Hi,
I've drafted a page for clock_nanosleep(), a system call added in
kernel 2.6. Would anyone on this list be prepared to check the draft
for readability or accuracy or both? A formatted version is shown
below; the raw groff is at the end of that.
Cheers,
Michael
==========
CLOCK_NANOSLEEP(2) Linux Programmer's Manual CLOCK_NANOSLEEP(2)
NAME
clock_nanosleep - high-resolution sleep with specifiable clock
SYNOPSIS
#include <time.h>
int clock_nanosleep(clockid_t clock_id, int flags,
const struct timespec *request,
struct timespec *remain);
Feature Test Macro Requirements for glibc (see fea-
ture_test_macros(7)):
clock_nanosleep(): _XOPEN_SOURCE >= 600
DESCRIPTION
Like nanosleep(2), clock_nanosleep() allows allows the caller
to sleep for an interval specified with nanosecond precision.
It differs in allowing the caller to select the clock against
which the sleep interval is to be measured, and in allowing the
sleep interval to be specified as either an absolute or a rela-
tive value.
The time values passed to and returned by this call are speci-
fied using timespec structures, defined as follows:
struct timespec {
time_t tv_sec; /* seconds */
long tv_nsec; /* nanoseconds [0 .. 999999999] */
};
The clock_id argument specifies the clock against which the
sleep interval is to be measured. This argument can have one
of the following values:
CLOCK_REALTIME A settable system-wide real-time clock.
CLOCK_MONOTONIC A non-settable, monotonically increasing clock
that measures time since some unspecified
point in the past that does not change after
system startup.
CLOCK_PROCESS_CPUTIME_ID
A settable per-process clock that measures CPU
time consumed by all threads in the process.
See clock_getres(3) for further details on these clocks.
If flags is 0, then the value specified in request is inter-
preted as an interval relative to the current value of the
clock specified by clock_id.
If flags is TIMER_ABSTIME, then request is interpreted as an
absolute time as measured by the clock, clock_id. If request
is less than or equal to the current value of the clock, then
clock_nanosleep() returns immediately without suspending the
calling thread.
clock_nanosleep() suspends the execution of the calling thread
until either at least the time specified by request has
elapsed, or a signal is delivered that causes a signal handler
to be called or that terminates the process.
If the call is interrupted by a signal handler,
clock_nanosleep() returns -1, and sets errno to EINTR. In
addition, if remain is not NULL, and flags was not
TIMER_ABSTIME, it returns the remaining unslept time in remain.
This value can then be used to call clock_nanosleep() again and
complete a (relative) sleep.
RETURN VALUE
On successfully sleeping for the requested interval,
clock_nanosleep() returns 0. If the call is interrupted by a
signal handler or encounters an error, then it returns a posi-
tive error number.
ERRORS
EFAULT request or remain specified an invalid address.
EINTR The sleep was interrupted by a signal handler.
EINVAL The value in the tv_nsec field was not in the range 0 to
999999999 or tv_sec was negative.
EINVAL clock_id was invalid. (CLOCK_THREAD_CPUTIME_ID is not a
permitted value for clock_id.)
VERSIONS
The clock_nanosleep(2) system call first appeared in Linux 2.6.
Support is available in glibc since version 2.1.
CONFORMING TO
POSIX.1-2001.
NOTES
If the interval specified in request is not an exact multiple
of the granularity underlying clock (see time(7)), then the
interval will be rounded up to the next multiple. Furthermore,
after the sleep completes, there may still be a delay before
the CPU becomes free to once again execute the calling thread.
Using an absolute timer is useful for preventing timer drift
problems of the type described in nanosleep(2). (Such problems
are exacerbated in programs that try to restart a relative
sleep that is repeatedly interrupted by signals.) To perform a
relative sleep that avoids these problems, call clock_get-
time(3) for the desired clock, add the desired interval to the
returned time value, and then call clock_nanosleep() with the
TIMER_ABSTIME flag.
clock_nanosleep() is never restarted after being interrupted by
a signal handler, regardless of the use of the sigaction(2)
SA_SIGACTION flag.
The remain argument is unused, and unnecessary, when flags is
TIMER_ABSTIME. (An absolute sleep can be restarted using the
same request argument.)
POSIX.1 specifies that clock_nanosleep() has no effect on sig-
nals dispositions or the signal mask.
POSIX.1 specifies that after changing the value of the
CLOCK_REALTIME clock via clock_settime(), the new clock value
shall be used to determine the time at which a thread blocked
on an absolute clock_nanosleep() will wake up; if the new clock
value falls past the end of the sleep interval, then the
clock_nanosleep() call will return immediately.
POSIX.1 specifies that changing the value of the CLOCK_REALTIME
clock via clock_settime() shall have no effect on a thread that
is blocked on a relative clock_nanosleep().
SEE ALSO
nanosleep(2), clock_getres(3), sleep(3), timer_create(3),
usleep(3), time(7)
Linux 2008-06-26 CLOCK_NANOSLEEP(2)
==========
.\" Copyright (c) 2008, Linux Foundation, written by Michael Kerrisk
.\" <mtk.manpages@xxxxxxxxx>
.\"
.\" Permission is granted to make and distribute verbatim copies of this
.\" manual provided the copyright notice and this permission notice are
.\" preserved on all copies.
.\"
.\" Permission is granted to copy and distribute modified versions of this
.\" manual under the conditions for verbatim copying, provided that the
.\" entire resulting derived work is distributed under the terms of a
.\" permission notice identical to this one.
.\"
.\" Since the Linux kernel and libraries are constantly changing, this
.\" manual page may be incorrect or out-of-date. The author(s) assume no
.\" responsibility for errors or omissions, or for damages resulting from
.\" the use of the information contained herein. The author(s) may not
.\" have taken the same level of care in the production of this manual,
.\" which is licensed free of charge, as they might when working
.\" professionally.
.\"
.\" Formatted or processed versions of this manual, if unaccompanied by
.\" the source, must acknowledge the copyright and authors of this work.
.\"
.TH CLOCK_NANOSLEEP 2 2008-06-26 "Linux" "Linux Programmer's Manual"
.SH NAME
clock_nanosleep \- high-resolution sleep with specifiable clock
.SH SYNOPSIS
.B #include <time.h>
.nf
.sp
.BI "int clock_nanosleep(clockid_t " clock_id ", int " flags ,
.BI " const struct timespec *" request ,
.BI " struct timespec *" remain );
.fi
.sp
.in -4n
Feature Test Macro Requirements for glibc (see
.BR feature_test_macros (7)):
.in
.sp
.BR clock_nanosleep ():
_XOPEN_SOURCE\ >=\ 600
.SH DESCRIPTION
Like
.BR nanosleep (2),
.BR clock_nanosleep ()
allows allows the caller to sleep for an interval specified
with nanosecond precision.
It differs in allowing the caller to select the clock against
which the sleep interval is to be measured,
and in allowing the sleep interval to be specified as
either an absolute or a relative value.
The time values passed to and returned by this call are specified using
.I timespec
structures, defined as follows:
.sp
.in +4n
.nf
struct timespec {
time_t tv_sec; /* seconds */
long tv_nsec; /* nanoseconds [0 .. 999999999] */
};
.fi
.in
The
.I clock_id
argument specifies the clock against which the sleep interval
is to be measured.
This argument can have one of the following values:
.TP 17
.BR CLOCK_REALTIME
A settable system-wide real-time clock.
.TP
.BR CLOCK_MONOTONIC
A non-settable, monotonically increasing clock that measures time
since some unspecified point in the past that does not change after
system startup.
.\" On Linux this clock measures time since boot.
.TP
.BR CLOCK_PROCESS_CPUTIME_ID
A settable per-process clock that measures CPU time consumed
by all threads in the process.
.\" There is some trickery between glibc and the kernel
.\" to deal with the CLOCK_PROCESS_CPUTIME_ID case.
.PP
See
.BR clock_getres (3)
for further details on these clocks.
If
.I flags
is 0, then the value specified in
.I request
is interpreted as an interval relative to the current
value of the clock specified by
.IR clock_id .
If
.I flags
is
.BR TIMER_ABSTIME ,
then
.I request
is interpreted as an absolute time as measured by the clock,
.IR clock_id .
If
.I request
is less than or equal to the current value of the clock,
then
.BR clock_nanosleep ()
returns immediately without suspending the calling thread.
.BR clock_nanosleep ()
suspends the execution of the calling thread
until either at least the time specified by
.IR request
has elapsed,
or a signal is delivered that causes a signal handler to be called or
that terminates the process.
If the call is interrupted by a signal handler,
.BR clock_nanosleep ()
returns \-1, and sets
.I errno
to
.BR EINTR .
In addition, if
.I remain
is not NULL, and
.I flags
was not
.BR TIMER_ABSTIME ,
it returns the remaining unslept time in
.IR remain .
This value can then be used to call
.BR clock_nanosleep ()
again and complete a (relative) sleep.
.SH "RETURN VALUE"
On successfully sleeping for the requested interval,
.BR clock_nanosleep ()
returns 0.
If the call is interrupted by a signal handler or encounters an error,
then it returns a positive error number.
.SH ERRORS
.B EFAULT
.I request
or
.I remain
specified an invalid address.
.TP
.B EINTR
The sleep was interrupted by a signal handler.
.TP
.B EINVAL
The value in the
.I tv_nsec
field was not in the range 0 to 999999999 or
.I tv_sec
was negative.
.TP
.B EINVAL
.I clock_id
was invalid.
.RB ( CLOCK_THREAD_CPUTIME_ID
is not a permitted value for
.IR clock_id .)
.SH VERSIONS
The
.BR clock_nanosleep (2)
system call first appeared in Linux 2.6.
Support is available in glibc since version 2.1.
.SH "CONFORMING TO"
POSIX.1-2001.
.SH NOTES
If the interval specified in
.I request
is not an exact multiple of the granularity underlying clock (see
.BR time (7)),
then the interval will be rounded up to the next multiple.
Furthermore, after the sleep completes, there may still be a delay before
the CPU becomes free to once again execute the calling thread.
Using an absolute timer is useful for preventing
timer drift problems of the type described in
.BR nanosleep (2).
(Such problems are exacerbated in programs that try to restart
a relative sleep that is repeatedly interrupted by signals.)
To perform a relative sleep that avoids these problems, call
.BR clock_gettime (3)
for the desired clock,
add the desired interval to the returned time value,
and then call
.BR clock_nanosleep ()
with the
.B TIMER_ABSTIME
flag.
.BR clock_nanosleep ()
is never restarted after being interrupted by a signal handler,
regardless of the use of the
.BR sigaction (2)
.B SA_SIGACTION
flag.
The
.I remain
argument is unused, and unnecessary, when
.I flags
is
.BR TIMER_ABSTIME .
(An absolute sleep can be restarted using the same
.I request
argument.)
POSIX.1 specifies that
.BR clock_nanosleep ()
has no effect on signals dispositions or the signal mask.
POSIX.1 specifies that after changing the value of the
.B CLOCK_REALTIME
clock via
.BR clock_settime (),
the new clock value shall be used to determine the time
at which a thread blocked on an absolute
.BR clock_nanosleep ()
will wake up;
if the new clock value falls past the end of the sleep interval, then the
.BR clock_nanosleep ()
call will return immediately.
POSIX.1 specifies that
changing the value of the
.B CLOCK_REALTIME
clock via
.BR clock_settime ()
shall have no effect on a thread that is blocked on a relative
.BR clock_nanosleep ().
.SH "SEE ALSO"
.BR nanosleep (2),
.BR clock_getres (3),
.BR sleep (3),
.BR timer_create (3),
.BR usleep (3),
.BR time (7)
--
To unsubscribe from this list: send the line "unsubscribe linux-man" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
