diff options
| author | Kent Overstreet <kent.overstreet@gmail.com> | 2021-04-27 14:03:13 -0400 |
|---|---|---|
| committer | Kent Overstreet <kent.overstreet@linux.dev> | 2023-10-22 17:09:02 -0400 |
| commit | d4b4422345fcb4e284260bd52166e189c137e846 (patch) | |
| tree | d21236145694b6485a0aa3610cc3ea6758a8057e /fs/bcachefs/movinggc.c | |
| parent | baa6502905df0acb94afbf6c93f51a1f2aa4c7ca (diff) | |
| download | lwn-d4b4422345fcb4e284260bd52166e189c137e846.tar.gz lwn-d4b4422345fcb4e284260bd52166e189c137e846.zip | |
bcachefs: Change copygc wait amount to be min of per device waits
We're seeing a filesystem get stuck when all devices but one have no
more reclaimable buckets - because the copygc wait amount is curretly
filesystem wide.
This patch should fix that, possibly at the expensive of running too
much when only one or a few devices is full and the rebalance thread
needs to move data around.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
Diffstat (limited to 'fs/bcachefs/movinggc.c')
| -rw-r--r-- | fs/bcachefs/movinggc.c | 10 |
1 files changed, 6 insertions, 4 deletions
diff --git a/fs/bcachefs/movinggc.c b/fs/bcachefs/movinggc.c index f9146ccd70ef..acb4d943db79 100644 --- a/fs/bcachefs/movinggc.c +++ b/fs/bcachefs/movinggc.c @@ -293,17 +293,19 @@ unsigned long bch2_copygc_wait_amount(struct bch_fs *c) { struct bch_dev *ca; unsigned dev_idx; - u64 fragmented_allowed = 0, fragmented = 0; + s64 wait = S64_MAX, fragmented_allowed, fragmented; for_each_rw_member(ca, c, dev_idx) { struct bch_dev_usage usage = bch2_dev_usage_read(ca); - fragmented_allowed += ((__dev_buckets_reclaimable(ca, usage) * + fragmented_allowed = ((__dev_buckets_reclaimable(ca, usage) * ca->mi.bucket_size) >> 1); - fragmented += usage.d[BCH_DATA_user].fragmented; + fragmented = usage.d[BCH_DATA_user].fragmented; + + wait = min(wait, max(0LL, fragmented_allowed - fragmented)); } - return max_t(s64, 0, fragmented_allowed - fragmented); + return wait; } static int bch2_copygc_thread(void *arg) |