Re: [PATCH 1/6] udf: Replace BKL with superblock's mutex s_lock

[Arnd Bergmann]

On Saturday 23 October 2010, Alessio Igor Bogani wrote:
> This work was supported by a hardware donation from the CE Linux Forum.
> 
> Signed-off-by: Alessio Igor Bogani <abogani@xxxxxxxxxx>

Hi Alessio,

Thanks a lot for joining in on the effort. Your patches look good for the
most part, but there are a few details you should change before they
go upstream. Most importantly, the subject lines of patches are currently
not unique and they tell very little about what is going on. You can
probably merge the three "remove BKL" patches into one and explain
why that is safe in a combined changelog.

> @@ -568,7 +568,7 @@ static int udf_remount_fs(struct super_block *sb, int *flags, char *options)
>  	if (!udf_parse_options(options, &uopt, true))
>  		return -EINVAL;
>  
> -	lock_kernel();
> +	mutex_lock(&sb->s_lock);
>  	sbi->s_flags = uopt.flags;
>  	sbi->s_uid   = uopt.uid;
>  	sbi->s_gid   = uopt.gid;

As I recently learned, there is also an effort to get rid of sb->s_lock, so
the change in super.c is not all that helpful. When you need a per-filesystem
mutex, please instead add a new one to udf_sb_info. There is already the
s_alloc_mutex. While I haven't looked at that, maybe it can be extended to
cover the other areas that are under s_lock with your patch. Watch out for
put_super freeing sb_info under the lock though.

	Arnd
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/