diff options
| author | Kent Overstreet <kent.overstreet@gmail.com> | 2021-08-30 01:34:37 +0200 |
|---|---|---|
| committer | Kent Overstreet <kent.overstreet@linux.dev> | 2023-10-22 23:09:11 +0200 |
| commit | 5f8077cca89bdcc3d7660567462b17831356826d (patch) | |
| tree | 20291e4a1674b0a38be2ff525bc66f16f4389597 /fs/bcachefs/move.c | |
| parent | bcachefs: Refactor bch2_trans_update_extent() (diff) | |
| download | linux-5f8077cca89bdcc3d7660567462b17831356826d.tar.xz linux-5f8077cca89bdcc3d7660567462b17831356826d.zip | |
bcachefs: Kill BTREE_ITER_SET_POS_AFTER_COMMIT
BTREE_ITER_SET_POS_AFTER_COMMIT is used internally to automagically
advance extent btree iterators on sucessful commit.
But with the upcomnig btree_path patch it's getting more awkward to
support, and it adds overhead to core data structures that's only used
in a few places, and can be easily done by the caller instead.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Diffstat (limited to '')
| -rw-r--r-- | fs/bcachefs/move.c | 9 |
1 files changed, 7 insertions, 2 deletions
diff --git a/fs/bcachefs/move.c b/fs/bcachefs/move.c index 8d28d8fc5395..c804af8b81de 100644 --- a/fs/bcachefs/move.c +++ b/fs/bcachefs/move.c @@ -79,6 +79,7 @@ int bch2_migrate_index_update(struct bch_write_op *op) struct bkey_i_extent *new; const union bch_extent_entry *entry; struct extent_ptr_decoded p; + struct bpos next_pos; bool did_work = false; bool extending = false, should_check_enospc; s64 i_sectors_delta = 0, disk_sectors_delta = 0; @@ -162,14 +163,18 @@ int bch2_migrate_index_update(struct bch_write_op *op) goto out; } + next_pos = insert->k.p; + ret = bch2_trans_update(&trans, iter, insert, 0) ?: bch2_trans_commit(&trans, &op->res, op_journal_seq(op), BTREE_INSERT_NOFAIL| m->data_opts.btree_insert_flags); -err: - if (!ret) + if (!ret) { + bch2_btree_iter_set_pos(iter, next_pos); atomic_long_inc(&c->extent_migrate_done); + } +err: if (ret == -EINTR) ret = 0; if (ret) |