linux-kernel - Re: [RFC PATCH 00/16] x86/split_lock: Enable #AC exception for split locked accesses

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: <20180622081924.GH2494@hirez.programming.kicks-ass.net>
Date:   Fri, 22 Jun 2018 10:19:24 +0200
From:   Peter Zijlstra <peterz@...radead.org>
To:     Fenghua Yu <fenghua.yu@...el.com>
Cc:     Thomas Gleixner <tglx@...utronix.de>, Ingo Molnar <mingo@...e.hu>,
        "H. Peter Anvin" <hpa@...ux.intel.com>,
        Ashok Raj <ashok.raj@...el.com>,
        Dave Hansen <dave.hansen@...el.com>,
        Rafael Wysocki <rafael.j.wysocki@...el.com>,
        Tony Luck <tony.luck@...el.com>,
        Alan Cox <alan@...ux.intel.com>,
        Ravi V Shankar <ravi.v.shankar@...el.com>,
        Arjan van de Ven <arjan@...radead.org>,
        linux-kernel <linux-kernel@...r.kernel.org>, x86 <x86@...nel.org>
Subject: Re: [RFC PATCH 00/16] x86/split_lock: Enable #AC exception for split
 locked accesses

On Thu, Jun 21, 2018 at 04:05:02PM -0700, Fenghua Yu wrote:

> But in a real case, when I enable #AC for split lock in kernel, reboot
> hits #AC because of split lock in firmware code and firmware handles #AC
> as fatal error and stops continuing to run.
> 
> It will take long time/forever for firmware to fix the split lock issue.
> Before the firmware issue is fixed, reboot or S4 cannot run if the feature
> is enabled by kernel.

Then it will take a long time / forever for people to use their machines
as they need. Don't you see that is a problem?

Either you take this real-time stuff serious and force BIOS monkeys to
get with the act, this includes no #AC but also very much force them
into abandoning SMM as a fail^Wfeature-platform.

Or it's all a joke and we'll continue to complain and shame..

> And if unlucky, I'm afraid the patch set even has no chance to be merged to
> upstream if maintainer's test machine has firmware split lock issue and the
> machine simply cannot reboot or go to S4 if the feature is enabled.

Well, then you'd better get those BIOS monkeys on board and fixing their
crap.