This fixes a bug of ("nilfs2: simplify handling of active state of
segments") patch. The patch did not take account that a base index is
increased in nilfs_sufile_get_suinfo() function if requested entries
go across block boundary on sufile.
Due to this bug, the active flag sometimes appears on wrong segments
and has induced malfunction of garbage collection.
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@xxxxxxxxxxxxx>
---
fs/nilfs2/sufile.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/fs/nilfs2/sufile.c b/fs/nilfs2/sufile.c
index c774cf3..1ef2b4d 100644
--- a/fs/nilfs2/sufile.c
+++ b/fs/nilfs2/sufile.c
@@ -625,7 +625,7 @@ ssize_t nilfs_sufile_get_suinfo(struct inode *sufile, __u64 segnum,
si[i + j].sui_nblocks = le32_to_cpu(su->su_nblocks);
si[i + j].sui_flags = le32_to_cpu(su->su_flags) &
~(1UL << NILFS_SEGMENT_USAGE_ACTIVE);
- if (nilfs_segment_is_active(nilfs, segnum + i + j))
+ if (nilfs_segment_is_active(nilfs, segnum + j))
si[i + j].sui_flags |=
(1UL << NILFS_SEGMENT_USAGE_ACTIVE);
}
--
1.5.6.5
--
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
