Limit InterProcessTimeTicksConverter to platforms that require it

base/time/time.h

 // Copyright (c) 2012 The Chromium Authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
 // Time represents an absolute point in coordinated universal time (UTC),
 // internally represented as microseconds (s/1,000,000) since the Windows epoch
 // (1601-01-01 00:00:00 UTC). System-dependent clock interface routines are
 // defined in time_PLATFORM.cc. Note that values for Time may skew and jump
 // around as the operating system makes adjustments to synchronize (e.g., with
 // NTP servers). Thus, client code that uses the Time class must account for
@@ skipping 689 matching lines @@
   // as coarse as ~15.6ms. Otherwise, the resolution should be no worse than one
   // microsecond.
   static TimeTicks Now();
 
   // Returns true if the high resolution clock is working on this system and
   // Now() will return high resolution values. Note that, on systems where the
   // high resolution clock works but is deemed inefficient, the low resolution
   // clock will be used instead.
   static bool IsHighResolution();
 
+  // Returns true if tick clock is consistent across processes meaning that
+  // tick counts taken on different processes can be safely compared with one
+  // another. Time values from different threads that differ 1 tick have an

[charliea (OOO until 10-5), 2016/06/29 22:38:54]

A couple suggestions:

1) s/tick clock/TimeTicks
2) Add a comma after processes
3) s/tick counts/timestamps
4) I'd change that last sentence to something like:

(Note that, even on platforms where this returns true, time values from
different threads that are within one tick of each other must be considered to
have an ambiguous ordering.)

[majidvp, 2016/07/05 01:45:37]

Done.

+  // ambiguous ordering.
+  static bool IsConsistentAcrossProcesses();

[charliea (OOO until 10-5), 2016/06/29 22:38:54]

It looks like, in all cases, IsConsistentAcrossProcesses == IsHighResolution().
Should we just provide an implementation in time.cc?

[majidvp, 2016/07/05 01:45:37]

This is true but that should be seen as a "coincidence" not the rule. I am
afraid if we move it into time.cc it may be seen as a rule and thus be a bit
confusing. Given that this method's exists only to document the semantic of
consistency across processes on different platforms I prefer to avoid any
confusion specially that it won't buy us much if we move it up.
 

[charliea (OOO until 10-5), 2016/07/06 17:43:52]

Acknowledged.

+
 #if defined(OS_WIN)
   // Translates an absolute QPC timestamp into a TimeTicks value. The returned
   // value has the same origin as Now(). Do NOT attempt to use this if
   // IsHighResolution() returns false.
   static TimeTicks FromQPCValue(LONGLONG qpc_value);
 #endif
 
   // Get an estimate of the TimeTick value at the time of the UnixEpoch. Because
   // Time and TimeTicks respond differently to user-set time and NTP
   // adjustments, this number is only an estimate. Nevertheless, this can be
@@ skipping 97 matching lines @@
   static void WaitUntilInitializedWin();
 #endif
 };
 
 // For logging use only.
 BASE_EXPORT std::ostream& operator<<(std::ostream& os, ThreadTicks time_ticks);
 
 }  // namespace base
 
 #endif  // BASE_TIME_TIME_H_