diff options
| author |  Kent Overstreet <kent.overstreet@gmail.com>
| 2021-02-10 16:13:57 -0500 |
|---|---|---|
| committer |  Kent Overstreet <kent.overstreet@linux.dev>
| 2023-10-22 17:08:53 -0400 |
| commit | 8042b5b715e6722fb26e40724b87f93b4b777acf (patch) | |
| tree | 5758d125a3fd04085b0dc7c5f24b2def52f8dce0 /fs/bcachefs/super-io.c | |
| parent | bcachefs: iter->real_pos (diff) | |
| download | linux-8042b5b715e6722fb26e40724b87f93b4b777acf.tar.gz linux-8042b5b715e6722fb26e40724b87f93b4b777acf.tar.bz2 linux-8042b5b715e6722fb26e40724b87f93b4b777acf.zip | |
bcachefs: Extents may now cross btree node boundaries
When snapshots arrive, we won't necessarily be able to arbitrarily split
existis - when we need to split an existing extent, we'll have to check
if the extent was overwritten in child snapshots and if so emit a
whiteout for the split in the child snapshot.
Because extents couldn't span btree nodes previously, journal replay
would sometimes have to split existing extents. That's no good anymore,
but fortunately since extent handling has already been lifted above most
of the btree code there's no real need for that rule anymore.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
Diffstat (limited to 'fs/bcachefs/super-io.c')
| -rw-r--r-- | fs/bcachefs/super-io.c | 4 |
1 files changed, 1 insertions, 3 deletions
diff --git a/fs/bcachefs/super-io.c b/fs/bcachefs/super-io.c index 767baab18807..79d03b18b5c8 100644 --- a/fs/bcachefs/super-io.c +++ b/fs/bcachefs/super-io.c @@ -956,9 +956,7 @@ int bch2_fs_mark_dirty(struct bch_fs *c) mutex_lock(&c->sb_lock); SET_BCH_SB_CLEAN(c->disk_sb.sb, false); - c->disk_sb.sb->features[0] |= 1ULL << BCH_FEATURE_new_extent_overwrite; - c->disk_sb.sb->features[0] |= 1ULL << BCH_FEATURE_extents_above_btree_updates; - c->disk_sb.sb->features[0] |= 1ULL << BCH_FEATURE_btree_updates_journalled; + c->disk_sb.sb->features[0] |= BCH_SB_FEATURES_ALWAYS; ret = bch2_write_super(c); mutex_unlock(&c->sb_lock); |