| Message ID | 0-v1-54e734311a7f+14f72-smmuv3_nesting_jgg@nvidia.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | Initial support for SMMUv3 nested translation | expand |
Hi Jason, On Tue, Aug 06, 2024 at 08:41:13PM -0300, Jason Gunthorpe wrote: > This brings support for the IOMMFD ioctls: > > - IOMMU_GET_HW_INFO > - IOMMU_HWPT_ALLOC_NEST_PARENT > - IOMMU_DOMAIN_NESTED > - ops->enforce_cache_coherency() > > This is quite straightforward as the nested STE can just be built in the > special NESTED domain op and fed through the generic update machinery. > > The design allows the user provided STE fragment to control several > aspects of the translation, including putting the STE into a "virtual > bypass" or a aborting state. This duplicates functionality available by > other means, but it allows trivially preserving the VMID in the STE as we > eventually move towards the VIOMMU owning the VMID. > > Nesting support requires the system to either support S2FWB or the > stronger CANWBS ACPI flag. This is to ensure the VM cannot bypass the > cache and view incoherent data, currently VFIO lacks any cache flushing > that would make this safe. > > Yan has a series to add some of the needed infrastructure for VFIO cache > flushing here: > > https://lore.kernel.org/linux-iommu/20240507061802.20184-1-yan.y.zhao@intel.com/ > > Which may someday allow relaxing this further. > > Remove VFIO_TYPE1_NESTING_IOMMU since it was never used and superseded by > this. > > This is the first series in what will be several to complete nesting > support. At least: > - IOMMU_RESV_SW_MSI related fixups > - VIOMMU object support to allow ATS invalidations > - vCMDQ hypervisor support for direct invalidation queue assignment > - KVM pinned VMID using VIOMMU for vBTM > - Cross instance S2 sharing > - Virtual Machine Structure using VIOMMU (for vMPAM?) > - Fault forwarding support through IOMMUFD's fault fd for vSVA > > It is enough to allow significant amounts of qemu work to progress. > Are there any qemu patches to tests this? As I am confused with some of the user space bits and that would help. Thanks, Mostafa > This is on github: https://github.com/jgunthorpe/linux/commits/smmuv3_nesting > > Jason Gunthorpe (5): > vfio: Remove VFIO_TYPE1_NESTING_IOMMU > iommu/arm-smmu-v3: Use S2FWB when available > iommu/arm-smmu-v3: Report IOMMU_CAP_ENFORCE_CACHE_COHERENCY for CANWBS > iommu/arm-smmu-v3: Implement IOMMU_HWPT_ALLOC_NEST_PARENT > iommu/arm-smmu-v3: Support IOMMU_DOMAIN_NESTED > > Nicolin Chen (3): > ACPI/IORT: Support CANWBS memory access flag > iommu/arm-smmu-v3: Support IOMMU_GET_HW_INFO via struct > arm_smmu_hw_info > iommu/arm-smmu-v3: Add arm_smmu_cache_invalidate_user > > drivers/acpi/arm64/iort.c | 13 + > drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 398 ++++++++++++++++++-- > drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h | 27 ++ > drivers/iommu/arm/arm-smmu/arm-smmu.c | 16 - > drivers/iommu/io-pgtable-arm.c | 24 +- > drivers/iommu/iommu.c | 10 - > drivers/iommu/iommufd/vfio_compat.c | 7 +- > drivers/vfio/vfio_iommu_type1.c | 12 +- > include/acpi/actbl2.h | 1 + > include/linux/io-pgtable.h | 2 + > include/linux/iommu.h | 54 ++- > include/uapi/linux/iommufd.h | 79 ++++ > include/uapi/linux/vfio.h | 2 +- > 13 files changed, 572 insertions(+), 73 deletions(-) > > > base-commit: e5e288d94186b266b062b3e44c82c285dfe68712 > -- > 2.46.0 >
On Tue, Aug 20, 2024 at 08:20:32AM +0000, Mostafa Saleh wrote: > > This is the first series in what will be several to complete nesting > > support. At least: > > - IOMMU_RESV_SW_MSI related fixups > > - VIOMMU object support to allow ATS invalidations > > - vCMDQ hypervisor support for direct invalidation queue assignment > > - KVM pinned VMID using VIOMMU for vBTM > > - Cross instance S2 sharing > > - Virtual Machine Structure using VIOMMU (for vMPAM?) > > - Fault forwarding support through IOMMUFD's fault fd for vSVA > > > > It is enough to allow significant amounts of qemu work to progress. > > > Are there any qemu patches to tests this? > As I am confused with some of the user space bits and that would help. I have the qemu patches, but am running some backlogs to keep it updated, and don't have one exactly fitting to test this series. I collected a few remarks from Jason regarding the VIOMMU series. And I am reworking on it. I plan to post a testable QEMU branch with the next VIOMMU version. Will CC you and more folks. Thanks Nicolin
This brings support for the IOMMFD ioctls: - IOMMU_GET_HW_INFO - IOMMU_HWPT_ALLOC_NEST_PARENT - IOMMU_DOMAIN_NESTED - ops->enforce_cache_coherency() This is quite straightforward as the nested STE can just be built in the special NESTED domain op and fed through the generic update machinery. The design allows the user provided STE fragment to control several aspects of the translation, including putting the STE into a "virtual bypass" or a aborting state. This duplicates functionality available by other means, but it allows trivially preserving the VMID in the STE as we eventually move towards the VIOMMU owning the VMID. Nesting support requires the system to either support S2FWB or the stronger CANWBS ACPI flag. This is to ensure the VM cannot bypass the cache and view incoherent data, currently VFIO lacks any cache flushing that would make this safe. Yan has a series to add some of the needed infrastructure for VFIO cache flushing here: https://lore.kernel.org/linux-iommu/20240507061802.20184-1-yan.y.zhao@intel.com/ Which may someday allow relaxing this further. Remove VFIO_TYPE1_NESTING_IOMMU since it was never used and superseded by this. This is the first series in what will be several to complete nesting support. At least: - IOMMU_RESV_SW_MSI related fixups - VIOMMU object support to allow ATS invalidations - vCMDQ hypervisor support for direct invalidation queue assignment - KVM pinned VMID using VIOMMU for vBTM - Cross instance S2 sharing - Virtual Machine Structure using VIOMMU (for vMPAM?) - Fault forwarding support through IOMMUFD's fault fd for vSVA It is enough to allow significant amounts of qemu work to progress. This is on github: https://github.com/jgunthorpe/linux/commits/smmuv3_nesting Jason Gunthorpe (5): vfio: Remove VFIO_TYPE1_NESTING_IOMMU iommu/arm-smmu-v3: Use S2FWB when available iommu/arm-smmu-v3: Report IOMMU_CAP_ENFORCE_CACHE_COHERENCY for CANWBS iommu/arm-smmu-v3: Implement IOMMU_HWPT_ALLOC_NEST_PARENT iommu/arm-smmu-v3: Support IOMMU_DOMAIN_NESTED Nicolin Chen (3): ACPI/IORT: Support CANWBS memory access flag iommu/arm-smmu-v3: Support IOMMU_GET_HW_INFO via struct arm_smmu_hw_info iommu/arm-smmu-v3: Add arm_smmu_cache_invalidate_user drivers/acpi/arm64/iort.c | 13 + drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 398 ++++++++++++++++++-- drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h | 27 ++ drivers/iommu/arm/arm-smmu/arm-smmu.c | 16 - drivers/iommu/io-pgtable-arm.c | 24 +- drivers/iommu/iommu.c | 10 - drivers/iommu/iommufd/vfio_compat.c | 7 +- drivers/vfio/vfio_iommu_type1.c | 12 +- include/acpi/actbl2.h | 1 + include/linux/io-pgtable.h | 2 + include/linux/iommu.h | 54 ++- include/uapi/linux/iommufd.h | 79 ++++ include/uapi/linux/vfio.h | 2 +- 13 files changed, 572 insertions(+), 73 deletions(-) base-commit: e5e288d94186b266b062b3e44c82c285dfe68712