[Bug libstdc++/80538] New: Probably unwanted thread yield for thread::sleep_for with < 1s

[zulliger at indel dot ch]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80538

            Bug ID: 80538
           Summary: Probably unwanted thread yield for thread::sleep_for
                    with < 1s
           Product: gcc
           Version: 7.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: libstdc++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: zulliger at indel dot ch
  Target Milestone: ---

libstdc++-v3 configured to NOT have _GLIBCXX_USE_NANOSLEEP, but to HAVE
_GLIBCXX_HAVE_SLEEP and _GLIBCXX_HAVE_USLEEP defined, causes the execution of
the following code:

    ::sleep(__s.count());
    if (__ns.count() > 0)
      {
        long __us = __ns.count() / 1000;
        if (__us == 0)
          __us = 1;
        ::usleep(__us);
      }

Which means that for values of __s.count() == 0, the current thread will always
"yield" (according to my understanding, sleep(0) yields), just before the
thread may actually fall asleep for some microseconds. Is this really the
desired behaviour?

Anyway, what's the desired effect of a thread::sleep_for(0.0)? should the
thread yield then? If not, then the code should look like this:

    if (__s.count() > 0) ::sleep(__s.count());
    if (__ns.count() > 0)
      {
        long __us = __ns.count() / 1000;
        if (__us == 0)
          __us = 1;
        ::usleep(__us);
      }

shouldn't it?