diff options
| author | Matthew Ahrens <[email protected]> | 2016-07-20 15:42:13 -0700 |
|---|---|---|
| committer | Brian Behlendorf <[email protected]> | 2016-08-19 12:48:03 -0700 |
| commit | 2bce8049c3d782f4feb72493564754c0595606bf (patch) | |
| tree | cbeb72318d3d772c4c689ad9e9cccc52a9229676 /module/zfs/dmu_tx.c | |
| parent | 8bea9815048e4b1a85905e0b381865dedd266f2d (diff) | |
OpenZFS 7004 - dmu_tx_hold_zap() does dnode_hold() 7x on same object
Using a benchmark which has 32 threads creating 2 million files in the
same directory, on a machine with 16 CPU cores, I observed poor
performance. I noticed that dmu_tx_hold_zap() was using about 30% of
all CPU, and doing dnode_hold() 7 times on the same object (the ZAP
object that is being held).
dmu_tx_hold_zap() keeps a hold on the dnode_t the entire time it is
running, in dmu_tx_hold_t:txh_dnode, so it would be nice to use the
dnode_t that we already have in hand, rather than repeatedly calling
dnode_hold(). To do this, we need to pass the dnode_t down through
all the intermediate calls that dmu_tx_hold_zap() makes, making these
routines take the dnode_t* rather than an objset_t* and a uint64_t
object number. In particular, the following routines will need to have
analogous *_by_dnode() variants created:
dmu_buf_hold_noread()
dmu_buf_hold()
zap_lookup()
zap_lookup_norm()
zap_count_write()
zap_lockdir()
zap_count_write()
This can improve performance on the benchmark described above by 100%,
from 30,000 file creations per second to 60,000. (This improvement is on
top of that provided by working around the object allocation issue. Peak
performance of ~90,000 creations per second was observed with 8 CPUs;
adding CPUs past that decreased performance due to lock contention.) The
CPU used by dmu_tx_hold_zap() was reduced by 88%, from 340 CPU-seconds
to 40 CPU-seconds.
Sponsored by: Intel Corp.
Signed-off-by: Matthew Ahrens <[email protected]>
Signed-off-by: Ned Bass <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
OpenZFS-issue: https://www.illumos.org/issues/7004
OpenZFS-commit: https://github.com/openzfs/openzfs/pull/109
Closes #4641
Closes #4972
Diffstat (limited to 'module/zfs/dmu_tx.c')
| -rw-r--r-- | module/zfs/dmu_tx.c | 7 |
1 files changed, 3 insertions, 4 deletions
diff --git a/module/zfs/dmu_tx.c b/module/zfs/dmu_tx.c index ed29bfbc6..c50f732a7 100644 --- a/module/zfs/dmu_tx.c +++ b/module/zfs/dmu_tx.c @@ -21,7 +21,7 @@ /* * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved. * Copyright 2011 Nexenta Systems, Inc. All rights reserved. - * Copyright (c) 2012, 2015 by Delphix. All rights reserved. + * Copyright (c) 2012, 2016 by Delphix. All rights reserved. */ #include <sys/dmu.h> @@ -794,15 +794,14 @@ dmu_tx_hold_zap(dmu_tx_t *tx, uint64_t object, int add, const char *name) * access the name in this fat-zap so that we'll check * for i/o errors to the leaf blocks, etc. */ - err = zap_lookup(dn->dn_objset, dn->dn_object, name, - 8, 0, NULL); + err = zap_lookup_by_dnode(dn, name, 8, 0, NULL); if (err == EIO) { tx->tx_err = err; return; } } - err = zap_count_write(dn->dn_objset, dn->dn_object, name, add, + err = zap_count_write_by_dnode(dn, name, add, &txh->txh_space_towrite, &txh->txh_space_tooverwrite); /* |