From: Casey Schaufler <casey@schaufler-ca.com>
To: "Mickaël Salaün" <mic@digikod.net>,
"Paul Moore" <paul@paul-moore.com>,
"Jeff Xu" <jeffxu@chromium.org>
Cc: Shervin Oloumi <enlightened@chromium.org>,
linux-security-module@vger.kernel.org, jorgelo@chromium.org,
keescook@chromium.org, groeck@chromium.org,
allenwebb@chromium.org, gnoack3000@gmail.com, areber@redhat.com,
criu@openvz.org, linux-api@vger.kernel.org, jannh@google.com,
brauner@kernel.org, Casey Schaufler <casey@schaufler-ca.com>
Subject: Re: [PATCH v2] lsm: adds process attribute getter for Landlock
Date: Thu, 25 May 2023 09:28:29 -0700 [thread overview]
Message-ID: <aa2e3c9c-eac4-237d-02d0-4574f602563d@schaufler-ca.com> (raw)
In-Reply-To: <b4825033-471c-ba32-530f-b0235356d55b@digikod.net>
On 5/24/2023 9:02 AM, Mickaël Salaün wrote:
>
> On 24/05/2023 17:38, Mickaël Salaün wrote:
>>
>> On 23/05/2023 23:12, Paul Moore wrote:
>>> On Tue, May 23, 2023 at 2:13 AM Jeff Xu <jeffxu@chromium.org> wrote:
>>>> On Mon, May 22, 2023 at 12:56 PM Paul Moore <paul@paul-moore.com>
>>>> wrote:
>>>>> On Thu, May 18, 2023 at 5:26 PM Casey Schaufler
>>>>> <casey@schaufler-ca.com> wrote:
>>>>>> On 5/18/2023 1:45 PM, Shervin Oloumi wrote:
>>>>>>> Adds a new getprocattr hook function to the Landlock LSM, which
>>>>>>> tracks
>>>>>>> the landlocked state of the process. This is invoked when
>>>>>>> user-space
>>>>>>> reads /proc/[pid]/attr/domain
>>>>>>
>>>>>> Please don't add a Landlock specific entry directly in the attr/
>>>>>> directory. Add it only to attr/landlock.
>>>>>>
>>>>>> Also be aware that the LSM maintainer (Paul Moore) wants to move
>>>>>> away from the /proc/.../attr interfaces in favor of a new system
>>>>>> call,
>>>>>> which is in review.
>>>>>
>>>>> What Casey said above.
>>>>>
>>>>> There is still some uncertainty around timing, and if we're perfectly
>>>>> honest, acceptance of the new syscalls at the Linus level, but yes, I
>>>>> would very much like to see the LSM infrastructure move away from
>>>>> procfs and towards a syscall API. Part of the reasoning is that the
>>>>> current procfs API is ill-suited to handle the multiple, stacked LSMs
>>>>> and the other part being the complexity of procfs in a namespaced
>>>>> system. If the syscall API is ultimately rejected, we will need to
>>>>> revisit the idea of a procfs API, but even then I think we'll need to
>>>>> make some changes to the current approach.
>>>>>
>>>>> As I believe we are in the latter stages of review for the syscall
>>>>> API, perhaps you could take a look and ensure that the current
>>>>> proposed API works for what you are envisioning with Landlock?
>>
>> I agree, and since the LSM syscalls are almost ready that should not
>> change much the timing. In fact, extending these syscalls might be
>> easier than tweaking the current procfs/attr API for Landlock specific
>> requirements (e.g. scoped visibility). We should ensure that these
>> syscalls would be a good fit to return file descriptors, but in the
>> short term we only need to know if a process is landlocked or not, so a
>> raw return value (0 or -errno) will be enough.
>>
>> Mentioning in the LSM syscalls patch series that they may deal with (and
>> return) file descriptors could help API reviewers though.
>
> It should be kept in mind that the current LSM syscalls only deal with
> the calling task, whereas the goal of this Landlock patch series is to
> inspect other tasks. A new LSM syscall would need to be created to
> handle pidfd e.g., named lsm_get_proc_attr() or lsm_get_pid_attr().
I think it would be lsm_get_pid_attr(). Yes, it's the obvious next step.
>
> I'm not sure if this should be a generic LSM syscall or a Landlock
> syscall though. I have plan to handle processes other than the caller
> (e.g. to restrict an existing process hierarchy), so thinking about a
> Landlock-specific syscall could make sense.
>
> To summarize, creating a new LSM syscall to deal with pidfd and to get
> LSM process "status/attr" looks OK. However, Landlock-specific
> syscalls to deal with Landlock specificities (e.g. ruleset or domain
> file descriptor) make more sense.
>
> Having one LSM-generic syscall to get minimal Landlock attributes
> (i.e. mainly to know if a process is sandboxed), and another
> Landlock-specific syscall to do more things (e.g. get the domain file
> descriptor, restrict a task) seems reasonable. The second one would
> overlap with the first one though. What do you think?
I find it difficult to think of a file descriptor as an attribute of
a process. To my (somewhat unorthodox) thinking a file descriptor is
a name for an object, not an attribute of the object. You can't access
an object by its attributes, but you can by its name. An attribute is
a description of the object. I'm perfectly happy with lsm_get_pid_attr()
returning an attribute that is a file descriptor if it describes the
process in some way, but not as a substitute for opening /proc/42.
next prev parent reply other threads:[~2023-05-25 16:30 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20230302185257.850681-1-enlightened@chromium.org>
2023-03-06 19:18 ` [PATCH 0/1] process attribute support for Landlock Mickaël Salaün
2023-03-07 14:16 ` Mickaël Salaün
2023-03-08 22:25 ` Shervin Oloumi
2023-03-15 9:56 ` Mickaël Salaün
2023-03-16 6:19 ` Günther Noack
2023-03-17 8:38 ` Mickaël Salaün
2023-05-18 20:44 ` Shervin Oloumi
2023-05-24 16:09 ` Mickaël Salaün
2023-05-24 16:21 ` Mickaël Salaün
2023-05-18 20:45 ` [PATCH v2] lsm: adds process attribute getter " Shervin Oloumi
2023-05-18 21:26 ` Casey Schaufler
2023-05-22 19:56 ` Paul Moore
2023-05-23 6:13 ` Jeff Xu
2023-05-23 15:32 ` Casey Schaufler
2023-05-30 18:02 ` Jeff Xu
2023-05-30 19:05 ` Casey Schaufler
2023-05-31 13:01 ` Mickaël Salaün
2023-06-01 20:45 ` Jeff Xu
2023-06-01 21:30 ` Casey Schaufler
2023-05-23 21:12 ` Paul Moore
2023-05-24 15:38 ` Mickaël Salaün
2023-05-24 16:02 ` Mickaël Salaün
2023-05-25 16:28 ` Casey Schaufler [this message]
2023-05-30 18:05 ` Jeff Xu
2023-05-30 19:19 ` Casey Schaufler
2023-05-31 13:26 ` Mickaël Salaün
2023-06-01 20:48 ` Jeff Xu
2023-06-01 21:34 ` Casey Schaufler
2023-06-01 22:08 ` Mickaël Salaün
2023-05-24 16:05 ` Mickaël Salaün
2023-05-19 5:22 ` kernel test robot
2023-05-24 16:48 ` Mickaël Salaün
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=aa2e3c9c-eac4-237d-02d0-4574f602563d@schaufler-ca.com \
--to=casey@schaufler-ca.com \
--cc=allenwebb@chromium.org \
--cc=areber@redhat.com \
--cc=brauner@kernel.org \
--cc=criu@openvz.org \
--cc=enlightened@chromium.org \
--cc=gnoack3000@gmail.com \
--cc=groeck@chromium.org \
--cc=jannh@google.com \
--cc=jeffxu@chromium.org \
--cc=jorgelo@chromium.org \
--cc=keescook@chromium.org \
--cc=linux-api@vger.kernel.org \
--cc=linux-security-module@vger.kernel.org \
--cc=mic@digikod.net \
--cc=paul@paul-moore.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 read-only IMAP folder(s) and NNTP newsgroup(s).