[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <87lfno5x0i.fsf@vitty.brq.redhat.com>
Date: Wed, 25 Mar 2020 15:28:45 +0100
From: Vitaly Kuznetsov <vkuznets@...hat.com>
To: Tianyu Lan <ltykernel@...il.com>
Cc: Yubo Xie <yuboxie@...rosoft.com>, linux-hyperv@...r.kernel.org,
linux-kernel@...r.kernel.org, stable@...r.kernel.org,
Tianyu Lan <Tianyu.Lan@...rosoft.com>, kys@...rosoft.com,
haiyangz@...rosoft.com, sthemmin@...rosoft.com,
liuwe@...rosoft.com, daniel.lezcano@...aro.org, tglx@...utronix.de,
michael.h.kelley@...rosoft.com
Subject: Re: [PATCH] x86/Hyper-V: Fix hv sched clock function return wrong time unit
Tianyu Lan <ltykernel@...il.com> writes:
>>> @@ -398,7 +399,8 @@ static u64 notrace read_hv_clock_msr_cs(struct clocksource *arg)
>>>
>>> static u64 read_hv_sched_clock_msr(void)
>>> {
>>> - return read_hv_clock_msr() - hv_sched_clock_offset;
>>> + return (read_hv_clock_msr() - hv_sched_clock_offset)
>>> + * (NSEC_PER_SEC / HV_CLOCK_HZ);
>>> }
>>
>> kvmclock seems to have the same (pre-patch) code ...
>
>
> kvm sched clock gets time from pvclock_clocksource_read() and
> the time unit is nanosecond. So there is such issue in KVM code.
>
Ah, true, kvmclock is always 1Ghz so it's reading 'naturally' converts
to ns.
--
Vitaly
Powered by blists - more mailing lists