io_uring/net: simplify zerocopy send user API

Following user feedback, this patch simplifies zerocopy send API. One of
the main complaints is that the current API is difficult with the
userspace managing notification slots, and then send retries with error
handling make it even worse.

Instead of keeping notification slots change it to the per-request
notifications model, which posts both completion and notification CQEs
for each request when any data has been sent, and only one CQE if it
fails. All notification CQEs will have IORING_CQE_F_NOTIF set and
IORING_CQE_F_MORE in completion CQEs indicates whether to wait a
notification or not.

IOSQE_CQE_SKIP_SUCCESS is disallowed with zerocopy sends for now.

This is less flexible, but greatly simplifies the user API and also the
kernel implementation. We reuse notif helpers in this patch, but in the
future there won't be need for keeping two requests.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/95287640ab98fc9417370afb16e310677c63e6ce.1662027856.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This commit is contained in:
Pavel Begunkov 2022-09-01 11:54:04 +01:00 committed by Jens Axboe
parent 57f332246a
commit b48c312be0
7 changed files with 47 additions and 76 deletions

View file

@ -71,8 +71,8 @@ struct io_uring_sqe {
__s32 splice_fd_in;
__u32 file_index;
struct {
__u16 notification_idx;
__u16 addr_len;
__u16 __pad3[1];
};
};
union {
@ -205,7 +205,7 @@ enum io_uring_op {
IORING_OP_GETXATTR,
IORING_OP_SOCKET,
IORING_OP_URING_CMD,
IORING_OP_SENDZC_NOTIF,
IORING_OP_SEND_ZC,
/* this goes last, obviously */
IORING_OP_LAST,
@ -326,10 +326,13 @@ 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
* IORING_CQE_F_NOTIF Set for notification CQEs. Can be used to distinct
* them from sends.
*/
#define IORING_CQE_F_BUFFER (1U << 0)
#define IORING_CQE_F_MORE (1U << 1)
#define IORING_CQE_F_SOCK_NONEMPTY (1U << 2)
#define IORING_CQE_F_NOTIF (1U << 3)
enum {
IORING_CQE_BUFFER_SHIFT = 16,

View file

@ -3923,8 +3923,8 @@ static int __init io_uring_init(void)
BUILD_BUG_SQE_ELEM(42, __u16, personality);
BUILD_BUG_SQE_ELEM(44, __s32, splice_fd_in);
BUILD_BUG_SQE_ELEM(44, __u32, file_index);
BUILD_BUG_SQE_ELEM(44, __u16, notification_idx);
BUILD_BUG_SQE_ELEM(46, __u16, addr_len);
BUILD_BUG_SQE_ELEM(44, __u16, addr_len);
BUILD_BUG_SQE_ELEM(46, __u16, __pad3[0]);
BUILD_BUG_SQE_ELEM(48, __u64, addr3);
BUILD_BUG_SQE_ELEM_SIZE(48, 0, cmd);
BUILD_BUG_SQE_ELEM(56, __u64, __pad2);

View file

@ -65,12 +65,12 @@ struct io_sendzc {
struct file *file;
void __user *buf;
size_t len;
u16 slot_idx;
unsigned msg_flags;
unsigned flags;
unsigned addr_len;
void __user *addr;
size_t done_io;
struct io_kiocb *notif;
};
#define IO_APOLL_MULTI_POLLED (REQ_F_APOLL_MULTISHOT | REQ_F_POLLED)
@ -879,12 +879,26 @@ int io_recv(struct io_kiocb *req, unsigned int issue_flags)
return ret;
}
void io_sendzc_cleanup(struct io_kiocb *req)
{
struct io_sendzc *zc = io_kiocb_to_cmd(req, struct io_sendzc);
zc->notif->flags |= REQ_F_CQE_SKIP;
io_notif_flush(zc->notif);
zc->notif = NULL;
}
int io_sendzc_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
{
struct io_sendzc *zc = io_kiocb_to_cmd(req, struct io_sendzc);
struct io_ring_ctx *ctx = req->ctx;
struct io_kiocb *notif;
if (READ_ONCE(sqe->__pad2[0]) || READ_ONCE(sqe->addr3))
if (READ_ONCE(sqe->__pad2[0]) || READ_ONCE(sqe->addr3) ||
READ_ONCE(sqe->__pad3[0]))
return -EINVAL;
/* we don't support IOSQE_CQE_SKIP_SUCCESS just yet */
if (req->flags & REQ_F_CQE_SKIP)
return -EINVAL;
zc->flags = READ_ONCE(sqe->ioprio);
@ -900,11 +914,17 @@ int io_sendzc_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
req->imu = READ_ONCE(ctx->user_bufs[idx]);
io_req_set_rsrc_node(req, ctx, 0);
}
notif = zc->notif = io_alloc_notif(ctx);
if (!notif)
return -ENOMEM;
notif->cqe.user_data = req->cqe.user_data;
notif->cqe.res = 0;
notif->cqe.flags = IORING_CQE_F_NOTIF;
req->flags |= REQ_F_NEED_CLEANUP;
zc->buf = u64_to_user_ptr(READ_ONCE(sqe->addr));
zc->len = READ_ONCE(sqe->len);
zc->msg_flags = READ_ONCE(sqe->msg_flags) | MSG_NOSIGNAL;
zc->slot_idx = READ_ONCE(sqe->notification_idx);
if (zc->msg_flags & MSG_DONTWAIT)
req->flags |= REQ_F_NOWAIT;
@ -976,33 +996,20 @@ static int io_sg_from_iter(struct sock *sk, struct sk_buff *skb,
int io_sendzc(struct io_kiocb *req, unsigned int issue_flags)
{
struct sockaddr_storage __address, *addr = NULL;
struct io_ring_ctx *ctx = req->ctx;
struct io_sendzc *zc = io_kiocb_to_cmd(req, struct io_sendzc);
struct io_notif_slot *notif_slot;
struct io_kiocb *notif;
struct msghdr msg;
struct iovec iov;
struct socket *sock;
unsigned msg_flags;
unsigned msg_flags, cflags;
int ret, min_ret = 0;
if (!(req->flags & REQ_F_POLLED) &&
(zc->flags & IORING_RECVSEND_POLL_FIRST))
return -EAGAIN;
if (issue_flags & IO_URING_F_UNLOCKED)
return -EAGAIN;
sock = sock_from_file(req->file);
if (unlikely(!sock))
return -ENOTSOCK;
notif_slot = io_get_notif_slot(ctx, zc->slot_idx);
if (!notif_slot)
return -EINVAL;
notif = io_get_notif(ctx, notif_slot);
if (!notif)
return -ENOMEM;
msg.msg_name = NULL;
msg.msg_control = NULL;
msg.msg_controllen = 0;
@ -1033,7 +1040,7 @@ int io_sendzc(struct io_kiocb *req, unsigned int issue_flags)
&msg.msg_iter);
if (unlikely(ret))
return ret;
ret = io_notif_account_mem(notif, zc->len);
ret = io_notif_account_mem(zc->notif, zc->len);
if (unlikely(ret))
return ret;
}
@ -1045,7 +1052,7 @@ int io_sendzc(struct io_kiocb *req, unsigned int issue_flags)
min_ret = iov_iter_count(&msg.msg_iter);
msg.msg_flags = msg_flags;
msg.msg_ubuf = &io_notif_to_data(notif)->uarg;
msg.msg_ubuf = &io_notif_to_data(zc->notif)->uarg;
msg.sg_from_iter = io_sg_from_iter;
ret = sock_sendmsg(sock, &msg);
@ -1060,6 +1067,8 @@ int io_sendzc(struct io_kiocb *req, unsigned int issue_flags)
req->flags |= REQ_F_PARTIAL_IO;
return io_setup_async_addr(req, addr, issue_flags);
}
if (ret < 0 && !zc->done_io)
zc->notif->flags |= REQ_F_CQE_SKIP;
if (ret == -ERESTARTSYS)
ret = -EINTR;
req_set_fail(req);
@ -1069,7 +1078,11 @@ int io_sendzc(struct io_kiocb *req, unsigned int issue_flags)
ret += zc->done_io;
else if (zc->done_io)
ret = zc->done_io;
io_req_set_res(req, ret, 0);
io_notif_flush(zc->notif);
req->flags &= ~REQ_F_NEED_CLEANUP;
cflags = ret >= 0 ? IORING_CQE_F_MORE : 0;
io_req_set_res(req, ret, cflags);
return IOU_OK;
}

View file

@ -55,6 +55,7 @@ int io_connect(struct io_kiocb *req, unsigned int issue_flags);
int io_sendzc(struct io_kiocb *req, unsigned int issue_flags);
int io_sendzc_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe);
void io_sendzc_cleanup(struct io_kiocb *req);
void io_netmsg_cache_free(struct io_cache_entry *entry);
#else

View file

@ -42,8 +42,7 @@ static void io_uring_tx_zerocopy_callback(struct sk_buff *skb,
}
}
struct io_kiocb *io_alloc_notif(struct io_ring_ctx *ctx,
struct io_notif_slot *slot)
struct io_kiocb *io_alloc_notif(struct io_ring_ctx *ctx)
__must_hold(&ctx->uring_lock)
{
struct io_kiocb *notif;
@ -59,27 +58,20 @@ struct io_kiocb *io_alloc_notif(struct io_ring_ctx *ctx,
io_get_task_refs(1);
notif->rsrc_node = NULL;
io_req_set_rsrc_node(notif, ctx, 0);
notif->cqe.user_data = slot->tag;
notif->cqe.flags = slot->seq++;
notif->cqe.res = 0;
nd = io_notif_to_data(notif);
nd->account_pages = 0;
nd->uarg.flags = SKBFL_ZEROCOPY_FRAG | SKBFL_DONT_ORPHAN;
nd->uarg.callback = io_uring_tx_zerocopy_callback;
/* master ref owned by io_notif_slot, will be dropped on flush */
refcount_set(&nd->uarg.refcnt, 1);
return notif;
}
void io_notif_slot_flush(struct io_notif_slot *slot)
void io_notif_flush(struct io_kiocb *notif)
__must_hold(&slot->notif->ctx->uring_lock)
{
struct io_kiocb *notif = slot->notif;
struct io_notif_data *nd = io_notif_to_data(notif);
slot->notif = NULL;
/* drop slot's master ref */
if (refcount_dec_and_test(&nd->uarg.refcnt)) {
notif->io_task_work.func = __io_notif_complete_tw;

View file

@ -15,53 +15,14 @@ struct io_notif_data {
unsigned long account_pages;
};
struct io_notif_slot {
/*
* Current/active notifier. A slot holds only one active notifier at a
* time and keeps one reference to it. Flush releases the reference and
* lazily replaces it with a new notifier.
*/
struct io_kiocb *notif;
/*
* Default ->user_data for this slot notifiers CQEs
*/
u64 tag;
/*
* Notifiers of a slot live in generations, we create a new notifier
* only after flushing the previous one. Track the sequential number
* for all notifiers and copy it into notifiers's cqe->cflags
*/
u32 seq;
};
void io_notif_slot_flush(struct io_notif_slot *slot);
struct io_kiocb *io_alloc_notif(struct io_ring_ctx *ctx,
struct io_notif_slot *slot);
void io_notif_flush(struct io_kiocb *notif);
struct io_kiocb *io_alloc_notif(struct io_ring_ctx *ctx);
static inline struct io_notif_data *io_notif_to_data(struct io_kiocb *notif)
{
return io_kiocb_to_cmd(notif, struct io_notif_data);
}
static inline struct io_kiocb *io_get_notif(struct io_ring_ctx *ctx,
struct io_notif_slot *slot)
{
if (!slot->notif)
slot->notif = io_alloc_notif(ctx, slot);
return slot->notif;
}
static inline struct io_notif_slot *io_get_notif_slot(struct io_ring_ctx *ctx,
unsigned idx)
__must_hold(&ctx->uring_lock)
{
if (idx >= ctx->nr_notif_slots)
return NULL;
idx = array_index_nospec(idx, ctx->nr_notif_slots);
return &ctx->notif_slots[idx];
}
static inline int io_notif_account_mem(struct io_kiocb *notif, unsigned len)
{
struct io_ring_ctx *ctx = notif->ctx;

View file

@ -470,7 +470,7 @@ const struct io_op_def io_op_defs[] = {
.issue = io_uring_cmd,
.prep_async = io_uring_cmd_prep_async,
},
[IORING_OP_SENDZC_NOTIF] = {
[IORING_OP_SEND_ZC] = {
.name = "SENDZC_NOTIF",
.needs_file = 1,
.unbound_nonreg_file = 1,
@ -483,6 +483,7 @@ const struct io_op_def io_op_defs[] = {
.prep = io_sendzc_prep,
.issue = io_sendzc,
.prep_async = io_sendzc_prep_async,
.cleanup = io_sendzc_cleanup,
#else
.prep = io_eopnotsupp_prep,
#endif