linux-kernel - Re: [PATCH] x86/Kconfig: make CFI_AUTO

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: <20250410132522.GD9833@noisy.programming.kicks-ass.net>
Date: Thu, 10 Apr 2025 15:25:22 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Paweł Anikiel <panikiel@...gle.com>
Cc: Sami Tolvanen <samitolvanen@...gle.com>, Kees Cook <kees@...nel.org>,
	Alex Gaynor <alex.gaynor@...il.com>, Borislav Petkov <bp@...en8.de>,
	Dave Hansen <dave.hansen@...ux.intel.com>,
	Ingo Molnar <mingo@...hat.com>,
	Josh Poimboeuf <jpoimboe@...nel.org>,
	Masahiro Yamada <masahiroy@...nel.org>,
	Miguel Ojeda <ojeda@...nel.org>,
	Thomas Gleixner <tglx@...utronix.de>,
	Alice Ryhl <aliceryhl@...gle.com>,
	Nathan Chancellor <nathan@...nel.org>, x86@...nel.org,
	linux-kernel@...r.kernel.org, rust-for-linux@...r.kernel.org
Subject: Re: [PATCH] x86/Kconfig: make CFI_AUTO_DEFAULT depend on !RUST

On Thu, Apr 10, 2025 at 03:12:54PM +0200, Paweł Anikiel wrote:

> I'll let the rust maintainers answer this one.
> 
> > >
> > > How is it still compatible with kCFI and not with FineIBT?
> 
> kCFI keeps the original function's endbr64. If a callsite disables
> CFI, it simply jumps to that function (without any magic value
> checks), the endbr64 is there, so no #CP.
> 
> With FineIBT, the original function's endbr64 gets replaced with a
> nop, and the callsite is supposed to be modified to jump to the CFI
> preamble. With CFI disabled, that modification doesn't happen
> (cfi_rewrite_callers() quietly ignores that callsite), so we jump to
> the nop, and we get a #CP.

Oh, you're attempting to do a no-cfi indirect branch? Yeah, you don't
get to do that.

I should get objtool to warn about those. They undermine the point of
CFI.

> > FWIW, CFI violations of any kind are a no-no. They're not accepted in C
> > and they're not accepted in Rust. This is clearly a Rust bug that needs
> > to be fixed ASAP. Disabling CFI is not an option.
> 
> This is a known issue, but it doesn't seem to have high priority:
> https://github.com/rust-lang/rust/issues/115199

Now you get to fix it :) Because such code is unacceptable for the
kernel.

Just because some crazy Rust person sticks this in their runtime,
doesn't mean you get to use it in the kernel.