linux-kernel - Re: [PATCH] perf/events: replace READ_ONCE() with standard page table accessors

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: <20251006084827.GR3245006@noisy.programming.kicks-ass.net>
Date: Mon, 6 Oct 2025 10:48:27 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Anshuman Khandual <anshuman.khandual@....com>
Cc: linux-perf-users@...r.kernel.org, Ingo Molnar <mingo@...hat.com>,
	Arnaldo Carvalho de Melo <acme@...nel.org>,
	Namhyung Kim <namhyung@...nel.org>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] perf/events: replace READ_ONCE() with standard page
 table accessors

On Mon, Oct 06, 2025 at 02:10:14PM +0530, Anshuman Khandual wrote:

> These accessors provide platforms the opportunity to override and have different
> implementations other than READ_ONCE() when required.

When would this be required? Traditionally the only case that was weird
was Alpha, since it needed the dereference barrier, but since we folded
that into READ_ONCE(), that has all gone away.

[ also, historically we trusted GCC to emit singe loads for word sized
accesses -- something we've walked back from as C compilers have started
to be more agressive ]

Anyway, READ_ONCE() is the only sane way to dereference a page-table
pointers, doubly so in the face of lockless access.

There should be no need for architectures to be 'funny' here.