diff options
| author | Joern Engel <joern@logfs.org> | 2010-04-13 19:46:37 +0400 |
|---|---|---|
| committer | Joern Engel <joern@logfs.org> | 2010-04-13 19:46:37 +0400 |
| commit | 032d8f7268444a0f5d4ee02d9513d682d5b8edfc (patch) | |
| tree | 57cd841514abb9ffe7df7d2569513663f551f960 /fs/logfs/journal.c | |
| parent | e05c378f4973674a16d5b9636f2310cf88aca5f2 (diff) | |
| download | linux-032d8f7268444a0f5d4ee02d9513d682d5b8edfc.tar.xz | |
[LogFS] Prevent memory corruption on large deletes
Removing sufficiently large files would create aliases for a large
number of segments. This in turn results in a large number of journal
entries and an overflow of s_je_array.
Cheap fix is to add a BUG_ON, turning memory corruption into something
annoying, but less dangerous. Real fix is to count the number of
affected segments and prevent the problem completely.
Signed-off-by: Joern Engel <joern@logfs.org>
Diffstat (limited to 'fs/logfs/journal.c')
| -rw-r--r-- | fs/logfs/journal.c | 3 |
1 files changed, 3 insertions, 0 deletions
diff --git a/fs/logfs/journal.c b/fs/logfs/journal.c index d57c7b07b60b..2c22a4ad5329 100644 --- a/fs/logfs/journal.c +++ b/fs/logfs/journal.c @@ -493,6 +493,8 @@ static void account_shadows(struct super_block *sb) btree_grim_visitor64(&tree->new, (unsigned long)sb, account_shadow); btree_grim_visitor64(&tree->old, (unsigned long)sb, account_shadow); + btree_grim_visitor32(&tree->segment_map, 0, NULL); + tree->no_shadowed_segments = 0; if (li->li_block) { /* @@ -660,6 +662,7 @@ static int logfs_write_je_buf(struct super_block *sb, void *buf, u16 type, if (ofs < 0) return ofs; logfs_buf_write(area, ofs, super->s_compressed_je, len); + BUG_ON(super->s_no_je >= MAX_JOURNAL_ENTRIES); super->s_je_array[super->s_no_je++] = cpu_to_be64(ofs); return 0; } |