xref: /aosp_15_r20/external/openscreen/third_party/abseil/src/absl/time/clock.cc (revision 3f982cf4871df8771c9d4abe6e9a6f8d829b2736)
1 // Copyright 2017 The Abseil Authors.
2 //
3 // Licensed under the Apache License, Version 2.0 (the "License");
4 // you may not use this file except in compliance with the License.
5 // You may obtain a copy of the License at
6 //
7 //      https://www.apache.org/licenses/LICENSE-2.0
8 //
9 // Unless required by applicable law or agreed to in writing, software
10 // distributed under the License is distributed on an "AS IS" BASIS,
11 // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12 // See the License for the specific language governing permissions and
13 // limitations under the License.
14 
15 #include "absl/time/clock.h"
16 
17 #include "absl/base/attributes.h"
18 
19 #ifdef _WIN32
20 #include <windows.h>
21 #endif
22 
23 #include <algorithm>
24 #include <atomic>
25 #include <cerrno>
26 #include <cstdint>
27 #include <ctime>
28 #include <limits>
29 
30 #include "absl/base/internal/spinlock.h"
31 #include "absl/base/internal/unscaledcycleclock.h"
32 #include "absl/base/macros.h"
33 #include "absl/base/port.h"
34 #include "absl/base/thread_annotations.h"
35 
36 namespace absl {
37 ABSL_NAMESPACE_BEGIN
Now()38 Time Now() {
39   // TODO(bww): Get a timespec instead so we don't have to divide.
40   int64_t n = absl::GetCurrentTimeNanos();
41   if (n >= 0) {
42     return time_internal::FromUnixDuration(
43         time_internal::MakeDuration(n / 1000000000, n % 1000000000 * 4));
44   }
45   return time_internal::FromUnixDuration(absl::Nanoseconds(n));
46 }
47 ABSL_NAMESPACE_END
48 }  // namespace absl
49 
50 // Decide if we should use the fast GetCurrentTimeNanos() algorithm
51 // based on the cyclecounter, otherwise just get the time directly
52 // from the OS on every call. This can be chosen at compile-time via
53 // -DABSL_USE_CYCLECLOCK_FOR_GET_CURRENT_TIME_NANOS=[0|1]
54 #ifndef ABSL_USE_CYCLECLOCK_FOR_GET_CURRENT_TIME_NANOS
55 #if ABSL_USE_UNSCALED_CYCLECLOCK
56 #define ABSL_USE_CYCLECLOCK_FOR_GET_CURRENT_TIME_NANOS 1
57 #else
58 #define ABSL_USE_CYCLECLOCK_FOR_GET_CURRENT_TIME_NANOS 0
59 #endif
60 #endif
61 
62 #if defined(__APPLE__) || defined(_WIN32)
63 #include "absl/time/internal/get_current_time_chrono.inc"
64 #else
65 #include "absl/time/internal/get_current_time_posix.inc"
66 #endif
67 
68 // Allows override by test.
69 #ifndef GET_CURRENT_TIME_NANOS_FROM_SYSTEM
70 #define GET_CURRENT_TIME_NANOS_FROM_SYSTEM() \
71   ::absl::time_internal::GetCurrentTimeNanosFromSystem()
72 #endif
73 
74 #if !ABSL_USE_CYCLECLOCK_FOR_GET_CURRENT_TIME_NANOS
75 namespace absl {
76 ABSL_NAMESPACE_BEGIN
GetCurrentTimeNanos()77 int64_t GetCurrentTimeNanos() { return GET_CURRENT_TIME_NANOS_FROM_SYSTEM(); }
78 ABSL_NAMESPACE_END
79 }  // namespace absl
80 #else  // Use the cyclecounter-based implementation below.
81 
82 // Allows override by test.
83 #ifndef GET_CURRENT_TIME_NANOS_CYCLECLOCK_NOW
84 #define GET_CURRENT_TIME_NANOS_CYCLECLOCK_NOW() \
85   ::absl::time_internal::UnscaledCycleClockWrapperForGetCurrentTime::Now()
86 #endif
87 
88 // The following counters are used only by the test code.
89 static int64_t stats_initializations;
90 static int64_t stats_reinitializations;
91 static int64_t stats_calibrations;
92 static int64_t stats_slow_paths;
93 static int64_t stats_fast_slow_paths;
94 
95 namespace absl {
96 ABSL_NAMESPACE_BEGIN
97 namespace time_internal {
98 // This is a friend wrapper around UnscaledCycleClock::Now()
99 // (needed to access UnscaledCycleClock).
100 class UnscaledCycleClockWrapperForGetCurrentTime {
101  public:
Now()102   static int64_t Now() { return base_internal::UnscaledCycleClock::Now(); }
103 };
104 }  // namespace time_internal
105 
106 // uint64_t is used in this module to provide an extra bit in multiplications
107 
108 // Return the time in ns as told by the kernel interface.  Place in *cycleclock
109 // the value of the cycleclock at about the time of the syscall.
110 // This call represents the time base that this module synchronizes to.
111 // Ensures that *cycleclock does not step back by up to (1 << 16) from
112 // last_cycleclock, to discard small backward counter steps.  (Larger steps are
113 // assumed to be complete resyncs, which shouldn't happen.  If they do, a full
114 // reinitialization of the outer algorithm should occur.)
GetCurrentTimeNanosFromKernel(uint64_t last_cycleclock,uint64_t * cycleclock)115 static int64_t GetCurrentTimeNanosFromKernel(uint64_t last_cycleclock,
116                                              uint64_t *cycleclock) {
117   // We try to read clock values at about the same time as the kernel clock.
118   // This value gets adjusted up or down as estimate of how long that should
119   // take, so we can reject attempts that take unusually long.
120   static std::atomic<uint64_t> approx_syscall_time_in_cycles{10 * 1000};
121 
122   uint64_t local_approx_syscall_time_in_cycles =  // local copy
123       approx_syscall_time_in_cycles.load(std::memory_order_relaxed);
124 
125   int64_t current_time_nanos_from_system;
126   uint64_t before_cycles;
127   uint64_t after_cycles;
128   uint64_t elapsed_cycles;
129   int loops = 0;
130   do {
131     before_cycles = GET_CURRENT_TIME_NANOS_CYCLECLOCK_NOW();
132     current_time_nanos_from_system = GET_CURRENT_TIME_NANOS_FROM_SYSTEM();
133     after_cycles = GET_CURRENT_TIME_NANOS_CYCLECLOCK_NOW();
134     // elapsed_cycles is unsigned, so is large on overflow
135     elapsed_cycles = after_cycles - before_cycles;
136     if (elapsed_cycles >= local_approx_syscall_time_in_cycles &&
137         ++loops == 20) {  // clock changed frequencies?  Back off.
138       loops = 0;
139       if (local_approx_syscall_time_in_cycles < 1000 * 1000) {
140         local_approx_syscall_time_in_cycles =
141             (local_approx_syscall_time_in_cycles + 1) << 1;
142       }
143       approx_syscall_time_in_cycles.store(
144           local_approx_syscall_time_in_cycles,
145           std::memory_order_relaxed);
146     }
147   } while (elapsed_cycles >= local_approx_syscall_time_in_cycles ||
148            last_cycleclock - after_cycles < (static_cast<uint64_t>(1) << 16));
149 
150   // Number of times in a row we've seen a kernel time call take substantially
151   // less than approx_syscall_time_in_cycles.
152   static std::atomic<uint32_t> seen_smaller{ 0 };
153 
154   // Adjust approx_syscall_time_in_cycles to be within a factor of 2
155   // of the typical time to execute one iteration of the loop above.
156   if ((local_approx_syscall_time_in_cycles >> 1) < elapsed_cycles) {
157     // measured time is no smaller than half current approximation
158     seen_smaller.store(0, std::memory_order_relaxed);
159   } else if (seen_smaller.fetch_add(1, std::memory_order_relaxed) >= 3) {
160     // smaller delays several times in a row; reduce approximation by 12.5%
161     const uint64_t new_approximation =
162         local_approx_syscall_time_in_cycles -
163         (local_approx_syscall_time_in_cycles >> 3);
164     approx_syscall_time_in_cycles.store(new_approximation,
165                                         std::memory_order_relaxed);
166     seen_smaller.store(0, std::memory_order_relaxed);
167   }
168 
169   *cycleclock = after_cycles;
170   return current_time_nanos_from_system;
171 }
172 
173 
174 // ---------------------------------------------------------------------
175 // An implementation of reader-write locks that use no atomic ops in the read
176 // case.  This is a generalization of Lamport's method for reading a multiword
177 // clock.  Increment a word on each write acquisition, using the low-order bit
178 // as a spinlock; the word is the high word of the "clock".  Readers read the
179 // high word, then all other data, then the high word again, and repeat the
180 // read if the reads of the high words yields different answers, or an odd
181 // value (either case suggests possible interference from a writer).
182 // Here we use a spinlock to ensure only one writer at a time, rather than
183 // spinning on the bottom bit of the word to benefit from SpinLock
184 // spin-delay tuning.
185 
186 // Acquire seqlock (*seq) and return the value to be written to unlock.
SeqAcquire(std::atomic<uint64_t> * seq)187 static inline uint64_t SeqAcquire(std::atomic<uint64_t> *seq) {
188   uint64_t x = seq->fetch_add(1, std::memory_order_relaxed);
189 
190   // We put a release fence between update to *seq and writes to shared data.
191   // Thus all stores to shared data are effectively release operations and
192   // update to *seq above cannot be re-ordered past any of them.  Note that
193   // this barrier is not for the fetch_add above.  A release barrier for the
194   // fetch_add would be before it, not after.
195   std::atomic_thread_fence(std::memory_order_release);
196 
197   return x + 2;   // original word plus 2
198 }
199 
200 // Release seqlock (*seq) by writing x to it---a value previously returned by
201 // SeqAcquire.
SeqRelease(std::atomic<uint64_t> * seq,uint64_t x)202 static inline void SeqRelease(std::atomic<uint64_t> *seq, uint64_t x) {
203   // The unlock store to *seq must have release ordering so that all
204   // updates to shared data must finish before this store.
205   seq->store(x, std::memory_order_release);  // release lock for readers
206 }
207 
208 // ---------------------------------------------------------------------
209 
210 // "nsscaled" is unit of time equal to a (2**kScale)th of a nanosecond.
211 enum { kScale = 30 };
212 
213 // The minimum interval between samples of the time base.
214 // We pick enough time to amortize the cost of the sample,
215 // to get a reasonably accurate cycle counter rate reading,
216 // and not so much that calculations will overflow 64-bits.
217 static const uint64_t kMinNSBetweenSamples = 2000 << 20;
218 
219 // We require that kMinNSBetweenSamples shifted by kScale
220 // have at least a bit left over for 64-bit calculations.
221 static_assert(((kMinNSBetweenSamples << (kScale + 1)) >> (kScale + 1)) ==
222                kMinNSBetweenSamples,
223                "cannot represent kMaxBetweenSamplesNSScaled");
224 
225 // A reader-writer lock protecting the static locations below.
226 // See SeqAcquire() and SeqRelease() above.
227 ABSL_CONST_INIT static absl::base_internal::SpinLock lock(
228     absl::kConstInit, base_internal::SCHEDULE_KERNEL_ONLY);
229 ABSL_CONST_INIT static std::atomic<uint64_t> seq(0);
230 
231 // data from a sample of the kernel's time value
232 struct TimeSampleAtomic {
233   std::atomic<uint64_t> raw_ns;              // raw kernel time
234   std::atomic<uint64_t> base_ns;             // our estimate of time
235   std::atomic<uint64_t> base_cycles;         // cycle counter reading
236   std::atomic<uint64_t> nsscaled_per_cycle;  // cycle period
237   // cycles before we'll sample again (a scaled reciprocal of the period,
238   // to avoid a division on the fast path).
239   std::atomic<uint64_t> min_cycles_per_sample;
240 };
241 // Same again, but with non-atomic types
242 struct TimeSample {
243   uint64_t raw_ns;                 // raw kernel time
244   uint64_t base_ns;                // our estimate of time
245   uint64_t base_cycles;            // cycle counter reading
246   uint64_t nsscaled_per_cycle;     // cycle period
247   uint64_t min_cycles_per_sample;  // approx cycles before next sample
248 };
249 
250 static struct TimeSampleAtomic last_sample;   // the last sample; under seq
251 
252 static int64_t GetCurrentTimeNanosSlowPath() ABSL_ATTRIBUTE_COLD;
253 
254 // Read the contents of *atomic into *sample.
255 // Each field is read atomically, but to maintain atomicity between fields,
256 // the access must be done under a lock.
ReadTimeSampleAtomic(const struct TimeSampleAtomic * atomic,struct TimeSample * sample)257 static void ReadTimeSampleAtomic(const struct TimeSampleAtomic *atomic,
258                                  struct TimeSample *sample) {
259   sample->base_ns = atomic->base_ns.load(std::memory_order_relaxed);
260   sample->base_cycles = atomic->base_cycles.load(std::memory_order_relaxed);
261   sample->nsscaled_per_cycle =
262       atomic->nsscaled_per_cycle.load(std::memory_order_relaxed);
263   sample->min_cycles_per_sample =
264       atomic->min_cycles_per_sample.load(std::memory_order_relaxed);
265   sample->raw_ns = atomic->raw_ns.load(std::memory_order_relaxed);
266 }
267 
268 // Public routine.
269 // Algorithm:  We wish to compute real time from a cycle counter.  In normal
270 // operation, we construct a piecewise linear approximation to the kernel time
271 // source, using the cycle counter value.  The start of each line segment is at
272 // the same point as the end of the last, but may have a different slope (that
273 // is, a different idea of the cycle counter frequency).  Every couple of
274 // seconds, the kernel time source is sampled and compared with the current
275 // approximation.  A new slope is chosen that, if followed for another couple
276 // of seconds, will correct the error at the current position.  The information
277 // for a sample is in the "last_sample" struct.  The linear approximation is
278 //   estimated_time = last_sample.base_ns +
279 //     last_sample.ns_per_cycle * (counter_reading - last_sample.base_cycles)
280 // (ns_per_cycle is actually stored in different units and scaled, to avoid
281 // overflow).  The base_ns of the next linear approximation is the
282 // estimated_time using the last approximation; the base_cycles is the cycle
283 // counter value at that time; the ns_per_cycle is the number of ns per cycle
284 // measured since the last sample, but adjusted so that most of the difference
285 // between the estimated_time and the kernel time will be corrected by the
286 // estimated time to the next sample.  In normal operation, this algorithm
287 // relies on:
288 // - the cycle counter and kernel time rates not changing a lot in a few
289 //   seconds.
290 // - the client calling into the code often compared to a couple of seconds, so
291 //   the time to the next correction can be estimated.
292 // Any time ns_per_cycle is not known, a major error is detected, or the
293 // assumption about frequent calls is violated, the implementation returns the
294 // kernel time.  It records sufficient data that a linear approximation can
295 // resume a little later.
296 
GetCurrentTimeNanos()297 int64_t GetCurrentTimeNanos() {
298   // read the data from the "last_sample" struct (but don't need raw_ns yet)
299   // The reads of "seq" and test of the values emulate a reader lock.
300   uint64_t base_ns;
301   uint64_t base_cycles;
302   uint64_t nsscaled_per_cycle;
303   uint64_t min_cycles_per_sample;
304   uint64_t seq_read0;
305   uint64_t seq_read1;
306 
307   // If we have enough information to interpolate, the value returned will be
308   // derived from this cycleclock-derived time estimate.  On some platforms
309   // (POWER) the function to retrieve this value has enough complexity to
310   // contribute to register pressure - reading it early before initializing
311   // the other pieces of the calculation minimizes spill/restore instructions,
312   // minimizing icache cost.
313   uint64_t now_cycles = GET_CURRENT_TIME_NANOS_CYCLECLOCK_NOW();
314 
315   // Acquire pairs with the barrier in SeqRelease - if this load sees that
316   // store, the shared-data reads necessarily see that SeqRelease's updates
317   // to the same shared data.
318   seq_read0 = seq.load(std::memory_order_acquire);
319 
320   base_ns = last_sample.base_ns.load(std::memory_order_relaxed);
321   base_cycles = last_sample.base_cycles.load(std::memory_order_relaxed);
322   nsscaled_per_cycle =
323       last_sample.nsscaled_per_cycle.load(std::memory_order_relaxed);
324   min_cycles_per_sample =
325       last_sample.min_cycles_per_sample.load(std::memory_order_relaxed);
326 
327   // This acquire fence pairs with the release fence in SeqAcquire.  Since it
328   // is sequenced between reads of shared data and seq_read1, the reads of
329   // shared data are effectively acquiring.
330   std::atomic_thread_fence(std::memory_order_acquire);
331 
332   // The shared-data reads are effectively acquire ordered, and the
333   // shared-data writes are effectively release ordered. Therefore if our
334   // shared-data reads see any of a particular update's shared-data writes,
335   // seq_read1 is guaranteed to see that update's SeqAcquire.
336   seq_read1 = seq.load(std::memory_order_relaxed);
337 
338   // Fast path.  Return if min_cycles_per_sample has not yet elapsed since the
339   // last sample, and we read a consistent sample.  The fast path activates
340   // only when min_cycles_per_sample is non-zero, which happens when we get an
341   // estimate for the cycle time.  The predicate will fail if now_cycles <
342   // base_cycles, or if some other thread is in the slow path.
343   //
344   // Since we now read now_cycles before base_ns, it is possible for now_cycles
345   // to be less than base_cycles (if we were interrupted between those loads and
346   // last_sample was updated). This is harmless, because delta_cycles will wrap
347   // and report a time much much bigger than min_cycles_per_sample. In that case
348   // we will take the slow path.
349   uint64_t delta_cycles = now_cycles - base_cycles;
350   if (seq_read0 == seq_read1 && (seq_read0 & 1) == 0 &&
351       delta_cycles < min_cycles_per_sample) {
352     return base_ns + ((delta_cycles * nsscaled_per_cycle) >> kScale);
353   }
354   return GetCurrentTimeNanosSlowPath();
355 }
356 
357 // Return (a << kScale)/b.
358 // Zero is returned if b==0.   Scaling is performed internally to
359 // preserve precision without overflow.
SafeDivideAndScale(uint64_t a,uint64_t b)360 static uint64_t SafeDivideAndScale(uint64_t a, uint64_t b) {
361   // Find maximum safe_shift so that
362   //  0 <= safe_shift <= kScale  and  (a << safe_shift) does not overflow.
363   int safe_shift = kScale;
364   while (((a << safe_shift) >> safe_shift) != a) {
365     safe_shift--;
366   }
367   uint64_t scaled_b = b >> (kScale - safe_shift);
368   uint64_t quotient = 0;
369   if (scaled_b != 0) {
370     quotient = (a << safe_shift) / scaled_b;
371   }
372   return quotient;
373 }
374 
375 static uint64_t UpdateLastSample(
376     uint64_t now_cycles, uint64_t now_ns, uint64_t delta_cycles,
377     const struct TimeSample *sample) ABSL_ATTRIBUTE_COLD;
378 
379 // The slow path of GetCurrentTimeNanos().  This is taken while gathering
380 // initial samples, when enough time has elapsed since the last sample, and if
381 // any other thread is writing to last_sample.
382 //
383 // Manually mark this 'noinline' to minimize stack frame size of the fast
384 // path.  Without this, sometimes a compiler may inline this big block of code
385 // into the fast path.  That causes lots of register spills and reloads that
386 // are unnecessary unless the slow path is taken.
387 //
388 // TODO(absl-team): Remove this attribute when our compiler is smart enough
389 // to do the right thing.
390 ABSL_ATTRIBUTE_NOINLINE
GetCurrentTimeNanosSlowPath()391 static int64_t GetCurrentTimeNanosSlowPath() ABSL_LOCKS_EXCLUDED(lock) {
392   // Serialize access to slow-path.  Fast-path readers are not blocked yet, and
393   // code below must not modify last_sample until the seqlock is acquired.
394   lock.Lock();
395 
396   // Sample the kernel time base.  This is the definition of
397   // "now" if we take the slow path.
398   static uint64_t last_now_cycles;  // protected by lock
399   uint64_t now_cycles;
400   uint64_t now_ns = GetCurrentTimeNanosFromKernel(last_now_cycles, &now_cycles);
401   last_now_cycles = now_cycles;
402 
403   uint64_t estimated_base_ns;
404 
405   // ----------
406   // Read the "last_sample" values again; this time holding the write lock.
407   struct TimeSample sample;
408   ReadTimeSampleAtomic(&last_sample, &sample);
409 
410   // ----------
411   // Try running the fast path again; another thread may have updated the
412   // sample between our run of the fast path and the sample we just read.
413   uint64_t delta_cycles = now_cycles - sample.base_cycles;
414   if (delta_cycles < sample.min_cycles_per_sample) {
415     // Another thread updated the sample.  This path does not take the seqlock
416     // so that blocked readers can make progress without blocking new readers.
417     estimated_base_ns = sample.base_ns +
418         ((delta_cycles * sample.nsscaled_per_cycle) >> kScale);
419     stats_fast_slow_paths++;
420   } else {
421     estimated_base_ns =
422         UpdateLastSample(now_cycles, now_ns, delta_cycles, &sample);
423   }
424 
425   lock.Unlock();
426 
427   return estimated_base_ns;
428 }
429 
430 // Main part of the algorithm.  Locks out readers, updates the approximation
431 // using the new sample from the kernel, and stores the result in last_sample
432 // for readers.  Returns the new estimated time.
UpdateLastSample(uint64_t now_cycles,uint64_t now_ns,uint64_t delta_cycles,const struct TimeSample * sample)433 static uint64_t UpdateLastSample(uint64_t now_cycles, uint64_t now_ns,
434                                  uint64_t delta_cycles,
435                                  const struct TimeSample *sample)
436     ABSL_EXCLUSIVE_LOCKS_REQUIRED(lock) {
437   uint64_t estimated_base_ns = now_ns;
438   uint64_t lock_value = SeqAcquire(&seq);  // acquire seqlock to block readers
439 
440   // The 5s in the next if-statement limits the time for which we will trust
441   // the cycle counter and our last sample to give a reasonable result.
442   // Errors in the rate of the source clock can be multiplied by the ratio
443   // between this limit and kMinNSBetweenSamples.
444   if (sample->raw_ns == 0 ||  // no recent sample, or clock went backwards
445       sample->raw_ns + static_cast<uint64_t>(5) * 1000 * 1000 * 1000 < now_ns ||
446       now_ns < sample->raw_ns || now_cycles < sample->base_cycles) {
447     // record this sample, and forget any previously known slope.
448     last_sample.raw_ns.store(now_ns, std::memory_order_relaxed);
449     last_sample.base_ns.store(estimated_base_ns, std::memory_order_relaxed);
450     last_sample.base_cycles.store(now_cycles, std::memory_order_relaxed);
451     last_sample.nsscaled_per_cycle.store(0, std::memory_order_relaxed);
452     last_sample.min_cycles_per_sample.store(0, std::memory_order_relaxed);
453     stats_initializations++;
454   } else if (sample->raw_ns + 500 * 1000 * 1000 < now_ns &&
455              sample->base_cycles + 50 < now_cycles) {
456     // Enough time has passed to compute the cycle time.
457     if (sample->nsscaled_per_cycle != 0) {  // Have a cycle time estimate.
458       // Compute time from counter reading, but avoiding overflow
459       // delta_cycles may be larger than on the fast path.
460       uint64_t estimated_scaled_ns;
461       int s = -1;
462       do {
463         s++;
464         estimated_scaled_ns = (delta_cycles >> s) * sample->nsscaled_per_cycle;
465       } while (estimated_scaled_ns / sample->nsscaled_per_cycle !=
466                (delta_cycles >> s));
467       estimated_base_ns = sample->base_ns +
468                           (estimated_scaled_ns >> (kScale - s));
469     }
470 
471     // Compute the assumed cycle time kMinNSBetweenSamples ns into the future
472     // assuming the cycle counter rate stays the same as the last interval.
473     uint64_t ns = now_ns - sample->raw_ns;
474     uint64_t measured_nsscaled_per_cycle = SafeDivideAndScale(ns, delta_cycles);
475 
476     uint64_t assumed_next_sample_delta_cycles =
477         SafeDivideAndScale(kMinNSBetweenSamples, measured_nsscaled_per_cycle);
478 
479     int64_t diff_ns = now_ns - estimated_base_ns;  // estimate low by this much
480 
481     // We want to set nsscaled_per_cycle so that our estimate of the ns time
482     // at the assumed cycle time is the assumed ns time.
483     // That is, we want to set nsscaled_per_cycle so:
484     //  kMinNSBetweenSamples + diff_ns  ==
485     //  (assumed_next_sample_delta_cycles * nsscaled_per_cycle) >> kScale
486     // But we wish to damp oscillations, so instead correct only most
487     // of our current error, by solving:
488     //  kMinNSBetweenSamples + diff_ns - (diff_ns / 16) ==
489     //  (assumed_next_sample_delta_cycles * nsscaled_per_cycle) >> kScale
490     ns = kMinNSBetweenSamples + diff_ns - (diff_ns / 16);
491     uint64_t new_nsscaled_per_cycle =
492         SafeDivideAndScale(ns, assumed_next_sample_delta_cycles);
493     if (new_nsscaled_per_cycle != 0 &&
494         diff_ns < 100 * 1000 * 1000 && -diff_ns < 100 * 1000 * 1000) {
495       // record the cycle time measurement
496       last_sample.nsscaled_per_cycle.store(
497           new_nsscaled_per_cycle, std::memory_order_relaxed);
498       uint64_t new_min_cycles_per_sample =
499           SafeDivideAndScale(kMinNSBetweenSamples, new_nsscaled_per_cycle);
500       last_sample.min_cycles_per_sample.store(
501           new_min_cycles_per_sample, std::memory_order_relaxed);
502       stats_calibrations++;
503     } else {  // something went wrong; forget the slope
504       last_sample.nsscaled_per_cycle.store(0, std::memory_order_relaxed);
505       last_sample.min_cycles_per_sample.store(0, std::memory_order_relaxed);
506       estimated_base_ns = now_ns;
507       stats_reinitializations++;
508     }
509     last_sample.raw_ns.store(now_ns, std::memory_order_relaxed);
510     last_sample.base_ns.store(estimated_base_ns, std::memory_order_relaxed);
511     last_sample.base_cycles.store(now_cycles, std::memory_order_relaxed);
512   } else {
513     // have a sample, but no slope; waiting for enough time for a calibration
514     stats_slow_paths++;
515   }
516 
517   SeqRelease(&seq, lock_value);  // release the readers
518 
519   return estimated_base_ns;
520 }
521 ABSL_NAMESPACE_END
522 }  // namespace absl
523 #endif  // ABSL_USE_CYCLECLOCK_FOR_GET_CURRENT_TIME_NANOS
524 
525 namespace absl {
526 ABSL_NAMESPACE_BEGIN
527 namespace {
528 
529 // Returns the maximum duration that SleepOnce() can sleep for.
MaxSleep()530 constexpr absl::Duration MaxSleep() {
531 #ifdef _WIN32
532   // Windows Sleep() takes unsigned long argument in milliseconds.
533   return absl::Milliseconds(
534       std::numeric_limits<unsigned long>::max());  // NOLINT(runtime/int)
535 #else
536   return absl::Seconds(std::numeric_limits<time_t>::max());
537 #endif
538 }
539 
540 // Sleeps for the given duration.
541 // REQUIRES: to_sleep <= MaxSleep().
SleepOnce(absl::Duration to_sleep)542 void SleepOnce(absl::Duration to_sleep) {
543 #ifdef _WIN32
544   Sleep(to_sleep / absl::Milliseconds(1));
545 #else
546   struct timespec sleep_time = absl::ToTimespec(to_sleep);
547   while (nanosleep(&sleep_time, &sleep_time) != 0 && errno == EINTR) {
548     // Ignore signals and wait for the full interval to elapse.
549   }
550 #endif
551 }
552 
553 }  // namespace
554 ABSL_NAMESPACE_END
555 }  // namespace absl
556 
557 extern "C" {
558 
AbslInternalSleepFor(absl::Duration duration)559 ABSL_ATTRIBUTE_WEAK void AbslInternalSleepFor(absl::Duration duration) {
560   while (duration > absl::ZeroDuration()) {
561     absl::Duration to_sleep = std::min(duration, absl::MaxSleep());
562     absl::SleepOnce(to_sleep);
563     duration -= to_sleep;
564   }
565 }
566 
567 }  // extern "C"
568