From: David Miller Subject: Re: [PATCH] bnx2: Fix panic in bnx2_poll_work(). Date: Wed, 06 May 2009 16:47:36 -0700 (PDT) Message-ID: <20090506.164736.36849446.davem@davemloft.net> References: <1241630636-21999-1-git-send-email-mchan@broadcom.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org, pmjdebruijn@pcode.nl, hnoefer@pironet-ndh.com To: mchan@broadcom.com Return-path: Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:43713 "EHLO sunset.davemloft.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752248AbZEFXrk (ORCPT ); Wed, 6 May 2009 19:47:40 -0400 In-Reply-To: <1241630636-21999-1-git-send-email-mchan@broadcom.com> Sender: netdev-owner@vger.kernel.org List-ID: From: "Michael Chan" Date: Wed, 6 May 2009 10:23:56 -0700 > Add barrier() to bnx2_get_hw_{tx|rx}_cons() to fix this issue: > > http://bugzilla.kernel.org/show_bug.cgi?id=12698 > > This issue was reported by multiple i386 users. Without barrier(), > the compiled code looks like the following where %eax contains the > address of the tx_cons or rx_cons in the DMA status block. The > status block contents can change between the cmpb and the movzwl > instruction. The driver would crash if the value was not 0xff during > the cmpb instruction, but changed to 0xff during the movzwl > instruction. ... > Signed-off-by: Michael Chan Fair enough, applied, thanks!