diff mbox series

jbd2: wake up journal waiters in FIFO order, not LIFO

Message ID 20220912180137.2308900-1-alexey.lyashkov@gmail.com
State Awaiting Upstream
Headers show
Series jbd2: wake up journal waiters in FIFO order, not LIFO | expand

Commit Message

Alexey Lyahkov Sept. 12, 2022, 6:01 p.m. UTC 
From: Andrew Perepechko <anserper@ya.ru>

LIFO wakeup order is unfair and sometimes leads to a journal
user not being able to get a journal handle for hundreds of
transactions in a row.

FIFO wakeup can make things more fair.

Reviewed-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
Signed-off-by: Alexey Lyashkov <alexey.lyashkov@gmail.com>
---
 fs/jbd2/commit.c      | 2 +-
 fs/jbd2/transaction.c | 6 +++---
 2 files changed, 4 insertions(+), 4 deletions(-)

Comments

Theodore Ts'o Sept. 30, 2022, 3:19 a.m. UTC | #1 
On Mon, 12 Sep 2022 21:01:37 +0300, Alexey Lyashkov wrote:
> From: Andrew Perepechko <anserper@ya.ru>
> 
> LIFO wakeup order is unfair and sometimes leads to a journal
> user not being able to get a journal handle for hundreds of
> transactions in a row.
> 
> FIFO wakeup can make things more fair.
> 
> [...]

Applied, thanks!

[1/1] jbd2: wake up journal waiters in FIFO order, not LIFO
      commit: 2bf6739025dcfca3a24fbc73b8f2dbd979c0e4c9

Best regards,
diff mbox series

Patch

diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index b2b2bc9b88d9..ec2b55879e3a 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -570,7 +570,7 @@  void jbd2_journal_commit_transaction(journal_t *journal)
 	journal->j_running_transaction = NULL;
 	start_time = ktime_get();
 	commit_transaction->t_log_start = journal->j_head;
-	wake_up(&journal->j_wait_transaction_locked);
+	wake_up_all(&journal->j_wait_transaction_locked);
 	write_unlock(&journal->j_state_lock);
 
 	jbd2_debug(3, "JBD2: commit phase 2a\n");
diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c
index e1be93ccd81c..6a404ac1c178 100644
--- a/fs/jbd2/transaction.c
+++ b/fs/jbd2/transaction.c
@@ -168,7 +168,7 @@  static void wait_transaction_locked(journal_t *journal)
 	int need_to_start;
 	tid_t tid = journal->j_running_transaction->t_tid;
 
-	prepare_to_wait(&journal->j_wait_transaction_locked, &wait,
+	prepare_to_wait_exclusive(&journal->j_wait_transaction_locked, &wait,
 			TASK_UNINTERRUPTIBLE);
 	need_to_start = !tid_geq(journal->j_commit_request, tid);
 	read_unlock(&journal->j_state_lock);
@@ -194,7 +194,7 @@  static void wait_transaction_switching(journal_t *journal)
 		read_unlock(&journal->j_state_lock);
 		return;
 	}
-	prepare_to_wait(&journal->j_wait_transaction_locked, &wait,
+	prepare_to_wait_exclusive(&journal->j_wait_transaction_locked, &wait,
 			TASK_UNINTERRUPTIBLE);
 	read_unlock(&journal->j_state_lock);
 	/*
@@ -920,7 +920,7 @@  void jbd2_journal_unlock_updates (journal_t *journal)
 	write_lock(&journal->j_state_lock);
 	--journal->j_barrier_count;
 	write_unlock(&journal->j_state_lock);
-	wake_up(&journal->j_wait_transaction_locked);
+	wake_up_all(&journal->j_wait_transaction_locked);
 }
 
 static void warn_dirty_buffer(struct buffer_head *bh)