path: root/man2/times.2

.\" Copyright (c) 1992 Drew Eckhardt (drew@cs.colorado.edu), March 28, 1992
.\"
.\" SPDX-License-Identifier: Linux-man-pages-copyleft
.\"
.\" Modified by Michael Haardt (michael@moria.de)
.\" Modified Sat Jul 24 14:29:17 1993 by Rik Faith (faith@cs.unc.edu)
.\" Modified 961203 and 001211 and 010326 by aeb@cwi.nl
.\" Modified 001213 by Michael Haardt (michael@moria.de)
.\" Modified 13 Jun 02, Michael Kerrisk <mtk.manpages@gmail.com>
.\"	Added note on nonstandard behavior when SIGCHLD is ignored.
.\" Modified 2004-11-16, mtk, Noted that the nonconformance when
.\"	SIGCHLD is being ignored is fixed in Linux 2.6.9; other minor changes
.\" Modified 2004-12-08, mtk, in Linux 2.6 times() return value changed
.\" 2005-04-13, mtk
.\"	Added notes on nonstandard behavior: Linux allows 'buf' to
.\"	be NULL, but POSIX.1 doesn't specify this and it's nonportable.
.\"
.TH times 2 2023-10-31 "Linux man-pages 6.7"
.SH NAME
times \- get process times
.SH LIBRARY
Standard C library
.RI ( libc ", " \-lc )
.SH SYNOPSIS
.nf
.B #include <sys/times.h>
.P
.BI "clock_t times(struct tms *" buf );
.fi
.SH DESCRIPTION
.BR times ()
stores the current process times in the
.I "struct tms"
that
.I buf
points to.
The
.I struct tms
is as defined in
.IR <sys/times.h> :
.P
.in +4n
.EX
struct tms {
    clock_t tms_utime;  /* user time */
    clock_t tms_stime;  /* system time */
    clock_t tms_cutime; /* user time of children */
    clock_t tms_cstime; /* system time of children */
};
.EE
.in
.P
The
.I tms_utime
field contains the CPU time spent executing instructions
of the calling process.
The
.I tms_stime
field contains the CPU time spent executing inside the kernel
while performing tasks on behalf of the calling process.
.P
The
.I tms_cutime
field contains the sum of the
.I tms_utime
and
.I tms_cutime
values for all waited-for terminated children.
The
.I tms_cstime
field contains the sum of the
.I tms_stime
and
.I tms_cstime
values for all waited-for terminated children.
.P
Times for terminated children (and their descendants)
are added in at the moment
.BR wait (2)
or
.BR waitpid (2)
returns their process ID.
In particular,
times of grandchildren
that the children did not wait for are never seen.
.P
All times reported are in clock ticks.
.SH RETURN VALUE
.BR times ()
returns the number of clock ticks that have elapsed since
an arbitrary point in the past.
The return value may overflow the possible range of type
.IR clock_t .
On error,
\fI(clock_t)\ \-1\fP is returned,
and
.I errno
is set to indicate the error.
.SH ERRORS
.TP
.B EFAULT
.I tms
points outside the process's address space.
.SH VERSIONS
On Linux,
the
.I buf
argument can be specified as NULL,
with the result that
.BR times ()
just returns a function result.
However,
POSIX does not specify this behavior,
and most
other UNIX implementations require a non-NULL value for
.IR buf .
.SH STANDARDS
POSIX.1-2008.
.SH HISTORY
POSIX.1-2001,
SVr4,
4.3BSD.
.P
In POSIX.1-1996 the symbol \fBCLK_TCK\fP (defined in
.IR <time.h> )
is mentioned as obsolescent.
It is obsolete now.
.P
Before Linux 2.6.9,
if the disposition of
.B SIGCHLD
is set to
.BR SIG_IGN ,
then the times of terminated children
are automatically included in the
.I tms_cstime
and
.I tms_cutime
fields,
although POSIX.1-2001 says that this should happen
only if the calling process
.BR wait (2)s
on its children.
This nonconformance is rectified in Linux 2.6.9 and later.
.\" See the description of times() in XSH, which says:
.\"	The times of a terminated child process are included... when wait()
.\"	or waitpid() returns the process ID of this terminated child.
.P
On Linux,
the \[lq]arbitrary point in the past\[rq]
from which the return value of
.BR times ()
is measured has varied across kernel versions.
On Linux 2.4 and earlier,
this point is the moment the system was booted.
Since Linux 2.6,
this point is \fI(2\[ha]32/HZ) \- 300\fP
seconds before system boot time.
This variability across kernel versions (and across UNIX implementations),
combined with the fact that the returned value may overflow the range of
.IR clock_t ,
means that a portable application would be wise to avoid using this value.
To measure changes in elapsed time,
use
.BR clock_gettime (2)
instead.
.\" .P
.\" On older systems the number of clock ticks per second is given
.\" by the variable HZ.
.P
SVr1-3 returns
.I long
and the struct members are of type
.I time_t
although they store clock ticks,
not seconds since the Epoch.
V7 used
.I long
for the struct members,
because it had no type
.I time_t
yet.
.SH NOTES
The number of clock ticks per second can be obtained using:
.P
.in +4n
.EX
sysconf(_SC_CLK_TCK);
.EE
.in
.P
Note that
.BR clock (3)
also returns a value of type
.IR clock_t ,
but this value is measured in units of
.BR CLOCKS_PER_SEC ,
not the clock ticks used by
.BR times ().
.SH BUGS
A limitation of the Linux system call conventions on some architectures
(notably i386) means that on Linux 2.6 there is a small time window
(41 seconds) soon after boot when
.BR times ()
can return \-1,
falsely indicating that an error occurred.
The same problem can occur when the return value wraps past
the maximum value that can be stored in
.BR clock_t .
.\" The problem is that a syscall return of -4095 to -1
.\" is interpreted by glibc as an error, and the wrapper converts
.\" the return value to -1.
.\" http://marc.info/?l=linux-kernel&m=119447727031225&w=2
.\" "compat_sys_times() bogus until jiffies >= 0"
.\" November 2007
.SH SEE ALSO
.BR time (1),
.BR getrusage (2),
.BR wait (2),
.BR clock (3),
.BR sysconf (3),
.BR time (7)