提交 914b2007 编写于 作者: J Jan Kara 提交者: David Sterba

btrfs: Fix busyloop in transaction_kthread()

When a filesystem got aborted due do error, transaction_kthread() will
busyloop.  Fix it by going to sleep in that case as well. Maybe we should
just stop transaction_kthread() when filesystem is aborted but that would be
more complex.
Signed-off-by: NJan Kara <jack@suse.cz>
上级 79787eaa
......@@ -1640,8 +1640,10 @@ static int transaction_kthread(void *arg)
u64 transid;
unsigned long now;
unsigned long delay;
bool cannot_commit;
do {
cannot_commit = false;
delay = HZ * 30;
vfs_check_frozen(root->fs_info->sb, SB_FREEZE_WRITE);
mutex_lock(&root->fs_info->transaction_kthread_mutex);
......@@ -1665,8 +1667,10 @@ static int transaction_kthread(void *arg)
/* If the file system is aborted, this will always fail. */
trans = btrfs_join_transaction(root);
if (IS_ERR(trans))
if (IS_ERR(trans)) {
cannot_commit = true;
goto sleep;
}
if (transid == trans->transid) {
btrfs_commit_transaction(trans, root);
} else {
......@@ -1679,7 +1683,8 @@ static int transaction_kthread(void *arg)
if (!try_to_freeze()) {
set_current_state(TASK_INTERRUPTIBLE);
if (!kthread_should_stop() &&
!btrfs_transaction_blocked(root->fs_info))
(!btrfs_transaction_blocked(root->fs_info) ||
cannot_commit))
schedule_timeout(delay);
__set_current_state(TASK_RUNNING);
}
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册