linux-kernel - Redundant barrier_nospec() added by commit 74e19ef0ff80 ("uaccess: Add speculation barrier to copy_from

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

Message-ID: <6f58923f-5ee2-41c6-bb73-8e0fb9ad9a96@csgroup.eu>
Date: Thu, 19 Jun 2025 10:42:13 +0200
From: Christophe Leroy <christophe.leroy@...roup.eu>
To: Dave Hansen <dave.hansen@...ux.intel.com>
Cc: Linus Torvalds <torvalds@...uxfoundation.org>,
 Thomas Gleixner <tglx@...utronix.de>, Daniel Borkmann
 <daniel@...earbox.net>, "linux-mm@...ck.org" <linux-mm@...ck.org>,
 "linuxppc-dev@...ts.ozlabs.org" <linuxppc-dev@...ts.ozlabs.org>,
 "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Redundant barrier_nospec() added by commit 74e19ef0ff80 ("uaccess:
 Add speculation barrier to copy_from_user()")

Hello Dave,

In commit 74e19ef0ff80 ("uaccess: Add speculation barrier to 
copy_from_user()") you add barrier_nospec() to copy_from_user().

There is already a barrier_nospec() in allow_read_from_user() which is 
called by raw_copy_from_user(), see commit 8524e2e76441 
("powerpc/uaccess: Perform barrier_nospec() in KUAP allowance helpers"). 
Why do you need an additional one ?

And why only copy_from_user() and not also other places like 
copy_from_user_iter() for instance ?

Thanks
Christophe