From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andrew Morton Subject: Re: [patch 20/21] forcedeth: fix locking bug with netconsole Date: Fri, 28 Mar 2008 16:42:09 -0700 Message-ID: <20080328164209.e8b723d8.akpm@linux-foundation.org> References: <200803282141.m2SLfUwU011858@imap1.linux-foundation.org> <20080328.154353.43255092.davem@davemloft.net> <47ED75CE.1080400@garzik.org> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: David Miller , netdev@vger.kernel.org, mingo@elte.hu, aabdulla@nvidia.com To: Jeff Garzik Return-path: Received: from smtp1.linux-foundation.org ([140.211.169.13]:35980 "EHLO smtp1.linux-foundation.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756444AbYC1Xm3 (ORCPT ); Fri, 28 Mar 2008 19:42:29 -0400 In-Reply-To: <47ED75CE.1080400@garzik.org> Sender: netdev-owner@vger.kernel.org List-ID: On Fri, 28 Mar 2008 18:48:46 -0400 Jeff Garzik wrote: > David Miller wrote: > > From: akpm@linux-foundation.org > > Date: Fri, 28 Mar 2008 14:41:30 -0700 > > > >> From: Ingo Molnar > >> > >> While using netconsole on forcedeth, lockdep noticed the following locking > >> bug: > > > > Please do not mix device driver changes with core networking > > changes. > > FWIW my plan was to snip the net/* stuff and only apply the forcedeth > portion. > > (leaving to pick up the pieces, should they so desire) > Just drop 'em completely I'd say. otoh, drivers do seem a bit flakey in the netpoll-support area, so a bit of extra debug wouldn't hurt. But checking irqs_disabled() either side of a udelay() was a bit paranoid ;)