diff options
| author | Pavel Begunkov <asml.silence@gmail.com> | 2020-01-25 00:40:24 +0300 |
|---|---|---|
| committer | Jens Axboe <axboe@kernel.dk> | 2020-01-28 01:36:30 +0300 |
| commit | 8cdf2193a3335b4cfb6e023b41ac293d0843d287 (patch) | |
| tree | 28b71b8e09fadea98f8906fe6fa9bf03438110a3 /fs | |
| parent | 980ad26304abf11e78caaa68023411b9c088b848 (diff) | |
| download | linux-8cdf2193a3335b4cfb6e023b41ac293d0843d287.tar.xz | |
io_uring: add comment for drain_next
Draining the middle of a link is tricky, so leave a comment there
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Diffstat (limited to 'fs')
| -rw-r--r-- | fs/io_uring.c | 7 |
1 files changed, 7 insertions, 0 deletions
diff --git a/fs/io_uring.c b/fs/io_uring.c index 4bde5da2c2f5..a700ee5fc89d 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -4629,6 +4629,13 @@ err_req: if (*link) { struct io_kiocb *head = *link; + /* + * Taking sequential execution of a link, draining both sides + * of the link also fullfils IOSQE_IO_DRAIN semantics for all + * requests in the link. So, it drains the head and the + * next after the link request. The last one is done via + * drain_next flag to persist the effect across calls. + */ if (sqe_flags & IOSQE_IO_DRAIN) { head->flags |= REQ_F_IO_DRAIN; ctx->drain_next = 1; |