linux-kernel - Re: [PATCH v2 0/3] x86/vdso: Add Hyper-V TSC page clocksource support

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 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]

Date:   Fri, 17 Feb 2017 18:55:02 +0100 (CET)
From:   Thomas Gleixner <tglx@...utronix.de>
To:     Andy Lutomirski <luto@...capital.net>
cc:     Vitaly Kuznetsov <vkuznets@...hat.com>,
        "K. Y. Srinivasan" <kys@...rosoft.com>, X86 ML <x86@...nel.org>,
        Ingo Molnar <mingo@...hat.com>,
        "H. Peter Anvin" <hpa@...or.com>,
        Haiyang Zhang <haiyangz@...rosoft.com>,
        Stephen Hemminger <sthemmin@...rosoft.com>,
        Dexuan Cui <decui@...rosoft.com>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        devel@...uxdriverproject.org,
        Linux Virtualization <virtualization@...ts.linux-foundation.org>
Subject: Re: [PATCH v2 0/3] x86/vdso: Add Hyper-V TSC page clocksource
 support

On Fri, 17 Feb 2017, Andy Lutomirski wrote:
> On Fri, Feb 17, 2017 at 2:14 AM, Vitaly Kuznetsov <vkuznets@...hat.com> wrote:
> > Not only. As far as I understand (and I *think* K. Y. pointed this out)
> > when VM is migrating to another host TSC page clocksource is disabled for
> > extended period of time so we're better off reading from MSR than
> > looping here. With regards to VDSO this means reverting to doing normal
> > syscall.
> 
> That's a crappy design.  The guest really ought to be able to
> distinguish "busy, try again" from "bail and use MSR".
> 
> Thomas, I can see valid reasons why the hypervisor might want to
> temporarily disable shared page-based timing, but I think it's silly
> that it's conflated with indicating "retry".

It's beyond silly if that's the case.

> But if this is indeed the ABI, we're stuck with it.

No argument.