From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx1.redhat.com ([209.132.183.28]:14334 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753780Ab2HDSJT (ORCPT ); Sat, 4 Aug 2012 14:09:19 -0400 From: Alex Williamson Subject: [PATCH 0/2] Fix ACS path checking To: joerg.roedel@amd.com, dwmw2@infradead.org Cc: iommu@lists.linux-foundation.org, linux-pci@vger.kernel.org, linux-kernel@vger.kernel.org, dsahern@gmail.com Date: Sat, 04 Aug 2012 12:08:46 -0600 Message-ID: <20120804180155.6144.83332.stgit@bling.home> MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Sender: linux-pci-owner@vger.kernel.org List-ID: David Ahern reported an oops caused by passing a NULL start device to pci_acs_path_enabled(). This can happen when a bus is created for sr-iov devices without an associated bridge. To handle this, skip over to the parent bus and look for a bridge there. Continue until we find something or reach the root bus. This works on my system, but still needs to be tested by David. I'd like to get this in for 3.6-rc2. Joerg, I notice that ACS isn't getting enabled on my AMD system on 3.6-rc1. I'll investigate more, but mention it in case you get to it first. Thanks, Alex --- Alex Williamson (2): amd-iommu: Fix ACS path checking intel-iommu: Fix ACS path checking drivers/iommu/amd_iommu.c | 25 ++++++++++++++++++++++--- drivers/iommu/intel-iommu.c | 25 ++++++++++++++++++++++--- 2 files changed, 44 insertions(+), 6 deletions(-)