io_uring: add POLL_FIRST support for send/sendmsg and recv/recvmsg

If IORING_RECVSEND_POLL_FIRST is set for recv/recvmsg or send/sendmsg,
then we arm poll first rather than attempt a receive or send upfront.
This can be useful if we expect there to be no data (or space) available
for the request, as we can then avoid wasting time on the initial
issue attempt.
Reviewed-by: Hao Xu <howeyxu@tencent.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

fs/io_uring.c

@@ -637,6 +637,7 @@ struct io_sr_msg {
 	int				bgid;
 	size_t				len;
 	size_t				done_io;
+	unsigned int			flags;
 };
 
 struct io_open {
@@ -5272,11 +5273,14 @@ static int io_sendmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 {
 	struct io_sr_msg *sr = &req->sr_msg;
 
-	if (unlikely(sqe->addr2 || sqe->file_index))
+	if (unlikely(sqe->file_index))
 		return -EINVAL;
 
 	sr->umsg = u64_to_user_ptr(READ_ONCE(sqe->addr));
 	sr->len = READ_ONCE(sqe->len);
+	sr->flags = READ_ONCE(sqe->addr2);
+	if (sr->flags & ~IORING_RECVSEND_POLL_FIRST)
+		return -EINVAL;
 	sr->msg_flags = READ_ONCE(sqe->msg_flags) | MSG_NOSIGNAL;
 	if (sr->msg_flags & MSG_DONTWAIT)
 		req->flags |= REQ_F_NOWAIT;
@@ -5311,6 +5315,10 @@ static int io_sendmsg(struct io_kiocb *req, unsigned int issue_flags)
 		kmsg = &iomsg;
 	}
 
+	if (!(req->flags & REQ_F_POLLED) &&
+	    (sr->flags & IORING_RECVSEND_POLL_FIRST))
+		return io_setup_async_msg(req, kmsg);
+
 	flags = req->sr_msg.msg_flags;
 	if (issue_flags & IO_URING_F_NONBLOCK)
 		flags |= MSG_DONTWAIT;
@@ -5353,6 +5361,10 @@ static int io_send(struct io_kiocb *req, unsigned int issue_flags)
 	int min_ret = 0;
 	int ret;
 
+	if (!(req->flags & REQ_F_POLLED) &&
+	    (sr->flags & IORING_RECVSEND_POLL_FIRST))
+		return -EAGAIN;
+
 	sock = sock_from_file(req->file);
 	if (unlikely(!sock))
 		return -ENOTSOCK;
@@ -5505,11 +5517,14 @@ static int io_recvmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 {
 	struct io_sr_msg *sr = &req->sr_msg;
 
-	if (unlikely(sqe->addr2 || sqe->file_index))
+	if (unlikely(sqe->file_index))
 		return -EINVAL;
 
 	sr->umsg = u64_to_user_ptr(READ_ONCE(sqe->addr));
 	sr->len = READ_ONCE(sqe->len);
+	sr->flags = READ_ONCE(sqe->addr2);
+	if (sr->flags & ~IORING_RECVSEND_POLL_FIRST)
+		return -EINVAL;
 	sr->bgid = READ_ONCE(sqe->buf_group);
 	sr->msg_flags = READ_ONCE(sqe->msg_flags) | MSG_NOSIGNAL;
 	if (sr->msg_flags & MSG_DONTWAIT)
@@ -5546,6 +5561,10 @@ static int io_recvmsg(struct io_kiocb *req, unsigned int issue_flags)
 		kmsg = &iomsg;
 	}
 
+	if (!(req->flags & REQ_F_POLLED) &&
+	    (sr->flags & IORING_RECVSEND_POLL_FIRST))
+		return io_setup_async_msg(req, kmsg);
+
 	if (req->flags & REQ_F_BUFFER_SELECT) {
 		kbuf = io_recv_buffer_select(req, issue_flags);
 		if (IS_ERR(kbuf))
@@ -5603,6 +5622,10 @@ static int io_recv(struct io_kiocb *req, unsigned int issue_flags)
 	int ret, min_ret = 0;
 	bool force_nonblock = issue_flags & IO_URING_F_NONBLOCK;
 
+	if (!(req->flags & REQ_F_POLLED) &&
+	    (sr->flags & IORING_RECVSEND_POLL_FIRST))
+		return -EAGAIN;
+
 	sock = sock_from_file(req->file);
 	if (unlikely(!sock))
 		return -ENOTSOCK;

include/uapi/linux/io_uring.h

@@ -213,6 +213,16 @@ enum {
 #define IORING_ASYNC_CANCEL_FD	(1U << 1)
 #define IORING_ASYNC_CANCEL_ANY	(1U << 2)
 
+/*
+ * send/sendmsg and recv/recvmsg flags (sqe->addr2)
+ *
+ * IORING_RECVSEND_POLL_FIRST	If set, instead of first attempting to send
+ *				or receive and arm poll if that yields an
+ *				-EAGAIN result, arm poll upfront and skip
+ *				the initial transfer attempt.
+ */
+#define IORING_RECVSEND_POLL_FIRST	(1U << 0)
+
 /*
  * IO completion data structure (Completion Queue Entry)
  */