linux-kernel - Re: [PATCH 3/3] x86/efi: Use efi_switch_mm() rather than manually twiddling with cr3

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: <20170821155913.7bmmy7zaihpd6s7h@hirez.programming.kicks-ass.net>
Date:   Mon, 21 Aug 2017 17:59:13 +0200
From:   Peter Zijlstra <peterz@...radead.org>
To:     Andy Lutomirski <luto@...capital.net>
Cc:     Andy Lutomirski <luto@...nel.org>,
        Will Deacon <will.deacon@....com>,
        Mark Rutland <mark.rutland@....com>,
        Matt Fleming <matt@...eblueprint.co.uk>,
        Ard Biesheuvel <ard.biesheuvel@...aro.org>,
        Sai Praneeth Prakhya <sai.praneeth.prakhya@...el.com>,
        "linux-efi@...r.kernel.org" <linux-efi@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        joeyli <jlee@...e.com>, Borislav Petkov <bp@...en8.de>,
        "Michael S. Tsirkin" <mst@...hat.com>,
        "Neri, Ricardo" <ricardo.neri@...el.com>,
        "Ravi V. Shankar" <ravi.v.shankar@...el.com>
Subject: Re: [PATCH 3/3] x86/efi: Use efi_switch_mm() rather than manually
 twiddling with cr3

On Mon, Aug 21, 2017 at 08:23:10AM -0700, Andy Lutomirski wrote:
> > Ah, but only root can create per-cpu events or attach events to kernel
> > threads (with sensible paranoia levels).
> 
> But this may not need to be percpu.  If a non root user can trigger, say, an EFI variable read in their own thread context, boom.

I was going by the proposed: "everything EFI in a kthread" model. But
yes, if that's not done, then you're quite right.

> > 
> >> The other is that, if EFI were to have IO memory mapped at a "user"
> >> address, perf could end up reading it.
> > 
> > Ah, but in neither mode does perf assume much, the LBR follows branches
> > the CPU took and thus we _know_ there was code there, not MMIO. And the
> > stack unwind simply follows the stack up, although I suppose it could be
> > 'tricked' into probing MMIO. We can certainly add an "->mm !=
> > ->active_mm" escape clause to the unwind code.
> > 
> > Although I don't see how we're currently avoiding the same problem with
> > existing userspace unwinds, userspace can equally have MMIO mapped.
> 
> But user space at least only has IO mapped to which the user program in question has rights.

Still, we should not mess it up just because we're trying to unwind
stacks.