Speed up fsck by caching the cylinder group maps in pass1 so

that they do not need to be read again in pass5. As this nearly
doubles the memory requirement for fsck, the cache is thrown away
if other memory needs in fsck would otherwise fail. Thus, the
memory footprint of fsck remains unchanged in memory constrained
environments.

This work was inspired by a paper presented at Usenix's FAST '13:
www.usenix.org/conference/fast13/ffsck-fast-file-system-checker

Details of this implementation appears in the April 2013 of ;login:
www.usenix.org/publications/login/april-2013-volume-38-number-2.
A copy of the April 2013 ;login: paper can also be downloaded
from: www.mckusick.com/publications/faster_fsck.pdf.

Reviewed by: kib
Tested by:   Peter Holm
MFC after:   4 weeks

1 files changed, 2 insertions, 2 deletions

diff --git a/sbin/fsck_ffs/suj.c b/sbin/fsck_ffs/suj.c
index ae47487..5fca0f5 100644
--- a/sbin/fsck_ffs/suj.c
+++ b/sbin/fsck_ffs/suj.c
@@ -161,7 +161,7 @@ errmalloc(size_t n)
 {
 	void *a;
 
-	a = malloc(n);
+	a = Malloc(n);
 	if (a == NULL)
 		err(EX_OSERR, "malloc(%zu)", n);
 	return (a);
@@ -194,7 +194,7 @@ opendisk(const char *devnam)
 {
 	if (disk != NULL)
 		return;
-	disk = malloc(sizeof(*disk));
+	disk = Malloc(sizeof(*disk));
 	if (disk == NULL)
 		err(EX_OSERR, "malloc(%zu)", sizeof(*disk));
 	if (ufs_disk_fillout(disk, devnam) == -1) {