IOMMUFD¶

Author: Jason Gunthorpe
Author: Kevin Tian

Overview¶

IOMMUFD is the user API to control the IOMMU subsystem as it relates to managing IO page tables from userspace using file descriptors. It intends to be general and consumable by any driver that wants to expose DMA to userspace. These drivers are eventually expected to deprecate any internal IOMMU logic they may already/historically implement (e.g. vfio_iommu_type1.c).

At minimum iommufd provides universal support of managing I/O address spaces and I/O page tables for all IOMMUs, with room in the design to add non-generic features to cater to specific hardware functionality.

In this context the capital letter (IOMMUFD) refers to the subsystem while the small letter (iommufd) refers to the file descriptors created via /dev/iommu for use by userspace.

Key Concepts¶

User Visible Objects¶

Following IOMMUFD objects are exposed to userspace:

IOMMUFD_OBJ_IOAS, representing an I/O address space (IOAS), allowing map/unmap of user space memory into ranges of I/O Virtual Address (IOVA).

The IOAS is a functional replacement for the VFIO container, and like the VFIO container it copies an IOVA map to a list of iommu_domains held within it.
IOMMUFD_OBJ_DEVICE, representing a device that is bound to iommufd by an external driver.
IOMMUFD_OBJ_HW_PAGETABLE, representing an actual hardware I/O page table (i.e. a single struct iommu_domain) managed by the iommu driver.

The IOAS has a list of HW_PAGETABLES that share the same IOVA mapping and it will synchronize its mapping with each member HW_PAGETABLE.

All user-visible objects are destroyed via the IOMMU_DESTROY uAPI.

The diagram below shows relationship between user-visible objects and kernel datastructures (external to iommufd), with numbers referred to operations creating the objects and links:

 _________________________________________________________
|                         iommufd                         |
|       [1]                                               |
|  _________________                                      |
| |                 |                                     |
| |                 |                                     |
| |                 |                                     |
| |                 |                                     |
| |                 |                                     |
| |                 |                                     |
| |                 |        [3]                 [2]      |
| |                 |    ____________         __________  |
| |      IOAS       |<--|            |<------|          | |
| |                 |   |HW_PAGETABLE|       |  DEVICE  | |
| |                 |   |____________|       |__________| |
| |                 |         |                   |       |
| |                 |         |                   |       |
| |                 |         |                   |       |
| |                 |         |                   |       |
| |                 |         |                   |       |
| |_________________|         |                   |       |
|         |                   |                   |       |
|_________|___________________|___________________|_______|
          |                   |                   |
          |              _____v______      _______v_____
          | PFN storage |            |    |             |
          |------------>|iommu_domain|    |struct device|
                        |____________|    |_____________|

IOMMUFD_OBJ_IOAS is created via the IOMMU_IOAS_ALLOC uAPI. An iommufd can hold multiple IOAS objects. IOAS is the most generic object and does not expose interfaces that are specific to single IOMMU drivers. All operations on the IOAS must operate equally on each of the iommu_domains inside of it.
IOMMUFD_OBJ_DEVICE is created when an external driver calls the IOMMUFD kAPI to bind a device to an iommufd. The driver is expected to implement a set of ioctls to allow userspace to initiate the binding operation. Successful completion of this operation establishes the desired DMA ownership over the device. The driver must also set the driver_managed_dma flag and must not touch the device until this operation succeeds.
IOMMUFD_OBJ_HW_PAGETABLE is created when an external driver calls the IOMMUFD kAPI to attach a bound device to an IOAS. Similarly the external driver uAPI allows userspace to initiate the attaching operation. If a compatible pagetable already exists then it is reused for the attachment. Otherwise a new pagetable object and iommu_domain is created. Successful completion of this operation sets up the linkages among IOAS, device and iommu_domain. Once this completes the device could do DMA.

Every iommu_domain inside the IOAS is also represented to userspace as a HW_PAGETABLE object.

Note

Future IOMMUFD updates will provide an API to create and manipulate the HW_PAGETABLE directly.

A device can only bind to an iommufd due to DMA ownership claim and attach to at most one IOAS object (no support of PASID yet).

Kernel Datastructure¶

User visible objects are backed by following datastructures:

iommufd_ioas for IOMMUFD_OBJ_IOAS.
iommufd_device for IOMMUFD_OBJ_DEVICE.
iommufd_hw_pagetable for IOMMUFD_OBJ_HW_PAGETABLE.

Several terminologies when looking at these datastructures:

Automatic domain - refers to an iommu domain created automatically when attaching a device to an IOAS object. This is compatible to the semantics of VFIO type1.
Manual domain - refers to an iommu domain designated by the user as the target pagetable to be attached to by a device. Though currently there are no uAPIs to directly create such domain, the datastructure and algorithms are ready for handling that use case.
In-kernel user - refers to something like a VFIO mdev that is using the IOMMUFD access interface to access the IOAS. This starts by creating an iommufd_access object that is similar to the domain binding a physical device would do. The access object will then allow converting IOVA ranges into struct page * lists, or doing direct read/write to an IOVA.

iommufd_ioas serves as the metadata datastructure to manage how IOVA ranges are mapped to memory pages, composed of:

struct io_pagetable holding the IOVA map
struct iopt_area’s representing populated portions of IOVA
struct iopt_pages representing the storage of PFNs
struct iommu_domain representing the IO page table in the IOMMU
struct iopt_pages_access representing in-kernel users of PFNs
struct xarray pinned_pfns holding a list of pages pinned by in-kernel users

Each iopt_pages represents a logical linear array of full PFNs. The PFNs are ultimately derived from userspace VAs via an mm_struct. Once they have been pinned the PFNs are stored in IOPTEs of an iommu_domain or inside the pinned_pfns xarray if they have been pinned through an iommufd_access.

PFN have to be copied between all combinations of storage locations, depending on what domains are present and what kinds of in-kernel “software access” users exist. The mechanism ensures that a page is pinned only once.

An io_pagetable is composed of iopt_areas pointing at iopt_pages, along with a list of iommu_domains that mirror the IOVA to PFN map.

Multiple io_pagetable-s, through their iopt_area-s, can share a single iopt_pages which avoids multi-pinning and double accounting of page consumption.

iommufd_ioas is shareable between subsystems, e.g. VFIO and VDPA, as long as devices managed by different subsystems are bound to a same iommufd.

IOMMUFD User API¶

General ioctl format

The ioctl interface follows a general format to allow for extensibility. Each ioctl is passed in a structure pointer as the argument providing the size of the structure in the first u32. The kernel checks that any structure space beyond what it understands is 0. This allows userspace to use the backward compatible portion while consistently using the newer, larger, structures.

ioctls use a standard meaning for common errnos:

ENOTTY: The IOCTL number itself is not supported at all

E2BIG: The IOCTL number is supported, but the provided structure has non-zero in a part the kernel does not understand.

EOPNOTSUPP: The IOCTL number is supported, and the structure is understood, however a known field has a value the kernel does not understand or support.

EINVAL: Everything about the IOCTL was understood, but a field is not correct.

ENOENT: An ID or IOVA provided does not exist.

ENOMEM: Out of memory.

EOVERFLOW: Mathematics overflowed.

As well as additional errnos, within specific ioctls.

struct iommu_destroy¶: ioctl(IOMMU_DESTROY)

Definition:

struct iommu_destroy {
    __u32 size;
    __u32 id;
};

Members

size: sizeof(struct iommu_destroy)
id: iommufd object ID to destroy. Can be any destroyable object type.

Description

Destroy any object held within iommufd.

struct iommu_ioas_alloc¶: ioctl(IOMMU_IOAS_ALLOC)

Definition:

struct iommu_ioas_alloc {
    __u32 size;
    __u32 flags;
    __u32 out_ioas_id;
};

Members

size: sizeof(struct iommu_ioas_alloc)
flags: Must be 0
out_ioas_id: Output IOAS ID for the allocated object

Description

Allocate an IO Address Space (IOAS) which holds an IO Virtual Address (IOVA) to memory mapping.

struct iommu_iova_range¶: ioctl(IOMMU_IOVA_RANGE)

Definition:

struct iommu_iova_range {
    __aligned_u64 start;
    __aligned_u64 last;
};

Members

start: First IOVA
last: Inclusive last IOVA

Description

An interval in IOVA space.

struct iommu_ioas_iova_ranges¶: ioctl(IOMMU_IOAS_IOVA_RANGES)

Definition:

struct iommu_ioas_iova_ranges {
    __u32 size;
    __u32 ioas_id;
    __u32 num_iovas;
    __u32 __reserved;
    __aligned_u64 allowed_iovas;
    __aligned_u64 out_iova_alignment;
};

Members

size: sizeof(struct iommu_ioas_iova_ranges)
ioas_id: IOAS ID to read ranges from
num_iovas: Input/Output total number of ranges in the IOAS
__reserved: Must be 0
allowed_iovas: Pointer to the output array of struct iommu_iova_range
out_iova_alignment: Minimum alignment required for mapping IOVA

Description

Query an IOAS for ranges of allowed IOVAs. Mapping IOVA outside these ranges is not allowed. num_iovas will be set to the total number of iovas and the allowed_iovas[] will be filled in as space permits.

The allowed ranges are dependent on the HW path the DMA operation takes, and can change during the lifetime of the IOAS. A fresh empty IOAS will have a full range, and each attached device will narrow the ranges based on that device’s HW restrictions. Detaching a device can widen the ranges. Userspace should query ranges after every attach/detach to know what IOVAs are valid for mapping.

On input num_iovas is the length of the allowed_iovas array. On output it is the total number of iovas filled in. The ioctl will return -EMSGSIZE and set num_iovas to the required value if num_iovas is too small. In this case the caller should allocate a larger output array and re-issue the ioctl.

out_iova_alignment returns the minimum IOVA alignment that can be given to IOMMU_IOAS_MAP/COPY. IOVA’s must satisfy:

starting_iova % out_iova_alignment == 0
(starting_iova + length) % out_iova_alignment == 0

out_iova_alignment can be 1 indicating any IOVA is allowed. It cannot be higher than the system PAGE_SIZE.

struct iommu_ioas_allow_iovas¶: ioctl(IOMMU_IOAS_ALLOW_IOVAS)

Definition:

struct iommu_ioas_allow_iovas {
    __u32 size;
    __u32 ioas_id;
    __u32 num_iovas;
    __u32 __reserved;
    __aligned_u64 allowed_iovas;
};

Members

size: sizeof(struct iommu_ioas_allow_iovas)
ioas_id: IOAS ID to allow IOVAs from
num_iovas: Input/Output total number of ranges in the IOAS
__reserved: Must be 0
allowed_iovas: Pointer to array of struct iommu_iova_range

Description

Ensure a range of IOVAs are always available for allocation. If this call succeeds then IOMMU_IOAS_IOVA_RANGES will never return a list of IOVA ranges that are narrower than the ranges provided here. This call will fail if IOMMU_IOAS_IOVA_RANGES is currently narrower than the given ranges.

When an IOAS is first created the IOVA_RANGES will be maximally sized, and as devices are attached the IOVA will narrow based on the device restrictions. When an allowed range is specified any narrowing will be refused, ie device attachment can fail if the device requires limiting within the allowed range.

Automatic IOVA allocation is also impacted by this call. MAP will only allocate within the allowed IOVAs if they are present.

This call replaces the entire allowed list with the given list.

enum iommufd_ioas_map_flags¶: Flags for map and copy

Constants

IOMMU_IOAS_MAP_FIXED_IOVA: If clear the kernel will compute an appropriate IOVA to place the mapping at
IOMMU_IOAS_MAP_WRITEABLE: DMA is allowed to write to this mapping
IOMMU_IOAS_MAP_READABLE: DMA is allowed to read from this mapping

struct iommu_ioas_map¶: ioctl(IOMMU_IOAS_MAP)

Definition:

struct iommu_ioas_map {
    __u32 size;
    __u32 flags;
    __u32 ioas_id;
    __u32 __reserved;
    __aligned_u64 user_va;
    __aligned_u64 length;
    __aligned_u64 iova;
};

Members

size: sizeof(struct iommu_ioas_map)
flags: Combination of enum iommufd_ioas_map_flags
ioas_id: IOAS ID to change the mapping of
__reserved: Must be 0
user_va: Userspace pointer to start mapping from
length: Number of bytes to map
iova: IOVA the mapping was placed at. If IOMMU_IOAS_MAP_FIXED_IOVA is set then this must be provided as input.

Description

Set an IOVA mapping from a user pointer. If FIXED_IOVA is specified then the mapping will be established at iova, otherwise a suitable location based on the reserved and allowed lists will be automatically selected and returned in iova.

If IOMMU_IOAS_MAP_FIXED_IOVA is specified then the iova range must currently be unused, existing IOVA cannot be replaced.

struct iommu_ioas_copy¶: ioctl(IOMMU_IOAS_COPY)

Definition:

struct iommu_ioas_copy {
    __u32 size;
    __u32 flags;
    __u32 dst_ioas_id;
    __u32 src_ioas_id;
    __aligned_u64 length;
    __aligned_u64 dst_iova;
    __aligned_u64 src_iova;
};

Members

size: sizeof(struct iommu_ioas_copy)
flags: Combination of enum iommufd_ioas_map_flags
dst_ioas_id: IOAS ID to change the mapping of
src_ioas_id: IOAS ID to copy from
length: Number of bytes to copy and map
dst_iova: IOVA the mapping was placed at. If IOMMU_IOAS_MAP_FIXED_IOVA is set then this must be provided as input.
src_iova: IOVA to start the copy

Description

Copy an already existing mapping from src_ioas_id and establish it in dst_ioas_id. The src iova/length must exactly match a range used with IOMMU_IOAS_MAP.

This may be used to efficiently clone a subset of an IOAS to another, or as a kind of ‘cache’ to speed up mapping. Copy has an efficiency advantage over establishing equivalent new mappings, as internal resources are shared, and the kernel will pin the user memory only once.

struct iommu_ioas_unmap¶: ioctl(IOMMU_IOAS_UNMAP)

Definition:

struct iommu_ioas_unmap {
    __u32 size;
    __u32 ioas_id;
    __aligned_u64 iova;
    __aligned_u64 length;
};

Members

size: sizeof(struct iommu_ioas_unmap)
ioas_id: IOAS ID to change the mapping of
iova: IOVA to start the unmapping at
length: Number of bytes to unmap, and return back the bytes unmapped

Description

Unmap an IOVA range. The iova/length must be a superset of a previously mapped range used with IOMMU_IOAS_MAP or IOMMU_IOAS_COPY. Splitting or truncating ranges is not allowed. The values 0 to U64_MAX will unmap everything.

enum iommufd_option¶: ioctl(IOMMU_OPTION_RLIMIT_MODE) and ioctl(IOMMU_OPTION_HUGE_PAGES)

Constants

IOMMU_OPTION_RLIMIT_MODE

Change how RLIMIT_MEMLOCK accounting works. The caller must have privilege to invoke this. Value 0 (default) is user based accouting, 1 uses process based accounting. Global option, object_id must be 0

IOMMU_OPTION_HUGE_PAGES

Value 1 (default) allows contiguous pages to be combined when generating iommu mappings. Value 0 disables combining, everything is mapped to PAGE_SIZE. This can be useful for benchmarking. This is a per-IOAS option, the object_id must be the IOAS ID.

enum iommufd_option_ops¶: ioctl(IOMMU_OPTION_OP_SET) and ioctl(IOMMU_OPTION_OP_GET)

Constants

IOMMU_OPTION_OP_SET: Set the option’s value
IOMMU_OPTION_OP_GET: Get the option’s value

struct iommu_option¶: iommu option multiplexer

Definition:

struct iommu_option {
    __u32 size;
    __u32 option_id;
    __u16 op;
    __u16 __reserved;
    __u32 object_id;
    __aligned_u64 val64;
};

Members

size: sizeof(struct iommu_option)
option_id: One of enum iommufd_option
op: One of enum iommufd_option_ops
__reserved: Must be 0
object_id: ID of the object if required
val64: Option value to set or value returned on get

Description

Change a simple option value. This multiplexor allows controlling options on objects. IOMMU_OPTION_OP_SET will load an option and IOMMU_OPTION_OP_GET will return the current value.

enum iommufd_vfio_ioas_op¶: IOMMU_VFIO_IOAS_* ioctls

Constants

IOMMU_VFIO_IOAS_GET: Get the current compatibility IOAS
IOMMU_VFIO_IOAS_SET: Change the current compatibility IOAS
IOMMU_VFIO_IOAS_CLEAR: Disable VFIO compatibility

struct iommu_vfio_ioas¶: ioctl(IOMMU_VFIO_IOAS)

Definition:

struct iommu_vfio_ioas {
    __u32 size;
    __u32 ioas_id;
    __u16 op;
    __u16 __reserved;
};

Members

size: sizeof(struct iommu_vfio_ioas)
ioas_id: For IOMMU_VFIO_IOAS_SET the input IOAS ID to set For IOMMU_VFIO_IOAS_GET will output the IOAS ID
op: One of enum iommufd_vfio_ioas_op
__reserved: Must be 0

Description

The VFIO compatibility support uses a single ioas because VFIO APIs do not support the ID field. Set or Get the IOAS that VFIO compatibility will use. When VFIO_GROUP_SET_CONTAINER is used on an iommufd it will get the compatibility ioas, either by taking what is already set, or auto creating one. From then on VFIO will continue to use that ioas and is not effected by this ioctl. SET or CLEAR does not destroy any auto-created IOAS.

IOMMUFD Kernel API¶

The IOMMUFD kAPI is device-centric with group-related tricks managed behind the scene. This allows the external drivers calling such kAPI to implement a simple device-centric uAPI for connecting its device to an iommufd, instead of explicitly imposing the group semantics in its uAPI as VFIO does.

struct iommufd_device *iommufd_device_bind(struct iommufd_ctx *ictx, struct device *dev, u32 *id)¶: Bind a physical device to an iommu fd

Parameters

struct iommufd_ctx *ictx: iommufd file descriptor
struct device *dev: Pointer to a physical device struct
u32 *id: Output ID number to return to userspace for this device

Description

A successful bind establishes an ownership over the device and returns struct iommufd_device pointer, otherwise returns error pointer.

A driver using this API must set driver_managed_dma and must not touch the device until this routine succeeds and establishes ownership.

Binding a PCI device places the entire RID under iommufd control.

The caller must undo this with iommufd_device_unbind()

void iommufd_device_unbind(struct iommufd_device *idev)¶: Undo iommufd_device_bind()

Parameters

struct iommufd_device *idev: Device returned by iommufd_device_bind()

Description

Release the device from iommufd control. The DMA ownership will return back to unowned with DMA controlled by the DMA API. This invalidates the iommufd_device pointer, other APIs that consume it must not be called concurrently.

int iommufd_device_attach(struct iommufd_device *idev, u32 *pt_id)¶: Connect a device from an iommu_domain

Parameters

struct iommufd_device *idev: device to attach
u32 *pt_id: Input a IOMMUFD_OBJ_IOAS, or IOMMUFD_OBJ_HW_PAGETABLE Output the IOMMUFD_OBJ_HW_PAGETABLE ID

Description

This connects the device to an iommu_domain, either automatically or manually selected. Once this completes the device could do DMA.

The caller should return the resulting pt_id back to userspace. This function is undone by calling iommufd_device_detach().

void iommufd_device_detach(struct iommufd_device *idev)¶: Disconnect a device to an iommu_domain

Parameters

struct iommufd_device *idev: device to detach

Description

Undo iommufd_device_attach(). This disconnects the idev from the previously attached pt_id. The device returns back to a blocked DMA translation.

struct iommufd_access *iommufd_access_create(struct iommufd_ctx *ictx, u32 ioas_id, const struct iommufd_access_ops *ops, void *data)¶: Create an iommufd_access

Parameters

struct iommufd_ctx *ictx: iommufd file descriptor
u32 ioas_id: ID for a IOMMUFD_OBJ_IOAS
const struct iommufd_access_ops *ops: Driver’s ops to associate with the access
void *data: Opaque data to pass into ops functions

Description

An iommufd_access allows a driver to read/write to the IOAS without using DMA. The underlying CPU memory can be accessed using the iommufd_access_pin_pages() or iommufd_access_rw() functions.

The provided ops are required to use iommufd_access_pin_pages().

void iommufd_access_destroy(struct iommufd_access *access)¶: Destroy an iommufd_access

Parameters

struct iommufd_access *access: The access to destroy

Description

The caller must stop using the access before destroying it.

void iommufd_access_unpin_pages(struct iommufd_access *access, unsigned long iova, unsigned long length)¶: Undo iommufd_access_pin_pages

Parameters

struct iommufd_access *access: IOAS access to act on
unsigned long iova: Starting IOVA
unsigned long length: Number of bytes to access

Description

Return the struct page’s. The caller must stop accessing them before calling this. The iova/length must exactly match the one provided to access_pages.

int iommufd_access_pin_pages(struct iommufd_access *access, unsigned long iova, unsigned long length, struct page **out_pages, unsigned int flags)¶: Return a list of pages under the iova

Parameters

struct iommufd_access *access: IOAS access to act on
unsigned long iova: Starting IOVA
unsigned long length: Number of bytes to access
struct page **out_pages: Output page list
unsigned int flags: IOPMMUFD_ACCESS_RW_* flags

Description

Reads length bytes starting at iova and returns the struct page * pointers. These can be kmap’d by the caller for CPU access.

The caller must perform iommufd_access_unpin_pages() when done to balance this.

This API always requires a page aligned iova. This happens naturally if the ioas alignment is >= PAGE_SIZE and the iova is PAGE_SIZE aligned. However smaller alignments have corner cases where this API can fail on otherwise aligned iova.

int iommufd_access_rw(struct iommufd_access *access, unsigned long iova, void *data, size_t length, unsigned int flags)¶: Read or write data under the iova

Parameters

struct iommufd_access *access: IOAS access to act on
unsigned long iova: Starting IOVA
void *data: Kernel buffer to copy to/from
size_t length: Number of bytes to access
unsigned int flags: IOMMUFD_ACCESS_RW_* flags

Description

Copy kernel to/from data into the range given by IOVA/length. If flags indicates IOMMUFD_ACCESS_RW_KTHREAD then a large copy can be optimized by changing it into copy_to/from_user().

void iommufd_ctx_get(struct iommufd_ctx *ictx)¶: Get a context reference

Parameters

struct iommufd_ctx *ictx: Context to get

Description

The caller must already hold a valid reference to ictx.

struct iommufd_ctx *iommufd_ctx_from_file(struct file *file)¶: Acquires a reference to the iommufd context

Parameters

struct file *file: File to obtain the reference from

Description

Returns a pointer to the iommufd_ctx, otherwise ERR_PTR. The struct file remains owned by the caller and the caller must still do fput. On success the caller is responsible to call iommufd_ctx_put().

void iommufd_ctx_put(struct iommufd_ctx *ictx)¶: Put back a reference

Parameters

struct iommufd_ctx *ictx: Context to put back

VFIO and IOMMUFD¶

Connecting a VFIO device to iommufd can be done in two ways.

First is a VFIO compatible way by directly implementing the /dev/vfio/vfio container IOCTLs by mapping them into io_pagetable operations. Doing so allows the use of iommufd in legacy VFIO applications by symlinking /dev/vfio/vfio to /dev/iommufd or extending VFIO to SET_CONTAINER using an iommufd instead of a container fd.

The second approach directly extends VFIO to support a new set of device-centric user API based on aforementioned IOMMUFD kernel API. It requires userspace change but better matches the IOMMUFD API semantics and easier to support new iommufd features when comparing it to the first approach.

Currently both approaches are still work-in-progress.

There are still a few gaps to be resolved to catch up with VFIO type1, as documented in iommufd_vfio_check_extension().

Future TODOs¶

Currently IOMMUFD supports only kernel-managed I/O page table, similar to VFIO type1. New features on the radar include:

Binding iommu_domain’s to PASID/SSID

Userspace page tables, for ARM, x86 and S390

Kernel bypass’d invalidation of user page tables

Re-use of the KVM page table in the IOMMU

Dirty page tracking in the IOMMU

Runtime Increase/Decrease of IOPTE size

PRI support with faults resolved in userspace

The Linux Kernel

Contents

This Page

IOMMUFD¶

Overview¶

Key Concepts¶

User Visible Objects¶

Kernel Datastructure¶

IOMMUFD User API¶

IOMMUFD Kernel API¶

VFIO and IOMMUFD¶

Future TODOs¶