Re: [kernel-hardening] [PATCH 1/2] security, perf: allow further restriction of perf_event_open

From: Daniel Micay <danielmicay@gmail.com>
To: kernel-hardening@lists.openwall.com
Cc: peterz@infradead.org, mingo@redhat.com, acme@kernel.org,
	alexander.shishkin@linux.intel.com, linux-doc@vger.kernel.org,
	linux-kernel@vger.kernel.org,
	Jeff Vander Stoep <jeffv@google.com>
Subject: Re: [kernel-hardening] [PATCH 1/2] security, perf: allow further restriction of perf_event_open
Date: Mon, 17 Oct 2016 10:54:33 -0400	[thread overview]
Message-ID: <1476716073.4032.5.camel@gmail.com> (raw)
In-Reply-To: <20161017134413.GK29095@leverpostej>

[-- Attachment #1: Type: text/plain, Size: 2274 bytes --]

On Mon, 2016-10-17 at 14:44 +0100, Mark Rutland wrote:
> Hi,
> 
> Attempt to revive discussions below...
> 
> On Wed, Jul 27, 2016 at 07:45:46AM -0700, Jeff Vander Stoep wrote:
> > When kernel.perf_event_paranoid is set to 3 (or greater), disallow
> > all access to performance events by users without CAP_SYS_ADMIN.
> > 
> > This new level of restriction is intended to reduce the attack
> > surface of the kernel. Perf is a valuable tool for developers but
> > is generally unnecessary and unused on production systems. Perf may
> > open up an attack vector to vulnerable device-specific drivers as
> > recently demonstrated in CVE-2016-0805, CVE-2016-0819,
> > CVE-2016-0843, CVE-2016-3768, and CVE-2016-3843. This new level of
> > restriction allows for a safe default to be set on production
> > systems
> > while leaving a simple means for developers to grant access [1].
> > 
> > This feature is derived from CONFIG_GRKERNSEC_PERF_HARDEN by Brad
> > Spengler. It is based on a patch by Ben Hutchings [2]. Ben's patches
> > have been modified and split up to address on-list feedback.
> > 
> > kernel.perf_event_paranoid=3 is the default on both Debian [2] and
> > Android [3].
> 
> While people weren't particularly happy with this global toggle
> approach, my understanding from face-to-face discussions at LSS2016
> was
> that people were happy with a more scoped restriction (e.g. using
> capabilities or some other access control mechanism), but no-one had
> the
> time to work on that.
> 
> Does that match everyone's understanding, or am I mistaken?
> 
> It's also my understanding that for Android, perf_event_paranoid is
> lowered when the user enables developer mode (rather than only when an
> external debugger is attached); is that correct?

It's exposed as a "system property" marked as writable by the shell
user, so the Android Debug Bridge shell can lower it. The debugging
tools learned how to toggle it off automatically when they're used. It
intentionally isn't a persist. prefixed property so the setting also
goes away on reboot.

ADB (incl. the shell user) isn't available until developer mode is
enabled + ADB is toggled on in the developer settings, and then it still
requires whitelisting keys.

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 833 bytes --]