Re: [PATCH] ext4: fix NULL pointer dereference from orig_data in fill_super and remount.

["Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com>]

On 11/06/2011 08:49 PM, Namjae Jeon wrote:
> Fix NULL pointer dereference from orig_data in fill_super and remount.
> 
> Signed-off-by: Namjae Jeon <linkinjeon@gmail.com>
> ---
>  fs/ext4/super.c |   12 ++++++++++--
>  1 files changed, 10 insertions(+), 2 deletions(-)
> 
> diff --git a/fs/ext4/super.c b/fs/ext4/super.c
> index 9953d80..3770d3f 100644
> --- a/fs/ext4/super.c
> +++ b/fs/ext4/super.c
> @@ -3102,7 +3102,6 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
>  				__releases(kernel_lock)
>  				__acquires(kernel_lock)
>  {
> -	char *orig_data = kstrdup(data, GFP_KERNEL);
>  	struct buffer_head *bh;
>  	struct ext4_super_block *es = NULL;
>  	struct ext4_sb_info *sbi;
> @@ -3124,6 +3123,10 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
>  	int err;
>  	unsigned int journal_ioprio = DEFAULT_JOURNAL_IOPRIO;
>  	ext4_group_t first_not_zeroed;
> +
> +	char *orig_data = kstrdup(data, GFP_KERNEL);
> +	if (!orig_data)
> +		return ret;
> 
>  	sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
>  	if (!sbi)
> @@ -4398,6 +4401,10 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data)
>  	int i;
>  #endif
>  	char *orig_data = kstrdup(data, GFP_KERNEL);
> +	if (!orig_data) {
> +		err = -ENOMEM;
> +		goto failed_alloc_orig;

I didn't quite get why 'failed_alloc_orig' is needed here.
Why can't we just return -ENOMEM? Anyway we haven't yet
clobbered any opts at this point, right?

See my comments below.

> +	}
> 
>  	/* Store the original options */
>  	lock_super(sb);
> @@ -4562,6 +4569,8 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data)
>  	return 0;
> 
>  restore_opts:
> +	kfree(orig_data);
> +failed_alloc_orig:
>  	sb->s_flags = old_sb_flags;
>  	sbi->s_mount_opt = old_opts.s_mount_opt;
>  	sbi->s_mount_opt2 = old_opts.s_mount_opt2;

This would put garbage values in sb->... and sbi->... when we jump
to 'failed_alloc_orig' upon 'orig_data' allocation failure, because
the old_* variables were still uninitialized at that point.

> @@ -4580,7 +4589,6 @@ restore_opts:
>  	}
>  #endif
>  	unlock_super(sb);
> -	kfree(orig_data);
>  	return err;
>  }
> 

Thanks,
Srivatsa S. Bhat