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 dpoll -2 offset 0

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 <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/hv/hv_util.c

@@ -27,6 +27,8 @@
 #include <linux/sysctl.h>
 #include <linux/reboot.h>
 #include <linux/hyperv.h>
+#include <linux/clockchips.h>
+#include <linux/ptp_clock_kernel.h>
 #include <asm/mshyperv.h>
 
 #include "hyperv_vmbus.h"
@@ -210,29 +212,15 @@ struct adj_time_work {
 
 static void hv_set_host_time(struct work_struct *work)
 {
-	struct adj_time_work	*wrk;
-	s64 host_tns;
-	u64 newtime;
+	struct adj_time_work *wrk;
 	struct timespec64 host_ts;
+	u64 reftime, 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
-		 * 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;
-
-		hv_get_current_tick(current_tick);
-		newtime += (current_tick - wrk->ref_time);
-	}
-	host_tns = (newtime - WLTIMEDELTA) * 100;
-	host_ts = ns_to_timespec64(host_tns);
+	reftime = hyperv_cs->read(hyperv_cs);
+	newtime = wrk->host_time + (reftime - wrk->ref_time);
+	host_ts = ns_to_timespec64((newtime - WLTIMEDELTA) * 100);
 
 	do_settimeofday64(&host_ts);
 }
@@ -251,22 +239,60 @@ 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;
+	struct system_time_snapshot	snap;
+	spinlock_t			lock;
+} host_ts;
+
+static inline void adj_guesttime(u64 hosttime, u64 reftime, u8 adj_flags)
 {
+	unsigned long flags;
+	u64 cur_reftime;
 
 	/*
 	 * 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;
-
-	wrk.host_time = hosttime;
-	wrk.ref_time = reftime;
-	wrk.flags = flags;
-	if ((flags & (ICTIMESYNCFLAG_SYNC | ICTIMESYNCFLAG_SAMPLE)) != 0) {
+	if (adj_flags & ICTIMESYNCFLAG_SYNC) {
+		/* Queue a job to do do_settimeofday64() */
+		if (work_pending(&wrk.work))
+			return;
+
+		wrk.host_time = hosttime;
+		wrk.ref_time = reftime;
+		wrk.flags = adj_flags;
 		schedule_work(&wrk.work);
+	} else {
+		/*
+		 * Save the adjusted time sample from the host and the snapshot
+		 * of the current system time for PTP device.
+		 */
+		spin_lock_irqsave(&host_ts.lock, flags);
+
+		cur_reftime = hyperv_cs->read(hyperv_cs);
+		host_ts.host_time = hosttime;
+		host_ts.ref_time = cur_reftime;
+		ktime_get_snapshot(&host_ts.snap);
+
+		/*
+		 * TimeSync v4 messages contain reference time (guest's Hyper-V
+		 * clocksource read when the time sample was generated), we can
+		 * improve the precision by adding the delta between now and the
+		 * time of generation.
+		 */
+		if (ts_srv_version > TS_VERSION_3)
+			host_ts.host_time += (cur_reftime - reftime);
+
+		spin_unlock_irqrestore(&host_ts.lock, flags);
 	}
 }
 
@@ -479,14 +505,113 @@ 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)
+{
+	unsigned long flags;
+	u64 newtime, reftime;
+
+	spin_lock_irqsave(&host_ts.lock, flags);
+	reftime = hyperv_cs->read(hyperv_cs);
+	newtime = host_ts.host_time + (reftime - host_ts.ref_time);
+	*ts = ns_to_timespec64((newtime - WLTIMEDELTA) * 100);
+	spin_unlock_irqrestore(&host_ts.lock, flags);
+
+	return 0;
+}
+
+static int hv_ptp_get_syncdevicetime(ktime_t *device,
+				     struct system_counterval_t *system,
+				     void *ctx)
+{
+	system->cs = hyperv_cs;
+	system->cycles = host_ts.ref_time;
+	*device = ns_to_ktime((host_ts.host_time - WLTIMEDELTA) * 100);
+
+	return 0;
+}
+
+static int hv_ptp_getcrosststamp(struct ptp_clock_info *ptp,
+				 struct system_device_crosststamp *xtstamp)
+{
+	unsigned long flags;
+	int ret;
+
+	spin_lock_irqsave(&host_ts.lock, flags);
+
+	/*
+	 * host_ts contains the last time sample from the host and the snapshot
+	 * of system time. We don't need to calculate the time delta between
+	 * the reception and now as get_device_system_crosststamp() does the
+	 * required interpolation.
+	 */
+	ret = get_device_system_crosststamp(hv_ptp_get_syncdevicetime,
+					    NULL, &host_ts.snap, xtstamp);
+
+	spin_unlock_irqrestore(&host_ts.lock, flags);
+
+	return ret;
+}
+
+static 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,
+	.getcrosststamp = hv_ptp_getcrosststamp,
+	.settime64      = hv_ptp_settime,
+	.owner		= THIS_MODULE,
+};
+
+static struct ptp_clock *hv_ptp_clock;
+
 static int hv_timesync_init(struct hv_util_service *srv)
 {
+	/* TimeSync requires Hyper-V clocksource. */
+	if (!hyperv_cs)
+		return -ENODEV;
+
 	INIT_WORK(&wrk.work, hv_set_host_time);
+
+	/*
+	 * ptp_clock_register() returns NULL when CONFIG_PTP_1588_CLOCK is
+	 * 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);
 }