From mboxrd@z Thu Jan 1 00:00:00 1970 From: Alex Williamson Subject: [RFC PATCH 0/2] pci/iommu: PCIe requester ID interface Date: Wed, 10 Jul 2013 16:10:32 -0600 Message-ID: <20130710215954.3045.89568.stgit@bling.home> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: 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: bhelgaas-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org Cc: linux-pci-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org, dwmw2-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org List-Id: iommu@lists.linux-foundation.org I took Bjorn's requester ID interface, or more appropriately my interpretation of the interface, through a full (apparently) working implementation. This is the result. It's quite a nice cleanup for intel-iommu. I think this kills two birds with one stone, we quirk PCIe-to-PCI bridges that don't have a PCIe capability (bz44881) and we also apply DMA quirks to all devices for dma_ops, which should fix devices like Ricoh and Marvell chips that do DMA from the wrong requester ID (still no ghost requester ID support, but that may be easier to add now). If this seems like the right direction I'll update other users of pci_find_upstream_pcie_bridge() and abolish that function. Thanks, Alex --- Alex Williamson (2): pci: Create PCIe requester ID interface iommu/intel: Make use of PCIe requester ID interface drivers/iommu/intel-iommu.c | 164 ++++++++++++++++------------------------- drivers/pci/search.c | 170 +++++++++++++++++++++++++++++++++++++++++++ include/linux/pci.h | 7 ++