From mboxrd@z Thu Jan 1 00:00:00 1970 From: Joerg Roedel Subject: Re: [PATCH 0/28] AMD IOMMU interrupt remapping support Date: Fri, 6 Jul 2012 16:08:27 +0200 Message-ID: <20120706140827.GA2412@8bytes.org> References: <1341491808-23083-1-git-send-email-joerg.roedel@amd.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Content-Disposition: inline In-Reply-To: <1341491808-23083-1-git-send-email-joerg.roedel-5C7GfCeVMHo@public.gmane.org> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: iommu-bounces-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org Errors-To: iommu-bounces-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org To: Joerg Roedel Cc: iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org, linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: iommu@lists.linux-foundation.org On Thu, Jul 05, 2012 at 02:36:20PM +0200, Joerg Roedel wrote: > Hi, > > this patch-set adds support for IRQ remapping to the AMD IOMMU driver in > Linux. It works similar to the already present IRQ remapping code for > VT-d. The IOAPIC, HPET and MSI interrupts have a fixed setup to index an > interrupt remapping table entry and the real vector and destination is > only configured in the IRQ remapping table. This means that also the IRQ > affinity is only changed in the IOMMU. The code was heavily > stress-tested on a lot of machines and no known issues are left. > > I also pushed this code into a branch of the IOMMU tree: > > git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu.git irq-remapping Okay, I updated this branch (non-fast-forward) with my current state. It includes the fixes for Joe's objection and for all problems reported by Fengguang Wu. While at it I added a patch to fix most of the other sparse warnings in the AMD IOMMU driver. Thanks again for the reports guys :) I will wait for more comments and fixes to come up (including the rewrite of patch 3 as requested by Ingo) and repost the patch-set some time next week. Thanks so far, Joerg