linux-kernel - RE: [PATCH v2 1/2] x86/split_lock: Make life miserable for split lockers

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: <532e45aad7c64349b52217bc548edaca@intel.com>
Date:   Thu, 17 Mar 2022 16:27:41 +0000
From:   "Luck, Tony" <tony.luck@...el.com>
To:     Pavel Machek <pavel@...x.de>
CC:     Thomas Gleixner <tglx@...utronix.de>,
        "Yu, Fenghua" <fenghua.yu@...el.com>,
        "x86@...nel.org" <x86@...nel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "patches@...ts.linux.dev" <patches@...ts.linux.dev>
Subject: RE: [PATCH v2 1/2] x86/split_lock: Make life miserable for split
 lockers

Pavel,

> I'm not sure what split locks are, and if you want applications to
> stop doing that maybe documentation would help.

See existing Documentation/x86/buslock.rst

> Anyway, you can't really introduce regressions to userspace to "get
> stuff fixed" in applications.

Applications can inflict pain on the system with atomic operations
on unaligned operands that cross cache line boundaries. This
is just pushing some pain back to the applications.

An alternate title for this patch series could have been:

	"Split-locks: The OS strikes back"

Note that there are few applications that actually split locks.
Normal compiler alignment rules generally avoid them.
Applications that run on non-x86 architectures have to
avoid them because few others allow atomic operations
on unaligned operands.

-Tony