fs,io_uring: add infrastructure for uring-cmd

file_operations->uring_cmd is a file private handler.
This is somewhat similar to ioctl but hopefully a lot more sane and
useful as it can be used to enable many io_uring capabilities for the
underlying operation.

IORING_OP_URING_CMD is a file private kind of request. io_uring doesn't
know what is in this command type, it's for the provider of ->uring_cmd()
to deal with.
Co-developed-by: Kanchan Joshi <joshi.k@samsung.com>
Signed-off-by: Kanchan Joshi <joshi.k@samsung.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20220511054750.20432-2-joshi.k@samsung.comSigned-off-by: Jens Axboe <axboe@kernel.dk>

fs/io_uring.c

@@ -202,13 +202,6 @@ struct io_rings {
 	struct io_uring_cqe	cqes[] ____cacheline_aligned_in_smp;
 };
 
-enum io_uring_cmd_flags {
-	IO_URING_F_COMPLETE_DEFER	= 1,
-	IO_URING_F_UNLOCKED		= 2,
-	/* int's last bit, sign checks are usually faster than a bit test */
-	IO_URING_F_NONBLOCK		= INT_MIN,
-};
-
 struct io_mapped_ubuf {
 	u64		ubuf;
 	u64		ubuf_end;
@@ -972,6 +965,7 @@ struct io_kiocb {
 		struct io_xattr		xattr;
 		struct io_socket	sock;
 		struct io_nop		nop;
+		struct io_uring_cmd	uring_cmd;
 	};
 
 	u8				opcode;
@@ -1050,6 +1044,14 @@ struct io_cancel_data {
 	int seq;
 };
 
+/*
+ * The URING_CMD payload starts at 'cmd' in the first sqe, and continues into
+ * the following sqe if SQE128 is used.
+ */
+#define uring_cmd_pdu_size(is_sqe128)				\
+	((1 + !!(is_sqe128)) * sizeof(struct io_uring_sqe) -	\
+		offsetof(struct io_uring_sqe, cmd))
+
 struct io_op_def {
 	/* needs req->file assigned */
 	unsigned		needs_file : 1;
@@ -1289,6 +1291,12 @@ static const struct io_op_def io_op_defs[] = {
 	[IORING_OP_SOCKET] = {
 		.audit_skip		= 1,
 	},
+	[IORING_OP_URING_CMD] = {
+		.needs_file		= 1,
+		.plug			= 1,
+		.needs_async_setup	= 1,
+		.async_size		= uring_cmd_pdu_size(1),
+	},
 };
 
 /* requests with any of those set should undergo io_disarm_next() */
@@ -1428,6 +1436,8 @@ const char *io_uring_get_opcode(u8 opcode)
 		return "GETXATTR";
 	case IORING_OP_SOCKET:
 		return "SOCKET";
+	case IORING_OP_URING_CMD:
+		return "URING_CMD";
 	case IORING_OP_LAST:
 		return "INVALID";
 	}
@@ -4507,10 +4517,6 @@ static int __io_getxattr_prep(struct io_kiocb *req,
 	const char __user *name;
 	int ret;
 
-	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
-		return -EINVAL;
-	if (unlikely(sqe->ioprio))
-		return -EINVAL;
 	if (unlikely(req->flags & REQ_F_FIXED_FILE))
 		return -EBADF;
 
@@ -4620,10 +4626,6 @@ static int __io_setxattr_prep(struct io_kiocb *req,
 	const char __user *name;
 	int ret;
 
-	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
-		return -EINVAL;
-	if (unlikely(sqe->ioprio))
-		return -EINVAL;
 	if (unlikely(req->flags & REQ_F_FIXED_FILE))
 		return -EBADF;
 
@@ -4910,6 +4912,96 @@ static int io_linkat(struct io_kiocb *req, unsigned int issue_flags)
 	return 0;
 }
 
+static void io_uring_cmd_work(struct io_kiocb *req, bool *locked)
+{
+	req->uring_cmd.task_work_cb(&req->uring_cmd);
+}
+
+void io_uring_cmd_complete_in_task(struct io_uring_cmd *ioucmd,
+			void (*task_work_cb)(struct io_uring_cmd *))
+{
+	struct io_kiocb *req = container_of(ioucmd, struct io_kiocb, uring_cmd);
+
+	req->uring_cmd.task_work_cb = task_work_cb;
+	req->io_task_work.func = io_uring_cmd_work;
+	io_req_task_work_add(req, !!(req->ctx->flags & IORING_SETUP_SQPOLL));
+}
+EXPORT_SYMBOL_GPL(io_uring_cmd_complete_in_task);
+
+/*
+ * Called by consumers of io_uring_cmd, if they originally returned
+ * -EIOCBQUEUED upon receiving the command.
+ */
+void io_uring_cmd_done(struct io_uring_cmd *ioucmd, ssize_t ret, ssize_t res2)
+{
+	struct io_kiocb *req = container_of(ioucmd, struct io_kiocb, uring_cmd);
+
+	if (ret < 0)
+		req_set_fail(req);
+	if (req->ctx->flags & IORING_SETUP_CQE32)
+		__io_req_complete32(req, 0, ret, 0, res2, 0);
+	else
+		io_req_complete(req, ret);
+}
+EXPORT_SYMBOL_GPL(io_uring_cmd_done);
+
+static int io_uring_cmd_prep_async(struct io_kiocb *req)
+{
+	size_t cmd_size;
+
+	cmd_size = uring_cmd_pdu_size(req->ctx->flags & IORING_SETUP_SQE128);
+
+	memcpy(req->async_data, req->uring_cmd.cmd, cmd_size);
+	return 0;
+}
+
+static int io_uring_cmd_prep(struct io_kiocb *req,
+			     const struct io_uring_sqe *sqe)
+{
+	struct io_uring_cmd *ioucmd = &req->uring_cmd;
+
+	if (sqe->rw_flags)
+		return -EINVAL;
+	ioucmd->cmd = sqe->cmd;
+	ioucmd->cmd_op = READ_ONCE(sqe->cmd_op);
+	return 0;
+}
+
+static int io_uring_cmd(struct io_kiocb *req, unsigned int issue_flags)
+{
+	struct io_uring_cmd *ioucmd = &req->uring_cmd;
+	struct io_ring_ctx *ctx = req->ctx;
+	struct file *file = req->file;
+	int ret;
+
+	if (!req->file->f_op->uring_cmd)
+		return -EOPNOTSUPP;
+
+	if (ctx->flags & IORING_SETUP_SQE128)
+		issue_flags |= IO_URING_F_SQE128;
+	if (ctx->flags & IORING_SETUP_CQE32)
+		issue_flags |= IO_URING_F_CQE32;
+	if (ctx->flags & IORING_SETUP_IOPOLL)
+		issue_flags |= IO_URING_F_IOPOLL;
+
+	if (req_has_async_data(req))
+		ioucmd->cmd = req->async_data;
+
+	ret = file->f_op->uring_cmd(ioucmd, issue_flags);
+	if (ret == -EAGAIN) {
+		if (!req_has_async_data(req)) {
+			if (io_alloc_async_data(req))
+				return -ENOMEM;
+			io_uring_cmd_prep_async(req);
+		}
+		return -EAGAIN;
+	}
+
+	if (ret != -EIOCBQUEUED)
+		io_uring_cmd_done(ioucmd, ret, 0);
+	return 0;
+}
+
 static int io_shutdown_prep(struct io_kiocb *req,
 			    const struct io_uring_sqe *sqe)
 {
@@ -6305,9 +6397,7 @@ static int io_socket_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 {
 	struct io_socket *sock = &req->sock;
 
-	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
-		return -EINVAL;
-	if (sqe->ioprio || sqe->addr || sqe->rw_flags || sqe->buf_index)
+	if (sqe->addr || sqe->rw_flags || sqe->buf_index)
 		return -EINVAL;
 
 	sock->domain = READ_ONCE(sqe->fd);
@@ -7755,6 +7845,8 @@ static int io_req_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 		return io_getxattr_prep(req, sqe);
 	case IORING_OP_SOCKET:
 		return io_socket_prep(req, sqe);
+	case IORING_OP_URING_CMD:
+		return io_uring_cmd_prep(req, sqe);
 	}
 
 	printk_once(KERN_WARNING "io_uring: unhandled opcode %d\n",
@@ -7787,6 +7879,8 @@ static int io_req_prep_async(struct io_kiocb *req)
 		return io_recvmsg_prep_async(req);
 	case IORING_OP_CONNECT:
 		return io_connect_prep_async(req);
+	case IORING_OP_URING_CMD:
+		return io_uring_cmd_prep_async(req);
 	}
 	printk_once(KERN_WARNING "io_uring: prep_async() bad opcode %d\n",
 		    req->opcode);
@@ -8081,6 +8175,9 @@ static int io_issue_sqe(struct io_kiocb *req, unsigned int issue_flags)
 	case IORING_OP_SOCKET:
 		ret = io_socket(req, issue_flags);
 		break;
+	case IORING_OP_URING_CMD:
+		ret = io_uring_cmd(req, issue_flags);
+		break;
 	default:
 		ret = -EINVAL;
 		break;
@@ -12699,6 +12796,8 @@ static int __init io_uring_init(void)
 
 	BUILD_BUG_ON(sizeof(atomic_t) != sizeof(u32));
 
+	BUILD_BUG_ON(sizeof(struct io_uring_cmd) > 64);
+
 	req_cachep = KMEM_CACHE(io_kiocb, SLAB_HWCACHE_ALIGN | SLAB_PANIC |
 				SLAB_ACCOUNT);
 	return 0;

include/linux/fs.h

@@ -1953,6 +1953,7 @@ struct dir_context {
 #define REMAP_FILE_ADVISORY		(REMAP_FILE_CAN_SHORTEN)
 
 struct iov_iter;
+struct io_uring_cmd;
 
 struct file_operations {
 	struct module *owner;
@@ -1995,6 +1996,7 @@ struct file_operations {
 				   struct file *file_out, loff_t pos_out,
 				   loff_t len, unsigned int remap_flags);
 	int (*fadvise)(struct file *, loff_t, loff_t, int);
+	int (*uring_cmd)(struct io_uring_cmd *ioucmd, unsigned int issue_flags);
 } __randomize_layout;
 
 struct inode_operations {

include/linux/io_uring.h

@@ -5,7 +5,32 @@
 #include <linux/sched.h>
 #include <linux/xarray.h>
 
+enum io_uring_cmd_flags {
+	IO_URING_F_COMPLETE_DEFER	= 1,
+	IO_URING_F_UNLOCKED		= 2,
+	/* int's last bit, sign checks are usually faster than a bit test */
+	IO_URING_F_NONBLOCK		= INT_MIN,
+
+	/* ctx state flags, for URING_CMD */
+	IO_URING_F_SQE128		= 4,
+	IO_URING_F_CQE32		= 8,
+	IO_URING_F_IOPOLL		= 16,
+};
+
+struct io_uring_cmd {
+	struct file	*file;
+	const void	*cmd;
+	/* callback to defer completions to task context */
+	void (*task_work_cb)(struct io_uring_cmd *cmd);
+	u32		cmd_op;
+	u32		pad;
+	u8		pdu[32]; /* available inline for free use */
+};
+
 #if defined(CONFIG_IO_URING)
+void io_uring_cmd_done(struct io_uring_cmd *cmd, ssize_t ret, ssize_t res2);
+void io_uring_cmd_complete_in_task(struct io_uring_cmd *ioucmd,
+			void (*task_work_cb)(struct io_uring_cmd *));
 struct sock *io_uring_get_socket(struct file *file);
 void __io_uring_cancel(bool cancel_all);
 void __io_uring_free(struct task_struct *tsk);
@@ -30,6 +55,14 @@ static inline void io_uring_free(struct task_struct *tsk)
 		__io_uring_free(tsk);
 }
 #else
+static inline void io_uring_cmd_done(struct io_uring_cmd *cmd, ssize_t ret,
+		ssize_t ret2)
+{
+}
+static inline void io_uring_cmd_complete_in_task(struct io_uring_cmd *ioucmd,
+			void (*task_work_cb)(struct io_uring_cmd *))
+{
+}
 static inline struct sock *io_uring_get_socket(struct file *file)
 {
 	return NULL;

include/uapi/linux/io_uring.h

@@ -22,6 +22,7 @@ struct io_uring_sqe {
 	union {
 		__u64	off;	/* offset into file */
 		__u64	addr2;
+		__u32	cmd_op;
 	};
 	union {
 		__u64	addr;	/* pointer to buffer or iovecs */
@@ -61,14 +62,17 @@ struct io_uring_sqe {
 		__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
-	 * contains 64-bytes of padding, doubling the size of the SQE.
+		 * If the ring is initialized with IORING_SETUP_SQE128, then
+		 * this field is used for 80 bytes of arbitrary command data
 		 */
-	__u64	__big_sqe_pad[0];
+		__u8	cmd[0];
+	};
 };
 
 enum {
@@ -175,6 +179,7 @@ enum io_uring_op {
 	IORING_OP_FGETXATTR,
 	IORING_OP_GETXATTR,
 	IORING_OP_SOCKET,
+	IORING_OP_URING_CMD,
 
 	/* this goes last, obviously */
 	IORING_OP_LAST,