From 0f295b0650c90362b4111f46d7f9149a0a4191be Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Fri, 13 Oct 2017 11:54:33 -0600 Subject: rtc: Allow rtc drivers to specify the tv_nsec value for ntp ntp is currently hardwired to try and call the rtc set when wall clock tv_nsec is 0.5 seconds. This historical behaviour works well with certain PC RTCs, but is not universal to all rtc hardware. Change how this works by introducing the driver specific concept of set_offset_nsec, the delay between current wall clock time and the target time to set (with a 0 tv_nsecs). For x86-style CMOS set_offset_nsec should be -0.5 s which causes the last second to be written 0.5 s after it has started. For compat with the old rtc_set_ntp_time, the value is defaulted to + 0.5 s, which causes the next second to be written 0.5s before it starts, as things were before this patch. Testing shows many non-x86 RTCs would like set_offset_nsec ~= 0, so ultimately each RTC driver should set the set_offset_nsec according to its needs, and non x86 architectures should stop using update_persistent_clock64 in order to access this feature. Future patches will revise the drivers as needed. Since CMOS and RTC now have very different handling they are split into two dedicated code paths, sharing the support code, and ifdefs are replaced with IS_ENABLED. Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Miroslav Lichvar Cc: Richard Cochran Cc: Prarit Bhargava Cc: Stephen Boyd Signed-off-by: Jason Gunthorpe Signed-off-by: John Stultz --- kernel/time/ntp.c | 166 +++++++++++++++++++++++++++++++++++++----------------- 1 file changed, 113 insertions(+), 53 deletions(-) (limited to 'kernel/time/ntp.c') diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c index edf19cc53140..bc19de1a0683 100644 --- a/kernel/time/ntp.c +++ b/kernel/time/ntp.c @@ -492,6 +492,67 @@ out: return leap; } +static void sync_hw_clock(struct work_struct *work); +static DECLARE_DELAYED_WORK(sync_work, sync_hw_clock); + +static void sched_sync_hw_clock(struct timespec64 now, + unsigned long target_nsec, bool fail) + +{ + struct timespec64 next; + + getnstimeofday64(&next); + if (!fail) + next.tv_sec = 659; + else { + /* + * Try again as soon as possible. Delaying long periods + * decreases the accuracy of the work queue timer. Due to this + * the algorithm is very likely to require a short-sleep retry + * after the above long sleep to synchronize ts_nsec. + */ + next.tv_sec = 0; + } + + /* Compute the needed delay that will get to tv_nsec == target_nsec */ + next.tv_nsec = target_nsec - next.tv_nsec; + if (next.tv_nsec <= 0) + next.tv_nsec += NSEC_PER_SEC; + if (next.tv_nsec >= NSEC_PER_SEC) { + next.tv_sec++; + next.tv_nsec -= NSEC_PER_SEC; + } + + queue_delayed_work(system_power_efficient_wq, &sync_work, + timespec64_to_jiffies(&next)); +} + +static void sync_rtc_clock(void) +{ + unsigned long target_nsec; + struct timespec64 adjust, now; + int rc; + + if (!IS_ENABLED(CONFIG_RTC_SYSTOHC)) + return; + + getnstimeofday64(&now); + + adjust = now; + if (persistent_clock_is_local) + adjust.tv_sec -= (sys_tz.tz_minuteswest * 60); + + /* + * The current RTC in use will provide the target_nsec it wants to be + * called at, and does rtc_tv_nsec_ok internally. + */ + rc = rtc_set_ntp_time(adjust, &target_nsec); + if (rc == -ENODEV) + return; + + sched_sync_hw_clock(now, target_nsec, rc); +} + #ifdef CONFIG_GENERIC_CMOS_UPDATE int __weak update_persistent_clock(struct timespec now) { @@ -507,76 +568,75 @@ int __weak update_persistent_clock64(struct timespec64 now64) } #endif -#if defined(CONFIG_GENERIC_CMOS_UPDATE) || defined(CONFIG_RTC_SYSTOHC) -static void sync_cmos_clock(struct work_struct *work); - -static DECLARE_DELAYED_WORK(sync_cmos_work, sync_cmos_clock); - -static void sync_cmos_clock(struct work_struct *work) +static bool sync_cmos_clock(void) { + static bool no_cmos; struct timespec64 now; - struct timespec64 next; - int fail = 1; + struct timespec64 adjust; + int rc = -EPROTO; + long target_nsec = NSEC_PER_SEC / 2; + + if (!IS_ENABLED(CONFIG_GENERIC_CMOS_UPDATE)) + return false; + + if (no_cmos) + return false; /* - * If we have an externally synchronized Linux clock, then update - * CMOS clock accordingly every ~11 minutes. Set_rtc_mmss() has to be - * called as close as possible to 500 ms before the new second starts. - * This code is run on a timer. If the clock is set, that timer - * may not expire at the correct time. Thus, we adjust... - * We want the clock to be within a couple of ticks from the target. + * Historically update_persistent_clock64() has followed x86 + * semantics, which match the MC146818A/etc RTC. This RTC will store + * 'adjust' and then in .5s it will advance once second. + * + * Architectures are strongly encouraged to use rtclib and not + * implement this legacy API. */ - if (!ntp_synced()) { - /* - * Not synced, exit, do not restart a timer (if one is - * running, let it run out). - */ - return; - } - getnstimeofday64(&now); - if (abs(now.tv_nsec - (NSEC_PER_SEC / 2)) <= tick_nsec * 5) { - struct timespec64 adjust = now; - - fail = -ENODEV; + if (rtc_tv_nsec_ok(-1 * target_nsec, &adjust, &now)) { if (persistent_clock_is_local) adjust.tv_sec -= (sys_tz.tz_minuteswest * 60); -#ifdef CONFIG_GENERIC_CMOS_UPDATE - fail = update_persistent_clock64(adjust); -#endif - -#ifdef CONFIG_RTC_SYSTOHC - if (fail == -ENODEV) - fail = rtc_set_ntp_time(adjust); -#endif + rc = update_persistent_clock64(adjust); + /* + * The machine does not support update_persistent_clock64 even + * though it defines CONFIG_GENERIC_CMOS_UPDATE. + */ + if (rc == -ENODEV) { + no_cmos = true; + return false; + } } - next.tv_nsec = (NSEC_PER_SEC / 2) - now.tv_nsec - (TICK_NSEC / 2); - if (next.tv_nsec <= 0) - next.tv_nsec += NSEC_PER_SEC; + sched_sync_hw_clock(now, target_nsec, rc); + return true; +} - if (!fail || fail == -ENODEV) - next.tv_sec = 659; - else - next.tv_sec = 0; +/* + * If we have an externally synchronized Linux clock, then update RTC clock + * accordingly every ~11 minutes. Generally RTCs can only store second + * precision, but many RTCs will adjust the phase of their second tick to + * match the moment of update. This infrastructure arranges to call to the RTC + * set at the correct moment to phase synchronize the RTC second tick over + * with the kernel clock. + */ +static void sync_hw_clock(struct work_struct *work) +{ + if (!ntp_synced()) + return; - if (next.tv_nsec >= NSEC_PER_SEC) { - next.tv_sec++; - next.tv_nsec -= NSEC_PER_SEC; - } - queue_delayed_work(system_power_efficient_wq, - &sync_cmos_work, timespec64_to_jiffies(&next)); + if (sync_cmos_clock()) + return; + + sync_rtc_clock(); } void ntp_notify_cmos_timer(void) { - queue_delayed_work(system_power_efficient_wq, &sync_cmos_work, 0); -} - -#else -void ntp_notify_cmos_timer(void) { } -#endif + if (!ntp_synced()) + return; + if (IS_ENABLED(CONFIG_GENERIC_CMOS_UPDATE) || + IS_ENABLED(CONFIG_RTC_SYSTOHC)) + queue_delayed_work(system_power_efficient_wq, &sync_work, 0); +} /* * Propagate a new txc->status value into the NTP state: -- cgit v1.2.3-59-g8ed1b From e0956dcc4ba74ec4b17e32fc9a156fcba1ef6610 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 19 Oct 2017 13:14:44 +0200 Subject: timekeeping: Consolidate timekeeping_inject_offset code The code to check the adjtimex() or clock_adjtime() arguments is spread out across multiple files for presumably only historic reasons. As a preparatation for a rework to get rid of the use of 'struct timeval' and 'struct timespec' in there, this moves all the portions into kernel/time/timekeeping.c and marks them as 'static'. The warp_clock() function here is not as closely related as the others, but I feel it still makes sense to move it here in order to consolidate all callers of timekeeping_inject_offset(). Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Miroslav Lichvar Cc: Richard Cochran Cc: Prarit Bhargava Cc: Stephen Boyd Signed-off-by: Arnd Bergmann [jstultz: Whitespace fixup] Signed-off-by: John Stultz --- include/linux/time.h | 26 ---------- kernel/time/ntp.c | 61 ---------------------- kernel/time/ntp_internal.h | 1 - kernel/time/time.c | 36 +------------ kernel/time/timekeeping.c | 123 ++++++++++++++++++++++++++++++++++++++++++++- kernel/time/timekeeping.h | 2 +- 6 files changed, 123 insertions(+), 126 deletions(-) (limited to 'kernel/time/ntp.c') diff --git a/include/linux/time.h b/include/linux/time.h index 9bc1f945777c..c0fbad08448f 100644 --- a/include/linux/time.h +++ b/include/linux/time.h @@ -134,32 +134,6 @@ static inline bool timeval_valid(const struct timeval *tv) extern struct timespec timespec_trunc(struct timespec t, unsigned gran); -/* - * Validates if a timespec/timeval used to inject a time offset is valid. - * Offsets can be postive or negative. The value of the timeval/timespec - * is the sum of its fields, but *NOTE*: the field tv_usec/tv_nsec must - * always be non-negative. - */ -static inline bool timeval_inject_offset_valid(const struct timeval *tv) -{ - /* We don't check the tv_sec as it can be positive or negative */ - - /* Can't have more microseconds then a second */ - if (tv->tv_usec < 0 || tv->tv_usec >= USEC_PER_SEC) - return false; - return true; -} - -static inline bool timespec_inject_offset_valid(const struct timespec *ts) -{ - /* We don't check the tv_sec as it can be positive or negative */ - - /* Can't have more nanoseconds then a second */ - if (ts->tv_nsec < 0 || ts->tv_nsec >= NSEC_PER_SEC) - return false; - return true; -} - /* Some architectures do not supply their own clocksource. * This is mainly the case in architectures that get their * inter-tick times by reading the counter on their interval diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c index bc19de1a0683..90f84582a076 100644 --- a/kernel/time/ntp.c +++ b/kernel/time/ntp.c @@ -713,67 +713,6 @@ static inline void process_adjtimex_modes(struct timex *txc, } - -/** - * ntp_validate_timex - Ensures the timex is ok for use in do_adjtimex - */ -int ntp_validate_timex(struct timex *txc) -{ - if (txc->modes & ADJ_ADJTIME) { - /* singleshot must not be used with any other mode bits */ - if (!(txc->modes & ADJ_OFFSET_SINGLESHOT)) - return -EINVAL; - if (!(txc->modes & ADJ_OFFSET_READONLY) && - !capable(CAP_SYS_TIME)) - return -EPERM; - } else { - /* In order to modify anything, you gotta be super-user! */ - if (txc->modes && !capable(CAP_SYS_TIME)) - return -EPERM; - /* - * if the quartz is off by more than 10% then - * something is VERY wrong! - */ - if (txc->modes & ADJ_TICK && - (txc->tick < 900000/USER_HZ || - txc->tick > 1100000/USER_HZ)) - return -EINVAL; - } - - if (txc->modes & ADJ_SETOFFSET) { - /* In order to inject time, you gotta be super-user! */ - if (!capable(CAP_SYS_TIME)) - return -EPERM; - - if (txc->modes & ADJ_NANO) { - struct timespec ts; - - ts.tv_sec = txc->time.tv_sec; - ts.tv_nsec = txc->time.tv_usec; - if (!timespec_inject_offset_valid(&ts)) - return -EINVAL; - - } else { - if (!timeval_inject_offset_valid(&txc->time)) - return -EINVAL; - } - } - - /* - * Check for potential multiplication overflows that can - * only happen on 64-bit systems: - */ - if ((txc->modes & ADJ_FREQUENCY) && (BITS_PER_LONG == 64)) { - if (LLONG_MIN / PPM_SCALE > txc->freq) - return -EINVAL; - if (LLONG_MAX / PPM_SCALE < txc->freq) - return -EINVAL; - } - - return 0; -} - - /* * adjtimex mainly allows reading (and writing, if superuser) of * kernel time-keeping variables. used by xntpd. diff --git a/kernel/time/ntp_internal.h b/kernel/time/ntp_internal.h index d8a7c11fa71a..74b52cd48209 100644 --- a/kernel/time/ntp_internal.h +++ b/kernel/time/ntp_internal.h @@ -7,7 +7,6 @@ extern void ntp_clear(void); extern u64 ntp_tick_length(void); extern ktime_t ntp_get_next_leap(void); extern int second_overflow(time64_t secs); -extern int ntp_validate_timex(struct timex *); extern int __do_adjtimex(struct timex *, struct timespec64 *, s32 *); extern void __hardpps(const struct timespec64 *, const struct timespec64 *); #endif /* _LINUX_NTP_INTERNAL_H */ diff --git a/kernel/time/time.c b/kernel/time/time.c index 44a8c1402133..04684e294f00 100644 --- a/kernel/time/time.c +++ b/kernel/time/time.c @@ -157,40 +157,6 @@ SYSCALL_DEFINE2(gettimeofday, struct timeval __user *, tv, return 0; } -/* - * Indicates if there is an offset between the system clock and the hardware - * clock/persistent clock/rtc. - */ -int persistent_clock_is_local; - -/* - * Adjust the time obtained from the CMOS to be UTC time instead of - * local time. - * - * This is ugly, but preferable to the alternatives. Otherwise we - * would either need to write a program to do it in /etc/rc (and risk - * confusion if the program gets run more than once; it would also be - * hard to make the program warp the clock precisely n hours) or - * compile in the timezone information into the kernel. Bad, bad.... - * - * - TYT, 1992-01-01 - * - * The best thing to do is to keep the CMOS clock in universal time (UTC) - * as real UNIX machines always do it. This avoids all headaches about - * daylight saving times and warping kernel clocks. - */ -static inline void warp_clock(void) -{ - if (sys_tz.tz_minuteswest != 0) { - struct timespec adjust; - - persistent_clock_is_local = 1; - adjust.tv_sec = sys_tz.tz_minuteswest * 60; - adjust.tv_nsec = 0; - timekeeping_inject_offset(&adjust); - } -} - /* * In case for some reason the CMOS clock has not already been running * in UTC, but in some local time: The first time we set the timezone, @@ -224,7 +190,7 @@ int do_sys_settimeofday64(const struct timespec64 *tv, const struct timezone *tz if (firsttime) { firsttime = 0; if (!tv) - warp_clock(); + timekeeping_warp_clock(); } } if (tv) diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 2cafb49aa65e..7d8e0e842484 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -1258,13 +1258,39 @@ out: } EXPORT_SYMBOL(do_settimeofday64); +/* + * Validates if a timespec/timeval used to inject a time offset is valid. + * Offsets can be postive or negative. The value of the timeval/timespec + * is the sum of its fields, but *NOTE*: the field tv_usec/tv_nsec must + * always be non-negative. + */ +static inline bool timeval_inject_offset_valid(const struct timeval *tv) +{ + /* We don't check the tv_sec as it can be positive or negative */ + + /* Can't have more microseconds then a second */ + if (tv->tv_usec < 0 || tv->tv_usec >= USEC_PER_SEC) + return false; + return true; +} + +static inline bool timespec_inject_offset_valid(const struct timespec *ts) +{ + /* We don't check the tv_sec as it can be positive or negative */ + + /* Can't have more nanoseconds then a second */ + if (ts->tv_nsec < 0 || ts->tv_nsec >= NSEC_PER_SEC) + return false; + return true; +} + /** * timekeeping_inject_offset - Adds or subtracts from the current time. * @tv: pointer to the timespec variable containing the offset * * Adds or subtracts an offset value from the current time. */ -int timekeeping_inject_offset(struct timespec *ts) +static int timekeeping_inject_offset(struct timespec *ts) { struct timekeeper *tk = &tk_core.timekeeper; unsigned long flags; @@ -1303,7 +1329,40 @@ error: /* even if we error out, we forwarded the time, so call update */ return ret; } -EXPORT_SYMBOL(timekeeping_inject_offset); + +/* + * Indicates if there is an offset between the system clock and the hardware + * clock/persistent clock/rtc. + */ +int persistent_clock_is_local; + +/* + * Adjust the time obtained from the CMOS to be UTC time instead of + * local time. + * + * This is ugly, but preferable to the alternatives. Otherwise we + * would either need to write a program to do it in /etc/rc (and risk + * confusion if the program gets run more than once; it would also be + * hard to make the program warp the clock precisely n hours) or + * compile in the timezone information into the kernel. Bad, bad.... + * + * - TYT, 1992-01-01 + * + * The best thing to do is to keep the CMOS clock in universal time (UTC) + * as real UNIX machines always do it. This avoids all headaches about + * daylight saving times and warping kernel clocks. + */ +void timekeeping_warp_clock(void) +{ + if (sys_tz.tz_minuteswest != 0) { + struct timespec adjust; + + persistent_clock_is_local = 1; + adjust.tv_sec = sys_tz.tz_minuteswest * 60; + adjust.tv_nsec = 0; + timekeeping_inject_offset(&adjust); + } +} /** * __timekeeping_set_tai_offset - Sets the TAI offset from UTC and monotonic @@ -2247,6 +2306,66 @@ ktime_t ktime_get_update_offsets_now(unsigned int *cwsseq, ktime_t *offs_real, return base; } +/** + * ntp_validate_timex - Ensures the timex is ok for use in do_adjtimex + */ +static int ntp_validate_timex(struct timex *txc) +{ + if (txc->modes & ADJ_ADJTIME) { + /* singleshot must not be used with any other mode bits */ + if (!(txc->modes & ADJ_OFFSET_SINGLESHOT)) + return -EINVAL; + if (!(txc->modes & ADJ_OFFSET_READONLY) && + !capable(CAP_SYS_TIME)) + return -EPERM; + } else { + /* In order to modify anything, you gotta be super-user! */ + if (txc->modes && !capable(CAP_SYS_TIME)) + return -EPERM; + /* + * if the quartz is off by more than 10% then + * something is VERY wrong! + */ + if (txc->modes & ADJ_TICK && + (txc->tick < 900000/USER_HZ || + txc->tick > 1100000/USER_HZ)) + return -EINVAL; + } + + if (txc->modes & ADJ_SETOFFSET) { + /* In order to inject time, you gotta be super-user! */ + if (!capable(CAP_SYS_TIME)) + return -EPERM; + + if (txc->modes & ADJ_NANO) { + struct timespec ts; + + ts.tv_sec = txc->time.tv_sec; + ts.tv_nsec = txc->time.tv_usec; + if (!timespec_inject_offset_valid(&ts)) + return -EINVAL; + + } else { + if (!timeval_inject_offset_valid(&txc->time)) + return -EINVAL; + } + } + + /* + * Check for potential multiplication overflows that can + * only happen on 64-bit systems: + */ + if ((txc->modes & ADJ_FREQUENCY) && (BITS_PER_LONG == 64)) { + if (LLONG_MIN / PPM_SCALE > txc->freq) + return -EINVAL; + if (LLONG_MAX / PPM_SCALE < txc->freq) + return -EINVAL; + } + + return 0; +} + + /** * do_adjtimex() - Accessor function to NTP __do_adjtimex function */ diff --git a/kernel/time/timekeeping.h b/kernel/time/timekeeping.h index d0914676d4c5..44aec7893cdd 100644 --- a/kernel/time/timekeeping.h +++ b/kernel/time/timekeeping.h @@ -10,7 +10,7 @@ extern ktime_t ktime_get_update_offsets_now(unsigned int *cwsseq, extern int timekeeping_valid_for_hres(void); extern u64 timekeeping_max_deferment(void); -extern int timekeeping_inject_offset(struct timespec *ts); +extern void timekeeping_warp_clock(void); extern int timekeeping_suspend(void); extern void timekeeping_resume(void); -- cgit v1.2.3-59-g8ed1b