summaryrefslogtreecommitdiff
path: root/drivers/iommu/amd
AgeCommit message (Collapse)AuthorFilesLines
2024-04-26Merge branch 'iommu/fixes' into x86/amdJoerg Roedel1-0/+4
2024-04-26iommu/amd: Add SVA domain supportVasant Hegde4-3/+93
- Allocate SVA domain and setup mmu notifier. In free path unregister mmu notifier and free protection domain. - Add mmu notifier callback function. It will retrieve SVA protection domain and invalidates IO/TLB. Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240418103400.6229-16-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-04-26iommu/amd: Initial SVA support for AMD IOMMUVasant Hegde6-6/+156
This includes : - Add data structure to track per protection domain dev/pasid binding details protection_domain->dev_data_list will track attached list of dev_data/PASIDs. - Move 'to_pdomain()' to header file - Add iommu_sva_set_dev_pasid(). It will check whether PASID is supported or not. Also adds PASID to SVA protection domain list as well as to device GCR3 table. - Add iommu_ops.remove_dev_pasid support. It will unbind PASID from device. Also remove pasid data from protection domain device list. - Add IOMMU_SVA as dependency to AMD_IOMMU driver For a given PASID, iommu_set_dev_pasid() will bind all devices to same SVA protection domain (1 PASID : 1 SVA protection domain : N devices). This protection domain is different from device protection domain (one that's mapped in attach_device() path). IOMMU uses domain ID for caching, invalidation, etc. In SVA mode it will use per-device-domain-ID. Hence in invalidation path we retrieve domain ID from gcr3_info_table structure and use that for invalidation. Co-developed-by: Wei Huang <wei.huang2@amd.com> Signed-off-by: Wei Huang <wei.huang2@amd.com> Co-developed-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240418103400.6229-14-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-04-26iommu/amd: Add support for enable/disable IOPFVasant Hegde4-8/+77
Return success from enable_feature(IOPF) path as this interface is going away. Instead we will enable/disable IOPF support in attach/detach device path. In attach device path, if device is capable of PRI, then we will add it to per IOMMU IOPF queue and enable PPR support in IOMMU. Also it will attach device to domain even if it fails to enable PRI or add device to IOPF queue as device can continue to work without PRI support. In detach device patch it follows following sequence: - Flush the queue for the given device - Disable PPR support in DTE[devid] - Remove device from IOPF queue - Disable device PRI Also add IOMMU_IOPF as dependency to AMD_IOMMU driver. Co-developed-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240418103400.6229-13-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-04-26iommu/amd: Add IO page fault notifier handlerWei Huang2-1/+107
Whenever there is a page fault IOMMU logs entry to ppr log and sends interrupt to host. We have to handle the page fault and respond to IOMMU. Add support to validate page fault request and hook it to core iommu page fault handler. Signed-off-by: Wei Huang <wei.huang2@amd.com> Co-developed-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Co-developed-by: Vasant Hegde <vasant.hegde@amd.com> Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240418103400.6229-12-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-04-26iommu/amd: Add support for page responseSuravee Suthikulpanit3-6/+13
This generates AMD IOMMU COMPLETE_PPR_REQUEST for the specified device with the specified PRI Response Code. Also update amd_iommu_complete_ppr() to accept 'struct device' instead of pdev as it just need device reference. Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Signed-off-by: Wei Huang <wei.huang2@amd.com> Co-developed-by: Vasant Hegde <vasant.hegde@amd.com> Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240418103400.6229-11-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-04-26iommu/amd: Define per-IOMMU iopf_queueSuravee Suthikulpanit4-2/+55
AMD IOMMU hardware supports PCI Peripheral Paging Request (PPR) using a PPR log, which is a circular buffer containing requests from downstream end-point devices. There is one PPR log per IOMMU instance. Therefore, allocate an iopf_queue per IOMMU instance during driver initialization, and free the queue during driver deinitialization. Also rename enable_iommus_v2() -> enable_iommus_ppr() to reflect its usage. And add amd_iommu_gt_ppr_supported() check before enabling PPR log. Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Co-developed-by: Vasant Hegde <vasant.hegde@amd.com> Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240418103400.6229-10-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-04-26iommu/amd: Enable PCI features based on attached domain capabilityVasant Hegde2-12/+14
Commit eda8c2860ab6 ("iommu/amd: Enable device ATS/PASID/PRI capabilities independently") changed the way it enables device capability while attaching devices. I missed to account the attached domain capability. Meaning if domain is not capable of handling PASID/PRI (ex: paging domain with v1 page table) then enabling device feature is not required. This patch enables PASID/PRI only if domain is capable of handling SVA. Also move pci feature enablement to do_attach() function so that we make SVA capability in one place. Finally make PRI enable/disable functions as static functions. Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240418103400.6229-9-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-04-26iommu/amd: Setup GCR3 table in advance if domain is SVA capableVasant Hegde1-15/+71
SVA can be supported if domain is in passthrough mode or paging domain with v2 page table. Current code sets up GCR3 table for domain with v2 page table only. Setup GCR3 table for all SVA capable domains. - Move GCR3 init/destroy to separate function. - Change default GCR3 table to use MAX supported PASIDs. Ideally it should use 1 level PASID table as its using PASID zero only. But we don't have support to extend PASID table yet. We will fix this later. - When domain is configured with passthrough mode, allocate default GCR3 table only if device is SVA capable. Note that in attach_device() path it will not know whether device will use SVA or not. If device is attached to passthrough domain and if it doesn't use SVA then GCR3 table will never be used. We will endup wasting memory allocated for GCR3 table. This is done to avoid DTE update when attaching PASID to device. Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240418103400.6229-8-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-04-26iommu/amd: Introduce iommu_dev_data.max_pasidsVasant Hegde2-0/+13
This variable will track the number of PASIDs supported by the device. If IOMMU or device doesn't support PASID then it will be zero. This will be used while allocating GCR3 table to decide required number of PASID table levels. Also in PASID bind path it will use this variable to check whether device supports PASID or not. Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240418103400.6229-7-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-04-26iommu/amd: Fix PPR interrupt processing logicVasant Hegde1-4/+0
* Do not re-read ppr head pointer as its just updated by the driver. * Do not read PPR buffer tail pointer inside while loop. If IOMMU generates PPR events continuously then completing interrupt processing takes long time. In worst case it may cause infinite loop. Suggested-by: Tom Lendacky <thomas.lendacky@amd.com> Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240418103400.6229-6-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-04-26iommu/amd: Move PPR-related functions into ppr.cSuravee Suthikulpanit5-114/+139
In preparation to subsequent PPR-related patches, and also remove static declaration for certain helper functions so that it can be reused in other files. Also rename below functions: alloc_ppr_log -> amd_iommu_alloc_ppr_log iommu_enable_ppr_log -> amd_iommu_enable_ppr_log free_ppr_log -> amd_iommu_free_ppr_log iommu_poll_ppr_log -> amd_iommu_poll_ppr_log Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Co-developed-by: Vasant Hegde <vasant.hegde@amd.com> Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240418103400.6229-5-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-04-26iommu/amd: Add support for enabling/disabling IOMMU featuresWei Huang1-0/+28
Add support for struct iommu_ops.dev_{enable/disable}_feat. Please note that the empty feature switches will be populated by subsequent patches. Signed-off-by: Wei Huang <wei.huang2@amd.com> Co-developed-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240418103400.6229-4-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-04-26iommu/amd: Introduce per device DTE update functionVasant Hegde2-8/+19
Consolidate per device update and flush logic into separate function. Also make it as global function as it will be used in subsequent series to update the DTE. Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240418103400.6229-3-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-04-26iommu/amd: Rename amd_iommu_v2_supported() as amd_iommu_pasid_supported()Vasant Hegde2-2/+2
To reflect its usage. No functional changes intended. Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240418103400.6229-2-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-04-26iommu/amd: Enhance def_domain_type to handle untrusted deviceVasant Hegde1-0/+4
Previously, IOMMU core layer was forcing IOMMU_DOMAIN_DMA domain for untrusted device. This always took precedence over driver's def_domain_type(). Commit 59ddce4418da ("iommu: Reorganize iommu_get_default_domain_type() to respect def_domain_type()") changed the behaviour. Current code calls def_domain_type() but if it doesn't return IOMMU_DOMAIN_DMA for untrusted device it throws error. This results in IOMMU group (and potentially IOMMU itself) in undetermined state. This patch adds untrusted check in AMD IOMMU driver code. So that it allows eGPUs behind Thunderbolt work again. Fine tuning amd_iommu_def_domain_type() will be done later. Reported-by: Eric Wagner <ewagner12@gmail.com> Link: https://lore.kernel.org/linux-iommu/CAHudX3zLH6CsRmLE-yb+gRjhh-v4bU5_1jW_xCcxOo_oUUZKYg@mail.gmail.com Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3182 Fixes: 59ddce4418da ("iommu: Reorganize iommu_get_default_domain_type() to respect def_domain_type()") Cc: Robin Murphy <robin.murphy@arm.com> Cc: Jason Gunthorpe <jgg@ziepe.ca> Cc: stable@kernel.org # v6.7+ Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Link: https://lore.kernel.org/r/20240423111725.5813-1-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-04-26iommu/dma: Centralise iommu_setup_dma_ops()Robin Murphy1-8/+0
It's somewhat hard to see, but arm64's arch_setup_dma_ops() should only ever call iommu_setup_dma_ops() after a successful iommu_probe_device(), which means there should be no harm in achieving the same order of operations by running it off the back of iommu_probe_device() itself. This then puts it in line with the x86 and s390 .probe_finalize bodges, letting us pull it all into the main flow properly. As a bonus this lets us fold in and de-scope the PCI workaround setup as well. At this point we can also then pull the call up inside the group mutex, and avoid having to think about whether iommu_group_store_type() could theoretically race and free the domain if iommu_setup_dma_ops() ran just *before* iommu_device_use_default_domain() claims it... Furthermore we replace one .probe_finalize call completely, since the only remaining implementations are now one which only needs to run once for the initial boot-time probe, and two which themselves render that path unreachable. This leaves us a big step closer to realistically being able to unpick the variety of different things that iommu_setup_dma_ops() has been muddling together, and further streamline iommu-dma into core API flows in future. Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com> # For Intel IOMMU Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Tested-by: Hanjun Guo <guohanjun@huawei.com> Signed-off-by: Robin Murphy <robin.murphy@arm.com> Acked-by: Catalin Marinas <catalin.marinas@arm.com> Link: https://lore.kernel.org/r/bebea331c1d688b34d9862eefd5ede47503961b8.1713523152.git.robin.murphy@arm.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-04-26iommu/vt-d: Allocate DMAR fault interrupts locallyDimitri Sivanich2-2/+2
The Intel IOMMU code currently tries to allocate all DMAR fault interrupt vectors on the boot cpu. On large systems with high DMAR counts this results in vector exhaustion, and most of the vectors are not initially allocated socket local. Instead, have a cpu on each node do the vector allocation for the DMARs on that node. The boot cpu still does the allocation for its node during its boot sequence. Signed-off-by: Dimitri Sivanich <sivanich@hpe.com> Reviewed-by: Kevin Tian <kevin.tian@intel.com> Link: https://lore.kernel.org/r/Zfydpp2Hm+as16TY@hpe.com Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-04-15iommu/amd: use page allocation function provided by iommu-pages.hPasha Tatashin5-79/+62
Convert iommu/amd/* files to use the new page allocation functions provided in iommu-pages.h. Signed-off-by: Pasha Tatashin <pasha.tatashin@soleen.com> Acked-by: David Rientjes <rientjes@google.com> Tested-by: Bagas Sanjaya <bagasdotme@gmail.com> Link: https://lore.kernel.org/r/20240413002522.1101315-4-pasha.tatashin@soleen.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-04-12iommu/amd: Change log message severityVasant Hegde1-2/+2
Use consistent log severity (pr_warn) to log all messages in SNP enable path. Suggested-by: Tom Lendacky <thomas.lendacky@amd.com> Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Link: https://lore.kernel.org/r/20240410101643.32309-1-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-04-12iommu/amd: Do not enable SNP when V2 page table is enabledVasant Hegde1-10/+11
DTE[Mode]=0 is not supported when SNP is enabled in the host. That means to support SNP, IOMMU must be configured with V1 page table (See IOMMU spec [1] for the details). If user passes kernel command line to configure IOMMU domains with v2 page table (amd_iommu=pgtbl_v2) then disable SNP as the user asked by not forcing the page table to v1. [1] https://www.amd.com/content/dam/amd/en/documents/processor-tech-docs/specifications/48882_IOMMU.pdf Cc: Ashish Kalra <ashish.kalra@amd.com> Cc: Michael Roth <michael.roth@amd.com> Cc: Tom Lendacky <thomas.lendacky@amd.com> Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com> Link: https://lore.kernel.org/r/20240410085702.31869-1-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-04-12iommu/amd: Fix possible irq lock inversion dependency issueVasant Hegde1-4/+7
LOCKDEP detector reported below warning: ---------------------------------------- [ 23.796949] ======================================================== [ 23.796950] WARNING: possible irq lock inversion dependency detected [ 23.796952] 6.8.0fix+ #811 Not tainted [ 23.796954] -------------------------------------------------------- [ 23.796954] kworker/0:1/8 just changed the state of lock: [ 23.796956] ff365325e084a9b8 (&domain->lock){..-.}-{3:3}, at: amd_iommu_flush_iotlb_all+0x1f/0x50 [ 23.796969] but this lock took another, SOFTIRQ-unsafe lock in the past: [ 23.796970] (pd_bitmap_lock){+.+.}-{3:3} [ 23.796972] and interrupts could create inverse lock ordering between them. [ 23.796973] other info that might help us debug this: [ 23.796974] Chain exists of: &domain->lock --> &dev_data->lock --> pd_bitmap_lock [ 23.796980] Possible interrupt unsafe locking scenario: [ 23.796981] CPU0 CPU1 [ 23.796982] ---- ---- [ 23.796983] lock(pd_bitmap_lock); [ 23.796985] local_irq_disable(); [ 23.796985] lock(&domain->lock); [ 23.796988] lock(&dev_data->lock); [ 23.796990] <Interrupt> [ 23.796991] lock(&domain->lock); Fix this issue by disabling interrupt when acquiring pd_bitmap_lock. Note that this is temporary fix. We have a plan to replace custom bitmap allocator with IDA allocator. Fixes: 87a6f1f22c97 ("iommu/amd: Introduce per-device domain ID to fix potential TLB aliasing issue") Reviewed-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Link: https://lore.kernel.org/r/20240404102717.6705-1-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-04-04x86/CPU/AMD: Track SNP host status with cc_platform_*()Borislav Petkov (AMD)1-1/+3
The host SNP worthiness can determined later, after alternatives have been patched, in snp_rmptable_init() depending on cmdline options like iommu=pt which is incompatible with SNP, for example. Which means that one cannot use X86_FEATURE_SEV_SNP and will need to have a special flag for that control. Use that newly added CC_ATTR_HOST_SEV_SNP in the appropriate places. Move kdump_sev_callback() to its rightful place, while at it. Fixes: 216d106c7ff7 ("x86/sev: Add SEV-SNP host initialization support") Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de> Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com> Tested-by: Srikanth Aithal <sraithal@amd.com> Link: https://lore.kernel.org/r/20240327154317.29909-6-bp@alien8.de
2024-03-13Merge tag 'iommu-updates-v6.9' of ↵Linus Torvalds5-405/+337
git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu Pull iommu updates from Joerg Roedel: "Core changes: - Constification of bus_type pointer - Preparations for user-space page-fault delivery - Use a named kmem_cache for IOVA magazines Intel VT-d changes from Lu Baolu: - Add RBTree to track iommu probed devices - Add Intel IOMMU debugfs document - Cleanup and refactoring ARM-SMMU Updates from Will Deacon: - Device-tree binding updates for a bunch of Qualcomm SoCs - SMMUv2: Support for Qualcomm X1E80100 MDSS - SMMUv3: Significant rework of the driver's STE manipulation and domain handling code. This is the initial part of a larger scale rework aiming to improve the driver's implementation of the IOMMU-API in preparation for hooking up IOMMUFD support. AMD-Vi Updates: - Refactor GCR3 table support for SVA - Cleanups Some smaller cleanups and fixes" * tag 'iommu-updates-v6.9' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu: (88 commits) iommu: Fix compilation without CONFIG_IOMMU_INTEL iommu/amd: Fix sleeping in atomic context iommu/dma: Document min_align_mask assumption iommu/vt-d: Remove scalabe mode in domain_context_clear_one() iommu/vt-d: Remove scalable mode context entry setup from attach_dev iommu/vt-d: Setup scalable mode context entry in probe path iommu/vt-d: Fix NULL domain on device release iommu: Add static iommu_ops->release_domain iommu/vt-d: Improve ITE fault handling if target device isn't present iommu/vt-d: Don't issue ATS Invalidation request when device is disconnected PCI: Make pci_dev_is_disconnected() helper public for other drivers iommu/vt-d: Use device rbtree in iopf reporting path iommu/vt-d: Use rbtree to track iommu probed devices iommu/vt-d: Merge intel_svm_bind_mm() into its caller iommu/vt-d: Remove initialization for dynamically heap-allocated rcu_head iommu/vt-d: Remove treatment for revoking PASIDs with pending page faults iommu/vt-d: Add the document for Intel IOMMU debugfs iommu/vt-d: Use kcalloc() instead of kzalloc() iommu/vt-d: Remove INTEL_IOMMU_BROKEN_GFX_WA iommu: re-use local fwnode variable in iommu_ops_from_fwnode() ...
2024-03-08iommu/amd: Fix sleeping in atomic contextVasant Hegde1-1/+1
Commit cf70873e3d01 ("iommu/amd: Refactor GCR3 table helper functions") changed GFP flag we use for GCR3 table. Original plan was to move GCR3 table allocation outside spinlock. But this requires complete rework of attach device path. Hence we didn't do it as part of SVA series. For now revert the GFP flag to ATOMIC (same as original code). Fixes: cf70873e3d01 ("iommu/amd: Refactor GCR3 table helper functions") Reported-by: Dan Carpenter <dan.carpenter@linaro.org> Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Link: https://lore.kernel.org/r/20240307052738.116035-1-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-16iommu/amd: Mark interrupt as managedMario Limonciello1-0/+3
On many systems that have an AMD IOMMU the following sequence of warnings is observed during bootup. ``` pci 0000:00:00.2 can't derive routing for PCI INT A pci 0000:00:00.2: PCI INT A: not connected ``` This series of events happens because of the IOMMU initialization sequence order and the lack of _PRT entries for the IOMMU. During initialization the IOMMU driver first enables the PCI device using pci_enable_device(). This will call acpi_pci_irq_enable() which will check if the interrupt is declared in a PCI routing table (_PRT) entry. According to the PCI spec [1] these routing entries are only required under PCI root bridges: The _PRT object is required under all PCI root bridges The IOMMU is directly connected to the root complex, so there is no parent bridge to look for a _PRT entry. The first warning is emitted since no entry could be found in the hierarchy. The second warning is then emitted because the interrupt hasn't yet been configured to any value. The pin was configured in pci_read_irq() but the byte in PCI_INTERRUPT_LINE return 0xff which means "Unknown". After that sequence of events pci_enable_msi() is called and this will allocate an interrupt. That is both of these warnings are totally harmless because the IOMMU uses MSI for interrupts. To avoid even trying to probe for a _PRT entry mark the IOMMU as IRQ managed. This avoids both warnings. Link: https://uefi.org/htmlspecs/ACPI_Spec_6_4_html/06_Device_Configuration/Device_Configuration.html?highlight=_prt#prt-pci-routing-table [1] Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Fixes: cffe0a2b5a34 ("x86, irq: Keep balance of IOAPIC pin reference count") Reviewed-by: Vasant Hegde <vasant.hegde@amd.com> Link: https://lore.kernel.org/r/20240122233400.1802-1-mario.limonciello@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-09iommu/amd: Introduce per-device domain ID to fix potential TLB aliasing issueVasant Hegde2-17/+64
With v1 page table, the AMD IOMMU spec states that the hardware must use the domain ID to tag its internal translation caches. I/O devices with different v1 page tables must be given different domain IDs. I/O devices that share the same v1 page table __may__ be given the same domain ID. This domain ID management policy is currently implemented by the AMD IOMMU driver. In this case, only the domain ID is needed when issuing the INVALIDATE_IOMMU_PAGES command to invalidate the IOMMU translation cache (TLB). With v2 page table, the hardware uses domain ID and PASID as parameters to tag and issue the INVALIDATE_IOMMU_PAGES command. Since the GCR3 table is setup per-device, and there is no guarantee for PASID to be unique across multiple devices. The same PASID for different devices could have different v2 page tables. In such case, if multiple devices share the same domain ID, IOMMU translation cache for these devices would be polluted due to TLB aliasing. Hence, avoid the TLB aliasing issue with v2 page table by allocating unique domain ID for each device even when multiple devices are sharing the same v1 page table. Please note that this fix would result in multiple INVALIDATE_IOMMU_PAGES commands (one per domain id) when unmapping a translation. Domain ID can be shared until device starts using PASID. We will enhance this code later where we will allocate per device domain ID only when its needed. Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240205115615.6053-18-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-09iommu/amd: Remove unused GCR3 table parameters from struct protection_domainSuravee Suthikulpanit2-13/+1
Since they are moved to struct iommu_dev_data, and the driver has been ported to use them. Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240205115615.6053-17-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-09iommu/amd: Rearrange device flush codeVasant Hegde1-49/+43
Consolidate all flush related code in one place so that its easy to maintain. No functional changes intended. Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240205115615.6053-16-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-09iommu/amd: Remove unused flush pasid functionsVasant Hegde2-103/+0
We have removed iommu_v2 module and converted v2 page table to use common flush functions. Also we have moved GCR3 table to per device. PASID related functions are not used. Hence remove these unused functions. Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240205115615.6053-15-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-09iommu/amd: Refactor GCR3 table helper functionsSuravee Suthikulpanit1-22/+31
To use the new per-device struct gcr3_tbl_info. Use GFP_KERNEL flag instead of GFP_ATOMIC for GCR3 table allocation. Also modify set_dte_entry() to use new per device GCR3 table. Also in free_gcr3_table() path replace BUG_ON with WARN_ON_ONCE(). Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Co-developed-by: Vasant Hegde <vasant.hegde@amd.com> Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240205115615.6053-14-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-09iommu/amd: Refactor protection_domain helper functionsSuravee Suthikulpanit1-11/+3
To removes the code to setup GCR3 table, and only handle domain create / destroy, since GCR3 is no longer part of a domain. Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240205115615.6053-13-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-09iommu/amd: Refactor attaching / detaching device functionsSuravee Suthikulpanit1-3/+27
If domain is configured with V2 page table then setup default GCR3 with domain GCR3 pointer. So that all devices in the domain uses same page table for translation. Also return page table setup status from do_attach() function. Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Co-developed-by: Vasant Hegde <vasant.hegde@amd.com> Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240205115615.6053-12-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-09iommu/amd: Refactor helper function for setting / clearing GCR3Suravee Suthikulpanit3-83/+61
Refactor GCR3 helper functions in preparation to use per device GCR3 table. * Add new function update_gcr3 to update per device GCR3 table * Remove per domain default GCR3 setup during v2 page table allocation. Subsequent patch will add support to setup default gcr3 while attaching device to domain. * Remove amd_iommu_domain_update() from V2 page table path as device detach path will take care of updating the domain. * Consolidate GCR3 table related code in one place so that its easy to maintain. * Rename functions to reflect its usage. Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Co-developed-by: Vasant Hegde <vasant.hegde@amd.com> Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240205115615.6053-11-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-09iommu/amd: Rearrange GCR3 table setup codeVasant Hegde1-32/+32
Consolidate GCR3 table related code in one place so that its easy to maintain. Note that this patch doesn't move __set_gcr3/__clear_gcr3. We are moving GCR3 table from per domain to per device. Following series will rework these functions. During that time I will move these functions as well. No functional changes intended. Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240205115615.6053-9-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-09iommu/amd: Add support for device based TLB invalidationVasant Hegde2-0/+28
Add support to invalidate TLB/IOTLB for the given device. These functions will be used in subsequent patches where we will introduce per device GCR3 table and SVA support. Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240205115615.6053-8-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-09iommu/amd: Use protection_domain.flags to check page table modeVasant Hegde1-2/+2
Page table mode (v1, v2 or pt) is per domain property. Recently we have enhanced protection_domain.pd_mode to track per domain page table mode. Use that variable to check the page table mode instead of global 'amd_iommu_pgtable' in {map/unmap}_pages path. Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240205115615.6053-7-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-09iommu/amd: Introduce per-device GCR3 tableSuravee Suthikulpanit1-0/+7
AMD IOMMU GCR3 table is indexed by PASID. Each entry stores guest CR3 register value, which is an address to the root of guest IO page table. The GCR3 table can be programmed per-device. However, Linux AMD IOMMU driver currently managing the table on a per-domain basis. PASID is a device feature. When SVA is enabled it will bind PASID to device, not domain. Hence it makes sense to have per device GCR3 table. Introduce struct iommu_dev_data.gcr3_tbl_info to keep track of GCR3 table configuration. This will eventually replaces gcr3 related variables in protection_domain structure. Suggested-by: Jason Gunthorpe <jgg@ziepe.ca> Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240205115615.6053-6-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-09iommu/amd: Introduce struct protection_domain.pd_modeSuravee Suthikulpanit2-0/+8
This enum variable is used to track the type of page table used by the protection domain. It will replace the protection_domain.flags in subsequent series. Suggested-by: Jason Gunthorpe <jgg@ziepe.ca> Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240205115615.6053-5-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-09iommu/amd: Introduce get_amd_iommu_from_dev()Suravee Suthikulpanit2-41/+28
Introduce get_amd_iommu_from_dev() and get_amd_iommu_from_dev_data(). And replace rlookup_amd_iommu() with the new helper function where applicable to avoid unnecessary loop to look up struct amd_iommu from struct device. Suggested-by: Jason Gunthorpe <jgg@ziepe.ca> Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240205115615.6053-4-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-09iommu/amd: Enable Guest Translation before registering devicesVasant Hegde1-3/+3
IOMMU Guest Translation (GT) feature needs to be enabled before invalidating guest translations (CMD_INV_IOMMU_PAGES with GN=1). Currently GT feature is enabled after setting up interrupt handler. So far it was fine as we were not invalidating guest page table before this point. Upcoming series will introduce per device GCR3 table and it will invalidate guest pages after configuring. Hence move GT feature enablement to early_enable_iommu(). Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240205115615.6053-3-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-09iommu/amd: Pass struct iommu_dev_data to set_dte_entry()Vasant Hegde1-10/+8
Pass iommu_dev_data structure instead of passing indivisual variables. No functional changes intended. Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240205115615.6053-2-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-09iommu/amd: Remove EXPORT_SYMBOL for perf counter related functionsVasant Hegde1-3/+0
.. as IOMMU perf counters are always built as part of kernel. No functional change intended. Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Reviewed-by: Alejandro Jimenez <alejandro.j.jimenez@oracle.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240118090105.5864-7-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-09iommu/amd: Remove redundant error check in amd_iommu_probe_device()Vasant Hegde1-2/+1
iommu_init_device() is not returning -ENOTSUPP since commit 61289cbaf6c8 ("iommu/amd: Remove old alias handling code"). No functional change intended. Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Reviewed-by: Alejandro Jimenez <alejandro.j.jimenez@oracle.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240118090105.5864-6-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-09iommu/amd: Remove duplicate function declarations from amd_iommu.hVasant Hegde1-8/+0
Perf counter related functions are defined in amd-iommu.h as well. Hence remove duplicate declarations. No functional change intended. Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Reviewed-by: Alejandro Jimenez <alejandro.j.jimenez@oracle.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240118090105.5864-5-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-09iommu/amd: Remove unused APERTURE_* macrosVasant Hegde1-8/+0
These macros are not used after commit 518d9b450387 ("iommu/amd: Remove special mapping code for dma_ops path"). No functional change intended. Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Reviewed-by: Alejandro Jimenez <alejandro.j.jimenez@oracle.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240118090105.5864-4-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-09iommu/amd: Remove unused IOVA_* macroVasant Hegde1-4/+0
These macros are not used after commit ac6d704679d343 ("iommu/dma: Pass address limit rather than size to iommu_setup_dma_ops()"). No functional change intended. Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Cc: Lu Baolu <baolu.lu@linux.intel.com> Reviewed-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Reviewed-by: Alejandro Jimenez <alejandro.j.jimenez@oracle.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240118090105.5864-3-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-09iommu/amd: Remove unused PPR_* macrosVasant Hegde1-4/+0
Commit 5a0b11a180a ("iommu/amd: Remove iommu_v2 module") missed to remove PPR_* macros. Remove these macros as its not used anymore. No functional change intended. Signed-off-by: Vasant Hegde <vasant.hegde@amd.com> Reviewed-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Reviewed-by: Alejandro Jimenez <alejandro.j.jimenez@oracle.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Link: https://lore.kernel.org/r/20240118090105.5864-2-vasant.hegde@amd.com Signed-off-by: Joerg Roedel <jroedel@suse.de>
2024-02-07iommu/amd: Fix failure return from snp_lookup_rmpentry()Ashish Kalra1-10/+15
Commit f366a8dac1b8: ("iommu/amd: Clean up RMP entries for IOMMU pages during SNP shutdown") leads to the following Smatch static checker warning: drivers/iommu/amd/init.c:3820 iommu_page_make_shared() error: uninitialized symbol 'assigned'. Fix it. [ bp: Address the other error cases too. ] Fixes: f366a8dac1b8 ("iommu/amd: Clean up RMP entries for IOMMU pages during SNP shutdown") Closes: https://lore.kernel.org/linux-iommu/1be69f6a-e7e1-45f9-9a74-b2550344f3fd@moroto.mountain Reported-by: Dan Carpenter <dan.carpenter@linaro.org> Signed-off-by: Ashish Kalra <ashish.kalra@amd.com> Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de> Cc: Joerg Roedel <jroedel@suse.com> Link: https://lore.kernel.org/lkml/20240126041126.1927228-20-michael.roth@amd.com
2024-01-29iommu/amd: Clean up RMP entries for IOMMU pages during SNP shutdownAshish Kalra1-0/+79
Add a new IOMMU API interface amd_iommu_snp_disable() to transition IOMMU pages to Hypervisor state from Reclaim state after SNP_SHUTDOWN_EX command. Invoke this API from the CCP driver after SNP_SHUTDOWN_EX command. Signed-off-by: Ashish Kalra <ashish.kalra@amd.com> Signed-off-by: Michael Roth <michael.roth@amd.com> Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de> Link: https://lore.kernel.org/r/20240126041126.1927228-20-michael.roth@amd.com
generated by cgit v1.2.3 (git 2.39.0) at 2025-03-14 21:45:42 +0300