Merge branch 'samsung/cleanup' into next/cleanup2

21 files changed, 648 insertions, 463 deletions

diff --git a/arch/arm/kernel/Makefile b/arch/arm/kernel/Makefile
index 16eed6aebfa4..43b740d0e374 100644
--- a/arch/arm/kernel/Makefile
+++ b/arch/arm/kernel/Makefile
@@ -13,7 +13,7 @@ CFLAGS_REMOVE_return_address.o = -pg
 
 # Object file lists.
 
-obj-y		:= elf.o entry-armv.o entry-common.o irq.o \
+obj-y		:= elf.o entry-armv.o entry-common.o irq.o opcodes.o \
 		   process.o ptrace.o return_address.o setup.o signal.o \
 		   sys_arm.o stacktrace.o time.o traps.o
 
diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S
index b145f16c91bc..3a456c6c7005 100644
--- a/arch/arm/kernel/entry-armv.S
+++ b/arch/arm/kernel/entry-armv.S
@@ -36,12 +36,11 @@
 #ifdef CONFIG_MULTI_IRQ_HANDLER
 	ldr	r1, =handle_arch_irq
 	mov	r0, sp
-	ldr	r1, [r1]
 	adr	lr, BSYM(9997f)
-	teq	r1, #0
-	movne	pc, r1
-#endif
+	ldr	pc, [r1]
+#else
 	arch_irq_handler_default
+#endif
 9997:
 	.endm
 
diff --git a/arch/arm/kernel/head.S b/arch/arm/kernel/head.S
index 08c82fd844a8..14e277d2ff91 100644
--- a/arch/arm/kernel/head.S
+++ b/arch/arm/kernel/head.S
@@ -39,8 +39,14 @@
 #error KERNEL_RAM_VADDR must start at 0xXXXX8000
 #endif
 
+#ifdef CONFIG_ARM_LPAE
+	/* LPAE requires an additional page for the PGD */
+#define PG_DIR_SIZE	0x5000
+#define PMD_ORDER	3
+#else
 #define PG_DIR_SIZE	0x4000
 #define PMD_ORDER	2
+#endif
 
 	.globl	swapper_pg_dir
 	.equ	swapper_pg_dir, KERNEL_RAM_VADDR - PG_DIR_SIZE
@@ -164,17 +170,36 @@ __create_page_tables:
 	teq	r0, r6
 	bne	1b
 
+#ifdef CONFIG_ARM_LPAE
+	/*
+	 * Build the PGD table (first level) to point to the PMD table. A PGD
+	 * entry is 64-bit wide.
+	 */
+	mov	r0, r4
+	add	r3, r4, #0x1000			@ first PMD table address
+	orr	r3, r3, #3			@ PGD block type
+	mov	r6, #4				@ PTRS_PER_PGD
+	mov	r7, #1 << (55 - 32)		@ L_PGD_SWAPPER
+1:	str	r3, [r0], #4			@ set bottom PGD entry bits
+	str	r7, [r0], #4			@ set top PGD entry bits
+	add	r3, r3, #0x1000			@ next PMD table
+	subs	r6, r6, #1
+	bne	1b
+
+	add	r4, r4, #0x1000			@ point to the PMD tables
+#endif
+
 	ldr	r7, [r10, #PROCINFO_MM_MMUFLAGS] @ mm_mmuflags
 
 	/*
 	 * Create identity mapping to cater for __enable_mmu.
 	 * This identity mapping will be removed by paging_init().
 	 */
-	adr	r0, __enable_mmu_loc
+	adr	r0, __turn_mmu_on_loc
 	ldmia	r0, {r3, r5, r6}
 	sub	r0, r0, r3			@ virt->phys offset
-	add	r5, r5, r0			@ phys __enable_mmu
-	add	r6, r6, r0			@ phys __enable_mmu_end
+	add	r5, r5, r0			@ phys __turn_mmu_on
+	add	r6, r6, r0			@ phys __turn_mmu_on_end
 	mov	r5, r5, lsr #SECTION_SHIFT
 	mov	r6, r6, lsr #SECTION_SHIFT
 
@@ -219,8 +244,8 @@ __create_page_tables:
 #endif
 
 	/*
-	 * Then map boot params address in r2 or
-	 * the first 1MB of ram if boot params address is not specified.
+	 * Then map boot params address in r2 or the first 1MB (2MB with LPAE)
+	 * of ram if boot params address is not specified.
 	 */
 	mov	r0, r2, lsr #SECTION_SHIFT
 	movs	r0, r0, lsl #SECTION_SHIFT
@@ -251,7 +276,15 @@ __create_page_tables:
 	mov	r3, r7, lsr #SECTION_SHIFT
 	ldr	r7, [r10, #PROCINFO_IO_MMUFLAGS] @ io_mmuflags
 	orr	r3, r7, r3, lsl #SECTION_SHIFT
+#ifdef CONFIG_ARM_LPAE
+	mov	r7, #1 << (54 - 32)		@ XN
+#else
+	orr	r3, r3, #PMD_SECT_XN
+#endif
 1:	str	r3, [r0], #4
+#ifdef CONFIG_ARM_LPAE
+	str	r7, [r0], #4
+#endif
 	add	r3, r3, #1 << SECTION_SHIFT
 	cmp	r0, r6
 	blo	1b
@@ -283,14 +316,17 @@ __create_page_tables:
 	str	r3, [r0]
 #endif
 #endif
+#ifdef CONFIG_ARM_LPAE
+	sub	r4, r4, #0x1000		@ point to the PGD table
+#endif
 	mov	pc, lr
 ENDPROC(__create_page_tables)
 	.ltorg
 	.align
-__enable_mmu_loc:
+__turn_mmu_on_loc:
 	.long	.
-	.long	__enable_mmu
-	.long	__enable_mmu_end
+	.long	__turn_mmu_on
+	.long	__turn_mmu_on_end
 
 #if defined(CONFIG_SMP)
 	__CPUINIT
@@ -374,12 +410,17 @@ __enable_mmu:
 #ifdef CONFIG_CPU_ICACHE_DISABLE
 	bic	r0, r0, #CR_I
 #endif
+#ifdef CONFIG_ARM_LPAE
+	mov	r5, #0
+	mcrr	p15, 0, r4, r5, c2		@ load TTBR0
+#else
 	mov	r5, #(domain_val(DOMAIN_USER, DOMAIN_MANAGER) | \
 		      domain_val(DOMAIN_KERNEL, DOMAIN_MANAGER) | \
 		      domain_val(DOMAIN_TABLE, DOMAIN_MANAGER) | \
 		      domain_val(DOMAIN_IO, DOMAIN_CLIENT))
 	mcr	p15, 0, r5, c3, c0, 0		@ load domain access register
 	mcr	p15, 0, r4, c2, c0, 0		@ load page table pointer
+#endif
 	b	__turn_mmu_on
 ENDPROC(__enable_mmu)
 
@@ -398,15 +439,19 @@ ENDPROC(__enable_mmu)
  * other registers depend on the function called upon completion
  */
 	.align	5
-__turn_mmu_on:
+	.pushsection	.idmap.text, "ax"
+ENTRY(__turn_mmu_on)
 	mov	r0, r0
+	instr_sync
 	mcr	p15, 0, r0, c1, c0, 0		@ write control reg
 	mrc	p15, 0, r3, c0, c0, 0		@ read id reg
+	instr_sync
 	mov	r3, r3
 	mov	r3, r13
 	mov	pc, r3
-__enable_mmu_end:
+__turn_mmu_on_end:
 ENDPROC(__turn_mmu_on)
+	.popsection
 
 
 #ifdef CONFIG_SMP_ON_UP
diff --git a/arch/arm/kernel/hw_breakpoint.c b/arch/arm/kernel/hw_breakpoint.c
index 814a52a9dc39..d6a95ef9131d 100644
--- a/arch/arm/kernel/hw_breakpoint.c
+++ b/arch/arm/kernel/hw_breakpoint.c
@@ -1016,10 +1016,10 @@ static int __init arch_hw_breakpoint_init(void)
 	}
 
 	/* Register debug fault handler. */
-	hook_fault_code(2, hw_breakpoint_pending, SIGTRAP, TRAP_HWBKPT,
-			"watchpoint debug exception");
-	hook_ifault_code(2, hw_breakpoint_pending, SIGTRAP, TRAP_HWBKPT,
-			"breakpoint debug exception");
+	hook_fault_code(FAULT_CODE_DEBUG, hw_breakpoint_pending, SIGTRAP,
+			TRAP_HWBKPT, "watchpoint debug exception");
+	hook_ifault_code(FAULT_CODE_DEBUG, hw_breakpoint_pending, SIGTRAP,
+			TRAP_HWBKPT, "breakpoint debug exception");
 
 	/* Register hotplug notifier. */
 	register_cpu_notifier(&dbg_reset_nb);
diff --git a/arch/arm/kernel/kprobes-test.c b/arch/arm/kernel/kprobes-test.c
index e17cdd6d90d8..1862d8f2fd44 100644
--- a/arch/arm/kernel/kprobes-test.c
+++ b/arch/arm/kernel/kprobes-test.c
@@ -202,6 +202,8 @@
 #include <linux/slab.h>
 #include <linux/kprobes.h>
 
+#include <asm/opcodes.h>
+
 #include "kprobes.h"
 #include "kprobes-test.h"
 
@@ -1050,65 +1052,9 @@ static int test_instance;
 
 static unsigned long test_check_cc(int cc, unsigned long cpsr)
 {
-	unsigned long temp;
-
-	switch (cc) {
-	case 0x0: /* eq */
-		return cpsr & PSR_Z_BIT;
-
-	case 0x1: /* ne */
-		return (~cpsr) & PSR_Z_BIT;
-
-	case 0x2: /* cs */
-		return cpsr & PSR_C_BIT;
-
-	case 0x3: /* cc */
-		return (~cpsr) & PSR_C_BIT;
-
-	case 0x4: /* mi */
-		return cpsr & PSR_N_BIT;
-
-	case 0x5: /* pl */
-		return (~cpsr) & PSR_N_BIT;
-
-	case 0x6: /* vs */
-		return cpsr & PSR_V_BIT;
-
-	case 0x7: /* vc */
-		return (~cpsr) & PSR_V_BIT;
+	int ret = arm_check_condition(cc << 28, cpsr);
 
-	case 0x8: /* hi */
-		cpsr &= ~(cpsr >> 1); /* PSR_C_BIT &= ~PSR_Z_BIT */
-		return cpsr & PSR_C_BIT;
-
-	case 0x9: /* ls */
-		cpsr &= ~(cpsr >> 1); /* PSR_C_BIT &= ~PSR_Z_BIT */
-		return (~cpsr) & PSR_C_BIT;
-
-	case 0xa: /* ge */
-		cpsr ^= (cpsr << 3); /* PSR_N_BIT ^= PSR_V_BIT */
-		return (~cpsr) & PSR_N_BIT;
-
-	case 0xb: /* lt */
-		cpsr ^= (cpsr << 3); /* PSR_N_BIT ^= PSR_V_BIT */
-		return cpsr & PSR_N_BIT;
-
-	case 0xc: /* gt */
-		temp = cpsr ^ (cpsr << 3); /* PSR_N_BIT ^= PSR_V_BIT */
-		temp |= (cpsr << 1);	   /* PSR_N_BIT |= PSR_Z_BIT */
-		return (~temp) & PSR_N_BIT;
-
-	case 0xd: /* le */
-		temp = cpsr ^ (cpsr << 3); /* PSR_N_BIT ^= PSR_V_BIT */
-		temp |= (cpsr << 1);	   /* PSR_N_BIT |= PSR_Z_BIT */
-		return temp & PSR_N_BIT;
-
-	case 0xe: /* al */
-	case 0xf: /* unconditional */
-		return true;
-	}
-	BUG();
-	return false;
+	return (ret != ARM_OPCODE_CONDTEST_FAIL);
 }
 
 static int is_last_scenario;
@@ -1128,7 +1074,9 @@ static unsigned long test_context_cpsr(int scenario)
 
 	if (!test_case_is_thumb) {
 		/* Testing ARM code */
-		probe_should_run = test_check_cc(current_instruction >> 28, cpsr) != 0;
+		int cc = current_instruction >> 28;
+
+		probe_should_run = test_check_cc(cc, cpsr) != 0;
 		if (scenario == 15)
 			is_last_scenario = true;
 
diff --git a/arch/arm/kernel/machine_kexec.c b/arch/arm/kernel/machine_kexec.c
index e59bbd496c39..764bd456d84f 100644
--- a/arch/arm/kernel/machine_kexec.c
+++ b/arch/arm/kernel/machine_kexec.c
@@ -12,12 +12,11 @@
 #include <asm/mmu_context.h>
 #include <asm/cacheflush.h>
 #include <asm/mach-types.h>
+#include <asm/system.h>
 
 extern const unsigned char relocate_new_kernel[];
 extern const unsigned int relocate_new_kernel_size;
 
-extern void setup_mm_for_reboot(char mode);
-
 extern unsigned long kexec_start_address;
 extern unsigned long kexec_indirection_page;
 extern unsigned long kexec_mach_type;
@@ -111,14 +110,6 @@ void machine_kexec(struct kimage *image)
 
 	if (kexec_reinit)
 		kexec_reinit();
-	local_irq_disable();
-	local_fiq_disable();
-	setup_mm_for_reboot(0); /* mode is not used, so just pass 0*/
-	flush_cache_all();
-	outer_flush_all();
-	outer_disable();
-	cpu_proc_fin();
-	outer_inv_all();
-	flush_cache_all();
-	cpu_reset(reboot_code_buffer_phys);
+
+	soft_restart(reboot_code_buffer_phys);
 }
diff --git a/arch/arm/kernel/opcodes.c b/arch/arm/kernel/opcodes.c
new file mode 100644
index 000000000000..f8179c6a817f
--- /dev/null
+++ b/arch/arm/kernel/opcodes.c
@@ -0,0 +1,72 @@
+/*
+ *  linux/arch/arm/kernel/opcodes.c
+ *
+ *  A32 condition code lookup feature moved from nwfpe/fpopcode.c
+ *
+ * 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.
+ */
+
+#include <linux/module.h>
+#include <asm/opcodes.h>
+
+#define ARM_OPCODE_CONDITION_UNCOND 0xf
+
+/*
+ * condition code lookup table
+ * index into the table is test code: EQ, NE, ... LT, GT, AL, NV
+ *
+ * bit position in short is condition code: NZCV
+ */
+static const unsigned short cc_map[16] = {
+	0xF0F0,			/* EQ == Z set            */
+	0x0F0F,			/* NE                     */
+	0xCCCC,			/* CS == C set            */
+	0x3333,			/* CC                     */
+	0xFF00,			/* MI == N set            */
+	0x00FF,			/* PL                     */
+	0xAAAA,			/* VS == V set            */
+	0x5555,			/* VC                     */
+	0x0C0C,			/* HI == C set && Z clear */
+	0xF3F3,			/* LS == C clear || Z set */
+	0xAA55,			/* GE == (N==V)           */
+	0x55AA,			/* LT == (N!=V)           */
+	0x0A05,			/* GT == (!Z && (N==V))   */
+	0xF5FA,			/* LE == (Z || (N!=V))    */
+	0xFFFF,			/* AL always              */
+	0			/* NV                     */
+};
+
+/*
+ * Returns:
+ * ARM_OPCODE_CONDTEST_FAIL   - if condition fails
+ * ARM_OPCODE_CONDTEST_PASS   - if condition passes (including AL)
+ * ARM_OPCODE_CONDTEST_UNCOND - if NV condition, or separate unconditional
+ *                              opcode space from v5 onwards
+ *
+ * Code that tests whether a conditional instruction would pass its condition
+ * check should check that return value == ARM_OPCODE_CONDTEST_PASS.
+ *
+ * Code that tests if a condition means that the instruction would be executed
+ * (regardless of conditional or unconditional) should instead check that the
+ * return value != ARM_OPCODE_CONDTEST_FAIL.
+ */
+asmlinkage unsigned int arm_check_condition(u32 opcode, u32 psr)
+{
+	u32 cc_bits  = opcode >> 28;
+	u32 psr_cond = psr >> 28;
+	unsigned int ret;
+
+	if (cc_bits != ARM_OPCODE_CONDITION_UNCOND) {
+		if ((cc_map[cc_bits] >> (psr_cond)) & 1)
+			ret = ARM_OPCODE_CONDTEST_PASS;
+		else
+			ret = ARM_OPCODE_CONDTEST_FAIL;
+	} else {
+		ret = ARM_OPCODE_CONDTEST_UNCOND;
+	}
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(arm_check_condition);
diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c
index 88b0941ce51e..5bb91bf3d47f 100644
--- a/arch/arm/kernel/perf_event.c
+++ b/arch/arm/kernel/perf_event.c
@@ -59,8 +59,7 @@ armpmu_get_pmu_id(void)
 }
 EXPORT_SYMBOL_GPL(armpmu_get_pmu_id);
 
-int
-armpmu_get_max_events(void)
+int perf_num_counters(void)
 {
 	int max_events = 0;
 
@@ -69,12 +68,6 @@ armpmu_get_max_events(void)
 
 	return max_events;
 }
-EXPORT_SYMBOL_GPL(armpmu_get_max_events);
-
-int perf_num_counters(void)
-{
-	return armpmu_get_max_events();
-}
 EXPORT_SYMBOL_GPL(perf_num_counters);
 
 #define HW_OP_UNSUPPORTED		0xFFFF
@@ -380,6 +373,8 @@ armpmu_release_hardware(struct arm_pmu *armpmu)
 {
 	int i, irq, irqs;
 	struct platform_device *pmu_device = armpmu->plat_device;
+	struct arm_pmu_platdata *plat =
+		dev_get_platdata(&pmu_device->dev);
 
 	irqs = min(pmu_device->num_resources, num_possible_cpus());
 
@@ -387,8 +382,11 @@ armpmu_release_hardware(struct arm_pmu *armpmu)
 		if (!cpumask_test_and_clear_cpu(i, &armpmu->active_irqs))
 			continue;
 		irq = platform_get_irq(pmu_device, i);
-		if (irq >= 0)
+		if (irq >= 0) {
+			if (plat && plat->disable_irq)
+				plat->disable_irq(irq);
 			free_irq(irq, armpmu);
+		}
 	}
 
 	release_pmu(armpmu->type);
@@ -448,7 +446,8 @@ armpmu_reserve_hardware(struct arm_pmu *armpmu)
 				irq);
 			armpmu_release_hardware(armpmu);
 			return err;
-		}
+		} else if (plat && plat->enable_irq)
+			plat->enable_irq(irq);
 
 		cpumask_set_cpu(i, &armpmu->active_irqs);
 	}
diff --git a/arch/arm/kernel/perf_event_v6.c b/arch/arm/kernel/perf_event_v6.c
index e63d8115c01b..533be9930ec2 100644
--- a/arch/arm/kernel/perf_event_v6.c
+++ b/arch/arm/kernel/perf_event_v6.c
@@ -65,13 +65,15 @@ enum armv6_counters {
  * accesses/misses in hardware.
  */
 static const unsigned armv6_perf_map[PERF_COUNT_HW_MAX] = {
-	[PERF_COUNT_HW_CPU_CYCLES]	    = ARMV6_PERFCTR_CPU_CYCLES,
-	[PERF_COUNT_HW_INSTRUCTIONS]	    = ARMV6_PERFCTR_INSTR_EXEC,
-	[PERF_COUNT_HW_CACHE_REFERENCES]    = HW_OP_UNSUPPORTED,
-	[PERF_COUNT_HW_CACHE_MISSES]	    = HW_OP_UNSUPPORTED,
-	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = ARMV6_PERFCTR_BR_EXEC,
-	[PERF_COUNT_HW_BRANCH_MISSES]	    = ARMV6_PERFCTR_BR_MISPREDICT,
-	[PERF_COUNT_HW_BUS_CYCLES]	    = HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_CPU_CYCLES]		= ARMV6_PERFCTR_CPU_CYCLES,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= ARMV6_PERFCTR_INSTR_EXEC,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_CACHE_MISSES]		= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= ARMV6_PERFCTR_BR_EXEC,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= ARMV6_PERFCTR_BR_MISPREDICT,
+	[PERF_COUNT_HW_BUS_CYCLES]		= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND]	= ARMV6_PERFCTR_IBUF_STALL,
+	[PERF_COUNT_HW_STALLED_CYCLES_BACKEND]	= ARMV6_PERFCTR_LSU_FULL_STALL,
 };
 
 static const unsigned armv6_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
@@ -218,13 +220,15 @@ enum armv6mpcore_perf_types {
  * accesses/misses in hardware.
  */
 static const unsigned armv6mpcore_perf_map[PERF_COUNT_HW_MAX] = {
-	[PERF_COUNT_HW_CPU_CYCLES]	    = ARMV6MPCORE_PERFCTR_CPU_CYCLES,
-	[PERF_COUNT_HW_INSTRUCTIONS]	    = ARMV6MPCORE_PERFCTR_INSTR_EXEC,
-	[PERF_COUNT_HW_CACHE_REFERENCES]    = HW_OP_UNSUPPORTED,
-	[PERF_COUNT_HW_CACHE_MISSES]	    = HW_OP_UNSUPPORTED,
-	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = ARMV6MPCORE_PERFCTR_BR_EXEC,
-	[PERF_COUNT_HW_BRANCH_MISSES]	    = ARMV6MPCORE_PERFCTR_BR_MISPREDICT,
-	[PERF_COUNT_HW_BUS_CYCLES]	    = HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_CPU_CYCLES]		= ARMV6MPCORE_PERFCTR_CPU_CYCLES,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= ARMV6MPCORE_PERFCTR_INSTR_EXEC,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_CACHE_MISSES]		= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= ARMV6MPCORE_PERFCTR_BR_EXEC,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= ARMV6MPCORE_PERFCTR_BR_MISPREDICT,
+	[PERF_COUNT_HW_BUS_CYCLES]		= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND]	= ARMV6MPCORE_PERFCTR_IBUF_STALL,
+	[PERF_COUNT_HW_STALLED_CYCLES_BACKEND]	= ARMV6MPCORE_PERFCTR_LSU_FULL_STALL,
 };
 
 static const unsigned armv6mpcore_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
diff --git a/arch/arm/kernel/perf_event_v7.c b/arch/arm/kernel/perf_event_v7.c
index 1ef6d0034b85..460bbbb6b885 100644
--- a/arch/arm/kernel/perf_event_v7.c
+++ b/arch/arm/kernel/perf_event_v7.c
@@ -28,165 +28,87 @@ static struct arm_pmu armv7pmu;
  * they are not available.
  */
 enum armv7_perf_types {
-	ARMV7_PERFCTR_PMNC_SW_INCR		= 0x00,
-	ARMV7_PERFCTR_IFETCH_MISS		= 0x01,
-	ARMV7_PERFCTR_ITLB_MISS			= 0x02,
-	ARMV7_PERFCTR_DCACHE_REFILL		= 0x03,	/* L1 */
-	ARMV7_PERFCTR_DCACHE_ACCESS		= 0x04,	/* L1 */
-	ARMV7_PERFCTR_DTLB_REFILL		= 0x05,
-	ARMV7_PERFCTR_DREAD			= 0x06,
-	ARMV7_PERFCTR_DWRITE			= 0x07,
-	ARMV7_PERFCTR_INSTR_EXECUTED		= 0x08,
-	ARMV7_PERFCTR_EXC_TAKEN			= 0x09,
-	ARMV7_PERFCTR_EXC_EXECUTED		= 0x0A,
-	ARMV7_PERFCTR_CID_WRITE			= 0x0B,
-	/* ARMV7_PERFCTR_PC_WRITE is equivalent to HW_BRANCH_INSTRUCTIONS.
+	ARMV7_PERFCTR_PMNC_SW_INCR			= 0x00,
+	ARMV7_PERFCTR_L1_ICACHE_REFILL			= 0x01,
+	ARMV7_PERFCTR_ITLB_REFILL			= 0x02,
+	ARMV7_PERFCTR_L1_DCACHE_REFILL			= 0x03,
+	ARMV7_PERFCTR_L1_DCACHE_ACCESS			= 0x04,
+	ARMV7_PERFCTR_DTLB_REFILL			= 0x05,
+	ARMV7_PERFCTR_MEM_READ				= 0x06,
+	ARMV7_PERFCTR_MEM_WRITE				= 0x07,
+	ARMV7_PERFCTR_INSTR_EXECUTED			= 0x08,
+	ARMV7_PERFCTR_EXC_TAKEN				= 0x09,
+	ARMV7_PERFCTR_EXC_EXECUTED			= 0x0A,
+	ARMV7_PERFCTR_CID_WRITE				= 0x0B,
+
+	/*
+	 * ARMV7_PERFCTR_PC_WRITE is equivalent to HW_BRANCH_INSTRUCTIONS.
 	 * It counts:
-	 *  - all branch instructions,
+	 *  - all (taken) branch instructions,
 	 *  - instructions that explicitly write the PC,
 	 *  - exception generating instructions.
 	 */
-	ARMV7_PERFCTR_PC_WRITE			= 0x0C,
-	ARMV7_PERFCTR_PC_IMM_BRANCH		= 0x0D,
-	ARMV7_PERFCTR_PC_PROC_RETURN		= 0x0E,
-	ARMV7_PERFCTR_UNALIGNED_ACCESS		= 0x0F,
+	ARMV7_PERFCTR_PC_WRITE				= 0x0C,
+	ARMV7_PERFCTR_PC_IMM_BRANCH			= 0x0D,
+	ARMV7_PERFCTR_PC_PROC_RETURN			= 0x0E,
+	ARMV7_PERFCTR_MEM_UNALIGNED_ACCESS		= 0x0F,
+	ARMV7_PERFCTR_PC_BRANCH_MIS_PRED		= 0x10,
+	ARMV7_PERFCTR_CLOCK_CYCLES			= 0x11,
+	ARMV7_PERFCTR_PC_BRANCH_PRED			= 0x12,
 
 	/* These events are defined by the PMUv2 supplement (ARM DDI 0457A). */
-	ARMV7_PERFCTR_PC_BRANCH_MIS_PRED	= 0x10,
-	ARMV7_PERFCTR_CLOCK_CYCLES		= 0x11,
-	ARMV7_PERFCTR_PC_BRANCH_PRED		= 0x12,
-	ARMV7_PERFCTR_MEM_ACCESS		= 0x13,
-	ARMV7_PERFCTR_L1_ICACHE_ACCESS		= 0x14,
-	ARMV7_PERFCTR_L1_DCACHE_WB		= 0x15,
-	ARMV7_PERFCTR_L2_DCACHE_ACCESS		= 0x16,
-	ARMV7_PERFCTR_L2_DCACHE_REFILL		= 0x17,
-	ARMV7_PERFCTR_L2_DCACHE_WB		= 0x18,
-	ARMV7_PERFCTR_BUS_ACCESS		= 0x19,
-	ARMV7_PERFCTR_MEMORY_ERROR		= 0x1A,
-	ARMV7_PERFCTR_INSTR_SPEC		= 0x1B,
-	ARMV7_PERFCTR_TTBR_WRITE		= 0x1C,
-	ARMV7_PERFCTR_BUS_CYCLES		= 0x1D,
-
-	ARMV7_PERFCTR_CPU_CYCLES		= 0xFF
+	ARMV7_PERFCTR_MEM_ACCESS			= 0x13,
+	ARMV7_PERFCTR_L1_ICACHE_ACCESS			= 0x14,
+	ARMV7_PERFCTR_L1_DCACHE_WB			= 0x15,
+	ARMV7_PERFCTR_L2_CACHE_ACCESS			= 0x16,
+	ARMV7_PERFCTR_L2_CACHE_REFILL			= 0x17,
+	ARMV7_PERFCTR_L2_CACHE_WB			= 0x18,
+	ARMV7_PERFCTR_BUS_ACCESS			= 0x19,
+	ARMV7_PERFCTR_MEM_ERROR				= 0x1A,
+	ARMV7_PERFCTR_INSTR_SPEC			= 0x1B,
+	ARMV7_PERFCTR_TTBR_WRITE			= 0x1C,
+	ARMV7_PERFCTR_BUS_CYCLES			= 0x1D,
+
+	ARMV7_PERFCTR_CPU_CYCLES			= 0xFF
 };
 
 /* ARMv7 Cortex-A8 specific event types */
 enum armv7_a8_perf_types {
-	ARMV7_PERFCTR_WRITE_BUFFER_FULL		= 0x40,
-	ARMV7_PERFCTR_L2_STORE_MERGED		= 0x41,
-	ARMV7_PERFCTR_L2_STORE_BUFF		= 0x42,
-	ARMV7_PERFCTR_L2_ACCESS			= 0x43,
-	ARMV7_PERFCTR_L2_CACH_MISS		= 0x44,
-	ARMV7_PERFCTR_AXI_READ_CYCLES		= 0x45,
-	ARMV7_PERFCTR_AXI_WRITE_CYCLES		= 0x46,
-	ARMV7_PERFCTR_MEMORY_REPLAY		= 0x47,
-	ARMV7_PERFCTR_UNALIGNED_ACCESS_REPLAY	= 0x48,
-	ARMV7_PERFCTR_L1_DATA_MISS		= 0x49,
-	ARMV7_PERFCTR_L1_INST_MISS		= 0x4A,
-	ARMV7_PERFCTR_L1_DATA_COLORING		= 0x4B,
-	ARMV7_PERFCTR_L1_NEON_DATA		= 0x4C,
-	ARMV7_PERFCTR_L1_NEON_CACH_DATA		= 0x4D,
-	ARMV7_PERFCTR_L2_NEON			= 0x4E,
-	ARMV7_PERFCTR_L2_NEON_HIT		= 0x4F,
-	ARMV7_PERFCTR_L1_INST			= 0x50,
-	ARMV7_PERFCTR_PC_RETURN_MIS_PRED	= 0x51,
-	ARMV7_PERFCTR_PC_BRANCH_FAILED		= 0x52,
-	ARMV7_PERFCTR_PC_BRANCH_TAKEN		= 0x53,
-	ARMV7_PERFCTR_PC_BRANCH_EXECUTED	= 0x54,
-	ARMV7_PERFCTR_OP_EXECUTED		= 0x55,
-	ARMV7_PERFCTR_CYCLES_INST_STALL		= 0x56,
-	ARMV7_PERFCTR_CYCLES_INST		= 0x57,
-	ARMV7_PERFCTR_CYCLES_NEON_DATA_STALL	= 0x58,
-	ARMV7_PERFCTR_CYCLES_NEON_INST_STALL	= 0x59,
-	ARMV7_PERFCTR_NEON_CYCLES		= 0x5A,
-
-	ARMV7_PERFCTR_PMU0_EVENTS		= 0x70,
-	ARMV7_PERFCTR_PMU1_EVENTS		= 0x71,
-	ARMV7_PERFCTR_PMU_EVENTS		= 0x72,
+	ARMV7_A8_PERFCTR_L2_CACHE_ACCESS		= 0x43,
+	ARMV7_A8_PERFCTR_L2_CACHE_REFILL		= 0x44,
+	ARMV7_A8_PERFCTR_L1_ICACHE_ACCESS		= 0x50,
+	ARMV7_A8_PERFCTR_STALL_ISIDE			= 0x56,
 };
 
 /* ARMv7 Cortex-A9 specific event types */
 enum armv7_a9_perf_types {
-	ARMV7_PERFCTR_JAVA_HW_BYTECODE_EXEC	= 0x40,
-	ARMV7_PERFCTR_JAVA_SW_BYTECODE_EXEC	= 0x41,
-	ARMV7_PERFCTR_JAZELLE_BRANCH_EXEC	= 0x42,
-
-	ARMV7_PERFCTR_COHERENT_LINE_MISS	= 0x50,
-	ARMV7_PERFCTR_COHERENT_LINE_HIT		= 0x51,
-
-	ARMV7_PERFCTR_ICACHE_DEP_STALL_CYCLES	= 0x60,
-	ARMV7_PERFCTR_DCACHE_DEP_STALL_CYCLES	= 0x61,
-	ARMV7_PERFCTR_TLB_MISS_DEP_STALL_CYCLES	= 0x62,
-	ARMV7_PERFCTR_STREX_EXECUTED_PASSED	= 0x63,
-	ARMV7_PERFCTR_STREX_EXECUTED_FAILED	= 0x64,
-	ARMV7_PERFCTR_DATA_EVICTION		= 0x65,
-	ARMV7_PERFCTR_ISSUE_STAGE_NO_INST	= 0x66,
-	ARMV7_PERFCTR_ISSUE_STAGE_EMPTY		= 0x67,
-	ARMV7_PERFCTR_INST_OUT_OF_RENAME_STAGE	= 0x68,
-
-	ARMV7_PERFCTR_PREDICTABLE_FUNCT_RETURNS	= 0x6E,
-
-	ARMV7_PERFCTR_MAIN_UNIT_EXECUTED_INST	= 0x70,
-	ARMV7_PERFCTR_SECOND_UNIT_EXECUTED_INST	= 0x71,
-	ARMV7_PERFCTR_LD_ST_UNIT_EXECUTED_INST	= 0x72,
-	ARMV7_PERFCTR_FP_EXECUTED_INST		= 0x73,
-	ARMV7_PERFCTR_NEON_EXECUTED_INST	= 0x74,
-
-	ARMV7_PERFCTR_PLD_FULL_DEP_STALL_CYCLES	= 0x80,
-	ARMV7_PERFCTR_DATA_WR_DEP_STALL_CYCLES	= 0x81,
-	ARMV7_PERFCTR_ITLB_MISS_DEP_STALL_CYCLES	= 0x82,
-	ARMV7_PERFCTR_DTLB_MISS_DEP_STALL_CYCLES	= 0x83,
-	ARMV7_PERFCTR_MICRO_ITLB_MISS_DEP_STALL_CYCLES	= 0x84,
-	ARMV7_PERFCTR_MICRO_DTLB_MISS_DEP_STALL_CYCLES	= 0x85,
-	ARMV7_PERFCTR_DMB_DEP_STALL_CYCLES	= 0x86,
-
-	ARMV7_PERFCTR_INTGR_CLK_ENABLED_CYCLES	= 0x8A,
-	ARMV7_PERFCTR_DATA_ENGINE_CLK_EN_CYCLES	= 0x8B,
-
-	ARMV7_PERFCTR_ISB_INST			= 0x90,
-	ARMV7_PERFCTR_DSB_INST			= 0x91,
-	ARMV7_PERFCTR_DMB_INST			= 0x92,
-	ARMV7_PERFCTR_EXT_INTERRUPTS		= 0x93,
-
-	ARMV7_PERFCTR_PLE_CACHE_LINE_RQST_COMPLETED	= 0xA0,
-	ARMV7_PERFCTR_PLE_CACHE_LINE_RQST_SKIPPED	= 0xA1,
-	ARMV7_PERFCTR_PLE_FIFO_FLUSH		= 0xA2,
-	ARMV7_PERFCTR_PLE_RQST_COMPLETED	= 0xA3,
-	ARMV7_PERFCTR_PLE_FIFO_OVERFLOW		= 0xA4,
-	ARMV7_PERFCTR_PLE_RQST_PROG		= 0xA5
+	ARMV7_A9_PERFCTR_INSTR_CORE_RENAME		= 0x68,
+	ARMV7_A9_PERFCTR_STALL_ICACHE			= 0x60,
+	ARMV7_A9_PERFCTR_STALL_DISPATCH			= 0x66,
 };
 
 /* ARMv7 Cortex-A5 specific event types */
 enum armv7_a5_perf_types {
-	ARMV7_PERFCTR_IRQ_TAKEN			= 0x86,
-	ARMV7_PERFCTR_FIQ_TAKEN			= 0x87,
-
-	ARMV7_PERFCTR_EXT_MEM_RQST		= 0xc0,
-	ARMV7_PERFCTR_NC_EXT_MEM_RQST		= 0xc1,
-	ARMV7_PERFCTR_PREFETCH_LINEFILL		= 0xc2,
-	ARMV7_PERFCTR_PREFETCH_LINEFILL_DROP	= 0xc3,
-	ARMV7_PERFCTR_ENTER_READ_ALLOC		= 0xc4,
-	ARMV7_PERFCTR_READ_ALLOC		= 0xc5,
-
-	ARMV7_PERFCTR_STALL_SB_FULL		= 0xc9,
+	ARMV7_A5_PERFCTR_PREFETCH_LINEFILL		= 0xc2,
+	ARMV7_A5_PERFCTR_PREFETCH_LINEFILL_DROP		= 0xc3,
 };
 
 /* ARMv7 Cortex-A15 specific event types */
 enum armv7_a15_perf_types {
-	ARMV7_PERFCTR_L1_DCACHE_READ_ACCESS	= 0x40,
-	ARMV7_PERFCTR_L1_DCACHE_WRITE_ACCESS	= 0x41,
-	ARMV7_PERFCTR_L1_DCACHE_READ_REFILL	= 0x42,
-	ARMV7_PERFCTR_L1_DCACHE_WRITE_REFILL	= 0x43,
+	ARMV7_A15_PERFCTR_L1_DCACHE_ACCESS_READ		= 0x40,
+	ARMV7_A15_PERFCTR_L1_DCACHE_ACCESS_WRITE	= 0x41,
+	ARMV7_A15_PERFCTR_L1_DCACHE_REFILL_READ		= 0x42,
+	ARMV7_A15_PERFCTR_L1_DCACHE_REFILL_WRITE	= 0x43,
 
-	ARMV7_PERFCTR_L1_DTLB_READ_REFILL	= 0x4C,
-	ARMV7_PERFCTR_L1_DTLB_WRITE_REFILL	= 0x4D,
+	ARMV7_A15_PERFCTR_DTLB_REFILL_L1_READ		= 0x4C,
+	ARMV7_A15_PERFCTR_DTLB_REFILL_L1_WRITE		= 0x4D,
 
-	ARMV7_PERFCTR_L2_DCACHE_READ_ACCESS	= 0x50,
-	ARMV7_PERFCTR_L2_DCACHE_WRITE_ACCESS	= 0x51,
-	ARMV7_PERFCTR_L2_DCACHE_READ_REFILL	= 0x52,
-	ARMV7_PERFCTR_L2_DCACHE_WRITE_REFILL	= 0x53,
+	ARMV7_A15_PERFCTR_L2_CACHE_ACCESS_READ		= 0x50,
+	ARMV7_A15_PERFCTR_L2_CACHE_ACCESS_WRITE		= 0x51,
+	ARMV7_A15_PERFCTR_L2_CACHE_REFILL_READ		= 0x52,
+	ARMV7_A15_PERFCTR_L2_CACHE_REFILL_WRITE		= 0x53,
 
-	ARMV7_PERFCTR_SPEC_PC_WRITE		= 0x76,
+	ARMV7_A15_PERFCTR_PC_WRITE_SPEC			= 0x76,
 };
 
 /*
@@ -197,13 +119,15 @@ enum armv7_a15_perf_types {
  * accesses/misses in hardware.
  */
 static const unsigned armv7_a8_perf_map[PERF_COUNT_HW_MAX] = {
-	[PERF_COUNT_HW_CPU_CYCLES]	    = ARMV7_PERFCTR_CPU_CYCLES,
-	[PERF_COUNT_HW_INSTRUCTIONS]	    = ARMV7_PERFCTR_INSTR_EXECUTED,
-	[PERF_COUNT_HW_CACHE_REFERENCES]    = HW_OP_UNSUPPORTED,
-	[PERF_COUNT_HW_CACHE_MISSES]	    = HW_OP_UNSUPPORTED,
-	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = ARMV7_PERFCTR_PC_WRITE,
-	[PERF_COUNT_HW_BRANCH_MISSES]	    = ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
-	[PERF_COUNT_HW_BUS_CYCLES]	    = ARMV7_PERFCTR_CLOCK_CYCLES,
+	[PERF_COUNT_HW_CPU_CYCLES]		= ARMV7_PERFCTR_CPU_CYCLES,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= ARMV7_PERFCTR_INSTR_EXECUTED,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= ARMV7_PERFCTR_L1_DCACHE_ACCESS,
+	[PERF_COUNT_HW_CACHE_MISSES]		= ARMV7_PERFCTR_L1_DCACHE_REFILL,
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= ARMV7_PERFCTR_PC_WRITE,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
+	[PERF_COUNT_HW_BUS_CYCLES]		= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND]	= ARMV7_A8_PERFCTR_STALL_ISIDE,
+	[PERF_COUNT_HW_STALLED_CYCLES_BACKEND]	= HW_OP_UNSUPPORTED,
 };
 
 static const unsigned armv7_a8_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
@@ -217,12 +141,12 @@ static const unsigned armv7_a8_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 		 * combined.
 		 */
 		[C(OP_READ)] = {
-			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_DCACHE_ACCESS,
-			[C(RESULT_MISS)]	= ARMV7_PERFCTR_DCACHE_REFILL,
+			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_L1_DCACHE_ACCESS,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_L1_DCACHE_REFILL,
 		},
 		[C(OP_WRITE)] = {
-			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_DCACHE_ACCESS,
-			[C(RESULT_MISS)]	= ARMV7_PERFCTR_DCACHE_REFILL,
+			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_L1_DCACHE_ACCESS,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_L1_DCACHE_REFILL,
 		},
 		[C(OP_PREFETCH)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
@@ -231,12 +155,12 @@ static const unsigned armv7_a8_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 	},
 	[C(L1I)] = {
 		[C(OP_READ)] = {
-			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_L1_INST,
-			[C(RESULT_MISS)]	= ARMV7_PERFCTR_L1_INST_MISS,
+			[C(RESULT_ACCESS)]	= ARMV7_A8_PERFCTR_L1_ICACHE_ACCESS,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_L1_ICACHE_REFILL,
 		},
 		[C(OP_WRITE)] = {
-			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_L1_INST,
-			[C(RESULT_MISS)]	= ARMV7_PERFCTR_L1_INST_MISS,
+			[C(RESULT_ACCESS)]	= ARMV7_A8_PERFCTR_L1_ICACHE_ACCESS,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_L1_ICACHE_REFILL,
 		},
 		[C(OP_PREFETCH)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
@@ -245,12 +169,12 @@ static const unsigned armv7_a8_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 	},
 	[C(LL)] = {
 		[C(OP_READ)] = {
-			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_L2_ACCESS,
-			[C(RESULT_MISS)]	= ARMV7_PERFCTR_L2_CACH_MISS,
+			[C(RESULT_ACCESS)]	= ARMV7_A8_PERFCTR_L2_CACHE_ACCESS,
+			[C(RESULT_MISS)]	= ARMV7_A8_PERFCTR_L2_CACHE_REFILL,
 		},
 		[C(OP_WRITE)] = {
-			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_L2_ACCESS,
-			[C(RESULT_MISS)]	= ARMV7_PERFCTR_L2_CACH_MISS,
+			[C(RESULT_ACCESS)]	= ARMV7_A8_PERFCTR_L2_CACHE_ACCESS,
+			[C(RESULT_MISS)]	= ARMV7_A8_PERFCTR_L2_CACHE_REFILL,
 		},
 		[C(OP_PREFETCH)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
@@ -274,11 +198,11 @@ static const unsigned armv7_a8_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 	[C(ITLB)] = {
 		[C(OP_READ)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
-			[C(RESULT_MISS)]	= ARMV7_PERFCTR_ITLB_MISS,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_ITLB_REFILL,
 		},
 		[C(OP_WRITE)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
-			[C(RESULT_MISS)]	= ARMV7_PERFCTR_ITLB_MISS,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_ITLB_REFILL,
 		},
 		[C(OP_PREFETCH)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
@@ -287,14 +211,12 @@ static const unsigned armv7_a8_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 	},
 	[C(BPU)] = {
 		[C(OP_READ)] = {
-			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_PC_WRITE,
-			[C(RESULT_MISS)]
-					= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
+			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_PC_BRANCH_PRED,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
 		},
 		[C(OP_WRITE)] = {
-			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_PC_WRITE,
-			[C(RESULT_MISS)]
-					= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
+			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_PC_BRANCH_PRED,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
 		},
 		[C(OP_PREFETCH)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
@@ -321,14 +243,15 @@ static const unsigned armv7_a8_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
  * Cortex-A9 HW events mapping
  */
 static const unsigned armv7_a9_perf_map[PERF_COUNT_HW_MAX] = {
-	[PERF_COUNT_HW_CPU_CYCLES]	    = ARMV7_PERFCTR_CPU_CYCLES,
-	[PERF_COUNT_HW_INSTRUCTIONS]	    =
-					ARMV7_PERFCTR_INST_OUT_OF_RENAME_STAGE,
-	[PERF_COUNT_HW_CACHE_REFERENCES]    = ARMV7_PERFCTR_DCACHE_ACCESS,
-	[PERF_COUNT_HW_CACHE_MISSES]	    = ARMV7_PERFCTR_DCACHE_REFILL,
-	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = ARMV7_PERFCTR_PC_WRITE,
-	[PERF_COUNT_HW_BRANCH_MISSES]	    = ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
-	[PERF_COUNT_HW_BUS_CYCLES]	    = ARMV7_PERFCTR_CLOCK_CYCLES,
+	[PERF_COUNT_HW_CPU_CYCLES]		= ARMV7_PERFCTR_CPU_CYCLES,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= ARMV7_A9_PERFCTR_INSTR_CORE_RENAME,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= ARMV7_PERFCTR_L1_DCACHE_ACCESS,
+	[PERF_COUNT_HW_CACHE_MISSES]		= ARMV7_PERFCTR_L1_DCACHE_REFILL,
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= ARMV7_PERFCTR_PC_WRITE,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
+	[PERF_COUNT_HW_BUS_CYCLES]		= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND]	= ARMV7_A9_PERFCTR_STALL_ICACHE,
+	[PERF_COUNT_HW_STALLED_CYCLES_BACKEND]	= ARMV7_A9_PERFCTR_STALL_DISPATCH,
 };
 
 static const unsigned armv7_a9_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
@@ -342,12 +265,12 @@ static const unsigned armv7_a9_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 		 * combined.
 		 */
 		[C(OP_READ)] = {
-			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_DCACHE_ACCESS,
-			[C(RESULT_MISS)]	= ARMV7_PERFCTR_DCACHE_REFILL,
+			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_L1_DCACHE_ACCESS,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_L1_DCACHE_REFILL,
 		},
 		[C(OP_WRITE)] = {
-			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_DCACHE_ACCESS,
-			[C(RESULT_MISS)]	= ARMV7_PERFCTR_DCACHE_REFILL,
+			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_L1_DCACHE_ACCESS,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_L1_DCACHE_REFILL,
 		},
 		[C(OP_PREFETCH)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
@@ -357,11 +280,11 @@ static const unsigned armv7_a9_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 	[C(L1I)] = {
 		[C(OP_READ)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
-			[C(RESULT_MISS)]	= ARMV7_PERFCTR_IFETCH_MISS,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_L1_ICACHE_REFILL,
 		},
 		[C(OP_WRITE)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
-			[C(RESULT_MISS)]	= ARMV7_PERFCTR_IFETCH_MISS,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_L1_ICACHE_REFILL,
 		},
 		[C(OP_PREFETCH)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
@@ -399,11 +322,11 @@ static const unsigned armv7_a9_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 	[C(ITLB)] = {
 		[C(OP_READ)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
-			[C(RESULT_MISS)]	= ARMV7_PERFCTR_ITLB_MISS,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_ITLB_REFILL,
 		},
 		[C(OP_WRITE)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
-			[C(RESULT_MISS)]	= ARMV7_PERFCTR_ITLB_MISS,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_ITLB_REFILL,
 		},
 		[C(OP_PREFETCH)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
@@ -412,14 +335,12 @@ static const unsigned armv7_a9_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 	},
 	[C(BPU)] = {
 		[C(OP_READ)] = {
-			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_PC_WRITE,
-			[C(RESULT_MISS)]
-					= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
+			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_PC_BRANCH_PRED,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
 		},
 		[C(OP_WRITE)] = {
-			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_PC_WRITE,
-			[C(RESULT_MISS)]
-					= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
+			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_PC_BRANCH_PRED,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
 		},
 		[C(OP_PREFETCH)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
@@ -446,13 +367,15 @@ static const unsigned armv7_a9_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
  * Cortex-A5 HW events mapping
  */
 static const unsigned armv7_a5_perf_map[PERF_COUNT_HW_MAX] = {
-	[PERF_COUNT_HW_CPU_CYCLES]	    = ARMV7_PERFCTR_CPU_CYCLES,
-	[PERF_COUNT_HW_INSTRUCTIONS]	    = ARMV7_PERFCTR_INSTR_EXECUTED,
-	[PERF_COUNT_HW_CACHE_REFERENCES]    = HW_OP_UNSUPPORTED,
-	[PERF_COUNT_HW_CACHE_MISSES]	    = HW_OP_UNSUPPORTED,
-	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = ARMV7_PERFCTR_PC_WRITE,
-	[PERF_COUNT_HW_BRANCH_MISSES]	    = ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
-	[PERF_COUNT_HW_BUS_CYCLES]	    = HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_CPU_CYCLES]		= ARMV7_PERFCTR_CPU_CYCLES,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= ARMV7_PERFCTR_INSTR_EXECUTED,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= ARMV7_PERFCTR_L1_DCACHE_ACCESS,
+	[PERF_COUNT_HW_CACHE_MISSES]		= ARMV7_PERFCTR_L1_DCACHE_REFILL,
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= ARMV7_PERFCTR_PC_WRITE,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
+	[PERF_COUNT_HW_BUS_CYCLES]		= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND]	= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_STALLED_CYCLES_BACKEND]	= HW_OP_UNSUPPORTED,
 };
 
 static const unsigned armv7_a5_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
@@ -460,42 +383,34 @@ static const unsigned armv7_a5_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 					[PERF_COUNT_HW_CACHE_RESULT_MAX] = {
 	[C(L1D)] = {
 		[C(OP_READ)] = {
-			[C(RESULT_ACCESS)]
-					= ARMV7_PERFCTR_DCACHE_ACCESS,
-			[C(RESULT_MISS)]
-					= ARMV7_PERFCTR_DCACHE_REFILL,
+			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_L1_DCACHE_ACCESS,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_L1_DCACHE_REFILL,
 		},
 		[C(OP_WRITE)] = {
-			[C(RESULT_ACCESS)]
-					= ARMV7_PERFCTR_DCACHE_ACCESS,
-			[C(RESULT_MISS)]
-					= ARMV7_PERFCTR_DCACHE_REFILL,
+			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_L1_DCACHE_ACCESS,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_L1_DCACHE_REFILL,
 		},
 		[C(OP_PREFETCH)] = {
-			[C(RESULT_ACCESS)]
-					= ARMV7_PERFCTR_PREFETCH_LINEFILL,
-			[C(RESULT_MISS)]
-					= ARMV7_PERFCTR_PREFETCH_LINEFILL_DROP,
+			[C(RESULT_ACCESS)]	= ARMV7_A5_PERFCTR_PREFETCH_LINEFILL,
+			[C(RESULT_MISS)]	= ARMV7_A5_PERFCTR_PREFETCH_LINEFILL_DROP,
 		},
 	},
 	[C(L1I)] = {
 		[C(OP_READ)] = {
 			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_L1_ICACHE_ACCESS,
-			[C(RESULT_MISS)]	= ARMV7_PERFCTR_IFETCH_MISS,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_L1_ICACHE_REFILL,
 		},
 		[C(OP_WRITE)] = {
 			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_L1_ICACHE_ACCESS,
-			[C(RESULT_MISS)]	= ARMV7_PERFCTR_IFETCH_MISS,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_L1_ICACHE_REFILL,
 		},
 		/*
 		 * The prefetch counters don't differentiate between the I
 		 * side and the D side.
 		 */
 		[C(OP_PREFETCH)] = {
-			[C(RESULT_ACCESS)]
-					= ARMV7_PERFCTR_PREFETCH_LINEFILL,
-			[C(RESULT_MISS)]
-					= ARMV7_PERFCTR_PREFETCH_LINEFILL_DROP,
+			[C(RESULT_ACCESS)]	= ARMV7_A5_PERFCTR_PREFETCH_LINEFILL,
+			[C(RESULT_MISS)]	= ARMV7_A5_PERFCTR_PREFETCH_LINEFILL_DROP,
 		},
 	},
 	[C(LL)] = {
@@ -529,11 +444,11 @@ static const unsigned armv7_a5_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 	[C(ITLB)] = {
 		[C(OP_READ)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
-			[C(RESULT_MISS)]	= ARMV7_PERFCTR_ITLB_MISS,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_ITLB_REFILL,
 		},
 		[C(OP_WRITE)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
-			[C(RESULT_MISS)]	= ARMV7_PERFCTR_ITLB_MISS,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_ITLB_REFILL,
 		},
 		[C(OP_PREFETCH)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
@@ -543,13 +458,11 @@ static const unsigned armv7_a5_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 	[C(BPU)] = {
 		[C(OP_READ)] = {
 			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_PC_BRANCH_PRED,
-			[C(RESULT_MISS)]
-					= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
 		},
 		[C(OP_WRITE)] = {
 			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_PC_BRANCH_PRED,
-			[C(RESULT_MISS)]
-					= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
 		},
 		[C(OP_PREFETCH)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
@@ -562,13 +475,15 @@ static const unsigned armv7_a5_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
  * Cortex-A15 HW events mapping
  */
 static const unsigned armv7_a15_perf_map[PERF_COUNT_HW_MAX] = {
-	[PERF_COUNT_HW_CPU_CYCLES]	    = ARMV7_PERFCTR_CPU_CYCLES,
-	[PERF_COUNT_HW_INSTRUCTIONS]	    = ARMV7_PERFCTR_INSTR_EXECUTED,
-	[PERF_COUNT_HW_CACHE_REFERENCES]    = HW_OP_UNSUPPORTED,
-	[PERF_COUNT_HW_CACHE_MISSES]	    = HW_OP_UNSUPPORTED,
-	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = ARMV7_PERFCTR_SPEC_PC_WRITE,
-	[PERF_COUNT_HW_BRANCH_MISSES]	    = ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
-	[PERF_COUNT_HW_BUS_CYCLES]	    = ARMV7_PERFCTR_BUS_CYCLES,
+	[PERF_COUNT_HW_CPU_CYCLES]		= ARMV7_PERFCTR_CPU_CYCLES,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= ARMV7_PERFCTR_INSTR_EXECUTED,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= ARMV7_PERFCTR_L1_DCACHE_ACCESS,
+	[PERF_COUNT_HW_CACHE_MISSES]		= ARMV7_PERFCTR_L1_DCACHE_REFILL,
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= ARMV7_A15_PERFCTR_PC_WRITE_SPEC,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
+	[PERF_COUNT_HW_BUS_CYCLES]		= ARMV7_PERFCTR_BUS_CYCLES,
+	[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND]	= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_STALLED_CYCLES_BACKEND]	= HW_OP_UNSUPPORTED,
 };
 
 static const unsigned armv7_a15_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
@@ -576,16 +491,12 @@ static const unsigned armv7_a15_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 					[PERF_COUNT_HW_CACHE_RESULT_MAX] = {
 	[C(L1D)] = {
 		[C(OP_READ)] = {
-			[C(RESULT_ACCESS)]
-					= ARMV7_PERFCTR_L1_DCACHE_READ_ACCESS,
-			[C(RESULT_MISS)]
-					= ARMV7_PERFCTR_L1_DCACHE_READ_REFILL,
+			[C(RESULT_ACCESS)]	= ARMV7_A15_PERFCTR_L1_DCACHE_ACCESS_READ,
+			[C(RESULT_MISS)]	= ARMV7_A15_PERFCTR_L1_DCACHE_REFILL_READ,
 		},
 		[C(OP_WRITE)] = {
-			[C(RESULT_ACCESS)]
-					= ARMV7_PERFCTR_L1_DCACHE_WRITE_ACCESS,
-			[C(RESULT_MISS)]
-					= ARMV7_PERFCTR_L1_DCACHE_WRITE_REFILL,
+			[C(RESULT_ACCESS)]	= ARMV7_A15_PERFCTR_L1_DCACHE_ACCESS_WRITE,
+			[C(RESULT_MISS)]	= ARMV7_A15_PERFCTR_L1_DCACHE_REFILL_WRITE,
 		},
 		[C(OP_PREFETCH)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
@@ -601,11 +512,11 @@ static const unsigned armv7_a15_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 		 */
 		[C(OP_READ)] = {
 			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_L1_ICACHE_ACCESS,
-			[C(RESULT_MISS)]	= ARMV7_PERFCTR_IFETCH_MISS,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_L1_ICACHE_REFILL,
 		},
 		[C(OP_WRITE)] = {
 			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_L1_ICACHE_ACCESS,
-			[C(RESULT_MISS)]	= ARMV7_PERFCTR_IFETCH_MISS,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_L1_ICACHE_REFILL,
 		},
 		[C(OP_PREFETCH)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
@@ -614,16 +525,12 @@ static const unsigned armv7_a15_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 	},
 	[C(LL)] = {
 		[C(OP_READ)] = {
-			[C(RESULT_ACCESS)]
-					= ARMV7_PERFCTR_L2_DCACHE_READ_ACCESS,
-			[C(RESULT_MISS)]
-					= ARMV7_PERFCTR_L2_DCACHE_READ_REFILL,
+			[C(RESULT_ACCESS)]	= ARMV7_A15_PERFCTR_L2_CACHE_ACCESS_READ,
+			[C(RESULT_MISS)]	= ARMV7_A15_PERFCTR_L2_CACHE_REFILL_READ,
 		},
 		[C(OP_WRITE)] = {
-			[C(RESULT_ACCESS)]
-					= ARMV7_PERFCTR_L2_DCACHE_WRITE_ACCESS,
-			[C(RESULT_MISS)]
-					= ARMV7_PERFCTR_L2_DCACHE_WRITE_REFILL,
+			[C(RESULT_ACCESS)]	= ARMV7_A15_PERFCTR_L2_CACHE_ACCESS_WRITE,
+			[C(RESULT_MISS)]	= ARMV7_A15_PERFCTR_L2_CACHE_REFILL_WRITE,
 		},
 		[C(OP_PREFETCH)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
@@ -633,13 +540,11 @@ static const unsigned armv7_a15_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 	[C(DTLB)] = {
 		[C(OP_READ)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
-			[C(RESULT_MISS)]
-					= ARMV7_PERFCTR_L1_DTLB_READ_REFILL,
+			[C(RESULT_MISS)]	= ARMV7_A15_PERFCTR_DTLB_REFILL_L1_READ,
 		},
 		[C(OP_WRITE)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
-			[C(RESULT_MISS)]
-					= ARMV7_PERFCTR_L1_DTLB_WRITE_REFILL,
+			[C(RESULT_MISS)]	= ARMV7_A15_PERFCTR_DTLB_REFILL_L1_WRITE,
 		},
 		[C(OP_PREFETCH)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
@@ -649,11 +554,11 @@ static const unsigned armv7_a15_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 	[C(ITLB)] = {
 		[C(OP_READ)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
-			[C(RESULT_MISS)]	= ARMV7_PERFCTR_ITLB_MISS,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_ITLB_REFILL,
 		},
 		[C(OP_WRITE)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
-			[C(RESULT_MISS)]	= ARMV7_PERFCTR_ITLB_MISS,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_ITLB_REFILL,
 		},
 		[C(OP_PREFETCH)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
@@ -663,13 +568,11 @@ static const unsigned armv7_a15_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 	[C(BPU)] = {
 		[C(OP_READ)] = {
 			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_PC_BRANCH_PRED,
-			[C(RESULT_MISS)]
-					= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
 		},
 		[C(OP_WRITE)] = {
 			[C(RESULT_ACCESS)]	= ARMV7_PERFCTR_PC_BRANCH_PRED,
-			[C(RESULT_MISS)]
-					= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
+			[C(RESULT_MISS)]	= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
 		},
 		[C(OP_PREFETCH)] = {
 			[C(RESULT_ACCESS)]	= CACHE_OP_UNSUPPORTED,
diff --git a/arch/arm/kernel/perf_event_xscale.c b/arch/arm/kernel/perf_event_xscale.c
index e0cca10a8411..3b99d8269829 100644
--- a/arch/arm/kernel/perf_event_xscale.c
+++ b/arch/arm/kernel/perf_event_xscale.c
@@ -48,13 +48,15 @@ enum xscale_counters {
 };
 
 static const unsigned xscale_perf_map[PERF_COUNT_HW_MAX] = {
-	[PERF_COUNT_HW_CPU_CYCLES]	    = XSCALE_PERFCTR_CCNT,
-	[PERF_COUNT_HW_INSTRUCTIONS]	    = XSCALE_PERFCTR_INSTRUCTION,
-	[PERF_COUNT_HW_CACHE_REFERENCES]    = HW_OP_UNSUPPORTED,
-	[PERF_COUNT_HW_CACHE_MISSES]	    = HW_OP_UNSUPPORTED,
-	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = XSCALE_PERFCTR_BRANCH,
-	[PERF_COUNT_HW_BRANCH_MISSES]	    = XSCALE_PERFCTR_BRANCH_MISS,
-	[PERF_COUNT_HW_BUS_CYCLES]	    = HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_CPU_CYCLES]		= XSCALE_PERFCTR_CCNT,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= XSCALE_PERFCTR_INSTRUCTION,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_CACHE_MISSES]		= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= XSCALE_PERFCTR_BRANCH,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= XSCALE_PERFCTR_BRANCH_MISS,
+	[PERF_COUNT_HW_BUS_CYCLES]		= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND]	= XSCALE_PERFCTR_ICACHE_NO_DELIVER,
+	[PERF_COUNT_HW_STALLED_CYCLES_BACKEND]	= HW_OP_UNSUPPORTED,
 };
 
 static const unsigned xscale_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c
index 3d0c6fb74ae4..b29776aa6586 100644
--- a/arch/arm/kernel/process.c
+++ b/arch/arm/kernel/process.c
@@ -57,7 +57,7 @@ static const char *isa_modes[] = {
   "ARM" , "Thumb" , "Jazelle", "ThumbEE"
 };
 
-extern void setup_mm_for_reboot(char mode);
+extern void setup_mm_for_reboot(void);
 
 static volatile int hlt_counter;
 
@@ -92,18 +92,24 @@ static int __init hlt_setup(char *__unused)
 __setup("nohlt", nohlt_setup);
 __setup("hlt", hlt_setup);
 
-void arm_machine_restart(char mode, const char *cmd)
+extern void call_with_stack(void (*fn)(void *), void *arg, void *sp);
+typedef void (*phys_reset_t)(unsigned long);
+
+/*
+ * A temporary stack to use for CPU reset. This is static so that we
+ * don't clobber it with the identity mapping. When running with this
+ * stack, any references to the current task *will not work* so you
+ * should really do as little as possible before jumping to your reset
+ * code.
+ */
+static u64 soft_restart_stack[16];
+
+static void __soft_restart(void *addr)
 {
-	/* Disable interrupts first */
-	local_irq_disable();
-	local_fiq_disable();
+	phys_reset_t phys_reset;
 
-	/*
-	 * Tell the mm system that we are going to reboot -
-	 * we may need it to insert some 1:1 mappings so that
-	 * soft boot works.
-	 */
-	setup_mm_for_reboot(mode);
+	/* Take out a flat memory mapping. */
+	setup_mm_for_reboot();
 
 	/* Clean and invalidate caches */
 	flush_cache_all();
@@ -114,18 +120,35 @@ void arm_machine_restart(char mode, const char *cmd)
 	/* Push out any further dirty data, and ensure cache is empty */
 	flush_cache_all();
 
-	/*
-	 * Now call the architecture specific reboot code.
-	 */
-	arch_reset(mode, cmd);
+	/* Switch to the identity mapping. */
+	phys_reset = (phys_reset_t)(unsigned long)virt_to_phys(cpu_reset);
+	phys_reset((unsigned long)addr);
 
-	/*
-	 * Whoops - the architecture was unable to reboot.
-	 * Tell the user!
-	 */
-	mdelay(1000);
-	printk("Reboot failed -- System halted\n");
-	while (1);
+	/* Should never get here. */
+	BUG();
+}
+
+void soft_restart(unsigned long addr)
+{
+	u64 *stack = soft_restart_stack + ARRAY_SIZE(soft_restart_stack);
+
+	/* Disable interrupts first */
+	local_irq_disable();
+	local_fiq_disable();
+
+	/* Disable the L2 if we're the last man standing. */
+	if (num_online_cpus() == 1)
+		outer_disable();
+
+	/* Change to the new stack and continue with the reset. */
+	call_with_stack(__soft_restart, (void *)addr, (void *)stack);
+
+	/* Should never get here. */
+	BUG();
+}
+
+static void null_restart(char mode, const char *cmd)
+{
 }
 
 /*
@@ -134,7 +157,7 @@ void arm_machine_restart(char mode, const char *cmd)
 void (*pm_power_off)(void);
 EXPORT_SYMBOL(pm_power_off);
 
-void (*arm_pm_restart)(char str, const char *cmd) = arm_machine_restart;
+void (*arm_pm_restart)(char str, const char *cmd) = null_restart;
 EXPORT_SYMBOL_GPL(arm_pm_restart);
 
 static void do_nothing(void *unused)
@@ -253,7 +276,15 @@ void machine_power_off(void)
 void machine_restart(char *cmd)
 {
 	machine_shutdown();
+
 	arm_pm_restart(reboot_mode, cmd);
+
+	/* Give a grace period for failure to restart of 1s */
+	mdelay(1000);
+
+	/* Whoops - the platform was unable to reboot. Tell the user! */
+	printk("Reboot failed -- System halted\n");
+	while (1);
 }
 
 void __show_regs(struct pt_regs *regs)
diff --git a/arch/arm/kernel/sched_clock.c b/arch/arm/kernel/sched_clock.c
index 9a46370fe9da..5416c7c12528 100644
--- a/arch/arm/kernel/sched_clock.c
+++ b/arch/arm/kernel/sched_clock.c
@@ -14,61 +14,153 @@
 
 #include <asm/sched_clock.h>
 
+struct clock_data {
+	u64 epoch_ns;
+	u32 epoch_cyc;
+	u32 epoch_cyc_copy;
+	u32 mult;
+	u32 shift;
+};
+
 static void sched_clock_poll(unsigned long wrap_ticks);
 static DEFINE_TIMER(sched_clock_timer, sched_clock_poll, 0, 0);
-static void (*sched_clock_update_fn)(void);
+
+static struct clock_data cd = {
+	.mult	= NSEC_PER_SEC / HZ,
+};
+
+static u32 __read_mostly sched_clock_mask = 0xffffffff;
+
+static u32 notrace jiffy_sched_clock_read(void)
+{
+	return (u32)(jiffies - INITIAL_JIFFIES);
+}
+
+static u32 __read_mostly (*read_sched_clock)(void) = jiffy_sched_clock_read;
+
+static inline u64 cyc_to_ns(u64 cyc, u32 mult, u32 shift)
+{
+	return (cyc * mult) >> shift;
+}
+
+static unsigned long long cyc_to_sched_clock(u32 cyc, u32 mask)
+{
+	u64 epoch_ns;
+	u32 epoch_cyc;
+
+	/*
+	 * Load the epoch_cyc and epoch_ns atomically.  We do this by
+	 * ensuring that we always write epoch_cyc, epoch_ns and
+	 * epoch_cyc_copy in strict order, and read them in strict order.
+	 * If epoch_cyc and epoch_cyc_copy are not equal, then we're in
+	 * the middle of an update, and we should repeat the load.
+	 */
+	do {
+		epoch_cyc = cd.epoch_cyc;
+		smp_rmb();
+		epoch_ns = cd.epoch_ns;
+		smp_rmb();
+	} while (epoch_cyc != cd.epoch_cyc_copy);
+
+	return epoch_ns + cyc_to_ns((cyc - epoch_cyc) & mask, cd.mult, cd.shift);
+}
+
+/*
+ * Atomically update the sched_clock epoch.
+ */
+static void notrace update_sched_clock(void)
+{
+	unsigned long flags;
+	u32 cyc;
+	u64 ns;
+
+	cyc = read_sched_clock();
+	ns = cd.epoch_ns +
+		cyc_to_ns((cyc - cd.epoch_cyc) & sched_clock_mask,
+			  cd.mult, cd.shift);
+	/*
+	 * Write epoch_cyc and epoch_ns in a way that the update is
+	 * detectable in cyc_to_fixed_sched_clock().
+	 */
+	raw_local_irq_save(flags);
+	cd.epoch_cyc = cyc;
+	smp_wmb();
+	cd.epoch_ns = ns;
+	smp_wmb();
+	cd.epoch_cyc_copy = cyc;
+	raw_local_irq_restore(flags);
+}
 
 static void sched_clock_poll(unsigned long wrap_ticks)
 {
 	mod_timer(&sched_clock_timer, round_jiffies(jiffies + wrap_ticks));
-	sched_clock_update_fn();
+	update_sched_clock();
 }
 
-void __init init_sched_clock(struct clock_data *cd, void (*update)(void),
-	unsigned int clock_bits, unsigned long rate)
+void __init setup_sched_clock(u32 (*read)(void), int bits, unsigned long rate)
 {
 	unsigned long r, w;
 	u64 res, wrap;
 	char r_unit;
 
-	sched_clock_update_fn = update;
+	BUG_ON(bits > 32);
+	WARN_ON(!irqs_disabled());
+	WARN_ON(read_sched_clock != jiffy_sched_clock_read);
+	read_sched_clock = read;
+	sched_clock_mask = (1 << bits) - 1;
 
 	/* calculate the mult/shift to convert counter ticks to ns. */
-	clocks_calc_mult_shift(&cd->mult, &cd->shift, rate, NSEC_PER_SEC, 0);
+	clocks_calc_mult_shift(&cd.mult, &cd.shift, rate, NSEC_PER_SEC, 0);
 
 	r = rate;
 	if (r >= 4000000) {
 		r /= 1000000;
 		r_unit = 'M';
-	} else {
+	} else if (r >= 1000) {
 		r /= 1000;
 		r_unit = 'k';
-	}
+	} else
+		r_unit = ' ';
 
 	/* calculate how many ns until we wrap */
-	wrap = cyc_to_ns((1ULL << clock_bits) - 1, cd->mult, cd->shift);
+	wrap = cyc_to_ns((1ULL << bits) - 1, cd.mult, cd.shift);
 	do_div(wrap, NSEC_PER_MSEC);
 	w = wrap;
 
 	/* calculate the ns resolution of this counter */
-	res = cyc_to_ns(1ULL, cd->mult, cd->shift);
+	res = cyc_to_ns(1ULL, cd.mult, cd.shift);
 	pr_info("sched_clock: %u bits at %lu%cHz, resolution %lluns, wraps every %lums\n",
-		clock_bits, r, r_unit, res, w);
+		bits, r, r_unit, res, w);
 
 	/*
 	 * Start the timer to keep sched_clock() properly updated and
 	 * sets the initial epoch.
 	 */
 	sched_clock_timer.data = msecs_to_jiffies(w - (w / 10));
-	update();
+	update_sched_clock();
 
 	/*
 	 * Ensure that sched_clock() starts off at 0ns
 	 */
-	cd->epoch_ns = 0;
+	cd.epoch_ns = 0;
+
+	pr_debug("Registered %pF as sched_clock source\n", read);
+}
+
+unsigned long long notrace sched_clock(void)
+{
+	u32 cyc = read_sched_clock();
+	return cyc_to_sched_clock(cyc, sched_clock_mask);
 }
 
 void __init sched_clock_postinit(void)
 {
+	/*
+	 * If no sched_clock function has been provided at that point,
+	 * make it the final one one.
+	 */
+	if (read_sched_clock == jiffy_sched_clock_read)
+		setup_sched_clock(jiffy_sched_clock_read, 32, HZ);
+
 	sched_clock_poll(sched_clock_timer.data);
 }
diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c
index 8fc2c8fcbdc6..095d6611c84e 100644
--- a/arch/arm/kernel/setup.c
+++ b/arch/arm/kernel/setup.c
@@ -31,6 +31,7 @@
 #include <linux/memblock.h>
 #include <linux/bug.h>
 #include <linux/compiler.h>
+#include <linux/sort.h>
 
 #include <asm/unified.h>
 #include <asm/cpu.h>
@@ -890,6 +891,12 @@ static struct machine_desc * __init setup_machine_tags(unsigned int nr)
 	return mdesc;
 }
 
+static int __init meminfo_cmp(const void *_a, const void *_b)
+{
+	const struct membank *a = _a, *b = _b;
+	long cmp = bank_pfn_start(a) - bank_pfn_start(b);
+	return cmp < 0 ? -1 : cmp > 0 ? 1 : 0;
+}
 
 void __init setup_arch(char **cmdline_p)
 {
@@ -908,8 +915,8 @@ void __init setup_arch(char **cmdline_p)
 		arm_dma_zone_size = mdesc->dma_zone_size;
 	}
 #endif
-	if (mdesc->soft_reboot)
-		reboot_setup("s");
+	if (mdesc->restart_mode)
+		reboot_setup(&mdesc->restart_mode);
 
 	init_mm.start_code = (unsigned long) _text;
 	init_mm.end_code   = (unsigned long) _etext;
@@ -922,12 +929,16 @@ void __init setup_arch(char **cmdline_p)
 
 	parse_early_param();
 
+	sort(&meminfo.bank, meminfo.nr_banks, sizeof(meminfo.bank[0]), meminfo_cmp, NULL);
 	sanity_check_meminfo();
 	arm_memblock_init(&meminfo, mdesc);
 
 	paging_init(mdesc);
 	request_standard_resources(mdesc);
 
+	if (mdesc->restart)
+		arm_pm_restart = mdesc->restart;
+
 	unflatten_device_tree();
 
 #ifdef CONFIG_SMP
diff --git a/arch/arm/kernel/sleep.S b/arch/arm/kernel/sleep.S
index 020e99c845e7..1f268bda4552 100644
--- a/arch/arm/kernel/sleep.S
+++ b/arch/arm/kernel/sleep.S
@@ -54,14 +54,18 @@ ENDPROC(cpu_suspend_abort)
  * r0 = control register value
  */
 	.align	5
+	.pushsection	.idmap.text,"ax"
 ENTRY(cpu_resume_mmu)
 	ldr	r3, =cpu_resume_after_mmu
+	instr_sync
 	mcr	p15, 0, r0, c1, c0, 0	@ turn on MMU, I-cache, etc
 	mrc	p15, 0, r0, c0, c0, 0	@ read id reg
+	instr_sync
 	mov	r0, r0
 	mov	r0, r0
 	mov	pc, r3			@ jump to virtual address
 ENDPROC(cpu_resume_mmu)
+	.popsection
 cpu_resume_after_mmu:
 	bl	cpu_init		@ restore the und/abt/irq banked regs
 	mov	r0, #0			@ return zero on success
diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
index ef5640b9e218..57db122a4f62 100644
--- a/arch/arm/kernel/smp.c
+++ b/arch/arm/kernel/smp.c
@@ -31,6 +31,7 @@
 #include <asm/cpu.h>
 #include <asm/cputype.h>
 #include <asm/exception.h>
+#include <asm/idmap.h>
 #include <asm/topology.h>
 #include <asm/mmu_context.h>
 #include <asm/pgtable.h>
@@ -61,7 +62,6 @@ int __cpuinit __cpu_up(unsigned int cpu)
 {
 	struct cpuinfo_arm *ci = &per_cpu(cpu_data, cpu);
 	struct task_struct *idle = ci->idle;
-	pgd_t *pgd;
 	int ret;
 
 	/*
@@ -84,29 +84,11 @@ int __cpuinit __cpu_up(unsigned int cpu)
 	}
 
 	/*
-	 * Allocate initial page tables to allow the new CPU to
-	 * enable the MMU safely.  This essentially means a set
-	 * of our "standard" page tables, with the addition of
-	 * a 1:1 mapping for the physical address of the kernel.
-	 */
-	pgd = pgd_alloc(&init_mm);
-	if (!pgd)
-		return -ENOMEM;
-
-	if (PHYS_OFFSET != PAGE_OFFSET) {
-#ifndef CONFIG_HOTPLUG_CPU
-		identity_mapping_add(pgd, __pa(__init_begin), __pa(__init_end));
-#endif
-		identity_mapping_add(pgd, __pa(_stext), __pa(_etext));
-		identity_mapping_add(pgd, __pa(_sdata), __pa(_edata));
-	}
-
-	/*
 	 * We need to tell the secondary core where to find
 	 * its stack and the page tables.
 	 */
 	secondary_data.stack = task_stack_page(idle) + THREAD_START_SP;
-	secondary_data.pgdir = virt_to_phys(pgd);
+	secondary_data.pgdir = virt_to_phys(idmap_pgd);
 	secondary_data.swapper_pg_dir = virt_to_phys(swapper_pg_dir);
 	__cpuc_flush_dcache_area(&secondary_data, sizeof(secondary_data));
 	outer_clean_range(__pa(&secondary_data), __pa(&secondary_data + 1));
@@ -142,16 +124,6 @@ int __cpuinit __cpu_up(unsigned int cpu)
 	secondary_data.stack = NULL;
 	secondary_data.pgdir = 0;
 
-	if (PHYS_OFFSET != PAGE_OFFSET) {
-#ifndef CONFIG_HOTPLUG_CPU
-		identity_mapping_del(pgd, __pa(__init_begin), __pa(__init_end));
-#endif
-		identity_mapping_del(pgd, __pa(_stext), __pa(_etext));
-		identity_mapping_del(pgd, __pa(_sdata), __pa(_edata));
-	}
-
-	pgd_free(&init_mm, pgd);
-
 	return ret;
 }
 
@@ -550,6 +522,10 @@ static void ipi_cpu_stop(unsigned int cpu)
 	local_fiq_disable();
 	local_irq_disable();
 
+#ifdef CONFIG_HOTPLUG_CPU
+	platform_cpu_kill(cpu);
+#endif
+
 	while (1)
 		cpu_relax();
 }
diff --git a/arch/arm/kernel/smp_twd.c b/arch/arm/kernel/smp_twd.c
index a8a6682d6b52..c8e938553d47 100644
--- a/arch/arm/kernel/smp_twd.c
+++ b/arch/arm/kernel/smp_twd.c
@@ -10,8 +10,11 @@
  */
 #include <linux/init.h>
 #include <linux/kernel.h>
+#include <linux/clk.h>
+#include <linux/cpufreq.h>
 #include <linux/delay.h>
 #include <linux/device.h>
+#include <linux/err.h>
 #include <linux/smp.h>
 #include <linux/jiffies.h>
 #include <linux/clockchips.h>
@@ -25,6 +28,7 @@
 /* set up by the platform code */
 void __iomem *twd_base;
 
+static struct clk *twd_clk;
 static unsigned long twd_timer_rate;
 
 static struct clock_event_device __percpu **twd_evt;
@@ -89,6 +93,52 @@ void twd_timer_stop(struct clock_event_device *clk)
 	disable_percpu_irq(clk->irq);
 }
 
+#ifdef CONFIG_CPU_FREQ
+
+/*
+ * Updates clockevent frequency when the cpu frequency changes.
+ * Called on the cpu that is changing frequency with interrupts disabled.
+ */
+static void twd_update_frequency(void *data)
+{
+	twd_timer_rate = clk_get_rate(twd_clk);
+
+	clockevents_update_freq(*__this_cpu_ptr(twd_evt), twd_timer_rate);
+}
+
+static int twd_cpufreq_transition(struct notifier_block *nb,
+	unsigned long state, void *data)
+{
+	struct cpufreq_freqs *freqs = data;
+
+	/*
+	 * The twd clock events must be reprogrammed to account for the new
+	 * frequency.  The timer is local to a cpu, so cross-call to the
+	 * changing cpu.
+	 */
+	if (state == CPUFREQ_POSTCHANGE || state == CPUFREQ_RESUMECHANGE)
+		smp_call_function_single(freqs->cpu, twd_update_frequency,
+			NULL, 1);
+
+	return NOTIFY_OK;
+}
+
+static struct notifier_block twd_cpufreq_nb = {
+	.notifier_call = twd_cpufreq_transition,
+};
+
+static int twd_cpufreq_init(void)
+{
+	if (!IS_ERR(twd_clk))
+		return cpufreq_register_notifier(&twd_cpufreq_nb,
+			CPUFREQ_TRANSITION_NOTIFIER);
+
+	return 0;
+}
+core_initcall(twd_cpufreq_init);
+
+#endif
+
 static void __cpuinit twd_calibrate_rate(void)
 {
 	unsigned long count;
@@ -140,6 +190,35 @@ static irqreturn_t twd_handler(int irq, void *dev_id)
 	return IRQ_NONE;
 }
 
+static struct clk *twd_get_clock(void)
+{
+	struct clk *clk;
+	int err;
+
+	clk = clk_get_sys("smp_twd", NULL);
+	if (IS_ERR(clk)) {
+		pr_err("smp_twd: clock not found: %d\n", (int)PTR_ERR(clk));
+		return clk;
+	}
+
+	err = clk_prepare(clk);
+	if (err) {
+		pr_err("smp_twd: clock failed to prepare: %d\n", err);
+		clk_put(clk);
+		return ERR_PTR(err);
+	}
+
+	err = clk_enable(clk);
+	if (err) {
+		pr_err("smp_twd: clock failed to enable: %d\n", err);
+		clk_unprepare(clk);
+		clk_put(clk);
+		return ERR_PTR(err);
+	}
+
+	return clk;
+}
+
 /*
  * Setup the local clock events for a CPU.
  */
@@ -165,7 +244,13 @@ void __cpuinit twd_timer_setup(struct clock_event_device *clk)
 		}
 	}
 
-	twd_calibrate_rate();
+	if (!twd_clk)
+		twd_clk = twd_get_clock();
+
+	if (!IS_ERR_OR_NULL(twd_clk))
+		twd_timer_rate = clk_get_rate(twd_clk);
+	else
+		twd_calibrate_rate();
 
 	clk->name = "local_timer";
 	clk->features = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT |
@@ -173,15 +258,11 @@ void __cpuinit twd_timer_setup(struct clock_event_device *clk)
 	clk->rating = 350;
 	clk->set_mode = twd_set_mode;
 	clk->set_next_event = twd_set_next_event;
-	clk->shift = 20;
-	clk->mult = div_sc(twd_timer_rate, NSEC_PER_SEC, clk->shift);
-	clk->max_delta_ns = clockevent_delta2ns(0xffffffff, clk);
-	clk->min_delta_ns = clockevent_delta2ns(0xf, clk);
 
 	this_cpu_clk = __this_cpu_ptr(twd_evt);
 	*this_cpu_clk = clk;
 
-	clockevents_register_device(clk);
-
+	clockevents_config_and_register(clk, twd_timer_rate,
+					0xf, 0xffffffff);
 	enable_percpu_irq(clk->irq, 0);
 }
diff --git a/arch/arm/kernel/suspend.c b/arch/arm/kernel/suspend.c
index 93a22d282c16..1794cc3b0f18 100644
--- a/arch/arm/kernel/suspend.c
+++ b/arch/arm/kernel/suspend.c
@@ -1,13 +1,12 @@
 #include <linux/init.h>
 
+#include <asm/idmap.h>
 #include <asm/pgalloc.h>
 #include <asm/pgtable.h>
 #include <asm/memory.h>
 #include <asm/suspend.h>
 #include <asm/tlbflush.h>
 
-static pgd_t *suspend_pgd;
-
 extern int __cpu_suspend(unsigned long, int (*)(unsigned long));
 extern void cpu_resume_mmu(void);
 
@@ -21,7 +20,7 @@ void __cpu_suspend_save(u32 *ptr, u32 ptrsz, u32 sp, u32 *save_ptr)
 	*save_ptr = virt_to_phys(ptr);
 
 	/* This must correspond to the LDM in cpu_resume() assembly */
-	*ptr++ = virt_to_phys(suspend_pgd);
+	*ptr++ = virt_to_phys(idmap_pgd);
 	*ptr++ = sp;
 	*ptr++ = virt_to_phys(cpu_do_resume);
 
@@ -42,7 +41,7 @@ int cpu_suspend(unsigned long arg, int (*fn)(unsigned long))
 	struct mm_struct *mm = current->active_mm;
 	int ret;
 
-	if (!suspend_pgd)
+	if (!idmap_pgd)
 		return -EINVAL;
 
 	/*
@@ -59,14 +58,3 @@ int cpu_suspend(unsigned long arg, int (*fn)(unsigned long))
 
 	return ret;
 }
-
-static int __init cpu_suspend_init(void)
-{
-	suspend_pgd = pgd_alloc(&init_mm);
-	if (suspend_pgd) {
-		unsigned long addr = virt_to_phys(cpu_resume_mmu);
-		identity_mapping_add(suspend_pgd, addr, addr + SECTION_SIZE);
-	}
-	return suspend_pgd ? 0 : -ENOMEM;
-}
-core_initcall(cpu_suspend_init);
diff --git a/arch/arm/kernel/swp_emulate.c b/arch/arm/kernel/swp_emulate.c
index 5f452f8fde05..df745188f5de 100644
--- a/arch/arm/kernel/swp_emulate.c
+++ b/arch/arm/kernel/swp_emulate.c
@@ -25,6 +25,7 @@
 #include <linux/syscalls.h>
 #include <linux/perf_event.h>
 
+#include <asm/opcodes.h>
 #include <asm/traps.h>
 #include <asm/uaccess.h>
 
@@ -185,6 +186,21 @@ static int swp_handler(struct pt_regs *regs, unsigned int instr)
 
 	perf_sw_event(PERF_COUNT_SW_EMULATION_FAULTS, 1, regs, regs->ARM_pc);
 
+	res = arm_check_condition(instr, regs->ARM_cpsr);
+	switch (res) {
+	case ARM_OPCODE_CONDTEST_PASS:
+		break;
+	case ARM_OPCODE_CONDTEST_FAIL:
+		/* Condition failed - return to next instruction */
+		regs->ARM_pc += 4;
+		return 0;
+	case ARM_OPCODE_CONDTEST_UNCOND:
+		/* If unconditional encoding - not a SWP, undef */
+		return -EFAULT;
+	default:
+		return -EINVAL;
+	}
+
 	if (current->pid != previous_pid) {
 		pr_debug("\"%s\" (%ld) uses deprecated SWP{B} instruction\n",
 			 current->comm, (unsigned long)current->pid);
diff --git a/arch/arm/kernel/tcm.c b/arch/arm/kernel/tcm.c
index 30e302d33e0a..01ec453bb924 100644
--- a/arch/arm/kernel/tcm.c
+++ b/arch/arm/kernel/tcm.c
@@ -180,9 +180,9 @@ static int __init setup_tcm_bank(u8 type, u8 bank, u8 banks,
  */
 void __init tcm_init(void)
 {
-	u32 tcm_status = read_cpuid_tcmstatus();
-	u8 dtcm_banks = (tcm_status >> 16) & 0x03;
-	u8 itcm_banks = (tcm_status & 0x03);
+	u32 tcm_status;
+	u8 dtcm_banks;
+	u8 itcm_banks;
 	size_t dtcm_code_sz = &__edtcm_data - &__sdtcm_data;
 	size_t itcm_code_sz = &__eitcm_text - &__sitcm_text;
 	char *start;
@@ -191,6 +191,22 @@ void __init tcm_init(void)
 	int ret;
 	int i;
 
+	/*
+	 * Prior to ARMv5 there is no TCM, and trying to read the status
+	 * register will hang the processor.
+	 */
+	if (cpu_architecture() < CPU_ARCH_ARMv5) {
+		if (dtcm_code_sz || itcm_code_sz)
+			pr_info("CPU TCM: %u bytes of DTCM and %u bytes of "
+				"ITCM code compiled in, but no TCM present "
+				"in pre-v5 CPU\n", dtcm_code_sz, itcm_code_sz);
+		return;
+	}
+
+	tcm_status = read_cpuid_tcmstatus();
+	dtcm_banks = (tcm_status >> 16) & 0x03;
+	itcm_banks = (tcm_status & 0x03);
+
 	/* Values greater than 2 for D/ITCM banks are "reserved" */
 	if (dtcm_banks > 2)
 		dtcm_banks = 0;
diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S
index 20b3041e0860..f76e75548670 100644
--- a/arch/arm/kernel/vmlinux.lds.S
+++ b/arch/arm/kernel/vmlinux.lds.S
@@ -13,6 +13,12 @@
 	*(.proc.info.init)						\
 	VMLINUX_SYMBOL(__proc_info_end) = .;
 
+#define IDMAP_TEXT							\
+	ALIGN_FUNCTION();						\
+	VMLINUX_SYMBOL(__idmap_text_start) = .;				\
+	*(.idmap.text)							\
+	VMLINUX_SYMBOL(__idmap_text_end) = .;
+
 #ifdef CONFIG_HOTPLUG_CPU
 #define ARM_CPU_DISCARD(x)
 #define ARM_CPU_KEEP(x)		x
@@ -92,6 +98,7 @@ SECTIONS
 			SCHED_TEXT
 			LOCK_TEXT
 			KPROBES_TEXT
+			IDMAP_TEXT
 #ifdef CONFIG_MMU
 			*(.fixup)
 #endif