Merge branch 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull timer fixes from Thomas Gleixner:
 "The timer departement delivers:

   - a regression fix for the NTP code along with a proper selftest
   - prevent a spurious timer interrupt in the NOHZ lowres code
   - a fix for user space interfaces returning the remaining time on
     architectures with CONFIG_TIME_LOW_RES=y
   - a few patches to fix COMPILE_TEST fallout"

* 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  tick/nohz: Set the correct expiry when switching to nohz/lowres mode
  clocksource: Fix dependencies for archs w/o HAS_IOMEM
  clocksource: Select CLKSRC_MMIO where needed
  tick/sched: Hide unused oneshot timer code
  kselftests: timers: Add adjtimex SETOFFSET validity tests
  ntp: Fix ADJ_SETOFFSET being used w/ ADJ_NANO
  itimers: Handle relative timers with CONFIG_TIME_LOW_RES proper
  posix-timers: Handle relative timers with CONFIG_TIME_LOW_RES proper
  timerfd: Handle relative timers with CONFIG_TIME_LOW_RES proper
  hrtimer: Handle remaining time proper for TIME_LOW_RES
  clockevents/tcb_clksrc: Prevent disabling an already disabled clock

drivers/clocksource/Kconfig

@@ -30,6 +30,8 @@ config CLKSRC_MMIO
 config DIGICOLOR_TIMER
 	bool "Digicolor timer driver" if COMPILE_TEST
 	depends on GENERIC_CLOCKEVENTS
+	select CLKSRC_MMIO
+	depends on HAS_IOMEM
 	help
 	  Enables the support for the digicolor timer driver.
 
@@ -55,6 +57,7 @@ config ARMADA_370_XP_TIMER
 	bool "Armada 370 and XP timer driver" if COMPILE_TEST
 	depends on ARM
 	select CLKSRC_OF
+	select CLKSRC_MMIO
 	help
 	  Enables the support for the Armada 370 and XP timer driver.
 
@@ -76,6 +79,7 @@ config ORION_TIMER
 config SUN4I_TIMER
 	bool "Sun4i timer driver" if COMPILE_TEST
 	depends on GENERIC_CLOCKEVENTS
+	depends on HAS_IOMEM
 	select CLKSRC_MMIO
 	help
 	  Enables support for the Sun4i timer.
@@ -89,6 +93,7 @@ config SUN5I_HSTIMER
 
 config TEGRA_TIMER
 	bool "Tegra timer driver" if COMPILE_TEST
+	select CLKSRC_MMIO
 	depends on ARM
 	help
 	  Enables support for the Tegra driver.
@@ -96,6 +101,7 @@ config TEGRA_TIMER
 config VT8500_TIMER
 	bool "VT8500 timer driver" if COMPILE_TEST
 	depends on GENERIC_CLOCKEVENTS
+	depends on HAS_IOMEM
 	help
 	  Enables support for the VT8500 driver.
 
@@ -131,6 +137,7 @@ config CLKSRC_NOMADIK_MTU_SCHED_CLOCK
 config CLKSRC_DBX500_PRCMU
 	bool "Clocksource PRCMU Timer" if COMPILE_TEST
 	depends on GENERIC_CLOCKEVENTS
+	depends on HAS_IOMEM
 	help
 	  Use the always on PRCMU Timer as clocksource
 
@@ -248,6 +255,7 @@ config CLKSRC_EXYNOS_MCT
 config CLKSRC_SAMSUNG_PWM
 	bool "PWM timer drvier for Samsung S3C, S5P" if COMPILE_TEST
 	depends on GENERIC_CLOCKEVENTS
+	depends on HAS_IOMEM
 	help
 	  This is a new clocksource driver for the PWM timer found in
 	  Samsung S3C, S5P and Exynos SoCs, replacing an earlier driver
@@ -257,12 +265,14 @@ config CLKSRC_SAMSUNG_PWM
 config FSL_FTM_TIMER
 	bool "Freescale FlexTimer Module driver" if COMPILE_TEST
 	depends on GENERIC_CLOCKEVENTS
+	depends on HAS_IOMEM
 	select CLKSRC_MMIO
 	help
 	  Support for Freescale FlexTimer Module (FTM) timer.
 
 config VF_PIT_TIMER
 	bool
+	select CLKSRC_MMIO
 	help
 	  Support for Period Interrupt Timer on Freescale Vybrid Family SoCs.
 
@@ -360,6 +370,7 @@ config CLKSRC_TANGO_XTAL
 config CLKSRC_PXA
 	bool "Clocksource for PXA or SA-11x0 platform" if COMPILE_TEST
 	depends on GENERIC_CLOCKEVENTS
+	depends on HAS_IOMEM
 	select CLKSRC_MMIO
 	help
 	  This enables OST0 support available on PXA and SA-11x0
@@ -394,6 +405,7 @@ config CLKSRC_ST_LPC
 	bool "Low power clocksource found in the LPC" if COMPILE_TEST
 	select CLKSRC_OF if OF
 	depends on HAS_IOMEM
+	select CLKSRC_MMIO
 	help
 	  Enable this option to use the Low Power controller timer
 	  as clocksource.

drivers/clocksource/tcb_clksrc.c

@@ -98,6 +98,7 @@ static int tc_shutdown(struct clock_event_device *d)
 
 	__raw_writel(0xff, regs + ATMEL_TC_REG(2, IDR));
 	__raw_writel(ATMEL_TC_CLKDIS, regs + ATMEL_TC_REG(2, CCR));
+	if (!clockevent_state_detached(d))
 		clk_disable(tcd->clk);
 
 	return 0;

fs/timerfd.c

@@ -153,7 +153,7 @@ static ktime_t timerfd_get_remaining(struct timerfd_ctx *ctx)
 	if (isalarm(ctx))
 		remaining = alarm_expires_remaining(&ctx->t.alarm);
 	else
-		remaining = hrtimer_expires_remaining(&ctx->t.tmr);
+		remaining = hrtimer_expires_remaining_adjusted(&ctx->t.tmr);
 
 	return remaining.tv64 < 0 ? ktime_set(0, 0): remaining;
 }

include/linux/hrtimer.h

@@ -87,6 +87,7 @@ enum hrtimer_restart {
  * @function:	timer expiry callback function
  * @base:	pointer to the timer base (per cpu and per clock)
  * @state:	state information (See bit values above)
+ * @is_rel:	Set if the timer was armed relative
  * @start_pid:  timer statistics field to store the pid of the task which
  *		started the timer
  * @start_site:	timer statistics field to store the site where the timer
@@ -101,7 +102,8 @@ struct hrtimer {
 	ktime_t				_softexpires;
 	enum hrtimer_restart		(*function)(struct hrtimer *);
 	struct hrtimer_clock_base	*base;
-	unsigned long			state;
+	u8				state;
+	u8				is_rel;
 #ifdef CONFIG_TIMER_STATS
 	int				start_pid;
 	void				*start_site;
@@ -321,6 +323,27 @@ static inline void clock_was_set_delayed(void) { }
 
 #endif
 
+static inline ktime_t
+__hrtimer_expires_remaining_adjusted(const struct hrtimer *timer, ktime_t now)
+{
+	ktime_t rem = ktime_sub(timer->node.expires, now);
+
+	/*
+	 * Adjust relative timers for the extra we added in
+	 * hrtimer_start_range_ns() to prevent short timeouts.
+	 */
+	if (IS_ENABLED(CONFIG_TIME_LOW_RES) && timer->is_rel)
+		rem.tv64 -= hrtimer_resolution;
+	return rem;
+}
+
+static inline ktime_t
+hrtimer_expires_remaining_adjusted(const struct hrtimer *timer)
+{
+	return __hrtimer_expires_remaining_adjusted(timer,
+						    timer->base->get_time());
+}
+
 extern void clock_was_set(void);
 #ifdef CONFIG_TIMERFD
 extern void timerfd_clock_was_set(void);
@@ -390,7 +413,12 @@ static inline void hrtimer_restart(struct hrtimer *timer)
 }
 
 /* Query timers: */
-extern ktime_t hrtimer_get_remaining(const struct hrtimer *timer);
+extern ktime_t __hrtimer_get_remaining(const struct hrtimer *timer, bool adjust);
+
+static inline ktime_t hrtimer_get_remaining(const struct hrtimer *timer)
+{
+	return __hrtimer_get_remaining(timer, false);
+}
 
 extern u64 hrtimer_get_next_event(void);
 

kernel/time/hrtimer.c

@@ -897,10 +897,10 @@ static int enqueue_hrtimer(struct hrtimer *timer,
  */
 static void __remove_hrtimer(struct hrtimer *timer,
 			     struct hrtimer_clock_base *base,
-			     unsigned long newstate, int reprogram)
+			     u8 newstate, int reprogram)
 {
 	struct hrtimer_cpu_base *cpu_base = base->cpu_base;
-	unsigned int state = timer->state;
+	u8 state = timer->state;
 
 	timer->state = newstate;
 	if (!(state & HRTIMER_STATE_ENQUEUED))
@@ -930,7 +930,7 @@ static inline int
 remove_hrtimer(struct hrtimer *timer, struct hrtimer_clock_base *base, bool restart)
 {
 	if (hrtimer_is_queued(timer)) {
-		unsigned long state = timer->state;
+		u8 state = timer->state;
 		int reprogram;
 
 		/*
@@ -954,6 +954,22 @@ remove_hrtimer(struct hrtimer *timer, struct hrtimer_clock_base *base, bool rest
 	return 0;
 }
 
+static inline ktime_t hrtimer_update_lowres(struct hrtimer *timer, ktime_t tim,
+					    const enum hrtimer_mode mode)
+{
+#ifdef CONFIG_TIME_LOW_RES
+	/*
+	 * CONFIG_TIME_LOW_RES indicates that the system has no way to return
+	 * granular time values. For relative timers we add hrtimer_resolution
+	 * (i.e. one jiffie) to prevent short timeouts.
+	 */
+	timer->is_rel = mode & HRTIMER_MODE_REL;
+	if (timer->is_rel)
+		tim = ktime_add_safe(tim, ktime_set(0, hrtimer_resolution));
+#endif
+	return tim;
+}
+
 /**
  * hrtimer_start_range_ns - (re)start an hrtimer on the current CPU
  * @timer:	the timer to be added
@@ -974,19 +990,10 @@ void hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim,
 	/* Remove an active timer from the queue: */
 	remove_hrtimer(timer, base, true);
 
-	if (mode & HRTIMER_MODE_REL) {
+	if (mode & HRTIMER_MODE_REL)
 		tim = ktime_add_safe(tim, base->get_time());
-		/*
-		 * CONFIG_TIME_LOW_RES is a temporary way for architectures
-		 * to signal that they simply return xtime in
-		 * do_gettimeoffset(). In this case we want to round up by
-		 * resolution when starting a relative timer, to avoid short
-		 * timeouts. This will go away with the GTOD framework.
-		 */
-#ifdef CONFIG_TIME_LOW_RES
-		tim = ktime_add_safe(tim, ktime_set(0, hrtimer_resolution));
-#endif
-	}
+
+	tim = hrtimer_update_lowres(timer, tim, mode);
 
 	hrtimer_set_expires_range_ns(timer, tim, delta_ns);
 
@@ -1074,19 +1081,23 @@ EXPORT_SYMBOL_GPL(hrtimer_cancel);
 /**
  * hrtimer_get_remaining - get remaining time for the timer
  * @timer:	the timer to read
+ * @adjust:	adjust relative timers when CONFIG_TIME_LOW_RES=y
  */
-ktime_t hrtimer_get_remaining(const struct hrtimer *timer)
+ktime_t __hrtimer_get_remaining(const struct hrtimer *timer, bool adjust)
 {
 	unsigned long flags;
 	ktime_t rem;
 
 	lock_hrtimer_base(timer, &flags);
+	if (IS_ENABLED(CONFIG_TIME_LOW_RES) && adjust)
+		rem = hrtimer_expires_remaining_adjusted(timer);
+	else
 		rem = hrtimer_expires_remaining(timer);
 	unlock_hrtimer_base(timer, &flags);
 
 	return rem;
 }
-EXPORT_SYMBOL_GPL(hrtimer_get_remaining);
+EXPORT_SYMBOL_GPL(__hrtimer_get_remaining);
 
 #ifdef CONFIG_NO_HZ_COMMON
 /**
@@ -1219,6 +1230,14 @@ static void __run_hrtimer(struct hrtimer_cpu_base *cpu_base,
 	timer_stats_account_hrtimer(timer);
 	fn = timer->function;
 
+	/*
+	 * Clear the 'is relative' flag for the TIME_LOW_RES case. If the
+	 * timer is restarted with a period then it becomes an absolute
+	 * timer. If its not restarted it does not matter.
+	 */
+	if (IS_ENABLED(CONFIG_TIME_LOW_RES))
+		timer->is_rel = false;
+
 	/*
 	 * Because we run timers from hardirq context, there is no chance
 	 * they get migrated to another cpu, therefore its safe to unlock

kernel/time/itimer.c

@@ -26,7 +26,7 @@
  */
 static struct timeval itimer_get_remtime(struct hrtimer *timer)
 {
-	ktime_t rem = hrtimer_get_remaining(timer);
+	ktime_t rem = __hrtimer_get_remaining(timer, true);
 
 	/*
 	 * Racy but safe: if the itimer expires after the above

kernel/time/ntp.c

@@ -685,9 +685,19 @@ int ntp_validate_timex(struct timex *txc)
 		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

kernel/time/posix-timers.c

@@ -760,7 +760,7 @@ common_timer_get(struct k_itimer *timr, struct itimerspec *cur_setting)
 	    (timr->it_sigev_notify & ~SIGEV_THREAD_ID) == SIGEV_NONE))
 		timr->it_overrun += (unsigned int) hrtimer_forward(timer, now, iv);
 
-	remaining = ktime_sub(hrtimer_get_expires(timer), now);
+	remaining = __hrtimer_expires_remaining_adjusted(timer, now);
 	/* Return 0 only, when the timer is expired and not pending */
 	if (remaining.tv64 <= 0) {
 		/*

kernel/time/tick-sched.c

@@ -36,16 +36,17 @@
  */
 static DEFINE_PER_CPU(struct tick_sched, tick_cpu_sched);
 
-/*
- * The time, when the last jiffy update happened. Protected by jiffies_lock.
- */
-static ktime_t last_jiffies_update;
-
 struct tick_sched *tick_get_tick_sched(int cpu)
 {
 	return &per_cpu(tick_cpu_sched, cpu);
 }
 
+#if defined(CONFIG_NO_HZ_COMMON) || defined(CONFIG_HIGH_RES_TIMERS)
+/*
+ * The time, when the last jiffy update happened. Protected by jiffies_lock.
+ */
+static ktime_t last_jiffies_update;
+
 /*
  * Must be called with interrupts disabled !
  */
@@ -151,6 +152,7 @@ static void tick_sched_handle(struct tick_sched *ts, struct pt_regs *regs)
 	update_process_times(user_mode(regs));
 	profile_tick(CPU_PROFILING);
 }
+#endif
 
 #ifdef CONFIG_NO_HZ_FULL
 cpumask_var_t tick_nohz_full_mask;
@@ -993,9 +995,9 @@ static void tick_nohz_switch_to_nohz(void)
 	/* Get the next period */
 	next = tick_init_jiffy_update();
 
-	hrtimer_forward_now(&ts->sched_timer, tick_period);
 	hrtimer_set_expires(&ts->sched_timer, next);
-	tick_program_event(next, 1);
+	hrtimer_forward_now(&ts->sched_timer, tick_period);
+	tick_program_event(hrtimer_get_expires(&ts->sched_timer), 1);
 	tick_nohz_activate(ts, NOHZ_MODE_LOWRES);
 }
 

kernel/time/timer_list.c

@@ -69,7 +69,7 @@ print_timer(struct seq_file *m, struct hrtimer *taddr, struct hrtimer *timer,
 	print_name_offset(m, taddr);
 	SEQ_printf(m, ", ");
 	print_name_offset(m, timer->function);
-	SEQ_printf(m, ", S:%02lx", timer->state);
+	SEQ_printf(m, ", S:%02x", timer->state);
 #ifdef CONFIG_TIMER_STATS
 	SEQ_printf(m, ", ");
 	print_name_offset(m, timer->start_site);

tools/testing/selftests/timers/valid-adjtimex.c

@@ -45,7 +45,17 @@ static inline int ksft_exit_fail(void)
 }
 #endif
 
-#define NSEC_PER_SEC 1000000000L
+#define NSEC_PER_SEC 1000000000LL
+#define USEC_PER_SEC 1000000LL
+
+#define ADJ_SETOFFSET 0x0100
+
+#include <sys/syscall.h>
+static int clock_adjtime(clockid_t id, struct timex *tx)
+{
+	return syscall(__NR_clock_adjtime, id, tx);
+}
+
 
 /* clear NTP time_status & time_state */
 int clear_time_state(void)
@@ -193,10 +203,137 @@ int validate_freq(void)
 }
 
 
+int set_offset(long long offset, int use_nano)
+{
+	struct timex tmx = {};
+	int ret;
+
+	tmx.modes = ADJ_SETOFFSET;
+	if (use_nano) {
+		tmx.modes |= ADJ_NANO;
+
+		tmx.time.tv_sec = offset / NSEC_PER_SEC;
+		tmx.time.tv_usec = offset % NSEC_PER_SEC;
+
+		if (offset < 0 && tmx.time.tv_usec) {
+			tmx.time.tv_sec -= 1;
+			tmx.time.tv_usec += NSEC_PER_SEC;
+		}
+	} else {
+		tmx.time.tv_sec = offset / USEC_PER_SEC;
+		tmx.time.tv_usec = offset % USEC_PER_SEC;
+
+		if (offset < 0 && tmx.time.tv_usec) {
+			tmx.time.tv_sec -= 1;
+			tmx.time.tv_usec += USEC_PER_SEC;
+		}
+	}
+
+	ret = clock_adjtime(CLOCK_REALTIME, &tmx);
+	if (ret < 0) {
+		printf("(sec: %ld  usec: %ld) ", tmx.time.tv_sec, tmx.time.tv_usec);
+		printf("[FAIL]\n");
+		return -1;
+	}
+	return 0;
+}
+
+int set_bad_offset(long sec, long usec, int use_nano)
+{
+	struct timex tmx = {};
+	int ret;
+
+	tmx.modes = ADJ_SETOFFSET;
+	if (use_nano)
+		tmx.modes |= ADJ_NANO;
+
+	tmx.time.tv_sec = sec;
+	tmx.time.tv_usec = usec;
+	ret = clock_adjtime(CLOCK_REALTIME, &tmx);
+	if (ret >= 0) {
+		printf("Invalid (sec: %ld  usec: %ld) did not fail! ", tmx.time.tv_sec, tmx.time.tv_usec);
+		printf("[FAIL]\n");
+		return -1;
+	}
+	return 0;
+}
+
+int validate_set_offset(void)
+{
+	printf("Testing ADJ_SETOFFSET... ");
+
+	/* Test valid values */
+	if (set_offset(NSEC_PER_SEC - 1, 1))
+		return -1;
+
+	if (set_offset(-NSEC_PER_SEC + 1, 1))
+		return -1;
+
+	if (set_offset(-NSEC_PER_SEC - 1, 1))
+		return -1;
+
+	if (set_offset(5 * NSEC_PER_SEC, 1))
+		return -1;
+
+	if (set_offset(-5 * NSEC_PER_SEC, 1))
+		return -1;
+
+	if (set_offset(5 * NSEC_PER_SEC + NSEC_PER_SEC / 2, 1))
+		return -1;
+
+	if (set_offset(-5 * NSEC_PER_SEC - NSEC_PER_SEC / 2, 1))
+		return -1;
+
+	if (set_offset(USEC_PER_SEC - 1, 0))
+		return -1;
+
+	if (set_offset(-USEC_PER_SEC + 1, 0))
+		return -1;
+
+	if (set_offset(-USEC_PER_SEC - 1, 0))
+		return -1;
+
+	if (set_offset(5 * USEC_PER_SEC, 0))
+		return -1;
+
+	if (set_offset(-5 * USEC_PER_SEC, 0))
+		return -1;
+
+	if (set_offset(5 * USEC_PER_SEC + USEC_PER_SEC / 2, 0))
+		return -1;
+
+	if (set_offset(-5 * USEC_PER_SEC - USEC_PER_SEC / 2, 0))
+		return -1;
+
+	/* Test invalid values */
+	if (set_bad_offset(0, -1, 1))
+		return -1;
+	if (set_bad_offset(0, -1, 0))
+		return -1;
+	if (set_bad_offset(0, 2 * NSEC_PER_SEC, 1))
+		return -1;
+	if (set_bad_offset(0, 2 * USEC_PER_SEC, 0))
+		return -1;
+	if (set_bad_offset(0, NSEC_PER_SEC, 1))
+		return -1;
+	if (set_bad_offset(0, USEC_PER_SEC, 0))
+		return -1;
+	if (set_bad_offset(0, -NSEC_PER_SEC, 1))
+		return -1;
+	if (set_bad_offset(0, -USEC_PER_SEC, 0))
+		return -1;
+
+	printf("[OK]\n");
+	return 0;
+}
+
 int main(int argc, char **argv)
 {
 	if (validate_freq())
 		return ksft_exit_fail();
 
+	if (validate_set_offset())
+		return ksft_exit_fail();
+
 	return ksft_exit_pass();
 }