diff --git a/arch/arm/boot/dts/zynq-7000.dtsi b/arch/arm/boot/dts/zynq-7000.dtsi
index e32b92b949d2..e7f73b2e4550 100644
--- a/arch/arm/boot/dts/zynq-7000.dtsi
+++ b/arch/arm/boot/dts/zynq-7000.dtsi
@@ -92,6 +92,14 @@
 			};
 		};
 
+		global_timer: timer@f8f00200 {
+			compatible = "arm,cortex-a9-global-timer";
+			reg = <0xf8f00200 0x20>;
+			interrupts = <1 11 0x301>;
+			interrupt-parent = <&intc>;
+			clocks = <&clkc 4>;
+		};
+
 		ttc0: ttc0@f8001000 {
 			interrupt-parent = <&intc>;
 			interrupts = < 0 10 4 0 11 4 0 12 4 >;
diff --git a/arch/arm/mach-msm/timer.c b/arch/arm/mach-msm/timer.c
index 696fb73296d0..1e9c3383daba 100644
--- a/arch/arm/mach-msm/timer.c
+++ b/arch/arm/mach-msm/timer.c
@@ -274,7 +274,6 @@ static void __init msm_dt_timer_init(struct device_node *np)
 		pr_err("Unknown frequency\n");
 		return;
 	}
-	of_node_put(np);
 
 	event_base = base + 0x4;
 	sts_base = base + 0x88;
diff --git a/arch/arm/mach-zynq/Kconfig b/arch/arm/mach-zynq/Kconfig
index 04f8a4a6e755..6b04260aa142 100644
--- a/arch/arm/mach-zynq/Kconfig
+++ b/arch/arm/mach-zynq/Kconfig
@@ -13,5 +13,6 @@ config ARCH_ZYNQ
 	select HAVE_SMP
 	select SPARSE_IRQ
 	select CADENCE_TTC_TIMER
+	select ARM_GLOBAL_TIMER
 	help
 	  Support for Xilinx Zynq ARM Cortex A9 Platform
diff --git a/drivers/clocksource/arm_arch_timer.c b/drivers/clocksource/arm_arch_timer.c
index fbd9ccd5e114..ce98d5e70927 100644
--- a/drivers/clocksource/arm_arch_timer.c
+++ b/drivers/clocksource/arm_arch_timer.c
@@ -389,7 +389,7 @@ static struct clocksource clocksource_counter = {
 	.rating	= 400,
 	.read	= arch_counter_read,
 	.mask	= CLOCKSOURCE_MASK(56),
-	.flags	= CLOCK_SOURCE_IS_CONTINUOUS,
+	.flags	= CLOCK_SOURCE_IS_CONTINUOUS | CLOCK_SOURCE_SUSPEND_NONSTOP,
 };
 
 static struct cyclecounter cyclecounter = {
diff --git a/drivers/clocksource/arm_global_timer.c b/drivers/clocksource/arm_global_timer.c
index b66c1f36066c..c639b1a9e996 100644
--- a/drivers/clocksource/arm_global_timer.c
+++ b/drivers/clocksource/arm_global_timer.c
@@ -169,7 +169,8 @@ static int gt_clockevents_init(struct clock_event_device *clk)
 	int cpu = smp_processor_id();
 
 	clk->name = "arm_global_timer";
-	clk->features = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT;
+	clk->features = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT |
+		CLOCK_EVT_FEAT_PERCPU;
 	clk->set_mode = gt_clockevent_set_mode;
 	clk->set_next_event = gt_clockevent_set_next_event;
 	clk->cpumask = cpumask_of(cpu);
diff --git a/drivers/clocksource/bcm2835_timer.c b/drivers/clocksource/bcm2835_timer.c
index 07ea7ce900dc..26ed331b1aad 100644
--- a/drivers/clocksource/bcm2835_timer.c
+++ b/drivers/clocksource/bcm2835_timer.c
@@ -49,7 +49,7 @@ struct bcm2835_timer {
 
 static void __iomem *system_clock __read_mostly;
 
-static u32 notrace bcm2835_sched_read(void)
+static u64 notrace bcm2835_sched_read(void)
 {
 	return readl_relaxed(system_clock);
 }
@@ -110,7 +110,7 @@ static void __init bcm2835_timer_init(struct device_node *node)
 		panic("Can't read clock-frequency");
 
 	system_clock = base + REG_COUNTER_LO;
-	setup_sched_clock(bcm2835_sched_read, 32, freq);
+	sched_clock_register(bcm2835_sched_read, 32, freq);
 
 	clocksource_mmio_init(base + REG_COUNTER_LO, node->name,
 		freq, 300, 32, clocksource_mmio_readl_up);
diff --git a/drivers/clocksource/clksrc-dbx500-prcmu.c b/drivers/clocksource/clksrc-dbx500-prcmu.c
index a9fd4ad25674..b375106844d8 100644
--- a/drivers/clocksource/clksrc-dbx500-prcmu.c
+++ b/drivers/clocksource/clksrc-dbx500-prcmu.c
@@ -53,7 +53,7 @@ static struct clocksource clocksource_dbx500_prcmu = {
 
 #ifdef CONFIG_CLKSRC_DBX500_PRCMU_SCHED_CLOCK
 
-static u32 notrace dbx500_prcmu_sched_clock_read(void)
+static u64 notrace dbx500_prcmu_sched_clock_read(void)
 {
 	if (unlikely(!clksrc_dbx500_timer_base))
 		return 0;
@@ -81,8 +81,7 @@ void __init clksrc_dbx500_prcmu_init(void __iomem *base)
 		       clksrc_dbx500_timer_base + PRCMU_TIMER_REF);
 	}
 #ifdef CONFIG_CLKSRC_DBX500_PRCMU_SCHED_CLOCK
-	setup_sched_clock(dbx500_prcmu_sched_clock_read,
-			 32, RATE_32K);
+	sched_clock_register(dbx500_prcmu_sched_clock_read, 32, RATE_32K);
 #endif
 	clocksource_register_hz(&clocksource_dbx500_prcmu, RATE_32K);
 }
diff --git a/drivers/clocksource/clksrc-of.c b/drivers/clocksource/clksrc-of.c
index b9ddd9e3a2f5..35639cf4e5a2 100644
--- a/drivers/clocksource/clksrc-of.c
+++ b/drivers/clocksource/clksrc-of.c
@@ -35,5 +35,6 @@ void __init clocksource_of_init(void)
 
 		init_func = match->data;
 		init_func(np);
+		of_node_put(np);
 	}
 }
diff --git a/drivers/clocksource/dw_apb_timer_of.c b/drivers/clocksource/dw_apb_timer_of.c
index 4cbae4f762b1..45ba8aecc729 100644
--- a/drivers/clocksource/dw_apb_timer_of.c
+++ b/drivers/clocksource/dw_apb_timer_of.c
@@ -23,7 +23,7 @@
 #include <linux/clk.h>
 #include <linux/sched_clock.h>
 
-static void timer_get_base_and_rate(struct device_node *np,
+static void __init timer_get_base_and_rate(struct device_node *np,
 				    void __iomem **base, u32 *rate)
 {
 	struct clk *timer_clk;
@@ -55,11 +55,11 @@ static void timer_get_base_and_rate(struct device_node *np,
 
 try_clock_freq:
 	if (of_property_read_u32(np, "clock-freq", rate) &&
-		of_property_read_u32(np, "clock-frequency", rate))
+	    of_property_read_u32(np, "clock-frequency", rate))
 		panic("No clock nor clock-frequency property for %s", np->name);
 }
 
-static void add_clockevent(struct device_node *event_timer)
+static void __init add_clockevent(struct device_node *event_timer)
 {
 	void __iomem *iobase;
 	struct dw_apb_clock_event_device *ced;
@@ -82,7 +82,7 @@ static void add_clockevent(struct device_node *event_timer)
 static void __iomem *sched_io_base;
 static u32 sched_rate;
 
-static void add_clocksource(struct device_node *source_timer)
+static void __init add_clocksource(struct device_node *source_timer)
 {
 	void __iomem *iobase;
 	struct dw_apb_clocksource *cs;
@@ -106,7 +106,7 @@ static void add_clocksource(struct device_node *source_timer)
 	sched_rate = rate;
 }
 
-static u32 read_sched_clock(void)
+static u64 read_sched_clock(void)
 {
 	return __raw_readl(sched_io_base);
 }
@@ -117,7 +117,7 @@ static const struct of_device_id sptimer_ids[] __initconst = {
 	{ /* Sentinel */ },
 };
 
-static void init_sched_clock(void)
+static void __init init_sched_clock(void)
 {
 	struct device_node *sched_timer;
 
@@ -128,7 +128,7 @@ static void init_sched_clock(void)
 		of_node_put(sched_timer);
 	}
 
-	setup_sched_clock(read_sched_clock, 32, sched_rate);
+	sched_clock_register(read_sched_clock, 32, sched_rate);
 }
 
 static int num_called;
@@ -138,12 +138,10 @@ static void __init dw_apb_timer_init(struct device_node *timer)
 	case 0:
 		pr_debug("%s: found clockevent timer\n", __func__);
 		add_clockevent(timer);
-		of_node_put(timer);
 		break;
 	case 1:
 		pr_debug("%s: found clocksource timer\n", __func__);
 		add_clocksource(timer);
-		of_node_put(timer);
 		init_sched_clock();
 		break;
 	default:
diff --git a/drivers/clocksource/mxs_timer.c b/drivers/clocksource/mxs_timer.c
index 0f5e65f74dc3..445b68a01dc5 100644
--- a/drivers/clocksource/mxs_timer.c
+++ b/drivers/clocksource/mxs_timer.c
@@ -222,7 +222,7 @@ static struct clocksource clocksource_mxs = {
 	.flags		= CLOCK_SOURCE_IS_CONTINUOUS,
 };
 
-static u32 notrace mxs_read_sched_clock_v2(void)
+static u64 notrace mxs_read_sched_clock_v2(void)
 {
 	return ~readl_relaxed(mxs_timrot_base + HW_TIMROT_RUNNING_COUNTn(1));
 }
@@ -236,7 +236,7 @@ static int __init mxs_clocksource_init(struct clk *timer_clk)
 	else {
 		clocksource_mmio_init(mxs_timrot_base + HW_TIMROT_RUNNING_COUNTn(1),
 			"mxs_timer", c, 200, 32, clocksource_mmio_readl_down);
-		setup_sched_clock(mxs_read_sched_clock_v2, 32, c);
+		sched_clock_register(mxs_read_sched_clock_v2, 32, c);
 	}
 
 	return 0;
diff --git a/drivers/clocksource/nomadik-mtu.c b/drivers/clocksource/nomadik-mtu.c
index 1b74bea12385..ed7b73b508e0 100644
--- a/drivers/clocksource/nomadik-mtu.c
+++ b/drivers/clocksource/nomadik-mtu.c
@@ -76,7 +76,7 @@ static struct delay_timer mtu_delay_timer;
  * local implementation which uses the clocksource to get some
  * better resolution when scheduling the kernel.
  */
-static u32 notrace nomadik_read_sched_clock(void)
+static u64 notrace nomadik_read_sched_clock(void)
 {
 	if (unlikely(!mtu_base))
 		return 0;
@@ -231,7 +231,7 @@ static void __init __nmdk_timer_init(void __iomem *base, int irq,
 		       "mtu_0");
 
 #ifdef CONFIG_CLKSRC_NOMADIK_MTU_SCHED_CLOCK
-	setup_sched_clock(nomadik_read_sched_clock, 32, rate);
+	sched_clock_register(nomadik_read_sched_clock, 32, rate);
 #endif
 
 	/* Timer 1 is used for events, register irq and clockevents */
diff --git a/drivers/clocksource/samsung_pwm_timer.c b/drivers/clocksource/samsung_pwm_timer.c
index ab29476ee5f9..85082e8d3052 100644
--- a/drivers/clocksource/samsung_pwm_timer.c
+++ b/drivers/clocksource/samsung_pwm_timer.c
@@ -331,7 +331,7 @@ static struct clocksource samsung_clocksource = {
  * this wraps around for now, since it is just a relative time
  * stamp. (Inspired by U300 implementation.)
  */
-static u32 notrace samsung_read_sched_clock(void)
+static u64 notrace samsung_read_sched_clock(void)
 {
 	return samsung_clocksource_read(NULL);
 }
@@ -357,7 +357,7 @@ static void __init samsung_clocksource_init(void)
 	else
 		pwm.source_reg = pwm.base + pwm.source_id * 0x0c + 0x14;
 
-	setup_sched_clock(samsung_read_sched_clock,
+	sched_clock_register(samsung_read_sched_clock,
 						pwm.variant.bits, clock_rate);
 
 	samsung_clocksource.mask = CLOCKSOURCE_MASK(pwm.variant.bits);
diff --git a/drivers/clocksource/tegra20_timer.c b/drivers/clocksource/tegra20_timer.c
index 93961703b887..642849256d82 100644
--- a/drivers/clocksource/tegra20_timer.c
+++ b/drivers/clocksource/tegra20_timer.c
@@ -98,7 +98,7 @@ static struct clock_event_device tegra_clockevent = {
 	.set_mode	= tegra_timer_set_mode,
 };
 
-static u32 notrace tegra_read_sched_clock(void)
+static u64 notrace tegra_read_sched_clock(void)
 {
 	return timer_readl(TIMERUS_CNTR_1US);
 }
@@ -181,8 +181,6 @@ static void __init tegra20_init_timer(struct device_node *np)
 		rate = clk_get_rate(clk);
 	}
 
-	of_node_put(np);
-
 	switch (rate) {
 	case 12000000:
 		timer_writel(0x000b, TIMERUS_USEC_CFG);
@@ -200,7 +198,7 @@ static void __init tegra20_init_timer(struct device_node *np)
 		WARN(1, "Unknown clock rate");
 	}
 
-	setup_sched_clock(tegra_read_sched_clock, 32, 1000000);
+	sched_clock_register(tegra_read_sched_clock, 32, 1000000);
 
 	if (clocksource_mmio_init(timer_reg_base + TIMERUS_CNTR_1US,
 		"timer_us", 1000000, 300, 32, clocksource_mmio_readl_up)) {
@@ -241,8 +239,6 @@ static void __init tegra20_init_rtc(struct device_node *np)
 	else
 		clk_prepare_enable(clk);
 
-	of_node_put(np);
-
 	register_persistent_clock(NULL, tegra_read_persistent_clock);
 }
 CLOCKSOURCE_OF_DECLARE(tegra20_rtc, "nvidia,tegra20-rtc", tegra20_init_rtc);
diff --git a/drivers/clocksource/time-armada-370-xp.c b/drivers/clocksource/time-armada-370-xp.c
index 0198504ef6b0..d8e47e502785 100644
--- a/drivers/clocksource/time-armada-370-xp.c
+++ b/drivers/clocksource/time-armada-370-xp.c
@@ -96,7 +96,7 @@ static void local_timer_ctrl_clrset(u32 clr, u32 set)
 		local_base + TIMER_CTRL_OFF);
 }
 
-static u32 notrace armada_370_xp_read_sched_clock(void)
+static u64 notrace armada_370_xp_read_sched_clock(void)
 {
 	return ~readl(timer_base + TIMER0_VAL_OFF);
 }
@@ -258,7 +258,7 @@ static void __init armada_370_xp_timer_common_init(struct device_node *np)
 	/*
 	 * Set scale and timer for sched_clock.
 	 */
-	setup_sched_clock(armada_370_xp_read_sched_clock, 32, timer_clk);
+	sched_clock_register(armada_370_xp_read_sched_clock, 32, timer_clk);
 
 	/*
 	 * Setup free-running clocksource timer (interrupts
diff --git a/drivers/clocksource/timer-prima2.c b/drivers/clocksource/timer-prima2.c
index ef3cfb269d8b..8a492d34ff9f 100644
--- a/drivers/clocksource/timer-prima2.c
+++ b/drivers/clocksource/timer-prima2.c
@@ -165,9 +165,9 @@ static struct irqaction sirfsoc_timer_irq = {
 };
 
 /* Overwrite weak default sched_clock with more precise one */
-static u32 notrace sirfsoc_read_sched_clock(void)
+static u64 notrace sirfsoc_read_sched_clock(void)
 {
-	return (u32)(sirfsoc_timer_read(NULL) & 0xffffffff);
+	return sirfsoc_timer_read(NULL);
 }
 
 static void __init sirfsoc_clockevent_init(void)
@@ -206,7 +206,7 @@ static void __init sirfsoc_prima2_timer_init(struct device_node *np)
 
 	BUG_ON(clocksource_register_hz(&sirfsoc_clocksource, CLOCK_TICK_RATE));
 
-	setup_sched_clock(sirfsoc_read_sched_clock, 32, CLOCK_TICK_RATE);
+	sched_clock_register(sirfsoc_read_sched_clock, 64, CLOCK_TICK_RATE);
 
 	BUG_ON(setup_irq(sirfsoc_timer_irq.irq, &sirfsoc_timer_irq));
 
diff --git a/drivers/clocksource/vf_pit_timer.c b/drivers/clocksource/vf_pit_timer.c
index 587e0202a70b..02821b06a39e 100644
--- a/drivers/clocksource/vf_pit_timer.c
+++ b/drivers/clocksource/vf_pit_timer.c
@@ -52,7 +52,7 @@ static inline void pit_irq_acknowledge(void)
 	__raw_writel(PITTFLG_TIF, clkevt_base + PITTFLG);
 }
 
-static unsigned int pit_read_sched_clock(void)
+static u64 pit_read_sched_clock(void)
 {
 	return __raw_readl(clksrc_base + PITCVAL);
 }
@@ -64,7 +64,7 @@ static int __init pit_clocksource_init(unsigned long rate)
 	__raw_writel(~0UL, clksrc_base + PITLDVAL);
 	__raw_writel(PITTCTRL_TEN, clksrc_base + PITTCTRL);
 
-	setup_sched_clock(pit_read_sched_clock, 32, rate);
+	sched_clock_register(pit_read_sched_clock, 32, rate);
 	return clocksource_mmio_init(clksrc_base + PITCVAL, "vf-pit", rate,
 			300, 32, clocksource_mmio_readl_down);
 }
diff --git a/drivers/clocksource/vt8500_timer.c b/drivers/clocksource/vt8500_timer.c
index 64f553f04fa4..ad3c0e83a779 100644
--- a/drivers/clocksource/vt8500_timer.c
+++ b/drivers/clocksource/vt8500_timer.c
@@ -137,14 +137,12 @@ static void __init vt8500_timer_init(struct device_node *np)
 	if (!regbase) {
 		pr_err("%s: Missing iobase description in Device Tree\n",
 								__func__);
-		of_node_put(np);
 		return;
 	}
 	timer_irq = irq_of_parse_and_map(np, 0);
 	if (!timer_irq) {
 		pr_err("%s: Missing irq description in Device Tree\n",
 								__func__);
-		of_node_put(np);
 		return;
 	}
 
diff --git a/include/linux/clockchips.h b/include/linux/clockchips.h
index 0857922e8ad0..493aa021c7a9 100644
--- a/include/linux/clockchips.h
+++ b/include/linux/clockchips.h
@@ -60,6 +60,7 @@ enum clock_event_mode {
  * Core shall set the interrupt affinity dynamically in broadcast mode
  */
 #define CLOCK_EVT_FEAT_DYNIRQ		0x000020
+#define CLOCK_EVT_FEAT_PERCPU		0x000040
 
 /**
  * struct clock_event_device - clock event device descriptor
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index 218bcb565fed..9532690daaa9 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -70,6 +70,7 @@ static bool tick_check_broadcast_device(struct clock_event_device *curdev,
 					struct clock_event_device *newdev)
 {
 	if ((newdev->features & CLOCK_EVT_FEAT_DUMMY) ||
+	    (newdev->features & CLOCK_EVT_FEAT_PERCPU) ||
 	    (newdev->features & CLOCK_EVT_FEAT_C3STOP))
 		return false;