Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux into rmobile-latest

19 files changed, 455 insertions, 296 deletions

diff --git a/arch/arm/mach-shmobile/Makefile.boot b/arch/arm/mach-shmobile/Makefile.boot
index 1c08ee9..498efd9 100644
--- a/arch/arm/mach-shmobile/Makefile.boot
+++ b/arch/arm/mach-shmobile/Makefile.boot
@@ -1,7 +1,7 @@
 __ZRELADDR	:= $(shell /bin/bash -c 'printf "0x%08x" \
 		     $$[$(CONFIG_MEMORY_START) + 0x8000]')
 
-   zreladdr-y   := $(__ZRELADDR)
+   zreladdr-y   += $(__ZRELADDR)
 
 # Unsupported legacy stuff
 #
diff --git a/arch/arm/mach-shmobile/board-ag5evm.c b/arch/arm/mach-shmobile/board-ag5evm.c
index cdfdd62..475342b 100644
--- a/arch/arm/mach-shmobile/board-ag5evm.c
+++ b/arch/arm/mach-shmobile/board-ag5evm.c
@@ -37,6 +37,7 @@
 #include <linux/mmc/sh_mobile_sdhi.h>
 #include <linux/mfd/tmio.h>
 #include <linux/sh_clk.h>
+#include <linux/dma-mapping.h>
 #include <video/sh_mobile_lcdc.h>
 #include <video/sh_mipi_dsi.h>
 #include <sound/sh_fsi.h>
@@ -354,14 +355,17 @@ static struct resource sdhi0_resources[] = {
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
+		.name	= SH_MOBILE_SDHI_IRQ_CARD_DETECT,
 		.start	= gic_spi(83),
 		.flags	= IORESOURCE_IRQ,
 	},
 	[2] = {
+		.name	= SH_MOBILE_SDHI_IRQ_SDCARD,
 		.start	= gic_spi(84),
 		.flags	= IORESOURCE_IRQ,
 	},
 	[3] = {
+		.name	= SH_MOBILE_SDHI_IRQ_SDIO,
 		.start	= gic_spi(85),
 		.flags	= IORESOURCE_IRQ,
 	},
@@ -397,14 +401,17 @@ static struct resource sdhi1_resources[] = {
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
+		.name	= SH_MOBILE_SDHI_IRQ_CARD_DETECT,
 		.start	= gic_spi(87),
 		.flags	= IORESOURCE_IRQ,
 	},
 	[2] = {
+		.name	= SH_MOBILE_SDHI_IRQ_SDCARD,
 		.start	= gic_spi(88),
 		.flags	= IORESOURCE_IRQ,
 	},
 	[3] = {
+		.name	= SH_MOBILE_SDHI_IRQ_SDIO,
 		.start	= gic_spi(89),
 		.flags	= IORESOURCE_IRQ,
 	},
@@ -447,6 +454,8 @@ static struct map_desc ag5evm_io_desc[] __initdata = {
 static void __init ag5evm_map_io(void)
 {
 	iotable_init(ag5evm_io_desc, ARRAY_SIZE(ag5evm_io_desc));
+	/* DMA memory at 0xf6000000 - 0xffdfffff */
+	init_consistent_dma_size(158 << 20);
 
 	/* setup early devices and console here as well */
 	sh73a0_add_early_devices();
diff --git a/arch/arm/mach-shmobile/board-ap4evb.c b/arch/arm/mach-shmobile/board-ap4evb.c
index 523f608..5b7edad 100644
--- a/arch/arm/mach-shmobile/board-ap4evb.c
+++ b/arch/arm/mach-shmobile/board-ap4evb.c
@@ -42,6 +42,8 @@
 #include <linux/leds.h>
 #include <linux/input/sh_keysc.h>
 #include <linux/usb/r8a66597.h>
+#include <linux/pm_clock.h>
+#include <linux/dma-mapping.h>
 
 #include <media/sh_mobile_ceu.h>
 #include <media/sh_mobile_csi2.h>
@@ -1170,6 +1172,8 @@ static struct map_desc ap4evb_io_desc[] __initdata = {
 static void __init ap4evb_map_io(void)
 {
 	iotable_init(ap4evb_io_desc, ARRAY_SIZE(ap4evb_io_desc));
+	/* DMA memory at 0xf6000000 - 0xffdfffff */
+	init_consistent_dma_size(158 << 20);
 
 	/* setup early devices and console here as well */
 	sh7372_add_early_devices();
@@ -1408,6 +1412,11 @@ static void __init ap4evb_init(void)
 	sh7372_add_device_to_domain(&sh7372_a4lc, &lcdc_device);
 	sh7372_add_device_to_domain(&sh7372_a4mp, &fsi_device);
 
+	sh7372_add_device_to_domain(&sh7372_a3sp, &sh_mmcif_device);
+	sh7372_add_device_to_domain(&sh7372_a3sp, &sdhi0_device);
+	sh7372_add_device_to_domain(&sh7372_a3sp, &sdhi1_device);
+	sh7372_add_device_to_domain(&sh7372_a4r, &ceu_device);
+
 	hdmi_init_pm_clock();
 	fsi_init_pm_clock();
 	sh7372_pm_init();
diff --git a/arch/arm/mach-shmobile/board-g3evm.c b/arch/arm/mach-shmobile/board-g3evm.c
index ef4613b..8b620bf 100644
--- a/arch/arm/mach-shmobile/board-g3evm.c
+++ b/arch/arm/mach-shmobile/board-g3evm.c
@@ -32,6 +32,7 @@
 #include <linux/gpio.h>
 #include <linux/input.h>
 #include <linux/input/sh_keysc.h>
+#include <linux/dma-mapping.h>
 #include <mach/sh7367.h>
 #include <mach/common.h>
 #include <asm/mach-types.h>
@@ -260,6 +261,8 @@ static struct map_desc g3evm_io_desc[] __initdata = {
 static void __init g3evm_map_io(void)
 {
 	iotable_init(g3evm_io_desc, ARRAY_SIZE(g3evm_io_desc));
+	/* DMA memory at 0xf6000000 - 0xffdfffff */
+	init_consistent_dma_size(158 << 20);
 
 	/* setup early devices and console here as well */
 	sh7367_add_early_devices();
diff --git a/arch/arm/mach-shmobile/board-g4evm.c b/arch/arm/mach-shmobile/board-g4evm.c
index 8e3c555..7719ddc 100644
--- a/arch/arm/mach-shmobile/board-g4evm.c
+++ b/arch/arm/mach-shmobile/board-g4evm.c
@@ -33,6 +33,7 @@
 #include <linux/mmc/host.h>
 #include <linux/mmc/sh_mobile_sdhi.h>
 #include <linux/gpio.h>
+#include <linux/dma-mapping.h>
 #include <mach/sh7377.h>
 #include <mach/common.h>
 #include <asm/mach-types.h>
@@ -274,6 +275,8 @@ static struct map_desc g4evm_io_desc[] __initdata = {
 static void __init g4evm_map_io(void)
 {
 	iotable_init(g4evm_io_desc, ARRAY_SIZE(g4evm_io_desc));
+	/* DMA memory at 0xf6000000 - 0xffdfffff */
+	init_consistent_dma_size(158 << 20);
 
 	/* setup early devices and console here as well */
 	sh7377_add_early_devices();
diff --git a/arch/arm/mach-shmobile/board-mackerel.c b/arch/arm/mach-shmobile/board-mackerel.c
index 17c19dc..3689ad2 100644
--- a/arch/arm/mach-shmobile/board-mackerel.c
+++ b/arch/arm/mach-shmobile/board-mackerel.c
@@ -39,12 +39,13 @@
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/partitions.h>
 #include <linux/mtd/physmap.h>
-#include <linux/pm_runtime.h>
+#include <linux/pm_clock.h>
 #include <linux/smsc911x.h>
 #include <linux/sh_intc.h>
 #include <linux/tca6416_keypad.h>
 #include <linux/usb/r8a66597.h>
 #include <linux/usb/renesas_usbhs.h>
+#include <linux/dma-mapping.h>
 
 #include <video/sh_mobile_hdmi.h>
 #include <video/sh_mobile_lcdc.h>
@@ -810,6 +811,7 @@ static struct usbhs_private usbhs1_private = {
 		},
 		.driver_param = {
 			.buswait_bwait	= 4,
+			.has_otg	= 1,
 			.pipe_type	= usbhs1_pipe_cfg,
 			.pipe_size	= ARRAY_SIZE(usbhs1_pipe_cfg),
 			.d0_tx_id	= SHDMA_SLAVE_USB1_TX,
@@ -1070,14 +1072,17 @@ static struct resource sdhi1_resources[] = {
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
+		.name	= SH_MOBILE_SDHI_IRQ_CARD_DETECT,
 		.start	= evt2irq(0x0e80), /* SDHI1_SDHI1I0 */
 		.flags	= IORESOURCE_IRQ,
 	},
 	[2] = {
+		.name	= SH_MOBILE_SDHI_IRQ_SDCARD,
 		.start	= evt2irq(0x0ea0), /* SDHI1_SDHI1I1 */
 		.flags	= IORESOURCE_IRQ,
 	},
 	[3] = {
+		.name	= SH_MOBILE_SDHI_IRQ_SDIO,
 		.start	= evt2irq(0x0ec0), /* SDHI1_SDHI1I2 */
 		.flags	= IORESOURCE_IRQ,
 	},
@@ -1121,14 +1126,17 @@ static struct resource sdhi2_resources[] = {
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
+		.name	= SH_MOBILE_SDHI_IRQ_CARD_DETECT,
 		.start	= evt2irq(0x1200), /* SDHI2_SDHI2I0 */
 		.flags	= IORESOURCE_IRQ,
 	},
 	[2] = {
+		.name	= SH_MOBILE_SDHI_IRQ_SDCARD,
 		.start	= evt2irq(0x1220), /* SDHI2_SDHI2I1 */
 		.flags	= IORESOURCE_IRQ,
 	},
 	[3] = {
+		.name	= SH_MOBILE_SDHI_IRQ_SDIO,
 		.start	= evt2irq(0x1240), /* SDHI2_SDHI2I2 */
 		.flags	= IORESOURCE_IRQ,
 	},
@@ -1381,6 +1389,8 @@ static struct map_desc mackerel_io_desc[] __initdata = {
 static void __init mackerel_map_io(void)
 {
 	iotable_init(mackerel_io_desc, ARRAY_SIZE(mackerel_io_desc));
+	/* DMA memory at 0xf6000000 - 0xffdfffff */
+	init_consistent_dma_size(158 << 20);
 
 	/* setup early devices and console here as well */
 	sh7372_add_early_devices();
@@ -1587,7 +1597,17 @@ static void __init mackerel_init(void)
 
 	sh7372_add_device_to_domain(&sh7372_a4lc, &lcdc_device);
 	sh7372_add_device_to_domain(&sh7372_a4lc, &hdmi_lcdc_device);
+	sh7372_add_device_to_domain(&sh7372_a4lc, &meram_device);
 	sh7372_add_device_to_domain(&sh7372_a4mp, &fsi_device);
+	sh7372_add_device_to_domain(&sh7372_a3sp, &usbhs0_device);
+	sh7372_add_device_to_domain(&sh7372_a3sp, &usbhs1_device);
+	sh7372_add_device_to_domain(&sh7372_a3sp, &sh_mmcif_device);
+	sh7372_add_device_to_domain(&sh7372_a3sp, &sdhi0_device);
+#if !defined(CONFIG_MMC_SH_MMCIF) && !defined(CONFIG_MMC_SH_MMCIF_MODULE)
+	sh7372_add_device_to_domain(&sh7372_a3sp, &sdhi1_device);
+#endif
+	sh7372_add_device_to_domain(&sh7372_a3sp, &sdhi2_device);
+	sh7372_add_device_to_domain(&sh7372_a4r, &ceu_device);
 
 	hdmi_init_pm_clock();
 	sh7372_pm_init();
diff --git a/arch/arm/mach-shmobile/entry-intc.S b/arch/arm/mach-shmobile/entry-intc.S
index cac0a7a..1a1c00c 100644
--- a/arch/arm/mach-shmobile/entry-intc.S
+++ b/arch/arm/mach-shmobile/entry-intc.S
@@ -51,7 +51,4 @@
 	.macro  test_for_ipi, irqnr, irqstat, base, tmp
 	.endm
 
-	.macro  test_for_ltirq, irqnr, irqstat, base, tmp
-	.endm
-
 	arch_irq_handler shmobile_handle_irq_intc
diff --git a/arch/arm/mach-shmobile/include/mach/common.h b/arch/arm/mach-shmobile/include/mach/common.h
index 06aecb3..c0cdbf9 100644
--- a/arch/arm/mach-shmobile/include/mach/common.h
+++ b/arch/arm/mach-shmobile/include/mach/common.h
@@ -35,8 +35,8 @@ extern void sh7372_add_standard_devices(void);
 extern void sh7372_clock_init(void);
 extern void sh7372_pinmux_init(void);
 extern void sh7372_pm_init(void);
-extern void sh7372_cpu_suspend(void);
-extern void sh7372_cpu_resume(void);
+extern void sh7372_resume_core_standby_a3sm(void);
+extern int sh7372_do_idle_a3sm(unsigned long unused);
 extern struct clk sh7372_extal1_clk;
 extern struct clk sh7372_extal2_clk;
 
diff --git a/arch/arm/mach-shmobile/include/mach/entry-macro.S b/arch/arm/mach-shmobile/include/mach/entry-macro.S
index d791f10..8d4a416 100644
--- a/arch/arm/mach-shmobile/include/mach/entry-macro.S
+++ b/arch/arm/mach-shmobile/include/mach/entry-macro.S
@@ -27,8 +27,5 @@
 	.macro  test_for_ipi, irqnr, irqstat, base, tmp
 	.endm
 
-	.macro  test_for_ltirq, irqnr, irqstat, base, tmp
-	.endm
-
 	.macro  arch_ret_to_user, tmp1, tmp2
 	.endm
diff --git a/arch/arm/mach-shmobile/include/mach/gpio.h b/arch/arm/mach-shmobile/include/mach/gpio.h
index 2b1bb9e..7bf0890 100644
--- a/arch/arm/mach-shmobile/include/mach/gpio.h
+++ b/arch/arm/mach-shmobile/include/mach/gpio.h
@@ -18,31 +18,15 @@
 
 #ifdef CONFIG_GPIOLIB
 
-static inline int gpio_get_value(unsigned gpio)
-{
-	return __gpio_get_value(gpio);
-}
-
-static inline void gpio_set_value(unsigned gpio, int value)
-{
-	__gpio_set_value(gpio, value);
-}
-
-static inline int gpio_cansleep(unsigned gpio)
-{
-	return __gpio_cansleep(gpio);
-}
-
-static inline int gpio_to_irq(unsigned gpio)
-{
-	return __gpio_to_irq(gpio);
-}
-
 static inline int irq_to_gpio(unsigned int irq)
 {
 	return -ENOSYS;
 }
 
+#else
+
+#define __ARM_GPIOLIB_COMPLEX
+
 #endif /* CONFIG_GPIOLIB */
 
 #endif /* __ASM_ARCH_GPIO_H */
diff --git a/arch/arm/mach-shmobile/include/mach/memory.h b/arch/arm/mach-shmobile/include/mach/memory.h
index ad00c3c..0ffbe81 100644
--- a/arch/arm/mach-shmobile/include/mach/memory.h
+++ b/arch/arm/mach-shmobile/include/mach/memory.h
@@ -4,7 +4,4 @@
 #define PLAT_PHYS_OFFSET	UL(CONFIG_MEMORY_START)
 #define MEM_SIZE	UL(CONFIG_MEMORY_SIZE)
 
-/* DMA memory at 0xf6000000 - 0xffdfffff */
-#define CONSISTENT_DMA_SIZE (158 << 20)
-
 #endif /* __ASM_MACH_MEMORY_H */
diff --git a/arch/arm/mach-shmobile/include/mach/sh7372.h b/arch/arm/mach-shmobile/include/mach/sh7372.h
index 24e63a8..84532f9 100644
--- a/arch/arm/mach-shmobile/include/mach/sh7372.h
+++ b/arch/arm/mach-shmobile/include/mach/sh7372.h
@@ -479,7 +479,12 @@ struct platform_device;
 
 struct sh7372_pm_domain {
 	struct generic_pm_domain genpd;
+	struct dev_power_governor *gov;
+	void (*suspend)(void);
+	void (*resume)(void);
 	unsigned int bit_shift;
+	bool no_debug;
+	bool stay_on;
 };
 
 static inline struct sh7372_pm_domain *to_sh7372_pd(struct generic_pm_domain *d)
@@ -491,16 +496,24 @@ static inline struct sh7372_pm_domain *to_sh7372_pd(struct generic_pm_domain *d)
 extern struct sh7372_pm_domain sh7372_a4lc;
 extern struct sh7372_pm_domain sh7372_a4mp;
 extern struct sh7372_pm_domain sh7372_d4;
+extern struct sh7372_pm_domain sh7372_a4r;
 extern struct sh7372_pm_domain sh7372_a3rv;
 extern struct sh7372_pm_domain sh7372_a3ri;
+extern struct sh7372_pm_domain sh7372_a3sp;
 extern struct sh7372_pm_domain sh7372_a3sg;
 
 extern void sh7372_init_pm_domain(struct sh7372_pm_domain *sh7372_pd);
 extern void sh7372_add_device_to_domain(struct sh7372_pm_domain *sh7372_pd,
 					struct platform_device *pdev);
+extern void sh7372_pm_add_subdomain(struct sh7372_pm_domain *sh7372_pd,
+				    struct sh7372_pm_domain *sh7372_sd);
 #else
 #define sh7372_init_pm_domain(pd) do { } while(0)
 #define sh7372_add_device_to_domain(pd, pdev) do { } while(0)
+#define sh7372_pm_add_subdomain(pd, sd) do { } while(0)
 #endif /* CONFIG_PM */
 
+extern void sh7372_intcs_suspend(void);
+extern void sh7372_intcs_resume(void);
+
 #endif /* __ASM_SH7372_H__ */
diff --git a/arch/arm/mach-shmobile/intc-sh7372.c b/arch/arm/mach-shmobile/intc-sh7372.c
index 739315e..29cdc05 100644
--- a/arch/arm/mach-shmobile/intc-sh7372.c
+++ b/arch/arm/mach-shmobile/intc-sh7372.c
@@ -606,9 +606,16 @@ static void intcs_demux(unsigned int irq, struct irq_desc *desc)
 	generic_handle_irq(intcs_evt2irq(evtcodeas));
 }
 
+static void __iomem *intcs_ffd2;
+static void __iomem *intcs_ffd5;
+
 void __init sh7372_init_irq(void)
 {
-	void __iomem *intevtsa = ioremap_nocache(0xffd20100, PAGE_SIZE);
+	void __iomem *intevtsa;
+
+	intcs_ffd2 = ioremap_nocache(0xffd20000, PAGE_SIZE);
+	intevtsa = intcs_ffd2 + 0x100;
+	intcs_ffd5 = ioremap_nocache(0xffd50000, PAGE_SIZE);
 
 	register_intc_controller(&intca_desc);
 	register_intc_controller(&intcs_desc);
@@ -617,3 +624,46 @@ void __init sh7372_init_irq(void)
 	irq_set_handler_data(evt2irq(0xf80), (void *)intevtsa);
 	irq_set_chained_handler(evt2irq(0xf80), intcs_demux);
 }
+
+static unsigned short ffd2[0x200];
+static unsigned short ffd5[0x100];
+
+void sh7372_intcs_suspend(void)
+{
+	int k;
+
+	for (k = 0x00; k <= 0x30; k += 4)
+		ffd2[k] = __raw_readw(intcs_ffd2 + k);
+
+	for (k = 0x80; k <= 0xb0; k += 4)
+		ffd2[k] = __raw_readb(intcs_ffd2 + k);
+
+	for (k = 0x180; k <= 0x188; k += 4)
+		ffd2[k] = __raw_readb(intcs_ffd2 + k);
+
+	for (k = 0x00; k <= 0x3c; k += 4)
+		ffd5[k] = __raw_readw(intcs_ffd5 + k);
+
+	for (k = 0x80; k <= 0x9c; k += 4)
+		ffd5[k] = __raw_readb(intcs_ffd5 + k);
+}
+
+void sh7372_intcs_resume(void)
+{
+	int k;
+
+	for (k = 0x00; k <= 0x30; k += 4)
+		__raw_writew(ffd2[k], intcs_ffd2 + k);
+
+	for (k = 0x80; k <= 0xb0; k += 4)
+		__raw_writeb(ffd2[k], intcs_ffd2 + k);
+
+	for (k = 0x180; k <= 0x188; k += 4)
+		__raw_writeb(ffd2[k], intcs_ffd2 + k);
+
+	for (k = 0x00; k <= 0x3c; k += 4)
+		__raw_writew(ffd5[k], intcs_ffd5 + k);
+
+	for (k = 0x80; k <= 0x9c; k += 4)
+		__raw_writeb(ffd5[k], intcs_ffd5 + k);
+}
diff --git a/arch/arm/mach-shmobile/platsmp.c b/arch/arm/mach-shmobile/platsmp.c
index 66f9806..e4e485f 100644
--- a/arch/arm/mach-shmobile/platsmp.c
+++ b/arch/arm/mach-shmobile/platsmp.c
@@ -56,6 +56,12 @@ void __init smp_init_cpus(void)
 	unsigned int ncores = shmobile_smp_get_core_count();
 	unsigned int i;
 
+	if (ncores > nr_cpu_ids) {
+		pr_warn("SMP: %u cores greater than maximum (%u), clipping\n",
+			ncores, nr_cpu_ids);
+		ncores = nr_cpu_ids;
+	}
+
 	for (i = 0; i < ncores; i++)
 		set_cpu_possible(i, true);
 
diff --git a/arch/arm/mach-shmobile/pm-sh7372.c b/arch/arm/mach-shmobile/pm-sh7372.c
index 933fb41..7961273 100644
--- a/arch/arm/mach-shmobile/pm-sh7372.c
+++ b/arch/arm/mach-shmobile/pm-sh7372.c
@@ -15,23 +15,61 @@
 #include <linux/list.h>
 #include <linux/err.h>
 #include <linux/slab.h>
-#include <linux/pm_runtime.h>
+#include <linux/pm_clock.h>
 #include <linux/platform_device.h>
 #include <linux/delay.h>
+#include <linux/irq.h>
+#include <linux/bitrev.h>
 #include <asm/system.h>
 #include <asm/io.h>
 #include <asm/tlbflush.h>
+#include <asm/suspend.h>
 #include <mach/common.h>
 #include <mach/sh7372.h>
 
-#define SMFRAM 0xe6a70000
-#define SYSTBCR 0xe6150024
-#define SBAR 0xe6180020
-#define APARMBAREA 0xe6f10020
+/* DBG */
+#define DBGREG1 0xe6100020
+#define DBGREG9 0xe6100040
 
+/* CPGA */
+#define SYSTBCR 0xe6150024
+#define MSTPSR0 0xe6150030
+#define MSTPSR1 0xe6150038
+#define MSTPSR2 0xe6150040
+#define MSTPSR3 0xe6150048
+#define MSTPSR4 0xe615004c
+#define PLLC01STPCR 0xe61500c8
+
+/* SYSC */
 #define SPDCR 0xe6180008
 #define SWUCR 0xe6180014
+#define SBAR 0xe6180020
+#define WUPRMSK 0xe6180028
+#define WUPSMSK 0xe618002c
+#define WUPSMSK2 0xe6180048
 #define PSTR 0xe6180080
+#define WUPSFAC 0xe6180098
+#define IRQCR 0xe618022c
+#define IRQCR2 0xe6180238
+#define IRQCR3 0xe6180244
+#define IRQCR4 0xe6180248
+#define PDNSEL 0xe6180254
+
+/* INTC */
+#define ICR1A 0xe6900000
+#define ICR2A 0xe6900004
+#define ICR3A 0xe6900008
+#define ICR4A 0xe690000c
+#define INTMSK00A 0xe6900040
+#define INTMSK10A 0xe6900044
+#define INTMSK20A 0xe6900048
+#define INTMSK30A 0xe690004c
+
+/* MFIS */
+#define SMFRAM 0xe6a70000
+
+/* AP-System Core */
+#define APARMBAREA 0xe6f10020
 
 #define PSTR_RETRIES 100
 #define PSTR_DELAY_US 10
@@ -43,6 +81,12 @@ static int pd_power_down(struct generic_pm_domain *genpd)
 	struct sh7372_pm_domain *sh7372_pd = to_sh7372_pd(genpd);
 	unsigned int mask = 1 << sh7372_pd->bit_shift;
 
+	if (sh7372_pd->suspend)
+		sh7372_pd->suspend();
+
+	if (sh7372_pd->stay_on)
+		return 0;
+
 	if (__raw_readl(PSTR) & mask) {
 		unsigned int retry_count;
 
@@ -55,8 +99,9 @@ static int pd_power_down(struct generic_pm_domain *genpd)
 		}
 	}
 
-	pr_debug("sh7372 power domain down 0x%08x -> PSTR = 0x%08x\n",
-		 mask, __raw_readl(PSTR));
+	if (!sh7372_pd->no_debug)
+		pr_debug("sh7372 power domain down 0x%08x -> PSTR = 0x%08x\n",
+			 mask, __raw_readl(PSTR));
 
 	return 0;
 }
@@ -68,6 +113,9 @@ static int pd_power_up(struct generic_pm_domain *genpd)
 	unsigned int retry_count;
 	int ret = 0;
 
+	if (sh7372_pd->stay_on)
+		goto out;
+
 	if (__raw_readl(PSTR) & mask)
 		goto out;
 
@@ -84,66 +132,48 @@ static int pd_power_up(struct generic_pm_domain *genpd)
 	if (__raw_readl(SWUCR) & mask)
 		ret = -EIO;
 
+	if (!sh7372_pd->no_debug)
+		pr_debug("sh7372 power domain up 0x%08x -> PSTR = 0x%08x\n",
+			 mask, __raw_readl(PSTR));
+
  out:
-	pr_debug("sh7372 power domain up 0x%08x -> PSTR = 0x%08x\n",
-		 mask, __raw_readl(PSTR));
+	if (ret == 0 && sh7372_pd->resume)
+		sh7372_pd->resume();
 
 	return ret;
 }
 
-static int pd_power_up_a3rv(struct generic_pm_domain *genpd)
+static void sh7372_a4r_suspend(void)
 {
-	int ret = pd_power_up(genpd);
-
-	/* force A4LC on after A3RV has been requested on */
-	pm_genpd_poweron(&sh7372_a4lc.genpd);
-
-	return ret;
+	sh7372_intcs_suspend();
+	__raw_writel(0x300fffff, WUPRMSK); /* avoid wakeup */
 }
 
-static int pd_power_down_a3rv(struct generic_pm_domain *genpd)
+static bool pd_active_wakeup(struct device *dev)
 {
-	int ret = pd_power_down(genpd);
-
-	/* try to power down A4LC after A3RV is requested off */
-	genpd_queue_power_off_work(&sh7372_a4lc.genpd);
-
-	return ret;
+	return true;
 }
 
-static int pd_power_down_a4lc(struct generic_pm_domain *genpd)
+static bool sh7372_power_down_forbidden(struct dev_pm_domain *domain)
 {
-	/* only power down A4LC if A3RV is off */
-	if (!(__raw_readl(PSTR) & (1 << sh7372_a3rv.bit_shift)))
-		return pd_power_down(genpd);
-
-	return -EBUSY;
+	return false;
 }
 
-static bool pd_active_wakeup(struct device *dev)
-{
-	return true;
-}
+struct dev_power_governor sh7372_always_on_gov = {
+	.power_down_ok = sh7372_power_down_forbidden,
+};
 
 void sh7372_init_pm_domain(struct sh7372_pm_domain *sh7372_pd)
 {
 	struct generic_pm_domain *genpd = &sh7372_pd->genpd;
 
-	pm_genpd_init(genpd, NULL, false);
+	pm_genpd_init(genpd, sh7372_pd->gov, false);
 	genpd->stop_device = pm_clk_suspend;
 	genpd->start_device = pm_clk_resume;
+	genpd->dev_irq_safe = true;
 	genpd->active_wakeup = pd_active_wakeup;
-
-	if (sh7372_pd == &sh7372_a4lc) {
-		genpd->power_off = pd_power_down_a4lc;
-		genpd->power_on = pd_power_up;
-	} else if (sh7372_pd == &sh7372_a3rv) {
-		genpd->power_off = pd_power_down_a3rv;
-		genpd->power_on = pd_power_up_a3rv;
-	} else {
-		genpd->power_off = pd_power_down;
-		genpd->power_on = pd_power_up;
-	}
+	genpd->power_off = pd_power_down;
+	genpd->power_on = pd_power_up;
 	genpd->power_on(&sh7372_pd->genpd);
 }
 
@@ -152,11 +182,15 @@ void sh7372_add_device_to_domain(struct sh7372_pm_domain *sh7372_pd,
 {
 	struct device *dev = &pdev->dev;
 
-	if (!dev->power.subsys_data) {
-		pm_clk_init(dev);
-		pm_clk_add(dev, NULL);
-	}
 	pm_genpd_add_device(&sh7372_pd->genpd, dev);
+	if (pm_clk_no_clocks(dev))
+		pm_clk_add(dev, NULL);
+}
+
+void sh7372_pm_add_subdomain(struct sh7372_pm_domain *sh7372_pd,
+			     struct sh7372_pm_domain *sh7372_sd)
+{
+	pm_genpd_add_subdomain(&sh7372_pd->genpd, &sh7372_sd->genpd);
 }
 
 struct sh7372_pm_domain sh7372_a4lc = {
@@ -171,6 +205,14 @@ struct sh7372_pm_domain sh7372_d4 = {
 	.bit_shift = 3,
 };
 
+struct sh7372_pm_domain sh7372_a4r = {
+	.bit_shift = 5,
+	.gov = &sh7372_always_on_gov,
+	.suspend = sh7372_a4r_suspend,
+	.resume = sh7372_intcs_resume,
+	.stay_on = true,
+};
+
 struct sh7372_pm_domain sh7372_a3rv = {
 	.bit_shift = 6,
 };
@@ -179,39 +221,187 @@ struct sh7372_pm_domain sh7372_a3ri = {
 	.bit_shift = 8,
 };
 
+struct sh7372_pm_domain sh7372_a3sp = {
+	.bit_shift = 11,
+	.gov = &sh7372_always_on_gov,
+	.no_debug = true,
+};
+
 struct sh7372_pm_domain sh7372_a3sg = {
 	.bit_shift = 13,
 };
 
 #endif /* CONFIG_PM */
 
+#if defined(CONFIG_SUSPEND) || defined(CONFIG_CPU_IDLE)
+static int sh7372_do_idle_core_standby(unsigned long unused)
+{
+	cpu_do_idle(); /* WFI when SYSTBCR == 0x10 -> Core Standby */
+	return 0;
+}
+
 static void sh7372_enter_core_standby(void)
 {
-	void __iomem *smfram = (void __iomem *)SMFRAM;
+	/* set reset vector, translate 4k */
+	__raw_writel(__pa(sh7372_resume_core_standby_a3sm), SBAR);
+	__raw_writel(0, APARMBAREA);
 
-	__raw_writel(0, APARMBAREA); /* translate 4k */
-	__raw_writel(__pa(sh7372_cpu_resume), SBAR); /* set reset vector */
-	__raw_writel(0x10, SYSTBCR); /* enable core standby */
+	/* enter sleep mode with SYSTBCR to 0x10 */
+	__raw_writel(0x10, SYSTBCR);
+	cpu_suspend(0, sh7372_do_idle_core_standby);
+	__raw_writel(0, SYSTBCR);
 
-	__raw_writel(0, smfram + 0x3c); /* clear page table address */
+	 /* disable reset vector translation */
+	__raw_writel(0, SBAR);
+}
+#endif
+
+#ifdef CONFIG_SUSPEND
+static void sh7372_enter_a3sm_common(int pllc0_on)
+{
+	/* set reset vector, translate 4k */
+	__raw_writel(__pa(sh7372_resume_core_standby_a3sm), SBAR);
+	__raw_writel(0, APARMBAREA);
+
+	if (pllc0_on)
+		__raw_writel(0, PLLC01STPCR);
+	else
+		__raw_writel(1 << 28, PLLC01STPCR);
+
+	__raw_writel(0, PDNSEL); /* power-down A3SM only, not A4S */
+	__raw_readl(WUPSFAC); /* read wakeup int. factor before sleep */
+	cpu_suspend(0, sh7372_do_idle_a3sm);
+	__raw_readl(WUPSFAC); /* read wakeup int. factor after wakeup */
+
+	 /* disable reset vector translation */
+	__raw_writel(0, SBAR);
+}
+
+static int sh7372_a3sm_valid(unsigned long *mskp, unsigned long *msk2p)
+{
+	unsigned long mstpsr0, mstpsr1, mstpsr2, mstpsr3, mstpsr4;
+	unsigned long msk, msk2;
+
+	/* check active clocks to determine potential wakeup sources */
+
+	mstpsr0 = __raw_readl(MSTPSR0);
+	if ((mstpsr0 & 0x00000003) != 0x00000003) {
+		pr_debug("sh7372 mstpsr0 0x%08lx\n", mstpsr0);
+		return 0;
+	}
+
+	mstpsr1 = __raw_readl(MSTPSR1);
+	if ((mstpsr1 & 0xff079b7f) != 0xff079b7f) {
+		pr_debug("sh7372 mstpsr1 0x%08lx\n", mstpsr1);
+		return 0;
+	}
 
-	sh7372_cpu_suspend();
-	cpu_init();
+	mstpsr2 = __raw_readl(MSTPSR2);
+	if ((mstpsr2 & 0x000741ff) != 0x000741ff) {
+		pr_debug("sh7372 mstpsr2 0x%08lx\n", mstpsr2);
+		return 0;
+	}
 
-	/* if page table address is non-NULL then we have been powered down */
-	if (__raw_readl(smfram + 0x3c)) {
-		__raw_writel(__raw_readl(smfram + 0x40),
-			     __va(__raw_readl(smfram + 0x3c)));
+	mstpsr3 = __raw_readl(MSTPSR3);
+	if ((mstpsr3 & 0x1a60f010) != 0x1a60f010) {
+		pr_debug("sh7372 mstpsr3 0x%08lx\n", mstpsr3);
+		return 0;
+	}
 
-		flush_tlb_all();
-		set_cr(__raw_readl(smfram + 0x38));
+	mstpsr4 = __raw_readl(MSTPSR4);
+	if ((mstpsr4 & 0x00008cf0) != 0x00008cf0) {
+		pr_debug("sh7372 mstpsr4 0x%08lx\n", mstpsr4);
+		return 0;
 	}
 
-	__raw_writel(0, SYSTBCR); /* disable core standby */
-	__raw_writel(0, SBAR); /* disable reset vector translation */
+	msk = 0;
+	msk2 = 0;
+
+	/* make bitmaps of limited number of wakeup sources */
+
+	if ((mstpsr2 & (1 << 23)) == 0) /* SPU2 */
+		msk |= 1 << 31;
+
+	if ((mstpsr2 & (1 << 12)) == 0) /* MFI_MFIM */
+		msk |= 1 << 21;
+
+	if ((mstpsr4 & (1 << 3)) == 0) /* KEYSC */
+		msk |= 1 << 2;
+
+	if ((mstpsr1 & (1 << 24)) == 0) /* CMT0 */
+		msk |= 1 << 1;
+
+	if ((mstpsr3 & (1 << 29)) == 0) /* CMT1 */
+		msk |= 1 << 1;
+
+	if ((mstpsr4 & (1 << 0)) == 0) /* CMT2 */
+		msk |= 1 << 1;
+
+	if ((mstpsr2 & (1 << 13)) == 0) /* MFI_MFIS */
+		msk2 |= 1 << 17;
+
+	*mskp = msk;
+	*msk2p = msk2;
+
+	return 1;
+}
+
+static void sh7372_icr_to_irqcr(unsigned long icr, u16 *irqcr1p, u16 *irqcr2p)
+{
+	u16 tmp, irqcr1, irqcr2;
+	int k;
+
+	irqcr1 = 0;
+	irqcr2 = 0;
+
+	/* convert INTCA ICR register layout to SYSC IRQCR+IRQCR2 */
+	for (k = 0; k <= 7; k++) {
+		tmp = (icr >> ((7 - k) * 4)) & 0xf;
+		irqcr1 |= (tmp & 0x03) << (k * 2);
+		irqcr2 |= (tmp >> 2) << (k * 2);
+	}
+
+	*irqcr1p = irqcr1;
+	*irqcr2p = irqcr2;
+}
+
+static void sh7372_setup_a3sm(unsigned long msk, unsigned long msk2)
+{
+	u16 irqcrx_low, irqcrx_high, irqcry_low, irqcry_high;
+	unsigned long tmp;
+
+	/* read IRQ0A -> IRQ15A mask */
+	tmp = bitrev8(__raw_readb(INTMSK00A));
+	tmp |= bitrev8(__raw_readb(INTMSK10A)) << 8;
+
+	/* setup WUPSMSK from clocks and external IRQ mask */
+	msk = (~msk & 0xc030000f) | (tmp << 4);
+	__raw_writel(msk, WUPSMSK);
+
+	/* propage level/edge trigger for external IRQ 0->15 */
+	sh7372_icr_to_irqcr(__raw_readl(ICR1A), &irqcrx_low, &irqcry_low);
+	sh7372_icr_to_irqcr(__raw_readl(ICR2A), &irqcrx_high, &irqcry_high);
+	__raw_writel((irqcrx_high << 16) | irqcrx_low, IRQCR);
+	__raw_writel((irqcry_high << 16) | irqcry_low, IRQCR2);
+
+	/* read IRQ16A -> IRQ31A mask */
+	tmp = bitrev8(__raw_readb(INTMSK20A));
+	tmp |= bitrev8(__raw_readb(INTMSK30A)) << 8;
+
+	/* setup WUPSMSK2 from clocks and external IRQ mask */
+	msk2 = (~msk2 & 0x00030000) | tmp;
+	__raw_writel(msk2, WUPSMSK2);
+
+	/* propage level/edge trigger for external IRQ 16->31 */
+	sh7372_icr_to_irqcr(__raw_readl(ICR3A), &irqcrx_low, &irqcry_low);
+	sh7372_icr_to_irqcr(__raw_readl(ICR4A), &irqcrx_high, &irqcry_high);
+	__raw_writel((irqcrx_high << 16) | irqcrx_low, IRQCR3);
+	__raw_writel((irqcry_high << 16) | irqcry_low, IRQCR4);
 }
+#endif
 
 #ifdef CONFIG_CPU_IDLE
+
 static void sh7372_cpuidle_setup(struct cpuidle_device *dev)
 {
 	struct cpuidle_state *state;
@@ -239,9 +429,25 @@ static void sh7372_cpuidle_init(void) {}
 #endif
 
 #ifdef CONFIG_SUSPEND
+
 static int sh7372_enter_suspend(suspend_state_t suspend_state)
 {
-	sh7372_enter_core_standby();
+	unsigned long msk, msk2;
+
+	/* check active clocks to determine potential wakeup sources */
+	if (sh7372_a3sm_valid(&msk, &msk2)) {
+
+		/* convert INTC mask and sense to SYSC mask and sense */
+		sh7372_setup_a3sm(msk, msk2);
+
+		/* enter A3SM sleep with PLLC0 off */
+		pr_debug("entering A3SM\n");
+		sh7372_enter_a3sm_common(0);
+	} else {
+		/* default to Core Standby that supports all wakeup sources */
+		pr_debug("entering Core Standby\n");
+		sh7372_enter_core_standby();
+	}
 	return 0;
 }
 
@@ -253,9 +459,6 @@ static void sh7372_suspend_init(void)
 static void sh7372_suspend_init(void) {}
 #endif
 
-#define DBGREG1 0xe6100020
-#define DBGREG9 0xe6100040
-
 void __init sh7372_pm_init(void)
 {
 	/* enable DBG hardware block to kick SYSC */
@@ -263,6 +466,9 @@ void __init sh7372_pm_init(void)
 	__raw_writel(0x0000a501, DBGREG9);
 	__raw_writel(0x00000000, DBGREG1);
 
+	/* do not convert A3SM, A3SP, A3SG, A4R power down into A4S */
+	__raw_writel(0, PDNSEL);
+
 	sh7372_suspend_init();
 	sh7372_cpuidle_init();
 }
diff --git a/arch/arm/mach-shmobile/pm_runtime.c b/arch/arm/mach-shmobile/pm_runtime.c
index 6ec454e..bd5c6a3 100644
--- a/arch/arm/mach-shmobile/pm_runtime.c
+++ b/arch/arm/mach-shmobile/pm_runtime.c
@@ -15,6 +15,7 @@
 #include <linux/io.h>
 #include <linux/pm_runtime.h>
 #include <linux/pm_domain.h>
+#include <linux/pm_clock.h>
 #include <linux/platform_device.h>
 #include <linux/clk.h>
 #include <linux/sh_clk.h>
diff --git a/arch/arm/mach-shmobile/setup-sh7372.c b/arch/arm/mach-shmobile/setup-sh7372.c
index 2d9b1b1..2380389 100644
--- a/arch/arm/mach-shmobile/setup-sh7372.c
+++ b/arch/arm/mach-shmobile/setup-sh7372.c
@@ -30,6 +30,7 @@
 #include <linux/sh_dma.h>
 #include <linux/sh_intc.h>
 #include <linux/sh_timer.h>
+#include <linux/pm_domain.h>
 #include <mach/hardware.h>
 #include <mach/sh7372.h>
 #include <asm/mach-types.h>
@@ -990,9 +991,14 @@ void __init sh7372_add_standard_devices(void)
 	sh7372_init_pm_domain(&sh7372_a4lc);
 	sh7372_init_pm_domain(&sh7372_a4mp);
 	sh7372_init_pm_domain(&sh7372_d4);
+	sh7372_init_pm_domain(&sh7372_a4r);
 	sh7372_init_pm_domain(&sh7372_a3rv);
 	sh7372_init_pm_domain(&sh7372_a3ri);
 	sh7372_init_pm_domain(&sh7372_a3sg);
+	sh7372_init_pm_domain(&sh7372_a3sp);
+
+	sh7372_pm_add_subdomain(&sh7372_a4lc, &sh7372_a3rv);
+	sh7372_pm_add_subdomain(&sh7372_a4r, &sh7372_a4lc);
 
 	platform_add_devices(sh7372_early_devices,
 			    ARRAY_SIZE(sh7372_early_devices));
@@ -1003,6 +1009,25 @@ void __init sh7372_add_standard_devices(void)
 	sh7372_add_device_to_domain(&sh7372_a3rv, &vpu_device);
 	sh7372_add_device_to_domain(&sh7372_a4mp, &spu0_device);
 	sh7372_add_device_to_domain(&sh7372_a4mp, &spu1_device);
+	sh7372_add_device_to_domain(&sh7372_a3sp, &scif0_device);
+	sh7372_add_device_to_domain(&sh7372_a3sp, &scif1_device);
+	sh7372_add_device_to_domain(&sh7372_a3sp, &scif2_device);
+	sh7372_add_device_to_domain(&sh7372_a3sp, &scif3_device);
+	sh7372_add_device_to_domain(&sh7372_a3sp, &scif4_device);
+	sh7372_add_device_to_domain(&sh7372_a3sp, &scif5_device);
+	sh7372_add_device_to_domain(&sh7372_a3sp, &scif6_device);
+	sh7372_add_device_to_domain(&sh7372_a3sp, &iic1_device);
+	sh7372_add_device_to_domain(&sh7372_a3sp, &dma0_device);
+	sh7372_add_device_to_domain(&sh7372_a3sp, &dma1_device);
+	sh7372_add_device_to_domain(&sh7372_a3sp, &dma2_device);
+	sh7372_add_device_to_domain(&sh7372_a3sp, &usb_dma0_device);
+	sh7372_add_device_to_domain(&sh7372_a3sp, &usb_dma1_device);
+	sh7372_add_device_to_domain(&sh7372_a4r, &iic0_device);
+	sh7372_add_device_to_domain(&sh7372_a4r, &veu0_device);
+	sh7372_add_device_to_domain(&sh7372_a4r, &veu1_device);
+	sh7372_add_device_to_domain(&sh7372_a4r, &veu2_device);
+	sh7372_add_device_to_domain(&sh7372_a4r, &veu3_device);
+	sh7372_add_device_to_domain(&sh7372_a4r, &jpu_device);
 }
 
 void __init sh7372_add_early_devices(void)
diff --git a/arch/arm/mach-shmobile/sleep-sh7372.S b/arch/arm/mach-shmobile/sleep-sh7372.S
index d37d3ca..f3ab3c5 100644
--- a/arch/arm/mach-shmobile/sleep-sh7372.S
+++ b/arch/arm/mach-shmobile/sleep-sh7372.S
@@ -30,58 +30,20 @@
  */
 
 #include <linux/linkage.h>
+#include <linux/init.h>
+#include <asm/memory.h>
 #include <asm/assembler.h>
 
-#define SMFRAM 0xe6a70000
-
-	.align
-kernel_flush:
-	.word	v7_flush_dcache_all
-
-	.align	3
-ENTRY(sh7372_cpu_suspend)
-	stmfd	sp!, {r0-r12, lr}	@ save registers on stack
-
-	ldr	r8, =SMFRAM
-
-	mov	r4, sp			@ Store sp
-	mrs	r5, spsr		@ Store spsr
-	mov	r6, lr			@ Store lr
-	stmia	r8!, {r4-r6}
-
-	mrc	p15, 0, r4, c1, c0, 2	@ Coprocessor access control register
-	mrc	p15, 0, r5, c2, c0, 0	@ TTBR0
-	mrc	p15, 0, r6, c2, c0, 1	@ TTBR1
-	mrc	p15, 0, r7, c2, c0, 2	@ TTBCR
-	stmia	r8!, {r4-r7}
-
-	mrc	p15, 0, r4, c3, c0, 0	@ Domain access Control Register
-	mrc	p15, 0, r5, c10, c2, 0	@ PRRR
-	mrc	p15, 0, r6, c10, c2, 1	@ NMRR
-	stmia	r8!,{r4-r6}
-
-	mrc	p15, 0, r4, c13, c0, 1	@ Context ID
-	mrc	p15, 0, r5, c13, c0, 2	@ User r/w thread and process ID
-	mrc	p15, 0, r6, c12, c0, 0	@ Secure or NS vector base address
-	mrs	r7, cpsr		@ Store current cpsr
-	stmia	r8!, {r4-r7}
-
-	mrc	p15, 0, r4, c1, c0, 0	@ save control register
-	stmia	r8!, {r4}
-
-	/*
-	 * jump out to kernel flush routine
-	 *  - reuse that code is better
-	 *  - it executes in a cached space so is faster than refetch per-block
-	 *  - should be faster and will change with kernel
-	 *  - 'might' have to copy address, load and jump to it
-	 * Flush all data from the L1 data cache before disabling
-	 * SCTLR.C bit.
-	 */
-	ldr	r1, kernel_flush
-	mov	lr, pc
-	bx	r1
+#if defined(CONFIG_SUSPEND) || defined(CONFIG_CPU_IDLE)
+	.align	12
+	.text
+	.global sh7372_resume_core_standby_a3sm
+sh7372_resume_core_standby_a3sm:
+	ldr     pc, 1f
+1:	.long   cpu_resume - PAGE_OFFSET + PLAT_PHYS_OFFSET
 
+	.global	sh7372_do_idle_a3sm
+sh7372_do_idle_a3sm:
 	/*
 	 * Clear the SCTLR.C bit to prevent further data cache
 	 * allocation. Clearing SCTLR.C would make all the data accesses
@@ -92,10 +54,13 @@ ENTRY(sh7372_cpu_suspend)
 	mcr	p15, 0, r0, c1, c0, 0
 	isb
 
+	/* disable L2 cache in the aux control register */
+	mrc     p15, 0, r10, c1, c0, 1
+	bic     r10, r10, #2
+	mcr     p15, 0, r10, c1, c0, 1
+
 	/*
-	 * Invalidate L1 data cache. Even though only invalidate is
-	 * necessary exported flush API is used here. Doing clean
-	 * on already clean cache would be almost NOP.
+	 * Invalidate data cache again.
 	 */
 	ldr	r1, kernel_flush
 	blx	r1
@@ -115,146 +80,16 @@ ENTRY(sh7372_cpu_suspend)
 	dsb
 	dmb
 
-/*
- * ===================================
- * == WFI instruction => Enter idle ==
- * ===================================
- */
-	wfi				@ wait for interrupt
-
-/*
- * ===================================
- * == Resume path for non-OFF modes ==
- * ===================================
- */
-	mrc	p15, 0, r0, c1, c0, 0
-	tst	r0, #(1 << 2)		@ Check C bit enabled?
-	orreq	r0, r0, #(1 << 2)	@ Enable the C bit if cleared
-	mcreq	p15, 0, r0, c1, c0, 0
-	isb
-
-/*
- * ===================================
- * == Exit point from non-OFF modes ==
- * ===================================
- */
-	ldmfd	sp!, {r0-r12, pc}	@ restore regs and return
+#define SPDCR 0xe6180008
+#define A3SM (1 << 12)
 
-	.pool
+	/* A3SM power down */
+	ldr     r0, =SPDCR
+	ldr     r1, =A3SM
+	str     r1, [r0]
+1:
+	b      1b
 
-	.align	12
-	.text
-	.global	sh7372_cpu_resume
-sh7372_cpu_resume:
-
-	mov	r1, #0
-	/*
-	 * Invalidate all instruction caches to PoU
-	 * and flush branch target cache
-	 */
-	mcr	p15, 0, r1, c7, c5, 0
-
-	ldr	r3, =SMFRAM
-
-	ldmia	r3!, {r4-r6}
-	mov	sp, r4			@ Restore sp
-	msr	spsr_cxsf, r5		@ Restore spsr
-	mov	lr, r6			@ Restore lr
-
-	ldmia	r3!, {r4-r7}
-	mcr	p15, 0, r4, c1, c0, 2	@ Coprocessor access Control Register
-	mcr	p15, 0, r5, c2, c0, 0	@ TTBR0
-	mcr	p15, 0, r6, c2, c0, 1	@ TTBR1
-	mcr	p15, 0, r7, c2, c0, 2	@ TTBCR
-
-	ldmia	r3!,{r4-r6}
-	mcr	p15, 0, r4, c3, c0, 0	@ Domain access Control Register
-	mcr	p15, 0, r5, c10, c2, 0	@ PRRR
-	mcr	p15, 0, r6, c10, c2, 1	@ NMRR
-
-	ldmia	r3!,{r4-r7}
-	mcr	p15, 0, r4, c13, c0, 1	@ Context ID
-	mcr	p15, 0, r5, c13, c0, 2	@ User r/w thread and process ID
-	mrc	p15, 0, r6, c12, c0, 0	@ Secure or NS vector base address
-	msr	cpsr, r7		@ store cpsr
-
-	/* Starting to enable MMU here */
-	mrc	p15, 0, r7, c2, c0, 2 	@ Read TTBRControl
-	/* Extract N (0:2) bits and decide whether to use TTBR0 or TTBR1 */
-	and	r7, #0x7
-	cmp	r7, #0x0
-	beq	usettbr0
-ttbr_error:
-	/*
-	 * More work needs to be done to support N[0:2] value other than 0
-	 * So looping here so that the error can be detected
-	 */
-	b	ttbr_error
-
-	.align
-cache_pred_disable_mask:
-	.word	0xFFFFE7FB
-ttbrbit_mask:
-	.word	0xFFFFC000
-table_index_mask:
-	.word	0xFFF00000
-table_entry:
-	.word	0x00000C02
-usettbr0:
-
-	mrc	p15, 0, r2, c2, c0, 0
-	ldr	r5, ttbrbit_mask
-	and	r2, r5
-	mov	r4, pc
-	ldr	r5, table_index_mask
-	and	r4, r5			@ r4 = 31 to 20 bits of pc
-	/* Extract the value to be written to table entry */
-	ldr	r6, table_entry
-	/* r6 has the value to be written to table entry */
-	add	r6, r6, r4
-	/* Getting the address of table entry to modify */
-	lsr	r4, #18
-	/* r2 has the location which needs to be modified */
-	add	r2, r4
-	ldr	r4, [r2]
-	str	r6, [r2] /* modify the table entry */
-
-	mov	r7, r6
-	mov	r5, r2
-	mov	r6, r4
-	/* r5 = original page table address */
-	/* r6 = original page table data */
-
-	mov	r0, #0
-	mcr	p15, 0, r0, c7, c5, 4	@ Flush prefetch buffer
-	mcr	p15, 0, r0, c7, c5, 6	@ Invalidate branch predictor array
-	mcr	p15, 0, r0, c8, c5, 0	@ Invalidate instruction TLB
-	mcr	p15, 0, r0, c8, c6, 0	@ Invalidate data TLB
-
-	/*
-	 * Restore control register. This enables the MMU.
-	 * The caches and prediction are not enabled here, they
-	 * will be enabled after restoring the MMU table entry.
-	 */
-	ldmia	r3!, {r4}
-	stmia	r3!, {r5} /* save original page table address */
-	stmia	r3!, {r6} /* save original page table data */
-	stmia	r3!, {r7} /* save modified page table data */
-
-	ldr	r2, cache_pred_disable_mask
-	and	r4, r2
-	mcr	p15, 0, r4, c1, c0, 0
-	dsb
-	isb
-
-	ldr     r0, =restoremmu_on
-	bx      r0
-
-/*
- * ==============================
- * == Exit point from OFF mode ==
- * ==============================
- */
-restoremmu_on:
-
-	ldmfd	sp!, {r0-r12, pc}	@ restore regs and return
+kernel_flush:
+	.word v7_flush_dcache_all
+#endif
diff --git a/arch/arm/mach-shmobile/smp-sh73a0.c b/arch/arm/mach-shmobile/smp-sh73a0.c
index 3ffdbc9..be1ade7 100644
--- a/arch/arm/mach-shmobile/smp-sh73a0.c
+++ b/arch/arm/mach-shmobile/smp-sh73a0.c
@@ -74,6 +74,8 @@ void __cpuinit sh73a0_secondary_init(unsigned int cpu)
 
 int __cpuinit sh73a0_boot_secondary(unsigned int cpu)
 {
+	cpu = cpu_logical_map(cpu);
+
 	/* enable cache coherency */
 	modify_scu_cpu_psr(0, 3 << (cpu * 8));
 
@@ -87,6 +89,8 @@ int __cpuinit sh73a0_boot_secondary(unsigned int cpu)
 
 void __init sh73a0_smp_prepare_cpus(void)
 {
+	int cpu = cpu_logical_map(0);
+
 	scu_enable(scu_base_addr());
 
 	/* Map the reset vector (in headsmp.S) */
@@ -94,5 +98,5 @@ void __init sh73a0_smp_prepare_cpus(void)
 	__raw_writel(__pa(shmobile_secondary_vector), __io(SBAR));
 
 	/* enable cache coherency on CPU0 */
-	modify_scu_cpu_psr(0, 3 << (0 * 8));
+	modify_scu_cpu_psr(0, 3 << (cpu * 8));
 }