From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jesse Brandeburg Date: Tue, 21 Nov 2017 09:39:34 -0800 Subject: [Intel-wired-lan] [PATCH v2 6/7] fm10k: Use smp_rmb rather than read_barrier_depends In-Reply-To: <1510938349-17608-7-git-send-email-brking@linux.vnet.ibm.com> References: <1510938349-17608-1-git-send-email-brking@linux.vnet.ibm.com> <1510938349-17608-7-git-send-email-brking@linux.vnet.ibm.com> Message-ID: <20171121093934.00003075@intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: intel-wired-lan@osuosl.org List-ID: On Fri, 17 Nov 2017 11:05:48 -0600 Brian King wrote: > The original issue being fixed in this patch was seen with the ixgbe > driver, but the same issue exists with fm10k as well, as the code is > very similar. read_barrier_depends is not sufficient to ensure > loads following it are not speculatively loaded out of order > by the CPU, which can result in stale data being loaded, causing > potential system crashes. Acked-by: Jesse Brandeburg From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga14.intel.com ([192.55.52.115]:60001 "EHLO mga14.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751191AbdKURjy (ORCPT ); Tue, 21 Nov 2017 12:39:54 -0500 Date: Tue, 21 Nov 2017 09:39:34 -0800 From: Jesse Brandeburg To: Brian King Cc: , , , , , , Subject: Re: [PATCH v2 6/7] fm10k: Use smp_rmb rather than read_barrier_depends Message-ID: <20171121093934.00003075@intel.com> In-Reply-To: <1510938349-17608-7-git-send-email-brking@linux.vnet.ibm.com> References: <1510938349-17608-1-git-send-email-brking@linux.vnet.ibm.com> <1510938349-17608-7-git-send-email-brking@linux.vnet.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: stable-owner@vger.kernel.org List-ID: On Fri, 17 Nov 2017 11:05:48 -0600 Brian King wrote: > The original issue being fixed in this patch was seen with the ixgbe > driver, but the same issue exists with fm10k as well, as the code is > very similar. read_barrier_depends is not sufficient to ensure > loads following it are not speculatively loaded out of order > by the CPU, which can result in stale data being loaded, causing > potential system crashes. Acked-by: Jesse Brandeburg