diff options
| author | Brian Foster <bfoster@redhat.com> | 2024-03-01 12:49:24 -0500 |
|---|---|---|
| committer | Kent Overstreet <kent.overstreet@linux.dev> | 2024-03-13 21:22:26 -0400 |
| commit | ada02c207c0168e0e9173983d00f82c9b5de90c8 (patch) | |
| tree | 473b55466587265a9dd86676a6e9a9ca96045228 | |
| parent | 2a68d611a17b8135ade5f7016144ef34af0a61a4 (diff) | |
| download | vfs-ada02c207c0168e0e9173983d00f82c9b5de90c8.tar.gz | |
bcachefs: fix lost journal buf wakeup due to improved pipelining
The journal_write_done() handler was reworked into a loop in commit
746a33c96b7a ("bcachefs: better journal pipelining"). As part of this,
the journal buffer wake was factored into a post-loop branch that
executes if at least one journal buffer has completed.
The journal buffer processing loop iterates on the journal buffer
pointer, however. This means that w refers to the last buffer processed
by the loop, which may or may not be done. This also means that if
multiple buffers are processed by the loop, only the last is awoken.
This lost wakeup behavior has lead to stalling problems in various CI
and fstests, such as generic/703.
Lift the wake into the loop so each done buffer sees a wake call as
it is processed.
Signed-off-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
| -rw-r--r-- | fs/bcachefs/journal_io.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/fs/bcachefs/journal_io.c b/fs/bcachefs/journal_io.c index 0b49b2383d825..8ded059591c7b 100644 --- a/fs/bcachefs/journal_io.c +++ b/fs/bcachefs/journal_io.c @@ -1672,6 +1672,7 @@ static CLOSURE_CALLBACK(journal_write_done) new.unwritten_idx++; } while ((v = atomic64_cmpxchg(&j->reservations.counter, old.v, new.v)) != old.v); + closure_wake_up(&w->wait); completed = true; } @@ -1682,7 +1683,6 @@ static CLOSURE_CALLBACK(journal_write_done) track_event_change(&c->times[BCH_TIME_blocked_journal_max_in_flight], &j->max_in_flight_start, false); - closure_wake_up(&w->wait); journal_wake(j); } |