diff options
Diffstat (limited to 'io_uring/net.c')
-rw-r--r-- | io_uring/net.c | 46 |
1 files changed, 39 insertions, 7 deletions
diff --git a/io_uring/net.c b/io_uring/net.c index 384915d931b7..0116cfaec848 100644 --- a/io_uring/net.c +++ b/io_uring/net.c @@ -76,12 +76,18 @@ struct io_sr_msg { /* initialised and used only by !msg send variants */ u16 addr_len; u16 buf_group; + unsigned short retry_flags; void __user *addr; 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 @@ -203,6 +209,7 @@ static inline void io_mshot_prep_retry(struct io_kiocb *req, req->flags &= ~REQ_F_BL_EMPTY; sr->done_io = 0; + sr->retry_flags = 0; sr->len = 0; /* get from the provided buffer */ req->buf_index = sr->buf_group; } @@ -409,6 +416,7 @@ 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_flags = 0; if (req->opcode == IORING_OP_SEND) { if (READ_ONCE(sqe->__pad3[0])) @@ -780,6 +788,7 @@ 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_flags = 0; if (unlikely(sqe->file_index || sqe->addr2)) return -EINVAL; @@ -828,6 +837,9 @@ int io_recvmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) return io_recvmsg_prep_setup(req); } +/* bits to clear in old and inherit in new cflags on bundle retry */ +#define CQE_F_MASK (IORING_CQE_F_SOCK_NONEMPTY|IORING_CQE_F_MORE) + /* * Finishes io_recv and io_recvmsg. * @@ -845,11 +857,27 @@ static inline bool io_recv_finish(struct io_kiocb *req, int *ret, cflags |= IORING_CQE_F_SOCK_NONEMPTY; if (sr->flags & IORING_RECVSEND_BUNDLE) { - cflags |= io_put_kbufs(req, *ret, io_bundle_nbufs(kmsg, *ret), + size_t this_ret = *ret - sr->done_io; + + cflags |= io_put_kbufs(req, this_ret, io_bundle_nbufs(kmsg, this_ret), issue_flags); + 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) goto finish; + /* + * If more is available AND it was a full transfer, retry and + * append to this one + */ + 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_flags |= IO_SR_MSG_RETRY; + return false; + } } else { cflags |= io_put_kbuf(req, *ret, issue_flags); } @@ -1088,13 +1116,21 @@ static int io_recv_buf_select(struct io_kiocb *req, struct io_async_msghdr *kmsg arg.mode |= KBUF_MODE_FREE; } - if (kmsg->msg.msg_inq > 0) + if (kmsg->msg.msg_inq > 1) arg.max_len = min_not_zero(sr->len, kmsg->msg.msg_inq); ret = io_buffers_peek(req, &arg); if (unlikely(ret < 0)) return ret; + if (arg.iovs != &kmsg->fast_iov && arg.iovs != kmsg->free_iov) { + kmsg->free_iov_nr = ret; + kmsg->free_iov = 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) { sr->buf = arg.iovs[0].iov_base; @@ -1103,11 +1139,6 @@ static int io_recv_buf_select(struct io_kiocb *req, struct io_async_msghdr *kmsg } iov_iter_init(&kmsg->msg.msg_iter, ITER_DEST, arg.iovs, ret, arg.out_len); - if (arg.iovs != &kmsg->fast_iov && arg.iovs != kmsg->free_iov) { - kmsg->free_iov_nr = ret; - kmsg->free_iov = arg.iovs; - req->flags |= REQ_F_NEED_CLEANUP; - } } else { void __user *buf; @@ -1228,6 +1259,7 @@ int io_send_zc_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) struct io_kiocb *notif; zc->done_io = 0; + zc->retry_flags = 0; req->flags |= REQ_F_POLL_NO_LAZY; if (unlikely(READ_ONCE(sqe->__pad2[0]) || READ_ONCE(sqe->addr3))) |