Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::3:2 as permitted sender) client-ip=2620:137:e000::3:2;
Date:   Mon, 6 Nov 2023 16:03:56 +0300
From:   "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
To:     "Nikunj A. Dadhania" <nikunj@amd.com>
Cc:     seanjc@google.com, pbonzini@redhat.com,
        Dave Hansen <dave.hansen@intel.com>,
        linux-kernel@vger.kernel.org, thomas.lendacky@amd.com,
        x86@kernel.org, kvm@vger.kernel.org, bp@alien8.de,
        mingo@redhat.com, tglx@linutronix.de, dave.hansen@linux.intel.com,
        dionnaglaze@google.com, pgonda@google.com
Subject: Re: [PATCH v5 13/14] x86/tsc: Mark Secure TSC as reliable clocksource
Message-ID: <20231106130356.5mppou4pzxmldy22@box.shutemov.name>
References: <20231030063652.68675-1-nikunj@amd.com>
 <20231030063652.68675-14-nikunj@amd.com>
 <57d63309-51cd-4138-889d-43fbdf5ec790@intel.com>
 <ae267e31-5722-4784-9146-28bb13ca7cf5@amd.com>
 <20231102103306.v7ydmrobd5ibs4yn@box.shutemov.name>
 <5d8040b2-c761-4cea-a2ec-39319603e94a@amd.com>
 <cf92b26e-d940-4dc8-a339-56903952cee2@amd.com>
 <20231102123851.jsdolkfz7sd3jys7@box>
 <b56a1eb7-4e31-4806-9f5e-31efe7212e04@amd.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <b56a1eb7-4e31-4806-9f5e-31efe7212e04@amd.com>
Precedence: bulk

On Mon, Nov 06, 2023 at 05:23:44PM +0530, Nikunj A. Dadhania wrote:
> > Maybe kvmclock rating has to be even lower after detecting sane TSC?
> 
> If I set kvmclock rating to 298, I do see exact behavior as you have seen on the bare-metal.
> 
> [    0.004520] clocksource: clocksource_enqueue: name kvm-clock rating 298
> [...]
> [    1.827422] clocksource: clocksource_enqueue: name tsc-early rating 299
> [...]
> [    3.485059] clocksource: Switched to clocksource tsc-early
> [...]
> [    3.623625] clocksource: clocksource_enqueue: name tsc rating 300
> [    3.628954] clocksource: Switched to clocksource tsc

This looks more reasonable to me. But I don't really understand
timekeeping. It would be nice to hear from someone who knows what he
saying.

-- 
  Kiryl Shutsemau / Kirill A. Shutemov