Re: [PATCH 1/1] hfsplus: fix longname handling

[Andrew Morton <akpm@linux-foundation.org>]

On Mon, 24 Feb 2014 21:28:27 +0200 Sougata Santra <sougata@tuxera.com> wrote:

> 
> -ENAMETOOLONG returned from hfsplus_asc2uni was not propaged to iops. This
> allowed hfsplus to create files/directories with HFSPLUS_MAX_STRLEN and
> incorrect keys, leaving the FS in an inconsistent state. This patch fixes
> this issue.
> 
> ...
>
> --- a/fs/hfsplus/hfsplus_fs.h
> +++ b/fs/hfsplus/hfsplus_fs.h
> @@ -443,8 +443,10 @@ int hfsplus_cat_case_cmp_key(const hfsplus_btree_key *,
>  		const hfsplus_btree_key *);
>  int hfsplus_cat_bin_cmp_key(const hfsplus_btree_key *,
>  		const hfsplus_btree_key *);
> -void hfsplus_cat_build_key(struct super_block *sb,
> +int hfsplus_cat_build_key(struct super_block *sb,
>  		hfsplus_btree_key *, u32, struct qstr *);
> +void hfsplus_cat_build_key_with_cnid(struct super_block *sb,
> +		hfsplus_btree_key *, u32);
>  int hfsplus_find_cat(struct super_block *, u32, struct hfs_find_data *);
>  int hfsplus_create_cat(u32, struct inode *, struct qstr *, struct inode *);
>  int hfsplus_delete_cat(u32, struct inode *, struct qstr *);

grumble.  Omitting the argument names from declarations makes them
unreadable and generally useless.  I mean, a bare u32?

And including the names of some arguments but omitting others is
downright bizarre.

However this isn't a thing which can or should be addressed within this
patch.