summaryrefslogtreecommitdiff
path: root/drivers/iommu
AgeCommit message (Collapse)Author
2015-05-11iommu/arm-smmu: Fix sign-extension of upstream bus addresses at stage 1Will Deacon
Stage 1 translation is controlled by two sets of page tables (TTBR0 and TTBR1) which grow up and down from zero respectively in the ARMv8 translation regime. For the SMMU, we only care about TTBR0 and, in the case of a 48-bit virtual space, we expect to map virtual addresses 0x0 through to 0xffff_ffff_ffff. Given that some masters may be incapable of emitting virtual addresses targetting TTBR1 (e.g. because they sit on a 48-bit bus), the SMMU architecture allows bit 47 to be sign-extended, halving the virtual range of TTBR0 but allowing TTBR1 to be used. This is controlled by the SEP field in TTBCR2. The SMMU driver incorrectly enables this sign-extension feature, which causes problems when userspace addresses are programmed into a master device with the SMMU expecting to map the incoming transactions via TTBR0; if the top bit of address is set, we will instead get a translation fault since TTBR1 walks are disabled in the TTBCR. This patch fixes the issue by disabling sign-extension of a fixed virtual address bit and instead basing the behaviour on the upstream bus size: the incoming address is zero extended unless the upstream bus is only 49 bits wide, in which case bit 48 is used as the sign bit and is replicated to the upper bits. Cc: <stable@vger.kernel.org> # v4.0+ Reported-by: Varun Sethi <varun.sethi@freescale.com> Signed-off-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-05-05iommu/rockchip: Fix build without CONFIG_OFArnd Bergmann
The rockchip iommu driver references its of_device_id table from the init function, which fails to build when the table is undefined: iommu/rockchip-iommu.c: In function 'rk_iommu_init': iommu/rockchip-iommu.c:1029:35: error: 'rk_iommu_dt_ids' undeclared (first use in this function) np = of_find_matching_node(NULL, rk_iommu_dt_ids); This removes the #ifdef and the corresponding of_match_ptr wrapper to make it build both with CONFIG_OF enabled or disabled. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Fixes: 425061b0f5074 ("iommu/rockchip: Play nice in multi-platform builds") Reviewed-by: Thierry Reding <treding@nvidia.com> Reviewed-by: Heiko Stuebner <heiko@sntech.de> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-05-04iommu/amd: Fix bug in put_pasid_state_waitOded Gabbay
This patch fixes a bug in put_pasid_state_wait that appeared in kernel 4.0 The bug is that pasid_state->count wasn't decremented before entering the wait_event. Thus, the condition in wait_event will never be true. The fix is to decrement (atomically) the pasid_state->count before the wait_event. Signed-off-by: Oded Gabbay <oded.gabbay@amd.com> Cc: stable@vger.kernel.org #v4.0 Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-04-26Merge git://git.infradead.org/intel-iommuLinus Torvalds
Pull intel iommu updates from David Woodhouse: "This lays a little of the groundwork for upcoming Shared Virtual Memory support — fixing some bogus #defines for capability bits and adding the new ones, and starting to use the new wider page tables where we can, in anticipation of actually filling in the new fields therein. It also allows graphics devices to be assigned to VM guests again. This got broken in 3.17 by disallowing assignment of RMRR-afflicted devices. Like USB, we do understand why there's an RMRR for graphics devices — and unlike USB, it's actually sane. So we can make an exception for graphics devices, just as we do USB controllers. Finally, tone down the warning about the X2APIC_OPT_OUT bit, due to persistent requests. X2APIC_OPT_OUT was added to the spec as a nasty hack to allow broken BIOSes to forbid us from using X2APIC when they do stupid and invasive things and would break if we did. Someone noticed that since Windows doesn't have full IOMMU support for DMA protection, setting the X2APIC_OPT_OUT bit made Windows avoid initialising the IOMMU on the graphics unit altogether. This means that it would be available for use in "driver mode", where the IOMMU registers are made available through a BAR of the graphics device and the graphics driver can do SVM all for itself. So they started setting the X2APIC_OPT_OUT bit on *all* platforms with SVM capabilities. And even the platforms which *might*, if the planets had been aligned correctly, possibly have had SVM capability but which in practice actually don't" * git://git.infradead.org/intel-iommu: iommu/vt-d: support extended root and context entries iommu/vt-d: Add new extended capabilities from v2.3 VT-d specification iommu/vt-d: Allow RMRR on graphics devices too iommu/vt-d: Print x2apic opt out info instead of printing a warning iommu/vt-d: kill bogus ecap_niotlb_iunits()
2015-04-20Merge tag 'iommu-updates-v4.1' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu Pull IOMMU updates from Joerg Roedel: "Not much this time, but the changes include: - moving domain allocation into the iommu drivers to prepare for the introduction of default domains for devices - fixing the IO page-table code in the AMD IOMMU driver to correctly encode large page sizes - extension of the PCI support in the ARM-SMMU driver - various fixes and cleanups" * tag 'iommu-updates-v4.1' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu: (34 commits) iommu/amd: Correctly encode huge pages in iommu page tables iommu/amd: Optimize amd_iommu_iova_to_phys for new fetch_pte interface iommu/amd: Optimize alloc_new_range for new fetch_pte interface iommu/amd: Optimize iommu_unmap_page for new fetch_pte interface iommu/amd: Return the pte page-size in fetch_pte iommu/amd: Add support for contiguous dma allocator iommu/amd: Don't allocate with __GFP_ZERO in alloc_coherent iommu/amd: Ignore BUS_NOTIFY_UNBOUND_DRIVER event iommu/amd: Use BUS_NOTIFY_REMOVED_DEVICE iommu/tegra: smmu: Compute PFN mask at runtime iommu/tegra: gart: Set aperture at domain initialization time iommu/tegra: Setup aperture iommu: Remove domain_init and domain_free iommu_ops iommu/fsl: Make use of domain_alloc and domain_free iommu/rockchip: Make use of domain_alloc and domain_free iommu/ipmmu-vmsa: Make use of domain_alloc and domain_free iommu/shmobile: Make use of domain_alloc and domain_free iommu/msm: Make use of domain_alloc and domain_free iommu/tegra-gart: Make use of domain_alloc and domain_free iommu/tegra-smmu: Make use of domain_alloc and domain_free ...
2015-04-14Merge tag 'pm+acpi-4.1-rc1' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm Pull power management and ACPI updates from Rafael Wysocki: "These are mostly fixes and cleanups all over, although there are a few items that sort of fall into the new feature category. First off, we have new callbacks for PM domains that should help us to handle some issues related to device initialization in a better way. There also is some consolidation in the unified device properties API area allowing us to use that inferface for accessing data coming from platform initialization code in addition to firmware-provided data. We have some new device/CPU IDs in a few drivers, support for new chips and a new cpufreq driver too. Specifics: - Generic PM domains support update including new PM domain callbacks to handle device initialization better (Russell King, Rafael J Wysocki, Kevin Hilman) - Unified device properties API update including a new mechanism for accessing data provided by platform initialization code (Rafael J Wysocki, Adrian Hunter) - ARM cpuidle update including ARM32/ARM64 handling consolidation (Daniel Lezcano) - intel_idle update including support for the Silvermont Core in the Baytrail SOC and for the Airmont Core in the Cherrytrail and Braswell SOCs (Len Brown, Mathias Krause) - New cpufreq driver for Hisilicon ACPU (Leo Yan) - intel_pstate update including support for the Knights Landing chip (Dasaratharaman Chandramouli, Kristen Carlson Accardi) - QorIQ cpufreq driver update (Tang Yuantian, Arnd Bergmann) - powernv cpufreq driver update (Shilpasri G Bhat) - devfreq update including Tegra support changes (Tomeu Vizoso, MyungJoo Ham, Chanwoo Choi) - powercap RAPL (Running-Average Power Limit) driver update including support for Intel Broadwell server chips (Jacob Pan, Mathias Krause) - ACPI device enumeration update related to the handling of the special PRP0001 device ID allowing DT-style 'compatible' property to be used for ACPI device identification (Rafael J Wysocki) - ACPI EC driver update including limited _DEP support (Lan Tianyu, Lv Zheng) - ACPI backlight driver update including a new mechanism to allow native backlight handling to be forced on non-Windows 8 systems and a new quirk for Lenovo Ideapad Z570 (Aaron Lu, Hans de Goede) - New Windows Vista compatibility quirk for Sony VGN-SR19XN (Chen Yu) - Assorted ACPI fixes and cleanups (Aaron Lu, Martin Kepplinger, Masanari Iida, Mika Westerberg, Nan Li, Rafael J Wysocki) - Fixes related to suspend-to-idle for the iTCO watchdog driver and the ACPI core system suspend/resume code (Rafael J Wysocki, Chen Yu) - PM tracing support for the suspend phase of system suspend/resume transitions (Zhonghui Fu) - Configurable delay for the system suspend/resume testing facility (Brian Norris) - PNP subsystem cleanups (Peter Huewe, Rafael J Wysocki)" * tag 'pm+acpi-4.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm: (74 commits) ACPI / scan: Fix NULL pointer dereference in acpi_companion_match() ACPI / scan: Rework modalias creation when "compatible" is present intel_idle: mark cpu id array as __initconst powercap / RAPL: mark rapl_ids array as __initconst powercap / RAPL: add ID for Broadwell server intel_pstate: Knights Landing support intel_pstate: remove MSR test cpufreq: fix qoriq uniprocessor build ACPI / scan: Take the PRP0001 position in the list of IDs into account ACPI / scan: Simplify acpi_match_device() ACPI / scan: Generalize of_compatible matching device property: Introduce firmware node type for platform data device property: Make it possible to use secondary firmware nodes PM / watchdog: iTCO: stop watchdog during system suspend cpufreq: hisilicon: add acpu driver ACPI / EC: Call acpi_walk_dep_device_list() after installing EC opregion handler cpufreq: powernv: Report cpu frequency throttling intel_idle: Add support for the Airmont Core in the Cherrytrail and Braswell SOCs intel_idle: Update support for Silvermont Core in Baytrail SOC PM / devfreq: tegra: Register governor on module init ...
2015-04-13Merge tag 'pci-v4.1-changes' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci Pull PCI changes from Bjorn Helgaas: "Enumeration - Read capability list as dwords, not bytes (Sean O. Stalley) Resource management - Don't check for PNP overlaps with unassigned PCI BARs (Bjorn Helgaas) - Mark invalid BARs as unassigned (Bjorn Helgaas) - Show driver, BAR#, and resource on pci_ioremap_bar() failure (Bjorn Helgaas) - Fail pci_ioremap_bar() on unassigned resources (Bjorn Helgaas) - Assign resources before drivers claim devices (Yijing Wang) - Claim bus resources before pci_bus_add_devices() (Yijing Wang) Power management - Optimize device state transition delays (Aaron Lu) - Don't clear ASPM bits when the FADT declares it's unsupported (Matthew Garrett) Virtualization - Add ACS quirks for Intel 1G NICs (Alex Williamson) IOMMU - Add ptr to OF node arg to of_iommu_configure() (Murali Karicheri) - Move of_dma_configure() to device.c to help re-use (Murali Karicheri) - Fix size when dma-range is not used (Murali Karicheri) - Add helper functions pci_get[put]_host_bridge_device() (Murali Karicheri) - Add of_pci_dma_configure() to update DMA configuration (Murali Karicheri) - Update DMA configuration from DT (Murali Karicheri) - dma-mapping: limit IOMMU mapping size (Murali Karicheri) - Calculate device DMA masks based on DT dma-range size (Murali Karicheri) ARM Versatile host bridge driver - Check for devm_ioremap_resource() failures (Jisheng Zhang) Broadcom iProc host bridge driver - Add Broadcom iProc PCIe driver (Ray Jui) Marvell MVEBU host bridge driver - Add suspend/resume support (Thomas Petazzoni) Renesas R-Car host bridge driver - Fix position of MSI enable bit (Nobuhiro Iwamatsu) - Write zeroes to reserved PCIEPARL bits (Nobuhiro Iwamatsu) - Change PCIEPARL and PCIEPARH to PCIEPALR and PCIEPAUR (Nobuhiro Iwamatsu) - Verify that mem_res is 64K-aligned (Nobuhiro Iwamatsu) Samsung Exynos host bridge driver - Fix INTx enablement statement termination error (Jaehoon Chung) Miscellaneous - Make a shareable UUID for PCI firmware ACPI _DSM (Aaron Lu) - Clarify policy for vendor IDs in pci.txt (Michael S. Tsirkin)" * tag 'pci-v4.1-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci: (36 commits) PCI: Read capability list as dwords, not bytes PCI: layerscape: Simplify platform_get_resource_byname() failure checking PCI: keystone: Don't dereference possible NULL pointer PCI: versatile: Check for devm_ioremap_resource() failures PCI: Don't clear ASPM bits when the FADT declares it's unsupported PCI: Clarify policy for vendor IDs in pci.txt PCI/ACPI: Optimize device state transition delays PCI: Export pci_find_host_bridge() for use inside PCI core PCI: Make a shareable UUID for PCI firmware ACPI _DSM PCI: Fix typo in Thunderbolt kernel message PCI: exynos: Fix INTx enablement statement termination error PCI: iproc: Add Broadcom iProc PCIe support PCI: iproc: Add DT docs for Broadcom iProc PCIe driver PCI: Export symbols required for loadable host driver modules PCI: Add ACS quirks for Intel 1G NICs PCI: mvebu: Add suspend/resume support PCI: Cleanup control flow sparc/PCI: Claim bus resources before pci_bus_add_devices() PCI: Assign resources before drivers claim devices (pci_scan_root_bus()) PCI: Fail pci_ioremap_bar() on unassigned resources ...
2015-04-13Merge branch 'device-properties'Rafael J. Wysocki
* device-properties: device property: Introduce firmware node type for platform data device property: Make it possible to use secondary firmware nodes driver core: Implement device property accessors through fwnode ones driver core: property: Update fwnode_property_read_string_array() driver core: Add comments about returning array counts ACPI: Introduce has_acpi_companion() driver core / ACPI: Represent ACPI companions using fwnode_handle
2015-04-02Merge branches 'iommu/fixes', 'x86/vt-d', 'x86/amd', 'arm/smmu', 'arm/tegra' ↵Joerg Roedel
and 'core' into next Conflicts: drivers/iommu/amd_iommu.c drivers/iommu/tegra-gart.c drivers/iommu/tegra-smmu.c
2015-04-02iommu/amd: Correctly encode huge pages in iommu page tablesJoerg Roedel
When a default page-size for given level should be mapped, the level encoding must be 0 rather than 7. This fixes an issue seen on IOMMUv2 hardware, where this encoding is enforced. Tested-by: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-04-02iommu/amd: Optimize amd_iommu_iova_to_phys for new fetch_pte interfaceJoerg Roedel
Now that fetch_pte returns the page-size of the pte, this function can be optimized too. Tested-by: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-04-02iommu/amd: Optimize alloc_new_range for new fetch_pte interfaceJoerg Roedel
Now that fetch_pte returns the page-size of the pte, the call in this function can also be optimized a little bit. Tested-by: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-04-02iommu/amd: Optimize iommu_unmap_page for new fetch_pte interfaceJoerg Roedel
Now that fetch_pte returns the page-size of the pte, this function can be optimized a lot. Tested-by: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-04-02iommu/amd: Return the pte page-size in fetch_pteJoerg Roedel
Extend the fetch_pte function to also return the page-size that is mapped by the returned pte. Tested-by: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-04-02iommu/amd: Add support for contiguous dma allocatorJoerg Roedel
Add code to allocate memory from the contiguous memory allocator to support coherent allocations larger than 8MB. Tested-by: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-04-02iommu/amd: Don't allocate with __GFP_ZERO in alloc_coherentJoerg Roedel
Don't explicitly add __GFP_ZERO to the allocator flags. Leave this up to the caller. Tested-by: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-04-02iommu/amd: Ignore BUS_NOTIFY_UNBOUND_DRIVER eventJoerg Roedel
Detaching a device from its domain at this event is problematic for several reasons: * The device might me in an alias group and detaching it will also detach all other devices in the group. This removes valid DMA mappings from the other devices causing io-page-faults and lets these devices fail. * Devices might have unity mappings specified by the IVRS table. These mappings are required for the device even when no device driver is attached. Detaching the device from its domain in driver unbind will also remove these unity mappings. This patch removes the handling of the BUS_NOTIFY_UNBOUND_DRIVER event to prevent these issues and align it better with the behavior of the VT-d driver. Tested-by: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-04-02iommu/amd: Use BUS_NOTIFY_REMOVED_DEVICEJoerg Roedel
Use the new device-notifier event instead of the old BUS_NOTIFY_DEL_DEVICE to make sure the device driver had a chance to uninit the device before all its mappings are teared down. Tested-by: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-31iommu/tegra: smmu: Compute PFN mask at runtimeThierry Reding
The SMMU on Tegra30 and Tegra114 supports addressing up to 4 GiB of physical memory. On Tegra124 the addressable physical memory was extended to 16 GiB. The page frame number stored in PTEs therefore requires 20 or 22 bits, depending on SoC generation. In order to cope with this, compute the proper value at runtime. Reported-by: Joseph Lo <josephl@nvidia.com> Cc: Hiroshi Doyu <hdoyu@nvidia.com> Signed-off-by: Thierry Reding <treding@nvidia.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-31iommu/tegra: gart: Set aperture at domain initialization timeThierry Reding
The aperture of the domain should always be available, otherwise drivers need to attach first before they can use the aperture geometry. Cc: Hiroshi Doyu <hdoyu@nvidia.com> Signed-off-by: Thierry Reding <treding@nvidia.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-31iommu/tegra: Setup apertureThierry Reding
Each address space in the Tegra SMMU provides 4 GiB worth of addresses. Cc: Hiroshi Doyu <hdoyu@nvidia.com> Signed-off-by: Thierry Reding <treding@nvidia.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-31iommu: Remove domain_init and domain_free iommu_opsJoerg Roedel
All drivers have been converted to the new domain_alloc and domain_free iommu-ops. So remove the old ones and get rid of iommu_domain->priv too, as this is no longer needed when the struct iommu_domain is embedded in the private structures of the iommu drivers. Tested-by: Thierry Reding <treding@nvidia.com> Tested-by: Heiko Stuebner <heiko@sntech.de> Reviewed-by: Alex Williamson <alex.williamson@redhat.com> Acked-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-31iommu/fsl: Make use of domain_alloc and domain_freeJoerg Roedel
Implement domain_alloc and domain_free iommu-ops as a replacement for domain_init/domain_destroy. Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-31iommu/rockchip: Make use of domain_alloc and domain_freeJoerg Roedel
Implement domain_alloc and domain_free iommu-ops as a replacement for domain_init/domain_destroy. Tested-by: Heiko Stuebner <heiko@sntech.de> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-31iommu/ipmmu-vmsa: Make use of domain_alloc and domain_freeJoerg Roedel
Implement domain_alloc and domain_free iommu-ops as a replacement for domain_init/domain_destroy. Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-31iommu/shmobile: Make use of domain_alloc and domain_freeJoerg Roedel
Implement domain_alloc and domain_free iommu-ops as a replacement for domain_init/domain_destroy. Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-31iommu/msm: Make use of domain_alloc and domain_freeJoerg Roedel
Implement domain_alloc and domain_free iommu-ops as a replacement for domain_init/domain_destroy. Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-31iommu/tegra-gart: Make use of domain_alloc and domain_freeJoerg Roedel
Implement domain_alloc and domain_free iommu-ops as a replacement for domain_init/domain_destroy. Tested-by: Thierry Reding <treding@nvidia.com> Acked-by: Thierry Reding <treding@nvidia.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-31iommu/tegra-smmu: Make use of domain_alloc and domain_freeJoerg Roedel
Implement domain_alloc and domain_free iommu-ops as a replacement for domain_init/domain_destroy. Tested-by: Thierry Reding <treding@nvidia.com> Acked-by: Thierry Reding <treding@nvidia.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-31iommu/exynos: Make use of domain_alloc and domain_freeJoerg Roedel
Implement domain_alloc and domain_free iommu-ops as a replacement for domain_init/domain_destroy. Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-31iommu/arm-smmu: Make use of domain_alloc and domain_freeJoerg Roedel
Implement domain_alloc and domain_free iommu-ops as a replacement for domain_init/domain_destroy. Acked-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-31iommu/omap: Make use of domain_alloc and domain_freeJoerg Roedel
Implement the new domain_alloc and domain_free call-backs and remove the old domain_init/destroy ones. Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-31iommu/vt-d: Make use of domain_alloc and domain_freeJoerg Roedel
Get rid of domain_init and domain_destroy and implement domain_alloc/domain_free instead. Reviewed-by: Alex Williamson <alex.williamson@redhat.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-31iommu/amd: Make use of domain_alloc and domain_freeJoerg Roedel
Implement the new iommu-ops function pointers and remove the obsolete domain_init and domain_destroy functions. Reviewed-by: Alex Williamson <alex.williamson@redhat.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-31iommu: Only allow iommu_map/unmap for paging domainsJoerg Roedel
Check for the new __IOMMU_DOMAIN_PAGING flag before calling into the iommu drivers ->map and ->unmap call-backs. Tested-by: Thierry Reding <treding@nvidia.com> Tested-by: Heiko Stuebner <heiko@sntech.de> Reviewed-by: Alex Williamson <alex.williamson@redhat.com> Acked-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-31iommu: Introduce iommu domain typesJoerg Roedel
This allows to handle domains differently based on their type in the future. An IOMMU driver can implement certain optimizations for DMA-API domains for example. The domain types can be extended later and some of the existing domain attributes can be migrated to become domain flags. Tested-by: Thierry Reding <treding@nvidia.com> Tested-by: Heiko Stuebner <heiko@sntech.de> Reviewed-by: Alex Williamson <alex.williamson@redhat.com> Acked-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-31iommu: Introduce domain_alloc and domain_free iommu_opsJoerg Roedel
These new call-backs defer the allocation and destruction of 'struct iommu_domain' to the iommu driver. This allows drivers to embed this struct into their private domain structures and to get rid of the domain_init and domain_destroy call-backs when all drivers have been converted. Tested-by: Thierry Reding <treding@nvidia.com> Tested-by: Heiko Stuebner <heiko@sntech.de> Reviewed-by: Alex Williamson <alex.williamson@redhat.com> Acked-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-27iommu/io-pgtable-arm: avoid speculative walks through TTBR1Will Deacon
Although we set TCR.T1SZ to 0, the input address range covered by TTBR1 is actually calculated using T0SZ in this case on the ARM SMMU. This could theoretically lead to speculative table walks through physical address zero, leading to all sorts of fun and games if we have MMIO regions down there. This patch avoids the issue by setting EPD1 to disable walks through the unused TTBR1 register. Signed-off-by: Will Deacon <will.deacon@arm.com>
2015-03-27iommu/arm-smmu: handle multi-alias IOMMU groups for PCI devicesWill Deacon
IOMMU groups for PCI devices can correspond to multiple DMA aliases due to things like ACS and PCI quirks. This patch extends the ARM SMMU ->add_device callback so that we consider all of the DMA aliases for a PCI IOMMU group, rather than creating a separate group for each Requester ID. Signed-off-by: Will Deacon <will.deacon@arm.com>
2015-03-27iommu/arm-smmu: set a more appropriate DMA maskRobin Murphy
Since we use dma_map_page() as an architecture-independent means of making page table updates visible to non-coherent SMMUs, we need to have a suitable DMA mask set to discourage the DMA mapping layer from creating bounce buffers and flushing those instead, if said page tables happen to lie outside the default 32-bit mask. Tested-by: Mitchel Humpherys <mitchelh@codeaurora.org> Signed-off-by: Robin Murphy <robin.murphy@arm.com> [will: added error checking] Signed-off-by: Will Deacon <will.deacon@arm.com>
2015-03-27iommu/arm-smmu: ensure CBA2R is initialised before CBAR on SMMUv2Will Deacon
The VMID16 (8.1) extension to SMMUv2 added a 16-bit VMID16 field to the CBA2R registers. Unfortunately, if software writes this field as zero after setting an 8-bit VMID in a stage-2 CBAR, then the VMID may also be overwritten with zero on some early implementations (the architecture was later updated to fix this issue). This patch ensures that we initialise CBA2R before CBAR, therefore ensuring that the VMID is set correctly. Tested-by: Manish Jaggi <mjaggi@caviumnetworks.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
2015-03-25iommu/vt-d: support extended root and context entriesDavid Woodhouse
Add a new function iommu_context_addr() which takes care of the differences and returns a pointer to a context entry which may be in either format. The formats are binary compatible for all the old fields anyway; the new one is just larger and some of the reserved bits in the original 128 are now meaningful. So far, nothing actually uses the new fields in the extended context entry. Modulo hardware bugs with interpreting the new-style tables, this should basically be a no-op. Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2015-03-25iommu/vt-d: Allow RMRR on graphics devices tooDavid Woodhouse
Commit c875d2c1 ("iommu/vt-d: Exclude devices using RMRRs from IOMMU API domains") prevents certain options for devices with RMRRs. This even prevents those devices from getting a 1:1 mapping with 'iommu=pt', because we don't have the code to handle *preserving* the RMRR regions when moving the device between domains. There's already an exclusion for USB devices, because we know the only reason for RMRRs there is a misguided desire to keep legacy keyboard/mouse emulation running in some theoretical OS which doesn't have support for USB in its own right... but which *does* enable the IOMMU. Add an exclusion for graphics devices too, so that 'iommu=pt' works there. We should be able to successfully assign graphics devices to guests too, as long as the initial handling of stolen memory is reconfigured appropriately. This has certainly worked in the past. Signed-off-by: David Woodhouse <David.Woodhouse@intel.com> Cc: stable@vger.kernel.org
2015-03-25iommu/vt-d: Print x2apic opt out info instead of printing a warningFenghua Yu
BIOS can set up x2apic_opt_out bit on some platforms, for various misguided reasons like insane SMM code with weird assumptions about what descriptors look like, or wanting Windows not to enable the IOMMU so that the graphics driver will take it over for SVM in "driver mode". A user can either disable the x2apic_opt_out bit in BIOS or by kernel parameter "no_x2apic_optout". Instead of printing a warning, we just print information of x2apic opt out. Signed-off-by: Fenghua Yu <fenghua.yu@intel.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2015-03-24iommu/vt-d: Remove unused variableAlex Williamson
Unused after commit 71684406905f ("iommu/vt-d: Detach domain *only* from attached iommus"). Reported by 0-day builder. Signed-off-by: Alex Williamson <alex.williamson@redhat.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-23iommu: ipmmu-vmsa: Add terminating entry for ipmmu_of_idsAxel Lin
The of_device_id table is supposed to be zero-terminated. Signed-off-by: Axel Lin <axel.lin@ingics.com> Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-23iommu/vt-d: Detach domain *only* from attached iommusAlex Williamson
Device domains never span IOMMU hardware units, which allows the domain ID space for each IOMMU to be an independent address space. Therefore we can have multiple, independent domains, each with the same domain->id, but attached to different hardware units. This is also why we need to do a heavy-weight search for VM domains since they can span multiple IOMMUs hardware units and we don't require a single global ID to use for all hardware units. Therefore, if we call iommu_detach_domain() across all active IOMMU hardware units for a non-VM domain, the result is that we clear domain IDs that are not associated with our domain, allowing them to be re-allocated and causing apparent coherency issues when the device cannot access IOVAs for the intended domain. This bug was introduced in commit fb170fb4c548 ("iommu/vt-d: Introduce helper functions to make code symmetric for readability"), but is significantly exacerbated by the more recent commit 62c22167dd70 ("iommu/vt-d: Fix dmar_domain leak in iommu_attach_device") which calls domain_exit() more frequently to resolve a domain leak. Fixes: fb170fb4c548 ("iommu/vt-d: Introduce helper functions to make code symmetric for readability") Signed-off-by: Alex Williamson <alex.williamson@redhat.com> Cc: Jiang Liu <jiang.liu@linux.intel.com> Cc: stable@vger.kernel.org # v3.17+ Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-23iommu/arm-smmu: fix ARM_SMMU_FEAT_TRANS_OPS conditionBaptiste Reynal
This patch is a fix to "iommu/arm-smmu: add support for iova_to_phys through ATS1PR". According to ARM documentation, translation registers are optional even in SMMUv1, so ID0_S1TS needs to be checked to verify their presence. Also, we check that the domain is a stage-1 domain. Signed-off-by: Baptiste Reynal <b.reynal@virtualopensystems.com> Acked-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-03-19Merge branches 'pci/iommu' and 'pci/resource' into nextBjorn Helgaas
* pci/iommu: of: Calculate device DMA masks based on DT dma-range size arm: dma-mapping: limit IOMMU mapping size PCI: Update DMA configuration from DT of/pci: Add of_pci_dma_configure() to update DMA configuration PCI: Add helper functions pci_get[put]_host_bridge_device() of: Fix size when dma-range is not used of: Move of_dma_configure() to device.c to help re-use of: iommu: Add ptr to OF node arg to of_iommu_configure() * pci/resource: PCI: Fail pci_ioremap_bar() on unassigned resources PCI: Show driver, BAR#, and resource on pci_ioremap_bar() failure PCI: Mark invalid BARs as unassigned PNP: Don't check for overlaps with unassigned PCI BARs
2015-03-16ACPI: Introduce has_acpi_companion()Rafael J. Wysocki
Now that the ACPI companions of devices are represented by pointers to struct fwnode_handle, it is not quite efficient to check whether or not an ACPI companion of a device is present by evaluating the ACPI_COMPANION() macro. For this reason, introduce a special static inline routine for that, has_acpi_companion(), and update the code to use it where applicable. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>