All of lore.kernel.org
 help / color / mirror / Atom feed
From: Vitaly Kuznetsov <vkuznets@redhat.com>
To: Thomas Gleixner <tglx@linutronix.de>
Cc: Stephen Hemminger <sthemmin@microsoft.com>,
	Haiyang Zhang <haiyangz@microsoft.com>,
	x86@kernel.org, linux-kernel@vger.kernel.org,
	Andy Lutomirski <luto@amacapital.net>,
	Ingo Molnar <mingo@redhat.com>, "H. Peter Anvin" <hpa@zytor.com>,
	devel@linuxdriverproject.org,
	virtualization@lists.linux-foundation.org
Subject: Re: [PATCH 2/2] x86/vdso: Add VCLOCK_HVCLOCK vDSO clock read method
Date: Fri, 10 Feb 2017 12:06:47 +0100	[thread overview]
Message-ID: <87lgteqp88.fsf@vitty.brq.redhat.com> (raw)
In-Reply-To: <alpine.DEB.2.20.1702091806400.3604@nanos> (Thomas Gleixner's message of "Thu, 9 Feb 2017 18:08:22 +0100 (CET)")

Thomas Gleixner <tglx@linutronix.de> writes:

> On Thu, 9 Feb 2017, Vitaly Kuznetsov wrote:
>> +#ifdef CONFIG_HYPERV_TSCPAGE
>> +static notrace u64 vread_hvclock(int *mode)
>> +{
>> +	const struct ms_hyperv_tsc_page *tsc_pg =
>> +		(const struct ms_hyperv_tsc_page *)&hvclock_page;
>> +	u64 sequence, scale, offset, current_tick, cur_tsc;
>> +
>> +	while (1) {
>> +		sequence = READ_ONCE(tsc_pg->tsc_sequence);
>> +		if (!sequence)
>> +			break;
>> +
>> +		scale = READ_ONCE(tsc_pg->tsc_scale);
>> +		offset = READ_ONCE(tsc_pg->tsc_offset);
>> +		rdtscll(cur_tsc);
>> +
>> +		current_tick = mul_u64_u64_shr(cur_tsc, scale, 64) + offset;
>> +
>> +		if (READ_ONCE(tsc_pg->tsc_sequence) == sequence)
>> +			return current_tick;
>
> That sequence stuff lacks still a sensible explanation. It's fundamentally
> different from the sequence counting we do in the kernel, so documentation
> for it is really required.

Sure, do you think the following would do?

diff --git a/arch/x86/entry/vdso/vclock_gettime.c b/arch/x86/entry/vdso/vclock_gettime.c
index 4af10b4..886b600 100644
--- a/arch/x86/entry/vdso/vclock_gettime.c
+++ b/arch/x86/entry/vdso/vclock_gettime.c
@@ -154,6 +154,22 @@ static notrace u64 vread_hvclock(int *mode)
                (const struct ms_hyperv_tsc_page *)&hvclock_page;
        u64 sequence, scale, offset, current_tick, cur_tsc;
 
+       /*
+        * The protocol for reading Hyper-V TSC page is specified in Hypervisor
+        * Top-Level Functional Specification ver. 3.0 and above. To get the
+        * reference time we must do the following:
+        * - READ ReferenceTscSequence
+        *   A special '0' value indicates the time source is unreliable and we
+        *   need to use something else. The currently published specification
+        *   versions (up to 4.0b) contain a mistake and wrongly claim '-1'
+        *   instead of '0' as the special value, see commit c35b82ef0294.
+        * - ReferenceTime =
+        *        ((RDTSC() * ReferenceTscScale) >> 64) + ReferenceTscOffset
+        * - READ ReferenceTscSequence again. In case its value has changed
+        *   since our first reading we need to discard ReferenceTime and repeat
+        *   the whole sequence as the hypervisor was updating the page in
+        *   between.
+        */
        while (1) {
                sequence = READ_ONCE(tsc_pg->tsc_sequence);
                if (!sequence)

-- 
  Vitaly

WARNING: multiple messages have this Message-ID (diff)
From: Vitaly Kuznetsov <vkuznets@redhat.com>
To: Thomas Gleixner <tglx@linutronix.de>
Cc: x86@kernel.org, Andy Lutomirski <luto@amacapital.net>,
	Ingo Molnar <mingo@redhat.com>, "H. Peter Anvin" <hpa@zytor.com>,
	"K. Y. Srinivasan" <kys@microsoft.com>,
	Haiyang Zhang <haiyangz@microsoft.com>,
	Stephen Hemminger <sthemmin@microsoft.com>,
	Dexuan Cui <decui@microsoft.com>,
	linux-kernel@vger.kernel.org, devel@linuxdriverproject.org,
	virtualization@lists.linux-foundation.org
Subject: Re: [PATCH 2/2] x86/vdso: Add VCLOCK_HVCLOCK vDSO clock read method
Date: Fri, 10 Feb 2017 12:06:47 +0100	[thread overview]
Message-ID: <87lgteqp88.fsf@vitty.brq.redhat.com> (raw)
In-Reply-To: <alpine.DEB.2.20.1702091806400.3604@nanos> (Thomas Gleixner's message of "Thu, 9 Feb 2017 18:08:22 +0100 (CET)")

Thomas Gleixner <tglx@linutronix.de> writes:

> On Thu, 9 Feb 2017, Vitaly Kuznetsov wrote:
>> +#ifdef CONFIG_HYPERV_TSCPAGE
>> +static notrace u64 vread_hvclock(int *mode)
>> +{
>> +	const struct ms_hyperv_tsc_page *tsc_pg =
>> +		(const struct ms_hyperv_tsc_page *)&hvclock_page;
>> +	u64 sequence, scale, offset, current_tick, cur_tsc;
>> +
>> +	while (1) {
>> +		sequence = READ_ONCE(tsc_pg->tsc_sequence);
>> +		if (!sequence)
>> +			break;
>> +
>> +		scale = READ_ONCE(tsc_pg->tsc_scale);
>> +		offset = READ_ONCE(tsc_pg->tsc_offset);
>> +		rdtscll(cur_tsc);
>> +
>> +		current_tick = mul_u64_u64_shr(cur_tsc, scale, 64) + offset;
>> +
>> +		if (READ_ONCE(tsc_pg->tsc_sequence) == sequence)
>> +			return current_tick;
>
> That sequence stuff lacks still a sensible explanation. It's fundamentally
> different from the sequence counting we do in the kernel, so documentation
> for it is really required.

Sure, do you think the following would do?

diff --git a/arch/x86/entry/vdso/vclock_gettime.c b/arch/x86/entry/vdso/vclock_gettime.c
index 4af10b4..886b600 100644
--- a/arch/x86/entry/vdso/vclock_gettime.c
+++ b/arch/x86/entry/vdso/vclock_gettime.c
@@ -154,6 +154,22 @@ static notrace u64 vread_hvclock(int *mode)
                (const struct ms_hyperv_tsc_page *)&hvclock_page;
        u64 sequence, scale, offset, current_tick, cur_tsc;
 
+       /*
+        * The protocol for reading Hyper-V TSC page is specified in Hypervisor
+        * Top-Level Functional Specification ver. 3.0 and above. To get the
+        * reference time we must do the following:
+        * - READ ReferenceTscSequence
+        *   A special '0' value indicates the time source is unreliable and we
+        *   need to use something else. The currently published specification
+        *   versions (up to 4.0b) contain a mistake and wrongly claim '-1'
+        *   instead of '0' as the special value, see commit c35b82ef0294.
+        * - ReferenceTime =
+        *        ((RDTSC() * ReferenceTscScale) >> 64) + ReferenceTscOffset
+        * - READ ReferenceTscSequence again. In case its value has changed
+        *   since our first reading we need to discard ReferenceTime and repeat
+        *   the whole sequence as the hypervisor was updating the page in
+        *   between.
+        */
        while (1) {
                sequence = READ_ONCE(tsc_pg->tsc_sequence);
                if (!sequence)

-- 
  Vitaly

  parent reply	other threads:[~2017-02-10 11:06 UTC|newest]

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-02-09 14:10 [PATCH 0/2] x86/vdso: Add Hyper-V TSC page clocksource support Vitaly Kuznetsov
2017-02-09 14:10 ` [PATCH 1/2] hyperv: implement hv_get_tsc_page() Vitaly Kuznetsov
2017-02-09 18:24   ` Stephen Hemminger via Virtualization
2017-02-09 18:24     ` Stephen Hemminger
2017-02-09 20:14     ` Thomas Gleixner
2017-02-09 20:14       ` Thomas Gleixner
2017-02-09 23:17       ` Stephen Hemminger
2017-02-09 23:17       ` Stephen Hemminger
2017-02-09 14:10 ` Vitaly Kuznetsov
2017-02-09 14:10 ` [PATCH 2/2] x86/vdso: Add VCLOCK_HVCLOCK vDSO clock read method Vitaly Kuznetsov
2017-02-09 14:10 ` Vitaly Kuznetsov
2017-02-09 17:08   ` Thomas Gleixner
2017-02-09 18:27     ` Stephen Hemminger via Virtualization
2017-02-09 18:27       ` Stephen Hemminger
2017-02-10 12:25       ` Vitaly Kuznetsov
2017-02-10 12:25         ` Vitaly Kuznetsov
2017-02-10 12:28         ` Thomas Gleixner
2017-02-10 12:28           ` Thomas Gleixner
2017-02-10 16:31           ` Stephen Hemminger via Virtualization
2017-02-10 16:31           ` Stephen Hemminger
2017-02-10 18:01             ` Thomas Gleixner
2017-02-10 18:01               ` Thomas Gleixner
2017-02-13  7:49               ` Dexuan Cui via Virtualization
2017-02-13  7:49                 ` Dexuan Cui
2017-02-13  9:27                 ` Thomas Gleixner
2017-02-13  9:27                   ` Thomas Gleixner
2017-02-13 19:06                 ` Andy Lutomirski
2017-02-13 19:06                   ` Andy Lutomirski
2017-02-13 19:28                   ` Thomas Gleixner
2017-02-13 19:28                   ` Thomas Gleixner
2017-02-09 20:45     ` KY Srinivasan
2017-02-09 22:55       ` Andy Lutomirski
2017-02-09 22:55         ` Andy Lutomirski
2017-02-09 23:15         ` Stephen Hemminger
2017-02-09 23:15         ` Stephen Hemminger
2017-02-10 12:15         ` Vitaly Kuznetsov
2017-02-10 12:15           ` Vitaly Kuznetsov
2017-02-09 20:45     ` KY Srinivasan via Virtualization
2017-02-10 11:06     ` Vitaly Kuznetsov [this message]
2017-02-10 11:06       ` Vitaly Kuznetsov
2017-02-10 11:15       ` Thomas Gleixner
2017-02-10 11:15         ` Thomas Gleixner
2017-02-09 17:08   ` Thomas Gleixner

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=87lgteqp88.fsf@vitty.brq.redhat.com \
    --to=vkuznets@redhat.com \
    --cc=devel@linuxdriverproject.org \
    --cc=haiyangz@microsoft.com \
    --cc=hpa@zytor.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=luto@amacapital.net \
    --cc=mingo@redhat.com \
    --cc=sthemmin@microsoft.com \
    --cc=tglx@linutronix.de \
    --cc=virtualization@lists.linux-foundation.org \
    --cc=x86@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.