Re: [RFC PATCH v9 7/8] ptp: arm64: Enable ptp_kvm for arm64

From: Marc Zyngier <maz@kernel.org>
To: Jianyong Wu <Jianyong.Wu@arm.com>
Cc: netdev@vger.kernel.org, yangbo.lu@nxp.com,
	john.stultz@linaro.org, tglx@linutronix.de, pbonzini@redhat.com,
	sean.j.christopherson@intel.com, richardcochran@gmail.com,
	Mark Rutland <Mark.Rutland@arm.com>,
	will@kernel.org, Suzuki Poulose <Suzuki.Poulose@arm.com>,
	Steven Price <Steven.Price@arm.com>,
	linux-kernel@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org,
	kvmarm@lists.cs.columbia.edu, kvm@vger.kernel.org,
	Steve Capper <Steve.Capper@arm.com>, Kaly Xin <Kaly.Xin@arm.com>,
	Justin He <Justin.He@arm.com>, nd <nd@arm.com>
Subject: Re: [RFC PATCH v9 7/8] ptp: arm64: Enable ptp_kvm for arm64
Date: Fri, 10 Jan 2020 10:35:02 +0000	[thread overview]
Message-ID: <a85deebc23c1fa77e6f70b6eaef22a34@kernel.org> (raw)
In-Reply-To: <HE1PR0801MB16765B52E5DCD8EA480EDABFF4380@HE1PR0801MB1676.eurprd08.prod.outlook.com>

Hi Jianyong,

On 2020-01-10 10:15, Jianyong Wu wrote:
> Hi Marc,

[...]

>> >> > +	ktime_overall = hvc_res.a0 << 32 | hvc_res.a1;
>> >> > +	*ts = ktime_to_timespec64(ktime_overall);
>> >> > +	*cycle = hvc_res.a2 << 32 | hvc_res.a3;
>> >>
>> >> So why isn't that just a read of the virtual counter, given that what
>> >> you do in the hypervisor seems to be "cntpct - cntvoff"?
>> >>
>> >> What am I missing here?
>> >>
>> > We need get clock time and counter cycle at the same time, so we can't
>> > just read virtual counter at guest and must get it from host.
>> 
>> See my comment in my reply to patch #6: *Must* seems like a very 
>> strong
>> word, and you don't explain *why* that's better than just computing 
>> the
>> total hypercall cost. Hint: given the frequency of the counter (in the 
>> few MHz
>> range) vs the frequency of a CPU (in the multiple GHz range, and with 
>> an IPC
>> close enough to 1), I doubt that you'll see the counter making much 
>> progress
>> across a hypercall.
>> 
> Sorry, I will avoid to use those strong words.
> 
> It's really the case that the hypercall won't across cycle in general.
> But sometimes, kernel preempt
> may happen in the middle of the hypercall which we can't assume how
> long before schedule back. so it's better capture them
> together at the same time.

Fair enough. Please document the rational, as I guess others will ask
the same questions.

Then the problem to solve is that of the reference counter, as you so 
far
assume the virtual counter. I guess you need to be able to let the guest
select the reference counter when calling the PTP service.

[...]

> By the way, does nested virtualization diff between arm64 and arm32?

There is no nested virt for 32bit (it is explicitly forbidden by the
architecture).

         M.
-- 
Jazz is not dead. It just smells funny...