OpenZFS 9337 - zfs get all is slow due to uncached metadata

This project's goal is to make read-heavy channel programs and zfs(1m)
administrative commands faster by caching all the metadata that they will
need in the dbuf layer. This will prevent the data from being evicted, so
that any future call to i.e. zfs get all won't have to go to disk (very
much). There are two parts:

The dbuf_metadata_cache. We identify what to put into the cache based on
the object type of each dbuf.  Caching objset properties os
{version,normalization,utf8only,casesensitivity} in the objset_t. The reason
these needed to be cached is that although they are queried frequently,
they aren't stored in a dbuf type which we can easily recognize and cache in
the dbuf layer; instead, we have to explicitly store them. There's already
existing infrastructure for maintaining cached properties in the objset
setup code, so I simply used that.

Performance Testing:

 - Disabled kmem_flags
 - Tuned dbuf_cache_max_bytes very low (128K)
 - Tuned zfs_arc_max very low (64M)

Created test pool with 400 filesystems, and 100 snapshots per filesystem.
Later on in testing, added 600 more filesystems (with no snapshots) to make
sure scaling didn't look different between snapshots and filesystems.

Results:

    | Test                   | Time (trunk / diff) | I/Os (trunk / diff) |
    +------------------------+---------------------+---------------------+
    | zpool import           |     0:05 / 0:06     |    12.9k / 12.9k    |
    | zfs get all (uncached) |     1:36 / 0:53     |    16.7k / 5.7k     |
    | zfs get all (cached)   |     1:36 / 0:51     |    16.0k / 6.0k     |

Authored by: Matthew Ahrens <[email protected]>
Reviewed by: Prakash Surya <[email protected]>
Reviewed by: George Wilson <[email protected]>
Reviewed by: Thomas Caputi <[email protected]>
Reviewed by: Brian Behlendorf <[email protected]>
Approved by: Richard Lowe <[email protected]>
Ported-by: Alek Pinchuk <[email protected]>
Signed-off-by: Alek Pinchuk <[email protected]>

OpenZFS-issue: https://illumos.org/issues/9337
OpenZFS-commit: https://github.com/openzfs/openzfs/commit/7dec52f
Closes #7668

1 files changed, 12 insertions, 4 deletions

diff --git a/include/sys/dbuf.h b/include/sys/dbuf.h
index 16dc61e7f..557d2af66 100644
--- a/include/sys/dbuf.h
+++ b/include/sys/dbuf.h
@@ -84,6 +84,13 @@ typedef enum dbuf_states {
 	DB_EVICTING
 } dbuf_states_t;
 
+typedef enum dbuf_cached_state {
+	DB_NO_CACHE = -1,
+	DB_DBUF_CACHE,
+	DB_DBUF_METADATA_CACHE,
+	DB_CACHE_MAX
+} dbuf_cached_state_t;
+
 struct dnode;
 struct dmu_tx;
 
@@ -240,11 +247,12 @@ typedef struct dmu_buf_impl {
 	 */
 	avl_node_t db_link;
 
-	/*
-	 * Link in dbuf_cache.
-	 */
+	/* Link in dbuf_cache or dbuf_metadata_cache */
 	multilist_node_t db_cache_link;
 
+	/* Tells us which dbuf cache this dbuf is in, if any */
+	dbuf_cached_state_t db_caching_status;
+
 	/* Data which is unique to data (leaf) blocks: */
 
 	/* User callback information. */
@@ -305,7 +313,7 @@ boolean_t dbuf_try_add_ref(dmu_buf_t *db, objset_t *os, uint64_t obj,
 uint64_t dbuf_refcount(dmu_buf_impl_t *db);
 
 void dbuf_rele(dmu_buf_impl_t *db, void *tag);
-void dbuf_rele_and_unlock(dmu_buf_impl_t *db, void *tag, boolean_t evicting);
+void dbuf_rele_and_unlock(dmu_buf_impl_t *db, void *tag);
 
 dmu_buf_impl_t *dbuf_find(struct objset *os, uint64_t object, uint8_t level,
     uint64_t blkid);