Bump CTL block backend threads from 14 to 32 per LUN.

This makes random read benchmarks look better on a wide ZFS pools.
I am not sure where the original value goes from, but it is there
for too long now.

MFC after:	1 week
This commit is contained in:
Alexander Motin 2021-02-23 10:58:56 -05:00
parent 0eaa97f0e8
commit 7d4c444374

View File

@ -226,7 +226,7 @@ struct ctl_be_block_io {
extern struct ctl_softc *control_softc;
static int cbb_num_threads = 14;
static int cbb_num_threads = 32;
SYSCTL_NODE(_kern_cam_ctl, OID_AUTO, block, CTLFLAG_RD | CTLFLAG_MPSAFE, 0,
"CAM Target Layer Block Backend");
SYSCTL_INT(_kern_cam_ctl_block, OID_AUTO, num_threads, CTLFLAG_RWTUN,