linux-kernel - Re: [PATCH] selinux: move user accesses in selinuxfs out of locked regions

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]

Date:   Tue, 26 Jun 2018 00:40:46 +0200
From:   Jann Horn <jannh@...gle.com>
To:     Paul Moore <paul@...l-moore.com>
Cc:     Stephen Smalley <sds@...ho.nsa.gov>,
        Eric Paris <eparis@...isplace.org>, selinux@...ho.nsa.gov,
        security@...nel.org, kernel list <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] selinux: move user accesses in selinuxfs out of locked regions

On Tue, Jun 26, 2018 at 12:36 AM Paul Moore <paul@...l-moore.com> wrote:
>
> On Mon, Jun 25, 2018 at 12:34 PM Jann Horn <jannh@...gle.com> wrote:
> > If a user is accessing a file in selinuxfs with a pointer to a userspace
> > buffer that is backed by e.g. a userfaultfd, the userspace access can
> > stall indefinitely, which can block fsi->mutex if it is held.
> >
> > For sel_read_policy(), remove the locking, since this method doesn't seem
> > to access anything that requires locking.
>
> Forgive me, I'm thinking about this quickly so I could be very wrong
> here, but isn't the mutex needed to prevent problems in multi-threaded
> apps hitting the same fd at the same time?

sel_read_policy() operates on a read-only copy of the policy, accessed
via ->private_data, allocated using vmalloc in sel_open_policy() via
security_read_policy(). As far as I can tell, nothing can write to
that read-only copy of the policy. None of the handlers in
sel_policy_ops write - they just mmap as readonly (in which case
you're already reading without locks, by the way) or read.