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: <202504161442.66CE2596@keescook>
Date: Wed, 16 Apr 2025 14:51:55 -0700
From: Kees Cook <kees@...nel.org>
To: Peter Zijlstra <peterz@...radead.org>
Cc: Miguel Ojeda <miguel.ojeda.sandonis@...il.com>,
	Alice Ryhl <aliceryhl@...gle.com>,
	Paweł Anikiel <panikiel@...gle.com>,
	Sami Tolvanen <samitolvanen@...gle.com>,
	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>,
	Nathan Chancellor <nathan@...nel.org>, x86@...nel.org,
	linux-kernel@...r.kernel.org, rust-for-linux@...r.kernel.org,
	Matthew Maurer <mmaurer@...gle.com>,
	Ramon de C Valle <rcvalle@...gle.com>
Subject: Re: [PATCH] x86/Kconfig: make CFI_AUTO_DEFAULT depend on !RUST

On Wed, Apr 16, 2025 at 10:20:40PM +0200, Peter Zijlstra wrote:
> On Tue, Apr 15, 2025 at 05:15:31PM +0200, Miguel Ojeda wrote:
> > On Thu, Apr 10, 2025 at 4:00 PM Alice Ryhl <aliceryhl@...gle.com> wrote:
> > >
> > > I submitted a PR that I believe should fix it:
> > > https://github.com/rust-lang/rust/pull/139632
> > 
> > This landed, scheduled for Rust 1.88 (2025-06-26) if all goes well.
> > 
> > Peter: are you OK with landing a patch like this? Well, modified to
> > look like this now that we know the version:
> > 
> >     depends on FINEIBT
> >     depends on !RUST || RUSTC_VERSION >= 108800
> > 
> > (assuming we confirm the nightly build works properly)
> 
> I don't much like it -- disabling FineIBT at config time like this also
> kills the CFI type rehash.

This isn't disabling CONFIG_FINEIBT (which gates cfi_rand), it's making
FineIBT not enabled by default at boot time. This is actually when I
created CONFIG_CFI_AUTO_DEFAULT: to be able to have kCFI _with_ type
rehashing still enabled.

> Not to mention that FineIBT is a Spectre-BHI
> mitigation, esp. with that arity thing on top.

That's certainly true, but the needed overlapping corner cases seem to
make this tolerable?

> I don't suppose we can simply mandate this rust version?

Perhaps capture the needed version in the Kconfig change at least, so it
becomes self-documenting.

-Kees

-- 
Kees Cook