Quelle TimeStamp.h Sprache: C

/* -*- Mode: C++; tab-width: 8; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
/* vim: set ts=8 sts=2 et sw=2 tw=80: */
/* This Source Code Form is subject to the terms of the Mozilla Public
* License, v. 2.0. If a copy of the MPL was not distributed with this
* file, You can obtain one at http://mozilla.org/MPL/2.0/. */

#ifndef mozilla_TimeStamp_h
#define mozilla_TimeStamp_h

#include "mozilla/Assertions.h"
#include "mozilla/Attributes.h"
#include "mozilla/FloatingPoint.h"
#include "mozilla/Types.h"
#include <algorithm>  // for std::min, std::max
#include <ostream>
#include <stdint.h>
#include <type_traits>

namespace IPC {
template <typename T>
struct ParamTraits;
}  // namespace IPC

#ifdef XP_WIN
// defines TimeStampValue as a complex value keeping both
// GetTickCount and QueryPerformanceCounter values
#  include "TimeStamp_windows.h"

#  include "mozilla/Maybe.h"  // For TimeStamp::RawQueryPerformanceCounterValue
#endif

namespace mozilla {

#ifndef XP_WIN
typedef uint64_t TimeStampValue;
#endif

class TimeStamp;
class TimeStampTests;

/**
* Platform-specific implementation details of BaseTimeDuration.
*/
class BaseTimeDurationPlatformUtils {
public:
  static MFBT_API double ToSeconds(int64_t aTicks);
  static MFBT_API double ToSecondsSigDigits(int64_t aTicks);
  static MFBT_API int64_t TicksFromMilliseconds(double aMilliseconds);
  static MFBT_API int64_t ResolutionInTicks();
};

/**
* Instances of this class represent the length of an interval of time.
* Negative durations are allowed, meaning the end is before the start.
*
* Internally the duration is stored as a int64_t in units of
* PR_TicksPerSecond() when building with NSPR interval timers, or a
* system-dependent unit when building with system clocks.  The
* system-dependent unit must be constant, otherwise the semantics of
* this class would be broken.
*
* The ValueCalculator template parameter determines how arithmetic
* operations are performed on the integer count of ticks (mValue).
*/
template <typename ValueCalculator>
class BaseTimeDuration {
public:
  // The default duration is 0.
  constexpr BaseTimeDuration() : mValue(0) {}
  // Allow construction using '0' as the initial value, for readability,
  // but no other numbers (so we don't have any implicit unit conversions).
  struct _SomethingVeryRandomHere;
  MOZ_IMPLICIT BaseTimeDuration(_SomethingVeryRandomHere* aZero) : mValue(0) {
    MOZ_ASSERT(!aZero, "Who's playing funny games here?");
  }
  // Default copy-constructor and assignment are OK

  // Converting copy-constructor and assignment operator
  template <typename E>
  explicit BaseTimeDuration(const BaseTimeDuration<E>& aOther)
      : mValue(aOther.mValue) {}

  template <typename E>
  BaseTimeDuration& operator=(const BaseTimeDuration<E>& aOther) {
    mValue = aOther.mValue;
    return *this;
  }

  double ToSeconds() const {
    if (mValue == INT64_MAX) {
      return PositiveInfinity<double>();
    }
    if (mValue == INT64_MIN) {
      return NegativeInfinity<double>();
    }
    return BaseTimeDurationPlatformUtils::ToSeconds(mValue);
  }
  // Return a duration value that includes digits of time we think to
  // be significant.  This method should be used when displaying a
  // time to humans.
  double ToSecondsSigDigits() const {
    if (mValue == INT64_MAX) {
      return PositiveInfinity<double>();
    }
    if (mValue == INT64_MIN) {
      return NegativeInfinity<double>();
    }
    return BaseTimeDurationPlatformUtils::ToSecondsSigDigits(mValue);
  }
  double ToMilliseconds() const { return ToSeconds() * 1000.0; }
  double ToMicroseconds() const { return ToMilliseconds() * 1000.0; }

  // Using a double here is safe enough; with 53 bits we can represent
  // durations up to over 280,000 years exactly.  If the units of
  // mValue do not allow us to represent durations of that length,
  // long durations are clamped to the max/min representable value
  // instead of overflowing.
  static inline BaseTimeDuration FromSeconds(double aSeconds) {
    return FromMilliseconds(aSeconds * 1000.0);
  }
  static BaseTimeDuration FromMilliseconds(double aMilliseconds) {
    if (aMilliseconds == PositiveInfinity<double>()) {
      return Forever();
    }
    if (aMilliseconds == NegativeInfinity<double>()) {
      return FromTicks(INT64_MIN);
    }
    return FromTicks(
        BaseTimeDurationPlatformUtils::TicksFromMilliseconds(aMilliseconds));
  }
  static inline BaseTimeDuration FromMicroseconds(double aMicroseconds) {
    return FromMilliseconds(aMicroseconds / 1000.0);
  }

  static constexpr BaseTimeDuration Zero() { return BaseTimeDuration(); }
  static constexpr BaseTimeDuration Forever() { return FromTicks(INT64_MAX); }

  BaseTimeDuration operator+(const BaseTimeDuration& aOther) const {
    return FromTicks(ValueCalculator::Add(mValue, aOther.mValue));
  }
  BaseTimeDuration operator-(const BaseTimeDuration& aOther) const {
    return FromTicks(ValueCalculator::Subtract(mValue, aOther.mValue));
  }
  BaseTimeDuration& operator+=(const BaseTimeDuration& aOther) {
    mValue = ValueCalculator::Add(mValue, aOther.mValue);
    return *this;
  }
  BaseTimeDuration& operator-=(const BaseTimeDuration& aOther) {
    mValue = ValueCalculator::Subtract(mValue, aOther.mValue);
    return *this;
  }
  BaseTimeDuration operator-() const {
    // We don't just use FromTicks(ValueCalculator::Subtract(0, mValue))
    // since that won't give the correct result for -TimeDuration::Forever().
    int64_t ticks;
    if (mValue == INT64_MAX) {
      ticks = INT64_MIN;
    } else if (mValue == INT64_MIN) {
      ticks = INT64_MAX;
    } else {
      ticks = -mValue;
    }

    return FromTicks(ticks);
  }

  static BaseTimeDuration Max(const BaseTimeDuration& aA,
                              const BaseTimeDuration& aB) {
    return FromTicks(std::max(aA.mValue, aB.mValue));
  }
  static BaseTimeDuration Min(const BaseTimeDuration& aA,
                              const BaseTimeDuration& aB) {
    return FromTicks(std::min(aA.mValue, aB.mValue));
  }

#if defined(DEBUG)
  int64_t GetValue() const { return mValue; }
#endif

private:
  // Block double multiplier (slower, imprecise if long duration) - Bug 853398.
  // If required, use MultDouble explicitly and with care.
  BaseTimeDuration operator*(const double aMultiplier) const = delete;

  // Block double divisor (for the same reason, and because dividing by
  // fractional values would otherwise invoke the int64_t variant, and rounding
  // the passed argument can then cause divide-by-zero) - Bug 1147491.
  BaseTimeDuration operator/(const double aDivisor) const = delete;

public:
  BaseTimeDuration MultDouble(double aMultiplier) const {
    return FromTicks(ValueCalculator::Multiply(mValue, aMultiplier));
  }
  BaseTimeDuration operator*(const int32_t aMultiplier) const {
    return FromTicks(ValueCalculator::Multiply(mValue, aMultiplier));
  }
  BaseTimeDuration operator*(const uint32_t aMultiplier) const {
    return FromTicks(ValueCalculator::Multiply(mValue, aMultiplier));
  }
  BaseTimeDuration operator*(const int64_t aMultiplier) const {
    return FromTicks(ValueCalculator::Multiply(mValue, aMultiplier));
  }
  BaseTimeDuration operator*(const uint64_t aMultiplier) const {
    if (aMultiplier > INT64_MAX) {
      return Forever();
    }
    return FromTicks(ValueCalculator::Multiply(mValue, aMultiplier));
  }
  BaseTimeDuration operator/(const int64_t aDivisor) const {
    MOZ_ASSERT(aDivisor != 0, "Division by zero");
    return FromTicks(ValueCalculator::Divide(mValue, aDivisor));
  }
  double operator/(const BaseTimeDuration& aOther) const {
    MOZ_ASSERT(aOther.mValue != 0, "Division by zero");
    return ValueCalculator::DivideDouble(mValue, aOther.mValue);
  }
  BaseTimeDuration operator%(const BaseTimeDuration& aOther) const {
    MOZ_ASSERT(aOther.mValue != 0, "Division by zero");
    return FromTicks(ValueCalculator::Modulo(mValue, aOther.mValue));
  }

  template <typename E>
  bool operator<(const BaseTimeDuration<E>& aOther) const {
    return mValue < aOther.mValue;
  }
  template <typename E>
  bool operator<=(const BaseTimeDuration<E>& aOther) const {
    return mValue <= aOther.mValue;
  }
  template <typename E>
  bool operator>=(const BaseTimeDuration<E>& aOther) const {
    return mValue >= aOther.mValue;
  }
  template <typename E>
  bool operator>(const BaseTimeDuration<E>& aOther) const {
    return mValue > aOther.mValue;
  }
  template <typename E>
  bool operator==(const BaseTimeDuration<E>& aOther) const {
    return mValue == aOther.mValue;
  }
  template <typename E>
  bool operator!=(const BaseTimeDuration<E>& aOther) const {
    return mValue != aOther.mValue;
  }
  bool IsZero() const { return mValue == 0; }
  explicit operator bool() const { return mValue != 0; }

  friend std::ostream& operator<<(std::ostream& aStream,
                                  const BaseTimeDuration& aDuration) {
    return aStream << aDuration.ToMilliseconds() << " ms";
  }

  // Return a best guess at the system's current timing resolution,
  // which might be variable.  BaseTimeDurations below this order of
  // magnitude are meaningless, and those at the same order of
  // magnitude or just above are suspect.
  static BaseTimeDuration Resolution() {
    return FromTicks(BaseTimeDurationPlatformUtils::ResolutionInTicks());
  }

  // We could define additional operators here:
  // -- convert to/from other time units
  // -- scale duration by a float
  // but let's do that on demand.
  // Comparing durations for equality will only lead to bugs on
  // platforms with high-resolution timers.

private:
  friend class TimeStamp;
  friend struct IPC::ParamTraits<mozilla::BaseTimeDuration<ValueCalculator>>;
  template <typename>
  friend class BaseTimeDuration;

  static constexpr BaseTimeDuration FromTicks(int64_t aTicks) {
    BaseTimeDuration t;
    t.mValue = aTicks;
    return t;
  }

  static BaseTimeDuration FromTicks(double aTicks) {
    // NOTE: this MUST be a >= test, because int64_t(double(INT64_MAX))
    // overflows and gives INT64_MIN.
    if (aTicks >= double(INT64_MAX)) {
      return FromTicks(INT64_MAX);
    }

    // This MUST be a <= test.
    if (aTicks <= double(INT64_MIN)) {
      return FromTicks(INT64_MIN);
    }

    return FromTicks(int64_t(aTicks));
  }

  // Duration, result is implementation-specific difference of two TimeStamps
  int64_t mValue;
};

/**
* Perform arithmetic operations on the value of a BaseTimeDuration without
* doing strict checks on the range of values.
*/
class TimeDurationValueCalculator {
public:
  static int64_t Add(int64_t aA, int64_t aB) { return aA + aB; }
  static int64_t Subtract(int64_t aA, int64_t aB) { return aA - aB; }

  template <typename T>
  static int64_t Multiply(int64_t aA, T aB) {
    static_assert(std::is_integral_v<T>,
                  "Using integer multiplication routine with non-integer type."
                  " Further specialization required");
    return aA * static_cast<int64_t>(aB);
  }

  static int64_t Divide(int64_t aA, int64_t aB) { return aA / aB; }
  static double DivideDouble(int64_t aA, int64_t aB) {
    return static_cast<double>(aA) / aB;
  }
  static int64_t Modulo(int64_t aA, int64_t aB) { return aA % aB; }
};

template <>
inline int64_t TimeDurationValueCalculator::Multiply<double>(int64_t aA,
                                                             double aB) {
  return static_cast<int64_t>(aA * aB);
}

/**
* Specialization of BaseTimeDuration that uses TimeDurationValueCalculator for
* arithmetic on the mValue member.
*
* Use this class for time durations that are *not* expected to hold values of
* Forever (or the negative equivalent) or when such time duration are *not*
* expected to be used in arithmetic operations.
*/
typedef BaseTimeDuration<TimeDurationValueCalculator> TimeDuration;

/**
* Instances of this class represent moments in time, or a special
* "null" moment. We do not use the non-monotonic system clock or
* local time, since they can be reset, causing apparent backward
* travel in time, which can confuse algorithms. Instead we measure
* elapsed time according to the system.  This time can never go
* backwards (i.e. it never wraps around, at least not in less than
* five million years of system elapsed time). It might not advance
* while the system is sleeping. If TimeStamp::SetNow() is not called
* at all for hours or days, we might not notice the passage of some
* of that time.
*
* We deliberately do not expose a way to convert TimeStamps to some
* particular unit. All you can do is compute a difference between two
* TimeStamps to get a TimeDuration. You can also add a TimeDuration
* to a TimeStamp to get a new TimeStamp. You can't do something
* meaningless like add two TimeStamps.
*
* Internally this is implemented as either a wrapper around
*   - high-resolution, monotonic, system clocks if they exist on this
*     platform
*   - PRIntervalTime otherwise.  We detect wraparounds of
*     PRIntervalTime and work around them.
*
* This class is similar to C++11's time_point, however it is
* explicitly nullable and provides an IsNull() method. time_point
* is initialized to the clock's epoch and provides a
* time_since_epoch() method that functions similiarly. i.e.
* t.IsNull() is equivalent to t.time_since_epoch() ==
* decltype(t)::duration::zero();
*
* Note that, since TimeStamp objects are small, prefer to pass them by value
* unless there is a specific reason not to do so.
*/
#if defined(XP_WIN)
// If this static_assert fails then possibly the warning comment below is no
// longer valid and should be removed.
static_assert(sizeof(TimeStampValue) > 8);
#endif
/*
* WARNING: On Windows, each TimeStamp is represented internally by two
* different raw values (one from GTC and one from QPC) and which value gets
* used for a given operation depends on whether both operands have QPC values
* or not. This duality of values can lead to some surprising results when
* mixing TimeStamps with and without QPC values, such as comparisons being
* non-transitive (ie, a > b > c might not imply a > c). See bug 1829983 for
* more details/an example.
*/
class TimeStamp {
public:
  using DurationType = TimeDuration;
  /**
   * Initialize to the "null" moment
   */
  constexpr TimeStamp() : mValue(0) {}
  // Default copy-constructor and assignment are OK

  /**
   * The system timestamps are the same as the TimeStamp
   * retrieved by mozilla::TimeStamp. Since we need this for
   * vsync timestamps, we enable the creation of mozilla::TimeStamps
   * on platforms that support vsync aligned refresh drivers / compositors
   * Verified true as of Jan 31, 2015: B2G and OS X
   * False on Windows 7
   * Android's event time uses CLOCK_MONOTONIC via SystemClock.uptimeMilles.
   * So it is same value of TimeStamp posix implementation.
   * Wayland/GTK event time also uses CLOCK_MONOTONIC on Weston/Mutter
   * compositors.
   * UNTESTED ON OTHER PLATFORMS
   */
#if defined(XP_DARWIN) || defined(MOZ_WIDGET_ANDROID) || defined(MOZ_WIDGET_GTK)
  static TimeStamp FromSystemTime(int64_t aSystemTime) {
    static_assert(sizeof(aSystemTime) == sizeof(TimeStampValue),
                  "System timestamp should be same units as TimeStampValue");
    return TimeStamp(aSystemTime);
  }
#endif

  /**
   * Return true if this is the "null" moment
   */
  constexpr bool IsNull() const { return mValue == 0; }

  /**
   * Return true if this is not the "null" moment, may be used in tests, e.g.:
   * |if (timestamp) { ... }|
   */
  explicit operator bool() const { return mValue != 0; }

  /**
   * Return a timestamp reflecting the current elapsed system time. This
   * is monotonically increasing (i.e., does not decrease) over the
   * lifetime of this process' XPCOM session.
   *
   * Now() is trying to ensure the best possible precision on each platform,
   * at least one millisecond.
   *
   * NowLoRes() has been introduced to workaround performance problems of
   * QueryPerformanceCounter on the Windows platform.  NowLoRes() is giving
   * lower precision, usually 15.6 ms, but with very good performance benefit.
   * Use it for measurements of longer times, like >200ms timeouts.
   */
  static TimeStamp Now() { return Now(true); }
  static TimeStamp NowLoRes() { return Now(false); }

  /**
   * Return a timestamp representing the time when the current process was
   * created which will be comparable with other timestamps taken with this
   * class.
   *
   * @returns A timestamp representing the time when the process was created
   */
  static MFBT_API TimeStamp ProcessCreation();

  /**
   * Return the very first timestamp that was taken. This can be used instead
   * of TimeStamp::ProcessCreation() by code that might not allow running the
   * complex logic required to compute the real process creation. This will
   * necessarily have been recorded sometimes after TimeStamp::ProcessCreation()
   * or at best should be equal to it.
   *
   * @returns The first tiemstamp that was taken by this process
   */
  static MFBT_API TimeStamp FirstTimeStamp();

  /**
   * Records a process restart. After this call ProcessCreation() will return
   * the time when the browser was restarted instead of the actual time when
   * the process was created.
   */
  static MFBT_API void RecordProcessRestart();

#ifdef XP_LINUX
  uint64_t RawClockMonotonicNanosecondsSinceBoot() const {
    return static_cast<uint64_t>(mValue);
  }
#endif

#ifdef XP_DARWIN
  // Returns the number of nanoseconds since the mach_absolute_time origin.
  MFBT_API uint64_t RawMachAbsoluteTimeNanoseconds() const;
#endif

#ifdef XP_WIN
  Maybe<uint64_t> RawQueryPerformanceCounterValue() const {
    // mQPC is stored in `mt` i.e. QueryPerformanceCounter * 1000
    // so divide out the 1000
    return mValue.mHasQPC ? Some(mValue.mQPC / 1000ULL) : Nothing();
  }
#endif

  /**
   * Compute the difference between two timestamps. Both must be non-null.
   */
  TimeDuration operator-(const TimeStamp& aOther) const {
    MOZ_ASSERT(!IsNull(), "Cannot compute with a null value");
    MOZ_ASSERT(!aOther.IsNull(), "Cannot compute with aOther null value");
    static_assert(-INT64_MAX > INT64_MIN, "int64_t sanity check");
    int64_t ticks = int64_t(mValue - aOther.mValue);
    // Check for overflow.
    if (mValue > aOther.mValue) {
      if (ticks < 0) {
        ticks = INT64_MAX;
      }
    } else {
      if (ticks > 0) {
        ticks = INT64_MIN;
      }
    }
    return TimeDuration::FromTicks(ticks);
  }

  TimeStamp operator+(const TimeDuration& aOther) const {
    TimeStamp result = *this;
    result += aOther;
    return result;
  }
  TimeStamp operator-(const TimeDuration& aOther) const {
    TimeStamp result = *this;
    result -= aOther;
    return result;
  }
  TimeStamp& operator+=(const TimeDuration& aOther) {
    MOZ_ASSERT(!IsNull(), "Cannot compute with a null value");
    TimeStampValue value = mValue + aOther.mValue;
    // Check for underflow.
    // (We don't check for overflow because it's not obvious what the error
    //  behavior should be in that case.)
    if (aOther.mValue < 0 && value > mValue) {
      value = 0;
    }
    mValue = value;
    return *this;
  }
  TimeStamp& operator-=(const TimeDuration& aOther) {
    MOZ_ASSERT(!IsNull(), "Cannot compute with a null value");
    TimeStampValue value = mValue - aOther.mValue;
    // Check for underflow.
    // (We don't check for overflow because it's not obvious what the error
    //  behavior should be in that case.)
    if (aOther.mValue > 0 && value > mValue) {
      value = 0;
    }
    mValue = value;
    return *this;
  }

  constexpr bool operator<(const TimeStamp& aOther) const {
    MOZ_ASSERT(!IsNull(), "Cannot compute with a null value");
    MOZ_ASSERT(!aOther.IsNull(), "Cannot compute with aOther null value");
    return mValue < aOther.mValue;
  }
  constexpr bool operator<=(const TimeStamp& aOther) const {
    MOZ_ASSERT(!IsNull(), "Cannot compute with a null value");
    MOZ_ASSERT(!aOther.IsNull(), "Cannot compute with aOther null value");
    return mValue <= aOther.mValue;
  }
  constexpr bool operator>=(const TimeStamp& aOther) const {
    MOZ_ASSERT(!IsNull(), "Cannot compute with a null value");
    MOZ_ASSERT(!aOther.IsNull(), "Cannot compute with aOther null value");
    return mValue >= aOther.mValue;
  }
  constexpr bool operator>(const TimeStamp& aOther) const {
    MOZ_ASSERT(!IsNull(), "Cannot compute with a null value");
    MOZ_ASSERT(!aOther.IsNull(), "Cannot compute with aOther null value");
    return mValue > aOther.mValue;
  }
  bool operator==(const TimeStamp& aOther) const {
    return IsNull() ? aOther.IsNull()
                    : !aOther.IsNull() && mValue == aOther.mValue;
  }
  bool operator!=(const TimeStamp& aOther) const { return !(*this == aOther); }

  // Comparing TimeStamps for equality should be discouraged. Adding
  // two TimeStamps, or scaling TimeStamps, is nonsense and must never
  // be allowed.

  static MFBT_API void Startup();
  static MFBT_API void Shutdown();

#if defined(DEBUG)
  TimeStampValue GetValue() const { return mValue; }
#endif

private:
  friend struct IPC::ParamTraits<mozilla::TimeStamp>;
  friend struct TimeStampInitialization;
  friend class TimeStampTests;

  constexpr MOZ_IMPLICIT TimeStamp(TimeStampValue aValue) : mValue(aValue) {}

  static MFBT_API TimeStamp Now(bool aHighResolution);

  /**
   * Computes the uptime of the current process in microseconds. The result
   * is platform-dependent and needs to be checked against existing timestamps
   * for consistency.
   *
   * @returns The number of microseconds since the calling process was started
   *          or 0 if an error was encountered while computing the uptime
   */
  static MFBT_API uint64_t ComputeProcessUptime();

  /**
   * When built with PRIntervalTime, a value of 0 means this instance
   * is "null". Otherwise, the low 32 bits represent a PRIntervalTime,
   * and the high 32 bits represent a counter of the number of
   * rollovers of PRIntervalTime that we've seen. This counter starts
   * at 1 to avoid a real time colliding with the "null" value.
   *
   * PR_INTERVAL_MAX is set at 100,000 ticks per second. So the minimum
   * time to wrap around is about 2^64/100000 seconds, i.e. about
   * 5,849,424 years.
   *
   * When using a system clock, a value is system dependent.
   */
  TimeStampValue mValue;
};

}  // namespace mozilla

#endif /* mozilla_TimeStamp_h */

quality100%

¤ Dauer der Verarbeitung: 0.15 Sekunden (vorverarbeitet) ¤

Wurzel

Suchen

Beweissystem der NASA

Beweissystem Isabelle

NIST Cobol Testsuite

Cephes Mathematical Library

Wiener Entwicklungsmethode

Haftungshinweis

Die Informationen auf dieser Webseite wurden nach bestem Wissen sorgfältig zusammengestellt. Es wird jedoch weder Vollständigkeit, noch Richtigkeit, noch Qualität der bereit gestellten Informationen zugesichert.

Bemerkung:

Die farbliche Syntaxdarstellung ist noch experimentell.