Received-SPF: pass (google.com: domain of linux-kernel+bounces-170236-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.80.249 as permitted sender) client-ip=147.75.80.249;
Date: Mon, 6 May 2024 10:36:06 -0700
From: Kees Cook <keescook@chromium.org>
To: Joao Moreira <joao@overdrivepizza.com>
Cc: Peter Zijlstra <peterz@infradead.org>, x86@kernel.org,
	Sami Tolvanen <samitolvanen@google.com>,
	linux-kernel@vger.kernel.org, Mark Rutland <mark.rutland@arm.com>,
	Josh Poimboeuf <jpoimboe@redhat.com>
Subject: Re: [PATCH] x86/ibt: Implement FineIBT
Message-ID: <202405061031.43EE1C4@keescook>
References: <Y06rtoE9BsERG9uv@hirez.programming.kicks-ass.net>
 <202210181020.79AF7F7@keescook>
 <Y08H8zJ5lQ62jel5@hirez.programming.kicks-ass.net>
 <c561dd8ec384bfc77998a6db6ed824e7@overdrivepizza.com>
 <Y08M4+GxoqvuZ+bq@hirez.programming.kicks-ass.net>
 <d219d61420c48a90a2e8bdc29cb8a579@overdrivepizza.com>
 <202210182218.56AD2871@keescook>
 <baced047981ff5fce633156e3e374dfd@overdrivepizza.com>
 <202211011437.F82B61B8C@keescook>
 <1f6069657f4630c36d60baab2e9b3d10@overdrivepizza.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1f6069657f4630c36d60baab2e9b3d10@overdrivepizza.com>

*thread necromancy*

On Tue, Nov 01, 2022 at 02:50:22PM -0700, Joao Moreira wrote:
> On 2022-11-01 14:39, Kees Cook wrote:
> > On Mon, Oct 31, 2022 at 12:13:50PM -0700, Joao Moreira wrote:
> > > On 2022-10-18 22:19, Kees Cook wrote:
> > > > On Tue, Oct 18, 2022 at 09:48:42PM -0700, Joao Moreira wrote:
> > > > > > > Is it useful to get the compiler to emit 0xcc with
> > > > > > > -fpatchable-function-entry under any circumstance? I can probably
> > > > > > > change
> > > > > > > that quickly if needed/useful.
> > > > > >
> > > > > > Having it emit 0xcc for the bytes in front of the symbol might be
> > > > > > interesting. It would mean a few kernel changes, but nothing too hard.
> > > 
> > > Should I push for this within clang? I have the patch semi-ready
> > > (below) and
> > > would have some cycles this week for polishing it.
> > 
> > Sure! While the NOP vs CC issue isn't very interesting when IBT is
> > available, it's nice for non-IBT to make attackers have to target
> > addresses precisely.
> > 
> > If it's really invasive or hard to maintain in Clang (or objtool),
> > then I'd say leave it as-is.
> 
> The Clang implementation is actually quite simple and, IIRC, I heard in the
> past someone mentioning that trapping instructions actually provide benefits
> for holding undesired straight-line speculation. Maybe someone can comment
> on that, or even if that is really relevant.
> 
> Meanwhile I'll work on pushing it then.

I happened to be looking at in-memory CFI preambles again and noticed
that there's still a NOP sled at every function that got a __cfi_...
target (which only matters in the non-IBT world).

I can't find a PR for your NOP->INT3 patch:
https://github.com/lvwr/llvm-project/commit/ca9029c4536d0544e35dff85e4806803e256841f

Are you able to get this refreshed and landed?

Thanks!

-Kees

-- 
Kees Cook