Re: [RFC][PATCH 03/12] digest_lists: Basic definitions

[Greg KH]

On Fri, Jun 25, 2021 at 06:56:05PM +0200, Roberto Sassu wrote:
> --- /dev/null
> +++ b/include/uapi/linux/digest_lists.h
> @@ -0,0 +1,43 @@
> +/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
> +/*
> + * Copyright (C) 2017-2021 Huawei Technologies Duesseldorf GmbH
> + *
> + * Author: Roberto Sassu <roberto.sassu@xxxxxxxxxx>
> + *
> + * This program is free software; you can redistribute it and/or
> + * modify it under the terms of the GNU General Public License as
> + * published by the Free Software Foundation, version 2 of the
> + * License.

As you already have the SPDX line up there, you do not need this
paragraph.  Please remove it from all of the new files you have added in
this series.

> + *
> + * File: digest_lists.h

We know the filename, no need to have it here again.

> + *      Digest list definitions exported to user space.

Now this is what probably needs more information...

> + */
> +
> +#ifndef _UAPI__LINUX_DIGEST_LISTS_H
> +#define _UAPI__LINUX_DIGEST_LISTS_H
> +
> +#include <linux/types.h>
> +#include <linux/hash_info.h>
> +
> +enum compact_types { COMPACT_KEY, COMPACT_PARSER, COMPACT_FILE,
> +		     COMPACT_METADATA, COMPACT_DIGEST_LIST, COMPACT__LAST };
> +
> +enum compact_modifiers { COMPACT_MOD_IMMUTABLE, COMPACT_MOD__LAST };
> +
> +enum compact_actions { COMPACT_ACTION_IMA_MEASURED,
> +		       COMPACT_ACTION_IMA_APPRAISED,
> +		       COMPACT_ACTION_IMA_APPRAISED_DIGSIG,
> +		       COMPACT_ACTION__LAST };
> +
> +enum ops { DIGEST_LIST_ADD, DIGEST_LIST_DEL, DIGEST_LIST_OP__LAST };
> +

For enums you export to userspace, you need to specify the values so
that all compilers get them right.

> +struct compact_list_hdr {
> +	__u8 version;

You should never need a version, that way lies madness.

> +	__u8 _reserved;

You better be testing this for 0, right?

> +	__le16 type;
> +	__le16 modifiers;
> +	__le16 algo;
> +	__le32 count;
> +	__le32 datalen;

Why are user/kernel apis specified in little endian format?  Why would
that matter?  Shouldn't they just be "native" endian?

thanks,

greg k-h