linux-kernel - Re: Question about qspinlock nest

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: <20190118102427.GA622@zn.tnic>
Date:   Fri, 18 Jan 2019 11:24:27 +0100
From:   Borislav Petkov <bp@...en8.de>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     James Morse <james.morse@....com>,
        Waiman Long <longman@...hat.com>,
        Zhenzhong Duan <zhenzhong.duan@...cle.com>,
        LKML <linux-kernel@...r.kernel.org>,
        SRINIVAS <srinivas.eeda@...cle.com>,
        Steven Rostedt <rostedt@...dmis.org>
Subject: Re: Question about qspinlock nest

On Fri, Jan 18, 2019 at 11:02:29AM +0100, Peter Zijlstra wrote:
> Well, x86 too has multiple non-maskable vectors, and afaik only the
> actual NMI vector is covered in tricky. But our MCE vector is
> non-maskable too (and I have vague memories of there being more).
> 
> Boris, Rostedt, WTH happens if our MCE code goes and hits a #BP ? (not
> unlikely with this proliferation of self-modifying code)
> 
> Anyway, the idea is that they can indeed not interrupt themselves, but I
> would not be surprised if the whole MCE thing is riddled with fail (on
> x86).

As we talked on IRC: we'll apply a mixture of "So don't do that
then!" hints i.e., kprobing MCE code etc, and fix the issues like the
schedule_work() invocation.

In general, I'd like to make/keep the #MC handler as simple and
as idiot-proof as possible.

-- 
Regards/Gruss,
    Boris.

Good mailing practices for 400: avoid top-posting and trim the reply.