linux-kernel - Re: [RFC 1/2] x86/tsc: add a timer to make sure tsc

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: <87fszy2d68.ffs@nanos.tec.linutronix.de>
Date:   Sat, 10 Apr 2021 20:43:27 +0200
From:   Thomas Gleixner <tglx@...utronix.de>
To:     Borislav Petkov <bp@...en8.de>
Cc:     Feng Tang <feng.tang@...el.com>, Ingo Molnar <mingo@...hat.com>,
        "H . Peter Anvin" <hpa@...or.com>,
        Peter Zijlstra <peterz@...radead.org>, x86@...nel.org,
        linux-kernel@...r.kernel.org, rui.zhang@...el.com,
        andi.kleen@...el.com, dave.hansen@...el.com, len.brown@...el.com
Subject: Re: [RFC 1/2] x86/tsc: add a timer to make sure tsc_adjust is always checked

On Sat, Apr 10 2021 at 11:47, Borislav Petkov wrote:

> On Sat, Apr 10, 2021 at 11:27:11AM +0200, Thomas Gleixner wrote:
>> On Tue, Mar 30 2021 at 16:25, Feng Tang wrote:
>> > Normally the tsc_sync will be checked every time system enters idle state,
>> > but there is still caveat that a system won't enter idle, either because
>> > it's too busy or configured purposely to not enter idle. Setup a periodic
>> > timer to make sure the check is always on.
>> 
>> Bah. I really hate the fact that we don't have a knob to disable writes
>> to the TSC/TSC_ADJUST msrs. That would spare this business alltogether.
>
> We have the MSR filtering and I'd *love* to add those MSRs to a
> permanent ban list of MSRs which will never ever be written to from
> luserspace.

That's good, but what I really want is a knob which prevents BIOS/SMM
from writing to it. The only reason why BIOS ever needs to write is for
physical hotplug and perhaps for 4+ socket machines on boot. After that
every write is a bug.

Thanks,

        tglx