Date: Fri, 8 Dec 2017 18:37:59 +0100 (CET)
From: Thomas Gleixner <tglx@linutronix.de>
To: Andy Lutomirski <luto@kernel.org>
cc: Ingo Molnar <mingo@kernel.org>, Borislav Petkov <bp@alien8.de>,
        X86 ML <x86@kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        Brian Gerst <brgerst@gmail.com>,
        David Laight <David.Laight@aculab.com>,
        Kees Cook <keescook@chromium.org>,
        Peter Zijlstra <peterz@infradead.org>
Subject: Re: [PATCH] LDT improvements
In-Reply-To: <CALCETrUgpFAuR4Y2x+QC1mOubCQr-m+s6_gK34A_ge5QQ9a-wg@mail.gmail.com>
Message-ID: <alpine.DEB.2.20.1712081835390.1840@nanos>
References: <48fe5cf1382d6a95c7b1837415882edcc81a9781.1512631324.git.luto@kernel.org> <20171207124347.p7kdj7q4qqs3ivri@pd.tnic> <CALCETrVrQRUn6PoBBGSD8+-wRrWp3mQtKY1sYnCRFyDBte5usA@mail.gmail.com> <alpine.DEB.2.20.1712071814200.1835@nanos>
 <665F1CA8-D012-4465-B5F7-E81E088847DB@amacapital.net> <20171208073454.dicyefwncsihq7sm@gmail.com> <alpine.DEB.2.20.1712081033040.1840@nanos> <20171208094400.wqnezwukq5yx4mgq@gmail.com> <alpine.DEB.2.20.1712081051060.1840@nanos> <20171208113128.hdpeznolztrdzjpf@gmail.com>
 <CALCETrUgpFAuR4Y2x+QC1mOubCQr-m+s6_gK34A_ge5QQ9a-wg@mail.gmail.com>
User-Agent: Alpine 2.20 (DEB 67 2015-01-07)
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1046
Lines: 28

On Fri, 8 Dec 2017, Andy Lutomirski wrote:
> Can we take a step back here?  I think there are four vaguely sane
> ways to make the LDT work:
> 
> 1. The way it is right now -- in vmalloc space.  The only real
> downside is that it requires exposing that part of vmalloc space in
> the user tables, which is a bit gross.
> 
> 2. In some fixmap-like space, which is what my patch does, albeit
> buggily.  This requires a PGD that we treat as per-mm, not per-cpu,
> but that's not so bad.
> 
> 3. In one of the user PGDs but above TASK_SIZE_MAX.  This is
> functionally almost identical to #2, except that there's more concern
> about exploits that write past TASK_SIZE_MAX.
> 
> 4. In an actual vma.  I don't see the benefit of doing this at all --
> it's just like #2 except way more error prone.  Hell, you have to make
> sure that you can't munmap or mremap it, which isn't a consideration
> at all with the other choices.

Why? You can unmap vdso or uprobe or whatever VMAs and you will simply
die. You get what you asked for.

Thanks,

	tglx