linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Kees Cook <kees.cook@canonical.com>
To: "Serge E. Hallyn" <serge@hallyn.com>
Cc: linux-security-module@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 2/2] Yama: add PTRACE exception tracking
Date: Tue, 29 Jun 2010 22:27:52 -0700	[thread overview]
Message-ID: <20100630052752.GL4837@outflux.net> (raw)
In-Reply-To: <20100630035609.GA16307@hallyn.com>

Hi Serge,

On Tue, Jun 29, 2010 at 10:56:09PM -0500, Serge E. Hallyn wrote:
> Quoting Kees Cook (kees.cook@canonical.com):
> > Some application suites have external crash handlers that depend on
> > being able to use PTRACE to generate crash reports (KDE, Chromium, etc).
> > Since the inferior process generally knows the PID of the debugger,
> > it can use PR_SET_PTRACER to allow a specific PID and its descendants
> > to perform the PTRACE instead of only a direct ancestor.
> > 
> > Signed-off-by: Kees Cook <kees.cook@canonical.com>
> > ---
> 
> Hi Kees - very nice, overall.  One little note though:

Thanks for looking it over!

> >  	rc = cap_ptrace_access_check(child, mode);
> 
> This means that if capable(CAP_SYS_PTRACE) we'll always shortcut
> here, so
> 
> > +	if (mode == PTRACE_MODE_ATTACH &&
> > +	    ptrace_scope &&
> > +	    !capable(CAP_SYS_PTRACE) &&
> > +	    !task_is_descendant(current, child) &&
> > +	    !ptracer_exception_found(current, child))
> > +		rc = -EPERM;
> 
> You don't need the CAP_SYS_PTRACE check here AFAICS.

I don't think that's true -- the capable(CAP_SYS_PTRACE) tests
are always done in the negative since we only ever abort with error
instead of forcing an early "okay" (see also __ptrace_may_access() in
kernel/ptrace.c, where capable(CAP_SYS_PTRACE) is called repeatedly while
evaluating various negative conditions).

For cap_ptrace_access_check, capable(CAP_SYS_PTRACE) is only tested if
the tracee's process permitted caps are not a subset of the tracer's.
i.e. cap_ptrace_access_check will return 0 when either cap_issubset
or capable.  In the case of normal user processes or a tracer
with greater caps, capable(CAP_SYS_PTRACE) will never be tested in
cap_ptrace_access_check.

As a result, I have to include it in the test here too.  I guess it's
arguable that I should move it to the end of the series of &&s, but it
logically doesn't really matter.

-Kees

(Interestingly, this means that having CAP_SYS_PTRACE means a process
effectively has all capabilities...)

-- 
Kees Cook
Ubuntu Security Team

  reply	other threads:[~2010-06-30  5:27 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-06-30  0:38 [PATCH 0/2] Yama: add PTRACE exception tracking Kees Cook
2010-06-30  0:39 ` [PATCH 1/2] security: create task_free security callback Kees Cook
2010-06-30  0:40 ` [PATCH 2/2] Yama: add PTRACE exception tracking Kees Cook
2010-06-30  1:09   ` Tetsuo Handa
2010-06-30  3:51     ` Kees Cook
2010-06-30  3:56   ` Serge E. Hallyn
2010-06-30  5:27     ` Kees Cook [this message]
2010-06-30 12:40       ` Serge E. Hallyn
2010-06-30 15:41   ` Eric Paris
2010-06-30 15:53     ` Kees Cook
2010-06-30 21:39       ` Tetsuo Handa
2010-06-30  7:31 ` [PATCH 0/2] " Christoph Hellwig
2010-06-30 15:45   ` Kees Cook
2010-07-01  1:39   ` James Morris
2010-07-01  4:44     ` Kees Cook
2010-07-01 13:20       ` Serge E. Hallyn
2010-07-01 15:22         ` Stephen Smalley
2010-07-01 17:16         ` Kees Cook
2010-07-01 19:41           ` Serge E. Hallyn
2010-07-01 19:57             ` Stephen Smalley

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=20100630052752.GL4837@outflux.net \
    --to=kees.cook@canonical.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-security-module@vger.kernel.org \
    --cc=serge@hallyn.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).