time: Add history to cross timestamp interface supporting slower devices

Another representative use case of time sync and the correlated
clocksource (in addition to PTP noted above) is PTP synchronized
audio.

In a streaming application, as an example, samples will be sent and/or
received by multiple devices with a presentation time that is in terms
of the PTP master clock. Synchronizing the audio output on these
devices requires correlating the audio clock with the PTP master
clock. The more precise this correlation is, the better the audio
quality (i.e. out of sync audio sounds bad).

From an application standpoint, to correlate the PTP master clock with
the audio device clock, the system clock is used as a intermediate
timebase. The transforms such an application would perform are:

    System Clock <-> Audio clock
    System Clock <-> Network Device Clock [<-> PTP Master Clock]

Modern Intel platforms can perform a more accurate cross timestamp in
hardware (ART,audio device clock).  The audio driver requires
ART->system time transforms -- the same as required for the network
driver. These platforms offload audio processing (including
cross-timestamps) to a DSP which to ensure uninterrupted audio
processing, communicates and response to the host only once every
millsecond. As a result is takes up to a millisecond for the DSP to
receive a request, the request is processed by the DSP, the audio
output hardware is polled for completion, the result is copied into
shared memory, and the host is notified. All of these operation occur
on a millisecond cadence.  This transaction requires about 2 ms, but
under heavier workloads it may take up to 4 ms.

Adding a history allows these slow devices the option of providing an
ART value outside of the current interval. In this case, the callback
provided is an accessor function for the previously obtained counter
value. If get_system_device_crosststamp() receives a counter value
previous to cycle_last, it consults the history provided as an
argument in history_ref and interpolates the realtime and monotonic
raw system time using the provided counter value. If there are any
clock discontinuities, e.g. from calling settimeofday(), the monotonic
raw time is interpolated in the usual way, but the realtime clock time
is adjusted by scaling the monotonic raw adjustment.

When an accessor function is used a history argument *must* be
provided. The history is initialized using ktime_get_snapshot() and
must be called before the counter values are read.

Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: kevin.b.stanton@intel.com
Cc: kevin.j.clarke@intel.com
Cc: hpa@zytor.com
Cc: jeffrey.t.kirsher@intel.com
Cc: netdev@vger.kernel.org
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Christopher S. Hall <christopher.s.hall@intel.com>
[jstultz: Fixed up cycles_t/cycle_t type confusion]
Signed-off-by: John Stultz <john.stultz@linaro.org>

include/linux/timekeeper_internal.h

@@ -50,6 +50,7 @@ struct tk_read_base {
  * @offs_tai:		Offset clock monotonic -> clock tai
  * @tai_offset:		The current UTC to TAI offset in seconds
  * @clock_was_set_seq:	The sequence number of clock was set events
+ * @cs_was_changed_seq:	The sequence number of clocksource change events
  * @next_leap_ktime:	CLOCK_MONOTONIC time value of a pending leap-second
  * @raw_time:		Monotonic raw base time in timespec64 format
  * @cycle_interval:	Number of clock cycles in one NTP interval
@@ -91,6 +92,7 @@ struct timekeeper {
 	ktime_t			offs_tai;
 	s32			tai_offset;
 	unsigned int		clock_was_set_seq;
+	u8			cs_was_changed_seq;
 	ktime_t			next_leap_ktime;
 	struct timespec64	raw_time;
 

include/linux/timekeeping.h

@@ -272,11 +272,15 @@ extern void ktime_get_raw_and_real_ts64(struct timespec64 *ts_raw,
  * @cycles:	Clocksource counter value to produce the system times
  * @real:	Realtime system time
  * @raw:	Monotonic raw system time
+ * @clock_was_set_seq:	The sequence number of clock was set events
+ * @cs_was_changed_seq:	The sequence number of clocksource change events
  */
 struct system_time_snapshot {
 	cycle_t		cycles;
 	ktime_t		real;
 	ktime_t		raw;
+	unsigned int	clock_was_set_seq;
+	u8		cs_was_changed_seq;
 };
 
 /*
@@ -312,6 +316,7 @@ extern int get_device_system_crosststamp(
 				struct system_counterval_t *system_counterval,
 				void *ctx),
 			void *ctx,
+			struct system_time_snapshot *history,
 			struct system_device_crosststamp *xtstamp);
 
 /*

kernel/time/timekeeping.c

@@ -233,6 +233,7 @@ static void tk_setup_internals(struct timekeeper *tk, struct clocksource *clock)
 	u64 tmp, ntpinterval;
 	struct clocksource *old_clock;
 
+	++tk->cs_was_changed_seq;
 	old_clock = tk->tkr_mono.clock;
 	tk->tkr_mono.clock = clock;
 	tk->tkr_mono.read = clock->read;
@@ -894,6 +895,8 @@ void ktime_get_snapshot(struct system_time_snapshot *systime_snapshot)
 		seq = read_seqcount_begin(&tk_core.seq);
 
 		now = tk->tkr_mono.read(tk->tkr_mono.clock);
+		systime_snapshot->cs_was_changed_seq = tk->cs_was_changed_seq;
+		systime_snapshot->clock_was_set_seq = tk->clock_was_set_seq;
 		base_real = ktime_add(tk->tkr_mono.base,
 				      tk_core.timekeeper.offs_real);
 		base_raw = tk->tkr_raw.base;
@@ -907,10 +910,123 @@ void ktime_get_snapshot(struct system_time_snapshot *systime_snapshot)
 }
 EXPORT_SYMBOL_GPL(ktime_get_snapshot);
 
+/* Scale base by mult/div checking for overflow */
+static int scale64_check_overflow(u64 mult, u64 div, u64 *base)
+{
+	u64 tmp, rem;
+
+	tmp = div64_u64_rem(*base, div, &rem);
+
+	if (((int)sizeof(u64)*8 - fls64(mult) < fls64(tmp)) ||
+	    ((int)sizeof(u64)*8 - fls64(mult) < fls64(rem)))
+		return -EOVERFLOW;
+	tmp *= mult;
+	rem *= mult;
+
+	do_div(rem, div);
+	*base = tmp + rem;
+	return 0;
+}
+
+/**
+ * adjust_historical_crosststamp - adjust crosstimestamp previous to current interval
+ * @history:			Snapshot representing start of history
+ * @partial_history_cycles:	Cycle offset into history (fractional part)
+ * @total_history_cycles:	Total history length in cycles
+ * @discontinuity:		True indicates clock was set on history period
+ * @ts:				Cross timestamp that should be adjusted using
+ *	partial/total ratio
+ *
+ * Helper function used by get_device_system_crosststamp() to correct the
+ * crosstimestamp corresponding to the start of the current interval to the
+ * system counter value (timestamp point) provided by the driver. The
+ * total_history_* quantities are the total history starting at the provided
+ * reference point and ending at the start of the current interval. The cycle
+ * count between the driver timestamp point and the start of the current
+ * interval is partial_history_cycles.
+ */
+static int adjust_historical_crosststamp(struct system_time_snapshot *history,
+					 cycle_t partial_history_cycles,
+					 cycle_t total_history_cycles,
+					 bool discontinuity,
+					 struct system_device_crosststamp *ts)
+{
+	struct timekeeper *tk = &tk_core.timekeeper;
+	u64 corr_raw, corr_real;
+	bool interp_forward;
+	int ret;
+
+	if (total_history_cycles == 0 || partial_history_cycles == 0)
+		return 0;
+
+	/* Interpolate shortest distance from beginning or end of history */
+	interp_forward = partial_history_cycles > total_history_cycles/2 ?
+		true : false;
+	partial_history_cycles = interp_forward ?
+		total_history_cycles - partial_history_cycles :
+		partial_history_cycles;
+
+	/*
+	 * Scale the monotonic raw time delta by:
+	 *	partial_history_cycles / total_history_cycles
+	 */
+	corr_raw = (u64)ktime_to_ns(
+		ktime_sub(ts->sys_monoraw, history->raw));
+	ret = scale64_check_overflow(partial_history_cycles,
+				     total_history_cycles, &corr_raw);
+	if (ret)
+		return ret;
+
+	/*
+	 * If there is a discontinuity in the history, scale monotonic raw
+	 *	correction by:
+	 *	mult(real)/mult(raw) yielding the realtime correction
+	 * Otherwise, calculate the realtime correction similar to monotonic
+	 *	raw calculation
+	 */
+	if (discontinuity) {
+		corr_real = mul_u64_u32_div
+			(corr_raw, tk->tkr_mono.mult, tk->tkr_raw.mult);
+	} else {
+		corr_real = (u64)ktime_to_ns(
+			ktime_sub(ts->sys_realtime, history->real));
+		ret = scale64_check_overflow(partial_history_cycles,
+					     total_history_cycles, &corr_real);
+		if (ret)
+			return ret;
+	}
+
+	/* Fixup monotonic raw and real time time values */
+	if (interp_forward) {
+		ts->sys_monoraw = ktime_add_ns(history->raw, corr_raw);
+		ts->sys_realtime = ktime_add_ns(history->real, corr_real);
+	} else {
+		ts->sys_monoraw = ktime_sub_ns(ts->sys_monoraw, corr_raw);
+		ts->sys_realtime = ktime_sub_ns(ts->sys_realtime, corr_real);
+	}
+
+	return 0;
+}
+
+/*
+ * cycle_between - true if test occurs chronologically between before and after
+ */
+static bool cycle_between(cycle_t before, cycle_t test, cycle_t after)
+{
+	if (test > before && test < after)
+		return true;
+	if (test < before && before > after)
+		return true;
+	return false;
+}
+
 /**
  * get_device_system_crosststamp - Synchronously capture system/device timestamp
- * @sync_devicetime:	Callback to get simultaneous device time and
+ * @get_time_fn:	Callback to get simultaneous device time and
  *	system counter from the device driver
+ * @ctx:		Context passed to get_time_fn()
+ * @history_begin:	Historical reference point used to interpolate system
+ *	time when counter provided by the driver is before the current interval
  * @xtstamp:		Receives simultaneously captured system and device time
  *
  * Reads a timestamp from a device and correlates it to system time
@@ -920,13 +1036,18 @@ int get_device_system_crosststamp(int (*get_time_fn)
 				   struct system_counterval_t *sys_counterval,
 				   void *ctx),
 				  void *ctx,
+				  struct system_time_snapshot *history_begin,
 				  struct system_device_crosststamp *xtstamp)
 {
 	struct system_counterval_t system_counterval;
 	struct timekeeper *tk = &tk_core.timekeeper;
+	cycle_t cycles, now, interval_start;
+	unsigned int clock_was_set_seq;
 	ktime_t base_real, base_raw;
 	s64 nsec_real, nsec_raw;
+	u8 cs_was_changed_seq;
 	unsigned long seq;
+	bool do_interp;
 	int ret;
 
 	do {
@@ -946,6 +1067,22 @@ int get_device_system_crosststamp(int (*get_time_fn)
 		 */
 		if (tk->tkr_mono.clock != system_counterval.cs)
 			return -ENODEV;
+		cycles = system_counterval.cycles;
+
+		/*
+		 * Check whether the system counter value provided by the
+		 * device driver is on the current timekeeping interval.
+		 */
+		now = tk->tkr_mono.read(tk->tkr_mono.clock);
+		interval_start = tk->tkr_mono.cycle_last;
+		if (!cycle_between(interval_start, cycles, now)) {
+			clock_was_set_seq = tk->clock_was_set_seq;
+			cs_was_changed_seq = tk->cs_was_changed_seq;
+			cycles = interval_start;
+			do_interp = true;
+		} else {
+			do_interp = false;
+		}
 
 		base_real = ktime_add(tk->tkr_mono.base,
 				      tk_core.timekeeper.offs_real);
@@ -959,6 +1096,38 @@ int get_device_system_crosststamp(int (*get_time_fn)
 
 	xtstamp->sys_realtime = ktime_add_ns(base_real, nsec_real);
 	xtstamp->sys_monoraw = ktime_add_ns(base_raw, nsec_raw);
+
+	/*
+	 * Interpolate if necessary, adjusting back from the start of the
+	 * current interval
+	 */
+	if (do_interp) {
+		cycle_t partial_history_cycles, total_history_cycles;
+		bool discontinuity;
+
+		/*
+		 * Check that the counter value occurs after the provided
+		 * history reference and that the history doesn't cross a
+		 * clocksource change
+		 */
+		if (!history_begin ||
+		    !cycle_between(history_begin->cycles,
+				   system_counterval.cycles, cycles) ||
+		    history_begin->cs_was_changed_seq != cs_was_changed_seq)
+			return -EINVAL;
+		partial_history_cycles = cycles - system_counterval.cycles;
+		total_history_cycles = cycles - history_begin->cycles;
+		discontinuity =
+			history_begin->clock_was_set_seq != clock_was_set_seq;
+
+		ret = adjust_historical_crosststamp(history_begin,
+						    partial_history_cycles,
+						    total_history_cycles,
+						    discontinuity, xtstamp);
+		if (ret)
+			return ret;
+	}
+
 	return 0;
 }
 EXPORT_SYMBOL_GPL(get_device_system_crosststamp);