Re: [RFC v2 09/10] landlock: Handle cgroups (program types)

From	Alexei Starovoitov <alexei.starovoitov@gmail.com>
Newsgroups	linux.kernel
Subject	Re: [RFC v2 09/10] landlock: Handle cgroups (program types)
Date	2016-08-27 20:30 +0200
Message-ID	<saQwF-7Xx-11@gated-at.bofh.it> (permalink)
References	(1 earlier) <sa0op-8hT-1@gated-at.bofh.it> <saf46-Zl-15@gated-at.bofh.it> <sar5f-nz-19@gated-at.bofh.it> <sayq5-56V-5@gated-at.bofh.it> <saMW6-5HE-7@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On Sat, Aug 27, 2016 at 04:34:55PM +0200, Mickaël Salaün wrote:
> 
> On 27/08/2016 01:05, Alexei Starovoitov wrote:
> > On Fri, Aug 26, 2016 at 05:10:40PM +0200, Mickaël Salaün wrote:
> >
> >>> As far as safety and type checking that bpf programs has to do,
> >>> I like the approach of patch 06/10:
> >>> +LANDLOCK_HOOK2(file_open, FILE_OPEN,
> >>> +       PTR_TO_STRUCT_FILE, struct file *, file,
> >>> +       PTR_TO_STRUCT_CRED, const struct cred *, cred
> >>> +)
> >>> teaching verifier to recognize struct file, cred, sockaddr
> >>> will let bpf program access them naturally without any overhead.
> >>> Though:
> >>> @@ -102,6 +102,9 @@ enum bpf_prog_type {
> >>>         BPF_PROG_TYPE_SCHED_CLS,
> >>>         BPF_PROG_TYPE_SCHED_ACT,
> >>>         BPF_PROG_TYPE_TRACEPOINT,
> >>> +       BPF_PROG_TYPE_LANDLOCK_FILE_OPEN,
> >>> +       BPF_PROG_TYPE_LANDLOCK_FILE_PERMISSION,
> >>> +       BPF_PROG_TYPE_LANDLOCK_MMAP_FILE,
> >>>  };
> >>> is a bit of overkill.
> >>> I think it would be cleaner to have single
> >>> BPF_PROG_TYPE_LSM and at program load time pass
> >>> lsm_hook_id as well, so that verifier can do safety checks
> >>> based on type info provided in LANDLOCK_HOOKs
> >>
> >> I first started with a unique BPF_PROG_TYPE but, the thing is, the BPF
> >> verifier check programs according to their types. If we need to check
> >> specific context value types (e.g. PTR_TO_STRUCT_FILE), we need a
> >> dedicated program types. I don't see any other way to do it with the
> >> current verifier code. Moreover it's the purpose of program types, right?
> > 
> > Adding new bpf program type for every lsm hook is not acceptable.
> > Either do one new program type + pass lsm_hook_id as suggested
> > or please come up with an alternative approach.
> 
> OK, so we have to modify the verifier to not only rely on the program
> type but on another value to check the context accesses. Do you have a
> hint from where this value could come from? Do we need to add a new bpf
> command to associate a program to a subtype?

It's another field prog_subtype (or prog_hook_id) in union bpf_attr.
Both prog_type and prog_hook_id are used during verification.
prog_type distinguishes the main aspects whereas prog_hook_id selects
which lsm_hook's argument definition to apply.
At the time of attaching to a hook, the prog_hook_id passed at the
load time should match lsm's hook_id.

Thread

[RFC v2 09/10] landlock: Handle cgroups Mickaël Salaün <mic@digikod.net> - 2016-08-25 12:50 +0200
  Re: [RFC v2 09/10] landlock: Handle cgroups Andy Lutomirski <luto@amacapital.net> - 2016-08-25 13:20 +0200
    Re: [RFC v2 09/10] landlock: Handle cgroups Mickaël Salaün <mic@digikod.net> - 2016-08-25 16:50 +0200
      Re: [RFC v2 09/10] landlock: Handle cgroups Tejun Heo <tj@kernel.org> - 2016-08-26 15:00 +0200
      Re: [RFC v2 09/10] landlock: Handle cgroups Andy Lutomirski <luto@amacapital.net> - 2016-08-26 16:30 +0200
        Re: [RFC v2 09/10] landlock: Handle cgroups Tejun Heo <tj@kernel.org> - 2016-08-26 18:00 +0200
  Re: [RFC v2 09/10] landlock: Handle cgroups Alexei Starovoitov <alexei.starovoitov@gmail.com> - 2016-08-26 04:30 +0200
    Re: [RFC v2 09/10] landlock: Handle cgroups Mickaël Salaün <mic@digikod.net> - 2016-08-26 17:20 +0200
      Re: [RFC v2 09/10] landlock: Handle cgroups Alexei Starovoitov <alexei.starovoitov@gmail.com> - 2016-08-27 01:10 +0200
        Re: [RFC v2 09/10] landlock: Handle cgroups Andy Lutomirski <luto@amacapital.net> - 2016-08-27 09:40 +0200
          Re: [RFC v2 09/10] landlock: Handle cgroups Alexei Starovoitov <alexei.starovoitov@gmail.com> - 2016-08-27 20:20 +0200
            Re: [RFC v2 09/10] landlock: Handle cgroups Andy Lutomirski <luto@amacapital.net> - 2016-08-28 10:20 +0200
        Re: [RFC v2 09/10] landlock: Handle cgroups (performance) Mickaël Salaün <mic@digikod.net> - 2016-08-27 16:20 +0200
          Re: [RFC v2 09/10] landlock: Handle cgroups (performance) Alexei Starovoitov <alexei.starovoitov@gmail.com> - 2016-08-27 20:10 +0200
            Re: [RFC v2 09/10] landlock: Handle cgroups (performance) Mickaël Salaün <mic@digikod.net> - 2016-08-27 21:40 +0200
              Re: [RFC v2 09/10] landlock: Handle cgroups (performance) Alexei Starovoitov <alexei.starovoitov@gmail.com> - 2016-08-27 22:50 +0200
                Re: [RFC v2 09/10] landlock: Handle cgroups (performance) Mickaël Salaün <mic@digikod.net> - 2016-08-27 23:20 +0200
                Re: [RFC v2 09/10] landlock: Handle cgroups (performance) Andy Lutomirski <luto@amacapital.net> - 2016-08-28 10:20 +0200
                Re: [RFC v2 09/10] landlock: Handle cgroups (performance) Mickaël Salaün <mic@digikod.net> - 2016-08-28 11:50 +0200
                Re: [RFC v2 09/10] landlock: Handle cgroups (performance) Andy Lutomirski <luto@amacapital.net> - 2016-08-30 21:00 +0200
                Re: [RFC v2 09/10] landlock: Handle cgroups (performance) Mickaël Salaün <mic@digikod.net> - 2016-08-30 22:30 +0200
                Re: [RFC v2 09/10] landlock: Handle cgroups (performance) Andy Lutomirski <luto@amacapital.net> - 2016-08-30 22:30 +0200
                Re: [RFC v2 09/10] landlock: Handle cgroups (performance) Mickaël Salaün <mic@digikod.net> - 2016-08-30 22:40 +0200
                Re: [RFC v2 09/10] landlock: Handle cgroups (performance) Alexei Starovoitov <alexei.starovoitov@gmail.com> - 2016-08-30 23:00 +0200
                Re: [RFC v2 09/10] landlock: Handle cgroups (performance) Andy Lutomirski <luto@amacapital.net> - 2016-08-30 23:50 +0200
                Re: [RFC v2 09/10] landlock: Handle cgroups (performance) Alexei Starovoitov <alexei.starovoitov@gmail.com> - 2016-08-31 03:40 +0200
                Re: [RFC v2 09/10] landlock: Handle cgroups (performance) Andy Lutomirski <luto@amacapital.net> - 2016-08-31 05:30 +0200
        Re: [RFC v2 09/10] landlock: Handle cgroups (netfilter match) Mickaël Salaün <mic@digikod.net> - 2016-08-27 16:30 +0200
          Re: [RFC v2 09/10] landlock: Handle cgroups (netfilter match) Alexei Starovoitov <alexei.starovoitov@gmail.com> - 2016-08-27 20:40 +0200
        Re: [RFC v2 09/10] landlock: Handle cgroups (program types) Mickaël Salaün <mic@digikod.net> - 2016-08-27 16:40 +0200
          Re: [RFC v2 09/10] landlock: Handle cgroups (program types) Alexei Starovoitov <alexei.starovoitov@gmail.com> - 2016-08-27 20:30 +0200
            Re: [RFC v2 09/10] landlock: Handle cgroups (program types) Mickaël Salaün <mic@digikod.net> - 2016-08-27 22:00 +0200
              Re: [RFC v2 09/10] landlock: Handle cgroups (program types) Alexei Starovoitov <alexei.starovoitov@gmail.com> - 2016-08-27 23:00 +0200
                Re: [RFC v2 09/10] landlock: Handle cgroups (program types) Mickaël Salaün <mic@digikod.net> - 2016-08-27 23:20 +0200

csiph-web