diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 173159e93c99..0ec034933cae 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -64,6 +64,7 @@ config ARM
 	select GENERIC_CLOCKEVENTS_BROADCAST if SMP
 	select GENERIC_IRQ_IPI if SMP
 	select GENERIC_CPU_AUTOPROBE
+	select GENERIC_CPU_DEVICES
 	select GENERIC_EARLY_IOREMAP
 	select GENERIC_IDLE_POLL_SETUP
 	select GENERIC_IRQ_MULTI_HANDLER
diff --git a/arch/arm/include/asm/cpu.h b/arch/arm/include/asm/cpu.h
index bd6fdb4b922d..9d8863537aa5 100644
--- a/arch/arm/include/asm/cpu.h
+++ b/arch/arm/include/asm/cpu.h
@@ -11,7 +11,6 @@
 #include <linux/cpu.h>
 
 struct cpuinfo_arm {
-	struct cpu	cpu;
 	u32		cpuid;
 #ifdef CONFIG_SMP
 	unsigned int	loops_per_jiffy;
diff --git a/arch/arm/include/asm/dma-iommu.h b/arch/arm/include/asm/dma-iommu.h
index 82ec1ccf1fee..2ce4c5683e6d 100644
--- a/arch/arm/include/asm/dma-iommu.h
+++ b/arch/arm/include/asm/dma-iommu.h
@@ -24,7 +24,7 @@ struct dma_iommu_mapping {
 };
 
 struct dma_iommu_mapping *
-arm_iommu_create_mapping(const struct bus_type *bus, dma_addr_t base, u64 size);
+arm_iommu_create_mapping(struct device *dev, dma_addr_t base, u64 size);
 
 void arm_iommu_release_mapping(struct dma_iommu_mapping *mapping);
 
diff --git a/arch/arm/include/asm/pgtable-3level-hwdef.h b/arch/arm/include/asm/pgtable-3level-hwdef.h
index dfab3e982cbf..944fc9955528 100644
--- a/arch/arm/include/asm/pgtable-3level-hwdef.h
+++ b/arch/arm/include/asm/pgtable-3level-hwdef.h
@@ -106,6 +106,11 @@
 
 /*
  * TTBCR register bits.
+ *
+ * The ORGN0 and IRGN0 bits enables different forms of caching when
+ * walking the translation table. Clearing these bits (which is claimed
+ * to be the reset default) means "normal memory, [outer|inner]
+ * non-cacheable"
  */
 #define TTBCR_EAE		(1 << 31)
 #define TTBCR_IMP		(1 << 30)
diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c
index 7b33b157fca0..e6a857bf0ce6 100644
--- a/arch/arm/kernel/setup.c
+++ b/arch/arm/kernel/setup.c
@@ -1201,20 +1201,10 @@ void __init setup_arch(char **cmdline_p)
 		mdesc->init_early();
 }
 
-
-static int __init topology_init(void)
+bool arch_cpu_is_hotpluggable(int num)
 {
-	int cpu;
-
-	for_each_possible_cpu(cpu) {
-		struct cpuinfo_arm *cpuinfo = &per_cpu(cpu_data, cpu);
-		cpuinfo->cpu.hotpluggable = platform_can_hotplug_cpu(cpu);
-		register_cpu(&cpuinfo->cpu, cpu);
-	}
-
-	return 0;
+	return platform_can_hotplug_cpu(num);
 }
-subsys_initcall(topology_init);
 
 #ifdef CONFIG_HAVE_PROC_CPU
 static int __init proc_cpu_init(void)
diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
index 5adf1769eee4..88c2d68a69c9 100644
--- a/arch/arm/mm/dma-mapping.c
+++ b/arch/arm/mm/dma-mapping.c
@@ -1532,7 +1532,7 @@ static const struct dma_map_ops iommu_ops = {
 
 /**
  * arm_iommu_create_mapping
- * @bus: pointer to the bus holding the client device (for IOMMU calls)
+ * @dev: pointer to the client device (for IOMMU calls)
  * @base: start address of the valid IO address space
  * @size: maximum size of the valid IO address space
  *
@@ -1544,7 +1544,7 @@ static const struct dma_map_ops iommu_ops = {
  * arm_iommu_attach_device function.
  */
 struct dma_iommu_mapping *
-arm_iommu_create_mapping(const struct bus_type *bus, dma_addr_t base, u64 size)
+arm_iommu_create_mapping(struct device *dev, dma_addr_t base, u64 size)
 {
 	unsigned int bits = size >> PAGE_SHIFT;
 	unsigned int bitmap_size = BITS_TO_LONGS(bits) * sizeof(long);
@@ -1585,9 +1585,11 @@ arm_iommu_create_mapping(const struct bus_type *bus, dma_addr_t base, u64 size)
 
 	spin_lock_init(&mapping->lock);
 
-	mapping->domain = iommu_domain_alloc(bus);
-	if (!mapping->domain)
+	mapping->domain = iommu_paging_domain_alloc(dev);
+	if (IS_ERR(mapping->domain)) {
+		err = PTR_ERR(mapping->domain);
 		goto err4;
+	}
 
 	kref_init(&mapping->kref);
 	return mapping;
@@ -1718,7 +1720,7 @@ static void arm_setup_iommu_dma_ops(struct device *dev)
 		dma_base = dma_range_map_min(dev->dma_range_map);
 		size = dma_range_map_max(dev->dma_range_map) - dma_base;
 	}
-	mapping = arm_iommu_create_mapping(dev->bus, dma_base, size);
+	mapping = arm_iommu_create_mapping(dev, dma_base, size);
 	if (IS_ERR(mapping)) {
 		pr_warn("Failed to create %llu-byte IOMMU mapping for device %s\n",
 				size, dev_name(dev));
diff --git a/arch/arm/mm/mmu.c b/arch/arm/mm/mmu.c
index 3f774856ca67..f85c177cdf8d 100644
--- a/arch/arm/mm/mmu.c
+++ b/arch/arm/mm/mmu.c
@@ -1638,7 +1638,7 @@ static void __init early_paging_init(const struct machine_desc *mdesc)
 {
 	pgtables_remap *lpae_pgtables_remap;
 	unsigned long pa_pgd;
-	unsigned int cr, ttbcr;
+	u32 cr, ttbcr, tmp;
 	long long offset;
 
 	if (!mdesc->pv_fixup)
@@ -1688,7 +1688,9 @@ static void __init early_paging_init(const struct machine_desc *mdesc)
 	cr = get_cr();
 	set_cr(cr & ~(CR_I | CR_C));
 	ttbcr = cpu_get_ttbcr();
-	cpu_set_ttbcr(ttbcr & ~(3 << 8 | 3 << 10));
+	/* Disable all kind of caching of the translation table */
+	tmp = ttbcr & ~(TTBCR_ORGN0_MASK | TTBCR_IRGN0_MASK);
+	cpu_set_ttbcr(tmp);
 	flush_cache_all();
 
 	/*
diff --git a/arch/arm/vfp/vfpinstr.h b/arch/arm/vfp/vfpinstr.h
index 3c7938fd40aa..32090b0fb250 100644
--- a/arch/arm/vfp/vfpinstr.h
+++ b/arch/arm/vfp/vfpinstr.h
@@ -64,33 +64,37 @@
 
 #ifdef CONFIG_AS_VFP_VMRS_FPINST
 
-#define fmrx(_vfp_) ({			\
-	u32 __v;			\
-	asm(".fpu	vfpv2\n"	\
-	    "vmrs	%0, " #_vfp_	\
-	    : "=r" (__v) : : "cc");	\
-	__v;				\
- })
+#define fmrx(_vfp_) ({				\
+	u32 __v;				\
+	asm volatile (".fpu	vfpv2\n"	\
+		      "vmrs	%0, " #_vfp_	\
+		     : "=r" (__v) : : "cc");	\
+	__v;					\
+})
 
-#define fmxr(_vfp_,_var_)		\
-	asm(".fpu	vfpv2\n"	\
-	    "vmsr	" #_vfp_ ", %0"	\
-	   : : "r" (_var_) : "cc")
+#define fmxr(_vfp_, _var_) ({			\
+	asm volatile (".fpu	vfpv2\n"	\
+		      "vmsr	" #_vfp_ ", %0"	\
+		     : : "r" (_var_) : "cc");	\
+})
 
 #else
 
 #define vfpreg(_vfp_) #_vfp_
 
-#define fmrx(_vfp_) ({			\
-	u32 __v;			\
-	asm("mrc p10, 7, %0, " vfpreg(_vfp_) ", cr0, 0 @ fmrx	%0, " #_vfp_	\
-	    : "=r" (__v) : : "cc");	\
-	__v;				\
- })
+#define fmrx(_vfp_) ({						\
+	u32 __v;						\
+	asm volatile ("mrc p10, 7, %0, " vfpreg(_vfp_) ","	\
+		      "cr0, 0 @ fmrx	%0, " #_vfp_		\
+		     : "=r" (__v) : : "cc");			\
+	__v;							\
+})
 
-#define fmxr(_vfp_,_var_)		\
-	asm("mcr p10, 7, %0, " vfpreg(_vfp_) ", cr0, 0 @ fmxr	" #_vfp_ ", %0"	\
-	   : : "r" (_var_) : "cc")
+#define fmxr(_vfp_, _var_) ({					\
+	asm volatile ("mcr p10, 7, %0, " vfpreg(_vfp_) ","	\
+		      "cr0, 0 @ fmxr	" #_vfp_ ", %0"		\
+		     : : "r" (_var_) : "cc");			\
+})
 
 #endif
 
diff --git a/drivers/amba/bus.c b/drivers/amba/bus.c
index 34bc880ca20b..0230c43377c1 100644
--- a/drivers/amba/bus.c
+++ b/drivers/amba/bus.c
@@ -435,7 +435,7 @@ static const struct dev_pm_ops amba_pm = {
  * DMA configuration for platform and AMBA bus is same. So here we reuse
  * platform's DMA config routine.
  */
-struct bus_type amba_bustype = {
+const struct bus_type amba_bustype = {
 	.name		= "amba",
 	.dev_groups	= amba_dev_groups,
 	.match		= amba_match,
diff --git a/drivers/gpu/drm/exynos/exynos_drm_dma.c b/drivers/gpu/drm/exynos/exynos_drm_dma.c
index e2c7373f20c6..6a6761935224 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_dma.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_dma.c
@@ -110,7 +110,7 @@ int exynos_drm_register_dma(struct drm_device *drm, struct device *dev,
 		void *mapping = NULL;
 
 		if (IS_ENABLED(CONFIG_ARM_DMA_USE_IOMMU))
-			mapping = arm_iommu_create_mapping(&platform_bus_type,
+			mapping = arm_iommu_create_mapping(dev,
 				EXYNOS_DEV_ADDR_START, EXYNOS_DEV_ADDR_SIZE);
 		else if (IS_ENABLED(CONFIG_IOMMU_DMA))
 			mapping = iommu_get_domain_for_dev(priv->dma_dev);
diff --git a/drivers/iommu/ipmmu-vmsa.c b/drivers/iommu/ipmmu-vmsa.c
index b657cc09605f..ff55b8c30712 100644
--- a/drivers/iommu/ipmmu-vmsa.c
+++ b/drivers/iommu/ipmmu-vmsa.c
@@ -804,8 +804,7 @@ static int ipmmu_init_arm_mapping(struct device *dev)
 	if (!mmu->mapping) {
 		struct dma_iommu_mapping *mapping;
 
-		mapping = arm_iommu_create_mapping(&platform_bus_type,
-						   SZ_1G, SZ_2G);
+		mapping = arm_iommu_create_mapping(dev, SZ_1G, SZ_2G);
 		if (IS_ERR(mapping)) {
 			dev_err(mmu->dev, "failed to create ARM IOMMU mapping\n");
 			ret = PTR_ERR(mapping);
diff --git a/drivers/iommu/mtk_iommu_v1.c b/drivers/iommu/mtk_iommu_v1.c
index c6ea5b4baff3..ee4e55b6b190 100644
--- a/drivers/iommu/mtk_iommu_v1.c
+++ b/drivers/iommu/mtk_iommu_v1.c
@@ -433,8 +433,7 @@ static int mtk_iommu_v1_create_mapping(struct device *dev,
 	mtk_mapping = data->mapping;
 	if (!mtk_mapping) {
 		/* MTK iommu support 4GB iova address space. */
-		mtk_mapping = arm_iommu_create_mapping(&platform_bus_type,
-						0, 1ULL << 32);
+		mtk_mapping = arm_iommu_create_mapping(dev, 0, 1ULL << 32);
 		if (IS_ERR(mtk_mapping))
 			return PTR_ERR(mtk_mapping);
 
diff --git a/drivers/media/platform/ti/omap3isp/isp.c b/drivers/media/platform/ti/omap3isp/isp.c
index 1cda23244c7b..91101ba88ef0 100644
--- a/drivers/media/platform/ti/omap3isp/isp.c
+++ b/drivers/media/platform/ti/omap3isp/isp.c
@@ -1965,7 +1965,7 @@ static int isp_attach_iommu(struct isp_device *isp)
 	 * Create the ARM mapping, used by the ARM DMA mapping core to allocate
 	 * VAs. This will allocate a corresponding IOMMU domain.
 	 */
-	mapping = arm_iommu_create_mapping(&platform_bus_type, SZ_1G, SZ_2G);
+	mapping = arm_iommu_create_mapping(isp->dev, SZ_1G, SZ_2G);
 	if (IS_ERR(mapping)) {
 		dev_err(isp->dev, "failed to create ARM IOMMU mapping\n");
 		return PTR_ERR(mapping);
diff --git a/include/linux/amba/bus.h b/include/linux/amba/bus.h
index 958a55bcc708..dda2f3ea89cb 100644
--- a/include/linux/amba/bus.h
+++ b/include/linux/amba/bus.h
@@ -105,7 +105,7 @@ enum amba_vendor {
 	AMBA_VENDOR_LSI = 0xb6,
 };
 
-extern struct bus_type amba_bustype;
+extern const struct bus_type amba_bustype;
 
 #define to_amba_device(d)	container_of_const(d, struct amba_device, dev)