diff options
| author | Kent Overstreet <kent.overstreet@gmail.com> | 2021-01-21 15:28:59 -0500 |
|---|---|---|
| committer | Kent Overstreet <kent.overstreet@linux.dev> | 2023-10-22 17:08:52 -0400 |
| commit | 2abe542087d9cb1bc7bb8ac7ae262afccbdb7aa6 (patch) | |
| tree | c33f5733a5cf3f091e59e5084be29ef5819a9f71 /fs/bcachefs/journal.c | |
| parent | 7f4e1d5d0faff0d72e9f6708bf98488d76533846 (diff) | |
bcachefs: Persist 64 bit io clocks
Originally, bcachefs - going back to bcache - stored, for each bucket, a
16 bit counter corresponding to how long it had been since the bucket
was read from. But, this required periodically rescaling counters on
every bucket to avoid wraparound. That wasn't an issue in bcache, where
we'd perodically rewrite the per bucket metadata all at once, but in
bcachefs we're trying to avoid having to walk every single bucket.
This patch switches to persisting 64 bit io clocks, corresponding to the
64 bit bucket timestaps introduced in the previous patch with
KEY_TYPE_alloc_v2.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
Diffstat (limited to 'fs/bcachefs/journal.c')
| -rw-r--r-- | fs/bcachefs/journal.c | 3 |
1 files changed, 3 insertions, 0 deletions
diff --git a/fs/bcachefs/journal.c b/fs/bcachefs/journal.c index ba37c78c01db..379b9ad2c0f9 100644 --- a/fs/bcachefs/journal.c +++ b/fs/bcachefs/journal.c @@ -1123,6 +1123,9 @@ int bch2_fs_journal_init(struct journal *j) j->entry_u64s_reserved += BTREE_ID_NR * (JSET_KEYS_U64s + BKEY_BTREE_PTR_U64s_MAX); + j->entry_u64s_reserved += + 2 * (sizeof(struct jset_entry_clock) / sizeof(u64)); + atomic64_set(&j->reservations.counter, ((union journal_res_state) { .cur_entry_offset = JOURNAL_ENTRY_CLOSED_VAL }).v); |