io_uring/kbuf: flag partial buffer mappings

A previous commit aborted mapping more for a non-incremental ring for
bundle peeking, but depending on where in the process this peeking
happened, it would not necessarily prevent a retry by the user. That can
create gaps in the received/read data.

Add struct buf_sel_arg->partial_map, which can pass this information
back. The networking side can then map that to internal state and use it
to gate retry as well.

Since this necessitates a new flag, change io_sr_msg->retry to a
retry_flags member, and store both the retry and partial map condition
in there.

Cc: [email protected]
Fixes: 26ec15e ("io_uring/kbuf: don't truncate end buffer for multiple buffer peeks")
Signed-off-by: Jens Axboe <[email protected]>

Author: axboe

io_uring/kbuf.c

@@ -271,6 +271,7 @@ static int io_ring_buffers_peek(struct io_kiocb *req, struct buf_sel_arg *arg,
 		if (len > arg->max_len) {
 			len = arg->max_len;
 			if (!(bl->flags & IOBL_INC)) {
+				arg->partial_map = 1;
 				if (iov != arg->iovs)
 					break;
 				buf->len = len;

io_uring/kbuf.h

@@ -58,7 +58,8 @@ struct buf_sel_arg {
 	size_t max_len;
 	unsigned short nr_iovs;
 	unsigned short mode;
-	unsigned buf_group;
+	unsigned short buf_group;
+	unsigned short partial_map;
 };
 
 void __user *io_buffer_select(struct io_kiocb *req, size_t *len,

io_uring/net.c

@@ -75,12 +75,17 @@ struct io_sr_msg {
 	u16				flags;
 	/* initialised and used only by !msg send variants */
 	u16				buf_group;
-	bool				retry;
+	unsigned short			retry_flags;
 	void __user			*msg_control;
 	/* used only for send zerocopy */
 	struct io_kiocb 		*notif;
 };
 
+enum sr_retry_flags {
+	IO_SR_MSG_RETRY		= 1,
+	IO_SR_MSG_PARTIAL_MAP	= 2,
+};
+
 /*
  * Number of times we'll try and do receives if there's more data. If we
  * exceed this limit, then add us to the back of the queue and retry from
@@ -187,7 +192,7 @@ static inline void io_mshot_prep_retry(struct io_kiocb *req,
 
 	req->flags &= ~REQ_F_BL_EMPTY;
 	sr->done_io = 0;
-	sr->retry = false;
+	sr->retry_flags = 0;
 	sr->len = 0; /* get from the provided buffer */
 }
 
@@ -397,7 +402,7 @@ int io_sendmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 	struct io_sr_msg *sr = io_kiocb_to_cmd(req, struct io_sr_msg);
 
 	sr->done_io = 0;
-	sr->retry = false;
+	sr->retry_flags = 0;
 	sr->len = READ_ONCE(sqe->len);
 	sr->flags = READ_ONCE(sqe->ioprio);
 	if (sr->flags & ~SENDMSG_FLAGS)
@@ -751,7 +756,7 @@ int io_recvmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 	struct io_sr_msg *sr = io_kiocb_to_cmd(req, struct io_sr_msg);
 
 	sr->done_io = 0;
-	sr->retry = false;
+	sr->retry_flags = 0;
 
 	if (unlikely(sqe->file_index || sqe->addr2))
 		return -EINVAL;
@@ -823,7 +828,7 @@ static inline bool io_recv_finish(struct io_kiocb *req, int *ret,
 
 		cflags |= io_put_kbufs(req, this_ret, io_bundle_nbufs(kmsg, this_ret),
 				      issue_flags);
-		if (sr->retry)
+		if (sr->retry_flags & IO_SR_MSG_RETRY)
 			cflags = req->cqe.flags | (cflags & CQE_F_MASK);
 		/* bundle with no more immediate buffers, we're done */
 		if (req->flags & REQ_F_BL_EMPTY)
@@ -832,12 +837,12 @@ static inline bool io_recv_finish(struct io_kiocb *req, int *ret,
 		 * If more is available AND it was a full transfer, retry and
 		 * append to this one
 		 */
-		if (!sr->retry && kmsg->msg.msg_inq > 1 && this_ret > 0 &&
+		if (!sr->retry_flags && kmsg->msg.msg_inq > 1 && this_ret > 0 &&
 		    !iov_iter_count(&kmsg->msg.msg_iter)) {
 			req->cqe.flags = cflags & ~CQE_F_MASK;
 			sr->len = kmsg->msg.msg_inq;
 			sr->done_io += this_ret;
-			sr->retry = true;
+			sr->retry_flags |= IO_SR_MSG_RETRY;
 			return false;
 		}
 	} else {
@@ -1082,6 +1087,8 @@ static int io_recv_buf_select(struct io_kiocb *req, struct io_async_msghdr *kmsg
 			kmsg->vec.iovec = arg.iovs;
 			req->flags |= REQ_F_NEED_CLEANUP;
 		}
+		if (arg.partial_map)
+			sr->retry_flags |= IO_SR_MSG_PARTIAL_MAP;
 
 		/* special case 1 vec, can be a fast path */
 		if (ret == 1) {
@@ -1276,7 +1283,7 @@ int io_send_zc_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 	int ret;
 
 	zc->done_io = 0;
-	zc->retry = false;
+	zc->retry_flags = 0;
 
 	if (unlikely(READ_ONCE(sqe->__pad2[0]) || READ_ONCE(sqe->addr3)))
 		return -EINVAL;