[fio.git] / os / linux / io_uring.h

/* SPDX-License-Identifier: (GPL-2.0 WITH Linux-syscall-note) OR MIT */
/*
 * Header file for the io_uring interface.
 *
 * Copyright (C) 2019 Jens Axboe
 * Copyright (C) 2019 Christoph Hellwig
 */
#ifndef LINUX_IO_URING_H
#define LINUX_IO_URING_H

#include <linux/fs.h>
#include <linux/types.h>

/*
 * IO submission data structure (Submission Queue Entry)
 */
struct io_uring_sqe {
	__u8	opcode;		/* type of operation for this sqe */
	__u8	flags;		/* IOSQE_ flags */
	__u16	ioprio;		/* ioprio for the request */
	__s32	fd;		/* file descriptor to do IO on */
	union {
		__u64	off;	/* offset into file */
		__u64	addr2;
		__u32	cmd_op;
	};
	union {
		__u64	addr;	/* pointer to buffer or iovecs */
		__u64	splice_off_in;
	};
	__u32	len;		/* buffer size or number of iovecs */
	union {
		__kernel_rwf_t	rw_flags;
		__u32		fsync_flags;
		__u16		poll_events;	/* compatibility */
		__u32		poll32_events;	/* word-reversed for BE */
		__u32		sync_range_flags;
		__u32		msg_flags;
		__u32		timeout_flags;
		__u32		accept_flags;
		__u32		cancel_flags;
		__u32		open_flags;
		__u32		statx_flags;
		__u32		fadvise_advice;
		__u32		splice_flags;
		__u32		rename_flags;
		__u32		unlink_flags;
		__u32		hardlink_flags;
		__u32		uring_cmd_flags;
	};
	__u64	user_data;	/* data to be passed back at completion time */
	/* pack this to avoid bogus arm OABI complaints */
	union {
		/* index into fixed buffers, if used */
		__u16	buf_index;
		/* for grouped buffer selection */
		__u16	buf_group;
	} __attribute__((packed));
	/* personality to use, if used */
	__u16	personality;
	union {
		__s32	splice_fd_in;
		__u32	file_index;
	};
	union {
		struct {
			__u64	addr3;
			__u64	__pad2[1];
		};
		/*
		 * If the ring is initialized with IORING_SETUP_SQE128, then
		 * this field is used for 80 bytes of arbitrary command data
		 */
		__u8	cmd[0];
	};
};

enum {
	IOSQE_FIXED_FILE_BIT,
	IOSQE_IO_DRAIN_BIT,
	IOSQE_IO_LINK_BIT,
	IOSQE_IO_HARDLINK_BIT,
	IOSQE_ASYNC_BIT,
	IOSQE_BUFFER_SELECT_BIT,
	IOSQE_CQE_SKIP_SUCCESS_BIT,
};

/*
 * sqe->flags
 */
/* use fixed fileset */
#define IOSQE_FIXED_FILE	(1U << IOSQE_FIXED_FILE_BIT)
/* issue after inflight IO */
#define IOSQE_IO_DRAIN		(1U << IOSQE_IO_DRAIN_BIT)
/* links next sqe */
#define IOSQE_IO_LINK		(1U << IOSQE_IO_LINK_BIT)
/* like LINK, but stronger */
#define IOSQE_IO_HARDLINK	(1U << IOSQE_IO_HARDLINK_BIT)
/* always go async */
#define IOSQE_ASYNC		(1U << IOSQE_ASYNC_BIT)
/* select buffer from sqe->buf_group */
#define IOSQE_BUFFER_SELECT	(1U << IOSQE_BUFFER_SELECT_BIT)
/* don't post CQE if request succeeded */
#define IOSQE_CQE_SKIP_SUCCESS	(1U << IOSQE_CQE_SKIP_SUCCESS_BIT)

/*
 * io_uring_setup() flags
 */
#define IORING_SETUP_IOPOLL	(1U << 0)	/* io_context is polled */
#define IORING_SETUP_SQPOLL	(1U << 1)	/* SQ poll thread */
#define IORING_SETUP_SQ_AFF	(1U << 2)	/* sq_thread_cpu is valid */
#define IORING_SETUP_CQSIZE	(1U << 3)	/* app defines CQ size */
#define IORING_SETUP_CLAMP	(1U << 4)	/* clamp SQ/CQ ring sizes */
#define IORING_SETUP_ATTACH_WQ	(1U << 5)	/* attach to existing wq */
#define IORING_SETUP_R_DISABLED	(1U << 6)	/* start with ring disabled */
#define IORING_SETUP_SUBMIT_ALL	(1U << 7)	/* continue submit on error */
/*
 * Cooperative task running. When requests complete, they often require
 * forcing the submitter to transition to the kernel to complete. If this
 * flag is set, work will be done when the task transitions anyway, rather
 * than force an inter-processor interrupt reschedule. This avoids interrupting
 * a task running in userspace, and saves an IPI.
 */
#define IORING_SETUP_COOP_TASKRUN	(1U << 8)
/*
 * If COOP_TASKRUN is set, get notified if task work is available for
 * running and a kernel transition would be needed to run it. This sets
 * IORING_SQ_TASKRUN in the sq ring flags. Not valid with COOP_TASKRUN.
 */
#define IORING_SETUP_TASKRUN_FLAG	(1U << 9)

#define IORING_SETUP_SQE128		(1U << 10) /* SQEs are 128 byte */
#define IORING_SETUP_CQE32		(1U << 11) /* CQEs are 32 byte */

/*
 * Only one task is allowed to submit requests
 */
#define IORING_SETUP_SINGLE_ISSUER	(1U << 12)

/*
 * Defer running task work to get events.
 * Rather than running bits of task work whenever the task transitions
 * try to do it just before it is needed.
 */
#define IORING_SETUP_DEFER_TASKRUN	(1U << 13)

enum {
	IORING_OP_NOP,
	IORING_OP_READV,
	IORING_OP_WRITEV,
	IORING_OP_FSYNC,
	IORING_OP_READ_FIXED,
	IORING_OP_WRITE_FIXED,
	IORING_OP_POLL_ADD,
	IORING_OP_POLL_REMOVE,
	IORING_OP_SYNC_FILE_RANGE,
	IORING_OP_SENDMSG,
	IORING_OP_RECVMSG,
	IORING_OP_TIMEOUT,
	IORING_OP_TIMEOUT_REMOVE,
	IORING_OP_ACCEPT,
	IORING_OP_ASYNC_CANCEL,
	IORING_OP_LINK_TIMEOUT,
	IORING_OP_CONNECT,
	IORING_OP_FALLOCATE,
	IORING_OP_OPENAT,
	IORING_OP_CLOSE,
	IORING_OP_FILES_UPDATE,
	IORING_OP_STATX,
	IORING_OP_READ,
	IORING_OP_WRITE,
	IORING_OP_FADVISE,
	IORING_OP_MADVISE,
	IORING_OP_SEND,
	IORING_OP_RECV,
	IORING_OP_OPENAT2,
	IORING_OP_EPOLL_CTL,
	IORING_OP_SPLICE,
	IORING_OP_PROVIDE_BUFFERS,
	IORING_OP_REMOVE_BUFFERS,
	IORING_OP_TEE,
	IORING_OP_SHUTDOWN,
	IORING_OP_RENAMEAT,
	IORING_OP_UNLINKAT,
	IORING_OP_MKDIRAT,
	IORING_OP_SYMLINKAT,
	IORING_OP_LINKAT,
	IORING_OP_MSG_RING,
	IORING_OP_FSETXATTR,
	IORING_OP_SETXATTR,
	IORING_OP_FGETXATTR,
	IORING_OP_GETXATTR,
	IORING_OP_SOCKET,
	IORING_OP_URING_CMD,


	/* this goes last, obviously */
	IORING_OP_LAST,
};

/*
 * sqe->uring_cmd_flags
 * IORING_URING_CMD_FIXED	use registered buffer; pass thig flag
 *				along with setting sqe->buf_index.
 */
#define IORING_URING_CMD_FIXED	(1U << 0)

/*
 * sqe->fsync_flags
 */
#define IORING_FSYNC_DATASYNC	(1U << 0)

/*
 * sqe->timeout_flags
 */
#define IORING_TIMEOUT_ABS		(1U << 0)
#define IORING_TIMEOUT_UPDATE		(1U << 1)
#define IORING_TIMEOUT_BOOTTIME		(1U << 2)
#define IORING_TIMEOUT_REALTIME		(1U << 3)
#define IORING_LINK_TIMEOUT_UPDATE	(1U << 4)
#define IORING_TIMEOUT_ETIME_SUCCESS	(1U << 5)
#define IORING_TIMEOUT_CLOCK_MASK	(IORING_TIMEOUT_BOOTTIME | IORING_TIMEOUT_REALTIME)
#define IORING_TIMEOUT_UPDATE_MASK	(IORING_TIMEOUT_UPDATE | IORING_LINK_TIMEOUT_UPDATE)
/*
 * sqe->splice_flags
 * extends splice(2) flags
 */
#define SPLICE_F_FD_IN_FIXED	(1U << 31) /* the last bit of __u32 */

/*
 * POLL_ADD flags. Note that since sqe->poll_events is the flag space, the
 * command flags for POLL_ADD are stored in sqe->len.
 *
 * IORING_POLL_ADD_MULTI	Multishot poll. Sets IORING_CQE_F_MORE if
 *				the poll handler will continue to report
 *				CQEs on behalf of the same SQE.
 *
 * IORING_POLL_UPDATE		Update existing poll request, matching
 *				sqe->addr as the old user_data field.
 */
#define IORING_POLL_ADD_MULTI	(1U << 0)
#define IORING_POLL_UPDATE_EVENTS	(1U << 1)
#define IORING_POLL_UPDATE_USER_DATA	(1U << 2)

/*
 * IO completion data structure (Completion Queue Entry)
 */
struct io_uring_cqe {
	__u64	user_data;	/* sqe->data submission passed back */
	__s32	res;		/* result code for this event */
	__u32	flags;

	/*
	 * If the ring is initialized with IORING_SETUP_CQE32, then this field
	 * contains 16-bytes of padding, doubling the size of the CQE.
	 */
	__u64 big_cqe[];
};

/*
 * cqe->flags
 *
 * IORING_CQE_F_BUFFER	If set, the upper 16 bits are the buffer ID
 * IORING_CQE_F_MORE	If set, parent SQE will generate more CQE entries
 */
#define IORING_CQE_F_BUFFER		(1U << 0)
#define IORING_CQE_F_MORE		(1U << 1)

enum {
	IORING_CQE_BUFFER_SHIFT		= 16,
};

/*
 * Magic offsets for the application to mmap the data it needs
 */
#define IORING_OFF_SQ_RING		0ULL
#define IORING_OFF_CQ_RING		0x8000000ULL
#define IORING_OFF_SQES			0x10000000ULL

/*
 * Filled with the offset for mmap(2)
 */
struct io_sqring_offsets {
	__u32 head;
	__u32 tail;
	__u32 ring_mask;
	__u32 ring_entries;
	__u32 flags;
	__u32 dropped;
	__u32 array;
	__u32 resv1;
	__u64 resv2;
};

/*
 * sq_ring->flags
 */
#define IORING_SQ_NEED_WAKEUP	(1U << 0) /* needs io_uring_enter wakeup */
#define IORING_SQ_CQ_OVERFLOW	(1U << 1) /* CQ ring is overflown */

struct io_cqring_offsets {
	__u32 head;
	__u32 tail;
	__u32 ring_mask;
	__u32 ring_entries;
	__u32 overflow;
	__u32 cqes;
	__u32 flags;
	__u32 resv1;
	__u64 resv2;
};

/*
 * cq_ring->flags
 */

/* disable eventfd notifications */
#define IORING_CQ_EVENTFD_DISABLED	(1U << 0)

/*
 * io_uring_enter(2) flags
 */
#define IORING_ENTER_GETEVENTS		(1U << 0)
#define IORING_ENTER_SQ_WAKEUP		(1U << 1)
#define IORING_ENTER_SQ_WAIT		(1U << 2)
#define IORING_ENTER_EXT_ARG		(1U << 3)
#define IORING_ENTER_REGISTERED_RING	(1U << 4)

/*
 * Passed in for io_uring_setup(2). Copied back with updated info on success
 */
struct io_uring_params {
	__u32 sq_entries;
	__u32 cq_entries;
	__u32 flags;
	__u32 sq_thread_cpu;
	__u32 sq_thread_idle;
	__u32 features;
	__u32 wq_fd;
	__u32 resv[3];
	struct io_sqring_offsets sq_off;
	struct io_cqring_offsets cq_off;
};

/*
 * io_uring_params->features flags
 */
#define IORING_FEAT_SINGLE_MMAP		(1U << 0)
#define IORING_FEAT_NODROP		(1U << 1)
#define IORING_FEAT_SUBMIT_STABLE	(1U << 2)
#define IORING_FEAT_RW_CUR_POS		(1U << 3)
#define IORING_FEAT_CUR_PERSONALITY	(1U << 4)
#define IORING_FEAT_FAST_POLL		(1U << 5)
#define IORING_FEAT_POLL_32BITS 	(1U << 6)
#define IORING_FEAT_SQPOLL_NONFIXED	(1U << 7)
#define IORING_FEAT_EXT_ARG		(1U << 8)
#define IORING_FEAT_NATIVE_WORKERS	(1U << 9)
#define IORING_FEAT_RSRC_TAGS		(1U << 10)
#define IORING_FEAT_CQE_SKIP		(1U << 11)

/*
 * io_uring_register(2) opcodes and arguments
 */
enum {
	IORING_REGISTER_BUFFERS			= 0,
	IORING_UNREGISTER_BUFFERS		= 1,
	IORING_REGISTER_FILES			= 2,
	IORING_UNREGISTER_FILES			= 3,
	IORING_REGISTER_EVENTFD			= 4,
	IORING_UNREGISTER_EVENTFD		= 5,
	IORING_REGISTER_FILES_UPDATE		= 6,
	IORING_REGISTER_EVENTFD_ASYNC		= 7,
	IORING_REGISTER_PROBE			= 8,
	IORING_REGISTER_PERSONALITY		= 9,
	IORING_UNREGISTER_PERSONALITY		= 10,
	IORING_REGISTER_RESTRICTIONS		= 11,
	IORING_REGISTER_ENABLE_RINGS		= 12,

	/* extended with tagging */
	IORING_REGISTER_FILES2			= 13,
	IORING_REGISTER_FILES_UPDATE2		= 14,
	IORING_REGISTER_BUFFERS2		= 15,
	IORING_REGISTER_BUFFERS_UPDATE		= 16,

	/* set/clear io-wq thread affinities */
	IORING_REGISTER_IOWQ_AFF		= 17,
	IORING_UNREGISTER_IOWQ_AFF		= 18,

	/* set/get max number of io-wq workers */
	IORING_REGISTER_IOWQ_MAX_WORKERS	= 19,

	/* register/unregister io_uring fd with the ring */
	IORING_REGISTER_RING_FDS		= 20,
	IORING_UNREGISTER_RING_FDS		= 21,

	/* this goes last */
	IORING_REGISTER_LAST
};

/* io-wq worker categories */
enum {
	IO_WQ_BOUND,
	IO_WQ_UNBOUND,
};

/* deprecated, see struct io_uring_rsrc_update */
struct io_uring_files_update {
	__u32 offset;
	__u32 resv;
	__aligned_u64 /* __s32 * */ fds;
};

struct io_uring_rsrc_register {
	__u32 nr;
	__u32 resv;
	__u64 resv2;
	__aligned_u64 data;
	__aligned_u64 tags;
};

struct io_uring_rsrc_update {
	__u32 offset;
	__u32 resv;
	__aligned_u64 data;
};

struct io_uring_rsrc_update2 {
	__u32 offset;
	__u32 resv;
	__aligned_u64 data;
	__aligned_u64 tags;
	__u32 nr;
	__u32 resv2;
};

/* Skip updating fd indexes set to this value in the fd table */
#define IORING_REGISTER_FILES_SKIP	(-2)

#define IO_URING_OP_SUPPORTED	(1U << 0)

struct io_uring_probe_op {
	__u8 op;
	__u8 resv;
	__u16 flags;	/* IO_URING_OP_* flags */
	__u32 resv2;
};

struct io_uring_probe {
	__u8 last_op;	/* last opcode supported */
	__u8 ops_len;	/* length of ops[] array below */
	__u16 resv;
	__u32 resv2[3];
	struct io_uring_probe_op ops[0];
};

struct io_uring_restriction {
	__u16 opcode;
	union {
		__u8 register_op; /* IORING_RESTRICTION_REGISTER_OP */
		__u8 sqe_op;      /* IORING_RESTRICTION_SQE_OP */
		__u8 sqe_flags;   /* IORING_RESTRICTION_SQE_FLAGS_* */
	};
	__u8 resv;
	__u32 resv2[3];
};

/*
 * io_uring_restriction->opcode values
 */
enum {
	/* Allow an io_uring_register(2) opcode */
	IORING_RESTRICTION_REGISTER_OP		= 0,

	/* Allow an sqe opcode */
	IORING_RESTRICTION_SQE_OP		= 1,

	/* Allow sqe flags */
	IORING_RESTRICTION_SQE_FLAGS_ALLOWED	= 2,

	/* Require sqe flags (these flags must be set on each submission) */
	IORING_RESTRICTION_SQE_FLAGS_REQUIRED	= 3,

	IORING_RESTRICTION_LAST
};

struct io_uring_getevents_arg {
	__u64	sigmask;
	__u32	sigmask_sz;
	__u32	pad;
	__u64	ts;
};

#endif
Commit	Line	Data
556d8415	1	/* SPDX-License-Identifier: (GPL-2.0 WITH Linux-syscall-note) OR MIT */
e2239016 JA	2	/*
	3	* Header file for the io_uring interface.
	4	*
	5	* Copyright (C) 2019 Jens Axboe
	6	* Copyright (C) 2019 Christoph Hellwig
	7	*/
	8	#ifndef LINUX_IO_URING_H
	9	#define LINUX_IO_URING_H
e31b8288	10
6e70fd30	11	#include <linux/fs.h>
e2239016	12	#include <linux/types.h>
6e70fd30	13
e31b8288	14	/*
f0403f94	15	* IO submission data structure (Submission Queue Entry)
e31b8288	16	*/
f0403f94	17	struct io_uring_sqe {
a7086591	18	__u8 opcode; /* type of operation for this sqe */
48e698fa	19	__u8 flags; /* IOSQE_ flags */
a7086591 JA	20	__u16 ioprio; /* ioprio for the request */
a7086591 JA	21	__s32 fd; /* file descriptor to do IO on */
b10b1e70 JA	22	union {
	23	__u64 off; /* offset into file */
	24	__u64 addr2;
855dc4d4	25	__u32 cmd_op;
b10b1e70	26	};
556d8415 JA	27	union {
	28	__u64 addr; /* pointer to buffer or iovecs */
	29	__u64 splice_off_in;
	30	};
a7086591	31	__u32 len; /* buffer size or number of iovecs */
e31b8288 JA	32	union {
e31b8288 JA	33	__kernel_rwf_t rw_flags;
48e698fa	34	__u32 fsync_flags;
556d8415 JA	35	__u16 poll_events; /* compatibility */
556d8415 JA	36	__u32 poll32_events; /* word-reversed for BE */
a819dfb6	37	__u32 sync_range_flags;
b10b1e70 JA	38	__u32 msg_flags;
	39	__u32 timeout_flags;
	40	__u32 accept_flags;
	41	__u32 cancel_flags;
	42	__u32 open_flags;
	43	__u32 statx_flags;
556d8415 JA	44	__u32 fadvise_advice;
556d8415 JA	45	__u32 splice_flags;
9f51d89c JA	46	__u32 rename_flags;
	47	__u32 unlink_flags;
	48	__u32 hardlink_flags;
021ce718	49	__u32 uring_cmd_flags;
e31b8288	50	};
48e698fa	51	__u64 user_data; /* data to be passed back at completion time */
9f51d89c	52	/* pack this to avoid bogus arm OABI complaints */
f3e769a4	53	union {
9f51d89c JA	54	/* index into fixed buffers, if used */
	55	__u16 buf_index;
	56	/* for grouped buffer selection */
	57	__u16 buf_group;
	58	} __attribute__((packed));
	59	/* personality to use, if used */
	60	__u16 personality;
	61	union {
	62	__s32 splice_fd_in;
	63	__u32 file_index;
f3e769a4	64	};
1bca8ad1 AG	65	union {
	66	struct {
	67	__u64 addr3;
	68	__u64 __pad2[1];
	69	};
	70	/*
	71	* If the ring is initialized with IORING_SETUP_SQE128, then
	72	* this field is used for 80 bytes of arbitrary command data
	73	*/
	74	__u8 cmd[0];
	75	};
e31b8288 JA	76	};
e31b8288 JA	77
556d8415 JA	78	enum {
	79	IOSQE_FIXED_FILE_BIT,
	80	IOSQE_IO_DRAIN_BIT,
	81	IOSQE_IO_LINK_BIT,
	82	IOSQE_IO_HARDLINK_BIT,
	83	IOSQE_ASYNC_BIT,
	84	IOSQE_BUFFER_SELECT_BIT,
788f19d4	85	IOSQE_CQE_SKIP_SUCCESS_BIT,
556d8415 JA	86	};
556d8415 JA	87
2ea53ca3 JA	88	/*
	89	* sqe->flags
	90	*/
556d8415 JA	91	/* use fixed fileset */
	92	#define IOSQE_FIXED_FILE (1U << IOSQE_FIXED_FILE_BIT)
	93	/* issue after inflight IO */
	94	#define IOSQE_IO_DRAIN (1U << IOSQE_IO_DRAIN_BIT)
	95	/* links next sqe */
	96	#define IOSQE_IO_LINK (1U << IOSQE_IO_LINK_BIT)
	97	/* like LINK, but stronger */
	98	#define IOSQE_IO_HARDLINK (1U << IOSQE_IO_HARDLINK_BIT)
	99	/* always go async */
	100	#define IOSQE_ASYNC (1U << IOSQE_ASYNC_BIT)
	101	/* select buffer from sqe->buf_group */
	102	#define IOSQE_BUFFER_SELECT (1U << IOSQE_BUFFER_SELECT_BIT)
788f19d4 JA	103	/* don't post CQE if request succeeded */
788f19d4 JA	104	#define IOSQE_CQE_SKIP_SUCCESS (1U << IOSQE_CQE_SKIP_SUCCESS_BIT)
2ea53ca3	105
e31b8288 JA	106	/*
	107	* io_uring_setup() flags
	108	*/
e9d4aa07 JA	109	#define IORING_SETUP_IOPOLL (1U << 0) /* io_context is polled */
	110	#define IORING_SETUP_SQPOLL (1U << 1) /* SQ poll thread */
	111	#define IORING_SETUP_SQ_AFF (1U << 2) /* sq_thread_cpu is valid */
b10b1e70	112	#define IORING_SETUP_CQSIZE (1U << 3) /* app defines CQ size */
556d8415 JA	113	#define IORING_SETUP_CLAMP (1U << 4) /* clamp SQ/CQ ring sizes */
556d8415 JA	114	#define IORING_SETUP_ATTACH_WQ (1U << 5) /* attach to existing wq */
9f51d89c	115	#define IORING_SETUP_R_DISABLED (1U << 6) /* start with ring disabled */
1bca8ad1 AG	116	#define IORING_SETUP_SUBMIT_ALL (1U << 7) /* continue submit on error */
	117	/*
	118	* Cooperative task running. When requests complete, they often require
	119	* forcing the submitter to transition to the kernel to complete. If this
	120	* flag is set, work will be done when the task transitions anyway, rather
	121	* than force an inter-processor interrupt reschedule. This avoids interrupting
	122	* a task running in userspace, and saves an IPI.
	123	*/
	124	#define IORING_SETUP_COOP_TASKRUN (1U << 8)
	125	/*
	126	* If COOP_TASKRUN is set, get notified if task work is available for
	127	* running and a kernel transition would be needed to run it. This sets
	128	* IORING_SQ_TASKRUN in the sq ring flags. Not valid with COOP_TASKRUN.
	129	*/
	130	#define IORING_SETUP_TASKRUN_FLAG (1U << 9)
	131
	132	#define IORING_SETUP_SQE128 (1U << 10) /* SQEs are 128 byte */
	133	#define IORING_SETUP_CQE32 (1U << 11) /* CQEs are 32 byte */
b10b1e70	134
e453f369 JA	135	/*
	136	* Only one task is allowed to submit requests
	137	*/
	138	#define IORING_SETUP_SINGLE_ISSUER (1U << 12)
	139
	140	/*
	141	* Defer running task work to get events.
	142	* Rather than running bits of task work whenever the task transitions
	143	* try to do it just before it is needed.
	144	*/
	145	#define IORING_SETUP_DEFER_TASKRUN (1U << 13)
	146
b10b1e70 JA	147	enum {
	148	IORING_OP_NOP,
	149	IORING_OP_READV,
	150	IORING_OP_WRITEV,
	151	IORING_OP_FSYNC,
	152	IORING_OP_READ_FIXED,
	153	IORING_OP_WRITE_FIXED,
	154	IORING_OP_POLL_ADD,
	155	IORING_OP_POLL_REMOVE,
	156	IORING_OP_SYNC_FILE_RANGE,
	157	IORING_OP_SENDMSG,
	158	IORING_OP_RECVMSG,
	159	IORING_OP_TIMEOUT,
	160	IORING_OP_TIMEOUT_REMOVE,
	161	IORING_OP_ACCEPT,
	162	IORING_OP_ASYNC_CANCEL,
	163	IORING_OP_LINK_TIMEOUT,
	164	IORING_OP_CONNECT,
	165	IORING_OP_FALLOCATE,
	166	IORING_OP_OPENAT,
	167	IORING_OP_CLOSE,
	168	IORING_OP_FILES_UPDATE,
	169	IORING_OP_STATX,
	170	IORING_OP_READ,
	171	IORING_OP_WRITE,
556d8415 JA	172	IORING_OP_FADVISE,
	173	IORING_OP_MADVISE,
	174	IORING_OP_SEND,
	175	IORING_OP_RECV,
	176	IORING_OP_OPENAT2,
	177	IORING_OP_EPOLL_CTL,
	178	IORING_OP_SPLICE,
	179	IORING_OP_PROVIDE_BUFFERS,
	180	IORING_OP_REMOVE_BUFFERS,
	181	IORING_OP_TEE,
9f51d89c JA	182	IORING_OP_SHUTDOWN,
	183	IORING_OP_RENAMEAT,
	184	IORING_OP_UNLINKAT,
	185	IORING_OP_MKDIRAT,
	186	IORING_OP_SYMLINKAT,
	187	IORING_OP_LINKAT,
855dc4d4 AG	188	IORING_OP_MSG_RING,
	189	IORING_OP_FSETXATTR,
	190	IORING_OP_SETXATTR,
	191	IORING_OP_FGETXATTR,
	192	IORING_OP_GETXATTR,
	193	IORING_OP_SOCKET,
	194	IORING_OP_URING_CMD,
	195
e31b8288	196
b10b1e70 JA	197	/* this goes last, obviously */
	198	IORING_OP_LAST,
	199	};
48e698fa	200
021ce718 JA	201	/*
	202	* sqe->uring_cmd_flags
	203	* IORING_URING_CMD_FIXED use registered buffer; pass thig flag
	204	* along with setting sqe->buf_index.
	205	*/
	206	#define IORING_URING_CMD_FIXED (1U << 0)
	207
48e698fa JA	208	/*
	209	* sqe->fsync_flags
	210	*/
e9d4aa07	211	#define IORING_FSYNC_DATASYNC (1U << 0)
e31b8288	212
b10b1e70 JA	213	/*
	214	* sqe->timeout_flags
	215	*/
9f51d89c JA	216	#define IORING_TIMEOUT_ABS (1U << 0)
	217	#define IORING_TIMEOUT_UPDATE (1U << 1)
	218	#define IORING_TIMEOUT_BOOTTIME (1U << 2)
	219	#define IORING_TIMEOUT_REALTIME (1U << 3)
	220	#define IORING_LINK_TIMEOUT_UPDATE (1U << 4)
	221	#define IORING_TIMEOUT_ETIME_SUCCESS (1U << 5)
	222	#define IORING_TIMEOUT_CLOCK_MASK (IORING_TIMEOUT_BOOTTIME \| IORING_TIMEOUT_REALTIME)
	223	#define IORING_TIMEOUT_UPDATE_MASK (IORING_TIMEOUT_UPDATE \| IORING_LINK_TIMEOUT_UPDATE)
556d8415 JA	224	/*
	225	* sqe->splice_flags
	226	* extends splice(2) flags
	227	*/
	228	#define SPLICE_F_FD_IN_FIXED (1U << 31) /* the last bit of __u32 */
	229
9f51d89c JA	230	/*
	231	* POLL_ADD flags. Note that since sqe->poll_events is the flag space, the
	232	* command flags for POLL_ADD are stored in sqe->len.
	233	*
	234	* IORING_POLL_ADD_MULTI Multishot poll. Sets IORING_CQE_F_MORE if
	235	* the poll handler will continue to report
	236	* CQEs on behalf of the same SQE.
	237	*
	238	* IORING_POLL_UPDATE Update existing poll request, matching
	239	* sqe->addr as the old user_data field.
	240	*/
	241	#define IORING_POLL_ADD_MULTI (1U << 0)
	242	#define IORING_POLL_UPDATE_EVENTS (1U << 1)
	243	#define IORING_POLL_UPDATE_USER_DATA (1U << 2)
	244
e31b8288	245	/*
f0403f94	246	* IO completion data structure (Completion Queue Entry)
e31b8288	247	*/
f0403f94	248	struct io_uring_cqe {
48e698fa	249	__u64 user_data; /* sqe->data submission passed back */
e2239016 JA	250	__s32 res; /* result code for this event */
e2239016 JA	251	__u32 flags;
1bca8ad1 AG	252
	253	/*
	254	* If the ring is initialized with IORING_SETUP_CQE32, then this field
	255	* contains 16-bytes of padding, doubling the size of the CQE.
	256	*/
	257	__u64 big_cqe[];
e31b8288 JA	258	};
e31b8288 JA	259
556d8415 JA	260	/*
	261	* cqe->flags
	262	*
	263	* IORING_CQE_F_BUFFER If set, the upper 16 bits are the buffer ID
9f51d89c	264	* IORING_CQE_F_MORE If set, parent SQE will generate more CQE entries
556d8415 JA	265	*/
556d8415 JA	266	#define IORING_CQE_F_BUFFER (1U << 0)
9f51d89c	267	#define IORING_CQE_F_MORE (1U << 1)
556d8415 JA	268
	269	enum {
	270	IORING_CQE_BUFFER_SHIFT = 16,
	271	};
	272
e31b8288 JA	273	/*
	274	* Magic offsets for the application to mmap the data it needs
	275	*/
	276	#define IORING_OFF_SQ_RING 0ULL
	277	#define IORING_OFF_CQ_RING 0x8000000ULL
f0403f94	278	#define IORING_OFF_SQES 0x10000000ULL
e31b8288 JA	279
	280	/*
	281	* Filled with the offset for mmap(2)
	282	*/
	283	struct io_sqring_offsets {
e2239016 JA	284	__u32 head;
	285	__u32 tail;
	286	__u32 ring_mask;
	287	__u32 ring_entries;
	288	__u32 flags;
	289	__u32 dropped;
	290	__u32 array;
66169493 JA	291	__u32 resv1;
66169493 JA	292	__u64 resv2;
e31b8288 JA	293	};
e31b8288 JA	294
48e698fa JA	295	/*
	296	* sq_ring->flags
	297	*/
e9d4aa07	298	#define IORING_SQ_NEED_WAKEUP (1U << 0) /* needs io_uring_enter wakeup */
556d8415	299	#define IORING_SQ_CQ_OVERFLOW (1U << 1) /* CQ ring is overflown */
e31b8288 JA	300
e31b8288 JA	301	struct io_cqring_offsets {
e2239016 JA	302	__u32 head;
	303	__u32 tail;
	304	__u32 ring_mask;
	305	__u32 ring_entries;
	306	__u32 overflow;
f0403f94	307	__u32 cqes;
556d8415 JA	308	__u32 flags;
	309	__u32 resv1;
	310	__u64 resv2;
e31b8288 JA	311	};
e31b8288 JA	312
556d8415 JA	313	/*
	314	* cq_ring->flags
	315	*/
	316
	317	/* disable eventfd notifications */
	318	#define IORING_CQ_EVENTFD_DISABLED (1U << 0)
	319
e2239016 JA	320	/*
	321	* io_uring_enter(2) flags
	322	*/
788f19d4 JA	323	#define IORING_ENTER_GETEVENTS (1U << 0)
	324	#define IORING_ENTER_SQ_WAKEUP (1U << 1)
	325	#define IORING_ENTER_SQ_WAIT (1U << 2)
	326	#define IORING_ENTER_EXT_ARG (1U << 3)
	327	#define IORING_ENTER_REGISTERED_RING (1U << 4)
e31b8288 JA	328
	329	/*
	330	* Passed in for io_uring_setup(2). Copied back with updated info on success
	331	*/
	332	struct io_uring_params {
e2239016 JA	333	__u32 sq_entries;
	334	__u32 cq_entries;
	335	__u32 flags;
e9d4aa07 JA	336	__u32 sq_thread_cpu;
e9d4aa07 JA	337	__u32 sq_thread_idle;
b10b1e70	338	__u32 features;
556d8415 JA	339	__u32 wq_fd;
556d8415 JA	340	__u32 resv[3];
e31b8288 JA	341	struct io_sqring_offsets sq_off;
	342	struct io_cqring_offsets cq_off;
	343	};
	344
b10b1e70 JA	345	/*
	346	* io_uring_params->features flags
	347	*/
	348	#define IORING_FEAT_SINGLE_MMAP (1U << 0)
	349	#define IORING_FEAT_NODROP (1U << 1)
	350	#define IORING_FEAT_SUBMIT_STABLE (1U << 2)
556d8415 JA	351	#define IORING_FEAT_RW_CUR_POS (1U << 3)
	352	#define IORING_FEAT_CUR_PERSONALITY (1U << 4)
	353	#define IORING_FEAT_FAST_POLL (1U << 5)
	354	#define IORING_FEAT_POLL_32BITS (1U << 6)
9f51d89c JA	355	#define IORING_FEAT_SQPOLL_NONFIXED (1U << 7)
	356	#define IORING_FEAT_EXT_ARG (1U << 8)
	357	#define IORING_FEAT_NATIVE_WORKERS (1U << 9)
	358	#define IORING_FEAT_RSRC_TAGS (1U << 10)
788f19d4	359	#define IORING_FEAT_CQE_SKIP (1U << 11)
b10b1e70	360
2ea53ca3 JA	361	/*
	362	* io_uring_register(2) opcodes and arguments
	363	*/
9f51d89c JA	364	enum {
	365	IORING_REGISTER_BUFFERS = 0,
	366	IORING_UNREGISTER_BUFFERS = 1,
	367	IORING_REGISTER_FILES = 2,
	368	IORING_UNREGISTER_FILES = 3,
	369	IORING_REGISTER_EVENTFD = 4,
	370	IORING_UNREGISTER_EVENTFD = 5,
	371	IORING_REGISTER_FILES_UPDATE = 6,
	372	IORING_REGISTER_EVENTFD_ASYNC = 7,
	373	IORING_REGISTER_PROBE = 8,
	374	IORING_REGISTER_PERSONALITY = 9,
	375	IORING_UNREGISTER_PERSONALITY = 10,
	376	IORING_REGISTER_RESTRICTIONS = 11,
	377	IORING_REGISTER_ENABLE_RINGS = 12,
	378
	379	/* extended with tagging */
	380	IORING_REGISTER_FILES2 = 13,
	381	IORING_REGISTER_FILES_UPDATE2 = 14,
	382	IORING_REGISTER_BUFFERS2 = 15,
	383	IORING_REGISTER_BUFFERS_UPDATE = 16,
	384
	385	/* set/clear io-wq thread affinities */
	386	IORING_REGISTER_IOWQ_AFF = 17,
	387	IORING_UNREGISTER_IOWQ_AFF = 18,
	388
	389	/* set/get max number of io-wq workers */
	390	IORING_REGISTER_IOWQ_MAX_WORKERS = 19,
b10b1e70	391
788f19d4 JA	392	/* register/unregister io_uring fd with the ring */
	393	IORING_REGISTER_RING_FDS = 20,
	394	IORING_UNREGISTER_RING_FDS = 21,
	395
9f51d89c JA	396	/* this goes last */
	397	IORING_REGISTER_LAST
	398	};
	399
	400	/* io-wq worker categories */
	401	enum {
	402	IO_WQ_BOUND,
	403	IO_WQ_UNBOUND,
	404	};
	405
	406	/* deprecated, see struct io_uring_rsrc_update */
b10b1e70 JA	407	struct io_uring_files_update {
b10b1e70 JA	408	__u32 offset;
556d8415 JA	409	__u32 resv;
556d8415 JA	410	__aligned_u64 /* __s32 * */ fds;
b10b1e70	411	};
2ea53ca3	412
9f51d89c JA	413	struct io_uring_rsrc_register {
	414	__u32 nr;
	415	__u32 resv;
	416	__u64 resv2;
	417	__aligned_u64 data;
	418	__aligned_u64 tags;
	419	};
	420
	421	struct io_uring_rsrc_update {
	422	__u32 offset;
	423	__u32 resv;
	424	__aligned_u64 data;
	425	};
	426
	427	struct io_uring_rsrc_update2 {
	428	__u32 offset;
	429	__u32 resv;
	430	__aligned_u64 data;
	431	__aligned_u64 tags;
	432	__u32 nr;
	433	__u32 resv2;
	434	};
	435
	436	/* Skip updating fd indexes set to this value in the fd table */
	437	#define IORING_REGISTER_FILES_SKIP (-2)
	438
556d8415 JA	439	#define IO_URING_OP_SUPPORTED (1U << 0)
	440
	441	struct io_uring_probe_op {
	442	__u8 op;
	443	__u8 resv;
	444	__u16 flags; /* IO_URING_OP_* flags */
	445	__u32 resv2;
	446	};
	447
	448	struct io_uring_probe {
	449	__u8 last_op; /* last opcode supported */
	450	__u8 ops_len; /* length of ops[] array below */
	451	__u16 resv;
	452	__u32 resv2[3];
	453	struct io_uring_probe_op ops[0];
	454	};
	455
9f51d89c JA	456	struct io_uring_restriction {
	457	__u16 opcode;
	458	union {
	459	__u8 register_op; /* IORING_RESTRICTION_REGISTER_OP */
	460	__u8 sqe_op; /* IORING_RESTRICTION_SQE_OP */
	461	__u8 sqe_flags; /* IORING_RESTRICTION_SQE_FLAGS_* */
	462	};
	463	__u8 resv;
	464	__u32 resv2[3];
	465	};
	466
	467	/*
	468	* io_uring_restriction->opcode values
	469	*/
	470	enum {
	471	/* Allow an io_uring_register(2) opcode */
	472	IORING_RESTRICTION_REGISTER_OP = 0,
	473
	474	/* Allow an sqe opcode */
	475	IORING_RESTRICTION_SQE_OP = 1,
	476
	477	/* Allow sqe flags */
	478	IORING_RESTRICTION_SQE_FLAGS_ALLOWED = 2,
	479
	480	/* Require sqe flags (these flags must be set on each submission) */
	481	IORING_RESTRICTION_SQE_FLAGS_REQUIRED = 3,
	482
	483	IORING_RESTRICTION_LAST
	484	};
	485
	486	struct io_uring_getevents_arg {
	487	__u64 sigmask;
	488	__u32 sigmask_sz;
	489	__u32 pad;
	490	__u64 ts;
	491	};
556d8415	492
e31b8288	493	#endif