From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx1.redhat.com ([209.132.183.28]:55735 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752834AbaFPPeX (ORCPT ); Mon, 16 Jun 2014 11:34:23 -0400 Message-ID: <1402932855.3707.40.camel@ul30vt.home> Subject: Re: [PATCH v4 00/16] PCI/iommu: Fix DMA alias problems From: Alex Williamson To: Joerg Roedel Cc: linux-pci@vger.kernel.org, iommu@lists.linux-foundation.org, bhelgaas@google.com, linux-kernel@vger.kernel.org, linux@horizon.com, eddy0596@gmail.com Date: Mon, 16 Jun 2014 09:34:15 -0600 In-Reply-To: <20140616144759.GC18986@8bytes.org> References: <20140522230230.2856.40017.stgit@bling.home> <20140616144759.GC18986@8bytes.org> Content-Type: text/plain; charset="UTF-8" Mime-Version: 1.0 Sender: linux-pci-owner@vger.kernel.org List-ID: On Mon, 2014-06-16 at 16:47 +0200, Joerg Roedel wrote: > Hi Alex, > > On Thu, May 22, 2014 at 05:07:23PM -0600, Alex Williamson wrote: > > Alex Williamson (16): > > PCI: Add DMA alias iterator > > PCI: define pci_dev_flags as bit shifts > > PCI: quirk pci_for_each_dma_alias() > > PCI: quirk dma_alias_devfn for Ricoh devices > > PCI: quirk dma_alias_devfn for Marvell devices > > PCI: Quirk pci_for_each_dma_alias() for bridges > > PCI: Add quirks for ASMedia and Tundra bridges > > iommu: Create central IOMMU group lookup/creation interface > > iommu/amd: Update to use PCI DMA aliases > > iommu/amd: Use iommu_group_get_for_dev() > > iommu/intel: Use iommu_group_get_for_dev() > > iommu/intel: Update to use PCI DMA aliases > > iommu/fsl: Use iommu_group_get_for_dev() for IOMMU groups > > iommu: Remove pci.h > > PCI: Remove pci_find_upstream_pcie_bridge() > > PCI: Remove pci_get_dma_source() > > Sorry for the delay, I had a look at the generic IOMMU and the AMD part > now. It looks good to me so far, but I still have to review the VT-d > changes and give it all some testing on my machines. I really like the > code simplification in the IOMMU drivers and also feel more comfortable > when the IVRS table is still taken into consideration for getting > aliases. Great, thanks Joerg. Let me know if you'd like a new series with just the IOMMU bits. Thanks, Alex