linux-kernel - Re: [RFC][PATCH]: x86: clarify/fix no-op barriers for text_poke

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: <20170731152617.q3nhiczhobrn67lk@hirez.programming.kicks-ass.net>
Date:   Mon, 31 Jul 2017 17:26:17 +0200
From:   Peter Zijlstra <peterz@...radead.org>
To:     Steven Rostedt <rostedt@...dmis.org>
Cc:     jkosina@...e.cz, masami.hiramatsu.pt@...achi.com,
        hpa@...ux.intel.com, linux-kernel@...r.kernel.org, x86@...nel.org
Subject: Re: [RFC][PATCH]: x86: clarify/fix no-op barriers for text_poke_bp()

On Mon, Jul 31, 2017 at 11:16:48AM -0400, Steven Rostedt wrote:
> On Mon, 31 Jul 2017 17:04:11 +0200
> Peter Zijlstra <peterz@...radead.org> wrote:
> 
> > Well, this is x86, its a NO-OP. The only reason to write barriers like
> 
> Is it really a nop?

smp_wmb() is ever since we killed OOSTORE, and smp_rmb() is effectively
(I don't think anybody uses PPRO_FENCE).

The normal x86 TSO memory model doesn't need read or write barriers --
only weird and wonderful (broken) x86 bits do.