slightly "friendlier" message? > > System: hp2510p (ICH8, Core2 Duo) running x86_64, Debian stable ("Lenny") > > 00:19.0 Ethernet controller [0200]: Intel Corporation 82566MM > Gigabit Network Connection [8086:1049] (rev 03) > Subsystem: Hewlett-Packard Company Device 30c9 > Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+ > Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast>TAbort-SERR- Latency: 0 > Interrupt: pin A routed to IRQ 26 > Region 0: Memory at e0620000 (32-bit, non-prefetchable) [size=128K] > Region 1: Memory at e0640000 (32-bit, non-prefetchable) [size=4K] > Region 2: I/O ports at 2060 [size=32] > Capabilities: [c8] Power Management version 2 > Flags: PMEClk- DSI+ D1- D2- AuxCurrent=0mA PME(D0+,D1-,D2-,D3hot+,D3cold+) > Status: D0 PME-Enable- DSel=0 DScale=1 PME- > Capabilities: [d0] Message Signalled Interrupts: Mask- 64bit+ Queue=0/0 Enable+ > Address: 00000000fee0300c Data: 41b1 > Kernel driver in use: e1000e > Kernel modules: e1000e > > I can provide full dmesg and kernel config if desired. > > Cheers, > FJP >