[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20b9015d-80ed-4864-811a-ff4b0e3774dd@huawei.com>
Date: Fri, 12 Jul 2024 18:04:53 +0800
From: "wangjianjian (C)" <wangjianjian3@...wei.com>
To: Luis Henriques <luis.henriques@...ux.dev>, Andreas Dilger
<adilger@...ger.ca>
CC: Wang Jianjian <wangjianjian0@...mail.com>, Theodore Ts'o <tytso@....edu>,
Jan Kara <jack@...e.cz>, Harshad Shirwadkar <harshadshirwadkar@...il.com>,
<linux-ext4@...r.kernel.org>, <linux-kernel@...r.kernel.org>
Subject: Re: [RFC PATCH] jbd2: make '0' an invalid transaction sequence
On 2024/7/12 17:53, Luis Henriques wrote:
> Since there's code (in fast-commit) that already handles a '0' tid as a
> special case, it's better to ensure that jbd2 never sets it to that value
> when journal->j_transaction_sequence increment wraps.
>
> Suggested-by: Andreas Dilger <adilger@...ger.ca>
> Signed-off-by: Luis Henriques (SUSE) <luis.henriques@...ux.dev>
> ---
> fs/jbd2/transaction.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c
> index 66513c18ca29..4dbdd37349c3 100644
> --- a/fs/jbd2/transaction.c
> +++ b/fs/jbd2/transaction.c
> @@ -84,6 +84,8 @@ static void jbd2_get_transaction(journal_t *journal,
> transaction->t_state = T_RUNNING;
> transaction->t_start_time = ktime_get();
> transaction->t_tid = journal->j_transaction_sequence++;
> + if (unlikely(transaction->t_tid == 0))
> + transaction->t_tid = journal->j_transaction_sequence++;
Do we need add j_transaction_sequence again here? if tansanction->t_tid
== 0, then journal->j_trnasaction_sequence must be 1
> transaction->t_expires = jiffies + journal->j_commit_interval;
> atomic_set(&transaction->t_updates, 0);
> atomic_set(&transaction->t_outstanding_credits,
--
Regards
Powered by blists - more mailing lists