Re: [PATCH] reiserfs: kill-the-BKL

[Ingo Molnar <mingo@elte.hu>]

* Frederic Weisbecker <fweisbec@gmail.com> wrote:

> +/*
> + * Utility function to force a BUG if it is called without the superblock
> + * write lock held.  caller is the string printed just before calling BUG()
> + */
> +void reiserfs_check_lock_depth(struct super_block *sb, char *caller)
> +{
> +	struct reiserfs_sb_info *sb_i = REISERFS_SB(sb);
> +
> +	if (sb_i->lock_depth < 0)
> +		reiserfs_panic(sb, "%s called without kernel lock held %d",
> +			       caller);

s/kernel lock/sb write lock/

> +	reiserfs_write_unlock(s);
> +	mutex_destroy(&REISERFS_SB(s)->lock);
>  	kfree(s->s_fs_info);
>  	s->s_fs_info = NULL;

ah, mutex_destroy() - it's a small detail but still nice ;-)

Anyway, it still looks good to me, after the earlier patch i 
reviewed. Would be nice to have the testing feedback of reiserfs 
users, and the locking review from anyone who knows this code.

	Ingo