File-copy from v4.4.100

This is the result of 'cp' from a linux-stable tree with the 'v4.4.100'
tag checked out (commit 26d6298789e695c9f627ce49a7bbd2286405798a) on
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git

Please refer to that tree for all history prior to this point.

Change-Id: I8a9ee2aea93cd29c52c847d0ce33091a73ae6afe
diff --git a/arch/microblaze/kernel/.gitignore b/arch/microblaze/kernel/.gitignore
new file mode 100644
index 0000000..c5f676c
--- /dev/null
+++ b/arch/microblaze/kernel/.gitignore
@@ -0,0 +1 @@
+vmlinux.lds
diff --git a/arch/microblaze/kernel/Makefile b/arch/microblaze/kernel/Makefile
new file mode 100644
index 0000000..f08baca
--- /dev/null
+++ b/arch/microblaze/kernel/Makefile
@@ -0,0 +1,32 @@
+#
+# Makefile
+#
+
+ifdef CONFIG_FUNCTION_TRACER
+# Do not trace early boot code and low level code
+CFLAGS_REMOVE_timer.o = -pg
+CFLAGS_REMOVE_intc.o = -pg
+CFLAGS_REMOVE_early_printk.o = -pg
+CFLAGS_REMOVE_heartbeat.o = -pg
+CFLAGS_REMOVE_ftrace.o = -pg
+CFLAGS_REMOVE_process.o = -pg
+endif
+
+extra-y := head.o vmlinux.lds
+
+obj-y += dma.o exceptions.o \
+	hw_exception_handler.o intc.o irq.o \
+	platform.o process.o prom.o ptrace.o \
+	reset.o setup.o signal.o sys_microblaze.o timer.o traps.o unwind.o
+
+obj-y += cpu/
+
+obj-$(CONFIG_EARLY_PRINTK)	+= early_printk.o
+obj-$(CONFIG_HEART_BEAT)	+= heartbeat.o
+obj-$(CONFIG_MODULES)		+= microblaze_ksyms.o module.o
+obj-$(CONFIG_MMU)		+= misc.o
+obj-$(CONFIG_STACKTRACE)	+= stacktrace.o
+obj-$(CONFIG_FUNCTION_TRACER)	+= ftrace.o mcount.o
+obj-$(CONFIG_KGDB)		+= kgdb.o
+
+obj-y	+= entry$(MMU).o
diff --git a/arch/microblaze/kernel/asm-offsets.c b/arch/microblaze/kernel/asm-offsets.c
new file mode 100644
index 0000000..c1b459c
--- /dev/null
+++ b/arch/microblaze/kernel/asm-offsets.c
@@ -0,0 +1,128 @@
+/*
+ * Copyright (C) 2007-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/init.h>
+#include <linux/stddef.h>
+#include <linux/sched.h>
+#include <linux/kernel_stat.h>
+#include <linux/ptrace.h>
+#include <linux/hardirq.h>
+#include <linux/thread_info.h>
+#include <linux/kbuild.h>
+#include <asm/cpuinfo.h>
+
+int main(int argc, char *argv[])
+{
+	/* struct pt_regs */
+	DEFINE(PT_SIZE, sizeof(struct pt_regs));
+	DEFINE(PT_MSR, offsetof(struct pt_regs, msr));
+	DEFINE(PT_EAR, offsetof(struct pt_regs, ear));
+	DEFINE(PT_ESR, offsetof(struct pt_regs, esr));
+	DEFINE(PT_FSR, offsetof(struct pt_regs, fsr));
+	DEFINE(PT_PC, offsetof(struct pt_regs, pc));
+	DEFINE(PT_R0, offsetof(struct pt_regs, r0));
+	DEFINE(PT_R1, offsetof(struct pt_regs, r1));
+	DEFINE(PT_R2, offsetof(struct pt_regs, r2));
+	DEFINE(PT_R3, offsetof(struct pt_regs, r3));
+	DEFINE(PT_R4, offsetof(struct pt_regs, r4));
+	DEFINE(PT_R5, offsetof(struct pt_regs, r5));
+	DEFINE(PT_R6, offsetof(struct pt_regs, r6));
+	DEFINE(PT_R7, offsetof(struct pt_regs, r7));
+	DEFINE(PT_R8, offsetof(struct pt_regs, r8));
+	DEFINE(PT_R9, offsetof(struct pt_regs, r9));
+	DEFINE(PT_R10, offsetof(struct pt_regs, r10));
+	DEFINE(PT_R11, offsetof(struct pt_regs, r11));
+	DEFINE(PT_R12, offsetof(struct pt_regs, r12));
+	DEFINE(PT_R13, offsetof(struct pt_regs, r13));
+	DEFINE(PT_R14, offsetof(struct pt_regs, r14));
+	DEFINE(PT_R15, offsetof(struct pt_regs, r15));
+	DEFINE(PT_R16, offsetof(struct pt_regs, r16));
+	DEFINE(PT_R17, offsetof(struct pt_regs, r17));
+	DEFINE(PT_R18, offsetof(struct pt_regs, r18));
+	DEFINE(PT_R19, offsetof(struct pt_regs, r19));
+	DEFINE(PT_R20, offsetof(struct pt_regs, r20));
+	DEFINE(PT_R21, offsetof(struct pt_regs, r21));
+	DEFINE(PT_R22, offsetof(struct pt_regs, r22));
+	DEFINE(PT_R23, offsetof(struct pt_regs, r23));
+	DEFINE(PT_R24, offsetof(struct pt_regs, r24));
+	DEFINE(PT_R25, offsetof(struct pt_regs, r25));
+	DEFINE(PT_R26, offsetof(struct pt_regs, r26));
+	DEFINE(PT_R27, offsetof(struct pt_regs, r27));
+	DEFINE(PT_R28, offsetof(struct pt_regs, r28));
+	DEFINE(PT_R29, offsetof(struct pt_regs, r29));
+	DEFINE(PT_R30, offsetof(struct pt_regs, r30));
+	DEFINE(PT_R31, offsetof(struct pt_regs, r31));
+	DEFINE(PT_MODE, offsetof(struct pt_regs, pt_mode));
+	BLANK();
+
+	/* Magic offsets for PTRACE PEEK/POKE etc */
+	DEFINE(PT_TEXT_ADDR, sizeof(struct pt_regs) + 1);
+	DEFINE(PT_TEXT_LEN, sizeof(struct pt_regs) + 2);
+	DEFINE(PT_DATA_ADDR, sizeof(struct pt_regs) + 3);
+	BLANK();
+
+	/* struct task_struct */
+	DEFINE(TS_THREAD_INFO, offsetof(struct task_struct, stack));
+#ifdef CONFIG_MMU
+	DEFINE(TASK_STATE, offsetof(struct task_struct, state));
+	DEFINE(TASK_FLAGS, offsetof(struct task_struct, flags));
+	DEFINE(TASK_PTRACE, offsetof(struct task_struct, ptrace));
+	DEFINE(TASK_BLOCKED, offsetof(struct task_struct, blocked));
+	DEFINE(TASK_MM, offsetof(struct task_struct, mm));
+	DEFINE(TASK_ACTIVE_MM, offsetof(struct task_struct, active_mm));
+	DEFINE(TASK_PID, offsetof(struct task_struct, pid));
+	DEFINE(TASK_THREAD, offsetof(struct task_struct, thread));
+	DEFINE(THREAD_KSP, offsetof(struct thread_struct, ksp));
+	BLANK();
+
+	DEFINE(PGDIR, offsetof(struct thread_struct, pgdir));
+	BLANK();
+#endif
+
+	/* struct thread_info */
+	DEFINE(TI_TASK, offsetof(struct thread_info, task));
+	DEFINE(TI_FLAGS, offsetof(struct thread_info, flags));
+	DEFINE(TI_ADDR_LIMIT, offsetof(struct thread_info, addr_limit));
+	DEFINE(TI_CPU_CONTEXT, offsetof(struct thread_info, cpu_context));
+	DEFINE(TI_PREEMPT_COUNT, offsetof(struct thread_info, preempt_count));
+	BLANK();
+
+	/* struct cpu_context */
+	DEFINE(CC_R1, offsetof(struct cpu_context, r1)); /* r1 */
+	DEFINE(CC_R2, offsetof(struct cpu_context, r2));
+	/* dedicated registers */
+	DEFINE(CC_R13, offsetof(struct cpu_context, r13));
+	DEFINE(CC_R14, offsetof(struct cpu_context, r14));
+	DEFINE(CC_R15, offsetof(struct cpu_context, r15));
+	DEFINE(CC_R16, offsetof(struct cpu_context, r16));
+	DEFINE(CC_R17, offsetof(struct cpu_context, r17));
+	DEFINE(CC_R18, offsetof(struct cpu_context, r18));
+	/* non-volatile registers */
+	DEFINE(CC_R19, offsetof(struct cpu_context, r19));
+	DEFINE(CC_R20, offsetof(struct cpu_context, r20));
+	DEFINE(CC_R21, offsetof(struct cpu_context, r21));
+	DEFINE(CC_R22, offsetof(struct cpu_context, r22));
+	DEFINE(CC_R23, offsetof(struct cpu_context, r23));
+	DEFINE(CC_R24, offsetof(struct cpu_context, r24));
+	DEFINE(CC_R25, offsetof(struct cpu_context, r25));
+	DEFINE(CC_R26, offsetof(struct cpu_context, r26));
+	DEFINE(CC_R27, offsetof(struct cpu_context, r27));
+	DEFINE(CC_R28, offsetof(struct cpu_context, r28));
+	DEFINE(CC_R29, offsetof(struct cpu_context, r29));
+	DEFINE(CC_R30, offsetof(struct cpu_context, r30));
+	/* special purpose registers */
+	DEFINE(CC_MSR, offsetof(struct cpu_context, msr));
+	DEFINE(CC_EAR, offsetof(struct cpu_context, ear));
+	DEFINE(CC_ESR, offsetof(struct cpu_context, esr));
+	DEFINE(CC_FSR, offsetof(struct cpu_context, fsr));
+	BLANK();
+
+	return 0;
+}
diff --git a/arch/microblaze/kernel/cpu/Makefile b/arch/microblaze/kernel/cpu/Makefile
new file mode 100644
index 0000000..fceed4e
--- /dev/null
+++ b/arch/microblaze/kernel/cpu/Makefile
@@ -0,0 +1,12 @@
+#
+# Build the appropriate CPU version support
+#
+
+ifdef CONFIG_FUNCTION_TRACER
+CFLAGS_REMOVE_cache.o = -pg
+endif
+
+ccflags-y := -DCPU_MAJOR=$(CPU_MAJOR) -DCPU_MINOR=$(CPU_MINOR) \
+		-DCPU_REV=$(CPU_REV)
+
+obj-y += cache.o cpuinfo.o cpuinfo-pvr-full.o cpuinfo-static.o mb.o pvr.o
diff --git a/arch/microblaze/kernel/cpu/cache.c b/arch/microblaze/kernel/cpu/cache.c
new file mode 100644
index 0000000..0bde47e
--- /dev/null
+++ b/arch/microblaze/kernel/cpu/cache.c
@@ -0,0 +1,655 @@
+/*
+ * Cache control for MicroBlaze cache memories
+ *
+ * Copyright (C) 2007-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007-2009 PetaLogix
+ * Copyright (C) 2007-2009 John Williams <john.williams@petalogix.com>
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License. See the file COPYING in the main directory of this
+ * archive for more details.
+ */
+
+#include <asm/cacheflush.h>
+#include <linux/cache.h>
+#include <asm/cpuinfo.h>
+#include <asm/pvr.h>
+
+static inline void __enable_icache_msr(void)
+{
+	__asm__ __volatile__ ("	 msrset	r0, %0;"	\
+				"nop;"			\
+			: : "i" (MSR_ICE) : "memory");
+}
+
+static inline void __disable_icache_msr(void)
+{
+	__asm__ __volatile__ ("	 msrclr	r0, %0;"	\
+				"nop;"			\
+			: : "i" (MSR_ICE) : "memory");
+}
+
+static inline void __enable_dcache_msr(void)
+{
+	__asm__ __volatile__ ("	 msrset	r0, %0;"	\
+				"nop;"			\
+			: : "i" (MSR_DCE) : "memory");
+}
+
+static inline void __disable_dcache_msr(void)
+{
+	__asm__ __volatile__ ("	 msrclr	r0, %0;"	\
+				"nop; "			\
+			: : "i" (MSR_DCE) : "memory");
+}
+
+static inline void __enable_icache_nomsr(void)
+{
+	__asm__ __volatile__ ("	 mfs	r12, rmsr;"	\
+				"nop;"			\
+				"ori	r12, r12, %0;"	\
+				"mts	rmsr, r12;"	\
+				"nop;"			\
+			: : "i" (MSR_ICE) : "memory", "r12");
+}
+
+static inline void __disable_icache_nomsr(void)
+{
+	__asm__ __volatile__ ("	 mfs	r12, rmsr;"	\
+				"nop;"			\
+				"andi	r12, r12, ~%0;"	\
+				"mts	rmsr, r12;"	\
+				"nop;"			\
+			: : "i" (MSR_ICE) : "memory", "r12");
+}
+
+static inline void __enable_dcache_nomsr(void)
+{
+	__asm__ __volatile__ ("	 mfs	r12, rmsr;"	\
+				"nop;"			\
+				"ori	r12, r12, %0;"	\
+				"mts	rmsr, r12;"	\
+				"nop;"			\
+			: : "i" (MSR_DCE) : "memory", "r12");
+}
+
+static inline void __disable_dcache_nomsr(void)
+{
+	__asm__ __volatile__ ("	 mfs	r12, rmsr;"	\
+				"nop;"			\
+				"andi	r12, r12, ~%0;"	\
+				"mts	rmsr, r12;"	\
+				"nop;"			\
+			: : "i" (MSR_DCE) : "memory", "r12");
+}
+
+
+/* Helper macro for computing the limits of cache range loops
+ *
+ * End address can be unaligned which is OK for C implementation.
+ * ASM implementation align it in ASM macros
+ */
+#define CACHE_LOOP_LIMITS(start, end, cache_line_length, cache_size)	\
+do {									\
+	int align = ~(cache_line_length - 1);				\
+	end = min(start + cache_size, end);				\
+	start &= align;							\
+} while (0)
+
+/*
+ * Helper macro to loop over the specified cache_size/line_length and
+ * execute 'op' on that cacheline
+ */
+#define CACHE_ALL_LOOP(cache_size, line_length, op)			\
+do {									\
+	unsigned int len = cache_size - line_length;			\
+	int step = -line_length;					\
+	WARN_ON(step >= 0);						\
+									\
+	__asm__ __volatile__ (" 1:      " #op " %0, r0;"		\
+					"bgtid   %0, 1b;"		\
+					"addk    %0, %0, %1;"		\
+					: : "r" (len), "r" (step)	\
+					: "memory");			\
+} while (0)
+
+/* Used for wdc.flush/clear which can use rB for offset which is not possible
+ * to use for simple wdc or wic.
+ *
+ * start address is cache aligned
+ * end address is not aligned, if end is aligned then I have to subtract
+ * cacheline length because I can't flush/invalidate the next cacheline.
+ * If is not, I align it because I will flush/invalidate whole line.
+ */
+#define CACHE_RANGE_LOOP_2(start, end, line_length, op)			\
+do {									\
+	int step = -line_length;					\
+	int align = ~(line_length - 1);					\
+	int count;							\
+	end = ((end & align) == end) ? end - line_length : end & align;	\
+	count = end - start;						\
+	WARN_ON(count < 0);						\
+									\
+	__asm__ __volatile__ (" 1:	" #op "	%0, %1;"		\
+					"bgtid	%1, 1b;"		\
+					"addk	%1, %1, %2;"		\
+					: : "r" (start), "r" (count),	\
+					"r" (step) : "memory");		\
+} while (0)
+
+/* It is used only first parameter for OP - for wic, wdc */
+#define CACHE_RANGE_LOOP_1(start, end, line_length, op)			\
+do {									\
+	unsigned int volatile temp = 0;						\
+	unsigned int align = ~(line_length - 1);					\
+	end = ((end & align) == end) ? end - line_length : end & align;	\
+	WARN_ON(end < start);					\
+									\
+	__asm__ __volatile__ (" 1:	" #op "	%1, r0;"		\
+					"cmpu	%0, %1, %2;"		\
+					"bgtid	%0, 1b;"		\
+					"addk	%1, %1, %3;"		\
+				: : "r" (temp), "r" (start), "r" (end),	\
+					"r" (line_length) : "memory");	\
+} while (0)
+
+#define ASM_LOOP
+
+static void __flush_icache_range_msr_irq(unsigned long start, unsigned long end)
+{
+	unsigned long flags;
+#ifndef ASM_LOOP
+	int i;
+#endif
+	pr_debug("%s: start 0x%x, end 0x%x\n", __func__,
+				(unsigned int)start, (unsigned int) end);
+
+	CACHE_LOOP_LIMITS(start, end,
+			cpuinfo.icache_line_length, cpuinfo.icache_size);
+
+	local_irq_save(flags);
+	__disable_icache_msr();
+
+#ifdef ASM_LOOP
+	CACHE_RANGE_LOOP_1(start, end, cpuinfo.icache_line_length, wic);
+#else
+	for (i = start; i < end; i += cpuinfo.icache_line_length)
+		__asm__ __volatile__ ("wic	%0, r0;"	\
+				: : "r" (i));
+#endif
+	__enable_icache_msr();
+	local_irq_restore(flags);
+}
+
+static void __flush_icache_range_nomsr_irq(unsigned long start,
+				unsigned long end)
+{
+	unsigned long flags;
+#ifndef ASM_LOOP
+	int i;
+#endif
+	pr_debug("%s: start 0x%x, end 0x%x\n", __func__,
+				(unsigned int)start, (unsigned int) end);
+
+	CACHE_LOOP_LIMITS(start, end,
+			cpuinfo.icache_line_length, cpuinfo.icache_size);
+
+	local_irq_save(flags);
+	__disable_icache_nomsr();
+
+#ifdef ASM_LOOP
+	CACHE_RANGE_LOOP_1(start, end, cpuinfo.icache_line_length, wic);
+#else
+	for (i = start; i < end; i += cpuinfo.icache_line_length)
+		__asm__ __volatile__ ("wic	%0, r0;"	\
+				: : "r" (i));
+#endif
+
+	__enable_icache_nomsr();
+	local_irq_restore(flags);
+}
+
+static void __flush_icache_range_noirq(unsigned long start,
+				unsigned long end)
+{
+#ifndef ASM_LOOP
+	int i;
+#endif
+	pr_debug("%s: start 0x%x, end 0x%x\n", __func__,
+				(unsigned int)start, (unsigned int) end);
+
+	CACHE_LOOP_LIMITS(start, end,
+			cpuinfo.icache_line_length, cpuinfo.icache_size);
+#ifdef ASM_LOOP
+	CACHE_RANGE_LOOP_1(start, end, cpuinfo.icache_line_length, wic);
+#else
+	for (i = start; i < end; i += cpuinfo.icache_line_length)
+		__asm__ __volatile__ ("wic	%0, r0;"	\
+				: : "r" (i));
+#endif
+}
+
+static void __flush_icache_all_msr_irq(void)
+{
+	unsigned long flags;
+#ifndef ASM_LOOP
+	int i;
+#endif
+	pr_debug("%s\n", __func__);
+
+	local_irq_save(flags);
+	__disable_icache_msr();
+#ifdef ASM_LOOP
+	CACHE_ALL_LOOP(cpuinfo.icache_size, cpuinfo.icache_line_length, wic);
+#else
+	for (i = 0; i < cpuinfo.icache_size;
+		 i += cpuinfo.icache_line_length)
+			__asm__ __volatile__ ("wic	%0, r0;" \
+					: : "r" (i));
+#endif
+	__enable_icache_msr();
+	local_irq_restore(flags);
+}
+
+static void __flush_icache_all_nomsr_irq(void)
+{
+	unsigned long flags;
+#ifndef ASM_LOOP
+	int i;
+#endif
+	pr_debug("%s\n", __func__);
+
+	local_irq_save(flags);
+	__disable_icache_nomsr();
+#ifdef ASM_LOOP
+	CACHE_ALL_LOOP(cpuinfo.icache_size, cpuinfo.icache_line_length, wic);
+#else
+	for (i = 0; i < cpuinfo.icache_size;
+		 i += cpuinfo.icache_line_length)
+			__asm__ __volatile__ ("wic	%0, r0;" \
+					: : "r" (i));
+#endif
+	__enable_icache_nomsr();
+	local_irq_restore(flags);
+}
+
+static void __flush_icache_all_noirq(void)
+{
+#ifndef ASM_LOOP
+	int i;
+#endif
+	pr_debug("%s\n", __func__);
+#ifdef ASM_LOOP
+	CACHE_ALL_LOOP(cpuinfo.icache_size, cpuinfo.icache_line_length, wic);
+#else
+	for (i = 0; i < cpuinfo.icache_size;
+		 i += cpuinfo.icache_line_length)
+			__asm__ __volatile__ ("wic	%0, r0;" \
+					: : "r" (i));
+#endif
+}
+
+static void __invalidate_dcache_all_msr_irq(void)
+{
+	unsigned long flags;
+#ifndef ASM_LOOP
+	int i;
+#endif
+	pr_debug("%s\n", __func__);
+
+	local_irq_save(flags);
+	__disable_dcache_msr();
+#ifdef ASM_LOOP
+	CACHE_ALL_LOOP(cpuinfo.dcache_size, cpuinfo.dcache_line_length, wdc);
+#else
+	for (i = 0; i < cpuinfo.dcache_size;
+		 i += cpuinfo.dcache_line_length)
+			__asm__ __volatile__ ("wdc	%0, r0;" \
+					: : "r" (i));
+#endif
+	__enable_dcache_msr();
+	local_irq_restore(flags);
+}
+
+static void __invalidate_dcache_all_nomsr_irq(void)
+{
+	unsigned long flags;
+#ifndef ASM_LOOP
+	int i;
+#endif
+	pr_debug("%s\n", __func__);
+
+	local_irq_save(flags);
+	__disable_dcache_nomsr();
+#ifdef ASM_LOOP
+	CACHE_ALL_LOOP(cpuinfo.dcache_size, cpuinfo.dcache_line_length, wdc);
+#else
+	for (i = 0; i < cpuinfo.dcache_size;
+		 i += cpuinfo.dcache_line_length)
+			__asm__ __volatile__ ("wdc	%0, r0;" \
+					: : "r" (i));
+#endif
+	__enable_dcache_nomsr();
+	local_irq_restore(flags);
+}
+
+static void __invalidate_dcache_all_noirq_wt(void)
+{
+#ifndef ASM_LOOP
+	int i;
+#endif
+	pr_debug("%s\n", __func__);
+#ifdef ASM_LOOP
+	CACHE_ALL_LOOP(cpuinfo.dcache_size, cpuinfo.dcache_line_length, wdc);
+#else
+	for (i = 0; i < cpuinfo.dcache_size;
+		 i += cpuinfo.dcache_line_length)
+			__asm__ __volatile__ ("wdc	%0, r0;" \
+					: : "r" (i));
+#endif
+}
+
+/*
+ * FIXME It is blindly invalidation as is expected
+ * but can't be called on noMMU in microblaze_cache_init below
+ *
+ * MS: noMMU kernel won't boot if simple wdc is used
+ * The reason should be that there are discared data which kernel needs
+ */
+static void __invalidate_dcache_all_wb(void)
+{
+#ifndef ASM_LOOP
+	int i;
+#endif
+	pr_debug("%s\n", __func__);
+#ifdef ASM_LOOP
+	CACHE_ALL_LOOP(cpuinfo.dcache_size, cpuinfo.dcache_line_length,
+					wdc);
+#else
+	for (i = 0; i < cpuinfo.dcache_size;
+		 i += cpuinfo.dcache_line_length)
+			__asm__ __volatile__ ("wdc	%0, r0;" \
+					: : "r" (i));
+#endif
+}
+
+static void __invalidate_dcache_range_wb(unsigned long start,
+						unsigned long end)
+{
+#ifndef ASM_LOOP
+	int i;
+#endif
+	pr_debug("%s: start 0x%x, end 0x%x\n", __func__,
+				(unsigned int)start, (unsigned int) end);
+
+	CACHE_LOOP_LIMITS(start, end,
+			cpuinfo.dcache_line_length, cpuinfo.dcache_size);
+#ifdef ASM_LOOP
+	CACHE_RANGE_LOOP_2(start, end, cpuinfo.dcache_line_length, wdc.clear);
+#else
+	for (i = start; i < end; i += cpuinfo.dcache_line_length)
+		__asm__ __volatile__ ("wdc.clear	%0, r0;"	\
+				: : "r" (i));
+#endif
+}
+
+static void __invalidate_dcache_range_nomsr_wt(unsigned long start,
+							unsigned long end)
+{
+#ifndef ASM_LOOP
+	int i;
+#endif
+	pr_debug("%s: start 0x%x, end 0x%x\n", __func__,
+				(unsigned int)start, (unsigned int) end);
+	CACHE_LOOP_LIMITS(start, end,
+			cpuinfo.dcache_line_length, cpuinfo.dcache_size);
+
+#ifdef ASM_LOOP
+	CACHE_RANGE_LOOP_1(start, end, cpuinfo.dcache_line_length, wdc);
+#else
+	for (i = start; i < end; i += cpuinfo.dcache_line_length)
+		__asm__ __volatile__ ("wdc	%0, r0;"	\
+				: : "r" (i));
+#endif
+}
+
+static void __invalidate_dcache_range_msr_irq_wt(unsigned long start,
+							unsigned long end)
+{
+	unsigned long flags;
+#ifndef ASM_LOOP
+	int i;
+#endif
+	pr_debug("%s: start 0x%x, end 0x%x\n", __func__,
+				(unsigned int)start, (unsigned int) end);
+	CACHE_LOOP_LIMITS(start, end,
+			cpuinfo.dcache_line_length, cpuinfo.dcache_size);
+
+	local_irq_save(flags);
+	__disable_dcache_msr();
+
+#ifdef ASM_LOOP
+	CACHE_RANGE_LOOP_1(start, end, cpuinfo.dcache_line_length, wdc);
+#else
+	for (i = start; i < end; i += cpuinfo.dcache_line_length)
+		__asm__ __volatile__ ("wdc	%0, r0;"	\
+				: : "r" (i));
+#endif
+
+	__enable_dcache_msr();
+	local_irq_restore(flags);
+}
+
+static void __invalidate_dcache_range_nomsr_irq(unsigned long start,
+							unsigned long end)
+{
+	unsigned long flags;
+#ifndef ASM_LOOP
+	int i;
+#endif
+	pr_debug("%s: start 0x%x, end 0x%x\n", __func__,
+				(unsigned int)start, (unsigned int) end);
+
+	CACHE_LOOP_LIMITS(start, end,
+			cpuinfo.dcache_line_length, cpuinfo.dcache_size);
+
+	local_irq_save(flags);
+	__disable_dcache_nomsr();
+
+#ifdef ASM_LOOP
+	CACHE_RANGE_LOOP_1(start, end, cpuinfo.dcache_line_length, wdc);
+#else
+	for (i = start; i < end; i += cpuinfo.dcache_line_length)
+		__asm__ __volatile__ ("wdc	%0, r0;"	\
+				: : "r" (i));
+#endif
+
+	__enable_dcache_nomsr();
+	local_irq_restore(flags);
+}
+
+static void __flush_dcache_all_wb(void)
+{
+#ifndef ASM_LOOP
+	int i;
+#endif
+	pr_debug("%s\n", __func__);
+#ifdef ASM_LOOP
+	CACHE_ALL_LOOP(cpuinfo.dcache_size, cpuinfo.dcache_line_length,
+				wdc.flush);
+#else
+	for (i = 0; i < cpuinfo.dcache_size;
+		 i += cpuinfo.dcache_line_length)
+			__asm__ __volatile__ ("wdc.flush	%0, r0;" \
+					: : "r" (i));
+#endif
+}
+
+static void __flush_dcache_range_wb(unsigned long start, unsigned long end)
+{
+#ifndef ASM_LOOP
+	int i;
+#endif
+	pr_debug("%s: start 0x%x, end 0x%x\n", __func__,
+				(unsigned int)start, (unsigned int) end);
+
+	CACHE_LOOP_LIMITS(start, end,
+			cpuinfo.dcache_line_length, cpuinfo.dcache_size);
+#ifdef ASM_LOOP
+	CACHE_RANGE_LOOP_2(start, end, cpuinfo.dcache_line_length, wdc.flush);
+#else
+	for (i = start; i < end; i += cpuinfo.dcache_line_length)
+		__asm__ __volatile__ ("wdc.flush	%0, r0;"	\
+				: : "r" (i));
+#endif
+}
+
+/* struct for wb caches and for wt caches */
+struct scache *mbc;
+
+/* new wb cache model */
+static const struct scache wb_msr = {
+	.ie = __enable_icache_msr,
+	.id = __disable_icache_msr,
+	.ifl = __flush_icache_all_noirq,
+	.iflr = __flush_icache_range_noirq,
+	.iin = __flush_icache_all_noirq,
+	.iinr = __flush_icache_range_noirq,
+	.de = __enable_dcache_msr,
+	.dd = __disable_dcache_msr,
+	.dfl = __flush_dcache_all_wb,
+	.dflr = __flush_dcache_range_wb,
+	.din = __invalidate_dcache_all_wb,
+	.dinr = __invalidate_dcache_range_wb,
+};
+
+/* There is only difference in ie, id, de, dd functions */
+static const struct scache wb_nomsr = {
+	.ie = __enable_icache_nomsr,
+	.id = __disable_icache_nomsr,
+	.ifl = __flush_icache_all_noirq,
+	.iflr = __flush_icache_range_noirq,
+	.iin = __flush_icache_all_noirq,
+	.iinr = __flush_icache_range_noirq,
+	.de = __enable_dcache_nomsr,
+	.dd = __disable_dcache_nomsr,
+	.dfl = __flush_dcache_all_wb,
+	.dflr = __flush_dcache_range_wb,
+	.din = __invalidate_dcache_all_wb,
+	.dinr = __invalidate_dcache_range_wb,
+};
+
+/* Old wt cache model with disabling irq and turn off cache */
+static const struct scache wt_msr = {
+	.ie = __enable_icache_msr,
+	.id = __disable_icache_msr,
+	.ifl = __flush_icache_all_msr_irq,
+	.iflr = __flush_icache_range_msr_irq,
+	.iin = __flush_icache_all_msr_irq,
+	.iinr = __flush_icache_range_msr_irq,
+	.de = __enable_dcache_msr,
+	.dd = __disable_dcache_msr,
+	.dfl = __invalidate_dcache_all_msr_irq,
+	.dflr = __invalidate_dcache_range_msr_irq_wt,
+	.din = __invalidate_dcache_all_msr_irq,
+	.dinr = __invalidate_dcache_range_msr_irq_wt,
+};
+
+static const struct scache wt_nomsr = {
+	.ie = __enable_icache_nomsr,
+	.id = __disable_icache_nomsr,
+	.ifl = __flush_icache_all_nomsr_irq,
+	.iflr = __flush_icache_range_nomsr_irq,
+	.iin = __flush_icache_all_nomsr_irq,
+	.iinr = __flush_icache_range_nomsr_irq,
+	.de = __enable_dcache_nomsr,
+	.dd = __disable_dcache_nomsr,
+	.dfl = __invalidate_dcache_all_nomsr_irq,
+	.dflr = __invalidate_dcache_range_nomsr_irq,
+	.din = __invalidate_dcache_all_nomsr_irq,
+	.dinr = __invalidate_dcache_range_nomsr_irq,
+};
+
+/* New wt cache model for newer Microblaze versions */
+static const struct scache wt_msr_noirq = {
+	.ie = __enable_icache_msr,
+	.id = __disable_icache_msr,
+	.ifl = __flush_icache_all_noirq,
+	.iflr = __flush_icache_range_noirq,
+	.iin = __flush_icache_all_noirq,
+	.iinr = __flush_icache_range_noirq,
+	.de = __enable_dcache_msr,
+	.dd = __disable_dcache_msr,
+	.dfl = __invalidate_dcache_all_noirq_wt,
+	.dflr = __invalidate_dcache_range_nomsr_wt,
+	.din = __invalidate_dcache_all_noirq_wt,
+	.dinr = __invalidate_dcache_range_nomsr_wt,
+};
+
+static const struct scache wt_nomsr_noirq = {
+	.ie = __enable_icache_nomsr,
+	.id = __disable_icache_nomsr,
+	.ifl = __flush_icache_all_noirq,
+	.iflr = __flush_icache_range_noirq,
+	.iin = __flush_icache_all_noirq,
+	.iinr = __flush_icache_range_noirq,
+	.de = __enable_dcache_nomsr,
+	.dd = __disable_dcache_nomsr,
+	.dfl = __invalidate_dcache_all_noirq_wt,
+	.dflr = __invalidate_dcache_range_nomsr_wt,
+	.din = __invalidate_dcache_all_noirq_wt,
+	.dinr = __invalidate_dcache_range_nomsr_wt,
+};
+
+/* CPU version code for 7.20.c - see arch/microblaze/kernel/cpu/cpuinfo.c */
+#define CPUVER_7_20_A	0x0c
+#define CPUVER_7_20_D	0x0f
+
+void microblaze_cache_init(void)
+{
+	if (cpuinfo.use_instr & PVR2_USE_MSR_INSTR) {
+		if (cpuinfo.dcache_wb) {
+			pr_info("wb_msr\n");
+			mbc = (struct scache *)&wb_msr;
+			if (cpuinfo.ver_code <= CPUVER_7_20_D) {
+				/* MS: problem with signal handling - hw bug */
+				pr_info("WB won't work properly\n");
+			}
+		} else {
+			if (cpuinfo.ver_code >= CPUVER_7_20_A) {
+				pr_info("wt_msr_noirq\n");
+				mbc = (struct scache *)&wt_msr_noirq;
+			} else {
+				pr_info("wt_msr\n");
+				mbc = (struct scache *)&wt_msr;
+			}
+		}
+	} else {
+		if (cpuinfo.dcache_wb) {
+			pr_info("wb_nomsr\n");
+			mbc = (struct scache *)&wb_nomsr;
+			if (cpuinfo.ver_code <= CPUVER_7_20_D) {
+				/* MS: problem with signal handling - hw bug */
+				pr_info("WB won't work properly\n");
+			}
+		} else {
+			if (cpuinfo.ver_code >= CPUVER_7_20_A) {
+				pr_info("wt_nomsr_noirq\n");
+				mbc = (struct scache *)&wt_nomsr_noirq;
+			} else {
+				pr_info("wt_nomsr\n");
+				mbc = (struct scache *)&wt_nomsr;
+			}
+		}
+	}
+	/*
+	 * FIXME Invalidation is done in U-BOOT
+	 * WT cache: Data is already written to main memory
+	 * WB cache: Discard data on noMMU which caused that kernel doesn't boot
+	 */
+	/* invalidate_dcache(); */
+	enable_dcache();
+
+	invalidate_icache();
+	enable_icache();
+}
diff --git a/arch/microblaze/kernel/cpu/cpuinfo-pvr-full.c b/arch/microblaze/kernel/cpu/cpuinfo-pvr-full.c
new file mode 100644
index 0000000..a32daec
--- /dev/null
+++ b/arch/microblaze/kernel/cpu/cpuinfo-pvr-full.c
@@ -0,0 +1,115 @@
+/*
+ * Support for MicroBlaze PVR (processor version register)
+ *
+ * Copyright (C) 2007-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007-2009 PetaLogix
+ * Copyright (C) 2007 John Williams <john.williams@petalogix.com>
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/init.h>
+#include <linux/string.h>
+#include <asm/pvr.h>
+#include <asm/cpuinfo.h>
+
+/*
+ * Helper macro to map between fields in our struct cpuinfo, and
+ * the PVR macros in pvr.h.
+ */
+
+#define CI(c, p) { ci->c = PVR_##p(pvr); }
+
+#if defined(CONFIG_EARLY_PRINTK) && defined(CONFIG_SERIAL_UARTLITE_CONSOLE)
+#define err_printk(x) \
+	early_printk("ERROR: Microblaze " x "-different for PVR and DTS\n");
+#else
+#define err_printk(x) \
+	pr_info("ERROR: Microblaze " x "-different for PVR and DTS\n");
+#endif
+
+void set_cpuinfo_pvr_full(struct cpuinfo *ci, struct device_node *cpu)
+{
+	struct pvr_s pvr;
+	u32 temp; /* for saving temp value */
+	get_pvr(&pvr);
+
+	CI(ver_code, VERSION);
+	if (!ci->ver_code) {
+		pr_err("ERROR: MB has broken PVR regs -> use DTS setting\n");
+		return;
+	}
+
+	temp = PVR_USE_BARREL(pvr) | PVR_USE_MSR_INSTR(pvr) |
+		PVR_USE_PCMP_INSTR(pvr) | PVR_USE_DIV(pvr);
+	if (ci->use_instr != temp)
+		err_printk("BARREL, MSR, PCMP or DIV");
+	ci->use_instr = temp;
+
+	temp = PVR_USE_HW_MUL(pvr) | PVR_USE_MUL64(pvr);
+	if (ci->use_mult != temp)
+		err_printk("HW_MUL");
+	ci->use_mult = temp;
+
+	temp = PVR_USE_FPU(pvr) | PVR_USE_FPU2(pvr);
+	if (ci->use_fpu != temp)
+		err_printk("HW_FPU");
+	ci->use_fpu = temp;
+
+	ci->use_exc = PVR_OPCODE_0x0_ILLEGAL(pvr) |
+			PVR_UNALIGNED_EXCEPTION(pvr) |
+			PVR_ILL_OPCODE_EXCEPTION(pvr) |
+			PVR_IOPB_BUS_EXCEPTION(pvr) |
+			PVR_DOPB_BUS_EXCEPTION(pvr) |
+			PVR_DIV_ZERO_EXCEPTION(pvr) |
+			PVR_FPU_EXCEPTION(pvr) |
+			PVR_FSL_EXCEPTION(pvr);
+
+	CI(pvr_user1, USER1);
+	CI(pvr_user2, USER2);
+
+	CI(mmu, USE_MMU);
+	CI(mmu_privins, MMU_PRIVINS);
+	CI(endian, ENDIAN);
+
+	CI(use_icache, USE_ICACHE);
+	CI(icache_tagbits, ICACHE_ADDR_TAG_BITS);
+	CI(icache_write, ICACHE_ALLOW_WR);
+	ci->icache_line_length = PVR_ICACHE_LINE_LEN(pvr) << 2;
+	CI(icache_size, ICACHE_BYTE_SIZE);
+	CI(icache_base, ICACHE_BASEADDR);
+	CI(icache_high, ICACHE_HIGHADDR);
+
+	CI(use_dcache, USE_DCACHE);
+	CI(dcache_tagbits, DCACHE_ADDR_TAG_BITS);
+	CI(dcache_write, DCACHE_ALLOW_WR);
+	ci->dcache_line_length = PVR_DCACHE_LINE_LEN(pvr) << 2;
+	CI(dcache_size, DCACHE_BYTE_SIZE);
+	CI(dcache_base, DCACHE_BASEADDR);
+	CI(dcache_high, DCACHE_HIGHADDR);
+
+	temp = PVR_DCACHE_USE_WRITEBACK(pvr);
+	if (ci->dcache_wb != temp)
+		err_printk("DCACHE WB");
+	ci->dcache_wb = temp;
+
+	CI(use_dopb, D_OPB);
+	CI(use_iopb, I_OPB);
+	CI(use_dlmb, D_LMB);
+	CI(use_ilmb, I_LMB);
+	CI(num_fsl, FSL_LINKS);
+
+	CI(irq_edge, INTERRUPT_IS_EDGE);
+	CI(irq_positive, EDGE_IS_POSITIVE);
+
+	CI(area_optimised, AREA_OPTIMISED);
+
+	CI(hw_debug, DEBUG_ENABLED);
+	CI(num_pc_brk, NUMBER_OF_PC_BRK);
+	CI(num_rd_brk, NUMBER_OF_RD_ADDR_BRK);
+	CI(num_wr_brk, NUMBER_OF_WR_ADDR_BRK);
+
+	CI(fpga_family_code, TARGET_FAMILY);
+}
diff --git a/arch/microblaze/kernel/cpu/cpuinfo-static.c b/arch/microblaze/kernel/cpu/cpuinfo-static.c
new file mode 100644
index 0000000..85dbda4
--- /dev/null
+++ b/arch/microblaze/kernel/cpu/cpuinfo-static.c
@@ -0,0 +1,145 @@
+/*
+ * Copyright (C) 2007-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007-2009 PetaLogix
+ * Copyright (C) 2007 John Williams <john.williams@petalogix.com>
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/string.h>
+#include <asm/cpuinfo.h>
+#include <asm/pvr.h>
+
+static const char family_string[] = CONFIG_XILINX_MICROBLAZE0_FAMILY;
+static const char cpu_ver_string[] = CONFIG_XILINX_MICROBLAZE0_HW_VER;
+
+#define err_printk(x) \
+	early_printk("ERROR: Microblaze " x "-different for kernel and DTS\n");
+
+void __init set_cpuinfo_static(struct cpuinfo *ci, struct device_node *cpu)
+{
+	u32 i = 0;
+
+	ci->use_instr =
+		(fcpu(cpu, "xlnx,use-barrel") ? PVR0_USE_BARREL_MASK : 0) |
+		(fcpu(cpu, "xlnx,use-msr-instr") ? PVR2_USE_MSR_INSTR : 0) |
+		(fcpu(cpu, "xlnx,use-pcmp-instr") ? PVR2_USE_PCMP_INSTR : 0) |
+		(fcpu(cpu, "xlnx,use-div") ? PVR0_USE_DIV_MASK : 0);
+	if (CONFIG_XILINX_MICROBLAZE0_USE_BARREL)
+		i |= PVR0_USE_BARREL_MASK;
+	if (CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR)
+		i |= PVR2_USE_MSR_INSTR;
+	if (CONFIG_XILINX_MICROBLAZE0_USE_PCMP_INSTR)
+		i |= PVR2_USE_PCMP_INSTR;
+	if (CONFIG_XILINX_MICROBLAZE0_USE_DIV)
+		i |= PVR0_USE_DIV_MASK;
+	if (ci->use_instr != i)
+		err_printk("BARREL, MSR, PCMP or DIV");
+
+	ci->use_mult = fcpu(cpu, "xlnx,use-hw-mul");
+	if (ci->use_mult != CONFIG_XILINX_MICROBLAZE0_USE_HW_MUL)
+		err_printk("HW_MUL");
+	ci->use_mult =
+		(ci->use_mult > 1 ?
+				(PVR2_USE_MUL64_MASK | PVR0_USE_HW_MUL_MASK) :
+				(ci->use_mult == 1 ? PVR0_USE_HW_MUL_MASK : 0));
+
+	ci->use_fpu = fcpu(cpu, "xlnx,use-fpu");
+	if (ci->use_fpu != CONFIG_XILINX_MICROBLAZE0_USE_FPU)
+		err_printk("HW_FPU");
+	ci->use_fpu = (ci->use_fpu > 1 ?
+				(PVR2_USE_FPU2_MASK | PVR0_USE_FPU_MASK) :
+				(ci->use_fpu == 1 ? PVR0_USE_FPU_MASK : 0));
+
+	ci->use_exc =
+		(fcpu(cpu, "xlnx,unaligned-exceptions") ?
+				PVR2_UNALIGNED_EXC_MASK : 0) |
+		(fcpu(cpu, "xlnx,ill-opcode-exception") ?
+				PVR2_ILL_OPCODE_EXC_MASK : 0) |
+		(fcpu(cpu, "xlnx,iopb-bus-exception") ?
+				PVR2_IOPB_BUS_EXC_MASK : 0) |
+		(fcpu(cpu, "xlnx,dopb-bus-exception") ?
+				PVR2_DOPB_BUS_EXC_MASK : 0) |
+		(fcpu(cpu, "xlnx,div-zero-exception") ?
+				PVR2_DIV_ZERO_EXC_MASK : 0) |
+		(fcpu(cpu, "xlnx,fpu-exception") ? PVR2_FPU_EXC_MASK : 0) |
+		(fcpu(cpu, "xlnx,fsl-exception") ? PVR2_USE_EXTEND_FSL : 0);
+
+	ci->use_icache = fcpu(cpu, "xlnx,use-icache");
+	ci->icache_tagbits = fcpu(cpu, "xlnx,addr-tag-bits");
+	ci->icache_write = fcpu(cpu, "xlnx,allow-icache-wr");
+	ci->icache_line_length = fcpu(cpu, "xlnx,icache-line-len") << 2;
+	if (!ci->icache_line_length) {
+		if (fcpu(cpu, "xlnx,icache-use-fsl"))
+			ci->icache_line_length = 4 << 2;
+		else
+			ci->icache_line_length = 1 << 2;
+	}
+	ci->icache_size = fcpu(cpu, "i-cache-size");
+	ci->icache_base = fcpu(cpu, "i-cache-baseaddr");
+	ci->icache_high = fcpu(cpu, "i-cache-highaddr");
+
+	ci->use_dcache = fcpu(cpu, "xlnx,use-dcache");
+	ci->dcache_tagbits = fcpu(cpu, "xlnx,dcache-addr-tag");
+	ci->dcache_write = fcpu(cpu, "xlnx,allow-dcache-wr");
+	ci->dcache_line_length = fcpu(cpu, "xlnx,dcache-line-len") << 2;
+	if (!ci->dcache_line_length) {
+		if (fcpu(cpu, "xlnx,dcache-use-fsl"))
+			ci->dcache_line_length = 4 << 2;
+		else
+			ci->dcache_line_length = 1 << 2;
+	}
+	ci->dcache_size = fcpu(cpu, "d-cache-size");
+	ci->dcache_base = fcpu(cpu, "d-cache-baseaddr");
+	ci->dcache_high = fcpu(cpu, "d-cache-highaddr");
+	ci->dcache_wb = fcpu(cpu, "xlnx,dcache-use-writeback");
+
+	ci->use_dopb = fcpu(cpu, "xlnx,d-opb");
+	ci->use_iopb = fcpu(cpu, "xlnx,i-opb");
+	ci->use_dlmb = fcpu(cpu, "xlnx,d-lmb");
+	ci->use_ilmb = fcpu(cpu, "xlnx,i-lmb");
+
+	ci->num_fsl = fcpu(cpu, "xlnx,fsl-links");
+	ci->irq_edge = fcpu(cpu, "xlnx,interrupt-is-edge");
+	ci->irq_positive = fcpu(cpu, "xlnx,edge-is-positive");
+	ci->area_optimised = 0;
+
+	ci->hw_debug = fcpu(cpu, "xlnx,debug-enabled");
+	ci->num_pc_brk = fcpu(cpu, "xlnx,number-of-pc-brk");
+	ci->num_rd_brk = fcpu(cpu, "xlnx,number-of-rd-addr-brk");
+	ci->num_wr_brk = fcpu(cpu, "xlnx,number-of-wr-addr-brk");
+
+	ci->pvr_user1 = fcpu(cpu, "xlnx,pvr-user1");
+	ci->pvr_user2 = fcpu(cpu, "xlnx,pvr-user2");
+
+	ci->mmu = fcpu(cpu, "xlnx,use-mmu");
+	ci->mmu_privins = fcpu(cpu, "xlnx,mmu-privileged-instr");
+	ci->endian = fcpu(cpu, "xlnx,endianness");
+
+	ci->ver_code = 0;
+	ci->fpga_family_code = 0;
+
+	/* Do various fixups based on CPU version and FPGA family strings */
+
+	/* Resolved the CPU version code */
+	for (i = 0; cpu_ver_lookup[i].s != NULL; i++) {
+		if (strcmp(cpu_ver_lookup[i].s, cpu_ver_string) == 0)
+			ci->ver_code = cpu_ver_lookup[i].k;
+	}
+
+	/* Resolved the fpga family code */
+	for (i = 0; family_string_lookup[i].s != NULL; i++) {
+		if (strcmp(family_string_lookup[i].s, family_string) == 0)
+			ci->fpga_family_code = family_string_lookup[i].k;
+	}
+
+	/* FIXME - mb3 and spartan2 do not exist in PVR */
+	/* This is mb3 and on a non Spartan2 */
+	if (ci->ver_code == 0x20 && ci->fpga_family_code != 0xf0)
+		/* Hardware Multiplier in use */
+		ci->use_mult = 1;
+}
diff --git a/arch/microblaze/kernel/cpu/cpuinfo.c b/arch/microblaze/kernel/cpu/cpuinfo.c
new file mode 100644
index 0000000..b70bb53
--- /dev/null
+++ b/arch/microblaze/kernel/cpu/cpuinfo.c
@@ -0,0 +1,133 @@
+/*
+ * Copyright (C) 2007-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007-2009 PetaLogix
+ * Copyright (C) 2007 John Williams <john.williams@petalogix.com>
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/clk.h>
+#include <linux/init.h>
+#include <asm/cpuinfo.h>
+#include <asm/pvr.h>
+
+const struct cpu_ver_key cpu_ver_lookup[] = {
+	/* These key value are as per MBV field in PVR0 */
+	{"5.00.a", 0x01},
+	{"5.00.b", 0x02},
+	{"5.00.c", 0x03},
+	{"6.00.a", 0x04},
+	{"6.00.b", 0x06},
+	{"7.00.a", 0x05},
+	{"7.00.b", 0x07},
+	{"7.10.a", 0x08},
+	{"7.10.b", 0x09},
+	{"7.10.c", 0x0a},
+	{"7.10.d", 0x0b},
+	{"7.20.a", 0x0c},
+	{"7.20.b", 0x0d},
+	{"7.20.c", 0x0e},
+	{"7.20.d", 0x0f},
+	{"7.30.a", 0x10},
+	{"7.30.b", 0x11},
+	{"8.00.a", 0x12},
+	{"8.00.b", 0x13},
+	{"8.10.a", 0x14},
+	{"8.20.a", 0x15},
+	{"8.20.b", 0x16},
+	{"8.30.a", 0x17},
+	{"8.40.a", 0x18},
+	{"8.40.b", 0x19},
+	{"8.50.a", 0x1a},
+	{"8.50.b", 0x1c},
+	{"8.50.c", 0x1e},
+	{"9.0", 0x1b},
+	{"9.1", 0x1d},
+	{"9.2", 0x1f},
+	{"9.3", 0x20},
+	{"9.4", 0x21},
+	{"9.5", 0x22},
+	{NULL, 0},
+};
+
+/*
+ * FIXME Not sure if the actual key is defined by Xilinx in the PVR
+ */
+const struct family_string_key family_string_lookup[] = {
+	{"virtex2", 0x4},
+	{"virtex2pro", 0x5},
+	{"spartan3", 0x6},
+	{"virtex4", 0x7},
+	{"virtex5", 0x8},
+	{"spartan3e", 0x9},
+	{"spartan3a", 0xa},
+	{"spartan3an", 0xb},
+	{"spartan3adsp", 0xc},
+	{"spartan6", 0xd},
+	{"virtex6", 0xe},
+	{"virtex7", 0xf},
+	/* FIXME There is no key code defined for spartan2 */
+	{"spartan2", 0xf0},
+	{"kintex7", 0x10},
+	{"artix7", 0x11},
+	{"zynq7000", 0x12},
+	{"UltraScale Virtex", 0x13},
+	{"UltraScale Kintex", 0x14},
+	{NULL, 0},
+};
+
+struct cpuinfo cpuinfo;
+static struct device_node *cpu;
+
+void __init setup_cpuinfo(void)
+{
+	cpu = (struct device_node *) of_find_node_by_type(NULL, "cpu");
+	if (!cpu)
+		pr_err("You don't have cpu!!!\n");
+
+	pr_info("%s: initialising\n", __func__);
+
+	switch (cpu_has_pvr()) {
+	case 0:
+		pr_warn("%s: No PVR support. Using static CPU info from FDT\n",
+			__func__);
+		set_cpuinfo_static(&cpuinfo, cpu);
+		break;
+/* FIXME I found weird behavior with MB 7.00.a/b 7.10.a
+ * please do not use FULL PVR with MMU */
+	case 1:
+		pr_info("%s: Using full CPU PVR support\n",
+			__func__);
+		set_cpuinfo_static(&cpuinfo, cpu);
+		set_cpuinfo_pvr_full(&cpuinfo, cpu);
+		break;
+	default:
+		pr_warn("%s: Unsupported PVR setting\n", __func__);
+		set_cpuinfo_static(&cpuinfo, cpu);
+	}
+
+	if (cpuinfo.mmu_privins)
+		pr_warn("%s: Stream instructions enabled"
+			" - USERSPACE CAN LOCK THIS KERNEL!\n", __func__);
+}
+
+void __init setup_cpuinfo_clk(void)
+{
+	struct clk *clk;
+
+	clk = of_clk_get(cpu, 0);
+	if (IS_ERR(clk)) {
+		pr_err("ERROR: CPU CCF input clock not found\n");
+		/* take timebase-frequency from DTS */
+		cpuinfo.cpu_clock_freq = fcpu(cpu, "timebase-frequency");
+	} else {
+		cpuinfo.cpu_clock_freq = clk_get_rate(clk);
+	}
+
+	if (!cpuinfo.cpu_clock_freq) {
+		pr_err("ERROR: CPU clock frequency not setup\n");
+		BUG();
+	}
+}
diff --git a/arch/microblaze/kernel/cpu/mb.c b/arch/microblaze/kernel/cpu/mb.c
new file mode 100644
index 0000000..9581d194
--- /dev/null
+++ b/arch/microblaze/kernel/cpu/mb.c
@@ -0,0 +1,158 @@
+/*
+ * CPU-version specific code
+ *
+ * Copyright (C) 2007-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2006-2009 PetaLogix
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/init.h>
+#include <linux/string.h>
+#include <linux/seq_file.h>
+#include <linux/cpu.h>
+#include <linux/initrd.h>
+
+#include <linux/bug.h>
+#include <asm/cpuinfo.h>
+#include <linux/delay.h>
+#include <linux/io.h>
+#include <asm/page.h>
+#include <linux/param.h>
+#include <asm/pvr.h>
+#include <asm/sections.h>
+#include <asm/setup.h>
+
+static int show_cpuinfo(struct seq_file *m, void *v)
+{
+	char *fpga_family = "Unknown";
+	char *cpu_ver = "Unknown";
+	int i;
+
+	/* Denormalised to get the fpga family string */
+	for (i = 0; family_string_lookup[i].s != NULL; i++) {
+		if (cpuinfo.fpga_family_code == family_string_lookup[i].k) {
+			fpga_family = (char *)family_string_lookup[i].s;
+			break;
+		}
+	}
+
+	/* Denormalised to get the hw version string */
+	for (i = 0; cpu_ver_lookup[i].s != NULL; i++) {
+		if (cpuinfo.ver_code == cpu_ver_lookup[i].k) {
+			cpu_ver = (char *)cpu_ver_lookup[i].s;
+			break;
+		}
+	}
+
+	seq_printf(m,
+		   "CPU-Family:	MicroBlaze\n"
+		   "FPGA-Arch:	%s\n"
+		   "CPU-Ver:	%s, %s endian\n"
+		   "CPU-MHz:	%d.%02d\n"
+		   "BogoMips:	%lu.%02lu\n",
+		   fpga_family,
+		   cpu_ver,
+		   cpuinfo.endian ? "little" : "big",
+		   cpuinfo.cpu_clock_freq / 1000000,
+		   cpuinfo.cpu_clock_freq % 1000000,
+		   loops_per_jiffy / (500000 / HZ),
+		   (loops_per_jiffy / (5000 / HZ)) % 100);
+
+	seq_printf(m,
+		   "HW:\n Shift:\t\t%s\n"
+		   " MSR:\t\t%s\n"
+		   " PCMP:\t\t%s\n"
+		   " DIV:\t\t%s\n",
+		   (cpuinfo.use_instr & PVR0_USE_BARREL_MASK) ? "yes" : "no",
+		   (cpuinfo.use_instr & PVR2_USE_MSR_INSTR) ? "yes" : "no",
+		   (cpuinfo.use_instr & PVR2_USE_PCMP_INSTR) ? "yes" : "no",
+		   (cpuinfo.use_instr & PVR0_USE_DIV_MASK) ? "yes" : "no");
+
+	seq_printf(m, " MMU:\t\t%x\n", cpuinfo.mmu);
+
+	seq_printf(m,
+		   " MUL:\t\t%s\n"
+		   " FPU:\t\t%s\n",
+		   (cpuinfo.use_mult & PVR2_USE_MUL64_MASK) ? "v2" :
+		   (cpuinfo.use_mult & PVR0_USE_HW_MUL_MASK) ? "v1" : "no",
+		   (cpuinfo.use_fpu & PVR2_USE_FPU2_MASK) ? "v2" :
+		   (cpuinfo.use_fpu & PVR0_USE_FPU_MASK) ? "v1" : "no");
+
+	seq_printf(m,
+		   " Exc:\t\t%s%s%s%s%s%s%s%s\n",
+		   (cpuinfo.use_exc & PVR2_OPCODE_0x0_ILL_MASK) ? "op0x0 " : "",
+		   (cpuinfo.use_exc & PVR2_UNALIGNED_EXC_MASK) ? "unal " : "",
+		   (cpuinfo.use_exc & PVR2_ILL_OPCODE_EXC_MASK) ? "ill " : "",
+		   (cpuinfo.use_exc & PVR2_IOPB_BUS_EXC_MASK) ? "iopb " : "",
+		   (cpuinfo.use_exc & PVR2_DOPB_BUS_EXC_MASK) ? "dopb " : "",
+		   (cpuinfo.use_exc & PVR2_DIV_ZERO_EXC_MASK) ? "zero " : "",
+		   (cpuinfo.use_exc & PVR2_FPU_EXC_MASK) ? "fpu " : "",
+		   (cpuinfo.use_exc & PVR2_USE_FSL_EXC) ? "fsl " : "");
+
+	seq_printf(m,
+		   "Stream-insns:\t%sprivileged\n",
+		   cpuinfo.mmu_privins ? "un" : "");
+
+	if (cpuinfo.use_icache)
+		seq_printf(m,
+			   "Icache:\t\t%ukB\tline length:\t%dB\n",
+			   cpuinfo.icache_size >> 10,
+			   cpuinfo.icache_line_length);
+	else
+		seq_puts(m, "Icache:\t\tno\n");
+
+	if (cpuinfo.use_dcache) {
+		seq_printf(m,
+			   "Dcache:\t\t%ukB\tline length:\t%dB\n",
+			   cpuinfo.dcache_size >> 10,
+			   cpuinfo.dcache_line_length);
+		seq_puts(m, "Dcache-Policy:\t");
+		if (cpuinfo.dcache_wb)
+			seq_puts(m, "write-back\n");
+		else
+			seq_puts(m, "write-through\n");
+	} else {
+		seq_puts(m, "Dcache:\t\tno\n");
+	}
+
+	seq_printf(m,
+		   "HW-Debug:\t%s\n",
+		   cpuinfo.hw_debug ? "yes" : "no");
+
+	seq_printf(m,
+		   "PVR-USR1:\t%02x\n"
+		   "PVR-USR2:\t%08x\n",
+		   cpuinfo.pvr_user1,
+		   cpuinfo.pvr_user2);
+
+	seq_printf(m, "Page size:\t%lu\n", PAGE_SIZE);
+
+	return 0;
+}
+
+static void *c_start(struct seq_file *m, loff_t *pos)
+{
+	int i = *pos;
+
+	return i < NR_CPUS ? (void *) (i + 1) : NULL;
+}
+
+static void *c_next(struct seq_file *m, void *v, loff_t *pos)
+{
+	++*pos;
+	return c_start(m, pos);
+}
+
+static void c_stop(struct seq_file *m, void *v)
+{
+}
+
+const struct seq_operations cpuinfo_op = {
+	.start = c_start,
+	.next = c_next,
+	.stop = c_stop,
+	.show = show_cpuinfo,
+};
diff --git a/arch/microblaze/kernel/cpu/pvr.c b/arch/microblaze/kernel/cpu/pvr.c
new file mode 100644
index 0000000..8d0dc6d
--- /dev/null
+++ b/arch/microblaze/kernel/cpu/pvr.c
@@ -0,0 +1,80 @@
+/*
+ * Support for MicroBlaze PVR (processor version register)
+ *
+ * Copyright (C) 2007-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007-2009 PetaLogix
+ * Copyright (C) 2007 John Williams <john.williams@petalogix.com>
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/kernel.h>
+#include <linux/compiler.h>
+#include <asm/exceptions.h>
+#include <asm/pvr.h>
+
+/*
+ * Until we get an assembler that knows about the pvr registers,
+ * this horrible cruft will have to do.
+ * That hardcoded opcode is mfs r3, rpvrNN
+ */
+
+#define get_single_pvr(pvrid, val)				\
+{								\
+	register unsigned tmp __asm__("r3");			\
+	tmp = 0x0;	/* Prevent warning about unused */	\
+	__asm__ __volatile__ (					\
+			"mfs	%0, rpvr" #pvrid ";"		\
+			: "=r" (tmp) : : "memory");		\
+	val = tmp;						\
+}
+
+/*
+ * Does the CPU support the PVR register?
+ * return value:
+ * 0: no PVR
+ * 1: simple PVR
+ * 2: full PVR
+ *
+ * This must work on all CPU versions, including those before the
+ * PVR was even an option.
+ */
+
+int cpu_has_pvr(void)
+{
+	unsigned long flags;
+	unsigned pvr0;
+
+	local_save_flags(flags);
+
+	/* PVR bit in MSR tells us if there is any support */
+	if (!(flags & PVR_MSR_BIT))
+		return 0;
+
+	get_single_pvr(0, pvr0);
+	pr_debug("%s: pvr0 is 0x%08x\n", __func__, pvr0);
+
+	if (pvr0 & PVR0_PVR_FULL_MASK)
+		return 1;
+
+	/* for partial PVR use static cpuinfo */
+	return 2;
+}
+
+void get_pvr(struct pvr_s *p)
+{
+	get_single_pvr(0, p->pvr[0]);
+	get_single_pvr(1, p->pvr[1]);
+	get_single_pvr(2, p->pvr[2]);
+	get_single_pvr(3, p->pvr[3]);
+	get_single_pvr(4, p->pvr[4]);
+	get_single_pvr(5, p->pvr[5]);
+	get_single_pvr(6, p->pvr[6]);
+	get_single_pvr(7, p->pvr[7]);
+	get_single_pvr(8, p->pvr[8]);
+	get_single_pvr(9, p->pvr[9]);
+	get_single_pvr(10, p->pvr[10]);
+	get_single_pvr(11, p->pvr[11]);
+}
diff --git a/arch/microblaze/kernel/dma.c b/arch/microblaze/kernel/dma.c
new file mode 100644
index 0000000..bf4dec2
--- /dev/null
+++ b/arch/microblaze/kernel/dma.c
@@ -0,0 +1,208 @@
+/*
+ * Copyright (C) 2009-2010 PetaLogix
+ * Copyright (C) 2006 Benjamin Herrenschmidt, IBM Corporation
+ *
+ * Provide default implementations of the DMA mapping callbacks for
+ * directly mapped busses.
+ */
+
+#include <linux/device.h>
+#include <linux/dma-mapping.h>
+#include <linux/gfp.h>
+#include <linux/dma-debug.h>
+#include <linux/export.h>
+#include <linux/bug.h>
+
+#define NOT_COHERENT_CACHE
+
+static void *dma_direct_alloc_coherent(struct device *dev, size_t size,
+				       dma_addr_t *dma_handle, gfp_t flag,
+				       struct dma_attrs *attrs)
+{
+#ifdef NOT_COHERENT_CACHE
+	return consistent_alloc(flag, size, dma_handle);
+#else
+	void *ret;
+	struct page *page;
+	int node = dev_to_node(dev);
+
+	/* ignore region specifiers */
+	flag  &= ~(__GFP_HIGHMEM);
+
+	page = alloc_pages_node(node, flag, get_order(size));
+	if (page == NULL)
+		return NULL;
+	ret = page_address(page);
+	memset(ret, 0, size);
+	*dma_handle = virt_to_phys(ret);
+
+	return ret;
+#endif
+}
+
+static void dma_direct_free_coherent(struct device *dev, size_t size,
+				     void *vaddr, dma_addr_t dma_handle,
+				     struct dma_attrs *attrs)
+{
+#ifdef NOT_COHERENT_CACHE
+	consistent_free(size, vaddr);
+#else
+	free_pages((unsigned long)vaddr, get_order(size));
+#endif
+}
+
+static int dma_direct_map_sg(struct device *dev, struct scatterlist *sgl,
+			     int nents, enum dma_data_direction direction,
+			     struct dma_attrs *attrs)
+{
+	struct scatterlist *sg;
+	int i;
+
+	/* FIXME this part of code is untested */
+	for_each_sg(sgl, sg, nents, i) {
+		sg->dma_address = sg_phys(sg);
+		__dma_sync(page_to_phys(sg_page(sg)) + sg->offset,
+							sg->length, direction);
+	}
+
+	return nents;
+}
+
+static int dma_direct_dma_supported(struct device *dev, u64 mask)
+{
+	return 1;
+}
+
+static inline dma_addr_t dma_direct_map_page(struct device *dev,
+					     struct page *page,
+					     unsigned long offset,
+					     size_t size,
+					     enum dma_data_direction direction,
+					     struct dma_attrs *attrs)
+{
+	__dma_sync(page_to_phys(page) + offset, size, direction);
+	return page_to_phys(page) + offset;
+}
+
+static inline void dma_direct_unmap_page(struct device *dev,
+					 dma_addr_t dma_address,
+					 size_t size,
+					 enum dma_data_direction direction,
+					 struct dma_attrs *attrs)
+{
+/* There is not necessary to do cache cleanup
+ *
+ * phys_to_virt is here because in __dma_sync_page is __virt_to_phys and
+ * dma_address is physical address
+ */
+	__dma_sync(dma_address, size, direction);
+}
+
+static inline void
+dma_direct_sync_single_for_cpu(struct device *dev,
+			       dma_addr_t dma_handle, size_t size,
+			       enum dma_data_direction direction)
+{
+	/*
+	 * It's pointless to flush the cache as the memory segment
+	 * is given to the CPU
+	 */
+
+	if (direction == DMA_FROM_DEVICE)
+		__dma_sync(dma_handle, size, direction);
+}
+
+static inline void
+dma_direct_sync_single_for_device(struct device *dev,
+				  dma_addr_t dma_handle, size_t size,
+				  enum dma_data_direction direction)
+{
+	/*
+	 * It's pointless to invalidate the cache if the device isn't
+	 * supposed to write to the relevant region
+	 */
+
+	if (direction == DMA_TO_DEVICE)
+		__dma_sync(dma_handle, size, direction);
+}
+
+static inline void
+dma_direct_sync_sg_for_cpu(struct device *dev,
+			   struct scatterlist *sgl, int nents,
+			   enum dma_data_direction direction)
+{
+	struct scatterlist *sg;
+	int i;
+
+	/* FIXME this part of code is untested */
+	if (direction == DMA_FROM_DEVICE)
+		for_each_sg(sgl, sg, nents, i)
+			__dma_sync(sg->dma_address, sg->length, direction);
+}
+
+static inline void
+dma_direct_sync_sg_for_device(struct device *dev,
+			      struct scatterlist *sgl, int nents,
+			      enum dma_data_direction direction)
+{
+	struct scatterlist *sg;
+	int i;
+
+	/* FIXME this part of code is untested */
+	if (direction == DMA_TO_DEVICE)
+		for_each_sg(sgl, sg, nents, i)
+			__dma_sync(sg->dma_address, sg->length, direction);
+}
+
+static
+int dma_direct_mmap_coherent(struct device *dev, struct vm_area_struct *vma,
+			     void *cpu_addr, dma_addr_t handle, size_t size,
+			     struct dma_attrs *attrs)
+{
+#ifdef CONFIG_MMU
+	unsigned long user_count = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT;
+	unsigned long count = PAGE_ALIGN(size) >> PAGE_SHIFT;
+	unsigned long off = vma->vm_pgoff;
+	unsigned long pfn;
+
+	if (off >= count || user_count > (count - off))
+		return -ENXIO;
+
+#ifdef NOT_COHERENT_CACHE
+	vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
+	pfn = consistent_virt_to_pfn(cpu_addr);
+#else
+	pfn = virt_to_pfn(cpu_addr);
+#endif
+	return remap_pfn_range(vma, vma->vm_start, pfn + off,
+			       vma->vm_end - vma->vm_start, vma->vm_page_prot);
+#else
+	return -ENXIO;
+#endif
+}
+
+struct dma_map_ops dma_direct_ops = {
+	.alloc		= dma_direct_alloc_coherent,
+	.free		= dma_direct_free_coherent,
+	.mmap		= dma_direct_mmap_coherent,
+	.map_sg		= dma_direct_map_sg,
+	.dma_supported	= dma_direct_dma_supported,
+	.map_page	= dma_direct_map_page,
+	.unmap_page	= dma_direct_unmap_page,
+	.sync_single_for_cpu		= dma_direct_sync_single_for_cpu,
+	.sync_single_for_device		= dma_direct_sync_single_for_device,
+	.sync_sg_for_cpu		= dma_direct_sync_sg_for_cpu,
+	.sync_sg_for_device		= dma_direct_sync_sg_for_device,
+};
+EXPORT_SYMBOL(dma_direct_ops);
+
+/* Number of entries preallocated for DMA-API debugging */
+#define PREALLOC_DMA_DEBUG_ENTRIES (1 << 16)
+
+static int __init dma_init(void)
+{
+	dma_debug_init(PREALLOC_DMA_DEBUG_ENTRIES);
+
+	return 0;
+}
+fs_initcall(dma_init);
diff --git a/arch/microblaze/kernel/early_printk.c b/arch/microblaze/kernel/early_printk.c
new file mode 100644
index 0000000..365f2d5
--- /dev/null
+++ b/arch/microblaze/kernel/early_printk.c
@@ -0,0 +1,184 @@
+/*
+ * Early printk support for Microblaze.
+ *
+ * Copyright (C) 2007-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007-2009 PetaLogix
+ * Copyright (C) 2003-2006 Yasushi SHOJI <yashi@atmark-techno.com>
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/console.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/string.h>
+#include <linux/tty.h>
+#include <linux/io.h>
+#include <asm/processor.h>
+#include <linux/fcntl.h>
+#include <asm/setup.h>
+#include <asm/prom.h>
+
+static u32 base_addr;
+
+#ifdef CONFIG_SERIAL_UARTLITE_CONSOLE
+static void early_printk_uartlite_putc(char c)
+{
+	/*
+	 * Limit how many times we'll spin waiting for TX FIFO status.
+	 * This will prevent lockups if the base address is incorrectly
+	 * set, or any other issue on the UARTLITE.
+	 * This limit is pretty arbitrary, unless we are at about 10 baud
+	 * we'll never timeout on a working UART.
+	 */
+
+	unsigned retries = 1000000;
+	/* read status bit - 0x8 offset */
+	while (--retries && (in_be32(base_addr + 8) & (1 << 3)))
+		;
+
+	/* Only attempt the iowrite if we didn't timeout */
+	/* write to TX_FIFO - 0x4 offset */
+	if (retries)
+		out_be32(base_addr + 4, c & 0xff);
+}
+
+static void early_printk_uartlite_write(struct console *unused,
+					const char *s, unsigned n)
+{
+	while (*s && n-- > 0) {
+		if (*s == '\n')
+			early_printk_uartlite_putc('\r');
+		early_printk_uartlite_putc(*s);
+		s++;
+	}
+}
+
+static struct console early_serial_uartlite_console = {
+	.name = "earlyser",
+	.write = early_printk_uartlite_write,
+	.flags = CON_PRINTBUFFER | CON_BOOT,
+	.index = -1,
+};
+#endif /* CONFIG_SERIAL_UARTLITE_CONSOLE */
+
+#ifdef CONFIG_SERIAL_8250_CONSOLE
+static void early_printk_uart16550_putc(char c)
+{
+	/*
+	 * Limit how many times we'll spin waiting for TX FIFO status.
+	 * This will prevent lockups if the base address is incorrectly
+	 * set, or any other issue on the UARTLITE.
+	 * This limit is pretty arbitrary, unless we are at about 10 baud
+	 * we'll never timeout on a working UART.
+	 */
+
+	#define UART_LSR_TEMT	0x40 /* Transmitter empty */
+	#define UART_LSR_THRE	0x20 /* Transmit-hold-register empty */
+	#define BOTH_EMPTY (UART_LSR_TEMT | UART_LSR_THRE)
+
+	unsigned retries = 10000;
+
+	while (--retries &&
+		!((in_be32(base_addr + 0x14) & BOTH_EMPTY) == BOTH_EMPTY))
+		;
+
+	if (retries)
+		out_be32(base_addr, c & 0xff);
+}
+
+static void early_printk_uart16550_write(struct console *unused,
+					const char *s, unsigned n)
+{
+	while (*s && n-- > 0) {
+		if (*s == '\n')
+			early_printk_uart16550_putc('\r');
+		early_printk_uart16550_putc(*s);
+		s++;
+	}
+}
+
+static struct console early_serial_uart16550_console = {
+	.name = "earlyser",
+	.write = early_printk_uart16550_write,
+	.flags = CON_PRINTBUFFER | CON_BOOT,
+	.index = -1,
+};
+#endif /* CONFIG_SERIAL_8250_CONSOLE */
+
+int __init setup_early_printk(char *opt)
+{
+	int version = 0;
+
+	if (early_console)
+		return 1;
+
+	base_addr = of_early_console(&version);
+	if (base_addr) {
+#ifdef CONFIG_MMU
+		early_console_reg_tlb_alloc(base_addr);
+#endif
+		switch (version) {
+#ifdef CONFIG_SERIAL_UARTLITE_CONSOLE
+		case UARTLITE:
+			pr_info("Early console on uartlite at 0x%08x\n",
+								base_addr);
+			early_console = &early_serial_uartlite_console;
+			break;
+#endif
+#ifdef CONFIG_SERIAL_8250_CONSOLE
+		case UART16550:
+			pr_info("Early console on uart16650 at 0x%08x\n",
+								base_addr);
+			early_console = &early_serial_uart16550_console;
+			break;
+#endif
+		default:
+			pr_info("Unsupported early console %d\n",
+								version);
+			return 1;
+		}
+
+		register_console(early_console);
+		return 0;
+	}
+	return 1;
+}
+
+/* Remap early console to virtual address and do not allocate one TLB
+ * only for early console because of performance degression */
+void __init remap_early_printk(void)
+{
+	if (!early_console)
+		return;
+	pr_info("early_printk_console remapping from 0x%x to ", base_addr);
+	base_addr = (u32) ioremap(base_addr, PAGE_SIZE);
+	pr_cont("0x%x\n", base_addr);
+
+#ifdef CONFIG_MMU
+	/*
+	 * Early console is on the top of skipped TLB entries
+	 * decrease tlb_skip size ensure that hardcoded TLB entry will be
+	 * used by generic algorithm
+	 * FIXME check if early console mapping is on the top by rereading
+	 * TLB entry and compare baseaddr
+	 *  mts  rtlbx, (tlb_skip - 1)
+	 *  nop
+	 *  mfs  rX, rtlblo
+	 *  nop
+	 *  cmp rX, orig_base_addr
+	 */
+	tlb_skip -= 1;
+#endif
+}
+
+void __init disable_early_printk(void)
+{
+	if (!early_console)
+		return;
+	pr_warn("disabling early console\n");
+	unregister_console(early_console);
+	early_console = NULL;
+}
diff --git a/arch/microblaze/kernel/entry-nommu.S b/arch/microblaze/kernel/entry-nommu.S
new file mode 100644
index 0000000..7e394fc
--- /dev/null
+++ b/arch/microblaze/kernel/entry-nommu.S
@@ -0,0 +1,622 @@
+/*
+ * Copyright (C) 2007-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/linkage.h>
+#include <asm/thread_info.h>
+#include <linux/errno.h>
+#include <asm/entry.h>
+#include <asm/asm-offsets.h>
+#include <asm/registers.h>
+#include <asm/unistd.h>
+#include <asm/percpu.h>
+#include <asm/signal.h>
+
+#if CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR
+	.macro	disable_irq
+	msrclr r0, MSR_IE
+	.endm
+
+	.macro	enable_irq
+	msrset r0, MSR_IE
+	.endm
+
+	.macro	clear_bip
+	msrclr r0, MSR_BIP
+	.endm
+#else
+	.macro	disable_irq
+	mfs r11, rmsr
+	andi r11, r11, ~MSR_IE
+	mts rmsr, r11
+	.endm
+
+	.macro	enable_irq
+	mfs r11, rmsr
+	ori r11, r11, MSR_IE
+	mts rmsr, r11
+	.endm
+
+	.macro	clear_bip
+	mfs r11, rmsr
+	andi r11, r11, ~MSR_BIP
+	mts rmsr, r11
+	.endm
+#endif
+
+ENTRY(_interrupt)
+	swi	r1, r0, PER_CPU(ENTRY_SP)	/* save the current sp */
+	swi	r11, r0, PER_CPU(R11_SAVE)	/* temporarily save r11 */
+	lwi	r11, r0, PER_CPU(KM)		/* load mode indicator */
+	beqid	r11, 1f
+	nop
+	brid	2f				/* jump over */
+	addik	r1, r1, (-PT_SIZE)	/* room for pt_regs (delay slot) */
+1:						/* switch to kernel stack */
+	lwi	r1, r0, PER_CPU(CURRENT_SAVE)	/* get the saved current */
+	lwi	r1, r1, TS_THREAD_INFO		/* get the thread info */
+	/* calculate kernel stack pointer */
+	addik	r1, r1, THREAD_SIZE - PT_SIZE
+2:
+	swi	r11, r1, PT_MODE		/* store the mode */
+	lwi	r11, r0, PER_CPU(R11_SAVE)	/* reload r11 */
+	swi	r2, r1, PT_R2
+	swi	r3, r1, PT_R3
+	swi	r4, r1, PT_R4
+	swi	r5, r1, PT_R5
+	swi	r6, r1, PT_R6
+	swi	r7, r1, PT_R7
+	swi	r8, r1, PT_R8
+	swi	r9, r1, PT_R9
+	swi	r10, r1, PT_R10
+	swi	r11, r1, PT_R11
+	swi	r12, r1, PT_R12
+	swi	r13, r1, PT_R13
+	swi	r14, r1, PT_R14
+	swi	r14, r1, PT_PC
+	swi	r15, r1, PT_R15
+	swi	r16, r1, PT_R16
+	swi	r17, r1, PT_R17
+	swi	r18, r1, PT_R18
+	swi	r19, r1, PT_R19
+	swi	r20, r1, PT_R20
+	swi	r21, r1, PT_R21
+	swi	r22, r1, PT_R22
+	swi	r23, r1, PT_R23
+	swi	r24, r1, PT_R24
+	swi	r25, r1, PT_R25
+	swi	r26, r1, PT_R26
+	swi	r27, r1, PT_R27
+	swi	r28, r1, PT_R28
+	swi	r29, r1, PT_R29
+	swi	r30, r1, PT_R30
+	swi	r31, r1, PT_R31
+	/* special purpose registers */
+	mfs	r11, rmsr
+	swi	r11, r1, PT_MSR
+	mfs	r11, rear
+	swi	r11, r1, PT_EAR
+	mfs	r11, resr
+	swi	r11, r1, PT_ESR
+	mfs	r11, rfsr
+	swi	r11, r1, PT_FSR
+	/* reload original stack pointer and save it */
+	lwi	r11, r0, PER_CPU(ENTRY_SP)
+	swi	r11, r1, PT_R1
+	/* update mode indicator we are in kernel mode */
+	addik	r11, r0, 1
+	swi	r11, r0, PER_CPU(KM)
+	/* restore r31 */
+	lwi	r31, r0, PER_CPU(CURRENT_SAVE)
+	/* prepare the link register, the argument and jump */
+	addik	r15, r0, ret_from_intr - 8
+	addk	r6, r0, r15
+	braid	do_IRQ
+	add	r5, r0, r1
+
+ret_from_intr:
+	lwi	r11, r1, PT_MODE
+	bneid	r11, no_intr_resched
+
+3:
+	lwi	r6, r31, TS_THREAD_INFO	/* get thread info */
+	lwi	r19, r6, TI_FLAGS	/* get flags in thread info */
+				/* do an extra work if any bits are set */
+
+	andi	r11, r19, _TIF_NEED_RESCHED
+	beqi	r11, 1f
+	bralid	r15, schedule
+	nop
+	bri	3b
+1:	andi	r11, r19, _TIF_SIGPENDING | _TIF_NOTIFY_RESUME
+	beqid	r11, no_intr_resched
+	addk	r5, r1, r0
+	bralid	r15, do_notify_resume
+	addk	r6, r0, r0
+	bri	3b
+
+no_intr_resched:
+	/* Disable interrupts, we are now committed to the state restore */
+	disable_irq
+
+	/* save mode indicator */
+	lwi	r11, r1, PT_MODE
+	swi	r11, r0, PER_CPU(KM)
+
+	/* save r31 */
+	swi	r31, r0, PER_CPU(CURRENT_SAVE)
+restore_context:
+	/* special purpose registers */
+	lwi	r11, r1, PT_FSR
+	mts	rfsr, r11
+	lwi	r11, r1, PT_ESR
+	mts	resr, r11
+	lwi	r11, r1, PT_EAR
+	mts	rear, r11
+	lwi	r11, r1, PT_MSR
+	mts	rmsr, r11
+
+	lwi	r31, r1, PT_R31
+	lwi	r30, r1, PT_R30
+	lwi	r29, r1, PT_R29
+	lwi	r28, r1, PT_R28
+	lwi	r27, r1, PT_R27
+	lwi	r26, r1, PT_R26
+	lwi	r25, r1, PT_R25
+	lwi	r24, r1, PT_R24
+	lwi	r23, r1, PT_R23
+	lwi	r22, r1, PT_R22
+	lwi	r21, r1, PT_R21
+	lwi	r20, r1, PT_R20
+	lwi	r19, r1, PT_R19
+	lwi	r18, r1, PT_R18
+	lwi	r17, r1, PT_R17
+	lwi	r16, r1, PT_R16
+	lwi	r15, r1, PT_R15
+	lwi	r14, r1, PT_PC
+	lwi	r13, r1, PT_R13
+	lwi	r12, r1, PT_R12
+	lwi	r11, r1, PT_R11
+	lwi	r10, r1, PT_R10
+	lwi	r9, r1, PT_R9
+	lwi	r8, r1, PT_R8
+	lwi	r7, r1, PT_R7
+	lwi	r6, r1, PT_R6
+	lwi	r5, r1, PT_R5
+	lwi	r4, r1, PT_R4
+	lwi	r3, r1, PT_R3
+	lwi	r2, r1, PT_R2
+	lwi	r1, r1, PT_R1
+	rtid	r14, 0
+	nop
+
+ENTRY(_reset)
+	brai	0;
+
+ENTRY(_user_exception)
+	swi	r1, r0, PER_CPU(ENTRY_SP)	/* save the current sp */
+	swi	r11, r0, PER_CPU(R11_SAVE)	/* temporarily save r11 */
+	lwi	r11, r0, PER_CPU(KM)		/* load mode indicator */
+	beqid	r11, 1f				/* Already in kernel mode? */
+	nop
+	brid	2f				/* jump over */
+	addik	r1, r1, (-PT_SIZE)	/* Room for pt_regs (delay slot) */
+1:						/* Switch to kernel stack */
+	lwi	r1, r0, PER_CPU(CURRENT_SAVE)	/* get the saved current */
+	lwi	r1, r1, TS_THREAD_INFO		/* get the thread info */
+	/* calculate kernel stack pointer */
+	addik	r1, r1, THREAD_SIZE - PT_SIZE
+2:
+	swi	r11, r1, PT_MODE		/* store the mode */
+	lwi	r11, r0, PER_CPU(R11_SAVE)	/* reload r11 */
+	/* save them on stack */
+	swi	r2, r1, PT_R2
+	swi	r3, r1, PT_R3 /* r3: _always_ in clobber list; see unistd.h */
+	swi	r4, r1, PT_R4 /* r4: _always_ in clobber list; see unistd.h */
+	swi	r5, r1, PT_R5
+	swi	r6, r1, PT_R6
+	swi	r7, r1, PT_R7
+	swi	r8, r1, PT_R8
+	swi	r9, r1, PT_R9
+	swi	r10, r1, PT_R10
+	swi	r11, r1, PT_R11
+	/* r12: _always_ in clobber list; see unistd.h */
+	swi	r12, r1, PT_R12
+	swi	r13, r1, PT_R13
+	/* r14: _always_ in clobber list; see unistd.h */
+	swi	r14, r1, PT_R14
+	/* but we want to return to the next inst. */
+	addik	r14, r14, 0x4
+	swi	r14, r1, PT_PC		/* increment by 4 and store in pc */
+	swi	r15, r1, PT_R15
+	swi	r16, r1, PT_R16
+	swi	r17, r1, PT_R17
+	swi	r18, r1, PT_R18
+	swi	r19, r1, PT_R19
+	swi	r20, r1, PT_R20
+	swi	r21, r1, PT_R21
+	swi	r22, r1, PT_R22
+	swi	r23, r1, PT_R23
+	swi	r24, r1, PT_R24
+	swi	r25, r1, PT_R25
+	swi	r26, r1, PT_R26
+	swi	r27, r1, PT_R27
+	swi	r28, r1, PT_R28
+	swi	r29, r1, PT_R29
+	swi	r30, r1, PT_R30
+	swi	r31, r1, PT_R31
+
+	disable_irq
+	nop		/* make sure IE bit is in effect */
+	clear_bip	/* once IE is in effect it is safe to clear BIP */
+	nop
+
+	/* special purpose registers */
+	mfs	r11, rmsr
+	swi	r11, r1, PT_MSR
+	mfs	r11, rear
+	swi	r11, r1, PT_EAR
+	mfs	r11, resr
+	swi	r11, r1, PT_ESR
+	mfs	r11, rfsr
+	swi	r11, r1, PT_FSR
+	/* reload original stack pointer and save it */
+	lwi	r11, r0, PER_CPU(ENTRY_SP)
+	swi	r11, r1, PT_R1
+	/* update mode indicator we are in kernel mode */
+	addik	r11, r0, 1
+	swi	r11, r0, PER_CPU(KM)
+	/* restore r31 */
+	lwi	r31, r0, PER_CPU(CURRENT_SAVE)
+	/* re-enable interrupts now we are in kernel mode */
+	enable_irq
+
+	/* See if the system call number is valid. */
+	addi	r11, r12, -__NR_syscalls
+	bgei	r11, 1f			/* return to user if not valid */
+	/* Figure out which function to use for this system call. */
+	/* Note Microblaze barrel shift is optional, so don't rely on it */
+	add	r12, r12, r12			/* convert num -> ptr */
+	addik	r30, r0, 1			/* restarts allowed */
+	add	r12, r12, r12
+	lwi	r12, r12, sys_call_table	/* Get function pointer */
+	addik	r15, r0, ret_to_user-8		/* set return address */
+	bra	r12				/* Make the system call. */
+	bri	0				/* won't reach here */
+1:
+	brid	ret_to_user			/* jump to syscall epilogue */
+	addi	r3, r0, -ENOSYS			/* set errno in delay slot */
+
+/*
+ * Debug traps are like a system call, but entered via brki r14, 0x60
+ * All we need to do is send the SIGTRAP signal to current, ptrace and
+ * do_notify_resume will handle the rest
+ */
+ENTRY(_debug_exception)
+	swi	r1, r0, PER_CPU(ENTRY_SP)	/* save the current sp */
+	lwi	r1, r0, PER_CPU(CURRENT_SAVE)	/* get the saved current */
+	lwi	r1, r1, TS_THREAD_INFO		/* get the thread info */
+	addik	r1, r1, THREAD_SIZE - PT_SIZE	/* get the kernel stack */
+	swi	r11, r0, PER_CPU(R11_SAVE)	/* temporarily save r11 */
+	lwi	r11, r0, PER_CPU(KM)		/* load mode indicator */
+//save_context:
+	swi	r11, r1, PT_MODE	/* store the mode */
+	lwi	r11, r0, PER_CPU(R11_SAVE)	/* reload r11 */
+	/* save them on stack */
+	swi	r2, r1, PT_R2
+	swi	r3, r1, PT_R3 /* r3: _always_ in clobber list; see unistd.h */
+	swi	r4, r1, PT_R4 /* r4: _always_ in clobber list; see unistd.h */
+	swi	r5, r1, PT_R5
+	swi	r6, r1, PT_R6
+	swi	r7, r1, PT_R7
+	swi	r8, r1, PT_R8
+	swi	r9, r1, PT_R9
+	swi	r10, r1, PT_R10
+	swi	r11, r1, PT_R11
+	/* r12: _always_ in clobber list; see unistd.h */
+	swi	r12, r1, PT_R12
+	swi	r13, r1, PT_R13
+	/* r14: _always_ in clobber list; see unistd.h */
+	swi	r14, r1, PT_R14
+	swi	r14, r1, PT_PC /* Will return to interrupted instruction */
+	swi	r15, r1, PT_R15
+	swi	r16, r1, PT_R16
+	swi	r17, r1, PT_R17
+	swi	r18, r1, PT_R18
+	swi	r19, r1, PT_R19
+	swi	r20, r1, PT_R20
+	swi	r21, r1, PT_R21
+	swi	r22, r1, PT_R22
+	swi	r23, r1, PT_R23
+	swi	r24, r1, PT_R24
+	swi	r25, r1, PT_R25
+	swi	r26, r1, PT_R26
+	swi	r27, r1, PT_R27
+	swi	r28, r1, PT_R28
+	swi	r29, r1, PT_R29
+	swi	r30, r1, PT_R30
+	swi	r31, r1, PT_R31
+
+	disable_irq
+	nop		/* make sure IE bit is in effect */
+	clear_bip	/* once IE is in effect it is safe to clear BIP */
+	nop
+
+	/* special purpose registers */
+	mfs	r11, rmsr
+	swi	r11, r1, PT_MSR
+	mfs	r11, rear
+	swi	r11, r1, PT_EAR
+	mfs	r11, resr
+	swi	r11, r1, PT_ESR
+	mfs	r11, rfsr
+	swi	r11, r1, PT_FSR
+	/* reload original stack pointer and save it */
+	lwi	r11, r0, PER_CPU(ENTRY_SP)
+	swi	r11, r1, PT_R1
+	/* update mode indicator we are in kernel mode */
+	addik	r11, r0, 1
+	swi	r11, r0, PER_CPU(KM)
+	/* restore r31 */
+	lwi	r31, r0, PER_CPU(CURRENT_SAVE)
+	/* re-enable interrupts now we are in kernel mode */
+	enable_irq
+
+	addi	r5, r0, SIGTRAP			/* sending the trap signal */
+	add	r6, r0, r31			/* to current */
+	bralid	r15, send_sig
+	add	r7, r0, r0			/* 3rd param zero */
+
+	addik	r30, r0, 1			/* restarts allowed ??? */
+	/* Restore r3/r4 to work around how ret_to_user works */
+	lwi	r3, r1, PT_R3
+	lwi	r4, r1, PT_R4
+	bri	ret_to_user
+
+ENTRY(_break)
+	bri	0
+
+/* struct task_struct *_switch_to(struct thread_info *prev,
+					struct thread_info *next); */
+ENTRY(_switch_to)
+	/* prepare return value */
+	addk	r3, r0, r31
+
+	/* save registers in cpu_context */
+	/* use r11 and r12, volatile registers, as temp register */
+	addik	r11, r5, TI_CPU_CONTEXT
+	swi	r1, r11, CC_R1
+	swi	r2, r11, CC_R2
+	/* skip volatile registers.
+	 * they are saved on stack when we jumped to _switch_to() */
+	/* dedicated registers */
+	swi	r13, r11, CC_R13
+	swi	r14, r11, CC_R14
+	swi	r15, r11, CC_R15
+	swi	r16, r11, CC_R16
+	swi	r17, r11, CC_R17
+	swi	r18, r11, CC_R18
+	/* save non-volatile registers */
+	swi	r19, r11, CC_R19
+	swi	r20, r11, CC_R20
+	swi	r21, r11, CC_R21
+	swi	r22, r11, CC_R22
+	swi	r23, r11, CC_R23
+	swi	r24, r11, CC_R24
+	swi	r25, r11, CC_R25
+	swi	r26, r11, CC_R26
+	swi	r27, r11, CC_R27
+	swi	r28, r11, CC_R28
+	swi	r29, r11, CC_R29
+	swi	r30, r11, CC_R30
+	/* special purpose registers */
+	mfs	r12, rmsr
+	swi	r12, r11, CC_MSR
+	mfs	r12, rear
+	swi	r12, r11, CC_EAR
+	mfs	r12, resr
+	swi	r12, r11, CC_ESR
+	mfs	r12, rfsr
+	swi	r12, r11, CC_FSR
+
+	/* update r31, the current */
+	lwi	r31, r6, TI_TASK
+	swi	r31, r0, PER_CPU(CURRENT_SAVE)
+
+	/* get new process' cpu context and restore */
+	addik	r11, r6, TI_CPU_CONTEXT
+
+	/* special purpose registers */
+	lwi	r12, r11, CC_FSR
+	mts	rfsr, r12
+	lwi	r12, r11, CC_ESR
+	mts	resr, r12
+	lwi	r12, r11, CC_EAR
+	mts	rear, r12
+	lwi	r12, r11, CC_MSR
+	mts	rmsr, r12
+	/* non-volatile registers */
+	lwi	r30, r11, CC_R30
+	lwi	r29, r11, CC_R29
+	lwi	r28, r11, CC_R28
+	lwi	r27, r11, CC_R27
+	lwi	r26, r11, CC_R26
+	lwi	r25, r11, CC_R25
+	lwi	r24, r11, CC_R24
+	lwi	r23, r11, CC_R23
+	lwi	r22, r11, CC_R22
+	lwi	r21, r11, CC_R21
+	lwi	r20, r11, CC_R20
+	lwi	r19, r11, CC_R19
+	/* dedicated registers */
+	lwi	r18, r11, CC_R18
+	lwi	r17, r11, CC_R17
+	lwi	r16, r11, CC_R16
+	lwi	r15, r11, CC_R15
+	lwi	r14, r11, CC_R14
+	lwi	r13, r11, CC_R13
+	/* skip volatile registers */
+	lwi	r2, r11, CC_R2
+	lwi	r1, r11, CC_R1
+
+	rtsd	r15, 8
+	nop
+
+ENTRY(ret_from_fork)
+	addk	r5, r0, r3
+	brlid	r15, schedule_tail
+	nop
+	swi	r31, r1, PT_R31		/* save r31 in user context. */
+			/* will soon be restored to r31 in ret_to_user */
+	addk	r3, r0, r0
+	brid	ret_to_user
+	nop
+
+ENTRY(ret_from_kernel_thread)
+	brlid	r15, schedule_tail
+	addk	r5, r0, r3
+	brald	r15, r20
+	addk	r5, r0, r19
+	brid	ret_to_user
+	addk	r3, r0, r0
+
+work_pending:
+	lwi	r11, r1, PT_MODE
+	bneid	r11, 2f
+3:
+	enable_irq
+	andi	r11, r19, _TIF_NEED_RESCHED
+	beqi	r11, 1f
+	bralid	r15, schedule
+	nop
+	bri	4f
+1:	andi	r11, r19, _TIF_SIGPENDING | _TIF_NOTIFY_RESUME
+	beqi	r11, no_work_pending
+	addk	r5, r30, r0
+	bralid	r15, do_notify_resume
+	addik	r6, r0, 1
+	addk	r30, r0, r0	/* no restarts from now on */
+4:
+	disable_irq
+	lwi	r6, r31, TS_THREAD_INFO /* get thread info */
+	lwi	r19, r6, TI_FLAGS /* get flags in thread info */
+	bri	3b
+
+ENTRY(ret_to_user)
+	disable_irq
+
+	swi	r4, r1, PT_R4		/* return val */
+	swi	r3, r1, PT_R3		/* return val */
+
+	lwi	r6, r31, TS_THREAD_INFO /* get thread info */
+	lwi	r19, r6, TI_FLAGS /* get flags in thread info */
+	bnei	r19, work_pending /* do an extra work if any bits are set */
+no_work_pending:
+	disable_irq
+
+2:
+	/* save r31 */
+	swi	r31, r0, PER_CPU(CURRENT_SAVE)
+	/* save mode indicator */
+	lwi	r18, r1, PT_MODE
+	swi	r18, r0, PER_CPU(KM)
+//restore_context:
+	/* special purpose registers */
+	lwi	r18, r1, PT_FSR
+	mts	rfsr, r18
+	lwi	r18, r1, PT_ESR
+	mts	resr, r18
+	lwi	r18, r1, PT_EAR
+	mts	rear, r18
+	lwi	r18, r1, PT_MSR
+	mts	rmsr, r18
+
+	lwi	r31, r1, PT_R31
+	lwi	r30, r1, PT_R30
+	lwi	r29, r1, PT_R29
+	lwi	r28, r1, PT_R28
+	lwi	r27, r1, PT_R27
+	lwi	r26, r1, PT_R26
+	lwi	r25, r1, PT_R25
+	lwi	r24, r1, PT_R24
+	lwi	r23, r1, PT_R23
+	lwi	r22, r1, PT_R22
+	lwi	r21, r1, PT_R21
+	lwi	r20, r1, PT_R20
+	lwi	r19, r1, PT_R19
+	lwi	r18, r1, PT_R18
+	lwi	r17, r1, PT_R17
+	lwi	r16, r1, PT_R16
+	lwi	r15, r1, PT_R15
+	lwi	r14, r1, PT_PC
+	lwi	r13, r1, PT_R13
+	lwi	r12, r1, PT_R12
+	lwi	r11, r1, PT_R11
+	lwi	r10, r1, PT_R10
+	lwi	r9, r1, PT_R9
+	lwi	r8, r1, PT_R8
+	lwi	r7, r1, PT_R7
+	lwi	r6, r1, PT_R6
+	lwi	r5, r1, PT_R5
+	lwi	r4, r1, PT_R4		/* return val */
+	lwi	r3, r1, PT_R3		/* return val */
+	lwi	r2, r1, PT_R2
+	lwi	r1, r1, PT_R1
+
+	rtid	r14, 0
+	nop
+
+sys_rt_sigreturn_wrapper:
+	addk	r30, r0, r0		/* no restarts for this one */
+	brid	sys_rt_sigreturn
+	addk	r5, r1, r0
+
+	/* Interrupt vector table */
+	.section	.init.ivt, "ax"
+	.org 0x0
+	brai	_reset
+	brai	_user_exception
+	brai	_interrupt
+	brai	_break
+	brai	_hw_exception_handler
+	.org 0x60
+	brai	_debug_exception
+
+.section .rodata,"a"
+#include "syscall_table.S"
+
+syscall_table_size=(.-sys_call_table)
+
+type_SYSCALL:
+	.ascii "SYSCALL\0"
+type_IRQ:
+	.ascii "IRQ\0"
+type_IRQ_PREEMPT:
+	.ascii "IRQ (PREEMPTED)\0"
+type_SYSCALL_PREEMPT:
+	.ascii " SYSCALL (PREEMPTED)\0"
+
+	/*
+	 * Trap decoding for stack unwinder
+	 * Tuples are (start addr, end addr, string)
+	 * If return address lies on [start addr, end addr],
+	 * unwinder displays 'string'
+	 */
+
+	.align 4
+.global microblaze_trap_handlers
+microblaze_trap_handlers:
+	/* Exact matches come first */
+	.word ret_to_user  ; .word ret_to_user    ; .word type_SYSCALL
+	.word ret_from_intr; .word ret_from_intr  ; .word type_IRQ
+	/* Fuzzy matches go here */
+	.word ret_from_intr; .word no_intr_resched; .word type_IRQ_PREEMPT
+	.word work_pending ; .word no_work_pending; .word type_SYSCALL_PREEMPT
+	/* End of table */
+	.word 0             ; .word 0               ; .word 0
diff --git a/arch/microblaze/kernel/entry.S b/arch/microblaze/kernel/entry.S
new file mode 100644
index 0000000..ef54851
--- /dev/null
+++ b/arch/microblaze/kernel/entry.S
@@ -0,0 +1,1005 @@
+/*
+ * Low-level system-call handling, trap handlers and context-switching
+ *
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ * Copyright (C) 2003		John Williams <jwilliams@itee.uq.edu.au>
+ * Copyright (C) 2001,2002	NEC Corporation
+ * Copyright (C) 2001,2002	Miles Bader <miles@gnu.org>
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License. See the file COPYING in the main directory of this
+ * archive for more details.
+ *
+ * Written by Miles Bader <miles@gnu.org>
+ * Heavily modified by John Williams for Microblaze
+ */
+
+#include <linux/sys.h>
+#include <linux/linkage.h>
+
+#include <asm/entry.h>
+#include <asm/current.h>
+#include <asm/processor.h>
+#include <asm/exceptions.h>
+#include <asm/asm-offsets.h>
+#include <asm/thread_info.h>
+
+#include <asm/page.h>
+#include <asm/unistd.h>
+
+#include <linux/errno.h>
+#include <asm/signal.h>
+
+#undef DEBUG
+
+#ifdef DEBUG
+/* Create space for syscalls counting. */
+.section .data
+.global syscall_debug_table
+.align 4
+syscall_debug_table:
+	.space	(__NR_syscalls * 4)
+#endif /* DEBUG */
+
+#define C_ENTRY(name)	.globl name; .align 4; name
+
+/*
+ * Various ways of setting and clearing BIP in flags reg.
+ * This is mucky, but necessary using microblaze version that
+ * allows msr ops to write to BIP
+ */
+#if CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR
+	.macro	clear_bip
+	msrclr	r0, MSR_BIP
+	.endm
+
+	.macro	set_bip
+	msrset	r0, MSR_BIP
+	.endm
+
+	.macro	clear_eip
+	msrclr	r0, MSR_EIP
+	.endm
+
+	.macro	set_ee
+	msrset	r0, MSR_EE
+	.endm
+
+	.macro	disable_irq
+	msrclr	r0, MSR_IE
+	.endm
+
+	.macro	enable_irq
+	msrset	r0, MSR_IE
+	.endm
+
+	.macro	set_ums
+	msrset	r0, MSR_UMS
+	msrclr	r0, MSR_VMS
+	.endm
+
+	.macro	set_vms
+	msrclr	r0, MSR_UMS
+	msrset	r0, MSR_VMS
+	.endm
+
+	.macro	clear_ums
+	msrclr	r0, MSR_UMS
+	.endm
+
+	.macro	clear_vms_ums
+	msrclr	r0, MSR_VMS | MSR_UMS
+	.endm
+#else
+	.macro	clear_bip
+	mfs	r11, rmsr
+	andi	r11, r11, ~MSR_BIP
+	mts	rmsr, r11
+	.endm
+
+	.macro	set_bip
+	mfs	r11, rmsr
+	ori	r11, r11, MSR_BIP
+	mts	rmsr, r11
+	.endm
+
+	.macro	clear_eip
+	mfs	r11, rmsr
+	andi	r11, r11, ~MSR_EIP
+	mts	rmsr, r11
+	.endm
+
+	.macro	set_ee
+	mfs	r11, rmsr
+	ori	r11, r11, MSR_EE
+	mts	rmsr, r11
+	.endm
+
+	.macro	disable_irq
+	mfs	r11, rmsr
+	andi	r11, r11, ~MSR_IE
+	mts	rmsr, r11
+	.endm
+
+	.macro	enable_irq
+	mfs	r11, rmsr
+	ori	r11, r11, MSR_IE
+	mts	rmsr, r11
+	.endm
+
+	.macro set_ums
+	mfs	r11, rmsr
+	ori	r11, r11, MSR_VMS
+	andni	r11, r11, MSR_UMS
+	mts	rmsr, r11
+	.endm
+
+	.macro	set_vms
+	mfs	r11, rmsr
+	ori	r11, r11, MSR_VMS
+	andni	r11, r11, MSR_UMS
+	mts	rmsr, r11
+	.endm
+
+	.macro	clear_ums
+	mfs	r11, rmsr
+	andni	r11, r11, MSR_UMS
+	mts	rmsr,r11
+	.endm
+
+	.macro	clear_vms_ums
+	mfs	r11, rmsr
+	andni	r11, r11, (MSR_VMS|MSR_UMS)
+	mts	rmsr,r11
+	.endm
+#endif
+
+/* Define how to call high-level functions. With MMU, virtual mode must be
+ * enabled when calling the high-level function. Clobbers R11.
+ * VM_ON, VM_OFF, DO_JUMP_BIPCLR, DO_CALL
+ */
+
+/* turn on virtual protected mode save */
+#define VM_ON		\
+	set_ums;	\
+	rted	r0, 2f;	\
+	nop; \
+2:
+
+/* turn off virtual protected mode save and user mode save*/
+#define VM_OFF			\
+	clear_vms_ums;		\
+	rted	r0, TOPHYS(1f);	\
+	nop; \
+1:
+
+#define SAVE_REGS \
+	swi	r2, r1, PT_R2;	/* Save SDA */			\
+	swi	r3, r1, PT_R3;					\
+	swi	r4, r1, PT_R4;					\
+	swi	r5, r1, PT_R5;					\
+	swi	r6, r1, PT_R6;					\
+	swi	r7, r1, PT_R7;					\
+	swi	r8, r1, PT_R8;					\
+	swi	r9, r1, PT_R9;					\
+	swi	r10, r1, PT_R10;					\
+	swi	r11, r1, PT_R11;	/* save clobbered regs after rval */\
+	swi	r12, r1, PT_R12;					\
+	swi	r13, r1, PT_R13;	/* Save SDA2 */			\
+	swi	r14, r1, PT_PC;	/* PC, before IRQ/trap */	\
+	swi	r15, r1, PT_R15;	/* Save LP */			\
+	swi	r16, r1, PT_R16;					\
+	swi	r17, r1, PT_R17;					\
+	swi	r18, r1, PT_R18;	/* Save asm scratch reg */	\
+	swi	r19, r1, PT_R19;					\
+	swi	r20, r1, PT_R20;					\
+	swi	r21, r1, PT_R21;					\
+	swi	r22, r1, PT_R22;					\
+	swi	r23, r1, PT_R23;					\
+	swi	r24, r1, PT_R24;					\
+	swi	r25, r1, PT_R25;					\
+	swi	r26, r1, PT_R26;					\
+	swi	r27, r1, PT_R27;					\
+	swi	r28, r1, PT_R28;					\
+	swi	r29, r1, PT_R29;					\
+	swi	r30, r1, PT_R30;					\
+	swi	r31, r1, PT_R31;	/* Save current task reg */	\
+	mfs	r11, rmsr;		/* save MSR */			\
+	swi	r11, r1, PT_MSR;
+
+#define RESTORE_REGS \
+	lwi	r11, r1, PT_MSR;					\
+	mts	rmsr , r11;						\
+	lwi	r2, r1, PT_R2;	/* restore SDA */		\
+	lwi	r3, r1, PT_R3;					\
+	lwi	r4, r1, PT_R4;					\
+	lwi	r5, r1, PT_R5;					\
+	lwi	r6, r1, PT_R6;					\
+	lwi	r7, r1, PT_R7;					\
+	lwi	r8, r1, PT_R8;					\
+	lwi	r9, r1, PT_R9;					\
+	lwi	r10, r1, PT_R10;					\
+	lwi	r11, r1, PT_R11;	/* restore clobbered regs after rval */\
+	lwi	r12, r1, PT_R12;					\
+	lwi	r13, r1, PT_R13;	/* restore SDA2 */		\
+	lwi	r14, r1, PT_PC;	/* RESTORE_LINK PC, before IRQ/trap */\
+	lwi	r15, r1, PT_R15;	/* restore LP */		\
+	lwi	r16, r1, PT_R16;					\
+	lwi	r17, r1, PT_R17;					\
+	lwi	r18, r1, PT_R18;	/* restore asm scratch reg */	\
+	lwi	r19, r1, PT_R19;					\
+	lwi	r20, r1, PT_R20;					\
+	lwi	r21, r1, PT_R21;					\
+	lwi	r22, r1, PT_R22;					\
+	lwi	r23, r1, PT_R23;					\
+	lwi	r24, r1, PT_R24;					\
+	lwi	r25, r1, PT_R25;					\
+	lwi	r26, r1, PT_R26;					\
+	lwi	r27, r1, PT_R27;					\
+	lwi	r28, r1, PT_R28;					\
+	lwi	r29, r1, PT_R29;					\
+	lwi	r30, r1, PT_R30;					\
+	lwi	r31, r1, PT_R31;	/* Restore cur task reg */
+
+#define SAVE_STATE	\
+	swi	r1, r0, TOPHYS(PER_CPU(ENTRY_SP)); /* save stack */	\
+	/* See if already in kernel mode.*/				\
+	mfs	r1, rmsr;						\
+	andi	r1, r1, MSR_UMS;					\
+	bnei	r1, 1f;						\
+	/* Kernel-mode state save.  */					\
+	/* Reload kernel stack-ptr. */					\
+	lwi	r1, r0, TOPHYS(PER_CPU(ENTRY_SP));			\
+	/* FIXME: I can add these two lines to one */			\
+	/* tophys(r1,r1); */						\
+	/* addik	r1, r1, -PT_SIZE; */				\
+	addik	r1, r1, CONFIG_KERNEL_BASE_ADDR - CONFIG_KERNEL_START - PT_SIZE; \
+	SAVE_REGS							\
+	brid	2f;							\
+	swi	r1, r1, PT_MODE; 	 				\
+1:	/* User-mode state save.  */					\
+	lwi	r1, r0, TOPHYS(PER_CPU(CURRENT_SAVE)); /* get saved current */\
+	tophys(r1,r1);							\
+	lwi	r1, r1, TS_THREAD_INFO;	/* get the thread info */	\
+	/* MS these three instructions can be added to one */		\
+	/* addik	r1, r1, THREAD_SIZE; */				\
+	/* tophys(r1,r1); */						\
+	/* addik	r1, r1, -PT_SIZE; */			\
+	addik r1, r1, THREAD_SIZE + CONFIG_KERNEL_BASE_ADDR - CONFIG_KERNEL_START - PT_SIZE; \
+	SAVE_REGS							\
+	lwi	r11, r0, TOPHYS(PER_CPU(ENTRY_SP));			\
+	swi	r11, r1, PT_R1; /* Store user SP.  */		\
+	swi	r0, r1, PT_MODE; /* Was in user-mode.  */		\
+	/* MS: I am clearing UMS even in case when I come from kernel space */ \
+	clear_ums; 							\
+2:	lwi	CURRENT_TASK, r0, TOPHYS(PER_CPU(CURRENT_SAVE));
+
+.text
+
+/*
+ * User trap.
+ *
+ * System calls are handled here.
+ *
+ * Syscall protocol:
+ * Syscall number in r12, args in r5-r10
+ * Return value in r3
+ *
+ * Trap entered via brki instruction, so BIP bit is set, and interrupts
+ * are masked. This is nice, means we don't have to CLI before state save
+ */
+C_ENTRY(_user_exception):
+	swi	r1, r0, TOPHYS(PER_CPU(ENTRY_SP)) /* save stack */
+	addi	r14, r14, 4	/* return address is 4 byte after call */
+
+	lwi	r1, r0, TOPHYS(PER_CPU(CURRENT_SAVE)); /* get saved current */
+	tophys(r1,r1);
+	lwi	r1, r1, TS_THREAD_INFO;	/* get stack from task_struct */
+/* calculate kernel stack pointer from task struct 8k */
+	addik	r1, r1, THREAD_SIZE;
+	tophys(r1,r1);
+
+	addik	r1, r1, -PT_SIZE; /* Make room on the stack.  */
+	SAVE_REGS
+	swi	r0, r1, PT_R3
+	swi	r0, r1, PT_R4
+
+	swi	r0, r1, PT_MODE;			/* Was in user-mode. */
+	lwi	r11, r0, TOPHYS(PER_CPU(ENTRY_SP));
+	swi	r11, r1, PT_R1;		/* Store user SP.  */
+	clear_ums;
+2:	lwi	CURRENT_TASK, r0, TOPHYS(PER_CPU(CURRENT_SAVE));
+	/* Save away the syscall number.  */
+	swi	r12, r1, PT_R0;
+	tovirt(r1,r1)
+
+/* where the trap should return need -8 to adjust for rtsd r15, 8*/
+/* Jump to the appropriate function for the system call number in r12
+ * (r12 is not preserved), or return an error if r12 is not valid. The LP
+ * register should point to the location where
+ * the called function should return.  [note that MAKE_SYS_CALL uses label 1] */
+
+	/* Step into virtual mode */
+	rtbd	r0, 3f
+	nop
+3:
+	lwi	r11, CURRENT_TASK, TS_THREAD_INFO /* get thread info */
+	lwi	r11, r11, TI_FLAGS	 /* get flags in thread info */
+	andi	r11, r11, _TIF_WORK_SYSCALL_MASK
+	beqi	r11, 4f
+
+	addik	r3, r0, -ENOSYS
+	swi	r3, r1, PT_R3
+	brlid	r15, do_syscall_trace_enter
+	addik	r5, r1, PT_R0
+
+	# do_syscall_trace_enter returns the new syscall nr.
+	addk	r12, r0, r3
+	lwi	r5, r1, PT_R5;
+	lwi	r6, r1, PT_R6;
+	lwi	r7, r1, PT_R7;
+	lwi	r8, r1, PT_R8;
+	lwi	r9, r1, PT_R9;
+	lwi	r10, r1, PT_R10;
+4:
+/* Jump to the appropriate function for the system call number in r12
+ * (r12 is not preserved), or return an error if r12 is not valid.
+ * The LP register should point to the location where the called function
+ * should return.  [note that MAKE_SYS_CALL uses label 1] */
+	/* See if the system call number is valid */
+	blti	r12, 5f
+	addi	r11, r12, -__NR_syscalls;
+	bgei	r11, 5f;
+	/* Figure out which function to use for this system call.  */
+	/* Note Microblaze barrel shift is optional, so don't rely on it */
+	add	r12, r12, r12;			/* convert num -> ptr */
+	add	r12, r12, r12;
+	addi	r30, r0, 1			/* restarts allowed */
+
+#ifdef DEBUG
+	/* Trac syscalls and stored them to syscall_debug_table */
+	/* The first syscall location stores total syscall number */
+	lwi	r3, r0, syscall_debug_table
+	addi	r3, r3, 1
+	swi	r3, r0, syscall_debug_table
+	lwi	r3, r12, syscall_debug_table
+	addi	r3, r3, 1
+	swi	r3, r12, syscall_debug_table
+#endif
+
+	# Find and jump into the syscall handler.
+	lwi	r12, r12, sys_call_table
+	/* where the trap should return need -8 to adjust for rtsd r15, 8 */
+	addi	r15, r0, ret_from_trap-8
+	bra	r12
+
+	/* The syscall number is invalid, return an error.  */
+5:
+	braid	ret_from_trap
+	addi	r3, r0, -ENOSYS;
+
+/* Entry point used to return from a syscall/trap */
+/* We re-enable BIP bit before state restore */
+C_ENTRY(ret_from_trap):
+	swi	r3, r1, PT_R3
+	swi	r4, r1, PT_R4
+
+	lwi	r11, r1, PT_MODE;
+/* See if returning to kernel mode, if so, skip resched &c.  */
+	bnei	r11, 2f;
+	/* We're returning to user mode, so check for various conditions that
+	 * trigger rescheduling. */
+	/* FIXME: Restructure all these flag checks. */
+	lwi	r11, CURRENT_TASK, TS_THREAD_INFO;	/* get thread info */
+	lwi	r11, r11, TI_FLAGS;		/* get flags in thread info */
+	andi	r11, r11, _TIF_WORK_SYSCALL_MASK
+	beqi	r11, 1f
+
+	brlid	r15, do_syscall_trace_leave
+	addik	r5, r1, PT_R0
+1:
+	/* We're returning to user mode, so check for various conditions that
+	 * trigger rescheduling. */
+	/* get thread info from current task */
+	lwi	r11, CURRENT_TASK, TS_THREAD_INFO;
+	lwi	r19, r11, TI_FLAGS;		/* get flags in thread info */
+	andi	r11, r19, _TIF_NEED_RESCHED;
+	beqi	r11, 5f;
+
+	bralid	r15, schedule;	/* Call scheduler */
+	nop;				/* delay slot */
+	bri	1b
+
+	/* Maybe handle a signal */
+5:
+	andi	r11, r19, _TIF_SIGPENDING | _TIF_NOTIFY_RESUME;
+	beqi	r11, 4f;		/* Signals to handle, handle them */
+
+	addik	r5, r1, 0;		/* Arg 1: struct pt_regs *regs */
+	bralid	r15, do_notify_resume;	/* Handle any signals */
+	add	r6, r30, r0;		/* Arg 2: int in_syscall */
+	add	r30, r0, r0		/* no more restarts */
+	bri	1b
+
+/* Finally, return to user state.  */
+4:	set_bip;			/*  Ints masked for state restore */
+	swi	CURRENT_TASK, r0, PER_CPU(CURRENT_SAVE); /* save current */
+	VM_OFF;
+	tophys(r1,r1);
+	RESTORE_REGS;
+	addik	r1, r1, PT_SIZE		/* Clean up stack space.  */
+	lwi	r1, r1, PT_R1 - PT_SIZE;/* Restore user stack pointer. */
+	bri	6f;
+
+/* Return to kernel state.  */
+2:	set_bip;			/*  Ints masked for state restore */
+	VM_OFF;
+	tophys(r1,r1);
+	RESTORE_REGS;
+	addik	r1, r1, PT_SIZE		/* Clean up stack space.  */
+	tovirt(r1,r1);
+6:
+TRAP_return:		/* Make global symbol for debugging */
+	rtbd	r14, 0;	/* Instructions to return from an IRQ */
+	nop;
+
+
+/* This the initial entry point for a new child thread, with an appropriate
+   stack in place that makes it look the the child is in the middle of an
+   syscall.  This function is actually `returned to' from switch_thread
+   (copy_thread makes ret_from_fork the return address in each new thread's
+   saved context).  */
+C_ENTRY(ret_from_fork):
+	bralid	r15, schedule_tail; /* ...which is schedule_tail's arg */
+	add	r5, r3, r0;	/* switch_thread returns the prev task */
+				/* ( in the delay slot ) */
+	brid	ret_from_trap;	/* Do normal trap return */
+	add	r3, r0, r0;	/* Child's fork call should return 0. */
+
+C_ENTRY(ret_from_kernel_thread):
+	bralid	r15, schedule_tail; /* ...which is schedule_tail's arg */
+	add	r5, r3, r0;	/* switch_thread returns the prev task */
+				/* ( in the delay slot ) */
+	brald	r15, r20	/* fn was left in r20 */
+	addk	r5, r0, r19	/* ... and argument - in r19 */
+	brid	ret_from_trap
+	add	r3, r0, r0
+
+C_ENTRY(sys_rt_sigreturn_wrapper):
+	addik	r30, r0, 0		/* no restarts */
+	brid	sys_rt_sigreturn	/* Do real work */
+	addik	r5, r1, 0;		/* add user context as 1st arg */
+
+/*
+ * HW EXCEPTION rutine start
+ */
+C_ENTRY(full_exception_trap):
+	/* adjust exception address for privileged instruction
+	 * for finding where is it */
+	addik	r17, r17, -4
+	SAVE_STATE /* Save registers */
+	/* PC, before IRQ/trap - this is one instruction above */
+	swi	r17, r1, PT_PC;
+	tovirt(r1,r1)
+	/* FIXME this can be store directly in PT_ESR reg.
+	 * I tested it but there is a fault */
+	/* where the trap should return need -8 to adjust for rtsd r15, 8 */
+	addik	r15, r0, ret_from_exc - 8
+	mfs	r6, resr
+	mfs	r7, rfsr;		/* save FSR */
+	mts	rfsr, r0;	/* Clear sticky fsr */
+	rted	r0, full_exception
+	addik	r5, r1, 0		 /* parameter struct pt_regs * regs */
+
+/*
+ * Unaligned data trap.
+ *
+ * Unaligned data trap last on 4k page is handled here.
+ *
+ * Trap entered via exception, so EE bit is set, and interrupts
+ * are masked.  This is nice, means we don't have to CLI before state save
+ *
+ * The assembler routine is in "arch/microblaze/kernel/hw_exception_handler.S"
+ */
+C_ENTRY(unaligned_data_trap):
+	/* MS: I have to save r11 value and then restore it because
+	 * set_bit, clear_eip, set_ee use r11 as temp register if MSR
+	 * instructions are not used. We don't need to do if MSR instructions
+	 * are used and they use r0 instead of r11.
+	 * I am using ENTRY_SP which should be primary used only for stack
+	 * pointer saving. */
+	swi	r11, r0, TOPHYS(PER_CPU(ENTRY_SP));
+	set_bip;        /* equalize initial state for all possible entries */
+	clear_eip;
+	set_ee;
+	lwi	r11, r0, TOPHYS(PER_CPU(ENTRY_SP));
+	SAVE_STATE		/* Save registers.*/
+	/* PC, before IRQ/trap - this is one instruction above */
+	swi	r17, r1, PT_PC;
+	tovirt(r1,r1)
+	/* where the trap should return need -8 to adjust for rtsd r15, 8 */
+	addik	r15, r0, ret_from_exc-8
+	mfs	r3, resr		/* ESR */
+	mfs	r4, rear		/* EAR */
+	rtbd	r0, _unaligned_data_exception
+	addik	r7, r1, 0		/* parameter struct pt_regs * regs */
+
+/*
+ * Page fault traps.
+ *
+ * If the real exception handler (from hw_exception_handler.S) didn't find
+ * the mapping for the process, then we're thrown here to handle such situation.
+ *
+ * Trap entered via exceptions, so EE bit is set, and interrupts
+ * are masked.  This is nice, means we don't have to CLI before state save
+ *
+ * Build a standard exception frame for TLB Access errors.  All TLB exceptions
+ * will bail out to this point if they can't resolve the lightweight TLB fault.
+ *
+ * The C function called is in "arch/microblaze/mm/fault.c", declared as:
+ * void do_page_fault(struct pt_regs *regs,
+ *				unsigned long address,
+ *				unsigned long error_code)
+ */
+/* data and intruction trap - which is choose is resolved int fault.c */
+C_ENTRY(page_fault_data_trap):
+	SAVE_STATE		/* Save registers.*/
+	/* PC, before IRQ/trap - this is one instruction above */
+	swi	r17, r1, PT_PC;
+	tovirt(r1,r1)
+	/* where the trap should return need -8 to adjust for rtsd r15, 8 */
+	addik	r15, r0, ret_from_exc-8
+	mfs	r6, rear		/* parameter unsigned long address */
+	mfs	r7, resr		/* parameter unsigned long error_code */
+	rted	r0, do_page_fault
+	addik	r5, r1, 0		/* parameter struct pt_regs * regs */
+
+C_ENTRY(page_fault_instr_trap):
+	SAVE_STATE		/* Save registers.*/
+	/* PC, before IRQ/trap - this is one instruction above */
+	swi	r17, r1, PT_PC;
+	tovirt(r1,r1)
+	/* where the trap should return need -8 to adjust for rtsd r15, 8 */
+	addik	r15, r0, ret_from_exc-8
+	mfs	r6, rear		/* parameter unsigned long address */
+	ori	r7, r0, 0		/* parameter unsigned long error_code */
+	rted	r0, do_page_fault
+	addik	r5, r1, 0		/* parameter struct pt_regs * regs */
+
+/* Entry point used to return from an exception.  */
+C_ENTRY(ret_from_exc):
+	lwi	r11, r1, PT_MODE;
+	bnei	r11, 2f;		/* See if returning to kernel mode, */
+					/* ... if so, skip resched &c.  */
+
+	/* We're returning to user mode, so check for various conditions that
+	   trigger rescheduling. */
+1:
+	lwi	r11, CURRENT_TASK, TS_THREAD_INFO;	/* get thread info */
+	lwi	r19, r11, TI_FLAGS;	/* get flags in thread info */
+	andi	r11, r19, _TIF_NEED_RESCHED;
+	beqi	r11, 5f;
+
+/* Call the scheduler before returning from a syscall/trap. */
+	bralid	r15, schedule;	/* Call scheduler */
+	nop;				/* delay slot */
+	bri	1b
+
+	/* Maybe handle a signal */
+5:	andi	r11, r19, _TIF_SIGPENDING | _TIF_NOTIFY_RESUME;
+	beqi	r11, 4f;		/* Signals to handle, handle them */
+
+	/*
+	 * Handle a signal return; Pending signals should be in r18.
+	 *
+	 * Not all registers are saved by the normal trap/interrupt entry
+	 * points (for instance, call-saved registers (because the normal
+	 * C-compiler calling sequence in the kernel makes sure they're
+	 * preserved), and call-clobbered registers in the case of
+	 * traps), but signal handlers may want to examine or change the
+	 * complete register state.  Here we save anything not saved by
+	 * the normal entry sequence, so that it may be safely restored
+	 * (in a possibly modified form) after do_notify_resume returns. */
+	addik	r5, r1, 0;		/* Arg 1: struct pt_regs *regs */
+	bralid	r15, do_notify_resume;	/* Handle any signals */
+	addi	r6, r0, 0;		/* Arg 2: int in_syscall */
+	bri	1b
+
+/* Finally, return to user state.  */
+4:	set_bip;			/* Ints masked for state restore */
+	swi	CURRENT_TASK, r0, PER_CPU(CURRENT_SAVE); /* save current */
+	VM_OFF;
+	tophys(r1,r1);
+
+	RESTORE_REGS;
+	addik	r1, r1, PT_SIZE		/* Clean up stack space.  */
+
+	lwi	r1, r1, PT_R1 - PT_SIZE; /* Restore user stack pointer. */
+	bri	6f;
+/* Return to kernel state.  */
+2:	set_bip;			/* Ints masked for state restore */
+	VM_OFF;
+	tophys(r1,r1);
+	RESTORE_REGS;
+	addik	r1, r1, PT_SIZE		/* Clean up stack space.  */
+
+	tovirt(r1,r1);
+6:
+EXC_return:		/* Make global symbol for debugging */
+	rtbd	r14, 0;	/* Instructions to return from an IRQ */
+	nop;
+
+/*
+ * HW EXCEPTION rutine end
+ */
+
+/*
+ * Hardware maskable interrupts.
+ *
+ * The stack-pointer (r1) should have already been saved to the memory
+ * location PER_CPU(ENTRY_SP).
+ */
+C_ENTRY(_interrupt):
+/* MS: we are in physical address */
+/* Save registers, switch to proper stack, convert SP to virtual.*/
+	swi	r1, r0, TOPHYS(PER_CPU(ENTRY_SP))
+	/* MS: See if already in kernel mode. */
+	mfs	r1, rmsr
+	nop
+	andi	r1, r1, MSR_UMS
+	bnei	r1, 1f
+
+/* Kernel-mode state save. */
+	lwi	r1, r0, TOPHYS(PER_CPU(ENTRY_SP))
+	tophys(r1,r1); /* MS: I have in r1 physical address where stack is */
+	/* save registers */
+/* MS: Make room on the stack -> activation record */
+	addik	r1, r1, -PT_SIZE;
+	SAVE_REGS
+	brid	2f;
+	swi	r1, r1, PT_MODE; /* 0 - user mode, 1 - kernel mode */
+1:
+/* User-mode state save. */
+ /* MS: get the saved current */
+	lwi	r1, r0, TOPHYS(PER_CPU(CURRENT_SAVE));
+	tophys(r1,r1);
+	lwi	r1, r1, TS_THREAD_INFO;
+	addik	r1, r1, THREAD_SIZE;
+	tophys(r1,r1);
+	/* save registers */
+	addik	r1, r1, -PT_SIZE;
+	SAVE_REGS
+	/* calculate mode */
+	swi	r0, r1, PT_MODE;
+	lwi	r11, r0, TOPHYS(PER_CPU(ENTRY_SP));
+	swi	r11, r1, PT_R1;
+	clear_ums;
+2:
+	lwi	CURRENT_TASK, r0, TOPHYS(PER_CPU(CURRENT_SAVE));
+	tovirt(r1,r1)
+	addik	r15, r0, irq_call;
+irq_call:rtbd	r0, do_IRQ;
+	addik	r5, r1, 0;
+
+/* MS: we are in virtual mode */
+ret_from_irq:
+	lwi	r11, r1, PT_MODE;
+	bnei	r11, 2f;
+
+1:
+	lwi	r11, CURRENT_TASK, TS_THREAD_INFO;
+	lwi	r19, r11, TI_FLAGS; /* MS: get flags from thread info */
+	andi	r11, r19, _TIF_NEED_RESCHED;
+	beqi	r11, 5f
+	bralid	r15, schedule;
+	nop; /* delay slot */
+	bri	1b
+
+    /* Maybe handle a signal */
+5:	andi	r11, r19, _TIF_SIGPENDING | _TIF_NOTIFY_RESUME;
+	beqid	r11, no_intr_resched
+/* Handle a signal return; Pending signals should be in r18. */
+	addik	r5, r1, 0; /* Arg 1: struct pt_regs *regs */
+	bralid	r15, do_notify_resume;	/* Handle any signals */
+	addi	r6, r0, 0; /* Arg 2: int in_syscall */
+	bri	1b
+
+/* Finally, return to user state. */
+no_intr_resched:
+    /* Disable interrupts, we are now committed to the state restore */
+	disable_irq
+	swi	CURRENT_TASK, r0, PER_CPU(CURRENT_SAVE);
+	VM_OFF;
+	tophys(r1,r1);
+	RESTORE_REGS
+	addik	r1, r1, PT_SIZE /* MS: Clean up stack space. */
+	lwi	r1, r1, PT_R1 - PT_SIZE;
+	bri	6f;
+/* MS: Return to kernel state. */
+2:
+#ifdef CONFIG_PREEMPT
+	lwi	r11, CURRENT_TASK, TS_THREAD_INFO;
+	/* MS: get preempt_count from thread info */
+	lwi	r5, r11, TI_PREEMPT_COUNT;
+	bgti	r5, restore;
+
+	lwi	r5, r11, TI_FLAGS;		/* get flags in thread info */
+	andi	r5, r5, _TIF_NEED_RESCHED;
+	beqi	r5, restore /* if zero jump over */
+
+preempt:
+	/* interrupts are off that's why I am calling preempt_chedule_irq */
+	bralid	r15, preempt_schedule_irq
+	nop
+	lwi	r11, CURRENT_TASK, TS_THREAD_INFO;	/* get thread info */
+	lwi	r5, r11, TI_FLAGS;		/* get flags in thread info */
+	andi	r5, r5, _TIF_NEED_RESCHED;
+	bnei	r5, preempt /* if non zero jump to resched */
+restore:
+#endif
+	VM_OFF /* MS: turn off MMU */
+	tophys(r1,r1)
+	RESTORE_REGS
+	addik	r1, r1, PT_SIZE	/* MS: Clean up stack space. */
+	tovirt(r1,r1);
+6:
+IRQ_return: /* MS: Make global symbol for debugging */
+	rtid	r14, 0
+	nop
+
+/*
+ * Debug trap for KGDB. Enter to _debug_exception by brki r16, 0x18
+ * and call handling function with saved pt_regs
+ */
+C_ENTRY(_debug_exception):
+	/* BIP bit is set on entry, no interrupts can occur */
+	swi	r1, r0, TOPHYS(PER_CPU(ENTRY_SP))
+
+	mfs	r1, rmsr
+	nop
+	andi	r1, r1, MSR_UMS
+	bnei	r1, 1f
+/* MS: Kernel-mode state save - kgdb */
+	lwi	r1, r0, TOPHYS(PER_CPU(ENTRY_SP)); /* Reload kernel stack-ptr*/
+
+	/* BIP bit is set on entry, no interrupts can occur */
+	addik   r1, r1, CONFIG_KERNEL_BASE_ADDR - CONFIG_KERNEL_START - PT_SIZE;
+	SAVE_REGS;
+	/* save all regs to pt_reg structure */
+	swi	r0, r1, PT_R0;	/* R0 must be saved too */
+	swi	r14, r1, PT_R14	/* rewrite saved R14 value */
+	swi	r16, r1, PT_PC; /* PC and r16 are the same */
+	/* save special purpose registers to pt_regs */
+	mfs	r11, rear;
+	swi	r11, r1, PT_EAR;
+	mfs	r11, resr;
+	swi	r11, r1, PT_ESR;
+	mfs	r11, rfsr;
+	swi	r11, r1, PT_FSR;
+
+	/* stack pointer is in physical address at it is decrease
+	 * by PT_SIZE but we need to get correct R1 value */
+	addik   r11, r1, CONFIG_KERNEL_START - CONFIG_KERNEL_BASE_ADDR + PT_SIZE;
+	swi	r11, r1, PT_R1
+	/* MS: r31 - current pointer isn't changed */
+	tovirt(r1,r1)
+#ifdef CONFIG_KGDB
+	addi	r5, r1, 0 /* pass pt_reg address as the first arg */
+	addik	r15, r0, dbtrap_call; /* return address */
+	rtbd	r0, microblaze_kgdb_break
+	nop;
+#endif
+	/* MS: Place handler for brki from kernel space if KGDB is OFF.
+	 * It is very unlikely that another brki instruction is called. */
+	bri 0
+
+/* MS: User-mode state save - gdb */
+1:	lwi	r1, r0, TOPHYS(PER_CPU(CURRENT_SAVE)); /* get saved current */
+	tophys(r1,r1);
+	lwi	r1, r1, TS_THREAD_INFO;	/* get the thread info */
+	addik	r1, r1, THREAD_SIZE;	/* calculate kernel stack pointer */
+	tophys(r1,r1);
+
+	addik	r1, r1, -PT_SIZE; /* Make room on the stack.  */
+	SAVE_REGS;
+	swi	r16, r1, PT_PC;	/* Save LP */
+	swi	r0, r1, PT_MODE; /* Was in user-mode.  */
+	lwi	r11, r0, TOPHYS(PER_CPU(ENTRY_SP));
+	swi	r11, r1, PT_R1; /* Store user SP.  */
+	lwi	CURRENT_TASK, r0, TOPHYS(PER_CPU(CURRENT_SAVE));
+	tovirt(r1,r1)
+	set_vms;
+	addik	r5, r1, 0;
+	addik	r15, r0, dbtrap_call;
+dbtrap_call: /* Return point for kernel/user entry + 8 because of rtsd r15, 8 */
+	rtbd	r0, sw_exception
+	nop
+
+	/* MS: The first instruction for the second part of the gdb/kgdb */
+	set_bip; /* Ints masked for state restore */
+	lwi	r11, r1, PT_MODE;
+	bnei	r11, 2f;
+/* MS: Return to user space - gdb */
+1:
+	/* Get current task ptr into r11 */
+	lwi	r11, CURRENT_TASK, TS_THREAD_INFO;	/* get thread info */
+	lwi	r19, r11, TI_FLAGS;	/* get flags in thread info */
+	andi	r11, r19, _TIF_NEED_RESCHED;
+	beqi	r11, 5f;
+
+	/* Call the scheduler before returning from a syscall/trap. */
+	bralid	r15, schedule;	/* Call scheduler */
+	nop;				/* delay slot */
+	bri	1b
+
+	/* Maybe handle a signal */
+5:	andi	r11, r19, _TIF_SIGPENDING | _TIF_NOTIFY_RESUME;
+	beqi	r11, 4f;		/* Signals to handle, handle them */
+
+	addik	r5, r1, 0;		/* Arg 1: struct pt_regs *regs */
+	bralid	r15, do_notify_resume;	/* Handle any signals */
+	addi  r6, r0, 0;	/* Arg 2: int in_syscall */
+	bri	1b
+
+/* Finally, return to user state.  */
+4:	swi	CURRENT_TASK, r0, PER_CPU(CURRENT_SAVE); /* save current */
+	VM_OFF;
+	tophys(r1,r1);
+	/* MS: Restore all regs */
+	RESTORE_REGS
+	addik	r1, r1, PT_SIZE	 /* Clean up stack space */
+	lwi	r1, r1, PT_R1 - PT_SIZE; /* Restore user stack pointer */
+DBTRAP_return_user: /* MS: Make global symbol for debugging */
+	rtbd	r16, 0; /* MS: Instructions to return from a debug trap */
+	nop;
+
+/* MS: Return to kernel state - kgdb */
+2:	VM_OFF;
+	tophys(r1,r1);
+	/* MS: Restore all regs */
+	RESTORE_REGS
+	lwi	r14, r1, PT_R14;
+	lwi	r16, r1, PT_PC;
+	addik	r1, r1, PT_SIZE; /* MS: Clean up stack space */
+	tovirt(r1,r1);
+DBTRAP_return_kernel: /* MS: Make global symbol for debugging */
+	rtbd	r16, 0; /* MS: Instructions to return from a debug trap */
+	nop;
+
+
+ENTRY(_switch_to)
+	/* prepare return value */
+	addk	r3, r0, CURRENT_TASK
+
+	/* save registers in cpu_context */
+	/* use r11 and r12, volatile registers, as temp register */
+	/* give start of cpu_context for previous process */
+	addik	r11, r5, TI_CPU_CONTEXT
+	swi	r1, r11, CC_R1
+	swi	r2, r11, CC_R2
+	/* skip volatile registers.
+	 * they are saved on stack when we jumped to _switch_to() */
+	/* dedicated registers */
+	swi	r13, r11, CC_R13
+	swi	r14, r11, CC_R14
+	swi	r15, r11, CC_R15
+	swi	r16, r11, CC_R16
+	swi	r17, r11, CC_R17
+	swi	r18, r11, CC_R18
+	/* save non-volatile registers */
+	swi	r19, r11, CC_R19
+	swi	r20, r11, CC_R20
+	swi	r21, r11, CC_R21
+	swi	r22, r11, CC_R22
+	swi	r23, r11, CC_R23
+	swi	r24, r11, CC_R24
+	swi	r25, r11, CC_R25
+	swi	r26, r11, CC_R26
+	swi	r27, r11, CC_R27
+	swi	r28, r11, CC_R28
+	swi	r29, r11, CC_R29
+	swi	r30, r11, CC_R30
+	/* special purpose registers */
+	mfs	r12, rmsr
+	swi	r12, r11, CC_MSR
+	mfs	r12, rear
+	swi	r12, r11, CC_EAR
+	mfs	r12, resr
+	swi	r12, r11, CC_ESR
+	mfs	r12, rfsr
+	swi	r12, r11, CC_FSR
+
+	/* update r31, the current-give me pointer to task which will be next */
+	lwi	CURRENT_TASK, r6, TI_TASK
+	/* stored it to current_save too */
+	swi	CURRENT_TASK, r0, PER_CPU(CURRENT_SAVE)
+
+	/* get new process' cpu context and restore */
+	/* give me start where start context of next task */
+	addik	r11, r6, TI_CPU_CONTEXT
+
+	/* non-volatile registers */
+	lwi	r30, r11, CC_R30
+	lwi	r29, r11, CC_R29
+	lwi	r28, r11, CC_R28
+	lwi	r27, r11, CC_R27
+	lwi	r26, r11, CC_R26
+	lwi	r25, r11, CC_R25
+	lwi	r24, r11, CC_R24
+	lwi	r23, r11, CC_R23
+	lwi	r22, r11, CC_R22
+	lwi	r21, r11, CC_R21
+	lwi	r20, r11, CC_R20
+	lwi	r19, r11, CC_R19
+	/* dedicated registers */
+	lwi	r18, r11, CC_R18
+	lwi	r17, r11, CC_R17
+	lwi	r16, r11, CC_R16
+	lwi	r15, r11, CC_R15
+	lwi	r14, r11, CC_R14
+	lwi	r13, r11, CC_R13
+	/* skip volatile registers */
+	lwi	r2, r11, CC_R2
+	lwi	r1, r11, CC_R1
+
+	/* special purpose registers */
+	lwi	r12, r11, CC_FSR
+	mts	rfsr, r12
+	lwi	r12, r11, CC_MSR
+	mts	rmsr, r12
+
+	rtsd	r15, 8
+	nop
+
+ENTRY(_reset)
+	brai	0; /* Jump to reset vector */
+
+	/* These are compiled and loaded into high memory, then
+	 * copied into place in mach_early_setup */
+	.section	.init.ivt, "ax"
+#if CONFIG_MANUAL_RESET_VECTOR
+	.org	0x0
+	brai	CONFIG_MANUAL_RESET_VECTOR
+#endif
+	.org	0x8
+	brai	TOPHYS(_user_exception); /* syscall handler */
+	.org	0x10
+	brai	TOPHYS(_interrupt);	/* Interrupt handler */
+	.org	0x18
+	brai	TOPHYS(_debug_exception);	/* debug trap handler */
+	.org	0x20
+	brai	TOPHYS(_hw_exception_handler);	/* HW exception handler */
+
+.section .rodata,"a"
+#include "syscall_table.S"
+
+syscall_table_size=(.-sys_call_table)
+
+type_SYSCALL:
+	.ascii "SYSCALL\0"
+type_IRQ:
+	.ascii "IRQ\0"
+type_IRQ_PREEMPT:
+	.ascii "IRQ (PREEMPTED)\0"
+type_SYSCALL_PREEMPT:
+	.ascii " SYSCALL (PREEMPTED)\0"
+
+	/*
+	 * Trap decoding for stack unwinder
+	 * Tuples are (start addr, end addr, string)
+	 * If return address lies on [start addr, end addr],
+	 * unwinder displays 'string'
+	 */
+
+	.align 4
+.global microblaze_trap_handlers
+microblaze_trap_handlers:
+	/* Exact matches come first */
+	.word ret_from_trap; .word ret_from_trap   ; .word type_SYSCALL
+	.word ret_from_irq ; .word ret_from_irq    ; .word type_IRQ
+	/* Fuzzy matches go here */
+	.word ret_from_irq ; .word no_intr_resched ; .word type_IRQ_PREEMPT
+	.word ret_from_trap; .word TRAP_return     ; .word type_SYSCALL_PREEMPT
+	/* End of table */
+	.word 0               ; .word 0               ; .word 0
diff --git a/arch/microblaze/kernel/exceptions.c b/arch/microblaze/kernel/exceptions.c
new file mode 100644
index 0000000..42dd12a
--- /dev/null
+++ b/arch/microblaze/kernel/exceptions.c
@@ -0,0 +1,154 @@
+/*
+ * HW exception handling
+ *
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008 PetaLogix
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License.  See the file COPYING in the main directory of this
+ * archive for more details.
+ */
+
+/*
+ * This file handles the architecture-dependent parts of hardware exceptions
+ */
+
+#include <linux/export.h>
+#include <linux/kernel.h>
+#include <linux/signal.h>
+#include <linux/sched.h>
+#include <linux/kallsyms.h>
+
+#include <asm/exceptions.h>
+#include <asm/entry.h>		/* For KM CPU var */
+#include <linux/uaccess.h>
+#include <linux/errno.h>
+#include <linux/ptrace.h>
+#include <asm/current.h>
+#include <asm/cacheflush.h>
+
+#define MICROBLAZE_ILL_OPCODE_EXCEPTION	0x02
+#define MICROBLAZE_IBUS_EXCEPTION	0x03
+#define MICROBLAZE_DBUS_EXCEPTION	0x04
+#define MICROBLAZE_DIV_ZERO_EXCEPTION	0x05
+#define MICROBLAZE_FPU_EXCEPTION	0x06
+#define MICROBLAZE_PRIVILEGED_EXCEPTION	0x07
+
+static DEFINE_SPINLOCK(die_lock);
+
+void die(const char *str, struct pt_regs *fp, long err)
+{
+	console_verbose();
+	spin_lock_irq(&die_lock);
+	pr_warn("Oops: %s, sig: %ld\n", str, err);
+	show_regs(fp);
+	spin_unlock_irq(&die_lock);
+	/* do_exit() should take care of panic'ing from an interrupt
+	 * context so we don't handle it here
+	 */
+	do_exit(err);
+}
+
+/* for user application debugging */
+asmlinkage void sw_exception(struct pt_regs *regs)
+{
+	_exception(SIGTRAP, regs, TRAP_BRKPT, regs->r16);
+	flush_dcache_range(regs->r16, regs->r16 + 0x4);
+	flush_icache_range(regs->r16, regs->r16 + 0x4);
+}
+
+void _exception(int signr, struct pt_regs *regs, int code, unsigned long addr)
+{
+	siginfo_t info;
+
+	if (kernel_mode(regs))
+		die("Exception in kernel mode", regs, signr);
+
+	info.si_signo = signr;
+	info.si_errno = 0;
+	info.si_code = code;
+	info.si_addr = (void __user *) addr;
+	force_sig_info(signr, &info, current);
+}
+
+asmlinkage void full_exception(struct pt_regs *regs, unsigned int type,
+							int fsr, int addr)
+{
+#ifdef CONFIG_MMU
+	addr = regs->pc;
+#endif
+
+#if 0
+	pr_warn("Exception %02x in %s mode, FSR=%08x PC=%08x ESR=%08x\n",
+			type, user_mode(regs) ? "user" : "kernel", fsr,
+			(unsigned int) regs->pc, (unsigned int) regs->esr);
+#endif
+
+	switch (type & 0x1F) {
+	case MICROBLAZE_ILL_OPCODE_EXCEPTION:
+		if (user_mode(regs)) {
+			pr_debug("Illegal opcode exception in user mode\n");
+			_exception(SIGILL, regs, ILL_ILLOPC, addr);
+			return;
+		}
+		pr_warn("Illegal opcode exception in kernel mode.\n");
+		die("opcode exception", regs, SIGBUS);
+		break;
+	case MICROBLAZE_IBUS_EXCEPTION:
+		if (user_mode(regs)) {
+			pr_debug("Instruction bus error exception in user mode\n");
+			_exception(SIGBUS, regs, BUS_ADRERR, addr);
+			return;
+		}
+		pr_warn("Instruction bus error exception in kernel mode.\n");
+		die("bus exception", regs, SIGBUS);
+		break;
+	case MICROBLAZE_DBUS_EXCEPTION:
+		if (user_mode(regs)) {
+			pr_debug("Data bus error exception in user mode\n");
+			_exception(SIGBUS, regs, BUS_ADRERR, addr);
+			return;
+		}
+		pr_warn("Data bus error exception in kernel mode.\n");
+		die("bus exception", regs, SIGBUS);
+		break;
+	case MICROBLAZE_DIV_ZERO_EXCEPTION:
+		if (user_mode(regs)) {
+			pr_debug("Divide by zero exception in user mode\n");
+			_exception(SIGFPE, regs, FPE_INTDIV, addr);
+			return;
+		}
+		pr_warn("Divide by zero exception in kernel mode.\n");
+		die("Divide by zero exception", regs, SIGBUS);
+		break;
+	case MICROBLAZE_FPU_EXCEPTION:
+		pr_debug("FPU exception\n");
+		/* IEEE FP exception */
+		/* I removed fsr variable and use code var for storing fsr */
+		if (fsr & FSR_IO)
+			fsr = FPE_FLTINV;
+		else if (fsr & FSR_OF)
+			fsr = FPE_FLTOVF;
+		else if (fsr & FSR_UF)
+			fsr = FPE_FLTUND;
+		else if (fsr & FSR_DZ)
+			fsr = FPE_FLTDIV;
+		else if (fsr & FSR_DO)
+			fsr = FPE_FLTRES;
+		_exception(SIGFPE, regs, fsr, addr);
+		break;
+
+#ifdef CONFIG_MMU
+	case MICROBLAZE_PRIVILEGED_EXCEPTION:
+		pr_debug("Privileged exception\n");
+		_exception(SIGILL, regs, ILL_PRVOPC, addr);
+		break;
+#endif
+	default:
+	/* FIXME what to do in unexpected exception */
+		pr_warn("Unexpected exception %02x PC=%08x in %s mode\n",
+			type, (unsigned int) addr,
+			kernel_mode(regs) ? "kernel" : "user");
+	}
+	return;
+}
diff --git a/arch/microblaze/kernel/ftrace.c b/arch/microblaze/kernel/ftrace.c
new file mode 100644
index 0000000..fc7b48a
--- /dev/null
+++ b/arch/microblaze/kernel/ftrace.c
@@ -0,0 +1,233 @@
+/*
+ * Ftrace support for Microblaze.
+ *
+ * Copyright (C) 2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2009 PetaLogix
+ *
+ * Based on MIPS and PowerPC ftrace code
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <asm/cacheflush.h>
+#include <linux/ftrace.h>
+
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+/*
+ * Hook the return address and push it in the stack of return addrs
+ * in current thread info.
+ */
+void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr)
+{
+	unsigned long old;
+	int faulted, err;
+	struct ftrace_graph_ent trace;
+	unsigned long return_hooker = (unsigned long)
+				&return_to_handler;
+
+	if (unlikely(ftrace_graph_is_dead()))
+		return;
+
+	if (unlikely(atomic_read(&current->tracing_graph_pause)))
+		return;
+
+	/*
+	 * Protect against fault, even if it shouldn't
+	 * happen. This tool is too much intrusive to
+	 * ignore such a protection.
+	 */
+	asm volatile("	1:	lwi	%0, %2, 0;"		\
+			"2:	swi	%3, %2, 0;"		\
+			"	addik	%1, r0, 0;"		\
+			"3:"					\
+			"	.section .fixup, \"ax\";"	\
+			"4:	brid	3b;"			\
+			"	addik	%1, r0, 1;"		\
+			"	.previous;"			\
+			"	.section __ex_table,\"a\";"	\
+			"	.word	1b,4b;"			\
+			"	.word	2b,4b;"			\
+			"	.previous;"			\
+			: "=&r" (old), "=r" (faulted)
+			: "r" (parent), "r" (return_hooker)
+	);
+
+	flush_dcache_range((u32)parent, (u32)parent + 4);
+	flush_icache_range((u32)parent, (u32)parent + 4);
+
+	if (unlikely(faulted)) {
+		ftrace_graph_stop();
+		WARN_ON(1);
+		return;
+	}
+
+	err = ftrace_push_return_trace(old, self_addr, &trace.depth, 0);
+	if (err == -EBUSY) {
+		*parent = old;
+		return;
+	}
+
+	trace.func = self_addr;
+	/* Only trace if the calling function expects to */
+	if (!ftrace_graph_entry(&trace)) {
+		current->curr_ret_stack--;
+		*parent = old;
+	}
+}
+#endif /* CONFIG_FUNCTION_GRAPH_TRACER */
+
+#ifdef CONFIG_DYNAMIC_FTRACE
+/* save value to addr - it is save to do it in asm */
+static int ftrace_modify_code(unsigned long addr, unsigned int value)
+{
+	int faulted = 0;
+
+	__asm__ __volatile__("	1:	swi	%2, %1, 0;"		\
+				"	addik	%0, r0, 0;"		\
+				"2:"					\
+				"	.section .fixup, \"ax\";"	\
+				"3:	brid	2b;"			\
+				"	addik	%0, r0, 1;"		\
+				"	.previous;"			\
+				"	.section __ex_table,\"a\";"	\
+				"	.word	1b,3b;"			\
+				"	.previous;"			\
+				: "=r" (faulted)
+				: "r" (addr), "r" (value)
+	);
+
+	if (unlikely(faulted))
+		return -EFAULT;
+
+	flush_dcache_range(addr, addr + 4);
+	flush_icache_range(addr, addr + 4);
+
+	return 0;
+}
+
+#define MICROBLAZE_NOP 0x80000000
+#define MICROBLAZE_BRI 0xb800000C
+
+static unsigned int recorded; /* if save was or not */
+static unsigned int imm; /* saving whole imm instruction */
+
+/* There are two approaches howto solve ftrace_make nop function - look below */
+#undef USE_FTRACE_NOP
+
+#ifdef USE_FTRACE_NOP
+static unsigned int bralid; /* saving whole bralid instruction */
+#endif
+
+int ftrace_make_nop(struct module *mod,
+			struct dyn_ftrace *rec, unsigned long addr)
+{
+	/* we have this part of code which we are working with
+	 * b000c000        imm     -16384
+	 * b9fc8e30        bralid  r15, -29136     // c0008e30 <_mcount>
+	 * 80000000        or      r0, r0, r0
+	 *
+	 * The first solution (!USE_FTRACE_NOP-could be called branch solution)
+	 * b000c000        bri	12 (0xC - jump to any other instruction)
+	 * b9fc8e30        bralid  r15, -29136     // c0008e30 <_mcount>
+	 * 80000000        or      r0, r0, r0
+	 * any other instruction
+	 *
+	 * The second solution (USE_FTRACE_NOP) - no jump just nops
+	 * 80000000        or      r0, r0, r0
+	 * 80000000        or      r0, r0, r0
+	 * 80000000        or      r0, r0, r0
+	 */
+	int ret = 0;
+
+	if (recorded == 0) {
+		recorded = 1;
+		imm = *(unsigned int *)rec->ip;
+		pr_debug("%s: imm:0x%x\n", __func__, imm);
+#ifdef USE_FTRACE_NOP
+		bralid = *(unsigned int *)(rec->ip + 4);
+		pr_debug("%s: bralid 0x%x\n", __func__, bralid);
+#endif /* USE_FTRACE_NOP */
+	}
+
+#ifdef USE_FTRACE_NOP
+	ret = ftrace_modify_code(rec->ip, MICROBLAZE_NOP);
+	ret += ftrace_modify_code(rec->ip + 4, MICROBLAZE_NOP);
+#else /* USE_FTRACE_NOP */
+	ret = ftrace_modify_code(rec->ip, MICROBLAZE_BRI);
+#endif /* USE_FTRACE_NOP */
+	return ret;
+}
+
+/* I believe that first is called ftrace_make_nop before this function */
+int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr)
+{
+	int ret;
+	pr_debug("%s: addr:0x%x, rec->ip: 0x%x, imm:0x%x\n",
+		__func__, (unsigned int)addr, (unsigned int)rec->ip, imm);
+	ret = ftrace_modify_code(rec->ip, imm);
+#ifdef USE_FTRACE_NOP
+	pr_debug("%s: bralid:0x%x\n", __func__, bralid);
+	ret += ftrace_modify_code(rec->ip + 4, bralid);
+#endif /* USE_FTRACE_NOP */
+	return ret;
+}
+
+int __init ftrace_dyn_arch_init(void)
+{
+	return 0;
+}
+
+int ftrace_update_ftrace_func(ftrace_func_t func)
+{
+	unsigned long ip = (unsigned long)(&ftrace_call);
+	unsigned int upper = (unsigned int)func;
+	unsigned int lower = (unsigned int)func;
+	int ret = 0;
+
+	/* create proper saving to ftrace_call poll */
+	upper = 0xb0000000 + (upper >> 16); /* imm func_upper */
+	lower = 0x32800000 + (lower & 0xFFFF); /* addik r20, r0, func_lower */
+
+	pr_debug("%s: func=0x%x, ip=0x%x, upper=0x%x, lower=0x%x\n",
+		__func__, (unsigned int)func, (unsigned int)ip, upper, lower);
+
+	/* save upper and lower code */
+	ret = ftrace_modify_code(ip, upper);
+	ret += ftrace_modify_code(ip + 4, lower);
+
+	/* We just need to replace the rtsd r15, 8 with NOP */
+	ret += ftrace_modify_code((unsigned long)&ftrace_caller,
+				  MICROBLAZE_NOP);
+
+	return ret;
+}
+
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+unsigned int old_jump; /* saving place for jump instruction */
+
+int ftrace_enable_ftrace_graph_caller(void)
+{
+	unsigned int ret;
+	unsigned long ip = (unsigned long)(&ftrace_call_graph);
+
+	old_jump = *(unsigned int *)ip; /* save jump over instruction */
+	ret = ftrace_modify_code(ip, MICROBLAZE_NOP);
+
+	pr_debug("%s: Replace instruction: 0x%x\n", __func__, old_jump);
+	return ret;
+}
+
+int ftrace_disable_ftrace_graph_caller(void)
+{
+	unsigned int ret;
+	unsigned long ip = (unsigned long)(&ftrace_call_graph);
+
+	ret = ftrace_modify_code(ip, old_jump);
+
+	pr_debug("%s\n", __func__);
+	return ret;
+}
+#endif /* CONFIG_FUNCTION_GRAPH_TRACER */
+#endif /* CONFIG_DYNAMIC_FTRACE */
diff --git a/arch/microblaze/kernel/head.S b/arch/microblaze/kernel/head.S
new file mode 100644
index 0000000..4655ff3
--- /dev/null
+++ b/arch/microblaze/kernel/head.S
@@ -0,0 +1,391 @@
+/*
+ * Copyright (C) 2007-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ *
+ * MMU code derived from arch/ppc/kernel/head_4xx.S:
+ *    Copyright (c) 1995-1996 Gary Thomas <gdt@linuxppc.org>
+ *      Initial PowerPC version.
+ *    Copyright (c) 1996 Cort Dougan <cort@cs.nmt.edu>
+ *      Rewritten for PReP
+ *    Copyright (c) 1996 Paul Mackerras <paulus@cs.anu.edu.au>
+ *      Low-level exception handers, MMU support, and rewrite.
+ *    Copyright (c) 1997 Dan Malek <dmalek@jlc.net>
+ *      PowerPC 8xx modifications.
+ *    Copyright (c) 1998-1999 TiVo, Inc.
+ *      PowerPC 403GCX modifications.
+ *    Copyright (c) 1999 Grant Erickson <grant@lcse.umn.edu>
+ *      PowerPC 403GCX/405GP modifications.
+ *    Copyright 2000 MontaVista Software Inc.
+ *	PPC405 modifications
+ *      PowerPC 403GCX/405GP modifications.
+ * 	Author: MontaVista Software, Inc.
+ *         	frank_rowand@mvista.com or source@mvista.com
+ * 	   	debbie_chu@mvista.com
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/init.h>
+#include <linux/linkage.h>
+#include <asm/thread_info.h>
+#include <asm/page.h>
+#include <linux/of_fdt.h>		/* for OF_DT_HEADER */
+
+#ifdef CONFIG_MMU
+#include <asm/setup.h> /* COMMAND_LINE_SIZE */
+#include <asm/mmu.h>
+#include <asm/processor.h>
+
+.section .data
+.global empty_zero_page
+.align 12
+empty_zero_page:
+	.space	PAGE_SIZE
+.global swapper_pg_dir
+swapper_pg_dir:
+	.space	PAGE_SIZE
+
+#endif /* CONFIG_MMU */
+
+.section .rodata
+.align 4
+endian_check:
+	.word	1
+
+	__HEAD
+ENTRY(_start)
+#if CONFIG_KERNEL_BASE_ADDR == 0
+	brai	TOPHYS(real_start)
+	.org	0x100
+real_start:
+#endif
+
+	mts	rmsr, r0
+/* Disable stack protection from bootloader */
+	mts	rslr, r0
+	addi	r8, r0, 0xFFFFFFFF
+	mts	rshr, r8
+/*
+ * According to Xilinx, msrclr instruction behaves like 'mfs rX,rpc'
+ * if the msrclr instruction is not enabled. We use this to detect
+ * if the opcode is available, by issuing msrclr and then testing the result.
+ * r8 == 0 - msr instructions are implemented
+ * r8 != 0 - msr instructions are not implemented
+ */
+	mfs	r1, rmsr
+	msrclr	r8, 0 /* clear nothing - just read msr for test */
+	cmpu	r8, r8, r1 /* r1 must contain msr reg content */
+
+/* r7 may point to an FDT, or there may be one linked in.
+   if it's in r7, we've got to save it away ASAP.
+   We ensure r7 points to a valid FDT, just in case the bootloader
+   is broken or non-existent */
+	beqi	r7, no_fdt_arg			/* NULL pointer?  don't copy */
+/* Does r7 point to a valid FDT? Load HEADER magic number */
+	/* Run time Big/Little endian platform */
+	/* Save 1 as word and load byte - 0 - BIG, 1 - LITTLE */
+	lbui	r11, r0, TOPHYS(endian_check)
+	beqid	r11, big_endian /* DO NOT break delay stop dependency */
+	lw	r11, r0, r7 /* Big endian load in delay slot */
+	lwr	r11, r0, r7 /* Little endian load */
+big_endian:
+	rsubi	r11, r11, OF_DT_HEADER	/* Check FDT header */
+	beqi	r11, _prepare_copy_fdt
+	or	r7, r0, r0		/* clear R7 when not valid DTB */
+	bnei	r11, no_fdt_arg			/* No - get out of here */
+_prepare_copy_fdt:
+	or	r11, r0, r0 /* incremment */
+	ori	r4, r0, TOPHYS(_fdt_start)
+	ori	r3, r0, (0x8000 - 4)
+_copy_fdt:
+	lw	r12, r7, r11 /* r12 = r7 + r11 */
+	sw	r12, r4, r11 /* addr[r4 + r11] = r12 */
+	addik	r11, r11, 4 /* increment counting */
+	bgtid	r3, _copy_fdt /* loop for all entries */
+	addik	r3, r3, -4 /* descrement loop */
+no_fdt_arg:
+
+#ifdef CONFIG_MMU
+
+#ifndef CONFIG_CMDLINE_BOOL
+/*
+ * handling command line
+ * copy command line directly to cmd_line placed in data section.
+ */
+	beqid	r5, skip	/* Skip if NULL pointer */
+	or	r11, r0, r0		/* incremment */
+	ori	r4, r0, cmd_line	/* load address of command line */
+	tophys(r4,r4)			/* convert to phys address */
+	ori	r3, r0, COMMAND_LINE_SIZE - 1 /* number of loops */
+_copy_command_line:
+	/* r2=r5+r6 - r5 contain pointer to command line */
+	lbu	r2, r5, r11
+	beqid	r2, skip		/* Skip if no data */
+	sb	r2, r4, r11		/* addr[r4+r6]= r2 */
+	addik	r11, r11, 1		/* increment counting */
+	bgtid	r3, _copy_command_line	/* loop for all entries       */
+	addik	r3, r3, -1		/* decrement loop */
+	addik	r5, r4, 0		/* add new space for command line */
+	tovirt(r5,r5)
+skip:
+#endif /* CONFIG_CMDLINE_BOOL */
+
+#ifdef NOT_COMPILE
+/* save bram context */
+	or	r11, r0, r0				/* incremment */
+	ori	r4, r0, TOPHYS(_bram_load_start)	/* save bram context */
+	ori	r3, r0, (LMB_SIZE - 4)
+_copy_bram:
+	lw	r7, r0, r11		/* r7 = r0 + r6 */
+	sw	r7, r4, r11		/* addr[r4 + r6] = r7 */
+	addik	r11, r11, 4		/* increment counting */
+	bgtid	r3, _copy_bram		/* loop for all entries */
+	addik	r3, r3, -4		/* descrement loop */
+#endif
+	/* We have to turn on the MMU right away. */
+
+	/*
+	 * Set up the initial MMU state so we can do the first level of
+	 * kernel initialization.  This maps the first 16 MBytes of memory 1:1
+	 * virtual to physical.
+	 */
+	nop
+	addik	r3, r0, MICROBLAZE_TLB_SIZE -1	/* Invalidate all TLB entries */
+_invalidate:
+	mts	rtlbx, r3
+	mts	rtlbhi, r0			/* flush: ensure V is clear   */
+	mts	rtlblo, r0
+	bgtid	r3, _invalidate		/* loop for all entries       */
+	addik	r3, r3, -1
+	/* sync */
+
+	/* Setup the kernel PID */
+	mts	rpid,r0			/* Load the kernel PID */
+	nop
+	bri	4
+
+	/*
+	 * We should still be executing code at physical address area
+	 * RAM_BASEADDR at this point. However, kernel code is at
+	 * a virtual address. So, set up a TLB mapping to cover this once
+	 * translation is enabled.
+	 */
+
+	addik	r3,r0, CONFIG_KERNEL_START /* Load the kernel virtual address */
+	tophys(r4,r3)			/* Load the kernel physical address */
+
+	/* start to do TLB calculation */
+	addik	r12, r0, _end
+	rsub	r12, r3, r12
+	addik	r12, r12, CONFIG_LOWMEM_SIZE >> PTE_SHIFT /* that's the pad */
+
+	or r9, r0, r0 /* TLB0 = 0 */
+	or r10, r0, r0 /* TLB1 = 0 */
+
+	addik	r11, r12, -0x1000000
+	bgei	r11, GT16 /* size is greater than 16MB */
+	addik	r11, r12, -0x0800000
+	bgei	r11, GT8 /* size is greater than 8MB */
+	addik	r11, r12, -0x0400000
+	bgei	r11, GT4 /* size is greater than 4MB */
+	/* size is less than 4MB */
+	addik	r11, r12, -0x0200000
+	bgei	r11, GT2 /* size is greater than 2MB */
+	addik	r9, r0, 0x0100000 /* TLB0 must be 1MB */
+	addik	r11, r12, -0x0100000
+	bgei	r11, GT1 /* size is greater than 1MB */
+	/* TLB1 is 0 which is setup above */
+	bri tlb_end
+GT4: /* r11 contains the rest - will be either 1 or 4 */
+	ori r9, r0, 0x400000 /* TLB0 is 4MB */
+	bri TLB1
+GT16: /* TLB0 is 16MB */
+	addik	r9, r0, 0x1000000 /* means TLB0 is 16MB */
+TLB1:
+	/* must be used r2 because of subtract if failed */
+	addik	r2, r11, -0x0400000
+	bgei	r2, GT20 /* size is greater than 16MB */
+	/* size is >16MB and <20MB */
+	addik	r11, r11, -0x0100000
+	bgei	r11, GT17 /* size is greater than 17MB */
+	/* kernel is >16MB and < 17MB */
+GT1:
+	addik	r10, r0, 0x0100000 /* means TLB1 is 1MB */
+	bri tlb_end
+GT2: /* TLB0 is 0 and TLB1 will be 4MB */
+GT17: /* TLB1 is 4MB - kernel size <20MB */
+	addik	r10, r0, 0x0400000 /* means TLB1 is 4MB */
+	bri tlb_end
+GT8: /* TLB0 is still zero that's why I can use only TLB1 */
+GT20: /* TLB1 is 16MB - kernel size >20MB */
+	addik	r10, r0, 0x1000000 /* means TLB1 is 16MB */
+tlb_end:
+
+	/*
+	 * Configure and load two entries into TLB slots 0 and 1.
+	 * In case we are pinning TLBs, these are reserved in by the
+	 * other TLB functions.  If not reserving, then it doesn't
+	 * matter where they are loaded.
+	 */
+	andi	r4,r4,0xfffffc00	/* Mask off the real page number */
+	ori	r4,r4,(TLB_WR | TLB_EX)	/* Set the write and execute bits */
+
+	/*
+	 * TLB0 is always used - check if is not zero (r9 stores TLB0 value)
+	 * if is use TLB1 value and clear it (r10 stores TLB1 value)
+	 */
+	bnei	r9, tlb0_not_zero
+	add	r9, r10, r0
+	add	r10, r0, r0
+tlb0_not_zero:
+
+	/* look at the code below */
+	ori	r30, r0, 0x200
+	andi	r29, r9, 0x100000
+	bneid	r29, 1f
+	addik	r30, r30, 0x80
+	andi	r29, r9, 0x400000
+	bneid	r29, 1f
+	addik	r30, r30, 0x80
+	andi	r29, r9, 0x1000000
+	bneid	r29, 1f
+	addik	r30, r30, 0x80
+1:
+	andi	r3,r3,0xfffffc00	/* Mask off the effective page number */
+	ori	r3,r3,(TLB_VALID)
+	or	r3, r3, r30
+
+	/* Load tlb_skip size value which is index to first unused TLB entry */
+	lwi	r11, r0, TOPHYS(tlb_skip)
+	mts     rtlbx,r11		/* TLB slow 0 */
+
+	mts	rtlblo,r4		/* Load the data portion of the entry */
+	mts	rtlbhi,r3		/* Load the tag portion of the entry */
+
+	/* Increase tlb_skip size */
+	addik	r11, r11, 1
+	swi	r11, r0, TOPHYS(tlb_skip)
+
+	/* TLB1 can be zeroes that's why we not setup it */
+	beqi	r10, jump_over2
+
+	/* look at the code below */
+	ori	r30, r0, 0x200
+	andi	r29, r10, 0x100000
+	bneid	r29, 1f
+	addik	r30, r30, 0x80
+	andi	r29, r10, 0x400000
+	bneid	r29, 1f
+	addik	r30, r30, 0x80
+	andi	r29, r10, 0x1000000
+	bneid	r29, 1f
+	addik	r30, r30, 0x80
+1:
+	addk	r4, r4, r9	/* previous addr + TLB0 size */
+	addk	r3, r3, r9
+
+	andi	r3,r3,0xfffffc00	/* Mask off the effective page number */
+	ori	r3,r3,(TLB_VALID)
+	or	r3, r3, r30
+
+	lwi	r11, r0, TOPHYS(tlb_skip)
+	mts     rtlbx, r11		/* r11 is used from TLB0 */
+
+	mts	rtlblo,r4		/* Load the data portion of the entry */
+	mts	rtlbhi,r3		/* Load the tag portion of the entry */
+
+	/* Increase tlb_skip size */
+	addik	r11, r11, 1
+	swi	r11, r0, TOPHYS(tlb_skip)
+
+jump_over2:
+	/*
+	 * Load a TLB entry for LMB, since we need access to
+	 * the exception vectors, using a 4k real==virtual mapping.
+	 */
+	/* Use temporary TLB_ID for LMB - clear this temporary mapping later */
+	ori	r11, r0, MICROBLAZE_LMB_TLB_ID
+	mts     rtlbx,r11
+
+	ori	r4,r0,(TLB_WR | TLB_EX)
+	ori	r3,r0,(TLB_VALID | TLB_PAGESZ(PAGESZ_4K))
+
+	mts	rtlblo,r4		/* Load the data portion of the entry */
+	mts	rtlbhi,r3		/* Load the tag portion of the entry */
+
+	/*
+	 * We now have the lower 16 Meg of RAM mapped into TLB entries, and the
+	 * caches ready to work.
+	 */
+turn_on_mmu:
+	ori	r15,r0,start_here
+	ori	r4,r0,MSR_KERNEL_VMS
+	mts	rmsr,r4
+	nop
+	rted	r15,0			/* enables MMU */
+	nop
+
+start_here:
+#endif /* CONFIG_MMU */
+
+	/* Initialize small data anchors */
+	addik	r13, r0, _KERNEL_SDA_BASE_
+	addik	r2, r0, _KERNEL_SDA2_BASE_
+
+	/* Initialize stack pointer */
+	addik	r1, r0, init_thread_union + THREAD_SIZE - 4
+
+	/* Initialize r31 with current task address */
+	addik	r31, r0, init_task
+
+	/*
+	 * Call platform dependent initialize function.
+	 * Please see $(ARCH)/mach-$(SUBARCH)/setup.c for
+	 * the function.
+	 */
+	addik	r11, r0, machine_early_init
+	brald	r15, r11
+	nop
+
+#ifndef CONFIG_MMU
+	addik	r15, r0, machine_halt
+	braid	start_kernel
+	nop
+#else
+	/*
+	 * Initialize the MMU.
+	 */
+	bralid	r15, mmu_init
+	nop
+
+	/* Go back to running unmapped so we can load up new values
+	 * and change to using our exception vectors.
+	 * On the MicroBlaze, all we invalidate the used TLB entries to clear
+	 * the old 16M byte TLB mappings.
+	 */
+	ori	r15,r0,TOPHYS(kernel_load_context)
+	ori	r4,r0,MSR_KERNEL
+	mts	rmsr,r4
+	nop
+	bri	4
+	rted	r15,0
+	nop
+
+	/* Load up the kernel context */
+kernel_load_context:
+	ori	r5, r0, MICROBLAZE_LMB_TLB_ID
+	mts     rtlbx,r5
+	nop
+	mts	rtlbhi,r0
+	nop
+	addi	r15, r0, machine_halt
+	ori	r17, r0, start_kernel
+	ori	r4, r0, MSR_KERNEL_VMS
+	mts	rmsr, r4
+	nop
+	rted	r17, 0		/* enable MMU and jump to start_kernel */
+	nop
+#endif /* CONFIG_MMU */
diff --git a/arch/microblaze/kernel/heartbeat.c b/arch/microblaze/kernel/heartbeat.c
new file mode 100644
index 0000000..4643e3a
--- /dev/null
+++ b/arch/microblaze/kernel/heartbeat.c
@@ -0,0 +1,71 @@
+/*
+ * Copyright (C) 2007-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/sched.h>
+#include <linux/io.h>
+
+#include <asm/setup.h>
+#include <asm/page.h>
+#include <asm/prom.h>
+
+static unsigned int base_addr;
+
+void microblaze_heartbeat(void)
+{
+	static unsigned int cnt, period, dist;
+
+	if (base_addr) {
+		if (cnt == 0 || cnt == dist)
+			out_be32(base_addr, 1);
+		else if (cnt == 7 || cnt == dist + 7)
+			out_be32(base_addr, 0);
+
+		if (++cnt > period) {
+			cnt = 0;
+			/*
+			 * The hyperbolic function below modifies the heartbeat
+			 * period length in dependency of the current (5min)
+			 * load. It goes through the points f(0)=126, f(1)=86,
+			 * f(5)=51, f(inf)->30.
+			 */
+			period = ((672 << FSHIFT) / (5 * avenrun[0] +
+						(7 << FSHIFT))) + 30;
+			dist = period / 4;
+		}
+	}
+}
+
+void microblaze_setup_heartbeat(void)
+{
+	struct device_node *gpio = NULL;
+	int *prop;
+	int j;
+	const char * const gpio_list[] = {
+		"xlnx,xps-gpio-1.00.a",
+		NULL
+	};
+
+	for (j = 0; gpio_list[j] != NULL; j++) {
+		gpio = of_find_compatible_node(NULL, NULL, gpio_list[j]);
+		if (gpio)
+			break;
+	}
+
+	if (gpio) {
+		base_addr = be32_to_cpup(of_get_property(gpio, "reg", NULL));
+		base_addr = (unsigned long) ioremap(base_addr, PAGE_SIZE);
+		pr_notice("Heartbeat GPIO at 0x%x\n", base_addr);
+
+		/* GPIO is configured as output */
+		prop = (int *) of_get_property(gpio, "xlnx,is-bidir", NULL);
+		if (prop)
+			out_be32(base_addr + 4, 0);
+	}
+}
diff --git a/arch/microblaze/kernel/hw_exception_handler.S b/arch/microblaze/kernel/hw_exception_handler.S
new file mode 100644
index 0000000..0b11a44
--- /dev/null
+++ b/arch/microblaze/kernel/hw_exception_handler.S
@@ -0,0 +1,1225 @@
+/*
+ * Exception handling for Microblaze
+ *
+ * Rewriten interrupt handling
+ *
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ *
+ * uClinux customisation (C) 2005 John Williams
+ *
+ * MMU code derived from arch/ppc/kernel/head_4xx.S:
+ *	Copyright (C) 1995-1996 Gary Thomas <gdt@linuxppc.org>
+ *		Initial PowerPC version.
+ *	Copyright (C) 1996 Cort Dougan <cort@cs.nmt.edu>
+ *		Rewritten for PReP
+ *	Copyright (C) 1996 Paul Mackerras <paulus@cs.anu.edu.au>
+ *		Low-level exception handers, MMU support, and rewrite.
+ *	Copyright (C) 1997 Dan Malek <dmalek@jlc.net>
+ *		PowerPC 8xx modifications.
+ *	Copyright (C) 1998-1999 TiVo, Inc.
+ *		PowerPC 403GCX modifications.
+ *	Copyright (C) 1999 Grant Erickson <grant@lcse.umn.edu>
+ *		PowerPC 403GCX/405GP modifications.
+ *	Copyright 2000 MontaVista Software Inc.
+ *		PPC405 modifications
+ *	PowerPC 403GCX/405GP modifications.
+ *		Author: MontaVista Software, Inc.
+ *		frank_rowand@mvista.com or source@mvista.com
+ *		debbie_chu@mvista.com
+ *
+ * Original code
+ * Copyright (C) 2004 Xilinx, Inc.
+ *
+ * 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.
+ */
+
+/*
+ * Here are the handlers which don't require enabling translation
+ * and calling other kernel code thus we can keep their design very simple
+ * and do all processing in real mode. All what they need is a valid current
+ * (that is an issue for the CONFIG_REGISTER_TASK_PTR case)
+ * This handlers use r3,r4,r5,r6 and optionally r[current] to work therefore
+ * these registers are saved/restored
+ * The handlers which require translation are in entry.S --KAA
+ *
+ * Microblaze HW Exception Handler
+ * - Non self-modifying exception handler for the following exception conditions
+ *   - Unalignment
+ *   - Instruction bus error
+ *   - Data bus error
+ *   - Illegal instruction opcode
+ *   - Divide-by-zero
+ *
+ *   - Privileged instruction exception (MMU)
+ *   - Data storage exception (MMU)
+ *   - Instruction storage exception (MMU)
+ *   - Data TLB miss exception (MMU)
+ *   - Instruction TLB miss exception (MMU)
+ *
+ * Note we disable interrupts during exception handling, otherwise we will
+ * possibly get multiple re-entrancy if interrupt handles themselves cause
+ * exceptions. JW
+ */
+
+#include <asm/exceptions.h>
+#include <asm/unistd.h>
+#include <asm/page.h>
+
+#include <asm/entry.h>
+#include <asm/current.h>
+#include <linux/linkage.h>
+
+#include <asm/mmu.h>
+#include <asm/pgtable.h>
+#include <asm/signal.h>
+#include <asm/registers.h>
+#include <asm/asm-offsets.h>
+
+#undef DEBUG
+
+/* Helpful Macros */
+#define NUM_TO_REG(num)		r ## num
+
+#ifdef CONFIG_MMU
+	#define RESTORE_STATE			\
+		lwi	r5, r1, 0;		\
+		mts	rmsr, r5;		\
+		nop;				\
+		lwi	r3, r1, PT_R3;		\
+		lwi	r4, r1, PT_R4;		\
+		lwi	r5, r1, PT_R5;		\
+		lwi	r6, r1, PT_R6;		\
+		lwi	r11, r1, PT_R11;	\
+		lwi	r31, r1, PT_R31;	\
+		lwi	r1, r1, PT_R1;
+#endif /* CONFIG_MMU */
+
+#define LWREG_NOP			\
+	bri	ex_handler_unhandled;	\
+	nop;
+
+#define SWREG_NOP			\
+	bri	ex_handler_unhandled;	\
+	nop;
+
+/* FIXME this is weird - for noMMU kernel is not possible to use brid
+ * instruction which can shorten executed time
+ */
+
+/* r3 is the source */
+#define R3_TO_LWREG_V(regnum)				\
+	swi	r3, r1, 4 * regnum;				\
+	bri	ex_handler_done;
+
+/* r3 is the source */
+#define R3_TO_LWREG(regnum)				\
+	or	NUM_TO_REG (regnum), r0, r3;		\
+	bri	ex_handler_done;
+
+/* r3 is the target */
+#define SWREG_TO_R3_V(regnum)				\
+	lwi	r3, r1, 4 * regnum;				\
+	bri	ex_sw_tail;
+
+/* r3 is the target */
+#define SWREG_TO_R3(regnum)				\
+	or	r3, r0, NUM_TO_REG (regnum);		\
+	bri	ex_sw_tail;
+
+#ifdef CONFIG_MMU
+	#define R3_TO_LWREG_VM_V(regnum)		\
+		brid	ex_lw_end_vm;			\
+		swi	r3, r7, 4 * regnum;
+
+	#define R3_TO_LWREG_VM(regnum)			\
+		brid	ex_lw_end_vm;			\
+		or	NUM_TO_REG (regnum), r0, r3;
+
+	#define SWREG_TO_R3_VM_V(regnum)		\
+		brid	ex_sw_tail_vm;			\
+		lwi	r3, r7, 4 * regnum;
+
+	#define SWREG_TO_R3_VM(regnum)			\
+		brid	ex_sw_tail_vm;			\
+		or	r3, r0, NUM_TO_REG (regnum);
+
+	/* Shift right instruction depending on available configuration */
+	#if CONFIG_XILINX_MICROBLAZE0_USE_BARREL == 0
+	/* Only the used shift constants defined here - add more if needed */
+	#define BSRLI2(rD, rA)				\
+		srl rD, rA;		/* << 1 */	\
+		srl rD, rD;		/* << 2 */
+	#define BSRLI4(rD, rA)		\
+		BSRLI2(rD, rA);		\
+		BSRLI2(rD, rD)
+	#define BSRLI10(rD, rA)				\
+		srl rD, rA;		/* << 1 */	\
+		srl rD, rD;		/* << 2 */	\
+		srl rD, rD;		/* << 3 */	\
+		srl rD, rD;		/* << 4 */	\
+		srl rD, rD;		/* << 5 */	\
+		srl rD, rD;		/* << 6 */	\
+		srl rD, rD;		/* << 7 */	\
+		srl rD, rD;		/* << 8 */	\
+		srl rD, rD;		/* << 9 */	\
+		srl rD, rD		/* << 10 */
+	#define BSRLI20(rD, rA)		\
+		BSRLI10(rD, rA);	\
+		BSRLI10(rD, rD)
+
+	.macro	bsrli, rD, rA, IMM
+	.if (\IMM) == 2
+		BSRLI2(\rD, \rA)
+	.elseif (\IMM) == 10
+		BSRLI10(\rD, \rA)
+	.elseif (\IMM) == 12
+		BSRLI2(\rD, \rA)
+		BSRLI10(\rD, \rD)
+	.elseif (\IMM) == 14
+		BSRLI4(\rD, \rA)
+		BSRLI10(\rD, \rD)
+	.elseif (\IMM) == 20
+		BSRLI20(\rD, \rA)
+	.elseif (\IMM) == 24
+		BSRLI4(\rD, \rA)
+		BSRLI20(\rD, \rD)
+	.elseif (\IMM) == 28
+		BSRLI4(\rD, \rA)
+		BSRLI4(\rD, \rD)
+		BSRLI20(\rD, \rD)
+	.else
+	.error "BSRLI shift macros \IMM"
+	.endif
+	.endm
+	#endif
+
+#endif /* CONFIG_MMU */
+
+.extern other_exception_handler /* Defined in exception.c */
+
+/*
+ * hw_exception_handler - Handler for exceptions
+ *
+ * Exception handler notes:
+ * - Handles all exceptions
+ * - Does not handle unaligned exceptions during load into r17, r1, r0.
+ * - Does not handle unaligned exceptions during store from r17 (cannot be
+ *   done) and r1 (slows down common case)
+ *
+ *  Relevant register structures
+ *
+ *  EAR - |----|----|----|----|----|----|----|----|
+ *      - <  ##   32 bit faulting address     ##  >
+ *
+ *  ESR - |----|----|----|----|----| - | - |-----|-----|
+ *      -                            W   S   REG   EXC
+ *
+ *
+ * STACK FRAME STRUCTURE (for CONFIG_MMU=n)
+ * ----------------------------------------
+ *
+ *      +-------------+         + 0
+ *      |     MSR     |
+ *      +-------------+         + 4
+ *      |     r1      |
+ *      |      .      |
+ *      |      .      |
+ *      |      .      |
+ *      |      .      |
+ *      |     r18     |
+ *      +-------------+         + 76
+ *      |      .      |
+ *      |      .      |
+ *
+ * MMU kernel uses the same 'pt_pool_space' pointed space
+ * which is used for storing register values - noMMu style was, that values were
+ * stored in stack but in case of failure you lost information about register.
+ * Currently you can see register value in memory in specific place.
+ * In compare to with previous solution the speed should be the same.
+ *
+ * MMU exception handler has different handling compare to no MMU kernel.
+ * Exception handler use jump table for directing of what happen. For MMU kernel
+ * is this approach better because MMU relate exception are handled by asm code
+ * in this file. In compare to with MMU expect of unaligned exception
+ * is everything handled by C code.
+ */
+
+/*
+ * every of these handlers is entered having R3/4/5/6/11/current saved on stack
+ * and clobbered so care should be taken to restore them if someone is going to
+ * return from exception
+ */
+
+/* wrappers to restore state before coming to entry.S */
+#ifdef CONFIG_MMU
+.section .data
+.align 4
+pt_pool_space:
+	.space	PT_SIZE
+
+#ifdef DEBUG
+/* Create space for exception counting. */
+.section .data
+.global exception_debug_table
+.align 4
+exception_debug_table:
+	/* Look at exception vector table. There is 32 exceptions * word size */
+	.space	(32 * 4)
+#endif /* DEBUG */
+
+.section .rodata
+.align 4
+_MB_HW_ExceptionVectorTable:
+/*  0 - Undefined */
+	.long	TOPHYS(ex_handler_unhandled)
+/*  1 - Unaligned data access exception */
+	.long	TOPHYS(handle_unaligned_ex)
+/*  2 - Illegal op-code exception */
+	.long	TOPHYS(full_exception_trapw)
+/*  3 - Instruction bus error exception */
+	.long	TOPHYS(full_exception_trapw)
+/*  4 - Data bus error exception */
+	.long	TOPHYS(full_exception_trapw)
+/*  5 - Divide by zero exception */
+	.long	TOPHYS(full_exception_trapw)
+/*  6 - Floating point unit exception */
+	.long	TOPHYS(full_exception_trapw)
+/*  7 - Privileged instruction exception */
+	.long	TOPHYS(full_exception_trapw)
+/*  8 - 15 - Undefined */
+	.long	TOPHYS(ex_handler_unhandled)
+	.long	TOPHYS(ex_handler_unhandled)
+	.long	TOPHYS(ex_handler_unhandled)
+	.long	TOPHYS(ex_handler_unhandled)
+	.long	TOPHYS(ex_handler_unhandled)
+	.long	TOPHYS(ex_handler_unhandled)
+	.long	TOPHYS(ex_handler_unhandled)
+	.long	TOPHYS(ex_handler_unhandled)
+/* 16 - Data storage exception */
+	.long	TOPHYS(handle_data_storage_exception)
+/* 17 - Instruction storage exception */
+	.long	TOPHYS(handle_instruction_storage_exception)
+/* 18 - Data TLB miss exception */
+	.long	TOPHYS(handle_data_tlb_miss_exception)
+/* 19 - Instruction TLB miss exception */
+	.long	TOPHYS(handle_instruction_tlb_miss_exception)
+/* 20 - 31 - Undefined */
+	.long	TOPHYS(ex_handler_unhandled)
+	.long	TOPHYS(ex_handler_unhandled)
+	.long	TOPHYS(ex_handler_unhandled)
+	.long	TOPHYS(ex_handler_unhandled)
+	.long	TOPHYS(ex_handler_unhandled)
+	.long	TOPHYS(ex_handler_unhandled)
+	.long	TOPHYS(ex_handler_unhandled)
+	.long	TOPHYS(ex_handler_unhandled)
+	.long	TOPHYS(ex_handler_unhandled)
+	.long	TOPHYS(ex_handler_unhandled)
+	.long	TOPHYS(ex_handler_unhandled)
+	.long	TOPHYS(ex_handler_unhandled)
+#endif
+
+.global _hw_exception_handler
+.section .text
+.align 4
+.ent _hw_exception_handler
+_hw_exception_handler:
+#ifndef CONFIG_MMU
+	addik	r1, r1, -(EX_HANDLER_STACK_SIZ); /* Create stack frame */
+#else
+	swi	r1, r0, TOPHYS(pt_pool_space + PT_R1); /* GET_SP */
+	/* Save date to kernel memory. Here is the problem
+	 * when you came from user space */
+	ori	r1, r0, TOPHYS(pt_pool_space);
+#endif
+	swi	r3, r1, PT_R3
+	swi	r4, r1, PT_R4
+	swi	r5, r1, PT_R5
+	swi	r6, r1, PT_R6
+
+#ifdef CONFIG_MMU
+	swi	r11, r1, PT_R11
+	swi	r31, r1, PT_R31
+	lwi	r31, r0, TOPHYS(PER_CPU(CURRENT_SAVE)) /* get saved current */
+#endif
+
+	mfs	r5, rmsr;
+	nop
+	swi	r5, r1, 0;
+	mfs	r4, resr
+	nop
+	mfs	r3, rear;
+	nop
+
+#ifndef CONFIG_MMU
+	andi	r5, r4, 0x1000;		/* Check ESR[DS] */
+	beqi	r5, not_in_delay_slot;	/* Branch if ESR[DS] not set */
+	mfs	r17, rbtr;	/* ESR[DS] set - return address in BTR */
+	nop
+not_in_delay_slot:
+	swi	r17, r1, PT_R17
+#endif
+
+	andi	r5, r4, 0x1F;		/* Extract ESR[EXC] */
+
+#ifdef CONFIG_MMU
+	/* Calculate exception vector offset = r5 << 2 */
+	addk	r6, r5, r5; /* << 1 */
+	addk	r6, r6, r6; /* << 2 */
+
+#ifdef DEBUG
+/* counting which exception happen */
+	lwi	r5, r0, TOPHYS(exception_debug_table)
+	addi	r5, r5, 1
+	swi	r5, r0, TOPHYS(exception_debug_table)
+	lwi	r5, r6, TOPHYS(exception_debug_table)
+	addi	r5, r5, 1
+	swi	r5, r6, TOPHYS(exception_debug_table)
+#endif
+/* end */
+	/* Load the HW Exception vector */
+	lwi	r6, r6, TOPHYS(_MB_HW_ExceptionVectorTable)
+	bra	r6
+
+full_exception_trapw:
+	RESTORE_STATE
+	bri	full_exception_trap
+#else
+	/* Exceptions enabled here. This will allow nested exceptions */
+	mfs	r6, rmsr;
+	nop
+	swi	r6, r1, 0; /* RMSR_OFFSET */
+	ori	r6, r6, 0x100; /* Turn ON the EE bit */
+	andi	r6, r6, ~2; /* Disable interrupts */
+	mts	rmsr, r6;
+	nop
+
+	xori	r6, r5, 1; /* 00001 = Unaligned Exception */
+	/* Jump to unalignment exception handler */
+	beqi	r6, handle_unaligned_ex;
+
+handle_other_ex: /* Handle Other exceptions here */
+	/* Save other volatiles before we make procedure calls below */
+	swi	r7, r1, PT_R7
+	swi	r8, r1, PT_R8
+	swi	r9, r1, PT_R9
+	swi	r10, r1, PT_R10
+	swi	r11, r1, PT_R11
+	swi	r12, r1, PT_R12
+	swi	r14, r1, PT_R14
+	swi	r15, r1, PT_R15
+	swi	r18, r1, PT_R18
+
+	or	r5, r1, r0
+	andi	r6, r4, 0x1F; /* Load ESR[EC] */
+	lwi	r7, r0, PER_CPU(KM) /* MS: saving current kernel mode to regs */
+	swi	r7, r1, PT_MODE
+	mfs	r7, rfsr
+	nop
+	addk	r8, r17, r0; /* Load exception address */
+	bralid	r15, full_exception; /* Branch to the handler */
+	nop;
+	mts	rfsr, r0;	/* Clear sticky fsr */
+	nop
+
+	/*
+	 * Trigger execution of the signal handler by enabling
+	 * interrupts and calling an invalid syscall.
+	 */
+	mfs	r5, rmsr;
+	nop
+	ori	r5, r5, 2;
+	mts	rmsr, r5; /* enable interrupt */
+	nop
+	addi	r12, r0, __NR_syscalls;
+	brki	r14, 0x08;
+	mfs	r5, rmsr; /* disable interrupt */
+	nop
+	andi	r5, r5, ~2;
+	mts	rmsr, r5;
+	nop
+
+	lwi	r7, r1, PT_R7
+	lwi	r8, r1, PT_R8
+	lwi	r9, r1, PT_R9
+	lwi	r10, r1, PT_R10
+	lwi	r11, r1, PT_R11
+	lwi	r12, r1, PT_R12
+	lwi	r14, r1, PT_R14
+	lwi	r15, r1, PT_R15
+	lwi	r18, r1, PT_R18
+
+	bri	ex_handler_done; /* Complete exception handling */
+#endif
+
+/* 0x01 - Unaligned data access exception
+ * This occurs when a word access is not aligned on a word boundary,
+ * or when a 16-bit access is not aligned on a 16-bit boundary.
+ * This handler perform the access, and returns, except for MMU when
+ * the unaligned address is last on a 4k page or the physical address is
+ * not found in the page table, in which case unaligned_data_trap is called.
+ */
+handle_unaligned_ex:
+	/* Working registers already saved: R3, R4, R5, R6
+	 *  R4 = ESR
+	 *  R3 = EAR
+	 */
+#ifdef CONFIG_MMU
+	andi	r6, r4, 0x1000			/* Check ESR[DS] */
+	beqi	r6, _no_delayslot		/* Branch if ESR[DS] not set */
+	mfs	r17, rbtr;	/* ESR[DS] set - return address in BTR */
+	nop
+_no_delayslot:
+	/* jump to high level unaligned handler */
+	RESTORE_STATE;
+	bri	unaligned_data_trap
+#endif
+	andi	r6, r4, 0x3E0; /* Mask and extract the register operand */
+	srl	r6, r6; /* r6 >> 5 */
+	srl	r6, r6;
+	srl	r6, r6;
+	srl	r6, r6;
+	srl	r6, r6;
+	/* Store the register operand in a temporary location */
+	sbi	r6, r0, TOPHYS(ex_reg_op);
+
+	andi	r6, r4, 0x400; /* Extract ESR[S] */
+	bnei	r6, ex_sw;
+ex_lw:
+	andi	r6, r4, 0x800; /* Extract ESR[W] */
+	beqi	r6, ex_lhw;
+	lbui	r5, r3, 0; /* Exception address in r3 */
+	/* Load a word, byte-by-byte from destination address
+		and save it in tmp space */
+	sbi	r5, r0, TOPHYS(ex_tmp_data_loc_0);
+	lbui	r5, r3, 1;
+	sbi	r5, r0, TOPHYS(ex_tmp_data_loc_1);
+	lbui	r5, r3, 2;
+	sbi	r5, r0, TOPHYS(ex_tmp_data_loc_2);
+	lbui	r5, r3, 3;
+	sbi	r5, r0, TOPHYS(ex_tmp_data_loc_3);
+	/* Get the destination register value into r4 */
+	lwi	r4, r0, TOPHYS(ex_tmp_data_loc_0);
+	bri	ex_lw_tail;
+ex_lhw:
+	lbui	r5, r3, 0; /* Exception address in r3 */
+	/* Load a half-word, byte-by-byte from destination
+		address and save it in tmp space */
+	sbi	r5, r0, TOPHYS(ex_tmp_data_loc_0);
+	lbui	r5, r3, 1;
+	sbi	r5, r0, TOPHYS(ex_tmp_data_loc_1);
+	/* Get the destination register value into r4 */
+	lhui	r4, r0, TOPHYS(ex_tmp_data_loc_0);
+ex_lw_tail:
+	/* Get the destination register number into r5 */
+	lbui	r5, r0, TOPHYS(ex_reg_op);
+	/* Form load_word jump table offset (lw_table + (8 * regnum)) */
+	addik	r6, r0, TOPHYS(lw_table);
+	addk	r5, r5, r5;
+	addk	r5, r5, r5;
+	addk	r5, r5, r5;
+	addk	r5, r5, r6;
+	bra	r5;
+ex_lw_end: /* Exception handling of load word, ends */
+ex_sw:
+	/* Get the destination register number into r5 */
+	lbui	r5, r0, TOPHYS(ex_reg_op);
+	/* Form store_word jump table offset (sw_table + (8 * regnum)) */
+	addik	r6, r0, TOPHYS(sw_table);
+	add	r5, r5, r5;
+	add	r5, r5, r5;
+	add	r5, r5, r5;
+	add	r5, r5, r6;
+	bra	r5;
+ex_sw_tail:
+	mfs	r6, resr;
+	nop
+	andi	r6, r6, 0x800; /* Extract ESR[W] */
+	beqi	r6, ex_shw;
+	/* Get the word - delay slot */
+	swi	r4, r0, TOPHYS(ex_tmp_data_loc_0);
+	/* Store the word, byte-by-byte into destination address */
+	lbui	r4, r0, TOPHYS(ex_tmp_data_loc_0);
+	sbi	r4, r3, 0;
+	lbui	r4, r0, TOPHYS(ex_tmp_data_loc_1);
+	sbi	r4, r3, 1;
+	lbui	r4, r0, TOPHYS(ex_tmp_data_loc_2);
+	sbi	r4, r3, 2;
+	lbui	r4, r0, TOPHYS(ex_tmp_data_loc_3);
+	sbi	r4, r3, 3;
+	bri	ex_handler_done;
+
+ex_shw:
+	/* Store the lower half-word, byte-by-byte into destination address */
+	swi	r4, r0, TOPHYS(ex_tmp_data_loc_0);
+	lbui	r4, r0, TOPHYS(ex_tmp_data_loc_2);
+	sbi	r4, r3, 0;
+	lbui	r4, r0, TOPHYS(ex_tmp_data_loc_3);
+	sbi	r4, r3, 1;
+ex_sw_end: /* Exception handling of store word, ends. */
+
+ex_handler_done:
+#ifndef CONFIG_MMU
+	lwi	r5, r1, 0 /* RMSR */
+	mts	rmsr, r5
+	nop
+	lwi	r3, r1, PT_R3
+	lwi	r4, r1, PT_R4
+	lwi	r5, r1, PT_R5
+	lwi	r6, r1, PT_R6
+	lwi	r17, r1, PT_R17
+
+	rted	r17, 0
+	addik	r1, r1, (EX_HANDLER_STACK_SIZ); /* Restore stack frame */
+#else
+	RESTORE_STATE;
+	rted	r17, 0
+	nop
+#endif
+
+#ifdef CONFIG_MMU
+	/* Exception vector entry code. This code runs with address translation
+	 * turned off (i.e. using physical addresses). */
+
+	/* Exception vectors. */
+
+	/* 0x10 - Data Storage Exception
+	 * This happens for just a few reasons. U0 set (but we don't do that),
+	 * or zone protection fault (user violation, write to protected page).
+	 * If this is just an update of modified status, we do that quickly
+	 * and exit. Otherwise, we call heavyweight functions to do the work.
+	 */
+	handle_data_storage_exception:
+		/* Working registers already saved: R3, R4, R5, R6
+		 * R3 = ESR
+		 */
+		mfs	r11, rpid
+		nop
+		/* If we are faulting a kernel address, we have to use the
+		 * kernel page tables.
+		 */
+		ori	r5, r0, CONFIG_KERNEL_START
+		cmpu	r5, r3, r5
+		bgti	r5, ex3
+		/* First, check if it was a zone fault (which means a user
+		 * tried to access a kernel or read-protected page - always
+		 * a SEGV). All other faults here must be stores, so no
+		 * need to check ESR_S as well. */
+		andi	r4, r4, ESR_DIZ		/* ESR_Z - zone protection */
+		bnei	r4, ex2
+
+		ori	r4, r0, swapper_pg_dir
+		mts	rpid, r0		/* TLB will have 0 TID */
+		nop
+		bri	ex4
+
+		/* Get the PGD for the current thread. */
+	ex3:
+		/* First, check if it was a zone fault (which means a user
+		 * tried to access a kernel or read-protected page - always
+		 * a SEGV). All other faults here must be stores, so no
+		 * need to check ESR_S as well. */
+		andi	r4, r4, ESR_DIZ		/* ESR_Z */
+		bnei	r4, ex2
+		/* get current task address */
+		addi	r4 ,CURRENT_TASK, TOPHYS(0);
+		lwi	r4, r4, TASK_THREAD+PGDIR
+	ex4:
+		tophys(r4,r4)
+		/* Create L1 (pgdir/pmd) address */
+		bsrli	r5, r3, PGDIR_SHIFT - 2
+		andi	r5, r5, PAGE_SIZE - 4
+/* Assume pgdir aligned on 4K boundary, no need for "andi r4,r4,0xfffff003" */
+		or	r4, r4, r5
+		lwi	r4, r4, 0		/* Get L1 entry */
+		andi	r5, r4, PAGE_MASK /* Extract L2 (pte) base address */
+		beqi	r5, ex2			/* Bail if no table */
+
+		tophys(r5,r5)
+		bsrli	r6, r3, PTE_SHIFT /* Compute PTE address */
+		andi	r6, r6, PAGE_SIZE - 4
+		or	r5, r5, r6
+		lwi	r4, r5, 0		/* Get Linux PTE */
+
+		andi	r6, r4, _PAGE_RW	/* Is it writeable? */
+		beqi	r6, ex2			/* Bail if not */
+
+		/* Update 'changed' */
+		ori	r4, r4, _PAGE_DIRTY|_PAGE_ACCESSED|_PAGE_HWWRITE
+		swi	r4, r5, 0		/* Update Linux page table */
+
+		/* Most of the Linux PTE is ready to load into the TLB LO.
+		 * We set ZSEL, where only the LS-bit determines user access.
+		 * We set execute, because we don't have the granularity to
+		 * properly set this at the page level (Linux problem).
+		 * If shared is set, we cause a zero PID->TID load.
+		 * Many of these bits are software only. Bits we don't set
+		 * here we (properly should) assume have the appropriate value.
+		 */
+/* Ignore memory coherent, just LSB on ZSEL is used + EX/WR */
+		andi	r4, r4, PAGE_MASK | TLB_EX | TLB_WR | \
+						TLB_ZSEL(1) | TLB_ATTR_MASK
+		ori	r4, r4, _PAGE_HWEXEC	/* make it executable */
+
+		/* find the TLB index that caused the fault. It has to be here*/
+		mts	rtlbsx, r3
+		nop
+		mfs	r5, rtlbx		/* DEBUG: TBD */
+		nop
+		mts	rtlblo, r4		/* Load TLB LO */
+		nop
+						/* Will sync shadow TLBs */
+
+		/* Done...restore registers and get out of here. */
+		mts	rpid, r11
+		nop
+		bri 4
+
+		RESTORE_STATE;
+		rted	r17, 0
+		nop
+	ex2:
+		/* The bailout. Restore registers to pre-exception conditions
+		 * and call the heavyweights to help us out. */
+		mts	rpid, r11
+		nop
+		bri 4
+		RESTORE_STATE;
+		bri	page_fault_data_trap
+
+
+	/* 0x11 - Instruction Storage Exception
+	 * This is caused by a fetch from non-execute or guarded pages. */
+	handle_instruction_storage_exception:
+		/* Working registers already saved: R3, R4, R5, R6
+		 * R3 = ESR
+		 */
+
+		RESTORE_STATE;
+		bri	page_fault_instr_trap
+
+	/* 0x12 - Data TLB Miss Exception
+	 * As the name implies, translation is not in the MMU, so search the
+	 * page tables and fix it. The only purpose of this function is to
+	 * load TLB entries from the page table if they exist.
+	 */
+	handle_data_tlb_miss_exception:
+		/* Working registers already saved: R3, R4, R5, R6
+		 * R3 = EAR, R4 = ESR
+		 */
+		mfs	r11, rpid
+		nop
+
+		/* If we are faulting a kernel address, we have to use the
+		 * kernel page tables. */
+		ori	r6, r0, CONFIG_KERNEL_START
+		cmpu	r4, r3, r6
+		bgti	r4, ex5
+		ori	r4, r0, swapper_pg_dir
+		mts	rpid, r0		/* TLB will have 0 TID */
+		nop
+		bri	ex6
+
+		/* Get the PGD for the current thread. */
+	ex5:
+		/* get current task address */
+		addi	r4 ,CURRENT_TASK, TOPHYS(0);
+		lwi	r4, r4, TASK_THREAD+PGDIR
+	ex6:
+		tophys(r4,r4)
+		/* Create L1 (pgdir/pmd) address */
+		bsrli	r5, r3, PGDIR_SHIFT - 2
+		andi	r5, r5, PAGE_SIZE - 4
+/* Assume pgdir aligned on 4K boundary, no need for "andi r4,r4,0xfffff003" */
+		or	r4, r4, r5
+		lwi	r4, r4, 0		/* Get L1 entry */
+		andi	r5, r4, PAGE_MASK /* Extract L2 (pte) base address */
+		beqi	r5, ex7			/* Bail if no table */
+
+		tophys(r5,r5)
+		bsrli	r6, r3, PTE_SHIFT /* Compute PTE address */
+		andi	r6, r6, PAGE_SIZE - 4
+		or	r5, r5, r6
+		lwi	r4, r5, 0		/* Get Linux PTE */
+
+		andi	r6, r4, _PAGE_PRESENT
+		beqi	r6, ex7
+
+		ori	r4, r4, _PAGE_ACCESSED
+		swi	r4, r5, 0
+
+		/* Most of the Linux PTE is ready to load into the TLB LO.
+		 * We set ZSEL, where only the LS-bit determines user access.
+		 * We set execute, because we don't have the granularity to
+		 * properly set this at the page level (Linux problem).
+		 * If shared is set, we cause a zero PID->TID load.
+		 * Many of these bits are software only. Bits we don't set
+		 * here we (properly should) assume have the appropriate value.
+		 */
+		brid	finish_tlb_load
+		andi	r4, r4, PAGE_MASK | TLB_EX | TLB_WR | \
+						TLB_ZSEL(1) | TLB_ATTR_MASK
+	ex7:
+		/* The bailout. Restore registers to pre-exception conditions
+		 * and call the heavyweights to help us out.
+		 */
+		mts	rpid, r11
+		nop
+		bri	4
+		RESTORE_STATE;
+		bri	page_fault_data_trap
+
+	/* 0x13 - Instruction TLB Miss Exception
+	 * Nearly the same as above, except we get our information from
+	 * different registers and bailout to a different point.
+	 */
+	handle_instruction_tlb_miss_exception:
+		/* Working registers already saved: R3, R4, R5, R6
+		 *  R3 = ESR
+		 */
+		mfs	r11, rpid
+		nop
+
+		/* If we are faulting a kernel address, we have to use the
+		 * kernel page tables.
+		 */
+		ori	r4, r0, CONFIG_KERNEL_START
+		cmpu	r4, r3, r4
+		bgti	r4, ex8
+		ori	r4, r0, swapper_pg_dir
+		mts	rpid, r0		/* TLB will have 0 TID */
+		nop
+		bri	ex9
+
+		/* Get the PGD for the current thread. */
+	ex8:
+		/* get current task address */
+		addi	r4 ,CURRENT_TASK, TOPHYS(0);
+		lwi	r4, r4, TASK_THREAD+PGDIR
+	ex9:
+		tophys(r4,r4)
+		/* Create L1 (pgdir/pmd) address */
+		bsrli	r5, r3, PGDIR_SHIFT - 2
+		andi	r5, r5, PAGE_SIZE - 4
+/* Assume pgdir aligned on 4K boundary, no need for "andi r4,r4,0xfffff003" */
+		or	r4, r4, r5
+		lwi	r4, r4, 0		/* Get L1 entry */
+		andi	r5, r4, PAGE_MASK /* Extract L2 (pte) base address */
+		beqi	r5, ex10		/* Bail if no table */
+
+		tophys(r5,r5)
+		bsrli	r6, r3, PTE_SHIFT /* Compute PTE address */
+		andi	r6, r6, PAGE_SIZE - 4
+		or	r5, r5, r6
+		lwi	r4, r5, 0		/* Get Linux PTE */
+
+		andi	r6, r4, _PAGE_PRESENT
+		beqi	r6, ex10
+
+		ori	r4, r4, _PAGE_ACCESSED
+		swi	r4, r5, 0
+
+		/* Most of the Linux PTE is ready to load into the TLB LO.
+		 * We set ZSEL, where only the LS-bit determines user access.
+		 * We set execute, because we don't have the granularity to
+		 * properly set this at the page level (Linux problem).
+		 * If shared is set, we cause a zero PID->TID load.
+		 * Many of these bits are software only. Bits we don't set
+		 * here we (properly should) assume have the appropriate value.
+		 */
+		brid	finish_tlb_load
+		andi	r4, r4, PAGE_MASK | TLB_EX | TLB_WR | \
+						TLB_ZSEL(1) | TLB_ATTR_MASK
+	ex10:
+		/* The bailout. Restore registers to pre-exception conditions
+		 * and call the heavyweights to help us out.
+		 */
+		mts	rpid, r11
+		nop
+		bri 4
+		RESTORE_STATE;
+		bri	page_fault_instr_trap
+
+/* Both the instruction and data TLB miss get to this point to load the TLB.
+ *	r3 - EA of fault
+ *	r4 - TLB LO (info from Linux PTE)
+ *	r5, r6 - available to use
+ *	PID - loaded with proper value when we get here
+ *	Upon exit, we reload everything and RFI.
+ * A common place to load the TLB.
+ */
+.section .data
+.align 4
+.global tlb_skip
+	tlb_skip:
+		.long	MICROBLAZE_TLB_SKIP
+	tlb_index:
+		/* MS: storing last used tlb index */
+		.long	MICROBLAZE_TLB_SIZE/2
+.previous
+	finish_tlb_load:
+		/* MS: load the last used TLB index. */
+		lwi	r5, r0, TOPHYS(tlb_index)
+		addik	r5, r5, 1 /* MS: inc tlb_index -> use next one */
+
+/* MS: FIXME this is potential fault, because this is mask not count */
+		andi	r5, r5, MICROBLAZE_TLB_SIZE - 1
+		ori	r6, r0, 1
+		cmp	r31, r5, r6
+		blti	r31, ex12
+		lwi	r5, r0, TOPHYS(tlb_skip)
+	ex12:
+		/* MS: save back current TLB index */
+		swi	r5, r0, TOPHYS(tlb_index)
+
+		ori	r4, r4, _PAGE_HWEXEC	/* make it executable */
+		mts	rtlbx, r5		/* MS: save current TLB */
+		nop
+		mts	rtlblo,	r4		/* MS: save to TLB LO */
+		nop
+
+		/* Create EPN. This is the faulting address plus a static
+		 * set of bits. These are size, valid, E, U0, and ensure
+		 * bits 20 and 21 are zero.
+		 */
+		andi	r3, r3, PAGE_MASK
+#ifdef CONFIG_MICROBLAZE_64K_PAGES
+		ori	r3, r3, TLB_VALID | TLB_PAGESZ(PAGESZ_64K)
+#elif CONFIG_MICROBLAZE_16K_PAGES
+		ori	r3, r3, TLB_VALID | TLB_PAGESZ(PAGESZ_16K)
+#else
+		ori	r3, r3, TLB_VALID | TLB_PAGESZ(PAGESZ_4K)
+#endif
+		mts	rtlbhi,	r3		/* Load TLB HI */
+		nop
+
+		/* Done...restore registers and get out of here. */
+		mts	rpid, r11
+		nop
+		bri 4
+		RESTORE_STATE;
+		rted	r17, 0
+		nop
+
+	/* extern void giveup_fpu(struct task_struct *prev)
+	 *
+	 * The MicroBlaze processor may have an FPU, so this should not just
+	 * return: TBD.
+	 */
+	.globl giveup_fpu;
+	.align 4;
+	giveup_fpu:
+		bralid	r15,0			/* TBD */
+		nop
+
+	/* At present, this routine just hangs. - extern void abort(void) */
+	.globl abort;
+	.align 4;
+	abort:
+		br	r0
+
+	.globl set_context;
+	.align 4;
+	set_context:
+		mts	rpid, r5	/* Shadow TLBs are automatically */
+		nop
+		bri	4		/* flushed by changing PID */
+		rtsd	r15,8
+		nop
+
+#endif
+.end _hw_exception_handler
+
+#ifdef CONFIG_MMU
+/* Unaligned data access exception last on a 4k page for MMU.
+ * When this is called, we are in virtual mode with exceptions enabled
+ * and registers 1-13,15,17,18 saved.
+ *
+ * R3 = ESR
+ * R4 = EAR
+ * R7 = pointer to saved registers (struct pt_regs *regs)
+ *
+ * This handler perform the access, and returns via ret_from_exc.
+ */
+.global _unaligned_data_exception
+.ent _unaligned_data_exception
+_unaligned_data_exception:
+	andi	r8, r3, 0x3E0;	/* Mask and extract the register operand */
+	bsrli   r8, r8, 2;		/* r8 >> 2 = register operand * 8 */
+	andi	r6, r3, 0x400;	/* Extract ESR[S] */
+	bneid	r6, ex_sw_vm;
+	andi	r6, r3, 0x800;	/* Extract ESR[W] - delay slot */
+ex_lw_vm:
+	beqid	r6, ex_lhw_vm;
+load1:	lbui	r5, r4, 0;	/* Exception address in r4 - delay slot */
+/* Load a word, byte-by-byte from destination address and save it in tmp space*/
+	addik	r6, r0, ex_tmp_data_loc_0;
+	sbi	r5, r6, 0;
+load2:	lbui	r5, r4, 1;
+	sbi	r5, r6, 1;
+load3:	lbui	r5, r4, 2;
+	sbi	r5, r6, 2;
+load4:	lbui	r5, r4, 3;
+	sbi	r5, r6, 3;
+	brid	ex_lw_tail_vm;
+/* Get the destination register value into r3 - delay slot */
+	lwi	r3, r6, 0;
+ex_lhw_vm:
+	/* Load a half-word, byte-by-byte from destination address and
+	 * save it in tmp space */
+	addik	r6, r0, ex_tmp_data_loc_0;
+	sbi	r5, r6, 0;
+load5:	lbui	r5, r4, 1;
+	sbi	r5, r6, 1;
+	lhui	r3, r6, 0;	/* Get the destination register value into r3 */
+ex_lw_tail_vm:
+	/* Form load_word jump table offset (lw_table_vm + (8 * regnum)) */
+	addik	r5, r8, lw_table_vm;
+	bra	r5;
+ex_lw_end_vm:			/* Exception handling of load word, ends */
+	brai	ret_from_exc;
+ex_sw_vm:
+/* Form store_word jump table offset (sw_table_vm + (8 * regnum)) */
+	addik	r5, r8, sw_table_vm;
+	bra	r5;
+ex_sw_tail_vm:
+	addik	r5, r0, ex_tmp_data_loc_0;
+	beqid	r6, ex_shw_vm;
+	swi	r3, r5, 0;	/* Get the word - delay slot */
+	/* Store the word, byte-by-byte into destination address */
+	lbui	r3, r5, 0;
+store1:	sbi	r3, r4, 0;
+	lbui	r3, r5, 1;
+store2:	sbi	r3, r4, 1;
+	lbui	r3, r5, 2;
+store3:	sbi	r3, r4, 2;
+	lbui	r3, r5, 3;
+	brid	ret_from_exc;
+store4:	sbi	r3, r4, 3;	/* Delay slot */
+ex_shw_vm:
+	/* Store the lower half-word, byte-by-byte into destination address */
+#ifdef __MICROBLAZEEL__
+	lbui	r3, r5, 0;
+store5:	sbi	r3, r4, 0;
+	lbui	r3, r5, 1;
+	brid	ret_from_exc;
+store6:	sbi	r3, r4, 1;	/* Delay slot */
+#else
+	lbui	r3, r5, 2;
+store5:	sbi	r3, r4, 0;
+	lbui	r3, r5, 3;
+	brid	ret_from_exc;
+store6:	sbi	r3, r4, 1;	/* Delay slot */
+#endif
+
+ex_sw_end_vm:			/* Exception handling of store word, ends. */
+
+/* We have to prevent cases that get/put_user macros get unaligned pointer
+ * to bad page area. We have to find out which origin instruction caused it
+ * and called fixup for that origin instruction not instruction in unaligned
+ * handler */
+ex_unaligned_fixup:
+	ori	r5, r7, 0 /* setup pointer to pt_regs */
+	lwi	r6, r7, PT_PC; /* faulting address is one instruction above */
+	addik	r6, r6, -4 /* for finding proper fixup */
+	swi	r6, r7, PT_PC; /* a save back it to PT_PC */
+	addik	r7, r0, SIGSEGV
+	/* call bad_page_fault for finding aligned fixup, fixup address is saved
+	 * in PT_PC which is used as return address from exception */
+	addik	r15, r0, ret_from_exc-8 /* setup return address */
+	brid	bad_page_fault
+	nop
+
+/* We prevent all load/store because it could failed any attempt to access */
+.section __ex_table,"a";
+	.word	load1,ex_unaligned_fixup;
+	.word	load2,ex_unaligned_fixup;
+	.word	load3,ex_unaligned_fixup;
+	.word	load4,ex_unaligned_fixup;
+	.word	load5,ex_unaligned_fixup;
+	.word	store1,ex_unaligned_fixup;
+	.word	store2,ex_unaligned_fixup;
+	.word	store3,ex_unaligned_fixup;
+	.word	store4,ex_unaligned_fixup;
+	.word	store5,ex_unaligned_fixup;
+	.word	store6,ex_unaligned_fixup;
+.previous;
+.end _unaligned_data_exception
+#endif /* CONFIG_MMU */
+
+.global ex_handler_unhandled
+ex_handler_unhandled:
+/* FIXME add handle function for unhandled exception - dump register */
+	bri 0
+
+/*
+ * hw_exception_handler Jump Table
+ * - Contains code snippets for each register that caused the unalign exception
+ * - Hence exception handler is NOT self-modifying
+ * - Separate table for load exceptions and store exceptions.
+ * - Each table is of size: (8 * 32) = 256 bytes
+ */
+
+.section .text
+.align 4
+lw_table:
+lw_r0:		R3_TO_LWREG	(0);
+lw_r1:		LWREG_NOP;
+lw_r2:		R3_TO_LWREG	(2);
+lw_r3:		R3_TO_LWREG_V	(3);
+lw_r4:		R3_TO_LWREG_V	(4);
+lw_r5:		R3_TO_LWREG_V	(5);
+lw_r6:		R3_TO_LWREG_V	(6);
+lw_r7:		R3_TO_LWREG	(7);
+lw_r8:		R3_TO_LWREG	(8);
+lw_r9:		R3_TO_LWREG	(9);
+lw_r10:		R3_TO_LWREG	(10);
+lw_r11:		R3_TO_LWREG	(11);
+lw_r12:		R3_TO_LWREG	(12);
+lw_r13:		R3_TO_LWREG	(13);
+lw_r14:		R3_TO_LWREG	(14);
+lw_r15:		R3_TO_LWREG	(15);
+lw_r16:		R3_TO_LWREG	(16);
+lw_r17:		LWREG_NOP;
+lw_r18:		R3_TO_LWREG	(18);
+lw_r19:		R3_TO_LWREG	(19);
+lw_r20:		R3_TO_LWREG	(20);
+lw_r21:		R3_TO_LWREG	(21);
+lw_r22:		R3_TO_LWREG	(22);
+lw_r23:		R3_TO_LWREG	(23);
+lw_r24:		R3_TO_LWREG	(24);
+lw_r25:		R3_TO_LWREG	(25);
+lw_r26:		R3_TO_LWREG	(26);
+lw_r27:		R3_TO_LWREG	(27);
+lw_r28:		R3_TO_LWREG	(28);
+lw_r29:		R3_TO_LWREG	(29);
+lw_r30:		R3_TO_LWREG	(30);
+#ifdef CONFIG_MMU
+lw_r31: 	R3_TO_LWREG_V	(31);
+#else
+lw_r31:		R3_TO_LWREG	(31);
+#endif
+
+sw_table:
+sw_r0:		SWREG_TO_R3	(0);
+sw_r1:		SWREG_NOP;
+sw_r2:		SWREG_TO_R3	(2);
+sw_r3:		SWREG_TO_R3_V	(3);
+sw_r4:		SWREG_TO_R3_V	(4);
+sw_r5:		SWREG_TO_R3_V	(5);
+sw_r6:		SWREG_TO_R3_V	(6);
+sw_r7:		SWREG_TO_R3	(7);
+sw_r8:		SWREG_TO_R3	(8);
+sw_r9:		SWREG_TO_R3	(9);
+sw_r10:		SWREG_TO_R3	(10);
+sw_r11:		SWREG_TO_R3	(11);
+sw_r12:		SWREG_TO_R3	(12);
+sw_r13:		SWREG_TO_R3	(13);
+sw_r14:		SWREG_TO_R3	(14);
+sw_r15:		SWREG_TO_R3	(15);
+sw_r16:		SWREG_TO_R3	(16);
+sw_r17:		SWREG_NOP;
+sw_r18:		SWREG_TO_R3	(18);
+sw_r19:		SWREG_TO_R3	(19);
+sw_r20:		SWREG_TO_R3	(20);
+sw_r21:		SWREG_TO_R3	(21);
+sw_r22:		SWREG_TO_R3	(22);
+sw_r23:		SWREG_TO_R3	(23);
+sw_r24:		SWREG_TO_R3	(24);
+sw_r25:		SWREG_TO_R3	(25);
+sw_r26:		SWREG_TO_R3	(26);
+sw_r27:		SWREG_TO_R3	(27);
+sw_r28:		SWREG_TO_R3	(28);
+sw_r29:		SWREG_TO_R3	(29);
+sw_r30:		SWREG_TO_R3	(30);
+#ifdef CONFIG_MMU
+sw_r31:		SWREG_TO_R3_V	(31);
+#else
+sw_r31:		SWREG_TO_R3	(31);
+#endif
+
+#ifdef CONFIG_MMU
+lw_table_vm:
+lw_r0_vm:	R3_TO_LWREG_VM		(0);
+lw_r1_vm:	R3_TO_LWREG_VM_V	(1);
+lw_r2_vm:	R3_TO_LWREG_VM_V	(2);
+lw_r3_vm:	R3_TO_LWREG_VM_V	(3);
+lw_r4_vm:	R3_TO_LWREG_VM_V	(4);
+lw_r5_vm:	R3_TO_LWREG_VM_V	(5);
+lw_r6_vm:	R3_TO_LWREG_VM_V	(6);
+lw_r7_vm:	R3_TO_LWREG_VM_V	(7);
+lw_r8_vm:	R3_TO_LWREG_VM_V	(8);
+lw_r9_vm:	R3_TO_LWREG_VM_V	(9);
+lw_r10_vm:	R3_TO_LWREG_VM_V	(10);
+lw_r11_vm:	R3_TO_LWREG_VM_V	(11);
+lw_r12_vm:	R3_TO_LWREG_VM_V	(12);
+lw_r13_vm:	R3_TO_LWREG_VM_V	(13);
+lw_r14_vm:	R3_TO_LWREG_VM_V	(14);
+lw_r15_vm:	R3_TO_LWREG_VM_V	(15);
+lw_r16_vm:	R3_TO_LWREG_VM_V	(16);
+lw_r17_vm:	R3_TO_LWREG_VM_V	(17);
+lw_r18_vm:	R3_TO_LWREG_VM_V	(18);
+lw_r19_vm:	R3_TO_LWREG_VM_V	(19);
+lw_r20_vm:	R3_TO_LWREG_VM_V	(20);
+lw_r21_vm:	R3_TO_LWREG_VM_V	(21);
+lw_r22_vm:	R3_TO_LWREG_VM_V	(22);
+lw_r23_vm:	R3_TO_LWREG_VM_V	(23);
+lw_r24_vm:	R3_TO_LWREG_VM_V	(24);
+lw_r25_vm:	R3_TO_LWREG_VM_V	(25);
+lw_r26_vm:	R3_TO_LWREG_VM_V	(26);
+lw_r27_vm:	R3_TO_LWREG_VM_V	(27);
+lw_r28_vm:	R3_TO_LWREG_VM_V	(28);
+lw_r29_vm:	R3_TO_LWREG_VM_V	(29);
+lw_r30_vm:	R3_TO_LWREG_VM_V	(30);
+lw_r31_vm:	R3_TO_LWREG_VM_V	(31);
+
+sw_table_vm:
+sw_r0_vm:	SWREG_TO_R3_VM		(0);
+sw_r1_vm:	SWREG_TO_R3_VM_V	(1);
+sw_r2_vm:	SWREG_TO_R3_VM_V	(2);
+sw_r3_vm:	SWREG_TO_R3_VM_V	(3);
+sw_r4_vm:	SWREG_TO_R3_VM_V	(4);
+sw_r5_vm:	SWREG_TO_R3_VM_V	(5);
+sw_r6_vm:	SWREG_TO_R3_VM_V	(6);
+sw_r7_vm:	SWREG_TO_R3_VM_V	(7);
+sw_r8_vm:	SWREG_TO_R3_VM_V	(8);
+sw_r9_vm:	SWREG_TO_R3_VM_V	(9);
+sw_r10_vm:	SWREG_TO_R3_VM_V	(10);
+sw_r11_vm:	SWREG_TO_R3_VM_V	(11);
+sw_r12_vm:	SWREG_TO_R3_VM_V	(12);
+sw_r13_vm:	SWREG_TO_R3_VM_V	(13);
+sw_r14_vm:	SWREG_TO_R3_VM_V	(14);
+sw_r15_vm:	SWREG_TO_R3_VM_V	(15);
+sw_r16_vm:	SWREG_TO_R3_VM_V	(16);
+sw_r17_vm:	SWREG_TO_R3_VM_V	(17);
+sw_r18_vm:	SWREG_TO_R3_VM_V	(18);
+sw_r19_vm:	SWREG_TO_R3_VM_V	(19);
+sw_r20_vm:	SWREG_TO_R3_VM_V	(20);
+sw_r21_vm:	SWREG_TO_R3_VM_V	(21);
+sw_r22_vm:	SWREG_TO_R3_VM_V	(22);
+sw_r23_vm:	SWREG_TO_R3_VM_V	(23);
+sw_r24_vm:	SWREG_TO_R3_VM_V	(24);
+sw_r25_vm:	SWREG_TO_R3_VM_V	(25);
+sw_r26_vm:	SWREG_TO_R3_VM_V	(26);
+sw_r27_vm:	SWREG_TO_R3_VM_V	(27);
+sw_r28_vm:	SWREG_TO_R3_VM_V	(28);
+sw_r29_vm:	SWREG_TO_R3_VM_V	(29);
+sw_r30_vm:	SWREG_TO_R3_VM_V	(30);
+sw_r31_vm:	SWREG_TO_R3_VM_V	(31);
+#endif /* CONFIG_MMU */
+
+/* Temporary data structures used in the handler */
+.section .data
+.align 4
+ex_tmp_data_loc_0:
+	.byte 0
+ex_tmp_data_loc_1:
+	.byte 0
+ex_tmp_data_loc_2:
+	.byte 0
+ex_tmp_data_loc_3:
+	.byte 0
+ex_reg_op:
+	.byte 0
diff --git a/arch/microblaze/kernel/intc.c b/arch/microblaze/kernel/intc.c
new file mode 100644
index 0000000..90bec7d
--- /dev/null
+++ b/arch/microblaze/kernel/intc.c
@@ -0,0 +1,196 @@
+/*
+ * Copyright (C) 2007-2013 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2012-2013 Xilinx, Inc.
+ * Copyright (C) 2007-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/irqdomain.h>
+#include <linux/irq.h>
+#include <linux/irqchip.h>
+#include <linux/of_address.h>
+#include <linux/io.h>
+#include <linux/bug.h>
+
+static void __iomem *intc_baseaddr;
+
+/* No one else should require these constants, so define them locally here. */
+#define ISR 0x00			/* Interrupt Status Register */
+#define IPR 0x04			/* Interrupt Pending Register */
+#define IER 0x08			/* Interrupt Enable Register */
+#define IAR 0x0c			/* Interrupt Acknowledge Register */
+#define SIE 0x10			/* Set Interrupt Enable bits */
+#define CIE 0x14			/* Clear Interrupt Enable bits */
+#define IVR 0x18			/* Interrupt Vector Register */
+#define MER 0x1c			/* Master Enable Register */
+
+#define MER_ME (1<<0)
+#define MER_HIE (1<<1)
+
+static unsigned int (*read_fn)(void __iomem *);
+static void (*write_fn)(u32, void __iomem *);
+
+static void intc_write32(u32 val, void __iomem *addr)
+{
+	iowrite32(val, addr);
+}
+
+static unsigned int intc_read32(void __iomem *addr)
+{
+	return ioread32(addr);
+}
+
+static void intc_write32_be(u32 val, void __iomem *addr)
+{
+	iowrite32be(val, addr);
+}
+
+static unsigned int intc_read32_be(void __iomem *addr)
+{
+	return ioread32be(addr);
+}
+
+static void intc_enable_or_unmask(struct irq_data *d)
+{
+	unsigned long mask = 1 << d->hwirq;
+
+	pr_debug("enable_or_unmask: %ld\n", d->hwirq);
+
+	/* ack level irqs because they can't be acked during
+	 * ack function since the handle_level_irq function
+	 * acks the irq before calling the interrupt handler
+	 */
+	if (irqd_is_level_type(d))
+		write_fn(mask, intc_baseaddr + IAR);
+
+	write_fn(mask, intc_baseaddr + SIE);
+}
+
+static void intc_disable_or_mask(struct irq_data *d)
+{
+	pr_debug("disable: %ld\n", d->hwirq);
+	write_fn(1 << d->hwirq, intc_baseaddr + CIE);
+}
+
+static void intc_ack(struct irq_data *d)
+{
+	pr_debug("ack: %ld\n", d->hwirq);
+	write_fn(1 << d->hwirq, intc_baseaddr + IAR);
+}
+
+static void intc_mask_ack(struct irq_data *d)
+{
+	unsigned long mask = 1 << d->hwirq;
+
+	pr_debug("disable_and_ack: %ld\n", d->hwirq);
+	write_fn(mask, intc_baseaddr + CIE);
+	write_fn(mask, intc_baseaddr + IAR);
+}
+
+static struct irq_chip intc_dev = {
+	.name = "Xilinx INTC",
+	.irq_unmask = intc_enable_or_unmask,
+	.irq_mask = intc_disable_or_mask,
+	.irq_ack = intc_ack,
+	.irq_mask_ack = intc_mask_ack,
+};
+
+static struct irq_domain *root_domain;
+
+unsigned int get_irq(void)
+{
+	unsigned int hwirq, irq = -1;
+
+	hwirq = read_fn(intc_baseaddr + IVR);
+	if (hwirq != -1U)
+		irq = irq_find_mapping(root_domain, hwirq);
+
+	pr_debug("get_irq: hwirq=%d, irq=%d\n", hwirq, irq);
+
+	return irq;
+}
+
+static int xintc_map(struct irq_domain *d, unsigned int irq, irq_hw_number_t hw)
+{
+	u32 intr_mask = (u32)d->host_data;
+
+	if (intr_mask & (1 << hw)) {
+		irq_set_chip_and_handler_name(irq, &intc_dev,
+						handle_edge_irq, "edge");
+		irq_clear_status_flags(irq, IRQ_LEVEL);
+	} else {
+		irq_set_chip_and_handler_name(irq, &intc_dev,
+						handle_level_irq, "level");
+		irq_set_status_flags(irq, IRQ_LEVEL);
+	}
+	return 0;
+}
+
+static const struct irq_domain_ops xintc_irq_domain_ops = {
+	.xlate = irq_domain_xlate_onetwocell,
+	.map = xintc_map,
+};
+
+static int __init xilinx_intc_of_init(struct device_node *intc,
+					     struct device_node *parent)
+{
+	u32 nr_irq, intr_mask;
+	int ret;
+
+	intc_baseaddr = of_iomap(intc, 0);
+	BUG_ON(!intc_baseaddr);
+
+	ret = of_property_read_u32(intc, "xlnx,num-intr-inputs", &nr_irq);
+	if (ret < 0) {
+		pr_err("%s: unable to read xlnx,num-intr-inputs\n", __func__);
+		return ret;
+	}
+
+	ret = of_property_read_u32(intc, "xlnx,kind-of-intr", &intr_mask);
+	if (ret < 0) {
+		pr_err("%s: unable to read xlnx,kind-of-intr\n", __func__);
+		return ret;
+	}
+
+	if (intr_mask >> nr_irq)
+		pr_warn("%s: mismatch in kind-of-intr param\n", __func__);
+
+	pr_info("%s: num_irq=%d, edge=0x%x\n",
+		intc->full_name, nr_irq, intr_mask);
+
+	write_fn = intc_write32;
+	read_fn = intc_read32;
+
+	/*
+	 * Disable all external interrupts until they are
+	 * explicity requested.
+	 */
+	write_fn(0, intc_baseaddr + IER);
+
+	/* Acknowledge any pending interrupts just in case. */
+	write_fn(0xffffffff, intc_baseaddr + IAR);
+
+	/* Turn on the Master Enable. */
+	write_fn(MER_HIE | MER_ME, intc_baseaddr + MER);
+	if (!(read_fn(intc_baseaddr + MER) & (MER_HIE | MER_ME))) {
+		write_fn = intc_write32_be;
+		read_fn = intc_read32_be;
+		write_fn(MER_HIE | MER_ME, intc_baseaddr + MER);
+	}
+
+	/* Yeah, okay, casting the intr_mask to a void* is butt-ugly, but I'm
+	 * lazy and Michal can clean it up to something nicer when he tests
+	 * and commits this patch.  ~~gcl */
+	root_domain = irq_domain_add_linear(intc, nr_irq, &xintc_irq_domain_ops,
+							(void *)intr_mask);
+
+	irq_set_default_host(root_domain);
+
+	return 0;
+}
+
+IRQCHIP_DECLARE(xilinx_intc, "xlnx,xps-intc-1.00.a", xilinx_intc_of_init);
diff --git a/arch/microblaze/kernel/irq.c b/arch/microblaze/kernel/irq.c
new file mode 100644
index 0000000..11e24de
--- /dev/null
+++ b/arch/microblaze/kernel/irq.c
@@ -0,0 +1,53 @@
+/*
+ * Copyright (C) 2007-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/init.h>
+#include <linux/ftrace.h>
+#include <linux/kernel.h>
+#include <linux/hardirq.h>
+#include <linux/interrupt.h>
+#include <linux/irqflags.h>
+#include <linux/seq_file.h>
+#include <linux/kernel_stat.h>
+#include <linux/irq.h>
+#include <linux/irqchip.h>
+#include <linux/of_irq.h>
+
+static u32 concurrent_irq;
+
+void __irq_entry do_IRQ(struct pt_regs *regs)
+{
+	unsigned int irq;
+	struct pt_regs *old_regs = set_irq_regs(regs);
+	trace_hardirqs_off();
+
+	irq_enter();
+	irq = get_irq();
+next_irq:
+	BUG_ON(!irq);
+	generic_handle_irq(irq);
+
+	irq = get_irq();
+	if (irq != -1U) {
+		pr_debug("next irq: %d\n", irq);
+		++concurrent_irq;
+		goto next_irq;
+	}
+
+	irq_exit();
+	set_irq_regs(old_regs);
+	trace_hardirqs_on();
+}
+
+void __init init_IRQ(void)
+{
+	/* process the entire interrupt tree in one go */
+	irqchip_init();
+}
diff --git a/arch/microblaze/kernel/kgdb.c b/arch/microblaze/kernel/kgdb.c
new file mode 100644
index 0000000..6366f69
--- /dev/null
+++ b/arch/microblaze/kernel/kgdb.c
@@ -0,0 +1,152 @@
+/*
+ * Microblaze KGDB support
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/kgdb.h>
+#include <linux/kdebug.h>
+#include <linux/irq.h>
+#include <linux/io.h>
+#include <asm/cacheflush.h>
+#include <asm/asm-offsets.h>
+#include <asm/kgdb.h>
+#include <asm/pvr.h>
+
+#define GDB_REG		0
+#define GDB_PC		32
+#define GDB_MSR		33
+#define GDB_EAR		34
+#define GDB_ESR		35
+#define GDB_FSR		36
+#define GDB_BTR		37
+#define GDB_PVR		38
+#define GDB_REDR	50
+#define GDB_RPID	51
+#define GDB_RZPR	52
+#define GDB_RTLBX	53
+#define GDB_RTLBSX	54 /* mfs can't read it */
+#define GDB_RTLBLO	55
+#define GDB_RTLBHI	56
+
+/* keep pvr separately because it is unchangeble */
+static struct pvr_s pvr;
+
+void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs)
+{
+	unsigned int i;
+	unsigned long *pt_regb = (unsigned long *)regs;
+	int temp;
+
+	/* registers r0 - r31, pc, msr, ear, esr, fsr + do not save pt_mode */
+	for (i = 0; i < (sizeof(struct pt_regs) / 4) - 1; i++)
+		gdb_regs[i] = pt_regb[i];
+
+	/* Branch target register can't be changed */
+	__asm__ __volatile__ ("mfs %0, rbtr;" : "=r"(temp) : );
+	gdb_regs[GDB_BTR] = temp;
+
+	/* pvr part  - we have 11 pvr regs */
+	for (i = 0; i < sizeof(struct pvr_s)/4; i++)
+		gdb_regs[GDB_PVR + i] = pvr.pvr[i];
+
+	/* read special registers - can't be changed */
+	__asm__ __volatile__ ("mfs %0, redr;" : "=r"(temp) : );
+	gdb_regs[GDB_REDR] = temp;
+	__asm__ __volatile__ ("mfs %0, rpid;" : "=r"(temp) : );
+	gdb_regs[GDB_RPID] = temp;
+	__asm__ __volatile__ ("mfs %0, rzpr;" : "=r"(temp) : );
+	gdb_regs[GDB_RZPR] = temp;
+	__asm__ __volatile__ ("mfs %0, rtlbx;" : "=r"(temp) : );
+	gdb_regs[GDB_RTLBX] = temp;
+	__asm__ __volatile__ ("mfs %0, rtlblo;" : "=r"(temp) : );
+	gdb_regs[GDB_RTLBLO] = temp;
+	__asm__ __volatile__ ("mfs %0, rtlbhi;" : "=r"(temp) : );
+	gdb_regs[GDB_RTLBHI] = temp;
+}
+
+void gdb_regs_to_pt_regs(unsigned long *gdb_regs, struct pt_regs *regs)
+{
+	unsigned int i;
+	unsigned long *pt_regb = (unsigned long *)regs;
+
+	/* pt_regs and gdb_regs have the same 37 values.
+	 * The rest of gdb_regs are unused and can't be changed.
+	 * r0 register value can't be changed too. */
+	for (i = 1; i < (sizeof(struct pt_regs) / 4) - 1; i++)
+		pt_regb[i] = gdb_regs[i];
+}
+
+asmlinkage void microblaze_kgdb_break(struct pt_regs *regs)
+{
+	if (kgdb_handle_exception(1, SIGTRAP, 0, regs) != 0)
+		return;
+
+	/* Jump over the first arch_kgdb_breakpoint which is barrier to
+	 * get kgdb work. The same solution is used for powerpc */
+	if (*(u32 *) (regs->pc) == *(u32 *) (&arch_kgdb_ops.gdb_bpt_instr))
+		regs->pc += BREAK_INSTR_SIZE;
+}
+
+/* untested */
+void sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *p)
+{
+	unsigned int i;
+	unsigned long *pt_regb = (unsigned long *)(p->thread.regs);
+
+	/* registers r0 - r31, pc, msr, ear, esr, fsr + do not save pt_mode */
+	for (i = 0; i < (sizeof(struct pt_regs) / 4) - 1; i++)
+		gdb_regs[i] = pt_regb[i];
+
+	/* pvr part  - we have 11 pvr regs */
+	for (i = 0; i < sizeof(struct pvr_s)/4; i++)
+		gdb_regs[GDB_PVR + i] = pvr.pvr[i];
+}
+
+void kgdb_arch_set_pc(struct pt_regs *regs, unsigned long ip)
+{
+	regs->pc = ip;
+}
+
+int kgdb_arch_handle_exception(int vector, int signo, int err_code,
+			       char *remcom_in_buffer, char *remcom_out_buffer,
+			       struct pt_regs *regs)
+{
+	char *ptr;
+	unsigned long address;
+
+	switch (remcom_in_buffer[0]) {
+	case 'c':
+		/* handle the optional parameter */
+		ptr = &remcom_in_buffer[1];
+		if (kgdb_hex2long(&ptr, &address))
+			regs->pc = address;
+
+		return 0;
+	}
+	return -1; /* this means that we do not want to exit from the handler */
+}
+
+int kgdb_arch_init(void)
+{
+	get_pvr(&pvr); /* Fill PVR structure */
+	return 0;
+}
+
+void kgdb_arch_exit(void)
+{
+	/* Nothing to do */
+}
+
+/*
+ * Global data
+ */
+struct kgdb_arch arch_kgdb_ops = {
+#ifdef __MICROBLAZEEL__
+	.gdb_bpt_instr = {0x18, 0x00, 0x0c, 0xba}, /* brki r16, 0x18 */
+#else
+	.gdb_bpt_instr = {0xba, 0x0c, 0x00, 0x18}, /* brki r16, 0x18 */
+#endif
+};
diff --git a/arch/microblaze/kernel/mcount.S b/arch/microblaze/kernel/mcount.S
new file mode 100644
index 0000000..fed9da5
--- /dev/null
+++ b/arch/microblaze/kernel/mcount.S
@@ -0,0 +1,165 @@
+/*
+ * Low-level ftrace handling
+ *
+ * Copyright (C) 2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2009 PetaLogix
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License. See the file COPYING in the main directory of this
+ * archive for more details.
+ */
+
+#include <linux/linkage.h>
+
+#define NOALIGN_ENTRY(name)	.globl name; name:
+
+/* FIXME MS: I think that I don't need to save all regs */
+#define SAVE_REGS		\
+	addik	r1, r1, -120;	\
+	swi	r2, r1, 4;	\
+	swi	r3, r1, 8;	\
+	swi	r4, r1, 12;	\
+	swi	r5, r1, 116;	\
+	swi	r6, r1, 16;	\
+	swi	r7, r1, 20;	\
+	swi	r8, r1, 24;	\
+	swi	r9, r1, 28;	\
+	swi	r10, r1, 32;	\
+	swi	r11, r1, 36;	\
+	swi	r12, r1, 40;	\
+	swi	r13, r1, 44;	\
+	swi	r14, r1, 48;	\
+	swi	r16, r1, 52;	\
+	swi	r17, r1, 56;	\
+	swi	r18, r1, 60;	\
+	swi	r19, r1, 64;	\
+	swi	r20, r1, 68;	\
+	swi	r21, r1, 72;	\
+	swi	r22, r1, 76;	\
+	swi	r23, r1, 80;	\
+	swi	r24, r1, 84;	\
+	swi	r25, r1, 88;	\
+	swi	r26, r1, 92;	\
+	swi	r27, r1, 96;	\
+	swi	r28, r1, 100;	\
+	swi	r29, r1, 104;	\
+	swi	r30, r1, 108;	\
+	swi	r31, r1, 112;
+
+#define RESTORE_REGS		\
+	lwi	r2, r1, 4;	\
+	lwi	r3, r1, 8;	\
+	lwi	r4, r1, 12;	\
+	lwi	r5, r1, 116;	\
+	lwi	r6, r1, 16;	\
+	lwi	r7, r1, 20;	\
+	lwi	r8, r1, 24;	\
+	lwi	r9, r1, 28;	\
+	lwi	r10, r1, 32;	\
+	lwi	r11, r1, 36;	\
+	lwi	r12, r1, 40;	\
+	lwi	r13, r1, 44;	\
+	lwi	r14, r1, 48;	\
+	lwi	r16, r1, 52;	\
+	lwi	r17, r1, 56;	\
+	lwi	r18, r1, 60;	\
+	lwi	r19, r1, 64;	\
+	lwi	r20, r1, 68;	\
+	lwi	r21, r1, 72;	\
+	lwi	r22, r1, 76;	\
+	lwi	r23, r1, 80;	\
+	lwi	r24, r1, 84;	\
+	lwi	r25, r1, 88;	\
+	lwi	r26, r1, 92;	\
+	lwi	r27, r1, 96;	\
+	lwi	r28, r1, 100;	\
+	lwi	r29, r1, 104;	\
+	lwi	r30, r1, 108;	\
+	lwi	r31, r1, 112;	\
+	addik	r1, r1, 120;
+
+ENTRY(ftrace_stub)
+	rtsd	r15, 8;
+	nop;
+
+ENTRY(_mcount)
+#ifdef CONFIG_DYNAMIC_FTRACE
+ENTRY(ftrace_caller)
+	/* MS: It is just barrier which is removed from C code */
+	rtsd	r15, 8
+	nop
+#endif /* CONFIG_DYNAMIC_FTRACE */
+	SAVE_REGS
+	swi	r15, r1, 0;
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+#ifndef CONFIG_DYNAMIC_FTRACE
+	lwi	r5, r0, ftrace_graph_return;
+	addik	r6, r0, ftrace_stub; /* asm implementation */
+	cmpu	r5, r5, r6; /* ftrace_graph_return != ftrace_stub */
+	beqid	r5, end_graph_tracer;
+	nop;
+
+	lwi	r6, r0, ftrace_graph_entry;
+	addik	r5, r0, ftrace_graph_entry_stub; /* implemented in C */
+	cmpu	r5, r5, r6; /* ftrace_graph_entry != ftrace_graph_entry_stub */
+	beqid	r5, end_graph_tracer;
+	nop;
+#else /* CONFIG_DYNAMIC_FTRACE */
+NOALIGN_ENTRY(ftrace_call_graph)
+	/* MS: jump over graph function - replaced from C code */
+	bri	end_graph_tracer
+#endif /* CONFIG_DYNAMIC_FTRACE */
+	addik	r5, r1, 120; /* MS: load parent addr */
+	addik	r6, r15, 0; /* MS: load current function addr */
+	bralid	r15, prepare_ftrace_return;
+	nop;
+	/* MS: graph was taken that's why - can jump over function trace */
+	brid	end;
+	nop;
+end_graph_tracer:
+#endif /* CONFIG_FUNCTION_GRAPH_TRACER */
+#ifndef CONFIG_DYNAMIC_FTRACE
+	/* MS: test function trace if is taken or not */
+	lwi	r20, r0, ftrace_trace_function;
+	addik	r6, r0, ftrace_stub;
+	cmpu	r5, r20, r6; /* ftrace_trace_function != ftrace_stub */
+	beqid	r5, end; /* MS: not taken -> jump over */
+	nop;
+#else /* CONFIG_DYNAMIC_FTRACE */
+NOALIGN_ENTRY(ftrace_call)
+/* instruction for setup imm FUNC_part1, addik r20, r0, FUNC_part2 */
+	nop
+	nop
+#endif /* CONFIG_DYNAMIC_FTRACE */
+/* static normal trace */
+	lwi	r6, r1, 120; /* MS: load parent addr */
+	addik	r5, r15, -4; /* MS: load current function addr */
+	/* MS: here is dependency on previous code */
+	brald	r15, r20; /* MS: jump to ftrace handler */
+	nop;
+end:
+	lwi	r15, r1, 0;
+	RESTORE_REGS
+
+	rtsd	r15, 8; /* MS: jump back */
+	nop;
+
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+ENTRY(return_to_handler)
+	nop; /* MS: just barrier for rtsd r15, 8 */
+	nop;
+	SAVE_REGS
+	swi	r15, r1, 0;
+
+	/* MS: find out returning address */
+	bralid	r15, ftrace_return_to_handler;
+	nop;
+
+	/* MS: return value from ftrace_return_to_handler is my returning addr
+	 * must be before restore regs because I have to restore r3 content */
+	addik	r15, r3, 0;
+	RESTORE_REGS
+
+	rtsd	r15, 8; /* MS: jump back */
+	nop;
+#endif	/* CONFIG_FUNCTION_TRACER */
diff --git a/arch/microblaze/kernel/microblaze_ksyms.c b/arch/microblaze/kernel/microblaze_ksyms.c
new file mode 100644
index 0000000..9f1d02c
--- /dev/null
+++ b/arch/microblaze/kernel/microblaze_ksyms.c
@@ -0,0 +1,55 @@
+/*
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ *
+ * 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/export.h>
+#include <linux/string.h>
+#include <linux/cryptohash.h>
+#include <linux/delay.h>
+#include <linux/in6.h>
+#include <linux/syscalls.h>
+
+#include <asm/checksum.h>
+#include <asm/cacheflush.h>
+#include <linux/io.h>
+#include <asm/page.h>
+#include <linux/ftrace.h>
+#include <linux/uaccess.h>
+
+#ifdef CONFIG_FUNCTION_TRACER
+extern void _mcount(void);
+EXPORT_SYMBOL(_mcount);
+#endif
+
+/*
+ * Assembly functions that may be used (directly or indirectly) by modules
+ */
+EXPORT_SYMBOL(__copy_tofrom_user);
+EXPORT_SYMBOL(__strncpy_user);
+
+#ifdef CONFIG_OPT_LIB_ASM
+EXPORT_SYMBOL(memcpy);
+EXPORT_SYMBOL(memmove);
+#endif
+
+#ifdef CONFIG_MMU
+EXPORT_SYMBOL(empty_zero_page);
+#endif
+
+EXPORT_SYMBOL(mbc);
+
+extern void __divsi3(void);
+EXPORT_SYMBOL(__divsi3);
+extern void __modsi3(void);
+EXPORT_SYMBOL(__modsi3);
+extern void __mulsi3(void);
+EXPORT_SYMBOL(__mulsi3);
+extern void __udivsi3(void);
+EXPORT_SYMBOL(__udivsi3);
+extern void __umodsi3(void);
+EXPORT_SYMBOL(__umodsi3);
diff --git a/arch/microblaze/kernel/misc.S b/arch/microblaze/kernel/misc.S
new file mode 100644
index 0000000..1dafdde
--- /dev/null
+++ b/arch/microblaze/kernel/misc.S
@@ -0,0 +1,100 @@
+/*
+ * Miscellaneous low-level MMU functions.
+ *
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ * Copyright (C) 2007 Xilinx, Inc.  All rights reserved.
+ *
+ * Derived from arch/ppc/kernel/misc.S
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License. See the file COPYING in the main directory of this
+ * archive for more details.
+ */
+
+#include <linux/linkage.h>
+#include <linux/sys.h>
+#include <asm/unistd.h>
+#include <linux/errno.h>
+#include <asm/mmu.h>
+#include <asm/page.h>
+
+	.text
+/*
+ * Flush MMU TLB
+ *
+ * We avoid flushing the pinned 0, 1 and possibly 2 entries.
+ */
+.globl _tlbia;
+.type  _tlbia, @function
+.align 4;
+_tlbia:
+	lwi	r12, r0, tlb_skip;
+	/* isync */
+_tlbia_1:
+	mts	rtlbx, r12
+	nop
+	mts	rtlbhi, r0 /* flush: ensure V is clear */
+	nop
+	rsubi	r11, r12, MICROBLAZE_TLB_SIZE - 1
+	bneid	r11, _tlbia_1 /* loop for all entries */
+	addik	r12, r12, 1
+	/* sync */
+	rtsd	r15, 8
+	nop
+	.size  _tlbia, . - _tlbia
+
+/*
+ * Flush MMU TLB for a particular address (in r5)
+ */
+.globl _tlbie;
+.type  _tlbie, @function
+.align 4;
+_tlbie:
+	mts	rtlbsx, r5 /* look up the address in TLB */
+	nop
+	mfs	r12, rtlbx /* Retrieve index */
+	nop
+	blti	r12, _tlbie_1 /* Check if found */
+	mts	rtlbhi, r0 /* flush: ensure V is clear */
+	nop
+_tlbie_1:
+	rtsd	r15, 8
+	nop
+
+	.size  _tlbie, . - _tlbie
+
+/*
+ * Allocate TLB entry for early console
+ */
+.globl early_console_reg_tlb_alloc;
+.type  early_console_reg_tlb_alloc, @function
+.align 4;
+early_console_reg_tlb_alloc:
+	/*
+	 * Load a TLB entry for the UART, so that microblaze_progress() can use
+	 * the UARTs nice and early.  We use a 4k real==virtual mapping.
+	 */
+	lwi	r4, r0, tlb_skip
+	mts	rtlbx, r4 /* TLB slot 63 */
+
+	or	r4,r5,r0
+	andi	r4,r4,0xfffff000
+	ori	r4,r4,(TLB_WR|TLB_I|TLB_M|TLB_G)
+
+	andi	r5,r5,0xfffff000
+	ori	r5,r5,(TLB_VALID | TLB_PAGESZ(PAGESZ_4K))
+
+	mts	rtlblo,r4 /* Load the data portion of the entry */
+	nop
+	mts	rtlbhi,r5 /* Load the tag portion of the entry */
+	nop
+
+	lwi	r5, r0, tlb_skip
+	addik	r5, r5, 1
+	swi	r5, r0, tlb_skip
+
+	rtsd	r15, 8
+	nop
+
+	.size  early_console_reg_tlb_alloc, . - early_console_reg_tlb_alloc
diff --git a/arch/microblaze/kernel/module.c b/arch/microblaze/kernel/module.c
new file mode 100644
index 0000000..182e6be
--- /dev/null
+++ b/arch/microblaze/kernel/module.c
@@ -0,0 +1,125 @@
+/*
+ * Copyright (C) 2007-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007-2009 PetaLogix
+ *
+ * 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/export.h>
+#include <linux/moduleloader.h>
+#include <linux/kernel.h>
+#include <linux/elf.h>
+#include <linux/vmalloc.h>
+#include <linux/fs.h>
+#include <linux/string.h>
+
+#include <asm/pgtable.h>
+#include <asm/cacheflush.h>
+
+int apply_relocate_add(Elf32_Shdr *sechdrs, const char *strtab,
+	unsigned int symindex, unsigned int relsec, struct module *module)
+{
+
+	unsigned int i;
+	Elf32_Rela *rela = (void *)sechdrs[relsec].sh_addr;
+	Elf32_Sym *sym;
+	unsigned long int *location;
+	unsigned long int value;
+#if __GNUC__ < 4
+	unsigned long int old_value;
+#endif
+
+	pr_debug("Applying add relocation section %u to %u\n",
+		relsec, sechdrs[relsec].sh_info);
+
+	for (i = 0; i < sechdrs[relsec].sh_size / sizeof(*rela); i++) {
+
+		location = (void *)sechdrs[sechdrs[relsec].sh_info].sh_addr +
+				rela[i].r_offset;
+		sym = (Elf32_Sym *)sechdrs[symindex].sh_addr +
+			ELF32_R_SYM(rela[i].r_info);
+		value = sym->st_value + rela[i].r_addend;
+
+		switch (ELF32_R_TYPE(rela[i].r_info)) {
+
+		/*
+		 * Be careful! mb-gcc / mb-ld splits the relocs between the
+		 * text and the reloc table. In general this means we must
+		 * read the current contents of (*location), add any offset
+		 * then store the result back in
+		 */
+
+		case R_MICROBLAZE_32:
+#if __GNUC__ < 4
+			old_value = *location;
+			*location = value + old_value;
+
+			pr_debug("R_MICROBLAZE_32 (%08lx->%08lx)\n",
+				old_value, value);
+#else
+			*location = value;
+#endif
+			break;
+
+		case R_MICROBLAZE_64:
+#if __GNUC__ < 4
+			/* Split relocs only required/used pre gcc4.1.1 */
+			old_value = ((location[0] & 0x0000FFFF) << 16) |
+					(location[1] & 0x0000FFFF);
+			value += old_value;
+#endif
+			location[0] = (location[0] & 0xFFFF0000) |
+					(value >> 16);
+			location[1] = (location[1] & 0xFFFF0000) |
+					(value & 0xFFFF);
+#if __GNUC__ < 4
+			pr_debug("R_MICROBLAZE_64 (%08lx->%08lx)\n",
+				old_value, value);
+#endif
+			break;
+
+		case R_MICROBLAZE_64_PCREL:
+#if __GNUC__ < 4
+			old_value = (location[0] & 0xFFFF) << 16 |
+				(location[1] & 0xFFFF);
+			value -= old_value;
+#endif
+			value -= (unsigned long int)(location) + 4;
+			location[0] = (location[0] & 0xFFFF0000) |
+					(value >> 16);
+			location[1] = (location[1] & 0xFFFF0000) |
+					(value & 0xFFFF);
+			pr_debug("R_MICROBLAZE_64_PCREL (%08lx)\n",
+				value);
+			break;
+
+		case R_MICROBLAZE_32_PCREL_LO:
+			pr_debug("R_MICROBLAZE_32_PCREL_LO\n");
+			break;
+
+		case R_MICROBLAZE_64_NONE:
+			pr_debug("R_MICROBLAZE_64_NONE\n");
+			break;
+
+		case R_MICROBLAZE_NONE:
+			pr_debug("R_MICROBLAZE_NONE\n");
+			break;
+
+		default:
+			pr_err("module %s: Unknown relocation: %u\n",
+				module->name,
+				ELF32_R_TYPE(rela[i].r_info));
+			return -ENOEXEC;
+		}
+	}
+	return 0;
+}
+
+int module_finalize(const Elf32_Ehdr *hdr, const Elf_Shdr *sechdrs,
+		struct module *module)
+{
+	flush_dcache();
+	return 0;
+}
diff --git a/arch/microblaze/kernel/platform.c b/arch/microblaze/kernel/platform.c
new file mode 100644
index 0000000..b9529ca
--- /dev/null
+++ b/arch/microblaze/kernel/platform.c
@@ -0,0 +1,30 @@
+/*
+ * Copyright 2008 Michal Simek <monstr@monstr.eu>
+ *
+ * based on virtex.c file
+ *
+ * Copyright 2007 Secret Lab Technologies Ltd.
+ *
+ * This file is licensed under the terms of the GNU General Public License
+ * version 2. This program is licensed "as is" without any warranty of any
+ * kind, whether express or implied.
+ */
+
+#include <linux/init.h>
+#include <linux/of_platform.h>
+#include <asm/prom.h>
+#include <asm/setup.h>
+
+static struct of_device_id xilinx_of_bus_ids[] __initdata = {
+	{ .compatible = "simple-bus", },
+	{ .compatible = "xlnx,compound", },
+	{}
+};
+
+static int __init microblaze_device_probe(void)
+{
+	of_platform_bus_probe(NULL, xilinx_of_bus_ids, NULL);
+	of_platform_reset_gpio_probe();
+	return 0;
+}
+device_initcall(microblaze_device_probe);
diff --git a/arch/microblaze/kernel/process.c b/arch/microblaze/kernel/process.c
new file mode 100644
index 0000000..b2dd371
--- /dev/null
+++ b/arch/microblaze/kernel/process.c
@@ -0,0 +1,168 @@
+/*
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/cpu.h>
+#include <linux/export.h>
+#include <linux/sched.h>
+#include <linux/pm.h>
+#include <linux/tick.h>
+#include <linux/bitops.h>
+#include <linux/ptrace.h>
+#include <asm/pgalloc.h>
+#include <linux/uaccess.h> /* for USER_DS macros */
+#include <asm/cacheflush.h>
+
+void show_regs(struct pt_regs *regs)
+{
+	show_regs_print_info(KERN_INFO);
+
+	pr_info(" Registers dump: mode=%X\r\n", regs->pt_mode);
+	pr_info(" r1=%08lX, r2=%08lX, r3=%08lX, r4=%08lX\n",
+				regs->r1, regs->r2, regs->r3, regs->r4);
+	pr_info(" r5=%08lX, r6=%08lX, r7=%08lX, r8=%08lX\n",
+				regs->r5, regs->r6, regs->r7, regs->r8);
+	pr_info(" r9=%08lX, r10=%08lX, r11=%08lX, r12=%08lX\n",
+				regs->r9, regs->r10, regs->r11, regs->r12);
+	pr_info(" r13=%08lX, r14=%08lX, r15=%08lX, r16=%08lX\n",
+				regs->r13, regs->r14, regs->r15, regs->r16);
+	pr_info(" r17=%08lX, r18=%08lX, r19=%08lX, r20=%08lX\n",
+				regs->r17, regs->r18, regs->r19, regs->r20);
+	pr_info(" r21=%08lX, r22=%08lX, r23=%08lX, r24=%08lX\n",
+				regs->r21, regs->r22, regs->r23, regs->r24);
+	pr_info(" r25=%08lX, r26=%08lX, r27=%08lX, r28=%08lX\n",
+				regs->r25, regs->r26, regs->r27, regs->r28);
+	pr_info(" r29=%08lX, r30=%08lX, r31=%08lX, rPC=%08lX\n",
+				regs->r29, regs->r30, regs->r31, regs->pc);
+	pr_info(" msr=%08lX, ear=%08lX, esr=%08lX, fsr=%08lX\n",
+				regs->msr, regs->ear, regs->esr, regs->fsr);
+}
+
+void (*pm_power_off)(void) = NULL;
+EXPORT_SYMBOL(pm_power_off);
+
+void flush_thread(void)
+{
+}
+
+int copy_thread(unsigned long clone_flags, unsigned long usp,
+		unsigned long arg, struct task_struct *p)
+{
+	struct pt_regs *childregs = task_pt_regs(p);
+	struct thread_info *ti = task_thread_info(p);
+
+	if (unlikely(p->flags & PF_KTHREAD)) {
+		/* if we're creating a new kernel thread then just zeroing all
+		 * the registers. That's OK for a brand new thread.*/
+		memset(childregs, 0, sizeof(struct pt_regs));
+		memset(&ti->cpu_context, 0, sizeof(struct cpu_context));
+		ti->cpu_context.r1  = (unsigned long)childregs;
+		ti->cpu_context.r20 = (unsigned long)usp; /* fn */
+		ti->cpu_context.r19 = (unsigned long)arg;
+		childregs->pt_mode = 1;
+		local_save_flags(childregs->msr);
+#ifdef CONFIG_MMU
+		ti->cpu_context.msr = childregs->msr & ~MSR_IE;
+#endif
+		ti->cpu_context.r15 = (unsigned long)ret_from_kernel_thread - 8;
+		return 0;
+	}
+	*childregs = *current_pt_regs();
+	if (usp)
+		childregs->r1 = usp;
+
+	memset(&ti->cpu_context, 0, sizeof(struct cpu_context));
+	ti->cpu_context.r1 = (unsigned long)childregs;
+#ifndef CONFIG_MMU
+	ti->cpu_context.msr = (unsigned long)childregs->msr;
+#else
+	childregs->msr |= MSR_UMS;
+
+	/* we should consider the fact that childregs is a copy of the parent
+	 * regs which were saved immediately after entering the kernel state
+	 * before enabling VM. This MSR will be restored in switch_to and
+	 * RETURN() and we want to have the right machine state there
+	 * specifically this state must have INTs disabled before and enabled
+	 * after performing rtbd
+	 * compose the right MSR for RETURN(). It will work for switch_to also
+	 * excepting for VM and UMS
+	 * don't touch UMS , CARRY and cache bits
+	 * right now MSR is a copy of parent one */
+	childregs->msr &= ~MSR_EIP;
+	childregs->msr |= MSR_IE;
+	childregs->msr &= ~MSR_VM;
+	childregs->msr |= MSR_VMS;
+	childregs->msr |= MSR_EE; /* exceptions will be enabled*/
+
+	ti->cpu_context.msr = (childregs->msr|MSR_VM);
+	ti->cpu_context.msr &= ~MSR_UMS; /* switch_to to kernel mode */
+	ti->cpu_context.msr &= ~MSR_IE;
+#endif
+	ti->cpu_context.r15 = (unsigned long)ret_from_fork - 8;
+
+	/*
+	 *  r21 is the thread reg, r10 is 6th arg to clone
+	 *  which contains TLS area
+	 */
+	if (clone_flags & CLONE_SETTLS)
+		childregs->r21 = childregs->r10;
+
+	return 0;
+}
+
+#ifndef CONFIG_MMU
+/*
+ * Return saved PC of a blocked thread.
+ */
+unsigned long thread_saved_pc(struct task_struct *tsk)
+{
+	struct cpu_context *ctx =
+		&(((struct thread_info *)(tsk->stack))->cpu_context);
+
+	/* Check whether the thread is blocked in resume() */
+	if (in_sched_functions(ctx->r15))
+		return (unsigned long)ctx->r15;
+	else
+		return ctx->r14;
+}
+#endif
+
+unsigned long get_wchan(struct task_struct *p)
+{
+/* TBD (used by procfs) */
+	return 0;
+}
+
+/* Set up a thread for executing a new program */
+void start_thread(struct pt_regs *regs, unsigned long pc, unsigned long usp)
+{
+	regs->pc = pc;
+	regs->r1 = usp;
+	regs->pt_mode = 0;
+#ifdef CONFIG_MMU
+	regs->msr |= MSR_UMS;
+	regs->msr &= ~MSR_VM;
+#endif
+}
+
+#ifdef CONFIG_MMU
+#include <linux/elfcore.h>
+/*
+ * Set up a thread for executing a new program
+ */
+int dump_fpu(struct pt_regs *regs, elf_fpregset_t *fpregs)
+{
+	return 0; /* MicroBlaze has no separate FPU registers */
+}
+#endif /* CONFIG_MMU */
+
+void arch_cpu_idle(void)
+{
+       local_irq_enable();
+}
diff --git a/arch/microblaze/kernel/prom.c b/arch/microblaze/kernel/prom.c
new file mode 100644
index 0000000..68f0999
--- /dev/null
+++ b/arch/microblaze/kernel/prom.c
@@ -0,0 +1,116 @@
+/*
+ * Procedures for creating, accessing and interpreting the device tree.
+ *
+ * Paul Mackerras	August 1996.
+ * Copyright (C) 1996-2005 Paul Mackerras.
+ *
+ *  Adapted for 64bit PowerPC by Dave Engebretsen and Peter Bergner.
+ *    {engebret|bergner}@us.ibm.com
+ *
+ *      This program is free software; you can redistribute it and/or
+ *      modify it under the terms of the GNU General Public License
+ *      as published by the Free Software Foundation; either version
+ *      2 of the License, or (at your option) any later version.
+ */
+
+#include <stdarg.h>
+#include <linux/export.h>
+#include <linux/kernel.h>
+#include <linux/string.h>
+#include <linux/init.h>
+#include <linux/threads.h>
+#include <linux/spinlock.h>
+#include <linux/types.h>
+#include <linux/pci.h>
+#include <linux/stringify.h>
+#include <linux/delay.h>
+#include <linux/initrd.h>
+#include <linux/bitops.h>
+#include <linux/kexec.h>
+#include <linux/debugfs.h>
+#include <linux/irq.h>
+#include <linux/memblock.h>
+#include <linux/of_fdt.h>
+
+#include <asm/prom.h>
+#include <asm/page.h>
+#include <asm/processor.h>
+#include <asm/irq.h>
+#include <linux/io.h>
+#include <asm/mmu.h>
+#include <asm/pgtable.h>
+#include <asm/sections.h>
+#include <asm/pci-bridge.h>
+
+#ifdef CONFIG_EARLY_PRINTK
+static const char *stdout;
+
+static int __init early_init_dt_scan_chosen_serial(unsigned long node,
+				const char *uname, int depth, void *data)
+{
+	int l;
+	const char *p;
+
+	pr_debug("%s: depth: %d, uname: %s\n", __func__, depth, uname);
+
+	if (depth == 1 && (strcmp(uname, "chosen") == 0 ||
+				strcmp(uname, "chosen@0") == 0)) {
+		p = of_get_flat_dt_prop(node, "linux,stdout-path", &l);
+		if (p != NULL && l > 0)
+			stdout = p; /* store pointer to stdout-path */
+	}
+
+	if (stdout && strstr(stdout, uname)) {
+		p = of_get_flat_dt_prop(node, "compatible", &l);
+		pr_debug("Compatible string: %s\n", p);
+
+		if ((strncmp(p, "xlnx,xps-uart16550", 18) == 0) ||
+			(strncmp(p, "xlnx,axi-uart16550", 18) == 0)) {
+			unsigned int addr;
+
+			*(u32 *)data = UART16550;
+
+			addr = *(u32 *)of_get_flat_dt_prop(node, "reg", &l);
+			addr += *(u32 *)of_get_flat_dt_prop(node,
+							"reg-offset", &l);
+			/* clear register offset */
+			return be32_to_cpu(addr) & ~3;
+		}
+		if ((strncmp(p, "xlnx,xps-uartlite", 17) == 0) ||
+				(strncmp(p, "xlnx,opb-uartlite", 17) == 0) ||
+				(strncmp(p, "xlnx,axi-uartlite", 17) == 0) ||
+				(strncmp(p, "xlnx,mdm", 8) == 0)) {
+			const unsigned int *addrp;
+
+			*(u32 *)data = UARTLITE;
+
+			addrp = of_get_flat_dt_prop(node, "reg", &l);
+			return be32_to_cpup(addrp); /* return address */
+		}
+	}
+	return 0;
+}
+
+/* this function is looking for early console - Microblaze specific */
+int __init of_early_console(void *version)
+{
+	return of_scan_flat_dt(early_init_dt_scan_chosen_serial, version);
+}
+#endif
+
+void __init early_init_devtree(void *params)
+{
+	pr_debug(" -> early_init_devtree(%p)\n", params);
+
+	early_init_dt_scan(params);
+	if (!strlen(boot_command_line))
+		strlcpy(boot_command_line, cmd_line, COMMAND_LINE_SIZE);
+
+	parse_early_param();
+
+	memblock_allow_resize();
+
+	pr_debug("Phys. mem: %lx\n", (unsigned long) memblock_phys_mem_size());
+
+	pr_debug(" <- early_init_devtree()\n");
+}
diff --git a/arch/microblaze/kernel/ptrace.c b/arch/microblaze/kernel/ptrace.c
new file mode 100644
index 0000000..8cfa98c
--- /dev/null
+++ b/arch/microblaze/kernel/ptrace.c
@@ -0,0 +1,169 @@
+/*
+ * `ptrace' system call
+ *
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007-2009 PetaLogix
+ * Copyright (C) 2004-2007 John Williams <john.williams@petalogix.com>
+ *
+ * derived from arch/v850/kernel/ptrace.c
+ *
+ * Copyright (C) 2002,03 NEC Electronics Corporation
+ * Copyright (C) 2002,03 Miles Bader <miles@gnu.org>
+ *
+ * Derived from arch/mips/kernel/ptrace.c:
+ *
+ * Copyright (C) 1992 Ross Biro
+ * Copyright (C) Linus Torvalds
+ * Copyright (C) 1994, 95, 96, 97, 98, 2000 Ralf Baechle
+ * Copyright (C) 1996 David S. Miller
+ * Kevin D. Kissell, kevink@mips.com and Carsten Langgaard, carstenl@mips.com
+ * Copyright (C) 1999 MIPS Technologies, Inc.
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License. See the file COPYING in the main directory of this
+ * archive for more details.
+ */
+
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/sched.h>
+#include <linux/ptrace.h>
+#include <linux/signal.h>
+#include <linux/elf.h>
+#include <linux/audit.h>
+#include <linux/seccomp.h>
+#include <linux/tracehook.h>
+
+#include <linux/errno.h>
+#include <asm/processor.h>
+#include <linux/uaccess.h>
+#include <asm/asm-offsets.h>
+#include <asm/cacheflush.h>
+#include <asm/syscall.h>
+#include <linux/io.h>
+
+/* Returns the address where the register at REG_OFFS in P is stashed away. */
+static microblaze_reg_t *reg_save_addr(unsigned reg_offs,
+					struct task_struct *t)
+{
+	struct pt_regs *regs;
+
+	/*
+	 * Three basic cases:
+	 *
+	 * (1)	A register normally saved before calling the scheduler, is
+	 *	available in the kernel entry pt_regs structure at the top
+	 *	of the kernel stack. The kernel trap/irq exit path takes
+	 *	care to save/restore almost all registers for ptrace'd
+	 *	processes.
+	 *
+	 * (2)	A call-clobbered register, where the process P entered the
+	 *	kernel via [syscall] trap, is not stored anywhere; that's
+	 *	OK, because such registers are not expected to be preserved
+	 *	when the trap returns anyway (so we don't actually bother to
+	 *	test for this case).
+	 *
+	 * (3)	A few registers not used at all by the kernel, and so
+	 *	normally never saved except by context-switches, are in the
+	 *	context switch state.
+	 */
+
+	/* Register saved during kernel entry (or not available). */
+	regs = task_pt_regs(t);
+
+	return (microblaze_reg_t *)((char *)regs + reg_offs);
+}
+
+long arch_ptrace(struct task_struct *child, long request,
+		 unsigned long addr, unsigned long data)
+{
+	int rval;
+	unsigned long val = 0;
+
+	switch (request) {
+	/* Read/write the word at location ADDR in the registers. */
+	case PTRACE_PEEKUSR:
+	case PTRACE_POKEUSR:
+		pr_debug("PEEKUSR/POKEUSR : 0x%08lx\n", addr);
+		rval = 0;
+		if (addr >= PT_SIZE && request == PTRACE_PEEKUSR) {
+			/*
+			 * Special requests that don't actually correspond
+			 * to offsets in struct pt_regs.
+			 */
+			if (addr == PT_TEXT_ADDR) {
+				val = child->mm->start_code;
+			} else if (addr == PT_DATA_ADDR) {
+				val = child->mm->start_data;
+			} else if (addr == PT_TEXT_LEN) {
+				val = child->mm->end_code
+					- child->mm->start_code;
+			} else {
+				rval = -EIO;
+			}
+		} else if (addr < PT_SIZE && (addr & 0x3) == 0) {
+			microblaze_reg_t *reg_addr = reg_save_addr(addr, child);
+			if (request == PTRACE_PEEKUSR)
+				val = *reg_addr;
+			else {
+#if 1
+				*reg_addr = data;
+#else
+				/* MS potential problem on WB system
+				 * Be aware that reg_addr is virtual address
+				 * virt_to_phys conversion is necessary.
+				 * This could be sensible solution.
+				 */
+				u32 paddr = virt_to_phys((u32)reg_addr);
+				invalidate_icache_range(paddr, paddr + 4);
+				*reg_addr = data;
+				flush_dcache_range(paddr, paddr + 4);
+#endif
+			}
+		} else
+			rval = -EIO;
+
+		if (rval == 0 && request == PTRACE_PEEKUSR)
+			rval = put_user(val, (unsigned long __user *)data);
+		break;
+	default:
+		rval = ptrace_request(child, request, addr, data);
+	}
+	return rval;
+}
+
+asmlinkage unsigned long do_syscall_trace_enter(struct pt_regs *regs)
+{
+	unsigned long ret = 0;
+
+	secure_computing_strict(regs->r12);
+
+	if (test_thread_flag(TIF_SYSCALL_TRACE) &&
+	    tracehook_report_syscall_entry(regs))
+		/*
+		 * Tracing decided this syscall should not happen.
+		 * We'll return a bogus call number to get an ENOSYS
+		 * error, but leave the original number in regs->regs[0].
+		 */
+		ret = -1L;
+
+	audit_syscall_entry(regs->r12, regs->r5, regs->r6, regs->r7, regs->r8);
+
+	return ret ?: regs->r12;
+}
+
+asmlinkage void do_syscall_trace_leave(struct pt_regs *regs)
+{
+	int step;
+
+	audit_syscall_exit(regs);
+
+	step = test_thread_flag(TIF_SINGLESTEP);
+	if (step || test_thread_flag(TIF_SYSCALL_TRACE))
+		tracehook_report_syscall_exit(regs, step);
+}
+
+void ptrace_disable(struct task_struct *child)
+{
+	/* nothing to do */
+}
diff --git a/arch/microblaze/kernel/reset.c b/arch/microblaze/kernel/reset.c
new file mode 100644
index 0000000..bab4c83
--- /dev/null
+++ b/arch/microblaze/kernel/reset.c
@@ -0,0 +1,107 @@
+/*
+ * Copyright (C) 2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2009 PetaLogix
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/init.h>
+#include <linux/of_platform.h>
+
+/* Trigger specific functions */
+#ifdef CONFIG_GPIOLIB
+
+#include <linux/of_gpio.h>
+
+static int handle; /* reset pin handle */
+static unsigned int reset_val;
+
+void of_platform_reset_gpio_probe(void)
+{
+	int ret;
+	handle = of_get_named_gpio(of_find_node_by_path("/"),
+				   "hard-reset-gpios", 0);
+
+	if (!gpio_is_valid(handle)) {
+		pr_info("Skipping unavailable RESET gpio %d (%s)\n",
+				handle, "reset");
+		return;
+	}
+
+	ret = gpio_request(handle, "reset");
+	if (ret < 0) {
+		pr_info("GPIO pin is already allocated\n");
+		return;
+	}
+
+	/* get current setup value */
+	reset_val = gpio_get_value(handle);
+	/* FIXME maybe worth to perform any action */
+	pr_debug("Reset: Gpio output state: 0x%x\n", reset_val);
+
+	/* Setup GPIO as output */
+	ret = gpio_direction_output(handle, 0);
+	if (ret < 0)
+		goto err;
+
+	/* Setup output direction */
+	gpio_set_value(handle, 0);
+
+	pr_info("RESET: Registered gpio device: %d, current val: %d\n",
+							handle, reset_val);
+	return;
+err:
+	gpio_free(handle);
+	return;
+}
+
+
+static void gpio_system_reset(void)
+{
+	if (gpio_is_valid(handle))
+		gpio_set_value(handle, 1 - reset_val);
+	else
+		pr_notice("Reset GPIO unavailable - halting!\n");
+}
+#else
+static void gpio_system_reset(void)
+{
+	pr_notice("No reset GPIO present - halting!\n");
+}
+
+void of_platform_reset_gpio_probe(void)
+{
+	return;
+}
+#endif
+
+void machine_restart(char *cmd)
+{
+	pr_notice("Machine restart...\n");
+	gpio_system_reset();
+	while (1)
+		;
+}
+
+void machine_shutdown(void)
+{
+	pr_notice("Machine shutdown...\n");
+	while (1)
+		;
+}
+
+void machine_halt(void)
+{
+	pr_notice("Machine halt...\n");
+	while (1)
+		;
+}
+
+void machine_power_off(void)
+{
+	pr_notice("Machine power off...\n");
+	while (1)
+		;
+}
diff --git a/arch/microblaze/kernel/setup.c b/arch/microblaze/kernel/setup.c
new file mode 100644
index 0000000..89a2a93
--- /dev/null
+++ b/arch/microblaze/kernel/setup.c
@@ -0,0 +1,227 @@
+/*
+ * Copyright (C) 2007-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/init.h>
+#include <linux/clk-provider.h>
+#include <linux/clocksource.h>
+#include <linux/string.h>
+#include <linux/seq_file.h>
+#include <linux/cpu.h>
+#include <linux/initrd.h>
+#include <linux/console.h>
+#include <linux/debugfs.h>
+#include <linux/of_fdt.h>
+
+#include <asm/setup.h>
+#include <asm/sections.h>
+#include <asm/page.h>
+#include <linux/io.h>
+#include <linux/bug.h>
+#include <linux/param.h>
+#include <linux/pci.h>
+#include <linux/cache.h>
+#include <linux/of_platform.h>
+#include <linux/dma-mapping.h>
+#include <asm/cacheflush.h>
+#include <asm/entry.h>
+#include <asm/cpuinfo.h>
+
+#include <asm/prom.h>
+#include <asm/pgtable.h>
+
+DEFINE_PER_CPU(unsigned int, KSP);	/* Saved kernel stack pointer */
+DEFINE_PER_CPU(unsigned int, KM);	/* Kernel/user mode */
+DEFINE_PER_CPU(unsigned int, ENTRY_SP);	/* Saved SP on kernel entry */
+DEFINE_PER_CPU(unsigned int, R11_SAVE);	/* Temp variable for entry */
+DEFINE_PER_CPU(unsigned int, CURRENT_SAVE);	/* Saved current pointer */
+
+unsigned int boot_cpuid;
+/*
+ * Placed cmd_line to .data section because can be initialized from
+ * ASM code. Default position is BSS section which is cleared
+ * in machine_early_init().
+ */
+char cmd_line[COMMAND_LINE_SIZE] __attribute__ ((section(".data")));
+
+void __init setup_arch(char **cmdline_p)
+{
+	*cmdline_p = boot_command_line;
+
+	console_verbose();
+
+	unflatten_device_tree();
+
+	setup_cpuinfo();
+
+	microblaze_cache_init();
+
+	setup_memory();
+
+#ifdef CONFIG_EARLY_PRINTK
+	/* remap early console to virtual address */
+	remap_early_printk();
+#endif
+
+	xilinx_pci_init();
+
+#if defined(CONFIG_DUMMY_CONSOLE)
+	conswitchp = &dummy_con;
+#endif
+}
+
+#ifdef CONFIG_MTD_UCLINUX
+/* Handle both romfs and cramfs types, without generating unnecessary
+ code (ie no point checking for CRAMFS if it's not even enabled) */
+inline unsigned get_romfs_len(unsigned *addr)
+{
+#ifdef CONFIG_ROMFS_FS
+	if (memcmp(&addr[0], "-rom1fs-", 8) == 0) /* romfs */
+		return be32_to_cpu(addr[2]);
+#endif
+
+#ifdef CONFIG_CRAMFS
+	if (addr[0] == le32_to_cpu(0x28cd3d45)) /* cramfs */
+		return le32_to_cpu(addr[1]);
+#endif
+	return 0;
+}
+#endif	/* CONFIG_MTD_UCLINUX_EBSS */
+
+unsigned long kernel_tlb;
+
+void __init machine_early_init(const char *cmdline, unsigned int ram,
+		unsigned int fdt, unsigned int msr, unsigned int tlb0,
+		unsigned int tlb1)
+{
+	unsigned long *src, *dst;
+	unsigned int offset = 0;
+
+	/* If CONFIG_MTD_UCLINUX is defined, assume ROMFS is at the
+	 * end of kernel. There are two position which we want to check.
+	 * The first is __init_end and the second __bss_start.
+	 */
+#ifdef CONFIG_MTD_UCLINUX
+	int romfs_size;
+	unsigned int romfs_base;
+	char *old_klimit = klimit;
+
+	romfs_base = (ram ? ram : (unsigned int)&__init_end);
+	romfs_size = PAGE_ALIGN(get_romfs_len((unsigned *)romfs_base));
+	if (!romfs_size) {
+		romfs_base = (unsigned int)&__bss_start;
+		romfs_size = PAGE_ALIGN(get_romfs_len((unsigned *)romfs_base));
+	}
+
+	/* Move ROMFS out of BSS before clearing it */
+	if (romfs_size > 0) {
+		memmove(&__bss_stop, (int *)romfs_base, romfs_size);
+		klimit += romfs_size;
+	}
+#endif
+
+/* clearing bss section */
+	memset(__bss_start, 0, __bss_stop-__bss_start);
+	memset(_ssbss, 0, _esbss-_ssbss);
+
+	lockdep_init();
+
+/* initialize device tree for usage in early_printk */
+	early_init_devtree(_fdt_start);
+
+#ifdef CONFIG_EARLY_PRINTK
+	setup_early_printk(NULL);
+#endif
+
+	/* setup kernel_tlb after BSS cleaning
+	 * Maybe worth to move to asm code */
+	kernel_tlb = tlb0 + tlb1;
+	/* printk("TLB1 0x%08x, TLB0 0x%08x, tlb 0x%x\n", tlb0,
+							tlb1, kernel_tlb); */
+
+	pr_info("Ramdisk addr 0x%08x, ", ram);
+	if (fdt)
+		pr_info("FDT at 0x%08x\n", fdt);
+	else
+		pr_info("Compiled-in FDT at %p\n", _fdt_start);
+
+#ifdef CONFIG_MTD_UCLINUX
+	pr_info("Found romfs @ 0x%08x (0x%08x)\n",
+			romfs_base, romfs_size);
+	pr_info("#### klimit %p ####\n", old_klimit);
+	BUG_ON(romfs_size < 0); /* What else can we do? */
+
+	pr_info("Moved 0x%08x bytes from 0x%08x to 0x%08x\n",
+			romfs_size, romfs_base, (unsigned)&__bss_stop);
+
+	pr_info("New klimit: 0x%08x\n", (unsigned)klimit);
+#endif
+
+#if CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR
+	if (msr) {
+		pr_info("!!!Your kernel has setup MSR instruction but ");
+		pr_cont("CPU don't have it %x\n", msr);
+	}
+#else
+	if (!msr) {
+		pr_info("!!!Your kernel not setup MSR instruction but ");
+		pr_cont("CPU have it %x\n", msr);
+	}
+#endif
+
+	/* Do not copy reset vectors. offset = 0x2 means skip the first
+	 * two instructions. dst is pointer to MB vectors which are placed
+	 * in block ram. If you want to copy reset vector setup offset to 0x0 */
+#if !CONFIG_MANUAL_RESET_VECTOR
+	offset = 0x2;
+#endif
+	dst = (unsigned long *) (offset * sizeof(u32));
+	for (src = __ivt_start + offset; src < __ivt_end; src++, dst++)
+		*dst = *src;
+
+	/* Initialize global data */
+	per_cpu(KM, 0) = 0x1;	/* We start in kernel mode */
+	per_cpu(CURRENT_SAVE, 0) = (unsigned long)current;
+}
+
+void __init time_init(void)
+{
+	of_clk_init(NULL);
+	setup_cpuinfo_clk();
+	clocksource_probe();
+}
+
+#ifdef CONFIG_DEBUG_FS
+struct dentry *of_debugfs_root;
+
+static int microblaze_debugfs_init(void)
+{
+	of_debugfs_root = debugfs_create_dir("microblaze", NULL);
+
+	return of_debugfs_root == NULL;
+}
+arch_initcall(microblaze_debugfs_init);
+
+# ifdef CONFIG_MMU
+static int __init debugfs_tlb(void)
+{
+	struct dentry *d;
+
+	if (!of_debugfs_root)
+		return -ENODEV;
+
+	d = debugfs_create_u32("tlb_skip", S_IRUGO, of_debugfs_root, &tlb_skip);
+	if (!d)
+		return -ENOMEM;
+
+	return 0;
+}
+device_initcall(debugfs_tlb);
+# endif
+#endif
diff --git a/arch/microblaze/kernel/signal.c b/arch/microblaze/kernel/signal.c
new file mode 100644
index 0000000..9700152
--- /dev/null
+++ b/arch/microblaze/kernel/signal.c
@@ -0,0 +1,325 @@
+/*
+ * Signal handling
+ *
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ * Copyright (C) 2003,2004 John Williams <jwilliams@itee.uq.edu.au>
+ * Copyright (C) 2001 NEC Corporation
+ * Copyright (C) 2001 Miles Bader <miles@gnu.org>
+ * Copyright (C) 1999,2000 Niibe Yutaka & Kaz Kojima
+ * Copyright (C) 1991,1992 Linus Torvalds
+ *
+ * 1997-11-28 Modified for POSIX.1b signals by Richard Henderson
+ *
+ * This file was was derived from the sh version, arch/sh/kernel/signal.c
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License. See the file COPYING in the main directory of this
+ * archive for more details.
+ */
+
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/smp.h>
+#include <linux/kernel.h>
+#include <linux/signal.h>
+#include <linux/errno.h>
+#include <linux/wait.h>
+#include <linux/ptrace.h>
+#include <linux/unistd.h>
+#include <linux/stddef.h>
+#include <linux/personality.h>
+#include <linux/percpu.h>
+#include <linux/linkage.h>
+#include <linux/tracehook.h>
+#include <asm/entry.h>
+#include <asm/ucontext.h>
+#include <linux/uaccess.h>
+#include <asm/pgtable.h>
+#include <asm/pgalloc.h>
+#include <linux/syscalls.h>
+#include <asm/cacheflush.h>
+#include <asm/syscalls.h>
+
+/*
+ * Do a signal return; undo the signal stack.
+ */
+struct sigframe {
+	struct sigcontext sc;
+	unsigned long extramask[_NSIG_WORDS-1];
+	unsigned long tramp[2];	/* signal trampoline */
+};
+
+struct rt_sigframe {
+	struct siginfo info;
+	struct ucontext uc;
+	unsigned long tramp[2];	/* signal trampoline */
+};
+
+static int restore_sigcontext(struct pt_regs *regs,
+				struct sigcontext __user *sc, int *rval_p)
+{
+	unsigned int err = 0;
+
+#define COPY(x)		{err |= __get_user(regs->x, &sc->regs.x); }
+	COPY(r0);
+	COPY(r1);
+	COPY(r2);	COPY(r3);	COPY(r4);	COPY(r5);
+	COPY(r6);	COPY(r7);	COPY(r8);	COPY(r9);
+	COPY(r10);	COPY(r11);	COPY(r12);	COPY(r13);
+	COPY(r14);	COPY(r15);	COPY(r16);	COPY(r17);
+	COPY(r18);	COPY(r19);	COPY(r20);	COPY(r21);
+	COPY(r22);	COPY(r23);	COPY(r24);	COPY(r25);
+	COPY(r26);	COPY(r27);	COPY(r28);	COPY(r29);
+	COPY(r30);	COPY(r31);
+	COPY(pc);	COPY(ear);	COPY(esr);	COPY(fsr);
+#undef COPY
+
+	*rval_p = regs->r3;
+
+	return err;
+}
+
+asmlinkage long sys_rt_sigreturn(struct pt_regs *regs)
+{
+	struct rt_sigframe __user *frame =
+		(struct rt_sigframe __user *)(regs->r1);
+
+	sigset_t set;
+	int rval;
+
+	/* Always make any pending restarted system calls return -EINTR */
+	current->restart_block.fn = do_no_restart_syscall;
+
+	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
+		goto badframe;
+
+	if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set)))
+		goto badframe;
+
+	set_current_blocked(&set);
+
+	if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &rval))
+		goto badframe;
+
+	if (restore_altstack(&frame->uc.uc_stack))
+		goto badframe;
+
+	return rval;
+
+badframe:
+	force_sig(SIGSEGV, current);
+	return 0;
+}
+
+/*
+ * Set up a signal frame.
+ */
+
+static int
+setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs,
+		unsigned long mask)
+{
+	int err = 0;
+
+#define COPY(x)		{err |= __put_user(regs->x, &sc->regs.x); }
+	COPY(r0);
+	COPY(r1);
+	COPY(r2);	COPY(r3);	COPY(r4);	COPY(r5);
+	COPY(r6);	COPY(r7);	COPY(r8);	COPY(r9);
+	COPY(r10);	COPY(r11);	COPY(r12);	COPY(r13);
+	COPY(r14);	COPY(r15);	COPY(r16);	COPY(r17);
+	COPY(r18);	COPY(r19);	COPY(r20);	COPY(r21);
+	COPY(r22);	COPY(r23);	COPY(r24);	COPY(r25);
+	COPY(r26);	COPY(r27);	COPY(r28);	COPY(r29);
+	COPY(r30);	COPY(r31);
+	COPY(pc);	COPY(ear);	COPY(esr);	COPY(fsr);
+#undef COPY
+
+	err |= __put_user(mask, &sc->oldmask);
+
+	return err;
+}
+
+/*
+ * Determine which stack to use..
+ */
+static inline void __user *
+get_sigframe(struct ksignal *ksig, struct pt_regs *regs, size_t frame_size)
+{
+	/* Default to using normal stack */
+	unsigned long sp = sigsp(regs->r1, ksig);
+
+	return (void __user *)((sp - frame_size) & -8UL);
+}
+
+static int setup_rt_frame(struct ksignal *ksig, sigset_t *set,
+			  struct pt_regs *regs)
+{
+	struct rt_sigframe __user *frame;
+	int err = 0, sig = ksig->sig;
+	unsigned long address = 0;
+#ifdef CONFIG_MMU
+	pmd_t *pmdp;
+	pte_t *ptep;
+#endif
+
+	frame = get_sigframe(ksig, regs, sizeof(*frame));
+
+	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
+		return -EFAULT;
+
+	if (ksig->ka.sa.sa_flags & SA_SIGINFO)
+		err |= copy_siginfo_to_user(&frame->info, &ksig->info);
+
+	/* Create the ucontext. */
+	err |= __put_user(0, &frame->uc.uc_flags);
+	err |= __put_user(NULL, &frame->uc.uc_link);
+	err |= __save_altstack(&frame->uc.uc_stack, regs->r1);
+	err |= setup_sigcontext(&frame->uc.uc_mcontext,
+			regs, set->sig[0]);
+	err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set));
+
+	/* Set up to return from userspace. If provided, use a stub
+	 already in userspace. */
+	/* minus 8 is offset to cater for "rtsd r15,8" */
+	/* addi r12, r0, __NR_sigreturn */
+	err |= __put_user(0x31800000 | __NR_rt_sigreturn ,
+			frame->tramp + 0);
+	/* brki r14, 0x8 */
+	err |= __put_user(0xb9cc0008, frame->tramp + 1);
+
+	/* Return from sighandler will jump to the tramp.
+	 Negative 8 offset because return is rtsd r15, 8 */
+	regs->r15 = ((unsigned long)frame->tramp)-8;
+
+	address = ((unsigned long)frame->tramp);
+#ifdef CONFIG_MMU
+	pmdp = pmd_offset(pud_offset(
+			pgd_offset(current->mm, address),
+					address), address);
+
+	preempt_disable();
+	ptep = pte_offset_map(pmdp, address);
+	if (pte_present(*ptep)) {
+		address = (unsigned long) page_address(pte_page(*ptep));
+		/* MS: I need add offset in page */
+		address += ((unsigned long)frame->tramp) & ~PAGE_MASK;
+		/* MS address is virtual */
+		address = __virt_to_phys(address);
+		invalidate_icache_range(address, address + 8);
+		flush_dcache_range(address, address + 8);
+	}
+	pte_unmap(ptep);
+	preempt_enable();
+#else
+	flush_icache_range(address, address + 8);
+	flush_dcache_range(address, address + 8);
+#endif
+	if (err)
+		return -EFAULT;
+
+	/* Set up registers for signal handler */
+	regs->r1 = (unsigned long) frame;
+
+	/* Signal handler args: */
+	regs->r5 = sig; /* arg 0: signum */
+	regs->r6 = (unsigned long) &frame->info; /* arg 1: siginfo */
+	regs->r7 = (unsigned long) &frame->uc; /* arg2: ucontext */
+	/* Offset to handle microblaze rtid r14, 0 */
+	regs->pc = (unsigned long)ksig->ka.sa.sa_handler;
+
+#ifdef DEBUG_SIG
+	pr_info("SIG deliver (%s:%d): sp=%p pc=%08lx\n",
+		current->comm, current->pid, frame, regs->pc);
+#endif
+
+	return 0;
+}
+
+/* Handle restarting system calls */
+static inline void
+handle_restart(struct pt_regs *regs, struct k_sigaction *ka, int has_handler)
+{
+	switch (regs->r3) {
+	case -ERESTART_RESTARTBLOCK:
+	case -ERESTARTNOHAND:
+		if (!has_handler)
+			goto do_restart;
+		regs->r3 = -EINTR;
+		break;
+	case -ERESTARTSYS:
+		if (has_handler && !(ka->sa.sa_flags & SA_RESTART)) {
+			regs->r3 = -EINTR;
+			break;
+	}
+	/* fallthrough */
+	case -ERESTARTNOINTR:
+do_restart:
+		/* offset of 4 bytes to re-execute trap (brki) instruction */
+		regs->pc -= 4;
+		break;
+	}
+}
+
+/*
+ * OK, we're invoking a handler
+ */
+
+static void
+handle_signal(struct ksignal *ksig, struct pt_regs *regs)
+{
+	sigset_t *oldset = sigmask_to_save();
+	int ret;
+
+	/* Set up the stack frame */
+	ret = setup_rt_frame(ksig, oldset, regs);
+
+	signal_setup_done(ret, ksig, test_thread_flag(TIF_SINGLESTEP));
+}
+
+/*
+ * Note that 'init' is a special process: it doesn't get signals it doesn't
+ * want to handle. Thus you cannot kill init even with a SIGKILL even by
+ * mistake.
+ *
+ * Note that we go through the signals twice: once to check the signals that
+ * the kernel can handle, and then we build all the user-level signal handling
+ * stack-frames in one go after that.
+ */
+static void do_signal(struct pt_regs *regs, int in_syscall)
+{
+	struct ksignal ksig;
+
+#ifdef DEBUG_SIG
+	pr_info("do signal: %p %d\n", regs, in_syscall);
+	pr_info("do signal2: %lx %lx %ld [%lx]\n", regs->pc, regs->r1,
+			regs->r12, current_thread_info()->flags);
+#endif
+
+	if (get_signal(&ksig)) {
+		/* Whee! Actually deliver the signal. */
+		if (in_syscall)
+			handle_restart(regs, &ksig.ka, 1);
+		handle_signal(&ksig, regs);
+		return;
+	}
+
+	if (in_syscall)
+		handle_restart(regs, NULL, 0);
+
+	/*
+	 * If there's no signal to deliver, we just put the saved sigmask
+	 * back.
+	 */
+	restore_saved_sigmask();
+}
+
+asmlinkage void do_notify_resume(struct pt_regs *regs, int in_syscall)
+{
+	if (test_thread_flag(TIF_SIGPENDING))
+		do_signal(regs, in_syscall);
+
+	if (test_and_clear_thread_flag(TIF_NOTIFY_RESUME))
+		tracehook_notify_resume(regs);
+}
diff --git a/arch/microblaze/kernel/stacktrace.c b/arch/microblaze/kernel/stacktrace.c
new file mode 100644
index 0000000..b4debe2
--- /dev/null
+++ b/arch/microblaze/kernel/stacktrace.c
@@ -0,0 +1,31 @@
+/*
+ * Stack trace support for Microblaze.
+ *
+ * Copyright (C) 2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2009 PetaLogix
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/export.h>
+#include <linux/sched.h>
+#include <linux/stacktrace.h>
+#include <linux/thread_info.h>
+#include <linux/ptrace.h>
+#include <asm/unwind.h>
+
+void save_stack_trace(struct stack_trace *trace)
+{
+	/* Exclude our helper functions from the trace*/
+	trace->skip += 2;
+	microblaze_unwind(NULL, trace);
+}
+EXPORT_SYMBOL_GPL(save_stack_trace);
+
+void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
+{
+	microblaze_unwind(tsk, trace);
+}
+EXPORT_SYMBOL_GPL(save_stack_trace_tsk);
diff --git a/arch/microblaze/kernel/sys_microblaze.c b/arch/microblaze/kernel/sys_microblaze.c
new file mode 100644
index 0000000..f1e1f66
--- /dev/null
+++ b/arch/microblaze/kernel/sys_microblaze.c
@@ -0,0 +1,55 @@
+/*
+ * Copyright (C) 2007-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007-2009 PetaLogix
+ * Copyright (C) 2007 John Williams <john.williams@petalogix.com>
+ *
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ *	Yasushi SHOJI <yashi@atmark-techno.com>
+ *	Tetsuya OHKAWA <tetsuya@atmark-techno.com>
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/errno.h>
+#include <linux/export.h>
+#include <linux/mm.h>
+#include <linux/smp.h>
+#include <linux/syscalls.h>
+#include <linux/sem.h>
+#include <linux/msg.h>
+#include <linux/shm.h>
+#include <linux/stat.h>
+#include <linux/mman.h>
+#include <linux/sys.h>
+#include <linux/ipc.h>
+#include <linux/file.h>
+#include <linux/err.h>
+#include <linux/fs.h>
+#include <linux/semaphore.h>
+#include <linux/uaccess.h>
+#include <linux/unistd.h>
+#include <linux/slab.h>
+#include <asm/syscalls.h>
+
+SYSCALL_DEFINE6(mmap, unsigned long, addr, unsigned long, len,
+		unsigned long, prot, unsigned long, flags, unsigned long, fd,
+		off_t, pgoff)
+{
+	if (pgoff & ~PAGE_MASK)
+		return -EINVAL;
+
+	return sys_mmap_pgoff(addr, len, prot, flags, fd, pgoff >> PAGE_SHIFT);
+}
+
+SYSCALL_DEFINE6(mmap2, unsigned long, addr, unsigned long, len,
+		unsigned long, prot, unsigned long, flags, unsigned long, fd,
+		unsigned long, pgoff)
+{
+	if (pgoff & (~PAGE_MASK >> 12))
+		return -EINVAL;
+
+	return sys_mmap_pgoff(addr, len, prot, flags, fd,
+			      pgoff >> (PAGE_SHIFT - 12));
+}
diff --git a/arch/microblaze/kernel/syscall_table.S b/arch/microblaze/kernel/syscall_table.S
new file mode 100644
index 0000000..29c8568
--- /dev/null
+++ b/arch/microblaze/kernel/syscall_table.S
@@ -0,0 +1,391 @@
+ENTRY(sys_call_table)
+	.long sys_restart_syscall	/* 0 - old "setup()" system call,
+					 * used for restarting */
+	.long sys_exit
+	.long sys_fork
+	.long sys_read
+	.long sys_write
+	.long sys_open			/* 5 */
+	.long sys_close
+	.long sys_waitpid
+	.long sys_creat
+	.long sys_link
+	.long sys_unlink		/* 10 */
+	.long sys_execve
+	.long sys_chdir
+	.long sys_time
+	.long sys_mknod
+	.long sys_chmod			/* 15 */
+	.long sys_lchown
+	.long sys_ni_syscall		/* old break syscall holder */
+	.long sys_ni_syscall		/* old stat */
+	.long sys_lseek
+	.long sys_getpid		/* 20 */
+	.long sys_mount
+	.long sys_oldumount
+	.long sys_setuid
+	.long sys_getuid
+	.long sys_stime			/* 25 */
+	.long sys_ptrace
+	.long sys_alarm
+	.long sys_ni_syscall		/* oldfstat */
+	.long sys_pause
+	.long sys_utime			/* 30 */
+	.long sys_ni_syscall		/* old stty syscall holder */
+	.long sys_ni_syscall		/* old gtty syscall holder */
+	.long sys_access
+	.long sys_nice
+	.long sys_ni_syscall		/* 35 - old ftime syscall holder */
+	.long sys_sync
+	.long sys_kill
+	.long sys_rename
+	.long sys_mkdir
+	.long sys_rmdir			/* 40 */
+	.long sys_dup
+	.long sys_pipe
+	.long sys_times
+	.long sys_ni_syscall		/* old prof syscall holder */
+	.long sys_brk			/* 45 */
+	.long sys_setgid
+	.long sys_getgid
+	.long sys_signal
+	.long sys_geteuid
+	.long sys_getegid		/* 50 */
+	.long sys_acct
+	.long sys_umount		/* recycled never used phys() */
+	.long sys_ni_syscall		/* old lock syscall holder */
+	.long sys_ioctl
+	.long sys_fcntl			/* 55 */
+	.long sys_ni_syscall		/* old mpx syscall holder */
+	.long sys_setpgid
+	.long sys_ni_syscall		/* old ulimit syscall holder */
+	.long sys_ni_syscall		/* olduname */
+	.long sys_umask			/* 60 */
+	.long sys_chroot
+	.long sys_ustat
+	.long sys_dup2
+	.long sys_getppid
+	.long sys_getpgrp		/* 65 */
+	.long sys_setsid
+	.long sys_ni_syscall		/* sys_sigaction */
+	.long sys_sgetmask
+	.long sys_ssetmask
+	.long sys_setreuid		/* 70 */
+	.long sys_setregid
+	.long sys_ni_syscall		/* sys_sigsuspend_wrapper */
+	.long sys_sigpending
+	.long sys_sethostname
+	.long sys_setrlimit		/* 75 */
+	.long sys_ni_syscall		/* old_getrlimit */
+	.long sys_getrusage
+	.long sys_gettimeofday
+	.long sys_settimeofday
+	.long sys_getgroups		/* 80 */
+	.long sys_setgroups
+	.long sys_ni_syscall		/* old_select */
+	.long sys_symlink
+	.long sys_ni_syscall		/* oldlstat */
+	.long sys_readlink		/* 85 */
+	.long sys_uselib
+	.long sys_swapon
+	.long sys_reboot
+	.long sys_ni_syscall		/* old_readdir */
+	.long sys_mmap			/* 90 */ /* old_mmap */
+	.long sys_munmap
+	.long sys_truncate
+	.long sys_ftruncate
+	.long sys_fchmod
+	.long sys_fchown		/* 95 */
+	.long sys_getpriority
+	.long sys_setpriority
+	.long sys_ni_syscall		/* old profil syscall holder */
+	.long sys_statfs
+	.long sys_fstatfs		/* 100 */
+	.long sys_ni_syscall		/* ioperm */
+	.long sys_socketcall
+	.long sys_syslog		/* operation with system console */
+	.long sys_setitimer
+	.long sys_getitimer		/* 105 */
+	.long sys_newstat
+	.long sys_newlstat
+	.long sys_newfstat
+	.long sys_ni_syscall		/* uname */
+	.long sys_ni_syscall		/* 110 */ /* iopl */
+	.long sys_vhangup
+	.long sys_ni_syscall		/* old "idle" system call */
+	.long sys_ni_syscall		/* old sys_vm86old */
+	.long sys_wait4
+	.long sys_swapoff		/* 115 */
+	.long sys_sysinfo
+	.long sys_ni_syscall		/* old sys_ipc */
+	.long sys_fsync
+	.long sys_ni_syscall		/* sys_sigreturn_wrapper */
+	.long sys_clone		/* 120 */
+	.long sys_setdomainname
+	.long sys_newuname
+	.long sys_ni_syscall		/* modify_ldt */
+	.long sys_adjtimex
+	.long sys_mprotect		/* 125:	sys_mprotect */
+	.long sys_sigprocmask
+	.long sys_ni_syscall		/* old "create_module" */
+	.long sys_init_module
+	.long sys_delete_module
+	.long sys_ni_syscall		/* 130:	old "get_kernel_syms" */
+	.long sys_quotactl
+	.long sys_getpgid
+	.long sys_fchdir
+	.long sys_bdflush
+	.long sys_sysfs			/* 135 */
+	.long sys_personality
+	.long sys_ni_syscall		/* reserved for afs_syscall */
+	.long sys_setfsuid
+	.long sys_setfsgid
+	.long sys_llseek		/* 140 */
+	.long sys_getdents
+	.long sys_select
+	.long sys_flock
+	.long sys_msync
+	.long sys_readv			/* 145 */
+	.long sys_writev
+	.long sys_getsid
+	.long sys_fdatasync
+	.long sys_sysctl
+	.long sys_mlock		/* 150:	sys_mlock */
+	.long sys_munlock
+	.long sys_mlockall
+	.long sys_munlockall
+	.long sys_sched_setparam
+	.long sys_sched_getparam	/* 155 */
+	.long sys_sched_setscheduler
+	.long sys_sched_getscheduler
+	.long sys_sched_yield
+	.long sys_sched_get_priority_max
+	.long sys_sched_get_priority_min	/* 160 */
+	.long sys_sched_rr_get_interval
+	.long sys_nanosleep
+	.long sys_mremap
+	.long sys_setresuid
+	.long sys_getresuid		/* 165 */
+	.long sys_ni_syscall		/* sys_vm86 */
+	.long sys_ni_syscall		/* Old sys_query_module */
+	.long sys_poll
+	.long sys_ni_syscall		/* old nfsservctl */
+	.long sys_setresgid		/* 170 */
+	.long sys_getresgid
+	.long sys_prctl
+	.long sys_rt_sigreturn_wrapper
+	.long sys_rt_sigaction
+	.long sys_rt_sigprocmask	/* 175 */
+	.long sys_rt_sigpending
+	.long sys_rt_sigtimedwait
+	.long sys_rt_sigqueueinfo
+	.long sys_rt_sigsuspend
+	.long sys_pread64		/* 180 */
+	.long sys_pwrite64
+	.long sys_chown
+	.long sys_getcwd
+	.long sys_capget
+	.long sys_capset		/* 185 */
+	.long sys_ni_syscall		/* sigaltstack */
+	.long sys_sendfile
+	.long sys_ni_syscall		/* reserved for streams1 */
+	.long sys_ni_syscall		/* reserved for streams2 */
+	.long sys_vfork		/* 190 */
+	.long sys_getrlimit
+	.long sys_mmap2
+	.long sys_truncate64
+	.long sys_ftruncate64
+	.long sys_stat64		/* 195 */
+	.long sys_lstat64
+	.long sys_fstat64
+	.long sys_lchown
+	.long sys_getuid
+	.long sys_getgid		/* 200 */
+	.long sys_geteuid
+	.long sys_getegid
+	.long sys_setreuid
+	.long sys_setregid
+	.long sys_getgroups		/* 205 */
+	.long sys_setgroups
+	.long sys_fchown
+	.long sys_setresuid
+	.long sys_getresuid
+	.long sys_setresgid		/* 210 */
+	.long sys_getresgid
+	.long sys_chown
+	.long sys_setuid
+	.long sys_setgid
+	.long sys_setfsuid		/* 215 */
+	.long sys_setfsgid
+	.long sys_pivot_root
+	.long sys_mincore
+	.long sys_madvise
+	.long sys_getdents64		/* 220 */
+	.long sys_fcntl64
+	.long sys_ni_syscall		/* reserved for TUX */
+	.long sys_ni_syscall
+	.long sys_gettid
+	.long sys_readahead		/* 225 */
+	.long sys_setxattr
+	.long sys_lsetxattr
+	.long sys_fsetxattr
+	.long sys_getxattr
+	.long sys_lgetxattr		/* 230 */
+	.long sys_fgetxattr
+	.long sys_listxattr
+	.long sys_llistxattr
+	.long sys_flistxattr
+	.long sys_removexattr		/* 235 */
+	.long sys_lremovexattr
+	.long sys_fremovexattr
+	.long sys_tkill
+	.long sys_sendfile64
+	.long sys_futex			/* 240 */
+	.long sys_sched_setaffinity
+	.long sys_sched_getaffinity
+	.long sys_ni_syscall		/* set_thread_area */
+	.long sys_ni_syscall		/* get_thread_area */
+	.long sys_io_setup		/* 245 */
+	.long sys_io_destroy
+	.long sys_io_getevents
+	.long sys_io_submit
+	.long sys_io_cancel
+	.long sys_fadvise64		/* 250 */
+	.long sys_ni_syscall
+	.long sys_exit_group
+	.long sys_lookup_dcookie
+	.long sys_epoll_create
+	.long sys_epoll_ctl		/* 255 */
+	.long sys_epoll_wait
+	.long sys_remap_file_pages
+	.long sys_set_tid_address
+	.long sys_timer_create
+	.long sys_timer_settime		/* 260 */
+	.long sys_timer_gettime
+	.long sys_timer_getoverrun
+	.long sys_timer_delete
+	.long sys_clock_settime
+	.long sys_clock_gettime		/* 265 */
+	.long sys_clock_getres
+	.long sys_clock_nanosleep
+	.long sys_statfs64
+	.long sys_fstatfs64
+	.long sys_tgkill		/* 270 */
+	.long sys_utimes
+	.long sys_fadvise64_64
+	.long sys_ni_syscall		/* sys_vserver */
+	.long sys_mbind
+	.long sys_get_mempolicy
+	.long sys_set_mempolicy
+	.long sys_mq_open
+	.long sys_mq_unlink
+	.long sys_mq_timedsend
+	.long sys_mq_timedreceive	/* 280 */
+	.long sys_mq_notify
+	.long sys_mq_getsetattr
+	.long sys_kexec_load
+	.long sys_waitid
+	.long sys_ni_syscall		/* 285 */ /* available */
+	.long sys_add_key
+	.long sys_request_key
+	.long sys_keyctl
+	.long sys_ioprio_set
+	.long sys_ioprio_get		/* 290 */
+	.long sys_inotify_init
+	.long sys_inotify_add_watch
+	.long sys_inotify_rm_watch
+	.long sys_ni_syscall		/* sys_migrate_pages */
+	.long sys_openat	/* 295 */
+	.long sys_mkdirat
+	.long sys_mknodat
+	.long sys_fchownat
+	.long sys_futimesat
+	.long sys_fstatat64		/* 300 */
+	.long sys_unlinkat
+	.long sys_renameat
+	.long sys_linkat
+	.long sys_symlinkat
+	.long sys_readlinkat		/* 305 */
+	.long sys_fchmodat
+	.long sys_faccessat
+	.long sys_pselect6
+	.long sys_ppoll
+	.long sys_unshare		/* 310 */
+	.long sys_set_robust_list
+	.long sys_get_robust_list
+	.long sys_splice
+	.long sys_sync_file_range
+	.long sys_tee 			/* 315 */
+	.long sys_vmsplice
+	.long sys_move_pages
+	.long sys_getcpu
+	.long sys_epoll_pwait
+	.long sys_utimensat		/* 320 */
+	.long sys_signalfd
+	.long sys_timerfd_create
+	.long sys_eventfd
+	.long sys_fallocate
+	.long sys_semtimedop		/* 325 */
+	.long sys_timerfd_settime
+	.long sys_timerfd_gettime
+	.long sys_semctl
+	.long sys_semget
+	.long sys_semop			/* 330 */
+	.long sys_msgctl
+	.long sys_msgget
+	.long sys_msgrcv
+	.long sys_msgsnd
+	.long sys_shmat			/* 335 */
+	.long sys_shmctl
+	.long sys_shmdt
+	.long sys_shmget
+	.long sys_signalfd4	/* new syscall */
+	.long sys_eventfd2		/* 340 */
+	.long sys_epoll_create1
+	.long sys_dup3
+	.long sys_pipe2
+	.long sys_inotify_init1
+	.long sys_socket		/* 345 */
+	.long sys_socketpair
+	.long sys_bind
+	.long sys_listen
+	.long sys_accept
+	.long sys_connect		/* 350 */
+	.long sys_getsockname
+	.long sys_getpeername
+	.long sys_sendto
+	.long sys_send
+	.long sys_recvfrom		/* 355 */
+	.long sys_recv
+	.long sys_setsockopt
+	.long sys_getsockopt
+	.long sys_shutdown
+	.long sys_sendmsg		/* 360 */
+	.long sys_recvmsg
+	.long sys_accept4
+	.long sys_preadv
+	.long sys_pwritev
+	.long sys_rt_tgsigqueueinfo	/* 365 */
+	.long sys_perf_event_open
+	.long sys_recvmmsg
+	.long sys_fanotify_init
+	.long sys_fanotify_mark
+	.long sys_prlimit64	/* 370 */
+	.long sys_name_to_handle_at
+	.long sys_open_by_handle_at
+	.long sys_clock_adjtime
+	.long sys_syncfs
+	.long sys_setns			/* 375 */
+	.long sys_sendmmsg
+	.long sys_process_vm_readv
+	.long sys_process_vm_writev
+	.long sys_kcmp
+	.long sys_finit_module		/* 380 */
+	.long sys_sched_setattr
+	.long sys_sched_getattr
+	.long sys_renameat2
+	.long sys_seccomp
+	.long sys_getrandom		/* 385 */
+	.long sys_memfd_create
+	.long sys_bpf
+	.long sys_execveat
diff --git a/arch/microblaze/kernel/timer.c b/arch/microblaze/kernel/timer.c
new file mode 100644
index 0000000..67e2ef4
--- /dev/null
+++ b/arch/microblaze/kernel/timer.c
@@ -0,0 +1,311 @@
+/*
+ * Copyright (C) 2007-2013 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2012-2013 Xilinx, Inc.
+ * Copyright (C) 2007-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/interrupt.h>
+#include <linux/delay.h>
+#include <linux/sched.h>
+#include <linux/sched_clock.h>
+#include <linux/clk.h>
+#include <linux/clockchips.h>
+#include <linux/of_address.h>
+#include <linux/of_irq.h>
+#include <linux/timecounter.h>
+#include <asm/cpuinfo.h>
+
+static void __iomem *timer_baseaddr;
+
+static unsigned int freq_div_hz;
+static unsigned int timer_clock_freq;
+
+#define TCSR0	(0x00)
+#define TLR0	(0x04)
+#define TCR0	(0x08)
+#define TCSR1	(0x10)
+#define TLR1	(0x14)
+#define TCR1	(0x18)
+
+#define TCSR_MDT	(1<<0)
+#define TCSR_UDT	(1<<1)
+#define TCSR_GENT	(1<<2)
+#define TCSR_CAPT	(1<<3)
+#define TCSR_ARHT	(1<<4)
+#define TCSR_LOAD	(1<<5)
+#define TCSR_ENIT	(1<<6)
+#define TCSR_ENT	(1<<7)
+#define TCSR_TINT	(1<<8)
+#define TCSR_PWMA	(1<<9)
+#define TCSR_ENALL	(1<<10)
+
+static unsigned int (*read_fn)(void __iomem *);
+static void (*write_fn)(u32, void __iomem *);
+
+static void timer_write32(u32 val, void __iomem *addr)
+{
+	iowrite32(val, addr);
+}
+
+static unsigned int timer_read32(void __iomem *addr)
+{
+	return ioread32(addr);
+}
+
+static void timer_write32_be(u32 val, void __iomem *addr)
+{
+	iowrite32be(val, addr);
+}
+
+static unsigned int timer_read32_be(void __iomem *addr)
+{
+	return ioread32be(addr);
+}
+
+static inline void xilinx_timer0_stop(void)
+{
+	write_fn(read_fn(timer_baseaddr + TCSR0) & ~TCSR_ENT,
+		 timer_baseaddr + TCSR0);
+}
+
+static inline void xilinx_timer0_start_periodic(unsigned long load_val)
+{
+	if (!load_val)
+		load_val = 1;
+	/* loading value to timer reg */
+	write_fn(load_val, timer_baseaddr + TLR0);
+
+	/* load the initial value */
+	write_fn(TCSR_LOAD, timer_baseaddr + TCSR0);
+
+	/* see timer data sheet for detail
+	 * !ENALL - don't enable 'em all
+	 * !PWMA - disable pwm
+	 * TINT - clear interrupt status
+	 * ENT- enable timer itself
+	 * ENIT - enable interrupt
+	 * !LOAD - clear the bit to let go
+	 * ARHT - auto reload
+	 * !CAPT - no external trigger
+	 * !GENT - no external signal
+	 * UDT - set the timer as down counter
+	 * !MDT0 - generate mode
+	 */
+	write_fn(TCSR_TINT|TCSR_ENIT|TCSR_ENT|TCSR_ARHT|TCSR_UDT,
+		 timer_baseaddr + TCSR0);
+}
+
+static inline void xilinx_timer0_start_oneshot(unsigned long load_val)
+{
+	if (!load_val)
+		load_val = 1;
+	/* loading value to timer reg */
+	write_fn(load_val, timer_baseaddr + TLR0);
+
+	/* load the initial value */
+	write_fn(TCSR_LOAD, timer_baseaddr + TCSR0);
+
+	write_fn(TCSR_TINT|TCSR_ENIT|TCSR_ENT|TCSR_ARHT|TCSR_UDT,
+		 timer_baseaddr + TCSR0);
+}
+
+static int xilinx_timer_set_next_event(unsigned long delta,
+					struct clock_event_device *dev)
+{
+	pr_debug("%s: next event, delta %x\n", __func__, (u32)delta);
+	xilinx_timer0_start_oneshot(delta);
+	return 0;
+}
+
+static int xilinx_timer_shutdown(struct clock_event_device *evt)
+{
+	pr_info("%s\n", __func__);
+	xilinx_timer0_stop();
+	return 0;
+}
+
+static int xilinx_timer_set_periodic(struct clock_event_device *evt)
+{
+	pr_info("%s\n", __func__);
+	xilinx_timer0_start_periodic(freq_div_hz);
+	return 0;
+}
+
+static struct clock_event_device clockevent_xilinx_timer = {
+	.name			= "xilinx_clockevent",
+	.features		= CLOCK_EVT_FEAT_ONESHOT |
+				  CLOCK_EVT_FEAT_PERIODIC,
+	.shift			= 8,
+	.rating			= 300,
+	.set_next_event		= xilinx_timer_set_next_event,
+	.set_state_shutdown	= xilinx_timer_shutdown,
+	.set_state_periodic	= xilinx_timer_set_periodic,
+};
+
+static inline void timer_ack(void)
+{
+	write_fn(read_fn(timer_baseaddr + TCSR0), timer_baseaddr + TCSR0);
+}
+
+static irqreturn_t timer_interrupt(int irq, void *dev_id)
+{
+	struct clock_event_device *evt = &clockevent_xilinx_timer;
+#ifdef CONFIG_HEART_BEAT
+	microblaze_heartbeat();
+#endif
+	timer_ack();
+	evt->event_handler(evt);
+	return IRQ_HANDLED;
+}
+
+static struct irqaction timer_irqaction = {
+	.handler = timer_interrupt,
+	.flags = IRQF_TIMER,
+	.name = "timer",
+	.dev_id = &clockevent_xilinx_timer,
+};
+
+static __init void xilinx_clockevent_init(void)
+{
+	clockevent_xilinx_timer.mult =
+		div_sc(timer_clock_freq, NSEC_PER_SEC,
+				clockevent_xilinx_timer.shift);
+	clockevent_xilinx_timer.max_delta_ns =
+		clockevent_delta2ns((u32)~0, &clockevent_xilinx_timer);
+	clockevent_xilinx_timer.min_delta_ns =
+		clockevent_delta2ns(1, &clockevent_xilinx_timer);
+	clockevent_xilinx_timer.cpumask = cpumask_of(0);
+	clockevents_register_device(&clockevent_xilinx_timer);
+}
+
+static u64 xilinx_clock_read(void)
+{
+	return read_fn(timer_baseaddr + TCR1);
+}
+
+static cycle_t xilinx_read(struct clocksource *cs)
+{
+	/* reading actual value of timer 1 */
+	return (cycle_t)xilinx_clock_read();
+}
+
+static struct timecounter xilinx_tc = {
+	.cc = NULL,
+};
+
+static cycle_t xilinx_cc_read(const struct cyclecounter *cc)
+{
+	return xilinx_read(NULL);
+}
+
+static struct cyclecounter xilinx_cc = {
+	.read = xilinx_cc_read,
+	.mask = CLOCKSOURCE_MASK(32),
+	.shift = 8,
+};
+
+static int __init init_xilinx_timecounter(void)
+{
+	xilinx_cc.mult = div_sc(timer_clock_freq, NSEC_PER_SEC,
+				xilinx_cc.shift);
+
+	timecounter_init(&xilinx_tc, &xilinx_cc, sched_clock());
+
+	return 0;
+}
+
+static struct clocksource clocksource_microblaze = {
+	.name		= "xilinx_clocksource",
+	.rating		= 300,
+	.read		= xilinx_read,
+	.mask		= CLOCKSOURCE_MASK(32),
+	.flags		= CLOCK_SOURCE_IS_CONTINUOUS,
+};
+
+static int __init xilinx_clocksource_init(void)
+{
+	if (clocksource_register_hz(&clocksource_microblaze, timer_clock_freq))
+		panic("failed to register clocksource");
+
+	/* stop timer1 */
+	write_fn(read_fn(timer_baseaddr + TCSR1) & ~TCSR_ENT,
+		 timer_baseaddr + TCSR1);
+	/* start timer1 - up counting without interrupt */
+	write_fn(TCSR_TINT|TCSR_ENT|TCSR_ARHT, timer_baseaddr + TCSR1);
+
+	/* register timecounter - for ftrace support */
+	init_xilinx_timecounter();
+	return 0;
+}
+
+static void __init xilinx_timer_init(struct device_node *timer)
+{
+	struct clk *clk;
+	static int initialized;
+	u32 irq;
+	u32 timer_num = 1;
+
+	if (initialized)
+		return;
+
+	initialized = 1;
+
+	timer_baseaddr = of_iomap(timer, 0);
+	if (!timer_baseaddr) {
+		pr_err("ERROR: invalid timer base address\n");
+		BUG();
+	}
+
+	write_fn = timer_write32;
+	read_fn = timer_read32;
+
+	write_fn(TCSR_MDT, timer_baseaddr + TCSR0);
+	if (!(read_fn(timer_baseaddr + TCSR0) & TCSR_MDT)) {
+		write_fn = timer_write32_be;
+		read_fn = timer_read32_be;
+	}
+
+	irq = irq_of_parse_and_map(timer, 0);
+
+	of_property_read_u32(timer, "xlnx,one-timer-only", &timer_num);
+	if (timer_num) {
+		pr_emerg("Please enable two timers in HW\n");
+		BUG();
+	}
+
+	pr_info("%s: irq=%d\n", timer->full_name, irq);
+
+	clk = of_clk_get(timer, 0);
+	if (IS_ERR(clk)) {
+		pr_err("ERROR: timer CCF input clock not found\n");
+		/* If there is clock-frequency property than use it */
+		of_property_read_u32(timer, "clock-frequency",
+				    &timer_clock_freq);
+	} else {
+		timer_clock_freq = clk_get_rate(clk);
+	}
+
+	if (!timer_clock_freq) {
+		pr_err("ERROR: Using CPU clock frequency\n");
+		timer_clock_freq = cpuinfo.cpu_clock_freq;
+	}
+
+	freq_div_hz = timer_clock_freq / HZ;
+
+	setup_irq(irq, &timer_irqaction);
+#ifdef CONFIG_HEART_BEAT
+	microblaze_setup_heartbeat();
+#endif
+	xilinx_clocksource_init();
+	xilinx_clockevent_init();
+
+	sched_clock_register(xilinx_clock_read, 32, timer_clock_freq);
+}
+
+CLOCKSOURCE_OF_DECLARE(xilinx_timer, "xlnx,xps-timer-1.00.a",
+		       xilinx_timer_init);
diff --git a/arch/microblaze/kernel/traps.c b/arch/microblaze/kernel/traps.c
new file mode 100644
index 0000000..cb61953
--- /dev/null
+++ b/arch/microblaze/kernel/traps.c
@@ -0,0 +1,77 @@
+/*
+ * Copyright (C) 2007-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/export.h>
+#include <linux/kernel.h>
+#include <linux/kallsyms.h>
+#include <linux/sched.h>
+#include <linux/debug_locks.h>
+
+#include <asm/exceptions.h>
+#include <asm/unwind.h>
+
+void trap_init(void)
+{
+	__enable_hw_exceptions();
+}
+
+static unsigned long kstack_depth_to_print;	/* 0 == entire stack */
+
+static int __init kstack_setup(char *s)
+{
+	return !kstrtoul(s, 0, &kstack_depth_to_print);
+}
+__setup("kstack=", kstack_setup);
+
+void show_stack(struct task_struct *task, unsigned long *sp)
+{
+	unsigned long words_to_show;
+	u32 fp = (u32) sp;
+
+	if (fp == 0) {
+		if (task) {
+			fp = ((struct thread_info *)
+				(task->stack))->cpu_context.r1;
+		} else {
+			/* Pick up caller of dump_stack() */
+			fp = (u32)&sp - 8;
+		}
+	}
+
+	words_to_show = (THREAD_SIZE - (fp & (THREAD_SIZE - 1))) >> 2;
+	if (kstack_depth_to_print && (words_to_show > kstack_depth_to_print))
+		words_to_show = kstack_depth_to_print;
+
+	pr_info("Kernel Stack:\n");
+
+	/*
+	 * Make the first line an 'odd' size if necessary to get
+	 * remaining lines to start at an address multiple of 0x10
+	 */
+	if (fp & 0xF) {
+		unsigned long line1_words = (0x10 - (fp & 0xF)) >> 2;
+		if (line1_words < words_to_show) {
+			print_hex_dump(KERN_INFO, "", DUMP_PREFIX_ADDRESS, 32,
+				       4, (void *)fp, line1_words << 2, 0);
+			fp += line1_words << 2;
+			words_to_show -= line1_words;
+		}
+	}
+	print_hex_dump(KERN_INFO, "", DUMP_PREFIX_ADDRESS, 32, 4, (void *)fp,
+		       words_to_show << 2, 0);
+	pr_info("\n\nCall Trace:\n");
+	microblaze_unwind(task, NULL);
+	pr_info("\n");
+
+	if (!task)
+		task = current;
+
+	debug_show_held_locks(task);
+}
diff --git a/arch/microblaze/kernel/unwind.c b/arch/microblaze/kernel/unwind.c
new file mode 100644
index 0000000..61c04ee
--- /dev/null
+++ b/arch/microblaze/kernel/unwind.c
@@ -0,0 +1,319 @@
+/*
+ * Backtrace support for Microblaze
+ *
+ * Copyright (C) 2010  Digital Design Corporation
+ *
+ * Based on arch/sh/kernel/cpu/sh5/unwind.c code which is:
+ * Copyright (C) 2004  Paul Mundt
+ * Copyright (C) 2004  Richard Curnow
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+/* #define DEBUG 1 */
+#include <linux/export.h>
+#include <linux/kallsyms.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/stacktrace.h>
+#include <linux/types.h>
+#include <linux/errno.h>
+#include <linux/io.h>
+#include <asm/sections.h>
+#include <asm/exceptions.h>
+#include <asm/unwind.h>
+#include <asm/switch_to.h>
+
+struct stack_trace;
+
+/*
+ * On Microblaze, finding the previous stack frame is a little tricky.
+ * At this writing (3/2010), Microblaze does not support CONFIG_FRAME_POINTERS,
+ * and even if it did, gcc (4.1.2) does not store the frame pointer at
+ * a consistent offset within each frame. To determine frame size, it is
+ * necessary to search for the assembly instruction that creates or reclaims
+ * the frame and extract the size from it.
+ *
+ * Microblaze stores the stack pointer in r1, and creates a frame via
+ *
+ *     addik r1, r1, -FRAME_SIZE
+ *
+ * The frame is reclaimed via
+ *
+ *     addik r1, r1, FRAME_SIZE
+ *
+ * Frame creation occurs at or near the top of a function.
+ * Depending on the compiler, reclaim may occur at the end, or before
+ * a mid-function return.
+ *
+ * A stack frame is usually not created in a leaf function.
+ *
+ */
+
+/**
+ * get_frame_size - Extract the stack adjustment from an
+ *                  "addik r1, r1, adjust" instruction
+ * @instr : Microblaze instruction
+ *
+ * Return - Number of stack bytes the instruction reserves or reclaims
+ */
+static inline long get_frame_size(unsigned long instr)
+{
+	return abs((s16)(instr & 0xFFFF));
+}
+
+/**
+ * find_frame_creation - Search backward to find the instruction that creates
+ *                       the stack frame (hopefully, for the same function the
+ *                       initial PC is in).
+ * @pc : Program counter at which to begin the search
+ *
+ * Return - PC at which stack frame creation occurs
+ *          NULL if this cannot be found, i.e. a leaf function
+ */
+static unsigned long *find_frame_creation(unsigned long *pc)
+{
+	int i;
+
+	/* NOTE: Distance to search is arbitrary
+	 *	 250 works well for most things,
+	 *	 750 picks up things like tcp_recvmsg(),
+	 *	1000 needed for fat_fill_super()
+	 */
+	for (i = 0; i < 1000; i++, pc--) {
+		unsigned long instr;
+		s16 frame_size;
+
+		if (!kernel_text_address((unsigned long) pc))
+			return NULL;
+
+		instr = *pc;
+
+		/* addik r1, r1, foo ? */
+		if ((instr & 0xFFFF0000) != 0x30210000)
+			continue;	/* No */
+
+		frame_size = get_frame_size(instr);
+		if ((frame_size < 8) || (frame_size & 3)) {
+			pr_debug("    Invalid frame size %d at 0x%p\n",
+				 frame_size, pc);
+			return NULL;
+		}
+
+		pr_debug("    Found frame creation at 0x%p, size %d\n", pc,
+			 frame_size);
+		return pc;
+	}
+
+	return NULL;
+}
+
+/**
+ * lookup_prev_stack_frame - Find the stack frame of the previous function.
+ * @fp          : Frame (stack) pointer for current function
+ * @pc          : Program counter within current function
+ * @leaf_return : r15 value within current function. If the current function
+ *		  is a leaf, this is the caller's return address.
+ * @pprev_fp    : On exit, set to frame (stack) pointer for previous function
+ * @pprev_pc    : On exit, set to current function caller's return address
+ *
+ * Return - 0 on success, -EINVAL if the previous frame cannot be found
+ */
+static int lookup_prev_stack_frame(unsigned long fp, unsigned long pc,
+				   unsigned long leaf_return,
+				   unsigned long *pprev_fp,
+				   unsigned long *pprev_pc)
+{
+	unsigned long *prologue = NULL;
+
+	/* _switch_to is a special leaf function */
+	if (pc != (unsigned long) &_switch_to)
+		prologue = find_frame_creation((unsigned long *)pc);
+
+	if (prologue) {
+		long frame_size = get_frame_size(*prologue);
+
+		*pprev_fp = fp + frame_size;
+		*pprev_pc = *(unsigned long *)fp;
+	} else {
+		if (!leaf_return)
+			return -EINVAL;
+		*pprev_pc = leaf_return;
+		*pprev_fp = fp;
+	}
+
+	/* NOTE: don't check kernel_text_address here, to allow display
+	 *	 of userland return address
+	 */
+	return (!*pprev_pc || (*pprev_pc & 3)) ? -EINVAL : 0;
+}
+
+static void microblaze_unwind_inner(struct task_struct *task,
+				    unsigned long pc, unsigned long fp,
+				    unsigned long leaf_return,
+				    struct stack_trace *trace);
+
+/**
+ * unwind_trap - Unwind through a system trap, that stored previous state
+ *		 on the stack.
+ */
+#ifdef CONFIG_MMU
+static inline void unwind_trap(struct task_struct *task, unsigned long pc,
+				unsigned long fp, struct stack_trace *trace)
+{
+	/* To be implemented */
+}
+#else
+static inline void unwind_trap(struct task_struct *task, unsigned long pc,
+				unsigned long fp, struct stack_trace *trace)
+{
+	const struct pt_regs *regs = (const struct pt_regs *) fp;
+	microblaze_unwind_inner(task, regs->pc, regs->r1, regs->r15, trace);
+}
+#endif
+
+/**
+ * microblaze_unwind_inner - Unwind the stack from the specified point
+ * @task  : Task whose stack we are to unwind (may be NULL)
+ * @pc    : Program counter from which we start unwinding
+ * @fp    : Frame (stack) pointer from which we start unwinding
+ * @leaf_return : Value of r15 at pc. If the function is a leaf, this is
+ *				  the caller's return address.
+ * @trace : Where to store stack backtrace (PC values).
+ *	    NULL == print backtrace to kernel log
+ */
+static void microblaze_unwind_inner(struct task_struct *task,
+			     unsigned long pc, unsigned long fp,
+			     unsigned long leaf_return,
+			     struct stack_trace *trace)
+{
+	int ofs = 0;
+
+	pr_debug("    Unwinding with PC=%p, FP=%p\n", (void *)pc, (void *)fp);
+	if (!pc || !fp || (pc & 3) || (fp & 3)) {
+		pr_debug("    Invalid state for unwind, aborting\n");
+		return;
+	}
+	for (; pc != 0;) {
+		unsigned long next_fp, next_pc = 0;
+		unsigned long return_to = pc +  2 * sizeof(unsigned long);
+		const struct trap_handler_info *handler =
+			&microblaze_trap_handlers;
+
+		/* Is previous function the HW exception handler? */
+		if ((return_to >= (unsigned long)&_hw_exception_handler)
+		    &&(return_to < (unsigned long)&ex_handler_unhandled)) {
+			/*
+			 * HW exception handler doesn't save all registers,
+			 * so we open-code a special case of unwind_trap()
+			 */
+#ifndef CONFIG_MMU
+			const struct pt_regs *regs =
+				(const struct pt_regs *) fp;
+#endif
+			pr_info("HW EXCEPTION\n");
+#ifndef CONFIG_MMU
+			microblaze_unwind_inner(task, regs->r17 - 4,
+						fp + EX_HANDLER_STACK_SIZ,
+						regs->r15, trace);
+#endif
+			return;
+		}
+
+		/* Is previous function a trap handler? */
+		for (; handler->start_addr; ++handler) {
+			if ((return_to >= handler->start_addr)
+			    && (return_to <= handler->end_addr)) {
+				if (!trace)
+					pr_info("%s\n", handler->trap_name);
+				unwind_trap(task, pc, fp, trace);
+				return;
+			}
+		}
+		pc -= ofs;
+
+		if (trace) {
+#ifdef CONFIG_STACKTRACE
+			if (trace->skip > 0)
+				trace->skip--;
+			else
+				trace->entries[trace->nr_entries++] = pc;
+
+			if (trace->nr_entries >= trace->max_entries)
+				break;
+#endif
+		} else {
+			/* Have we reached userland? */
+			if (unlikely(pc == task_pt_regs(task)->pc)) {
+				pr_info("[<%p>] PID %lu [%s]\n",
+					(void *) pc,
+					(unsigned long) task->pid,
+					task->comm);
+				break;
+			} else
+				print_ip_sym(pc);
+		}
+
+		/* Stop when we reach anything not part of the kernel */
+		if (!kernel_text_address(pc))
+			break;
+
+		if (lookup_prev_stack_frame(fp, pc, leaf_return, &next_fp,
+					    &next_pc) == 0) {
+			ofs = sizeof(unsigned long);
+			pc = next_pc & ~3;
+			fp = next_fp;
+			leaf_return = 0;
+		} else {
+			pr_debug("    Failed to find previous stack frame\n");
+			break;
+		}
+
+		pr_debug("    Next PC=%p, next FP=%p\n",
+			 (void *)next_pc, (void *)next_fp);
+	}
+}
+
+/**
+ * microblaze_unwind - Stack unwinder for Microblaze (external entry point)
+ * @task  : Task whose stack we are to unwind (NULL == current)
+ * @trace : Where to store stack backtrace (PC values).
+ *	    NULL == print backtrace to kernel log
+ */
+void microblaze_unwind(struct task_struct *task, struct stack_trace *trace)
+{
+	if (task) {
+		if (task == current) {
+			const struct pt_regs *regs = task_pt_regs(task);
+			microblaze_unwind_inner(task, regs->pc, regs->r1,
+						regs->r15, trace);
+		} else {
+			struct thread_info *thread_info =
+				(struct thread_info *)(task->stack);
+			const struct cpu_context *cpu_context =
+				&thread_info->cpu_context;
+
+			microblaze_unwind_inner(task,
+						(unsigned long) &_switch_to,
+						cpu_context->r1,
+						cpu_context->r15, trace);
+		}
+	} else {
+		unsigned long pc, fp;
+
+		__asm__ __volatile__ ("or %0, r1, r0" : "=r" (fp));
+
+		__asm__ __volatile__ (
+			"brlid %0, 0f;"
+			"nop;"
+			"0:"
+			: "=r" (pc)
+		);
+
+		/* Since we are not a leaf function, use leaf_return = 0 */
+		microblaze_unwind_inner(current, pc, fp, 0, trace);
+	}
+}
+
diff --git a/arch/microblaze/kernel/vmlinux.lds.S b/arch/microblaze/kernel/vmlinux.lds.S
new file mode 100644
index 0000000..be9488d
--- /dev/null
+++ b/arch/microblaze/kernel/vmlinux.lds.S
@@ -0,0 +1,140 @@
+/*
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+OUTPUT_ARCH(microblaze)
+ENTRY(microblaze_start)
+
+#include <asm/page.h>
+#include <asm-generic/vmlinux.lds.h>
+#include <asm/thread_info.h>
+
+#ifdef __MICROBLAZEEL__
+jiffies = jiffies_64;
+#else
+jiffies = jiffies_64 + 4;
+#endif
+
+SECTIONS {
+	. = CONFIG_KERNEL_START;
+	microblaze_start = CONFIG_KERNEL_BASE_ADDR;
+	.text : AT(ADDR(.text) - LOAD_OFFSET) {
+		_text = . ;
+		_stext = . ;
+		HEAD_TEXT
+		TEXT_TEXT
+		*(.fixup)
+		EXIT_TEXT
+		EXIT_CALL
+		SCHED_TEXT
+		LOCK_TEXT
+		KPROBES_TEXT
+		IRQENTRY_TEXT
+		. = ALIGN (4) ;
+		_etext = . ;
+	}
+
+	. = ALIGN (4) ;
+	__fdt_blob : AT(ADDR(__fdt_blob) - LOAD_OFFSET) {
+		_fdt_start = . ;		/* place for fdt blob */
+		*(__fdt_blob) ;			/* Any link-placed DTB */
+	        . = _fdt_start + 0x8000;	/* Pad up to 32kbyte */
+		_fdt_end = . ;
+	}
+
+	. = ALIGN(16);
+	RODATA
+	EXCEPTION_TABLE(16)
+	NOTES
+
+	/*
+	 * sdata2 section can go anywhere, but must be word aligned
+	 * and SDA2_BASE must point to the middle of it
+	 */
+	.sdata2 : AT(ADDR(.sdata2) - LOAD_OFFSET) {
+		_ssrw = .;
+		. = ALIGN(PAGE_SIZE); /* page aligned when MMU used */
+		*(.sdata2)
+	. = ALIGN(8);
+	_essrw = .;
+	_ssrw_size = _essrw - _ssrw;
+	_KERNEL_SDA2_BASE_ = _ssrw + (_ssrw_size / 2);
+	}
+
+	_sdata = . ;
+	RW_DATA_SECTION(32, PAGE_SIZE, THREAD_SIZE)
+	_edata = . ;
+
+	/* Under the microblaze ABI, .sdata and .sbss must be contiguous */
+	. = ALIGN(8);
+	.sdata : AT(ADDR(.sdata) - LOAD_OFFSET) {
+		_ssro = .;
+		*(.sdata)
+	}
+
+	.sbss :	AT(ADDR(.sbss) - LOAD_OFFSET) {
+		_ssbss = .;
+		*(.sbss)
+		_esbss = .;
+		_essro = .;
+		_ssro_size = _essro - _ssro ;
+		_KERNEL_SDA_BASE_ = _ssro + (_ssro_size / 2) ;
+	}
+
+	. = ALIGN(PAGE_SIZE);
+	__init_begin = .;
+
+	INIT_TEXT_SECTION(PAGE_SIZE)
+
+	.init.data : AT(ADDR(.init.data) - LOAD_OFFSET) {
+		INIT_DATA
+	}
+
+	. = ALIGN(4);
+	.init.ivt : AT(ADDR(.init.ivt) - LOAD_OFFSET) {
+		__ivt_start = .;
+		*(.init.ivt)
+		__ivt_end = .;
+	}
+
+	.init.setup : AT(ADDR(.init.setup) - LOAD_OFFSET) {
+		INIT_SETUP(0)
+	}
+
+	.initcall.init : AT(ADDR(.initcall.init) - LOAD_OFFSET ) {
+		INIT_CALLS
+	}
+
+	.con_initcall.init : AT(ADDR(.con_initcall.init) - LOAD_OFFSET) {
+		CON_INITCALL
+	}
+
+	SECURITY_INIT
+
+	__init_end_before_initramfs = .;
+
+	.init.ramfs : AT(ADDR(.init.ramfs) - LOAD_OFFSET) {
+		INIT_RAM_FS
+	}
+
+	__init_end = .;
+
+	.bss ALIGN (PAGE_SIZE) : AT(ADDR(.bss) - LOAD_OFFSET) {
+		/* page aligned when MMU used */
+		__bss_start = . ;
+			*(.bss*)
+			*(COMMON)
+		. = ALIGN (4) ;
+		__bss_stop = . ;
+	}
+	. = ALIGN(PAGE_SIZE);
+	_end = .;
+
+	DISCARDS
+}