diff options
| author | Pavel Begunkov <asml.silence@gmail.com> | 2020-07-06 17:59:30 +0300 |
|---|---|---|
| committer | Jens Axboe <axboe@kernel.dk> | 2020-07-06 09:06:20 -0600 |
| commit | eba0a4dd2aa5c47ca5b0c56ffb6d6665e047ff72 (patch) | |
| tree | e36780c1ba1eb115af4a80853ce48c5b25618b37 /fs/io_uring.c | |
| parent | 3aadc23e6054353ca056bf14e87250c79efbd7ed (diff) | |
| download | lwn-eba0a4dd2aa5c47ca5b0c56ffb6d6665e047ff72.tar.gz lwn-eba0a4dd2aa5c47ca5b0c56ffb6d6665e047ff72.zip | |
io_uring: fix stopping iopoll'ing too early
Nobody adjusts *nr_events (number of completed requests) before calling
io_iopoll_getevents(), so the passed @min shouldn't be adjusted as well.
Othewise it can return less than initially asked @min without hitting
need_resched().
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Diffstat (limited to 'fs/io_uring.c')
| -rw-r--r-- | fs/io_uring.c | 9 |
1 files changed, 2 insertions, 7 deletions
diff --git a/fs/io_uring.c b/fs/io_uring.c index 60f1a81c6c35..332008f346e3 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -2044,7 +2044,7 @@ static int io_iopoll_getevents(struct io_ring_ctx *ctx, unsigned int *nr_events, ret = io_do_iopoll(ctx, nr_events, min); if (ret < 0) return ret; - if (!min || *nr_events >= min) + if (*nr_events >= min) return 0; } @@ -2087,8 +2087,6 @@ static int io_iopoll_check(struct io_ring_ctx *ctx, unsigned *nr_events, */ mutex_lock(&ctx->uring_lock); do { - int tmin = 0; - /* * Don't enter poll loop if we already have events pending. * If we do, we can potentially be spinning for commands that @@ -2113,10 +2111,7 @@ static int io_iopoll_check(struct io_ring_ctx *ctx, unsigned *nr_events, mutex_lock(&ctx->uring_lock); } - if (*nr_events < min) - tmin = min - *nr_events; - - ret = io_iopoll_getevents(ctx, nr_events, tmin); + ret = io_iopoll_getevents(ctx, nr_events, min); if (ret <= 0) break; ret = 0; |