Merge branch 'next' into for-linus

11 files changed, 312 insertions, 90 deletions

diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
index 20979564e7ee..83c025e72ceb 100644
--- a/arch/arm/mm/Kconfig
+++ b/arch/arm/mm/Kconfig
@@ -391,7 +391,7 @@ config CPU_FEROCEON_OLD_ID
 
 # ARMv6
 config CPU_V6
-	bool "Support ARM V6 processor" if ARCH_INTEGRATOR || MACH_REALVIEW_EB
+	bool "Support ARM V6 processor" if ARCH_INTEGRATOR || MACH_REALVIEW_EB || MACH_REALVIEW_PBX
 	select CPU_32v6
 	select CPU_ABRT_EV6
 	select CPU_PABRT_NOIFAR
@@ -416,7 +416,7 @@ config CPU_32v6K
 
 # ARMv7
 config CPU_V7
-	bool "Support ARM V7 processor" if ARCH_INTEGRATOR || MACH_REALVIEW_EB
+	bool "Support ARM V7 processor" if ARCH_INTEGRATOR || MACH_REALVIEW_EB || MACH_REALVIEW_PBX
 	select CPU_32v6K
 	select CPU_32v7
 	select CPU_ABRT_EV7
@@ -639,10 +639,23 @@ config CPU_BIG_ENDIAN
 	  port must properly enable any big-endian related features
 	  of your chipset/board/processor.
 
+config CPU_ENDIAN_BE8
+	bool
+	depends on CPU_BIG_ENDIAN
+	default CPU_V6 || CPU_V7
+	help
+	  Support for the BE-8 (big-endian) mode on ARMv6 and ARMv7 processors.
+
+config CPU_ENDIAN_BE32
+	bool
+	depends on CPU_BIG_ENDIAN
+	default !CPU_ENDIAN_BE8
+	help
+	  Support for the BE-32 (big-endian) mode on pre-ARMv6 processors.
+
 config CPU_HIGH_VECTOR
 	depends on !MMU && CPU_CP15 && !CPU_ARM740T
 	bool "Select the High exception vector"
-	default n
 	help
 	  Say Y here to select high exception vector(0xFFFF0000~).
 	  The exception vector can be vary depending on the platform
@@ -726,7 +739,6 @@ config NEEDS_SYSCALL_FOR_CMPXCHG
 
 config OUTER_CACHE
 	bool
-	default n
 
 config CACHE_FEROCEON_L2
 	bool "Enable the Feroceon L2 cache controller"
@@ -739,7 +751,6 @@ config CACHE_FEROCEON_L2
 config CACHE_FEROCEON_L2_WRITETHROUGH
 	bool "Force Feroceon L2 cache write through"
 	depends on CACHE_FEROCEON_L2
-	default n
 	help
 	  Say Y here to use the Feroceon L2 cache in writethrough mode.
 	  Unless you specifically require this, say N for writeback mode.
@@ -747,7 +758,7 @@ config CACHE_FEROCEON_L2_WRITETHROUGH
 config CACHE_L2X0
 	bool "Enable the L2x0 outer cache controller"
 	depends on REALVIEW_EB_ARM11MP || MACH_REALVIEW_PB11MP || MACH_REALVIEW_PB1176 || \
-		   REALVIEW_EB_A9MP || ARCH_MX35 || ARCH_MX31
+		   REALVIEW_EB_A9MP || ARCH_MX35 || ARCH_MX31 || MACH_REALVIEW_PBX
 	default y
 	select OUTER_CACHE
 	help
diff --git a/arch/arm/mm/abort-ev6.S b/arch/arm/mm/abort-ev6.S
index 6f7e70907e44..f332df7f0d37 100644
--- a/arch/arm/mm/abort-ev6.S
+++ b/arch/arm/mm/abort-ev6.S
@@ -37,6 +37,9 @@ ENTRY(v6_early_abort)
 	movne	pc, lr
 	do_thumb_abort
 	ldreq	r3, [r2]			@ read aborted ARM instruction
+#ifdef CONFIG_CPU_ENDIAN_BE8
+	reveq	r3, r3
+#endif
 	do_ldrd_abort
 	tst	r3, #1 << 20			@ L = 0 -> write
 	orreq	r1, r1, #1 << 11		@ yes.
diff --git a/arch/arm/mm/alignment.c b/arch/arm/mm/alignment.c
index 3a398befed41..03cd27d917b9 100644
--- a/arch/arm/mm/alignment.c
+++ b/arch/arm/mm/alignment.c
@@ -62,6 +62,12 @@
 #define SHIFT_ASR	0x40
 #define SHIFT_RORRRX	0x60
 
+#define BAD_INSTR 	0xdeadc0de
+
+/* Thumb-2 32 bit format per ARMv7 DDI0406A A6.3, either f800h,e800h,f800h */
+#define IS_T32(hi16) \
+	(((hi16) & 0xe000) == 0xe000 && ((hi16) & 0x1800))
+
 static unsigned long ai_user;
 static unsigned long ai_sys;
 static unsigned long ai_skipped;
@@ -332,38 +338,48 @@ do_alignment_ldrdstrd(unsigned long addr, unsigned long instr,
 		      struct pt_regs *regs)
 {
 	unsigned int rd = RD_BITS(instr);
-
-	if (((rd & 1) == 1) || (rd == 14))
+	unsigned int rd2;
+	int load;
+
+	if ((instr & 0xfe000000) == 0xe8000000) {
+		/* ARMv7 Thumb-2 32-bit LDRD/STRD */
+		rd2 = (instr >> 8) & 0xf;
+		load = !!(LDST_L_BIT(instr));
+	} else if (((rd & 1) == 1) || (rd == 14))
 		goto bad;
+	else {
+		load = ((instr & 0xf0) == 0xd0);
+		rd2 = rd + 1;
+	}
 
 	ai_dword += 1;
 
 	if (user_mode(regs))
 		goto user;
 
-	if ((instr & 0xf0) == 0xd0) {
+	if (load) {
 		unsigned long val;
 		get32_unaligned_check(val, addr);
 		regs->uregs[rd] = val;
 		get32_unaligned_check(val, addr + 4);
-		regs->uregs[rd + 1] = val;
+		regs->uregs[rd2] = val;
 	} else {
 		put32_unaligned_check(regs->uregs[rd], addr);
-		put32_unaligned_check(regs->uregs[rd + 1], addr + 4);
+		put32_unaligned_check(regs->uregs[rd2], addr + 4);
 	}
 
 	return TYPE_LDST;
 
  user:
-	if ((instr & 0xf0) == 0xd0) {
+	if (load) {
 		unsigned long val;
 		get32t_unaligned_check(val, addr);
 		regs->uregs[rd] = val;
 		get32t_unaligned_check(val, addr + 4);
-		regs->uregs[rd + 1] = val;
+		regs->uregs[rd2] = val;
 	} else {
 		put32t_unaligned_check(regs->uregs[rd], addr);
-		put32t_unaligned_check(regs->uregs[rd + 1], addr + 4);
+		put32t_unaligned_check(regs->uregs[rd2], addr + 4);
 	}
 
 	return TYPE_LDST;
@@ -616,8 +632,72 @@ thumb2arm(u16 tinstr)
 		/* Else fall through for illegal instruction case */
 
 	default:
-		return 0xdeadc0de;
+		return BAD_INSTR;
+	}
+}
+
+/*
+ * Convert Thumb-2 32 bit LDM, STM, LDRD, STRD to equivalent instruction
+ * handlable by ARM alignment handler, also find the corresponding handler,
+ * so that we can reuse ARM userland alignment fault fixups for Thumb.
+ *
+ * @pinstr: original Thumb-2 instruction; returns new handlable instruction
+ * @regs: register context.
+ * @poffset: return offset from faulted addr for later writeback
+ *
+ * NOTES:
+ * 1. Comments below refer to ARMv7 DDI0406A Thumb Instruction sections.
+ * 2. Register name Rt from ARMv7 is same as Rd from ARMv6 (Rd is Rt)
+ */
+static void *
+do_alignment_t32_to_handler(unsigned long *pinstr, struct pt_regs *regs,
+			    union offset_union *poffset)
+{
+	unsigned long instr = *pinstr;
+	u16 tinst1 = (instr >> 16) & 0xffff;
+	u16 tinst2 = instr & 0xffff;
+	poffset->un = 0;
+
+	switch (tinst1 & 0xffe0) {
+	/* A6.3.5 Load/Store multiple */
+	case 0xe880:		/* STM/STMIA/STMEA,LDM/LDMIA, PUSH/POP T2 */
+	case 0xe8a0:		/* ...above writeback version */
+	case 0xe900:		/* STMDB/STMFD, LDMDB/LDMEA */
+	case 0xe920:		/* ...above writeback version */
+		/* no need offset decision since handler calculates it */
+		return do_alignment_ldmstm;
+
+	case 0xf840:		/* POP/PUSH T3 (single register) */
+		if (RN_BITS(instr) == 13 && (tinst2 & 0x09ff) == 0x0904) {
+			u32 L = !!(LDST_L_BIT(instr));
+			const u32 subset[2] = {
+				0xe92d0000,	/* STMDB sp!,{registers} */
+				0xe8bd0000,	/* LDMIA sp!,{registers} */
+			};
+			*pinstr = subset[L] | (1<<RD_BITS(instr));
+			return do_alignment_ldmstm;
+		}
+		/* Else fall through for illegal instruction case */
+		break;
+
+	/* A6.3.6 Load/store double, STRD/LDRD(immed, lit, reg) */
+	case 0xe860:
+	case 0xe960:
+	case 0xe8e0:
+	case 0xe9e0:
+		poffset->un = (tinst2 & 0xff) << 2;
+	case 0xe940:
+	case 0xe9c0:
+		return do_alignment_ldrdstrd;
+
+	/*
+	 * No need to handle load/store instructions up to word size
+	 * since ARMv6 and later CPUs can perform unaligned accesses.
+	 */
+	default:
+		break;
 	}
+	return NULL;
 }
 
 static int
@@ -630,6 +710,8 @@ do_alignment(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
 	mm_segment_t fs;
 	unsigned int fault;
 	u16 tinstr = 0;
+	int isize = 4;
+	int thumb2_32b = 0;
 
 	instrptr = instruction_pointer(regs);
 
@@ -637,8 +719,19 @@ do_alignment(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
 	set_fs(KERNEL_DS);
 	if (thumb_mode(regs)) {
 		fault = __get_user(tinstr, (u16 *)(instrptr & ~1));
-		if (!(fault))
-			instr = thumb2arm(tinstr);
+		if (!fault) {
+			if (cpu_architecture() >= CPU_ARCH_ARMv7 &&
+			    IS_T32(tinstr)) {
+				/* Thumb-2 32-bit */
+				u16 tinst2 = 0;
+				fault = __get_user(tinst2, (u16 *)(instrptr+2));
+				instr = (tinstr << 16) | tinst2;
+				thumb2_32b = 1;
+			} else {
+				isize = 2;
+				instr = thumb2arm(tinstr);
+			}
+		}
 	} else
 		fault = __get_user(instr, (u32 *)instrptr);
 	set_fs(fs);
@@ -655,7 +748,7 @@ do_alignment(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
 
  fixup:
 
-	regs->ARM_pc += thumb_mode(regs) ? 2 : 4;
+	regs->ARM_pc += isize;
 
 	switch (CODING_BITS(instr)) {
 	case 0x00000000:	/* 3.13.4 load/store instruction extensions */
@@ -714,18 +807,25 @@ do_alignment(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
 		handler = do_alignment_ldrstr;
 		break;
 
-	case 0x08000000:	/* ldm or stm */
-		handler = do_alignment_ldmstm;
+	case 0x08000000:	/* ldm or stm, or thumb-2 32bit instruction */
+		if (thumb2_32b)
+			handler = do_alignment_t32_to_handler(&instr, regs, &offset);
+		else
+			handler = do_alignment_ldmstm;
 		break;
 
 	default:
 		goto bad;
 	}
 
+	if (!handler)
+		goto bad;
 	type = handler(addr, instr, regs);
 
-	if (type == TYPE_ERROR || type == TYPE_FAULT)
+	if (type == TYPE_ERROR || type == TYPE_FAULT) {
+		regs->ARM_pc -= isize;
 		goto bad_or_fault;
+	}
 
 	if (type == TYPE_LDST)
 		do_alignment_finish_ldst(addr, instr, regs, offset);
@@ -735,7 +835,6 @@ do_alignment(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
  bad_or_fault:
 	if (type == TYPE_ERROR)
 		goto bad;
-	regs->ARM_pc -= thumb_mode(regs) ? 2 : 4;
 	/*
 	 * We got a fault - fix it up, or die.
 	 */
@@ -751,8 +850,8 @@ do_alignment(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
 	 */
 	printk(KERN_ERR "Alignment trap: not handling instruction "
 		"%0*lx at [<%08lx>]\n",
-		thumb_mode(regs) ? 4 : 8,
-		thumb_mode(regs) ? tinstr : instr, instrptr);
+		isize << 1,
+		isize == 2 ? tinstr : instr, instrptr);
 	ai_skipped += 1;
 	return 1;
 
@@ -763,8 +862,8 @@ do_alignment(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
 		printk("Alignment trap: %s (%d) PC=0x%08lx Instr=0x%0*lx "
 		       "Address=0x%08lx FSR 0x%03x\n", current->comm,
 			task_pid_nr(current), instrptr,
-		        thumb_mode(regs) ? 4 : 8,
-		        thumb_mode(regs) ? tinstr : instr,
+			isize << 1,
+			isize == 2 ? tinstr : instr,
 		        addr, fsr);
 
 	if (ai_usermode & UM_FIXUP)
diff --git a/arch/arm/mm/fault.c b/arch/arm/mm/fault.c
index 0455557a2899..6fdcbb709827 100644
--- a/arch/arm/mm/fault.c
+++ b/arch/arm/mm/fault.c
@@ -208,7 +208,7 @@ good_area:
 	 * than endlessly redo the fault.
 	 */
 survive:
-	fault = handle_mm_fault(mm, vma, addr & PAGE_MASK, fsr & (1 << 11));
+	fault = handle_mm_fault(mm, vma, addr & PAGE_MASK, (fsr & (1 << 11)) ? FAULT_FLAG_WRITE : 0);
 	if (unlikely(fault & VM_FAULT_ERROR)) {
 		if (fault & VM_FAULT_OOM)
 			goto out_of_memory;
diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c
index 8277802ec859..3a7279c1ce5e 100644
--- a/arch/arm/mm/init.c
+++ b/arch/arm/mm/init.c
@@ -120,6 +120,32 @@ void show_mem(void)
 	printk("%d pages swap cached\n", cached);
 }
 
+static void __init find_node_limits(int node, struct meminfo *mi,
+	unsigned long *min, unsigned long *max_low, unsigned long *max_high)
+{
+	int i;
+
+	*min = -1UL;
+	*max_low = *max_high = 0;
+
+	for_each_nodebank(i, mi, node) {
+		struct membank *bank = &mi->bank[i];
+		unsigned long start, end;
+
+		start = bank_pfn_start(bank);
+		end = bank_pfn_end(bank);
+
+		if (*min > start)
+			*min = start;
+		if (*max_high < end)
+			*max_high = end;
+		if (bank->highmem)
+			continue;
+		if (*max_low < end)
+			*max_low = end;
+	}
+}
+
 /*
  * FIXME: We really want to avoid allocating the bootmap bitmap
  * over the top of the initrd.  Hopefully, this is located towards
@@ -210,41 +236,25 @@ static inline void map_memory_bank(struct membank *bank)
 #endif
 }
 
-static unsigned long __init bootmem_init_node(int node, struct meminfo *mi)
+static void __init bootmem_init_node(int node, struct meminfo *mi,
+	unsigned long start_pfn, unsigned long end_pfn)
 {
-	unsigned long start_pfn, end_pfn, boot_pfn;
+	unsigned long boot_pfn;
 	unsigned int boot_pages;
 	pg_data_t *pgdat;
 	int i;
 
-	start_pfn = -1UL;
-	end_pfn = 0;
-
 	/*
-	 * Calculate the pfn range, and map the memory banks for this node.
+	 * Map the memory banks for this node.
 	 */
 	for_each_nodebank(i, mi, node) {
 		struct membank *bank = &mi->bank[i];
-		unsigned long start, end;
 
-		start = bank_pfn_start(bank);
-		end = bank_pfn_end(bank);
-
-		if (start_pfn > start)
-			start_pfn = start;
-		if (end_pfn < end)
-			end_pfn = end;
-
-		map_memory_bank(bank);
+		if (!bank->highmem)
+			map_memory_bank(bank);
 	}
 
 	/*
-	 * If there is no memory in this node, ignore it.
-	 */
-	if (end_pfn == 0)
-		return end_pfn;
-
-	/*
 	 * Allocate the bootmem bitmap page.
 	 */
 	boot_pages = bootmem_bootmap_pages(end_pfn - start_pfn);
@@ -260,7 +270,8 @@ static unsigned long __init bootmem_init_node(int node, struct meminfo *mi)
 
 	for_each_nodebank(i, mi, node) {
 		struct membank *bank = &mi->bank[i];
-		free_bootmem_node(pgdat, bank_phys_start(bank), bank_phys_size(bank));
+		if (!bank->highmem)
+			free_bootmem_node(pgdat, bank_phys_start(bank), bank_phys_size(bank));
 		memory_present(node, bank_pfn_start(bank), bank_pfn_end(bank));
 	}
 
@@ -269,8 +280,6 @@ static unsigned long __init bootmem_init_node(int node, struct meminfo *mi)
 	 */
 	reserve_bootmem_node(pgdat, boot_pfn << PAGE_SHIFT,
 			     boot_pages << PAGE_SHIFT, BOOTMEM_DEFAULT);
-
-	return end_pfn;
 }
 
 static void __init bootmem_reserve_initrd(int node)
@@ -297,33 +306,39 @@ static void __init bootmem_reserve_initrd(int node)
 static void __init bootmem_free_node(int node, struct meminfo *mi)
 {
 	unsigned long zone_size[MAX_NR_ZONES], zhole_size[MAX_NR_ZONES];
-	unsigned long start_pfn, end_pfn;
-	pg_data_t *pgdat = NODE_DATA(node);
+	unsigned long min, max_low, max_high;
 	int i;
 
-	start_pfn = pgdat->bdata->node_min_pfn;
-	end_pfn = pgdat->bdata->node_low_pfn;
+	find_node_limits(node, mi, &min, &max_low, &max_high);
 
 	/*
 	 * initialise the zones within this node.
 	 */
 	memset(zone_size, 0, sizeof(zone_size));
-	memset(zhole_size, 0, sizeof(zhole_size));
 
 	/*
 	 * The size of this node has already been determined.  If we need
 	 * to do anything fancy with the allocation of this memory to the
 	 * zones, now is the time to do it.
 	 */
-	zone_size[0] = end_pfn - start_pfn;
+	zone_size[0] = max_low - min;
+#ifdef CONFIG_HIGHMEM
+	zone_size[ZONE_HIGHMEM] = max_high - max_low;
+#endif
 
 	/*
 	 * For each bank in this node, calculate the size of the holes.
 	 *  holes = node_size - sum(bank_sizes_in_node)
 	 */
-	zhole_size[0] = zone_size[0];
-	for_each_nodebank(i, mi, node)
-		zhole_size[0] -= bank_pfn_size(&mi->bank[i]);
+	memcpy(zhole_size, zone_size, sizeof(zhole_size));
+	for_each_nodebank(i, mi, node) {
+		int idx = 0;
+#ifdef CONFIG_HIGHMEM
+		if (mi->bank[i].highmem)
+			idx = ZONE_HIGHMEM;
+#endif
+		zhole_size[idx] -= bank_pfn_size(&mi->bank[i]);
+	}
 
 	/*
 	 * Adjust the sizes according to any special requirements for
@@ -331,13 +346,13 @@ static void __init bootmem_free_node(int node, struct meminfo *mi)
 	 */
 	arch_adjust_zones(node, zone_size, zhole_size);
 
-	free_area_init_node(node, zone_size, start_pfn, zhole_size);
+	free_area_init_node(node, zone_size, min, zhole_size);
 }
 
 void __init bootmem_init(void)
 {
 	struct meminfo *mi = &meminfo;
-	unsigned long memend_pfn = 0;
+	unsigned long min, max_low, max_high;
 	int node, initrd_node;
 
 	/*
@@ -345,11 +360,29 @@ void __init bootmem_init(void)
 	 */
 	initrd_node = check_initrd(mi);
 
+	max_low = max_high = 0;
+
 	/*
 	 * Run through each node initialising the bootmem allocator.
 	 */
 	for_each_node(node) {
-		unsigned long end_pfn = bootmem_init_node(node, mi);
+		unsigned long node_low, node_high;
+
+		find_node_limits(node, mi, &min, &node_low, &node_high);
+
+		if (node_low > max_low)
+			max_low = node_low;
+		if (node_high > max_high)
+			max_high = node_high;
+
+		/*
+		 * If there is no memory in this node, ignore it.
+		 * (We can't have nodes which have no lowmem)
+		 */
+		if (node_low == 0)
+			continue;
+
+		bootmem_init_node(node, mi, min, node_low);
 
 		/*
 		 * Reserve any special node zero regions.
@@ -362,12 +395,6 @@ void __init bootmem_init(void)
 		 */
 		if (node == initrd_node)
 			bootmem_reserve_initrd(node);
-
-		/*
-		 * Remember the highest memory PFN.
-		 */
-		if (end_pfn > memend_pfn)
-			memend_pfn = end_pfn;
 	}
 
 	/*
@@ -383,7 +410,7 @@ void __init bootmem_init(void)
 	for_each_node(node)
 		bootmem_free_node(node, mi);
 
-	high_memory = __va((memend_pfn << PAGE_SHIFT) - 1) + 1;
+	high_memory = __va((max_low << PAGE_SHIFT) - 1) + 1;
 
 	/*
 	 * This doesn't seem to be used by the Linux memory manager any
@@ -393,7 +420,8 @@ void __init bootmem_init(void)
 	 * Note: max_low_pfn and max_pfn reflect the number of _pages_ in
 	 * the system, not the maximum PFN.
 	 */
-	max_pfn = max_low_pfn = memend_pfn - PHYS_PFN_OFFSET;
+	max_low_pfn = max_low - PHYS_PFN_OFFSET;
+	max_pfn = max_high - PHYS_PFN_OFFSET;
 }
 
 static inline int free_area(unsigned long pfn, unsigned long end, char *s)
diff --git a/arch/arm/mm/ioremap.c b/arch/arm/mm/ioremap.c
index 9f88dd3be601..0ab75c60f7cf 100644
--- a/arch/arm/mm/ioremap.c
+++ b/arch/arm/mm/ioremap.c
@@ -110,6 +110,12 @@ static int remap_area_pages(unsigned long start, unsigned long pfn,
 	return err;
 }
 
+int ioremap_page(unsigned long virt, unsigned long phys,
+		 const struct mem_type *mtype)
+{
+	return remap_area_pages(virt, __phys_to_pfn(phys), PAGE_SIZE, mtype);
+}
+EXPORT_SYMBOL(ioremap_page);
 
 void __check_kvm_seq(struct mm_struct *mm)
 {
diff --git a/arch/arm/mm/mmu.c b/arch/arm/mm/mmu.c
index e6344ece00ce..4426ee67ceca 100644
--- a/arch/arm/mm/mmu.c
+++ b/arch/arm/mm/mmu.c
@@ -255,6 +255,7 @@ const struct mem_type *get_mem_type(unsigned int type)
 {
 	return type < ARRAY_SIZE(mem_types) ? &mem_types[type] : NULL;
 }
+EXPORT_SYMBOL(get_mem_type);
 
 /*
  * Adjust the PMD section entries according to the CPU in use.
@@ -686,13 +687,19 @@ __early_param("vmalloc=", early_vmalloc);
 
 static void __init sanity_check_meminfo(void)
 {
-	int i, j;
+	int i, j, highmem = 0;
 
 	for (i = 0, j = 0; i < meminfo.nr_banks; i++) {
 		struct membank *bank = &meminfo.bank[j];
 		*bank = meminfo.bank[i];
 
 #ifdef CONFIG_HIGHMEM
+		if (__va(bank->start) > VMALLOC_MIN ||
+		    __va(bank->start) < (void *)PAGE_OFFSET)
+			highmem = 1;
+
+		bank->highmem = highmem;
+
 		/*
 		 * Split those memory banks which are partially overlapping
 		 * the vmalloc area greatly simplifying things later.
@@ -713,6 +720,7 @@ static void __init sanity_check_meminfo(void)
 				i++;
 				bank[1].size -= VMALLOC_MIN - __va(bank->start);
 				bank[1].start = __pa(VMALLOC_MIN - 1) + 1;
+				bank[1].highmem = highmem = 1;
 				j++;
 			}
 			bank->size = VMALLOC_MIN - __va(bank->start);
@@ -835,10 +843,31 @@ void __init reserve_node_zero(pg_data_t *pgdat)
 				BOOTMEM_EXCLUSIVE);
 	}
 
+	if (machine_is_treo680()) {
+		reserve_bootmem_node(pgdat, 0xa0000000, 0x1000,
+				BOOTMEM_EXCLUSIVE);
+		reserve_bootmem_node(pgdat, 0xa2000000, 0x1000,
+				BOOTMEM_EXCLUSIVE);
+	}
+
 	if (machine_is_palmt5())
 		reserve_bootmem_node(pgdat, 0xa0200000, 0x1000,
 				BOOTMEM_EXCLUSIVE);
 
+	/*
+	 * U300 - This platform family can share physical memory
+	 * between two ARM cpus, one running Linux and the other
+	 * running another OS.
+	 */
+	if (machine_is_u300()) {
+#ifdef CONFIG_MACH_U300_SINGLE_RAM
+#if ((CONFIG_MACH_U300_ACCESS_MEM_SIZE & 1) == 1) &&	\
+	CONFIG_MACH_U300_2MB_ALIGNMENT_FIX
+		res_size = 0x00100000;
+#endif
+#endif
+	}
+
 #ifdef CONFIG_SA1111
 	/*
 	 * Because of the SA1111 DMA bug, we want to preserve our
diff --git a/arch/arm/mm/proc-syms.c b/arch/arm/mm/proc-syms.c
index 195e48edd8c2..ac5c80062b70 100644
--- a/arch/arm/mm/proc-syms.c
+++ b/arch/arm/mm/proc-syms.c
@@ -27,6 +27,7 @@ EXPORT_SYMBOL(__cpuc_flush_kern_all);
 EXPORT_SYMBOL(__cpuc_flush_user_all);
 EXPORT_SYMBOL(__cpuc_flush_user_range);
 EXPORT_SYMBOL(__cpuc_coherent_kern_range);
+EXPORT_SYMBOL(__cpuc_flush_dcache_page);
 EXPORT_SYMBOL(dmac_inv_range);  /* because of flush_ioremap_region() */
 #else
 EXPORT_SYMBOL(cpu_cache);
diff --git a/arch/arm/mm/proc-v6.S b/arch/arm/mm/proc-v6.S
index 087e239704df..524ddae92595 100644
--- a/arch/arm/mm/proc-v6.S
+++ b/arch/arm/mm/proc-v6.S
@@ -170,6 +170,9 @@ __v6_setup:
 #endif /* CONFIG_MMU */
 	adr	r5, v6_crval
 	ldmia	r5, {r5, r6}
+#ifdef CONFIG_CPU_ENDIAN_BE8
+	orr	r6, r6, #1 << 25		@ big-endian page tables
+#endif
 	mrc	p15, 0, r0, c1, c0, 0		@ read control register
 	bic	r0, r0, r5			@ clear bits them
 	orr	r0, r0, r6			@ set them
diff --git a/arch/arm/mm/proc-v7.S b/arch/arm/mm/proc-v7.S
index a08d9d2380d3..180a08d03a03 100644
--- a/arch/arm/mm/proc-v7.S
+++ b/arch/arm/mm/proc-v7.S
@@ -19,17 +19,23 @@
 
 #include "proc-macros.S"
 
-#define TTB_C		(1 << 0)
 #define TTB_S		(1 << 1)
 #define TTB_RGN_NC	(0 << 3)
 #define TTB_RGN_OC_WBWA	(1 << 3)
 #define TTB_RGN_OC_WT	(2 << 3)
 #define TTB_RGN_OC_WB	(3 << 3)
+#define TTB_NOS		(1 << 5)
+#define TTB_IRGN_NC	((0 << 0) | (0 << 6))
+#define TTB_IRGN_WBWA	((0 << 0) | (1 << 6))
+#define TTB_IRGN_WT	((1 << 0) | (0 << 6))
+#define TTB_IRGN_WB	((1 << 0) | (1 << 6))
 
 #ifndef CONFIG_SMP
-#define TTB_FLAGS	TTB_C|TTB_RGN_OC_WB		@ mark PTWs cacheable, outer WB
+/* PTWs cacheable, inner WB not shareable, outer WB not shareable */
+#define TTB_FLAGS	TTB_IRGN_WB|TTB_RGN_OC_WB
 #else
-#define TTB_FLAGS	TTB_C|TTB_S|TTB_RGN_OC_WBWA	@ mark PTWs cacheable and shared, outer WBWA
+/* PTWs cacheable, inner WBWA shareable, outer WBWA not shareable */
+#define TTB_FLAGS	TTB_IRGN_WBWA|TTB_S|TTB_NOS|TTB_RGN_OC_WBWA
 #endif
 
 ENTRY(cpu_v7_proc_init)
@@ -176,8 +182,8 @@ cpu_v7_name:
  */
 __v7_setup:
 #ifdef CONFIG_SMP
-	mrc	p15, 0, r0, c1, c0, 1		@ Enable SMP/nAMP mode
-	orr	r0, r0, #(0x1 << 6)
+	mrc	p15, 0, r0, c1, c0, 1		@ Enable SMP/nAMP mode and
+	orr	r0, r0, #(1 << 6) | (1 << 0)	@ TLB ops broadcasting
 	mcr	p15, 0, r0, c1, c0, 1
 #endif
 	adr	r12, __v7_setup_stack		@ the local stack
@@ -227,12 +233,43 @@ __v7_setup:
 	mov	r10, #0x1f			@ domains 0, 1 = manager
 	mcr	p15, 0, r10, c3, c0, 0		@ load domain access register
 #endif
-	ldr	r5, =0xff0aa1a8
-	ldr	r6, =0x40e040e0
+	/*
+	 * Memory region attributes with SCTLR.TRE=1
+	 *
+	 *   n = TEX[0],C,B
+	 *   TR = PRRR[2n+1:2n]		- memory type
+	 *   IR = NMRR[2n+1:2n]		- inner cacheable property
+	 *   OR = NMRR[2n+17:2n+16]	- outer cacheable property
+	 *
+	 *			n	TR	IR	OR
+	 *   UNCACHED		000	00
+	 *   BUFFERABLE		001	10	00	00
+	 *   WRITETHROUGH	010	10	10	10
+	 *   WRITEBACK		011	10	11	11
+	 *   reserved		110
+	 *   WRITEALLOC		111	10	01	01
+	 *   DEV_SHARED		100	01
+	 *   DEV_NONSHARED	100	01
+	 *   DEV_WC		001	10
+	 *   DEV_CACHED		011	10
+	 *
+	 * Other attributes:
+	 *
+	 *   DS0 = PRRR[16] = 0		- device shareable property
+	 *   DS1 = PRRR[17] = 1		- device shareable property
+	 *   NS0 = PRRR[18] = 0		- normal shareable property
+	 *   NS1 = PRRR[19] = 1		- normal shareable property
+	 *   NOS = PRRR[24+n] = 1	- not outer shareable
+	 */
+	ldr	r5, =0xff0a81a8			@ PRRR
+	ldr	r6, =0x40e040e0			@ NMRR
 	mcr	p15, 0, r5, c10, c2, 0		@ write PRRR
 	mcr	p15, 0, r6, c10, c2, 1		@ write NMRR
 	adr	r5, v7_crval
 	ldmia	r5, {r5, r6}
+#ifdef CONFIG_CPU_ENDIAN_BE8
+	orr	r6, r6, #1 << 25		@ big-endian page tables
+#endif
    	mrc	p15, 0, r0, c1, c0, 0		@ read control register
 	bic	r0, r0, r5			@ clear bits them
 	orr	r0, r0, r6			@ set them
@@ -240,14 +277,14 @@ __v7_setup:
 ENDPROC(__v7_setup)
 
 	/*   AT
-	 *  TFR   EV X F   I D LR
-	 * .EEE ..EE PUI. .T.T 4RVI ZFRS BLDP WCAM
+	 *  TFR   EV X F   I D LR    S
+	 * .EEE ..EE PUI. .T.T 4RVI ZWRS BLDP WCAM
 	 * rxxx rrxx xxx0 0101 xxxx xxxx x111 xxxx < forced
-	 *    1    0 110       0011 1.00 .111 1101 < we want
+	 *    1    0 110       0011 1100 .111 1101 < we want
 	 */
 	.type	v7_crval, #object
 v7_crval:
-	crval	clear=0x0120c302, mmuset=0x10c0387d, ucset=0x00c0187c
+	crval	clear=0x0120c302, mmuset=0x10c03c7d, ucset=0x00c01c7c
 
 __v7_setup_stack:
 	.space	4 * 11				@ 11 registers
diff --git a/arch/arm/mm/tlb-v7.S b/arch/arm/mm/tlb-v7.S
index b637e7380ab7..a26a605b73bd 100644
--- a/arch/arm/mm/tlb-v7.S
+++ b/arch/arm/mm/tlb-v7.S
@@ -42,9 +42,11 @@ ENTRY(v7wbi_flush_user_tlb_range)
 	mov	r1, r1, lsl #PAGE_SHIFT
 	vma_vm_flags r2, r2			@ get vma->vm_flags
 1:
-	mcr	p15, 0, r0, c8, c6, 1		@ TLB invalidate D MVA (was 1)
-	tst	r2, #VM_EXEC			@ Executable area ?
-	mcrne	p15, 0, r0, c8, c5, 1		@ TLB invalidate I MVA (was 1)
+#ifdef CONFIG_SMP
+	mcr	p15, 0, r0, c8, c3, 1		@ TLB invalidate U MVA (shareable) 
+#else
+	mcr	p15, 0, r0, c8, c7, 1		@ TLB invalidate U MVA
+#endif
 	add	r0, r0, #PAGE_SZ
 	cmp	r0, r1
 	blo	1b
@@ -69,8 +71,11 @@ ENTRY(v7wbi_flush_kern_tlb_range)
 	mov	r0, r0, lsl #PAGE_SHIFT
 	mov	r1, r1, lsl #PAGE_SHIFT
 1:
-	mcr	p15, 0, r0, c8, c6, 1		@ TLB invalidate D MVA
-	mcr	p15, 0, r0, c8, c5, 1		@ TLB invalidate I MVA
+#ifdef CONFIG_SMP
+	mcr	p15, 0, r0, c8, c3, 1		@ TLB invalidate U MVA (shareable)
+#else
+	mcr	p15, 0, r0, c8, c7, 1		@ TLB invalidate U MVA
+#endif
 	add	r0, r0, #PAGE_SZ
 	cmp	r0, r1
 	blo	1b
@@ -87,5 +92,5 @@ ENDPROC(v7wbi_flush_kern_tlb_range)
 ENTRY(v7wbi_tlb_fns)
 	.long	v7wbi_flush_user_tlb_range
 	.long	v7wbi_flush_kern_tlb_range
-	.long	v6wbi_tlb_flags
+	.long	v7wbi_tlb_flags
 	.size	v7wbi_tlb_fns, . - v7wbi_tlb_fns