From: "Mickaël Salaün" <mic@digikod.net>
To: "Günther Noack" <gnoack@google.com>
Cc: linux-security-module@vger.kernel.org,
Matt Bobrowski <repnop@google.com>,
Alejandro Colomar <alx@kernel.org>,
Konstantin Meskhidze <konstantin.meskhidze@huawei.com>
Subject: Re: [PATCH] landlock: Clarify documentation for struct landlock_ruleset_attr
Date: Wed, 10 Jul 2024 16:15:47 +0200 [thread overview]
Message-ID: <20240710.te8ceiPhav6e@digikod.net> (raw)
In-Reply-To: <20240710120134.1926158-1-gnoack@google.com>
On Wed, Jul 10, 2024 at 12:01:34PM +0000, Günther Noack wrote:
> The explanation for @handled_access_fs and @handled_access_net has
> significant overlap and is better explained together. I tried to clarify
> the wording and break up longer sentences as well. I am putting emphasis
> on the word "handled" to make it clearer that "handled" is a term with
> special meaning in the context of Landlock.
>
> I'd like to transfer this wording into the man pages as well.
Thanks for working on this.
>
> Signed-off-by: Günther Noack <gnoack@google.com>
> Cc: Alejandro Colomar <alx@kernel.org>
> Cc: Mickaël Salaün <mic@digikod.net>
> Cc: Konstantin Meskhidze <konstantin.meskhidze@huawei.com>
> Cc: linux-security-module@vger.kernel.org
> ---
> include/uapi/linux/landlock.h | 42 ++++++++++++++++++-----------------
> 1 file changed, 22 insertions(+), 20 deletions(-)
>
> diff --git a/include/uapi/linux/landlock.h b/include/uapi/linux/landlock.h
> index 68625e728f43..6f1b05c6995b 100644
> --- a/include/uapi/linux/landlock.h
> +++ b/include/uapi/linux/landlock.h
> @@ -12,30 +12,32 @@
> #include <linux/types.h>
>
> /**
> - * struct landlock_ruleset_attr - Ruleset definition
> + * struct landlock_ruleset_attr - Ruleset definition.
> *
> - * Argument of sys_landlock_create_ruleset(). This structure can grow in
> - * future versions.
> + * @handled_access_fs: Bitmask of handled filesystem actions (cf. `Filesystem flags`_)
> + * @handled_access_net: Bitmask of handled network actions (cf. `Network flags`_)
These @handled_* lines should be kept close the the related fields to
ease maintenance and consistency. It looks like the Sphinx rendering
would be the same.
> + *
> + * Argument of sys_landlock_create_ruleset().
> + *
> + * This struct defines a set of *handled access rights*, a set of actions on
> + * different object types, which should be denied by default when the ruleset is
> + * enacted. Vice versa, access rights that are not specifically listed here are
> + * going to be allowed when the ruleset is enacted.
They could still be denied because of other access controls or parent
Landlock domains.
> + *
> + * One exception is the %LANDLOCK_ACCESS_FS_REFER access right, which is always
> + * implicitly *handled*, even when its bit is not set in @handled_access_fs.
I wrote this sentence but I now think it might be confusing.
LANDLOCK_ACCESS_FS_REFER couldn't be handled before it was introduced
(with Linux 5.19). I couldn't find a better way to explain it though.
> + * However, in order to add new rules with this access right, the bit must still
> + * be set explicitly.
> + *
> + * The explicit listing of *handled access rights* is required for backwards
> + * compatibility reasons. In most use cases, processes that use Landlock will
> + * *handle* a wide range or all access rights that they know about at build
> + * time.
...and that they tested with a kernel supporting all of them.
> + *
> + * This structure can grow in future Landlock versions.
> */
> struct landlock_ruleset_attr {
> - /**
> - * @handled_access_fs: Bitmask of actions (cf. `Filesystem flags`_)
> - * that is handled by this ruleset and should then be forbidden if no
> - * rule explicitly allow them: it is a deny-by-default list that should
> - * contain as much Landlock access rights as possible. Indeed, all
> - * Landlock filesystem access rights that are not part of
> - * handled_access_fs are allowed. This is needed for backward
> - * compatibility reasons. One exception is the
> - * %LANDLOCK_ACCESS_FS_REFER access right, which is always implicitly
> - * handled, but must still be explicitly handled to add new rules with
> - * this access right.
> - */
> __u64 handled_access_fs;
> - /**
> - * @handled_access_net: Bitmask of actions (cf. `Network flags`_)
> - * that is handled by this ruleset and should then be forbidden if no
> - * rule explicitly allow them.
> - */
> __u64 handled_access_net;
> };
>
> --
> 2.45.2.803.g4e1b14247a-goog
>
>
next prev parent reply other threads:[~2024-07-10 14:15 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-07-10 12:01 [PATCH] landlock: Clarify documentation for struct landlock_ruleset_attr Günther Noack
2024-07-10 12:15 ` Alejandro Colomar
2024-07-10 14:15 ` Mickaël Salaün [this message]
2024-07-11 16:50 ` Günther Noack
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20240710.te8ceiPhav6e@digikod.net \
--to=mic@digikod.net \
--cc=alx@kernel.org \
--cc=gnoack@google.com \
--cc=konstantin.meskhidze@huawei.com \
--cc=linux-security-module@vger.kernel.org \
--cc=repnop@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).