From f94fd25cb0aaf77fd7453f31c5d394a1a68ecf60 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 28 Apr 2022 18:45:06 -0600 Subject: [PATCH 1/2] tcp: pass back data left in socket after receive This is currently done for CMSG_INQ, add an ability to do so via struct msghdr as well and have CMSG_INQ use that too. If the caller sets msghdr->msg_get_inq, then we'll pass back the hint in msghdr->msg_inq. Rearrange struct msghdr a bit so we can add this member while shrinking it at the same time. On a 64-bit build, it was 96 bytes before this change and 88 bytes afterwards. Reviewed-by: Eric Dumazet Signed-off-by: Jens Axboe Link: https://lore.kernel.org/r/650c22ca-cffc-0255-9a05-2413a1e20826@kernel.dk Signed-off-by: Jakub Kicinski --- include/linux/socket.h | 6 +++++- net/ipv4/tcp.c | 16 ++++++++++------ 2 files changed, 15 insertions(+), 7 deletions(-) diff --git a/include/linux/socket.h b/include/linux/socket.h index 6f85f5d957ef..12085c9a8544 100644 --- a/include/linux/socket.h +++ b/include/linux/socket.h @@ -50,6 +50,9 @@ struct linger { struct msghdr { void *msg_name; /* ptr to socket address structure */ int msg_namelen; /* size of socket address structure */ + + int msg_inq; /* output, data left in socket */ + struct iov_iter msg_iter; /* data */ /* @@ -62,8 +65,9 @@ struct msghdr { void __user *msg_control_user; }; bool msg_control_is_user : 1; - __kernel_size_t msg_controllen; /* ancillary data buffer length */ + bool msg_get_inq : 1;/* return INQ after receive */ unsigned int msg_flags; /* flags on received message */ + __kernel_size_t msg_controllen; /* ancillary data buffer length */ struct kiocb *msg_iocb; /* ptr to iocb for async requests */ }; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index cf18fbcbf123..bb7ef45408e1 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2335,8 +2335,10 @@ static int tcp_recvmsg_locked(struct sock *sk, struct msghdr *msg, size_t len, if (sk->sk_state == TCP_LISTEN) goto out; - if (tp->recvmsg_inq) + if (tp->recvmsg_inq) { *cmsg_flags = TCP_CMSG_INQ; + msg->msg_get_inq = 1; + } timeo = sock_rcvtimeo(sk, nonblock); /* Urgent data needs to be handled specially. */ @@ -2559,7 +2561,7 @@ recv_sndq: int tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int nonblock, int flags, int *addr_len) { - int cmsg_flags = 0, ret, inq; + int cmsg_flags = 0, ret; struct scm_timestamping_internal tss; if (unlikely(flags & MSG_ERRQUEUE)) @@ -2576,12 +2578,14 @@ int tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int nonblock, release_sock(sk); sk_defer_free_flush(sk); - if (cmsg_flags && ret >= 0) { + if ((cmsg_flags || msg->msg_get_inq) && ret >= 0) { if (cmsg_flags & TCP_CMSG_TS) tcp_recv_timestamp(msg, sk, &tss); - if (cmsg_flags & TCP_CMSG_INQ) { - inq = tcp_inq_hint(sk); - put_cmsg(msg, SOL_TCP, TCP_CM_INQ, sizeof(inq), &inq); + if (msg->msg_get_inq) { + msg->msg_inq = tcp_inq_hint(sk); + if (cmsg_flags & TCP_CMSG_INQ) + put_cmsg(msg, SOL_TCP, TCP_CM_INQ, + sizeof(msg->msg_inq), &msg->msg_inq); } } return ret; From f548a12efd5ab97e6b1fb332e5634ce44b3d9328 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Tue, 26 Apr 2022 17:39:50 -0600 Subject: [PATCH 2/2] io_uring: return hint on whether more data is available after receive For now just use a CQE flag for this, with big CQE support we could return the actual number of bytes left. Signed-off-by: Jens Axboe --- fs/io_uring.c | 21 ++++++++++++++++----- include/uapi/linux/io_uring.h | 2 ++ 2 files changed, 18 insertions(+), 5 deletions(-) diff --git a/fs/io_uring.c b/fs/io_uring.c index 5a0388bac42c..20c5d29e5b6c 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -5948,6 +5948,7 @@ static int io_recvmsg(struct io_kiocb *req, unsigned int issue_flags) struct io_sr_msg *sr = &req->sr_msg; struct socket *sock; struct io_buffer *kbuf; + unsigned int cflags; unsigned flags; int ret, min_ret = 0; bool force_nonblock = issue_flags & IO_URING_F_NONBLOCK; @@ -5981,6 +5982,8 @@ static int io_recvmsg(struct io_kiocb *req, unsigned int issue_flags) if (flags & MSG_WAITALL) min_ret = iov_iter_count(&kmsg->msg.msg_iter); + kmsg->msg.msg_get_inq = 1; + ret = __sys_recvmsg_sock(sock, &kmsg->msg, req->sr_msg.umsg, kmsg->uaddr, flags); if (ret < min_ret) { @@ -6006,7 +6009,10 @@ static int io_recvmsg(struct io_kiocb *req, unsigned int issue_flags) ret += sr->done_io; else if (sr->done_io) ret = sr->done_io; - __io_req_complete(req, issue_flags, ret, io_put_kbuf(req, issue_flags)); + cflags = io_put_kbuf(req, issue_flags); + if (kmsg->msg.msg_inq) + cflags |= IORING_CQE_F_SOCK_NONEMPTY; + __io_req_complete(req, issue_flags, ret, cflags); return 0; } @@ -6018,6 +6024,7 @@ static int io_recv(struct io_kiocb *req, unsigned int issue_flags) void __user *buf = sr->buf; struct socket *sock; struct iovec iov; + unsigned int cflags; unsigned flags; int ret, min_ret = 0; bool force_nonblock = issue_flags & IO_URING_F_NONBLOCK; @@ -6038,11 +6045,12 @@ static int io_recv(struct io_kiocb *req, unsigned int issue_flags) goto out_free; msg.msg_name = NULL; - msg.msg_control = NULL; - msg.msg_controllen = 0; msg.msg_namelen = 0; - msg.msg_iocb = NULL; + msg.msg_control = NULL; + msg.msg_get_inq = 1; msg.msg_flags = 0; + msg.msg_controllen = 0; + msg.msg_iocb = NULL; flags = req->sr_msg.msg_flags; if (force_nonblock) @@ -6073,7 +6081,10 @@ out_free: ret += sr->done_io; else if (sr->done_io) ret = sr->done_io; - __io_req_complete(req, issue_flags, ret, io_put_kbuf(req, issue_flags)); + cflags = io_put_kbuf(req, issue_flags); + if (msg.msg_inq) + cflags |= IORING_CQE_F_SOCK_NONEMPTY; + __io_req_complete(req, issue_flags, ret, cflags); return 0; } diff --git a/include/uapi/linux/io_uring.h b/include/uapi/linux/io_uring.h index 49d1f3994f8d..92d1799892b2 100644 --- a/include/uapi/linux/io_uring.h +++ b/include/uapi/linux/io_uring.h @@ -220,9 +220,11 @@ struct io_uring_cqe { * * IORING_CQE_F_BUFFER If set, the upper 16 bits are the buffer ID * IORING_CQE_F_MORE If set, parent SQE will generate more CQE entries + * IORING_CQE_F_SOCK_NONEMPTY If set, more data to read after socket recv */ #define IORING_CQE_F_BUFFER (1U << 0) #define IORING_CQE_F_MORE (1U << 1) +#define IORING_CQE_F_SOCK_NONEMPTY (1U << 2) enum { IORING_CQE_BUFFER_SHIFT = 16,