linux-kernel - Re: [RFC][PATCH 2/2 v3] security: Add task_settimerslack/task

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CALAqxLXx3kDWs2SzuEumaqK6=f6=dJYZMRMYA5urHgo7JDtqMg@mail.gmail.com>
Date:	Wed, 20 Jul 2016 22:54:53 -0700
From:	John Stultz <john.stultz@...aro.org>
To:	James Morris <jmorris@...ei.org>
Cc:	lkml <linux-kernel@...r.kernel.org>,
	Kees Cook <keescook@...omium.org>,
	"Serge E. Hallyn" <serge@...lyn.com>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Thomas Gleixner <tglx@...utronix.de>,
	Arjan van de Ven <arjan@...ux.intel.com>,
	Oren Laadan <orenl@...lrox.com>,
	Ruchi Kandoi <kandoiruchi@...gle.com>,
	Rom Lemarchand <romlem@...roid.com>,
	Todd Kjos <tkjos@...gle.com>, Colin Cross <ccross@...roid.com>,
	Nick Kralevich <nnk@...gle.com>,
	Dmitry Shmidt <dimitrysh@...gle.com>,
	Elliott Hughes <enh@...gle.com>,
	Android Kernel Team <kernel-team@...roid.com>,
	LSM List <linux-security-module@...r.kernel.org>,
	SELinux <selinux@...ho.nsa.gov>
Subject: Re: [RFC][PATCH 2/2 v3] security: Add task_settimerslack/task_gettimerslack
 LSM hook

On Tue, Jul 19, 2016 at 11:12 PM, James Morris <jmorris@...ei.org> wrote:
> On Mon, 18 Jul 2016, John Stultz wrote:
>
>> As requested, this patch implements a task_settimerslack and
>> task_gettimerslack LSM hooks so that the /proc/<tid>/timerslack_ns
>> interface can have finer grained security policies applied to it.
>>
>> I've kept the CAP_SYS_NICE check in the timerslack_ns_write/show
>> functions, as hiding it in the LSM hook seems too opaque, and doesn't
>> seem like a widely enough adopted practice.
>>
>
> I may have missed something in the earlier discussion, but why do we need
> new LSM hooks here vs. calling the existing set/getscheduler hooks?

Mostly since adding a new hook was suggested originally. I don't think
there's much difference as it stands, but I guess more fine grained
checks could be added on the slack amounts, etc.

I can rework it, so let me know if using the existing hooks would be
preferred, but otherwise I'll be sending out the non-rfc patches
tomorrow.

thanks
-john