io_uring: enable READ/WRITE to use deferred completions

A bit more surgery required here, as completions are generally done
through the kiocb->ki_complete() callback, even if they complete inline.
This enables the regular read/write path to use the io_comp_state
logic to batch inline completions.

Signed-off-by: Jens Axboe <axboe@kernel.dk>

diff --git a/fs/io_uring.c b/fs/io_uring.c
index 6c9ca4f..0bba12e 100644 (file)
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -2019,7 +2019,8 @@ static inline void req_set_fail_links(struct io_kiocb *req)
                req->flags |= REQ_F_FAIL_LINK;
 }
 
-static void io_complete_rw_common(struct kiocb *kiocb, long res)
+static void io_complete_rw_common(struct kiocb *kiocb, long res,
+                                 struct io_comp_state *cs)
 {
        struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw.kiocb);
        int cflags = 0;
@@ -2031,7 +2032,7 @@ static void io_complete_rw_common(struct kiocb *kiocb, long res)
                req_set_fail_links(req);
        if (req->flags & REQ_F_BUFFER_SELECTED)
                cflags = io_put_kbuf(req);
-       io_cqring_add_event(req, res, cflags);
+       __io_req_complete(req, res, cflags, cs);
 }
 
 static void io_sq_thread_drop_mm(struct io_ring_ctx *ctx)
@@ -2141,14 +2142,18 @@ static bool io_rw_reissue(struct io_kiocb *req, long res)
        return false;
 }
 
+static void __io_complete_rw(struct io_kiocb *req, long res, long res2,
+                            struct io_comp_state *cs)
+{
+       if (!io_rw_reissue(req, res))
+               io_complete_rw_common(&req->rw.kiocb, res, cs);
+}
+
 static void io_complete_rw(struct kiocb *kiocb, long res, long res2)
 {
        struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw.kiocb);
 
-       if (!io_rw_reissue(req, res)) {
-               io_complete_rw_common(kiocb, res);
-               io_put_req(req);
-       }
+       __io_complete_rw(req, res, res2, NULL);
 }
 
 static void io_complete_rw_iopoll(struct kiocb *kiocb, long res, long res2)
@@ -2382,14 +2387,15 @@ static inline void io_rw_done(struct kiocb *kiocb, ssize_t ret)
        }
 }
 
-static void kiocb_done(struct kiocb *kiocb, ssize_t ret)
+static void kiocb_done(struct kiocb *kiocb, ssize_t ret,
+                      struct io_comp_state *cs)
 {
        struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw.kiocb);
 
        if (req->flags & REQ_F_CUR_POS)
                req->file->f_pos = kiocb->ki_pos;
        if (ret >= 0 && kiocb->ki_complete == io_complete_rw)
-               io_complete_rw(kiocb, ret, 0);
+               __io_complete_rw(req, ret, 0, cs);
        else
                io_rw_done(kiocb, ret);
 }
@@ -2925,7 +2931,8 @@ static int io_iter_do_read(struct io_kiocb *req, struct iov_iter *iter)
        return loop_rw_iter(READ, req->file, &req->rw.kiocb, iter);
 }
 
-static int io_read(struct io_kiocb *req, bool force_nonblock)
+static int io_read(struct io_kiocb *req, bool force_nonblock,
+                  struct io_comp_state *cs)
 {
        struct iovec inline_vecs[UIO_FASTIOV], *iovec = inline_vecs;
        struct kiocb *kiocb = &req->rw.kiocb;
@@ -2960,7 +2967,7 @@ static int io_read(struct io_kiocb *req, bool force_nonblock)
 
                /* Catch -EAGAIN return for forced non-blocking submission */
                if (!force_nonblock || (ret2 != -EAGAIN && ret2 != -EIO)) {
-                       kiocb_done(kiocb, ret2);
+                       kiocb_done(kiocb, ret2, cs);
                } else {
                        iter.count = iov_count;
                        iter.nr_segs = nr_segs;
@@ -2975,7 +2982,7 @@ copy_iov:
                                if (ret2 == -EIOCBQUEUED) {
                                        goto out_free;
                                } else if (ret2 != -EAGAIN) {
-                                       kiocb_done(kiocb, ret2);
+                                       kiocb_done(kiocb, ret2, cs);
                                        goto out_free;
                                }
                        }
@@ -3021,7 +3028,8 @@ static int io_write_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe,
        return 0;
 }
 
-static int io_write(struct io_kiocb *req, bool force_nonblock)
+static int io_write(struct io_kiocb *req, bool force_nonblock,
+                   struct io_comp_state *cs)
 {
        struct iovec inline_vecs[UIO_FASTIOV], *iovec = inline_vecs;
        struct kiocb *kiocb = &req->rw.kiocb;
@@ -3090,7 +3098,7 @@ static int io_write(struct io_kiocb *req, bool force_nonblock)
                if (ret2 == -EOPNOTSUPP && (kiocb->ki_flags & IOCB_NOWAIT))
                        ret2 = -EAGAIN;
                if (!force_nonblock || ret2 != -EAGAIN) {
-                       kiocb_done(kiocb, ret2);
+                       kiocb_done(kiocb, ret2, cs);
                } else {
                        iter.count = iov_count;
                        iter.nr_segs = nr_segs;
@@ -5416,7 +5424,7 @@ static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe,
                        if (ret < 0)
                                break;
                }
-               ret = io_read(req, force_nonblock);
+               ret = io_read(req, force_nonblock, cs);
                break;
        case IORING_OP_WRITEV:
        case IORING_OP_WRITE_FIXED:
@@ -5426,7 +5434,7 @@ static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe,
                        if (ret < 0)
                                break;
                }
-               ret = io_write(req, force_nonblock);
+               ret = io_write(req, force_nonblock, cs);
                break;
        case IORING_OP_FSYNC:
                if (sqe) {