Commit f2f87370 authored by Pavel Begunkov's avatar Pavel Begunkov Committed by Jens Axboe
Browse files

io_uring: link requests with singly linked list 



Singly linked list for keeping linked requests is enough, because we
almost always operate on the head and traverse forward with the
exception of linked timeouts going 1 hop backwards.

Replace ->link_list with a handmade singly linked list. Also kill
REQ_F_LINK_HEAD in favour of checking a newly added ->list for NULL
directly.

That saves 8B in io_kiocb, is not as heavy as list fixup, makes better
use of cache by not touching a previous request (i.e. last request of
the link) each time on list modification and optimises cache use further
in the following patch, and actually makes travesal easier removing in
the end some lines. Also, keeping invariant in ->list instead of having
REQ_F_LINK_HEAD is less error-prone.

Signed-off-by: default avatarPavel Begunkov <asml.silence@gmail.com>
Signed-off-by: default avatarJens Axboe <axboe@kernel.dk>
parent 90cd7e42

fs/io_uring.c

+52 −81
Original line number Diff line number Diff line
@@ -600,7 +600,6 @@ enum { 
	REQ_F_FORCE_ASYNC_BIT	= IOSQE_ASYNC_BIT,

	REQ_F_BUFFER_SELECT_BIT	= IOSQE_BUFFER_SELECT_BIT,



	REQ_F_LINK_HEAD_BIT,

	REQ_F_FAIL_LINK_BIT,

	REQ_F_INFLIGHT_BIT,

	REQ_F_CUR_POS_BIT,
@@ -632,8 +631,6 @@ enum { 
	/* IOSQE_BUFFER_SELECT */

	REQ_F_BUFFER_SELECT	= BIT(REQ_F_BUFFER_SELECT_BIT),



	/* head of a link */

	REQ_F_LINK_HEAD		= BIT(REQ_F_LINK_HEAD_BIT),

	/* fail rest of links */

	REQ_F_FAIL_LINK		= BIT(REQ_F_FAIL_LINK_BIT),

	/* on inflight list */
@@ -715,7 +712,7 @@ struct io_kiocb { 
	struct task_struct		*task;

	u64				user_data;



	struct list_head		link_list;

	struct io_kiocb			*link;



	/*

	 * 1. used with ctx->iopoll_list with reads/writes
@@ -1023,6 +1020,9 @@ struct sock *io_uring_get_socket(struct file *file) 
}

EXPORT_SYMBOL(io_uring_get_socket);



#define io_for_each_link(pos, head) \

	for (pos = (head); pos; pos = pos->link)



static inline void io_clean_op(struct io_kiocb *req)

{

	if (req->flags & (REQ_F_NEED_CLEANUP | REQ_F_BUFFER_SELECTED |
@@ -1501,9 +1501,7 @@ static void io_prep_async_link(struct io_kiocb *req) 
{

	struct io_kiocb *cur;



	io_prep_async_work(req);

	if (req->flags & REQ_F_LINK_HEAD)

		list_for_each_entry(cur, &req->link_list, link_list)

	io_for_each_link(cur, req)

		io_prep_async_work(cur);

}
@@ -1687,20 +1685,15 @@ static inline bool __io_match_files(struct io_kiocb *req, 
		req->work.identity->files == files;

}



static bool io_match_files(struct io_kiocb *req,

			   struct files_struct *files)

static bool io_match_files(struct io_kiocb *head, struct files_struct *files)

{

	struct io_kiocb *link;

	struct io_kiocb *req;



	if (!files)

		return true;

	io_for_each_link(req, head) {

		if (__io_match_files(req, files))

			return true;

	if (req->flags & REQ_F_LINK_HEAD) {

		list_for_each_entry(link, &req->link_list, link_list) {

			if (__io_match_files(link, files))

				return true;

		}

	}

	return false;

}
@@ -1967,6 +1960,14 @@ static void __io_free_req(struct io_kiocb *req) 
	percpu_ref_put(&ctx->refs);

}



static inline void io_remove_next_linked(struct io_kiocb *req)

{

	struct io_kiocb *nxt = req->link;



	req->link = nxt->link;

	nxt->link = NULL;

}



static void io_kill_linked_timeout(struct io_kiocb *req)

{

	struct io_ring_ctx *ctx = req->ctx;
@@ -1975,8 +1976,8 @@ static void io_kill_linked_timeout(struct io_kiocb *req) 
	unsigned long flags;



	spin_lock_irqsave(&ctx->completion_lock, flags);

	link = list_first_entry_or_null(&req->link_list, struct io_kiocb,

					link_list);

	link = req->link;



	/*

	 * Can happen if a linked timeout fired and link had been like

	 * req -> link t-out -> link t-out [-> ...]
@@ -1985,7 +1986,7 @@ static void io_kill_linked_timeout(struct io_kiocb *req) 
		struct io_timeout_data *io = link->async_data;

		int ret;



		list_del_init(&link->link_list);

		io_remove_next_linked(req);

		link->timeout.head = NULL;

		ret = hrtimer_try_to_cancel(&io->timer);

		if (ret != -1) {
@@ -2003,41 +2004,22 @@ static void io_kill_linked_timeout(struct io_kiocb *req) 
	}

}



static struct io_kiocb *io_req_link_next(struct io_kiocb *req)

{

	struct io_kiocb *nxt;



	/*

	 * The list should never be empty when we are called here. But could

	 * potentially happen if the chain is messed up, check to be on the

	 * safe side.

	 */

	if (unlikely(list_empty(&req->link_list)))

		return NULL;



	nxt = list_first_entry(&req->link_list, struct io_kiocb, link_list);

	list_del_init(&req->link_list);

	if (!list_empty(&nxt->link_list))

		nxt->flags |= REQ_F_LINK_HEAD;

	return nxt;

}



/*

 * Called if REQ_F_LINK_HEAD is set, and we fail the head request

 */

static void io_fail_links(struct io_kiocb *req)

{

	struct io_kiocb *link, *nxt;

	struct io_ring_ctx *ctx = req->ctx;

	unsigned long flags;



	spin_lock_irqsave(&ctx->completion_lock, flags);

	while (!list_empty(&req->link_list)) {

		struct io_kiocb *link = list_first_entry(&req->link_list,

						struct io_kiocb, link_list);

	link = req->link;

	req->link = NULL;



		list_del_init(&link->link_list);

		trace_io_uring_fail_link(req, link);

	while (link) {

		nxt = link->link;

		link->link = NULL;



		trace_io_uring_fail_link(req, link);

		io_cqring_fill_event(link, -ECANCELED);



		/*
@@ -2049,8 +2031,8 @@ static void io_fail_links(struct io_kiocb *req) 
			io_put_req_deferred(link, 2);

		else

			io_double_put_req(link);

		link = nxt;

	}



	io_commit_cqring(ctx);

	spin_unlock_irqrestore(&ctx->completion_lock, flags);
@@ -2059,7 +2041,6 @@ static void io_fail_links(struct io_kiocb *req) 


static struct io_kiocb *__io_req_find_next(struct io_kiocb *req)

{

	req->flags &= ~REQ_F_LINK_HEAD;

	if (req->flags & REQ_F_LINK_TIMEOUT)

		io_kill_linked_timeout(req);
@@ -2069,15 +2050,19 @@ static struct io_kiocb *__io_req_find_next(struct io_kiocb *req) 
	 * dependencies to the next request. In case of failure, fail the rest

	 * of the chain.

	 */

	if (likely(!(req->flags & REQ_F_FAIL_LINK)))

		return io_req_link_next(req);

	if (likely(!(req->flags & REQ_F_FAIL_LINK))) {

		struct io_kiocb *nxt = req->link;



		req->link = NULL;

		return nxt;

	}

	io_fail_links(req);

	return NULL;

}



static struct io_kiocb *io_req_find_next(struct io_kiocb *req)

static inline struct io_kiocb *io_req_find_next(struct io_kiocb *req)

{

	if (likely(!(req->flags & REQ_F_LINK_HEAD)))

	if (likely(!(req->link) && !(req->flags & REQ_F_LINK_TIMEOUT)))

		return NULL;

	return __io_req_find_next(req);

}
@@ -2173,7 +2158,7 @@ static void io_req_task_queue(struct io_kiocb *req) 
	}

}



static void io_queue_next(struct io_kiocb *req)

static inline void io_queue_next(struct io_kiocb *req)

{

	struct io_kiocb *nxt = io_req_find_next(req);
@@ -2230,7 +2215,6 @@ static void io_req_free_batch(struct req_batch *rb, struct io_kiocb *req) 
		io_free_req(req);

		return;

	}

	if (req->flags & REQ_F_LINK_HEAD)

	io_queue_next(req);



	if (req->task != rb->task) {
@@ -6015,10 +5999,9 @@ static u32 io_get_sequence(struct io_kiocb *req) 
{

	struct io_kiocb *pos;

	struct io_ring_ctx *ctx = req->ctx;

	u32 total_submitted, nr_reqs = 1;

	u32 total_submitted, nr_reqs = 0;



	if (req->flags & REQ_F_LINK_HEAD)

		list_for_each_entry(pos, &req->link_list, link_list)

	io_for_each_link(pos, req)

		nr_reqs++;



	total_submitted = ctx->cached_sq_head - ctx->cached_sq_dropped;
@@ -6374,7 +6357,7 @@ static enum hrtimer_restart io_link_timeout_fn(struct hrtimer *timer) 
	 * race with the completion of the linked work.

	 */

	if (prev && refcount_inc_not_zero(&prev->refs))

		list_del_init(&req->link_list);

		io_remove_next_linked(prev);

	else

		prev = NULL;

	spin_unlock_irqrestore(&ctx->completion_lock, flags);
@@ -6392,8 +6375,8 @@ static enum hrtimer_restart io_link_timeout_fn(struct hrtimer *timer) 
static void __io_queue_linked_timeout(struct io_kiocb *req)

{

	/*

	 * If the list is now empty, then our linked request finished before

	 * we got a chance to setup the timer

	 * If the back reference is NULL, then our linked request finished

	 * before we got a chance to setup the timer

	 */

	if (req->timeout.head) {

		struct io_timeout_data *data = req->async_data;
@@ -6418,16 +6401,10 @@ static void io_queue_linked_timeout(struct io_kiocb *req) 


static struct io_kiocb *io_prep_linked_timeout(struct io_kiocb *req)

{

	struct io_kiocb *nxt;



	if (!(req->flags & REQ_F_LINK_HEAD))

		return NULL;

	if (req->flags & REQ_F_LINK_TIMEOUT)

		return NULL;

	struct io_kiocb *nxt = req->link;



	nxt = list_first_entry_or_null(&req->link_list, struct io_kiocb,

					link_list);

	if (!nxt || nxt->opcode != IORING_OP_LINK_TIMEOUT)

	if (!nxt || (req->flags & REQ_F_LINK_TIMEOUT) ||

	    nxt->opcode != IORING_OP_LINK_TIMEOUT)

		return NULL;



	nxt->timeout.head = req;
@@ -6575,7 +6552,7 @@ static int io_submit_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, 
			return ret;

		}

		trace_io_uring_link(ctx, req, head);

		list_add_tail(&req->link_list, &head->link_list);

		link->last->link = req;

		link->last = req;



		/* last request of a link, enqueue the link */
@@ -6589,9 +6566,6 @@ static int io_submit_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, 
			ctx->drain_next = 0;

		}

		if (req->flags & (REQ_F_LINK | REQ_F_HARDLINK)) {

			req->flags |= REQ_F_LINK_HEAD;

			INIT_LIST_HEAD(&req->link_list);



			ret = io_req_defer_prep(req, sqe);

			if (unlikely(ret))

				req->flags |= REQ_F_FAIL_LINK;
@@ -6724,6 +6698,7 @@ static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req, 
	req->file = NULL;

	req->ctx = ctx;

	req->flags = 0;

	req->link = NULL;

	/* one is dropped after submission, the other at completion */

	refcount_set(&req->refs, 2);

	req->task = current;
@@ -8682,14 +8657,10 @@ static bool io_match_link(struct io_kiocb *preq, struct io_kiocb *req) 
{

	struct io_kiocb *link;



	if (!(preq->flags & REQ_F_LINK_HEAD))

		return false;



	list_for_each_entry(link, &preq->link_list, link_list) {

	io_for_each_link(link, preq->link) {

		if (link == req)

			return true;

	}



	return false;

}

CRA Git | Maintained and supported by SUSTech CRA and CCSE