thread.c: use rb_hrtime_t scalar for high-resolution time operations

Relying on "struct timespec" was too annoying API-wise and
used more stack space.  "double" was a bit wacky w.r.t rounding
in the past, so now we'll switch to using a 64-bit type.

Unsigned 64-bit integer is able to give us over nearly 585
years of range with nanoseconds.  This range is good enough
for the Linux kernel internal time representation, so it
ought to be good enough for us.

This reduces the stack usage of functions while GVL is held
(and thus subject to marking) on x86-64 Linux (with ppoll):

	rb_wait_for_single_fd    120 => 104
	do_select                120 => 88

[ruby-core:88582] [Misc #15014]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64533 b2dd03c8-39d4-4d8f-98ff-823fe69b080e

hrtime.h

@@ -0,0 +1,120 @@
+#ifndef RB_HRTIME_H
+#define RB_HRTIME_H
+#include "ruby/ruby.h"
+#include <time.h>
+#if defined(HAVE_SYS_TIME_H)
+#  include <sys/time.h>
+#endif
+
+/*
+ * Hi-res monotonic clock.  It is currently nsec resolution, which has over
+ * 500 years of range.
+ *
+ * TBD: Is nsec even necessary? usec resolution seems enough for userspace
+ * and it'll be suitable for use with devices lasting over 500,000 years
+ * (maybe some devices designed for long-term space travel)
+ */
+#define RB_HRTIME_PER_USEC ((rb_hrtime_t)1000)
+#define RB_HRTIME_PER_MSEC (RB_HRTIME_PER_USEC * (rb_hrtime_t)1000)
+#define RB_HRTIME_PER_SEC  (RB_HRTIME_PER_MSEC * (rb_hrtime_t)1000)
+#define RB_HRTIME_MAX      UINT64_MAX
+
+/*
+ * Lets try to support time travelers.  Lets assume anybody with a time machine
+ * also has access to a modern gcc or clang with 128-bit int support
+ */
+#ifdef MY_RUBY_BUILD_MAY_TIME_TRAVEL
+typedef int128_t rb_hrtime_t;
+#else
+typedef uint64_t rb_hrtime_t;
+#endif
+
+/* thread.c */
+rb_hrtime_t rb_hrtime_now(void);
+
+static inline rb_hrtime_t
+rb_hrtime_mul(rb_hrtime_t a, rb_hrtime_t b)
+{
+    rb_hrtime_t c;
+
+#ifdef HAVE_BUILTIN___BUILTIN_MUL_OVERFLOW
+    if (__builtin_mul_overflow(a, b, &c))
+        return RB_HRTIME_MAX;
+#else
+    if (b != 0 && b > RB_HRTIME_MAX / b) /* overflow */
+        return RB_HRTIME_MAX;
+    c = a * b;
+#endif
+    return c;
+}
+
+static inline rb_hrtime_t
+rb_hrtime_add(rb_hrtime_t a, rb_hrtime_t b)
+{
+    rb_hrtime_t c;
+
+#ifdef HAVE_BUILTIN___BUILTIN_ADD_OVERFLOW
+    if (__builtin_add_overflow(a, b, &c))
+        return RB_HRTIME_MAX;
+#else
+    c = a + b;
+    if (c < a) /* overflow */
+        return RB_HRTIME_MAX;
+#endif
+    return c;
+}
+
+static inline rb_hrtime_t
+rb_timeval2hrtime(const struct timeval *tv)
+{
+    rb_hrtime_t s = rb_hrtime_mul((rb_hrtime_t)tv->tv_sec, RB_HRTIME_PER_SEC);
+    rb_hrtime_t u = rb_hrtime_mul((rb_hrtime_t)tv->tv_usec, RB_HRTIME_PER_USEC);
+
+    return rb_hrtime_add(s, u);
+}
+
+static inline rb_hrtime_t
+rb_timespec2hrtime(const struct timespec *ts)
+{
+    rb_hrtime_t s = rb_hrtime_mul((rb_hrtime_t)ts->tv_sec, RB_HRTIME_PER_SEC);
+
+    return rb_hrtime_add(s, (rb_hrtime_t)ts->tv_nsec);
+}
+
+static inline rb_hrtime_t
+rb_msec2hrtime(unsigned long msec)
+{
+    return rb_hrtime_mul((rb_hrtime_t)msec, RB_HRTIME_PER_MSEC);
+}
+
+static inline rb_hrtime_t
+rb_sec2hrtime(time_t sec)
+{
+    if (sec <= 0) return 0;
+
+    return rb_hrtime_mul((rb_hrtime_t)sec, RB_HRTIME_PER_SEC);
+}
+
+static inline struct timespec *
+rb_hrtime2timespec(struct timespec *ts, const rb_hrtime_t *hrt)
+{
+    if (hrt) {
+        ts->tv_sec = *hrt / RB_HRTIME_PER_SEC;
+        ts->tv_nsec = (int32_t)(*hrt % RB_HRTIME_PER_SEC);
+        return ts;
+    }
+    return 0;
+}
+
+static inline struct timeval *
+rb_hrtime2timeval(struct timeval *tv, const rb_hrtime_t *hrt)
+{
+    if (hrt) {
+        tv->tv_sec = *hrt / RB_HRTIME_PER_SEC;
+        tv->tv_usec = (int32_t)((*hrt % RB_HRTIME_PER_SEC)/RB_HRTIME_PER_USEC);
+
+        return tv;
+    }
+    return 0;
+}
+#endif /* RB_HRTIME_H */