[PATCH 3/4] io_uring: let fast poll support multishot

[Hao Xu]

From: Hao Xu <howeyxu@xxxxxxxxxxx>

For operations like accept, multishot is a useful feature, since we can
reduce a number of accept sqe. Let's integrate it to fast poll, it may
be good for other operations in the future.

Signed-off-by: Hao Xu <howeyxu@xxxxxxxxxxx>
[fold in cleaning up]
Signed-off-by: Jens Axboe <axboe@xxxxxxxxx>
---
 fs/io_uring.c | 47 ++++++++++++++++++++++++++++++++---------------
 1 file changed, 32 insertions(+), 15 deletions(-)

diff --git a/fs/io_uring.c b/fs/io_uring.c
index c2ee184ac693..e0d12af04cd1 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -5955,6 +5955,7 @@ static void io_poll_remove_entries(struct io_kiocb *req)
 	rcu_read_unlock();
 }
 
+static int io_issue_sqe(struct io_kiocb *req, unsigned int issue_flags);
 /*
  * All poll tw should go through this. Checks for poll events, manages
  * references, does rewait, etc.
@@ -5963,10 +5964,10 @@ static void io_poll_remove_entries(struct io_kiocb *req)
  * either spurious wakeup or multishot CQE is served. 0 when it's done with
  * the request, then the mask is stored in req->cqe.res.
  */
-static int io_poll_check_events(struct io_kiocb *req, bool locked)
+static int io_poll_check_events(struct io_kiocb *req, bool *locked)
 {
 	struct io_ring_ctx *ctx = req->ctx;
-	int v;
+	int v, ret;
 
 	/* req->task == current here, checking PF_EXITING is safe */
 	if (unlikely(req->task->flags & PF_EXITING))
@@ -5990,23 +5991,37 @@ static int io_poll_check_events(struct io_kiocb *req, bool locked)
 			req->cqe.res = vfs_poll(req->file, &pt) & req->apoll_events;
 		}
 
-		/* multishot, just fill an CQE and proceed */
-		if (req->cqe.res && !(req->apoll_events & EPOLLONESHOT)) {
-			__poll_t mask = mangle_poll(req->cqe.res & req->apoll_events);
+		if ((unlikely(!req->cqe.res)))
+			continue;
+		if (req->apoll_events & EPOLLONESHOT)
+			return 0;
+
+		/* multishot, just fill a CQE and proceed */
+		if (!(req->flags & REQ_F_APOLL_MULTISHOT)) {
+			__poll_t mask = mangle_poll(req->cqe.res &
+						    req->apoll_events);
 			bool filled;
 
 			spin_lock(&ctx->completion_lock);
-			filled = io_fill_cqe_aux(ctx, req->cqe.user_data, mask,
-						 IORING_CQE_F_MORE);
+			filled = io_fill_cqe_aux(ctx, req->cqe.user_data,
+						 mask, IORING_CQE_F_MORE);
 			io_commit_cqring(ctx);
 			spin_unlock(&ctx->completion_lock);
-			if (unlikely(!filled))
-				return -ECANCELED;
-			io_cqring_ev_posted(ctx);
-		} else if (req->cqe.res) {
-			return 0;
+			if (filled) {
+				io_cqring_ev_posted(ctx);
+				continue;
+			}
+			return -ECANCELED;
 		}
 
+		io_tw_lock(req->ctx, locked);
+		if (unlikely(req->task->flags & PF_EXITING))
+			return -EFAULT;
+		ret = io_issue_sqe(req,
+				   IO_URING_F_NONBLOCK|IO_URING_F_COMPLETE_DEFER);
+		if (ret)
+			return ret;
+
 		/*
 		 * Release all references, retry if someone tried to restart
 		 * task_work while we were executing it.
@@ -6021,7 +6036,7 @@ static void io_poll_task_func(struct io_kiocb *req, bool *locked)
 	struct io_ring_ctx *ctx = req->ctx;
 	int ret;
 
-	ret = io_poll_check_events(req, *locked);
+	ret = io_poll_check_events(req, locked);
 	if (ret > 0)
 		return;
 
@@ -6046,7 +6061,7 @@ static void io_apoll_task_func(struct io_kiocb *req, bool *locked)
 	struct io_ring_ctx *ctx = req->ctx;
 	int ret;
 
-	ret = io_poll_check_events(req, *locked);
+	ret = io_poll_check_events(req, locked);
 	if (ret > 0)
 		return;
 
@@ -6286,7 +6301,7 @@ static int io_arm_poll_handler(struct io_kiocb *req, unsigned issue_flags)
 	struct io_ring_ctx *ctx = req->ctx;
 	struct async_poll *apoll;
 	struct io_poll_table ipt;
-	__poll_t mask = EPOLLONESHOT | POLLERR | POLLPRI;
+	__poll_t mask = POLLERR | POLLPRI;
 	int ret;
 
 	if (!def->pollin && !def->pollout)
@@ -6295,6 +6310,8 @@ static int io_arm_poll_handler(struct io_kiocb *req, unsigned issue_flags)
 		return IO_APOLL_ABORTED;
 	if ((req->flags & (REQ_F_POLLED|REQ_F_PARTIAL_IO)) == REQ_F_POLLED)
 		return IO_APOLL_ABORTED;
+	if (!(req->flags & REQ_F_APOLL_MULTISHOT))
+		mask |= EPOLLONESHOT;
 
 	if (def->pollin) {
 		mask |= POLLIN | POLLRDNORM;
-- 
2.36.0