SELinux Archive mirror
 help / color / mirror / Atom feed
From: James Carter <jwcart2@gmail.com>
To: Stephen Smalley <stephen.smalley.work@gmail.com>
Cc: "Christian Göttsche" <cgzones@googlemail.com>, selinux@vger.kernel.org
Subject: Re: [RFC PATCH] libsepol: validate permission identifier length
Date: Fri, 8 Dec 2023 12:39:45 -0500	[thread overview]
Message-ID: <CAP+JOzRF-HcrVjihnngqXRbuPYTn2Ra8AU6swP4NXj7S7=n7Hw@mail.gmail.com> (raw)
In-Reply-To: <CAEjxPJ4htnqzDyHFNR_aNJH-hrz_wbM=Z=NN7BjUQDrKBa8e9A@mail.gmail.com>

On Fri, Dec 8, 2023 at 11:06 AM Stephen Smalley
<stephen.smalley.work@gmail.com> wrote:
>
> On Thu, Dec 7, 2023 at 11:53 AM Christian Göttsche
> <cgzones@googlemail.com> wrote:
> >
> > Limit the maximum length of permission identifiers.  Otherwise
> > formatting an access vector might fail in the common used helper
> > sepol_av_to_string().
> >
> > The current longest permission within the Reference Policy is
> > x_application_data { paste_after_confirm } with a length of 19.
>
> Android has longer permission names. I'd rather just fix
> sepol_av_to_string() than impose some arbitrary limit here.
>

I agree.

Christian, something like your recent patch to avoid a fixed size
format buffer for xperms would be better.

Thanks for all your work.
Jim


> >
> > Reported-by: oss-fuzz (issue 64832)
> > Signed-off-by: Christian Göttsche <cgzones@googlemail.com>
> > ---
> > The oss-fuzz issue got closed by the latest fuzzer updates (due to
> > unrelated changes that invalid the current reproducer), but the issue is
> > still valid.
> > ---
> >  libsepol/src/policydb_validate.c | 20 ++++++++++++++++++++
> >  1 file changed, 20 insertions(+)
> >
> > diff --git a/libsepol/src/policydb_validate.c b/libsepol/src/policydb_validate.c
> > index bd8e9f8f..72681120 100644
> > --- a/libsepol/src/policydb_validate.c
> > +++ b/libsepol/src/policydb_validate.c
> > @@ -366,12 +366,30 @@ bad:
> >         return -1;
> >  }
> >
> > +static int validate_permission_wrapper(hashtab_key_t k, __attribute__((unused)) hashtab_datum_t d, void *args)
> > +{
> > +       sepol_handle_t *handle = args;
> > +       const char *name = k;
> > +       size_t len = strlen(name);
> > +
> > +       if (len == 0 || len >= 32)
> > +               goto bad;
> > +
> > +       return 0;
> > +
> > +bad:
> > +       ERR(handle, "Invalid permission");
> > +       return -1;
> > +}
> > +
> >  static int validate_common_datum(sepol_handle_t *handle, const common_datum_t *common, validate_t flavors[])
> >  {
> >         if (validate_value(common->s.value, &flavors[SYM_COMMONS]))
> >                 goto bad;
> >         if (common->permissions.table->nel == 0 || common->permissions.nprim > PERM_SYMTAB_SIZE)
> >                 goto bad;
> > +       if (hashtab_map(common->permissions.table, validate_permission_wrapper, handle))
> > +               goto bad;
> >
> >         return 0;
> >
> > @@ -395,6 +413,8 @@ static int validate_class_datum(sepol_handle_t *handle, const class_datum_t *cla
> >                 goto bad;
> >         if (class->permissions.nprim > PERM_SYMTAB_SIZE)
> >                 goto bad;
> > +       if (hashtab_map(class->permissions.table, validate_permission_wrapper, handle))
> > +               goto bad;
> >         if (validate_constraint_nodes(handle, class->permissions.nprim, class->constraints, flavors))
> >                 goto bad;
> >         if (validate_constraint_nodes(handle, 0, class->validatetrans, flavors))
> > --
> > 2.43.0
> >
> >
>

      parent reply	other threads:[~2023-12-08 17:39 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-12-07 16:53 [RFC PATCH] libsepol: validate permission identifier length Christian Göttsche
2023-12-08 16:05 ` Stephen Smalley
2023-12-08 16:23   ` Stephen Smalley
2023-12-08 17:39   ` James Carter [this message]

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='CAP+JOzRF-HcrVjihnngqXRbuPYTn2Ra8AU6swP4NXj7S7=n7Hw@mail.gmail.com' \
    --to=jwcart2@gmail.com \
    --cc=cgzones@googlemail.com \
    --cc=selinux@vger.kernel.org \
    --cc=stephen.smalley.work@gmail.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).