Re: [RFC PATCH v5 03/19] fscrypt: export fscrypt_fname_encrypt and fscrypt_fname_encrypted_size

[Eric Biggers <ebiggers@kernel.org>]

On Fri, Mar 26, 2021 at 01:32:11PM -0400, Jeff Layton wrote:
> For ceph, we want to use our own scheme for handling filenames that are
> are longer than NAME_MAX after encryption and base64 encoding. This
> allows us to have a consistent view of the encrypted filenames for
> clients that don't support fscrypt and clients that do but that don't
> have the key.
> 
> Export fscrypt_fname_encrypt. Rename fscrypt_fname_encrypted_size to
> __fscrypt_fname_encrypted_size and add a new wrapper called
> fscrypt_fname_encrypted_size that takes an inode argument rahter than
> a pointer to a fscrypt_policy union.

This explanation seems to be missing a logical connection between the first and
second paragraphs.  I think it's missing something along the lines of:
"Currently, fs/crypto/ only supports filenames encryption using
fscrypt_setup_filename(), which also handles decoding no-key names.  Ceph can't
use that because it needs to handle no-key names in a different way.  So, we
need to export the functions needed to encrypt filenames separately."

(I might have gotten the explanation a bit wrong... Point is, it's the type of
thing that seems to be missing here.)

- Eric