From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH] [NET]: Fix Ooops of napi net_rx_action. Date: Tue, 11 Dec 2007 04:41:49 -0800 (PST) Message-ID: <20071211.044149.217878270.davem@davemloft.net> References: <20071211.023218.31965443.davem@davemloft.net> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: joonwpark81@gmail.com, netdev@vger.kernel.org, linux-kernel@vger.kernel.org To: herbert@gondor.apana.org.au Return-path: Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:53622 "EHLO sunset.davemloft.net" rhost-flags-OK-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1751728AbXLKMlt (ORCPT ); Tue, 11 Dec 2007 07:41:49 -0500 In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: From: Herbert Xu Date: Tue, 11 Dec 2007 20:36:21 +0800 > David Miller wrote: > > > > How can the NAPI_STATE_SCHED bit be cleared externally yet we take > > this list_move_tail() code path? > > His driver is probably buggy. When we had two drivers beginning > with e100 we often forgot to apply fixes to the both of them. Now > that we have three it's even more confusing. > > I just checked and indeed e1000e seems to be missing the NAPI fix > that was applied to e1000. Of course it doesn't rule out the > possibility of another NAPI bug in e1000. Thanks for checking. Indeed I stuck the huge comment there in net_rx_action() above the list move to try and explain things to people, so that if you saw a crash in the list manipulation, you're go check the driver first.