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

1 files changed, 43 insertions, 0 deletions

diff --git a/module/zfs/dmu.c b/module/zfs/dmu.c
index e1dfb41ff..6ed4743ee 100644
--- a/module/zfs/dmu.c
+++ b/module/zfs/dmu.c
@@ -128,6 +128,26 @@ const dmu_object_byteswap_info_t dmu_ot_byteswap[DMU_BSWAP_NUMFUNCS] = {
 };
 
 int
+dmu_buf_hold_noread_by_dnode(dnode_t *dn, uint64_t offset,
+    void *tag, dmu_buf_t **dbp)
+{
+	uint64_t blkid;
+	dmu_buf_impl_t *db;
+
+	blkid = dbuf_whichblock(dn, 0, offset);
+	rw_enter(&dn->dn_struct_rwlock, RW_READER);
+	db = dbuf_hold(dn, blkid, tag);
+	rw_exit(&dn->dn_struct_rwlock);
+
+	if (db == NULL) {
+		*dbp = NULL;
+		return (SET_ERROR(EIO));
+	}
+
+	*dbp = &db->db;
+	return (0);
+}
+int
 dmu_buf_hold_noread(objset_t *os, uint64_t object, uint64_t offset,
     void *tag, dmu_buf_t **dbp)
 {
@@ -155,6 +175,29 @@ dmu_buf_hold_noread(objset_t *os, uint64_t object, uint64_t offset,
 }
 
 int
+dmu_buf_hold_by_dnode(dnode_t *dn, uint64_t offset,
+    void *tag, dmu_buf_t **dbp, int flags)
+{
+	int err;
+	int db_flags = DB_RF_CANFAIL;
+
+	if (flags & DMU_READ_NO_PREFETCH)
+		db_flags |= DB_RF_NOPREFETCH;
+
+	err = dmu_buf_hold_noread_by_dnode(dn, offset, tag, dbp);
+	if (err == 0) {
+		dmu_buf_impl_t *db = (dmu_buf_impl_t *)(*dbp);
+		err = dbuf_read(db, NULL, db_flags);
+		if (err != 0) {
+			dbuf_rele(db, tag);
+			*dbp = NULL;
+		}
+	}
+
+	return (err);
+}
+
+int
 dmu_buf_hold(objset_t *os, uint64_t object, uint64_t offset,
     void *tag, dmu_buf_t **dbp, int flags)
 {