Add module parameter to disable prefetch in zfs_readdir

Add paramter zfs_readdir_dnode_prefetch_limit, defaulting to 0, to
control whether zfs_readdir prefetched metadata for objects it look at
when reading a directory. If zero, metadata is prefetched for all
directory entries. If non-zero, metadata is prefetched only if
directory has fewer entries than this.

Setting it to non-0 can be important for NFS servers with directories
containing many subdirectories.

Signed-off-by: Charles Hedrick <[email protected]>
Co-authored-by: Chris Siebenmann<[email protected]>

man/man4/zfs.4

@@ -1774,6 +1774,24 @@ intact.
 Unlike predictive prefetch, prescient prefetch never issues I/O
 that ends up not being needed, so it can't hurt performance.
 .
+.It Sy zfs_readdir_dnode_prefetch_limit Ns = Ns Sy 0 Pq u64
+Disable prefetches in readdir for large directories.
+When readdir searches a directory, it normally prefetches metadata for
+all objects in the directory it checks, even if it's just
+looking for a single object.
+Setting this to a non-zero value disables that prefetching for directories
+with a greater size than that value.
+Disabling it for large directories can greatly lower CPU usage on NFS servers
+where directories have a very large number of subdirectories.
+Directory size in this case is the size returned from calling
+.Sy stat
+on the directory (stat.st_size).
+On ZFS, this directory size value is approximately the number of files
+and subdirectories in the directory.
+A reasonable value would be 20000.
+A zero value (the default) means no limit on directory metadata prefetching.
+This parameter only applies on Linux.
+.
 .It Sy zfs_qat_checksum_disable Ns = Ns Sy 0 Ns | Ns 1 Pq int
 Disable QAT hardware acceleration for SHA256 checksums.
 May be unset after the ZFS modules have been loaded to initialize the QAT

module/os/linux/zfs/zfs_vnops_os.c

@@ -1504,6 +1504,7 @@ zfs_rmdir(znode_t *dzp, char *name, znode_t *cwd, cred_t *cr,
  * We use 0 for '.', and 1 for '..'.  If this is the root of the filesystem,
  * we use the offset 2 for the '.zfs' directory.
  */
+static ulong_t zfs_readdir_dnode_prefetch_limit = 0UL;
 int
 zfs_readdir(struct inode *ip, zpl_dir_context_t *ctx, cred_t *cr)
 {
@@ -1537,6 +1538,9 @@ zfs_readdir(struct inode *ip, zpl_dir_context_t *ctx, cred_t *cr)
 	os = zfsvfs->z_os;
 	offset = ctx->pos;
 	prefetch = zp->z_zn_prefetch;
+	if (zfs_readdir_dnode_prefetch_limit &&
+	    zp->z_size > zfs_readdir_dnode_prefetch_limit)
+		prefetch = B_FALSE;
 
 	/*
 	 * Initialize the iterator cursor.
@@ -4252,4 +4256,9 @@ EXPORT_SYMBOL(zfs_map);
 /* CSTYLED */
 module_param(zfs_delete_blocks, ulong, 0644);
 MODULE_PARM_DESC(zfs_delete_blocks, "Delete files larger than N blocks async");
+
+/* CSTYLED */
+module_param(zfs_readdir_dnode_prefetch_limit, ulong, 0644);
+MODULE_PARM_DESC(zfs_readdir_dnode_prefetch_limit,
+	"No zfs_readdir prefetch if non-zero and size > this");
 #endif