hv_util: adjust system time smoothly

Hi Vitaly,

[auto build test WARNING on linus/master]
[also build test WARNING on v4.10-rc4 next-20170119]
[if your patch is applied to the wrong git tree, please drop us a note to help improve the system]

url: https://github.com/0day-ci/linux/commits/Vitaly-Kuznetsov/hv_util-adjust-system-time-smoothly/20170120-011342
config: x86_64-rhel (attached as .config)
compiler: gcc-6 (Debian 6.2.0-3) 6.2.0 20160901 reproduce:

        # save the attached .config to linux build tree
        make ARCH=x86_64 

All warnings (new ones prefixed by >>):

drivers/hv/hv_util.c: In function 'hv_timesync_init':

Quoted 1 line

Quoted 1 line

drivers/hv/hv_util.c:561:36: warning: passing argument 1 of 'ptp_clock_register' discards 'const' qualifier from pointer target type [-Wdiscarded-qualifiers]

     hv_ptp_clock = ptp_clock_register(&ptp_hyperv_info, NULL);
                                       ^

In file included from drivers/hv/hv_util.c:30:0:
include/linux/ptp_clock_kernel.h:172:26: note: expected 'struct ptp_clock_info *' but argument is of type 'const struct ptp_clock_info *'
extern struct ptp_clock *ptp_clock_register(struct ptp_clock_info *info,
^~~~~~~~~~~~~~~~~~

With TimeSync version 4 protocol support we started updating system time continuously through the whole lifetime of Hyper-V guests. Every 5 seconds there is a time sample from the host which triggers do_settimeofday[64](). While the time from the host is very accurate such adjustments may cause issues:
- Time is jumping forward and backward, some applications may misbehave.
- In case an NTP server runs in parallel and uses something else for time sync (network, PTP,...) system time will never converge.
- Systemd starts annoying you by printing "Time has been changed" every 5 seconds to the system log.

Instead of doing in-kernel time adjustments offload the work to an NTP client by exposing TimeSync messages as a PTP device. Users may now decide what they want to use as a source.

I tested the solution with chrony, the config was:

refclock PHC /dev/ptp0 poll 3 precision 1e-9

The result I'm seeing is accurate enough, the time delta between the guest and the host is almost always within [-10us, +10us], the in-kernel solution was giving us comparable results.

I also tried implementing PPS device instead of PTP by using not currently used Hyper-V synthetic timers (we use only one of four for clockevent) but with PPS source only chrony wasn't able to give me the required accuracy, the delta often more that 100us.

Signed-off-by Vitaly Kuznetsov <[email protected]>

---

2017-01-19 15:16+0100, Vitaly Kuznetsov:

Quoted 23 lines

With TimeSync version 4 protocol support we started updating system time continuously through the whole lifetime of Hyper-V guests. Every 5 seconds there is a time sample from the host which triggers do_settimeofday[64](). While the time from the host is very accurate such adjustments may cause issues:
- Time is jumping forward and backward, some applications may misbehave.
- In case an NTP server runs in parallel and uses something else for time sync (network, PTP,...) system time will never converge.
- Systemd starts annoying you by printing "Time has been changed" every 5 seconds to the system log.

Instead of doing in-kernel time adjustments offload the work to an NTP client by exposing TimeSync messages as a PTP device. Users may now decide what they want to use as a source.

I tested the solution with chrony, the config was:

refclock PHC /dev/ptp0 poll 3 precision 1e-9

The result I'm seeing is accurate enough, the time delta between the guest and the host is almost always within [-10us, +10us], the in-kernel solution was giving us comparable results.

Quoted 7 lines

I also tried implementing PPS device instead of PTP by using not currently used Hyper-V synthetic timers (we use only one of four for clockevent) but with PPS source only chrony wasn't able to give me the required accuracy, the delta often more that 100us.

Signed-off-by Vitaly Kuznetsov <[email protected]>

---

It is a nice coincidence that KVM is working on a PTP driver as well,
https://lkml.org/lkml/2017/1/20/247, and it uses more precise/accurate method of converting the host time that Hyper-V could also use.

Hyper-V provides {host_time, ref_time} tuple, but gettime64() requires that you return just host_time and a new "ref_time" is then computed to be in the middle of two guest_time reads.
I recommend you use getcrosststamp PTP callback, which allows you to provide the tuple. Userspace can then use PTP_SYS_OFFSET_PRECISE ioctl.

KVM patches also proposes to change PTP_SYS_OFFSET, so when gettime64 callback is not implemented, the ioctl uses getcrosststamp instead, which would avoid code duplication and improve precision/accuracy.

-----Original Message-----
From: Vitaly Kuznetsov [mailto:[email protected]]
Sent: Thursday, January 19, 2017 6:17 AM
To: [email protected]
Cc: Thomas Gleixner <[email protected]>; [email protected]; Haiyang Zhang <[email protected]>; KY Srinivasan <[email protected]>; John Stultz <[email protected]>; Alex Ng (LIS) <[email protected]>; Stephen Hemminger <[email protected]>; Olaf Hering <[email protected]>; Richard Cochran <[email protected]>; Radim Krcmar <[email protected]>
Subject: [PATCH v4 2/2] hv_utils: implement Hyper-V PTP source

With TimeSync version 4 protocol support we started updating system time continuously through the whole lifetime of Hyper-V guests. Every 5 seconds there is a time sample from the host which triggers do_settimeofday[64](). While the time from the host is very accurate such adjustments may cause issues:
- Time is jumping forward and backward, some applications may misbehave.
- In case an NTP server runs in parallel and uses something else for time sync (network, PTP,...) system time will never converge.
- Systemd starts annoying you by printing "Time has been changed" every 5 seconds to the system log.

Instead of doing in-kernel time adjustments offload the work to an NTP client by exposing TimeSync messages as a PTP device. Users may now decide what they want to use as a source.

I tested the solution with chrony, the config was:

refclock PHC /dev/ptp0 poll 3 precision 1e-9

The result I'm seeing is accurate enough, the time delta between the guest and the host is almost always within [-10us, +10us], the in-kernel solution was giving us comparable results.

I also tried implementing PPS device instead of PTP by using not currently used Hyper-V synthetic timers (we use only one of four for clockevent) but with PPS source only chrony wasn't able to give me the required accuracy, the delta often more that 100us.

Signed-off-by Vitaly Kuznetsov <[email protected]>

---
 drivers/hv/hv_util.c | 144
++++++++++++++++++++++++++++++++++++++++++---------
 1 file changed, 119 insertions(+), 25 deletions(-)

drivers/hv/hv_util.c +20 -6

index 94719eb..e1211b8 100644
@@ -27,6 +27,7 @@
 #include <linux/sysctl.h>
 #include <linux/reboot.h>
 #include <linux/hyperv.h>
+#include <linux/ptp_clock_kernel.h>
 
 #include "hyperv_vmbus.h"
 

@@ -179,31 +180,34 @@ struct adj_time_work {
 	u8	flags;
 };
 
+static u64 get_timeadj_latency(u64 ref_time)
+{
+	u64 current_tick;
+
+	if (ts_srv_version <= TS_VERSION_3)
+		return 0;
+
+	/*
+	 * Some latency has been introduced since Hyper-V generated
+	 * its time sample. Take that latency into account before
+	 * using TSC reference time sample from Hyper-V.
+	 *
+	 * This sample is given by TimeSync v4 and above hosts.
+	 */
+	rdmsrl(HV_X64_MSR_TIME_REF_COUNT, current_tick);
+	return current_tick - ref_time;
+}
+
 static void hv_set_host_time(struct work_struct *work)
 {
 	struct adj_time_work	*wrk;
-	s64 host_tns;
-	u64 newtime;
 	struct timespec64 host_ts;
+	u64 newtime;
 
 	wrk = container_of(work, struct adj_time_work, work);
 
-	newtime = wrk->host_time;
-	if (ts_srv_version > TS_VERSION_3) {
-		/*
-		 * Some latency has been introduced since Hyper-V

generated

drivers/hv/hv_util.c +29 -19

-		 * its time sample. Take that latency into account before
-		 * using TSC reference time sample from Hyper-V.
-		 *
-		 * This sample is given by TimeSync v4 and above hosts.
-		 */
-		u64 current_tick;
-
-		rdmsrl(HV_X64_MSR_TIME_REF_COUNT, current_tick);
-		newtime += (current_tick - wrk->ref_time);
-	}
-	host_tns = (newtime - WLTIMEDELTA) * 100;
-	host_ts = ns_to_timespec64(host_tns);
+	newtime = wrk->host_time + get_timeadj_latency(wrk->ref_time);
+	host_ts = ns_to_timespec64((newtime - WLTIMEDELTA) * 100);
 
 	do_settimeofday64(&host_ts);
 }

@@ -222,22 +226,52 @@ static void hv_set_host_time(struct work_struct
 *work)
  * to discipline the clock.
  */
 static struct adj_time_work  wrk;
-static inline void adj_guesttime(u64 hosttime, u64 reftime, u8 flags)
+
+/*
+ * The last time sample, received from the host. PTP device responds to
+ * requests by using this data and the current partition-wide time reference
+ * count.
+ */
+static struct {
+	u64		host_time;
+	u64		ref_time;
+	spinlock_t	lock;
+} host_ts;
+
+static inline void adj_guesttime(u64 hosttime, u64 reftime, u8 adj_flags)
 {
+	unsigned long flags;
 
 	/*
 	 * This check is safe since we are executing in the
 	 * interrupt context and time synch messages arre always
 	 * delivered on the same CPU.
 	 */
-	if (work_pending(&wrk.work))
-		return;
+	if (adj_flags & ICTIMESYNCFLAG_SYNC) {
+		if (work_pending(&wrk.work))
+			return;
 
-	wrk.host_time = hosttime;
-	wrk.ref_time = reftime;
-	wrk.flags = flags;
-	if ((flags & (ICTIMESYNCFLAG_SYNC | ICTIMESYNCFLAG_SAMPLE)) !=
 0) {
+		wrk.host_time = hosttime;
+		wrk.ref_time = reftime;
+		wrk.flags = adj_flags;
 		schedule_work(&wrk.work);
+	} else {
+		spin_lock_irqsave(&host_ts.lock, flags);
+		host_ts.host_time = hosttime;
+
+		/*
+		 * Prior to version 4 TimeSync messages from the host don't
+		 * contain any reference time (the time when the time

sample

drivers/hv/hv_util.c +3 -0

+		 * was generated), save the current time reference count
+		 * instead. This adds a small delta between the time sample
+		 * generation and the reception of the sample here to the

result

drivers/hv/hv_util.c +4 -0

+		 * but it's the best thing we can do.
+		 */
+		if (ts_srv_version <= TS_VERSION_3)
+			rdmsrl(HV_X64_MSR_TIME_REF_COUNT,

host_ts.ref_time);

drivers/hv/hv_util.c +51 -0

+		else
+			host_ts.ref_time = reftime;
+		spin_unlock_irqrestore(&host_ts.lock, flags);
 	}
 }

@@ -470,14 +504,74 @@ static  struct hv_driver util_drv = {
 	.remove =  util_remove,
 };
 
+static int hv_ptp_enable(struct ptp_clock_info *info,
+			 struct ptp_clock_request *request, int on)
+{
+	return -EOPNOTSUPP;
+}
+
+static int hv_ptp_settime(struct ptp_clock_info *p, const struct timespec64
 *ts)
+{
+	return -EOPNOTSUPP;
+}
+
+static int hv_ptp_adjfreq(struct ptp_clock_info *ptp, s32 delta)
+{
+	return -EOPNOTSUPP;
+}
+static int hv_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta)
+{
+	return -EOPNOTSUPP;
+}
+
+static int hv_ptp_gettime(struct ptp_clock_info *info, struct timespec64
 *ts)
+{
+	u64 newtime;
+	unsigned long flags;
+
+	spin_lock_irqsave(&host_ts.lock, flags);
+	newtime = host_ts.host_time +
 get_timeadj_latency(host_ts.ref_time);
+	*ts = ns_to_timespec64((newtime - WLTIMEDELTA) * 100);
+	spin_unlock_irqrestore(&host_ts.lock, flags);
+
+	return 0;
+}
+
+static const struct ptp_clock_info ptp_hyperv_info = {
+	.name		= "hyperv",
+	.enable         = hv_ptp_enable,
+	.adjtime        = hv_ptp_adjtime,
+	.adjfreq        = hv_ptp_adjfreq,
+	.gettime64      = hv_ptp_gettime,
+	.settime64      = hv_ptp_settime,
+	.owner		= THIS_MODULE,
+};
+
+static struct ptp_clock *hv_ptp_clock;
+
 static int hv_timesync_init(struct hv_util_service *srv)
 {
 	INIT_WORK(&wrk.work, hv_set_host_time);
+
+	/*
+	 * ptp_clock_register() returns NULL when

CONFIG_PTP_1588_CLOCK is

drivers/hv/hv_util.c +12 -1

+	 * disabled but the driver is still useful without the PTP device
+	 * as it still handles the ICTIMESYNCFLAG_SYNC case.
+	 */
+	hv_ptp_clock = ptp_clock_register(&ptp_hyperv_info, NULL);
+	if (IS_ERR_OR_NULL(hv_ptp_clock)) {
+		pr_err("cannot register PTP clock: %ld\n",
+		       PTR_ERR(hv_ptp_clock));
+		hv_ptp_clock = NULL;
+	}
+
 	return 0;
 }
 
 static void hv_timesync_deinit(void)
 {
+	if (hv_ptp_clock)
+		ptp_clock_unregister(hv_ptp_clock);
 	cancel_work_sync(&wrk.work);
 }
 
--

2.9.3

Quoted 41 lines

-----Original Message-----
From: Vitaly Kuznetsov [mailto:[email protected]]
Sent: Thursday, January 19, 2017 6:17 AM
To: [email protected]
Cc: Thomas Gleixner <[email protected]>; [email protected]; Haiyang Zhang <[email protected]>; KY Srinivasan <[email protected]>; John Stultz <[email protected]>; Alex Ng (LIS) <[email protected]>; Stephen Hemminger <[email protected]>; Olaf Hering <[email protected]>; Richard Cochran <[email protected]>; Radim Krcmar <[email protected]>
Subject: [PATCH v4 2/2] hv_utils: implement Hyper-V PTP source

With TimeSync version 4 protocol support we started updating system time continuously through the whole lifetime of Hyper-V guests. Every 5 seconds there is a time sample from the host which triggers do_settimeofday[64](). While the time from the host is very accurate such adjustments may cause issues:
- Time is jumping forward and backward, some applications may misbehave.
- In case an NTP server runs in parallel and uses something else for time sync (network, PTP,...) system time will never converge.
- Systemd starts annoying you by printing "Time has been changed" every 5 seconds to the system log.

Instead of doing in-kernel time adjustments offload the work to an NTP client by exposing TimeSync messages as a PTP device. Users may now decide what they want to use as a source.

I tested the solution with chrony, the config was:

refclock PHC /dev/ptp0 poll 3 precision 1e-9

The result I'm seeing is accurate enough, the time delta between the guest and the host is almost always within [-10us, +10us], the in-kernel solution was giving us comparable results.

I also tried implementing PPS device instead of PTP by using not currently used Hyper-V synthetic timers (we use only one of four for clockevent) but with PPS source only chrony wasn't able to give me the required accuracy, the delta often more that 100us.

Quoted 1 line

Signed-off-by Vitaly Kuznetsov <[email protected]>

Vitaly,

This patch does not apply. Please rebase and send.