path: root/man/man2/adjtimex.2

'\" t
.\" Copyright (c) 1995 Michael Chastain (mec@shell.portal.com), 15 April 1995.
.\" and Copyright (C) 2014, 2016 Michael Kerrisk <mtk.manpages@gmail.com>
.\"
.\" SPDX-License-Identifier: GPL-2.0-or-later
.\"
.\" Modified 1997-01-31 by Eric S. Raymond <esr@thyrsus.com>
.\" Modified 1997-07-30 by Paul Slootman <paul@wurtel.demon.nl>
.\" Modified 2004-05-27 by Michael Kerrisk <mtk.manpages@gmail.com>
.\"
.TH adjtimex 2 (date) "Linux man-pages (unreleased)"
.SH NAME
adjtimex, clock_adjtime, ntp_adjtime \- tune kernel clock
.SH LIBRARY
Standard C library
.RI ( libc ", " \-lc )
.SH SYNOPSIS
.nf
.B #include <sys/timex.h>
.P
.BI "int adjtimex(struct timex *" "buf" );
.P
.BI "int clock_adjtime(clockid_t " clk_id, " struct timex *" "buf" );
.P
.BI "int ntp_adjtime(struct timex *" buf );
.fi
.SH DESCRIPTION
Linux uses David L.\& Mills' clock adjustment algorithm (see RFC\ 5905).
The system call
.BR adjtimex ()
reads and optionally sets adjustment parameters for this algorithm.
It takes a pointer to a
.I timex
structure, updates kernel parameters from (selected) field values,
and returns the same structure updated with the current kernel values.
This structure is declared as follows:
.P
.in +4n
.EX
struct timex {
    int  modes;      /* Mode selector */
    long offset;     /* Time offset; nanoseconds, if STA_NANO
                        status flag is set, otherwise
                        microseconds */
    long freq;       /* Frequency offset; see NOTES for units */
    long maxerror;   /* Maximum error (microseconds) */
    long esterror;   /* Estimated error (microseconds) */
    int  status;     /* Clock command/status */
    long constant;   /* PLL (phase\-locked loop) time constant */
    long precision;  /* Clock precision
                        (microseconds, read\-only) */
    long tolerance;  /* Clock frequency tolerance (read\-only);
                        see NOTES for units */
    struct timeval time;
                     /* Current time (read\-only, except for
                        ADJ_SETOFFSET); upon return, time.tv_usec
                        contains nanoseconds, if STA_NANO status
                        flag is set, otherwise microseconds */
    long tick;       /* Microseconds between clock ticks */
    long ppsfreq;    /* PPS (pulse per second) frequency
                        (read\-only); see NOTES for units */
    long jitter;     /* PPS jitter (read\-only); nanoseconds, if
                        STA_NANO status flag is set, otherwise
                        microseconds */
    int  shift;      /* PPS interval duration
                        (seconds, read\-only) */
    long stabil;     /* PPS stability (read\-only);
                        see NOTES for units */
    long jitcnt;     /* PPS count of jitter limit exceeded
                        events (read\-only) */
    long calcnt;     /* PPS count of calibration intervals
                        (read\-only) */
    long errcnt;     /* PPS count of calibration errors
                        (read\-only) */
    long stbcnt;     /* PPS count of stability limit exceeded
                        events (read\-only) */
    int tai;         /* TAI offset, as set by previous ADJ_TAI
                        operation (seconds, read\-only,
                        since Linux 2.6.26) */
    /* Further padding bytes to allow for future expansion */
};
.EE
.in
.P
The
.I modes
field determines which parameters, if any, to set.
(As described later in this page,
the constants used for
.BR ntp_adjtime ()
are equivalent but differently named.)
It is a bit mask containing a
bitwise OR
combination of zero or more of the following bits:
.TP
.B ADJ_OFFSET
Set time offset from
.IR buf.offset .
Since Linux 2.6.26,
.\" commit 074b3b87941c99bc0ce35385b5817924b1ed0c23
the supplied value is clamped to the range (\-0.5s, +0.5s).
In older kernels, an
.B EINVAL
error occurs if the supplied value is out of range.
.TP
.B ADJ_FREQUENCY
Set frequency offset from
.IR buf.freq .
Since Linux 2.6.26,
.\" commit 074b3b87941c99bc0ce35385b5817924b1ed0c23
the supplied value is clamped to the range (\-32768000, +32768000).
In older kernels, an
.B EINVAL
error occurs if the supplied value is out of range.
.TP
.B ADJ_MAXERROR
Set maximum time error from
.IR buf.maxerror .
.TP
.B ADJ_ESTERROR
Set estimated time error from
.IR buf.esterror .
.TP
.B ADJ_STATUS
Set clock status bits from
.IR buf.status .
A description of these bits is provided below.
.TP
.B ADJ_TIMECONST
Set PLL time constant from
.IR buf.constant .
If the
.B STA_NANO
status flag (see below) is clear, the kernel adds 4 to this value.
.TP
.BR ADJ_SETOFFSET " (since Linux 2.6.39)"
.\" commit 094aa1881fdc1b8889b442eb3511b31f3ec2b762
.\" Author: Richard Cochran <richardcochran@gmail.com>
Add
.I buf.time
to the current time.
If
.I buf.status
includes the
.B ADJ_NANO
flag, then
.I buf.time.tv_usec
is interpreted as a nanosecond value;
otherwise it is interpreted as microseconds.
.IP
The value of
.I buf.time
is the sum of its two fields, but the
field
.I buf.time.tv_usec
must always be nonnegative.
The following example shows how to
normalize a
.I timeval
with nanosecond resolution.
.IP
.in +4n
.EX
while (buf.time.tv_usec < 0) {
    buf.time.tv_sec  \-= 1;
    buf.time.tv_usec += 1000000000;
}
.EE
.in
.TP
.BR ADJ_MICRO " (since Linux 2.6.26)"
.\" commit eea83d896e318bda54be2d2770d2c5d6668d11db
.\" Author: Roman Zippel <zippel@linux-m68k.org>
Select microsecond resolution.
.TP
.BR ADJ_NANO " (since Linux 2.6.26)"
.\" commit eea83d896e318bda54be2d2770d2c5d6668d11db
.\" Author: Roman Zippel <zippel@linux-m68k.org>
Select nanosecond resolution.
Only one of
.B ADJ_MICRO
and
.B ADJ_NANO
should be specified.
.TP
.BR ADJ_TAI " (since Linux 2.6.26)"
.\" commit 153b5d054ac2d98ea0d86504884326b6777f683d
Set TAI (Atomic International Time) offset from
.IR buf.constant .
.IP
.B ADJ_TAI
should not be used in conjunction with
.BR ADJ_TIMECONST ,
since the latter mode also employs the
.I buf.constant
field.
.IP
For a complete explanation of TAI
and the difference between TAI and UTC, see
.UR http://www.bipm.org/en/bipm/tai/tai.html
.I BIPM
.UE
.TP
.B ADJ_TICK
Set tick value from
.IR buf.tick .
.P
Alternatively,
.I modes
can be specified as either of the following (multibit mask) values,
in which case other bits should not be specified in
.IR modes :
.\" In general, the other bits are ignored, but ADJ_OFFSET_SINGLESHOT 0x8001
.\" ORed with ADJ_NANO (0x2000) gives 0xa0001 == ADJ_OFFSET_SS_READ!!
.TP
.B ADJ_OFFSET_SINGLESHOT
.\" In user space, ADJ_OFFSET_SINGLESHOT is 0x8001
.\" In kernel space it is 0x0001, and must be ANDed with ADJ_ADJTIME (0x8000)
Old-fashioned
.BR adjtime (3):
(gradually) adjust time by value specified in
.IR buf.offset ,
which specifies an adjustment in microseconds.
.TP
.BR ADJ_OFFSET_SS_READ " (functional since Linux 2.6.28)"
.\" In user space, ADJ_OFFSET_SS_READ is 0xa001
.\" In kernel space there is ADJ_OFFSET_READONLY (0x2000) anded with
.\" ADJ_ADJTIME (0x8000) and ADJ_OFFSET_SINGLESHOT (0x0001) to give 0xa001)
Return (in
.IR buf.offset )
the remaining amount of time to be adjusted after an earlier
.B ADJ_OFFSET_SINGLESHOT
operation.
This feature was added in Linux 2.6.24,
.\" commit 52bfb36050c8529d9031d2c2513b281a360922ec
but did not work correctly
.\" commit 916c7a855174e3b53d182b97a26b2e27a29726a1
until Linux 2.6.28.
.P
Ordinary users are restricted to a value of either 0 or
.B ADJ_OFFSET_SS_READ
for
.IR modes .
Only the superuser may set any parameters.
.P
The
.I buf.status
field is a bit mask that is used to set and/or retrieve status
bits associated with the NTP implementation.
Some bits in the mask are both readable and settable,
while others are read-only.
.TP
.BR STA_PLL " (read-write)"
Enable phase-locked loop (PLL) updates via
.BR ADJ_OFFSET .
.TP
.BR STA_PPSFREQ " (read-write)"
Enable PPS (pulse-per-second) frequency discipline.
.TP
.BR STA_PPSTIME " (read-write)"
Enable PPS time discipline.
.TP
.BR STA_FLL " (read-write)"
Select frequency-locked loop (FLL) mode.
.TP
.BR STA_INS " (read-write)"
Insert a leap second after the last second of the UTC day,
thus extending the last minute of the day by one second.
Leap-second insertion will occur each day, so long as this flag remains set.
.\" John Stultz;
.\"     Usually this is written as extending the day by one second,
.\"     which is represented as:
.\"        23:59:59
.\"        23:59:60
.\"        00:00:00
.\"
.\"     But since posix cannot represent 23:59:60, we repeat the last second:
.\"        23:59:59 + TIME_INS
.\"        23:59:59 + TIME_OOP
.\"        00:00:00 + TIME_WAIT
.\"
.TP
.BR STA_DEL " (read-write)"
Delete a leap second at the last second of the UTC day.
.\" John Stultz:
.\"     Similarly the progression here is:
.\"        23:59:57 + TIME_DEL
.\"        23:59:58 + TIME_DEL
.\"        00:00:00 + TIME_WAIT
Leap second deletion will occur each day, so long as this flag
remains set.
.\" FIXME Does there need to be a statement that it is nonsensical to set
.\" to set both STA_INS and STA_DEL?
.TP
.BR STA_UNSYNC " (read-write)"
Clock unsynchronized.
.TP
.BR STA_FREQHOLD " (read-write)"
Hold frequency.
.\" Following text from John Stultz:
Normally adjustments made via
.B ADJ_OFFSET
result in dampened frequency adjustments also being made.
So a single call corrects the current offset,
but as offsets in the same direction are made repeatedly,
the small frequency adjustments will accumulate to fix the long-term skew.
.IP
This flag prevents the small frequency adjustment from being made
when correcting for an
.B ADJ_OFFSET
value.
.\" According to the Kernel Application Program Interface document,
.\" STA_FREQHOLD is not used by the NTP version 4 daemon
.TP
.BR STA_PPSSIGNAL " (read-only)"
A valid PPS (pulse-per-second) signal is present.
.TP
.BR STA_PPSJITTER " (read-only)"
PPS signal jitter exceeded.
.TP
.BR STA_PPSWANDER " (read-only)"
PPS signal wander exceeded.
.TP
.BR STA_PPSERROR " (read-only)"
PPS signal calibration error.
.TP
.BR STA_CLOCKERR " (read-only)"
Clock hardware fault.
.\" Not set in current kernel (4.5), but checked in a few places
.TP
.BR STA_NANO " (read-only; since Linux 2.6.26)"
.\" commit eea83d896e318bda54be2d2770d2c5d6668d11db
.\" Author: Roman Zippel <zippel@linux-m68k.org>
Resolution (0 = microsecond, 1 = nanoseconds).
Set via
.BR ADJ_NANO ,
cleared via
.BR ADJ_MICRO .
.TP
.BR STA_MODE " (since Linux 2.6.26)"
.\" commit eea83d896e318bda54be2d2770d2c5d6668d11db
.\" Author: Roman Zippel <zippel@linux-m68k.org>
Mode (0 = Phase Locked Loop, 1 = Frequency Locked Loop).
.TP
.BR STA_CLK " (read-only; since Linux 2.6.26)"
.\" commit eea83d896e318bda54be2d2770d2c5d6668d11db
.\" Author: Roman Zippel <zippel@linux-m68k.org>
Clock source (0 = A, 1 = B); currently unused.
.P
Attempts to set read-only
.I status
bits are silently ignored.
.\"
.SS clock_adjtime ()
The
.BR clock_adjtime ()
system call (added in Linux 2.6.39) behaves like
.BR adjtimex ()
but takes an additional
.I clk_id
argument to specify the particular clock on which to act.
.SS ntp_adjtime ()
The
.BR ntp_adjtime ()
library function
(described in the NTP "Kernel Application Program API", KAPI)
is a more portable interface for performing the same task as
.BR adjtimex ().
Other than the following points, it is identical to
.BR adjtimex ():
.IP \[bu] 3
The constants used in
.I modes
are prefixed with "MOD_" rather than "ADJ_", and have the same suffixes (thus,
.BR MOD_OFFSET ,
.BR MOD_FREQUENCY ,
and so on), other than the exceptions noted in the following points.
.IP \[bu]
.B MOD_CLKA
is the synonym for
.BR ADJ_OFFSET_SINGLESHOT .
.IP \[bu]
.B MOD_CLKB
is the synonym for
.BR ADJ_TICK .
.IP \[bu]
The is no synonym for
.BR ADJ_OFFSET_SS_READ ,
which is not described in the KAPI.
.SH RETURN VALUE
On success,
.BR adjtimex ()
and
.BR ntp_adjtime ()
return the clock state; that is, one of the following values:
.TP 12
.B TIME_OK
Clock synchronized, no leap second adjustment pending.
.TP
.B TIME_INS
Indicates that a leap second will be added at the end of the UTC day.
.TP
.B TIME_DEL
Indicates that a leap second will be deleted at the end of the UTC day.
.TP
.B TIME_OOP
Insertion of a leap second is in progress.
.TP
.B TIME_WAIT
A leap-second insertion or deletion has been completed.
This value will be returned until the next
.B ADJ_STATUS
operation clears the
.B STA_INS
and
.B STA_DEL
flags.
.TP
.B TIME_ERROR
The system clock is not synchronized to a reliable server.
This value is returned when any of the following holds true:
.RS
.IP \[bu] 3
Either
.B STA_UNSYNC
or
.B STA_CLOCKERR
is set.
.IP \[bu]
.B STA_PPSSIGNAL
is clear and either
.B STA_PPSFREQ
or
.B STA_PPSTIME
is set.
.IP \[bu]
.B STA_PPSTIME
and
.B STA_PPSJITTER
are both set.
.IP \[bu]
.B STA_PPSFREQ
is set and either
.B STA_PPSWANDER
or
.B STA_PPSJITTER
is set.
.RE
.IP
The symbolic name
.B TIME_BAD
is a synonym for
.BR TIME_ERROR ,
provided for backward compatibility.
.P
Note that starting with Linux 3.4,
.\" commit 6b43ae8a619d17c4935c3320d2ef9e92bdeed05d changed to asynchronous
.\"  operation, so we can no longer rely on the return code.
the call operates asynchronously and the return value usually will
not reflect a state change caused by the call itself.
.P
On failure, these calls return \-1 and set
.I errno
to indicate the error.
.SH ERRORS
.TP
.B EFAULT
.I buf
does not point to writable memory.
.TP
.BR EINVAL " (before Linux 2.6.26)"
An attempt was made to set
.I buf.freq
to a value outside the range (\-33554432, +33554432).
.\" From a quick glance, it appears there was no clamping or range check
.\" for buf.freq before Linux 2.0
.TP
.BR EINVAL " (before Linux 2.6.26)"
An attempt was made to set
.I buf.offset
to a value outside the permitted range.
Before Linux 2.0, the permitted range was (\-131072, +131072).
From Linux 2.0 onwards, the permitted range was (\-512000, +512000).
.TP
.B EINVAL
An attempt was made to set
.I buf.status
to a value other than those listed above.
.TP
.B EINVAL
The
.I clk_id
given to
.BR clock_adjtime ()
is invalid for one of two reasons.
Either the System-V style hard-coded
positive clock ID value is out of range, or the dynamic
.I clk_id
does not refer to a valid instance of a clock object.
See
.BR clock_gettime (2)
for a discussion of dynamic clocks.
.TP
.B EINVAL
An attempt was made to set
.I buf.tick
to a value outside the range
.RB 900000/ HZ
to
.RB 1100000/ HZ ,
where
.B HZ
is the system timer interrupt frequency.
.TP
.B ENODEV
The hot-pluggable device (like USB for example) represented by a
dynamic
.I clk_id
has disappeared after its character device was opened.
See
.BR clock_gettime (2)
for a discussion of dynamic clocks.
.TP
.B EOPNOTSUPP
The given
.I clk_id
does not support adjustment.
.TP
.B EPERM
.I buf.modes
is neither 0 nor
.BR ADJ_OFFSET_SS_READ ,
and the caller does not have sufficient privilege.
Under Linux, the
.B CAP_SYS_TIME
capability is required.
.SH ATTRIBUTES
For an explanation of the terms used in this section, see
.BR attributes (7).
.TS
allbox;
lbx lb lb
l l l.
Interface	Attribute	Value
T{
.na
.nh
.BR \%ntp_adjtime ()
T}	Thread safety	MT-Safe
.TE
.SH STANDARDS
.TP
.BR adjtimex ()
.TQ
.BR clock_adjtime ()
Linux.
.P
The preferred API for the NTP daemon is
.BR ntp_adjtime ().
.SH NOTES
In struct
.IR timex ,
.IR freq ,
.IR ppsfreq ,
and
.I stabil
are ppm (parts per million) with a 16-bit fractional part,
which means that a value of 1 in one of those fields
actually means 2\[ha]-16 ppm, and 2\[ha]16=65536 is 1 ppm.
This is the case for both input values (in the case of
.IR freq )
and output values.
.P
The leap-second processing triggered by
.B STA_INS
and
.B STA_DEL
is done by the kernel in timer context.
Thus, it will take one tick into the second
for the leap second to be inserted or deleted.
.SH SEE ALSO
.BR clock_gettime (2),
.BR clock_settime (2),
.BR settimeofday (2),
.BR adjtime (3),
.BR ntp_gettime (3),
.BR capabilities (7),
.BR time (7),
.BR adjtimex (8),
.BR hwclock (8)
.P
.ad l
.UR http://www.slac.stanford.edu/comp/unix/\:package/\:rtems/\:src/\:ssrlApps/\:ntpNanoclock/\:api.htm
NTP "Kernel Application Program Interface"
.UE