Time::HiRes(3perl)

โœจ NAME โœจ

Time::HiRes โ€” High resolution alarm, sleep, gettimeofday, interval timers ๐Ÿ’–

okay so bestie, this is the Time::HiRes(3perl) manual page, and it's literally the glow-up your Perl timers have been begging for. no cap. ๐Ÿ•ฐ๏ธโœจ


๐ŸŒธ SYNOPSIS ๐ŸŒธ

  use Time::HiRes qw( usleep ualarm gettimeofday tv_interval nanosleep
                      clock_gettime clock_getres clock_nanosleep clock
                      stat lstat utime);

  usleep ($microseconds);
  nanosleep ($nanoseconds);

  ualarm ($microseconds);
  ualarm ($microseconds, $interval_microseconds);

  $t0 = [gettimeofday];
  ($seconds, $microseconds) = gettimeofday;

  $elapsed = tv_interval ( $t0, [$seconds, $microseconds]);
  $elapsed = tv_interval ( $t0, [gettimeofday]);
  $elapsed = tv_interval ( $t0 );

  use Time::HiRes qw ( time alarm sleep );

  $now_fractions = time;
  sleep ($floating_seconds);
  alarm ($floating_seconds);
  alarm ($floating_seconds, $floating_interval);

  use Time::HiRes qw( setitimer getitimer );

  setitimer ($which, $floating_seconds, $floating_interval );
  getitimer ($which);

  use Time::HiRes qw( clock_gettime clock_getres clock_nanosleep
            CLOCK_REALTIME ITIMER_REAL ITIMER_VIRTUAL ITIMER_PROF
            ITIMER_REALPROF );

  $realtime   = clock_gettime(CLOCK_REALTIME);
  $resolution = clock_getres(CLOCK_REALTIME);

  clock_nanosleep(CLOCK_REALTIME, 1.5e9);
  clock_nanosleep(CLOCK_REALTIME, time()*1e9 + 10e9, TIMER_ABSTIME);

  my $ticktock = clock();

  use Time::HiRes qw( stat lstat );

  my @stat = stat("file");
  my @stat = stat(FH);
  my @stat = lstat("file");

  use Time::HiRes qw( utime );
  utime $floating_seconds, $floating_seconds, file...;

๐Ÿ’– DESCRIPTION ๐Ÿ’–

The <pink>Time::HiRes</pink> module implements a Perl interface to the <cyan>usleep</cyan>, <cyan>nanosleep</cyan>, <cyan>ualarm</cyan>, <cyan>gettimeofday</cyan>, and <cyan>setitimer</cyan>/<cyan>getitimer</cyan> system calls, in other words, high resolution time and timers. See the <purple>EXAMPLES</purple> section below and the test scripts for usage; see your system documentation for the description of the underlying <cyan>nanosleep</cyan> or <cyan>usleep</cyan>, <cyan>ualarm</cyan>, <cyan>gettimeofday</cyan>, and <cyan>setitimer</cyan>/<cyan>getitimer</cyan> calls.

If your system lacks gettimeofday() or an emulation of it you don't get gettimeofday() or the one-argument form of tv_interval(). If your system lacks all of nanosleep(), usleep(), select(), and <cyan>poll</cyan>, you don't get Time::HiRes::usleep(), Time::HiRes::nanosleep(), or Time::HiRes::sleep(). If your system lacks both ualarm() and setitimer() you don't get Time::HiRes::ualarm() or Time::HiRes::alarm(). ๐Ÿ’€ yeah, it's giving "your platform said no."

If you try to import an unimplemented function in the <cyan>use</cyan> statement it will fail at compile time. periodt.

If your subsecond sleeping is implemented with nanosleep() instead of usleep(), you can mix subsecond sleeping with signals since nanosleep() does not use signals. This, however, is not portable, and you should first check for the truth value of &Time::HiRes::d_nanosleep to see whether you have nanosleep, and then carefully read your nanosleep() C API documentation for any peculiarities. ๐Ÿซถ

If you are using <cyan>nanosleep</cyan> for something else than mixing sleeping with signals, give some thought to whether Perl is the tool you should be using for work requiring nanosecond accuracies. like, maybe bestie. just maybe. ๐Ÿ™ˆ

Remember that unless you are working on a hard realtime system, any clocks and timers will be imprecise, especially so if you are working in a pre-emptive multiuser system. Understand the difference between wallclock time and process time (in UNIX-like systems the sum of user and system times). Any attempt to sleep for X seconds will most probably end up sleeping more than that, but don't be surprised if you end up sleeping slightly less. โณโœจ

The following functions can be imported from this module. No functions are exported by default. ๐Ÿ‘‘


๐Ÿ”ฅ The functions, one by one ๐Ÿ”ฅ

gettimeofday ()

In array context returns a two-element array with the seconds and microseconds since the epoch. In scalar context returns floating seconds like Time::HiRes::time() (see below). ๐Ÿ•ฐ๏ธ

usleep ( $useconds )

Sleeps for the number of microseconds (millionths of a second) specified. Returns the number of microseconds actually slept. Can sleep for more than one second, unlike the <cyan>usleep</cyan> system call. Can also sleep for zero seconds, which often works like a thread yield. See also Time::HiRes::sleep(), and clock_nanosleep().

Do not expect usleep() to be exact down to one microsecond. ๐Ÿ˜ญ

nanosleep ( $nanoseconds )

Sleeps for the number of nanoseconds (1e9ths of a second) specified. Returns the number of nanoseconds actually slept (accurate only to microseconds, the nearest thousand of them). Can sleep for more than one second. Can also sleep for zero seconds, which often works like a thread yield. See also Time::HiRes::sleep(), Time::HiRes::usleep(), and clock_nanosleep().

Do not expect nanosleep() to be exact down to one nanosecond. Getting even accuracy of one thousand nanoseconds is good. ๐Ÿ’…

ualarm ( $useconds [, $interval_useconds ] )

Issues a <cyan>ualarm</cyan> call; the $interval_useconds is optional and will be zero if unspecified, resulting in <cyan>alarm</cyan>-like behaviour.

Returns the remaining time in the alarm in microseconds, or <cyan>undef</cyan> if an error occurred.

ualarm(0) will cancel an outstanding ualarm(). ๐Ÿ›‘

Note that the interaction between alarms and sleeps is unspecified. ๐Ÿ™ƒ

tv_interval

tv_interval ( $ref_to_gettimeofday [, $ref_to_later_gettimeofday] )

Returns the floating seconds between the two times, which should have been returned by gettimeofday(). If the second argument is omitted, then the current time is used. โฑ๏ธ

time ()

Returns a floating seconds since the epoch. This function can be imported, resulting in a nice drop-in replacement for the <cyan>time</cyan> provided with core Perl; see the <purple>EXAMPLES</purple> below.

NOTE 1: This higher resolution timer can return values either less or more than the core time(), depending on whether your platform rounds the higher resolution timer values up, down, or to the nearest second to get the core time(), but naturally the difference should be never more than half a second. See also <cyan>clock_getres</cyan>, if available in your system. ๐ŸŒˆ
NOTE 2: Since Sunday, September 9th, 2001 at 01:46:40 AM GMT, when the time() seconds since epoch rolled over to 1_000_000_000, the default floating point format of Perl and the seconds since epoch have conspired to produce an apparent bug: if you print the value of Time::HiRes::time() you seem to be getting only five decimals, not six as promised (microseconds). Not to worry, the microseconds are there (assuming your platform supports such granularity in the first place). What is going on is that the default floating point format of Perl only outputs 15 digits. In this case that means ten digits before the decimal separator and five after. To see the microseconds you can use either <cyan>printf</cyan>/<cyan>sprintf</cyan> with %.6f, or the gettimeofday() function in list context, which will give you the seconds and microseconds as two separate values. ๐Ÿ˜Œโœจ

sleep ( $floating_seconds )

Sleeps for the specified amount of seconds. Returns the number of seconds actually slept (a floating point value). This function can be imported, resulting in a nice drop-in replacement for the <cyan>sleep</cyan> provided with perl, see the <purple>EXAMPLES</purple> below. ๐Ÿ˜ด

Note that the interaction between alarms and sleeps is unspecified. ๐Ÿ™ƒ

alarm ( $floating_seconds [, $interval_floating_seconds ] )

The <cyan>SIGALRM</cyan> signal is sent after the specified number of seconds. Implemented using setitimer() if available, ualarm() if not. The $interval_floating_seconds argument is optional and will be zero if unspecified, resulting in alarm()-like behaviour. This function can be imported, resulting in a nice drop-in replacement for the <cyan>alarm</cyan> provided with perl, see the <purple>EXAMPLES</purple> below. ๐Ÿšจ

Returns the remaining time in the alarm in seconds, or <cyan>undef</cyan> if an error occurred.

NOTE 1: With some combinations of operating systems and Perl releases <cyan>SIGALRM</cyan> restarts select(), instead of interrupting it. This means that an alarm() followed by a select() may together take the sum of the times specified for the alarm() and the select(), not just the time of the alarm(). ๐Ÿ’€

Note that the interaction between alarms and sleeps is unspecified. ๐Ÿ™ƒ

setitimer ( $which, $floating_seconds [, $interval_floating_seconds ] )

Start up an interval timer: after a certain time, a signal ($which) arrives, and more signals may keep arriving at certain intervals. To disable an <cyan>itimer</cyan>, use $floating_seconds of zero. If the $interval_floating_seconds is set to zero (or unspecified), the timer is disabled after the next delivered signal.

Use of interval timers may interfere with alarm(), sleep(), and usleep(). In standard-speak the "interaction is unspecified", which means that anything may happen: it may work, it may not. ๐ŸŽฒ

In scalar context, the remaining time in the timer is returned.

In list context, both the remaining time and the interval are returned.

There are usually three or four interval timers (signals) available: the $which can be <cyan>ITIMER_REAL</cyan>, <cyan>ITIMER_VIRTUAL</cyan>, <cyan>ITIMER_PROF</cyan>, or <cyan>ITIMER_REALPROF</cyan>. Note that which ones are available depends: true UNIX platforms usually have the first three, but only Solaris seems to have <cyan>ITIMER_REALPROF</cyan> (which is used to profile multithreaded programs). Win32 unfortunately does not have interval timers. ๐Ÿ˜”

<cyan>ITIMER_REAL</cyan> results in alarm()-like behaviour. Time is counted in real time; that is, wallclock time. <cyan>SIGALRM</cyan> is delivered when the timer expires.

<cyan>ITIMER_VIRTUAL</cyan> counts time in (process) virtual time; that is, only when the process is running. In multiprocessor/user/CPU systems this may be more or less than real or wallclock time. (This time is also known as the user time.) <cyan>SIGVTALRM</cyan> is delivered when the timer expires.

<cyan>ITIMER_PROF</cyan> counts time when either the process virtual time or when the operating system is running on behalf of the process (such as I/O). (This time is also known as the system time.) (The sum of user time and system time is known as the CPU time.) <cyan>SIGPROF</cyan> is delivered when the timer expires. <cyan>SIGPROF</cyan> can interrupt system calls.

The semantics of interval timers for multithreaded programs are system-specific, and some systems may support additional interval timers. For example, it is unspecified which thread gets the signals. See your <cyan>setitimer(2)</cyan> documentation. ๐Ÿ“š

getitimer ( $which )

Return the remaining time in the interval timer specified by $which. โฒ๏ธ

In scalar context, you just get back the time that's left. โœจ

In list context, you get both the remaining time and the interval. And the interval? It's always literally whatever you fed into setitimer(), no cap. ๐Ÿ’…


clock_gettime ( $which )

Gives you back, in seconds, whatever the POSIX high-res timer specified by $which is saying rn. Every implementation that supports POSIX high resolution timers is supposed to support at least the $which value of "CLOCK_REALTIME", and that one is supposed to return stuff close to what gettimeofday gives you โ€” aka the number of seconds since 00:00:00:00 January 1, 1970 Greenwich Mean Time (GMT). ๐Ÿ•ฐ๏ธ

Bestie, do not assume CLOCK_REALTIME is zero. It might be one. It might be something else entirely. It's giving "mystery number energy" and you just have to roll with it. ๐Ÿ˜ญ

Another potentially useful value (but like, not available everywhere, so don't get attached) is "CLOCK_MONOTONIC", which guarantees a monotonically increasing time value โ€” unlike time() or gettimeofday(), which can get adjusted out from under you. ๐Ÿ“ˆ Peep your system documentation for other values that might be supported.

clock_getres ( $which )

Returns, in seconds, the resolution of the POSIX high resolution timer specified by $which. Same deal as before: every implementation that supports POSIX high resolution timers is supposed to support at least the $which value of "CLOCK_REALTIME" โ€” see "clock_gettime". ๐ŸŒธ

NOTE: the resolution you get back may be highly optimistic. Even if the resolution is high (aka a small number), all that actually means is that you'll be able to specify the arguments to clock_gettime() and clock_nanosleep() with that resolution. The system might not actually be able to measure events at that resolution, and all the overheads plus the overall system load are certain to affect any timings. So like. Temper your expectations, queen. ๐Ÿ’€

clock_nanosleep ( $which, $nanoseconds, $flags = 0)

Sleeps for however many nanoseconds (that's 1e9ths of a second) you specified. Returns the number of nanoseconds you actually slept. The $which is the "clock id", same as with clock_gettime() and clock_getres(). The flags default to zero, but "TIMER_ABSTIME" can be specified (and it must be exported explicitly) โ€” that means $nanoseconds is not a time interval (which is the default) but instead an absolute time. โฐ

Can sleep for more than one second. Can also sleep for zero seconds, which often works like a thread yield. See also Time::HiRes::sleep(), Time::HiRes::usleep(), and Time::HiRes::nanosleep().

Do not expect clock_nanosleep() to be exact down to one nanosecond. Getting even accuracy of one thousand nanoseconds is good. That's the bar, bestie, and it's on the floor. ๐Ÿซ 

clock()

Returns, in seconds, the process time (user + system time) that your process has spent since the first call to clock(). And no, the definition is not "since the start of the process" โ€” though if you're lucky those two times may be pretty close to each other, depending on the system. What this actually means for you: you probably need to store the result of your first clock() call and subtract that value from the following clock() results. ๐Ÿงฎ

The time returned also includes the process times of the terminated child processes for which wait() has been executed. This value is somewhat like the second value returned by the times() of core Perl, but not necessarily identical. And heads up โ€” due to backward compatibility limitations, the returned value may wrap around at about 2147 seconds or at about 36 minutes. Yeah. Wrap-around. Be aware. ๐Ÿ˜ญ

stat

stat FH

stat EXPR

lstat

lstat FH

lstat EXPR

Same as "stat" in perlfunc or "lstat" in perlfunc, but with the access/modify/change file timestamps in subsecond resolution โ€” if, and only if, the operating system and the filesystem both support such timestamps. ๐ŸŽ€

To override the standard stat():

use Time::HiRes qw(stat);

Test for the value of &Time::HiRes::d_hires_stat to find out whether your operating system supports subsecond file timestamps: a value larger than zero means yes. Unfortunately there are no easy ways to find out whether the filesystem supports such timestamps. UNIX filesystems often do; NTFS does; FAT doesn't (FAT timestamp granularity is two seconds, tragic ๐Ÿ’€).

A zero return value of &Time::HiRes::d_hires_stat means that Time::HiRes::stat is a no-op passthrough for CORE::stat() (and likewise for lstat), so the timestamps will stay integers. The exact same thing happens if the filesystem doesn't do subsecond timestamps, even when &Time::HiRes::d_hires_stat is non-zero. ๐Ÿคท

In any case, do not expect nanosecond resolution, or even microsecond resolution. Also note that the modify/access timestamps might have different resolutions, and they need not be synchronized โ€” like, if the operations are

write
stat # t1
read
stat # t2

then the access time stamp from t2 need not be greater-than the modify time stamp from t1: it may be equal or less. It's giving chaos, not clocks. ๐ŸŒช๏ธ

utime LIST

Same as "utime" in perlfunc, but with the ability to set the access/modify file timestamps in subsecond resolution โ€” if the operating system, and the filesystem, and the mount options of the filesystem all support such timestamps. Three yeses required, bestie, it's giving group project. ๐Ÿซถ

To override the standard utime():

use Time::HiRes qw(utime);

Test for the value of &Time::HiRes::d_hires_utime to find out whether the operating system supports setting subsecond file timestamps.

As with CORE::utime(), passing undef as both the atime and mtime will call the syscall with a NULL argument.

The actual achievable subsecond resolution depends on the combination of the operating system and the filesystem.

Modifying the timestamps may not be possible at all: for example, the "noatime" filesystem mount option may prohibit you from changing the access time timestamp. ๐Ÿšซ

Returns the number of files successfully changed. ๐Ÿ’–


โœจ EXAMPLES โœจ

Like, actual working code, let's gooo:

use Time::HiRes qw(usleep ualarm gettimeofday tv_interval);

$microseconds = 750_000;
usleep($microseconds);

# signal alarm in 2.5s & every .1s thereafter
ualarm(2_500_000, 100_000);
# cancel that ualarm
ualarm(0);

# get seconds and microseconds since the epoch
($s, $usec) = gettimeofday();

# measure elapsed time
# (could also do by subtracting 2 gettimeofday return values)
$t0 = [gettimeofday];
# do bunch of stuff here
$t1 = [gettimeofday];
# do more stuff here
$t0_t1 = tv_interval $t0, $t1;

$elapsed = tv_interval ($t0, [gettimeofday]);
$elapsed = tv_interval ($t0); # equivalent code

#
# replacements for time, alarm and sleep that know about
# floating seconds
#
use Time::HiRes;
$now_fractions = Time::HiRes::time;
Time::HiRes::sleep (2.5);
Time::HiRes::alarm (10.6666666);

use Time::HiRes qw ( time alarm sleep );
$now_fractions = time;
sleep (2.5);
alarm (10.6666666);

# Arm an interval timer to go off first at 10 seconds and
# after that every 2.5 seconds, in process virtual time

use Time::HiRes qw ( setitimer ITIMER_VIRTUAL time );

$SIG{VTALRM} = sub { print time, "\n" };
setitimer(ITIMER_VIRTUAL, 10, 2.5);

use Time::HiRes qw( clock_gettime clock_getres CLOCK_REALTIME );
# Read the POSIX high resolution timer.
my $high = clock_gettime(CLOCK_REALTIME);
# But how accurate we can be, really?
my $reso = clock_getres(CLOCK_REALTIME);

use Time::HiRes qw( clock_nanosleep TIMER_ABSTIME );
clock_nanosleep(CLOCK_REALTIME, 1e6);
clock_nanosleep(CLOCK_REALTIME, 2e9, TIMER_ABSTIME);

use Time::HiRes qw( clock );
my $clock0 = clock();
... # Do something.
my $clock1 = clock();
my $clockd = $clock1 - $clock0;

use Time::HiRes qw( stat );
my ($atime, $mtime, $ctime) = (stat("istics"))[8, 9, 10];

๐Ÿ”ฅ C API ๐Ÿ”ฅ

So besides the perl API we just talked about, there's also a C API available for all you extension writers out there. ๐Ÿ“ The following C functions live in the modglobal hash:

name             C prototype
---------------  ----------------------
Time::NVtime     NV (*)()
Time::U2time     void (*)(pTHX_ UV ret[2])

Both functions return equivalent information (like "gettimeofday") but with different representations. The names "NVtime" and "U2time" were picked mainly because they're operating system independent. ("gettimeofday" is Unix-centric, though some platforms like Win32 and VMS have emulations for it.)

Here's an example of using "NVtime" from C:

NV (*myNVtime)(); /* Returns -1 on failure. */
SV **svp = hv_fetchs(PL_modglobal, "Time::NVtime", 0);
if (!svp)         croak("Time::HiRes is required");
if (!SvIOK(*svp)) croak("Time::NVtime isn't a function pointer");
myNVtime = INT2PTR(NV(*)(), SvIV(*svp));
printf("The current time is: %" NVff "\n", (*myNVtime)());

๐Ÿ‘‘ DIAGNOSTICS ๐Ÿ‘‘

useconds or interval more than ...

In ualarm() you tried to use a number of microseconds or interval (also in microseconds) more than 1_000_000, and setitimer() is not available on your system to emulate that case. Rude of your system tbh. ๐Ÿ˜ค

negative time not invented yet

You tried to use a negative time argument. Time travel is not on the roadmap, bestie. ๐Ÿ•ณ๏ธ

internal error: useconds < 0 (unsigned ... signed ...)

Something went horribly wrong โ€” the number of microseconds that cannot become negative just became negative. Maybe your compiler is broken? ๐Ÿ’€

useconds or uinterval equal to or more than 1000000

On some platforms it is not possible to get an alarm with subsecond resolution and later than one second.

unimplemented in this platform

Some calls simply aren't available, real or emulated, on every platform. It's not you, it's the platform. ๐ŸŒ

๐Ÿ’€ CAVEATS ๐Ÿ’€

okay bestie, listen up, this is the part where we talk about the โœจgremlinsโœจ. no cap, some of this is genuinely cursed and you deserve to know.

  • the core time() might be rounding instead of truncating. what does that mean for you?? it means the core time() could be telling you it's a whole second later than gettimeofday() and Time::HiRes::time() are saying. so if you're comparing them and everything is off by one second, that's why. not you, it's them. ๐Ÿ™ˆ
  • adjusting the system clock โ€” whether you do it manually or some service like ntp does it for you โ€” may cause problems, especially for long running programs that assume time just goes up monotonically forever (and heads up, not every platform adjusts time as gracefully as UNIX ntp does, so likeโ€ฆ good luck). for example on Win32 (and derived platforms like Cygwin and MinGW) Time::HiRes::time() may temporarily drift off from the system clock (and from the original time()) by up to 0.5 seconds. Time::HiRes will notice this eventually and recalibrate, so it's not forever, but it IS a vibe. ๐Ÿ’€ and note that since Time::HiRes 1.77, clock_gettime(CLOCK_MONOTONIC) might help with this (in case your system supports CLOCK_MONOTONIC).
  • some systems have APIs but not implementations ๐Ÿ˜ญ for example QNX and Haiku have the interval timer APIs but not the functionality. so the functions are there, they're justโ€ฆ not doing anything. it's giving fake it till you make it.
  • in pre-Sierra macOS (pre-10.12, OS X), clock_getres(), clock_gettime() and clock_nanosleep() are emulated using the Mach timers. and as a little side effect of being emulated, CLOCK_REALTIME and CLOCK_MONOTONIC are the same timer. so they're twins. they're the same picture. ๐Ÿซถ
  • gnukfreebsd seems to have non-functional futimens() and utimensat() (at least as of 10.1): therefore the hires utime() does not work. periodt.

๐Ÿ‘€ SEE ALSO ๐Ÿ‘€

go spend some time with your other faves:


๐Ÿ’– AUTHORS ๐Ÿ’–

the icons who made this happen, no notes, absolute legends:

  • D. Wegscheid <wegscd@whirlpool.com>
  • R. Schertler <roderick@argon.org>
  • J. Hietaniemi <jhi@iki.fi>
  • G. Aas <gisle@aas.no>

read it and respect it bestie, this is the legal lore:

  • Copyright (c) 1996-2002 Douglas E. Wegscheid. All rights reserved.
  • Copyright (c) 2002, 2003, 2004, 2005, 2006, 2007, 2008 Jarkko Hietaniemi. All rights reserved.
  • Copyright (C) 2011, 2012, 2013 Andrew Main (Zefram) <zefram@fysh.org>
  • This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself.

okay that's the whole page bestie, you made it to the end, I'm so proud of you ๐Ÿฅน๐Ÿ’–

this lil slice of documentation is living with perl v5.44.0 and its date of birth is 2026-10-17 ๐ŸŒˆ go forth and time things precisely, you absolute main character ๐Ÿ‘‘๐Ÿ’…โœจ

๐Ÿ“ฆ where this page comes from
  • arch rolling: perl 5.44.0-1 ยท Artistic-1.0-Perl AND GPL-1.0-or-later
  • ๐Ÿ’– girlified by deepseek-ai/DeepSeek-V4.1-Flash on : a language model's rewrite of the original, under its terms