mkfs.xfs: make lazy-count=1 the default

With the xfstests fixed up, we should be able to do this now,
and should add some performance benefits to the default
mkfs options.

Signed-off-by: Eric Sandeen <sandeen@sandeen.net>
Reviewed-by: Christoph Hellwig <hch@lst.de>

diff --git a/man/man8/mkfs.xfs.8 b/man/man8/mkfs.xfs.8
index 40ea1bd707d6fcaf47facf67bcff41087aa2d054..4b147e979fb7bd8576ca527e1140fe028b8da778 100644 (file)
--- a/man/man8/mkfs.xfs.8
+++ b/man/man8/mkfs.xfs.8
@@ -457,9 +457,10 @@ counter values without needed to keep them in the superblock.
 This gives significant improvements in performance on some configurations.
 The default
 .I value
-is 0 (off) so you must specify
-.B lazy-count=1
-if you want to make use of this feature.
+is 1 (on) so you must specify
+.B lazy-count=0
+if you want to disable this feature for older kernels which don't support
+it.
 .RE
 .TP
 .BI \-n " naming_options"

diff --git a/mkfs/xfs_mkfs.c b/mkfs/xfs_mkfs.c
index bd92bc00fdbe6afa68f2ba1c66836c0275278c24..d3ed00ae4232fa22b404711bc99678c8f0e8a1d5 100644 (file)
--- a/mkfs/xfs_mkfs.c
+++ b/mkfs/xfs_mkfs.c
@@ -878,7 +878,7 @@ main(
        nodsflag = norsflag = 0;
        force_overwrite = 0;
        worst_freelist = 0;
-       lazy_sb_counters = 0;
+       lazy_sb_counters = 1;
        memset(&fsx, 0, sizeof(fsx));
 
        memset(&xi, 0, sizeof(xi));