[linux-block.git] / include / linux / vfio.h

/* SPDX-License-Identifier: GPL-2.0-only */
/*
 * VFIO API definition
 *
 * Copyright (C) 2012 Red Hat, Inc.  All rights reserved.
 *     Author: Alex Williamson <alex.williamson@redhat.com>
 */
#ifndef VFIO_H
#define VFIO_H


#include <linux/iommu.h>
#include <linux/mm.h>
#include <linux/workqueue.h>
#include <linux/poll.h>
#include <uapi/linux/vfio.h>

/*
 * VFIO devices can be placed in a set, this allows all devices to share this
 * structure and the VFIO core will provide a lock that is held around
 * open_device()/close_device() for all devices in the set.
 */
struct vfio_device_set {
	void *set_id;
	struct mutex lock;
	struct list_head device_list;
	unsigned int device_count;
};

struct vfio_device {
	struct device *dev;
	const struct vfio_device_ops *ops;
	struct vfio_group *group;
	struct vfio_device_set *dev_set;
	struct list_head dev_set_list;
	unsigned int migration_flags;

	/* Members below here are private, not for driver use */
	refcount_t refcount;
	unsigned int open_count;
	struct completion comp;
	struct list_head group_next;
};

/**
 * struct vfio_device_ops - VFIO bus driver device callbacks
 *
 * @open_device: Called when the first file descriptor is opened for this device
 * @close_device: Opposite of open_device
 * @read: Perform read(2) on device file descriptor
 * @write: Perform write(2) on device file descriptor
 * @ioctl: Perform ioctl(2) on device file descriptor, supporting VFIO_DEVICE_*
 *         operations documented below
 * @mmap: Perform mmap(2) on a region of the device file descriptor
 * @request: Request for the bus driver to release the device
 * @match: Optional device name match callback (return: 0 for no-match, >0 for
 *         match, -errno for abort (ex. match with insufficient or incorrect
 *         additional args)
 * @device_feature: Optional, fill in the VFIO_DEVICE_FEATURE ioctl
 * @migration_set_state: Optional callback to change the migration state for
 *         devices that support migration. It's mandatory for
 *         VFIO_DEVICE_FEATURE_MIGRATION migration support.
 *         The returned FD is used for data transfer according to the FSM
 *         definition. The driver is responsible to ensure that FD reaches end
 *         of stream or error whenever the migration FSM leaves a data transfer
 *         state or before close_device() returns.
 * @migration_get_state: Optional callback to get the migration state for
 *         devices that support migration. It's mandatory for
 *         VFIO_DEVICE_FEATURE_MIGRATION migration support.
 */
struct vfio_device_ops {
	char	*name;
	int	(*open_device)(struct vfio_device *vdev);
	void	(*close_device)(struct vfio_device *vdev);
	ssize_t	(*read)(struct vfio_device *vdev, char __user *buf,
			size_t count, loff_t *ppos);
	ssize_t	(*write)(struct vfio_device *vdev, const char __user *buf,
			 size_t count, loff_t *size);
	long	(*ioctl)(struct vfio_device *vdev, unsigned int cmd,
			 unsigned long arg);
	int	(*mmap)(struct vfio_device *vdev, struct vm_area_struct *vma);
	void	(*request)(struct vfio_device *vdev, unsigned int count);
	int	(*match)(struct vfio_device *vdev, char *buf);
	int	(*device_feature)(struct vfio_device *device, u32 flags,
				  void __user *arg, size_t argsz);
	struct file *(*migration_set_state)(
		struct vfio_device *device,
		enum vfio_device_mig_state new_state);
	int (*migration_get_state)(struct vfio_device *device,
				   enum vfio_device_mig_state *curr_state);
};

/**
 * vfio_check_feature - Validate user input for the VFIO_DEVICE_FEATURE ioctl
 * @flags: Arg from the device_feature op
 * @argsz: Arg from the device_feature op
 * @supported_ops: Combination of VFIO_DEVICE_FEATURE_GET and SET the driver
 *                 supports
 * @minsz: Minimum data size the driver accepts
 *
 * For use in a driver's device_feature op. Checks that the inputs to the
 * VFIO_DEVICE_FEATURE ioctl are correct for the driver's feature. Returns 1 if
 * the driver should execute the get or set, otherwise the relevant
 * value should be returned.
 */
static inline int vfio_check_feature(u32 flags, size_t argsz, u32 supported_ops,
				    size_t minsz)
{
	if ((flags & (VFIO_DEVICE_FEATURE_GET | VFIO_DEVICE_FEATURE_SET)) &
	    ~supported_ops)
		return -EINVAL;
	if (flags & VFIO_DEVICE_FEATURE_PROBE)
		return 0;
	/* Without PROBE one of GET or SET must be requested */
	if (!(flags & (VFIO_DEVICE_FEATURE_GET | VFIO_DEVICE_FEATURE_SET)))
		return -EINVAL;
	if (argsz < minsz)
		return -EINVAL;
	return 1;
}

void vfio_init_group_dev(struct vfio_device *device, struct device *dev,
			 const struct vfio_device_ops *ops);
void vfio_uninit_group_dev(struct vfio_device *device);
int vfio_register_group_dev(struct vfio_device *device);
int vfio_register_emulated_iommu_dev(struct vfio_device *device);
void vfio_unregister_group_dev(struct vfio_device *device);
extern struct vfio_device *vfio_device_get_from_dev(struct device *dev);
extern void vfio_device_put(struct vfio_device *device);

int vfio_assign_device_set(struct vfio_device *device, void *set_id);

int vfio_mig_get_next_state(struct vfio_device *device,
			    enum vfio_device_mig_state cur_fsm,
			    enum vfio_device_mig_state new_fsm,
			    enum vfio_device_mig_state *next_fsm);

/*
 * External user API
 */
extern struct vfio_group *vfio_group_get_external_user(struct file *filep);
extern void vfio_group_put_external_user(struct vfio_group *group);
extern struct vfio_group *vfio_group_get_external_user_from_dev(struct device
								*dev);
extern bool vfio_external_group_match_file(struct vfio_group *group,
					   struct file *filep);
extern int vfio_external_user_iommu_id(struct vfio_group *group);
extern long vfio_external_check_extension(struct vfio_group *group,
					  unsigned long arg);

#define VFIO_PIN_PAGES_MAX_ENTRIES	(PAGE_SIZE/sizeof(unsigned long))

extern int vfio_pin_pages(struct device *dev, unsigned long *user_pfn,
			  int npage, int prot, unsigned long *phys_pfn);
extern int vfio_unpin_pages(struct device *dev, unsigned long *user_pfn,
			    int npage);

extern int vfio_group_pin_pages(struct vfio_group *group,
				unsigned long *user_iova_pfn, int npage,
				int prot, unsigned long *phys_pfn);
extern int vfio_group_unpin_pages(struct vfio_group *group,
				  unsigned long *user_iova_pfn, int npage);

extern int vfio_dma_rw(struct vfio_group *group, dma_addr_t user_iova,
		       void *data, size_t len, bool write);

extern struct iommu_domain *vfio_group_iommu_domain(struct vfio_group *group);

/* each type has independent events */
enum vfio_notify_type {
	VFIO_IOMMU_NOTIFY = 0,
	VFIO_GROUP_NOTIFY = 1,
};

/* events for VFIO_IOMMU_NOTIFY */
#define VFIO_IOMMU_NOTIFY_DMA_UNMAP	BIT(0)

/* events for VFIO_GROUP_NOTIFY */
#define VFIO_GROUP_NOTIFY_SET_KVM	BIT(0)

extern int vfio_register_notifier(struct device *dev,
				  enum vfio_notify_type type,
				  unsigned long *required_events,
				  struct notifier_block *nb);
extern int vfio_unregister_notifier(struct device *dev,
				    enum vfio_notify_type type,
				    struct notifier_block *nb);

struct kvm;
extern void vfio_group_set_kvm(struct vfio_group *group, struct kvm *kvm);

/*
 * Sub-module helpers
 */
struct vfio_info_cap {
	struct vfio_info_cap_header *buf;
	size_t size;
};
extern struct vfio_info_cap_header *vfio_info_cap_add(
		struct vfio_info_cap *caps, size_t size, u16 id, u16 version);
extern void vfio_info_cap_shift(struct vfio_info_cap *caps, size_t offset);

extern int vfio_info_add_capability(struct vfio_info_cap *caps,
				    struct vfio_info_cap_header *cap,
				    size_t size);

extern int vfio_set_irqs_validate_and_prepare(struct vfio_irq_set *hdr,
					      int num_irqs, int max_irq_type,
					      size_t *data_size);

struct pci_dev;
#if IS_ENABLED(CONFIG_VFIO_SPAPR_EEH)
extern void vfio_spapr_pci_eeh_open(struct pci_dev *pdev);
extern void vfio_spapr_pci_eeh_release(struct pci_dev *pdev);
extern long vfio_spapr_iommu_eeh_ioctl(struct iommu_group *group,
				       unsigned int cmd,
				       unsigned long arg);
#else
static inline void vfio_spapr_pci_eeh_open(struct pci_dev *pdev)
{
}

static inline void vfio_spapr_pci_eeh_release(struct pci_dev *pdev)
{
}

static inline long vfio_spapr_iommu_eeh_ioctl(struct iommu_group *group,
					      unsigned int cmd,
					      unsigned long arg)
{
	return -ENOTTY;
}
#endif /* CONFIG_VFIO_SPAPR_EEH */

/*
 * IRQfd - generic
 */
struct virqfd {
	void			*opaque;
	struct eventfd_ctx	*eventfd;
	int			(*handler)(void *, void *);
	void			(*thread)(void *, void *);
	void			*data;
	struct work_struct	inject;
	wait_queue_entry_t		wait;
	poll_table		pt;
	struct work_struct	shutdown;
	struct virqfd		**pvirqfd;
};

extern int vfio_virqfd_enable(void *opaque,
			      int (*handler)(void *, void *),
			      void (*thread)(void *, void *),
			      void *data, struct virqfd **pvirqfd, int fd);
extern void vfio_virqfd_disable(struct virqfd **pvirqfd);

#endif /* VFIO_H */
Commit	Line	Data
d2912cb1	1	/* SPDX-License-Identifier: GPL-2.0-only */
cba3345c AW	2	/*
	3	* VFIO API definition
	4	*
	5	* Copyright (C) 2012 Red Hat, Inc. All rights reserved.
	6	* Author: Alex Williamson <alex.williamson@redhat.com>
cba3345c AW	7	*/
	8	#ifndef VFIO_H
	9	#define VFIO_H
	10
cba3345c AW	11
	12	#include <linux/iommu.h>
	13	#include <linux/mm.h>
7e992d69 AM	14	#include <linux/workqueue.h>
7e992d69 AM	15	#include <linux/poll.h>
607ca46e	16	#include <uapi/linux/vfio.h>
cba3345c	17
2fd585f4 JG	18	/*
	19	* VFIO devices can be placed in a set, this allows all devices to share this
	20	* structure and the VFIO core will provide a lock that is held around
	21	* open_device()/close_device() for all devices in the set.
	22	*/
	23	struct vfio_device_set {
	24	void *set_id;
	25	struct mutex lock;
	26	struct list_head device_list;
	27	unsigned int device_count;
	28	};
	29
0bfc6a4e JG	30	struct vfio_device {
	31	struct device *dev;
	32	const struct vfio_device_ops *ops;
	33	struct vfio_group *group;
2fd585f4 JG	34	struct vfio_device_set *dev_set;
2fd585f4 JG	35	struct list_head dev_set_list;
8cb3d83b	36	unsigned int migration_flags;
0bfc6a4e JG	37
	38	/* Members below here are private, not for driver use */
	39	refcount_t refcount;
2fd585f4	40	unsigned int open_count;
0bfc6a4e JG	41	struct completion comp;
0bfc6a4e JG	42	struct list_head group_next;
0bfc6a4e JG	43	};
0bfc6a4e JG	44
cba3345c AW	45	/**
	46	* struct vfio_device_ops - VFIO bus driver device callbacks
	47	*
2fd585f4 JG	48	* @open_device: Called when the first file descriptor is opened for this device
2fd585f4 JG	49	* @close_device: Opposite of open_device
cba3345c AW	50	* @read: Perform read(2) on device file descriptor
	51	* @write: Perform write(2) on device file descriptor
	52	* @ioctl: Perform ioctl(2) on device file descriptor, supporting VFIO_DEVICE_*
	53	* operations documented below
	54	* @mmap: Perform mmap(2) on a region of the device file descriptor
13060b64	55	* @request: Request for the bus driver to release the device
5f3874c2 AW	56	* @match: Optional device name match callback (return: 0 for no-match, >0 for
	57	* match, -errno for abort (ex. match with insufficient or incorrect
	58	* additional args)
445ad495	59	* @device_feature: Optional, fill in the VFIO_DEVICE_FEATURE ioctl
115dcec6 JG	60	* @migration_set_state: Optional callback to change the migration state for
	61	* devices that support migration. It's mandatory for
	62	* VFIO_DEVICE_FEATURE_MIGRATION migration support.
	63	* The returned FD is used for data transfer according to the FSM
	64	* definition. The driver is responsible to ensure that FD reaches end
	65	* of stream or error whenever the migration FSM leaves a data transfer
	66	* state or before close_device() returns.
	67	* @migration_get_state: Optional callback to get the migration state for
	68	* devices that support migration. It's mandatory for
	69	* VFIO_DEVICE_FEATURE_MIGRATION migration support.
cba3345c AW	70	*/
	71	struct vfio_device_ops {
	72	char *name;
2fd585f4 JG	73	int (open_device)(struct vfio_device vdev);
2fd585f4 JG	74	void (close_device)(struct vfio_device vdev);
6df62c5b	75	ssize_t (read)(struct vfio_device vdev, char __user *buf,
cba3345c	76	size_t count, loff_t *ppos);
6df62c5b	77	ssize_t (write)(struct vfio_device vdev, const char __user *buf,
cba3345c	78	size_t count, loff_t *size);
6df62c5b	79	long (ioctl)(struct vfio_device vdev, unsigned int cmd,
cba3345c	80	unsigned long arg);
6df62c5b JG	81	int (mmap)(struct vfio_device vdev, struct vm_area_struct *vma);
	82	void (request)(struct vfio_device vdev, unsigned int count);
	83	int (match)(struct vfio_device vdev, char *buf);
445ad495 JG	84	int (device_feature)(struct vfio_device device, u32 flags,
445ad495 JG	85	void __user *arg, size_t argsz);
115dcec6 JG	86	struct file (migration_set_state)(
	87	struct vfio_device *device,
	88	enum vfio_device_mig_state new_state);
	89	int (migration_get_state)(struct vfio_device device,
	90	enum vfio_device_mig_state *curr_state);
cba3345c AW	91	};
cba3345c AW	92
445ad495 JG	93	/**
	94	* vfio_check_feature - Validate user input for the VFIO_DEVICE_FEATURE ioctl
	95	* @flags: Arg from the device_feature op
	96	* @argsz: Arg from the device_feature op
	97	* @supported_ops: Combination of VFIO_DEVICE_FEATURE_GET and SET the driver
	98	* supports
	99	* @minsz: Minimum data size the driver accepts
	100	*
	101	* For use in a driver's device_feature op. Checks that the inputs to the
	102	* VFIO_DEVICE_FEATURE ioctl are correct for the driver's feature. Returns 1 if
	103	* the driver should execute the get or set, otherwise the relevant
	104	* value should be returned.
	105	*/
	106	static inline int vfio_check_feature(u32 flags, size_t argsz, u32 supported_ops,
	107	size_t minsz)
	108	{
	109	if ((flags & (VFIO_DEVICE_FEATURE_GET \| VFIO_DEVICE_FEATURE_SET)) &
	110	~supported_ops)
	111	return -EINVAL;
	112	if (flags & VFIO_DEVICE_FEATURE_PROBE)
	113	return 0;
	114	/* Without PROBE one of GET or SET must be requested */
	115	if (!(flags & (VFIO_DEVICE_FEATURE_GET \| VFIO_DEVICE_FEATURE_SET)))
	116	return -EINVAL;
	117	if (argsz < minsz)
	118	return -EINVAL;
	119	return 1;
	120	}
	121
0bfc6a4e	122	void vfio_init_group_dev(struct vfio_device device, struct device dev,
1e04ec14	123	const struct vfio_device_ops *ops);
ae03c377	124	void vfio_uninit_group_dev(struct vfio_device *device);
0bfc6a4e	125	int vfio_register_group_dev(struct vfio_device *device);
c68ea0d0	126	int vfio_register_emulated_iommu_dev(struct vfio_device *device);
0bfc6a4e	127	void vfio_unregister_group_dev(struct vfio_device *device);
44f50716 VMP	128	extern struct vfio_device vfio_device_get_from_dev(struct device dev);
44f50716 VMP	129	extern void vfio_device_put(struct vfio_device *device);
cba3345c	130
2fd585f4 JG	131	int vfio_assign_device_set(struct vfio_device device, void set_id);
2fd585f4 JG	132
115dcec6 JG	133	int vfio_mig_get_next_state(struct vfio_device *device,
	134	enum vfio_device_mig_state cur_fsm,
	135	enum vfio_device_mig_state new_fsm,
	136	enum vfio_device_mig_state *next_fsm);
	137
6cdd9782 AK	138	/*
	139	* External user API
	140	*/
	141	extern struct vfio_group vfio_group_get_external_user(struct file filep);
	142	extern void vfio_group_put_external_user(struct vfio_group *group);
c0560f51 YZ	143	extern struct vfio_group *vfio_group_get_external_user_from_dev(struct device
c0560f51 YZ	144	*dev);
5d6dee80 AW	145	extern bool vfio_external_group_match_file(struct vfio_group *group,
5d6dee80 AW	146	struct file *filep);
6cdd9782	147	extern int vfio_external_user_iommu_id(struct vfio_group *group);
88d7ab89 AW	148	extern long vfio_external_check_extension(struct vfio_group *group,
88d7ab89 AW	149	unsigned long arg);
6cdd9782	150
2169037d KW	151	#define VFIO_PIN_PAGES_MAX_ENTRIES (PAGE_SIZE/sizeof(unsigned long))
	152
	153	extern int vfio_pin_pages(struct device dev, unsigned long user_pfn,
	154	int npage, int prot, unsigned long *phys_pfn);
	155	extern int vfio_unpin_pages(struct device dev, unsigned long user_pfn,
	156	int npage);
	157
40280cf7 YZ	158	extern int vfio_group_pin_pages(struct vfio_group *group,
	159	unsigned long *user_iova_pfn, int npage,
	160	int prot, unsigned long *phys_pfn);
	161	extern int vfio_group_unpin_pages(struct vfio_group *group,
	162	unsigned long *user_iova_pfn, int npage);
	163
8d46c0cc YZ	164	extern int vfio_dma_rw(struct vfio_group *group, dma_addr_t user_iova,
	165	void *data, size_t len, bool write);
	166
bdfae1c9 LB	167	extern struct iommu_domain vfio_group_iommu_domain(struct vfio_group group);
bdfae1c9 LB	168
22195cbd JS	169	/* each type has independent events */
	170	enum vfio_notify_type {
	171	VFIO_IOMMU_NOTIFY = 0,
ccd46dba	172	VFIO_GROUP_NOTIFY = 1,
22195cbd JS	173	};
	174
	175	/* events for VFIO_IOMMU_NOTIFY */
	176	#define VFIO_IOMMU_NOTIFY_DMA_UNMAP BIT(0)
c086de81	177
ccd46dba JS	178	/* events for VFIO_GROUP_NOTIFY */
	179	#define VFIO_GROUP_NOTIFY_SET_KVM BIT(0)
	180
c086de81	181	extern int vfio_register_notifier(struct device *dev,
22195cbd JS	182	enum vfio_notify_type type,
22195cbd JS	183	unsigned long *required_events,
c086de81	184	struct notifier_block *nb);
c086de81	185	extern int vfio_unregister_notifier(struct device *dev,
22195cbd	186	enum vfio_notify_type type,
c086de81 KW	187	struct notifier_block *nb);
c086de81 KW	188
ccd46dba JS	189	struct kvm;
	190	extern void vfio_group_set_kvm(struct vfio_group group, struct kvm kvm);
	191
d7a8d5ed AW	192	/*
	193	* Sub-module helpers
	194	*/
	195	struct vfio_info_cap {
	196	struct vfio_info_cap_header *buf;
	197	size_t size;
	198	};
	199	extern struct vfio_info_cap_header *vfio_info_cap_add(
	200	struct vfio_info_cap *caps, size_t size, u16 id, u16 version);
	201	extern void vfio_info_cap_shift(struct vfio_info_cap *caps, size_t offset);
	202
b3c0a866	203	extern int vfio_info_add_capability(struct vfio_info_cap *caps,
dda01f78 AW	204	struct vfio_info_cap_header *cap,
dda01f78 AW	205	size_t size);
b3c0a866	206
c747f08a KW	207	extern int vfio_set_irqs_validate_and_prepare(struct vfio_irq_set *hdr,
	208	int num_irqs, int max_irq_type,
	209	size_t *data_size);
	210
92d18a68	211	struct pci_dev;
bb67b496	212	#if IS_ENABLED(CONFIG_VFIO_SPAPR_EEH)
9b936c96	213	extern void vfio_spapr_pci_eeh_open(struct pci_dev *pdev);
1b69be5e GS	214	extern void vfio_spapr_pci_eeh_release(struct pci_dev *pdev);
	215	extern long vfio_spapr_iommu_eeh_ioctl(struct iommu_group *group,
	216	unsigned int cmd,
	217	unsigned long arg);
	218	#else
9b936c96	219	static inline void vfio_spapr_pci_eeh_open(struct pci_dev *pdev)
1b69be5e	220	{
1b69be5e GS	221	}
	222
	223	static inline void vfio_spapr_pci_eeh_release(struct pci_dev *pdev)
	224	{
	225	}
	226
	227	static inline long vfio_spapr_iommu_eeh_ioctl(struct iommu_group *group,
	228	unsigned int cmd,
	229	unsigned long arg)
	230	{
	231	return -ENOTTY;
	232	}
bb67b496	233	#endif /* CONFIG_VFIO_SPAPR_EEH */
7e992d69 AM	234
	235	/*
	236	* IRQfd - generic
	237	*/
	238	struct virqfd {
	239	void *opaque;
	240	struct eventfd_ctx *eventfd;
	241	int (handler)(void , void *);
	242	void (thread)(void , void *);
	243	void *data;
	244	struct work_struct inject;
ac6424b9	245	wait_queue_entry_t wait;
7e992d69 AM	246	poll_table pt;
	247	struct work_struct shutdown;
	248	struct virqfd **pvirqfd;
	249	};
	250
7e992d69 AM	251	extern int vfio_virqfd_enable(void *opaque,
	252	int (handler)(void , void *),
	253	void (thread)(void , void *),
	254	void data, struct virqfd *pvirqfd, int fd);
	255	extern void vfio_virqfd_disable(struct virqfd **pvirqfd);
	256
cba3345c	257	#endif /* VFIO_H */