Merge branch 'next' (accumulated 3.16 merge window patches) into master

Now that 3.15 is released, this merges the 'next' branch into 'master',
bringing us to the normal situation where my 'master' branch is the
merge window.

* accumulated work in next: (6809 commits)
  ufs: sb mutex merge + mutex_destroy
  powerpc: update comments for generic idle conversion
  cris: update comments for generic idle conversion
  idle: remove cpu_idle() forward declarations
  nbd: zero from and len fields in NBD_CMD_DISCONNECT.
  mm: convert some level-less printks to pr_*
  MAINTAINERS: adi-buildroot-devel is moderated
  MAINTAINERS: add linux-api for review of API/ABI changes
  mm/kmemleak-test.c: use pr_fmt for logging
  fs/dlm/debug_fs.c: replace seq_printf by seq_puts
  fs/dlm/lockspace.c: convert simple_str to kstr
  fs/dlm/config.c: convert simple_str to kstr
  mm: mark remap_file_pages() syscall as deprecated
  mm: memcontrol: remove unnecessary memcg argument from soft limit functions
  mm: memcontrol: clean up memcg zoneinfo lookup
  mm/memblock.c: call kmemleak directly from memblock_(alloc|free)
  mm/mempool.c: update the kmemleak stack trace for mempool allocations
  lib/radix-tree.c: update the kmemleak stack trace for radix tree allocations
  mm: introduce kmemleak_update_trace()
  mm/kmemleak.c: use %u to print ->checksum
  ...

19 files changed, 1828 insertions, 748 deletions

diff --git a/drivers/clocksource/Kconfig b/drivers/clocksource/Kconfig
index 96918e1f26a3..065131cbfcc0 100644
--- a/drivers/clocksource/Kconfig
+++ b/drivers/clocksource/Kconfig
@@ -136,6 +136,11 @@ config CLKSRC_SAMSUNG_PWM
 	  for all devicetree enabled platforms. This driver will be
 	  needed only on systems that do not have the Exynos MCT available.
 
+config FSL_FTM_TIMER
+	bool
+	help
+	  Support for Freescale FlexTimer Module (FTM) timer.
+
 config VF_PIT_TIMER
 	bool
 	help
@@ -191,3 +196,14 @@ config EM_TIMER_STI
 
 config CLKSRC_QCOM
 	bool
+
+config CLKSRC_VERSATILE
+	bool "ARM Versatile (Express) reference platforms clock source"
+	depends on GENERIC_SCHED_CLOCK && !ARCH_USES_GETTIMEOFFSET
+	select CLKSRC_OF
+	default y if MFD_VEXPRESS_SYSREG
+	help
+	  This option enables clock source based on free running
+	  counter available in the "System Registers" block of
+	  ARM Versatile, RealView and Versatile Express reference
+	  platforms.
diff --git a/drivers/clocksource/Makefile b/drivers/clocksource/Makefile
index 98cb6c51aa87..800b1303c236 100644
--- a/drivers/clocksource/Makefile
+++ b/drivers/clocksource/Makefile
@@ -31,6 +31,7 @@ obj-$(CONFIG_CADENCE_TTC_TIMER)	+= cadence_ttc_timer.o
 obj-$(CONFIG_CLKSRC_EFM32)	+= time-efm32.o
 obj-$(CONFIG_CLKSRC_EXYNOS_MCT)	+= exynos_mct.o
 obj-$(CONFIG_CLKSRC_SAMSUNG_PWM)	+= samsung_pwm_timer.o
+obj-$(CONFIG_FSL_FTM_TIMER)	+= fsl_ftm_timer.o
 obj-$(CONFIG_VF_PIT_TIMER)	+= vf_pit_timer.o
 obj-$(CONFIG_CLKSRC_QCOM)	+= qcom-timer.o
 
@@ -39,3 +40,4 @@ obj-$(CONFIG_ARM_GLOBAL_TIMER)		+= arm_global_timer.o
 obj-$(CONFIG_CLKSRC_METAG_GENERIC)	+= metag_generic.o
 obj-$(CONFIG_ARCH_HAS_TICK_BROADCAST)	+= dummy_timer.o
 obj-$(CONFIG_ARCH_KEYSTONE)		+= timer-keystone.o
+obj-$(CONFIG_CLKSRC_VERSATILE)		+= versatile.o
diff --git a/drivers/clocksource/arm_global_timer.c b/drivers/clocksource/arm_global_timer.c
index 0fc31d029e52..60e5a170c4d2 100644
--- a/drivers/clocksource/arm_global_timer.c
+++ b/drivers/clocksource/arm_global_timer.c
@@ -246,11 +246,12 @@ static void __init global_timer_of_register(struct device_node *np)
 	int err = 0;
 
 	/*
-	 * In r2p0 the comparators for each processor with the global timer
+	 * In A9 r2p0 the comparators for each processor with the global timer
 	 * fire when the timer value is greater than or equal to. In previous
 	 * revisions the comparators fired when the timer value was equal to.
 	 */
-	if ((read_cpuid_id() & 0xf0000f) < 0x200000) {
+	if (read_cpuid_part_number() == ARM_CPU_PART_CORTEX_A9
+	    && (read_cpuid_id() & 0xf0000f) < 0x200000) {
 		pr_warn("global-timer: non support for this cpu version.\n");
 		return;
 	}
diff --git a/drivers/clocksource/cadence_ttc_timer.c b/drivers/clocksource/cadence_ttc_timer.c
index 49fbe2847c84..7a08811df9aa 100644
--- a/drivers/clocksource/cadence_ttc_timer.c
+++ b/drivers/clocksource/cadence_ttc_timer.c
@@ -118,11 +118,11 @@ static void ttc_set_interval(struct ttc_timer *timer,
 	u32 ctrl_reg;
 
 	/* Disable the counter, set the counter value  and re-enable counter */
-	ctrl_reg = __raw_readl(timer->base_addr + TTC_CNT_CNTRL_OFFSET);
+	ctrl_reg = readl_relaxed(timer->base_addr + TTC_CNT_CNTRL_OFFSET);
 	ctrl_reg |= TTC_CNT_CNTRL_DISABLE_MASK;
-	__raw_writel(ctrl_reg, timer->base_addr + TTC_CNT_CNTRL_OFFSET);
+	writel_relaxed(ctrl_reg, timer->base_addr + TTC_CNT_CNTRL_OFFSET);
 
-	__raw_writel(cycles, timer->base_addr + TTC_INTR_VAL_OFFSET);
+	writel_relaxed(cycles, timer->base_addr + TTC_INTR_VAL_OFFSET);
 
 	/*
 	 * Reset the counter (0x10) so that it starts from 0, one-shot
@@ -130,7 +130,7 @@ static void ttc_set_interval(struct ttc_timer *timer,
 	 */
 	ctrl_reg |= CNT_CNTRL_RESET;
 	ctrl_reg &= ~TTC_CNT_CNTRL_DISABLE_MASK;
-	__raw_writel(ctrl_reg, timer->base_addr + TTC_CNT_CNTRL_OFFSET);
+	writel_relaxed(ctrl_reg, timer->base_addr + TTC_CNT_CNTRL_OFFSET);
 }
 
 /**
@@ -147,7 +147,7 @@ static irqreturn_t ttc_clock_event_interrupt(int irq, void *dev_id)
 	struct ttc_timer *timer = &ttce->ttc;
 
 	/* Acknowledge the interrupt and call event handler */
-	__raw_readl(timer->base_addr + TTC_ISR_OFFSET);
+	readl_relaxed(timer->base_addr + TTC_ISR_OFFSET);
 
 	ttce->ce.event_handler(&ttce->ce);
 
@@ -163,13 +163,13 @@ static cycle_t __ttc_clocksource_read(struct clocksource *cs)
 {
 	struct ttc_timer *timer = &to_ttc_timer_clksrc(cs)->ttc;
 
-	return (cycle_t)__raw_readl(timer->base_addr +
+	return (cycle_t)readl_relaxed(timer->base_addr +
 				TTC_COUNT_VAL_OFFSET);
 }
 
 static u64 notrace ttc_sched_clock_read(void)
 {
-	return __raw_readl(ttc_sched_clock_val_reg);
+	return readl_relaxed(ttc_sched_clock_val_reg);
 }
 
 /**
@@ -211,17 +211,17 @@ static void ttc_set_mode(enum clock_event_mode mode,
 	case CLOCK_EVT_MODE_ONESHOT:
 	case CLOCK_EVT_MODE_UNUSED:
 	case CLOCK_EVT_MODE_SHUTDOWN:
-		ctrl_reg = __raw_readl(timer->base_addr +
+		ctrl_reg = readl_relaxed(timer->base_addr +
 					TTC_CNT_CNTRL_OFFSET);
 		ctrl_reg |= TTC_CNT_CNTRL_DISABLE_MASK;
-		__raw_writel(ctrl_reg,
+		writel_relaxed(ctrl_reg,
 				timer->base_addr + TTC_CNT_CNTRL_OFFSET);
 		break;
 	case CLOCK_EVT_MODE_RESUME:
-		ctrl_reg = __raw_readl(timer->base_addr +
+		ctrl_reg = readl_relaxed(timer->base_addr +
 					TTC_CNT_CNTRL_OFFSET);
 		ctrl_reg &= ~TTC_CNT_CNTRL_DISABLE_MASK;
-		__raw_writel(ctrl_reg,
+		writel_relaxed(ctrl_reg,
 				timer->base_addr + TTC_CNT_CNTRL_OFFSET);
 		break;
 	}
@@ -266,8 +266,8 @@ static int ttc_rate_change_clocksource_cb(struct notifier_block *nb,
 		 * of an abort.
 		 */
 		ttccs->scale_clk_ctrl_reg_old =
-			__raw_readl(ttccs->ttc.base_addr +
-					TTC_CLK_CNTRL_OFFSET);
+			readl_relaxed(ttccs->ttc.base_addr +
+			TTC_CLK_CNTRL_OFFSET);
 
 		psv = (ttccs->scale_clk_ctrl_reg_old &
 				TTC_CLK_CNTRL_PSV_MASK) >>
@@ -291,8 +291,8 @@ static int ttc_rate_change_clocksource_cb(struct notifier_block *nb,
 			return NOTIFY_DONE;
 
 		/* scale up: adjust divider now - before frequency change */
-		__raw_writel(ttccs->scale_clk_ctrl_reg_new,
-				ttccs->ttc.base_addr + TTC_CLK_CNTRL_OFFSET);
+		writel_relaxed(ttccs->scale_clk_ctrl_reg_new,
+			       ttccs->ttc.base_addr + TTC_CLK_CNTRL_OFFSET);
 		break;
 	}
 	case POST_RATE_CHANGE:
@@ -301,8 +301,8 @@ static int ttc_rate_change_clocksource_cb(struct notifier_block *nb,
 			return NOTIFY_OK;
 
 		/* scale down: adjust divider now - after frequency change */
-		__raw_writel(ttccs->scale_clk_ctrl_reg_new,
-				ttccs->ttc.base_addr + TTC_CLK_CNTRL_OFFSET);
+		writel_relaxed(ttccs->scale_clk_ctrl_reg_new,
+			       ttccs->ttc.base_addr + TTC_CLK_CNTRL_OFFSET);
 		break;
 
 	case ABORT_RATE_CHANGE:
@@ -311,8 +311,8 @@ static int ttc_rate_change_clocksource_cb(struct notifier_block *nb,
 			return NOTIFY_OK;
 
 		/* restore original register value */
-		__raw_writel(ttccs->scale_clk_ctrl_reg_old,
-				ttccs->ttc.base_addr + TTC_CLK_CNTRL_OFFSET);
+		writel_relaxed(ttccs->scale_clk_ctrl_reg_old,
+			       ttccs->ttc.base_addr + TTC_CLK_CNTRL_OFFSET);
 		/* fall through */
 	default:
 		return NOTIFY_DONE;
@@ -359,10 +359,10 @@ static void __init ttc_setup_clocksource(struct clk *clk, void __iomem *base)
 	 * with no interrupt and it rolls over at 0xFFFF. Pre-scale
 	 * it by 32 also. Let it start running now.
 	 */
-	__raw_writel(0x0,  ttccs->ttc.base_addr + TTC_IER_OFFSET);
-	__raw_writel(CLK_CNTRL_PRESCALE | CLK_CNTRL_PRESCALE_EN,
+	writel_relaxed(0x0,  ttccs->ttc.base_addr + TTC_IER_OFFSET);
+	writel_relaxed(CLK_CNTRL_PRESCALE | CLK_CNTRL_PRESCALE_EN,
 		     ttccs->ttc.base_addr + TTC_CLK_CNTRL_OFFSET);
-	__raw_writel(CNT_CNTRL_RESET,
+	writel_relaxed(CNT_CNTRL_RESET,
 		     ttccs->ttc.base_addr + TTC_CNT_CNTRL_OFFSET);
 
 	err = clocksource_register_hz(&ttccs->cs, ttccs->ttc.freq / PRESCALE);
@@ -438,10 +438,10 @@ static void __init ttc_setup_clockevent(struct clk *clk,
 	 * is prescaled by 32 using the interval interrupt. Leave it
 	 * disabled for now.
 	 */
-	__raw_writel(0x23, ttcce->ttc.base_addr + TTC_CNT_CNTRL_OFFSET);
-	__raw_writel(CLK_CNTRL_PRESCALE | CLK_CNTRL_PRESCALE_EN,
+	writel_relaxed(0x23, ttcce->ttc.base_addr + TTC_CNT_CNTRL_OFFSET);
+	writel_relaxed(CLK_CNTRL_PRESCALE | CLK_CNTRL_PRESCALE_EN,
 		     ttcce->ttc.base_addr + TTC_CLK_CNTRL_OFFSET);
-	__raw_writel(0x1,  ttcce->ttc.base_addr + TTC_IER_OFFSET);
+	writel_relaxed(0x1,  ttcce->ttc.base_addr + TTC_IER_OFFSET);
 
 	err = request_irq(irq, ttc_clock_event_interrupt,
 			  IRQF_TIMER, ttcce->ce.name, ttcce);
@@ -490,7 +490,7 @@ static void __init ttc_timer_init(struct device_node *timer)
 		BUG();
 	}
 
-	clksel = __raw_readl(timer_baseaddr + TTC_CLK_CNTRL_OFFSET);
+	clksel = readl_relaxed(timer_baseaddr + TTC_CLK_CNTRL_OFFSET);
 	clksel = !!(clksel & TTC_CLK_CNTRL_CSRC_MASK);
 	clk_cs = of_clk_get(timer, clksel);
 	if (IS_ERR(clk_cs)) {
@@ -498,7 +498,7 @@ static void __init ttc_timer_init(struct device_node *timer)
 		BUG();
 	}
 
-	clksel = __raw_readl(timer_baseaddr + 4 + TTC_CLK_CNTRL_OFFSET);
+	clksel = readl_relaxed(timer_baseaddr + 4 + TTC_CLK_CNTRL_OFFSET);
 	clksel = !!(clksel & TTC_CLK_CNTRL_CSRC_MASK);
 	clk_ce = of_clk_get(timer, clksel);
 	if (IS_ERR(clk_ce)) {
diff --git a/drivers/clocksource/clksrc-of.c b/drivers/clocksource/clksrc-of.c
index ae2e4278c42a..0093a8e49e14 100644
--- a/drivers/clocksource/clksrc-of.c
+++ b/drivers/clocksource/clksrc-of.c
@@ -27,7 +27,7 @@ void __init clocksource_of_init(void)
 {
 	struct device_node *np;
 	const struct of_device_id *match;
-	clocksource_of_init_fn init_func;
+	of_init_fn_1 init_func;
 	unsigned clocksources = 0;
 
 	for_each_matching_node_and_match(np, __clksrc_of_table, &match) {
diff --git a/drivers/clocksource/dw_apb_timer_of.c b/drivers/clocksource/dw_apb_timer_of.c
index 2a2ea2717f3a..d305fb089767 100644
--- a/drivers/clocksource/dw_apb_timer_of.c
+++ b/drivers/clocksource/dw_apb_timer_of.c
@@ -106,7 +106,7 @@ static void __init add_clocksource(struct device_node *source_timer)
 	sched_rate = rate;
 }
 
-static u64 read_sched_clock(void)
+static u64 notrace read_sched_clock(void)
 {
 	return ~__raw_readl(sched_io_base);
 }
diff --git a/drivers/clocksource/em_sti.c b/drivers/clocksource/em_sti.c
index 9d170834fcf3..d0a7bd66b8b9 100644
--- a/drivers/clocksource/em_sti.c
+++ b/drivers/clocksource/em_sti.c
@@ -318,10 +318,8 @@ static int em_sti_probe(struct platform_device *pdev)
 	int irq;
 
 	p = devm_kzalloc(&pdev->dev, sizeof(*p), GFP_KERNEL);
-	if (p == NULL) {
-		dev_err(&pdev->dev, "failed to allocate driver data\n");
+	if (p == NULL)
 		return -ENOMEM;
-	}
 
 	p->pdev = pdev;
 	platform_set_drvdata(pdev, p);
diff --git a/drivers/clocksource/exynos_mct.c b/drivers/clocksource/exynos_mct.c
index acf5a329d538..8d6420013a04 100644
--- a/drivers/clocksource/exynos_mct.c
+++ b/drivers/clocksource/exynos_mct.c
@@ -24,6 +24,7 @@
 #include <linux/of_irq.h>
 #include <linux/of_address.h>
 #include <linux/clocksource.h>
+#include <linux/sched_clock.h>
 
 #define EXYNOS4_MCTREG(x)		(x)
 #define EXYNOS4_MCT_G_CNT_L		EXYNOS4_MCTREG(0x100)
@@ -192,12 +193,19 @@ struct clocksource mct_frc = {
 	.resume		= exynos4_frc_resume,
 };
 
+static u64 notrace exynos4_read_sched_clock(void)
+{
+	return exynos4_frc_read(&mct_frc);
+}
+
 static void __init exynos4_clocksource_init(void)
 {
 	exynos4_mct_frc_start(0, 0);
 
 	if (clocksource_register_hz(&mct_frc, clk_rate))
 		panic("%s: can't register clocksource\n", mct_frc.name);
+
+	sched_clock_register(exynos4_read_sched_clock, 64, clk_rate);
 }
 
 static void exynos4_mct_comp0_stop(void)
diff --git a/drivers/clocksource/fsl_ftm_timer.c b/drivers/clocksource/fsl_ftm_timer.c
new file mode 100644
index 000000000000..454227d4f895
--- /dev/null
+++ b/drivers/clocksource/fsl_ftm_timer.c
@@ -0,0 +1,367 @@
+/*
+ * Freescale FlexTimer Module (FTM) timer driver.
+ *
+ * Copyright 2014 Freescale Semiconductor, Inc.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version 2
+ * of the License, or (at your option) any later version.
+ */
+
+#include <linux/clk.h>
+#include <linux/clockchips.h>
+#include <linux/clocksource.h>
+#include <linux/err.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/of_address.h>
+#include <linux/of_irq.h>
+#include <linux/sched_clock.h>
+#include <linux/slab.h>
+
+#define FTM_SC		0x00
+#define FTM_SC_CLK_SHIFT	3
+#define FTM_SC_CLK_MASK	(0x3 << FTM_SC_CLK_SHIFT)
+#define FTM_SC_CLK(c)	((c) << FTM_SC_CLK_SHIFT)
+#define FTM_SC_PS_MASK	0x7
+#define FTM_SC_TOIE	BIT(6)
+#define FTM_SC_TOF	BIT(7)
+
+#define FTM_CNT		0x04
+#define FTM_MOD		0x08
+#define FTM_CNTIN	0x4C
+
+#define FTM_PS_MAX	7
+
+struct ftm_clock_device {
+	void __iomem *clksrc_base;
+	void __iomem *clkevt_base;
+	unsigned long periodic_cyc;
+	unsigned long ps;
+	bool big_endian;
+};
+
+static struct ftm_clock_device *priv;
+
+static inline u32 ftm_readl(void __iomem *addr)
+{
+	if (priv->big_endian)
+		return ioread32be(addr);
+	else
+		return ioread32(addr);
+}
+
+static inline void ftm_writel(u32 val, void __iomem *addr)
+{
+	if (priv->big_endian)
+		iowrite32be(val, addr);
+	else
+		iowrite32(val, addr);
+}
+
+static inline void ftm_counter_enable(void __iomem *base)
+{
+	u32 val;
+
+	/* select and enable counter clock source */
+	val = ftm_readl(base + FTM_SC);
+	val &= ~(FTM_SC_PS_MASK | FTM_SC_CLK_MASK);
+	val |= priv->ps | FTM_SC_CLK(1);
+	ftm_writel(val, base + FTM_SC);
+}
+
+static inline void ftm_counter_disable(void __iomem *base)
+{
+	u32 val;
+
+	/* disable counter clock source */
+	val = ftm_readl(base + FTM_SC);
+	val &= ~(FTM_SC_PS_MASK | FTM_SC_CLK_MASK);
+	ftm_writel(val, base + FTM_SC);
+}
+
+static inline void ftm_irq_acknowledge(void __iomem *base)
+{
+	u32 val;
+
+	val = ftm_readl(base + FTM_SC);
+	val &= ~FTM_SC_TOF;
+	ftm_writel(val, base + FTM_SC);
+}
+
+static inline void ftm_irq_enable(void __iomem *base)
+{
+	u32 val;
+
+	val = ftm_readl(base + FTM_SC);
+	val |= FTM_SC_TOIE;
+	ftm_writel(val, base + FTM_SC);
+}
+
+static inline void ftm_irq_disable(void __iomem *base)
+{
+	u32 val;
+
+	val = ftm_readl(base + FTM_SC);
+	val &= ~FTM_SC_TOIE;
+	ftm_writel(val, base + FTM_SC);
+}
+
+static inline void ftm_reset_counter(void __iomem *base)
+{
+	/*
+	 * The CNT register contains the FTM counter value.
+	 * Reset clears the CNT register. Writing any value to COUNT
+	 * updates the counter with its initial value, CNTIN.
+	 */
+	ftm_writel(0x00, base + FTM_CNT);
+}
+
+static u64 ftm_read_sched_clock(void)
+{
+	return ftm_readl(priv->clksrc_base + FTM_CNT);
+}
+
+static int ftm_set_next_event(unsigned long delta,
+				struct clock_event_device *unused)
+{
+	/*
+	 * The CNNIN and MOD are all double buffer registers, writing
+	 * to the MOD register latches the value into a buffer. The MOD
+	 * register is updated with the value of its write buffer with
+	 * the following scenario:
+	 * a, the counter source clock is diabled.
+	 */
+	ftm_counter_disable(priv->clkevt_base);
+
+	/* Force the value of CNTIN to be loaded into the FTM counter */
+	ftm_reset_counter(priv->clkevt_base);
+
+	/*
+	 * The counter increments until the value of MOD is reached,
+	 * at which point the counter is reloaded with the value of CNTIN.
+	 * The TOF (the overflow flag) bit is set when the FTM counter
+	 * changes from MOD to CNTIN. So we should using the delta - 1.
+	 */
+	ftm_writel(delta - 1, priv->clkevt_base + FTM_MOD);
+
+	ftm_counter_enable(priv->clkevt_base);
+
+	ftm_irq_enable(priv->clkevt_base);
+
+	return 0;
+}
+
+static void ftm_set_mode(enum clock_event_mode mode,
+				struct clock_event_device *evt)
+{
+	switch (mode) {
+	case CLOCK_EVT_MODE_PERIODIC:
+		ftm_set_next_event(priv->periodic_cyc, evt);
+		break;
+	case CLOCK_EVT_MODE_ONESHOT:
+		ftm_counter_disable(priv->clkevt_base);
+		break;
+	default:
+		return;
+	}
+}
+
+static irqreturn_t ftm_evt_interrupt(int irq, void *dev_id)
+{
+	struct clock_event_device *evt = dev_id;
+
+	ftm_irq_acknowledge(priv->clkevt_base);
+
+	if (likely(evt->mode == CLOCK_EVT_MODE_ONESHOT)) {
+		ftm_irq_disable(priv->clkevt_base);
+		ftm_counter_disable(priv->clkevt_base);
+	}
+
+	evt->event_handler(evt);
+
+	return IRQ_HANDLED;
+}
+
+static struct clock_event_device ftm_clockevent = {
+	.name		= "Freescale ftm timer",
+	.features	= CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
+	.set_mode	= ftm_set_mode,
+	.set_next_event	= ftm_set_next_event,
+	.rating		= 300,
+};
+
+static struct irqaction ftm_timer_irq = {
+	.name		= "Freescale ftm timer",
+	.flags		= IRQF_TIMER | IRQF_IRQPOLL,
+	.handler	= ftm_evt_interrupt,
+	.dev_id		= &ftm_clockevent,
+};
+
+static int __init ftm_clockevent_init(unsigned long freq, int irq)
+{
+	int err;
+
+	ftm_writel(0x00, priv->clkevt_base + FTM_CNTIN);
+	ftm_writel(~0UL, priv->clkevt_base + FTM_MOD);
+
+	ftm_reset_counter(priv->clkevt_base);
+
+	err = setup_irq(irq, &ftm_timer_irq);
+	if (err) {
+		pr_err("ftm: setup irq failed: %d\n", err);
+		return err;
+	}
+
+	ftm_clockevent.cpumask = cpumask_of(0);
+	ftm_clockevent.irq = irq;
+
+	clockevents_config_and_register(&ftm_clockevent,
+					freq / (1 << priv->ps),
+					1, 0xffff);
+
+	ftm_counter_enable(priv->clkevt_base);
+
+	return 0;
+}
+
+static int __init ftm_clocksource_init(unsigned long freq)
+{
+	int err;
+
+	ftm_writel(0x00, priv->clksrc_base + FTM_CNTIN);
+	ftm_writel(~0UL, priv->clksrc_base + FTM_MOD);
+
+	ftm_reset_counter(priv->clksrc_base);
+
+	sched_clock_register(ftm_read_sched_clock, 16, freq / (1 << priv->ps));
+	err = clocksource_mmio_init(priv->clksrc_base + FTM_CNT, "fsl-ftm",
+				    freq / (1 << priv->ps), 300, 16,
+				    clocksource_mmio_readl_up);
+	if (err) {
+		pr_err("ftm: init clock source mmio failed: %d\n", err);
+		return err;
+	}
+
+	ftm_counter_enable(priv->clksrc_base);
+
+	return 0;
+}
+
+static int __init __ftm_clk_init(struct device_node *np, char *cnt_name,
+				 char *ftm_name)
+{
+	struct clk *clk;
+	int err;
+
+	clk = of_clk_get_by_name(np, cnt_name);
+	if (IS_ERR(clk)) {
+		pr_err("ftm: Cannot get \"%s\": %ld\n", cnt_name, PTR_ERR(clk));
+		return PTR_ERR(clk);
+	}
+	err = clk_prepare_enable(clk);
+	if (err) {
+		pr_err("ftm: clock failed to prepare+enable \"%s\": %d\n",
+			cnt_name, err);
+		return err;
+	}
+
+	clk = of_clk_get_by_name(np, ftm_name);
+	if (IS_ERR(clk)) {
+		pr_err("ftm: Cannot get \"%s\": %ld\n", ftm_name, PTR_ERR(clk));
+		return PTR_ERR(clk);
+	}
+	err = clk_prepare_enable(clk);
+	if (err)
+		pr_err("ftm: clock failed to prepare+enable \"%s\": %d\n",
+			ftm_name, err);
+
+	return clk_get_rate(clk);
+}
+
+static unsigned long __init ftm_clk_init(struct device_node *np)
+{
+	unsigned long freq;
+
+	freq = __ftm_clk_init(np, "ftm-evt-counter-en", "ftm-evt");
+	if (freq <= 0)
+		return 0;
+
+	freq = __ftm_clk_init(np, "ftm-src-counter-en", "ftm-src");
+	if (freq <= 0)
+		return 0;
+
+	return freq;
+}
+
+static int __init ftm_calc_closest_round_cyc(unsigned long freq)
+{
+	priv->ps = 0;
+
+	/* The counter register is only using the lower 16 bits, and
+	 * if the 'freq' value is to big here, then the periodic_cyc
+	 * may exceed 0xFFFF.
+	 */
+	do {
+		priv->periodic_cyc = DIV_ROUND_CLOSEST(freq,
+						HZ * (1 << priv->ps++));
+	} while (priv->periodic_cyc > 0xFFFF);
+
+	if (priv->ps > FTM_PS_MAX) {
+		pr_err("ftm: the prescaler is %lu > %d\n",
+				priv->ps, FTM_PS_MAX);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static void __init ftm_timer_init(struct device_node *np)
+{
+	unsigned long freq;
+	int irq;
+
+	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return;
+
+	priv->clkevt_base = of_iomap(np, 0);
+	if (!priv->clkevt_base) {
+		pr_err("ftm: unable to map event timer registers\n");
+		goto err;
+	}
+
+	priv->clksrc_base = of_iomap(np, 1);
+	if (!priv->clksrc_base) {
+		pr_err("ftm: unable to map source timer registers\n");
+		goto err;
+	}
+
+	irq = irq_of_parse_and_map(np, 0);
+	if (irq <= 0) {
+		pr_err("ftm: unable to get IRQ from DT, %d\n", irq);
+		goto err;
+	}
+
+	priv->big_endian = of_property_read_bool(np, "big-endian");
+
+	freq = ftm_clk_init(np);
+	if (!freq)
+		goto err;
+
+	if (ftm_calc_closest_round_cyc(freq))
+		goto err;
+
+	if (ftm_clocksource_init(freq))
+		goto err;
+
+	if (ftm_clockevent_init(freq, irq))
+		goto err;
+
+	return;
+
+err:
+	kfree(priv);
+}
+CLOCKSOURCE_OF_DECLARE(flextimer, "fsl,ftm-timer", ftm_timer_init);
diff --git a/drivers/clocksource/mmio.c b/drivers/clocksource/mmio.c
index c0e25125a55e..1593ade2a815 100644
--- a/drivers/clocksource/mmio.c
+++ b/drivers/clocksource/mmio.c
@@ -22,22 +22,22 @@ static inline struct clocksource_mmio *to_mmio_clksrc(struct clocksource *c)
 
 cycle_t clocksource_mmio_readl_up(struct clocksource *c)
 {
-	return readl_relaxed(to_mmio_clksrc(c)->reg);
+	return (cycle_t)readl_relaxed(to_mmio_clksrc(c)->reg);
 }
 
 cycle_t clocksource_mmio_readl_down(struct clocksource *c)
 {
-	return ~readl_relaxed(to_mmio_clksrc(c)->reg);
+	return ~(cycle_t)readl_relaxed(to_mmio_clksrc(c)->reg) & c->mask;
 }
 
 cycle_t clocksource_mmio_readw_up(struct clocksource *c)
 {
-	return readw_relaxed(to_mmio_clksrc(c)->reg);
+	return (cycle_t)readw_relaxed(to_mmio_clksrc(c)->reg);
 }
 
 cycle_t clocksource_mmio_readw_down(struct clocksource *c)
 {
-	return ~(unsigned)readw_relaxed(to_mmio_clksrc(c)->reg);
+	return ~(cycle_t)readw_relaxed(to_mmio_clksrc(c)->reg) & c->mask;
 }
 
 /**
diff --git a/drivers/clocksource/qcom-timer.c b/drivers/clocksource/qcom-timer.c
index e807acf4c665..8d115db1e651 100644
--- a/drivers/clocksource/qcom-timer.c
+++ b/drivers/clocksource/qcom-timer.c
@@ -26,6 +26,8 @@
 #include <linux/of_irq.h>
 #include <linux/sched_clock.h>
 
+#include <asm/delay.h>
+
 #define TIMER_MATCH_VAL			0x0000
 #define TIMER_COUNT_VAL			0x0004
 #define TIMER_ENABLE			0x0008
@@ -179,6 +181,15 @@ static u64 notrace msm_sched_clock_read(void)
 	return msm_clocksource.read(&msm_clocksource);
 }
 
+static unsigned long msm_read_current_timer(void)
+{
+	return msm_clocksource.read(&msm_clocksource);
+}
+
+static struct delay_timer msm_delay_timer = {
+	.read_current_timer = msm_read_current_timer,
+};
+
 static void __init msm_timer_init(u32 dgt_hz, int sched_bits, int irq,
 				  bool percpu)
 {
@@ -217,6 +228,8 @@ err:
 	if (res)
 		pr_err("clocksource_register failed\n");
 	sched_clock_register(msm_sched_clock_read, sched_bits, dgt_hz);
+	msm_delay_timer.freq = dgt_hz;
+	register_current_timer_delay(&msm_delay_timer);
 }
 
 #ifdef CONFIG_ARCH_QCOM
diff --git a/drivers/clocksource/sh_cmt.c b/drivers/clocksource/sh_cmt.c
index 0b1836a6c539..dfa780396b91 100644
--- a/drivers/clocksource/sh_cmt.c
+++ b/drivers/clocksource/sh_cmt.c
@@ -11,40 +11,93 @@
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
  */
 
+#include <linux/clk.h>
+#include <linux/clockchips.h>
+#include <linux/clocksource.h>
+#include <linux/delay.h>
+#include <linux/err.h>
 #include <linux/init.h>
-#include <linux/platform_device.h>
-#include <linux/spinlock.h>
 #include <linux/interrupt.h>
-#include <linux/ioport.h>
 #include <linux/io.h>
-#include <linux/clk.h>
+#include <linux/ioport.h>
 #include <linux/irq.h>
-#include <linux/err.h>
-#include <linux/delay.h>
-#include <linux/clocksource.h>
-#include <linux/clockchips.h>
-#include <linux/sh_timer.h>
-#include <linux/slab.h>
 #include <linux/module.h>
+#include <linux/platform_device.h>
 #include <linux/pm_domain.h>
 #include <linux/pm_runtime.h>
+#include <linux/sh_timer.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+
+struct sh_cmt_device;
+
+/*
+ * The CMT comes in 5 different identified flavours, depending not only on the
+ * SoC but also on the particular instance. The following table lists the main
+ * characteristics of those flavours.
+ *
+ *			16B	32B	32B-F	48B	48B-2
+ * -----------------------------------------------------------------------------
+ * Channels		2	1/4	1	6	2/8
+ * Control Width	16	16	16	16	32
+ * Counter Width	16	32	32	32/48	32/48
+ * Shared Start/Stop	Y	Y	Y	Y	N
+ *
+ * The 48-bit gen2 version has a per-channel start/stop register located in the
+ * channel registers block. All other versions have a shared start/stop register
+ * located in the global space.
+ *
+ * Channels are indexed from 0 to N-1 in the documentation. The channel index
+ * infers the start/stop bit position in the control register and the channel
+ * registers block address. Some CMT instances have a subset of channels
+ * available, in which case the index in the documentation doesn't match the
+ * "real" index as implemented in hardware. This is for instance the case with
+ * CMT0 on r8a7740, which is a 32-bit variant with a single channel numbered 0
+ * in the documentation but using start/stop bit 5 and having its registers
+ * block at 0x60.
+ *
+ * Similarly CMT0 on r8a73a4, r8a7790 and r8a7791, while implementing 32-bit
+ * channels only, is a 48-bit gen2 CMT with the 48-bit channels unavailable.
+ */
+
+enum sh_cmt_model {
+	SH_CMT_16BIT,
+	SH_CMT_32BIT,
+	SH_CMT_32BIT_FAST,
+	SH_CMT_48BIT,
+	SH_CMT_48BIT_GEN2,
+};
+
+struct sh_cmt_info {
+	enum sh_cmt_model model;
 
-struct sh_cmt_priv {
-	void __iomem *mapbase;
-	void __iomem *mapbase_str;
-	struct clk *clk;
 	unsigned long width; /* 16 or 32 bit version of hardware block */
 	unsigned long overflow_bit;
 	unsigned long clear_bits;
-	struct irqaction irqaction;
-	struct platform_device *pdev;
 
+	/* callbacks for CMSTR and CMCSR access */
+	unsigned long (*read_control)(void __iomem *base, unsigned long offs);
+	void (*write_control)(void __iomem *base, unsigned long offs,
+			      unsigned long value);
+
+	/* callbacks for CMCNT and CMCOR access */
+	unsigned long (*read_count)(void __iomem *base, unsigned long offs);
+	void (*write_count)(void __iomem *base, unsigned long offs,
+			    unsigned long value);
+};
+
+struct sh_cmt_channel {
+	struct sh_cmt_device *cmt;
+
+	unsigned int index;	/* Index in the documentation */
+	unsigned int hwidx;	/* Real hardware index */
+
+	void __iomem *iostart;
+	void __iomem *ioctrl;
+
+	unsigned int timer_bit;
 	unsigned long flags;
 	unsigned long match_value;
 	unsigned long next_match_value;
@@ -55,38 +108,52 @@ struct sh_cmt_priv {
 	struct clocksource cs;
 	unsigned long total_cycles;
 	bool cs_enabled;
+};
 
-	/* callbacks for CMSTR and CMCSR access */
-	unsigned long (*read_control)(void __iomem *base, unsigned long offs);
-	void (*write_control)(void __iomem *base, unsigned long offs,
-			      unsigned long value);
+struct sh_cmt_device {
+	struct platform_device *pdev;
 
-	/* callbacks for CMCNT and CMCOR access */
-	unsigned long (*read_count)(void __iomem *base, unsigned long offs);
-	void (*write_count)(void __iomem *base, unsigned long offs,
-			    unsigned long value);
+	const struct sh_cmt_info *info;
+	bool legacy;
+
+	void __iomem *mapbase_ch;
+	void __iomem *mapbase;
+	struct clk *clk;
+
+	struct sh_cmt_channel *channels;
+	unsigned int num_channels;
+
+	bool has_clockevent;
+	bool has_clocksource;
 };
 
-/* Examples of supported CMT timer register layouts and I/O access widths:
- *
- * "16-bit counter and 16-bit control" as found on sh7263:
- * CMSTR 0xfffec000 16-bit
- * CMCSR 0xfffec002 16-bit
- * CMCNT 0xfffec004 16-bit
- * CMCOR 0xfffec006 16-bit
- *
- * "32-bit counter and 16-bit control" as found on sh7372, sh73a0, r8a7740:
- * CMSTR 0xffca0000 16-bit
- * CMCSR 0xffca0060 16-bit
- * CMCNT 0xffca0064 32-bit
- * CMCOR 0xffca0068 32-bit
- *
- * "32-bit counter and 32-bit control" as found on r8a73a4 and r8a7790:
- * CMSTR 0xffca0500 32-bit
- * CMCSR 0xffca0510 32-bit
- * CMCNT 0xffca0514 32-bit
- * CMCOR 0xffca0518 32-bit
- */
+#define SH_CMT16_CMCSR_CMF		(1 << 7)
+#define SH_CMT16_CMCSR_CMIE		(1 << 6)
+#define SH_CMT16_CMCSR_CKS8		(0 << 0)
+#define SH_CMT16_CMCSR_CKS32		(1 << 0)
+#define SH_CMT16_CMCSR_CKS128		(2 << 0)
+#define SH_CMT16_CMCSR_CKS512		(3 << 0)
+#define SH_CMT16_CMCSR_CKS_MASK		(3 << 0)
+
+#define SH_CMT32_CMCSR_CMF		(1 << 15)
+#define SH_CMT32_CMCSR_OVF		(1 << 14)
+#define SH_CMT32_CMCSR_WRFLG		(1 << 13)
+#define SH_CMT32_CMCSR_STTF		(1 << 12)
+#define SH_CMT32_CMCSR_STPF		(1 << 11)
+#define SH_CMT32_CMCSR_SSIE		(1 << 10)
+#define SH_CMT32_CMCSR_CMS		(1 << 9)
+#define SH_CMT32_CMCSR_CMM		(1 << 8)
+#define SH_CMT32_CMCSR_CMTOUT_IE	(1 << 7)
+#define SH_CMT32_CMCSR_CMR_NONE		(0 << 4)
+#define SH_CMT32_CMCSR_CMR_DMA		(1 << 4)
+#define SH_CMT32_CMCSR_CMR_IRQ		(2 << 4)
+#define SH_CMT32_CMCSR_CMR_MASK		(3 << 4)
+#define SH_CMT32_CMCSR_DBGIVD		(1 << 3)
+#define SH_CMT32_CMCSR_CKS_RCLK8	(4 << 0)
+#define SH_CMT32_CMCSR_CKS_RCLK32	(5 << 0)
+#define SH_CMT32_CMCSR_CKS_RCLK128	(6 << 0)
+#define SH_CMT32_CMCSR_CKS_RCLK1	(7 << 0)
+#define SH_CMT32_CMCSR_CKS_MASK		(7 << 0)
 
 static unsigned long sh_cmt_read16(void __iomem *base, unsigned long offs)
 {
@@ -110,64 +177,123 @@ static void sh_cmt_write32(void __iomem *base, unsigned long offs,
 	iowrite32(value, base + (offs << 2));
 }
 
+static const struct sh_cmt_info sh_cmt_info[] = {
+	[SH_CMT_16BIT] = {
+		.model = SH_CMT_16BIT,
+		.width = 16,
+		.overflow_bit = SH_CMT16_CMCSR_CMF,
+		.clear_bits = ~SH_CMT16_CMCSR_CMF,
+		.read_control = sh_cmt_read16,
+		.write_control = sh_cmt_write16,
+		.read_count = sh_cmt_read16,
+		.write_count = sh_cmt_write16,
+	},
+	[SH_CMT_32BIT] = {
+		.model = SH_CMT_32BIT,
+		.width = 32,
+		.overflow_bit = SH_CMT32_CMCSR_CMF,
+		.clear_bits = ~(SH_CMT32_CMCSR_CMF | SH_CMT32_CMCSR_OVF),
+		.read_control = sh_cmt_read16,
+		.write_control = sh_cmt_write16,
+		.read_count = sh_cmt_read32,
+		.write_count = sh_cmt_write32,
+	},
+	[SH_CMT_32BIT_FAST] = {
+		.model = SH_CMT_32BIT_FAST,
+		.width = 32,
+		.overflow_bit = SH_CMT32_CMCSR_CMF,
+		.clear_bits = ~(SH_CMT32_CMCSR_CMF | SH_CMT32_CMCSR_OVF),
+		.read_control = sh_cmt_read16,
+		.write_control = sh_cmt_write16,
+		.read_count = sh_cmt_read32,
+		.write_count = sh_cmt_write32,
+	},
+	[SH_CMT_48BIT] = {
+		.model = SH_CMT_48BIT,
+		.width = 32,
+		.overflow_bit = SH_CMT32_CMCSR_CMF,
+		.clear_bits = ~(SH_CMT32_CMCSR_CMF | SH_CMT32_CMCSR_OVF),
+		.read_control = sh_cmt_read32,
+		.write_control = sh_cmt_write32,
+		.read_count = sh_cmt_read32,
+		.write_count = sh_cmt_write32,
+	},
+	[SH_CMT_48BIT_GEN2] = {
+		.model = SH_CMT_48BIT_GEN2,
+		.width = 32,
+		.overflow_bit = SH_CMT32_CMCSR_CMF,
+		.clear_bits = ~(SH_CMT32_CMCSR_CMF | SH_CMT32_CMCSR_OVF),
+		.read_control = sh_cmt_read32,
+		.write_control = sh_cmt_write32,
+		.read_count = sh_cmt_read32,
+		.write_count = sh_cmt_write32,
+	},
+};
+
 #define CMCSR 0 /* channel register */
 #define CMCNT 1 /* channel register */
 #define CMCOR 2 /* channel register */
 
-static inline unsigned long sh_cmt_read_cmstr(struct sh_cmt_priv *p)
+static inline unsigned long sh_cmt_read_cmstr(struct sh_cmt_channel *ch)
 {
-	return p->read_control(p->mapbase_str, 0);
+	if (ch->iostart)
+		return ch->cmt->info->read_control(ch->iostart, 0);
+	else
+		return ch->cmt->info->read_control(ch->cmt->mapbase, 0);
 }
 
-static inline unsigned long sh_cmt_read_cmcsr(struct sh_cmt_priv *p)
+static inline void sh_cmt_write_cmstr(struct sh_cmt_channel *ch,
+				      unsigned long value)
 {
-	return p->read_control(p->mapbase, CMCSR);
+	if (ch->iostart)
+		ch->cmt->info->write_control(ch->iostart, 0, value);
+	else
+		ch->cmt->info->write_control(ch->cmt->mapbase, 0, value);
 }
 
-static inline unsigned long sh_cmt_read_cmcnt(struct sh_cmt_priv *p)
+static inline unsigned long sh_cmt_read_cmcsr(struct sh_cmt_channel *ch)
 {
-	return p->read_count(p->mapbase, CMCNT);
+	return ch->cmt->info->read_control(ch->ioctrl, CMCSR);
 }
 
-static inline void sh_cmt_write_cmstr(struct sh_cmt_priv *p,
+static inline void sh_cmt_write_cmcsr(struct sh_cmt_channel *ch,
 				      unsigned long value)
 {
-	p->write_control(p->mapbase_str, 0, value);
+	ch->cmt->info->write_control(ch->ioctrl, CMCSR, value);
 }
 
-static inline void sh_cmt_write_cmcsr(struct sh_cmt_priv *p,
-				      unsigned long value)
+static inline unsigned long sh_cmt_read_cmcnt(struct sh_cmt_channel *ch)
 {
-	p->write_control(p->mapbase, CMCSR, value);
+	return ch->cmt->info->read_count(ch->ioctrl, CMCNT);
 }
 
-static inline void sh_cmt_write_cmcnt(struct sh_cmt_priv *p,
+static inline void sh_cmt_write_cmcnt(struct sh_cmt_channel *ch,
 				      unsigned long value)
 {
-	p->write_count(p->mapbase, CMCNT, value);
+	ch->cmt->info->write_count(ch->ioctrl, CMCNT, value);
 }
 
-static inline void sh_cmt_write_cmcor(struct sh_cmt_priv *p,
+static inline void sh_cmt_write_cmcor(struct sh_cmt_channel *ch,
 				      unsigned long value)
 {
-	p->write_count(p->mapbase, CMCOR, value);
+	ch->cmt->info->write_count(ch->ioctrl, CMCOR, value);
 }
 
-static unsigned long sh_cmt_get_counter(struct sh_cmt_priv *p,
+static unsigned long sh_cmt_get_counter(struct sh_cmt_channel *ch,
 					int *has_wrapped)
 {
 	unsigned long v1, v2, v3;
 	int o1, o2;
 
-	o1 = sh_cmt_read_cmcsr(p) & p->overflow_bit;
+	o1 = sh_cmt_read_cmcsr(ch) & ch->cmt->info->overflow_bit;
 
 	/* Make sure the timer value is stable. Stolen from acpi_pm.c */
 	do {
 		o2 = o1;
-		v1 = sh_cmt_read_cmcnt(p);
-		v2 = sh_cmt_read_cmcnt(p);
-		v3 = sh_cmt_read_cmcnt(p);
-		o1 = sh_cmt_read_cmcsr(p) & p->overflow_bit;
+		v1 = sh_cmt_read_cmcnt(ch);
+		v2 = sh_cmt_read_cmcnt(ch);
+		v3 = sh_cmt_read_cmcnt(ch);
+		o1 = sh_cmt_read_cmcsr(ch) & ch->cmt->info->overflow_bit;
 	} while (unlikely((o1 != o2) || (v1 > v2 && v1 < v3)
 			  || (v2 > v3 && v2 < v1) || (v3 > v1 && v3 < v2)));
 
@@ -177,52 +303,56 @@ static unsigned long sh_cmt_get_counter(struct sh_cmt_priv *p,
 
 static DEFINE_RAW_SPINLOCK(sh_cmt_lock);
 
-static void sh_cmt_start_stop_ch(struct sh_cmt_priv *p, int start)
+static void sh_cmt_start_stop_ch(struct sh_cmt_channel *ch, int start)
 {
-	struct sh_timer_config *cfg = p->pdev->dev.platform_data;
 	unsigned long flags, value;
 
 	/* start stop register shared by multiple timer channels */
 	raw_spin_lock_irqsave(&sh_cmt_lock, flags);
-	value = sh_cmt_read_cmstr(p);
+	value = sh_cmt_read_cmstr(ch);
 
 	if (start)
-		value |= 1 << cfg->timer_bit;
+		value |= 1 << ch->timer_bit;
 	else
-		value &= ~(1 << cfg->timer_bit);
+		value &= ~(1 << ch->timer_bit);
 
-	sh_cmt_write_cmstr(p, value);
+	sh_cmt_write_cmstr(ch, value);
 	raw_spin_unlock_irqrestore(&sh_cmt_lock, flags);
 }
 
-static int sh_cmt_enable(struct sh_cmt_priv *p, unsigned long *rate)
+static int sh_cmt_enable(struct sh_cmt_channel *ch, unsigned long *rate)
 {
 	int k, ret;
 
-	pm_runtime_get_sync(&p->pdev->dev);
-	dev_pm_syscore_device(&p->pdev->dev, true);
+	pm_runtime_get_sync(&ch->cmt->pdev->dev);
+	dev_pm_syscore_device(&ch->cmt->pdev->dev, true);
 
 	/* enable clock */
-	ret = clk_enable(p->clk);
+	ret = clk_enable(ch->cmt->clk);
 	if (ret) {
-		dev_err(&p->pdev->dev, "cannot enable clock\n");
+		dev_err(&ch->cmt->pdev->dev, "ch%u: cannot enable clock\n",
+			ch->index);
 		goto err0;
 	}
 
 	/* make sure channel is disabled */
-	sh_cmt_start_stop_ch(p, 0);
+	sh_cmt_start_stop_ch(ch, 0);
 
 	/* configure channel, periodic mode and maximum timeout */
-	if (p->width == 16) {
-		*rate = clk_get_rate(p->clk) / 512;
-		sh_cmt_write_cmcsr(p, 0x43);
+	if (ch->cmt->info->width == 16) {
+		*rate = clk_get_rate(ch->cmt->clk) / 512;
+		sh_cmt_write_cmcsr(ch, SH_CMT16_CMCSR_CMIE |
+				   SH_CMT16_CMCSR_CKS512);
 	} else {
-		*rate = clk_get_rate(p->clk) / 8;
-		sh_cmt_write_cmcsr(p, 0x01a4);
+		*rate = clk_get_rate(ch->cmt->clk) / 8;
+		sh_cmt_write_cmcsr(ch, SH_CMT32_CMCSR_CMM |
+				   SH_CMT32_CMCSR_CMTOUT_IE |
+				   SH_CMT32_CMCSR_CMR_IRQ |
+				   SH_CMT32_CMCSR_CKS_RCLK8);
 	}
 
-	sh_cmt_write_cmcor(p, 0xffffffff);
-	sh_cmt_write_cmcnt(p, 0);
+	sh_cmt_write_cmcor(ch, 0xffffffff);
+	sh_cmt_write_cmcnt(ch, 0);
 
 	/*
 	 * According to the sh73a0 user's manual, as CMCNT can be operated
@@ -236,41 +366,42 @@ static int sh_cmt_enable(struct sh_cmt_priv *p, unsigned long *rate)
 	 * take RCLKx2 at maximum.
 	 */
 	for (k = 0; k < 100; k++) {
-		if (!sh_cmt_read_cmcnt(p))
+		if (!sh_cmt_read_cmcnt(ch))
 			break;
 		udelay(1);
 	}
 
-	if (sh_cmt_read_cmcnt(p)) {
-		dev_err(&p->pdev->dev, "cannot clear CMCNT\n");
+	if (sh_cmt_read_cmcnt(ch)) {
+		dev_err(&ch->cmt->pdev->dev, "ch%u: cannot clear CMCNT\n",
+			ch->index);
 		ret = -ETIMEDOUT;
 		goto err1;
 	}
 
 	/* enable channel */
-	sh_cmt_start_stop_ch(p, 1);
+	sh_cmt_start_stop_ch(ch, 1);
 	return 0;
  err1:
 	/* stop clock */
-	clk_disable(p->clk);
+	clk_disable(ch->cmt->clk);
 
  err0:
 	return ret;
 }
 
-static void sh_cmt_disable(struct sh_cmt_priv *p)
+static void sh_cmt_disable(struct sh_cmt_channel *ch)
 {
 	/* disable channel */
-	sh_cmt_start_stop_ch(p, 0);
+	sh_cmt_start_stop_ch(ch, 0);
 
 	/* disable interrupts in CMT block */
-	sh_cmt_write_cmcsr(p, 0);
+	sh_cmt_write_cmcsr(ch, 0);
 
 	/* stop clock */
-	clk_disable(p->clk);
+	clk_disable(ch->cmt->clk);
 
-	dev_pm_syscore_device(&p->pdev->dev, false);
-	pm_runtime_put(&p->pdev->dev);
+	dev_pm_syscore_device(&ch->cmt->pdev->dev, false);
+	pm_runtime_put(&ch->cmt->pdev->dev);
 }
 
 /* private flags */
@@ -280,24 +411,24 @@ static void sh_cmt_disable(struct sh_cmt_priv *p)
 #define FLAG_SKIPEVENT (1 << 3)
 #define FLAG_IRQCONTEXT (1 << 4)
 
-static void sh_cmt_clock_event_program_verify(struct sh_cmt_priv *p,
+static void sh_cmt_clock_event_program_verify(struct sh_cmt_channel *ch,
 					      int absolute)
 {
 	unsigned long new_match;
-	unsigned long value = p->next_match_value;
+	unsigned long value = ch->next_match_value;
 	unsigned long delay = 0;
 	unsigned long now = 0;
 	int has_wrapped;
 
-	now = sh_cmt_get_counter(p, &has_wrapped);
-	p->flags |= FLAG_REPROGRAM; /* force reprogram */
+	now = sh_cmt_get_counter(ch, &has_wrapped);
+	ch->flags |= FLAG_REPROGRAM; /* force reprogram */
 
 	if (has_wrapped) {
 		/* we're competing with the interrupt handler.
 		 *  -> let the interrupt handler reprogram the timer.
 		 *  -> interrupt number two handles the event.
 		 */
-		p->flags |= FLAG_SKIPEVENT;
+		ch->flags |= FLAG_SKIPEVENT;
 		return;
 	}
 
@@ -309,20 +440,20 @@ static void sh_cmt_clock_event_program_verify(struct sh_cmt_priv *p,
 		 * but don't save the new match value yet.
 		 */
 		new_match = now + value + delay;
-		if (new_match > p->max_match_value)
-			new_match = p->max_match_value;
+		if (new_match > ch->max_match_value)
+			new_match = ch->max_match_value;
 
-		sh_cmt_write_cmcor(p, new_match);
+		sh_cmt_write_cmcor(ch, new_match);
 
-		now = sh_cmt_get_counter(p, &has_wrapped);
-		if (has_wrapped && (new_match > p->match_value)) {
+		now = sh_cmt_get_counter(ch, &has_wrapped);
+		if (has_wrapped && (new_match > ch->match_value)) {
 			/* we are changing to a greater match value,
 			 * so this wrap must be caused by the counter
 			 * matching the old value.
 			 * -> first interrupt reprograms the timer.
 			 * -> interrupt number two handles the event.
 			 */
-			p->flags |= FLAG_SKIPEVENT;
+			ch->flags |= FLAG_SKIPEVENT;
 			break;
 		}
 
@@ -333,7 +464,7 @@ static void sh_cmt_clock_event_program_verify(struct sh_cmt_priv *p,
 			 * -> save programmed match value.
 			 * -> let isr handle the event.
 			 */
-			p->match_value = new_match;
+			ch->match_value = new_match;
 			break;
 		}
 
@@ -344,7 +475,7 @@ static void sh_cmt_clock_event_program_verify(struct sh_cmt_priv *p,
 			 * -> save programmed match value.
 			 * -> let isr handle the event.
 			 */
-			p->match_value = new_match;
+			ch->match_value = new_match;
 			break;
 		}
 
@@ -360,138 +491,141 @@ static void sh_cmt_clock_event_program_verify(struct sh_cmt_priv *p,
 			delay = 1;
 
 		if (!delay)
-			dev_warn(&p->pdev->dev, "too long delay\n");
+			dev_warn(&ch->cmt->pdev->dev, "ch%u: too long delay\n",
+				 ch->index);
 
 	} while (delay);
 }
 
-static void __sh_cmt_set_next(struct sh_cmt_priv *p, unsigned long delta)
+static void __sh_cmt_set_next(struct sh_cmt_channel *ch, unsigned long delta)
 {
-	if (delta > p->max_match_value)
-		dev_warn(&p->pdev->dev, "delta out of range\n");
+	if (delta > ch->max_match_value)
+		dev_warn(&ch->cmt->pdev->dev, "ch%u: delta out of range\n",
+			 ch->index);
 
-	p->next_match_value = delta;
-	sh_cmt_clock_event_program_verify(p, 0);
+	ch->next_match_value = delta;
+	sh_cmt_clock_event_program_verify(ch, 0);
 }
 
-static void sh_cmt_set_next(struct sh_cmt_priv *p, unsigned long delta)
+static void sh_cmt_set_next(struct sh_cmt_channel *ch, unsigned long delta)
 {
 	unsigned long flags;
 
-	raw_spin_lock_irqsave(&p->lock, flags);
-	__sh_cmt_set_next(p, delta);
-	raw_spin_unlock_irqrestore(&p->lock, flags);
+	raw_spin_lock_irqsave(&ch->lock, flags);
+	__sh_cmt_set_next(ch, delta);
+	raw_spin_unlock_irqrestore(&ch->lock, flags);
 }
 
 static irqreturn_t sh_cmt_interrupt(int irq, void *dev_id)
 {
-	struct sh_cmt_priv *p = dev_id;
+	struct sh_cmt_channel *ch = dev_id;
 
 	/* clear flags */
-	sh_cmt_write_cmcsr(p, sh_cmt_read_cmcsr(p) & p->clear_bits);
+	sh_cmt_write_cmcsr(ch, sh_cmt_read_cmcsr(ch) &
+			   ch->cmt->info->clear_bits);
 
 	/* update clock source counter to begin with if enabled
 	 * the wrap flag should be cleared by the timer specific
 	 * isr before we end up here.
 	 */
-	if (p->flags & FLAG_CLOCKSOURCE)
-		p->total_cycles += p->match_value + 1;
+	if (ch->flags & FLAG_CLOCKSOURCE)
+		ch->total_cycles += ch->match_value + 1;
 
-	if (!(p->flags & FLAG_REPROGRAM))
-		p->next_match_value = p->max_match_value;
+	if (!(ch->flags & FLAG_REPROGRAM))
+		ch->next_match_value = ch->max_match_value;
 
-	p->flags |= FLAG_IRQCONTEXT;
+	ch->flags |= FLAG_IRQCONTEXT;
 
-	if (p->flags & FLAG_CLOCKEVENT) {
-		if (!(p->flags & FLAG_SKIPEVENT)) {
-			if (p->ced.mode == CLOCK_EVT_MODE_ONESHOT) {
-				p->next_match_value = p->max_match_value;
-				p->flags |= FLAG_REPROGRAM;
+	if (ch->flags & FLAG_CLOCKEVENT) {
+		if (!(ch->flags & FLAG_SKIPEVENT)) {
+			if (ch->ced.mode == CLOCK_EVT_MODE_ONESHOT) {
+				ch->next_match_value = ch->max_match_value;
+				ch->flags |= FLAG_REPROGRAM;
 			}
 
-			p->ced.event_handler(&p->ced);
+			ch->ced.event_handler(&ch->ced);
 		}
 	}
 
-	p->flags &= ~FLAG_SKIPEVENT;
+	ch->flags &= ~FLAG_SKIPEVENT;
 
-	if (p->flags & FLAG_REPROGRAM) {
-		p->flags &= ~FLAG_REPROGRAM;
-		sh_cmt_clock_event_program_verify(p, 1);
+	if (ch->flags & FLAG_REPROGRAM) {
+		ch->flags &= ~FLAG_REPROGRAM;
+		sh_cmt_clock_event_program_verify(ch, 1);
 
-		if (p->flags & FLAG_CLOCKEVENT)
-			if ((p->ced.mode == CLOCK_EVT_MODE_SHUTDOWN)
-			    || (p->match_value == p->next_match_value))
-				p->flags &= ~FLAG_REPROGRAM;
+		if (ch->flags & FLAG_CLOCKEVENT)
+			if ((ch->ced.mode == CLOCK_EVT_MODE_SHUTDOWN)
+			    || (ch->match_value == ch->next_match_value))
+				ch->flags &= ~FLAG_REPROGRAM;
 	}
 
-	p->flags &= ~FLAG_IRQCONTEXT;
+	ch->flags &= ~FLAG_IRQCONTEXT;
 
 	return IRQ_HANDLED;
 }
 
-static int sh_cmt_start(struct sh_cmt_priv *p, unsigned long flag)
+static int sh_cmt_start(struct sh_cmt_channel *ch, unsigned long flag)
 {
 	int ret = 0;
 	unsigned long flags;
 
-	raw_spin_lock_irqsave(&p->lock, flags);
+	raw_spin_lock_irqsave(&ch->lock, flags);
 
-	if (!(p->flags & (FLAG_CLOCKEVENT | FLAG_CLOCKSOURCE)))
-		ret = sh_cmt_enable(p, &p->rate);
+	if (!(ch->flags & (FLAG_CLOCKEVENT | FLAG_CLOCKSOURCE)))
+		ret = sh_cmt_enable(ch, &ch->rate);
 
 	if (ret)
 		goto out;
-	p->flags |= flag;
+	ch->flags |= flag;
 
 	/* setup timeout if no clockevent */
-	if ((flag == FLAG_CLOCKSOURCE) && (!(p->flags & FLAG_CLOCKEVENT)))
-		__sh_cmt_set_next(p, p->max_match_value);
+	if ((flag == FLAG_CLOCKSOURCE) && (!(ch->flags & FLAG_CLOCKEVENT)))
+		__sh_cmt_set_next(ch, ch->max_match_value);
  out:
-	raw_spin_unlock_irqrestore(&p->lock, flags);
+	raw_spin_unlock_irqrestore(&ch->lock, flags);
 
 	return ret;
 }
 
-static void sh_cmt_stop(struct sh_cmt_priv *p, unsigned long flag)
+static void sh_cmt_stop(struct sh_cmt_channel *ch, unsigned long flag)
 {
 	unsigned long flags;
 	unsigned long f;
 
-	raw_spin_lock_irqsave(&p->lock, flags);
+	raw_spin_lock_irqsave(&ch->lock, flags);
 
-	f = p->flags & (FLAG_CLOCKEVENT | FLAG_CLOCKSOURCE);
-	p->flags &= ~flag;
+	f = ch->flags & (FLAG_CLOCKEVENT | FLAG_CLOCKSOURCE);
+	ch->flags &= ~flag;
 
-	if (f && !(p->flags & (FLAG_CLOCKEVENT | FLAG_CLOCKSOURCE)))
-		sh_cmt_disable(p);
+	if (f && !(ch->flags & (FLAG_CLOCKEVENT | FLAG_CLOCKSOURCE)))
+		sh_cmt_disable(ch);
 
 	/* adjust the timeout to maximum if only clocksource left */
-	if ((flag == FLAG_CLOCKEVENT) && (p->flags & FLAG_CLOCKSOURCE))
-		__sh_cmt_set_next(p, p->max_match_value);
+	if ((flag == FLAG_CLOCKEVENT) && (ch->flags & FLAG_CLOCKSOURCE))
+		__sh_cmt_set_next(ch, ch->max_match_value);
 
-	raw_spin_unlock_irqrestore(&p->lock, flags);
+	raw_spin_unlock_irqrestore(&ch->lock, flags);
 }
 
-static struct sh_cmt_priv *cs_to_sh_cmt(struct clocksource *cs)
+static struct sh_cmt_channel *cs_to_sh_cmt(struct clocksource *cs)
 {
-	return container_of(cs, struct sh_cmt_priv, cs);
+	return container_of(cs, struct sh_cmt_channel, cs);
 }
 
 static cycle_t sh_cmt_clocksource_read(struct clocksource *cs)
 {
-	struct sh_cmt_priv *p = cs_to_sh_cmt(cs);
+	struct sh_cmt_channel *ch = cs_to_sh_cmt(cs);
 	unsigned long flags, raw;
 	unsigned long value;
 	int has_wrapped;
 
-	raw_spin_lock_irqsave(&p->lock, flags);
-	value = p->total_cycles;
-	raw = sh_cmt_get_counter(p, &has_wrapped);
+	raw_spin_lock_irqsave(&ch->lock, flags);
+	value = ch->total_cycles;
+	raw = sh_cmt_get_counter(ch, &has_wrapped);
 
 	if (unlikely(has_wrapped))
-		raw += p->match_value + 1;
-	raw_spin_unlock_irqrestore(&p->lock, flags);
+		raw += ch->match_value + 1;
+	raw_spin_unlock_irqrestore(&ch->lock, flags);
 
 	return value + raw;
 }
@@ -499,53 +633,53 @@ static cycle_t sh_cmt_clocksource_read(struct clocksource *cs)
 static int sh_cmt_clocksource_enable(struct clocksource *cs)
 {
 	int ret;
-	struct sh_cmt_priv *p = cs_to_sh_cmt(cs);
+	struct sh_cmt_channel *ch = cs_to_sh_cmt(cs);
 
-	WARN_ON(p->cs_enabled);
+	WARN_ON(ch->cs_enabled);
 
-	p->total_cycles = 0;
+	ch->total_cycles = 0;
 
-	ret = sh_cmt_start(p, FLAG_CLOCKSOURCE);
+	ret = sh_cmt_start(ch, FLAG_CLOCKSOURCE);
 	if (!ret) {
-		__clocksource_updatefreq_hz(cs, p->rate);
-		p->cs_enabled = true;
+		__clocksource_updatefreq_hz(cs, ch->rate);
+		ch->cs_enabled = true;
 	}
 	return ret;
 }
 
 static void sh_cmt_clocksource_disable(struct clocksource *cs)
 {
-	struct sh_cmt_priv *p = cs_to_sh_cmt(cs);
+	struct sh_cmt_channel *ch = cs_to_sh_cmt(cs);
 
-	WARN_ON(!p->cs_enabled);
+	WARN_ON(!ch->cs_enabled);
 
-	sh_cmt_stop(p, FLAG_CLOCKSOURCE);
-	p->cs_enabled = false;
+	sh_cmt_stop(ch, FLAG_CLOCKSOURCE);
+	ch->cs_enabled = false;
 }
 
 static void sh_cmt_clocksource_suspend(struct clocksource *cs)
 {
-	struct sh_cmt_priv *p = cs_to_sh_cmt(cs);
+	struct sh_cmt_channel *ch = cs_to_sh_cmt(cs);
 
-	sh_cmt_stop(p, FLAG_CLOCKSOURCE);
-	pm_genpd_syscore_poweroff(&p->pdev->dev);
+	sh_cmt_stop(ch, FLAG_CLOCKSOURCE);
+	pm_genpd_syscore_poweroff(&ch->cmt->pdev->dev);
 }
 
 static void sh_cmt_clocksource_resume(struct clocksource *cs)
 {
-	struct sh_cmt_priv *p = cs_to_sh_cmt(cs);
+	struct sh_cmt_channel *ch = cs_to_sh_cmt(cs);
 
-	pm_genpd_syscore_poweron(&p->pdev->dev);
-	sh_cmt_start(p, FLAG_CLOCKSOURCE);
+	pm_genpd_syscore_poweron(&ch->cmt->pdev->dev);
+	sh_cmt_start(ch, FLAG_CLOCKSOURCE);
 }
 
-static int sh_cmt_register_clocksource(struct sh_cmt_priv *p,
-				       char *name, unsigned long rating)
+static int sh_cmt_register_clocksource(struct sh_cmt_channel *ch,
+				       const char *name)
 {
-	struct clocksource *cs = &p->cs;
+	struct clocksource *cs = &ch->cs;
 
 	cs->name = name;
-	cs->rating = rating;
+	cs->rating = 125;
 	cs->read = sh_cmt_clocksource_read;
 	cs->enable = sh_cmt_clocksource_enable;
 	cs->disable = sh_cmt_clocksource_disable;
@@ -554,47 +688,48 @@ static int sh_cmt_register_clocksource(struct sh_cmt_priv *p,
 	cs->mask = CLOCKSOURCE_MASK(sizeof(unsigned long) * 8);
 	cs->flags = CLOCK_SOURCE_IS_CONTINUOUS;
 
-	dev_info(&p->pdev->dev, "used as clock source\n");
+	dev_info(&ch->cmt->pdev->dev, "ch%u: used as clock source\n",
+		 ch->index);
 
 	/* Register with dummy 1 Hz value, gets updated in ->enable() */
 	clocksource_register_hz(cs, 1);
 	return 0;
 }
 
-static struct sh_cmt_priv *ced_to_sh_cmt(struct clock_event_device *ced)
+static struct sh_cmt_channel *ced_to_sh_cmt(struct clock_event_device *ced)
 {
-	return container_of(ced, struct sh_cmt_priv, ced);
+	return container_of(ced, struct sh_cmt_channel, ced);
 }
 
-static void sh_cmt_clock_event_start(struct sh_cmt_priv *p, int periodic)
+static void sh_cmt_clock_event_start(struct sh_cmt_channel *ch, int periodic)
 {
-	struct clock_event_device *ced = &p->ced;
+	struct clock_event_device *ced = &ch->ced;
 
-	sh_cmt_start(p, FLAG_CLOCKEVENT);
+	sh_cmt_start(ch, FLAG_CLOCKEVENT);
 
 	/* TODO: calculate good shift from rate and counter bit width */
 
 	ced->shift = 32;
-	ced->mult = div_sc(p->rate, NSEC_PER_SEC, ced->shift);
-	ced->max_delta_ns = clockevent_delta2ns(p->max_match_value, ced);
+	ced->mult = div_sc(ch->rate, NSEC_PER_SEC, ced->shift);
+	ced->max_delta_ns = clockevent_delta2ns(ch->max_match_value, ced);
 	ced->min_delta_ns = clockevent_delta2ns(0x1f, ced);
 
 	if (periodic)
-		sh_cmt_set_next(p, ((p->rate + HZ/2) / HZ) - 1);
+		sh_cmt_set_next(ch, ((ch->rate + HZ/2) / HZ) - 1);
 	else
-		sh_cmt_set_next(p, p->max_match_value);
+		sh_cmt_set_next(ch, ch->max_match_value);
 }
 
 static void sh_cmt_clock_event_mode(enum clock_event_mode mode,
 				    struct clock_event_device *ced)
 {
-	struct sh_cmt_priv *p = ced_to_sh_cmt(ced);
+	struct sh_cmt_channel *ch = ced_to_sh_cmt(ced);
 
 	/* deal with old setting first */
 	switch (ced->mode) {
 	case CLOCK_EVT_MODE_PERIODIC:
 	case CLOCK_EVT_MODE_ONESHOT:
-		sh_cmt_stop(p, FLAG_CLOCKEVENT);
+		sh_cmt_stop(ch, FLAG_CLOCKEVENT);
 		break;
 	default:
 		break;
@@ -602,16 +737,18 @@ static void sh_cmt_clock_event_mode(enum clock_event_mode mode,
 
 	switch (mode) {
 	case CLOCK_EVT_MODE_PERIODIC:
-		dev_info(&p->pdev->dev, "used for periodic clock events\n");
-		sh_cmt_clock_event_start(p, 1);
+		dev_info(&ch->cmt->pdev->dev,
+			 "ch%u: used for periodic clock events\n", ch->index);
+		sh_cmt_clock_event_start(ch, 1);
 		break;
 	case CLOCK_EVT_MODE_ONESHOT:
-		dev_info(&p->pdev->dev, "used for oneshot clock events\n");
-		sh_cmt_clock_event_start(p, 0);
+		dev_info(&ch->cmt->pdev->dev,
+			 "ch%u: used for oneshot clock events\n", ch->index);
+		sh_cmt_clock_event_start(ch, 0);
 		break;
 	case CLOCK_EVT_MODE_SHUTDOWN:
 	case CLOCK_EVT_MODE_UNUSED:
-		sh_cmt_stop(p, FLAG_CLOCKEVENT);
+		sh_cmt_stop(ch, FLAG_CLOCKEVENT);
 		break;
 	default:
 		break;
@@ -621,196 +758,341 @@ static void sh_cmt_clock_event_mode(enum clock_event_mode mode,
 static int sh_cmt_clock_event_next(unsigned long delta,
 				   struct clock_event_device *ced)
 {
-	struct sh_cmt_priv *p = ced_to_sh_cmt(ced);
+	struct sh_cmt_channel *ch = ced_to_sh_cmt(ced);
 
 	BUG_ON(ced->mode != CLOCK_EVT_MODE_ONESHOT);
-	if (likely(p->flags & FLAG_IRQCONTEXT))
-		p->next_match_value = delta - 1;
+	if (likely(ch->flags & FLAG_IRQCONTEXT))
+		ch->next_match_value = delta - 1;
 	else
-		sh_cmt_set_next(p, delta - 1);
+		sh_cmt_set_next(ch, delta - 1);
 
 	return 0;
 }
 
 static void sh_cmt_clock_event_suspend(struct clock_event_device *ced)
 {
-	struct sh_cmt_priv *p = ced_to_sh_cmt(ced);
+	struct sh_cmt_channel *ch = ced_to_sh_cmt(ced);
 
-	pm_genpd_syscore_poweroff(&p->pdev->dev);
-	clk_unprepare(p->clk);
+	pm_genpd_syscore_poweroff(&ch->cmt->pdev->dev);
+	clk_unprepare(ch->cmt->clk);
 }
 
 static void sh_cmt_clock_event_resume(struct clock_event_device *ced)
 {
-	struct sh_cmt_priv *p = ced_to_sh_cmt(ced);
+	struct sh_cmt_channel *ch = ced_to_sh_cmt(ced);
 
-	clk_prepare(p->clk);
-	pm_genpd_syscore_poweron(&p->pdev->dev);
+	clk_prepare(ch->cmt->clk);
+	pm_genpd_syscore_poweron(&ch->cmt->pdev->dev);
 }
 
-static void sh_cmt_register_clockevent(struct sh_cmt_priv *p,
-				       char *name, unsigned long rating)
+static int sh_cmt_register_clockevent(struct sh_cmt_channel *ch,
+				      const char *name)
 {
-	struct clock_event_device *ced = &p->ced;
+	struct clock_event_device *ced = &ch->ced;
+	int irq;
+	int ret;
 
-	memset(ced, 0, sizeof(*ced));
+	irq = platform_get_irq(ch->cmt->pdev, ch->cmt->legacy ? 0 : ch->index);
+	if (irq < 0) {
+		dev_err(&ch->cmt->pdev->dev, "ch%u: failed to get irq\n",
+			ch->index);
+		return irq;
+	}
+
+	ret = request_irq(irq, sh_cmt_interrupt,
+			  IRQF_TIMER | IRQF_IRQPOLL | IRQF_NOBALANCING,
+			  dev_name(&ch->cmt->pdev->dev), ch);
+	if (ret) {
+		dev_err(&ch->cmt->pdev->dev, "ch%u: failed to request irq %d\n",
+			ch->index, irq);
+		return ret;
+	}
 
 	ced->name = name;
 	ced->features = CLOCK_EVT_FEAT_PERIODIC;
 	ced->features |= CLOCK_EVT_FEAT_ONESHOT;
-	ced->rating = rating;
-	ced->cpumask = cpumask_of(0);
+	ced->rating = 125;
+	ced->cpumask = cpu_possible_mask;
 	ced->set_next_event = sh_cmt_clock_event_next;
 	ced->set_mode = sh_cmt_clock_event_mode;
 	ced->suspend = sh_cmt_clock_event_suspend;
 	ced->resume = sh_cmt_clock_event_resume;
 
-	dev_info(&p->pdev->dev, "used for clock events\n");
+	dev_info(&ch->cmt->pdev->dev, "ch%u: used for clock events\n",
+		 ch->index);
 	clockevents_register_device(ced);
+
+	return 0;
 }
 
-static int sh_cmt_register(struct sh_cmt_priv *p, char *name,
-			   unsigned long clockevent_rating,
-			   unsigned long clocksource_rating)
+static int sh_cmt_register(struct sh_cmt_channel *ch, const char *name,
+			   bool clockevent, bool clocksource)
 {
-	if (clockevent_rating)
-		sh_cmt_register_clockevent(p, name, clockevent_rating);
+	int ret;
 
-	if (clocksource_rating)
-		sh_cmt_register_clocksource(p, name, clocksource_rating);
+	if (clockevent) {
+		ch->cmt->has_clockevent = true;
+		ret = sh_cmt_register_clockevent(ch, name);
+		if (ret < 0)
+			return ret;
+	}
+
+	if (clocksource) {
+		ch->cmt->has_clocksource = true;
+		sh_cmt_register_clocksource(ch, name);
+	}
 
 	return 0;
 }
 
-static int sh_cmt_setup(struct sh_cmt_priv *p, struct platform_device *pdev)
+static int sh_cmt_setup_channel(struct sh_cmt_channel *ch, unsigned int index,
+				unsigned int hwidx, bool clockevent,
+				bool clocksource, struct sh_cmt_device *cmt)
 {
-	struct sh_timer_config *cfg = pdev->dev.platform_data;
-	struct resource *res, *res2;
-	int irq, ret;
-	ret = -ENXIO;
+	int ret;
 
-	memset(p, 0, sizeof(*p));
-	p->pdev = pdev;
+	/* Skip unused channels. */
+	if (!clockevent && !clocksource)
+		return 0;
 
-	if (!cfg) {
-		dev_err(&p->pdev->dev, "missing platform data\n");
-		goto err0;
+	ch->cmt = cmt;
+	ch->index = index;
+	ch->hwidx = hwidx;
+
+	/*
+	 * Compute the address of the channel control register block. For the
+	 * timers with a per-channel start/stop register, compute its address
+	 * as well.
+	 *
+	 * For legacy configuration the address has been mapped explicitly.
+	 */
+	if (cmt->legacy) {
+		ch->ioctrl = cmt->mapbase_ch;
+	} else {
+		switch (cmt->info->model) {
+		case SH_CMT_16BIT:
+			ch->ioctrl = cmt->mapbase + 2 + ch->hwidx * 6;
+			break;
+		case SH_CMT_32BIT:
+		case SH_CMT_48BIT:
+			ch->ioctrl = cmt->mapbase + 0x10 + ch->hwidx * 0x10;
+			break;
+		case SH_CMT_32BIT_FAST:
+			/*
+			 * The 32-bit "fast" timer has a single channel at hwidx
+			 * 5 but is located at offset 0x40 instead of 0x60 for
+			 * some reason.
+			 */
+			ch->ioctrl = cmt->mapbase + 0x40;
+			break;
+		case SH_CMT_48BIT_GEN2:
+			ch->iostart = cmt->mapbase + ch->hwidx * 0x100;
+			ch->ioctrl = ch->iostart + 0x10;
+			break;
+		}
 	}
 
-	res = platform_get_resource(p->pdev, IORESOURCE_MEM, 0);
-	if (!res) {
-		dev_err(&p->pdev->dev, "failed to get I/O memory\n");
-		goto err0;
+	if (cmt->info->width == (sizeof(ch->max_match_value) * 8))
+		ch->max_match_value = ~0;
+	else
+		ch->max_match_value = (1 << cmt->info->width) - 1;
+
+	ch->match_value = ch->max_match_value;
+	raw_spin_lock_init(&ch->lock);
+
+	if (cmt->legacy) {
+		ch->timer_bit = ch->hwidx;
+	} else {
+		ch->timer_bit = cmt->info->model == SH_CMT_48BIT_GEN2
+			      ? 0 : ch->hwidx;
 	}
 
-	/* optional resource for the shared timer start/stop register */
-	res2 = platform_get_resource(p->pdev, IORESOURCE_MEM, 1);
+	ret = sh_cmt_register(ch, dev_name(&cmt->pdev->dev),
+			      clockevent, clocksource);
+	if (ret) {
+		dev_err(&cmt->pdev->dev, "ch%u: registration failed\n",
+			ch->index);
+		return ret;
+	}
+	ch->cs_enabled = false;
 
-	irq = platform_get_irq(p->pdev, 0);
-	if (irq < 0) {
-		dev_err(&p->pdev->dev, "failed to get irq\n");
-		goto err0;
+	return 0;
+}
+
+static int sh_cmt_map_memory(struct sh_cmt_device *cmt)
+{
+	struct resource *mem;
+
+	mem = platform_get_resource(cmt->pdev, IORESOURCE_MEM, 0);
+	if (!mem) {
+		dev_err(&cmt->pdev->dev, "failed to get I/O memory\n");
+		return -ENXIO;
 	}
 
-	/* map memory, let mapbase point to our channel */
-	p->mapbase = ioremap_nocache(res->start, resource_size(res));
-	if (p->mapbase == NULL) {
-		dev_err(&p->pdev->dev, "failed to remap I/O memory\n");
-		goto err0;
+	cmt->mapbase = ioremap_nocache(mem->start, resource_size(mem));
+	if (cmt->mapbase == NULL) {
+		dev_err(&cmt->pdev->dev, "failed to remap I/O memory\n");
+		return -ENXIO;
 	}
 
-	/* map second resource for CMSTR */
-	p->mapbase_str = ioremap_nocache(res2 ? res2->start :
-					 res->start - cfg->channel_offset,
-					 res2 ? resource_size(res2) : 2);
-	if (p->mapbase_str == NULL) {
-		dev_err(&p->pdev->dev, "failed to remap I/O second memory\n");
-		goto err1;
+	return 0;
+}
+
+static int sh_cmt_map_memory_legacy(struct sh_cmt_device *cmt)
+{
+	struct sh_timer_config *cfg = cmt->pdev->dev.platform_data;
+	struct resource *res, *res2;
+
+	/* map memory, let mapbase_ch point to our channel */
+	res = platform_get_resource(cmt->pdev, IORESOURCE_MEM, 0);
+	if (!res) {
+		dev_err(&cmt->pdev->dev, "failed to get I/O memory\n");
+		return -ENXIO;
 	}
 
-	/* request irq using setup_irq() (too early for request_irq()) */
-	p->irqaction.name = dev_name(&p->pdev->dev);
-	p->irqaction.handler = sh_cmt_interrupt;
-	p->irqaction.dev_id = p;
-	p->irqaction.flags = IRQF_TIMER | IRQF_IRQPOLL | IRQF_NOBALANCING;
-
-	/* get hold of clock */
-	p->clk = clk_get(&p->pdev->dev, "cmt_fck");
-	if (IS_ERR(p->clk)) {
-		dev_err(&p->pdev->dev, "cannot get clock\n");
-		ret = PTR_ERR(p->clk);
-		goto err2;
+	cmt->mapbase_ch = ioremap_nocache(res->start, resource_size(res));
+	if (cmt->mapbase_ch == NULL) {
+		dev_err(&cmt->pdev->dev, "failed to remap I/O memory\n");
+		return -ENXIO;
 	}
 
-	ret = clk_prepare(p->clk);
-	if (ret < 0)
-		goto err3;
+	/* optional resource for the shared timer start/stop register */
+	res2 = platform_get_resource(cmt->pdev, IORESOURCE_MEM, 1);
 
-	if (res2 && (resource_size(res2) == 4)) {
-		/* assume both CMSTR and CMCSR to be 32-bit */
-		p->read_control = sh_cmt_read32;
-		p->write_control = sh_cmt_write32;
-	} else {
-		p->read_control = sh_cmt_read16;
-		p->write_control = sh_cmt_write16;
+	/* map second resource for CMSTR */
+	cmt->mapbase = ioremap_nocache(res2 ? res2->start :
+				       res->start - cfg->channel_offset,
+				       res2 ? resource_size(res2) : 2);
+	if (cmt->mapbase == NULL) {
+		dev_err(&cmt->pdev->dev, "failed to remap I/O second memory\n");
+		iounmap(cmt->mapbase_ch);
+		return -ENXIO;
 	}
 
-	if (resource_size(res) == 6) {
-		p->width = 16;
-		p->read_count = sh_cmt_read16;
-		p->write_count = sh_cmt_write16;
-		p->overflow_bit = 0x80;
-		p->clear_bits = ~0x80;
-	} else {
-		p->width = 32;
-		p->read_count = sh_cmt_read32;
-		p->write_count = sh_cmt_write32;
-		p->overflow_bit = 0x8000;
-		p->clear_bits = ~0xc000;
+	/* identify the model based on the resources */
+	if (resource_size(res) == 6)
+		cmt->info = &sh_cmt_info[SH_CMT_16BIT];
+	else if (res2 && (resource_size(res2) == 4))
+		cmt->info = &sh_cmt_info[SH_CMT_48BIT_GEN2];
+	else
+		cmt->info = &sh_cmt_info[SH_CMT_32BIT];
+
+	return 0;
+}
+
+static void sh_cmt_unmap_memory(struct sh_cmt_device *cmt)
+{
+	iounmap(cmt->mapbase);
+	if (cmt->mapbase_ch)
+		iounmap(cmt->mapbase_ch);
+}
+
+static int sh_cmt_setup(struct sh_cmt_device *cmt, struct platform_device *pdev)
+{
+	struct sh_timer_config *cfg = pdev->dev.platform_data;
+	const struct platform_device_id *id = pdev->id_entry;
+	unsigned int hw_channels;
+	int ret;
+
+	memset(cmt, 0, sizeof(*cmt));
+	cmt->pdev = pdev;
+
+	if (!cfg) {
+		dev_err(&cmt->pdev->dev, "missing platform data\n");
+		return -ENXIO;
 	}
 
-	if (p->width == (sizeof(p->max_match_value) * 8))
-		p->max_match_value = ~0;
+	cmt->info = (const struct sh_cmt_info *)id->driver_data;
+	cmt->legacy = cmt->info ? false : true;
+
+	/* Get hold of clock. */
+	cmt->clk = clk_get(&cmt->pdev->dev, cmt->legacy ? "cmt_fck" : "fck");
+	if (IS_ERR(cmt->clk)) {
+		dev_err(&cmt->pdev->dev, "cannot get clock\n");
+		return PTR_ERR(cmt->clk);
+	}
+
+	ret = clk_prepare(cmt->clk);
+	if (ret < 0)
+		goto err_clk_put;
+
+	/*
+	 * Map the memory resource(s). We need to support both the legacy
+	 * platform device configuration (with one device per channel) and the
+	 * new version (with multiple channels per device).
+	 */
+	if (cmt->legacy)
+		ret = sh_cmt_map_memory_legacy(cmt);
 	else
-		p->max_match_value = (1 << p->width) - 1;
+		ret = sh_cmt_map_memory(cmt);
 
-	p->match_value = p->max_match_value;
-	raw_spin_lock_init(&p->lock);
+	if (ret < 0)
+		goto err_clk_unprepare;
 
-	ret = sh_cmt_register(p, (char *)dev_name(&p->pdev->dev),
-			      cfg->clockevent_rating,
-			      cfg->clocksource_rating);
-	if (ret) {
-		dev_err(&p->pdev->dev, "registration failed\n");
-		goto err4;
+	/* Allocate and setup the channels. */
+	if (cmt->legacy) {
+		cmt->num_channels = 1;
+		hw_channels = 0;
+	} else {
+		cmt->num_channels = hweight8(cfg->channels_mask);
+		hw_channels = cfg->channels_mask;
 	}
-	p->cs_enabled = false;
 
-	ret = setup_irq(irq, &p->irqaction);
-	if (ret) {
-		dev_err(&p->pdev->dev, "failed to request irq %d\n", irq);
-		goto err4;
+	cmt->channels = kzalloc(cmt->num_channels * sizeof(*cmt->channels),
+				GFP_KERNEL);
+	if (cmt->channels == NULL) {
+		ret = -ENOMEM;
+		goto err_unmap;
 	}
 
-	platform_set_drvdata(pdev, p);
+	if (cmt->legacy) {
+		ret = sh_cmt_setup_channel(&cmt->channels[0],
+					   cfg->timer_bit, cfg->timer_bit,
+					   cfg->clockevent_rating != 0,
+					   cfg->clocksource_rating != 0, cmt);
+		if (ret < 0)
+			goto err_unmap;
+	} else {
+		unsigned int mask = hw_channels;
+		unsigned int i;
+
+		/*
+		 * Use the first channel as a clock event device and the second
+		 * channel as a clock source. If only one channel is available
+		 * use it for both.
+		 */
+		for (i = 0; i < cmt->num_channels; ++i) {
+			unsigned int hwidx = ffs(mask) - 1;
+			bool clocksource = i == 1 || cmt->num_channels == 1;
+			bool clockevent = i == 0;
+
+			ret = sh_cmt_setup_channel(&cmt->channels[i], i, hwidx,
+						   clockevent, clocksource,
+						   cmt);
+			if (ret < 0)
+				goto err_unmap;
+
+			mask &= ~(1 << hwidx);
+		}
+	}
+
+	platform_set_drvdata(pdev, cmt);
 
 	return 0;
-err4:
-	clk_unprepare(p->clk);
-err3:
-	clk_put(p->clk);
-err2:
-	iounmap(p->mapbase_str);
-err1:
-	iounmap(p->mapbase);
-err0:
+
+err_unmap:
+	kfree(cmt->channels);
+	sh_cmt_unmap_memory(cmt);
+err_clk_unprepare:
+	clk_unprepare(cmt->clk);
+err_clk_put:
+	clk_put(cmt->clk);
 	return ret;
 }
 
 static int sh_cmt_probe(struct platform_device *pdev)
 {
-	struct sh_cmt_priv *p = platform_get_drvdata(pdev);
-	struct sh_timer_config *cfg = pdev->dev.platform_data;
+	struct sh_cmt_device *cmt = platform_get_drvdata(pdev);
 	int ret;
 
 	if (!is_early_platform_device(pdev)) {
@@ -818,20 +1100,18 @@ static int sh_cmt_probe(struct platform_device *pdev)
 		pm_runtime_enable(&pdev->dev);
 	}
 
-	if (p) {
+	if (cmt) {
 		dev_info(&pdev->dev, "kept as earlytimer\n");
 		goto out;
 	}
 
-	p = kmalloc(sizeof(*p), GFP_KERNEL);
-	if (p == NULL) {
-		dev_err(&pdev->dev, "failed to allocate driver data\n");
+	cmt = kzalloc(sizeof(*cmt), GFP_KERNEL);
+	if (cmt == NULL)
 		return -ENOMEM;
-	}
 
-	ret = sh_cmt_setup(p, pdev);
+	ret = sh_cmt_setup(cmt, pdev);
 	if (ret) {
-		kfree(p);
+		kfree(cmt);
 		pm_runtime_idle(&pdev->dev);
 		return ret;
 	}
@@ -839,7 +1119,7 @@ static int sh_cmt_probe(struct platform_device *pdev)
 		return 0;
 
  out:
-	if (cfg->clockevent_rating || cfg->clocksource_rating)
+	if (cmt->has_clockevent || cmt->has_clocksource)
 		pm_runtime_irq_safe(&pdev->dev);
 	else
 		pm_runtime_idle(&pdev->dev);
@@ -852,12 +1132,24 @@ static int sh_cmt_remove(struct platform_device *pdev)
 	return -EBUSY; /* cannot unregister clockevent and clocksource */
 }
 
+static const struct platform_device_id sh_cmt_id_table[] = {
+	{ "sh_cmt", 0 },
+	{ "sh-cmt-16", (kernel_ulong_t)&sh_cmt_info[SH_CMT_16BIT] },
+	{ "sh-cmt-32", (kernel_ulong_t)&sh_cmt_info[SH_CMT_32BIT] },
+	{ "sh-cmt-32-fast", (kernel_ulong_t)&sh_cmt_info[SH_CMT_32BIT_FAST] },
+	{ "sh-cmt-48", (kernel_ulong_t)&sh_cmt_info[SH_CMT_48BIT] },
+	{ "sh-cmt-48-gen2", (kernel_ulong_t)&sh_cmt_info[SH_CMT_48BIT_GEN2] },
+	{ }
+};
+MODULE_DEVICE_TABLE(platform, sh_cmt_id_table);
+
 static struct platform_driver sh_cmt_device_driver = {
 	.probe		= sh_cmt_probe,
 	.remove		= sh_cmt_remove,
 	.driver		= {
 		.name	= "sh_cmt",
-	}
+	},
+	.id_table	= sh_cmt_id_table,
 };
 
 static int __init sh_cmt_init(void)
diff --git a/drivers/clocksource/sh_mtu2.c b/drivers/clocksource/sh_mtu2.c
index e30d76e0a6fa..188d4e092efc 100644
--- a/drivers/clocksource/sh_mtu2.c
+++ b/drivers/clocksource/sh_mtu2.c
@@ -11,37 +11,48 @@
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
  */
 
+#include <linux/clk.h>
+#include <linux/clockchips.h>
+#include <linux/delay.h>
+#include <linux/err.h>
 #include <linux/init.h>
-#include <linux/platform_device.h>
-#include <linux/spinlock.h>
 #include <linux/interrupt.h>
-#include <linux/ioport.h>
-#include <linux/delay.h>
 #include <linux/io.h>
-#include <linux/clk.h>
+#include <linux/ioport.h>
 #include <linux/irq.h>
-#include <linux/err.h>
-#include <linux/clockchips.h>
-#include <linux/sh_timer.h>
-#include <linux/slab.h>
 #include <linux/module.h>
+#include <linux/platform_device.h>
 #include <linux/pm_domain.h>
 #include <linux/pm_runtime.h>
+#include <linux/sh_timer.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+
+struct sh_mtu2_device;
+
+struct sh_mtu2_channel {
+	struct sh_mtu2_device *mtu;
+	unsigned int index;
+
+	void __iomem *base;
+	int irq;
+
+	struct clock_event_device ced;
+};
+
+struct sh_mtu2_device {
+	struct platform_device *pdev;
 
-struct sh_mtu2_priv {
 	void __iomem *mapbase;
 	struct clk *clk;
-	struct irqaction irqaction;
-	struct platform_device *pdev;
-	unsigned long rate;
-	unsigned long periodic;
-	struct clock_event_device ced;
+
+	struct sh_mtu2_channel *channels;
+	unsigned int num_channels;
+
+	bool legacy;
+	bool has_clockevent;
 };
 
 static DEFINE_RAW_SPINLOCK(sh_mtu2_lock);
@@ -55,6 +66,88 @@ static DEFINE_RAW_SPINLOCK(sh_mtu2_lock);
 #define TCNT 5 /* channel register */
 #define TGR  6 /* channel register */
 
+#define TCR_CCLR_NONE		(0 << 5)
+#define TCR_CCLR_TGRA		(1 << 5)
+#define TCR_CCLR_TGRB		(2 << 5)
+#define TCR_CCLR_SYNC		(3 << 5)
+#define TCR_CCLR_TGRC		(5 << 5)
+#define TCR_CCLR_TGRD		(6 << 5)
+#define TCR_CCLR_MASK		(7 << 5)
+#define TCR_CKEG_RISING		(0 << 3)
+#define TCR_CKEG_FALLING	(1 << 3)
+#define TCR_CKEG_BOTH		(2 << 3)
+#define TCR_CKEG_MASK		(3 << 3)
+/* Values 4 to 7 are channel-dependent */
+#define TCR_TPSC_P1		(0 << 0)
+#define TCR_TPSC_P4		(1 << 0)
+#define TCR_TPSC_P16		(2 << 0)
+#define TCR_TPSC_P64		(3 << 0)
+#define TCR_TPSC_CH0_TCLKA	(4 << 0)
+#define TCR_TPSC_CH0_TCLKB	(5 << 0)
+#define TCR_TPSC_CH0_TCLKC	(6 << 0)
+#define TCR_TPSC_CH0_TCLKD	(7 << 0)
+#define TCR_TPSC_CH1_TCLKA	(4 << 0)
+#define TCR_TPSC_CH1_TCLKB	(5 << 0)
+#define TCR_TPSC_CH1_P256	(6 << 0)
+#define TCR_TPSC_CH1_TCNT2	(7 << 0)
+#define TCR_TPSC_CH2_TCLKA	(4 << 0)
+#define TCR_TPSC_CH2_TCLKB	(5 << 0)
+#define TCR_TPSC_CH2_TCLKC	(6 << 0)
+#define TCR_TPSC_CH2_P1024	(7 << 0)
+#define TCR_TPSC_CH34_P256	(4 << 0)
+#define TCR_TPSC_CH34_P1024	(5 << 0)
+#define TCR_TPSC_CH34_TCLKA	(6 << 0)
+#define TCR_TPSC_CH34_TCLKB	(7 << 0)
+#define TCR_TPSC_MASK		(7 << 0)
+
+#define TMDR_BFE		(1 << 6)
+#define TMDR_BFB		(1 << 5)
+#define TMDR_BFA		(1 << 4)
+#define TMDR_MD_NORMAL		(0 << 0)
+#define TMDR_MD_PWM_1		(2 << 0)
+#define TMDR_MD_PWM_2		(3 << 0)
+#define TMDR_MD_PHASE_1		(4 << 0)
+#define TMDR_MD_PHASE_2		(5 << 0)
+#define TMDR_MD_PHASE_3		(6 << 0)
+#define TMDR_MD_PHASE_4		(7 << 0)
+#define TMDR_MD_PWM_SYNC	(8 << 0)
+#define TMDR_MD_PWM_COMP_CREST	(13 << 0)
+#define TMDR_MD_PWM_COMP_TROUGH	(14 << 0)
+#define TMDR_MD_PWM_COMP_BOTH	(15 << 0)
+#define TMDR_MD_MASK		(15 << 0)
+
+#define TIOC_IOCH(n)		((n) << 4)
+#define TIOC_IOCL(n)		((n) << 0)
+#define TIOR_OC_RETAIN		(0 << 0)
+#define TIOR_OC_0_CLEAR		(1 << 0)
+#define TIOR_OC_0_SET		(2 << 0)
+#define TIOR_OC_0_TOGGLE	(3 << 0)
+#define TIOR_OC_1_CLEAR		(5 << 0)
+#define TIOR_OC_1_SET		(6 << 0)
+#define TIOR_OC_1_TOGGLE	(7 << 0)
+#define TIOR_IC_RISING		(8 << 0)
+#define TIOR_IC_FALLING		(9 << 0)
+#define TIOR_IC_BOTH		(10 << 0)
+#define TIOR_IC_TCNT		(12 << 0)
+#define TIOR_MASK		(15 << 0)
+
+#define TIER_TTGE		(1 << 7)
+#define TIER_TTGE2		(1 << 6)
+#define TIER_TCIEU		(1 << 5)
+#define TIER_TCIEV		(1 << 4)
+#define TIER_TGIED		(1 << 3)
+#define TIER_TGIEC		(1 << 2)
+#define TIER_TGIEB		(1 << 1)
+#define TIER_TGIEA		(1 << 0)
+
+#define TSR_TCFD		(1 << 7)
+#define TSR_TCFU		(1 << 5)
+#define TSR_TCFV		(1 << 4)
+#define TSR_TGFD		(1 << 3)
+#define TSR_TGFC		(1 << 2)
+#define TSR_TGFB		(1 << 1)
+#define TSR_TGFA		(1 << 0)
+
 static unsigned long mtu2_reg_offs[] = {
 	[TCR] = 0,
 	[TMDR] = 1,
@@ -65,135 +158,143 @@ static unsigned long mtu2_reg_offs[] = {
 	[TGR] = 8,
 };
 
-static inline unsigned long sh_mtu2_read(struct sh_mtu2_priv *p, int reg_nr)
+static inline unsigned long sh_mtu2_read(struct sh_mtu2_channel *ch, int reg_nr)
 {
-	struct sh_timer_config *cfg = p->pdev->dev.platform_data;
-	void __iomem *base = p->mapbase;
 	unsigned long offs;
 
-	if (reg_nr == TSTR)
-		return ioread8(base + cfg->channel_offset);
+	if (reg_nr == TSTR) {
+		if (ch->mtu->legacy)
+			return ioread8(ch->mtu->mapbase);
+		else
+			return ioread8(ch->mtu->mapbase + 0x280);
+	}
 
 	offs = mtu2_reg_offs[reg_nr];
 
 	if ((reg_nr == TCNT) || (reg_nr == TGR))
-		return ioread16(base + offs);
+		return ioread16(ch->base + offs);
 	else
-		return ioread8(base + offs);
+		return ioread8(ch->base + offs);
 }
 
-static inline void sh_mtu2_write(struct sh_mtu2_priv *p, int reg_nr,
+static inline void sh_mtu2_write(struct sh_mtu2_channel *ch, int reg_nr,
 				unsigned long value)
 {
-	struct sh_timer_config *cfg = p->pdev->dev.platform_data;
-	void __iomem *base = p->mapbase;
 	unsigned long offs;
 
 	if (reg_nr == TSTR) {
-		iowrite8(value, base + cfg->channel_offset);
-		return;
+		if (ch->mtu->legacy)
+			return iowrite8(value, ch->mtu->mapbase);
+		else
+			return iowrite8(value, ch->mtu->mapbase + 0x280);
 	}
 
 	offs = mtu2_reg_offs[reg_nr];
 
 	if ((reg_nr == TCNT) || (reg_nr == TGR))
-		iowrite16(value, base + offs);
+		iowrite16(value, ch->base + offs);
 	else
-		iowrite8(value, base + offs);
+		iowrite8(value, ch->base + offs);
 }
 
-static void sh_mtu2_start_stop_ch(struct sh_mtu2_priv *p, int start)
+static void sh_mtu2_start_stop_ch(struct sh_mtu2_channel *ch, int start)
 {
-	struct sh_timer_config *cfg = p->pdev->dev.platform_data;
 	unsigned long flags, value;
 
 	/* start stop register shared by multiple timer channels */
 	raw_spin_lock_irqsave(&sh_mtu2_lock, flags);
-	value = sh_mtu2_read(p, TSTR);
+	value = sh_mtu2_read(ch, TSTR);
 
 	if (start)
-		value |= 1 << cfg->timer_bit;
+		value |= 1 << ch->index;
 	else
-		value &= ~(1 << cfg->timer_bit);
+		value &= ~(1 << ch->index);
 
-	sh_mtu2_write(p, TSTR, value);
+	sh_mtu2_write(ch, TSTR, value);
 	raw_spin_unlock_irqrestore(&sh_mtu2_lock, flags);
 }
 
-static int sh_mtu2_enable(struct sh_mtu2_priv *p)
+static int sh_mtu2_enable(struct sh_mtu2_channel *ch)
 {
+	unsigned long periodic;
+	unsigned long rate;
 	int ret;
 
-	pm_runtime_get_sync(&p->pdev->dev);
-	dev_pm_syscore_device(&p->pdev->dev, true);
+	pm_runtime_get_sync(&ch->mtu->pdev->dev);
+	dev_pm_syscore_device(&ch->mtu->pdev->dev, true);
 
 	/* enable clock */
-	ret = clk_enable(p->clk);
+	ret = clk_enable(ch->mtu->clk);
 	if (ret) {
-		dev_err(&p->pdev->dev, "cannot enable clock\n");
+		dev_err(&ch->mtu->pdev->dev, "ch%u: cannot enable clock\n",
+			ch->index);
 		return ret;
 	}
 
 	/* make sure channel is disabled */
-	sh_mtu2_start_stop_ch(p, 0);
-
-	p->rate = clk_get_rate(p->clk) / 64;
-	p->periodic = (p->rate + HZ/2) / HZ;
-
-	/* "Periodic Counter Operation" */
-	sh_mtu2_write(p, TCR, 0x23); /* TGRA clear, divide clock by 64 */
-	sh_mtu2_write(p, TIOR, 0);
-	sh_mtu2_write(p, TGR, p->periodic);
-	sh_mtu2_write(p, TCNT, 0);
-	sh_mtu2_write(p, TMDR, 0);
-	sh_mtu2_write(p, TIER, 0x01);
+	sh_mtu2_start_stop_ch(ch, 0);
+
+	rate = clk_get_rate(ch->mtu->clk) / 64;
+	periodic = (rate + HZ/2) / HZ;
+
+	/*
+	 * "Periodic Counter Operation"
+	 * Clear on TGRA compare match, divide clock by 64.
+	 */
+	sh_mtu2_write(ch, TCR, TCR_CCLR_TGRA | TCR_TPSC_P64);
+	sh_mtu2_write(ch, TIOR, TIOC_IOCH(TIOR_OC_0_CLEAR) |
+		      TIOC_IOCL(TIOR_OC_0_CLEAR));
+	sh_mtu2_write(ch, TGR, periodic);
+	sh_mtu2_write(ch, TCNT, 0);
+	sh_mtu2_write(ch, TMDR, TMDR_MD_NORMAL);
+	sh_mtu2_write(ch, TIER, TIER_TGIEA);
 
 	/* enable channel */
-	sh_mtu2_start_stop_ch(p, 1);
+	sh_mtu2_start_stop_ch(ch, 1);
 
 	return 0;
 }
 
-static void sh_mtu2_disable(struct sh_mtu2_priv *p)
+static void sh_mtu2_disable(struct sh_mtu2_channel *ch)
 {
 	/* disable channel */
-	sh_mtu2_start_stop_ch(p, 0);
+	sh_mtu2_start_stop_ch(ch, 0);
 
 	/* stop clock */
-	clk_disable(p->clk);
+	clk_disable(ch->mtu->clk);
 
-	dev_pm_syscore_device(&p->pdev->dev, false);
-	pm_runtime_put(&p->pdev->dev);
+	dev_pm_syscore_device(&ch->mtu->pdev->dev, false);
+	pm_runtime_put(&ch->mtu->pdev->dev);
 }
 
 static irqreturn_t sh_mtu2_interrupt(int irq, void *dev_id)
 {
-	struct sh_mtu2_priv *p = dev_id;
+	struct sh_mtu2_channel *ch = dev_id;
 
 	/* acknowledge interrupt */
-	sh_mtu2_read(p, TSR);
-	sh_mtu2_write(p, TSR, 0xfe);
+	sh_mtu2_read(ch, TSR);
+	sh_mtu2_write(ch, TSR, ~TSR_TGFA);
 
 	/* notify clockevent layer */
-	p->ced.event_handler(&p->ced);
+	ch->ced.event_handler(&ch->ced);
 	return IRQ_HANDLED;
 }
 
-static struct sh_mtu2_priv *ced_to_sh_mtu2(struct clock_event_device *ced)
+static struct sh_mtu2_channel *ced_to_sh_mtu2(struct clock_event_device *ced)
 {
-	return container_of(ced, struct sh_mtu2_priv, ced);
+	return container_of(ced, struct sh_mtu2_channel, ced);
 }
 
 static void sh_mtu2_clock_event_mode(enum clock_event_mode mode,
 				    struct clock_event_device *ced)
 {
-	struct sh_mtu2_priv *p = ced_to_sh_mtu2(ced);
+	struct sh_mtu2_channel *ch = ced_to_sh_mtu2(ced);
 	int disabled = 0;
 
 	/* deal with old setting first */
 	switch (ced->mode) {
 	case CLOCK_EVT_MODE_PERIODIC:
-		sh_mtu2_disable(p);
+		sh_mtu2_disable(ch);
 		disabled = 1;
 		break;
 	default:
@@ -202,12 +303,13 @@ static void sh_mtu2_clock_event_mode(enum clock_event_mode mode,
 
 	switch (mode) {
 	case CLOCK_EVT_MODE_PERIODIC:
-		dev_info(&p->pdev->dev, "used for periodic clock events\n");
-		sh_mtu2_enable(p);
+		dev_info(&ch->mtu->pdev->dev,
+			 "ch%u: used for periodic clock events\n", ch->index);
+		sh_mtu2_enable(ch);
 		break;
 	case CLOCK_EVT_MODE_UNUSED:
 		if (!disabled)
-			sh_mtu2_disable(p);
+			sh_mtu2_disable(ch);
 		break;
 	case CLOCK_EVT_MODE_SHUTDOWN:
 	default:
@@ -217,125 +319,207 @@ static void sh_mtu2_clock_event_mode(enum clock_event_mode mode,
 
 static void sh_mtu2_clock_event_suspend(struct clock_event_device *ced)
 {
-	pm_genpd_syscore_poweroff(&ced_to_sh_mtu2(ced)->pdev->dev);
+	pm_genpd_syscore_poweroff(&ced_to_sh_mtu2(ced)->mtu->pdev->dev);
 }
 
 static void sh_mtu2_clock_event_resume(struct clock_event_device *ced)
 {
-	pm_genpd_syscore_poweron(&ced_to_sh_mtu2(ced)->pdev->dev);
+	pm_genpd_syscore_poweron(&ced_to_sh_mtu2(ced)->mtu->pdev->dev);
 }
 
-static void sh_mtu2_register_clockevent(struct sh_mtu2_priv *p,
-				       char *name, unsigned long rating)
+static void sh_mtu2_register_clockevent(struct sh_mtu2_channel *ch,
+					const char *name)
 {
-	struct clock_event_device *ced = &p->ced;
+	struct clock_event_device *ced = &ch->ced;
 	int ret;
 
-	memset(ced, 0, sizeof(*ced));
-
 	ced->name = name;
 	ced->features = CLOCK_EVT_FEAT_PERIODIC;
-	ced->rating = rating;
-	ced->cpumask = cpumask_of(0);
+	ced->rating = 200;
+	ced->cpumask = cpu_possible_mask;
 	ced->set_mode = sh_mtu2_clock_event_mode;
 	ced->suspend = sh_mtu2_clock_event_suspend;
 	ced->resume = sh_mtu2_clock_event_resume;
 
-	dev_info(&p->pdev->dev, "used for clock events\n");
+	dev_info(&ch->mtu->pdev->dev, "ch%u: used for clock events\n",
+		 ch->index);
 	clockevents_register_device(ced);
 
-	ret = setup_irq(p->irqaction.irq, &p->irqaction);
+	ret = request_irq(ch->irq, sh_mtu2_interrupt,
+			  IRQF_TIMER | IRQF_IRQPOLL | IRQF_NOBALANCING,
+			  dev_name(&ch->mtu->pdev->dev), ch);
 	if (ret) {
-		dev_err(&p->pdev->dev, "failed to request irq %d\n",
-			p->irqaction.irq);
+		dev_err(&ch->mtu->pdev->dev, "ch%u: failed to request irq %d\n",
+			ch->index, ch->irq);
 		return;
 	}
 }
 
-static int sh_mtu2_register(struct sh_mtu2_priv *p, char *name,
-			    unsigned long clockevent_rating)
+static int sh_mtu2_register(struct sh_mtu2_channel *ch, const char *name,
+			    bool clockevent)
 {
-	if (clockevent_rating)
-		sh_mtu2_register_clockevent(p, name, clockevent_rating);
+	if (clockevent) {
+		ch->mtu->has_clockevent = true;
+		sh_mtu2_register_clockevent(ch, name);
+	}
 
 	return 0;
 }
 
-static int sh_mtu2_setup(struct sh_mtu2_priv *p, struct platform_device *pdev)
+static int sh_mtu2_setup_channel(struct sh_mtu2_channel *ch, unsigned int index,
+				 struct sh_mtu2_device *mtu)
 {
-	struct sh_timer_config *cfg = pdev->dev.platform_data;
-	struct resource *res;
-	int irq, ret;
-	ret = -ENXIO;
+	static const unsigned int channel_offsets[] = {
+		0x300, 0x380, 0x000,
+	};
+	bool clockevent;
+
+	ch->mtu = mtu;
+
+	if (mtu->legacy) {
+		struct sh_timer_config *cfg = mtu->pdev->dev.platform_data;
+
+		clockevent = cfg->clockevent_rating != 0;
 
-	memset(p, 0, sizeof(*p));
-	p->pdev = pdev;
+		ch->irq = platform_get_irq(mtu->pdev, 0);
+		ch->base = mtu->mapbase - cfg->channel_offset;
+		ch->index = cfg->timer_bit;
+	} else {
+		char name[6];
 
-	if (!cfg) {
-		dev_err(&p->pdev->dev, "missing platform data\n");
-		goto err0;
+		clockevent = true;
+
+		sprintf(name, "tgi%ua", index);
+		ch->irq = platform_get_irq_byname(mtu->pdev, name);
+		ch->base = mtu->mapbase + channel_offsets[index];
+		ch->index = index;
 	}
 
-	platform_set_drvdata(pdev, p);
+	if (ch->irq < 0) {
+		/* Skip channels with no declared interrupt. */
+		if (!mtu->legacy)
+			return 0;
+
+		dev_err(&mtu->pdev->dev, "ch%u: failed to get irq\n",
+			ch->index);
+		return ch->irq;
+	}
+
+	return sh_mtu2_register(ch, dev_name(&mtu->pdev->dev), clockevent);
+}
 
-	res = platform_get_resource(p->pdev, IORESOURCE_MEM, 0);
+static int sh_mtu2_map_memory(struct sh_mtu2_device *mtu)
+{
+	struct resource *res;
+
+	res = platform_get_resource(mtu->pdev, IORESOURCE_MEM, 0);
 	if (!res) {
-		dev_err(&p->pdev->dev, "failed to get I/O memory\n");
-		goto err0;
+		dev_err(&mtu->pdev->dev, "failed to get I/O memory\n");
+		return -ENXIO;
+	}
+
+	mtu->mapbase = ioremap_nocache(res->start, resource_size(res));
+	if (mtu->mapbase == NULL)
+		return -ENXIO;
+
+	/*
+	 * In legacy platform device configuration (with one device per channel)
+	 * the resource points to the channel base address.
+	 */
+	if (mtu->legacy) {
+		struct sh_timer_config *cfg = mtu->pdev->dev.platform_data;
+		mtu->mapbase += cfg->channel_offset;
 	}
 
-	irq = platform_get_irq(p->pdev, 0);
-	if (irq < 0) {
-		dev_err(&p->pdev->dev, "failed to get irq\n");
-		goto err0;
+	return 0;
+}
+
+static void sh_mtu2_unmap_memory(struct sh_mtu2_device *mtu)
+{
+	if (mtu->legacy) {
+		struct sh_timer_config *cfg = mtu->pdev->dev.platform_data;
+		mtu->mapbase -= cfg->channel_offset;
 	}
 
-	/* map memory, let mapbase point to our channel */
-	p->mapbase = ioremap_nocache(res->start, resource_size(res));
-	if (p->mapbase == NULL) {
-		dev_err(&p->pdev->dev, "failed to remap I/O memory\n");
-		goto err0;
+	iounmap(mtu->mapbase);
+}
+
+static int sh_mtu2_setup(struct sh_mtu2_device *mtu,
+			 struct platform_device *pdev)
+{
+	struct sh_timer_config *cfg = pdev->dev.platform_data;
+	const struct platform_device_id *id = pdev->id_entry;
+	unsigned int i;
+	int ret;
+
+	mtu->pdev = pdev;
+	mtu->legacy = id->driver_data;
+
+	if (mtu->legacy && !cfg) {
+		dev_err(&mtu->pdev->dev, "missing platform data\n");
+		return -ENXIO;
 	}
 
-	/* setup data for setup_irq() (too early for request_irq()) */
-	p->irqaction.name = dev_name(&p->pdev->dev);
-	p->irqaction.handler = sh_mtu2_interrupt;
-	p->irqaction.dev_id = p;
-	p->irqaction.irq = irq;
-	p->irqaction.flags = IRQF_TIMER | IRQF_IRQPOLL | IRQF_NOBALANCING;
-
-	/* get hold of clock */
-	p->clk = clk_get(&p->pdev->dev, "mtu2_fck");
-	if (IS_ERR(p->clk)) {
-		dev_err(&p->pdev->dev, "cannot get clock\n");
-		ret = PTR_ERR(p->clk);
-		goto err1;
+	/* Get hold of clock. */
+	mtu->clk = clk_get(&mtu->pdev->dev, mtu->legacy ? "mtu2_fck" : "fck");
+	if (IS_ERR(mtu->clk)) {
+		dev_err(&mtu->pdev->dev, "cannot get clock\n");
+		return PTR_ERR(mtu->clk);
 	}
 
-	ret = clk_prepare(p->clk);
+	ret = clk_prepare(mtu->clk);
 	if (ret < 0)
-		goto err2;
+		goto err_clk_put;
 
-	ret = sh_mtu2_register(p, (char *)dev_name(&p->pdev->dev),
-			       cfg->clockevent_rating);
-	if (ret < 0)
-		goto err3;
+	/* Map the memory resource. */
+	ret = sh_mtu2_map_memory(mtu);
+	if (ret < 0) {
+		dev_err(&mtu->pdev->dev, "failed to remap I/O memory\n");
+		goto err_clk_unprepare;
+	}
+
+	/* Allocate and setup the channels. */
+	if (mtu->legacy)
+		mtu->num_channels = 1;
+	else
+		mtu->num_channels = 3;
+
+	mtu->channels = kzalloc(sizeof(*mtu->channels) * mtu->num_channels,
+				GFP_KERNEL);
+	if (mtu->channels == NULL) {
+		ret = -ENOMEM;
+		goto err_unmap;
+	}
+
+	if (mtu->legacy) {
+		ret = sh_mtu2_setup_channel(&mtu->channels[0], 0, mtu);
+		if (ret < 0)
+			goto err_unmap;
+	} else {
+		for (i = 0; i < mtu->num_channels; ++i) {
+			ret = sh_mtu2_setup_channel(&mtu->channels[i], i, mtu);
+			if (ret < 0)
+				goto err_unmap;
+		}
+	}
+
+	platform_set_drvdata(pdev, mtu);
 
 	return 0;
- err3:
-	clk_unprepare(p->clk);
- err2:
-	clk_put(p->clk);
- err1:
-	iounmap(p->mapbase);
- err0:
+
+err_unmap:
+	kfree(mtu->channels);
+	sh_mtu2_unmap_memory(mtu);
+err_clk_unprepare:
+	clk_unprepare(mtu->clk);
+err_clk_put:
+	clk_put(mtu->clk);
 	return ret;
 }
 
 static int sh_mtu2_probe(struct platform_device *pdev)
 {
-	struct sh_mtu2_priv *p = platform_get_drvdata(pdev);
-	struct sh_timer_config *cfg = pdev->dev.platform_data;
+	struct sh_mtu2_device *mtu = platform_get_drvdata(pdev);
 	int ret;
 
 	if (!is_early_platform_device(pdev)) {
@@ -343,20 +527,18 @@ static int sh_mtu2_probe(struct platform_device *pdev)
 		pm_runtime_enable(&pdev->dev);
 	}
 
-	if (p) {
+	if (mtu) {
 		dev_info(&pdev->dev, "kept as earlytimer\n");
 		goto out;
 	}
 
-	p = kmalloc(sizeof(*p), GFP_KERNEL);
-	if (p == NULL) {
-		dev_err(&pdev->dev, "failed to allocate driver data\n");
+	mtu = kzalloc(sizeof(*mtu), GFP_KERNEL);
+	if (mtu == NULL)
 		return -ENOMEM;
-	}
 
-	ret = sh_mtu2_setup(p, pdev);
+	ret = sh_mtu2_setup(mtu, pdev);
 	if (ret) {
-		kfree(p);
+		kfree(mtu);
 		pm_runtime_idle(&pdev->dev);
 		return ret;
 	}
@@ -364,7 +546,7 @@ static int sh_mtu2_probe(struct platform_device *pdev)
 		return 0;
 
  out:
-	if (cfg->clockevent_rating)
+	if (mtu->has_clockevent)
 		pm_runtime_irq_safe(&pdev->dev);
 	else
 		pm_runtime_idle(&pdev->dev);
@@ -377,12 +559,20 @@ static int sh_mtu2_remove(struct platform_device *pdev)
 	return -EBUSY; /* cannot unregister clockevent */
 }
 
+static const struct platform_device_id sh_mtu2_id_table[] = {
+	{ "sh_mtu2", 1 },
+	{ "sh-mtu2", 0 },
+	{ },
+};
+MODULE_DEVICE_TABLE(platform, sh_mtu2_id_table);
+
 static struct platform_driver sh_mtu2_device_driver = {
 	.probe		= sh_mtu2_probe,
 	.remove		= sh_mtu2_remove,
 	.driver		= {
 		.name	= "sh_mtu2",
-	}
+	},
+	.id_table	= sh_mtu2_id_table,
 };
 
 static int __init sh_mtu2_init(void)
diff --git a/drivers/clocksource/sh_tmu.c b/drivers/clocksource/sh_tmu.c
index ecd7b60bfdfa..6bd17a8f3dd4 100644
--- a/drivers/clocksource/sh_tmu.c
+++ b/drivers/clocksource/sh_tmu.c
@@ -11,35 +11,41 @@
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
  */
 
+#include <linux/clk.h>
+#include <linux/clockchips.h>
+#include <linux/clocksource.h>
+#include <linux/delay.h>
+#include <linux/err.h>
 #include <linux/init.h>
-#include <linux/platform_device.h>
-#include <linux/spinlock.h>
 #include <linux/interrupt.h>
-#include <linux/ioport.h>
-#include <linux/delay.h>
 #include <linux/io.h>
-#include <linux/clk.h>
+#include <linux/ioport.h>
 #include <linux/irq.h>
-#include <linux/err.h>
-#include <linux/clocksource.h>
-#include <linux/clockchips.h>
-#include <linux/sh_timer.h>
-#include <linux/slab.h>
 #include <linux/module.h>
+#include <linux/platform_device.h>
 #include <linux/pm_domain.h>
 #include <linux/pm_runtime.h>
+#include <linux/sh_timer.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+
+enum sh_tmu_model {
+	SH_TMU_LEGACY,
+	SH_TMU,
+	SH_TMU_SH3,
+};
+
+struct sh_tmu_device;
+
+struct sh_tmu_channel {
+	struct sh_tmu_device *tmu;
+	unsigned int index;
+
+	void __iomem *base;
+	int irq;
 
-struct sh_tmu_priv {
-	void __iomem *mapbase;
-	struct clk *clk;
-	struct irqaction irqaction;
-	struct platform_device *pdev;
 	unsigned long rate;
 	unsigned long periodic;
 	struct clock_event_device ced;
@@ -48,6 +54,21 @@ struct sh_tmu_priv {
 	unsigned int enable_count;
 };
 
+struct sh_tmu_device {
+	struct platform_device *pdev;
+
+	void __iomem *mapbase;
+	struct clk *clk;
+
+	enum sh_tmu_model model;
+
+	struct sh_tmu_channel *channels;
+	unsigned int num_channels;
+
+	bool has_clockevent;
+	bool has_clocksource;
+};
+
 static DEFINE_RAW_SPINLOCK(sh_tmu_lock);
 
 #define TSTR -1 /* shared register */
@@ -55,189 +76,208 @@ static DEFINE_RAW_SPINLOCK(sh_tmu_lock);
 #define TCNT 1 /* channel register */
 #define TCR 2 /* channel register */
 
-static inline unsigned long sh_tmu_read(struct sh_tmu_priv *p, int reg_nr)
+#define TCR_UNF			(1 << 8)
+#define TCR_UNIE		(1 << 5)
+#define TCR_TPSC_CLK4		(0 << 0)
+#define TCR_TPSC_CLK16		(1 << 0)
+#define TCR_TPSC_CLK64		(2 << 0)
+#define TCR_TPSC_CLK256		(3 << 0)
+#define TCR_TPSC_CLK1024	(4 << 0)
+#define TCR_TPSC_MASK		(7 << 0)
+
+static inline unsigned long sh_tmu_read(struct sh_tmu_channel *ch, int reg_nr)
 {
-	struct sh_timer_config *cfg = p->pdev->dev.platform_data;
-	void __iomem *base = p->mapbase;
 	unsigned long offs;
 
-	if (reg_nr == TSTR)
-		return ioread8(base - cfg->channel_offset);
+	if (reg_nr == TSTR) {
+		switch (ch->tmu->model) {
+		case SH_TMU_LEGACY:
+			return ioread8(ch->tmu->mapbase);
+		case SH_TMU_SH3:
+			return ioread8(ch->tmu->mapbase + 2);
+		case SH_TMU:
+			return ioread8(ch->tmu->mapbase + 4);
+		}
+	}
 
 	offs = reg_nr << 2;
 
 	if (reg_nr == TCR)
-		return ioread16(base + offs);
+		return ioread16(ch->base + offs);
 	else
-		return ioread32(base + offs);
+		return ioread32(ch->base + offs);
 }
 
-static inline void sh_tmu_write(struct sh_tmu_priv *p, int reg_nr,
+static inline void sh_tmu_write(struct sh_tmu_channel *ch, int reg_nr,
 				unsigned long value)
 {
-	struct sh_timer_config *cfg = p->pdev->dev.platform_data;
-	void __iomem *base = p->mapbase;
 	unsigned long offs;
 
 	if (reg_nr == TSTR) {
-		iowrite8(value, base - cfg->channel_offset);
-		return;
+		switch (ch->tmu->model) {
+		case SH_TMU_LEGACY:
+			return iowrite8(value, ch->tmu->mapbase);
+		case SH_TMU_SH3:
+			return iowrite8(value, ch->tmu->mapbase + 2);
+		case SH_TMU:
+			return iowrite8(value, ch->tmu->mapbase + 4);
+		}
 	}
 
 	offs = reg_nr << 2;
 
 	if (reg_nr == TCR)
-		iowrite16(value, base + offs);
+		iowrite16(value, ch->base + offs);
 	else
-		iowrite32(value, base + offs);
+		iowrite32(value, ch->base + offs);
 }
 
-static void sh_tmu_start_stop_ch(struct sh_tmu_priv *p, int start)
+static void sh_tmu_start_stop_ch(struct sh_tmu_channel *ch, int start)
 {
-	struct sh_timer_config *cfg = p->pdev->dev.platform_data;
 	unsigned long flags, value;
 
 	/* start stop register shared by multiple timer channels */
 	raw_spin_lock_irqsave(&sh_tmu_lock, flags);
-	value = sh_tmu_read(p, TSTR);
+	value = sh_tmu_read(ch, TSTR);
 
 	if (start)
-		value |= 1 << cfg->timer_bit;
+		value |= 1 << ch->index;
 	else
-		value &= ~(1 << cfg->timer_bit);
+		value &= ~(1 << ch->index);
 
-	sh_tmu_write(p, TSTR, value);
+	sh_tmu_write(ch, TSTR, value);
 	raw_spin_unlock_irqrestore(&sh_tmu_lock, flags);
 }
 
-static int __sh_tmu_enable(struct sh_tmu_priv *p)
+static int __sh_tmu_enable(struct sh_tmu_channel *ch)
 {
 	int ret;
 
 	/* enable clock */
-	ret = clk_enable(p->clk);
+	ret = clk_enable(ch->tmu->clk);
 	if (ret) {
-		dev_err(&p->pdev->dev, "cannot enable clock\n");
+		dev_err(&ch->tmu->pdev->dev, "ch%u: cannot enable clock\n",
+			ch->index);
 		return ret;
 	}
 
 	/* make sure channel is disabled */
-	sh_tmu_start_stop_ch(p, 0);
+	sh_tmu_start_stop_ch(ch, 0);
 
 	/* maximum timeout */
-	sh_tmu_write(p, TCOR, 0xffffffff);
-	sh_tmu_write(p, TCNT, 0xffffffff);
+	sh_tmu_write(ch, TCOR, 0xffffffff);
+	sh_tmu_write(ch, TCNT, 0xffffffff);
 
 	/* configure channel to parent clock / 4, irq off */
-	p->rate = clk_get_rate(p->clk) / 4;
-	sh_tmu_write(p, TCR, 0x0000);
+	ch->rate = clk_get_rate(ch->tmu->clk) / 4;
+	sh_tmu_write(ch, TCR, TCR_TPSC_CLK4);
 
 	/* enable channel */
-	sh_tmu_start_stop_ch(p, 1);
+	sh_tmu_start_stop_ch(ch, 1);
 
 	return 0;
 }
 
-static int sh_tmu_enable(struct sh_tmu_priv *p)
+static int sh_tmu_enable(struct sh_tmu_channel *ch)
 {
-	if (p->enable_count++ > 0)
+	if (ch->enable_count++ > 0)
 		return 0;
 
-	pm_runtime_get_sync(&p->pdev->dev);
-	dev_pm_syscore_device(&p->pdev->dev, true);
+	pm_runtime_get_sync(&ch->tmu->pdev->dev);
+	dev_pm_syscore_device(&ch->tmu->pdev->dev, true);
 
-	return __sh_tmu_enable(p);
+	return __sh_tmu_enable(ch);
 }
 
-static void __sh_tmu_disable(struct sh_tmu_priv *p)
+static void __sh_tmu_disable(struct sh_tmu_channel *ch)
 {
 	/* disable channel */
-	sh_tmu_start_stop_ch(p, 0);
+	sh_tmu_start_stop_ch(ch, 0);
 
 	/* disable interrupts in TMU block */
-	sh_tmu_write(p, TCR, 0x0000);
+	sh_tmu_write(ch, TCR, TCR_TPSC_CLK4);
 
 	/* stop clock */
-	clk_disable(p->clk);
+	clk_disable(ch->tmu->clk);
 }
 
-static void sh_tmu_disable(struct sh_tmu_priv *p)
+static void sh_tmu_disable(struct sh_tmu_channel *ch)
 {
-	if (WARN_ON(p->enable_count == 0))
+	if (WARN_ON(ch->enable_count == 0))
 		return;
 
-	if (--p->enable_count > 0)
+	if (--ch->enable_count > 0)
 		return;
 
-	__sh_tmu_disable(p);
+	__sh_tmu_disable(ch);
 
-	dev_pm_syscore_device(&p->pdev->dev, false);
-	pm_runtime_put(&p->pdev->dev);
+	dev_pm_syscore_device(&ch->tmu->pdev->dev, false);
+	pm_runtime_put(&ch->tmu->pdev->dev);
 }
 
-static void sh_tmu_set_next(struct sh_tmu_priv *p, unsigned long delta,
+static void sh_tmu_set_next(struct sh_tmu_channel *ch, unsigned long delta,
 			    int periodic)
 {
 	/* stop timer */
-	sh_tmu_start_stop_ch(p, 0);
+	sh_tmu_start_stop_ch(ch, 0);
 
 	/* acknowledge interrupt */
-	sh_tmu_read(p, TCR);
+	sh_tmu_read(ch, TCR);
 
 	/* enable interrupt */
-	sh_tmu_write(p, TCR, 0x0020);
+	sh_tmu_write(ch, TCR, TCR_UNIE | TCR_TPSC_CLK4);
 
 	/* reload delta value in case of periodic timer */
 	if (periodic)
-		sh_tmu_write(p, TCOR, delta);
+		sh_tmu_write(ch, TCOR, delta);
 	else
-		sh_tmu_write(p, TCOR, 0xffffffff);
+		sh_tmu_write(ch, TCOR, 0xffffffff);
 
-	sh_tmu_write(p, TCNT, delta);
+	sh_tmu_write(ch, TCNT, delta);
 
 	/* start timer */
-	sh_tmu_start_stop_ch(p, 1);
+	sh_tmu_start_stop_ch(ch, 1);
 }
 
 static irqreturn_t sh_tmu_interrupt(int irq, void *dev_id)
 {
-	struct sh_tmu_priv *p = dev_id;
+	struct sh_tmu_channel *ch = dev_id;
 
 	/* disable or acknowledge interrupt */
-	if (p->ced.mode == CLOCK_EVT_MODE_ONESHOT)
-		sh_tmu_write(p, TCR, 0x0000);
+	if (ch->ced.mode == CLOCK_EVT_MODE_ONESHOT)
+		sh_tmu_write(ch, TCR, TCR_TPSC_CLK4);
 	else
-		sh_tmu_write(p, TCR, 0x0020);
+		sh_tmu_write(ch, TCR, TCR_UNIE | TCR_TPSC_CLK4);
 
 	/* notify clockevent layer */
-	p->ced.event_handler(&p->ced);
+	ch->ced.event_handler(&ch->ced);
 	return IRQ_HANDLED;
 }
 
-static struct sh_tmu_priv *cs_to_sh_tmu(struct clocksource *cs)
+static struct sh_tmu_channel *cs_to_sh_tmu(struct clocksource *cs)
 {
-	return container_of(cs, struct sh_tmu_priv, cs);
+	return container_of(cs, struct sh_tmu_channel, cs);
 }
 
 static cycle_t sh_tmu_clocksource_read(struct clocksource *cs)
 {
-	struct sh_tmu_priv *p = cs_to_sh_tmu(cs);
+	struct sh_tmu_channel *ch = cs_to_sh_tmu(cs);
 
-	return sh_tmu_read(p, TCNT) ^ 0xffffffff;
+	return sh_tmu_read(ch, TCNT) ^ 0xffffffff;
 }
 
 static int sh_tmu_clocksource_enable(struct clocksource *cs)
 {
-	struct sh_tmu_priv *p = cs_to_sh_tmu(cs);
+	struct sh_tmu_channel *ch = cs_to_sh_tmu(cs);
 	int ret;
 
-	if (WARN_ON(p->cs_enabled))
+	if (WARN_ON(ch->cs_enabled))
 		return 0;
 
-	ret = sh_tmu_enable(p);
+	ret = sh_tmu_enable(ch);
 	if (!ret) {
-		__clocksource_updatefreq_hz(cs, p->rate);
-		p->cs_enabled = true;
+		__clocksource_updatefreq_hz(cs, ch->rate);
+		ch->cs_enabled = true;
 	}
 
 	return ret;
@@ -245,48 +285,48 @@ static int sh_tmu_clocksource_enable(struct clocksource *cs)
 
 static void sh_tmu_clocksource_disable(struct clocksource *cs)
 {
-	struct sh_tmu_priv *p = cs_to_sh_tmu(cs);
+	struct sh_tmu_channel *ch = cs_to_sh_tmu(cs);
 
-	if (WARN_ON(!p->cs_enabled))
+	if (WARN_ON(!ch->cs_enabled))
 		return;
 
-	sh_tmu_disable(p);
-	p->cs_enabled = false;
+	sh_tmu_disable(ch);
+	ch->cs_enabled = false;
 }
 
 static void sh_tmu_clocksource_suspend(struct clocksource *cs)
 {
-	struct sh_tmu_priv *p = cs_to_sh_tmu(cs);
+	struct sh_tmu_channel *ch = cs_to_sh_tmu(cs);
 
-	if (!p->cs_enabled)
+	if (!ch->cs_enabled)
 		return;
 
-	if (--p->enable_count == 0) {
-		__sh_tmu_disable(p);
-		pm_genpd_syscore_poweroff(&p->pdev->dev);
+	if (--ch->enable_count == 0) {
+		__sh_tmu_disable(ch);
+		pm_genpd_syscore_poweroff(&ch->tmu->pdev->dev);
 	}
 }
 
 static void sh_tmu_clocksource_resume(struct clocksource *cs)
 {
-	struct sh_tmu_priv *p = cs_to_sh_tmu(cs);
+	struct sh_tmu_channel *ch = cs_to_sh_tmu(cs);
 
-	if (!p->cs_enabled)
+	if (!ch->cs_enabled)
 		return;
 
-	if (p->enable_count++ == 0) {
-		pm_genpd_syscore_poweron(&p->pdev->dev);
-		__sh_tmu_enable(p);
+	if (ch->enable_count++ == 0) {
+		pm_genpd_syscore_poweron(&ch->tmu->pdev->dev);
+		__sh_tmu_enable(ch);
 	}
 }
 
-static int sh_tmu_register_clocksource(struct sh_tmu_priv *p,
-				       char *name, unsigned long rating)
+static int sh_tmu_register_clocksource(struct sh_tmu_channel *ch,
+				       const char *name)
 {
-	struct clocksource *cs = &p->cs;
+	struct clocksource *cs = &ch->cs;
 
 	cs->name = name;
-	cs->rating = rating;
+	cs->rating = 200;
 	cs->read = sh_tmu_clocksource_read;
 	cs->enable = sh_tmu_clocksource_enable;
 	cs->disable = sh_tmu_clocksource_disable;
@@ -295,43 +335,44 @@ static int sh_tmu_register_clocksource(struct sh_tmu_priv *p,
 	cs->mask = CLOCKSOURCE_MASK(32);
 	cs->flags = CLOCK_SOURCE_IS_CONTINUOUS;
 
-	dev_info(&p->pdev->dev, "used as clock source\n");
+	dev_info(&ch->tmu->pdev->dev, "ch%u: used as clock source\n",
+		 ch->index);
 
 	/* Register with dummy 1 Hz value, gets updated in ->enable() */
 	clocksource_register_hz(cs, 1);
 	return 0;
 }
 
-static struct sh_tmu_priv *ced_to_sh_tmu(struct clock_event_device *ced)
+static struct sh_tmu_channel *ced_to_sh_tmu(struct clock_event_device *ced)
 {
-	return container_of(ced, struct sh_tmu_priv, ced);
+	return container_of(ced, struct sh_tmu_channel, ced);
 }
 
-static void sh_tmu_clock_event_start(struct sh_tmu_priv *p, int periodic)
+static void sh_tmu_clock_event_start(struct sh_tmu_channel *ch, int periodic)
 {
-	struct clock_event_device *ced = &p->ced;
+	struct clock_event_device *ced = &ch->ced;
 
-	sh_tmu_enable(p);
+	sh_tmu_enable(ch);
 
-	clockevents_config(ced, p->rate);
+	clockevents_config(ced, ch->rate);
 
 	if (periodic) {
-		p->periodic = (p->rate + HZ/2) / HZ;
-		sh_tmu_set_next(p, p->periodic, 1);
+		ch->periodic = (ch->rate + HZ/2) / HZ;
+		sh_tmu_set_next(ch, ch->periodic, 1);
 	}
 }
 
 static void sh_tmu_clock_event_mode(enum clock_event_mode mode,
 				    struct clock_event_device *ced)
 {
-	struct sh_tmu_priv *p = ced_to_sh_tmu(ced);
+	struct sh_tmu_channel *ch = ced_to_sh_tmu(ced);
 	int disabled = 0;
 
 	/* deal with old setting first */
 	switch (ced->mode) {
 	case CLOCK_EVT_MODE_PERIODIC:
 	case CLOCK_EVT_MODE_ONESHOT:
-		sh_tmu_disable(p);
+		sh_tmu_disable(ch);
 		disabled = 1;
 		break;
 	default:
@@ -340,16 +381,18 @@ static void sh_tmu_clock_event_mode(enum clock_event_mode mode,
 
 	switch (mode) {
 	case CLOCK_EVT_MODE_PERIODIC:
-		dev_info(&p->pdev->dev, "used for periodic clock events\n");
-		sh_tmu_clock_event_start(p, 1);
+		dev_info(&ch->tmu->pdev->dev,
+			 "ch%u: used for periodic clock events\n", ch->index);
+		sh_tmu_clock_event_start(ch, 1);
 		break;
 	case CLOCK_EVT_MODE_ONESHOT:
-		dev_info(&p->pdev->dev, "used for oneshot clock events\n");
-		sh_tmu_clock_event_start(p, 0);
+		dev_info(&ch->tmu->pdev->dev,
+			 "ch%u: used for oneshot clock events\n", ch->index);
+		sh_tmu_clock_event_start(ch, 0);
 		break;
 	case CLOCK_EVT_MODE_UNUSED:
 		if (!disabled)
-			sh_tmu_disable(p);
+			sh_tmu_disable(ch);
 		break;
 	case CLOCK_EVT_MODE_SHUTDOWN:
 	default:
@@ -360,147 +403,234 @@ static void sh_tmu_clock_event_mode(enum clock_event_mode mode,
 static int sh_tmu_clock_event_next(unsigned long delta,
 				   struct clock_event_device *ced)
 {
-	struct sh_tmu_priv *p = ced_to_sh_tmu(ced);
+	struct sh_tmu_channel *ch = ced_to_sh_tmu(ced);
 
 	BUG_ON(ced->mode != CLOCK_EVT_MODE_ONESHOT);
 
 	/* program new delta value */
-	sh_tmu_set_next(p, delta, 0);
+	sh_tmu_set_next(ch, delta, 0);
 	return 0;
 }
 
 static void sh_tmu_clock_event_suspend(struct clock_event_device *ced)
 {
-	pm_genpd_syscore_poweroff(&ced_to_sh_tmu(ced)->pdev->dev);
+	pm_genpd_syscore_poweroff(&ced_to_sh_tmu(ced)->tmu->pdev->dev);
 }
 
 static void sh_tmu_clock_event_resume(struct clock_event_device *ced)
 {
-	pm_genpd_syscore_poweron(&ced_to_sh_tmu(ced)->pdev->dev);
+	pm_genpd_syscore_poweron(&ced_to_sh_tmu(ced)->tmu->pdev->dev);
 }
 
-static void sh_tmu_register_clockevent(struct sh_tmu_priv *p,
-				       char *name, unsigned long rating)
+static void sh_tmu_register_clockevent(struct sh_tmu_channel *ch,
+				       const char *name)
 {
-	struct clock_event_device *ced = &p->ced;
+	struct clock_event_device *ced = &ch->ced;
 	int ret;
 
-	memset(ced, 0, sizeof(*ced));
-
 	ced->name = name;
 	ced->features = CLOCK_EVT_FEAT_PERIODIC;
 	ced->features |= CLOCK_EVT_FEAT_ONESHOT;
-	ced->rating = rating;
+	ced->rating = 200;
 	ced->cpumask = cpumask_of(0);
 	ced->set_next_event = sh_tmu_clock_event_next;
 	ced->set_mode = sh_tmu_clock_event_mode;
 	ced->suspend = sh_tmu_clock_event_suspend;
 	ced->resume = sh_tmu_clock_event_resume;
 
-	dev_info(&p->pdev->dev, "used for clock events\n");
+	dev_info(&ch->tmu->pdev->dev, "ch%u: used for clock events\n",
+		 ch->index);
 
 	clockevents_config_and_register(ced, 1, 0x300, 0xffffffff);
 
-	ret = setup_irq(p->irqaction.irq, &p->irqaction);
+	ret = request_irq(ch->irq, sh_tmu_interrupt,
+			  IRQF_TIMER | IRQF_IRQPOLL | IRQF_NOBALANCING,
+			  dev_name(&ch->tmu->pdev->dev), ch);
 	if (ret) {
-		dev_err(&p->pdev->dev, "failed to request irq %d\n",
-			p->irqaction.irq);
+		dev_err(&ch->tmu->pdev->dev, "ch%u: failed to request irq %d\n",
+			ch->index, ch->irq);
 		return;
 	}
 }
 
-static int sh_tmu_register(struct sh_tmu_priv *p, char *name,
-		    unsigned long clockevent_rating,
-		    unsigned long clocksource_rating)
+static int sh_tmu_register(struct sh_tmu_channel *ch, const char *name,
+			   bool clockevent, bool clocksource)
 {
-	if (clockevent_rating)
-		sh_tmu_register_clockevent(p, name, clockevent_rating);
-	else if (clocksource_rating)
-		sh_tmu_register_clocksource(p, name, clocksource_rating);
+	if (clockevent) {
+		ch->tmu->has_clockevent = true;
+		sh_tmu_register_clockevent(ch, name);
+	} else if (clocksource) {
+		ch->tmu->has_clocksource = true;
+		sh_tmu_register_clocksource(ch, name);
+	}
 
 	return 0;
 }
 
-static int sh_tmu_setup(struct sh_tmu_priv *p, struct platform_device *pdev)
+static int sh_tmu_channel_setup(struct sh_tmu_channel *ch, unsigned int index,
+				bool clockevent, bool clocksource,
+				struct sh_tmu_device *tmu)
 {
-	struct sh_timer_config *cfg = pdev->dev.platform_data;
-	struct resource *res;
-	int irq, ret;
-	ret = -ENXIO;
+	/* Skip unused channels. */
+	if (!clockevent && !clocksource)
+		return 0;
 
-	memset(p, 0, sizeof(*p));
-	p->pdev = pdev;
+	ch->tmu = tmu;
+
+	if (tmu->model == SH_TMU_LEGACY) {
+		struct sh_timer_config *cfg = tmu->pdev->dev.platform_data;
+
+		/*
+		 * The SH3 variant (SH770x, SH7705, SH7710 and SH7720) maps
+		 * channel registers blocks at base + 2 + 12 * index, while all
+		 * other variants map them at base + 4 + 12 * index. We can
+		 * compute the index by just dividing by 12, the 2 bytes or 4
+		 * bytes offset being hidden by the integer division.
+		 */
+		ch->index = cfg->channel_offset / 12;
+		ch->base = tmu->mapbase + cfg->channel_offset;
+	} else {
+		ch->index = index;
+
+		if (tmu->model == SH_TMU_SH3)
+			ch->base = tmu->mapbase + 4 + ch->index * 12;
+		else
+			ch->base = tmu->mapbase + 8 + ch->index * 12;
+	}
 
-	if (!cfg) {
-		dev_err(&p->pdev->dev, "missing platform data\n");
-		goto err0;
+	ch->irq = platform_get_irq(tmu->pdev, index);
+	if (ch->irq < 0) {
+		dev_err(&tmu->pdev->dev, "ch%u: failed to get irq\n",
+			ch->index);
+		return ch->irq;
 	}
 
-	platform_set_drvdata(pdev, p);
+	ch->cs_enabled = false;
+	ch->enable_count = 0;
+
+	return sh_tmu_register(ch, dev_name(&tmu->pdev->dev),
+			       clockevent, clocksource);
+}
+
+static int sh_tmu_map_memory(struct sh_tmu_device *tmu)
+{
+	struct resource *res;
 
-	res = platform_get_resource(p->pdev, IORESOURCE_MEM, 0);
+	res = platform_get_resource(tmu->pdev, IORESOURCE_MEM, 0);
 	if (!res) {
-		dev_err(&p->pdev->dev, "failed to get I/O memory\n");
-		goto err0;
+		dev_err(&tmu->pdev->dev, "failed to get I/O memory\n");
+		return -ENXIO;
+	}
+
+	tmu->mapbase = ioremap_nocache(res->start, resource_size(res));
+	if (tmu->mapbase == NULL)
+		return -ENXIO;
+
+	/*
+	 * In legacy platform device configuration (with one device per channel)
+	 * the resource points to the channel base address.
+	 */
+	if (tmu->model == SH_TMU_LEGACY) {
+		struct sh_timer_config *cfg = tmu->pdev->dev.platform_data;
+		tmu->mapbase -= cfg->channel_offset;
 	}
 
-	irq = platform_get_irq(p->pdev, 0);
-	if (irq < 0) {
-		dev_err(&p->pdev->dev, "failed to get irq\n");
-		goto err0;
+	return 0;
+}
+
+static void sh_tmu_unmap_memory(struct sh_tmu_device *tmu)
+{
+	if (tmu->model == SH_TMU_LEGACY) {
+		struct sh_timer_config *cfg = tmu->pdev->dev.platform_data;
+		tmu->mapbase += cfg->channel_offset;
 	}
 
-	/* map memory, let mapbase point to our channel */
-	p->mapbase = ioremap_nocache(res->start, resource_size(res));
-	if (p->mapbase == NULL) {
-		dev_err(&p->pdev->dev, "failed to remap I/O memory\n");
-		goto err0;
+	iounmap(tmu->mapbase);
+}
+
+static int sh_tmu_setup(struct sh_tmu_device *tmu, struct platform_device *pdev)
+{
+	struct sh_timer_config *cfg = pdev->dev.platform_data;
+	const struct platform_device_id *id = pdev->id_entry;
+	unsigned int i;
+	int ret;
+
+	if (!cfg) {
+		dev_err(&tmu->pdev->dev, "missing platform data\n");
+		return -ENXIO;
 	}
 
-	/* setup data for setup_irq() (too early for request_irq()) */
-	p->irqaction.name = dev_name(&p->pdev->dev);
-	p->irqaction.handler = sh_tmu_interrupt;
-	p->irqaction.dev_id = p;
-	p->irqaction.irq = irq;
-	p->irqaction.flags = IRQF_TIMER | IRQF_IRQPOLL | IRQF_NOBALANCING;
-
-	/* get hold of clock */
-	p->clk = clk_get(&p->pdev->dev, "tmu_fck");
-	if (IS_ERR(p->clk)) {
-		dev_err(&p->pdev->dev, "cannot get clock\n");
-		ret = PTR_ERR(p->clk);
-		goto err1;
+	tmu->pdev = pdev;
+	tmu->model = id->driver_data;
+
+	/* Get hold of clock. */
+	tmu->clk = clk_get(&tmu->pdev->dev,
+			   tmu->model == SH_TMU_LEGACY ? "tmu_fck" : "fck");
+	if (IS_ERR(tmu->clk)) {
+		dev_err(&tmu->pdev->dev, "cannot get clock\n");
+		return PTR_ERR(tmu->clk);
 	}
 
-	ret = clk_prepare(p->clk);
+	ret = clk_prepare(tmu->clk);
 	if (ret < 0)
-		goto err2;
+		goto err_clk_put;
 
-	p->cs_enabled = false;
-	p->enable_count = 0;
+	/* Map the memory resource. */
+	ret = sh_tmu_map_memory(tmu);
+	if (ret < 0) {
+		dev_err(&tmu->pdev->dev, "failed to remap I/O memory\n");
+		goto err_clk_unprepare;
+	}
 
-	ret = sh_tmu_register(p, (char *)dev_name(&p->pdev->dev),
-			      cfg->clockevent_rating,
-			      cfg->clocksource_rating);
-	if (ret < 0)
-		goto err3;
+	/* Allocate and setup the channels. */
+	if (tmu->model == SH_TMU_LEGACY)
+		tmu->num_channels = 1;
+	else
+		tmu->num_channels = hweight8(cfg->channels_mask);
+
+	tmu->channels = kzalloc(sizeof(*tmu->channels) * tmu->num_channels,
+				GFP_KERNEL);
+	if (tmu->channels == NULL) {
+		ret = -ENOMEM;
+		goto err_unmap;
+	}
+
+	if (tmu->model == SH_TMU_LEGACY) {
+		ret = sh_tmu_channel_setup(&tmu->channels[0], 0,
+					   cfg->clockevent_rating != 0,
+					   cfg->clocksource_rating != 0, tmu);
+		if (ret < 0)
+			goto err_unmap;
+	} else {
+		/*
+		 * Use the first channel as a clock event device and the second
+		 * channel as a clock source.
+		 */
+		for (i = 0; i < tmu->num_channels; ++i) {
+			ret = sh_tmu_channel_setup(&tmu->channels[i], i,
+						   i == 0, i == 1, tmu);
+			if (ret < 0)
+				goto err_unmap;
+		}
+	}
+
+	platform_set_drvdata(pdev, tmu);
 
 	return 0;
 
- err3:
-	clk_unprepare(p->clk);
- err2:
-	clk_put(p->clk);
- err1:
-	iounmap(p->mapbase);
- err0:
+err_unmap:
+	kfree(tmu->channels);
+	sh_tmu_unmap_memory(tmu);
+err_clk_unprepare:
+	clk_unprepare(tmu->clk);
+err_clk_put:
+	clk_put(tmu->clk);
 	return ret;
 }
 
 static int sh_tmu_probe(struct platform_device *pdev)
 {
-	struct sh_tmu_priv *p = platform_get_drvdata(pdev);
-	struct sh_timer_config *cfg = pdev->dev.platform_data;
+	struct sh_tmu_device *tmu = platform_get_drvdata(pdev);
 	int ret;
 
 	if (!is_early_platform_device(pdev)) {
@@ -508,20 +638,18 @@ static int sh_tmu_probe(struct platform_device *pdev)
 		pm_runtime_enable(&pdev->dev);
 	}
 
-	if (p) {
+	if (tmu) {
 		dev_info(&pdev->dev, "kept as earlytimer\n");
 		goto out;
 	}
 
-	p = kmalloc(sizeof(*p), GFP_KERNEL);
-	if (p == NULL) {
-		dev_err(&pdev->dev, "failed to allocate driver data\n");
+	tmu = kzalloc(sizeof(*tmu), GFP_KERNEL);
+	if (tmu == NULL)
 		return -ENOMEM;
-	}
 
-	ret = sh_tmu_setup(p, pdev);
+	ret = sh_tmu_setup(tmu, pdev);
 	if (ret) {
-		kfree(p);
+		kfree(tmu);
 		pm_runtime_idle(&pdev->dev);
 		return ret;
 	}
@@ -529,7 +657,7 @@ static int sh_tmu_probe(struct platform_device *pdev)
 		return 0;
 
  out:
-	if (cfg->clockevent_rating || cfg->clocksource_rating)
+	if (tmu->has_clockevent || tmu->has_clocksource)
 		pm_runtime_irq_safe(&pdev->dev);
 	else
 		pm_runtime_idle(&pdev->dev);
@@ -542,12 +670,21 @@ static int sh_tmu_remove(struct platform_device *pdev)
 	return -EBUSY; /* cannot unregister clockevent and clocksource */
 }
 
+static const struct platform_device_id sh_tmu_id_table[] = {
+	{ "sh_tmu", SH_TMU_LEGACY },
+	{ "sh-tmu", SH_TMU },
+	{ "sh-tmu-sh3", SH_TMU_SH3 },
+	{ }
+};
+MODULE_DEVICE_TABLE(platform, sh_tmu_id_table);
+
 static struct platform_driver sh_tmu_device_driver = {
 	.probe		= sh_tmu_probe,
 	.remove		= sh_tmu_remove,
 	.driver		= {
 		.name	= "sh_tmu",
-	}
+	},
+	.id_table	= sh_tmu_id_table,
 };
 
 static int __init sh_tmu_init(void)
diff --git a/drivers/clocksource/time-efm32.c b/drivers/clocksource/time-efm32.c
index 1a6205b7bed3..bba62f9deefb 100644
--- a/drivers/clocksource/time-efm32.c
+++ b/drivers/clocksource/time-efm32.c
@@ -272,4 +272,5 @@ static void __init efm32_timer_init(struct device_node *np)
 		}
 	}
 }
-CLOCKSOURCE_OF_DECLARE(efm32, "efm32,timer", efm32_timer_init);
+CLOCKSOURCE_OF_DECLARE(efm32compat, "efm32,timer", efm32_timer_init);
+CLOCKSOURCE_OF_DECLARE(efm32, "energymicro,efm32-timer", efm32_timer_init);
diff --git a/drivers/clocksource/timer-marco.c b/drivers/clocksource/timer-marco.c
index 7f5374dbefd9..dbd30398222a 100644
--- a/drivers/clocksource/timer-marco.c
+++ b/drivers/clocksource/timer-marco.c
@@ -252,15 +252,13 @@ static void __init sirfsoc_clockevent_init(void)
 }
 
 /* initialize the kernel jiffy timer source */
-static void __init sirfsoc_marco_timer_init(void)
+static void __init sirfsoc_marco_timer_init(struct device_node *np)
 {
 	unsigned long rate;
 	u32 timer_div;
 	struct clk *clk;
 
-	/* timer's input clock is io clock */
-	clk = clk_get_sys("io", NULL);
-
+	clk = of_clk_get(np, 0);
 	BUG_ON(IS_ERR(clk));
 	rate = clk_get_rate(clk);
 
@@ -303,6 +301,6 @@ static void __init sirfsoc_of_timer_init(struct device_node *np)
 	if (!sirfsoc_timer1_irq.irq)
 		panic("No irq passed for timer1 via DT\n");
 
-	sirfsoc_marco_timer_init();
+	sirfsoc_marco_timer_init(np);
 }
 CLOCKSOURCE_OF_DECLARE(sirfsoc_marco_timer, "sirf,marco-tick", sirfsoc_of_timer_init );
diff --git a/drivers/clocksource/timer-prima2.c b/drivers/clocksource/timer-prima2.c
index 1a6b2d6356d6..a722aac7ac02 100644
--- a/drivers/clocksource/timer-prima2.c
+++ b/drivers/clocksource/timer-prima2.c
@@ -61,7 +61,8 @@ static irqreturn_t sirfsoc_timer_interrupt(int irq, void *dev_id)
 {
 	struct clock_event_device *ce = dev_id;
 
-	WARN_ON(!(readl_relaxed(sirfsoc_timer_base + SIRFSOC_TIMER_STATUS) & BIT(0)));
+	WARN_ON(!(readl_relaxed(sirfsoc_timer_base + SIRFSOC_TIMER_STATUS) &
+		BIT(0)));
 
 	/* clear timer0 interrupt */
 	writel_relaxed(BIT(0), sirfsoc_timer_base + SIRFSOC_TIMER_STATUS);
@@ -77,9 +78,11 @@ static cycle_t sirfsoc_timer_read(struct clocksource *cs)
 	u64 cycles;
 
 	/* latch the 64-bit timer counter */
-	writel_relaxed(SIRFSOC_TIMER_LATCH_BIT, sirfsoc_timer_base + SIRFSOC_TIMER_LATCH);
+	writel_relaxed(SIRFSOC_TIMER_LATCH_BIT,
+		sirfsoc_timer_base + SIRFSOC_TIMER_LATCH);
 	cycles = readl_relaxed(sirfsoc_timer_base + SIRFSOC_TIMER_LATCHED_HI);
-	cycles = (cycles << 32) | readl_relaxed(sirfsoc_timer_base + SIRFSOC_TIMER_LATCHED_LO);
+	cycles = (cycles << 32) |
+		readl_relaxed(sirfsoc_timer_base + SIRFSOC_TIMER_LATCHED_LO);
 
 	return cycles;
 }
@@ -89,11 +92,13 @@ static int sirfsoc_timer_set_next_event(unsigned long delta,
 {
 	unsigned long now, next;
 
-	writel_relaxed(SIRFSOC_TIMER_LATCH_BIT, sirfsoc_timer_base + SIRFSOC_TIMER_LATCH);
+	writel_relaxed(SIRFSOC_TIMER_LATCH_BIT,
+		sirfsoc_timer_base + SIRFSOC_TIMER_LATCH);
 	now = readl_relaxed(sirfsoc_timer_base + SIRFSOC_TIMER_LATCHED_LO);
 	next = now + delta;
 	writel_relaxed(next, sirfsoc_timer_base + SIRFSOC_TIMER_MATCH_0);
-	writel_relaxed(SIRFSOC_TIMER_LATCH_BIT, sirfsoc_timer_base + SIRFSOC_TIMER_LATCH);
+	writel_relaxed(SIRFSOC_TIMER_LATCH_BIT,
+		sirfsoc_timer_base + SIRFSOC_TIMER_LATCH);
 	now = readl_relaxed(sirfsoc_timer_base + SIRFSOC_TIMER_LATCHED_LO);
 
 	return next - now > delta ? -ETIME : 0;
@@ -108,10 +113,12 @@ static void sirfsoc_timer_set_mode(enum clock_event_mode mode,
 		WARN_ON(1);
 		break;
 	case CLOCK_EVT_MODE_ONESHOT:
-		writel_relaxed(val | BIT(0), sirfsoc_timer_base + SIRFSOC_TIMER_INT_EN);
+		writel_relaxed(val | BIT(0),
+			sirfsoc_timer_base + SIRFSOC_TIMER_INT_EN);
 		break;
 	case CLOCK_EVT_MODE_SHUTDOWN:
-		writel_relaxed(val & ~BIT(0), sirfsoc_timer_base + SIRFSOC_TIMER_INT_EN);
+		writel_relaxed(val & ~BIT(0),
+			sirfsoc_timer_base + SIRFSOC_TIMER_INT_EN);
 		break;
 	case CLOCK_EVT_MODE_UNUSED:
 	case CLOCK_EVT_MODE_RESUME:
@@ -123,10 +130,13 @@ static void sirfsoc_clocksource_suspend(struct clocksource *cs)
 {
 	int i;
 
-	writel_relaxed(SIRFSOC_TIMER_LATCH_BIT, sirfsoc_timer_base + SIRFSOC_TIMER_LATCH);
+	writel_relaxed(SIRFSOC_TIMER_LATCH_BIT,
+		sirfsoc_timer_base + SIRFSOC_TIMER_LATCH);
 
 	for (i = 0; i < SIRFSOC_TIMER_REG_CNT; i++)
-		sirfsoc_timer_reg_val[i] = readl_relaxed(sirfsoc_timer_base + sirfsoc_timer_reg_list[i]);
+		sirfsoc_timer_reg_val[i] =
+			readl_relaxed(sirfsoc_timer_base +
+				sirfsoc_timer_reg_list[i]);
 }
 
 static void sirfsoc_clocksource_resume(struct clocksource *cs)
@@ -134,10 +144,13 @@ static void sirfsoc_clocksource_resume(struct clocksource *cs)
 	int i;
 
 	for (i = 0; i < SIRFSOC_TIMER_REG_CNT - 2; i++)
-		writel_relaxed(sirfsoc_timer_reg_val[i], sirfsoc_timer_base + sirfsoc_timer_reg_list[i]);
+		writel_relaxed(sirfsoc_timer_reg_val[i],
+			sirfsoc_timer_base + sirfsoc_timer_reg_list[i]);
 
-	writel_relaxed(sirfsoc_timer_reg_val[SIRFSOC_TIMER_REG_CNT - 2], sirfsoc_timer_base + SIRFSOC_TIMER_COUNTER_LO);
-	writel_relaxed(sirfsoc_timer_reg_val[SIRFSOC_TIMER_REG_CNT - 1], sirfsoc_timer_base + SIRFSOC_TIMER_COUNTER_HI);
+	writel_relaxed(sirfsoc_timer_reg_val[SIRFSOC_TIMER_REG_CNT - 2],
+		sirfsoc_timer_base + SIRFSOC_TIMER_COUNTER_LO);
+	writel_relaxed(sirfsoc_timer_reg_val[SIRFSOC_TIMER_REG_CNT - 1],
+		sirfsoc_timer_base + SIRFSOC_TIMER_COUNTER_HI);
 }
 
 static struct clock_event_device sirfsoc_clockevent = {
@@ -185,11 +198,8 @@ static void __init sirfsoc_prima2_timer_init(struct device_node *np)
 	unsigned long rate;
 	struct clk *clk;
 
-	/* timer's input clock is io clock */
-	clk = clk_get_sys("io", NULL);
-
+	clk = of_clk_get(np, 0);
 	BUG_ON(IS_ERR(clk));
-
 	rate = clk_get_rate(clk);
 
 	BUG_ON(rate < PRIMA2_CLOCK_FREQ);
@@ -202,7 +212,7 @@ static void __init sirfsoc_prima2_timer_init(struct device_node *np)
 	sirfsoc_timer_irq.irq = irq_of_parse_and_map(np, 0);
 
 	writel_relaxed(rate / PRIMA2_CLOCK_FREQ / 2 - 1,
-		       sirfsoc_timer_base + SIRFSOC_TIMER_DIV);
+		sirfsoc_timer_base + SIRFSOC_TIMER_DIV);
 	writel_relaxed(0, sirfsoc_timer_base + SIRFSOC_TIMER_COUNTER_LO);
 	writel_relaxed(0, sirfsoc_timer_base + SIRFSOC_TIMER_COUNTER_HI);
 	writel_relaxed(BIT(0), sirfsoc_timer_base + SIRFSOC_TIMER_STATUS);
@@ -216,4 +226,5 @@ static void __init sirfsoc_prima2_timer_init(struct device_node *np)
 
 	sirfsoc_clockevent_init();
 }
-CLOCKSOURCE_OF_DECLARE(sirfsoc_prima2_timer, "sirf,prima2-tick", sirfsoc_prima2_timer_init);
+CLOCKSOURCE_OF_DECLARE(sirfsoc_prima2_timer,
+	"sirf,prima2-tick", sirfsoc_prima2_timer_init);
diff --git a/drivers/clocksource/timer-sun5i.c b/drivers/clocksource/timer-sun5i.c
index deebcd6469fc..02268448dc85 100644
--- a/drivers/clocksource/timer-sun5i.c
+++ b/drivers/clocksource/timer-sun5i.c
@@ -16,6 +16,7 @@
 #include <linux/interrupt.h>
 #include <linux/irq.h>
 #include <linux/irqreturn.h>
+#include <linux/reset.h>
 #include <linux/sched_clock.h>
 #include <linux/of.h>
 #include <linux/of_address.h>
@@ -143,6 +144,7 @@ static u64 sun5i_timer_sched_read(void)
 
 static void __init sun5i_timer_init(struct device_node *node)
 {
+	struct reset_control *rstc;
 	unsigned long rate;
 	struct clk *clk;
 	int ret, irq;
@@ -162,6 +164,10 @@ static void __init sun5i_timer_init(struct device_node *node)
 	clk_prepare_enable(clk);
 	rate = clk_get_rate(clk);
 
+	rstc = of_reset_control_get(node, NULL);
+	if (!IS_ERR(rstc))
+		reset_control_deassert(rstc);
+
 	writel(~0, timer_base + TIMER_INTVAL_LO_REG(1));
 	writel(TIMER_CTL_ENABLE | TIMER_CTL_RELOAD,
 	       timer_base + TIMER_CTL_REG(1));
diff --git a/drivers/clocksource/versatile.c b/drivers/clocksource/versatile.c
new file mode 100644
index 000000000000..2798e7492234
--- /dev/null
+++ b/drivers/clocksource/versatile.c
@@ -0,0 +1,40 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * Copyright (C) 2014 ARM Limited
+ */
+
+#include <linux/clocksource.h>
+#include <linux/io.h>
+#include <linux/of_address.h>
+#include <linux/sched_clock.h>
+
+#define SYS_24MHZ 0x05c
+
+static void __iomem *versatile_sys_24mhz;
+
+static u64 notrace versatile_sys_24mhz_read(void)
+{
+	return readl(versatile_sys_24mhz);
+}
+
+static void __init versatile_sched_clock_init(struct device_node *node)
+{
+	void __iomem *base = of_iomap(node, 0);
+
+	if (!base)
+		return;
+
+	versatile_sys_24mhz = base + SYS_24MHZ;
+
+	sched_clock_register(versatile_sys_24mhz_read, 32, 24000000);
+}
+CLOCKSOURCE_OF_DECLARE(versatile, "arm,vexpress-sysreg",
+		       versatile_sched_clock_init);