|
| 1 | +.. _timeutil_api: |
| 2 | + |
| 3 | +Time Utilities |
| 4 | +############## |
| 5 | + |
| 6 | +Overview |
| 7 | +******** |
| 8 | + |
| 9 | +:ref:`kernel_timing_uptime` in Zephyr is based on the a tick counter. With |
| 10 | +the default :option:`CONFIG_TICKLESS_KERNEL` this counter advances at a |
| 11 | +nominally constant rate from zero at the instant the system started. The POSIX |
| 12 | +equivalent to this counter is something like ``CLOCK_MONOTONIC`` or, in Linux, |
| 13 | +``CLOCK_MONOTONIC_RAW``. :c:func:`k_uptime_get()` provides a millisecond |
| 14 | +representation of this time. |
| 15 | + |
| 16 | +Applications often need to correlate the Zephyr internal time with external |
| 17 | +time scales used in daily life, such as local time or Coordinated Universal |
| 18 | +Time. These systems interpret time in different ways and may have |
| 19 | +discontinuities due to `leap seconds <https://what-if.xkcd.com/26/>`__ and |
| 20 | +local time offsets like daylight saving time. |
| 21 | + |
| 22 | +Because of these discontinuities, as well as significant inaccuracies in the |
| 23 | +clocks underlying the cycle counter, the offset between time estimated from |
| 24 | +the Zephyr clock and the actual time in a "real" civil time scale is not |
| 25 | +constant and can vary widely over the runtime of a Zephyr application. |
| 26 | + |
| 27 | +The time utilities API supports: |
| 28 | + |
| 29 | +* :ref:`converting between time representations <timeutil_repr>` |
| 30 | +* :ref:`synchronizing and aligning time scales <timeutil_sync>` |
| 31 | + |
| 32 | +For terminology and concepts that support these functions see |
| 33 | +:ref:`timeutil_concepts`. |
| 34 | + |
| 35 | +Time Utility APIs |
| 36 | +***************** |
| 37 | + |
| 38 | +.. _timeutil_repr: |
| 39 | + |
| 40 | +Representation Transformation |
| 41 | +============================= |
| 42 | + |
| 43 | +Time scale instants can be represented in multiple ways including: |
| 44 | + |
| 45 | +* Seconds since an epoch. POSIX representations of time in this form include |
| 46 | + ``time_t`` and ``struct timespec``, which are generally interpreted as a |
| 47 | + representation of `"UNIX Time" |
| 48 | + <https://tools.ietf.org/html/rfc8536#section-2>`__. |
| 49 | + |
| 50 | +* Calendar time as a year, month, day, hour, minutes, and seconds relative to |
| 51 | + an epoch. POSIX representations of time in this form include ``struct tm``. |
| 52 | + |
| 53 | +Keep in mind that these are simply time representations that must be |
| 54 | +interpreted relative to a time scale which may be local time, UTC, or some |
| 55 | +other continuous or discontinuous scale. |
| 56 | + |
| 57 | +Some necessary transformations are available in standard C library |
| 58 | +routines. For example, ``time_t`` measuring seconds since the POSIX EPOCH is |
| 59 | +converted to ``struct tm`` representing calendar time with `gmtime() |
| 60 | +<https://pubs.opengroup.org/onlinepubs/9699919799/functions/gmtime.html>`__. |
| 61 | +Sub-second timestamps like ``struct timespec`` can also use this to produce |
| 62 | +the calendar time representation and deal with sub-second offsets separately. |
| 63 | + |
| 64 | +The inverse transformation is not standardized: APIs like ``mktime()`` expect |
| 65 | +information about time zones. Zephyr provides this transformation with |
| 66 | +:c:func:`timeutil_timegm` and :c:func:`timeutil_timegm64`. |
| 67 | + |
| 68 | +.. doxygengroup:: timeutil_repr_apis |
| 69 | + :project: Zephyr |
| 70 | + |
| 71 | +.. _timeutil_sync: |
| 72 | + |
| 73 | +Time Scale Synchronization |
| 74 | +========================== |
| 75 | + |
| 76 | +There are several factors that affect synchronizing time scales: |
| 77 | + |
| 78 | +* The rate of discrete instant representation change. For example Zephyr |
| 79 | + uptime is tracked in ticks which advance at events that nominally occur at |
| 80 | + :option:`CONFIG_SYS_CLOCK_TICKS_PER_SEC` Hertz, while an external time |
| 81 | + source may provide data in whole or fractional seconds (e.g. microseconds). |
| 82 | +* The absolute offset required to align the two scales at a single instant. |
| 83 | +* The relative error between observable instants in each scale, required to |
| 84 | + align multiple instants consistently. For example a reference clock that's |
| 85 | + conditioned by a 1-pulse-per-second GPS signal will be much more accurate |
| 86 | + than a Zephyr system clock driven by a RC oscillator with a +/- 250 ppm |
| 87 | + error. |
| 88 | + |
| 89 | +Synchronization or alignment between time scales is done with a multi-step |
| 90 | +process: |
| 91 | + |
| 92 | +* An instant in a time scale is represented by an (unsigned) 64-bit integer, |
| 93 | + assumed to advance at a fixed nominal rate. |
| 94 | +* :c:struct:`timeutil_sync_config` records the nominal rates of a reference |
| 95 | + time scale/source (e.g. TAI) and a local time source |
| 96 | + (e.g. :c:func:`k_uptime_ticks`). |
| 97 | +* :c:struct:`timeutil_sync_instant` records the representation of a single |
| 98 | + instant in both the reference and local time scales. |
| 99 | +* :c:struct:`timeutil_sync_state` provides storage for an initial instant, a |
| 100 | + recently received second observation, and a skew that can adjust for |
| 101 | + relative errors in the actual rate of each time scale. |
| 102 | +* :c:func:`timeutil_sync_ref_from_local()` and |
| 103 | + :c:func:`timeutil_sync_local_from_ref()` convert instants in one time scale |
| 104 | + to another taking into account skew that can be estimated from the two |
| 105 | + instances stored in the state structure by |
| 106 | + :c:func:`timeutil_sync_estimate_skew`. |
| 107 | + |
| 108 | +.. doxygengroup:: timeutil_sync_apis |
| 109 | + :project: Zephyr |
| 110 | + |
| 111 | +.. _timeutil_concepts: |
| 112 | + |
| 113 | +Concepts Underlying Time Support in Zephyr |
| 114 | +****************************************** |
| 115 | + |
| 116 | +Terms from `ISO/TC 154/WG 5 N0038 |
| 117 | +<https://www.loc.gov/standards/datetime/iso-tc154-wg5_n0038_iso_wd_8601-1_2016-02-16.pdf>`__ |
| 118 | +(ISO/WD 8601-1) and elsewhere: |
| 119 | + |
| 120 | +* A *time axis* is a representation of time as an ordered sequence of |
| 121 | + instants. |
| 122 | +* A *time scale* is a way of representing an instant relative to an origin |
| 123 | + that serves as the epoch. |
| 124 | +* A time scale is *monotonic* (increasing) if the representation of successive |
| 125 | + time instants never decreases in value. |
| 126 | +* A time scale is *continuous* if the representation has no abrupt changes in |
| 127 | + value, e.g. jumping forward or back when going between successive instants. |
| 128 | +* `Civil time <https://en.wikipedia.org/wiki/Civil_time>`__ generally refers |
| 129 | + to time scales that legally defined by civil authorities, like local |
| 130 | + governments, often to align local midnight to solar time. |
| 131 | + |
| 132 | +Relevant Time Scales |
| 133 | +==================== |
| 134 | + |
| 135 | +`International Atomic Time |
| 136 | +<https://en.wikipedia.org/wiki/International_Atomic_Time>`__ (TAI) is a time |
| 137 | +scale based on averaging clocks that count in SI seconds. TAI is a montonic |
| 138 | +and continuous time scale. |
| 139 | + |
| 140 | +`Universal Time <https://en.wikipedia.org/wiki/Universal_Time>`__ (UT) is a |
| 141 | +time scale based on Earth’s rotation. UT is a discontinuous time scale as it |
| 142 | +requires occasional adjustments (`leap seconds |
| 143 | +<https://en.wikipedia.org/wiki/Leap_second>`__) to maintain alignment to |
| 144 | +changes in Earth’s rotation. Thus the difference between TAI and UT varies |
| 145 | +over time. There are several variants of UT, with `UTC |
| 146 | +<https://en.wikipedia.org/wiki/Coordinated_Universal_Time>`__ being the most |
| 147 | +common. |
| 148 | + |
| 149 | +UT times are independent of location. UT is the basis for Standard Time |
| 150 | +(or "local time") which is the time at a particular location. Standard |
| 151 | +time has a fixed offset from UT at any given instant, primarily |
| 152 | +influenced by longitude, but the offset may be adjusted ("daylight |
| 153 | +saving time") to align standard time to the local solar time. In a sense |
| 154 | +local time is "more discontinuous" than UT. |
| 155 | + |
| 156 | +`POSIX Time <https://tools.ietf.org/html/rfc8536#section-2>`__ is a time scale |
| 157 | +that counts seconds since the "POSIX epoch" at 1970-01-01T00:00:00Z (i.e. the |
| 158 | +start of 1970 UTC). `UNIX Time |
| 159 | +<https://tools.ietf.org/html/rfc8536#section-2>`__ is an extension of POSIX |
| 160 | +time using negative values to represent times before the POSIX epoch. Both of |
| 161 | +these scales assume that every day has exactly 86400 seconds. In normal use |
| 162 | +instants in these scales correspond to times in the UTC scale, so they inherit |
| 163 | +the discontinuity. |
| 164 | + |
| 165 | +The continuous analogue is `UNIX Leap Time |
| 166 | +<https://tools.ietf.org/html/rfc8536#section-2>`__ which is UNIX time plus all |
| 167 | +leap-second corrections added after the POSIX epoch (when TAI-UTC was 8 s). |
| 168 | + |
| 169 | +Example of Time Scale Differences |
| 170 | +--------------------------------- |
| 171 | + |
| 172 | +A positive leap second was introduced at the end of 2016, increasing the |
| 173 | +difference between TAI and UTC from 36 seconds to 37 seconds. There was |
| 174 | +no leap second introduced at the end of 1999, when the difference |
| 175 | +between TAI and UTC was only 32 seconds. The following table shows |
| 176 | +relevant civil and epoch times in several scales: |
| 177 | + |
| 178 | +==================== ========== =================== ======= ============== |
| 179 | +UTC Date UNIX time TAI Date TAI-UTC UNIX Leap Time |
| 180 | +==================== ========== =================== ======= ============== |
| 181 | +1970-01-01T00:00:00Z 0 1970-01-01T00:00:08 +8 0 |
| 182 | +1999-12-31T23:59:28Z 946684768 2000-01-01T00:00:00 +32 946684792 |
| 183 | +1999-12-31T23:59:59Z 946684799 2000-01-01T00:00:31 +32 946684823 |
| 184 | +2000-01-01T00:00:00Z 946684800 2000-01-01T00:00:32 +32 946684824 |
| 185 | +2016-12-31T23:59:59Z 1483228799 2017-01-01T00:00:35 +36 1483228827 |
| 186 | +2016-12-31T23:59:60Z undefined 2017-01-01T00:00:36 +36 1483228828 |
| 187 | +2017-01-01T00:00:00Z 1483228800 2017-01-01T00:00:37 +37 1483228829 |
| 188 | +==================== ========== =================== ======= ============== |
| 189 | + |
| 190 | +Functional Requirements |
| 191 | +----------------------- |
| 192 | + |
| 193 | +The Zephyr tick counter has no concept of leap seconds or standard time |
| 194 | +offsets and is a continuous time scale. However it can be relatively |
| 195 | +inaccurate, with drifts as much as three minutes per hour (assuming an RC |
| 196 | +timer with 5% tolerance). |
| 197 | + |
| 198 | +There are two stages required to support conversion between Zephyr time and |
| 199 | +common human time scales: |
| 200 | + |
| 201 | +* Translation between the continuous but inaccurate Zephyr time scale and an |
| 202 | + accurate external stable time scale; |
| 203 | +* Translation between the stable time scale and the (possibly discontinuous) |
| 204 | + civil time scale. |
| 205 | + |
| 206 | +The API around :c:func:`timeutil_sync_state_update()` supports the first step |
| 207 | +of converting between continuous time scales. |
| 208 | + |
| 209 | +The second step requires external information including schedules of leap |
| 210 | +seconds and local time offset changes. This may be best provided by an |
| 211 | +external library, and is not currently part of the time utility APIs. |
| 212 | + |
| 213 | +Selecting an External Source and Time Scale |
| 214 | +------------------------------------------- |
| 215 | + |
| 216 | +If an application requires civil time accuracy within several seconds then UTC |
| 217 | +could be used as the stable time source. However, if the external source |
| 218 | +adjusts to a leap second there will be a discontinuity: the elapsed time |
| 219 | +between two observations taken at 1 Hz is not equal to the numeric difference |
| 220 | +between their timestamps. |
| 221 | + |
| 222 | +For precise activities a continuous scale that is independent of local and |
| 223 | +solar adjustments simplifies things considerably. Suitable continuous scales |
| 224 | +include: |
| 225 | + |
| 226 | +- GPS time: epoch of 1980-01-06T00:00:00Z, continuous following TAI with an |
| 227 | + offset of TAI-GPS=19 s. |
| 228 | +- Bluetooth mesh time: epoch of 2000-01-01T00:00:00Z, continuous following TAI |
| 229 | + with an offset of -32. |
| 230 | +- UNIX Leap Time: epoch of 1970-01-01T00:00:00Z, continuous following TAI with |
| 231 | + an offset of -8. |
| 232 | + |
| 233 | +Because C and Zephyr library functions support conversion between integral and |
| 234 | +calendar time representations using the UNIX epoch, UNIX Leap Time is an ideal |
| 235 | +choice for the external time scale. |
| 236 | + |
| 237 | +The mechanism used to populate synchronization points is not relevant: it may |
| 238 | +involve reading from a local high-precision RTC peripheral, exchanging packets |
| 239 | +over a network using a protocol like NTP or PTP, or processing NMEA messages |
| 240 | +received a GPS with or without a 1pps signal. |
0 commit comments