STRUCT IOMMU_HWPT_INVALIDATE(9)
π NAME π
struct iommu_hwpt_invalidate β the ioctl(IOMMU_HWPT_INVALIDATE) situation, explained by your favorite kernel-fluent bestie β¨ (welcome to the Kernel API Manual, pull up a chair π«Ά)
β¨ SYNOPSIS β¨
struct iommu_hwpt_invalidate {
__u32 size;
__u32 hwpt_id;
__aligned_u64 data_uptr;
__u32 data_type;
__u32 entry_len;
__u32 entry_num;
__u32 __reserved;
};
Likeβ¦ look at her. She's a whole struct. Iconic. π
π Members π
Okay so let's meet the cast, one field at a time:
sizeβ literally justsizeof(struct iommu_hwpt_invalidate). She knows her own measurements, periodt πhwpt_idβ the ID of a nested HWPT or a vIOMMU, used for cache invalidation. Pick your fighter π₯data_uptrβ a user pointer to an array of driver-specific cache invalidation data. It's giving "the actual receipts" π§Ύdata_typeβ one ofenum iommu_hwpt_invalidate_data_type, and it defines the data type of all the entries in the invalidation request array. It has to be a type supported by the hwpt thathwpt_idpoints at, no exceptions π¨entry_lenβ length (in bytes) of a request entry in the request array πentry_numβ on input, the number of cache invalidation requests in the array. On output, the number of requests the kernel successfully handled. She does both, she's versatile πββοΈ__reservedβ must be0. Zero. Nothing. Empty. Nada. Don't be cute with it π
π¦ Description π¦
So what does she do? She invalidates the iommu cache for a user-managed page table or a vIOMMU. That's the whole personality πΈ
Modifications on a user-managed page table should be followed by this operation β but only if a HWPT is passed in via hwpt_id. And other caches, like device cache or descriptor cache, can be flushed if a vIOMMU is passed in via the hwpt_id field. iykyk π₯
Each ioctl can support one or more cache invalidation requests in the array that has a total size of entry_len * entry_num. Math girlie behavior, we love to see it β¨
And here's the cute part: an empty invalidation request array by settingentry_num==0is allowed! In that caseentry_lenanddata_uptrwould be ignored entirely. Which honestly slays, because you can use it to check if the givendata_typeis supported or not by the kernel π
π SEE ALSO π
Start with the kernel file itself, bestie: ./include/uapi/linux/iommufd.h β then the whole extended friend group:
- General ioctl format(9)
struct iommu_destroy(9)struct iommu_ioas_alloc(9)struct iommu_iova_range(9)struct iommu_ioas_iova_ranges(9)struct iommu_ioas_allow_iovas(9)enum iommufd_ioas_map_flags(9)struct iommu_ioas_map(9)struct iommu_ioas_noiommu_get_pa(9)struct iommu_ioas_map_file(9)struct iommu_ioas_copy(9)struct iommu_ioas_unmap(9)enum iommufd_option(9)enum iommufd_option_ops(9)struct iommu_option(9)enum iommufd_vfio_ioas_op(9)struct iommu_vfio_ioas(9)enum iommufd_hwpt_alloc_flags(9)enum iommu_hwpt_vtd_s1_flags(9)struct iommu_hwpt_vtd_s1(9)struct iommu_hwpt_arm_smmuv3(9)struct iommu_hwpt_amd_guest(9)enum iommu_hwpt_data_type(9)struct iommu_hwpt_alloc(9)enum iommu_hw_info_vtd_flags(9)struct iommu_hw_info_vtd(9)enum iommu_hw_info_arm_smmuv3_flags(9)struct iommu_hw_info_arm_smmuv3(9)struct iommu_hw_info_tegra241_cmdqv(9)struct iommu_hw_info_amd(9)enum iommu_hw_info_type(9)enum iommufd_hw_capabilities(9)enum iommufd_hw_info_flags(9)struct iommu_hw_info(9)struct iommu_hwpt_set_dirty_tracking(9)enum iommufd_hwpt_get_dirty_bitmap_flags(9)struct iommu_hwpt_get_dirty_bitmap(9)enum iommu_hwpt_invalidate_data_type(9)enum iommu_hwpt_vtd_s1_invalidate_flags(9)struct iommu_hwpt_vtd_s1_invalidate(9)struct iommu_viommu_arm_smmuv3_invalidate(9)enum iommu_hwpt_pgfault_flags(9)enum iommu_hwpt_pgfault_perm(9)struct iommu_hwpt_pgfault(9)enum iommufd_page_response_code(9)struct iommu_hwpt_page_response(9)struct iommu_fault_alloc(9)enum iommu_viommu_type(9)struct iommu_viommu_tegra241_cmdqv(9)struct iommu_viommu_alloc(9)struct iommu_vdevice_alloc(9)struct iommu_ioas_change_process(9)enum iommu_veventq_flag(9)struct iommufd_vevent_header(9)enum iommu_veventq_type(9)struct iommu_vevent_arm_smmuv3(9)struct iommu_vevent_tegra241_cmdqv(9)struct iommu_veventq_alloc(9)enum iommu_hw_queue_type(9)struct iommu_hw_queue_alloc(9)
Okay that's the whole page, bestie, you ate that π this one lives at ./include/uapi/linux, dated October 2026, and it's literally titled struct iommu...t_invalidate(9) β go forth and invalidate those caches, no cap π§ββοΈπ
π¦ where this page comes from
- nixpkgs unstable: linux-manual 7.3.1
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on : a language model's rewrite of the original, under its terms