jbd2: use J_ASSERT instead of BUG_ON for checkpoint mutex

[PATCH v1] jbd2: use J_ASSERT instead of BUG_ON for checkpoint mutex

Posted by Baolin Liu 2 months, 3 weeks ago

From: Baolin Liu <liubaolin@kylinos.cn>

Replace BUG_ON with J_ASSERT to write the assertion in a positive form.

No functional change.

Signed-off-by: Baolin Liu <liubaolin@kylinos.cn>
---
 fs/jbd2/journal.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index d480b94117cd..c858ade1f7f3 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -1058,7 +1058,7 @@ int __jbd2_update_log_tail(journal_t *journal, tid_t tid, unsigned long block)
 	unsigned long freed;
 	int ret;
 
-	BUG_ON(!mutex_is_locked(&journal->j_checkpoint_mutex));
+	J_ASSERT(mutex_is_locked(&journal->j_checkpoint_mutex));
 
 	/*
 	 * We cannot afford for write to remain in drive's caches since as
@@ -1863,7 +1863,7 @@ int jbd2_journal_update_sb_log_tail(journal_t *journal, tid_t tail_tid,
 		return -EIO;
 	}
 
-	BUG_ON(!mutex_is_locked(&journal->j_checkpoint_mutex));
+	J_ASSERT(mutex_is_locked(&journal->j_checkpoint_mutex));
 	jbd2_debug(1, "JBD2: updating superblock (start %lu, seq %u)\n",
 		  tail_block, tail_tid);
 
@@ -1897,7 +1897,7 @@ static void jbd2_mark_journal_empty(journal_t *journal, blk_opf_t write_flags)
 	journal_superblock_t *sb = journal->j_superblock;
 	bool had_fast_commit = false;
 
-	BUG_ON(!mutex_is_locked(&journal->j_checkpoint_mutex));
+	J_ASSERT(mutex_is_locked(&journal->j_checkpoint_mutex));
 	lock_buffer(journal->j_sb_buffer);
 	if (sb->s_start == 0) {		/* Is it already empty? */
 		unlock_buffer(journal->j_sb_buffer);
-- 
2.39.2

Re: [PATCH v1] jbd2: use J_ASSERT instead of BUG_ON for checkpoint mutex

Posted by Jan Kara 2 months, 3 weeks ago

Hello!

On Wed 12-11-25 18:29:14, Baolin Liu wrote:
> From: Baolin Liu <liubaolin@kylinos.cn>
> 
> Replace BUG_ON with J_ASSERT to write the assertion in a positive form.
> 
> No functional change.
> 
> Signed-off-by: Baolin Liu <liubaolin@kylinos.cn>

Thanks for the patch I don't think this is really useful. It is IMO just
a pointless churn... If something, we could get rid of J_ASSERT,
J_ASSERT_JH and J_ASSERT_BH macros which are mostly pointless obfuscation
but even that will cause newer patches to not apply to later stable kernels
and thus I don't think it's really worth it.

								Honza

> ---
>  fs/jbd2/journal.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
> index d480b94117cd..c858ade1f7f3 100644
> --- a/fs/jbd2/journal.c
> +++ b/fs/jbd2/journal.c
> @@ -1058,7 +1058,7 @@ int __jbd2_update_log_tail(journal_t *journal, tid_t tid, unsigned long block)
>  	unsigned long freed;
>  	int ret;
>  
> -	BUG_ON(!mutex_is_locked(&journal->j_checkpoint_mutex));
> +	J_ASSERT(mutex_is_locked(&journal->j_checkpoint_mutex));
>  
>  	/*
>  	 * We cannot afford for write to remain in drive's caches since as
> @@ -1863,7 +1863,7 @@ int jbd2_journal_update_sb_log_tail(journal_t *journal, tid_t tail_tid,
>  		return -EIO;
>  	}
>  
> -	BUG_ON(!mutex_is_locked(&journal->j_checkpoint_mutex));
> +	J_ASSERT(mutex_is_locked(&journal->j_checkpoint_mutex));
>  	jbd2_debug(1, "JBD2: updating superblock (start %lu, seq %u)\n",
>  		  tail_block, tail_tid);
>  
> @@ -1897,7 +1897,7 @@ static void jbd2_mark_journal_empty(journal_t *journal, blk_opf_t write_flags)
>  	journal_superblock_t *sb = journal->j_superblock;
>  	bool had_fast_commit = false;
>  
> -	BUG_ON(!mutex_is_locked(&journal->j_checkpoint_mutex));
> +	J_ASSERT(mutex_is_locked(&journal->j_checkpoint_mutex));
>  	lock_buffer(journal->j_sb_buffer);
>  	if (sb->s_start == 0) {		/* Is it already empty? */
>  		unlock_buffer(journal->j_sb_buffer);
> -- 
> 2.39.2
> 
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR

Re: [PATCH v1] jbd2: use J_ASSERT instead of BUG_ON for checkpoint mutex

Posted by liubaolin 2 months, 3 weeks ago

> Sorry, my intention was to unify all the BUG_ON statements in the jbd2 code to J_ASSERT in order to improve readability. 
> You made a good point — this change could indeed cause conflicts with future patches. 
> I’ll withdraw this patch and not submit it. 
> Thank you for your review.



在 2025/11/13 3:29, Jan Kara 写道:
> J_ASSERT,
> J_ASSERT_JH and J_ASSERT_BH