d/freebsd-skq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
42d92058c3
freebsd-skq/sys/kern/kern_ntptime.c
kib 42d92058c3 Currently the ntptime code and resettodr() are Giant-locked. In 
particular, the Giant is supposed to protect against parallel
ntp_adjtime(2) invocations.  But, for instance, sys_ntp_adjtime() does
copyout(9) under Giant and then examines time_status to return syscall
result.  Since copyout(9) could sleep, the syscall result might be
inconsistent.

Another and more important issue is that if PPS is configured,
hardpps(9) is executed without any protection against the parallel
top-level code invocation. Potentially, this may result in the
inconsistent state of the ntptime state variables, but I cannot say
how serious such distortion is. The non-functional splclock() call in
sys_ntp_adjtime() protected against clock interrupts calling hardpps()
in the pre-SMP era.

Modernize the locking. A mutex protects ntptime data.  Due to the
hardpps() KPI legitimately serving from the interrupt filters (and
e.g. uart(4) does call it from filter), the lock cannot be sleepable
mutex if PPS_SYNC is defined.  Otherwise, use normal sleepable mutex
to reduce interrupt latency.

Reviewed by:	  imp, jhb
Sponsored by:	  The FreeBSD Foundation
Approved by:	  re (gjb)
Differential revision:	https://reviews.freebsd.org/D6825
2016-06-28 16:43:23 +00:00

1084 lines
33 KiB
C
Raw Blame History

/*-
***********************************************************************
* *
* Copyright (c) David L. Mills 1993-2001 *
* *
* Permission to use, copy, modify, and distribute this software and *
* its documentation for any purpose and without fee is hereby *
* granted, provided that the above copyright notice appears in all *
* copies and that both the copyright notice and this permission *
* notice appear in supporting documentation, and that the name *
* University of Delaware not be used in advertising or publicity *
* pertaining to distribution of the software without specific, *
* written prior permission. The University of Delaware makes no *
* representations about the suitability this software for any *
* purpose. It is provided "as is" without express or implied *
* warranty. *
* *
**********************************************************************/
/*
* Adapted from the original sources for FreeBSD and timecounters by:
* Poul-Henning Kamp <phk@FreeBSD.org>.
*
* The 32bit version of the "LP" macros seems a bit past its "sell by"
* date so I have retained only the 64bit version and included it directly
* in this file.
*
* Only minor changes done to interface with the timecounters over in
* sys/kern/kern_clock.c. Some of the comments below may be (even more)
* confusing and/or plain wrong in that context.
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include "opt_ntp.h"
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/sysproto.h>
#include <sys/eventhandler.h>
#include <sys/kernel.h>
#include <sys/priv.h>
#include <sys/proc.h>
#include <sys/lock.h>
#include <sys/mutex.h>
#include <sys/time.h>
#include <sys/timex.h>
#include <sys/timetc.h>
#include <sys/timepps.h>
#include <sys/syscallsubr.h>
#include <sys/sysctl.h>
#ifdef PPS_SYNC
FEATURE(pps_sync, "Support usage of external PPS signal by kernel PLL");
#endif
/*
* Single-precision macros for 64-bit machines
*/
typedef int64_t l_fp;
#define L_ADD(v, u) ((v) += (u))
#define L_SUB(v, u) ((v) -= (u))
#define L_ADDHI(v, a) ((v) += (int64_t)(a) << 32)
#define L_NEG(v) ((v) = -(v))
#define L_RSHIFT(v, n) \
do { \
if ((v) < 0) \
(v) = -(-(v) >> (n)); \
else \
(v) = (v) >> (n); \
} while (0)
#define L_MPY(v, a) ((v) *= (a))
#define L_CLR(v) ((v) = 0)
#define L_ISNEG(v) ((v) < 0)
#define L_LINT(v, a) ((v) = (int64_t)(a) << 32)
#define L_GINT(v) ((v) < 0 ? -(-(v) >> 32) : (v) >> 32)
/*
* Generic NTP kernel interface
*
* These routines constitute the Network Time Protocol (NTP) interfaces
* for user and daemon application programs. The ntp_gettime() routine
* provides the time, maximum error (synch distance) and estimated error
* (dispersion) to client user application programs. The ntp_adjtime()
* routine is used by the NTP daemon to adjust the system clock to an
* externally derived time. The time offset and related variables set by
* this routine are used by other routines in this module to adjust the
* phase and frequency of the clock discipline loop which controls the
* system clock.
*
* When the kernel time is reckoned directly in nanoseconds (NTP_NANO
* defined), the time at each tick interrupt is derived directly from
* the kernel time variable. When the kernel time is reckoned in
* microseconds, (NTP_NANO undefined), the time is derived from the
* kernel time variable together with a variable representing the
* leftover nanoseconds at the last tick interrupt. In either case, the
* current nanosecond time is reckoned from these values plus an
* interpolated value derived by the clock routines in another
* architecture-specific module. The interpolation can use either a
* dedicated counter or a processor cycle counter (PCC) implemented in
* some architectures.
*
* Note that all routines must run at priority splclock or higher.
*/
/*
* Phase/frequency-lock loop (PLL/FLL) definitions
*
* The nanosecond clock discipline uses two variable types, time
* variables and frequency variables. Both types are represented as 64-
* bit fixed-point quantities with the decimal point between two 32-bit
* halves. On a 32-bit machine, each half is represented as a single
* word and mathematical operations are done using multiple-precision
* arithmetic. On a 64-bit machine, ordinary computer arithmetic is
* used.
*
* A time variable is a signed 64-bit fixed-point number in ns and
* fraction. It represents the remaining time offset to be amortized
* over succeeding tick interrupts. The maximum time offset is about
* 0.5 s and the resolution is about 2.3e-10 ns.
*
* 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 3 3
* 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
* +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
* |s s s| ns |
* +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
* | fraction |
* +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
*
* A frequency variable is a signed 64-bit fixed-point number in ns/s
* and fraction. It represents the ns and fraction to be added to the
* kernel time variable at each second. The maximum frequency offset is
* about +-500000 ns/s and the resolution is about 2.3e-10 ns/s.
*
* 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 3 3
* 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
* +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
* |s s s s s s s s s s s s s| ns/s |
* +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
* | fraction |
* +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
*/
/*
* The following variables establish the state of the PLL/FLL and the
* residual time and frequency offset of the local clock.
*/
#define SHIFT_PLL 4 /* PLL loop gain (shift) */
#define SHIFT_FLL 2 /* FLL loop gain (shift) */
static int time_state = TIME_OK; /* clock state */
int time_status = STA_UNSYNC; /* clock status bits */
static long time_tai; /* TAI offset (s) */
static long time_monitor; /* last time offset scaled (ns) */
static long time_constant; /* poll interval (shift) (s) */
static long time_precision = 1; /* clock precision (ns) */
static long time_maxerror = MAXPHASE / 1000; /* maximum error (us) */
long time_esterror = MAXPHASE / 1000; /* estimated error (us) */
static long time_reftime; /* uptime at last adjustment (s) */
static l_fp time_offset; /* time offset (ns) */
static l_fp time_freq; /* frequency offset (ns/s) */
static l_fp time_adj; /* tick adjust (ns/s) */
static int64_t time_adjtime; /* correction from adjtime(2) (usec) */
static struct mtx ntpadj_lock;
MTX_SYSINIT(ntpadj, &ntpadj_lock, "ntpadj",
#ifdef PPS_SYNC
MTX_SPIN
#else
MTX_DEF
#endif
);
/*
* When PPS_SYNC is defined, hardpps() function is provided which can
* be legitimately called from interrupt filters. Due to this, use
* spinlock for ntptime state protection, otherwise sleepable mutex is
* adequate.
*/
#ifdef PPS_SYNC
#define NTPADJ_LOCK() mtx_lock_spin(&ntpadj_lock)
#define NTPADJ_UNLOCK() mtx_unlock_spin(&ntpadj_lock)
#else
#define NTPADJ_LOCK() mtx_lock(&ntpadj_lock)
#define NTPADJ_UNLOCK() mtx_unlock(&ntpadj_lock)
#endif
#define NTPADJ_ASSERT_LOCKED() mtx_assert(&ntpadj_lock, MA_OWNED)
#ifdef PPS_SYNC
/*
* The following variables are used when a pulse-per-second (PPS) signal
* is available and connected via a modem control lead. They establish
* the engineering parameters of the clock discipline loop when
* controlled by the PPS signal.
*/
#define PPS_FAVG 2 /* min freq avg interval (s) (shift) */
#define PPS_FAVGDEF 8 /* default freq avg int (s) (shift) */
#define PPS_FAVGMAX 15 /* max freq avg interval (s) (shift) */
#define PPS_PAVG 4 /* phase avg interval (s) (shift) */
#define PPS_VALID 120 /* PPS signal watchdog max (s) */
#define PPS_MAXWANDER 100000 /* max PPS wander (ns/s) */
#define PPS_POPCORN 2 /* popcorn spike threshold (shift) */
static struct timespec pps_tf[3]; /* phase median filter */
static l_fp pps_freq; /* scaled frequency offset (ns/s) */
static long pps_fcount; /* frequency accumulator */
static long pps_jitter; /* nominal jitter (ns) */
static long pps_stabil; /* nominal stability (scaled ns/s) */
static long pps_lastsec; /* time at last calibration (s) */
static int pps_valid; /* signal watchdog counter */
static int pps_shift = PPS_FAVG; /* interval duration (s) (shift) */
static int pps_shiftmax = PPS_FAVGDEF; /* max interval duration (s) (shift) */
static int pps_intcnt; /* wander counter */
/*
* PPS signal quality monitors
*/
static long pps_calcnt; /* calibration intervals */
static long pps_jitcnt; /* jitter limit exceeded */
static long pps_stbcnt; /* stability limit exceeded */
static long pps_errcnt; /* calibration errors */
#endif /* PPS_SYNC */
/*
* End of phase/frequency-lock loop (PLL/FLL) definitions
*/
static void ntp_init(void);
static void hardupdate(long offset);
static void ntp_gettime1(struct ntptimeval *ntvp);
static bool ntp_is_time_error(int tsl);
static bool
ntp_is_time_error(int tsl)
{
/*
* Status word error decode. If any of these conditions occur,
* an error is returned, instead of the status word. Most
* applications will care only about the fact the system clock
* may not be trusted, not about the details.
*
* Hardware or software error
*/
if ((tsl & (STA_UNSYNC | STA_CLOCKERR)) ||
/*
* PPS signal lost when either time or frequency synchronization
* requested
*/
(tsl & (STA_PPSFREQ | STA_PPSTIME) &&
!(tsl & STA_PPSSIGNAL)) ||
/*
* PPS jitter exceeded when time synchronization requested
*/
(tsl & STA_PPSTIME && tsl & STA_PPSJITTER) ||
/*
* PPS wander exceeded or calibration error when frequency
* synchronization requested
*/
(tsl & STA_PPSFREQ &&
tsl & (STA_PPSWANDER | STA_PPSERROR)))
return (true);
return (false);
}
static void
ntp_gettime1(struct ntptimeval *ntvp)
{
struct timespec atv; /* nanosecond time */
NTPADJ_ASSERT_LOCKED();
nanotime(&atv);
ntvp->time.tv_sec = atv.tv_sec;
ntvp->time.tv_nsec = atv.tv_nsec;
ntvp->maxerror = time_maxerror;
ntvp->esterror = time_esterror;
ntvp->tai = time_tai;
ntvp->time_state = time_state;
if (ntp_is_time_error(time_status))
ntvp->time_state = TIME_ERROR;
}
/*
* ntp_gettime() - NTP user application interface
*
* See the timex.h header file for synopsis and API description. Note that
* the TAI offset is returned in the ntvtimeval.tai structure member.
*/
#ifndef _SYS_SYSPROTO_H_
struct ntp_gettime_args {
struct ntptimeval *ntvp;
};
#endif
/* ARGSUSED */
int
sys_ntp_gettime(struct thread *td, struct ntp_gettime_args *uap)
{
struct ntptimeval ntv;
NTPADJ_LOCK();
ntp_gettime1(&ntv);
NTPADJ_UNLOCK();
td->td_retval[0] = ntv.time_state;
return (copyout(&ntv, uap->ntvp, sizeof(ntv)));
}
static int
ntp_sysctl(SYSCTL_HANDLER_ARGS)
{
struct ntptimeval ntv; /* temporary structure */
NTPADJ_LOCK();
ntp_gettime1(&ntv);
NTPADJ_UNLOCK();
return (sysctl_handle_opaque(oidp, &ntv, sizeof(ntv), req));
}
SYSCTL_NODE(_kern, OID_AUTO, ntp_pll, CTLFLAG_RW, 0, "");
SYSCTL_PROC(_kern_ntp_pll, OID_AUTO, gettime, CTLTYPE_OPAQUE | CTLFLAG_RD |
CTLFLAG_MPSAFE, 0, sizeof(struct ntptimeval) , ntp_sysctl, "S,ntptimeval",
"");
#ifdef PPS_SYNC
SYSCTL_INT(_kern_ntp_pll, OID_AUTO, pps_shiftmax, CTLFLAG_RW,
&pps_shiftmax, 0, "Max interval duration (sec) (shift)");
SYSCTL_INT(_kern_ntp_pll, OID_AUTO, pps_shift, CTLFLAG_RW,
&pps_shift, 0, "Interval duration (sec) (shift)");
SYSCTL_LONG(_kern_ntp_pll, OID_AUTO, time_monitor, CTLFLAG_RD,
&time_monitor, 0, "Last time offset scaled (ns)");
SYSCTL_S64(_kern_ntp_pll, OID_AUTO, pps_freq, CTLFLAG_RD | CTLFLAG_MPSAFE,
&pps_freq, 0,
"Scaled frequency offset (ns/sec)");
SYSCTL_S64(_kern_ntp_pll, OID_AUTO, time_freq, CTLFLAG_RD | CTLFLAG_MPSAFE,
&time_freq, 0,
"Frequency offset (ns/sec)");
#endif
/*
* ntp_adjtime() - NTP daemon application interface
*
* See the timex.h header file for synopsis and API description. Note that
* the timex.constant structure member has a dual purpose to set the time
* constant and to set the TAI offset.
*/
#ifndef _SYS_SYSPROTO_H_
struct ntp_adjtime_args {
struct timex *tp;
};
#endif
int
sys_ntp_adjtime(struct thread *td, struct ntp_adjtime_args *uap)
{
struct timex ntv; /* temporary structure */
long freq; /* frequency ns/s) */
int modes; /* mode bits from structure */
int error, retval;
error = copyin((caddr_t)uap->tp, (caddr_t)&ntv, sizeof(ntv));
if (error)
return (error);
/*
* Update selected clock variables - only the superuser can
* change anything. Note that there is no error checking here on
* the assumption the superuser should know what it is doing.
* Note that either the time constant or TAI offset are loaded
* from the ntv.constant member, depending on the mode bits. If
* the STA_PLL bit in the status word is cleared, the state and
* status words are reset to the initial values at boot.
*/
modes = ntv.modes;
if (modes)
error = priv_check(td, PRIV_NTP_ADJTIME);
if (error != 0)
return (error);
NTPADJ_LOCK();
if (modes & MOD_MAXERROR)
time_maxerror = ntv.maxerror;
if (modes & MOD_ESTERROR)
time_esterror = ntv.esterror;
if (modes & MOD_STATUS) {
if (time_status & STA_PLL && !(ntv.status & STA_PLL)) {
time_state = TIME_OK;
time_status = STA_UNSYNC;
#ifdef PPS_SYNC
pps_shift = PPS_FAVG;
#endif /* PPS_SYNC */
}
time_status &= STA_RONLY;
time_status |= ntv.status & ~STA_RONLY;
}
if (modes & MOD_TIMECONST) {
if (ntv.constant < 0)
time_constant = 0;
else if (ntv.constant > MAXTC)
time_constant = MAXTC;
else
time_constant = ntv.constant;
}
if (modes & MOD_TAI) {
if (ntv.constant > 0) /* XXX zero & negative numbers ? */
time_tai = ntv.constant;
}
#ifdef PPS_SYNC
if (modes & MOD_PPSMAX) {
if (ntv.shift < PPS_FAVG)
pps_shiftmax = PPS_FAVG;
else if (ntv.shift > PPS_FAVGMAX)
pps_shiftmax = PPS_FAVGMAX;
else
pps_shiftmax = ntv.shift;
}
#endif /* PPS_SYNC */
if (modes & MOD_NANO)
time_status |= STA_NANO;
if (modes & MOD_MICRO)
time_status &= ~STA_NANO;
if (modes & MOD_CLKB)
time_status |= STA_CLK;
if (modes & MOD_CLKA)
time_status &= ~STA_CLK;
if (modes & MOD_FREQUENCY) {
freq = (ntv.freq * 1000LL) >> 16;
if (freq > MAXFREQ)
L_LINT(time_freq, MAXFREQ);
else if (freq < -MAXFREQ)
L_LINT(time_freq, -MAXFREQ);
else {
/*
* ntv.freq is [PPM * 2^16] = [us/s * 2^16]
* time_freq is [ns/s * 2^32]
*/
time_freq = ntv.freq * 1000LL * 65536LL;
}
#ifdef PPS_SYNC
pps_freq = time_freq;
#endif /* PPS_SYNC */
}
if (modes & MOD_OFFSET) {
if (time_status & STA_NANO)
hardupdate(ntv.offset);
else
hardupdate(ntv.offset * 1000);
}
/*
* Retrieve all clock variables. Note that the TAI offset is
* returned only by ntp_gettime();
*/
if (time_status & STA_NANO)
ntv.offset = L_GINT(time_offset);
else
ntv.offset = L_GINT(time_offset) / 1000; /* XXX rounding ? */
ntv.freq = L_GINT((time_freq / 1000LL) << 16);
ntv.maxerror = time_maxerror;
ntv.esterror = time_esterror;
ntv.status = time_status;
ntv.constant = time_constant;
if (time_status & STA_NANO)
ntv.precision = time_precision;
else
ntv.precision = time_precision / 1000;
ntv.tolerance = MAXFREQ * SCALE_PPM;
#ifdef PPS_SYNC
ntv.shift = pps_shift;
ntv.ppsfreq = L_GINT((pps_freq / 1000LL) << 16);
if (time_status & STA_NANO)
ntv.jitter = pps_jitter;
else
ntv.jitter = pps_jitter / 1000;
ntv.stabil = pps_stabil;
ntv.calcnt = pps_calcnt;
ntv.errcnt = pps_errcnt;
ntv.jitcnt = pps_jitcnt;
ntv.stbcnt = pps_stbcnt;
#endif /* PPS_SYNC */
retval = ntp_is_time_error(time_status) ? TIME_ERROR : time_state;
NTPADJ_UNLOCK();
error = copyout((caddr_t)&ntv, (caddr_t)uap->tp, sizeof(ntv));
if (error == 0)
td->td_retval[0] = retval;
return (error);
}
/*
* second_overflow() - called after ntp_tick_adjust()
*
* This routine is ordinarily called immediately following the above
* routine ntp_tick_adjust(). While these two routines are normally
* combined, they are separated here only for the purposes of
* simulation.
*/
void
ntp_update_second(int64_t *adjustment, time_t *newsec)
{
int tickrate;
l_fp ftemp; /* 32/64-bit temporary */
/*
* On rollover of the second both the nanosecond and microsecond
* clocks are updated and the state machine cranked as
* necessary. The phase adjustment to be used for the next
* second is calculated and the maximum error is increased by
* the tolerance.
*/
time_maxerror += MAXFREQ / 1000;
/*
* Leap second processing. If in leap-insert state at
* the end of the day, the system clock is set back one
* second; if in leap-delete state, the system clock is
* set ahead one second. The nano_time() routine or
* external clock driver will insure that reported time
* is always monotonic.
*/
switch (time_state) {
/*
* No warning.
*/
case TIME_OK:
if (time_status & STA_INS)
time_state = TIME_INS;
else if (time_status & STA_DEL)
time_state = TIME_DEL;
break;
/*
* Insert second 23:59:60 following second
* 23:59:59.
*/
case TIME_INS:
if (!(time_status & STA_INS))
time_state = TIME_OK;
else if ((*newsec) % 86400 == 0) {
(*newsec)--;
time_state = TIME_OOP;
time_tai++;
}
break;
/*
* Delete second 23:59:59.
*/
case TIME_DEL:
if (!(time_status & STA_DEL))
time_state = TIME_OK;
else if (((*newsec) + 1) % 86400 == 0) {
(*newsec)++;
time_tai--;
time_state = TIME_WAIT;
}
break;
/*
* Insert second in progress.
*/
case TIME_OOP:
time_state = TIME_WAIT;
break;
/*
* Wait for status bits to clear.
*/
case TIME_WAIT:
if (!(time_status & (STA_INS | STA_DEL)))
time_state = TIME_OK;
}
/*
* Compute the total time adjustment for the next second
* in ns. The offset is reduced by a factor depending on
* whether the PPS signal is operating. Note that the
* value is in effect scaled by the clock frequency,
* since the adjustment is added at each tick interrupt.
*/
ftemp = time_offset;
#ifdef PPS_SYNC
/* XXX even if PPS signal dies we should finish adjustment ? */
if (time_status & STA_PPSTIME && time_status &
STA_PPSSIGNAL)
L_RSHIFT(ftemp, pps_shift);
else
L_RSHIFT(ftemp, SHIFT_PLL + time_constant);
#else
L_RSHIFT(ftemp, SHIFT_PLL + time_constant);
#endif /* PPS_SYNC */
time_adj = ftemp;
L_SUB(time_offset, ftemp);
L_ADD(time_adj, time_freq);
/*
* Apply any correction from adjtime(2). If more than one second
* off we slew at a rate of 5ms/s (5000 PPM) else 500us/s (500PPM)
* until the last second is slewed the final < 500 usecs.
*/
if (time_adjtime != 0) {
if (time_adjtime > 1000000)
tickrate = 5000;
else if (time_adjtime < -1000000)
tickrate = -5000;
else if (time_adjtime > 500)
tickrate = 500;
else if (time_adjtime < -500)
tickrate = -500;
else
tickrate = time_adjtime;
time_adjtime -= tickrate;
L_LINT(ftemp, tickrate * 1000);
L_ADD(time_adj, ftemp);
}
*adjustment = time_adj;
#ifdef PPS_SYNC
if (pps_valid > 0)
pps_valid--;
else
time_status &= ~STA_PPSSIGNAL;
#endif /* PPS_SYNC */
}
/*
* ntp_init() - initialize variables and structures
*
* This routine must be called after the kernel variables hz and tick
* are set or changed and before the next tick interrupt. In this
* particular implementation, these values are assumed set elsewhere in
* the kernel. The design allows the clock frequency and tick interval
* to be changed while the system is running. So, this routine should
* probably be integrated with the code that does that.
*/
static void
ntp_init(void)
{
/*
* The following variables are initialized only at startup. Only
* those structures not cleared by the compiler need to be
* initialized, and these only in the simulator. In the actual
* kernel, any nonzero values here will quickly evaporate.
*/
L_CLR(time_offset);
L_CLR(time_freq);
#ifdef PPS_SYNC
pps_tf[0].tv_sec = pps_tf[0].tv_nsec = 0;
pps_tf[1].tv_sec = pps_tf[1].tv_nsec = 0;
pps_tf[2].tv_sec = pps_tf[2].tv_nsec = 0;
pps_fcount = 0;
L_CLR(pps_freq);
#endif /* PPS_SYNC */
}
SYSINIT(ntpclocks, SI_SUB_CLOCKS, SI_ORDER_MIDDLE, ntp_init, NULL);
/*
* hardupdate() - local clock update
*
* This routine is called by ntp_adjtime() to update the local clock
* phase and frequency. The implementation is of an adaptive-parameter,
* hybrid phase/frequency-lock loop (PLL/FLL). The routine computes new
* time and frequency offset estimates for each call. If the kernel PPS
* discipline code is configured (PPS_SYNC), the PPS signal itself
* determines the new time offset, instead of the calling argument.
* Presumably, calls to ntp_adjtime() occur only when the caller
* believes the local clock is valid within some bound (+-128 ms with
* NTP). If the caller's time is far different than the PPS time, an
* argument will ensue, and it's not clear who will lose.
*
* For uncompensated quartz crystal oscillators and nominal update
* intervals less than 256 s, operation should be in phase-lock mode,
* where the loop is disciplined to phase. For update intervals greater
* than 1024 s, operation should be in frequency-lock mode, where the
* loop is disciplined to frequency. Between 256 s and 1024 s, the mode
* is selected by the STA_MODE status bit.
*/
static void
hardupdate(offset)
long offset; /* clock offset (ns) */
{
long mtemp;
l_fp ftemp;
NTPADJ_ASSERT_LOCKED();
/*
* Select how the phase is to be controlled and from which
* source. If the PPS signal is present and enabled to
* discipline the time, the PPS offset is used; otherwise, the
* argument offset is used.
*/
if (!(time_status & STA_PLL))
return;
if (!(time_status & STA_PPSTIME && time_status &
STA_PPSSIGNAL)) {
if (offset > MAXPHASE)
time_monitor = MAXPHASE;
else if (offset < -MAXPHASE)
time_monitor = -MAXPHASE;
else
time_monitor = offset;
L_LINT(time_offset, time_monitor);
}
/*
* Select how the frequency is to be controlled and in which
* mode (PLL or FLL). If the PPS signal is present and enabled
* to discipline the frequency, the PPS frequency is used;
* otherwise, the argument offset is used to compute it.
*/
if (time_status & STA_PPSFREQ && time_status & STA_PPSSIGNAL) {
time_reftime = time_uptime;
return;
}
if (time_status & STA_FREQHOLD || time_reftime == 0)
time_reftime = time_uptime;
mtemp = time_uptime - time_reftime;
L_LINT(ftemp, time_monitor);
L_RSHIFT(ftemp, (SHIFT_PLL + 2 + time_constant) << 1);
L_MPY(ftemp, mtemp);
L_ADD(time_freq, ftemp);
time_status &= ~STA_MODE;
if (mtemp >= MINSEC && (time_status & STA_FLL || mtemp >
MAXSEC)) {
L_LINT(ftemp, (time_monitor << 4) / mtemp);
L_RSHIFT(ftemp, SHIFT_FLL + 4);
L_ADD(time_freq, ftemp);
time_status |= STA_MODE;
}
time_reftime = time_uptime;
if (L_GINT(time_freq) > MAXFREQ)
L_LINT(time_freq, MAXFREQ);
else if (L_GINT(time_freq) < -MAXFREQ)
L_LINT(time_freq, -MAXFREQ);
}
#ifdef PPS_SYNC
/*
* hardpps() - discipline CPU clock oscillator to external PPS signal
*
* This routine is called at each PPS interrupt in order to discipline
* the CPU clock oscillator to the PPS signal. There are two independent
* first-order feedback loops, one for the phase, the other for the
* frequency. The phase loop measures and grooms the PPS phase offset
* and leaves it in a handy spot for the seconds overflow routine. The
* frequency loop averages successive PPS phase differences and
* calculates the PPS frequency offset, which is also processed by the
* seconds overflow routine. The code requires the caller to capture the
* time and architecture-dependent hardware counter values in
* nanoseconds at the on-time PPS signal transition.
*
* Note that, on some Unix systems this routine runs at an interrupt
* priority level higher than the timer interrupt routine hardclock().
* Therefore, the variables used are distinct from the hardclock()
* variables, except for the actual time and frequency variables, which
* are determined by this routine and updated atomically.
*/
void
hardpps(tsp, nsec)
struct timespec *tsp; /* time at PPS */
long nsec; /* hardware counter at PPS */
{
long u_sec, u_nsec, v_nsec; /* temps */
l_fp ftemp;
NTPADJ_LOCK();
/*
* The signal is first processed by a range gate and frequency
* discriminator. The range gate rejects noise spikes outside
* the range +-500 us. The frequency discriminator rejects input
* signals with apparent frequency outside the range 1 +-500
* PPM. If two hits occur in the same second, we ignore the
* later hit; if not and a hit occurs outside the range gate,
* keep the later hit for later comparison, but do not process
* it.
*/
time_status |= STA_PPSSIGNAL | STA_PPSJITTER;
time_status &= ~(STA_PPSWANDER | STA_PPSERROR);
pps_valid = PPS_VALID;
u_sec = tsp->tv_sec;
u_nsec = tsp->tv_nsec;
if (u_nsec >= (NANOSECOND >> 1)) {
u_nsec -= NANOSECOND;
u_sec++;
}
v_nsec = u_nsec - pps_tf[0].tv_nsec;
if (u_sec == pps_tf[0].tv_sec && v_nsec < NANOSECOND - MAXFREQ)
goto out;
pps_tf[2] = pps_tf[1];
pps_tf[1] = pps_tf[0];
pps_tf[0].tv_sec = u_sec;
pps_tf[0].tv_nsec = u_nsec;
/*
* Compute the difference between the current and previous
* counter values. If the difference exceeds 0.5 s, assume it
* has wrapped around, so correct 1.0 s. If the result exceeds
* the tick interval, the sample point has crossed a tick
* boundary during the last second, so correct the tick. Very
* intricate.
*/
u_nsec = nsec;
if (u_nsec > (NANOSECOND >> 1))
u_nsec -= NANOSECOND;
else if (u_nsec < -(NANOSECOND >> 1))
u_nsec += NANOSECOND;
pps_fcount += u_nsec;
if (v_nsec > MAXFREQ || v_nsec < -MAXFREQ)
goto out;
time_status &= ~STA_PPSJITTER;
/*
* A three-stage median filter is used to help denoise the PPS
* time. The median sample becomes the time offset estimate; the
* difference between the other two samples becomes the time
* dispersion (jitter) estimate.
*/
if (pps_tf[0].tv_nsec > pps_tf[1].tv_nsec) {
if (pps_tf[1].tv_nsec > pps_tf[2].tv_nsec) {
v_nsec = pps_tf[1].tv_nsec; /* 0 1 2 */
u_nsec = pps_tf[0].tv_nsec - pps_tf[2].tv_nsec;
} else if (pps_tf[2].tv_nsec > pps_tf[0].tv_nsec) {
v_nsec = pps_tf[0].tv_nsec; /* 2 0 1 */
u_nsec = pps_tf[2].tv_nsec - pps_tf[1].tv_nsec;
} else {
v_nsec = pps_tf[2].tv_nsec; /* 0 2 1 */
u_nsec = pps_tf[0].tv_nsec - pps_tf[1].tv_nsec;
}
} else {
if (pps_tf[1].tv_nsec < pps_tf[2].tv_nsec) {
v_nsec = pps_tf[1].tv_nsec; /* 2 1 0 */
u_nsec = pps_tf[2].tv_nsec - pps_tf[0].tv_nsec;
} else if (pps_tf[2].tv_nsec < pps_tf[0].tv_nsec) {
v_nsec = pps_tf[0].tv_nsec; /* 1 0 2 */
u_nsec = pps_tf[1].tv_nsec - pps_tf[2].tv_nsec;
} else {
v_nsec = pps_tf[2].tv_nsec; /* 1 2 0 */
u_nsec = pps_tf[1].tv_nsec - pps_tf[0].tv_nsec;
}
}
/*
* Nominal jitter is due to PPS signal noise and interrupt
* latency. If it exceeds the popcorn threshold, the sample is
* discarded. otherwise, if so enabled, the time offset is
* updated. We can tolerate a modest loss of data here without
* much degrading time accuracy.
*
* The measurements being checked here were made with the system
* timecounter, so the popcorn threshold is not allowed to fall below
* the number of nanoseconds in two ticks of the timecounter. For a
* timecounter running faster than 1 GHz the lower bound is 2ns, just
* to avoid a nonsensical threshold of zero.
*/
if (u_nsec > lmax(pps_jitter << PPS_POPCORN,
2 * (NANOSECOND / (long)qmin(NANOSECOND, tc_getfrequency())))) {
time_status |= STA_PPSJITTER;
pps_jitcnt++;
} else if (time_status & STA_PPSTIME) {
time_monitor = -v_nsec;
L_LINT(time_offset, time_monitor);
}
pps_jitter += (u_nsec - pps_jitter) >> PPS_FAVG;
u_sec = pps_tf[0].tv_sec - pps_lastsec;
if (u_sec < (1 << pps_shift))
goto out;
/*
* At the end of the calibration interval the difference between
* the first and last counter values becomes the scaled
* frequency. It will later be divided by the length of the
* interval to determine the frequency update. If the frequency
* exceeds a sanity threshold, or if the actual calibration
* interval is not equal to the expected length, the data are
* discarded. We can tolerate a modest loss of data here without
* much degrading frequency accuracy.
*/
pps_calcnt++;
v_nsec = -pps_fcount;
pps_lastsec = pps_tf[0].tv_sec;
pps_fcount = 0;
u_nsec = MAXFREQ << pps_shift;
if (v_nsec > u_nsec || v_nsec < -u_nsec || u_sec != (1 << pps_shift)) {
time_status |= STA_PPSERROR;
pps_errcnt++;
goto out;
}
/*
* Here the raw frequency offset and wander (stability) is
* calculated. If the wander is less than the wander threshold
* for four consecutive averaging intervals, the interval is
* doubled; if it is greater than the threshold for four
* consecutive intervals, the interval is halved. The scaled
* frequency offset is converted to frequency offset. The
* stability metric is calculated as the average of recent
* frequency changes, but is used only for performance
* monitoring.
*/
L_LINT(ftemp, v_nsec);
L_RSHIFT(ftemp, pps_shift);
L_SUB(ftemp, pps_freq);
u_nsec = L_GINT(ftemp);
if (u_nsec > PPS_MAXWANDER) {
L_LINT(ftemp, PPS_MAXWANDER);
pps_intcnt--;
time_status |= STA_PPSWANDER;
pps_stbcnt++;
} else if (u_nsec < -PPS_MAXWANDER) {
L_LINT(ftemp, -PPS_MAXWANDER);
pps_intcnt--;
time_status |= STA_PPSWANDER;
pps_stbcnt++;
} else {
pps_intcnt++;
}
if (pps_intcnt >= 4) {
pps_intcnt = 4;
if (pps_shift < pps_shiftmax) {
pps_shift++;
pps_intcnt = 0;
}
} else if (pps_intcnt <= -4 || pps_shift > pps_shiftmax) {
pps_intcnt = -4;
if (pps_shift > PPS_FAVG) {
pps_shift--;
pps_intcnt = 0;
}
}
if (u_nsec < 0)
u_nsec = -u_nsec;
pps_stabil += (u_nsec * SCALE_PPM - pps_stabil) >> PPS_FAVG;
/*
* The PPS frequency is recalculated and clamped to the maximum
* MAXFREQ. If enabled, the system clock frequency is updated as
* well.
*/
L_ADD(pps_freq, ftemp);
u_nsec = L_GINT(pps_freq);
if (u_nsec > MAXFREQ)
L_LINT(pps_freq, MAXFREQ);
else if (u_nsec < -MAXFREQ)
L_LINT(pps_freq, -MAXFREQ);
if (time_status & STA_PPSFREQ)
time_freq = pps_freq;
out:
NTPADJ_UNLOCK();
}
#endif /* PPS_SYNC */
#ifndef _SYS_SYSPROTO_H_
struct adjtime_args {
struct timeval *delta;
struct timeval *olddelta;
};
#endif
/* ARGSUSED */
int
sys_adjtime(struct thread *td, struct adjtime_args *uap)
{
struct timeval delta, olddelta, *deltap;
int error;
if (uap->delta) {
error = copyin(uap->delta, &delta, sizeof(delta));
if (error)
return (error);
deltap = &delta;
} else
deltap = NULL;
error = kern_adjtime(td, deltap, &olddelta);
if (uap->olddelta && error == 0)
error = copyout(&olddelta, uap->olddelta, sizeof(olddelta));
return (error);
}
int
kern_adjtime(struct thread *td, struct timeval *delta, struct timeval *olddelta)
{
struct timeval atv;
int64_t ltr, ltw;
int error;
if (delta != NULL) {
error = priv_check(td, PRIV_ADJTIME);
if (error != 0)
return (error);
ltw = (int64_t)delta->tv_sec * 1000000 + delta->tv_usec;
}
NTPADJ_LOCK();
ltr = time_adjtime;
if (delta != NULL)
time_adjtime = ltw;
NTPADJ_UNLOCK();
if (olddelta != NULL) {
atv.tv_sec = ltr / 1000000;
atv.tv_usec = ltr % 1000000;
if (atv.tv_usec < 0) {
atv.tv_usec += 1000000;
atv.tv_sec--;
}
*olddelta = atv;
}
return (0);
}
static struct callout resettodr_callout;
static int resettodr_period = 1800;
static void
periodic_resettodr(void *arg __unused)
{
/*
* Read of time_status is lock-less, which is fine since
* ntp_is_time_error() operates on the consistent read value.
*/
if (!ntp_is_time_error(time_status))
resettodr();
if (resettodr_period > 0)
callout_schedule(&resettodr_callout, resettodr_period * hz);
}
static void
shutdown_resettodr(void *arg __unused, int howto __unused)
{
callout_drain(&resettodr_callout);
/* Another unlocked read of time_status */
if (resettodr_period > 0 && !ntp_is_time_error(time_status))
resettodr();
}
static int
sysctl_resettodr_period(SYSCTL_HANDLER_ARGS)
{
int error;
error = sysctl_handle_int(oidp, oidp->oid_arg1, oidp->oid_arg2, req);
if (error || !req->newptr)
return (error);
if (cold)
goto done;
if (resettodr_period == 0)
callout_stop(&resettodr_callout);
else
callout_reset(&resettodr_callout, resettodr_period * hz,
periodic_resettodr, NULL);
done:
return (0);
}
SYSCTL_PROC(_machdep, OID_AUTO, rtc_save_period, CTLTYPE_INT | CTLFLAG_RWTUN |
CTLFLAG_MPSAFE, &resettodr_period, 1800, sysctl_resettodr_period, "I",
"Save system time to RTC with this period (in seconds)");
static void
start_periodic_resettodr(void *arg __unused)
{
EVENTHANDLER_REGISTER(shutdown_pre_sync, shutdown_resettodr, NULL,
SHUTDOWN_PRI_FIRST);
callout_init(&resettodr_callout, 1);
if (resettodr_period == 0)
return;
callout_reset(&resettodr_callout, resettodr_period * hz,
periodic_resettodr, NULL);
}
SYSINIT(periodic_resettodr, SI_SUB_LAST, SI_ORDER_MIDDLE,
start_periodic_resettodr, NULL);
Reference in New Issue View Git Blame Copy Permalink