linux-kernel - Re: TSC to Mono-raw Drift

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]

Date:   Fri, 2 Nov 2018 12:20:06 +0100
From:   Miroslav Lichvar <mlichvar@...hat.com>
To:     Thomas Gleixner <tglx@...utronix.de>
Cc:     John Stultz <john.stultz@...aro.org>,
        Christopher Hall <christopher.s.hall@...el.com>,
        "H. Peter Anvin" <hpa@...or.com>,
        linux-rt-users <linux-rt-users@...r.kernel.org>,
        jesus.sanchez-palencia@...el.com,
        Gavin Hindman <gavin.hindman@...el.com>,
        liam.r.girdwood@...el.com, Peter Zijlstra <peterz@...radead.org>,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: TSC to Mono-raw Drift

On Thu, Nov 01, 2018 at 07:03:37PM +0100, Thomas Gleixner wrote:
> On Thu, 1 Nov 2018, John Stultz wrote:
> > On Thu, Nov 1, 2018 at 10:44 AM, Thomas Gleixner <tglx@...utronix.de> wrote:
> > > On Tue, 23 Oct 2018, John Stultz wrote:
> > >> However, to be correct, the ntp adjustments made would have to be made
> > >> to both the base interval + error, which mucks the math up a fair bit.
> > >
> > > Hmm, confused as usual. Why would you need to do anything like that?
> > 
> > Because the NTP adjustment is done off of what is now the raw clock.
> > If the raw clock is "corrected" the ppb adjustment has to be done off
> > of that corrected rate.
> 
> Sure, but why would that require any change? Right now the raw clock is
> slightly off and you correct clock monotonic against NTP. So with that
> extra correction you just see a slightly different raw clock slew and work
> from there.

It makes sense to me.

I think there are basically two different ways how it could be done.
One is to correct the frequency of the raw clock, on which sits the
mono/real clock. The other is to create a new raw clock which is
separate from the mono/real clock, and add an offset to the NTP
frequency to match the frequencies of the two clocks when not
synchronized by NTP/PTP. The latter would provide a more stable
mono/real clock.

clocksource -> MONOTONIC_RAW -> MONOTONIC/REALTIME

or 

clocksource -> ? -> MONOTONIC_RAW
                 -> MONOTONIC/REALTIME

-- 
Miroslav Lichvar