linux-kernel - Re: [PATCH] futex: don't leak robust

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] [day] [month] [year] [list]

Message-ID: <87pldhioov.ffs@tglx>
Date: Wed, 30 Jul 2025 19:16:48 +0200
From: Thomas Gleixner <tglx@...utronix.de>
To: Pranav Tyagi <pranav.tyagi03@...il.com>
Cc: Ingo Molnar <mingo@...hat.com>, Peter Zijlstra <peterz@...radead.org>,
 Darren Hart <dvhart@...radead.org>, Davidlohr Bueso <dave@...olabs.net>,
 André Almeida <andrealmeid@...lia.com>,
 linux-kernel@...r.kernel.org,
 jann@...jh.net, keescook@...omium.org, skhan@...uxfoundation.org,
 linux-kernel-mentees@...ts.linux.dev
Subject: Re: [PATCH] futex: don't leak robust_list pointer on exec race

On Wed, Jul 30 2025 at 20:21, Pranav Tyagi wrote:
> On Fri, Jun 13, 2025 at 6:38 PM Thomas Gleixner <tglx@...utronix.de> wrote:
> I face a small issue while refactoring the common code in a helper.
>
> The main obstacle to a full refactor is that the native and compat
> syscalls use different user-visible types (size_t vs compat_size_t,
> struct robust_list_head * vs compat_uptr_t). Because put_user() is
> type-checked at compile-time, I can’t unify both into one function
> without either unsafe casting or weakening type safety (this is as far
> as I understand).
>
> The best I can do is refactor the common task lookup/permission
> logic into a helper, and leave ABI-specific put_user() calls in thin wrappers.

These are two different SYSCALL() implementations, so they
can deal with the difference sizes of put_user() there.

Thanks,

        tglx