Alexander Motin 56eee7cbb4 Reimplement BIO_ORDERED handling in nvd(4).
This fixes BIO_ORDERED semantics while also improving performance by:
 - sleeping also before BIO_ORDERED bio, as defined, not only after;
 - not queueing BIO_ORDERED bio to taskqueue if no other bios running;
 - waking up sleeping taskqueue explicitly rather then rely on polling.

On Samsung SSD 970 PRO this shows sync write latency, measured with
`diskinfo -wS`, reduction from ~2ms to ~1.1ms by not sleeping without
reason till next HZ tick.

On the same device ZFS pool with 8 ZVOLs synchronously writing 4KB blocks
shows ~950 IOPS instead of ~750 IOPS before.  I suspect ZFS does not need
BIO_ORDERED on BIO_FLUSH at all, but that will be next question.

MFC after:	2 weeks
Sponsored by:	iXsystems, Inc.
2019-01-29 20:35:09 +00:00
..
2019-01-19 05:20:31 +00:00
2018-12-13 04:40:53 +00:00
2019-01-11 09:40:34 +00:00