#include <sys/time.h> /* for 'struct timeval' */
\f
-/* The type to use to represent temporal intervals. Its address can be passed
- as the timeout argument to the pselect system call. */
+/* The type to use to represent non-negative temporal intervals. Its
+ address can be passed as the timeout argument to the pselect system
+ call. */
typedef struct timespec EMACS_TIME;
/* Resolution of EMACS_TIME time stamps (in units per second), and log
/* Return the result of adding A to B, or of subtracting B from A.
On overflow, store an extremal value: ergo, if time_t is unsigned,
- return 0 if the true answer would be negative. */
+ return 0 if the true answer would be negative.
+
+ WARNING: These are NOT general-purpose macros for adding or
+ subtracting arbitrary time values! They are generally intended to
+ be used with their first argument an absolute time since the epoch
+ and the second argument a non-negative offset. Do NOT use them for
+ anything else. */
static inline EMACS_TIME
add_emacs_time (EMACS_TIME a, EMACS_TIME b)
{
return timespec_sub (a, b);
}
-/* Return the sign of the valid time stamp TIME, either -1, 0, or 1. */
+/* Return the sign of the valid time stamp TIME, either -1, 0, or 1.
+ Note: this can only return a negative value if time_t is a signed
+ data type. */
static inline int
EMACS_TIME_SIGN (EMACS_TIME t)
{
}
/* Convert the double D to the greatest EMACS_TIME not greater than D.
- On overflow, return an extremal value. Return the minimum
- EMACS_TIME if D is not a number. */
+ On overflow, return an extremal value; in particular, if time_t is
+ an unsigned data type and D is negative, return zero. Return the
+ minimum EMACS_TIME if D is not a number. */
static inline EMACS_TIME
EMACS_TIME_FROM_DOUBLE (double d)
{