io_uring: defer file assignment

If an application uses direct open or accept, it knows in advance what
direct descriptor value it will get as it picks it itself. This allows
combined requests such as:

sqe = io_uring_get_sqe(ring);
io_uring_prep_openat_direct(sqe, ..., file_slot);
sqe->flags |= IOSQE_IO_LINK | IOSQE_CQE_SKIP_SUCCESS;

sqe = io_uring_get_sqe(ring);
io_uring_prep_read(sqe,file_slot, buf, buf_size, 0);
sqe->flags |= IOSQE_FIXED_FILE;

io_uring_submit(ring);

where we prepare both a file open and read, and only get a completion
event for the read when both have completed successfully.

Currently links are fully prepared before the head is issued, but that
fails if the dependent link needs a file assigned that isn't valid until
the head has completed.

Conversely, if the same chain is performed but the fixed file slot is
already valid, then we would be unexpectedly returning data from the
old file slot rather than the newly opened one. Make sure we're
consistent here.

Allow deferral of file setup, which makes this documented case work.

Cc: [email protected] # v5.15+
Signed-off-by: Jens Axboe <[email protected]>

Author: axboe

fs/io-wq.h

@@ -155,6 +155,7 @@ struct io_wq_work_node *wq_stack_extract(struct io_wq_work_node *stack)
 struct io_wq_work {
 	struct io_wq_work_node list;
 	unsigned flags;
+	int fd;
 };
 
 static inline struct io_wq_work *wq_next_work(struct io_wq_work *work)

fs/io_uring.c

@@ -7240,6 +7240,23 @@ static void io_clean_op(struct io_kiocb *req)
 	req->flags &= ~IO_REQ_CLEAN_FLAGS;
 }
 
+static bool io_assign_file(struct io_kiocb *req, unsigned int issue_flags)
+{
+	if (req->file || !io_op_defs[req->opcode].needs_file)
+		return true;
+
+	if (req->flags & REQ_F_FIXED_FILE)
+		req->file = io_file_get_fixed(req, req->work.fd, issue_flags);
+	else
+		req->file = io_file_get_normal(req, req->work.fd);
+	if (req->file)
+		return true;
+
+	req_set_fail(req);
+	req->result = -EBADF;
+	return false;
+}
+
 static int io_issue_sqe(struct io_kiocb *req, unsigned int issue_flags)
 {
 	const struct cred *creds = NULL;
@@ -7250,6 +7267,8 @@ static int io_issue_sqe(struct io_kiocb *req, unsigned int issue_flags)
 
 	if (!io_op_defs[req->opcode].audit_skip)
 		audit_uring_entry(req->opcode);
+	if (unlikely(!io_assign_file(req, issue_flags)))
+		return -EBADF;
 
 	switch (req->opcode) {
 	case IORING_OP_NOP:
@@ -7394,10 +7413,11 @@ static struct io_wq_work *io_wq_free_work(struct io_wq_work *work)
 static void io_wq_submit_work(struct io_wq_work *work)
 {
 	struct io_kiocb *req = container_of(work, struct io_kiocb, work);
+	const struct io_op_def *def = &io_op_defs[req->opcode];
 	unsigned int issue_flags = IO_URING_F_UNLOCKED;
 	bool needs_poll = false;
 	struct io_kiocb *timeout;
-	int ret = 0;
+	int ret = 0, err = -ECANCELED;
 
 	/* one will be dropped by ->io_free_work() after returning to io-wq */
 	if (!(req->flags & REQ_F_REFCOUNT))
@@ -7409,14 +7429,18 @@ static void io_wq_submit_work(struct io_wq_work *work)
 	if (timeout)
 		io_queue_linked_timeout(timeout);
 
+	if (!io_assign_file(req, issue_flags)) {
+		err = -EBADF;
+		work->flags |= IO_WQ_WORK_CANCEL;
+	}
+
 	/* either cancelled or io-wq is dying, so don't touch tctx->iowq */
 	if (work->flags & IO_WQ_WORK_CANCEL) {
-		io_req_task_queue_fail(req, -ECANCELED);
+		io_req_task_queue_fail(req, err);
 		return;
 	}
 
 	if (req->flags & REQ_F_FORCE_ASYNC) {
-		const struct io_op_def *def = &io_op_defs[req->opcode];
 		bool opcode_poll = def->pollin || def->pollout;
 
 		if (opcode_poll && file_can_poll(req->file)) {
@@ -7749,6 +7773,8 @@ static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req,
 	if (io_op_defs[opcode].needs_file) {
 		struct io_submit_state *state = &ctx->submit_state;
 
+		req->work.fd = READ_ONCE(sqe->fd);
+
 		/*
 		 * Plug now if we have more than 2 IO left after this, and the
 		 * target is potentially a read/write to block based storage.
@@ -7758,13 +7784,6 @@ static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req,
 			state->need_plug = false;
 			blk_start_plug_nr_ios(&state->plug, state->submit_nr);
 		}
-
-		if (req->flags & REQ_F_FIXED_FILE)
-			req->file = io_file_get_fixed(req, READ_ONCE(sqe->fd), 0);
-		else
-			req->file = io_file_get_normal(req, READ_ONCE(sqe->fd));
-		if (unlikely(!req->file))
-			return -EBADF;
 	}
 
 	personality = READ_ONCE(sqe->personality);