Fix thread hang when using async engines (libaio,etc.) when too low of a iops rate is specified.
Rate limiting logic was using thread_data->cur_depth to decide the
min_evts number to ask for during its "flush" prior to sleeping.
td->cur_depth, however, does not properly track in-flight IOs submitted
to the async engines. Added field to thread_data structure and use
that, instead, to track IOs currently in flight.
Signed-off-by: Ryan Marchand <rmarchan@amazon.com>
Signed-off-by: Steven Noonan <snoonan@amazon.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
diff --git a/fio.h b/fio.h
index a9286b8..c8b94f6 100644
--- a/fio.h
+++ b/fio.h
@@ -347,10 +347,23 @@
struct ioengine_ops *io_ops;
/*
- * Current IO depth and list of free and busy io_u's.
+ * Queue depth of io_u's that fio MIGHT do
*/
unsigned int cur_depth;
+
+ /*
+ * io_u's about to be committed
+ */
unsigned int io_u_queued;
+
+ /*
+ * io_u's submitted but not completed yet
+ */
+ unsigned int io_u_in_flight;
+
+ /*
+ * List of free and busy io_u's
+ */
struct flist_head io_u_freelist;
struct flist_head io_u_busylist;
struct flist_head io_u_requeues;