diff options
| author | Joern Engel <joern@logfs.org> | 2010-03-04 23:30:58 +0300 |
|---|---|---|
| committer | Joern Engel <joern@logfs.org> | 2010-03-04 23:30:58 +0300 |
| commit | 9421502b4fc894cc477be8fc49776830e37ca157 (patch) | |
| tree | 9c9b1bfa42b2acdf4b5e080a256c3cd37852a94f /fs/logfs/segment.c | |
| parent | 5c564c2a04d4bb6ba79eeb83bd06de584479f362 (diff) | |
| download | linux-9421502b4fc894cc477be8fc49776830e37ca157.tar.xz | |
[LogFS] Fix bdev erases
Erases for block devices were always just emulated by writing 0xff.
Some time back the write was removed and only the page cache was
changed to 0xff. Superficialy a good idea with two problems:
1. Touching the page cache isn't necessary either.
2. However, writing out 0xff _is_ necessary for the journal. As the
journal is scanned linearly, an old non-overwritten commit entry
can be used on next mount and cause havoc.
This should fix both aspects.
Diffstat (limited to 'fs/logfs/segment.c')
| -rw-r--r-- | fs/logfs/segment.c | 6 |
1 files changed, 3 insertions, 3 deletions
diff --git a/fs/logfs/segment.c b/fs/logfs/segment.c index 5f58b74516ca..664cd0dd3576 100644 --- a/fs/logfs/segment.c +++ b/fs/logfs/segment.c @@ -25,14 +25,14 @@ static int logfs_mark_segment_bad(struct super_block *sb, u32 segno) return 0; } -int logfs_erase_segment(struct super_block *sb, u32 segno) +int logfs_erase_segment(struct super_block *sb, u32 segno, int ensure_erase) { struct logfs_super *super = logfs_super(sb); super->s_gec++; return super->s_devops->erase(sb, (u64)segno << super->s_segshift, - super->s_segsize); + super->s_segsize, ensure_erase); } static s64 logfs_get_free_bytes(struct logfs_area *area, size_t bytes) @@ -798,7 +798,7 @@ static int ostore_erase_segment(struct logfs_area *area) u64 ofs; int err; - err = logfs_erase_segment(sb, area->a_segno); + err = logfs_erase_segment(sb, area->a_segno, 0); if (err) return err; |