[PATCH mptcp-next v10 7/9] mptcp: implement .splice_read

Geliang Tang posted 9 patches 2 weeks, 1 day ago
There is a newer version of this series
[PATCH mptcp-next v10 7/9] mptcp: implement .splice_read
Posted by Geliang Tang 2 weeks, 1 day ago
From: Geliang Tang <tanggeliang@kylinos.cn>

This patch implements .splice_read interface of mptcp struct proto_ops
as mptcp_splice_read() with reference to tcp_splice_read().

Corresponding to __tcp_splice_read(), __mptcp_splice_read() is defined,
invoking mptcp_read_sock() instead of tcp_read_sock().

mptcp_splice_read() is almost the same as tcp_splice_read(), except for
sock_rps_record_flow() and __mptcp_move_skbs().

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
---
 net/mptcp/protocol.c | 94 ++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 94 insertions(+)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 6574f82d59d1..701295f6ae1b 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -4091,6 +4091,98 @@ static int mptcp_read_sock(struct sock *sk, read_descriptor_t *desc,
 	return copied;
 }
 
+static int __mptcp_splice_read(struct sock *sk, struct tcp_splice_state *tss)
+{
+	/* Store TCP splice context information in read_descriptor_t. */
+	read_descriptor_t rd_desc = {
+		.arg.data = tss,
+		.count	  = tss->len,
+	};
+
+	return mptcp_read_sock(sk, &rd_desc, tcp_splice_data_recv);
+}
+
+/**
+ *  mptcp_splice_read - splice data from MPTCP socket to a pipe
+ * @sock:	socket to splice from
+ * @ppos:	position (not valid)
+ * @pipe:	pipe to splice to
+ * @len:	number of bytes to splice
+ * @flags:	splice modifier flags
+ *
+ * Description:
+ *    Will read pages from given socket and fill them into a pipe.
+ *
+ **/
+static ssize_t mptcp_splice_read(struct socket *sock, loff_t *ppos,
+				 struct pipe_inode_info *pipe, size_t len,
+				 unsigned int flags)
+{
+	struct tcp_splice_state tss = {
+		.pipe	= pipe,
+		.len	= len,
+		.flags	= flags,
+	};
+	struct sock *sk = sock->sk;
+	ssize_t spliced = 0;
+	int ret = 0, err;
+	long timeo;
+
+	/*
+	 * We can't seek on a socket input
+	 */
+	if (unlikely(*ppos))
+		return -ESPIPE;
+
+	lock_sock(sk);
+
+	timeo = sock_rcvtimeo(sk, sock->file->f_flags & O_NONBLOCK);
+	while (tss.len) {
+		ret = __mptcp_splice_read(sk, &tss);
+		if (ret < 0) {
+			break;
+		} else if (!ret) {
+			if (spliced)
+				break;
+			err = tcp_recv_should_stop(sk, timeo);
+			if (err < 0) {
+				if (err == -ESHUTDOWN) {
+					if (__mptcp_move_skbs(sk))
+						continue;
+					break;
+				}
+				ret = err;
+				break;
+			}
+			/* if __mptcp_splice_read() got nothing while we have
+			 * an skb in receive queue, we do not want to loop.
+			 * This might happen with URG data.
+			 */
+			if (!skb_queue_empty(&sk->sk_receive_queue))
+				break;
+			ret = sk_wait_data(sk, &timeo, NULL);
+			if (ret < 0)
+				break;
+			continue;
+		}
+		tss.len -= ret;
+		spliced += ret;
+
+		if (!tss.len)
+			break;
+
+		if (tcp_recv_should_stop(sk, timeo))
+			break;
+	}
+
+	release_sock(sk);
+
+	if (spliced)
+		return spliced;
+
+	return ret;
+}
+
 static const struct proto_ops mptcp_stream_ops = {
 	.family		   = PF_INET,
 	.owner		   = THIS_MODULE,
@@ -4112,6 +4204,7 @@ static const struct proto_ops mptcp_stream_ops = {
 	.mmap		   = sock_no_mmap,
 	.set_rcvlowat	   = mptcp_set_rcvlowat,
 	.read_sock	   = mptcp_read_sock,
+	.splice_read	   = mptcp_splice_read,
 };
 
 static struct inet_protosw mptcp_protosw = {
@@ -4217,6 +4310,7 @@ static const struct proto_ops mptcp_v6_stream_ops = {
 #endif
 	.set_rcvlowat	   = mptcp_set_rcvlowat,
 	.read_sock	   = mptcp_read_sock,
+	.splice_read	   = mptcp_splice_read,
 };
 
 static struct proto mptcp_v6_prot;
-- 
2.48.1
Re: [PATCH mptcp-next v10 7/9] mptcp: implement .splice_read
Posted by Mat Martineau 4 days, 17 hours ago
On Tue, 2 Sep 2025, Geliang Tang wrote:

> From: Geliang Tang <tanggeliang@kylinos.cn>
>
> This patch implements .splice_read interface of mptcp struct proto_ops
> as mptcp_splice_read() with reference to tcp_splice_read().
>
> Corresponding to __tcp_splice_read(), __mptcp_splice_read() is defined,
> invoking mptcp_read_sock() instead of tcp_read_sock().
>
> mptcp_splice_read() is almost the same as tcp_splice_read(), except for
> sock_rps_record_flow() and __mptcp_move_skbs().
>

Hi Geliang -

Please take a look at the recent addition of mptcp_rps_record_subflows()! 
This should be added to mptcp_splice_read.

- Mat


> Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
> ---
> net/mptcp/protocol.c | 94 ++++++++++++++++++++++++++++++++++++++++++++
> 1 file changed, 94 insertions(+)
>
> diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
> index 6574f82d59d1..701295f6ae1b 100644
> --- a/net/mptcp/protocol.c
> +++ b/net/mptcp/protocol.c
> @@ -4091,6 +4091,98 @@ static int mptcp_read_sock(struct sock *sk, read_descriptor_t *desc,
> 	return copied;
> }
>
> +static int __mptcp_splice_read(struct sock *sk, struct tcp_splice_state *tss)
> +{
> +	/* Store TCP splice context information in read_descriptor_t. */
> +	read_descriptor_t rd_desc = {
> +		.arg.data = tss,
> +		.count	  = tss->len,
> +	};
> +
> +	return mptcp_read_sock(sk, &rd_desc, tcp_splice_data_recv);
> +}
> +
> +/**
> + *  mptcp_splice_read - splice data from MPTCP socket to a pipe
> + * @sock:	socket to splice from
> + * @ppos:	position (not valid)
> + * @pipe:	pipe to splice to
> + * @len:	number of bytes to splice
> + * @flags:	splice modifier flags
> + *
> + * Description:
> + *    Will read pages from given socket and fill them into a pipe.
> + *
> + **/
> +static ssize_t mptcp_splice_read(struct socket *sock, loff_t *ppos,
> +				 struct pipe_inode_info *pipe, size_t len,
> +				 unsigned int flags)
> +{
> +	struct tcp_splice_state tss = {
> +		.pipe	= pipe,
> +		.len	= len,
> +		.flags	= flags,
> +	};
> +	struct sock *sk = sock->sk;
> +	ssize_t spliced = 0;
> +	int ret = 0, err;
> +	long timeo;
> +
> +	/*
> +	 * We can't seek on a socket input
> +	 */
> +	if (unlikely(*ppos))
> +		return -ESPIPE;
> +
> +	lock_sock(sk);
> +
> +	timeo = sock_rcvtimeo(sk, sock->file->f_flags & O_NONBLOCK);
> +	while (tss.len) {
> +		ret = __mptcp_splice_read(sk, &tss);
> +		if (ret < 0) {
> +			break;
> +		} else if (!ret) {
> +			if (spliced)
> +				break;
> +			err = tcp_recv_should_stop(sk, timeo);
> +			if (err < 0) {
> +				if (err == -ESHUTDOWN) {
> +					if (__mptcp_move_skbs(sk))
> +						continue;
> +					break;
> +				}
> +				ret = err;
> +				break;
> +			}
> +			/* if __mptcp_splice_read() got nothing while we have
> +			 * an skb in receive queue, we do not want to loop.
> +			 * This might happen with URG data.
> +			 */
> +			if (!skb_queue_empty(&sk->sk_receive_queue))
> +				break;
> +			ret = sk_wait_data(sk, &timeo, NULL);
> +			if (ret < 0)
> +				break;
> +			continue;
> +		}
> +		tss.len -= ret;
> +		spliced += ret;
> +
> +		if (!tss.len)
> +			break;
> +
> +		if (tcp_recv_should_stop(sk, timeo))
> +			break;
> +	}
> +
> +	release_sock(sk);
> +
> +	if (spliced)
> +		return spliced;
> +
> +	return ret;
> +}
> +
> static const struct proto_ops mptcp_stream_ops = {
> 	.family		   = PF_INET,
> 	.owner		   = THIS_MODULE,
> @@ -4112,6 +4204,7 @@ static const struct proto_ops mptcp_stream_ops = {
> 	.mmap		   = sock_no_mmap,
> 	.set_rcvlowat	   = mptcp_set_rcvlowat,
> 	.read_sock	   = mptcp_read_sock,
> +	.splice_read	   = mptcp_splice_read,
> };
>
> static struct inet_protosw mptcp_protosw = {
> @@ -4217,6 +4310,7 @@ static const struct proto_ops mptcp_v6_stream_ops = {
> #endif
> 	.set_rcvlowat	   = mptcp_set_rcvlowat,
> 	.read_sock	   = mptcp_read_sock,
> +	.splice_read	   = mptcp_splice_read,
> };
>
> static struct proto mptcp_v6_prot;
> -- 
> 2.48.1
>
>
>
Re: [PATCH mptcp-next v10 7/9] mptcp: implement .splice_read
Posted by Geliang Tang 4 days, 15 hours ago
On Fri, 2025-09-12 at 16:41 -0700, Mat Martineau wrote:
> On Tue, 2 Sep 2025, Geliang Tang wrote:
> 
> > From: Geliang Tang <tanggeliang@kylinos.cn>
> > 
> > This patch implements .splice_read interface of mptcp struct
> > proto_ops
> > as mptcp_splice_read() with reference to tcp_splice_read().
> > 
> > Corresponding to __tcp_splice_read(), __mptcp_splice_read() is
> > defined,
> > invoking mptcp_read_sock() instead of tcp_read_sock().
> > 
> > mptcp_splice_read() is almost the same as tcp_splice_read(), except
> > for
> > sock_rps_record_flow() and __mptcp_move_skbs().
> > 
> 
> Hi Geliang -
> 
> Please take a look at the recent addition of
> mptcp_rps_record_subflows()! 
> This should be added to mptcp_splice_read.

Yes, indeed. It was included in v11.

Thanks,
-Geliang

> 
> - Mat
> 
> 
> > Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
> > ---
> > net/mptcp/protocol.c | 94
> > ++++++++++++++++++++++++++++++++++++++++++++
> > 1 file changed, 94 insertions(+)
> > 
> > diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
> > index 6574f82d59d1..701295f6ae1b 100644
> > --- a/net/mptcp/protocol.c
> > +++ b/net/mptcp/protocol.c
> > @@ -4091,6 +4091,98 @@ static int mptcp_read_sock(struct sock *sk,
> > read_descriptor_t *desc,
> > 	return copied;
> > }
> > 
> > +static int __mptcp_splice_read(struct sock *sk, struct
> > tcp_splice_state *tss)
> > +{
> > +	/* Store TCP splice context information in
> > read_descriptor_t. */
> > +	read_descriptor_t rd_desc = {
> > +		.arg.data = tss,
> > +		.count	  = tss->len,
> > +	};
> > +
> > +	return mptcp_read_sock(sk, &rd_desc,
> > tcp_splice_data_recv);
> > +}
> > +
> > +/**
> > + *  mptcp_splice_read - splice data from MPTCP socket to a pipe
> > + * @sock:	socket to splice from
> > + * @ppos:	position (not valid)
> > + * @pipe:	pipe to splice to
> > + * @len:	number of bytes to splice
> > + * @flags:	splice modifier flags
> > + *
> > + * Description:
> > + *    Will read pages from given socket and fill them into a pipe.
> > + *
> > + **/
> > +static ssize_t mptcp_splice_read(struct socket *sock, loff_t
> > *ppos,
> > +				 struct pipe_inode_info *pipe,
> > size_t len,
> > +				 unsigned int flags)
> > +{
> > +	struct tcp_splice_state tss = {
> > +		.pipe	= pipe,
> > +		.len	= len,
> > +		.flags	= flags,
> > +	};
> > +	struct sock *sk = sock->sk;
> > +	ssize_t spliced = 0;
> > +	int ret = 0, err;
> > +	long timeo;
> > +
> > +	/*
> > +	 * We can't seek on a socket input
> > +	 */
> > +	if (unlikely(*ppos))
> > +		return -ESPIPE;
> > +
> > +	lock_sock(sk);
> > +
> > +	timeo = sock_rcvtimeo(sk, sock->file->f_flags &
> > O_NONBLOCK);
> > +	while (tss.len) {
> > +		ret = __mptcp_splice_read(sk, &tss);
> > +		if (ret < 0) {
> > +			break;
> > +		} else if (!ret) {
> > +			if (spliced)
> > +				break;
> > +			err = tcp_recv_should_stop(sk, timeo);
> > +			if (err < 0) {
> > +				if (err == -ESHUTDOWN) {
> > +					if (__mptcp_move_skbs(sk))
> > +						continue;
> > +					break;
> > +				}
> > +				ret = err;
> > +				break;
> > +			}
> > +			/* if __mptcp_splice_read() got nothing
> > while we have
> > +			 * an skb in receive queue, we do not want
> > to loop.
> > +			 * This might happen with URG data.
> > +			 */
> > +			if (!skb_queue_empty(&sk-
> > >sk_receive_queue))
> > +				break;
> > +			ret = sk_wait_data(sk, &timeo, NULL);
> > +			if (ret < 0)
> > +				break;
> > +			continue;
> > +		}
> > +		tss.len -= ret;
> > +		spliced += ret;
> > +
> > +		if (!tss.len)
> > +			break;
> > +
> > +		if (tcp_recv_should_stop(sk, timeo))
> > +			break;
> > +	}
> > +
> > +	release_sock(sk);
> > +
> > +	if (spliced)
> > +		return spliced;
> > +
> > +	return ret;
> > +}
> > +
> > static const struct proto_ops mptcp_stream_ops = {
> > 	.family		   = PF_INET,
> > 	.owner		   = THIS_MODULE,
> > @@ -4112,6 +4204,7 @@ static const struct proto_ops
> > mptcp_stream_ops = {
> > 	.mmap		   = sock_no_mmap,
> > 	.set_rcvlowat	   = mptcp_set_rcvlowat,
> > 	.read_sock	   = mptcp_read_sock,
> > +	.splice_read	   = mptcp_splice_read,
> > };
> > 
> > static struct inet_protosw mptcp_protosw = {
> > @@ -4217,6 +4310,7 @@ static const struct proto_ops
> > mptcp_v6_stream_ops = {
> > #endif
> > 	.set_rcvlowat	   = mptcp_set_rcvlowat,
> > 	.read_sock	   = mptcp_read_sock,
> > +	.splice_read	   = mptcp_splice_read,
> > };
> > 
> > static struct proto mptcp_v6_prot;
> > -- 
> > 2.48.1
> > 
> > 
> >