Hi all, Compared to v1[1] this is more or less a repost of the core parts, although patch 1 is new. I still need to take the horrible SMMUv2 code out back and shoot it (there turned out to be some subtle nasties in v1), so in the meantime I picked up the SMMUv3 ticket to fill in as it was rather more straightforward. I'll be reworking SMMUv2 on top of these patches (modulo any feedback) to post in another week or so. Note that patch 6 might not be plausible to queue just yet as it ends up looking like a regression due to deficiencies elsewhere - the virtio block device on the SMMUv3 Fast Model blows up because virtio doesn't use the DMA API appropriately on a host (although hacking vring_use_dma_api() suffices), and MSIs are still an open problem - I could really do with focusing on that with Eric, so it'd be nice to get as much of this out of the way as I can :) Branch at git://linux-arm.org/linux-rm iommu/generic-v2 Robin. [1]:http://thread.gmane.org/gmane.linux.kernel.iommu/12454 Mark Rutland (1): Docs: dt: add PCI IOMMU map bindings Robin Murphy (6): iommu/of: Respect disabled IOMMUs of/irq: Break out msi-map lookup (again) iommu/of: Handle iommu-map property for PCI iommu/arm-smmu: Implement of_xlate() for SMMUv3 iommu/arm-smmu: Finish off SMMUv3 default domain support iommu/arm-smmu: Support non-PCI devices with SMMUv3 .../devicetree/bindings/pci/pci-iommu.txt | 171 ++++++++++++ drivers/iommu/Kconfig | 2 +- drivers/iommu/arm-smmu-v3.c | 300 +++++++++------------ drivers/iommu/of_iommu.c | 73 +++-- drivers/of/irq.c | 70 +---- drivers/of/of_pci.c | 102 +++++++ include/linux/of_pci.h | 8 + 7 files changed, 460 insertions(+), 266 deletions(-) create mode 100644 Documentation/devicetree/bindings/pci/pci-iommu.txt -- 2.8.1.dirty -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
