diff --git a/arch/microblaze/Kconfig b/arch/microblaze/Kconfig
new file mode 100644
index 0000000..0bce820
--- /dev/null
+++ b/arch/microblaze/Kconfig
@@ -0,0 +1,297 @@
+config MICROBLAZE
+	def_bool y
+	select ARCH_HAS_GCOV_PROFILE_ALL
+	select ARCH_MIGHT_HAVE_PC_PARPORT
+	select ARCH_WANT_IPC_PARSE_VERSION
+	select ARCH_WANT_OPTIONAL_GPIOLIB
+	select BUILDTIME_EXTABLE_SORT
+	select CLKSRC_OF
+	select CLONE_BACKWARDS3
+	select COMMON_CLK
+	select GENERIC_ATOMIC64
+	select GENERIC_CLOCKEVENTS
+	select GENERIC_CPU_DEVICES
+	select GENERIC_IDLE_POLL_SETUP
+	select GENERIC_IRQ_PROBE
+	select GENERIC_IRQ_SHOW
+	select GENERIC_PCI_IOMAP
+	select GENERIC_SCHED_CLOCK
+	select HAVE_ARCH_KGDB
+	select HAVE_DEBUG_KMEMLEAK
+	select HAVE_DMA_API_DEBUG
+	select HAVE_DMA_ATTRS
+	select HAVE_DYNAMIC_FTRACE
+	select HAVE_FTRACE_MCOUNT_RECORD
+	select HAVE_FUNCTION_GRAPH_TRACER
+	select HAVE_FUNCTION_TRACER
+	select HAVE_MEMBLOCK
+	select HAVE_MEMBLOCK_NODE_MAP
+	select HAVE_OPROFILE
+	select IRQ_DOMAIN
+	select MODULES_USE_ELF_RELA
+	select OF
+	select OF_EARLY_FLATTREE
+	select TRACING_SUPPORT
+	select VIRT_TO_BUS
+
+config SWAP
+	def_bool n
+
+config RWSEM_GENERIC_SPINLOCK
+	def_bool y
+
+config ZONE_DMA
+	def_bool y
+
+config RWSEM_XCHGADD_ALGORITHM
+	bool
+
+config ARCH_HAS_ILOG2_U32
+	def_bool n
+
+config ARCH_HAS_ILOG2_U64
+	def_bool n
+
+config GENERIC_HWEIGHT
+	def_bool y
+
+config GENERIC_CALIBRATE_DELAY
+	def_bool y
+
+config GENERIC_CSUM
+	def_bool y
+
+config STACKTRACE_SUPPORT
+	def_bool y
+
+config LOCKDEP_SUPPORT
+	def_bool y
+
+config HAVE_LATENCYTOP_SUPPORT
+	def_bool y
+
+source "init/Kconfig"
+
+source "kernel/Kconfig.freezer"
+
+source "arch/microblaze/Kconfig.platform"
+
+menu "Processor type and features"
+
+source "kernel/Kconfig.preempt"
+
+source "kernel/Kconfig.hz"
+
+config MMU
+	bool "MMU support"
+	default n
+
+comment "Boot options"
+
+config CMDLINE_BOOL
+	bool "Default bootloader kernel arguments"
+
+config CMDLINE
+	string "Default kernel command string"
+	depends on CMDLINE_BOOL
+	default "console=ttyUL0,115200"
+	help
+	  On some architectures there is currently no way for the boot loader
+	  to pass arguments to the kernel. For these architectures, you should
+	  supply some command-line options at build time by entering them
+	  here.
+
+config CMDLINE_FORCE
+	bool "Force default kernel command string"
+	depends on CMDLINE_BOOL
+	default n
+	help
+	  Set this to have arguments from the default kernel command string
+	  override those passed by the boot loader.
+
+config SECCOMP
+	bool "Enable seccomp to safely compute untrusted bytecode"
+	depends on PROC_FS
+	default y
+	help
+	  This kernel feature is useful for number crunching applications
+	  that may need to compute untrusted bytecode during their
+	  execution. By using pipes or other transports made available to
+	  the process as file descriptors supporting the read/write
+	  syscalls, it's possible to isolate those applications in
+	  their own address space using seccomp. Once seccomp is
+	  enabled via /proc/<pid>/seccomp, it cannot be disabled
+	  and the task is only allowed to execute a few safe syscalls
+	  defined by each seccomp mode.
+
+	  If unsure, say Y. Only embedded should say N here.
+
+endmenu
+
+menu "Kernel features"
+
+config NR_CPUS
+	int
+	default "1"
+
+config ADVANCED_OPTIONS
+	bool "Prompt for advanced kernel configuration options"
+	help
+	  This option will enable prompting for a variety of advanced kernel
+	  configuration options.  These options can cause the kernel to not
+	  work if they are set incorrectly, but can be used to optimize certain
+	  aspects of kernel memory management.
+
+	  Unless you know what you are doing, say N here.
+
+comment "Default settings for advanced configuration options are used"
+	depends on !ADVANCED_OPTIONS
+
+config XILINX_UNCACHED_SHADOW
+	bool "Are you using uncached shadow for RAM ?"
+	depends on ADVANCED_OPTIONS && !MMU
+	default n
+	help
+	  This is needed to be able to allocate uncachable memory regions.
+	  The feature requires the design to define the RAM memory controller
+	  window to be twice as large as the actual physical memory.
+
+config HIGHMEM
+	bool "High memory support"
+	depends on MMU
+	help
+	  The address space of Microblaze processors is only 4 Gigabytes large
+	  and it has to accommodate user address space, kernel address
+	  space as well as some memory mapped IO. That means that, if you
+	  have a large amount of physical memory and/or IO, not all of the
+	  memory can be "permanently mapped" by the kernel. The physical
+	  memory that is not permanently mapped is called "high memory".
+
+	  If unsure, say n.
+
+config LOWMEM_SIZE_BOOL
+	bool "Set maximum low memory"
+	depends on ADVANCED_OPTIONS && MMU
+	help
+	  This option allows you to set the maximum amount of memory which
+	  will be used as "low memory", that is, memory which the kernel can
+	  access directly, without having to set up a kernel virtual mapping.
+	  This can be useful in optimizing the layout of kernel virtual
+	  memory.
+
+	  Say N here unless you know what you are doing.
+
+config LOWMEM_SIZE
+	hex "Maximum low memory size (in bytes)" if LOWMEM_SIZE_BOOL
+	default "0x30000000"
+
+config MANUAL_RESET_VECTOR
+	hex "Microblaze reset vector address setup"
+	default "0x0"
+	help
+	  Set this option to have the kernel override the CPU Reset vector.
+	  If zero, no change will be made to the MicroBlaze reset vector at
+	  address 0x0.
+	  If non-zero, a jump instruction to this address, will be written
+	  to the reset vector at address 0x0.
+	  If you are unsure, set it to default value 0x0.
+
+config KERNEL_START_BOOL
+	bool "Set custom kernel base address"
+	depends on ADVANCED_OPTIONS
+	help
+	  This option allows you to set the kernel virtual address at which
+	  the kernel will map low memory (the kernel image will be linked at
+	  this address).  This can be useful in optimizing the virtual memory
+	  layout of the system.
+
+	  Say N here unless you know what you are doing.
+
+config KERNEL_START
+	hex "Virtual address of kernel base" if KERNEL_START_BOOL
+	default "0xc0000000" if MMU
+	default KERNEL_BASE_ADDR if !MMU
+
+config TASK_SIZE_BOOL
+	bool "Set custom user task size"
+	depends on ADVANCED_OPTIONS && MMU
+	help
+	  This option allows you to set the amount of virtual address space
+	  allocated to user tasks.  This can be useful in optimizing the
+	  virtual memory layout of the system.
+
+	  Say N here unless you know what you are doing.
+
+config TASK_SIZE
+	hex "Size of user task space" if TASK_SIZE_BOOL
+	default "0x80000000"
+
+choice
+	prompt "Page size"
+	default MICROBLAZE_4K_PAGES
+	depends on ADVANCED_OPTIONS && !MMU
+	help
+	  Select the kernel logical page size. Increasing the page size
+	  will reduce software overhead at each page boundary, allow
+	  hardware prefetch mechanisms to be more effective, and allow
+	  larger dma transfers increasing IO efficiency and reducing
+	  overhead. However the utilization of memory will increase.
+	  For example, each cached file will using a multiple of the
+	  page size to hold its contents and the difference between the
+	  end of file and the end of page is wasted.
+
+	  If unsure, choose 4K_PAGES.
+
+config MICROBLAZE_4K_PAGES
+	bool "4k page size"
+
+config MICROBLAZE_16K_PAGES
+	bool "16k page size"
+
+config MICROBLAZE_64K_PAGES
+	bool "64k page size"
+
+endchoice
+
+source "mm/Kconfig"
+
+endmenu
+
+menu "Executable file formats"
+
+source "fs/Kconfig.binfmt"
+
+endmenu
+
+menu "Bus Options"
+
+config PCI
+	bool "PCI support"
+
+config PCI_DOMAINS
+	def_bool PCI
+
+config PCI_SYSCALL
+	def_bool PCI
+
+config PCI_XILINX
+	bool "Xilinx PCI host bridge support"
+	depends on PCI
+
+source "drivers/pci/Kconfig"
+
+endmenu
+
+source "net/Kconfig"
+
+source "drivers/Kconfig"
+
+source "fs/Kconfig"
+
+source "arch/microblaze/Kconfig.debug"
+
+source "security/Kconfig"
+
+source "crypto/Kconfig"
+
+source "lib/Kconfig"
diff --git a/arch/microblaze/Kconfig.debug b/arch/microblaze/Kconfig.debug
new file mode 100644
index 0000000..012e377
--- /dev/null
+++ b/arch/microblaze/Kconfig.debug
@@ -0,0 +1,26 @@
+# For a description of the syntax of this configuration file,
+# see Documentation/kbuild/kconfig-language.txt.
+
+menu "Kernel hacking"
+
+config TRACE_IRQFLAGS_SUPPORT
+	def_bool y
+
+source "lib/Kconfig.debug"
+
+config EARLY_PRINTK
+	bool "Early printk function for kernel"
+	depends on SERIAL_UARTLITE_CONSOLE || SERIAL_8250_CONSOLE
+	default n
+	help
+	  This option turns on/off early printk messages to console.
+	  First Uartlite node is taken.
+
+config HEART_BEAT
+	bool "Heart beat function for kernel"
+	default n
+	help
+	  This option turns on/off heart beat kernel functionality.
+	  First GPIO node is taken.
+
+endmenu
diff --git a/arch/microblaze/Kconfig.platform b/arch/microblaze/Kconfig.platform
new file mode 100644
index 0000000..1b3d8c8
--- /dev/null
+++ b/arch/microblaze/Kconfig.platform
@@ -0,0 +1,69 @@
+# For a description of the syntax of this configuration file,
+# see Documentation/kbuild/kconfig-language.txt.
+#
+# Platform selection Kconfig menu for MicroBlaze targets
+#
+
+menu "Platform options"
+
+config OPT_LIB_FUNCTION
+	bool "Optimalized lib function"
+	default y
+	help
+	  Allows turn on optimalized library function (memcpy and memmove).
+	  They are optimized by using word alignment. This will work
+	  fine if both source and destination are aligned on the same
+	  boundary. However, if they are aligned on different boundaries
+	  shifts will be necessary. This might result in bad performance
+	  on MicroBlaze systems without a barrel shifter.
+
+config OPT_LIB_ASM
+	bool "Optimalized lib function ASM"
+	depends on OPT_LIB_FUNCTION && (XILINX_MICROBLAZE0_USE_BARREL = 1)
+	default n
+	help
+	  Allows turn on optimalized library function (memcpy and memmove).
+	  Function are written in asm code.
+
+# Definitions for MICROBLAZE0
+comment "Definitions for MICROBLAZE0"
+
+config KERNEL_BASE_ADDR
+	hex "Physical address where Linux Kernel is"
+	default "0x90000000"
+	help
+	  BASE Address for kernel
+
+config XILINX_MICROBLAZE0_FAMILY
+	string "Targeted FPGA family"
+	default "virtex5"
+
+config XILINX_MICROBLAZE0_USE_MSR_INSTR
+	int "USE_MSR_INSTR range (0:1)"
+	default 0
+
+config XILINX_MICROBLAZE0_USE_PCMP_INSTR
+	int "USE_PCMP_INSTR range (0:1)"
+	default 0
+
+config XILINX_MICROBLAZE0_USE_BARREL
+	int "USE_BARREL range (0:1)"
+	default 0
+
+config XILINX_MICROBLAZE0_USE_DIV
+	int "USE_DIV range (0:1)"
+	default 0
+
+config XILINX_MICROBLAZE0_USE_HW_MUL
+	int "USE_HW_MUL values (0=NONE, 1=MUL32, 2=MUL64)"
+	default 0
+
+config XILINX_MICROBLAZE0_USE_FPU
+	int "USE_FPU values (0=NONE, 1=BASIC, 2=EXTENDED)"
+	default 0
+
+config XILINX_MICROBLAZE0_HW_VER
+	string "Core version number"
+	default 7.10.d
+
+endmenu
diff --git a/arch/microblaze/Makefile b/arch/microblaze/Makefile
new file mode 100644
index 0000000..740f2b8
--- /dev/null
+++ b/arch/microblaze/Makefile
@@ -0,0 +1,98 @@
+KBUILD_DEFCONFIG := mmu_defconfig
+
+ifeq ($(CONFIG_MMU),y)
+UTS_SYSNAME = -DUTS_SYSNAME=\"Linux\"
+else
+UTS_SYSNAME = -DUTS_SYSNAME=\"uClinux\"
+endif
+
+# What CPU vesion are we building for, and crack it open
+# as major.minor.rev
+CPU_VER   := $(shell echo $(CONFIG_XILINX_MICROBLAZE0_HW_VER))
+CPU_MAJOR := $(shell echo $(CPU_VER) | cut -d '.' -f 1)
+CPU_MINOR := $(shell echo $(CPU_VER) | cut -d '.' -f 2)
+CPU_REV   := $(shell echo $(CPU_VER) | cut -d '.' -f 3)
+
+export CPU_VER CPU_MAJOR CPU_MINOR CPU_REV
+
+# Use cpu-related CONFIG_ vars to set compile options.
+# The various CONFIG_XILINX cpu features options are integers 0/1/2...
+# rather than bools y/n
+
+# Work out HW multipler support. This is tricky.
+# 1. Spartan2 has no HW multipliers.
+# 2. MicroBlaze v3.x always uses them, except in Spartan 2
+# 3. All other FPGa/CPU ver combos, we can trust the CONFIG_ settings
+ifeq (,$(findstring spartan2,$(CONFIG_XILINX_MICROBLAZE0_FAMILY)))
+  ifeq ($(CPU_MAJOR),3)
+    CPUFLAGS-1 += -mno-xl-soft-mul
+  else
+    # USE_HW_MUL can be 0, 1, or 2, defining a hierarchy of HW Mul support.
+    CPUFLAGS-$(subst 1,,$(CONFIG_XILINX_MICROBLAZE0_USE_HW_MUL)) += -mxl-multiply-high
+    CPUFLAGS-$(CONFIG_XILINX_MICROBLAZE0_USE_HW_MUL) += -mno-xl-soft-mul
+  endif
+endif
+CPUFLAGS-$(CONFIG_XILINX_MICROBLAZE0_USE_DIV) += -mno-xl-soft-div
+CPUFLAGS-$(CONFIG_XILINX_MICROBLAZE0_USE_BARREL) += -mxl-barrel-shift
+CPUFLAGS-$(CONFIG_XILINX_MICROBLAZE0_USE_PCMP_INSTR) += -mxl-pattern-compare
+
+CPUFLAGS-1 += $(call cc-option,-mcpu=v$(CPU_VER))
+
+# r31 holds current when in kernel mode
+KBUILD_CFLAGS += -ffixed-r31 $(CPUFLAGS-1) $(CPUFLAGS-2)
+
+LDFLAGS		:=
+LDFLAGS_vmlinux	:=
+
+head-y := arch/microblaze/kernel/head.o
+libs-y += arch/microblaze/lib/
+core-y += arch/microblaze/kernel/
+core-y += arch/microblaze/mm/
+core-$(CONFIG_PCI) += arch/microblaze/pci/
+
+drivers-$(CONFIG_OPROFILE) += arch/microblaze/oprofile/
+
+boot := arch/microblaze/boot
+
+# Are we making a simpleImage.<boardname> target? If so, crack out the boardname
+DTB:=$(subst simpleImage.,,$(filter simpleImage.%, $(MAKECMDGOALS)))
+
+ifneq ($(DTB),)
+	core-y	+= $(boot)/dts/
+endif
+
+# defines filename extension depending memory management type
+ifeq ($(CONFIG_MMU),)
+MMU := -nommu
+endif
+
+export MMU DTB
+
+all: linux.bin
+
+archclean:
+	$(Q)$(MAKE) $(clean)=$(boot)
+
+linux.bin linux.bin.gz linux.bin.ub: vmlinux
+	$(Q)$(MAKE) $(build)=$(boot) $(boot)/$@
+
+simpleImage.%: vmlinux
+	$(Q)$(MAKE) $(build)=$(boot) $(boot)/$@
+
+define archhelp
+  echo '* linux.bin    - Create raw binary'
+  echo '  linux.bin.gz - Create compressed raw binary'
+  echo '  linux.bin.ub - Create U-Boot wrapped raw binary'
+  echo '  simpleImage.<dt> - ELF image with $(arch)/boot/dts/<dt>.dts linked in'
+  echo '                   - stripped elf with fdt blob'
+  echo '  simpleImage.<dt>.unstrip - full ELF image with fdt blob'
+  echo '  *_defconfig      - Select default config from arch/microblaze/configs'
+  echo ''
+  echo '  Targets with <dt> embed a device tree blob inside the image'
+  echo '  These targets support board with firmware that does not'
+  echo '  support passing a device tree directly. Replace <dt> with the'
+  echo '  name of a dts file from the arch/microblaze/boot/dts/ directory'
+  echo '  (minus the .dts extension).'
+endef
+
+MRPROPER_FILES += $(boot)/simpleImage.*
diff --git a/arch/microblaze/boot/.gitignore b/arch/microblaze/boot/.gitignore
new file mode 100644
index 0000000..bf04591
--- /dev/null
+++ b/arch/microblaze/boot/.gitignore
@@ -0,0 +1,3 @@
+*.dtb
+linux.bin*
+simpleImage.*
diff --git a/arch/microblaze/boot/Makefile b/arch/microblaze/boot/Makefile
new file mode 100644
index 0000000..91d2068
--- /dev/null
+++ b/arch/microblaze/boot/Makefile
@@ -0,0 +1,37 @@
+#
+# arch/microblaze/boot/Makefile
+#
+
+targets := linux.bin linux.bin.gz linux.bin.ub simpleImage.%
+
+OBJCOPYFLAGS := -R .note -R .comment -R .note.gnu.build-id -O binary
+
+$(obj)/linux.bin: vmlinux FORCE
+	$(call if_changed,objcopy)
+	@echo 'Kernel: $@ is ready' ' (#'`cat .version`')'
+
+$(obj)/linux.bin.ub: $(obj)/linux.bin FORCE
+	$(call if_changed,uimage)
+	@echo 'Kernel: $@ is ready' ' (#'`cat .version`')'
+
+$(obj)/linux.bin.gz: $(obj)/linux.bin FORCE
+	$(call if_changed,gzip)
+	@echo 'Kernel: $@ is ready' ' (#'`cat .version`')'
+
+quiet_cmd_cp = CP      $< $@$2
+	cmd_cp = cat $< >$@$2 || (rm -f $@ && echo false)
+
+quiet_cmd_strip = STRIP   $@
+	cmd_strip = $(STRIP) -K microblaze_start -K _end -K __log_buf \
+				-K _fdt_start vmlinux -o $@
+
+UIMAGE_LOADADDR = $(CONFIG_KERNEL_BASE_ADDR)
+
+$(obj)/simpleImage.%: vmlinux FORCE
+	$(call if_changed,cp,.unstrip)
+	$(call if_changed,objcopy)
+	$(call if_changed,uimage)
+	$(call if_changed,strip)
+	@echo 'Kernel: $@ is ready' ' (#'`cat .version`')'
+
+clean-files += simpleImage.*.unstrip linux.bin.ub dts/*.dtb
diff --git a/arch/microblaze/boot/dts/Makefile b/arch/microblaze/boot/dts/Makefile
new file mode 100644
index 0000000..a3d2e42
--- /dev/null
+++ b/arch/microblaze/boot/dts/Makefile
@@ -0,0 +1,18 @@
+#
+
+obj-y += linked_dtb.o
+
+# Ensure system.dtb exists
+$(obj)/linked_dtb.o: $(obj)/system.dtb
+
+# Generate system.dtb from $(DTB).dtb
+ifneq ($(DTB),system)
+$(obj)/system.dtb: $(obj)/$(DTB).dtb
+	$(call if_changed,cp)
+endif
+
+quiet_cmd_cp = CP      $< $@$2
+	cmd_cp = cat $< >$@$2 || (rm -f $@ && echo false)
+
+# Rule to build device tree blobs
+DTC_FLAGS := -p 1024
diff --git a/arch/microblaze/boot/dts/linked_dtb.S b/arch/microblaze/boot/dts/linked_dtb.S
new file mode 100644
index 0000000..23345af
--- /dev/null
+++ b/arch/microblaze/boot/dts/linked_dtb.S
@@ -0,0 +1,2 @@
+.section __fdt_blob,"a"
+.incbin "arch/microblaze/boot/dts/system.dtb"
diff --git a/arch/microblaze/boot/dts/system.dts b/arch/microblaze/boot/dts/system.dts
new file mode 100644
index 0000000..b620da2
--- /dev/null
+++ b/arch/microblaze/boot/dts/system.dts
@@ -0,0 +1,366 @@
+/*
+ * Device Tree Generator version: 1.1
+ *
+ * (C) Copyright 2007-2008 Xilinx, Inc.
+ * (C) Copyright 2007-2009 Michal Simek
+ *
+ * Michal SIMEK <monstr@monstr.eu>
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston,
+ * MA 02111-1307 USA
+ *
+ * CAUTION: This file is automatically generated by libgen.
+ * Version: Xilinx EDK 10.1.03 EDK_K_SP3.6
+ *
+ * XPS project directory: Xilinx-ML505-ll_temac-sgdma-MMU-FDT-edk101
+ */
+
+/dts-v1/;
+/ {
+	#address-cells = <1>;
+	#size-cells = <1>;
+	compatible = "xlnx,microblaze";
+	hard-reset-gpios = <&LEDs_8Bit 2 1>;
+	model = "testing";
+	DDR2_SDRAM: memory@90000000 {
+		device_type = "memory";
+		reg = < 0x90000000 0x10000000 >;
+	} ;
+	aliases {
+		ethernet0 = &Hard_Ethernet_MAC;
+		serial0 = &RS232_Uart_1;
+	} ;
+	chosen {
+		bootargs = "console=ttyUL0,115200 highres=on";
+		linux,stdout-path = "/plb@0/serial@84000000";
+	} ;
+	cpus {
+		#address-cells = <1>;
+		#cpus = <0x1>;
+		#size-cells = <0>;
+		microblaze_0: cpu@0 {
+			clock-frequency = <125000000>;
+			compatible = "xlnx,microblaze-7.10.d";
+			d-cache-baseaddr = <0x90000000>;
+			d-cache-highaddr = <0x9fffffff>;
+			d-cache-line-size = <0x10>;
+			d-cache-size = <0x2000>;
+			device_type = "cpu";
+			i-cache-baseaddr = <0x90000000>;
+			i-cache-highaddr = <0x9fffffff>;
+			i-cache-line-size = <0x10>;
+			i-cache-size = <0x2000>;
+			model = "microblaze,7.10.d";
+			reg = <0>;
+			timebase-frequency = <125000000>;
+			xlnx,addr-tag-bits = <0xf>;
+			xlnx,allow-dcache-wr = <0x1>;
+			xlnx,allow-icache-wr = <0x1>;
+			xlnx,area-optimized = <0x0>;
+			xlnx,cache-byte-size = <0x2000>;
+			xlnx,d-lmb = <0x1>;
+			xlnx,d-opb = <0x0>;
+			xlnx,d-plb = <0x1>;
+			xlnx,data-size = <0x20>;
+			xlnx,dcache-addr-tag = <0xf>;
+			xlnx,dcache-always-used = <0x1>;
+			xlnx,dcache-byte-size = <0x2000>;
+			xlnx,dcache-line-len = <0x4>;
+			xlnx,dcache-use-fsl = <0x1>;
+			xlnx,debug-enabled = <0x1>;
+			xlnx,div-zero-exception = <0x1>;
+			xlnx,dopb-bus-exception = <0x0>;
+			xlnx,dynamic-bus-sizing = <0x1>;
+			xlnx,edge-is-positive = <0x1>;
+			xlnx,family = "virtex5";
+			xlnx,endianness = <0x1>;
+			xlnx,fpu-exception = <0x1>;
+			xlnx,fsl-data-size = <0x20>;
+			xlnx,fsl-exception = <0x0>;
+			xlnx,fsl-links = <0x0>;
+			xlnx,i-lmb = <0x1>;
+			xlnx,i-opb = <0x0>;
+			xlnx,i-plb = <0x1>;
+			xlnx,icache-always-used = <0x1>;
+			xlnx,icache-line-len = <0x4>;
+			xlnx,icache-use-fsl = <0x1>;
+			xlnx,ill-opcode-exception = <0x1>;
+			xlnx,instance = "microblaze_0";
+			xlnx,interconnect = <0x1>;
+			xlnx,interrupt-is-edge = <0x0>;
+			xlnx,iopb-bus-exception = <0x0>;
+			xlnx,mmu-dtlb-size = <0x4>;
+			xlnx,mmu-itlb-size = <0x2>;
+			xlnx,mmu-tlb-access = <0x3>;
+			xlnx,mmu-zones = <0x10>;
+			xlnx,number-of-pc-brk = <0x1>;
+			xlnx,number-of-rd-addr-brk = <0x0>;
+			xlnx,number-of-wr-addr-brk = <0x0>;
+			xlnx,opcode-0x0-illegal = <0x1>;
+			xlnx,pvr = <0x2>;
+			xlnx,pvr-user1 = <0x0>;
+			xlnx,pvr-user2 = <0x0>;
+			xlnx,reset-msr = <0x0>;
+			xlnx,sco = <0x0>;
+			xlnx,unaligned-exceptions = <0x1>;
+			xlnx,use-barrel = <0x1>;
+			xlnx,use-dcache = <0x1>;
+			xlnx,use-div = <0x1>;
+			xlnx,use-ext-brk = <0x1>;
+			xlnx,use-ext-nm-brk = <0x1>;
+			xlnx,use-extended-fsl-instr = <0x0>;
+			xlnx,use-fpu = <0x2>;
+			xlnx,use-hw-mul = <0x2>;
+			xlnx,use-icache = <0x1>;
+			xlnx,use-interrupt = <0x1>;
+			xlnx,use-mmu = <0x3>;
+			xlnx,use-msr-instr = <0x1>;
+			xlnx,use-pcmp-instr = <0x1>;
+		} ;
+	} ;
+	mb_plb: plb@0 {
+		#address-cells = <1>;
+		#size-cells = <1>;
+		compatible = "xlnx,plb-v46-1.03.a", "xlnx,plb-v46-1.00.a", "simple-bus";
+		ranges ;
+		FLASH: flash@a0000000 {
+			bank-width = <2>;
+			compatible = "xlnx,xps-mch-emc-2.00.a", "cfi-flash";
+			reg = < 0xa0000000 0x2000000 >;
+			xlnx,family = "virtex5";
+			xlnx,include-datawidth-matching-0 = <0x1>;
+			xlnx,include-datawidth-matching-1 = <0x0>;
+			xlnx,include-datawidth-matching-2 = <0x0>;
+			xlnx,include-datawidth-matching-3 = <0x0>;
+			xlnx,include-negedge-ioregs = <0x0>;
+			xlnx,include-plb-ipif = <0x1>;
+			xlnx,include-wrbuf = <0x1>;
+			xlnx,max-mem-width = <0x10>;
+			xlnx,mch-native-dwidth = <0x20>;
+			xlnx,mch-plb-clk-period-ps = <0x1f40>;
+			xlnx,mch-splb-awidth = <0x20>;
+			xlnx,mch0-accessbuf-depth = <0x10>;
+			xlnx,mch0-protocol = <0x0>;
+			xlnx,mch0-rddatabuf-depth = <0x10>;
+			xlnx,mch1-accessbuf-depth = <0x10>;
+			xlnx,mch1-protocol = <0x0>;
+			xlnx,mch1-rddatabuf-depth = <0x10>;
+			xlnx,mch2-accessbuf-depth = <0x10>;
+			xlnx,mch2-protocol = <0x0>;
+			xlnx,mch2-rddatabuf-depth = <0x10>;
+			xlnx,mch3-accessbuf-depth = <0x10>;
+			xlnx,mch3-protocol = <0x0>;
+			xlnx,mch3-rddatabuf-depth = <0x10>;
+			xlnx,mem0-width = <0x10>;
+			xlnx,mem1-width = <0x20>;
+			xlnx,mem2-width = <0x20>;
+			xlnx,mem3-width = <0x20>;
+			xlnx,num-banks-mem = <0x1>;
+			xlnx,num-channels = <0x0>;
+			xlnx,priority-mode = <0x0>;
+			xlnx,synch-mem-0 = <0x0>;
+			xlnx,synch-mem-1 = <0x0>;
+			xlnx,synch-mem-2 = <0x0>;
+			xlnx,synch-mem-3 = <0x0>;
+			xlnx,synch-pipedelay-0 = <0x2>;
+			xlnx,synch-pipedelay-1 = <0x2>;
+			xlnx,synch-pipedelay-2 = <0x2>;
+			xlnx,synch-pipedelay-3 = <0x2>;
+			xlnx,tavdv-ps-mem-0 = <0x1adb0>;
+			xlnx,tavdv-ps-mem-1 = <0x3a98>;
+			xlnx,tavdv-ps-mem-2 = <0x3a98>;
+			xlnx,tavdv-ps-mem-3 = <0x3a98>;
+			xlnx,tcedv-ps-mem-0 = <0x1adb0>;
+			xlnx,tcedv-ps-mem-1 = <0x3a98>;
+			xlnx,tcedv-ps-mem-2 = <0x3a98>;
+			xlnx,tcedv-ps-mem-3 = <0x3a98>;
+			xlnx,thzce-ps-mem-0 = <0x88b8>;
+			xlnx,thzce-ps-mem-1 = <0x1b58>;
+			xlnx,thzce-ps-mem-2 = <0x1b58>;
+			xlnx,thzce-ps-mem-3 = <0x1b58>;
+			xlnx,thzoe-ps-mem-0 = <0x1b58>;
+			xlnx,thzoe-ps-mem-1 = <0x1b58>;
+			xlnx,thzoe-ps-mem-2 = <0x1b58>;
+			xlnx,thzoe-ps-mem-3 = <0x1b58>;
+			xlnx,tlzwe-ps-mem-0 = <0x88b8>;
+			xlnx,tlzwe-ps-mem-1 = <0x0>;
+			xlnx,tlzwe-ps-mem-2 = <0x0>;
+			xlnx,tlzwe-ps-mem-3 = <0x0>;
+			xlnx,twc-ps-mem-0 = <0x2af8>;
+			xlnx,twc-ps-mem-1 = <0x3a98>;
+			xlnx,twc-ps-mem-2 = <0x3a98>;
+			xlnx,twc-ps-mem-3 = <0x3a98>;
+			xlnx,twp-ps-mem-0 = <0x11170>;
+			xlnx,twp-ps-mem-1 = <0x2ee0>;
+			xlnx,twp-ps-mem-2 = <0x2ee0>;
+			xlnx,twp-ps-mem-3 = <0x2ee0>;
+			xlnx,xcl0-linesize = <0x4>;
+			xlnx,xcl0-writexfer = <0x1>;
+			xlnx,xcl1-linesize = <0x4>;
+			xlnx,xcl1-writexfer = <0x1>;
+			xlnx,xcl2-linesize = <0x4>;
+			xlnx,xcl2-writexfer = <0x1>;
+			xlnx,xcl3-linesize = <0x4>;
+			xlnx,xcl3-writexfer = <0x1>;
+		} ;
+		Hard_Ethernet_MAC: xps-ll-temac@81c00000 {
+			#address-cells = <1>;
+			#size-cells = <1>;
+			compatible = "xlnx,compound";
+			ranges ;
+			ethernet@81c00000 {
+				compatible = "xlnx,xps-ll-temac-1.01.b", "xlnx,xps-ll-temac-1.00.a";
+				interrupt-parent = <&xps_intc_0>;
+				interrupts = < 5 2 >;
+				llink-connected = <&PIM3>;
+				local-mac-address = [ 00 0a 35 00 00 00 ];
+				reg = < 0x81c00000 0x40 >;
+				xlnx,bus2core-clk-ratio = <0x1>;
+				xlnx,phy-type = <0x1>;
+				xlnx,phyaddr = <0x1>;
+				xlnx,rxcsum = <0x0>;
+				xlnx,rxfifo = <0x1000>;
+				xlnx,temac-type = <0x0>;
+				xlnx,txcsum = <0x0>;
+				xlnx,txfifo = <0x1000>;
+			} ;
+		} ;
+		IIC_EEPROM: i2c@81600000 {
+			compatible = "xlnx,xps-iic-2.00.a";
+			interrupt-parent = <&xps_intc_0>;
+			interrupts = < 6 2 >;
+			reg = < 0x81600000 0x10000 >;
+			xlnx,clk-freq = <0x7735940>;
+			xlnx,family = "virtex5";
+			xlnx,gpo-width = <0x1>;
+			xlnx,iic-freq = <0x186a0>;
+			xlnx,scl-inertial-delay = <0x0>;
+			xlnx,sda-inertial-delay = <0x0>;
+			xlnx,ten-bit-adr = <0x0>;
+		} ;
+		LEDs_8Bit: gpio@81400000 {
+			compatible = "xlnx,xps-gpio-1.00.a";
+			interrupt-parent = <&xps_intc_0>;
+			interrupts = < 7 2 >;
+			reg = < 0x81400000 0x10000 >;
+			xlnx,all-inputs = <0x0>;
+			xlnx,all-inputs-2 = <0x0>;
+			xlnx,dout-default = <0x0>;
+			xlnx,dout-default-2 = <0x0>;
+			xlnx,family = "virtex5";
+			xlnx,gpio-width = <0x8>;
+			xlnx,interrupt-present = <0x1>;
+			xlnx,is-bidir = <0x1>;
+			xlnx,is-bidir-2 = <0x1>;
+			xlnx,is-dual = <0x0>;
+			xlnx,tri-default = <0xffffffff>;
+			xlnx,tri-default-2 = <0xffffffff>;
+			#gpio-cells = <2>;
+			gpio-controller;
+		} ;
+
+		gpio-leds {
+			compatible = "gpio-leds";
+
+			heartbeat {
+				label = "Heartbeat";
+				gpios = <&LEDs_8Bit 4 1>;
+				linux,default-trigger = "heartbeat";
+			};
+
+			yellow {
+				label = "Yellow";
+				gpios = <&LEDs_8Bit 5 1>;
+			};
+
+			red {
+				label = "Red";
+				gpios = <&LEDs_8Bit 6 1>;
+			};
+
+			green {
+				label = "Green";
+				gpios = <&LEDs_8Bit 7 1>;
+			};
+		} ;
+		RS232_Uart_1: serial@84000000 {
+			clock-frequency = <125000000>;
+			compatible = "xlnx,xps-uartlite-1.00.a";
+			current-speed = <115200>;
+			device_type = "serial";
+			interrupt-parent = <&xps_intc_0>;
+			interrupts = < 8 0 >;
+			port-number = <0>;
+			reg = < 0x84000000 0x10000 >;
+			xlnx,baudrate = <0x1c200>;
+			xlnx,data-bits = <0x8>;
+			xlnx,family = "virtex5";
+			xlnx,odd-parity = <0x0>;
+			xlnx,use-parity = <0x0>;
+		} ;
+		SysACE_CompactFlash: sysace@83600000 {
+			compatible = "xlnx,xps-sysace-1.00.a";
+			interrupt-parent = <&xps_intc_0>;
+			interrupts = < 4 2 >;
+			reg = < 0x83600000 0x10000 >;
+			xlnx,family = "virtex5";
+			xlnx,mem-width = <0x10>;
+		} ;
+		debug_module: debug@84400000 {
+			compatible = "xlnx,mdm-1.00.d";
+			reg = < 0x84400000 0x10000 >;
+			xlnx,family = "virtex5";
+			xlnx,interconnect = <0x1>;
+			xlnx,jtag-chain = <0x2>;
+			xlnx,mb-dbg-ports = <0x1>;
+			xlnx,uart-width = <0x8>;
+			xlnx,use-uart = <0x1>;
+			xlnx,write-fsl-ports = <0x0>;
+		} ;
+		mpmc@90000000 {
+			#address-cells = <1>;
+			#size-cells = <1>;
+			compatible = "xlnx,mpmc-4.02.a";
+			ranges ;
+			PIM3: sdma@84600180 {
+				compatible = "xlnx,ll-dma-1.00.a";
+				interrupt-parent = <&xps_intc_0>;
+				interrupts = < 2 2 1 2 >;
+				reg = < 0x84600180 0x80 >;
+			} ;
+		} ;
+		xps_intc_0: interrupt-controller@81800000 {
+			#interrupt-cells = <0x2>;
+			compatible = "xlnx,xps-intc-1.00.a";
+			interrupt-controller ;
+			reg = < 0x81800000 0x10000 >;
+			xlnx,kind-of-intr = <0x100>;
+			xlnx,num-intr-inputs = <0x9>;
+		} ;
+		xps_timer_1: timer@83c00000 {
+			compatible = "xlnx,xps-timer-1.00.a";
+			interrupt-parent = <&xps_intc_0>;
+			interrupts = < 3 2 >;
+			reg = < 0x83c00000 0x10000 >;
+			xlnx,count-width = <0x20>;
+			xlnx,family = "virtex5";
+			xlnx,gen0-assert = <0x1>;
+			xlnx,gen1-assert = <0x1>;
+			xlnx,one-timer-only = <0x0>;
+			xlnx,trig0-assert = <0x1>;
+			xlnx,trig1-assert = <0x1>;
+		} ;
+	} ;
+}  ;
diff --git a/arch/microblaze/configs/mmu_defconfig b/arch/microblaze/configs/mmu_defconfig
new file mode 100644
index 0000000..e2f6543
--- /dev/null
+++ b/arch/microblaze/configs/mmu_defconfig
@@ -0,0 +1,91 @@
+CONFIG_SYSVIPC=y
+CONFIG_POSIX_MQUEUE=y
+CONFIG_FHANDLE=y
+CONFIG_AUDIT=y
+CONFIG_AUDIT_LOGINUID_IMMUTABLE=y
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
+CONFIG_SYSFS_DEPRECATED=y
+CONFIG_SYSFS_DEPRECATED_V2=y
+CONFIG_KALLSYMS_ALL=y
+# CONFIG_BASE_FULL is not set
+CONFIG_EMBEDDED=y
+CONFIG_SLAB=y
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+# CONFIG_BLK_DEV_BSG is not set
+CONFIG_PARTITION_ADVANCED=y
+# CONFIG_EFI_PARTITION is not set
+CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR=1
+CONFIG_XILINX_MICROBLAZE0_USE_PCMP_INSTR=1
+CONFIG_XILINX_MICROBLAZE0_USE_BARREL=1
+CONFIG_XILINX_MICROBLAZE0_USE_DIV=1
+CONFIG_XILINX_MICROBLAZE0_USE_HW_MUL=2
+CONFIG_XILINX_MICROBLAZE0_USE_FPU=2
+CONFIG_HZ_100=y
+CONFIG_MMU=y
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE_FORCE=y
+CONFIG_HIGHMEM=y
+CONFIG_PCI=y
+CONFIG_PCI_XILINX=y
+CONFIG_NET=y
+CONFIG_PACKET=y
+CONFIG_UNIX=y
+CONFIG_INET=y
+# CONFIG_INET_LRO is not set
+# CONFIG_IPV6 is not set
+CONFIG_MTD=y
+CONFIG_PROC_DEVICETREE=y
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_SIZE=8192
+CONFIG_NETDEVICES=y
+CONFIG_XILINX_EMACLITE=y
+CONFIG_XILINX_LL_TEMAC=y
+# CONFIG_INPUT is not set
+# CONFIG_SERIO is not set
+# CONFIG_VT is not set
+CONFIG_SERIAL_8250=y
+CONFIG_SERIAL_8250_CONSOLE=y
+CONFIG_SERIAL_UARTLITE=y
+CONFIG_SERIAL_UARTLITE_CONSOLE=y
+CONFIG_SERIAL_OF_PLATFORM=y
+# CONFIG_HW_RANDOM is not set
+CONFIG_XILINX_HWICAP=y
+CONFIG_I2C=y
+CONFIG_I2C_XILINX=y
+CONFIG_SPI=y
+CONFIG_SPI_XILINX=y
+CONFIG_GPIOLIB=y
+CONFIG_GPIO_SYSFS=y
+CONFIG_GPIO_XILINX=y
+# CONFIG_HWMON is not set
+CONFIG_WATCHDOG=y
+CONFIG_XILINX_WATCHDOG=y
+CONFIG_FB=y
+CONFIG_FB_XILINX=y
+# CONFIG_USB_SUPPORT is not set
+CONFIG_UIO=y
+CONFIG_UIO_PDRV=y
+CONFIG_UIO_PDRV_GENIRQ=y
+CONFIG_UIO_DMEM_GENIRQ=y
+CONFIG_EXT2_FS=y
+# CONFIG_DNOTIFY is not set
+CONFIG_CRAMFS=y
+CONFIG_ROMFS_FS=y
+CONFIG_NFS_FS=y
+CONFIG_CIFS=y
+CONFIG_CIFS_STATS=y
+CONFIG_CIFS_STATS2=y
+CONFIG_DETECT_HUNG_TASK=y
+CONFIG_DEBUG_SLAB=y
+CONFIG_DEBUG_SPINLOCK=y
+CONFIG_DEBUG_INFO=y
+CONFIG_KGDB=y
+CONFIG_KGDB_TESTS=y
+CONFIG_KGDB_KDB=y
+CONFIG_EARLY_PRINTK=y
+CONFIG_KEYS=y
+CONFIG_ENCRYPTED_KEYS=y
+CONFIG_KEYS_DEBUG_PROC_KEYS=y
+# CONFIG_CRYPTO_ANSI_CPRNG is not set
diff --git a/arch/microblaze/configs/nommu_defconfig b/arch/microblaze/configs/nommu_defconfig
new file mode 100644
index 0000000..a29ebd4
--- /dev/null
+++ b/arch/microblaze/configs/nommu_defconfig
@@ -0,0 +1,101 @@
+CONFIG_EXPERIMENTAL=y
+CONFIG_SYSVIPC=y
+CONFIG_POSIX_MQUEUE=y
+CONFIG_FHANDLE=y
+CONFIG_AUDIT=y
+CONFIG_AUDIT_LOGINUID_IMMUTABLE=y
+CONFIG_BSD_PROCESS_ACCT=y
+CONFIG_BSD_PROCESS_ACCT_V3=y
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
+CONFIG_SYSFS_DEPRECATED=y
+CONFIG_SYSFS_DEPRECATED_V2=y
+CONFIG_KALLSYMS_ALL=y
+# CONFIG_BASE_FULL is not set
+CONFIG_EMBEDDED=y
+CONFIG_SLAB=y
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+# CONFIG_BLK_DEV_BSG is not set
+CONFIG_PARTITION_ADVANCED=y
+# CONFIG_EFI_PARTITION is not set
+CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR=1
+CONFIG_XILINX_MICROBLAZE0_USE_PCMP_INSTR=1
+CONFIG_XILINX_MICROBLAZE0_USE_BARREL=1
+CONFIG_XILINX_MICROBLAZE0_USE_DIV=1
+CONFIG_XILINX_MICROBLAZE0_USE_HW_MUL=2
+CONFIG_XILINX_MICROBLAZE0_USE_FPU=2
+CONFIG_HZ_100=y
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE_FORCE=y
+CONFIG_PCI=y
+CONFIG_PCI_XILINX=y
+CONFIG_NET=y
+CONFIG_PACKET=y
+CONFIG_UNIX=y
+CONFIG_INET=y
+# CONFIG_INET_LRO is not set
+# CONFIG_IPV6 is not set
+CONFIG_MTD=y
+CONFIG_MTD_CMDLINE_PARTS=y
+CONFIG_MTD_CHAR=y
+CONFIG_MTD_BLOCK=y
+CONFIG_MTD_CFI=y
+CONFIG_MTD_CFI_INTELEXT=y
+CONFIG_MTD_CFI_AMDSTD=y
+CONFIG_MTD_RAM=y
+CONFIG_MTD_UCLINUX=y
+CONFIG_PROC_DEVICETREE=y
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_SIZE=8192
+CONFIG_NETDEVICES=y
+CONFIG_XILINX_EMACLITE=y
+CONFIG_XILINX_LL_TEMAC=y
+# CONFIG_INPUT is not set
+# CONFIG_SERIO is not set
+# CONFIG_VT is not set
+CONFIG_SERIAL_8250=y
+CONFIG_SERIAL_8250_CONSOLE=y
+CONFIG_SERIAL_UARTLITE=y
+CONFIG_SERIAL_UARTLITE_CONSOLE=y
+CONFIG_SERIAL_OF_PLATFORM=y
+# CONFIG_HW_RANDOM is not set
+CONFIG_XILINX_HWICAP=y
+CONFIG_I2C=y
+CONFIG_I2C_XILINX=y
+CONFIG_SPI=y
+CONFIG_SPI_XILINX=y
+CONFIG_GPIOLIB=y
+CONFIG_GPIO_SYSFS=y
+CONFIG_GPIO_XILINX=y
+# CONFIG_HWMON is not set
+CONFIG_WATCHDOG=y
+CONFIG_XILINX_WATCHDOG=y
+CONFIG_FB=y
+CONFIG_FB_XILINX=y
+# CONFIG_USB_SUPPORT is not set
+CONFIG_UIO=y
+CONFIG_UIO_PDRV=y
+CONFIG_UIO_PDRV_GENIRQ=y
+CONFIG_UIO_DMEM_GENIRQ=y
+CONFIG_EXT2_FS=y
+# CONFIG_DNOTIFY is not set
+CONFIG_CRAMFS=y
+CONFIG_ROMFS_FS=y
+CONFIG_NFS_FS=y
+CONFIG_NFS_V3_ACL=y
+CONFIG_NLS=y
+CONFIG_DETECT_HUNG_TASK=y
+CONFIG_DEBUG_SLAB=y
+CONFIG_DEBUG_SPINLOCK=y
+CONFIG_DEBUG_INFO=y
+CONFIG_EARLY_PRINTK=y
+CONFIG_KEYS=y
+CONFIG_ENCRYPTED_KEYS=y
+CONFIG_KEYS_DEBUG_PROC_KEYS=y
+CONFIG_CRYPTO_ECB=y
+CONFIG_CRYPTO_MD4=y
+CONFIG_CRYPTO_MD5=y
+CONFIG_CRYPTO_ARC4=y
+CONFIG_CRYPTO_DES=y
+# CONFIG_CRYPTO_ANSI_CPRNG is not set
diff --git a/arch/microblaze/include/asm/Kbuild b/arch/microblaze/include/asm/Kbuild
new file mode 100644
index 0000000..b0ae88c
--- /dev/null
+++ b/arch/microblaze/include/asm/Kbuild
@@ -0,0 +1,13 @@
+
+generic-y += barrier.h
+generic-y += clkdev.h
+generic-y += cputime.h
+generic-y += device.h
+generic-y += exec.h
+generic-y += irq_work.h
+generic-y += mcs_spinlock.h
+generic-y += mm-arch-hooks.h
+generic-y += preempt.h
+generic-y += syscalls.h
+generic-y += trace_clock.h
+generic-y += word-at-a-time.h
diff --git a/arch/microblaze/include/asm/asm-compat.h b/arch/microblaze/include/asm/asm-compat.h
new file mode 100644
index 0000000..e7bc9dc
--- /dev/null
+++ b/arch/microblaze/include/asm/asm-compat.h
@@ -0,0 +1,17 @@
+#ifndef _ASM_MICROBLAZE_ASM_COMPAT_H
+#define _ASM_MICROBLAZE_ASM_COMPAT_H
+
+#include <asm/types.h>
+
+#ifdef __ASSEMBLY__
+#  define stringify_in_c(...)	__VA_ARGS__
+#  define ASM_CONST(x)		x
+#else
+/* This version of stringify will deal with commas... */
+#  define __stringify_in_c(...)	#__VA_ARGS__
+#  define stringify_in_c(...)	__stringify_in_c(__VA_ARGS__) " "
+#  define __ASM_CONST(x)	x##UL
+#  define ASM_CONST(x)		__ASM_CONST(x)
+#endif
+
+#endif /* _ASM_MICROBLAZE_ASM_COMPAT_H */
diff --git a/arch/microblaze/include/asm/asm-offsets.h b/arch/microblaze/include/asm/asm-offsets.h
new file mode 100644
index 0000000..d370ee3
--- /dev/null
+++ b/arch/microblaze/include/asm/asm-offsets.h
@@ -0,0 +1 @@
+#include <generated/asm-offsets.h>
diff --git a/arch/microblaze/include/asm/atomic.h b/arch/microblaze/include/asm/atomic.h
new file mode 100644
index 0000000..42ac382
--- /dev/null
+++ b/arch/microblaze/include/asm/atomic.h
@@ -0,0 +1,27 @@
+#ifndef _ASM_MICROBLAZE_ATOMIC_H
+#define _ASM_MICROBLAZE_ATOMIC_H
+
+#include <asm/cmpxchg.h>
+#include <asm-generic/atomic.h>
+#include <asm-generic/atomic64.h>
+
+/*
+ * Atomically test *v and decrement if it is greater than 0.
+ * The function returns the old value of *v minus 1.
+ */
+static inline int atomic_dec_if_positive(atomic_t *v)
+{
+	unsigned long flags;
+	int res;
+
+	local_irq_save(flags);
+	res = v->counter - 1;
+	if (res >= 0)
+		v->counter = res;
+	local_irq_restore(flags);
+
+	return res;
+}
+#define atomic_dec_if_positive atomic_dec_if_positive
+
+#endif /* _ASM_MICROBLAZE_ATOMIC_H */
diff --git a/arch/microblaze/include/asm/bitops.h b/arch/microblaze/include/asm/bitops.h
new file mode 100644
index 0000000..a72468f
--- /dev/null
+++ b/arch/microblaze/include/asm/bitops.h
@@ -0,0 +1 @@
+#include <asm-generic/bitops.h>
diff --git a/arch/microblaze/include/asm/bug.h b/arch/microblaze/include/asm/bug.h
new file mode 100644
index 0000000..b12fd89
--- /dev/null
+++ b/arch/microblaze/include/asm/bug.h
@@ -0,0 +1 @@
+#include <asm-generic/bug.h>
diff --git a/arch/microblaze/include/asm/bugs.h b/arch/microblaze/include/asm/bugs.h
new file mode 100644
index 0000000..61791e1
--- /dev/null
+++ b/arch/microblaze/include/asm/bugs.h
@@ -0,0 +1 @@
+#include <asm-generic/bugs.h>
diff --git a/arch/microblaze/include/asm/cache.h b/arch/microblaze/include/asm/cache.h
new file mode 100644
index 0000000..4efe96a
--- /dev/null
+++ b/arch/microblaze/include/asm/cache.h
@@ -0,0 +1,24 @@
+/*
+ * Cache operations
+ *
+ * Copyright (C) 2007-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007-2009 PetaLogix
+ * Copyright (C) 2003 John Williams <jwilliams@itee.uq.edu.au>
+ *
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_CACHE_H
+#define _ASM_MICROBLAZE_CACHE_H
+
+#include <asm/registers.h>
+
+#define L1_CACHE_SHIFT 5
+/* word-granular cache in microblaze */
+#define L1_CACHE_BYTES	(1 << L1_CACHE_SHIFT)
+
+#define SMP_CACHE_BYTES	L1_CACHE_BYTES
+
+#endif /* _ASM_MICROBLAZE_CACHE_H */
diff --git a/arch/microblaze/include/asm/cacheflush.h b/arch/microblaze/include/asm/cacheflush.h
new file mode 100644
index 0000000..ffea82a
--- /dev/null
+++ b/arch/microblaze/include/asm/cacheflush.h
@@ -0,0 +1,124 @@
+/*
+ * Copyright (C) 2007-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007-2009 PetaLogix
+ * Copyright (C) 2007 John Williams <john.williams@petalogix.com>
+ * based on v850 version which was
+ * Copyright (C) 2001,02,03 NEC Electronics Corporation
+ * Copyright (C) 2001,02,03 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.
+ *
+ */
+
+#ifndef _ASM_MICROBLAZE_CACHEFLUSH_H
+#define _ASM_MICROBLAZE_CACHEFLUSH_H
+
+/* Somebody depends on this; sigh... */
+#include <linux/mm.h>
+#include <linux/io.h>
+
+/* Look at Documentation/cachetlb.txt */
+
+/*
+ * Cache handling functions.
+ * Microblaze has a write-through data cache, meaning that the data cache
+ * never needs to be flushed.  The only flushing operations that are
+ * implemented are to invalidate the instruction cache.  These are called
+ * after loading a user application into memory, we must invalidate the
+ * instruction cache to make sure we don't fetch old, bad code.
+ */
+
+/* struct cache, d=dcache, i=icache, fl = flush, iv = invalidate,
+ * suffix r = range */
+struct scache {
+	/* icache */
+	void (*ie)(void); /* enable */
+	void (*id)(void); /* disable */
+	void (*ifl)(void); /* flush */
+	void (*iflr)(unsigned long a, unsigned long b);
+	void (*iin)(void); /* invalidate */
+	void (*iinr)(unsigned long a, unsigned long b);
+	/* dcache */
+	void (*de)(void); /* enable */
+	void (*dd)(void); /* disable */
+	void (*dfl)(void); /* flush */
+	void (*dflr)(unsigned long a, unsigned long b);
+	void (*din)(void); /* invalidate */
+	void (*dinr)(unsigned long a, unsigned long b);
+};
+
+/* microblaze cache */
+extern struct scache *mbc;
+
+void microblaze_cache_init(void);
+
+#define enable_icache()					mbc->ie();
+#define disable_icache()				mbc->id();
+#define flush_icache()					mbc->ifl();
+#define flush_icache_range(start, end)			mbc->iflr(start, end);
+#define invalidate_icache()				mbc->iin();
+#define invalidate_icache_range(start, end)		mbc->iinr(start, end);
+
+#define flush_icache_user_range(vma, pg, adr, len)	flush_icache();
+#define flush_icache_page(vma, pg)			do { } while (0)
+
+#define enable_dcache()					mbc->de();
+#define disable_dcache()				mbc->dd();
+/* FIXME for LL-temac driver */
+#define invalidate_dcache()				mbc->din();
+#define invalidate_dcache_range(start, end)		mbc->dinr(start, end);
+#define flush_dcache()					mbc->dfl();
+#define flush_dcache_range(start, end)			mbc->dflr(start, end);
+
+#define ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE 1
+/* MS: We have to implement it because of rootfs-jffs2 issue on WB */
+#define flush_dcache_page(page) \
+do { \
+	unsigned long addr = (unsigned long) page_address(page); /* virtual */ \
+	addr = (u32)virt_to_phys((void *)addr); \
+	flush_dcache_range((unsigned) (addr), (unsigned) (addr) + PAGE_SIZE); \
+} while (0);
+
+#define flush_dcache_mmap_lock(mapping)		do { } while (0)
+#define flush_dcache_mmap_unlock(mapping)	do { } while (0)
+
+#define flush_cache_dup_mm(mm)				do { } while (0)
+#define flush_cache_vmap(start, end)			do { } while (0)
+#define flush_cache_vunmap(start, end)			do { } while (0)
+#define flush_cache_mm(mm)			do { } while (0)
+
+#define flush_cache_page(vma, vmaddr, pfn) \
+	flush_dcache_range(pfn << PAGE_SHIFT, (pfn << PAGE_SHIFT) + PAGE_SIZE);
+
+/* MS: kgdb code use this macro, wrong len with FLASH */
+#if 0
+#define flush_cache_range(vma, start, len)	{	\
+	flush_icache_range((unsigned) (start), (unsigned) (start) + (len)); \
+	flush_dcache_range((unsigned) (start), (unsigned) (start) + (len)); \
+}
+#endif
+
+#define flush_cache_range(vma, start, len) do { } while (0)
+
+static inline void copy_to_user_page(struct vm_area_struct *vma,
+				     struct page *page, unsigned long vaddr,
+				     void *dst, void *src, int len)
+{
+	u32 addr = virt_to_phys(dst);
+	memcpy(dst, src, len);
+	if (vma->vm_flags & VM_EXEC) {
+		invalidate_icache_range(addr, addr + PAGE_SIZE);
+		flush_dcache_range(addr, addr + PAGE_SIZE);
+	}
+}
+
+static inline void copy_from_user_page(struct vm_area_struct *vma,
+				       struct page *page, unsigned long vaddr,
+				       void *dst, void *src, int len)
+{
+	memcpy(dst, src, len);
+}
+
+#endif /* _ASM_MICROBLAZE_CACHEFLUSH_H */
diff --git a/arch/microblaze/include/asm/checksum.h b/arch/microblaze/include/asm/checksum.h
new file mode 100644
index 0000000..0185cbe
--- /dev/null
+++ b/arch/microblaze/include/asm/checksum.h
@@ -0,0 +1,39 @@
+/*
+ * Copyright (C) 2008 Michal Simek <monstr@monstr.eu>
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_CHECKSUM_H
+#define _ASM_MICROBLAZE_CHECKSUM_H
+
+/*
+ * computes the checksum of the TCP/UDP pseudo-header
+ * returns a 16-bit checksum, already complemented
+ */
+#define csum_tcpudp_nofold	csum_tcpudp_nofold
+static inline __wsum
+csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
+		   unsigned short proto, __wsum sum)
+{
+	__asm__("add %0, %0, %1\n\t"
+		"addc %0, %0, %2\n\t"
+		"addc %0, %0, %3\n\t"
+		"addc %0, %0, r0\n\t"
+		: "+&d" (sum)
+		: "d" (saddr), "d" (daddr),
+#ifdef __MICROBLAZEEL__
+	"d" ((len + proto) << 8)
+#else
+	"d" (len + proto)
+#endif
+);
+	return sum;
+}
+
+#include <asm-generic/checksum.h>
+
+#endif /* _ASM_MICROBLAZE_CHECKSUM_H */
diff --git a/arch/microblaze/include/asm/cmpxchg.h b/arch/microblaze/include/asm/cmpxchg.h
new file mode 100644
index 0000000..538afc0
--- /dev/null
+++ b/arch/microblaze/include/asm/cmpxchg.h
@@ -0,0 +1,42 @@
+#ifndef _ASM_MICROBLAZE_CMPXCHG_H
+#define _ASM_MICROBLAZE_CMPXCHG_H
+
+#include <linux/irqflags.h>
+
+void __bad_xchg(volatile void *ptr, int size);
+
+static inline unsigned long __xchg(unsigned long x, volatile void *ptr,
+								int size)
+{
+	unsigned long ret;
+	unsigned long flags;
+
+	switch (size) {
+	case 1:
+		local_irq_save(flags);
+		ret = *(volatile unsigned char *)ptr;
+		*(volatile unsigned char *)ptr = x;
+		local_irq_restore(flags);
+		break;
+
+	case 4:
+		local_irq_save(flags);
+		ret = *(volatile unsigned long *)ptr;
+		*(volatile unsigned long *)ptr = x;
+		local_irq_restore(flags);
+		break;
+	default:
+		__bad_xchg(ptr, size), ret = 0;
+		break;
+	}
+
+	return ret;
+}
+
+#define xchg(ptr, x) \
+	((__typeof__(*(ptr))) __xchg((unsigned long)(x), (ptr), sizeof(*(ptr))))
+
+#include <asm-generic/cmpxchg.h>
+#include <asm-generic/cmpxchg-local.h>
+
+#endif /* _ASM_MICROBLAZE_CMPXCHG_H */
diff --git a/arch/microblaze/include/asm/cpuinfo.h b/arch/microblaze/include/asm/cpuinfo.h
new file mode 100644
index 0000000..3337417
--- /dev/null
+++ b/arch/microblaze/include/asm/cpuinfo.h
@@ -0,0 +1,108 @@
+/*
+ * Generic support for queying CPU info
+ *
+ * Copyright (C) 2007-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007-2009 PetaLogix
+ * Copyright (C) 2007 John Williams <jwilliams@itee.uq.edu.au>
+ *
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_CPUINFO_H
+#define _ASM_MICROBLAZE_CPUINFO_H
+
+#include <asm/prom.h>
+
+/* CPU Version and FPGA Family code conversion table type */
+struct cpu_ver_key {
+	const char *s;
+	const unsigned k;
+};
+
+extern const struct cpu_ver_key cpu_ver_lookup[];
+
+struct family_string_key {
+	const char *s;
+	const unsigned k;
+};
+
+extern const struct family_string_key family_string_lookup[];
+
+struct cpuinfo {
+	/* Core CPU configuration */
+	u32 use_instr;
+	u32 use_mult;
+	u32 use_fpu;
+	u32 use_exc;
+	u32 ver_code;
+	u32 mmu;
+	u32 mmu_privins;
+	u32 endian;
+
+	/* CPU caches */
+	u32 use_icache;
+	u32 icache_tagbits;
+	u32 icache_write;
+	u32 icache_line_length;
+	u32 icache_size;
+	unsigned long icache_base;
+	unsigned long icache_high;
+
+	u32 use_dcache;
+	u32 dcache_tagbits;
+	u32 dcache_write;
+	u32 dcache_line_length;
+	u32 dcache_size;
+	u32 dcache_wb;
+	unsigned long dcache_base;
+	unsigned long dcache_high;
+
+	/* Bus connections */
+	u32 use_dopb;
+	u32 use_iopb;
+	u32 use_dlmb;
+	u32 use_ilmb;
+	u32 num_fsl;
+
+	/* CPU interrupt line info */
+	u32 irq_edge;
+	u32 irq_positive;
+
+	u32 area_optimised;
+
+	/* HW debug support */
+	u32 hw_debug;
+	u32 num_pc_brk;
+	u32 num_rd_brk;
+	u32 num_wr_brk;
+	u32 cpu_clock_freq; /* store real freq of cpu */
+
+	/* FPGA family */
+	u32 fpga_family_code;
+
+	/* User define */
+	u32 pvr_user1;
+	u32 pvr_user2;
+};
+
+extern struct cpuinfo cpuinfo;
+
+/* fwd declarations of the various CPUinfo populators */
+void setup_cpuinfo(void);
+void setup_cpuinfo_clk(void);
+
+void set_cpuinfo_static(struct cpuinfo *ci, struct device_node *cpu);
+void set_cpuinfo_pvr_full(struct cpuinfo *ci, struct device_node *cpu);
+
+static inline unsigned int fcpu(struct device_node *cpu, char *n)
+{
+	u32 val = 0;
+
+	of_property_read_u32(cpu, n, &val);
+
+	return val;
+}
+
+#endif /* _ASM_MICROBLAZE_CPUINFO_H */
diff --git a/arch/microblaze/include/asm/cputable.h b/arch/microblaze/include/asm/cputable.h
new file mode 100644
index 0000000..8b13789
--- /dev/null
+++ b/arch/microblaze/include/asm/cputable.h
@@ -0,0 +1 @@
+
diff --git a/arch/microblaze/include/asm/current.h b/arch/microblaze/include/asm/current.h
new file mode 100644
index 0000000..29303ed
--- /dev/null
+++ b/arch/microblaze/include/asm/current.h
@@ -0,0 +1,29 @@
+/*
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_CURRENT_H
+#define _ASM_MICROBLAZE_CURRENT_H
+
+/*
+ * Register used to hold the current task pointer while in the kernel.
+ * Any `call clobbered' register without a special meaning should be OK,
+ * but check asm/microblaze/kernel/entry.S to be sure.
+ */
+#define CURRENT_TASK	r31
+# ifndef __ASSEMBLY__
+/*
+ * Dedicate r31 to keeping the current task pointer
+ */
+register struct task_struct *current asm("r31");
+
+# define get_current()	current
+# endif /* __ASSEMBLY__ */
+
+#endif /* _ASM_MICROBLAZE_CURRENT_H */
diff --git a/arch/microblaze/include/asm/delay.h b/arch/microblaze/include/asm/delay.h
new file mode 100644
index 0000000..ea2a9cd
--- /dev/null
+++ b/arch/microblaze/include/asm/delay.h
@@ -0,0 +1,90 @@
+/*
+ * include/asm-microblaze/delay.h
+ *
+ * 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.
+ *
+ * Copyright (C) 2008 Michal Simek
+ * Copyright (C) 2007 John Williams
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+
+#ifndef _ASM_MICROBLAZE_DELAY_H
+#define _ASM_MICROBLAZE_DELAY_H
+
+#include <linux/param.h>
+
+static inline void __delay(unsigned long loops)
+{
+	asm volatile ("# __delay		\n\t"		\
+			"1: addi	%0, %0, -1\t\n"		\
+			"bneid	%0, 1b		\t\n"		\
+			"nop			\t\n"
+			: "=r" (loops)
+			: "0" (loops));
+}
+
+/*
+ * Note that 19 * 226 == 4294 ==~ 2^32 / 10^6, so
+ * loops = (4294 * usecs * loops_per_jiffy * HZ) / 2^32.
+ *
+ * The mul instruction gives us loops = (a * b) / 2^32.
+ * We choose a = usecs * 19 * HZ and b = loops_per_jiffy * 226
+ * because this lets us support a wide range of HZ and
+ * loops_per_jiffy values without either a or b overflowing 2^32.
+ * Thus we need usecs * HZ <= (2^32 - 1) / 19 = 226050910 and
+ * loops_per_jiffy <= (2^32 - 1) / 226 = 19004280
+ * (which corresponds to ~3800 bogomips at HZ = 100).
+ * -- paulus
+ */
+#define __MAX_UDELAY	(226050910UL/HZ)	/* maximum udelay argument */
+#define __MAX_NDELAY	(4294967295UL/HZ)	/* maximum ndelay argument */
+
+extern unsigned long loops_per_jiffy;
+
+static inline void __udelay(unsigned int x)
+{
+
+	unsigned long long tmp =
+		(unsigned long long)x * (unsigned long long)loops_per_jiffy \
+			* 226LL;
+	unsigned loops = tmp >> 32;
+
+/*
+	__asm__("mulxuu %0,%1,%2" : "=r" (loops) :
+		"r" (x), "r" (loops_per_jiffy * 226));
+*/
+	__delay(loops);
+}
+
+extern void __bad_udelay(void);		/* deliberately undefined */
+extern void __bad_ndelay(void);		/* deliberately undefined */
+
+#define udelay(n)						\
+	({							\
+		if (__builtin_constant_p(n)) {			\
+			if ((n) / __MAX_UDELAY >= 1)		\
+				__bad_udelay();			\
+			else					\
+				__udelay((n) * (19 * HZ));	\
+		} else {					\
+			__udelay((n) * (19 * HZ));		\
+		}						\
+	})
+
+#define ndelay(n)						\
+	({							\
+		if (__builtin_constant_p(n)) {			\
+			if ((n) / __MAX_NDELAY >= 1)		\
+				__bad_ndelay();			\
+			else					\
+				__udelay((n) * HZ);		\
+		} else {					\
+			__udelay((n) * HZ);			\
+		}						\
+	})
+
+#define muldiv(a, b, c)		(((a)*(b))/(c))
+
+#endif /* _ASM_MICROBLAZE_DELAY_H */
diff --git a/arch/microblaze/include/asm/div64.h b/arch/microblaze/include/asm/div64.h
new file mode 100644
index 0000000..6cd978c
--- /dev/null
+++ b/arch/microblaze/include/asm/div64.h
@@ -0,0 +1 @@
+#include <asm-generic/div64.h>
diff --git a/arch/microblaze/include/asm/dma-mapping.h b/arch/microblaze/include/asm/dma-mapping.h
new file mode 100644
index 0000000..24b1297
--- /dev/null
+++ b/arch/microblaze/include/asm/dma-mapping.h
@@ -0,0 +1,72 @@
+/*
+ * Implements the generic device dma API for microblaze and the pci
+ *
+ * Copyright (C) 2009-2010 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2009-2010 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 is base on powerpc and x86 dma-mapping.h versions
+ * Copyright (C) 2004 IBM
+ */
+
+#ifndef _ASM_MICROBLAZE_DMA_MAPPING_H
+#define _ASM_MICROBLAZE_DMA_MAPPING_H
+
+/*
+ * See Documentation/DMA-API-HOWTO.txt and
+ * Documentation/DMA-API.txt for documentation.
+ */
+
+#include <linux/types.h>
+#include <linux/cache.h>
+#include <linux/mm.h>
+#include <linux/scatterlist.h>
+#include <linux/dma-debug.h>
+#include <linux/dma-attrs.h>
+#include <asm/io.h>
+#include <asm/cacheflush.h>
+
+#define DMA_ERROR_CODE		(~(dma_addr_t)0x0)
+
+#define __dma_alloc_coherent(dev, gfp, size, handle)	NULL
+#define __dma_free_coherent(size, addr)		((void)0)
+
+/*
+ * Available generic sets of operations
+ */
+extern struct dma_map_ops dma_direct_ops;
+
+static inline struct dma_map_ops *get_dma_ops(struct device *dev)
+{
+	return &dma_direct_ops;
+}
+
+#include <asm-generic/dma-mapping-common.h>
+
+static inline void __dma_sync(unsigned long paddr,
+			      size_t size, enum dma_data_direction direction)
+{
+	switch (direction) {
+	case DMA_TO_DEVICE:
+	case DMA_BIDIRECTIONAL:
+		flush_dcache_range(paddr, paddr + size);
+		break;
+	case DMA_FROM_DEVICE:
+		invalidate_dcache_range(paddr, paddr + size);
+		break;
+	default:
+		BUG();
+	}
+}
+
+static inline void dma_cache_sync(struct device *dev, void *vaddr, size_t size,
+		enum dma_data_direction direction)
+{
+	BUG_ON(direction == DMA_NONE);
+	__dma_sync(virt_to_phys(vaddr), size, (int)direction);
+}
+
+#endif	/* _ASM_MICROBLAZE_DMA_MAPPING_H */
diff --git a/arch/microblaze/include/asm/dma.h b/arch/microblaze/include/asm/dma.h
new file mode 100644
index 0000000..0d73d0c
--- /dev/null
+++ b/arch/microblaze/include/asm/dma.h
@@ -0,0 +1,27 @@
+/*
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_DMA_H
+#define _ASM_MICROBLAZE_DMA_H
+
+#ifndef CONFIG_MMU
+/* we don't have dma address limit. define it as zero to be
+ * unlimited. */
+#define MAX_DMA_ADDRESS		(0)
+#else
+/* Virtual address corresponding to last available physical memory address.  */
+#define MAX_DMA_ADDRESS (CONFIG_KERNEL_START + memory_size - 1)
+#endif
+
+#ifdef CONFIG_PCI
+extern int isa_dma_bridge_buggy;
+#else
+#define isa_dma_bridge_buggy     (0)
+#endif
+
+#endif /* _ASM_MICROBLAZE_DMA_H */
diff --git a/arch/microblaze/include/asm/elf.h b/arch/microblaze/include/asm/elf.h
new file mode 100644
index 0000000..6590244
--- /dev/null
+++ b/arch/microblaze/include/asm/elf.h
@@ -0,0 +1,30 @@
+/*
+ * 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.
+ */
+#ifndef _ASM_MICROBLAZE_ELF_H
+#define _ASM_MICROBLAZE_ELF_H
+
+#include <uapi/asm/elf.h>
+
+#ifndef __uClinux__
+#ifndef ELF_GREG_T
+#endif
+#ifndef ELF_NGREG
+#endif
+#ifndef ELF_GREGSET_T
+#endif
+#ifndef ELF_FPREGSET_T
+#endif
+#ifdef __MICROBLAZEEL__
+#else
+#endif
+#define SET_PERSONALITY(ex) \
+	set_personality(PER_LINUX_32BIT | (current->personality & (~PER_MASK)))
+#endif /* __uClinux__ */
+#endif /* _ASM_MICROBLAZE_ELF_H */
diff --git a/arch/microblaze/include/asm/emergency-restart.h b/arch/microblaze/include/asm/emergency-restart.h
new file mode 100644
index 0000000..3711bd9
--- /dev/null
+++ b/arch/microblaze/include/asm/emergency-restart.h
@@ -0,0 +1 @@
+#include <asm-generic/emergency-restart.h>
diff --git a/arch/microblaze/include/asm/entry.h b/arch/microblaze/include/asm/entry.h
new file mode 100644
index 0000000..596e485
--- /dev/null
+++ b/arch/microblaze/include/asm/entry.h
@@ -0,0 +1,37 @@
+/*
+ * Definitions used by low-level trap handlers
+ *
+ * Copyright (C) 2008-2009 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_ENTRY_H
+#define _ASM_MICROBLAZE_ENTRY_H
+
+#include <asm/percpu.h>
+#include <asm/ptrace.h>
+#include <linux/linkage.h>
+
+/*
+ * These are per-cpu variables required in entry.S, among other
+ * places
+ */
+
+#define PER_CPU(var) var
+
+# ifndef __ASSEMBLY__
+DECLARE_PER_CPU(unsigned int, KSP); /* Saved kernel stack pointer */
+DECLARE_PER_CPU(unsigned int, KM); /* Kernel/user mode */
+DECLARE_PER_CPU(unsigned int, ENTRY_SP); /* Saved SP on kernel entry */
+DECLARE_PER_CPU(unsigned int, R11_SAVE); /* Temp variable for entry */
+DECLARE_PER_CPU(unsigned int, CURRENT_SAVE); /* Saved current pointer */
+
+extern asmlinkage void do_notify_resume(struct pt_regs *regs, int in_syscall);
+# endif /* __ASSEMBLY__ */
+
+#endif /* _ASM_MICROBLAZE_ENTRY_H */
diff --git a/arch/microblaze/include/asm/exceptions.h b/arch/microblaze/include/asm/exceptions.h
new file mode 100644
index 0000000..e6a8dde
--- /dev/null
+++ b/arch/microblaze/include/asm/exceptions.h
@@ -0,0 +1,77 @@
+/*
+ * Preliminary support for HW exception handing for Microblaze
+ *
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ * Copyright (C) 2005 John Williams <jwilliams@itee.uq.edu.au>
+ *
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_EXCEPTIONS_H
+#define _ASM_MICROBLAZE_EXCEPTIONS_H
+
+#ifdef __KERNEL__
+
+#ifndef CONFIG_MMU
+#define EX_HANDLER_STACK_SIZ	(4*19)
+#endif
+
+#ifndef __ASSEMBLY__
+
+/* Macros to enable and disable HW exceptions in the MSR */
+/* Define MSR enable bit for HW exceptions */
+#define HWEX_MSR_BIT (1 << 8)
+
+#if CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR
+#define __enable_hw_exceptions()					\
+	__asm__ __volatile__ ("	msrset	r0, %0;				\
+				nop;"					\
+				:					\
+				: "i" (HWEX_MSR_BIT)			\
+				: "memory")
+
+#define __disable_hw_exceptions()					\
+	__asm__ __volatile__ ("	msrclr r0, %0;				\
+				nop;"					\
+				:					\
+				: "i" (HWEX_MSR_BIT)			\
+				: "memory")
+#else /* !CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR */
+#define __enable_hw_exceptions()					\
+	__asm__ __volatile__ ("						\
+				mfs	r12, rmsr;			\
+				nop;					\
+				ori	r12, r12, %0;			\
+				mts	rmsr, r12;			\
+				nop;"					\
+				:					\
+				: "i" (HWEX_MSR_BIT)			\
+				: "memory", "r12")
+
+#define __disable_hw_exceptions()					\
+	__asm__ __volatile__ ("						\
+				mfs	r12, rmsr;			\
+				nop;					\
+				andi	r12, r12, ~%0;			\
+				mts	rmsr, r12;			\
+				nop;"					\
+				:					\
+				: "i" (HWEX_MSR_BIT)			\
+				: "memory", "r12")
+#endif /* CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR */
+
+asmlinkage void full_exception(struct pt_regs *regs, unsigned int type,
+							int fsr, int addr);
+
+asmlinkage void sw_exception(struct pt_regs *regs);
+void bad_page_fault(struct pt_regs *regs, unsigned long address, int sig);
+
+void die(const char *str, struct pt_regs *fp, long err);
+void _exception(int signr, struct pt_regs *regs, int code, unsigned long addr);
+
+#endif /*__ASSEMBLY__ */
+#endif /* __KERNEL__ */
+#endif /* _ASM_MICROBLAZE_EXCEPTIONS_H */
diff --git a/arch/microblaze/include/asm/fb.h b/arch/microblaze/include/asm/fb.h
new file mode 100644
index 0000000..3a4988e
--- /dev/null
+++ b/arch/microblaze/include/asm/fb.h
@@ -0,0 +1 @@
+#include <asm-generic/fb.h>
diff --git a/arch/microblaze/include/asm/fixmap.h b/arch/microblaze/include/asm/fixmap.h
new file mode 100644
index 0000000..06c0e2b
--- /dev/null
+++ b/arch/microblaze/include/asm/fixmap.h
@@ -0,0 +1,69 @@
+/*
+ * fixmap.h: compile-time virtual memory allocation
+ *
+ * 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.
+ *
+ * Copyright (C) 1998 Ingo Molnar
+ *
+ * Copyright 2008 Freescale Semiconductor Inc.
+ *   Port to powerpc added by Kumar Gala
+ *
+ * Copyright 2011 Michal Simek <monstr@monstr.eu>
+ * Copyright 2011 PetaLogix Qld Pty Ltd
+ *   Port to Microblaze
+ */
+
+#ifndef _ASM_FIXMAP_H
+#define _ASM_FIXMAP_H
+
+#ifndef __ASSEMBLY__
+#include <linux/kernel.h>
+#include <asm/page.h>
+#ifdef CONFIG_HIGHMEM
+#include <linux/threads.h>
+#include <asm/kmap_types.h>
+#endif
+
+#define FIXADDR_TOP	((unsigned long)(-PAGE_SIZE))
+
+/*
+ * Here we define all the compile-time 'special' virtual
+ * addresses. The point is to have a constant address at
+ * compile time, but to set the physical address only
+ * in the boot process. We allocate these special addresses
+ * from the end of virtual memory (0xfffff000) backwards.
+ * Also this lets us do fail-safe vmalloc(), we
+ * can guarantee that these special addresses and
+ * vmalloc()-ed addresses never overlap.
+ *
+ * these 'compile-time allocated' memory buffers are
+ * fixed-size 4k pages. (or larger if used with an increment
+ * highger than 1) use fixmap_set(idx,phys) to associate
+ * physical memory with fixmap indices.
+ *
+ * TLB entries of such buffers will not be flushed across
+ * task switches.
+ */
+enum fixed_addresses {
+	FIX_HOLE,
+#ifdef CONFIG_HIGHMEM
+	FIX_KMAP_BEGIN,	/* reserved pte's for temporary kernel mappings */
+	FIX_KMAP_END = FIX_KMAP_BEGIN + (KM_TYPE_NR * num_possible_cpus()) - 1,
+#endif
+	__end_of_fixed_addresses
+};
+
+extern void __set_fixmap(enum fixed_addresses idx,
+					phys_addr_t phys, pgprot_t flags);
+
+#define __FIXADDR_SIZE	(__end_of_fixed_addresses << PAGE_SHIFT)
+#define FIXADDR_START		(FIXADDR_TOP - __FIXADDR_SIZE)
+
+#define FIXMAP_PAGE_NOCACHE PAGE_KERNEL_CI
+
+#include <asm-generic/fixmap.h>
+
+#endif /* !__ASSEMBLY__ */
+#endif
diff --git a/arch/microblaze/include/asm/flat.h b/arch/microblaze/include/asm/flat.h
new file mode 100644
index 0000000..6847c15
--- /dev/null
+++ b/arch/microblaze/include/asm/flat.h
@@ -0,0 +1,89 @@
+/*
+ * uClinux flat-format executables
+ *
+ * Copyright (C) 2005 John Williams <jwilliams@itee.uq.edu.au>
+ *
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_FLAT_H
+#define _ASM_MICROBLAZE_FLAT_H
+
+#include <asm/unaligned.h>
+
+#define	flat_argvp_envp_on_stack()	0
+#define	flat_old_ram_flag(flags)	(flags)
+#define	flat_reloc_valid(reloc, size)	((reloc) <= (size))
+#define	flat_set_persistent(relval, p)		0
+
+/*
+ * Microblaze works a little differently from other arches, because
+ * of the MICROBLAZE_64 reloc type. Here, a 32 bit address is split
+ * over two instructions, an 'imm' instruction which provides the top
+ * 16 bits, then the instruction "proper" which provides the low 16
+ * bits.
+ */
+
+/*
+ * Crack open a symbol reference and extract the address to be
+ * relocated. rp is a potentially unaligned pointer to the
+ * reference
+ */
+
+static inline unsigned long
+flat_get_addr_from_rp(unsigned long *rp, unsigned long relval,
+			unsigned long flags, unsigned long *persistent)
+{
+	unsigned long addr;
+	(void)flags;
+
+	/* Is it a split 64/32 reference? */
+	if (relval & 0x80000000) {
+		/* Grab the two halves of the reference */
+		unsigned long val_hi, val_lo;
+
+		val_hi = get_unaligned(rp);
+		val_lo = get_unaligned(rp+1);
+
+		/* Crack the address out */
+		addr = ((val_hi & 0xffff) << 16) + (val_lo & 0xffff);
+	} else {
+		/* Get the address straight out */
+		addr = get_unaligned(rp);
+	}
+
+	return addr;
+}
+
+/*
+ * Insert an address into the symbol reference at rp. rp is potentially
+ * unaligned.
+ */
+
+static inline void
+flat_put_addr_at_rp(unsigned long *rp, unsigned long addr, unsigned long relval)
+{
+	/* Is this a split 64/32 reloc? */
+	if (relval & 0x80000000) {
+		/* Get the two "halves" */
+		unsigned long val_hi = get_unaligned(rp);
+		unsigned long val_lo = get_unaligned(rp + 1);
+
+		/* insert the address */
+		val_hi = (val_hi & 0xffff0000) | addr >> 16;
+		val_lo = (val_lo & 0xffff0000) | (addr & 0xffff);
+
+		/* store the two halves back into memory */
+		put_unaligned(val_hi, rp);
+		put_unaligned(val_lo, rp+1);
+	} else {
+		/* Put it straight in, no messing around */
+		put_unaligned(addr, rp);
+	}
+}
+
+#define	flat_get_relocate_addr(rel)	(rel & 0x7fffffff)
+
+#endif /* _ASM_MICROBLAZE_FLAT_H */
diff --git a/arch/microblaze/include/asm/ftrace.h b/arch/microblaze/include/asm/ftrace.h
new file mode 100644
index 0000000..da0144f
--- /dev/null
+++ b/arch/microblaze/include/asm/ftrace.h
@@ -0,0 +1,26 @@
+#ifndef _ASM_MICROBLAZE_FTRACE
+#define _ASM_MICROBLAZE_FTRACE
+
+#ifdef CONFIG_FUNCTION_TRACER
+
+#define MCOUNT_ADDR		((unsigned long)(_mcount))
+#define MCOUNT_INSN_SIZE	8 /* sizeof mcount call */
+
+#ifndef __ASSEMBLY__
+extern void _mcount(void);
+extern void ftrace_call_graph(void);
+#endif
+
+#ifdef CONFIG_DYNAMIC_FTRACE
+/* reloction of mcount call site is the same as the address */
+static inline unsigned long ftrace_call_adjust(unsigned long addr)
+{
+	return addr;
+}
+
+struct dyn_arch_ftrace {
+};
+#endif /* CONFIG_DYNAMIC_FTRACE */
+
+#endif /* CONFIG_FUNCTION_TRACER */
+#endif /* _ASM_MICROBLAZE_FTRACE */
diff --git a/arch/microblaze/include/asm/futex.h b/arch/microblaze/include/asm/futex.h
new file mode 100644
index 0000000..01848f0
--- /dev/null
+++ b/arch/microblaze/include/asm/futex.h
@@ -0,0 +1,129 @@
+#ifndef _ASM_MICROBLAZE_FUTEX_H
+#define _ASM_MICROBLAZE_FUTEX_H
+
+#ifdef __KERNEL__
+
+#include <linux/futex.h>
+#include <linux/uaccess.h>
+#include <asm/errno.h>
+
+#define __futex_atomic_op(insn, ret, oldval, uaddr, oparg) \
+({									\
+	__asm__ __volatile__ (						\
+			"1:	lwx	%0, %2, r0; "			\
+				insn					\
+			"2:	swx	%1, %2, r0;			\
+				addic	%1, r0, 0;			\
+				bnei	%1, 1b;				\
+			3:						\
+			.section .fixup,\"ax\";				\
+			4:	brid	3b;				\
+				addik	%1, r0, %3;			\
+			.previous;					\
+			.section __ex_table,\"a\";			\
+			.word	1b,4b,2b,4b;				\
+			.previous;"					\
+	: "=&r" (oldval), "=&r" (ret)					\
+	: "r" (uaddr), "i" (-EFAULT), "r" (oparg)			\
+	);								\
+})
+
+static inline int
+futex_atomic_op_inuser(int encoded_op, u32 __user *uaddr)
+{
+	int op = (encoded_op >> 28) & 7;
+	int cmp = (encoded_op >> 24) & 15;
+	int oparg = (encoded_op << 8) >> 20;
+	int cmparg = (encoded_op << 20) >> 20;
+	int oldval = 0, ret;
+	if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28))
+		oparg = 1 << oparg;
+
+	if (!access_ok(VERIFY_WRITE, uaddr, sizeof(u32)))
+		return -EFAULT;
+
+	pagefault_disable();
+
+	switch (op) {
+	case FUTEX_OP_SET:
+		__futex_atomic_op("or %1,%4,%4;", ret, oldval, uaddr, oparg);
+		break;
+	case FUTEX_OP_ADD:
+		__futex_atomic_op("add %1,%0,%4;", ret, oldval, uaddr, oparg);
+		break;
+	case FUTEX_OP_OR:
+		__futex_atomic_op("or %1,%0,%4;", ret, oldval, uaddr, oparg);
+		break;
+	case FUTEX_OP_ANDN:
+		__futex_atomic_op("andn %1,%0,%4;", ret, oldval, uaddr, oparg);
+		break;
+	case FUTEX_OP_XOR:
+		__futex_atomic_op("xor %1,%0,%4;", ret, oldval, uaddr, oparg);
+		break;
+	default:
+		ret = -ENOSYS;
+	}
+
+	pagefault_enable();
+
+	if (!ret) {
+		switch (cmp) {
+		case FUTEX_OP_CMP_EQ:
+			ret = (oldval == cmparg);
+			break;
+		case FUTEX_OP_CMP_NE:
+			ret = (oldval != cmparg);
+			break;
+		case FUTEX_OP_CMP_LT:
+			ret = (oldval < cmparg);
+			break;
+		case FUTEX_OP_CMP_GE:
+			ret = (oldval >= cmparg);
+			break;
+		case FUTEX_OP_CMP_LE:
+			ret = (oldval <= cmparg);
+			break;
+		case FUTEX_OP_CMP_GT:
+			ret = (oldval > cmparg);
+			break;
+		default:
+			ret = -ENOSYS;
+		}
+	}
+	return ret;
+}
+
+static inline int
+futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr,
+			      u32 oldval, u32 newval)
+{
+	int ret = 0, cmp;
+	u32 prev;
+
+	if (!access_ok(VERIFY_WRITE, uaddr, sizeof(u32)))
+		return -EFAULT;
+
+	__asm__ __volatile__ ("1:	lwx	%1, %3, r0;		\
+					cmp	%2, %1, %4;		\
+					bnei	%2, 3f;			\
+				2:	swx	%5, %3, r0;		\
+					addic	%2, r0, 0;		\
+					bnei	%2, 1b;			\
+				3:					\
+				.section .fixup,\"ax\";			\
+				4:	brid	3b;			\
+					addik	%0, r0, %6;		\
+				.previous;				\
+				.section __ex_table,\"a\";		\
+				.word	1b,4b,2b,4b;			\
+				.previous;"				\
+		: "+r" (ret), "=&r" (prev), "=&r"(cmp)	\
+		: "r" (uaddr), "r" (oldval), "r" (newval), "i" (-EFAULT));
+
+	*uval = prev;
+	return ret;
+}
+
+#endif /* __KERNEL__ */
+
+#endif
diff --git a/arch/microblaze/include/asm/gpio.h b/arch/microblaze/include/asm/gpio.h
new file mode 100644
index 0000000..b3799d8
--- /dev/null
+++ b/arch/microblaze/include/asm/gpio.h
@@ -0,0 +1,4 @@
+#ifndef __LINUX_GPIO_H
+#warning Include linux/gpio.h instead of asm/gpio.h
+#include <linux/gpio.h>
+#endif
diff --git a/arch/microblaze/include/asm/hardirq.h b/arch/microblaze/include/asm/hardirq.h
new file mode 100644
index 0000000..fb3c05a
--- /dev/null
+++ b/arch/microblaze/include/asm/hardirq.h
@@ -0,0 +1 @@
+#include <asm-generic/hardirq.h>
diff --git a/arch/microblaze/include/asm/highmem.h b/arch/microblaze/include/asm/highmem.h
new file mode 100644
index 0000000..67925ef
--- /dev/null
+++ b/arch/microblaze/include/asm/highmem.h
@@ -0,0 +1,83 @@
+/*
+ * highmem.h: virtual kernel memory mappings for high memory
+ *
+ * Used in CONFIG_HIGHMEM systems for memory pages which
+ * are not addressable by direct kernel virtual addresses.
+ *
+ * Copyright (C) 1999 Gerhard Wichert, Siemens AG
+ *		      Gerhard.Wichert@pdb.siemens.de
+ *
+ *
+ * Redesigned the x86 32-bit VM architecture to deal with
+ * up to 16 Terabyte physical memory. With current x86 CPUs
+ * we now support up to 64 Gigabytes physical RAM.
+ *
+ * Copyright (C) 1999 Ingo Molnar <mingo@redhat.com>
+ */
+#ifndef _ASM_HIGHMEM_H
+#define _ASM_HIGHMEM_H
+
+#ifdef __KERNEL__
+
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/uaccess.h>
+#include <asm/fixmap.h>
+
+extern pte_t *kmap_pte;
+extern pgprot_t kmap_prot;
+extern pte_t *pkmap_page_table;
+
+/*
+ * Right now we initialize only a single pte table. It can be extended
+ * easily, subsequent pte tables have to be allocated in one physical
+ * chunk of RAM.
+ */
+/*
+ * We use one full pte table with 4K pages. And with 16K/64K/256K pages pte
+ * table covers enough memory (32MB/512MB/2GB resp.), so that both FIXMAP
+ * and PKMAP can be placed in a single pte table. We use 512 pages for PKMAP
+ * in case of 16K/64K/256K page sizes.
+ */
+
+#define PKMAP_ORDER	PTE_SHIFT
+#define LAST_PKMAP	(1 << PKMAP_ORDER)
+
+#define PKMAP_BASE	((FIXADDR_START - PAGE_SIZE * (LAST_PKMAP + 1)) \
+								& PMD_MASK)
+
+#define LAST_PKMAP_MASK	(LAST_PKMAP - 1)
+#define PKMAP_NR(virt)  ((virt - PKMAP_BASE) >> PAGE_SHIFT)
+#define PKMAP_ADDR(nr)  (PKMAP_BASE + ((nr) << PAGE_SHIFT))
+
+extern void *kmap_high(struct page *page);
+extern void kunmap_high(struct page *page);
+extern void *kmap_atomic_prot(struct page *page, pgprot_t prot);
+extern void __kunmap_atomic(void *kvaddr);
+
+static inline void *kmap(struct page *page)
+{
+	might_sleep();
+	if (!PageHighMem(page))
+		return page_address(page);
+	return kmap_high(page);
+}
+
+static inline void kunmap(struct page *page)
+{
+	BUG_ON(in_interrupt());
+	if (!PageHighMem(page))
+		return;
+	kunmap_high(page);
+}
+
+static inline void *kmap_atomic(struct page *page)
+{
+	return kmap_atomic_prot(page, kmap_prot);
+}
+
+#define flush_cache_kmaps()	{ flush_icache(); flush_dcache(); }
+
+#endif /* __KERNEL__ */
+
+#endif /* _ASM_HIGHMEM_H */
diff --git a/arch/microblaze/include/asm/hw_irq.h b/arch/microblaze/include/asm/hw_irq.h
new file mode 100644
index 0000000..8b13789
--- /dev/null
+++ b/arch/microblaze/include/asm/hw_irq.h
@@ -0,0 +1 @@
+
diff --git a/arch/microblaze/include/asm/io.h b/arch/microblaze/include/asm/io.h
new file mode 100644
index 0000000..39b6315
--- /dev/null
+++ b/arch/microblaze/include/asm/io.h
@@ -0,0 +1,72 @@
+/*
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_IO_H
+#define _ASM_MICROBLAZE_IO_H
+
+#include <asm/byteorder.h>
+#include <asm/page.h>
+#include <linux/types.h>
+#include <linux/mm.h>          /* Get struct page {...} */
+
+#ifndef CONFIG_PCI
+#define _IO_BASE	0
+#define _ISA_MEM_BASE	0
+#else
+#define _IO_BASE	isa_io_base
+#define _ISA_MEM_BASE	isa_mem_base
+struct pci_dev;
+extern void pci_iounmap(struct pci_dev *dev, void __iomem *);
+#define pci_iounmap pci_iounmap
+
+extern unsigned long isa_io_base;
+extern resource_size_t isa_mem_base;
+#endif
+
+#define PCI_IOBASE	((void __iomem *)_IO_BASE)
+#define IO_SPACE_LIMIT (0xFFFFFFFF)
+
+#ifdef CONFIG_MMU
+#define page_to_bus(page)	(page_to_phys(page))
+
+extern void iounmap(void __iomem *addr);
+
+extern void __iomem *ioremap(phys_addr_t address, unsigned long size);
+#define ioremap_nocache(addr, size)		ioremap((addr), (size))
+#define ioremap_fullcache(addr, size)		ioremap((addr), (size))
+#define ioremap_wc(addr, size)			ioremap((addr), (size))
+#define ioremap_wt(addr, size)			ioremap((addr), (size))
+
+#endif /* CONFIG_MMU */
+
+/* Big Endian */
+#define out_be32(a, v) __raw_writel((v), (void __iomem __force *)(a))
+#define out_be16(a, v) __raw_writew((v), (a))
+
+#define in_be32(a) __raw_readl((const void __iomem __force *)(a))
+#define in_be16(a) __raw_readw(a)
+
+#define writel_be(v, a)	out_be32((__force unsigned *)a, v)
+#define readl_be(a)	in_be32((__force unsigned *)a)
+
+/* Little endian */
+#define out_le32(a, v) __raw_writel(__cpu_to_le32(v), (a))
+#define out_le16(a, v) __raw_writew(__cpu_to_le16(v), (a))
+
+#define in_le32(a) __le32_to_cpu(__raw_readl(a))
+#define in_le16(a) __le16_to_cpu(__raw_readw(a))
+
+/* Byte ops */
+#define out_8(a, v) __raw_writeb((v), (a))
+#define in_8(a) __raw_readb(a)
+
+#include <asm-generic/io.h>
+
+#endif /* _ASM_MICROBLAZE_IO_H */
diff --git a/arch/microblaze/include/asm/irq.h b/arch/microblaze/include/asm/irq.h
new file mode 100644
index 0000000..bab3b13
--- /dev/null
+++ b/arch/microblaze/include/asm/irq.h
@@ -0,0 +1,21 @@
+/*
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_IRQ_H
+#define _ASM_MICROBLAZE_IRQ_H
+
+#define NR_IRQS		(32 + 1)
+#include <asm-generic/irq.h>
+
+struct pt_regs;
+extern void do_IRQ(struct pt_regs *regs);
+
+/* should be defined in each interrupt controller driver */
+extern unsigned int get_irq(void);
+
+#endif /* _ASM_MICROBLAZE_IRQ_H */
diff --git a/arch/microblaze/include/asm/irq_regs.h b/arch/microblaze/include/asm/irq_regs.h
new file mode 100644
index 0000000..3dd9c0b
--- /dev/null
+++ b/arch/microblaze/include/asm/irq_regs.h
@@ -0,0 +1 @@
+#include <asm-generic/irq_regs.h>
diff --git a/arch/microblaze/include/asm/irqflags.h b/arch/microblaze/include/asm/irqflags.h
new file mode 100644
index 0000000..c9a6262
--- /dev/null
+++ b/arch/microblaze/include/asm/irqflags.h
@@ -0,0 +1,122 @@
+/*
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_IRQFLAGS_H
+#define _ASM_MICROBLAZE_IRQFLAGS_H
+
+#include <linux/types.h>
+#include <asm/registers.h>
+
+#if CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR
+
+static inline notrace unsigned long arch_local_irq_save(void)
+{
+	unsigned long flags;
+	asm volatile("	msrclr %0, %1	\n"
+		     "	nop		\n"
+		     : "=r"(flags)
+		     : "i"(MSR_IE)
+		     : "memory");
+	return flags;
+}
+
+static inline notrace void arch_local_irq_disable(void)
+{
+	/* this uses r0 without declaring it - is that correct? */
+	asm volatile("	msrclr r0, %0	\n"
+		     "	nop		\n"
+		     :
+		     : "i"(MSR_IE)
+		     : "memory");
+}
+
+static inline notrace void arch_local_irq_enable(void)
+{
+	/* this uses r0 without declaring it - is that correct? */
+	asm volatile("	msrset	r0, %0	\n"
+		     "	nop		\n"
+		     :
+		     : "i"(MSR_IE)
+		     : "memory");
+}
+
+#else /* !CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR */
+
+static inline notrace unsigned long arch_local_irq_save(void)
+{
+	unsigned long flags, tmp;
+	asm volatile ("	mfs	%0, rmsr	\n"
+		      "	nop			\n"
+		      "	andi	%1, %0, %2	\n"
+		      "	mts	rmsr, %1	\n"
+		      "	nop			\n"
+		      : "=r"(flags), "=r"(tmp)
+		      : "i"(~MSR_IE)
+		      : "memory");
+	return flags;
+}
+
+static inline notrace void arch_local_irq_disable(void)
+{
+	unsigned long tmp;
+	asm volatile("	mfs	%0, rmsr	\n"
+		     "	nop			\n"
+		     "	andi	%0, %0, %1	\n"
+		     "	mts	rmsr, %0	\n"
+		     "	nop			\n"
+		     : "=r"(tmp)
+		     : "i"(~MSR_IE)
+		     : "memory");
+}
+
+static inline notrace void arch_local_irq_enable(void)
+{
+	unsigned long tmp;
+	asm volatile("	mfs	%0, rmsr	\n"
+		     "	nop			\n"
+		     "	ori	%0, %0, %1	\n"
+		     "	mts	rmsr, %0	\n"
+		     "	nop			\n"
+		     : "=r"(tmp)
+		     : "i"(MSR_IE)
+		     : "memory");
+}
+
+#endif /* CONFIG_XILINX_MICROBLAZE0_USE_MSR_INSTR */
+
+static inline notrace unsigned long arch_local_save_flags(void)
+{
+	unsigned long flags;
+	asm volatile("	mfs	%0, rmsr	\n"
+		     "	nop			\n"
+		     : "=r"(flags)
+		     :
+		     : "memory");
+	return flags;
+}
+
+static inline notrace void arch_local_irq_restore(unsigned long flags)
+{
+	asm volatile("	mts	rmsr, %0	\n"
+		     "	nop			\n"
+		     :
+		     : "r"(flags)
+		     : "memory");
+}
+
+static inline notrace bool arch_irqs_disabled_flags(unsigned long flags)
+{
+	return (flags & MSR_IE) == 0;
+}
+
+static inline notrace bool arch_irqs_disabled(void)
+{
+	return arch_irqs_disabled_flags(arch_local_save_flags());
+}
+
+#endif /* _ASM_MICROBLAZE_IRQFLAGS_H */
diff --git a/arch/microblaze/include/asm/kdebug.h b/arch/microblaze/include/asm/kdebug.h
new file mode 100644
index 0000000..6ece1b0
--- /dev/null
+++ b/arch/microblaze/include/asm/kdebug.h
@@ -0,0 +1 @@
+#include <asm-generic/kdebug.h>
diff --git a/arch/microblaze/include/asm/kgdb.h b/arch/microblaze/include/asm/kgdb.h
new file mode 100644
index 0000000..ad27acb
--- /dev/null
+++ b/arch/microblaze/include/asm/kgdb.h
@@ -0,0 +1,31 @@
+#ifdef __KERNEL__
+#ifndef __MICROBLAZE_KGDB_H__
+#define __MICROBLAZE_KGDB_H__
+
+#ifndef __ASSEMBLY__
+
+#define CACHE_FLUSH_IS_SAFE	1
+#define BUFMAX			2048
+
+/*
+ * 32 32-bit general purpose registers (r0-r31)
+ *  6 32-bit special registers (pc, msr, ear, esr, fsr, btr)
+ * 12 32-bit PVR
+ *   7 32-bit MMU Regs (redr, rpid, rzpr, rtlbx, rtlbsx, rtlblo, rtlbhi)
+ * ------
+ *  57 registers
+ */
+#define NUMREGBYTES	(57 * 4)
+
+#define BREAK_INSTR_SIZE	4
+static inline void arch_kgdb_breakpoint(void)
+{
+	__asm__ __volatile__("brki r16, 0x18;");
+}
+
+struct pt_regs;
+asmlinkage void microblaze_kgdb_break(struct pt_regs *regs);
+
+#endif /* __ASSEMBLY__ */
+#endif /* __MICROBLAZE_KGDB_H__ */
+#endif /* __KERNEL__ */
diff --git a/arch/microblaze/include/asm/kmap_types.h b/arch/microblaze/include/asm/kmap_types.h
new file mode 100644
index 0000000..2597525
--- /dev/null
+++ b/arch/microblaze/include/asm/kmap_types.h
@@ -0,0 +1,6 @@
+#ifndef _ASM_MICROBLAZE_KMAP_TYPES_H
+#define _ASM_MICROBLAZE_KMAP_TYPES_H
+
+#include <asm-generic/kmap_types.h>
+
+#endif /* _ASM_MICROBLAZE_KMAP_TYPES_H */
diff --git a/arch/microblaze/include/asm/linkage.h b/arch/microblaze/include/asm/linkage.h
new file mode 100644
index 0000000..0540bba
--- /dev/null
+++ b/arch/microblaze/include/asm/linkage.h
@@ -0,0 +1 @@
+#include <asm-generic/linkage.h>
diff --git a/arch/microblaze/include/asm/local.h b/arch/microblaze/include/asm/local.h
new file mode 100644
index 0000000..c11c530
--- /dev/null
+++ b/arch/microblaze/include/asm/local.h
@@ -0,0 +1 @@
+#include <asm-generic/local.h>
diff --git a/arch/microblaze/include/asm/local64.h b/arch/microblaze/include/asm/local64.h
new file mode 100644
index 0000000..36c93b5
--- /dev/null
+++ b/arch/microblaze/include/asm/local64.h
@@ -0,0 +1 @@
+#include <asm-generic/local64.h>
diff --git a/arch/microblaze/include/asm/mmu.h b/arch/microblaze/include/asm/mmu.h
new file mode 100644
index 0000000..1f9eddd
--- /dev/null
+++ b/arch/microblaze/include/asm/mmu.h
@@ -0,0 +1,126 @@
+/*
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_MMU_H
+#define _ASM_MICROBLAZE_MMU_H
+
+# ifndef CONFIG_MMU
+#  include <asm-generic/mmu.h>
+# else /* CONFIG_MMU */
+#  ifdef __KERNEL__
+#   ifndef __ASSEMBLY__
+
+/* Default "unsigned long" context */
+typedef unsigned long mm_context_t;
+
+/* Hardware Page Table Entry */
+typedef struct _PTE {
+	unsigned long    v:1;	/* Entry is valid */
+	unsigned long vsid:24;	/* Virtual segment identifier */
+	unsigned long    h:1;	/* Hash algorithm indicator */
+	unsigned long  api:6;	/* Abbreviated page index */
+	unsigned long  rpn:20;	/* Real (physical) page number */
+	unsigned long     :3;	/* Unused */
+	unsigned long    r:1;	/* Referenced */
+	unsigned long    c:1;	/* Changed */
+	unsigned long    w:1;	/* Write-thru cache mode */
+	unsigned long    i:1;	/* Cache inhibited */
+	unsigned long    m:1;	/* Memory coherence */
+	unsigned long    g:1;	/* Guarded */
+	unsigned long     :1;	/* Unused */
+	unsigned long   pp:2;	/* Page protection */
+} PTE;
+
+/* Values for PP (assumes Ks=0, Kp=1) */
+#  define PP_RWXX	0 /* Supervisor read/write, User none */
+#  define PP_RWRX	1 /* Supervisor read/write, User read */
+#  define PP_RWRW	2 /* Supervisor read/write, User read/write */
+#  define PP_RXRX	3 /* Supervisor read,       User read */
+
+/* Segment Register */
+typedef struct _SEGREG {
+	unsigned long    t:1;	/* Normal or I/O  type */
+	unsigned long   ks:1;	/* Supervisor 'key' (normally 0) */
+	unsigned long   kp:1;	/* User 'key' (normally 1) */
+	unsigned long    n:1;	/* No-execute */
+	unsigned long     :4;	/* Unused */
+	unsigned long vsid:24;	/* Virtual Segment Identifier */
+} SEGREG;
+
+extern void _tlbie(unsigned long va);	/* invalidate a TLB entry */
+extern void _tlbia(void);		/* invalidate all TLB entries */
+
+/*
+ * tlb_skip size stores actual number skipped TLBs from TLB0 - every directy TLB
+ * mapping has to increase tlb_skip size.
+ */
+extern u32 tlb_skip;
+#   endif /* __ASSEMBLY__ */
+
+/*
+ * The MicroBlaze processor has a TLB architecture identical to PPC-40x. The
+ * instruction and data sides share a unified, 64-entry, semi-associative
+ * TLB which is maintained totally under software control. In addition, the
+ * instruction side has a hardware-managed, 2,4, or 8-entry, fully-associative
+ * TLB which serves as a first level to the shared TLB. These two TLBs are
+ * known as the UTLB and ITLB, respectively.
+ */
+
+#  define MICROBLAZE_TLB_SIZE 64
+
+/* For cases when you want to skip some TLB entries */
+#  define MICROBLAZE_TLB_SKIP 0
+
+/* Use the last TLB for temporary access to LMB */
+#  define MICROBLAZE_LMB_TLB_ID 63
+
+/*
+ * TLB entries are defined by a "high" tag portion and a "low" data
+ * portion. The data portion is 32-bits.
+ *
+ * TLB entries are managed entirely under software control by reading,
+ * writing, and searching using the MTS and MFS instructions.
+ */
+
+#  define TLB_LO		1
+#  define TLB_HI		0
+#  define TLB_DATA		TLB_LO
+#  define TLB_TAG		TLB_HI
+
+/* Tag portion */
+#  define TLB_EPN_MASK		0xFFFFFC00 /* Effective Page Number */
+#  define TLB_PAGESZ_MASK	0x00000380
+#  define TLB_PAGESZ(x)		(((x) & 0x7) << 7)
+#  define PAGESZ_1K		0
+#  define PAGESZ_4K		1
+#  define PAGESZ_16K		2
+#  define PAGESZ_64K		3
+#  define PAGESZ_256K		4
+#  define PAGESZ_1M		5
+#  define PAGESZ_4M		6
+#  define PAGESZ_16M		7
+#  define TLB_VALID		0x00000040 /* Entry is valid */
+
+/* Data portion */
+#  define TLB_RPN_MASK		0xFFFFFC00 /* Real Page Number */
+#  define TLB_PERM_MASK		0x00000300
+#  define TLB_EX		0x00000200 /* Instruction execution allowed */
+#  define TLB_WR		0x00000100 /* Writes permitted */
+#  define TLB_ZSEL_MASK		0x000000F0
+#  define TLB_ZSEL(x)		(((x) & 0xF) << 4)
+#  define TLB_ATTR_MASK		0x0000000F
+#  define TLB_W			0x00000008 /* Caching is write-through */
+#  define TLB_I			0x00000004 /* Caching is inhibited */
+#  define TLB_M			0x00000002 /* Memory is coherent */
+#  define TLB_G			0x00000001 /* Memory is guarded from prefetch */
+
+#  endif /* __KERNEL__ */
+# endif /* CONFIG_MMU */
+#endif /* _ASM_MICROBLAZE_MMU_H */
diff --git a/arch/microblaze/include/asm/mmu_context.h b/arch/microblaze/include/asm/mmu_context.h
new file mode 100644
index 0000000..0ccd8c4
--- /dev/null
+++ b/arch/microblaze/include/asm/mmu_context.h
@@ -0,0 +1,5 @@
+#ifdef CONFIG_MMU
+# include <asm/mmu_context_mm.h>
+#else
+# include <asm-generic/mmu_context.h>
+#endif
diff --git a/arch/microblaze/include/asm/mmu_context_mm.h b/arch/microblaze/include/asm/mmu_context_mm.h
new file mode 100644
index 0000000..d686477
--- /dev/null
+++ b/arch/microblaze/include/asm/mmu_context_mm.h
@@ -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.
+ */
+
+#ifndef _ASM_MICROBLAZE_MMU_CONTEXT_H
+#define _ASM_MICROBLAZE_MMU_CONTEXT_H
+
+#include <linux/atomic.h>
+#include <asm/bitops.h>
+#include <asm/mmu.h>
+#include <asm-generic/mm_hooks.h>
+
+# ifdef __KERNEL__
+/*
+ * This function defines the mapping from contexts to VSIDs (virtual
+ * segment IDs).  We use a skew on both the context and the high 4 bits
+ * of the 32-bit virtual address (the "effective segment ID") in order
+ * to spread out the entries in the MMU hash table.
+ */
+# define CTX_TO_VSID(ctx, va)	(((ctx) * (897 * 16) + ((va) >> 28) * 0x111) \
+				 & 0xffffff)
+
+/*
+   MicroBlaze has 256 contexts, so we can just rotate through these
+   as a way of "switching" contexts.  If the TID of the TLB is zero,
+   the PID/TID comparison is disabled, so we can use a TID of zero
+   to represent all kernel pages as shared among all contexts.
+ */
+
+static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
+{
+}
+
+# define NO_CONTEXT	256
+# define LAST_CONTEXT	255
+# define FIRST_CONTEXT	1
+
+/*
+ * Set the current MMU context.
+ * This is done byloading up the segment registers for the user part of the
+ * address space.
+ *
+ * Since the PGD is immediately available, it is much faster to simply
+ * pass this along as a second parameter, which is required for 8xx and
+ * can be used for debugging on all processors (if you happen to have
+ * an Abatron).
+ */
+extern void set_context(mm_context_t context, pgd_t *pgd);
+
+/*
+ * Bitmap of contexts in use.
+ * The size of this bitmap is LAST_CONTEXT + 1 bits.
+ */
+extern unsigned long context_map[];
+
+/*
+ * This caches the next context number that we expect to be free.
+ * Its use is an optimization only, we can't rely on this context
+ * number to be free, but it usually will be.
+ */
+extern mm_context_t next_mmu_context;
+
+/*
+ * Since we don't have sufficient contexts to give one to every task
+ * that could be in the system, we need to be able to steal contexts.
+ * These variables support that.
+ */
+extern atomic_t nr_free_contexts;
+extern struct mm_struct *context_mm[LAST_CONTEXT+1];
+extern void steal_context(void);
+
+/*
+ * Get a new mmu context for the address space described by `mm'.
+ */
+static inline void get_mmu_context(struct mm_struct *mm)
+{
+	mm_context_t ctx;
+
+	if (mm->context != NO_CONTEXT)
+		return;
+	while (atomic_dec_if_positive(&nr_free_contexts) < 0)
+		steal_context();
+	ctx = next_mmu_context;
+	while (test_and_set_bit(ctx, context_map)) {
+		ctx = find_next_zero_bit(context_map, LAST_CONTEXT+1, ctx);
+		if (ctx > LAST_CONTEXT)
+			ctx = 0;
+	}
+	next_mmu_context = (ctx + 1) & LAST_CONTEXT;
+	mm->context = ctx;
+	context_mm[ctx] = mm;
+}
+
+/*
+ * Set up the context for a new address space.
+ */
+# define init_new_context(tsk, mm)	(((mm)->context = NO_CONTEXT), 0)
+
+/*
+ * We're finished using the context for an address space.
+ */
+static inline void destroy_context(struct mm_struct *mm)
+{
+	if (mm->context != NO_CONTEXT) {
+		clear_bit(mm->context, context_map);
+		mm->context = NO_CONTEXT;
+		atomic_inc(&nr_free_contexts);
+	}
+}
+
+static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
+			     struct task_struct *tsk)
+{
+	tsk->thread.pgdir = next->pgd;
+	get_mmu_context(next);
+	set_context(next->context, next->pgd);
+}
+
+/*
+ * After we have set current->mm to a new value, this activates
+ * the context for the new mm so we see the new mappings.
+ */
+static inline void activate_mm(struct mm_struct *active_mm,
+			struct mm_struct *mm)
+{
+	current->thread.pgdir = mm->pgd;
+	get_mmu_context(mm);
+	set_context(mm->context, mm->pgd);
+}
+
+extern void mmu_context_init(void);
+
+# endif /* __KERNEL__ */
+#endif /* _ASM_MICROBLAZE_MMU_CONTEXT_H */
diff --git a/arch/microblaze/include/asm/module.h b/arch/microblaze/include/asm/module.h
new file mode 100644
index 0000000..7be1347
--- /dev/null
+++ b/arch/microblaze/include/asm/module.h
@@ -0,0 +1,31 @@
+/*
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_MODULE_H
+#define _ASM_MICROBLAZE_MODULE_H
+
+#include <asm-generic/module.h>
+
+/* Microblaze Relocations */
+#define R_MICROBLAZE_NONE 0
+#define R_MICROBLAZE_32 1
+#define R_MICROBLAZE_32_PCREL 2
+#define R_MICROBLAZE_64_PCREL 3
+#define R_MICROBLAZE_32_PCREL_LO 4
+#define R_MICROBLAZE_64 5
+#define R_MICROBLAZE_32_LO 6
+#define R_MICROBLAZE_SRO32 7
+#define R_MICROBLAZE_SRW32 8
+#define R_MICROBLAZE_64_NONE 9
+#define R_MICROBLAZE_32_SYM_OP_SYM 10
+/* Keep this the last entry. */
+#define R_MICROBLAZE_NUM 11
+
+typedef struct { volatile int counter; } module_t;
+
+#endif /* _ASM_MICROBLAZE_MODULE_H */
diff --git a/arch/microblaze/include/asm/mutex.h b/arch/microblaze/include/asm/mutex.h
new file mode 100644
index 0000000..ff6101a
--- /dev/null
+++ b/arch/microblaze/include/asm/mutex.h
@@ -0,0 +1 @@
+#include <asm-generic/mutex-dec.h>
diff --git a/arch/microblaze/include/asm/page.h b/arch/microblaze/include/asm/page.h
new file mode 100644
index 0000000..fd85087
--- /dev/null
+++ b/arch/microblaze/include/asm/page.h
@@ -0,0 +1,211 @@
+/*
+ * VM ops
+ *
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ * Changes for MMU support:
+ *    Copyright (C) 2007 Xilinx, Inc.  All rights reserved.
+ *
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_PAGE_H
+#define _ASM_MICROBLAZE_PAGE_H
+
+#include <linux/pfn.h>
+#include <asm/setup.h>
+#include <asm/asm-compat.h>
+#include <linux/const.h>
+
+#ifdef __KERNEL__
+
+/* PAGE_SHIFT determines the page size */
+#if defined(CONFIG_MICROBLAZE_64K_PAGES)
+#define PAGE_SHIFT		16
+#elif defined(CONFIG_MICROBLAZE_16K_PAGES)
+#define PAGE_SHIFT		14
+#else
+#define PAGE_SHIFT		12
+#endif
+#define PAGE_SIZE	(ASM_CONST(1) << PAGE_SHIFT)
+#define PAGE_MASK	(~(PAGE_SIZE-1))
+
+#define LOAD_OFFSET	ASM_CONST((CONFIG_KERNEL_START-CONFIG_KERNEL_BASE_ADDR))
+
+#define PTE_SHIFT	(PAGE_SHIFT - 2)	/* 1024 ptes per page */
+
+#ifndef __ASSEMBLY__
+
+/* MS be sure that SLAB allocates aligned objects */
+#define ARCH_DMA_MINALIGN	L1_CACHE_BYTES
+
+#define ARCH_SLAB_MINALIGN	L1_CACHE_BYTES
+
+#define PAGE_UP(addr)	(((addr)+((PAGE_SIZE)-1))&(~((PAGE_SIZE)-1)))
+#define PAGE_DOWN(addr)	((addr)&(~((PAGE_SIZE)-1)))
+
+#ifndef CONFIG_MMU
+/*
+ * PAGE_OFFSET -- the first address of the first page of memory. When not
+ * using MMU this corresponds to the first free page in physical memory (aligned
+ * on a page boundary).
+ */
+extern unsigned int __page_offset;
+#define PAGE_OFFSET __page_offset
+
+#else /* CONFIG_MMU */
+
+/*
+ * PAGE_OFFSET -- the first address of the first page of memory. With MMU
+ * it is set to the kernel start address (aligned on a page boundary).
+ *
+ * CONFIG_KERNEL_START is defined in arch/microblaze/config.in and used
+ * in arch/microblaze/Makefile.
+ */
+#define PAGE_OFFSET	CONFIG_KERNEL_START
+
+/*
+ * The basic type of a PTE - 32 bit physical addressing.
+ */
+typedef unsigned long pte_basic_t;
+#define PTE_FMT		"%.8lx"
+
+#endif /* CONFIG_MMU */
+
+# define copy_page(to, from)			memcpy((to), (from), PAGE_SIZE)
+# define clear_page(pgaddr)			memset((pgaddr), 0, PAGE_SIZE)
+
+# define clear_user_page(pgaddr, vaddr, page)	memset((pgaddr), 0, PAGE_SIZE)
+# define copy_user_page(vto, vfrom, vaddr, topg) \
+			memcpy((vto), (vfrom), PAGE_SIZE)
+
+/*
+ * These are used to make use of C type-checking..
+ */
+typedef struct page *pgtable_t;
+typedef struct { unsigned long	pte; }		pte_t;
+typedef struct { unsigned long	pgprot; }	pgprot_t;
+/* FIXME this can depend on linux kernel version */
+#   ifdef CONFIG_MMU
+typedef struct { unsigned long pmd; } pmd_t;
+typedef struct { unsigned long pgd; } pgd_t;
+#   else /* CONFIG_MMU */
+typedef struct { unsigned long	ste[64]; }	pmd_t;
+typedef struct { pmd_t		pue[1]; }	pud_t;
+typedef struct { pud_t		pge[1]; }	pgd_t;
+#   endif /* CONFIG_MMU */
+
+# define pte_val(x)	((x).pte)
+# define pgprot_val(x)	((x).pgprot)
+
+#   ifdef CONFIG_MMU
+#   define pmd_val(x)      ((x).pmd)
+#   define pgd_val(x)      ((x).pgd)
+#   else  /* CONFIG_MMU */
+#   define pmd_val(x)	((x).ste[0])
+#   define pud_val(x)	((x).pue[0])
+#   define pgd_val(x)	((x).pge[0])
+#   endif  /* CONFIG_MMU */
+
+# define __pte(x)	((pte_t) { (x) })
+# define __pmd(x)	((pmd_t) { (x) })
+# define __pgd(x)	((pgd_t) { (x) })
+# define __pgprot(x)	((pgprot_t) { (x) })
+
+/**
+ * Conversions for virtual address, physical address, pfn, and struct
+ * page are defined in the following files.
+ *
+ * virt -+
+ *	 | asm-microblaze/page.h
+ * phys -+
+ *	 | linux/pfn.h
+ *  pfn -+
+ *	 | asm-generic/memory_model.h
+ * page -+
+ *
+ */
+
+extern unsigned long max_low_pfn;
+extern unsigned long min_low_pfn;
+extern unsigned long max_pfn;
+
+extern unsigned long memory_start;
+extern unsigned long memory_size;
+extern unsigned long lowmem_size;
+
+extern unsigned long kernel_tlb;
+
+extern int page_is_ram(unsigned long pfn);
+
+# define phys_to_pfn(phys)	(PFN_DOWN(phys))
+# define pfn_to_phys(pfn)	(PFN_PHYS(pfn))
+
+# define virt_to_pfn(vaddr)	(phys_to_pfn((__pa(vaddr))))
+# define pfn_to_virt(pfn)	__va(pfn_to_phys((pfn)))
+
+#  ifdef CONFIG_MMU
+
+#  define virt_to_page(kaddr)	(pfn_to_page(__pa(kaddr) >> PAGE_SHIFT))
+#  define page_to_virt(page)   __va(page_to_pfn(page) << PAGE_SHIFT)
+#  define page_to_phys(page)     (page_to_pfn(page) << PAGE_SHIFT)
+
+#  else /* CONFIG_MMU */
+#  define virt_to_page(vaddr)	(pfn_to_page(virt_to_pfn(vaddr)))
+#  define page_to_virt(page)	(pfn_to_virt(page_to_pfn(page)))
+#  define page_to_phys(page)	(pfn_to_phys(page_to_pfn(page)))
+#  define page_to_bus(page)	(page_to_phys(page))
+#  define phys_to_page(paddr)	(pfn_to_page(phys_to_pfn(paddr)))
+#  endif /* CONFIG_MMU */
+
+#  ifndef CONFIG_MMU
+#  define pfn_valid(pfn)	(((pfn) >= min_low_pfn) && \
+				((pfn) <= (min_low_pfn + max_mapnr)))
+#  define ARCH_PFN_OFFSET	(PAGE_OFFSET >> PAGE_SHIFT)
+#  else /* CONFIG_MMU */
+#  define ARCH_PFN_OFFSET	(memory_start >> PAGE_SHIFT)
+#  define pfn_valid(pfn)	((pfn) < (max_mapnr + ARCH_PFN_OFFSET))
+#  endif /* CONFIG_MMU */
+
+# endif /* __ASSEMBLY__ */
+
+#define	virt_addr_valid(vaddr)	(pfn_valid(virt_to_pfn(vaddr)))
+
+# define __pa(x)	__virt_to_phys((unsigned long)(x))
+# define __va(x)	((void *)__phys_to_virt((unsigned long)(x)))
+
+/* Convert between virtual and physical address for MMU. */
+/* Handle MicroBlaze processor with virtual memory. */
+#ifndef CONFIG_MMU
+#define __virt_to_phys(addr)	addr
+#define __phys_to_virt(addr)	addr
+#define tophys(rd, rs)	addik rd, rs, 0
+#define tovirt(rd, rs)	addik rd, rs, 0
+#else
+#define __virt_to_phys(addr) \
+	((addr) + CONFIG_KERNEL_BASE_ADDR - CONFIG_KERNEL_START)
+#define __phys_to_virt(addr) \
+	((addr) + CONFIG_KERNEL_START - CONFIG_KERNEL_BASE_ADDR)
+#define tophys(rd, rs) \
+	addik rd, rs, (CONFIG_KERNEL_BASE_ADDR - CONFIG_KERNEL_START)
+#define tovirt(rd, rs) \
+	addik rd, rs, (CONFIG_KERNEL_START - CONFIG_KERNEL_BASE_ADDR)
+#endif /* CONFIG_MMU */
+
+#define TOPHYS(addr)  __virt_to_phys(addr)
+
+#ifdef CONFIG_MMU
+
+#define VM_DATA_DEFAULT_FLAGS	(VM_READ | VM_WRITE | VM_EXEC | \
+				 VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
+#endif /* CONFIG_MMU */
+
+#endif /* __KERNEL__ */
+
+#include <asm-generic/memory_model.h>
+#include <asm-generic/getorder.h>
+
+#endif /* _ASM_MICROBLAZE_PAGE_H */
diff --git a/arch/microblaze/include/asm/parport.h b/arch/microblaze/include/asm/parport.h
new file mode 100644
index 0000000..cf252af
--- /dev/null
+++ b/arch/microblaze/include/asm/parport.h
@@ -0,0 +1 @@
+#include <asm-generic/parport.h>
diff --git a/arch/microblaze/include/asm/pci-bridge.h b/arch/microblaze/include/asm/pci-bridge.h
new file mode 100644
index 0000000..cb5d397
--- /dev/null
+++ b/arch/microblaze/include/asm/pci-bridge.h
@@ -0,0 +1,144 @@
+#ifndef _ASM_MICROBLAZE_PCI_BRIDGE_H
+#define _ASM_MICROBLAZE_PCI_BRIDGE_H
+#ifdef __KERNEL__
+/*
+ * 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 <linux/pci.h>
+#include <linux/list.h>
+#include <linux/ioport.h>
+
+struct device_node;
+
+#ifdef CONFIG_PCI
+extern struct list_head hose_list;
+extern int pcibios_vaddr_is_ioport(void __iomem *address);
+#else
+static inline int pcibios_vaddr_is_ioport(void __iomem *address)
+{
+	return 0;
+}
+#endif
+
+/*
+ * Structure of a PCI controller (host bridge)
+ */
+struct pci_controller {
+	struct pci_bus *bus;
+	char is_dynamic;
+	struct device_node *dn;
+	struct list_head list_node;
+	struct device *parent;
+
+	int first_busno;
+	int last_busno;
+
+	int self_busno;
+
+	void __iomem *io_base_virt;
+	resource_size_t io_base_phys;
+
+	resource_size_t pci_io_size;
+
+	/* Some machines (PReP) have a non 1:1 mapping of
+	 * the PCI memory space in the CPU bus space
+	 */
+	resource_size_t pci_mem_offset;
+
+	/* Some machines have a special region to forward the ISA
+	 * "memory" cycles such as VGA memory regions. Left to 0
+	 * if unsupported
+	 */
+	resource_size_t isa_mem_phys;
+	resource_size_t isa_mem_size;
+
+	struct pci_ops *ops;
+	unsigned int __iomem *cfg_addr;
+	void __iomem *cfg_data;
+
+	/*
+	 * Used for variants of PCI indirect handling and possible quirks:
+	 *  SET_CFG_TYPE - used on 4xx or any PHB that does explicit type0/1
+	 *  EXT_REG - provides access to PCI-e extended registers
+	 *  SURPRESS_PRIMARY_BUS - we suppress the setting of PCI_PRIMARY_BUS
+	 *   on Freescale PCI-e controllers since they used the PCI_PRIMARY_BUS
+	 *   to determine which bus number to match on when generating type0
+	 *   config cycles
+	 *  NO_PCIE_LINK - the Freescale PCI-e controllers have issues with
+	 *   hanging if we don't have link and try to do config cycles to
+	 *   anything but the PHB.  Only allow talking to the PHB if this is
+	 *   set.
+	 *  BIG_ENDIAN - cfg_addr is a big endian register
+	 *  BROKEN_MRM - the 440EPx/GRx chips have an errata that causes hangs
+	 *   on the PLB4.  Effectively disable MRM commands by setting this.
+	 */
+#define INDIRECT_TYPE_SET_CFG_TYPE		0x00000001
+#define INDIRECT_TYPE_EXT_REG		0x00000002
+#define INDIRECT_TYPE_SURPRESS_PRIMARY_BUS	0x00000004
+#define INDIRECT_TYPE_NO_PCIE_LINK		0x00000008
+#define INDIRECT_TYPE_BIG_ENDIAN		0x00000010
+#define INDIRECT_TYPE_BROKEN_MRM		0x00000020
+	u32 indirect_type;
+
+	/* Currently, we limit ourselves to 1 IO range and 3 mem
+	 * ranges since the common pci_bus structure can't handle more
+	 */
+	struct resource io_resource;
+	struct resource mem_resources[3];
+	int global_number;	/* PCI domain number */
+};
+
+#ifdef CONFIG_PCI
+static inline struct pci_controller *pci_bus_to_host(const struct pci_bus *bus)
+{
+	return bus->sysdata;
+}
+
+static inline int isa_vaddr_is_ioport(void __iomem *address)
+{
+	/* No specific ISA handling on ppc32 at this stage, it
+	 * all goes through PCI
+	 */
+	return 0;
+}
+#endif /* CONFIG_PCI */
+
+/* These are used for config access before all the PCI probing
+   has been done. */
+extern int early_read_config_byte(struct pci_controller *hose, int bus,
+			int dev_fn, int where, u8 *val);
+extern int early_read_config_word(struct pci_controller *hose, int bus,
+			int dev_fn, int where, u16 *val);
+extern int early_read_config_dword(struct pci_controller *hose, int bus,
+			int dev_fn, int where, u32 *val);
+extern int early_write_config_byte(struct pci_controller *hose, int bus,
+			int dev_fn, int where, u8 val);
+extern int early_write_config_word(struct pci_controller *hose, int bus,
+			int dev_fn, int where, u16 val);
+extern int early_write_config_dword(struct pci_controller *hose, int bus,
+			int dev_fn, int where, u32 val);
+
+extern int early_find_capability(struct pci_controller *hose, int bus,
+				 int dev_fn, int cap);
+
+extern void setup_indirect_pci(struct pci_controller *hose,
+			       resource_size_t cfg_addr,
+			       resource_size_t cfg_data, u32 flags);
+
+/* Get the PCI host controller for an OF device */
+extern struct pci_controller *pci_find_hose_for_OF_device(
+			struct device_node *node);
+
+/* Fill up host controller resources from the OF node */
+extern void pci_process_bridge_OF_ranges(struct pci_controller *hose,
+			struct device_node *dev, int primary);
+
+/* Allocate & free a PCI host bridge structure */
+extern struct pci_controller *pcibios_alloc_controller(struct device_node *dev);
+extern void pcibios_free_controller(struct pci_controller *phb);
+
+#endif	/* __KERNEL__ */
+#endif	/* _ASM_MICROBLAZE_PCI_BRIDGE_H */
diff --git a/arch/microblaze/include/asm/pci.h b/arch/microblaze/include/asm/pci.h
new file mode 100644
index 0000000..dc9eb66
--- /dev/null
+++ b/arch/microblaze/include/asm/pci.h
@@ -0,0 +1,102 @@
+/*
+ * 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.
+ *
+ * Based on powerpc version
+ */
+
+#ifndef __ASM_MICROBLAZE_PCI_H
+#define __ASM_MICROBLAZE_PCI_H
+#ifdef __KERNEL__
+
+#include <linux/types.h>
+#include <linux/slab.h>
+#include <linux/string.h>
+#include <linux/dma-mapping.h>
+#include <linux/pci.h>
+#include <linux/scatterlist.h>
+
+#include <asm/io.h>
+#include <asm/prom.h>
+#include <asm/pci-bridge.h>
+
+#include <asm-generic/pci-dma-compat.h>
+
+#define PCIBIOS_MIN_IO		0x1000
+#define PCIBIOS_MIN_MEM		0x10000000
+
+/* Values for the `which' argument to sys_pciconfig_iobase syscall.  */
+#define IOBASE_BRIDGE_NUMBER	0
+#define IOBASE_MEMORY		1
+#define IOBASE_IO		2
+#define IOBASE_ISA_IO		3
+#define IOBASE_ISA_MEM		4
+
+#define pcibios_scan_all_fns(a, b)	0
+
+/*
+ * Set this to 1 if you want the kernel to re-assign all PCI
+ * bus numbers (don't do that on ppc64 yet !)
+ */
+#define pcibios_assign_all_busses()	0
+
+extern int pci_domain_nr(struct pci_bus *bus);
+
+/* Decide whether to display the domain number in /proc */
+extern int pci_proc_domain(struct pci_bus *bus);
+
+struct vm_area_struct;
+/* Map a range of PCI memory or I/O space for a device into user space */
+int pci_mmap_page_range(struct pci_dev *pdev, struct vm_area_struct *vma,
+			enum pci_mmap_state mmap_state, int write_combine);
+
+/* Tell drivers/pci/proc.c that we have pci_mmap_page_range() */
+#define HAVE_PCI_MMAP	1
+
+extern int pci_legacy_read(struct pci_bus *bus, loff_t port, u32 *val,
+			   size_t count);
+extern int pci_legacy_write(struct pci_bus *bus, loff_t port, u32 val,
+			   size_t count);
+extern int pci_mmap_legacy_page_range(struct pci_bus *bus,
+				      struct vm_area_struct *vma,
+				      enum pci_mmap_state mmap_state);
+
+#define HAVE_PCI_LEGACY	1
+
+/* The PCI address space does equal the physical memory
+ * address space (no IOMMU).  The IDE and SCSI device layers use
+ * this boolean for bounce buffer decisions.
+ */
+#define PCI_DMA_BUS_IS_PHYS     (1)
+
+extern void pcibios_claim_one_bus(struct pci_bus *b);
+
+extern void pcibios_finish_adding_to_bus(struct pci_bus *bus);
+
+extern void pcibios_resource_survey(void);
+
+struct file;
+extern pgprot_t	pci_phys_mem_access_prot(struct file *file,
+					 unsigned long pfn,
+					 unsigned long size,
+					 pgprot_t prot);
+
+#define HAVE_ARCH_PCI_RESOURCE_TO_USER
+extern void pci_resource_to_user(const struct pci_dev *dev, int bar,
+				 const struct resource *rsrc,
+				 resource_size_t *start, resource_size_t *end);
+
+extern void pcibios_setup_bus_devices(struct pci_bus *bus);
+extern void pcibios_setup_bus_self(struct pci_bus *bus);
+
+/* This part of code was originally in xilinx-pci.h */
+#ifdef CONFIG_PCI_XILINX
+extern void __init xilinx_pci_init(void);
+#else
+static inline void __init xilinx_pci_init(void) { return; }
+#endif
+
+#endif	/* __KERNEL__ */
+#endif /* __ASM_MICROBLAZE_PCI_H */
diff --git a/arch/microblaze/include/asm/percpu.h b/arch/microblaze/include/asm/percpu.h
new file mode 100644
index 0000000..06a959d
--- /dev/null
+++ b/arch/microblaze/include/asm/percpu.h
@@ -0,0 +1 @@
+#include <asm-generic/percpu.h>
diff --git a/arch/microblaze/include/asm/pgalloc.h b/arch/microblaze/include/asm/pgalloc.h
new file mode 100644
index 0000000..61436d6
--- /dev/null
+++ b/arch/microblaze/include/asm/pgalloc.h
@@ -0,0 +1,195 @@
+/*
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_PGALLOC_H
+#define _ASM_MICROBLAZE_PGALLOC_H
+
+#ifdef CONFIG_MMU
+
+#include <linux/kernel.h>	/* For min/max macros */
+#include <linux/highmem.h>
+#include <asm/setup.h>
+#include <asm/io.h>
+#include <asm/page.h>
+#include <asm/cache.h>
+#include <asm/pgtable.h>
+
+#define PGDIR_ORDER	0
+
+/*
+ * This is handled very differently on MicroBlaze since out page tables
+ * are all 0's and I want to be able to use these zero'd pages elsewhere
+ * as well - it gives us quite a speedup.
+ * -- Cort
+ */
+extern struct pgtable_cache_struct {
+	unsigned long *pgd_cache;
+	unsigned long *pte_cache;
+	unsigned long pgtable_cache_sz;
+} quicklists;
+
+#define pgd_quicklist		(quicklists.pgd_cache)
+#define pmd_quicklist		((unsigned long *)0)
+#define pte_quicklist		(quicklists.pte_cache)
+#define pgtable_cache_size	(quicklists.pgtable_cache_sz)
+
+extern unsigned long *zero_cache; /* head linked list of pre-zero'd pages */
+extern atomic_t zero_sz; /* # currently pre-zero'd pages */
+extern atomic_t zeropage_hits; /* # zero'd pages request that we've done */
+extern atomic_t zeropage_calls; /* # zero'd pages request that've been made */
+extern atomic_t zerototal; /* # pages zero'd over time */
+
+#define zero_quicklist		(zero_cache)
+#define zero_cache_sz	 	(zero_sz)
+#define zero_cache_calls	(zeropage_calls)
+#define zero_cache_hits		(zeropage_hits)
+#define zero_cache_total	(zerototal)
+
+/*
+ * return a pre-zero'd page from the list,
+ * return NULL if none available -- Cort
+ */
+extern unsigned long get_zero_page_fast(void);
+
+extern void __bad_pte(pmd_t *pmd);
+
+static inline pgd_t *get_pgd_slow(void)
+{
+	pgd_t *ret;
+
+	ret = (pgd_t *)__get_free_pages(GFP_KERNEL, PGDIR_ORDER);
+	if (ret != NULL)
+		clear_page(ret);
+	return ret;
+}
+
+static inline pgd_t *get_pgd_fast(void)
+{
+	unsigned long *ret;
+
+	ret = pgd_quicklist;
+	if (ret != NULL) {
+		pgd_quicklist = (unsigned long *)(*ret);
+		ret[0] = 0;
+		pgtable_cache_size--;
+	} else
+		ret = (unsigned long *)get_pgd_slow();
+	return (pgd_t *)ret;
+}
+
+static inline void free_pgd_fast(pgd_t *pgd)
+{
+	*(unsigned long **)pgd = pgd_quicklist;
+	pgd_quicklist = (unsigned long *) pgd;
+	pgtable_cache_size++;
+}
+
+static inline void free_pgd_slow(pgd_t *pgd)
+{
+	free_page((unsigned long)pgd);
+}
+
+#define pgd_free(mm, pgd)        free_pgd_fast(pgd)
+#define pgd_alloc(mm)		get_pgd_fast()
+
+#define pmd_pgtable(pmd)	pmd_page(pmd)
+
+/*
+ * We don't have any real pmd's, and this code never triggers because
+ * the pgd will always be present..
+ */
+#define pmd_alloc_one_fast(mm, address)	({ BUG(); ((pmd_t *)1); })
+#define pmd_alloc_one(mm, address)	({ BUG(); ((pmd_t *)2); })
+
+extern pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long addr);
+
+static inline struct page *pte_alloc_one(struct mm_struct *mm,
+		unsigned long address)
+{
+	struct page *ptepage;
+
+#ifdef CONFIG_HIGHPTE
+	int flags = GFP_KERNEL | __GFP_HIGHMEM | __GFP_REPEAT;
+#else
+	int flags = GFP_KERNEL | __GFP_REPEAT;
+#endif
+
+	ptepage = alloc_pages(flags, 0);
+	if (!ptepage)
+		return NULL;
+	clear_highpage(ptepage);
+	if (!pgtable_page_ctor(ptepage)) {
+		__free_page(ptepage);
+		return NULL;
+	}
+	return ptepage;
+}
+
+static inline pte_t *pte_alloc_one_fast(struct mm_struct *mm,
+		unsigned long address)
+{
+	unsigned long *ret;
+
+	ret = pte_quicklist;
+	if (ret != NULL) {
+		pte_quicklist = (unsigned long *)(*ret);
+		ret[0] = 0;
+		pgtable_cache_size--;
+	}
+	return (pte_t *)ret;
+}
+
+static inline void pte_free_fast(pte_t *pte)
+{
+	*(unsigned long **)pte = pte_quicklist;
+	pte_quicklist = (unsigned long *) pte;
+	pgtable_cache_size++;
+}
+
+static inline void pte_free_kernel(struct mm_struct *mm, pte_t *pte)
+{
+	free_page((unsigned long)pte);
+}
+
+static inline void pte_free_slow(struct page *ptepage)
+{
+	__free_page(ptepage);
+}
+
+static inline void pte_free(struct mm_struct *mm, struct page *ptepage)
+{
+	pgtable_page_dtor(ptepage);
+	__free_page(ptepage);
+}
+
+#define __pte_free_tlb(tlb, pte, addr)	pte_free((tlb)->mm, (pte))
+
+#define pmd_populate(mm, pmd, pte) \
+			(pmd_val(*(pmd)) = (unsigned long)page_address(pte))
+
+#define pmd_populate_kernel(mm, pmd, pte) \
+		(pmd_val(*(pmd)) = (unsigned long) (pte))
+
+/*
+ * We don't have any real pmd's, and this code never triggers because
+ * the pgd will always be present..
+ */
+#define pmd_alloc_one(mm, address)	({ BUG(); ((pmd_t *)2); })
+#define pmd_free(mm, x)			do { } while (0)
+#define __pmd_free_tlb(tlb, x, addr)	pmd_free((tlb)->mm, x)
+#define pgd_populate(mm, pmd, pte)	BUG()
+
+extern int do_check_pgt_cache(int, int);
+
+#endif /* CONFIG_MMU */
+
+#define check_pgt_cache()		do { } while (0)
+
+#endif /* _ASM_MICROBLAZE_PGALLOC_H */
diff --git a/arch/microblaze/include/asm/pgtable.h b/arch/microblaze/include/asm/pgtable.h
new file mode 100644
index 0000000..e53b853
--- /dev/null
+++ b/arch/microblaze/include/asm/pgtable.h
@@ -0,0 +1,564 @@
+/*
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_PGTABLE_H
+#define _ASM_MICROBLAZE_PGTABLE_H
+
+#include <asm/setup.h>
+
+#ifndef __ASSEMBLY__
+extern int mem_init_done;
+#endif
+
+#ifndef CONFIG_MMU
+
+#define pgd_present(pgd)	(1) /* pages are always present on non MMU */
+#define pgd_none(pgd)		(0)
+#define pgd_bad(pgd)		(0)
+#define pgd_clear(pgdp)
+#define kern_addr_valid(addr)	(1)
+#define	pmd_offset(a, b)	((void *) 0)
+
+#define PAGE_NONE		__pgprot(0) /* these mean nothing to non MMU */
+#define PAGE_SHARED		__pgprot(0) /* these mean nothing to non MMU */
+#define PAGE_COPY		__pgprot(0) /* these mean nothing to non MMU */
+#define PAGE_READONLY		__pgprot(0) /* these mean nothing to non MMU */
+#define PAGE_KERNEL		__pgprot(0) /* these mean nothing to non MMU */
+
+#define pgprot_noncached(x)	(x)
+
+#define __swp_type(x)		(0)
+#define __swp_offset(x)		(0)
+#define __swp_entry(typ, off)	((swp_entry_t) { ((typ) | ((off) << 7)) })
+#define __pte_to_swp_entry(pte)	((swp_entry_t) { pte_val(pte) })
+#define __swp_entry_to_pte(x)	((pte_t) { (x).val })
+
+#define ZERO_PAGE(vaddr)	({ BUG(); NULL; })
+
+#define swapper_pg_dir ((pgd_t *) NULL)
+
+#define pgtable_cache_init()	do {} while (0)
+
+#define arch_enter_lazy_cpu_mode()	do {} while (0)
+
+#define pgprot_noncached_wc(prot)	prot
+
+/*
+ * All 32bit addresses are effectively valid for vmalloc...
+ * Sort of meaningless for non-VM targets.
+ */
+#define	VMALLOC_START	0
+#define	VMALLOC_END	0xffffffff
+
+#else /* CONFIG_MMU */
+
+#include <asm-generic/4level-fixup.h>
+
+#define __PAGETABLE_PMD_FOLDED
+
+#ifdef __KERNEL__
+#ifndef __ASSEMBLY__
+
+#include <linux/sched.h>
+#include <linux/threads.h>
+#include <asm/processor.h>		/* For TASK_SIZE */
+#include <asm/mmu.h>
+#include <asm/page.h>
+
+#define FIRST_USER_ADDRESS	0UL
+
+extern unsigned long va_to_phys(unsigned long address);
+extern pte_t *va_to_pte(unsigned long address);
+
+/*
+ * The following only work if pte_present() is true.
+ * Undefined behaviour if not..
+ */
+
+static inline int pte_special(pte_t pte)	{ return 0; }
+
+static inline pte_t pte_mkspecial(pte_t pte)	{ return pte; }
+
+/* Start and end of the vmalloc area. */
+/* Make sure to map the vmalloc area above the pinned kernel memory area
+   of 32Mb.  */
+#define VMALLOC_START	(CONFIG_KERNEL_START + CONFIG_LOWMEM_SIZE)
+#define VMALLOC_END	ioremap_bot
+
+#endif /* __ASSEMBLY__ */
+
+/*
+ * Macro to mark a page protection value as "uncacheable".
+ */
+
+#define _PAGE_CACHE_CTL	(_PAGE_GUARDED | _PAGE_NO_CACHE | \
+							_PAGE_WRITETHRU)
+
+#define pgprot_noncached(prot) \
+			(__pgprot((pgprot_val(prot) & ~_PAGE_CACHE_CTL) | \
+					_PAGE_NO_CACHE | _PAGE_GUARDED))
+
+#define pgprot_noncached_wc(prot) \
+			 (__pgprot((pgprot_val(prot) & ~_PAGE_CACHE_CTL) | \
+							_PAGE_NO_CACHE))
+
+/*
+ * The MicroBlaze MMU is identical to the PPC-40x MMU, and uses a hash
+ * table containing PTEs, together with a set of 16 segment registers, to
+ * define the virtual to physical address mapping.
+ *
+ * We use the hash table as an extended TLB, i.e. a cache of currently
+ * active mappings.  We maintain a two-level page table tree, much
+ * like that used by the i386, for the sake of the Linux memory
+ * management code.  Low-level assembler code in hashtable.S
+ * (procedure hash_page) is responsible for extracting ptes from the
+ * tree and putting them into the hash table when necessary, and
+ * updating the accessed and modified bits in the page table tree.
+ */
+
+/*
+ * The MicroBlaze processor has a TLB architecture identical to PPC-40x. The
+ * instruction and data sides share a unified, 64-entry, semi-associative
+ * TLB which is maintained totally under software control. In addition, the
+ * instruction side has a hardware-managed, 2,4, or 8-entry, fully-associative
+ * TLB which serves as a first level to the shared TLB. These two TLBs are
+ * known as the UTLB and ITLB, respectively (see "mmu.h" for definitions).
+ */
+
+/*
+ * The normal case is that PTEs are 32-bits and we have a 1-page
+ * 1024-entry pgdir pointing to 1-page 1024-entry PTE pages.  -- paulus
+ *
+ */
+
+/* PMD_SHIFT determines the size of the area mapped by the PTE pages */
+#define PMD_SHIFT	(PAGE_SHIFT + PTE_SHIFT)
+#define PMD_SIZE	(1UL << PMD_SHIFT)
+#define PMD_MASK	(~(PMD_SIZE-1))
+
+/* PGDIR_SHIFT determines what a top-level page table entry can map */
+#define PGDIR_SHIFT	PMD_SHIFT
+#define PGDIR_SIZE	(1UL << PGDIR_SHIFT)
+#define PGDIR_MASK	(~(PGDIR_SIZE-1))
+
+/*
+ * entries per page directory level: our page-table tree is two-level, so
+ * we don't really have any PMD directory.
+ */
+#define PTRS_PER_PTE	(1 << PTE_SHIFT)
+#define PTRS_PER_PMD	1
+#define PTRS_PER_PGD	(1 << (32 - PGDIR_SHIFT))
+
+#define USER_PTRS_PER_PGD	(TASK_SIZE / PGDIR_SIZE)
+#define FIRST_USER_PGD_NR	0
+
+#define USER_PGD_PTRS (PAGE_OFFSET >> PGDIR_SHIFT)
+#define KERNEL_PGD_PTRS (PTRS_PER_PGD-USER_PGD_PTRS)
+
+#define pte_ERROR(e) \
+	printk(KERN_ERR "%s:%d: bad pte "PTE_FMT".\n", \
+		__FILE__, __LINE__, pte_val(e))
+#define pmd_ERROR(e) \
+	printk(KERN_ERR "%s:%d: bad pmd %08lx.\n", \
+		__FILE__, __LINE__, pmd_val(e))
+#define pgd_ERROR(e) \
+	printk(KERN_ERR "%s:%d: bad pgd %08lx.\n", \
+		__FILE__, __LINE__, pgd_val(e))
+
+/*
+ * Bits in a linux-style PTE.  These match the bits in the
+ * (hardware-defined) PTE as closely as possible.
+ */
+
+/* There are several potential gotchas here.  The hardware TLBLO
+ * field looks like this:
+ *
+ * 0  1  2  3  4  ... 18 19 20 21 22 23 24 25 26 27 28 29 30 31
+ * RPN.....................  0  0 EX WR ZSEL.......  W  I  M  G
+ *
+ * Where possible we make the Linux PTE bits match up with this
+ *
+ * - bits 20 and 21 must be cleared, because we use 4k pages (4xx can
+ * support down to 1k pages), this is done in the TLBMiss exception
+ * handler.
+ * - We use only zones 0 (for kernel pages) and 1 (for user pages)
+ * of the 16 available.  Bit 24-26 of the TLB are cleared in the TLB
+ * miss handler.  Bit 27 is PAGE_USER, thus selecting the correct
+ * zone.
+ * - PRESENT *must* be in the bottom two bits because swap cache
+ * entries use the top 30 bits.  Because 4xx doesn't support SMP
+ * anyway, M is irrelevant so we borrow it for PAGE_PRESENT.  Bit 30
+ * is cleared in the TLB miss handler before the TLB entry is loaded.
+ * - All other bits of the PTE are loaded into TLBLO without
+ *  * modification, leaving us only the bits 20, 21, 24, 25, 26, 30 for
+ * software PTE bits.  We actually use use bits 21, 24, 25, and
+ * 30 respectively for the software bits: ACCESSED, DIRTY, RW, and
+ * PRESENT.
+ */
+
+/* Definitions for MicroBlaze. */
+#define	_PAGE_GUARDED	0x001	/* G: page is guarded from prefetch */
+#define _PAGE_PRESENT	0x002	/* software: PTE contains a translation */
+#define	_PAGE_NO_CACHE	0x004	/* I: caching is inhibited */
+#define	_PAGE_WRITETHRU	0x008	/* W: caching is write-through */
+#define	_PAGE_USER	0x010	/* matches one of the zone permission bits */
+#define	_PAGE_RW	0x040	/* software: Writes permitted */
+#define	_PAGE_DIRTY	0x080	/* software: dirty page */
+#define _PAGE_HWWRITE	0x100	/* hardware: Dirty & RW, set in exception */
+#define _PAGE_HWEXEC	0x200	/* hardware: EX permission */
+#define _PAGE_ACCESSED	0x400	/* software: R: page referenced */
+#define _PMD_PRESENT	PAGE_MASK
+
+/*
+ * Some bits are unused...
+ */
+#ifndef _PAGE_HASHPTE
+#define _PAGE_HASHPTE	0
+#endif
+#ifndef _PTE_NONE_MASK
+#define _PTE_NONE_MASK	0
+#endif
+#ifndef _PAGE_SHARED
+#define _PAGE_SHARED	0
+#endif
+#ifndef _PAGE_EXEC
+#define _PAGE_EXEC	0
+#endif
+
+#define _PAGE_CHG_MASK	(PAGE_MASK | _PAGE_ACCESSED | _PAGE_DIRTY)
+
+/*
+ * Note: the _PAGE_COHERENT bit automatically gets set in the hardware
+ * PTE if CONFIG_SMP is defined (hash_page does this); there is no need
+ * to have it in the Linux PTE, and in fact the bit could be reused for
+ * another purpose.  -- paulus.
+ */
+#define _PAGE_BASE	(_PAGE_PRESENT | _PAGE_ACCESSED)
+#define _PAGE_WRENABLE	(_PAGE_RW | _PAGE_DIRTY | _PAGE_HWWRITE)
+
+#define _PAGE_KERNEL \
+	(_PAGE_BASE | _PAGE_WRENABLE | _PAGE_SHARED | _PAGE_HWEXEC)
+
+#define _PAGE_IO	(_PAGE_KERNEL | _PAGE_NO_CACHE | _PAGE_GUARDED)
+
+#define PAGE_NONE	__pgprot(_PAGE_BASE)
+#define PAGE_READONLY	__pgprot(_PAGE_BASE | _PAGE_USER)
+#define PAGE_READONLY_X	__pgprot(_PAGE_BASE | _PAGE_USER | _PAGE_EXEC)
+#define PAGE_SHARED	__pgprot(_PAGE_BASE | _PAGE_USER | _PAGE_RW)
+#define PAGE_SHARED_X \
+		__pgprot(_PAGE_BASE | _PAGE_USER | _PAGE_RW | _PAGE_EXEC)
+#define PAGE_COPY	__pgprot(_PAGE_BASE | _PAGE_USER)
+#define PAGE_COPY_X	__pgprot(_PAGE_BASE | _PAGE_USER | _PAGE_EXEC)
+
+#define PAGE_KERNEL	__pgprot(_PAGE_KERNEL)
+#define PAGE_KERNEL_RO	__pgprot(_PAGE_BASE | _PAGE_SHARED)
+#define PAGE_KERNEL_CI	__pgprot(_PAGE_IO)
+
+/*
+ * We consider execute permission the same as read.
+ * Also, write permissions imply read permissions.
+ */
+#define __P000	PAGE_NONE
+#define __P001	PAGE_READONLY_X
+#define __P010	PAGE_COPY
+#define __P011	PAGE_COPY_X
+#define __P100	PAGE_READONLY
+#define __P101	PAGE_READONLY_X
+#define __P110	PAGE_COPY
+#define __P111	PAGE_COPY_X
+
+#define __S000	PAGE_NONE
+#define __S001	PAGE_READONLY_X
+#define __S010	PAGE_SHARED
+#define __S011	PAGE_SHARED_X
+#define __S100	PAGE_READONLY
+#define __S101	PAGE_READONLY_X
+#define __S110	PAGE_SHARED
+#define __S111	PAGE_SHARED_X
+
+#ifndef __ASSEMBLY__
+/*
+ * ZERO_PAGE is a global shared page that is always zero: used
+ * for zero-mapped memory areas etc..
+ */
+extern unsigned long empty_zero_page[1024];
+#define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page))
+
+#endif /* __ASSEMBLY__ */
+
+#define pte_none(pte)		((pte_val(pte) & ~_PTE_NONE_MASK) == 0)
+#define pte_present(pte)	(pte_val(pte) & _PAGE_PRESENT)
+#define pte_clear(mm, addr, ptep) \
+	do { set_pte_at((mm), (addr), (ptep), __pte(0)); } while (0)
+
+#define pmd_none(pmd)		(!pmd_val(pmd))
+#define	pmd_bad(pmd)		((pmd_val(pmd) & _PMD_PRESENT) == 0)
+#define	pmd_present(pmd)	((pmd_val(pmd) & _PMD_PRESENT) != 0)
+#define	pmd_clear(pmdp)		do { pmd_val(*(pmdp)) = 0; } while (0)
+
+#define pte_page(x)		(mem_map + (unsigned long) \
+				((pte_val(x) - memory_start) >> PAGE_SHIFT))
+#define PFN_SHIFT_OFFSET	(PAGE_SHIFT)
+
+#define pte_pfn(x)		(pte_val(x) >> PFN_SHIFT_OFFSET)
+
+#define pfn_pte(pfn, prot) \
+	__pte(((pte_basic_t)(pfn) << PFN_SHIFT_OFFSET) | pgprot_val(prot))
+
+#ifndef __ASSEMBLY__
+/*
+ * The "pgd_xxx()" functions here are trivial for a folded two-level
+ * setup: the pgd is never bad, and a pmd always exists (as it's folded
+ * into the pgd entry)
+ */
+static inline int pgd_none(pgd_t pgd)		{ return 0; }
+static inline int pgd_bad(pgd_t pgd)		{ return 0; }
+static inline int pgd_present(pgd_t pgd)	{ return 1; }
+#define pgd_clear(xp)				do { } while (0)
+#define pgd_page(pgd) \
+	((unsigned long) __va(pgd_val(pgd) & PAGE_MASK))
+
+/*
+ * The following only work if pte_present() is true.
+ * Undefined behaviour if not..
+ */
+static inline int pte_read(pte_t pte)  { return pte_val(pte) & _PAGE_USER; }
+static inline int pte_write(pte_t pte) { return pte_val(pte) & _PAGE_RW; }
+static inline int pte_exec(pte_t pte)  { return pte_val(pte) & _PAGE_EXEC; }
+static inline int pte_dirty(pte_t pte) { return pte_val(pte) & _PAGE_DIRTY; }
+static inline int pte_young(pte_t pte) { return pte_val(pte) & _PAGE_ACCESSED; }
+
+static inline void pte_uncache(pte_t pte) { pte_val(pte) |= _PAGE_NO_CACHE; }
+static inline void pte_cache(pte_t pte)   { pte_val(pte) &= ~_PAGE_NO_CACHE; }
+
+static inline pte_t pte_rdprotect(pte_t pte) \
+		{ pte_val(pte) &= ~_PAGE_USER; return pte; }
+static inline pte_t pte_wrprotect(pte_t pte) \
+	{ pte_val(pte) &= ~(_PAGE_RW | _PAGE_HWWRITE); return pte; }
+static inline pte_t pte_exprotect(pte_t pte) \
+	{ pte_val(pte) &= ~_PAGE_EXEC; return pte; }
+static inline pte_t pte_mkclean(pte_t pte) \
+	{ pte_val(pte) &= ~(_PAGE_DIRTY | _PAGE_HWWRITE); return pte; }
+static inline pte_t pte_mkold(pte_t pte) \
+	{ pte_val(pte) &= ~_PAGE_ACCESSED; return pte; }
+
+static inline pte_t pte_mkread(pte_t pte) \
+	{ pte_val(pte) |= _PAGE_USER; return pte; }
+static inline pte_t pte_mkexec(pte_t pte) \
+	{ pte_val(pte) |= _PAGE_USER | _PAGE_EXEC; return pte; }
+static inline pte_t pte_mkwrite(pte_t pte) \
+	{ pte_val(pte) |= _PAGE_RW; return pte; }
+static inline pte_t pte_mkdirty(pte_t pte) \
+	{ pte_val(pte) |= _PAGE_DIRTY; return pte; }
+static inline pte_t pte_mkyoung(pte_t pte) \
+	{ pte_val(pte) |= _PAGE_ACCESSED; return pte; }
+
+/*
+ * Conversion functions: convert a page and protection to a page entry,
+ * and a page entry and page directory to the page they refer to.
+ */
+
+static inline pte_t mk_pte_phys(phys_addr_t physpage, pgprot_t pgprot)
+{
+	pte_t pte;
+	pte_val(pte) = physpage | pgprot_val(pgprot);
+	return pte;
+}
+
+#define mk_pte(page, pgprot) \
+({									   \
+	pte_t pte;							   \
+	pte_val(pte) = (((page - mem_map) << PAGE_SHIFT) + memory_start) |  \
+			pgprot_val(pgprot);				   \
+	pte;								   \
+})
+
+static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
+{
+	pte_val(pte) = (pte_val(pte) & _PAGE_CHG_MASK) | pgprot_val(newprot);
+	return pte;
+}
+
+/*
+ * Atomic PTE updates.
+ *
+ * pte_update clears and sets bit atomically, and returns
+ * the old pte value.
+ * The ((unsigned long)(p+1) - 4) hack is to get to the least-significant
+ * 32 bits of the PTE regardless of whether PTEs are 32 or 64 bits.
+ */
+static inline unsigned long pte_update(pte_t *p, unsigned long clr,
+				unsigned long set)
+{
+	unsigned long flags, old, tmp;
+
+	raw_local_irq_save(flags);
+
+	__asm__ __volatile__(	"lw	%0, %2, r0	\n"
+				"andn	%1, %0, %3	\n"
+				"or	%1, %1, %4	\n"
+				"sw	%1, %2, r0	\n"
+			: "=&r" (old), "=&r" (tmp)
+			: "r" ((unsigned long)(p + 1) - 4), "r" (clr), "r" (set)
+			: "cc");
+
+	raw_local_irq_restore(flags);
+
+	return old;
+}
+
+/*
+ * set_pte stores a linux PTE into the linux page table.
+ */
+static inline void set_pte(struct mm_struct *mm, unsigned long addr,
+		pte_t *ptep, pte_t pte)
+{
+	*ptep = pte;
+}
+
+static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
+		pte_t *ptep, pte_t pte)
+{
+	*ptep = pte;
+}
+
+#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
+static inline int ptep_test_and_clear_young(struct vm_area_struct *vma,
+		unsigned long address, pte_t *ptep)
+{
+	return (pte_update(ptep, _PAGE_ACCESSED, 0) & _PAGE_ACCESSED) != 0;
+}
+
+static inline int ptep_test_and_clear_dirty(struct mm_struct *mm,
+		unsigned long addr, pte_t *ptep)
+{
+	return (pte_update(ptep, \
+		(_PAGE_DIRTY | _PAGE_HWWRITE), 0) & _PAGE_DIRTY) != 0;
+}
+
+#define __HAVE_ARCH_PTEP_GET_AND_CLEAR
+static inline pte_t ptep_get_and_clear(struct mm_struct *mm,
+		unsigned long addr, pte_t *ptep)
+{
+	return __pte(pte_update(ptep, ~_PAGE_HASHPTE, 0));
+}
+
+/*static inline void ptep_set_wrprotect(struct mm_struct *mm,
+		unsigned long addr, pte_t *ptep)
+{
+	pte_update(ptep, (_PAGE_RW | _PAGE_HWWRITE), 0);
+}*/
+
+static inline void ptep_mkdirty(struct mm_struct *mm,
+		unsigned long addr, pte_t *ptep)
+{
+	pte_update(ptep, 0, _PAGE_DIRTY);
+}
+
+/*#define pte_same(A,B)	(((pte_val(A) ^ pte_val(B)) & ~_PAGE_HASHPTE) == 0)*/
+
+/* Convert pmd entry to page */
+/* our pmd entry is an effective address of pte table*/
+/* returns effective address of the pmd entry*/
+#define pmd_page_kernel(pmd)	((unsigned long) (pmd_val(pmd) & PAGE_MASK))
+
+/* returns struct *page of the pmd entry*/
+#define pmd_page(pmd)	(pfn_to_page(__pa(pmd_val(pmd)) >> PAGE_SHIFT))
+
+/* to find an entry in a kernel page-table-directory */
+#define pgd_offset_k(address) pgd_offset(&init_mm, address)
+
+/* to find an entry in a page-table-directory */
+#define pgd_index(address)	 ((address) >> PGDIR_SHIFT)
+#define pgd_offset(mm, address)	 ((mm)->pgd + pgd_index(address))
+
+/* Find an entry in the second-level page table.. */
+static inline pmd_t *pmd_offset(pgd_t *dir, unsigned long address)
+{
+	return (pmd_t *) dir;
+}
+
+/* Find an entry in the third-level page table.. */
+#define pte_index(address)		\
+	(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
+#define pte_offset_kernel(dir, addr)	\
+	((pte_t *) pmd_page_kernel(*(dir)) + pte_index(addr))
+#define pte_offset_map(dir, addr)		\
+	((pte_t *) kmap_atomic(pmd_page(*(dir))) + pte_index(addr))
+
+#define pte_unmap(pte)		kunmap_atomic(pte)
+
+extern pgd_t swapper_pg_dir[PTRS_PER_PGD];
+
+/*
+ * Encode and decode a swap entry.
+ * Note that the bits we use in a PTE for representing a swap entry
+ * must not include the _PAGE_PRESENT bit, or the _PAGE_HASHPTE bit
+ * (if used).  -- paulus
+ */
+#define __swp_type(entry)		((entry).val & 0x3f)
+#define __swp_offset(entry)	((entry).val >> 6)
+#define __swp_entry(type, offset) \
+		((swp_entry_t) { (type) | ((offset) << 6) })
+#define __pte_to_swp_entry(pte)	((swp_entry_t) { pte_val(pte) >> 2 })
+#define __swp_entry_to_pte(x)	((pte_t) { (x).val << 2 })
+
+extern unsigned long iopa(unsigned long addr);
+
+/* Values for nocacheflag and cmode */
+/* These are not used by the APUS kernel_map, but prevents
+ * compilation errors.
+ */
+#define	IOMAP_FULL_CACHING	0
+#define	IOMAP_NOCACHE_SER	1
+#define	IOMAP_NOCACHE_NONSER	2
+#define	IOMAP_NO_COPYBACK	3
+
+/* Needs to be defined here and not in linux/mm.h, as it is arch dependent */
+#define kern_addr_valid(addr)	(1)
+
+/*
+ * No page table caches to initialise
+ */
+#define pgtable_cache_init()	do { } while (0)
+
+void do_page_fault(struct pt_regs *regs, unsigned long address,
+		   unsigned long error_code);
+
+void mapin_ram(void);
+int map_page(unsigned long va, phys_addr_t pa, int flags);
+
+extern int mem_init_done;
+
+asmlinkage void __init mmu_init(void);
+
+void __init *early_get_page(void);
+
+#endif /* __ASSEMBLY__ */
+#endif /* __KERNEL__ */
+
+#endif /* CONFIG_MMU */
+
+#ifndef __ASSEMBLY__
+#include <asm-generic/pgtable.h>
+
+extern unsigned long ioremap_bot, ioremap_base;
+
+void *consistent_alloc(gfp_t gfp, size_t size, dma_addr_t *dma_handle);
+void consistent_free(size_t size, void *vaddr);
+void consistent_sync(void *vaddr, size_t size, int direction);
+void consistent_sync_page(struct page *page, unsigned long offset,
+	size_t size, int direction);
+unsigned long consistent_virt_to_pfn(void *vaddr);
+
+void setup_memory(void);
+#endif /* __ASSEMBLY__ */
+
+#endif /* _ASM_MICROBLAZE_PGTABLE_H */
diff --git a/arch/microblaze/include/asm/processor.h b/arch/microblaze/include/asm/processor.h
new file mode 100644
index 0000000..497a988
--- /dev/null
+++ b/arch/microblaze/include/asm/processor.h
@@ -0,0 +1,169 @@
+/*
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_PROCESSOR_H
+#define _ASM_MICROBLAZE_PROCESSOR_H
+
+#include <asm/ptrace.h>
+#include <asm/setup.h>
+#include <asm/registers.h>
+#include <asm/entry.h>
+#include <asm/current.h>
+
+# ifndef __ASSEMBLY__
+/* from kernel/cpu/mb.c */
+extern const struct seq_operations cpuinfo_op;
+
+# define cpu_relax()		barrier()
+# define cpu_relax_lowlatency()	cpu_relax()
+
+#define task_pt_regs(tsk) \
+		(((struct pt_regs *)(THREAD_SIZE + task_stack_page(tsk))) - 1)
+
+/* Do necessary setup to start up a newly executed thread. */
+void start_thread(struct pt_regs *regs, unsigned long pc, unsigned long usp);
+
+extern void ret_from_fork(void);
+extern void ret_from_kernel_thread(void);
+
+# endif /* __ASSEMBLY__ */
+
+# ifndef CONFIG_MMU
+/*
+ * User space process size: memory size
+ *
+ * TASK_SIZE on MMU cpu is usually 1GB. However, on no-MMU arch, both
+ * user processes and the kernel is on the same memory region. They
+ * both share the memory space and that is limited by the amount of
+ * physical memory. thus, we set TASK_SIZE == amount of total memory.
+ */
+# define TASK_SIZE	(0x81000000 - 0x80000000)
+
+/*
+ * Default implementation of macro that returns current
+ * instruction pointer ("program counter").
+ */
+# define current_text_addr() ({ __label__ _l; _l: &&_l; })
+
+/*
+ * This decides where the kernel will search for a free chunk of vm
+ * space during mmap's. We won't be using it
+ */
+# define TASK_UNMAPPED_BASE	0
+
+/* definition in include/linux/sched.h */
+struct task_struct;
+
+/* thread_struct is gone. use thread_info instead. */
+struct thread_struct { };
+# define INIT_THREAD	{ }
+
+/* Free all resources held by a thread. */
+static inline void release_thread(struct task_struct *dead_task)
+{
+}
+
+/* Free all resources held by a thread. */
+static inline void exit_thread(void)
+{
+}
+
+extern unsigned long thread_saved_pc(struct task_struct *t);
+
+extern unsigned long get_wchan(struct task_struct *p);
+
+# define KSTK_EIP(tsk)	(0)
+# define KSTK_ESP(tsk)	(0)
+
+# else /* CONFIG_MMU */
+
+/*
+ * This is used to define STACK_TOP, and with MMU it must be below
+ * kernel base to select the correct PGD when handling MMU exceptions.
+ */
+# define TASK_SIZE	(CONFIG_KERNEL_START)
+
+/*
+ * This decides where the kernel will search for a free chunk of vm
+ * space during mmap's.
+ */
+# define TASK_UNMAPPED_BASE	(TASK_SIZE / 8 * 3)
+
+# define THREAD_KSP	0
+
+#  ifndef __ASSEMBLY__
+
+/*
+ * Default implementation of macro that returns current
+ * instruction pointer ("program counter").
+ */
+#  define current_text_addr()	({ __label__ _l; _l: &&_l; })
+
+/* If you change this, you must change the associated assembly-languages
+ * constants defined below, THREAD_*.
+ */
+struct thread_struct {
+	/* kernel stack pointer (must be first field in structure) */
+	unsigned long	ksp;
+	unsigned long	ksp_limit;	/* if ksp <= ksp_limit stack overflow */
+	void		*pgdir;		/* root of page-table tree */
+	struct pt_regs	*regs;		/* Pointer to saved register state */
+};
+
+#  define INIT_THREAD { \
+	.ksp   = sizeof init_stack + (unsigned long)init_stack, \
+	.pgdir = swapper_pg_dir, \
+}
+
+/* Free all resources held by a thread. */
+static inline void release_thread(struct task_struct *dead_task)
+{
+}
+
+/* Free current thread data structures etc.  */
+static inline void exit_thread(void)
+{
+}
+
+/* Return saved (kernel) PC of a blocked thread.  */
+#  define thread_saved_pc(tsk)	\
+	((tsk)->thread.regs ? (tsk)->thread.regs->r15 : 0)
+
+unsigned long get_wchan(struct task_struct *p);
+
+/* The size allocated for kernel stacks. This _must_ be a power of two! */
+# define KERNEL_STACK_SIZE	0x2000
+
+/* Return some info about the user process TASK.  */
+#  define task_tos(task)	((unsigned long)(task) + KERNEL_STACK_SIZE)
+#  define task_regs(task) ((struct pt_regs *)task_tos(task) - 1)
+
+#  define task_pt_regs_plus_args(tsk) \
+	((void *)task_pt_regs(tsk))
+
+#  define task_sp(task)	(task_regs(task)->r1)
+#  define task_pc(task)	(task_regs(task)->pc)
+/* Grotty old names for some.  */
+#  define KSTK_EIP(task)	(task_pc(task))
+#  define KSTK_ESP(task)	(task_sp(task))
+
+/* FIXME */
+#  define deactivate_mm(tsk, mm)	do { } while (0)
+
+#  define STACK_TOP	TASK_SIZE
+#  define STACK_TOP_MAX	STACK_TOP
+
+#ifdef CONFIG_DEBUG_FS
+extern struct dentry *of_debugfs_root;
+#endif
+
+#  endif /* __ASSEMBLY__ */
+# endif /* CONFIG_MMU */
+#endif /* _ASM_MICROBLAZE_PROCESSOR_H */
diff --git a/arch/microblaze/include/asm/prom.h b/arch/microblaze/include/asm/prom.h
new file mode 100644
index 0000000..2f03ac8
--- /dev/null
+++ b/arch/microblaze/include/asm/prom.h
@@ -0,0 +1,27 @@
+/*
+ * Definitions for talking to the Open Firmware PROM on
+ * Power Macintosh computers.
+ *
+ * Copyright (C) 1996-2005 Paul Mackerras.
+ *
+ * Updates for PPC64 by Peter Bergner & David Engebretsen, IBM Corp.
+ *
+ * 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.
+ */
+#ifndef _ASM_MICROBLAZE_PROM_H
+#define _ASM_MICROBLAZE_PROM_H
+
+#include <linux/of.h>
+
+/* Other Prototypes */
+enum early_consoles {
+	UARTLITE = 1,
+	UART16550 = 2,
+};
+
+extern int of_early_console(void *version);
+
+#endif /* _ASM_MICROBLAZE_PROM_H */
diff --git a/arch/microblaze/include/asm/ptrace.h b/arch/microblaze/include/asm/ptrace.h
new file mode 100644
index 0000000..5b18ec1
--- /dev/null
+++ b/arch/microblaze/include/asm/ptrace.h
@@ -0,0 +1,27 @@
+/*
+ * 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.
+ */
+#ifndef _ASM_MICROBLAZE_PTRACE_H
+#define _ASM_MICROBLAZE_PTRACE_H
+
+#include <uapi/asm/ptrace.h>
+
+#ifndef __ASSEMBLY__
+#define kernel_mode(regs)		((regs)->pt_mode)
+#define user_mode(regs)			(!kernel_mode(regs))
+
+#define instruction_pointer(regs)	((regs)->pc)
+#define profile_pc(regs)		instruction_pointer(regs)
+#define user_stack_pointer(regs)	((regs)->r1)
+
+static inline long regs_return_value(struct pt_regs *regs)
+{
+	return regs->r3;
+}
+
+#endif /* __ASSEMBLY__ */
+#endif /* _ASM_MICROBLAZE_PTRACE_H */
diff --git a/arch/microblaze/include/asm/pvr.h b/arch/microblaze/include/asm/pvr.h
new file mode 100644
index 0000000..4bbdb4c
--- /dev/null
+++ b/arch/microblaze/include/asm/pvr.h
@@ -0,0 +1,227 @@
+/*
+ * Support for the MicroBlaze PVR (Processor Version Register)
+ *
+ * Copyright (C) 2009 - 2011 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007 John Williams <john.williams@petalogix.com>
+ * Copyright (C) 2007 - 2011 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_PVR_H
+#define _ASM_MICROBLAZE_PVR_H
+
+#define PVR_MSR_BIT 0x400
+
+struct pvr_s {
+	unsigned pvr[12];
+};
+
+/* The following taken from Xilinx's standalone BSP pvr.h */
+
+/* Basic PVR mask */
+#define PVR0_PVR_FULL_MASK		0x80000000
+#define PVR0_USE_BARREL_MASK		0x40000000
+#define PVR0_USE_DIV_MASK		0x20000000
+#define PVR0_USE_HW_MUL_MASK		0x10000000
+#define PVR0_USE_FPU_MASK		0x08000000
+#define PVR0_USE_EXC_MASK		0x04000000
+#define PVR0_USE_ICACHE_MASK		0x02000000
+#define PVR0_USE_DCACHE_MASK		0x01000000
+#define PVR0_USE_MMU			0x00800000
+#define PVR0_USE_BTC			0x00400000
+#define PVR0_ENDI			0x00200000
+#define PVR0_VERSION_MASK		0x0000FF00
+#define PVR0_USER1_MASK			0x000000FF
+
+/* User 2 PVR mask */
+#define PVR1_USER2_MASK			0xFFFFFFFF
+
+/* Configuration PVR masks */
+#define PVR2_D_OPB_MASK			0x80000000 /* or AXI */
+#define PVR2_D_LMB_MASK			0x40000000
+#define PVR2_I_OPB_MASK			0x20000000 /* or AXI */
+#define PVR2_I_LMB_MASK			0x10000000
+#define PVR2_INTERRUPT_IS_EDGE_MASK	0x08000000
+#define PVR2_EDGE_IS_POSITIVE_MASK	0x04000000
+#define PVR2_D_PLB_MASK			0x02000000 /* new */
+#define PVR2_I_PLB_MASK			0x01000000 /* new */
+#define PVR2_INTERCONNECT		0x00800000 /* new */
+#define PVR2_USE_EXTEND_FSL		0x00080000 /* new */
+#define PVR2_USE_FSL_EXC		0x00040000 /* new */
+#define PVR2_USE_MSR_INSTR		0x00020000
+#define PVR2_USE_PCMP_INSTR		0x00010000
+#define PVR2_AREA_OPTIMISED		0x00008000
+#define PVR2_USE_BARREL_MASK		0x00004000
+#define PVR2_USE_DIV_MASK		0x00002000
+#define PVR2_USE_HW_MUL_MASK		0x00001000
+#define PVR2_USE_FPU_MASK		0x00000800
+#define PVR2_USE_MUL64_MASK		0x00000400
+#define PVR2_USE_FPU2_MASK		0x00000200 /* new */
+#define PVR2_USE_IPLBEXC 		0x00000100
+#define PVR2_USE_DPLBEXC		0x00000080
+#define PVR2_OPCODE_0x0_ILL_MASK	0x00000040
+#define PVR2_UNALIGNED_EXC_MASK		0x00000020
+#define PVR2_ILL_OPCODE_EXC_MASK	0x00000010
+#define PVR2_IOPB_BUS_EXC_MASK		0x00000008 /* or AXI */
+#define PVR2_DOPB_BUS_EXC_MASK		0x00000004 /* or AXI */
+#define PVR2_DIV_ZERO_EXC_MASK		0x00000002
+#define PVR2_FPU_EXC_MASK		0x00000001
+
+/* Debug and exception PVR masks */
+#define PVR3_DEBUG_ENABLED_MASK		0x80000000
+#define PVR3_NUMBER_OF_PC_BRK_MASK	0x1E000000
+#define PVR3_NUMBER_OF_RD_ADDR_BRK_MASK	0x00380000
+#define PVR3_NUMBER_OF_WR_ADDR_BRK_MASK	0x0000E000
+#define PVR3_FSL_LINKS_MASK		0x00000380
+
+/* ICache config PVR masks */
+#define PVR4_USE_ICACHE_MASK		0x80000000 /* ICU */
+#define PVR4_ICACHE_ADDR_TAG_BITS_MASK	0x7C000000 /* ICTS */
+#define PVR4_ICACHE_ALLOW_WR_MASK	0x01000000 /* ICW */
+#define PVR4_ICACHE_LINE_LEN_MASK	0x00E00000 /* ICLL */
+#define PVR4_ICACHE_BYTE_SIZE_MASK	0x001F0000 /* ICBS */
+#define PVR4_ICACHE_ALWAYS_USED		0x00008000 /* IAU */
+#define PVR4_ICACHE_INTERFACE		0x00002000 /* ICI */
+
+/* DCache config PVR masks */
+#define PVR5_USE_DCACHE_MASK		0x80000000 /* DCU */
+#define PVR5_DCACHE_ADDR_TAG_BITS_MASK	0x7C000000 /* DCTS */
+#define PVR5_DCACHE_ALLOW_WR_MASK	0x01000000 /* DCW */
+#define PVR5_DCACHE_LINE_LEN_MASK	0x00E00000 /* DCLL */
+#define PVR5_DCACHE_BYTE_SIZE_MASK	0x001F0000 /* DCBS */
+#define PVR5_DCACHE_ALWAYS_USED		0x00008000 /* DAU */
+#define PVR5_DCACHE_USE_WRITEBACK	0x00004000 /* DWB */
+#define PVR5_DCACHE_INTERFACE		0x00002000 /* DCI */
+
+/* ICache base address PVR mask */
+#define PVR6_ICACHE_BASEADDR_MASK	0xFFFFFFFF
+
+/* ICache high address PVR mask */
+#define PVR7_ICACHE_HIGHADDR_MASK	0xFFFFFFFF
+
+/* DCache base address PVR mask */
+#define PVR8_DCACHE_BASEADDR_MASK	0xFFFFFFFF
+
+/* DCache high address PVR mask */
+#define PVR9_DCACHE_HIGHADDR_MASK	0xFFFFFFFF
+
+/* Target family PVR mask */
+#define PVR10_TARGET_FAMILY_MASK	0xFF000000
+
+/* MMU description */
+#define PVR11_USE_MMU			0xC0000000
+#define PVR11_MMU_ITLB_SIZE		0x38000000
+#define PVR11_MMU_DTLB_SIZE		0x07000000
+#define PVR11_MMU_TLB_ACCESS		0x00C00000
+#define PVR11_MMU_ZONES			0x003C0000
+#define PVR11_MMU_PRIVINS		0x00010000
+/* MSR Reset value PVR mask */
+#define PVR11_MSR_RESET_VALUE_MASK	0x000007FF
+
+/* PVR access macros */
+#define PVR_IS_FULL(_pvr)	(_pvr.pvr[0] & PVR0_PVR_FULL_MASK)
+#define PVR_USE_BARREL(_pvr)	(_pvr.pvr[0] & PVR0_USE_BARREL_MASK)
+#define PVR_USE_DIV(_pvr)	(_pvr.pvr[0] & PVR0_USE_DIV_MASK)
+#define PVR_USE_HW_MUL(_pvr)	(_pvr.pvr[0] & PVR0_USE_HW_MUL_MASK)
+#define PVR_USE_FPU(_pvr)	(_pvr.pvr[0] & PVR0_USE_FPU_MASK)
+#define PVR_USE_FPU2(_pvr)	(_pvr.pvr[2] & PVR2_USE_FPU2_MASK)
+#define PVR_USE_ICACHE(_pvr)	(_pvr.pvr[0] & PVR0_USE_ICACHE_MASK)
+#define PVR_USE_DCACHE(_pvr)	(_pvr.pvr[0] & PVR0_USE_DCACHE_MASK)
+#define PVR_VERSION(_pvr)	((_pvr.pvr[0] & PVR0_VERSION_MASK) >> 8)
+#define PVR_USER1(_pvr)		(_pvr.pvr[0] & PVR0_USER1_MASK)
+#define PVR_USER2(_pvr)		(_pvr.pvr[1] & PVR1_USER2_MASK)
+
+#define PVR_D_OPB(_pvr)		(_pvr.pvr[2] & PVR2_D_OPB_MASK)
+#define PVR_D_LMB(_pvr)		(_pvr.pvr[2] & PVR2_D_LMB_MASK)
+#define PVR_I_OPB(_pvr)		(_pvr.pvr[2] & PVR2_I_OPB_MASK)
+#define PVR_I_LMB(_pvr)		(_pvr.pvr[2] & PVR2_I_LMB_MASK)
+#define PVR_INTERRUPT_IS_EDGE(_pvr) \
+			(_pvr.pvr[2] & PVR2_INTERRUPT_IS_EDGE_MASK)
+#define PVR_EDGE_IS_POSITIVE(_pvr) \
+			(_pvr.pvr[2] & PVR2_EDGE_IS_POSITIVE_MASK)
+#define PVR_USE_MSR_INSTR(_pvr)		(_pvr.pvr[2] & PVR2_USE_MSR_INSTR)
+#define PVR_USE_PCMP_INSTR(_pvr)	(_pvr.pvr[2] & PVR2_USE_PCMP_INSTR)
+#define PVR_AREA_OPTIMISED(_pvr)	(_pvr.pvr[2] & PVR2_AREA_OPTIMISED)
+#define PVR_USE_MUL64(_pvr)		(_pvr.pvr[2] & PVR2_USE_MUL64_MASK)
+#define PVR_OPCODE_0x0_ILLEGAL(_pvr) \
+			(_pvr.pvr[2] & PVR2_OPCODE_0x0_ILL_MASK)
+#define PVR_UNALIGNED_EXCEPTION(_pvr) \
+			(_pvr.pvr[2] & PVR2_UNALIGNED_EXC_MASK)
+#define PVR_ILL_OPCODE_EXCEPTION(_pvr) \
+			(_pvr.pvr[2] & PVR2_ILL_OPCODE_EXC_MASK)
+#define PVR_IOPB_BUS_EXCEPTION(_pvr) \
+			(_pvr.pvr[2] & PVR2_IOPB_BUS_EXC_MASK)
+#define PVR_DOPB_BUS_EXCEPTION(_pvr) \
+			(_pvr.pvr[2] & PVR2_DOPB_BUS_EXC_MASK)
+#define PVR_DIV_ZERO_EXCEPTION(_pvr) \
+			(_pvr.pvr[2] & PVR2_DIV_ZERO_EXC_MASK)
+#define PVR_FPU_EXCEPTION(_pvr)		(_pvr.pvr[2] & PVR2_FPU_EXC_MASK)
+#define PVR_FSL_EXCEPTION(_pvr)		(_pvr.pvr[2] & PVR2_USE_EXTEND_FSL)
+
+#define PVR_DEBUG_ENABLED(_pvr)		(_pvr.pvr[3] & PVR3_DEBUG_ENABLED_MASK)
+#define PVR_NUMBER_OF_PC_BRK(_pvr) \
+			((_pvr.pvr[3] & PVR3_NUMBER_OF_PC_BRK_MASK) >> 25)
+#define PVR_NUMBER_OF_RD_ADDR_BRK(_pvr) \
+			((_pvr.pvr[3] & PVR3_NUMBER_OF_RD_ADDR_BRK_MASK) >> 19)
+#define PVR_NUMBER_OF_WR_ADDR_BRK(_pvr) \
+			((_pvr.pvr[3] & PVR3_NUMBER_OF_WR_ADDR_BRK_MASK) >> 13)
+#define PVR_FSL_LINKS(_pvr)	((_pvr.pvr[3] & PVR3_FSL_LINKS_MASK) >> 7)
+
+#define PVR_ICACHE_ADDR_TAG_BITS(_pvr) \
+		((_pvr.pvr[4] & PVR4_ICACHE_ADDR_TAG_BITS_MASK) >> 26)
+#define PVR_ICACHE_USE_FSL(_pvr) \
+		(_pvr.pvr[4] & PVR4_ICACHE_USE_FSL_MASK)
+#define PVR_ICACHE_ALLOW_WR(_pvr) \
+		(_pvr.pvr[4] & PVR4_ICACHE_ALLOW_WR_MASK)
+#define PVR_ICACHE_LINE_LEN(_pvr) \
+		(1 << ((_pvr.pvr[4] & PVR4_ICACHE_LINE_LEN_MASK) >> 21))
+#define PVR_ICACHE_BYTE_SIZE(_pvr) \
+		(1 << ((_pvr.pvr[4] & PVR4_ICACHE_BYTE_SIZE_MASK) >> 16))
+
+#define PVR_DCACHE_ADDR_TAG_BITS(_pvr) \
+			((_pvr.pvr[5] & PVR5_DCACHE_ADDR_TAG_BITS_MASK) >> 26)
+#define PVR_DCACHE_USE_FSL(_pvr)	(_pvr.pvr[5] & PVR5_DCACHE_USE_FSL_MASK)
+#define PVR_DCACHE_ALLOW_WR(_pvr) \
+			(_pvr.pvr[5] & PVR5_DCACHE_ALLOW_WR_MASK)
+/* FIXME two shifts on one line needs any comment */
+#define PVR_DCACHE_LINE_LEN(_pvr) \
+		(1 << ((_pvr.pvr[5] & PVR5_DCACHE_LINE_LEN_MASK) >> 21))
+#define PVR_DCACHE_BYTE_SIZE(_pvr) \
+		(1 << ((_pvr.pvr[5] & PVR5_DCACHE_BYTE_SIZE_MASK) >> 16))
+
+#define PVR_DCACHE_USE_WRITEBACK(_pvr) \
+			((_pvr.pvr[5] & PVR5_DCACHE_USE_WRITEBACK) >> 14)
+
+#define PVR_ICACHE_BASEADDR(_pvr) \
+			(_pvr.pvr[6] & PVR6_ICACHE_BASEADDR_MASK)
+#define PVR_ICACHE_HIGHADDR(_pvr) \
+			(_pvr.pvr[7] & PVR7_ICACHE_HIGHADDR_MASK)
+#define PVR_DCACHE_BASEADDR(_pvr) \
+			(_pvr.pvr[8] & PVR8_DCACHE_BASEADDR_MASK)
+#define PVR_DCACHE_HIGHADDR(_pvr) \
+			(_pvr.pvr[9] & PVR9_DCACHE_HIGHADDR_MASK)
+
+#define PVR_TARGET_FAMILY(_pvr) \
+			((_pvr.pvr[10] & PVR10_TARGET_FAMILY_MASK) >> 24)
+
+#define PVR_MSR_RESET_VALUE(_pvr) \
+			(_pvr.pvr[11] & PVR11_MSR_RESET_VALUE_MASK)
+
+/* mmu */
+#define PVR_USE_MMU(_pvr)		((_pvr.pvr[11] & PVR11_USE_MMU) >> 30)
+#define PVR_MMU_ITLB_SIZE(_pvr)		(_pvr.pvr[11] & PVR11_MMU_ITLB_SIZE)
+#define PVR_MMU_DTLB_SIZE(_pvr)		(_pvr.pvr[11] & PVR11_MMU_DTLB_SIZE)
+#define PVR_MMU_TLB_ACCESS(_pvr)	(_pvr.pvr[11] & PVR11_MMU_TLB_ACCESS)
+#define PVR_MMU_ZONES(_pvr)		(_pvr.pvr[11] & PVR11_MMU_ZONES)
+#define PVR_MMU_PRIVINS(pvr)		(pvr.pvr[11] & PVR11_MMU_PRIVINS)
+
+/* endian */
+#define PVR_ENDIAN(_pvr)	(_pvr.pvr[0] & PVR0_ENDI)
+
+int cpu_has_pvr(void);
+void get_pvr(struct pvr_s *pvr);
+
+#endif /* _ASM_MICROBLAZE_PVR_H */
diff --git a/arch/microblaze/include/asm/registers.h b/arch/microblaze/include/asm/registers.h
new file mode 100644
index 0000000..68c3afb
--- /dev/null
+++ b/arch/microblaze/include/asm/registers.h
@@ -0,0 +1,50 @@
+/*
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_REGISTERS_H
+#define _ASM_MICROBLAZE_REGISTERS_H
+
+#define MSR_BE	(1<<0) /* 0x001 */
+#define MSR_IE	(1<<1) /* 0x002 */
+#define MSR_C	(1<<2) /* 0x004 */
+#define MSR_BIP	(1<<3) /* 0x008 */
+#define MSR_FSL	(1<<4) /* 0x010 */
+#define MSR_ICE	(1<<5) /* 0x020 */
+#define MSR_DZ	(1<<6) /* 0x040 */
+#define MSR_DCE	(1<<7) /* 0x080 */
+#define MSR_EE	(1<<8) /* 0x100 */
+#define MSR_EIP	(1<<9) /* 0x200 */
+#define MSR_CC	(1<<31)
+
+/* Floating Point Status Register (FSR) Bits */
+#define FSR_IO		(1<<4) /* Invalid operation */
+#define FSR_DZ		(1<<3) /* Divide-by-zero */
+#define FSR_OF		(1<<2) /* Overflow */
+#define FSR_UF		(1<<1) /* Underflow */
+#define FSR_DO		(1<<0) /* Denormalized operand error */
+
+# ifdef CONFIG_MMU
+/* Machine State Register (MSR) Fields */
+# define MSR_UM		(1<<11) /* User Mode */
+# define MSR_UMS	(1<<12) /* User Mode Save */
+# define MSR_VM		(1<<13) /* Virtual Mode */
+# define MSR_VMS	(1<<14) /* Virtual Mode Save */
+
+# define MSR_KERNEL	(MSR_EE | MSR_VM)
+/* # define MSR_USER	(MSR_KERNEL | MSR_UM | MSR_IE) */
+# define MSR_KERNEL_VMS	(MSR_EE | MSR_VMS)
+/* # define MSR_USER_VMS	(MSR_KERNEL_VMS | MSR_UMS | MSR_IE) */
+
+/* Exception State Register (ESR) Fields */
+# define	  ESR_DIZ	(1<<11) /* Zone Protection */
+# define	  ESR_S		(1<<10) /* Store instruction */
+
+# endif /* CONFIG_MMU */
+#endif /* _ASM_MICROBLAZE_REGISTERS_H */
diff --git a/arch/microblaze/include/asm/seccomp.h b/arch/microblaze/include/asm/seccomp.h
new file mode 100644
index 0000000..204618a
--- /dev/null
+++ b/arch/microblaze/include/asm/seccomp.h
@@ -0,0 +1,10 @@
+#ifndef _ASM_MICROBLAZE_SECCOMP_H
+#define _ASM_MICROBLAZE_SECCOMP_H
+
+#include <linux/unistd.h>
+
+#define __NR_seccomp_sigreturn		__NR_sigreturn
+
+#include <asm-generic/seccomp.h>
+
+#endif	/* _ASM_MICROBLAZE_SECCOMP_H */
diff --git a/arch/microblaze/include/asm/sections.h b/arch/microblaze/include/asm/sections.h
new file mode 100644
index 0000000..1b281d3
--- /dev/null
+++ b/arch/microblaze/include/asm/sections.h
@@ -0,0 +1,23 @@
+/*
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_SECTIONS_H
+#define _ASM_MICROBLAZE_SECTIONS_H
+
+#include <asm-generic/sections.h>
+
+# ifndef __ASSEMBLY__
+extern char _ssbss[], _esbss[];
+extern unsigned long __ivt_start[], __ivt_end[];
+
+extern u32 _fdt_start[], _fdt_end[];
+
+# endif /* !__ASSEMBLY__ */
+#endif /* _ASM_MICROBLAZE_SECTIONS_H */
diff --git a/arch/microblaze/include/asm/serial.h b/arch/microblaze/include/asm/serial.h
new file mode 100644
index 0000000..a0cb0ca
--- /dev/null
+++ b/arch/microblaze/include/asm/serial.h
@@ -0,0 +1 @@
+#include <asm-generic/serial.h>
diff --git a/arch/microblaze/include/asm/setup.h b/arch/microblaze/include/asm/setup.h
new file mode 100644
index 0000000..be84a4d
--- /dev/null
+++ b/arch/microblaze/include/asm/setup.h
@@ -0,0 +1,51 @@
+/*
+ * 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.
+ */
+#ifndef _ASM_MICROBLAZE_SETUP_H
+#define _ASM_MICROBLAZE_SETUP_H
+
+#include <uapi/asm/setup.h>
+
+# ifndef __ASSEMBLY__
+extern unsigned int boot_cpuid; /* move to smp.h */
+
+extern char cmd_line[COMMAND_LINE_SIZE];
+
+extern char *klimit;
+
+int setup_early_printk(char *opt);
+void remap_early_printk(void);
+void disable_early_printk(void);
+
+void microblaze_heartbeat(void);
+void microblaze_setup_heartbeat(void);
+
+#   ifdef CONFIG_MMU
+extern void mmu_reset(void);
+extern void early_console_reg_tlb_alloc(unsigned int addr);
+#   endif /* CONFIG_MMU */
+
+extern void of_platform_reset_gpio_probe(void);
+
+void time_init(void);
+void init_IRQ(void);
+void machine_early_init(const char *cmdline, unsigned int ram,
+		unsigned int fdt, unsigned int msr, unsigned int tlb0,
+		unsigned int tlb1);
+
+void machine_restart(char *cmd);
+void machine_shutdown(void);
+void machine_halt(void);
+void machine_power_off(void);
+
+extern void *alloc_maybe_bootmem(size_t size, gfp_t mask);
+extern void *zalloc_maybe_bootmem(size_t size, gfp_t mask);
+
+# endif /* __ASSEMBLY__ */
+#endif /* _ASM_MICROBLAZE_SETUP_H */
diff --git a/arch/microblaze/include/asm/shmparam.h b/arch/microblaze/include/asm/shmparam.h
new file mode 100644
index 0000000..93f30de
--- /dev/null
+++ b/arch/microblaze/include/asm/shmparam.h
@@ -0,0 +1 @@
+#include <asm-generic/shmparam.h>
diff --git a/arch/microblaze/include/asm/string.h b/arch/microblaze/include/asm/string.h
new file mode 100644
index 0000000..aec2f59
--- /dev/null
+++ b/arch/microblaze/include/asm/string.h
@@ -0,0 +1,24 @@
+/*
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_STRING_H
+#define _ASM_MICROBLAZE_STRING_H
+
+#ifdef __KERNEL__
+
+#define __HAVE_ARCH_MEMSET
+#define __HAVE_ARCH_MEMCPY
+#define __HAVE_ARCH_MEMMOVE
+
+extern void *memset(void *, int, __kernel_size_t);
+extern void *memcpy(void *, const void *, __kernel_size_t);
+extern void *memmove(void *, const void *, __kernel_size_t);
+
+#endif /* __KERNEL__ */
+
+#endif /* _ASM_MICROBLAZE_STRING_H */
diff --git a/arch/microblaze/include/asm/switch_to.h b/arch/microblaze/include/asm/switch_to.h
new file mode 100644
index 0000000..f45baa2
--- /dev/null
+++ b/arch/microblaze/include/asm/switch_to.h
@@ -0,0 +1,24 @@
+/*
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_SWITCH_TO_H
+#define _ASM_MICROBLAZE_SWITCH_TO_H
+
+struct task_struct;
+struct thread_info;
+
+extern struct task_struct *_switch_to(struct thread_info *prev,
+					struct thread_info *next);
+
+#define switch_to(prev, next, last)					\
+	do {								\
+		(last) = _switch_to(task_thread_info(prev),		\
+					task_thread_info(next));	\
+	} while (0)
+
+#endif /* _ASM_MICROBLAZE_SWITCH_TO_H */
diff --git a/arch/microblaze/include/asm/syscall.h b/arch/microblaze/include/asm/syscall.h
new file mode 100644
index 0000000..04a5bec
--- /dev/null
+++ b/arch/microblaze/include/asm/syscall.h
@@ -0,0 +1,107 @@
+#ifndef __ASM_MICROBLAZE_SYSCALL_H
+#define __ASM_MICROBLAZE_SYSCALL_H
+
+#include <uapi/linux/audit.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <asm/ptrace.h>
+
+/* The system call number is given by the user in R12 */
+static inline long syscall_get_nr(struct task_struct *task,
+				  struct pt_regs *regs)
+{
+	return regs->r12;
+}
+
+static inline void syscall_rollback(struct task_struct *task,
+				    struct pt_regs *regs)
+{
+	/* TODO.  */
+}
+
+static inline long syscall_get_error(struct task_struct *task,
+				     struct pt_regs *regs)
+{
+	return IS_ERR_VALUE(regs->r3) ? regs->r3 : 0;
+}
+
+static inline long syscall_get_return_value(struct task_struct *task,
+					    struct pt_regs *regs)
+{
+	return regs->r3;
+}
+
+static inline void syscall_set_return_value(struct task_struct *task,
+					    struct pt_regs *regs,
+					    int error, long val)
+{
+	if (error)
+		regs->r3 = -error;
+	else
+		regs->r3 = val;
+}
+
+static inline microblaze_reg_t microblaze_get_syscall_arg(struct pt_regs *regs,
+							  unsigned int n)
+{
+	switch (n) {
+	case 5: return regs->r10;
+	case 4: return regs->r9;
+	case 3: return regs->r8;
+	case 2: return regs->r7;
+	case 1: return regs->r6;
+	case 0: return regs->r5;
+	default:
+		BUG();
+	}
+	return ~0;
+}
+
+static inline void microblaze_set_syscall_arg(struct pt_regs *regs,
+					      unsigned int n,
+					      unsigned long val)
+{
+	switch (n) {
+	case 5:
+		regs->r10 = val;
+	case 4:
+		regs->r9 = val;
+	case 3:
+		regs->r8 = val;
+	case 2:
+		regs->r7 = val;
+	case 1:
+		regs->r6 = val;
+	case 0:
+		regs->r5 = val;
+	default:
+		BUG();
+	}
+}
+
+static inline void syscall_get_arguments(struct task_struct *task,
+					 struct pt_regs *regs,
+					 unsigned int i, unsigned int n,
+					 unsigned long *args)
+{
+	while (n--)
+		*args++ = microblaze_get_syscall_arg(regs, i++);
+}
+
+static inline void syscall_set_arguments(struct task_struct *task,
+					 struct pt_regs *regs,
+					 unsigned int i, unsigned int n,
+					 const unsigned long *args)
+{
+	while (n--)
+		microblaze_set_syscall_arg(regs, i++, *args++);
+}
+
+asmlinkage unsigned long do_syscall_trace_enter(struct pt_regs *regs);
+asmlinkage void do_syscall_trace_leave(struct pt_regs *regs);
+
+static inline int syscall_get_arch(void)
+{
+	return AUDIT_ARCH_MICROBLAZE;
+}
+#endif /* __ASM_MICROBLAZE_SYSCALL_H */
diff --git a/arch/microblaze/include/asm/thread_info.h b/arch/microblaze/include/asm/thread_info.h
new file mode 100644
index 0000000..383f387
--- /dev/null
+++ b/arch/microblaze/include/asm/thread_info.h
@@ -0,0 +1,180 @@
+/*
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_THREAD_INFO_H
+#define _ASM_MICROBLAZE_THREAD_INFO_H
+
+#ifdef __KERNEL__
+
+/* we have 8k stack */
+#define THREAD_SHIFT		13
+#define THREAD_SIZE		(1 << THREAD_SHIFT)
+#define THREAD_SIZE_ORDER	1
+
+#ifndef __ASSEMBLY__
+# include <linux/types.h>
+# include <asm/processor.h>
+
+/*
+ * low level task data that entry.S needs immediate access to
+ * - this struct should fit entirely inside of one cache line
+ * - this struct shares the supervisor stack pages
+ * - if the contents of this structure are changed, the assembly constants
+ *	 must also be changed
+ */
+
+struct cpu_context {
+	__u32	r1; /* stack pointer */
+	__u32	r2;
+	/* dedicated registers */
+	__u32	r13;
+	__u32	r14;
+	__u32	r15;
+	__u32	r16;
+	__u32	r17;
+	__u32	r18;
+	/* non-volatile registers */
+	__u32	r19;
+	__u32	r20;
+	__u32	r21;
+	__u32	r22;
+	__u32	r23;
+	__u32	r24;
+	__u32	r25;
+	__u32	r26;
+	__u32	r27;
+	__u32	r28;
+	__u32	r29;
+	__u32	r30;
+	/* r31 is used as current task pointer */
+	/* special purpose registers */
+	__u32	msr;
+	__u32	ear;
+	__u32	esr;
+	__u32	fsr;
+};
+
+typedef struct {
+	unsigned long seg;
+} mm_segment_t;
+
+struct thread_info {
+	struct task_struct	*task; /* main task structure */
+	unsigned long		flags; /* low level flags */
+	unsigned long		status; /* thread-synchronous flags */
+	__u32			cpu; /* current CPU */
+	__s32			preempt_count; /* 0 => preemptable,< 0 => BUG*/
+	mm_segment_t		addr_limit; /* thread address space */
+
+	struct cpu_context	cpu_context;
+};
+
+/*
+ * macros/functions for gaining access to the thread information structure
+ */
+#define INIT_THREAD_INFO(tsk)			\
+{						\
+	.task		= &tsk,			\
+	.flags		= 0,			\
+	.cpu		= 0,			\
+	.preempt_count	= INIT_PREEMPT_COUNT,	\
+	.addr_limit	= KERNEL_DS,		\
+}
+
+#define init_thread_info	(init_thread_union.thread_info)
+#define init_stack		(init_thread_union.stack)
+
+/* how to get the thread information struct from C */
+static inline struct thread_info *current_thread_info(void)
+{
+	register unsigned long sp asm("r1");
+
+	return (struct thread_info *)(sp & ~(THREAD_SIZE-1));
+}
+
+/* thread information allocation */
+#endif /* __ASSEMBLY__ */
+
+/*
+ * thread information flags
+ * - these are process state flags that various assembly files may
+ *   need to access
+ * - pending work-to-be-done flags are in LSW
+ * - other flags in MSW
+ */
+#define TIF_SYSCALL_TRACE	0 /* syscall trace active */
+#define TIF_NOTIFY_RESUME	1 /* resumption notification requested */
+#define TIF_SIGPENDING		2 /* signal pending */
+#define TIF_NEED_RESCHED	3 /* rescheduling necessary */
+/* restore singlestep on return to user mode */
+#define TIF_SINGLESTEP		4
+#define TIF_MEMDIE		6	/* is terminating due to OOM killer */
+#define TIF_SYSCALL_AUDIT	9       /* syscall auditing active */
+#define TIF_SECCOMP		10      /* secure computing */
+
+/* true if poll_idle() is polling TIF_NEED_RESCHED */
+#define TIF_POLLING_NRFLAG	16
+
+#define _TIF_SYSCALL_TRACE	(1 << TIF_SYSCALL_TRACE)
+#define _TIF_NOTIFY_RESUME	(1 << TIF_NOTIFY_RESUME)
+#define _TIF_SIGPENDING		(1 << TIF_SIGPENDING)
+#define _TIF_NEED_RESCHED	(1 << TIF_NEED_RESCHED)
+#define _TIF_SINGLESTEP		(1 << TIF_SINGLESTEP)
+#define _TIF_POLLING_NRFLAG	(1 << TIF_POLLING_NRFLAG)
+#define _TIF_SYSCALL_AUDIT	(1 << TIF_SYSCALL_AUDIT)
+#define _TIF_SECCOMP		(1 << TIF_SECCOMP)
+
+/* work to do in syscall trace */
+#define _TIF_WORK_SYSCALL_MASK  (_TIF_SYSCALL_TRACE | _TIF_SINGLESTEP | \
+				 _TIF_SYSCALL_AUDIT | _TIF_SECCOMP)
+
+/* work to do on interrupt/exception return */
+#define _TIF_WORK_MASK		0x0000FFFE
+
+/* work to do on any return to u-space */
+#define _TIF_ALLWORK_MASK	0x0000FFFF
+
+/*
+ * Thread-synchronous status.
+ *
+ * This is different from the flags in that nobody else
+ * ever touches our thread-synchronous status, so we don't
+ * have to worry about atomic accesses.
+ */
+/* FPU was used by this task this quantum (SMP) */
+#define TS_USEDFPU		0x0001
+#define TS_RESTORE_SIGMASK	0x0002
+
+#ifndef __ASSEMBLY__
+#define HAVE_SET_RESTORE_SIGMASK 1
+static inline void set_restore_sigmask(void)
+{
+	struct thread_info *ti = current_thread_info();
+	ti->status |= TS_RESTORE_SIGMASK;
+	WARN_ON(!test_bit(TIF_SIGPENDING, (unsigned long *)&ti->flags));
+}
+static inline void clear_restore_sigmask(void)
+{
+	current_thread_info()->status &= ~TS_RESTORE_SIGMASK;
+}
+static inline bool test_restore_sigmask(void)
+{
+	return current_thread_info()->status & TS_RESTORE_SIGMASK;
+}
+static inline bool test_and_clear_restore_sigmask(void)
+{
+	struct thread_info *ti = current_thread_info();
+	if (!(ti->status & TS_RESTORE_SIGMASK))
+		return false;
+	ti->status &= ~TS_RESTORE_SIGMASK;
+	return true;
+}
+#endif
+
+#endif /* __KERNEL__ */
+#endif /* _ASM_MICROBLAZE_THREAD_INFO_H */
diff --git a/arch/microblaze/include/asm/timex.h b/arch/microblaze/include/asm/timex.h
new file mode 100644
index 0000000..befcf3d
--- /dev/null
+++ b/arch/microblaze/include/asm/timex.h
@@ -0,0 +1,16 @@
+/*
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_TIMEX_H
+#define _ASM_MICROBLAZE_TIMEX_H
+
+#include <asm-generic/timex.h>
+
+#define CLOCK_TICK_RATE 1000 /* Timer input freq. */
+
+#endif /* _ASM_TIMEX_H */
diff --git a/arch/microblaze/include/asm/tlb.h b/arch/microblaze/include/asm/tlb.h
new file mode 100644
index 0000000..99b6ded
--- /dev/null
+++ b/arch/microblaze/include/asm/tlb.h
@@ -0,0 +1,26 @@
+/*
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_TLB_H
+#define _ASM_MICROBLAZE_TLB_H
+
+#define tlb_flush(tlb)	flush_tlb_mm((tlb)->mm)
+
+#include <linux/pagemap.h>
+
+#ifdef CONFIG_MMU
+#define tlb_start_vma(tlb, vma)		do { } while (0)
+#define tlb_end_vma(tlb, vma)		do { } while (0)
+#define __tlb_remove_tlb_entry(tlb, pte, address) do { } while (0)
+#endif
+
+#include <asm-generic/tlb.h>
+
+#endif /* _ASM_MICROBLAZE_TLB_H */
diff --git a/arch/microblaze/include/asm/tlbflush.h b/arch/microblaze/include/asm/tlbflush.h
new file mode 100644
index 0000000..2e1353c
--- /dev/null
+++ b/arch/microblaze/include/asm/tlbflush.h
@@ -0,0 +1,69 @@
+/*
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_TLBFLUSH_H
+#define _ASM_MICROBLAZE_TLBFLUSH_H
+
+#ifdef CONFIG_MMU
+
+#include <linux/sched.h>
+#include <linux/threads.h>
+#include <asm/processor.h>	/* For TASK_SIZE */
+#include <asm/mmu.h>
+#include <asm/page.h>
+#include <asm/pgalloc.h>
+
+extern void _tlbie(unsigned long address);
+extern void _tlbia(void);
+
+#define __tlbia()	{ preempt_disable(); _tlbia(); preempt_enable(); }
+#define __tlbie(x)	{ _tlbie(x); }
+
+static inline void local_flush_tlb_all(void)
+	{ __tlbia(); }
+static inline void local_flush_tlb_mm(struct mm_struct *mm)
+	{ __tlbia(); }
+static inline void local_flush_tlb_page(struct vm_area_struct *vma,
+				unsigned long vmaddr)
+	{ __tlbie(vmaddr); }
+static inline void local_flush_tlb_range(struct vm_area_struct *vma,
+		unsigned long start, unsigned long end)
+	{ __tlbia(); }
+
+#define flush_tlb_kernel_range(start, end)	do { } while (0)
+
+#define update_mmu_cache(vma, addr, ptep)	do { } while (0)
+
+#define flush_tlb_all local_flush_tlb_all
+#define flush_tlb_mm local_flush_tlb_mm
+#define flush_tlb_page local_flush_tlb_page
+#define flush_tlb_range local_flush_tlb_range
+
+/*
+ * This is called in munmap when we have freed up some page-table
+ * pages.  We don't need to do anything here, there's nothing special
+ * about our page-table pages.  -- paulus
+ */
+static inline void flush_tlb_pgtables(struct mm_struct *mm,
+	unsigned long start, unsigned long end) { }
+
+#else /* CONFIG_MMU */
+
+#define flush_tlb()				BUG()
+#define flush_tlb_all()				BUG()
+#define flush_tlb_mm(mm)			BUG()
+#define flush_tlb_page(vma, addr)		BUG()
+#define flush_tlb_range(mm, start, end)		BUG()
+#define flush_tlb_pgtables(mm, start, end)	BUG()
+#define flush_tlb_kernel_range(start, end)	BUG()
+
+#endif /* CONFIG_MMU */
+
+#endif /* _ASM_MICROBLAZE_TLBFLUSH_H */
diff --git a/arch/microblaze/include/asm/topology.h b/arch/microblaze/include/asm/topology.h
new file mode 100644
index 0000000..5428f33
--- /dev/null
+++ b/arch/microblaze/include/asm/topology.h
@@ -0,0 +1 @@
+#include <asm-generic/topology.h>
diff --git a/arch/microblaze/include/asm/uaccess.h b/arch/microblaze/include/asm/uaccess.h
new file mode 100644
index 0000000..8266767
--- /dev/null
+++ b/arch/microblaze/include/asm/uaccess.h
@@ -0,0 +1,431 @@
+/*
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_UACCESS_H
+#define _ASM_MICROBLAZE_UACCESS_H
+
+#ifdef __KERNEL__
+#ifndef __ASSEMBLY__
+
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/sched.h> /* RLIMIT_FSIZE */
+#include <linux/mm.h>
+
+#include <asm/mmu.h>
+#include <asm/page.h>
+#include <asm/pgtable.h>
+#include <linux/string.h>
+
+#define VERIFY_READ	0
+#define VERIFY_WRITE	1
+
+/*
+ * On Microblaze the fs value is actually the top of the corresponding
+ * address space.
+ *
+ * The fs value determines whether argument validity checking should be
+ * performed or not. If get_fs() == USER_DS, checking is performed, with
+ * get_fs() == KERNEL_DS, checking is bypassed.
+ *
+ * For historical reasons, these macros are grossly misnamed.
+ *
+ * For non-MMU arch like Microblaze, KERNEL_DS and USER_DS is equal.
+ */
+# define MAKE_MM_SEG(s)       ((mm_segment_t) { (s) })
+
+#  ifndef CONFIG_MMU
+#  define KERNEL_DS	MAKE_MM_SEG(0)
+#  define USER_DS	KERNEL_DS
+#  else
+#  define KERNEL_DS	MAKE_MM_SEG(0xFFFFFFFF)
+#  define USER_DS	MAKE_MM_SEG(TASK_SIZE - 1)
+#  endif
+
+# define get_ds()	(KERNEL_DS)
+# define get_fs()	(current_thread_info()->addr_limit)
+# define set_fs(val)	(current_thread_info()->addr_limit = (val))
+
+# define segment_eq(a, b)	((a).seg == (b).seg)
+
+/*
+ * The exception table consists of pairs of addresses: the first is the
+ * address of an instruction that is allowed to fault, and the second is
+ * the address at which the program should continue. No registers are
+ * modified, so it is entirely up to the continuation code to figure out
+ * what to do.
+ *
+ * All the routines below use bits of fixup code that are out of line
+ * with the main instruction path. This means when everything is well,
+ * we don't even have to jump over them. Further, they do not intrude
+ * on our cache or tlb entries.
+ */
+struct exception_table_entry {
+	unsigned long insn, fixup;
+};
+
+/* Returns 0 if exception not found and fixup otherwise.  */
+extern unsigned long search_exception_table(unsigned long);
+
+#ifndef CONFIG_MMU
+
+/* Check against bounds of physical memory */
+static inline int ___range_ok(unsigned long addr, unsigned long size)
+{
+	return ((addr < memory_start) ||
+		((addr + size - 1) > (memory_start + memory_size - 1)));
+}
+
+#define __range_ok(addr, size) \
+		___range_ok((unsigned long)(addr), (unsigned long)(size))
+
+#define access_ok(type, addr, size) (__range_ok((addr), (size)) == 0)
+
+#else
+
+static inline int access_ok(int type, const void __user *addr,
+							unsigned long size)
+{
+	if (!size)
+		goto ok;
+
+	if ((get_fs().seg < ((unsigned long)addr)) ||
+			(get_fs().seg < ((unsigned long)addr + size - 1))) {
+		pr_devel("ACCESS fail: %s at 0x%08x (size 0x%x), seg 0x%08x\n",
+			type ? "WRITE" : "READ ", (__force u32)addr, (u32)size,
+			(u32)get_fs().seg);
+		return 0;
+	}
+ok:
+	pr_devel("ACCESS OK: %s at 0x%08x (size 0x%x), seg 0x%08x\n",
+			type ? "WRITE" : "READ ", (__force u32)addr, (u32)size,
+			(u32)get_fs().seg);
+	return 1;
+}
+#endif
+
+#ifdef CONFIG_MMU
+# define __FIXUP_SECTION	".section .fixup,\"ax\"\n"
+# define __EX_TABLE_SECTION	".section __ex_table,\"a\"\n"
+#else
+# define __FIXUP_SECTION	".section .discard,\"ax\"\n"
+# define __EX_TABLE_SECTION	".section .discard,\"ax\"\n"
+#endif
+
+extern unsigned long __copy_tofrom_user(void __user *to,
+		const void __user *from, unsigned long size);
+
+/* Return: number of not copied bytes, i.e. 0 if OK or non-zero if fail. */
+static inline unsigned long __must_check __clear_user(void __user *to,
+							unsigned long n)
+{
+	/* normal memset with two words to __ex_table */
+	__asm__ __volatile__ (				\
+			"1:	sb	r0, %1, r0;"	\
+			"	addik	%0, %0, -1;"	\
+			"	bneid	%0, 1b;"	\
+			"	addik	%1, %1, 1;"	\
+			"2:			"	\
+			__EX_TABLE_SECTION		\
+			".word	1b,2b;"			\
+			".previous;"			\
+		: "=r"(n), "=r"(to)			\
+		: "0"(n), "1"(to)
+	);
+	return n;
+}
+
+static inline unsigned long __must_check clear_user(void __user *to,
+							unsigned long n)
+{
+	might_fault();
+	if (unlikely(!access_ok(VERIFY_WRITE, to, n)))
+		return n;
+
+	return __clear_user(to, n);
+}
+
+/* put_user and get_user macros */
+extern long __user_bad(void);
+
+#define __get_user_asm(insn, __gu_ptr, __gu_val, __gu_err)	\
+({								\
+	__asm__ __volatile__ (					\
+			"1:"	insn	" %1, %2, r0;"		\
+			"	addk	%0, r0, r0;"		\
+			"2:			"		\
+			__FIXUP_SECTION				\
+			"3:	brid	2b;"			\
+			"	addik	%0, r0, %3;"		\
+			".previous;"				\
+			__EX_TABLE_SECTION			\
+			".word	1b,3b;"				\
+			".previous;"				\
+		: "=&r"(__gu_err), "=r"(__gu_val)		\
+		: "r"(__gu_ptr), "i"(-EFAULT)			\
+	);							\
+})
+
+/**
+ * get_user: - Get a simple variable from user space.
+ * @x:   Variable to store result.
+ * @ptr: Source address, in user space.
+ *
+ * Context: User context only. This function may sleep if pagefaults are
+ *          enabled.
+ *
+ * This macro copies a single simple variable from user space to kernel
+ * space.  It supports simple types like char and int, but not larger
+ * data types like structures or arrays.
+ *
+ * @ptr must have pointer-to-simple-variable type, and the result of
+ * dereferencing @ptr must be assignable to @x without a cast.
+ *
+ * Returns zero on success, or -EFAULT on error.
+ * On error, the variable @x is set to zero.
+ */
+#define get_user(x, ptr)						\
+	__get_user_check((x), (ptr), sizeof(*(ptr)))
+
+#define __get_user_check(x, ptr, size)					\
+({									\
+	unsigned long __gu_val = 0;					\
+	const typeof(*(ptr)) __user *__gu_addr = (ptr);			\
+	int __gu_err = 0;						\
+									\
+	if (access_ok(VERIFY_READ, __gu_addr, size)) {			\
+		switch (size) {						\
+		case 1:							\
+			__get_user_asm("lbu", __gu_addr, __gu_val,	\
+				       __gu_err);			\
+			break;						\
+		case 2:							\
+			__get_user_asm("lhu", __gu_addr, __gu_val,	\
+				       __gu_err);			\
+			break;						\
+		case 4:							\
+			__get_user_asm("lw", __gu_addr, __gu_val,	\
+				       __gu_err);			\
+			break;						\
+		default:						\
+			__gu_err = __user_bad();			\
+			break;						\
+		}							\
+	} else {							\
+		__gu_err = -EFAULT;					\
+	}								\
+	x = (__force typeof(*(ptr)))__gu_val;				\
+	__gu_err;							\
+})
+
+#define __get_user(x, ptr)						\
+({									\
+	unsigned long __gu_val = 0;					\
+	/*unsigned long __gu_ptr = (unsigned long)(ptr);*/		\
+	long __gu_err;							\
+	switch (sizeof(*(ptr))) {					\
+	case 1:								\
+		__get_user_asm("lbu", (ptr), __gu_val, __gu_err);	\
+		break;							\
+	case 2:								\
+		__get_user_asm("lhu", (ptr), __gu_val, __gu_err);	\
+		break;							\
+	case 4:								\
+		__get_user_asm("lw", (ptr), __gu_val, __gu_err);	\
+		break;							\
+	default:							\
+		/* __gu_val = 0; __gu_err = -EINVAL;*/ __gu_err = __user_bad();\
+	}								\
+	x = (__force __typeof__(*(ptr))) __gu_val;			\
+	__gu_err;							\
+})
+
+
+#define __put_user_asm(insn, __gu_ptr, __gu_val, __gu_err)	\
+({								\
+	__asm__ __volatile__ (					\
+			"1:"	insn	" %1, %2, r0;"		\
+			"	addk	%0, r0, r0;"		\
+			"2:			"		\
+			__FIXUP_SECTION				\
+			"3:	brid	2b;"			\
+			"	addik	%0, r0, %3;"		\
+			".previous;"				\
+			__EX_TABLE_SECTION			\
+			".word	1b,3b;"				\
+			".previous;"				\
+		: "=&r"(__gu_err)				\
+		: "r"(__gu_val), "r"(__gu_ptr), "i"(-EFAULT)	\
+	);							\
+})
+
+#define __put_user_asm_8(__gu_ptr, __gu_val, __gu_err)		\
+({								\
+	__asm__ __volatile__ ("	lwi	%0, %1, 0;"		\
+			"1:	swi	%0, %2, 0;"		\
+			"	lwi	%0, %1, 4;"		\
+			"2:	swi	%0, %2, 4;"		\
+			"	addk	%0, r0, r0;"		\
+			"3:			"		\
+			__FIXUP_SECTION				\
+			"4:	brid	3b;"			\
+			"	addik	%0, r0, %3;"		\
+			".previous;"				\
+			__EX_TABLE_SECTION			\
+			".word	1b,4b,2b,4b;"			\
+			".previous;"				\
+		: "=&r"(__gu_err)				\
+		: "r"(&__gu_val), "r"(__gu_ptr), "i"(-EFAULT)	\
+		);						\
+})
+
+/**
+ * put_user: - Write a simple value into user space.
+ * @x:   Value to copy to user space.
+ * @ptr: Destination address, in user space.
+ *
+ * Context: User context only. This function may sleep if pagefaults are
+ *          enabled.
+ *
+ * This macro copies a single simple value from kernel space to user
+ * space.  It supports simple types like char and int, but not larger
+ * data types like structures or arrays.
+ *
+ * @ptr must have pointer-to-simple-variable type, and @x must be assignable
+ * to the result of dereferencing @ptr.
+ *
+ * Returns zero on success, or -EFAULT on error.
+ */
+#define put_user(x, ptr)						\
+	__put_user_check((x), (ptr), sizeof(*(ptr)))
+
+#define __put_user_check(x, ptr, size)					\
+({									\
+	typeof(*(ptr)) volatile __pu_val = x;				\
+	typeof(*(ptr)) __user *__pu_addr = (ptr);			\
+	int __pu_err = 0;						\
+									\
+	if (access_ok(VERIFY_WRITE, __pu_addr, size)) {			\
+		switch (size) {						\
+		case 1:							\
+			__put_user_asm("sb", __pu_addr, __pu_val,	\
+				       __pu_err);			\
+			break;						\
+		case 2:							\
+			__put_user_asm("sh", __pu_addr, __pu_val,	\
+				       __pu_err);			\
+			break;						\
+		case 4:							\
+			__put_user_asm("sw", __pu_addr, __pu_val,	\
+				       __pu_err);			\
+			break;						\
+		case 8:							\
+			__put_user_asm_8(__pu_addr, __pu_val, __pu_err);\
+			break;						\
+		default:						\
+			__pu_err = __user_bad();			\
+			break;						\
+		}							\
+	} else {							\
+		__pu_err = -EFAULT;					\
+	}								\
+	__pu_err;							\
+})
+
+#define __put_user(x, ptr)						\
+({									\
+	__typeof__(*(ptr)) volatile __gu_val = (x);			\
+	long __gu_err = 0;						\
+	switch (sizeof(__gu_val)) {					\
+	case 1:								\
+		__put_user_asm("sb", (ptr), __gu_val, __gu_err);	\
+		break;							\
+	case 2:								\
+		__put_user_asm("sh", (ptr), __gu_val, __gu_err);	\
+		break;							\
+	case 4:								\
+		__put_user_asm("sw", (ptr), __gu_val, __gu_err);	\
+		break;							\
+	case 8:								\
+		__put_user_asm_8((ptr), __gu_val, __gu_err);		\
+		break;							\
+	default:							\
+		/*__gu_err = -EINVAL;*/	__gu_err = __user_bad();	\
+	}								\
+	__gu_err;							\
+})
+
+
+/* copy_to_from_user */
+#define __copy_from_user(to, from, n)	\
+	__copy_tofrom_user((__force void __user *)(to), \
+				(void __user *)(from), (n))
+#define __copy_from_user_inatomic(to, from, n) \
+		__copy_from_user((to), (from), (n))
+
+static inline long copy_from_user(void *to,
+		const void __user *from, unsigned long n)
+{
+	unsigned long res = n;
+	might_fault();
+	if (likely(access_ok(VERIFY_READ, from, n)))
+		res = __copy_from_user(to, from, n);
+	if (unlikely(res))
+		memset(to + (n - res), 0, res);
+	return res;
+}
+
+#define __copy_to_user(to, from, n)	\
+		__copy_tofrom_user((void __user *)(to), \
+			(__force const void __user *)(from), (n))
+#define __copy_to_user_inatomic(to, from, n) __copy_to_user((to), (from), (n))
+
+static inline long copy_to_user(void __user *to,
+		const void *from, unsigned long n)
+{
+	might_fault();
+	if (access_ok(VERIFY_WRITE, to, n))
+		return __copy_to_user(to, from, n);
+	return n;
+}
+
+/*
+ * Copy a null terminated string from userspace.
+ */
+extern int __strncpy_user(char *to, const char __user *from, int len);
+
+#define __strncpy_from_user	__strncpy_user
+
+static inline long
+strncpy_from_user(char *dst, const char __user *src, long count)
+{
+	if (!access_ok(VERIFY_READ, src, 1))
+		return -EFAULT;
+	return __strncpy_from_user(dst, src, count);
+}
+
+/*
+ * Return the size of a string (including the ending 0)
+ *
+ * Return 0 on exception, a value greater than N if too long
+ */
+extern int __strnlen_user(const char __user *sstr, int len);
+
+static inline long strnlen_user(const char __user *src, long n)
+{
+	if (!access_ok(VERIFY_READ, src, 1))
+		return 0;
+	return __strnlen_user(src, n);
+}
+
+#endif  /* __ASSEMBLY__ */
+#endif /* __KERNEL__ */
+
+#endif /* _ASM_MICROBLAZE_UACCESS_H */
diff --git a/arch/microblaze/include/asm/ucontext.h b/arch/microblaze/include/asm/ucontext.h
new file mode 100644
index 0000000..9bc07b9
--- /dev/null
+++ b/arch/microblaze/include/asm/ucontext.h
@@ -0,0 +1 @@
+#include <asm-generic/ucontext.h>
diff --git a/arch/microblaze/include/asm/unaligned.h b/arch/microblaze/include/asm/unaligned.h
new file mode 100644
index 0000000..b162ed8
--- /dev/null
+++ b/arch/microblaze/include/asm/unaligned.h
@@ -0,0 +1,30 @@
+/*
+ * Copyright (C) 2008 Michal Simek <monstr@monstr.eu>
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_UNALIGNED_H
+#define _ASM_MICROBLAZE_UNALIGNED_H
+
+# ifdef __KERNEL__
+
+#  ifdef __MICROBLAZEEL__
+#   include <linux/unaligned/le_struct.h>
+#   include <linux/unaligned/be_byteshift.h>
+#   define get_unaligned	__get_unaligned_le
+#   define put_unaligned	__put_unaligned_le
+#  else
+#   include <linux/unaligned/be_struct.h>
+#   include <linux/unaligned/le_byteshift.h>
+#   define get_unaligned	__get_unaligned_be
+#   define put_unaligned	__put_unaligned_be
+#  endif
+
+# include <linux/unaligned/generic.h>
+
+# endif	/* __KERNEL__ */
+#endif /* _ASM_MICROBLAZE_UNALIGNED_H */
diff --git a/arch/microblaze/include/asm/unistd.h b/arch/microblaze/include/asm/unistd.h
new file mode 100644
index 0000000..76ed17b
--- /dev/null
+++ b/arch/microblaze/include/asm/unistd.h
@@ -0,0 +1,43 @@
+/*
+ * Copyright (C) 2007-2008 Michal Simek <monstr@monstr.eu>
+ * 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.
+ */
+#ifndef _ASM_MICROBLAZE_UNISTD_H
+#define _ASM_MICROBLAZE_UNISTD_H
+
+#include <uapi/asm/unistd.h>
+
+#ifndef __ASSEMBLY__
+
+/* #define __ARCH_WANT_OLD_READDIR */
+/* #define __ARCH_WANT_OLD_STAT */
+#define __ARCH_WANT_STAT64
+#define __ARCH_WANT_SYS_ALARM
+#define __ARCH_WANT_SYS_GETHOSTNAME
+#define __ARCH_WANT_SYS_PAUSE
+#define __ARCH_WANT_SYS_SIGNAL
+#define __ARCH_WANT_SYS_TIME
+#define __ARCH_WANT_SYS_UTIME
+#define __ARCH_WANT_SYS_WAITPID
+#define __ARCH_WANT_SYS_SOCKETCALL
+#define __ARCH_WANT_SYS_FADVISE64
+#define __ARCH_WANT_SYS_GETPGRP
+#define __ARCH_WANT_SYS_LLSEEK
+#define __ARCH_WANT_SYS_NICE
+/* #define __ARCH_WANT_SYS_OLD_GETRLIMIT */
+#define __ARCH_WANT_SYS_OLDUMOUNT
+#define __ARCH_WANT_SYS_SIGPENDING
+#define __ARCH_WANT_SYS_SIGPROCMASK
+#define __ARCH_WANT_SYS_CLONE
+#define __ARCH_WANT_SYS_VFORK
+#define __ARCH_WANT_SYS_FORK
+
+#endif /* __ASSEMBLY__ */
+
+#define __NR_syscalls         389
+
+#endif /* _ASM_MICROBLAZE_UNISTD_H */
diff --git a/arch/microblaze/include/asm/unwind.h b/arch/microblaze/include/asm/unwind.h
new file mode 100644
index 0000000..d248b7d
--- /dev/null
+++ b/arch/microblaze/include/asm/unwind.h
@@ -0,0 +1,29 @@
+/*
+ * Backtrace support for Microblaze
+ *
+ * Copyright (C) 2010  Digital Design Corporation
+ *
+ * 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.
+ */
+
+#ifndef __MICROBLAZE_UNWIND_H
+#define __MICROBLAZE_UNWIND_H
+
+struct stack_trace;
+
+struct trap_handler_info {
+	unsigned long start_addr;
+	unsigned long end_addr;
+	const char    *trap_name;
+};
+extern struct trap_handler_info microblaze_trap_handlers;
+
+extern const char _hw_exception_handler;
+extern const char ex_handler_unhandled;
+
+void microblaze_unwind(struct task_struct *task, struct stack_trace *trace);
+
+#endif	/* __MICROBLAZE_UNWIND_H */
+
diff --git a/arch/microblaze/include/asm/user.h b/arch/microblaze/include/asm/user.h
new file mode 100644
index 0000000..8b13789
--- /dev/null
+++ b/arch/microblaze/include/asm/user.h
@@ -0,0 +1 @@
+
diff --git a/arch/microblaze/include/asm/vga.h b/arch/microblaze/include/asm/vga.h
new file mode 100644
index 0000000..89d82fd
--- /dev/null
+++ b/arch/microblaze/include/asm/vga.h
@@ -0,0 +1 @@
+#include <asm-generic/vga.h>
diff --git a/arch/microblaze/include/asm/xor.h b/arch/microblaze/include/asm/xor.h
new file mode 100644
index 0000000..c82eb12
--- /dev/null
+++ b/arch/microblaze/include/asm/xor.h
@@ -0,0 +1 @@
+#include <asm-generic/xor.h>
diff --git a/arch/microblaze/include/uapi/asm/Kbuild b/arch/microblaze/include/uapi/asm/Kbuild
new file mode 100644
index 0000000..1aac99f
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/Kbuild
@@ -0,0 +1,36 @@
+# UAPI Header export list
+include include/uapi/asm-generic/Kbuild.asm
+
+generic-y += types.h
+
+header-y += auxvec.h
+header-y += bitsperlong.h
+header-y += byteorder.h
+header-y += elf.h
+header-y += errno.h
+header-y += fcntl.h
+header-y += ioctl.h
+header-y += ioctls.h
+header-y += ipcbuf.h
+header-y += kvm_para.h
+header-y += mman.h
+header-y += msgbuf.h
+header-y += param.h
+header-y += poll.h
+header-y += posix_types.h
+header-y += ptrace.h
+header-y += resource.h
+header-y += sembuf.h
+header-y += setup.h
+header-y += shmbuf.h
+header-y += sigcontext.h
+header-y += siginfo.h
+header-y += signal.h
+header-y += socket.h
+header-y += sockios.h
+header-y += stat.h
+header-y += statfs.h
+header-y += swab.h
+header-y += termbits.h
+header-y += termios.h
+header-y += unistd.h
diff --git a/arch/microblaze/include/uapi/asm/auxvec.h b/arch/microblaze/include/uapi/asm/auxvec.h
new file mode 100644
index 0000000..8b13789
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/auxvec.h
@@ -0,0 +1 @@
+
diff --git a/arch/microblaze/include/uapi/asm/bitsperlong.h b/arch/microblaze/include/uapi/asm/bitsperlong.h
new file mode 100644
index 0000000..6dc0bb0
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/bitsperlong.h
@@ -0,0 +1 @@
+#include <asm-generic/bitsperlong.h>
diff --git a/arch/microblaze/include/uapi/asm/byteorder.h b/arch/microblaze/include/uapi/asm/byteorder.h
new file mode 100644
index 0000000..3190276
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/byteorder.h
@@ -0,0 +1,10 @@
+#ifndef _ASM_MICROBLAZE_BYTEORDER_H
+#define _ASM_MICROBLAZE_BYTEORDER_H
+
+#ifdef __MICROBLAZEEL__
+#include <linux/byteorder/little_endian.h>
+#else
+#include <linux/byteorder/big_endian.h>
+#endif
+
+#endif /* _ASM_MICROBLAZE_BYTEORDER_H */
diff --git a/arch/microblaze/include/uapi/asm/elf.h b/arch/microblaze/include/uapi/asm/elf.h
new file mode 100644
index 0000000..e9bcdb6
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/elf.h
@@ -0,0 +1,122 @@
+/*
+ * 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.
+ */
+
+#ifndef _UAPI_ASM_MICROBLAZE_ELF_H
+#define _UAPI_ASM_MICROBLAZE_ELF_H
+
+#include <linux/elf-em.h>
+
+/*
+ * Note there is no "official" ELF designation for Microblaze.
+ * I've snaffled the value from the microblaze binutils source code
+ * /binutils/microblaze/include/elf/microblaze.h
+ */
+#define EM_MICROBLAZE_OLD	0xbaab
+#define ELF_ARCH		EM_MICROBLAZE
+
+/*
+ * This is used to ensure we don't load something for the wrong architecture.
+ */
+#define elf_check_arch(x)	((x)->e_machine == EM_MICROBLAZE \
+				 || (x)->e_machine == EM_MICROBLAZE_OLD)
+
+/*
+ * These are used to set parameters in the core dumps.
+ */
+#define ELF_CLASS	ELFCLASS32
+
+#ifndef __uClinux__
+
+/*
+ * ELF register definitions..
+ */
+
+#include <asm/ptrace.h>
+#include <asm/byteorder.h>
+
+#ifndef ELF_GREG_T
+#define ELF_GREG_T
+typedef unsigned long elf_greg_t;
+#endif
+
+#ifndef ELF_NGREG
+#define ELF_NGREG (sizeof(struct pt_regs) / sizeof(elf_greg_t))
+#endif
+
+#ifndef ELF_GREGSET_T
+#define ELF_GREGSET_T
+typedef elf_greg_t elf_gregset_t[ELF_NGREG];
+#endif
+
+#ifndef ELF_FPREGSET_T
+#define ELF_FPREGSET_T
+
+/* TBD */
+#define ELF_NFPREG	33	/* includes fsr */
+typedef unsigned long elf_fpreg_t;
+typedef elf_fpreg_t elf_fpregset_t[ELF_NFPREG];
+
+/* typedef struct user_fpu_struct elf_fpregset_t; */
+#endif
+
+/* This is the location that an ET_DYN program is loaded if exec'ed.  Typical
+ * use of this is to invoke "./ld.so someprog" to test out a new version of
+ * the loader.  We need to make sure that it is out of the way of the program
+ * that it will "exec", and that there is sufficient room for the brk.
+ */
+
+#define ELF_ET_DYN_BASE         (0x08000000)
+
+#ifdef __MICROBLAZEEL__
+#define ELF_DATA	ELFDATA2LSB
+#else
+#define ELF_DATA	ELFDATA2MSB
+#endif
+
+#define ELF_EXEC_PAGESIZE	PAGE_SIZE
+
+
+#define ELF_CORE_COPY_REGS(_dest, _regs)			\
+	memcpy((char *) &_dest, (char *) _regs,		\
+	sizeof(struct pt_regs));
+
+/* This yields a mask that user programs can use to figure out what
+ * instruction set this CPU supports.  This could be done in user space,
+ * but it's not easy, and we've already done it here.
+ */
+#define ELF_HWCAP	(0)
+
+/* This yields a string that ld.so will use to load implementation
+ * specific libraries for optimization.  This is more specific in
+ * intent than poking at uname or /proc/cpuinfo.
+
+ * For the moment, we have only optimizations for the Intel generations,
+ * but that could change...
+ */
+#define ELF_PLATFORM  (NULL)
+
+/* Added _f parameter. Is this definition correct: TBD */
+#define ELF_PLAT_INIT(_r, _f)				\
+do {							\
+	_r->r0 =  _r->r1 =  _r->r2 =  _r->r3 =		\
+	_r->r4 =  _r->r5 =  _r->r6 =  _r->r7 =		\
+	_r->r8 =  _r->r9 =  _r->r10 = _r->r11 =		\
+	_r->r12 = _r->r13 = _r->r14 = _r->r15 =		\
+	_r->r16 = _r->r17 = _r->r18 = _r->r19 =		\
+	_r->r20 = _r->r21 = _r->r22 = _r->r23 =		\
+	_r->r24 = _r->r25 = _r->r26 = _r->r27 =		\
+	_r->r28 = _r->r29 = _r->r30 = _r->r31 =		\
+	0;						\
+} while (0)
+
+
+#endif /* __uClinux__ */
+
+#endif /* _UAPI_ASM_MICROBLAZE_ELF_H */
diff --git a/arch/microblaze/include/uapi/asm/errno.h b/arch/microblaze/include/uapi/asm/errno.h
new file mode 100644
index 0000000..4c82b50
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/errno.h
@@ -0,0 +1 @@
+#include <asm-generic/errno.h>
diff --git a/arch/microblaze/include/uapi/asm/fcntl.h b/arch/microblaze/include/uapi/asm/fcntl.h
new file mode 100644
index 0000000..46ab12d
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/fcntl.h
@@ -0,0 +1 @@
+#include <asm-generic/fcntl.h>
diff --git a/arch/microblaze/include/uapi/asm/ioctl.h b/arch/microblaze/include/uapi/asm/ioctl.h
new file mode 100644
index 0000000..b279fe0
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/ioctl.h
@@ -0,0 +1 @@
+#include <asm-generic/ioctl.h>
diff --git a/arch/microblaze/include/uapi/asm/ioctls.h b/arch/microblaze/include/uapi/asm/ioctls.h
new file mode 100644
index 0000000..ec34c76
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/ioctls.h
@@ -0,0 +1 @@
+#include <asm-generic/ioctls.h>
diff --git a/arch/microblaze/include/uapi/asm/ipcbuf.h b/arch/microblaze/include/uapi/asm/ipcbuf.h
new file mode 100644
index 0000000..84c7e51
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/ipcbuf.h
@@ -0,0 +1 @@
+#include <asm-generic/ipcbuf.h>
diff --git a/arch/microblaze/include/uapi/asm/kvm_para.h b/arch/microblaze/include/uapi/asm/kvm_para.h
new file mode 100644
index 0000000..14fab8f
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/kvm_para.h
@@ -0,0 +1 @@
+#include <asm-generic/kvm_para.h>
diff --git a/arch/microblaze/include/uapi/asm/mman.h b/arch/microblaze/include/uapi/asm/mman.h
new file mode 100644
index 0000000..8eebf89
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/mman.h
@@ -0,0 +1 @@
+#include <asm-generic/mman.h>
diff --git a/arch/microblaze/include/uapi/asm/msgbuf.h b/arch/microblaze/include/uapi/asm/msgbuf.h
new file mode 100644
index 0000000..809134c
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/msgbuf.h
@@ -0,0 +1 @@
+#include <asm-generic/msgbuf.h>
diff --git a/arch/microblaze/include/uapi/asm/param.h b/arch/microblaze/include/uapi/asm/param.h
new file mode 100644
index 0000000..965d454
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/param.h
@@ -0,0 +1 @@
+#include <asm-generic/param.h>
diff --git a/arch/microblaze/include/uapi/asm/poll.h b/arch/microblaze/include/uapi/asm/poll.h
new file mode 100644
index 0000000..c98509d
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/poll.h
@@ -0,0 +1 @@
+#include <asm-generic/poll.h>
diff --git a/arch/microblaze/include/uapi/asm/posix_types.h b/arch/microblaze/include/uapi/asm/posix_types.h
new file mode 100644
index 0000000..0e15039
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/posix_types.h
@@ -0,0 +1,9 @@
+#ifndef _ASM_MICROBLAZE_POSIX_TYPES_H
+#define _ASM_MICROBLAZE_POSIX_TYPES_H
+
+typedef unsigned short	__kernel_mode_t;
+#define __kernel_mode_t __kernel_mode_t
+
+#include <asm-generic/posix_types.h>
+
+#endif /* _ASM_MICROBLAZE_POSIX_TYPES_H */
diff --git a/arch/microblaze/include/uapi/asm/ptrace.h b/arch/microblaze/include/uapi/asm/ptrace.h
new file mode 100644
index 0000000..d31238a
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/ptrace.h
@@ -0,0 +1,72 @@
+/*
+ * 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.
+ */
+
+#ifndef _UAPI_ASM_MICROBLAZE_PTRACE_H
+#define _UAPI_ASM_MICROBLAZE_PTRACE_H
+
+#ifndef __ASSEMBLY__
+
+typedef unsigned long microblaze_reg_t;
+
+struct pt_regs {
+	microblaze_reg_t r0;
+	microblaze_reg_t r1;
+	microblaze_reg_t r2;
+	microblaze_reg_t r3;
+	microblaze_reg_t r4;
+	microblaze_reg_t r5;
+	microblaze_reg_t r6;
+	microblaze_reg_t r7;
+	microblaze_reg_t r8;
+	microblaze_reg_t r9;
+	microblaze_reg_t r10;
+	microblaze_reg_t r11;
+	microblaze_reg_t r12;
+	microblaze_reg_t r13;
+	microblaze_reg_t r14;
+	microblaze_reg_t r15;
+	microblaze_reg_t r16;
+	microblaze_reg_t r17;
+	microblaze_reg_t r18;
+	microblaze_reg_t r19;
+	microblaze_reg_t r20;
+	microblaze_reg_t r21;
+	microblaze_reg_t r22;
+	microblaze_reg_t r23;
+	microblaze_reg_t r24;
+	microblaze_reg_t r25;
+	microblaze_reg_t r26;
+	microblaze_reg_t r27;
+	microblaze_reg_t r28;
+	microblaze_reg_t r29;
+	microblaze_reg_t r30;
+	microblaze_reg_t r31;
+	microblaze_reg_t pc;
+	microblaze_reg_t msr;
+	microblaze_reg_t ear;
+	microblaze_reg_t esr;
+	microblaze_reg_t fsr;
+	int pt_mode;
+};
+
+#ifndef __KERNEL__
+
+/* pt_regs offsets used by gdbserver etc in ptrace syscalls */
+#define PT_GPR(n)	((n) * sizeof(microblaze_reg_t))
+#define PT_PC		(32 * sizeof(microblaze_reg_t))
+#define PT_MSR		(33 * sizeof(microblaze_reg_t))
+#define PT_EAR		(34 * sizeof(microblaze_reg_t))
+#define PT_ESR		(35 * sizeof(microblaze_reg_t))
+#define PT_FSR		(36 * sizeof(microblaze_reg_t))
+#define PT_KERNEL_MODE	(37 * sizeof(microblaze_reg_t))
+
+#endif /* __KERNEL */
+
+#endif /* __ASSEMBLY__ */
+
+#endif /* _UAPI_ASM_MICROBLAZE_PTRACE_H */
diff --git a/arch/microblaze/include/uapi/asm/resource.h b/arch/microblaze/include/uapi/asm/resource.h
new file mode 100644
index 0000000..04bc4db
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/resource.h
@@ -0,0 +1 @@
+#include <asm-generic/resource.h>
diff --git a/arch/microblaze/include/uapi/asm/sembuf.h b/arch/microblaze/include/uapi/asm/sembuf.h
new file mode 100644
index 0000000..7673b83
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/sembuf.h
@@ -0,0 +1 @@
+#include <asm-generic/sembuf.h>
diff --git a/arch/microblaze/include/uapi/asm/setup.h b/arch/microblaze/include/uapi/asm/setup.h
new file mode 100644
index 0000000..76bc2ac
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/setup.h
@@ -0,0 +1,19 @@
+/*
+ * 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.
+ */
+
+#ifndef _UAPI_ASM_MICROBLAZE_SETUP_H
+#define _UAPI_ASM_MICROBLAZE_SETUP_H
+
+#define COMMAND_LINE_SIZE	256
+
+# ifndef __ASSEMBLY__
+
+# endif /* __ASSEMBLY__ */
+#endif /* _UAPI_ASM_MICROBLAZE_SETUP_H */
diff --git a/arch/microblaze/include/uapi/asm/shmbuf.h b/arch/microblaze/include/uapi/asm/shmbuf.h
new file mode 100644
index 0000000..83c05fc
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/shmbuf.h
@@ -0,0 +1 @@
+#include <asm-generic/shmbuf.h>
diff --git a/arch/microblaze/include/uapi/asm/sigcontext.h b/arch/microblaze/include/uapi/asm/sigcontext.h
new file mode 100644
index 0000000..55873c8
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/sigcontext.h
@@ -0,0 +1,20 @@
+/*
+ * 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.
+ */
+
+#ifndef _ASM_MICROBLAZE_SIGCONTEXT_H
+#define _ASM_MICROBLAZE_SIGCONTEXT_H
+
+/* FIXME should be linux/ptrace.h */
+#include <asm/ptrace.h>
+
+struct sigcontext {
+	struct pt_regs regs;
+	unsigned long oldmask;
+};
+
+#endif /* _ASM_MICROBLAZE_SIGCONTEXT_H */
diff --git a/arch/microblaze/include/uapi/asm/siginfo.h b/arch/microblaze/include/uapi/asm/siginfo.h
new file mode 100644
index 0000000..0815d29
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/siginfo.h
@@ -0,0 +1 @@
+#include <asm-generic/siginfo.h>
diff --git a/arch/microblaze/include/uapi/asm/signal.h b/arch/microblaze/include/uapi/asm/signal.h
new file mode 100644
index 0000000..7b1573c
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/signal.h
@@ -0,0 +1 @@
+#include <asm-generic/signal.h>
diff --git a/arch/microblaze/include/uapi/asm/socket.h b/arch/microblaze/include/uapi/asm/socket.h
new file mode 100644
index 0000000..6b71384
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/socket.h
@@ -0,0 +1 @@
+#include <asm-generic/socket.h>
diff --git a/arch/microblaze/include/uapi/asm/sockios.h b/arch/microblaze/include/uapi/asm/sockios.h
new file mode 100644
index 0000000..def6d47
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/sockios.h
@@ -0,0 +1 @@
+#include <asm-generic/sockios.h>
diff --git a/arch/microblaze/include/uapi/asm/stat.h b/arch/microblaze/include/uapi/asm/stat.h
new file mode 100644
index 0000000..3dc90fa
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/stat.h
@@ -0,0 +1 @@
+#include <asm-generic/stat.h>
diff --git a/arch/microblaze/include/uapi/asm/statfs.h b/arch/microblaze/include/uapi/asm/statfs.h
new file mode 100644
index 0000000..0b91fe1
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/statfs.h
@@ -0,0 +1 @@
+#include <asm-generic/statfs.h>
diff --git a/arch/microblaze/include/uapi/asm/swab.h b/arch/microblaze/include/uapi/asm/swab.h
new file mode 100644
index 0000000..7847e56
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/swab.h
@@ -0,0 +1 @@
+#include <asm-generic/swab.h>
diff --git a/arch/microblaze/include/uapi/asm/termbits.h b/arch/microblaze/include/uapi/asm/termbits.h
new file mode 100644
index 0000000..3935b10
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/termbits.h
@@ -0,0 +1 @@
+#include <asm-generic/termbits.h>
diff --git a/arch/microblaze/include/uapi/asm/termios.h b/arch/microblaze/include/uapi/asm/termios.h
new file mode 100644
index 0000000..280d78a
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/termios.h
@@ -0,0 +1 @@
+#include <asm-generic/termios.h>
diff --git a/arch/microblaze/include/uapi/asm/unistd.h b/arch/microblaze/include/uapi/asm/unistd.h
new file mode 100644
index 0000000..32850c7
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/unistd.h
@@ -0,0 +1,408 @@
+/*
+ * Copyright (C) 2007-2008 Michal Simek <monstr@monstr.eu>
+ * 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.
+ */
+
+#ifndef _UAPI_ASM_MICROBLAZE_UNISTD_H
+#define _UAPI_ASM_MICROBLAZE_UNISTD_H
+
+#define __NR_restart_syscall	0 /* ok */
+#define __NR_exit		1 /* ok */
+#define __NR_fork		2 /* not for no MMU - weird */
+#define __NR_read		3 /* ok */
+#define __NR_write		4 /* ok */
+#define __NR_open		5 /* openat */
+#define __NR_close		6 /* ok */
+#define __NR_waitpid		7 /* waitid */
+#define __NR_creat		8 /* openat */
+#define __NR_link		9 /* linkat */
+#define __NR_unlink		10 /* unlinkat */
+#define __NR_execve		11 /* ok */
+#define __NR_chdir		12 /* ok */
+#define __NR_time		13 /* obsolete -> sys_gettimeofday */
+#define __NR_mknod		14 /* mknodat */
+#define __NR_chmod		15 /* fchmodat */
+#define __NR_lchown		16 /* ok */
+#define __NR_break		17 /* don't know */
+#define __NR_oldstat		18 /* remove */
+#define __NR_lseek		19 /* ok */
+#define __NR_getpid		20 /* ok */
+#define __NR_mount		21 /* ok */
+#define __NR_umount		22 /* ok */  /* use only umount2 */
+#define __NR_setuid		23 /* ok */
+#define __NR_getuid		24 /* ok */
+#define __NR_stime		25 /* obsolete -> sys_settimeofday */
+#define __NR_ptrace		26 /* ok */
+#define __NR_alarm		27 /* obsolete -> sys_setitimer */
+#define __NR_oldfstat		28 /* remove */
+#define __NR_pause		29 /* obsolete -> sys_rt_sigtimedwait */
+#define __NR_utime		30 /* obsolete -> sys_utimesat */
+#define __NR_stty		31 /* remove */
+#define __NR_gtty		32 /* remove */
+#define __NR_access		33 /* faccessat */
+/* can be implemented by sys_setpriority */
+#define __NR_nice		34
+#define __NR_ftime		35 /* remove */
+#define __NR_sync		36 /* ok */
+#define __NR_kill		37 /* ok */
+#define __NR_rename		38 /* renameat */
+#define __NR_mkdir		39 /* mkdirat */
+#define __NR_rmdir		40 /* unlinkat */
+#define __NR_dup		41 /* ok */
+#define __NR_pipe		42 /* ok */
+#define __NR_times		43 /* ok */
+#define __NR_prof		44 /* remove */
+#define __NR_brk		45 /* ok -mmu, nommu specific */
+#define __NR_setgid		46 /* ok */
+#define __NR_getgid		47 /* ok */
+#define __NR_signal		48 /* obsolete -> sys_rt_sigaction */
+#define __NR_geteuid		49 /* ok */
+#define __NR_getegid		50 /* ok */
+#define __NR_acct		51 /* add it and then I can disable it */
+#define __NR_umount2		52 /* remove */
+#define __NR_lock		53 /* remove */
+#define __NR_ioctl		54 /* ok */
+#define __NR_fcntl		55 /* ok -> 64bit version*/
+#define __NR_mpx		56 /* remove */
+#define __NR_setpgid		57 /* ok */
+#define __NR_ulimit		58 /* remove */
+#define __NR_oldolduname	59 /* remove */
+#define __NR_umask		60 /* ok */
+#define __NR_chroot		61 /* ok */
+#define __NR_ustat		62 /* obsolete -> statfs64 */
+#define __NR_dup2		63 /* ok */
+#define __NR_getppid		64 /* ok */
+#define __NR_getpgrp		65 /* obsolete -> sys_getpgid */
+#define __NR_setsid		66 /* ok */
+#define __NR_sigaction		67 /* obsolete -> rt_sigaction */
+#define __NR_sgetmask		68 /* obsolete -> sys_rt_sigprocmask */
+#define __NR_ssetmask		69 /* obsolete ->sys_rt_sigprocmask */
+#define __NR_setreuid		70 /* ok */
+#define __NR_setregid		71 /* ok */
+#define __NR_sigsuspend		72 /* obsolete -> rt_sigsuspend */
+#define __NR_sigpending		73 /* obsolete -> sys_rt_sigpending */
+#define __NR_sethostname	74 /* ok */
+#define __NR_setrlimit		75 /* ok */
+#define __NR_getrlimit		76 /* ok Back compatible 2G limited rlimit */
+#define __NR_getrusage		77 /* ok */
+#define __NR_gettimeofday	78 /* ok */
+#define __NR_settimeofday	79 /* ok */
+#define __NR_getgroups		80 /* ok */
+#define __NR_setgroups		81 /* ok */
+#define __NR_select		82 /* obsolete -> sys_pselect6 */
+#define __NR_symlink		83 /* symlinkat */
+#define __NR_oldlstat		84 /* remove */
+#define __NR_readlink		85 /* obsolete -> sys_readlinkat */
+#define __NR_uselib		86 /* remove */
+#define __NR_swapon		87 /* ok */
+#define __NR_reboot		88 /* ok */
+#define __NR_readdir		89 /* remove ? */
+#define __NR_mmap		90 /* obsolete -> sys_mmap2 */
+#define __NR_munmap		91 /* ok - mmu and nommu */
+#define __NR_truncate		92 /* ok or truncate64 */
+#define __NR_ftruncate		93 /* ok or ftruncate64 */
+#define __NR_fchmod		94 /* ok */
+#define __NR_fchown		95 /* ok */
+#define __NR_getpriority	96 /* ok */
+#define __NR_setpriority	97 /* ok */
+#define __NR_profil		98 /* remove */
+#define __NR_statfs		99 /* ok or statfs64 */
+#define __NR_fstatfs		100  /* ok or fstatfs64 */
+#define __NR_ioperm		101 /* remove */
+#define __NR_socketcall		102 /* remove */
+#define __NR_syslog		103 /* ok */
+#define __NR_setitimer		104 /* ok */
+#define __NR_getitimer		105 /* ok */
+#define __NR_stat		106 /* remove */
+#define __NR_lstat		107 /* remove */
+#define __NR_fstat		108 /* remove */
+#define __NR_olduname		109 /* remove */
+#define __NR_iopl		110 /* remove */
+#define __NR_vhangup		111 /* ok */
+#define __NR_idle		112 /* remove */
+#define __NR_vm86old		113 /* remove */
+#define __NR_wait4		114 /* obsolete -> waitid */
+#define __NR_swapoff		115 /* ok */
+#define __NR_sysinfo		116 /* ok */
+#define __NR_ipc		117 /* remove - direct call */
+#define __NR_fsync		118 /* ok */
+#define __NR_sigreturn		119 /* obsolete -> sys_rt_sigreturn */
+#define __NR_clone		120 /* ok */
+#define __NR_setdomainname	121 /* ok */
+#define __NR_uname		122 /* remove */
+#define __NR_modify_ldt		123 /* remove */
+#define __NR_adjtimex		124 /* ok */
+#define __NR_mprotect		125 /* remove */
+#define __NR_sigprocmask	126 /* obsolete -> sys_rt_sigprocmask */
+#define __NR_create_module	127 /* remove */
+#define __NR_init_module	128 /* ok */
+#define __NR_delete_module	129 /* ok */
+#define __NR_get_kernel_syms	130 /* remove */
+#define __NR_quotactl		131 /* ok */
+#define __NR_getpgid		132 /* ok */
+#define __NR_fchdir		133 /* ok */
+#define __NR_bdflush		134 /* remove */
+#define __NR_sysfs		135 /* needed for busybox */
+#define __NR_personality	136 /* ok */
+#define __NR_afs_syscall	137 /* Syscall for Andrew File System */
+#define __NR_setfsuid		138 /* ok */
+#define __NR_setfsgid		139 /* ok */
+#define __NR__llseek		140 /* remove only lseek */
+#define __NR_getdents		141 /* ok or getdents64 */
+#define __NR__newselect		142 /* remove */
+#define __NR_flock		143 /* ok */
+#define __NR_msync		144 /* remove */
+#define __NR_readv		145 /* ok */
+#define __NR_writev		146 /* ok */
+#define __NR_getsid		147 /* ok */
+#define __NR_fdatasync		148 /* ok */
+#define __NR__sysctl		149 /* remove */
+#define __NR_mlock		150 /* ok - nommu or mmu */
+#define __NR_munlock		151 /* ok - nommu or mmu */
+#define __NR_mlockall		152 /* ok - nommu or mmu */
+#define __NR_munlockall		153 /* ok - nommu or mmu */
+#define __NR_sched_setparam		154 /* ok */
+#define __NR_sched_getparam		155 /* ok */
+#define __NR_sched_setscheduler		156 /* ok */
+#define __NR_sched_getscheduler		157 /* ok */
+#define __NR_sched_yield		158 /* ok */
+#define __NR_sched_get_priority_max	159 /* ok */
+#define __NR_sched_get_priority_min	160 /* ok */
+#define __NR_sched_rr_get_interval	161 /* ok */
+#define __NR_nanosleep		162 /* ok */
+#define __NR_mremap		163 /* ok - nommu or mmu */
+#define __NR_setresuid		164 /* ok */
+#define __NR_getresuid		165 /* ok */
+#define __NR_vm86		166 /* remove */
+#define __NR_query_module	167 /* ok */
+#define __NR_poll		168 /* obsolete -> sys_ppoll */
+#define __NR_nfsservctl		169 /* ok */
+#define __NR_setresgid		170 /* ok */
+#define __NR_getresgid		171 /* ok */
+#define __NR_prctl		172 /* ok */
+#define __NR_rt_sigreturn	173 /* ok */
+#define __NR_rt_sigaction	174 /* ok */
+#define __NR_rt_sigprocmask	175 /* ok */
+#define __NR_rt_sigpending	176 /* ok */
+#define __NR_rt_sigtimedwait	177 /* ok */
+#define __NR_rt_sigqueueinfo	178 /* ok */
+#define __NR_rt_sigsuspend	179 /* ok */
+#define __NR_pread64		180 /* ok */
+#define __NR_pwrite64		181 /* ok */
+#define __NR_chown		182 /* obsolete -> fchownat */
+#define __NR_getcwd		183 /* ok */
+#define __NR_capget		184 /* ok */
+#define __NR_capset		185 /* ok */
+#define __NR_sigaltstack	186 /* remove */
+#define __NR_sendfile		187 /* ok -> exist 64bit version*/
+#define __NR_getpmsg		188 /* remove */
+/* remove - some people actually want streams */
+#define __NR_putpmsg		189
+/* for noMMU - group with clone -> maybe remove */
+#define __NR_vfork		190
+#define __NR_ugetrlimit		191 /* remove - SuS compliant getrlimit */
+#define __NR_mmap2		192 /* ok */
+#define __NR_truncate64		193 /* ok */
+#define __NR_ftruncate64	194 /* ok */
+#define __NR_stat64		195 /* remove _ARCH_WANT_STAT64 */
+#define __NR_lstat64		196 /* remove _ARCH_WANT_STAT64 */
+#define __NR_fstat64		197 /* remove _ARCH_WANT_STAT64 */
+#define __NR_lchown32		198 /* ok - without 32 */
+#define __NR_getuid32		199 /* ok - without 32 */
+#define __NR_getgid32		200 /* ok - without 32 */
+#define __NR_geteuid32		201 /* ok - without 32 */
+#define __NR_getegid32		202 /* ok - without 32 */
+#define __NR_setreuid32		203 /* ok - without 32 */
+#define __NR_setregid32		204 /* ok - without 32 */
+#define __NR_getgroups32	205 /* ok - without 32 */
+#define __NR_setgroups32	206 /* ok - without 32 */
+#define __NR_fchown32		207 /* ok - without 32 */
+#define __NR_setresuid32	208 /* ok - without 32 */
+#define __NR_getresuid32	209 /* ok - without 32 */
+#define __NR_setresgid32	210 /* ok - without 32 */
+#define __NR_getresgid32	211 /* ok - without 32 */
+#define __NR_chown32		212 /* ok - without 32 -obsolete -> fchownat */
+#define __NR_setuid32		213 /* ok - without 32 */
+#define __NR_setgid32		214 /* ok - without 32 */
+#define __NR_setfsuid32		215 /* ok - without 32 */
+#define __NR_setfsgid32		216 /* ok - without 32 */
+#define __NR_pivot_root		217 /* ok */
+#define __NR_mincore		218 /* ok */
+#define __NR_madvise		219 /* ok */
+#define __NR_getdents64		220 /* ok */
+#define __NR_fcntl64		221 /* ok */
+/* 223 is unused */
+#define __NR_gettid		224 /* ok */
+#define __NR_readahead		225 /* ok */
+#define __NR_setxattr		226 /* ok */
+#define __NR_lsetxattr		227 /* ok */
+#define __NR_fsetxattr		228 /* ok */
+#define __NR_getxattr		229 /* ok */
+#define __NR_lgetxattr		230 /* ok */
+#define __NR_fgetxattr		231 /* ok */
+#define __NR_listxattr		232 /* ok */
+#define __NR_llistxattr		233 /* ok */
+#define __NR_flistxattr		234 /* ok */
+#define __NR_removexattr	235 /* ok */
+#define __NR_lremovexattr	236 /* ok */
+#define __NR_fremovexattr	237 /* ok */
+#define __NR_tkill		238 /* ok */
+#define __NR_sendfile64		239 /* ok */
+#define __NR_futex		240 /* ok */
+#define __NR_sched_setaffinity	241 /* ok */
+#define __NR_sched_getaffinity	242 /* ok */
+#define __NR_set_thread_area	243 /* remove */
+#define __NR_get_thread_area	244 /* remove */
+#define __NR_io_setup		245 /* ok */
+#define __NR_io_destroy		246 /* ok */
+#define __NR_io_getevents	247 /* ok */
+#define __NR_io_submit		248 /* ok */
+#define __NR_io_cancel		249 /* ok */
+#define __NR_fadvise64		250 /* remove -> sys_fadvise64_64 */
+/* 251 is available for reuse (was briefly sys_set_zone_reclaim) */
+#define __NR_exit_group		252 /* ok */
+#define __NR_lookup_dcookie	253 /* ok */
+#define __NR_epoll_create	254 /* ok */
+#define __NR_epoll_ctl		255 /* ok */
+#define __NR_epoll_wait		256 /* obsolete -> sys_epoll_pwait */
+#define __NR_remap_file_pages	257 /* only for mmu */
+#define __NR_set_tid_address	258 /* ok */
+#define __NR_timer_create	259 /* ok */
+#define __NR_timer_settime	(__NR_timer_create+1) /* 260 */ /* ok */
+#define __NR_timer_gettime	(__NR_timer_create+2) /* 261 */ /* ok */
+#define __NR_timer_getoverrun	(__NR_timer_create+3) /* 262 */ /* ok */
+#define __NR_timer_delete	(__NR_timer_create+4) /* 263 */ /* ok */
+#define __NR_clock_settime	(__NR_timer_create+5) /* 264 */ /* ok */
+#define __NR_clock_gettime	(__NR_timer_create+6) /* 265 */ /* ok */
+#define __NR_clock_getres	(__NR_timer_create+7) /* 266 */ /* ok */
+#define __NR_clock_nanosleep	(__NR_timer_create+8) /* 267 */ /* ok */
+#define __NR_statfs64		268 /* ok */
+#define __NR_fstatfs64		269 /* ok */
+#define __NR_tgkill		270 /* ok */
+#define __NR_utimes		271 /* obsolete -> sys_futimesat */
+#define __NR_fadvise64_64	272 /* ok */
+#define __NR_vserver		273 /* ok */
+#define __NR_mbind		274 /* only for mmu */
+#define __NR_get_mempolicy	275 /* only for mmu */
+#define __NR_set_mempolicy	276 /* only for mmu */
+#define __NR_mq_open		277 /* ok */
+#define __NR_mq_unlink		(__NR_mq_open+1) /* 278 */ /* ok */
+#define __NR_mq_timedsend	(__NR_mq_open+2) /* 279 */ /* ok */
+#define __NR_mq_timedreceive	(__NR_mq_open+3) /* 280 */ /* ok */
+#define __NR_mq_notify		(__NR_mq_open+4) /* 281 */ /* ok */
+#define __NR_mq_getsetattr	(__NR_mq_open+5) /* 282 */ /* ok */
+#define __NR_kexec_load		283 /* ok */
+#define __NR_waitid		284 /* ok */
+/* #define __NR_sys_setaltroot	285 */
+#define __NR_add_key		286 /* ok */
+#define __NR_request_key	287 /* ok */
+#define __NR_keyctl		288 /* ok */
+#define __NR_ioprio_set		289 /* ok */
+#define __NR_ioprio_get		290 /* ok */
+#define __NR_inotify_init	291 /* ok */
+#define __NR_inotify_add_watch	292 /* ok */
+#define __NR_inotify_rm_watch	293 /* ok */
+#define __NR_migrate_pages	294 /* mmu */
+#define __NR_openat		295 /* ok */
+#define __NR_mkdirat		296 /* ok */
+#define __NR_mknodat		297 /* ok */
+#define __NR_fchownat		298 /* ok */
+#define __NR_futimesat		299 /* obsolete -> sys_utimesat */
+#define __NR_fstatat64		300 /* stat64 */
+#define __NR_unlinkat		301 /* ok */
+#define __NR_renameat		302 /* ok */
+#define __NR_linkat		303 /* ok */
+#define __NR_symlinkat		304 /* ok */
+#define __NR_readlinkat		305 /* ok */
+#define __NR_fchmodat		306 /* ok */
+#define __NR_faccessat		307 /* ok */
+#define __NR_pselect6		308 /* ok */
+#define __NR_ppoll		309 /* ok */
+#define __NR_unshare		310 /* ok */
+#define __NR_set_robust_list	311 /* ok */
+#define __NR_get_robust_list	312 /* ok */
+#define __NR_splice		313 /* ok */
+#define __NR_sync_file_range	314 /* ok */
+#define __NR_tee		315 /* ok */
+#define __NR_vmsplice		316 /* ok */
+#define __NR_move_pages		317 /* mmu */
+#define __NR_getcpu		318 /* ok */
+#define __NR_epoll_pwait	319 /* ok */
+#define __NR_utimensat		320 /* ok */
+#define __NR_signalfd		321 /* ok */
+#define __NR_timerfd_create	322 /* ok */
+#define __NR_eventfd		323 /* ok */
+#define __NR_fallocate		324 /* ok */
+#define __NR_semtimedop		325 /* ok - semaphore group */
+#define __NR_timerfd_settime	326 /* ok */
+#define __NR_timerfd_gettime	327 /* ok */
+/* sysv ipc syscalls */
+#define __NR_semctl		328 /* ok */
+#define __NR_semget		329 /* ok */
+#define __NR_semop		330 /* ok */
+#define __NR_msgctl		331 /* ok */
+#define __NR_msgget		332 /* ok */
+#define __NR_msgrcv		333 /* ok */
+#define __NR_msgsnd		334 /* ok */
+#define __NR_shmat		335 /* ok */
+#define __NR_shmctl		336 /* ok */
+#define __NR_shmdt		337 /* ok */
+#define __NR_shmget		338 /* ok */
+
+
+#define __NR_signalfd4		339 /* new */
+#define __NR_eventfd2		340 /* new */
+#define __NR_epoll_create1	341 /* new */
+#define __NR_dup3		342 /* new */
+#define __NR_pipe2		343 /* new */
+#define __NR_inotify_init1	344 /* new */
+#define __NR_socket		345 /* new */
+#define __NR_socketpair		346 /* new */
+#define __NR_bind		347 /* new */
+#define __NR_listen		348 /* new */
+#define __NR_accept		349 /* new */
+#define __NR_connect		350 /* new */
+#define __NR_getsockname	351 /* new */
+#define __NR_getpeername	352 /* new */
+#define __NR_sendto		353 /* new */
+#define __NR_send		354 /* new */
+#define __NR_recvfrom		355 /* new */
+#define __NR_recv		356 /* new */
+#define __NR_setsockopt		357 /* new */
+#define __NR_getsockopt		358 /* new */
+#define __NR_shutdown		359 /* new */
+#define __NR_sendmsg		360 /* new */
+#define __NR_recvmsg		361 /* new */
+#define __NR_accept4		362 /* new */
+#define __NR_preadv		363 /* new */
+#define __NR_pwritev		364 /* new */
+#define __NR_rt_tgsigqueueinfo	365 /* new */
+#define __NR_perf_event_open	366 /* new */
+#define __NR_recvmmsg		367 /* new */
+#define __NR_fanotify_init	368
+#define __NR_fanotify_mark	369
+#define __NR_prlimit64		370
+#define __NR_name_to_handle_at	371
+#define __NR_open_by_handle_at	372
+#define __NR_clock_adjtime	373
+#define __NR_syncfs		374
+#define __NR_setns		375
+#define __NR_sendmmsg		376
+#define __NR_process_vm_readv	377
+#define __NR_process_vm_writev	378
+#define __NR_kcmp		379
+#define __NR_finit_module	380
+#define __NR_sched_setattr	381
+#define __NR_sched_getattr	382
+#define __NR_renameat2		383
+#define __NR_seccomp		384
+#define __NR_getrandom		385
+#define __NR_memfd_create	386
+#define __NR_bpf		387
+#define __NR_execveat		388
+
+#endif /* _UAPI_ASM_MICROBLAZE_UNISTD_H */
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
+}
diff --git a/arch/microblaze/lib/Makefile b/arch/microblaze/lib/Makefile
new file mode 100644
index 0000000..70c7ae6
--- /dev/null
+++ b/arch/microblaze/lib/Makefile
@@ -0,0 +1,23 @@
+#
+# Makefile
+#
+
+ifdef CONFIG_FUNCTION_TRACER
+CFLAGS_REMOVE_ashldi3.o = -pg
+CFLAGS_REMOVE_ashrdi3.o = -pg
+CFLAGS_REMOVE_lshrdi3.o = -pg
+endif
+
+lib-y :=  memset.o
+
+ifeq ($(CONFIG_OPT_LIB_ASM),y)
+lib-y += fastcopy.o
+else
+lib-y += memcpy.o memmove.o
+endif
+
+lib-y += uaccess_old.o
+
+# libgcc-style stuff needed in the kernel
+obj-y += ashldi3.o ashrdi3.o cmpdi2.o divsi3.o lshrdi3.o modsi3.o
+obj-y += muldi3.o mulsi3.o ucmpdi2.o udivsi3.o umodsi3.o
diff --git a/arch/microblaze/lib/ashldi3.c b/arch/microblaze/lib/ashldi3.c
new file mode 100644
index 0000000..1af904c
--- /dev/null
+++ b/arch/microblaze/lib/ashldi3.c
@@ -0,0 +1,28 @@
+#include <linux/export.h>
+
+#include "libgcc.h"
+
+long long __ashldi3(long long u, word_type b)
+{
+	DWunion uu, w;
+	word_type bm;
+
+	if (b == 0)
+		return u;
+
+	uu.ll = u;
+	bm = 32 - b;
+
+	if (bm <= 0) {
+		w.s.low = 0;
+		w.s.high = (unsigned int) uu.s.low << -bm;
+	} else {
+		const unsigned int carries = (unsigned int) uu.s.low >> bm;
+
+		w.s.low = (unsigned int) uu.s.low << b;
+		w.s.high = ((unsigned int) uu.s.high << b) | carries;
+	}
+
+	return w.ll;
+}
+EXPORT_SYMBOL(__ashldi3);
diff --git a/arch/microblaze/lib/ashrdi3.c b/arch/microblaze/lib/ashrdi3.c
new file mode 100644
index 0000000..32c334c
--- /dev/null
+++ b/arch/microblaze/lib/ashrdi3.c
@@ -0,0 +1,30 @@
+#include <linux/export.h>
+
+#include "libgcc.h"
+
+long long __ashrdi3(long long u, word_type b)
+{
+	DWunion uu, w;
+	word_type bm;
+
+	if (b == 0)
+		return u;
+
+	uu.ll = u;
+	bm = 32 - b;
+
+	if (bm <= 0) {
+		/* w.s.high = 1..1 or 0..0 */
+		w.s.high =
+		    uu.s.high >> 31;
+		w.s.low = uu.s.high >> -bm;
+	} else {
+		const unsigned int carries = (unsigned int) uu.s.high << bm;
+
+		w.s.high = uu.s.high >> b;
+		w.s.low = ((unsigned int) uu.s.low >> b) | carries;
+	}
+
+	return w.ll;
+}
+EXPORT_SYMBOL(__ashrdi3);
diff --git a/arch/microblaze/lib/cmpdi2.c b/arch/microblaze/lib/cmpdi2.c
new file mode 100644
index 0000000..67abc9a
--- /dev/null
+++ b/arch/microblaze/lib/cmpdi2.c
@@ -0,0 +1,26 @@
+#include <linux/export.h>
+
+#include "libgcc.h"
+
+word_type __cmpdi2(long long a, long long b)
+{
+	const DWunion au = {
+		.ll = a
+	};
+	const DWunion bu = {
+		.ll = b
+	};
+
+	if (au.s.high < bu.s.high)
+		return 0;
+	else if (au.s.high > bu.s.high)
+		return 2;
+
+	if ((unsigned int) au.s.low < (unsigned int) bu.s.low)
+		return 0;
+	else if ((unsigned int) au.s.low > (unsigned int) bu.s.low)
+		return 2;
+
+	return 1;
+}
+EXPORT_SYMBOL(__cmpdi2);
diff --git a/arch/microblaze/lib/divsi3.S b/arch/microblaze/lib/divsi3.S
new file mode 100644
index 0000000..595b02d
--- /dev/null
+++ b/arch/microblaze/lib/divsi3.S
@@ -0,0 +1,73 @@
+#include <linux/linkage.h>
+
+/*
+* Divide operation for 32 bit integers.
+*	Input :	Dividend in Reg r5
+*		Divisor in Reg r6
+*	Output: Result in Reg r3
+*/
+	.text
+	.globl	__divsi3
+	.type __divsi3, @function
+	.ent __divsi3
+__divsi3:
+	.frame	r1, 0, r15
+
+	addik	r1, r1, -16
+	swi	r28, r1, 0
+	swi	r29, r1, 4
+	swi	r30, r1, 8
+	swi	r31, r1, 12
+
+	beqi	r6, div_by_zero /* div_by_zero - division error */
+	beqi	r5, result_is_zero /* result is zero */
+	bgeid	r5, r5_pos
+	xor	r28, r5, r6 /* get the sign of the result */
+	rsubi	r5, r5, 0 /* make r5 positive */
+r5_pos:
+	bgei	r6, r6_pos
+	rsubi	r6, r6, 0 /* make r6 positive */
+r6_pos:
+	addik	r30, r0, 0 /* clear mod */
+	addik	r3, r0, 0 /* clear div */
+	addik	r29, r0, 32 /* initialize the loop count */
+
+	/* first part try to find the first '1' in the r5 */
+div0:
+	blti	r5, div2 /* this traps r5 == 0x80000000 */
+div1:
+	add	r5, r5, r5 /* left shift logical r5 */
+	bgtid	r5, div1
+	addik	r29, r29, -1
+div2:
+	/* left shift logical r5 get the '1' into the carry */
+	add	r5, r5, r5
+	addc	r30, r30, r30 /* move that bit into the mod register */
+	rsub	r31, r6, r30 /* try to subtract (r30 a r6) */
+	blti	r31, mod_too_small
+	/* move the r31 to mod since the result was positive */
+	or	r30, r0, r31
+	addik	r3, r3, 1
+mod_too_small:
+	addik	r29, r29, -1
+	beqi	r29, loop_end
+	add	r3, r3, r3 /* shift in the '1' into div */
+	bri	div2 /* div2 */
+loop_end:
+	bgei	r28, return_here
+	brid	return_here
+	rsubi	r3, r3, 0 /* negate the result */
+div_by_zero:
+result_is_zero:
+	or	r3, r0, r0 /* set result to 0 */
+return_here:
+/* restore values of csrs and that of r3 and the divisor and the dividend */
+	lwi	r28, r1, 0
+	lwi	r29, r1, 4
+	lwi	r30, r1, 8
+	lwi	r31, r1, 12
+	rtsd	r15, 8
+	addik	r1, r1, 16
+
+.size __divsi3, . - __divsi3
+.end __divsi3
diff --git a/arch/microblaze/lib/fastcopy.S b/arch/microblaze/lib/fastcopy.S
new file mode 100644
index 0000000..62021d7
--- /dev/null
+++ b/arch/microblaze/lib/fastcopy.S
@@ -0,0 +1,670 @@
+/*
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ * Copyright (C) 2008 Jim Law - Iris LP  All rights reserved.
+ *
+ * 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 Jim Law <jlaw@irispower.com>
+ *
+ * intended to replace:
+ *	memcpy in memcpy.c and
+ *	memmove in memmove.c
+ * ... in arch/microblaze/lib
+ *
+ *
+ * assly_fastcopy.S
+ *
+ * Attempt at quicker memcpy and memmove for MicroBlaze
+ *	Input :	Operand1 in Reg r5 - destination address
+ *		Operand2 in Reg r6 - source address
+ *		Operand3 in Reg r7 - number of bytes to transfer
+ *	Output: Result in Reg r3 - starting destinaition address
+ *
+ *
+ * Explanation:
+ *	Perform (possibly unaligned) copy of a block of memory
+ *	between mem locations with size of xfer spec'd in bytes
+ */
+
+#ifdef __MICROBLAZEEL__
+#error Microblaze LE not support ASM optimized lib func. Disable OPT_LIB_ASM.
+#endif
+
+#include <linux/linkage.h>
+	.text
+	.globl	memcpy
+	.type  memcpy, @function
+	.ent	memcpy
+
+memcpy:
+fast_memcpy_ascending:
+	/* move d to return register as value of function */
+	addi	r3, r5, 0
+
+	addi	r4, r0, 4	/* n = 4 */
+	cmpu	r4, r4, r7	/* n = c - n  (unsigned) */
+	blti	r4, a_xfer_end	/* if n < 0, less than one word to transfer */
+
+	/* transfer first 0~3 bytes to get aligned dest address */
+	andi	r4, r5, 3		/* n = d & 3 */
+	/* if zero, destination already aligned */
+	beqi	r4, a_dalign_done
+	/* n = 4 - n (yields 3, 2, 1 transfers for 1, 2, 3 addr offset) */
+	rsubi	r4, r4, 4
+	rsub	r7, r4, r7		/* c = c - n adjust c */
+
+a_xfer_first_loop:
+	/* if no bytes left to transfer, transfer the bulk */
+	beqi	r4, a_dalign_done
+	lbui	r11, r6, 0		/* h = *s */
+	sbi	r11, r5, 0		/* *d = h */
+	addi	r6, r6, 1		/* s++ */
+	addi	r5, r5, 1		/* d++ */
+	brid	a_xfer_first_loop	/* loop */
+	addi	r4, r4, -1		/* n-- (IN DELAY SLOT) */
+
+a_dalign_done:
+	addi	r4, r0, 32		/* n = 32 */
+	cmpu	r4, r4, r7		/* n = c - n  (unsigned) */
+	/* if n < 0, less than one block to transfer */
+	blti	r4, a_block_done
+
+a_block_xfer:
+	andi	r4, r7, 0xffffffe0	/* n = c & ~31 */
+	rsub	r7, r4, r7		/* c = c - n */
+
+	andi	r9, r6, 3		/* t1 = s & 3 */
+	/* if temp != 0, unaligned transfers needed */
+	bnei	r9, a_block_unaligned
+
+a_block_aligned:
+	lwi	r9, r6, 0		/* t1 = *(s + 0) */
+	lwi	r10, r6, 4		/* t2 = *(s + 4) */
+	lwi	r11, r6, 8		/* t3 = *(s + 8) */
+	lwi	r12, r6, 12		/* t4 = *(s + 12) */
+	swi	r9, r5, 0		/* *(d + 0) = t1 */
+	swi	r10, r5, 4		/* *(d + 4) = t2 */
+	swi	r11, r5, 8		/* *(d + 8) = t3 */
+	swi	r12, r5, 12		/* *(d + 12) = t4 */
+	lwi	r9, r6, 16		/* t1 = *(s + 16) */
+	lwi	r10, r6, 20		/* t2 = *(s + 20) */
+	lwi	r11, r6, 24		/* t3 = *(s + 24) */
+	lwi	r12, r6, 28		/* t4 = *(s + 28) */
+	swi	r9, r5, 16		/* *(d + 16) = t1 */
+	swi	r10, r5, 20		/* *(d + 20) = t2 */
+	swi	r11, r5, 24		/* *(d + 24) = t3 */
+	swi	r12, r5, 28		/* *(d + 28) = t4 */
+	addi	r6, r6, 32		/* s = s + 32 */
+	addi	r4, r4, -32		/* n = n - 32 */
+	bneid	r4, a_block_aligned	/* while (n) loop */
+	addi	r5, r5, 32		/* d = d + 32 (IN DELAY SLOT) */
+	bri	a_block_done
+
+a_block_unaligned:
+	andi	r8, r6, 0xfffffffc	/* as = s & ~3 */
+	add	r6, r6, r4		/* s = s + n */
+	lwi	r11, r8, 0		/* h = *(as + 0) */
+
+	addi	r9, r9, -1
+	beqi	r9, a_block_u1		/* t1 was 1 => 1 byte offset */
+	addi	r9, r9, -1
+	beqi	r9, a_block_u2		/* t1 was 2 => 2 byte offset */
+
+a_block_u3:
+	bslli	r11, r11, 24	/* h = h << 24 */
+a_bu3_loop:
+	lwi	r12, r8, 4	/* v = *(as + 4) */
+	bsrli	r9, r12, 8	/* t1 = v >> 8 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 0	/* *(d + 0) = t1 */
+	bslli	r11, r12, 24	/* h = v << 24 */
+	lwi	r12, r8, 8	/* v = *(as + 8) */
+	bsrli	r9, r12, 8	/* t1 = v >> 8 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 4	/* *(d + 4) = t1 */
+	bslli	r11, r12, 24	/* h = v << 24 */
+	lwi	r12, r8, 12	/* v = *(as + 12) */
+	bsrli	r9, r12, 8	/* t1 = v >> 8 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 8	/* *(d + 8) = t1 */
+	bslli	r11, r12, 24	/* h = v << 24 */
+	lwi	r12, r8, 16	/* v = *(as + 16) */
+	bsrli	r9, r12, 8	/* t1 = v >> 8 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 12	/* *(d + 12) = t1 */
+	bslli	r11, r12, 24	/* h = v << 24 */
+	lwi	r12, r8, 20	/* v = *(as + 20) */
+	bsrli	r9, r12, 8	/* t1 = v >> 8 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 16	/* *(d + 16) = t1 */
+	bslli	r11, r12, 24	/* h = v << 24 */
+	lwi	r12, r8, 24	/* v = *(as + 24) */
+	bsrli	r9, r12, 8	/* t1 = v >> 8 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 20	/* *(d + 20) = t1 */
+	bslli	r11, r12, 24	/* h = v << 24 */
+	lwi	r12, r8, 28	/* v = *(as + 28) */
+	bsrli	r9, r12, 8	/* t1 = v >> 8 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 24	/* *(d + 24) = t1 */
+	bslli	r11, r12, 24	/* h = v << 24 */
+	lwi	r12, r8, 32	/* v = *(as + 32) */
+	bsrli	r9, r12, 8	/* t1 = v >> 8 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 28	/* *(d + 28) = t1 */
+	bslli	r11, r12, 24	/* h = v << 24 */
+	addi	r8, r8, 32	/* as = as + 32 */
+	addi	r4, r4, -32	/* n = n - 32 */
+	bneid	r4, a_bu3_loop	/* while (n) loop */
+	addi	r5, r5, 32	/* d = d + 32 (IN DELAY SLOT) */
+	bri	a_block_done
+
+a_block_u1:
+	bslli	r11, r11, 8	/* h = h << 8 */
+a_bu1_loop:
+	lwi	r12, r8, 4	/* v = *(as + 4) */
+	bsrli	r9, r12, 24	/* t1 = v >> 24 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 0	/* *(d + 0) = t1 */
+	bslli	r11, r12, 8	/* h = v << 8 */
+	lwi	r12, r8, 8	/* v = *(as + 8) */
+	bsrli	r9, r12, 24	/* t1 = v >> 24 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 4	/* *(d + 4) = t1 */
+	bslli	r11, r12, 8	/* h = v << 8 */
+	lwi	r12, r8, 12	/* v = *(as + 12) */
+	bsrli	r9, r12, 24	/* t1 = v >> 24 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 8	/* *(d + 8) = t1 */
+	bslli	r11, r12, 8	/* h = v << 8 */
+	lwi	r12, r8, 16	/* v = *(as + 16) */
+	bsrli	r9, r12, 24	/* t1 = v >> 24 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 12	/* *(d + 12) = t1 */
+	bslli	r11, r12, 8	/* h = v << 8 */
+	lwi	r12, r8, 20	/* v = *(as + 20) */
+	bsrli	r9, r12, 24	/* t1 = v >> 24 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 16	/* *(d + 16) = t1 */
+	bslli	r11, r12, 8	/* h = v << 8 */
+	lwi	r12, r8, 24	/* v = *(as + 24) */
+	bsrli	r9, r12, 24	/* t1 = v >> 24 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 20	/* *(d + 20) = t1 */
+	bslli	r11, r12, 8	/* h = v << 8 */
+	lwi	r12, r8, 28	/* v = *(as + 28) */
+	bsrli	r9, r12, 24	/* t1 = v >> 24 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 24	/* *(d + 24) = t1 */
+	bslli	r11, r12, 8	/* h = v << 8 */
+	lwi	r12, r8, 32	/* v = *(as + 32) */
+	bsrli	r9, r12, 24	/* t1 = v >> 24 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 28	/* *(d + 28) = t1 */
+	bslli	r11, r12, 8	/* h = v << 8 */
+	addi	r8, r8, 32	/* as = as + 32 */
+	addi	r4, r4, -32	/* n = n - 32 */
+	bneid	r4, a_bu1_loop	/* while (n) loop */
+	addi	r5, r5, 32	/* d = d + 32 (IN DELAY SLOT) */
+	bri	a_block_done
+
+a_block_u2:
+	bslli	r11, r11, 16	/* h = h << 16 */
+a_bu2_loop:
+	lwi	r12, r8, 4	/* v = *(as + 4) */
+	bsrli	r9, r12, 16	/* t1 = v >> 16 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 0	/* *(d + 0) = t1 */
+	bslli	r11, r12, 16	/* h = v << 16 */
+	lwi	r12, r8, 8	/* v = *(as + 8) */
+	bsrli	r9, r12, 16	/* t1 = v >> 16 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 4	/* *(d + 4) = t1 */
+	bslli	r11, r12, 16	/* h = v << 16 */
+	lwi	r12, r8, 12	/* v = *(as + 12) */
+	bsrli	r9, r12, 16	/* t1 = v >> 16 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 8	/* *(d + 8) = t1 */
+	bslli	r11, r12, 16	/* h = v << 16 */
+	lwi	r12, r8, 16	/* v = *(as + 16) */
+	bsrli	r9, r12, 16	/* t1 = v >> 16 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 12	/* *(d + 12) = t1 */
+	bslli	r11, r12, 16	/* h = v << 16 */
+	lwi	r12, r8, 20	/* v = *(as + 20) */
+	bsrli	r9, r12, 16	/* t1 = v >> 16 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 16	/* *(d + 16) = t1 */
+	bslli	r11, r12, 16	/* h = v << 16 */
+	lwi	r12, r8, 24	/* v = *(as + 24) */
+	bsrli	r9, r12, 16	/* t1 = v >> 16 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 20	/* *(d + 20) = t1 */
+	bslli	r11, r12, 16	/* h = v << 16 */
+	lwi	r12, r8, 28	/* v = *(as + 28) */
+	bsrli	r9, r12, 16	/* t1 = v >> 16 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 24	/* *(d + 24) = t1 */
+	bslli	r11, r12, 16	/* h = v << 16 */
+	lwi	r12, r8, 32	/* v = *(as + 32) */
+	bsrli	r9, r12, 16	/* t1 = v >> 16 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 28	/* *(d + 28) = t1 */
+	bslli	r11, r12, 16	/* h = v << 16 */
+	addi	r8, r8, 32	/* as = as + 32 */
+	addi	r4, r4, -32	/* n = n - 32 */
+	bneid	r4, a_bu2_loop	/* while (n) loop */
+	addi	r5, r5, 32	/* d = d + 32 (IN DELAY SLOT) */
+
+a_block_done:
+	addi	r4, r0, 4	/* n = 4 */
+	cmpu	r4, r4, r7	/* n = c - n  (unsigned) */
+	blti	r4, a_xfer_end	/* if n < 0, less than one word to transfer */
+
+a_word_xfer:
+	andi	r4, r7, 0xfffffffc	/* n = c & ~3 */
+	addi	r10, r0, 0		/* offset = 0 */
+
+	andi	r9, r6, 3		/* t1 = s & 3 */
+	/* if temp != 0, unaligned transfers needed */
+	bnei	r9, a_word_unaligned
+
+a_word_aligned:
+	lw	r9, r6, r10		/* t1 = *(s+offset) */
+	sw	r9, r5, r10		/* *(d+offset) = t1 */
+	addi	r4, r4,-4		/* n-- */
+	bneid	r4, a_word_aligned	/* loop */
+	addi	r10, r10, 4		/* offset++ (IN DELAY SLOT) */
+
+	bri	a_word_done
+
+a_word_unaligned:
+	andi	r8, r6, 0xfffffffc	/* as = s & ~3 */
+	lwi	r11, r8, 0		/* h = *(as + 0) */
+	addi	r8, r8, 4		/* as = as + 4 */
+
+	addi	r9, r9, -1
+	beqi	r9, a_word_u1		/* t1 was 1 => 1 byte offset */
+	addi	r9, r9, -1
+	beqi	r9, a_word_u2		/* t1 was 2 => 2 byte offset */
+
+a_word_u3:
+	bslli	r11, r11, 24	/* h = h << 24 */
+a_wu3_loop:
+	lw	r12, r8, r10	/* v = *(as + offset) */
+	bsrli	r9, r12, 8	/* t1 = v >> 8 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	sw	r9, r5, r10	/* *(d + offset) = t1 */
+	bslli	r11, r12, 24	/* h = v << 24 */
+	addi	r4, r4,-4	/* n = n - 4 */
+	bneid	r4, a_wu3_loop	/* while (n) loop */
+	addi	r10, r10, 4	/* offset = ofset + 4 (IN DELAY SLOT) */
+
+	bri	a_word_done
+
+a_word_u1:
+	bslli	r11, r11, 8	/* h = h << 8 */
+a_wu1_loop:
+	lw	r12, r8, r10	/* v = *(as + offset) */
+	bsrli	r9, r12, 24	/* t1 = v >> 24 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	sw	r9, r5, r10	/* *(d + offset) = t1 */
+	bslli	r11, r12, 8	/* h = v << 8 */
+	addi	r4, r4,-4	/* n = n - 4 */
+	bneid	r4, a_wu1_loop	/* while (n) loop */
+	addi	r10, r10, 4	/* offset = ofset + 4 (IN DELAY SLOT) */
+
+	bri	a_word_done
+
+a_word_u2:
+	bslli	r11, r11, 16	/* h = h << 16 */
+a_wu2_loop:
+	lw	r12, r8, r10	/* v = *(as + offset) */
+	bsrli	r9, r12, 16	/* t1 = v >> 16 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	sw	r9, r5, r10	/* *(d + offset) = t1 */
+	bslli	r11, r12, 16	/* h = v << 16 */
+	addi	r4, r4,-4	/* n = n - 4 */
+	bneid	r4, a_wu2_loop	/* while (n) loop */
+	addi	r10, r10, 4	/* offset = ofset + 4 (IN DELAY SLOT) */
+
+a_word_done:
+	add	r5, r5, r10	/* d = d + offset */
+	add	r6, r6, r10	/* s = s + offset */
+	rsub	r7, r10, r7	/* c = c - offset */
+
+a_xfer_end:
+a_xfer_end_loop:
+	beqi	r7, a_done		/* while (c) */
+	lbui	r9, r6, 0		/* t1 = *s */
+	addi	r6, r6, 1		/* s++ */
+	sbi	r9, r5, 0		/* *d = t1 */
+	addi	r7, r7, -1		/* c-- */
+	brid	a_xfer_end_loop		/* loop */
+	addi	r5, r5, 1		/* d++ (IN DELAY SLOT) */
+
+a_done:
+	rtsd	r15, 8
+	nop
+
+.size  memcpy, . - memcpy
+.end memcpy
+/*----------------------------------------------------------------------------*/
+	.globl	memmove
+	.type  memmove, @function
+	.ent	memmove
+
+memmove:
+	cmpu	r4, r5, r6	/* n = s - d */
+	bgei	r4,fast_memcpy_ascending
+
+fast_memcpy_descending:
+	/* move d to return register as value of function */
+	addi	r3, r5, 0
+
+	add	r5, r5, r7	/* d = d + c */
+	add	r6, r6, r7	/* s = s + c */
+
+	addi	r4, r0, 4	/* n = 4 */
+	cmpu	r4, r4, r7	/* n = c - n  (unsigned) */
+	blti	r4,d_xfer_end	/* if n < 0, less than one word to transfer */
+
+	/* transfer first 0~3 bytes to get aligned dest address */
+	andi	r4, r5, 3		/* n = d & 3 */
+	/* if zero, destination already aligned */
+	beqi	r4,d_dalign_done
+	rsub	r7, r4, r7		/* c = c - n adjust c */
+
+d_xfer_first_loop:
+	/* if no bytes left to transfer, transfer the bulk */
+	beqi	r4,d_dalign_done
+	addi	r6, r6, -1		/* s-- */
+	addi	r5, r5, -1		/* d-- */
+	lbui	r11, r6, 0		/* h = *s */
+	sbi	r11, r5, 0		/* *d = h */
+	brid	d_xfer_first_loop	/* loop */
+	addi	r4, r4, -1		/* n-- (IN DELAY SLOT) */
+
+d_dalign_done:
+	addi	r4, r0, 32	/* n = 32 */
+	cmpu	r4, r4, r7	/* n = c - n  (unsigned) */
+	/* if n < 0, less than one block to transfer */
+	blti	r4, d_block_done
+
+d_block_xfer:
+	andi	r4, r7, 0xffffffe0	/* n = c & ~31 */
+	rsub	r7, r4, r7		/* c = c - n */
+
+	andi	r9, r6, 3		/* t1 = s & 3 */
+	/* if temp != 0, unaligned transfers needed */
+	bnei	r9, d_block_unaligned
+
+d_block_aligned:
+	addi	r6, r6, -32		/* s = s - 32 */
+	addi	r5, r5, -32		/* d = d - 32 */
+	lwi	r9, r6, 28		/* t1 = *(s + 28) */
+	lwi	r10, r6, 24		/* t2 = *(s + 24) */
+	lwi	r11, r6, 20		/* t3 = *(s + 20) */
+	lwi	r12, r6, 16		/* t4 = *(s + 16) */
+	swi	r9, r5, 28		/* *(d + 28) = t1 */
+	swi	r10, r5, 24		/* *(d + 24) = t2 */
+	swi	r11, r5, 20		/* *(d + 20) = t3 */
+	swi	r12, r5, 16		/* *(d + 16) = t4 */
+	lwi	r9, r6, 12		/* t1 = *(s + 12) */
+	lwi	r10, r6, 8		/* t2 = *(s + 8) */
+	lwi	r11, r6, 4		/* t3 = *(s + 4) */
+	lwi	r12, r6, 0		/* t4 = *(s + 0) */
+	swi	r9, r5, 12		/* *(d + 12) = t1 */
+	swi	r10, r5, 8		/* *(d + 8) = t2 */
+	swi	r11, r5, 4		/* *(d + 4) = t3 */
+	addi	r4, r4, -32		/* n = n - 32 */
+	bneid	r4, d_block_aligned	/* while (n) loop */
+	swi	r12, r5, 0		/* *(d + 0) = t4 (IN DELAY SLOT) */
+	bri	d_block_done
+
+d_block_unaligned:
+	andi	r8, r6, 0xfffffffc	/* as = s & ~3 */
+	rsub	r6, r4, r6		/* s = s - n */
+	lwi	r11, r8, 0		/* h = *(as + 0) */
+
+	addi	r9, r9, -1
+	beqi	r9,d_block_u1		/* t1 was 1 => 1 byte offset */
+	addi	r9, r9, -1
+	beqi	r9,d_block_u2		/* t1 was 2 => 2 byte offset */
+
+d_block_u3:
+	bsrli	r11, r11, 8	/* h = h >> 8 */
+d_bu3_loop:
+	addi	r8, r8, -32	/* as = as - 32 */
+	addi	r5, r5, -32	/* d = d - 32 */
+	lwi	r12, r8, 28	/* v = *(as + 28) */
+	bslli	r9, r12, 24	/* t1 = v << 24 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 28	/* *(d + 28) = t1 */
+	bsrli	r11, r12, 8	/* h = v >> 8 */
+	lwi	r12, r8, 24	/* v = *(as + 24) */
+	bslli	r9, r12, 24	/* t1 = v << 24 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 24	/* *(d + 24) = t1 */
+	bsrli	r11, r12, 8	/* h = v >> 8 */
+	lwi	r12, r8, 20	/* v = *(as + 20) */
+	bslli	r9, r12, 24	/* t1 = v << 24 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 20	/* *(d + 20) = t1 */
+	bsrli	r11, r12, 8	/* h = v >> 8 */
+	lwi	r12, r8, 16	/* v = *(as + 16) */
+	bslli	r9, r12, 24	/* t1 = v << 24 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 16	/* *(d + 16) = t1 */
+	bsrli	r11, r12, 8	/* h = v >> 8 */
+	lwi	r12, r8, 12	/* v = *(as + 12) */
+	bslli	r9, r12, 24	/* t1 = v << 24 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 12	/* *(d + 112) = t1 */
+	bsrli	r11, r12, 8	/* h = v >> 8 */
+	lwi	r12, r8, 8	/* v = *(as + 8) */
+	bslli	r9, r12, 24	/* t1 = v << 24 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 8	/* *(d + 8) = t1 */
+	bsrli	r11, r12, 8	/* h = v >> 8 */
+	lwi	r12, r8, 4	/* v = *(as + 4) */
+	bslli	r9, r12, 24	/* t1 = v << 24 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 4	/* *(d + 4) = t1 */
+	bsrli	r11, r12, 8	/* h = v >> 8 */
+	lwi	r12, r8, 0	/* v = *(as + 0) */
+	bslli	r9, r12, 24	/* t1 = v << 24 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 0	/* *(d + 0) = t1 */
+	addi	r4, r4, -32	/* n = n - 32 */
+	bneid	r4, d_bu3_loop	/* while (n) loop */
+	bsrli	r11, r12, 8	/* h = v >> 8 (IN DELAY SLOT) */
+	bri	d_block_done
+
+d_block_u1:
+	bsrli	r11, r11, 24	/* h = h >> 24 */
+d_bu1_loop:
+	addi	r8, r8, -32	/* as = as - 32 */
+	addi	r5, r5, -32	/* d = d - 32 */
+	lwi	r12, r8, 28	/* v = *(as + 28) */
+	bslli	r9, r12, 8	/* t1 = v << 8 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 28	/* *(d + 28) = t1 */
+	bsrli	r11, r12, 24	/* h = v >> 24 */
+	lwi	r12, r8, 24	/* v = *(as + 24) */
+	bslli	r9, r12, 8	/* t1 = v << 8 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 24	/* *(d + 24) = t1 */
+	bsrli	r11, r12, 24	/* h = v >> 24 */
+	lwi	r12, r8, 20	/* v = *(as + 20) */
+	bslli	r9, r12, 8	/* t1 = v << 8 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 20	/* *(d + 20) = t1 */
+	bsrli	r11, r12, 24	/* h = v >> 24 */
+	lwi	r12, r8, 16	/* v = *(as + 16) */
+	bslli	r9, r12, 8	/* t1 = v << 8 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 16	/* *(d + 16) = t1 */
+	bsrli	r11, r12, 24	/* h = v >> 24 */
+	lwi	r12, r8, 12	/* v = *(as + 12) */
+	bslli	r9, r12, 8	/* t1 = v << 8 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 12	/* *(d + 112) = t1 */
+	bsrli	r11, r12, 24	/* h = v >> 24 */
+	lwi	r12, r8, 8	/* v = *(as + 8) */
+	bslli	r9, r12, 8	/* t1 = v << 8 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 8	/* *(d + 8) = t1 */
+	bsrli	r11, r12, 24	/* h = v >> 24 */
+	lwi	r12, r8, 4	/* v = *(as + 4) */
+	bslli	r9, r12, 8	/* t1 = v << 8 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 4	/* *(d + 4) = t1 */
+	bsrli	r11, r12, 24	/* h = v >> 24 */
+	lwi	r12, r8, 0	/* v = *(as + 0) */
+	bslli	r9, r12, 8	/* t1 = v << 8 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 0	/* *(d + 0) = t1 */
+	addi	r4, r4, -32	/* n = n - 32 */
+	bneid	r4, d_bu1_loop	/* while (n) loop */
+	bsrli	r11, r12, 24	/* h = v >> 24 (IN DELAY SLOT) */
+	bri	d_block_done
+
+d_block_u2:
+	bsrli	r11, r11, 16	/* h = h >> 16 */
+d_bu2_loop:
+	addi	r8, r8, -32	/* as = as - 32 */
+	addi	r5, r5, -32	/* d = d - 32 */
+	lwi	r12, r8, 28	/* v = *(as + 28) */
+	bslli	r9, r12, 16	/* t1 = v << 16 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 28	/* *(d + 28) = t1 */
+	bsrli	r11, r12, 16	/* h = v >> 16 */
+	lwi	r12, r8, 24	/* v = *(as + 24) */
+	bslli	r9, r12, 16	/* t1 = v << 16 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 24	/* *(d + 24) = t1 */
+	bsrli	r11, r12, 16	/* h = v >> 16 */
+	lwi	r12, r8, 20	/* v = *(as + 20) */
+	bslli	r9, r12, 16	/* t1 = v << 16 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 20	/* *(d + 20) = t1 */
+	bsrli	r11, r12, 16	/* h = v >> 16 */
+	lwi	r12, r8, 16	/* v = *(as + 16) */
+	bslli	r9, r12, 16	/* t1 = v << 16 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 16	/* *(d + 16) = t1 */
+	bsrli	r11, r12, 16	/* h = v >> 16 */
+	lwi	r12, r8, 12	/* v = *(as + 12) */
+	bslli	r9, r12, 16	/* t1 = v << 16 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 12	/* *(d + 112) = t1 */
+	bsrli	r11, r12, 16	/* h = v >> 16 */
+	lwi	r12, r8, 8	/* v = *(as + 8) */
+	bslli	r9, r12, 16	/* t1 = v << 16 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 8	/* *(d + 8) = t1 */
+	bsrli	r11, r12, 16	/* h = v >> 16 */
+	lwi	r12, r8, 4	/* v = *(as + 4) */
+	bslli	r9, r12, 16	/* t1 = v << 16 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 4	/* *(d + 4) = t1 */
+	bsrli	r11, r12, 16	/* h = v >> 16 */
+	lwi	r12, r8, 0	/* v = *(as + 0) */
+	bslli	r9, r12, 16	/* t1 = v << 16 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	swi	r9, r5, 0	/* *(d + 0) = t1 */
+	addi	r4, r4, -32	/* n = n - 32 */
+	bneid	r4, d_bu2_loop	/* while (n) loop */
+	bsrli	r11, r12, 16	/* h = v >> 16 (IN DELAY SLOT) */
+
+d_block_done:
+	addi	r4, r0, 4	/* n = 4 */
+	cmpu	r4, r4, r7	/* n = c - n  (unsigned) */
+	blti	r4,d_xfer_end	/* if n < 0, less than one word to transfer */
+
+d_word_xfer:
+	andi	r4, r7, 0xfffffffc	/* n = c & ~3 */
+	rsub	r5, r4, r5		/* d = d - n */
+	rsub	r6, r4, r6		/* s = s - n */
+	rsub	r7, r4, r7		/* c = c - n */
+
+	andi	r9, r6, 3		/* t1 = s & 3 */
+	/* if temp != 0, unaligned transfers needed */
+	bnei	r9, d_word_unaligned
+
+d_word_aligned:
+	addi	r4, r4,-4		/* n-- */
+	lw	r9, r6, r4		/* t1 = *(s+n) */
+	bneid	r4, d_word_aligned	/* loop */
+	sw	r9, r5, r4		/* *(d+n) = t1 (IN DELAY SLOT) */
+
+	bri	d_word_done
+
+d_word_unaligned:
+	andi	r8, r6, 0xfffffffc	/* as = s & ~3 */
+	lw	r11, r8, r4		/* h = *(as + n) */
+
+	addi	r9, r9, -1
+	beqi	r9,d_word_u1		/* t1 was 1 => 1 byte offset */
+	addi	r9, r9, -1
+	beqi	r9,d_word_u2		/* t1 was 2 => 2 byte offset */
+
+d_word_u3:
+	bsrli	r11, r11, 8	/* h = h >> 8 */
+d_wu3_loop:
+	addi	r4, r4,-4	/* n = n - 4 */
+	lw	r12, r8, r4	/* v = *(as + n) */
+	bslli	r9, r12, 24	/* t1 = v << 24 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	sw	r9, r5, r4	/* *(d + n) = t1 */
+	bneid	r4, d_wu3_loop	/* while (n) loop */
+	bsrli	r11, r12, 8	/* h = v >> 8 (IN DELAY SLOT) */
+
+	bri	d_word_done
+
+d_word_u1:
+	bsrli	r11, r11, 24	/* h = h >> 24 */
+d_wu1_loop:
+	addi	r4, r4,-4	/* n = n - 4 */
+	lw	r12, r8, r4	/* v = *(as + n) */
+	bslli	r9, r12, 8	/* t1 = v << 8 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	sw	r9, r5, r4	/* *(d + n) = t1 */
+	bneid	r4, d_wu1_loop	/* while (n) loop */
+	bsrli	r11, r12, 24	/* h = v >> 24 (IN DELAY SLOT) */
+
+	bri	d_word_done
+
+d_word_u2:
+	bsrli	r11, r11, 16	/* h = h >> 16 */
+d_wu2_loop:
+	addi	r4, r4,-4	/* n = n - 4 */
+	lw	r12, r8, r4	/* v = *(as + n) */
+	bslli	r9, r12, 16	/* t1 = v << 16 */
+	or	r9, r11, r9	/* t1 = h | t1 */
+	sw	r9, r5, r4	/* *(d + n) = t1 */
+	bneid	r4, d_wu2_loop	/* while (n) loop */
+	bsrli	r11, r12, 16	/* h = v >> 16 (IN DELAY SLOT) */
+
+d_word_done:
+
+d_xfer_end:
+d_xfer_end_loop:
+	beqi	r7, a_done		/* while (c) */
+	addi	r6, r6, -1		/* s-- */
+	lbui	r9, r6, 0		/* t1 = *s */
+	addi	r5, r5, -1		/* d-- */
+	sbi	r9, r5, 0		/* *d = t1 */
+	brid	d_xfer_end_loop		/* loop */
+	addi	r7, r7, -1		/* c-- (IN DELAY SLOT) */
+
+d_done:
+	rtsd	r15, 8
+	nop
+
+.size  memmove, . - memmove
+.end memmove
diff --git a/arch/microblaze/lib/libgcc.h b/arch/microblaze/lib/libgcc.h
new file mode 100644
index 0000000..ab077ef
--- /dev/null
+++ b/arch/microblaze/lib/libgcc.h
@@ -0,0 +1,32 @@
+#ifndef __ASM_LIBGCC_H
+#define __ASM_LIBGCC_H
+
+#include <asm/byteorder.h>
+
+typedef int word_type __attribute__ ((mode (__word__)));
+
+#ifdef __BIG_ENDIAN
+struct DWstruct {
+	int high, low;
+};
+#elif defined(__LITTLE_ENDIAN)
+struct DWstruct {
+	int low, high;
+};
+#else
+#error I feel sick.
+#endif
+
+typedef union {
+	struct DWstruct s;
+	long long ll;
+} DWunion;
+
+extern long long __ashldi3(long long u, word_type b);
+extern long long __ashrdi3(long long u, word_type b);
+extern word_type __cmpdi2(long long a, long long b);
+extern long long __lshrdi3(long long u, word_type b);
+extern long long __muldi3(long long u, long long v);
+extern word_type __ucmpdi2(unsigned long long a, unsigned long long b);
+
+#endif /* __ASM_LIBGCC_H */
diff --git a/arch/microblaze/lib/lshrdi3.c b/arch/microblaze/lib/lshrdi3.c
new file mode 100644
index 0000000..adcb253
--- /dev/null
+++ b/arch/microblaze/lib/lshrdi3.c
@@ -0,0 +1,28 @@
+#include <linux/export.h>
+
+#include "libgcc.h"
+
+long long __lshrdi3(long long u, word_type b)
+{
+	DWunion uu, w;
+	word_type bm;
+
+	if (b == 0)
+		return u;
+
+	uu.ll = u;
+	bm = 32 - b;
+
+	if (bm <= 0) {
+		w.s.high = 0;
+		w.s.low = (unsigned int) uu.s.high >> -bm;
+	} else {
+		const unsigned int carries = (unsigned int) uu.s.high << bm;
+
+		w.s.high = (unsigned int) uu.s.high >> b;
+		w.s.low = ((unsigned int) uu.s.low >> b) | carries;
+	}
+
+	return w.ll;
+}
+EXPORT_SYMBOL(__lshrdi3);
diff --git a/arch/microblaze/lib/memcpy.c b/arch/microblaze/lib/memcpy.c
new file mode 100644
index 0000000..f536e81
--- /dev/null
+++ b/arch/microblaze/lib/memcpy.c
@@ -0,0 +1,189 @@
+/*
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ * Copyright (C) 2007 John Williams
+ *
+ * Reasonably optimised generic C-code for memcpy on Microblaze
+ * This is generic C code to do efficient, alignment-aware memcpy.
+ *
+ * It is based on demo code originally Copyright 2001 by Intel Corp, taken from
+ * http://www.embedded.com/showArticle.jhtml?articleID=19205567
+ *
+ * Attempts were made, unsuccessfully, to contact the original
+ * author of this code (Michael Morrow, Intel).  Below is the original
+ * copyright notice.
+ *
+ * This software has been developed by Intel Corporation.
+ * Intel specifically disclaims all warranties, express or
+ * implied, and all liability, including consequential and
+ * other indirect damages, for the use of this program, including
+ * liability for infringement of any proprietary rights,
+ * and including the warranties of merchantability and fitness
+ * for a particular purpose. Intel does not assume any
+ * responsibility for and errors which may appear in this program
+ * not any responsibility to update it.
+ */
+
+#include <linux/export.h>
+#include <linux/types.h>
+#include <linux/stddef.h>
+#include <linux/compiler.h>
+
+#include <linux/string.h>
+
+#ifdef __HAVE_ARCH_MEMCPY
+#ifndef CONFIG_OPT_LIB_FUNCTION
+void *memcpy(void *v_dst, const void *v_src, __kernel_size_t c)
+{
+	const char *src = v_src;
+	char *dst = v_dst;
+
+	/* Simple, byte oriented memcpy. */
+	while (c--)
+		*dst++ = *src++;
+
+	return v_dst;
+}
+#else /* CONFIG_OPT_LIB_FUNCTION */
+void *memcpy(void *v_dst, const void *v_src, __kernel_size_t c)
+{
+	const char *src = v_src;
+	char *dst = v_dst;
+
+	/* The following code tries to optimize the copy by using unsigned
+	 * alignment. This will work fine if both source and destination are
+	 * aligned on the same boundary. However, if they are aligned on
+	 * different boundaries shifts will be necessary. This might result in
+	 * bad performance on MicroBlaze systems without a barrel shifter.
+	 */
+	const uint32_t *i_src;
+	uint32_t *i_dst;
+
+	if (likely(c >= 4)) {
+		unsigned  value, buf_hold;
+
+		/* Align the destination to a word boundary. */
+		/* This is done in an endian independent manner. */
+		switch ((unsigned long)dst & 3) {
+		case 1:
+			*dst++ = *src++;
+			--c;
+		case 2:
+			*dst++ = *src++;
+			--c;
+		case 3:
+			*dst++ = *src++;
+			--c;
+		}
+
+		i_dst = (void *)dst;
+
+		/* Choose a copy scheme based on the source */
+		/* alignment relative to destination. */
+		switch ((unsigned long)src & 3) {
+		case 0x0:	/* Both byte offsets are aligned */
+			i_src  = (const void *)src;
+
+			for (; c >= 4; c -= 4)
+				*i_dst++ = *i_src++;
+
+			src  = (const void *)i_src;
+			break;
+		case 0x1:	/* Unaligned - Off by 1 */
+			/* Word align the source */
+			i_src = (const void *) ((unsigned)src & ~3);
+#ifndef __MICROBLAZEEL__
+			/* Load the holding buffer */
+			buf_hold = *i_src++ << 8;
+
+			for (; c >= 4; c -= 4) {
+				value = *i_src++;
+				*i_dst++ = buf_hold | value >> 24;
+				buf_hold = value << 8;
+			}
+#else
+			/* Load the holding buffer */
+			buf_hold = (*i_src++ & 0xFFFFFF00) >> 8;
+
+			for (; c >= 4; c -= 4) {
+				value = *i_src++;
+				*i_dst++ = buf_hold | ((value & 0xFF) << 24);
+				buf_hold = (value & 0xFFFFFF00) >> 8;
+			}
+#endif
+			/* Realign the source */
+			src = (const void *)i_src;
+			src -= 3;
+			break;
+		case 0x2:	/* Unaligned - Off by 2 */
+			/* Word align the source */
+			i_src = (const void *) ((unsigned)src & ~3);
+#ifndef __MICROBLAZEEL__
+			/* Load the holding buffer */
+			buf_hold = *i_src++ << 16;
+
+			for (; c >= 4; c -= 4) {
+				value = *i_src++;
+				*i_dst++ = buf_hold | value >> 16;
+				buf_hold = value << 16;
+			}
+#else
+			/* Load the holding buffer */
+			buf_hold = (*i_src++ & 0xFFFF0000) >> 16;
+
+			for (; c >= 4; c -= 4) {
+				value = *i_src++;
+				*i_dst++ = buf_hold | ((value & 0xFFFF) << 16);
+				buf_hold = (value & 0xFFFF0000) >> 16;
+			}
+#endif
+			/* Realign the source */
+			src = (const void *)i_src;
+			src -= 2;
+			break;
+		case 0x3:	/* Unaligned - Off by 3 */
+			/* Word align the source */
+			i_src = (const void *) ((unsigned)src & ~3);
+#ifndef __MICROBLAZEEL__
+			/* Load the holding buffer */
+			buf_hold = *i_src++ << 24;
+
+			for (; c >= 4; c -= 4) {
+				value = *i_src++;
+				*i_dst++ = buf_hold | value >> 8;
+				buf_hold = value << 24;
+			}
+#else
+			/* Load the holding buffer */
+			buf_hold = (*i_src++ & 0xFF000000) >> 24;
+
+			for (; c >= 4; c -= 4) {
+				value = *i_src++;
+				*i_dst++ = buf_hold | ((value & 0xFFFFFF) << 8);
+				buf_hold = (value & 0xFF000000) >> 24;
+			}
+#endif
+			/* Realign the source */
+			src = (const void *)i_src;
+			src -= 1;
+			break;
+		}
+		dst = (void *)i_dst;
+	}
+
+	/* Finish off any remaining bytes */
+	/* simple fast copy, ... unless a cache boundary is crossed */
+	switch (c) {
+	case 3:
+		*dst++ = *src++;
+	case 2:
+		*dst++ = *src++;
+	case 1:
+		*dst++ = *src++;
+	}
+
+	return v_dst;
+}
+#endif /* CONFIG_OPT_LIB_FUNCTION */
+EXPORT_SYMBOL(memcpy);
+#endif /* __HAVE_ARCH_MEMCPY */
diff --git a/arch/microblaze/lib/memmove.c b/arch/microblaze/lib/memmove.c
new file mode 100644
index 0000000..3611ce7
--- /dev/null
+++ b/arch/microblaze/lib/memmove.c
@@ -0,0 +1,215 @@
+/*
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ * Copyright (C) 2007 John Williams
+ *
+ * Reasonably optimised generic C-code for memcpy on Microblaze
+ * This is generic C code to do efficient, alignment-aware memmove.
+ *
+ * It is based on demo code originally Copyright 2001 by Intel Corp, taken from
+ * http://www.embedded.com/showArticle.jhtml?articleID=19205567
+ *
+ * Attempts were made, unsuccessfully, to contact the original
+ * author of this code (Michael Morrow, Intel).  Below is the original
+ * copyright notice.
+ *
+ * This software has been developed by Intel Corporation.
+ * Intel specifically disclaims all warranties, express or
+ * implied, and all liability, including consequential and
+ * other indirect damages, for the use of this program, including
+ * liability for infringement of any proprietary rights,
+ * and including the warranties of merchantability and fitness
+ * for a particular purpose. Intel does not assume any
+ * responsibility for and errors which may appear in this program
+ * not any responsibility to update it.
+ */
+
+#include <linux/export.h>
+#include <linux/types.h>
+#include <linux/stddef.h>
+#include <linux/compiler.h>
+#include <linux/string.h>
+
+#ifdef __HAVE_ARCH_MEMMOVE
+#ifndef CONFIG_OPT_LIB_FUNCTION
+void *memmove(void *v_dst, const void *v_src, __kernel_size_t c)
+{
+	const char *src = v_src;
+	char *dst = v_dst;
+
+	if (!c)
+		return v_dst;
+
+	/* Use memcpy when source is higher than dest */
+	if (v_dst <= v_src)
+		return memcpy(v_dst, v_src, c);
+
+	/* copy backwards, from end to beginning */
+	src += c;
+	dst += c;
+
+	/* Simple, byte oriented memmove. */
+	while (c--)
+		*--dst = *--src;
+
+	return v_dst;
+}
+#else /* CONFIG_OPT_LIB_FUNCTION */
+void *memmove(void *v_dst, const void *v_src, __kernel_size_t c)
+{
+	const char *src = v_src;
+	char *dst = v_dst;
+	const uint32_t *i_src;
+	uint32_t *i_dst;
+
+	if (!c)
+		return v_dst;
+
+	/* Use memcpy when source is higher than dest */
+	if (v_dst <= v_src)
+		return memcpy(v_dst, v_src, c);
+
+	/* The following code tries to optimize the copy by using unsigned
+	 * alignment. This will work fine if both source and destination are
+	 * aligned on the same boundary. However, if they are aligned on
+	 * different boundaries shifts will be necessary. This might result in
+	 * bad performance on MicroBlaze systems without a barrel shifter.
+	 */
+	/* FIXME this part needs more test */
+	/* Do a descending copy - this is a bit trickier! */
+	dst += c;
+	src += c;
+
+	if (c >= 4) {
+		unsigned  value, buf_hold;
+
+		/* Align the destination to a word boundary. */
+		/* This is done in an endian independent manner. */
+
+		switch ((unsigned long)dst & 3) {
+		case 3:
+			*--dst = *--src;
+			--c;
+		case 2:
+			*--dst = *--src;
+			--c;
+		case 1:
+			*--dst = *--src;
+			--c;
+		}
+
+		i_dst = (void *)dst;
+		/* Choose a copy scheme based on the source */
+		/* alignment relative to dstination. */
+		switch ((unsigned long)src & 3) {
+		case 0x0:	/* Both byte offsets are aligned */
+
+			i_src  = (const void *)src;
+
+			for (; c >= 4; c -= 4)
+				*--i_dst = *--i_src;
+
+			src  = (const void *)i_src;
+			break;
+		case 0x1:	/* Unaligned - Off by 1 */
+			/* Word align the source */
+			i_src = (const void *) (((unsigned)src + 4) & ~3);
+#ifndef __MICROBLAZEEL__
+			/* Load the holding buffer */
+			buf_hold = *--i_src >> 24;
+
+			for (; c >= 4; c -= 4) {
+				value = *--i_src;
+				*--i_dst = buf_hold << 8 | value;
+				buf_hold = value >> 24;
+			}
+#else
+			/* Load the holding buffer */
+			buf_hold = (*--i_src & 0xFF) << 24;
+
+			for (; c >= 4; c -= 4) {
+				value = *--i_src;
+				*--i_dst = buf_hold |
+						((value & 0xFFFFFF00) >> 8);
+				buf_hold = (value  & 0xFF) << 24;
+			}
+#endif
+			/* Realign the source */
+			src = (const void *)i_src;
+			src += 1;
+			break;
+		case 0x2:	/* Unaligned - Off by 2 */
+			/* Word align the source */
+			i_src = (const void *) (((unsigned)src + 4) & ~3);
+#ifndef __MICROBLAZEEL__
+			/* Load the holding buffer */
+			buf_hold = *--i_src >> 16;
+
+			for (; c >= 4; c -= 4) {
+				value = *--i_src;
+				*--i_dst = buf_hold << 16 | value;
+				buf_hold = value >> 16;
+			}
+#else
+			/* Load the holding buffer */
+			buf_hold = (*--i_src & 0xFFFF) << 16;
+
+			for (; c >= 4; c -= 4) {
+				value = *--i_src;
+				*--i_dst = buf_hold |
+						((value & 0xFFFF0000) >> 16);
+				buf_hold = (value & 0xFFFF) << 16;
+			}
+#endif
+			/* Realign the source */
+			src = (const void *)i_src;
+			src += 2;
+			break;
+		case 0x3:	/* Unaligned - Off by 3 */
+			/* Word align the source */
+			i_src = (const void *) (((unsigned)src + 4) & ~3);
+#ifndef __MICROBLAZEEL__
+			/* Load the holding buffer */
+			buf_hold = *--i_src >> 8;
+
+			for (; c >= 4; c -= 4) {
+				value = *--i_src;
+				*--i_dst = buf_hold << 24 | value;
+				buf_hold = value >> 8;
+			}
+#else
+			/* Load the holding buffer */
+			buf_hold = (*--i_src & 0xFFFFFF) << 8;
+
+			for (; c >= 4; c -= 4) {
+				value = *--i_src;
+				*--i_dst = buf_hold |
+						((value & 0xFF000000) >> 24);
+				buf_hold = (value & 0xFFFFFF) << 8;
+			}
+#endif
+			/* Realign the source */
+			src = (const void *)i_src;
+			src += 3;
+			break;
+		}
+		dst = (void *)i_dst;
+	}
+
+	/* simple fast copy, ... unless a cache boundary is crossed */
+	/* Finish off any remaining bytes */
+	switch (c) {
+	case 4:
+		*--dst = *--src;
+	case 3:
+		*--dst = *--src;
+	case 2:
+		*--dst = *--src;
+	case 1:
+		*--dst = *--src;
+	}
+	return v_dst;
+}
+#endif /* CONFIG_OPT_LIB_FUNCTION */
+EXPORT_SYMBOL(memmove);
+#endif /* __HAVE_ARCH_MEMMOVE */
diff --git a/arch/microblaze/lib/memset.c b/arch/microblaze/lib/memset.c
new file mode 100644
index 0000000..04ea72c
--- /dev/null
+++ b/arch/microblaze/lib/memset.c
@@ -0,0 +1,97 @@
+/*
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ * Copyright (C) 2007 John Williams
+ *
+ * Reasonably optimised generic C-code for memset on Microblaze
+ * This is generic C code to do efficient, alignment-aware memcpy.
+ *
+ * It is based on demo code originally Copyright 2001 by Intel Corp, taken from
+ * http://www.embedded.com/showArticle.jhtml?articleID=19205567
+ *
+ * Attempts were made, unsuccessfully, to contact the original
+ * author of this code (Michael Morrow, Intel).  Below is the original
+ * copyright notice.
+ *
+ * This software has been developed by Intel Corporation.
+ * Intel specifically disclaims all warranties, express or
+ * implied, and all liability, including consequential and
+ * other indirect damages, for the use of this program, including
+ * liability for infringement of any proprietary rights,
+ * and including the warranties of merchantability and fitness
+ * for a particular purpose. Intel does not assume any
+ * responsibility for and errors which may appear in this program
+ * not any responsibility to update it.
+ */
+
+#include <linux/export.h>
+#include <linux/types.h>
+#include <linux/stddef.h>
+#include <linux/compiler.h>
+#include <linux/string.h>
+
+#ifdef __HAVE_ARCH_MEMSET
+#ifndef CONFIG_OPT_LIB_FUNCTION
+void *memset(void *v_src, int c, __kernel_size_t n)
+{
+	char *src = v_src;
+
+	/* Truncate c to 8 bits */
+	c = (c & 0xFF);
+
+	/* Simple, byte oriented memset or the rest of count. */
+	while (n--)
+		*src++ = c;
+
+	return v_src;
+}
+#else /* CONFIG_OPT_LIB_FUNCTION */
+void *memset(void *v_src, int c, __kernel_size_t n)
+{
+	char *src = v_src;
+	uint32_t *i_src;
+	uint32_t w32 = 0;
+
+	/* Truncate c to 8 bits */
+	c = (c & 0xFF);
+
+	if (unlikely(c)) {
+		/* Make a repeating word out of it */
+		w32 = c;
+		w32 |= w32 << 8;
+		w32 |= w32 << 16;
+	}
+
+	if (likely(n >= 4)) {
+		/* Align the destination to a word boundary */
+		/* This is done in an endian independent manner */
+		switch ((unsigned) src & 3) {
+		case 1:
+			*src++ = c;
+			--n;
+		case 2:
+			*src++ = c;
+			--n;
+		case 3:
+			*src++ = c;
+			--n;
+		}
+
+		i_src  = (void *)src;
+
+		/* Do as many full-word copies as we can */
+		for (; n >= 4; n -= 4)
+			*i_src++ = w32;
+
+		src  = (void *)i_src;
+	}
+
+	/* Simple, byte oriented memset or the rest of count. */
+	while (n--)
+		*src++ = c;
+
+	return v_src;
+}
+#endif /* CONFIG_OPT_LIB_FUNCTION */
+EXPORT_SYMBOL(memset);
+#endif /* __HAVE_ARCH_MEMSET */
diff --git a/arch/microblaze/lib/modsi3.S b/arch/microblaze/lib/modsi3.S
new file mode 100644
index 0000000..84e0bee
--- /dev/null
+++ b/arch/microblaze/lib/modsi3.S
@@ -0,0 +1,73 @@
+#include <linux/linkage.h>
+
+/*
+* modulo operation for 32 bit integers.
+*	Input :	op1 in Reg r5
+*		op2 in Reg r6
+*	Output: op1 mod op2 in Reg r3
+*/
+
+	.text
+	.globl	__modsi3
+	.type __modsi3,  @function
+	.ent __modsi3
+
+__modsi3:
+	.frame	r1, 0, r15
+
+	addik	r1, r1, -16
+	swi	r28, r1, 0
+	swi	r29, r1, 4
+	swi	r30, r1, 8
+	swi	r31, r1, 12
+
+	beqi	r6, div_by_zero /* div_by_zero division error */
+	beqi	r5, result_is_zero /* result is zero */
+	bgeid	r5, r5_pos
+	/* get the sign of the result [ depends only on the first arg] */
+	add	r28, r5, r0
+	rsubi	r5, r5, 0	 /* make r5 positive */
+r5_pos:
+	bgei	r6, r6_pos
+	rsubi	r6, r6, 0	 /* make r6 positive */
+r6_pos:
+	addik	r3, r0, 0 /* clear mod */
+	addik	r30, r0, 0 /* clear div */
+	addik	r29, r0, 32 /* initialize the loop count */
+/* first part try to find the first '1' in the r5 */
+div1:
+	add	r5, r5, r5 /* left shift logical r5 */
+	bgeid	r5, div1
+	addik	r29, r29, -1
+div2:
+	/* left shift logical r5 get the '1' into the carry */
+	add	r5, r5, r5
+	addc	r3, r3, r3 /* move that bit into the mod register */
+	rsub	r31, r6, r3 /* try to subtract (r30 a r6) */
+	blti	r31, mod_too_small
+	/* move the r31 to mod since the result was positive */
+	or	r3, r0, r31
+	addik	r30, r30, 1
+mod_too_small:
+	addik	r29, r29, -1
+	beqi	r29, loop_end
+	add	r30, r30, r30 /* shift in the '1' into div */
+	bri	div2 /* div2 */
+loop_end:
+	bgei	r28, return_here
+	brid	return_here
+	rsubi	r3, r3, 0 /* negate the result */
+div_by_zero:
+result_is_zero:
+	or	r3, r0, r0 /* set result to 0 [both mod as well as div are 0] */
+return_here:
+/* restore values of csrs and that of r3 and the divisor and the dividend */
+	lwi	r28, r1, 0
+	lwi	r29, r1, 4
+	lwi	r30, r1, 8
+	lwi	r31, r1, 12
+	rtsd	r15, 8
+	addik	r1, r1, 16
+
+.size __modsi3,  . - __modsi3
+.end __modsi3
diff --git a/arch/microblaze/lib/muldi3.c b/arch/microblaze/lib/muldi3.c
new file mode 100644
index 0000000..a3f9a03
--- /dev/null
+++ b/arch/microblaze/lib/muldi3.c
@@ -0,0 +1,57 @@
+#include <linux/export.h>
+
+#include "libgcc.h"
+
+#define W_TYPE_SIZE 32
+
+#define __ll_B ((unsigned long) 1 << (W_TYPE_SIZE / 2))
+#define __ll_lowpart(t) ((unsigned long) (t) & (__ll_B - 1))
+#define __ll_highpart(t) ((unsigned long) (t) >> (W_TYPE_SIZE / 2))
+
+/* If we still don't have umul_ppmm, define it using plain C.  */
+#if !defined(umul_ppmm)
+#define umul_ppmm(w1, w0, u, v)						\
+	do {								\
+		unsigned long __x0, __x1, __x2, __x3;			\
+		unsigned short __ul, __vl, __uh, __vh;			\
+									\
+		__ul = __ll_lowpart(u);					\
+		__uh = __ll_highpart(u);				\
+		__vl = __ll_lowpart(v);					\
+		__vh = __ll_highpart(v);				\
+									\
+		__x0 = (unsigned long) __ul * __vl;			\
+		__x1 = (unsigned long) __ul * __vh;			\
+		__x2 = (unsigned long) __uh * __vl;			\
+		__x3 = (unsigned long) __uh * __vh;			\
+									\
+		__x1 += __ll_highpart(__x0); /* this can't give carry */\
+		__x1 += __x2; /* but this indeed can */			\
+		if (__x1 < __x2) /* did we get it? */			\
+		__x3 += __ll_B; /* yes, add it in the proper pos */	\
+									\
+		(w1) = __x3 + __ll_highpart(__x1);			\
+		(w0) = __ll_lowpart(__x1) * __ll_B + __ll_lowpart(__x0);\
+	} while (0)
+#endif
+
+#if !defined(__umulsidi3)
+#define __umulsidi3(u, v) ({				\
+	DWunion __w;					\
+	umul_ppmm(__w.s.high, __w.s.low, u, v);		\
+	__w.ll;						\
+	})
+#endif
+
+long long __muldi3(long long u, long long v)
+{
+	const DWunion uu = {.ll = u};
+	const DWunion vv = {.ll = v};
+	DWunion w = {.ll = __umulsidi3(uu.s.low, vv.s.low)};
+
+	w.s.high += ((unsigned long) uu.s.low * (unsigned long) vv.s.high
+		+ (unsigned long) uu.s.high * (unsigned long) vv.s.low);
+
+	return w.ll;
+}
+EXPORT_SYMBOL(__muldi3);
diff --git a/arch/microblaze/lib/mulsi3.S b/arch/microblaze/lib/mulsi3.S
new file mode 100644
index 0000000..90bd7b9
--- /dev/null
+++ b/arch/microblaze/lib/mulsi3.S
@@ -0,0 +1,46 @@
+#include <linux/linkage.h>
+
+/*
+ * Multiply operation for 32 bit integers.
+ *	Input :	Operand1 in Reg r5
+ *		Operand2 in Reg r6
+ *	Output: Result [op1 * op2] in Reg r3
+ */
+	.text
+	.globl	__mulsi3
+	.type __mulsi3,  @function
+	.ent __mulsi3
+
+__mulsi3:
+	.frame	r1, 0, r15
+	add	r3, r0, r0
+	beqi	r5, result_is_zero /* multiply by zero */
+	beqi	r6, result_is_zero /* multiply by zero */
+	bgeid	r5, r5_pos
+	xor	r4, r5, r6 /* get the sign of the result */
+	rsubi	r5, r5, 0 /* make r5 positive */
+r5_pos:
+	bgei	r6, r6_pos
+	rsubi	r6, r6, 0 /* make r6 positive */
+r6_pos:
+	bri	l1
+l2:
+	add	r5, r5, r5
+l1:
+	srl	r6, r6
+	addc	r7, r0, r0
+	beqi	r7, l2
+	bneid	r6, l2
+	add	r3, r3, r5
+	blti	r4, negateresult
+	rtsd	r15, 8
+	nop
+negateresult:
+	rtsd	r15, 8
+	rsub	r3, r3, r0
+result_is_zero:
+	rtsd	r15, 8
+	addi	r3, r0, 0
+
+.size __mulsi3,  . - __mulsi3
+.end __mulsi3
diff --git a/arch/microblaze/lib/uaccess_old.S b/arch/microblaze/lib/uaccess_old.S
new file mode 100644
index 0000000..0e8cc27
--- /dev/null
+++ b/arch/microblaze/lib/uaccess_old.S
@@ -0,0 +1,266 @@
+/*
+ * Copyright (C) 2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2009 PetaLogix
+ * Copyright (C) 2007 LynuxWorks, 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/errno.h>
+#include <linux/linkage.h>
+#include <asm/page.h>
+
+/*
+ * int __strncpy_user(char *to, char *from, int len);
+ *
+ * Returns:
+ *  -EFAULT  for an exception
+ *  len      if we hit the buffer limit
+ *  bytes copied
+ */
+
+	.text
+.globl __strncpy_user;
+.type  __strncpy_user, @function
+.align 4;
+__strncpy_user:
+
+	/*
+	 * r5 - to
+	 * r6 - from
+	 * r7 - len
+	 * r3 - temp count
+	 * r4 - temp val
+	 */
+	beqid	r7,3f
+	addik	r3,r7,0		/* temp_count = len */
+1:
+	lbu	r4,r6,r0
+	beqid	r4,2f
+	sb	r4,r5,r0
+
+	addik	r5,r5,1
+	addik	r6,r6,1		/* delay slot */
+
+	addik	r3,r3,-1
+	bnei	r3,1b		/* break on len */
+2:
+	rsubk	r3,r3,r7	/* temp_count = len - temp_count */
+3:
+	rtsd	r15,8
+	nop
+	.size   __strncpy_user, . - __strncpy_user
+
+	.section	.fixup, "ax"
+	.align	2
+4:
+	brid	3b
+	addik	r3,r0, -EFAULT
+
+	.section	__ex_table, "a"
+	.word	1b,4b
+
+/*
+ * int __strnlen_user(char __user *str, int maxlen);
+ *
+ * Returns:
+ *  0 on error
+ *  maxlen + 1  if no NUL byte found within maxlen bytes
+ *  size of the string (including NUL byte)
+ */
+
+	.text
+.globl __strnlen_user;
+.type  __strnlen_user, @function
+.align 4;
+__strnlen_user:
+	beqid	r6,3f
+	addik	r3,r6,0
+1:
+	lbu	r4,r5,r0
+	beqid	r4,2f		/* break on NUL */
+	addik	r3,r3,-1	/* delay slot */
+
+	bneid	r3,1b
+	addik	r5,r5,1		/* delay slot */
+
+	addik	r3,r3,-1	/* for break on len */
+2:
+	rsubk	r3,r3,r6
+3:
+	rtsd	r15,8
+	nop
+	.size   __strnlen_user, . - __strnlen_user
+
+	.section	.fixup,"ax"
+4:
+	brid	3b
+	addk	r3,r0,r0
+
+	.section	__ex_table,"a"
+	.word	1b,4b
+
+/* Loop unrolling for __copy_tofrom_user */
+#define COPY(offset)	\
+1:	lwi	r4 , r6, 0x0000 + offset;	\
+2:	lwi	r19, r6, 0x0004 + offset;	\
+3:	lwi	r20, r6, 0x0008 + offset;	\
+4:	lwi	r21, r6, 0x000C + offset;	\
+5:	lwi	r22, r6, 0x0010 + offset;	\
+6:	lwi	r23, r6, 0x0014 + offset;	\
+7:	lwi	r24, r6, 0x0018 + offset;	\
+8:	lwi	r25, r6, 0x001C + offset;	\
+9:	swi	r4 , r5, 0x0000 + offset;	\
+10:	swi	r19, r5, 0x0004 + offset;	\
+11:	swi	r20, r5, 0x0008 + offset;	\
+12:	swi	r21, r5, 0x000C + offset;	\
+13:	swi	r22, r5, 0x0010 + offset;	\
+14:	swi	r23, r5, 0x0014 + offset;	\
+15:	swi	r24, r5, 0x0018 + offset;	\
+16:	swi	r25, r5, 0x001C + offset;	\
+	.section __ex_table,"a";		\
+	.word	1b, 33f;			\
+	.word	2b, 33f;			\
+	.word	3b, 33f;			\
+	.word	4b, 33f;			\
+	.word	5b, 33f;			\
+	.word	6b, 33f;			\
+	.word	7b, 33f;			\
+	.word	8b, 33f;			\
+	.word	9b, 33f;			\
+	.word	10b, 33f;			\
+	.word	11b, 33f;			\
+	.word	12b, 33f;			\
+	.word	13b, 33f;			\
+	.word	14b, 33f;			\
+	.word	15b, 33f;			\
+	.word	16b, 33f;			\
+	.text
+
+#define COPY_80(offset)	\
+	COPY(0x00 + offset);\
+	COPY(0x20 + offset);\
+	COPY(0x40 + offset);\
+	COPY(0x60 + offset);
+
+/*
+ * int __copy_tofrom_user(char *to, char *from, int len)
+ * Return:
+ *   0 on success
+ *   number of not copied bytes on error
+ */
+	.text
+.globl __copy_tofrom_user;
+.type  __copy_tofrom_user, @function
+.align 4;
+__copy_tofrom_user:
+	/*
+	 * r5 - to
+	 * r6 - from
+	 * r7, r3 - count
+	 * r4 - tempval
+	 */
+	beqid	r7, 0f /* zero size is not likely */
+	or	r3, r5, r6 /* find if is any to/from unaligned */
+	or	r3, r3, r7 /* find if count is unaligned */
+	andi	r3, r3, 0x3 /* mask last 3 bits */
+	bneid	r3, bu1 /* if r3 is not zero then byte copying */
+	or	r3, r0, r0
+
+	rsubi	r3, r7, PAGE_SIZE /* detect PAGE_SIZE */
+	beqid	r3, page;
+	or	r3, r0, r0
+
+w1:	lw	r4, r6, r3 /* at least one 4 byte copy */
+w2:	sw	r4, r5, r3
+	addik	r7, r7, -4
+	bneid	r7, w1
+	addik	r3, r3, 4
+	addik	r3, r7, 0
+	rtsd	r15, 8
+	nop
+
+	.section	__ex_table,"a"
+	.word	w1, 0f;
+	.word	w2, 0f;
+	.text
+
+.align 4 /* Alignment is important to keep icache happy */
+page:	/* Create room on stack and save registers for storign values */
+	addik   r1, r1, -40
+	swi	r5, r1, 0
+	swi	r6, r1, 4
+	swi	r7, r1, 8
+	swi	r19, r1, 12
+	swi	r20, r1, 16
+	swi	r21, r1, 20
+	swi	r22, r1, 24
+	swi	r23, r1, 28
+	swi	r24, r1, 32
+	swi	r25, r1, 36
+loop:	/* r4, r19, r20, r21, r22, r23, r24, r25 are used for storing values */
+	/* Loop unrolling to get performance boost */
+	COPY_80(0x000);
+	COPY_80(0x080);
+	COPY_80(0x100);
+	COPY_80(0x180);
+	/* copy loop */
+	addik   r6, r6, 0x200
+	addik   r7, r7, -0x200
+	bneid   r7, loop
+	addik   r5, r5, 0x200
+
+	/* Restore register content */
+	lwi	r5, r1, 0
+	lwi	r6, r1, 4
+	lwi	r7, r1, 8
+	lwi	r19, r1, 12
+	lwi	r20, r1, 16
+	lwi	r21, r1, 20
+	lwi	r22, r1, 24
+	lwi	r23, r1, 28
+	lwi	r24, r1, 32
+	lwi	r25, r1, 36
+	addik   r1, r1, 40
+	/* return back */
+	addik	r3, r0, 0
+	rtsd	r15, 8
+	nop
+
+/* Fault case - return temp count */
+33:
+	addik	r3, r7, 0
+	/* Restore register content */
+	lwi	r5, r1, 0
+	lwi	r6, r1, 4
+	lwi	r7, r1, 8
+	lwi	r19, r1, 12
+	lwi	r20, r1, 16
+	lwi	r21, r1, 20
+	lwi	r22, r1, 24
+	lwi	r23, r1, 28
+	lwi	r24, r1, 32
+	lwi	r25, r1, 36
+	addik   r1, r1, 40
+	/* return back */
+	rtsd	r15, 8
+	nop
+
+.align 4 /* Alignment is important to keep icache happy */
+bu1:	lbu	r4,r6,r3
+bu2:	sb	r4,r5,r3
+	addik	r7,r7,-1
+	bneid	r7,bu1
+	addik	r3,r3,1		/* delay slot */
+0:
+	addik	r3,r7,0
+	rtsd	r15,8
+	nop
+	.size   __copy_tofrom_user, . - __copy_tofrom_user
+
+	.section	__ex_table,"a"
+	.word	bu1, 0b;
+	.word	bu2, 0b;
+	.text
diff --git a/arch/microblaze/lib/ucmpdi2.c b/arch/microblaze/lib/ucmpdi2.c
new file mode 100644
index 0000000..d05f158
--- /dev/null
+++ b/arch/microblaze/lib/ucmpdi2.c
@@ -0,0 +1,20 @@
+#include <linux/export.h>
+
+#include "libgcc.h"
+
+word_type __ucmpdi2(unsigned long long a, unsigned long long b)
+{
+	const DWunion au = {.ll = a};
+	const DWunion bu = {.ll = b};
+
+	if ((unsigned int) au.s.high < (unsigned int) bu.s.high)
+		return 0;
+	else if ((unsigned int) au.s.high > (unsigned int) bu.s.high)
+		return 2;
+	if ((unsigned int) au.s.low < (unsigned int) bu.s.low)
+		return 0;
+	else if ((unsigned int) au.s.low > (unsigned int) bu.s.low)
+		return 2;
+	return 1;
+}
+EXPORT_SYMBOL(__ucmpdi2);
diff --git a/arch/microblaze/lib/udivsi3.S b/arch/microblaze/lib/udivsi3.S
new file mode 100644
index 0000000..64cf57e
--- /dev/null
+++ b/arch/microblaze/lib/udivsi3.S
@@ -0,0 +1,84 @@
+#include <linux/linkage.h>
+
+/*
+* Unsigned divide operation.
+*	Input :	Divisor in Reg r5
+*		Dividend in Reg r6
+*	Output: Result in Reg r3
+*/
+
+	.text
+	.globl	__udivsi3
+	.type __udivsi3, @function
+	.ent __udivsi3
+
+__udivsi3:
+
+	.frame	r1, 0, r15
+
+	addik	r1, r1, -12
+	swi	r29, r1, 0
+	swi	r30, r1, 4
+	swi	r31, r1, 8
+
+	beqi	r6, div_by_zero /* div_by_zero /* division error */
+	beqid	r5, result_is_zero /* result is zero */
+	addik	r30, r0, 0 /* clear mod */
+	addik	r29, r0, 32 /* initialize the loop count */
+
+/* check if r6 and r5 are equal - if yes, return 1 */
+	rsub	r18, r5, r6
+	beqid	r18, return_here
+	addik	r3, r0, 1
+
+/* check if (uns)r6 is greater than (uns)r5. in that case, just return 0 */
+	xor	r18, r5, r6
+	bgeid	r18, 16
+	add	r3, r0, r0 /* we would anyways clear r3 */
+	blti	r6, return_here /* r6[bit 31 = 1] hence is greater */
+	bri	checkr6
+	rsub	r18, r6, r5 /* microblazecmp */
+	blti	r18, return_here
+
+/* if r6 [bit 31] is set, then return result as 1 */
+checkr6:
+	bgti	r6, div0
+	brid	return_here
+	addik	r3, r0, 1
+
+/* first part try to find the first '1' in the r5 */
+div0:
+	blti	r5, div2
+div1:
+	add	r5, r5, r5 /* left shift logical r5 */
+	bgtid	r5, div1
+	addik	r29, r29, -1
+div2:
+/* left shift logical r5 get the '1' into the carry */
+	add	r5, r5, r5
+	addc	r30, r30, r30 /* move that bit into the mod register */
+	rsub	r31, r6, r30 /* try to subtract (r30 a r6) */
+	blti	r31, mod_too_small
+/* move the r31 to mod since the result was positive */
+	or	r30, r0, r31
+	addik	r3, r3, 1
+mod_too_small:
+	addik	r29, r29, -1
+	beqi	r29, loop_end
+	add	r3, r3, r3 /* shift in the '1' into div */
+	bri	div2 /* div2 */
+loop_end:
+	bri	return_here
+div_by_zero:
+result_is_zero:
+	or	r3, r0, r0 /* set result to 0 */
+return_here:
+/* restore values of csrs and that of r3 and the divisor and the dividend */
+	lwi	r29, r1, 0
+	lwi	r30, r1, 4
+	lwi	r31, r1, 8
+	rtsd	r15, 8
+	addik	r1, r1, 12
+
+.size __udivsi3, . - __udivsi3
+.end __udivsi3
diff --git a/arch/microblaze/lib/umodsi3.S b/arch/microblaze/lib/umodsi3.S
new file mode 100644
index 0000000..17d16ba
--- /dev/null
+++ b/arch/microblaze/lib/umodsi3.S
@@ -0,0 +1,86 @@
+#include <linux/linkage.h>
+
+/*
+ * Unsigned modulo operation for 32 bit integers.
+ *	Input :	op1 in Reg r5
+ *		op2 in Reg r6
+ *	Output: op1 mod op2 in Reg r3
+ */
+
+	.text
+	.globl	__umodsi3
+	.type __umodsi3, @function
+	.ent __umodsi3
+
+__umodsi3:
+	.frame	r1, 0, r15
+
+	addik	r1, r1, -12
+	swi	r29, r1, 0
+	swi	r30, r1, 4
+	swi	r31, r1, 8
+
+	beqi	r6, div_by_zero /* div_by_zero - division error */
+	beqid	r5, result_is_zero /* result is zero */
+	addik	r3, r0, 0 /* clear div */
+	addik	r30, r0, 0 /* clear mod */
+	addik	r29, r0, 32 /* initialize the loop count */
+
+/* check if r6 and r5 are equal /* if yes, return 0 */
+	rsub	r18, r5, r6
+	beqi	r18, return_here
+
+/* check if (uns)r6 is greater than (uns)r5. in that case, just return r5 */
+	xor	r18, r5, r6
+	bgeid	r18, 16
+	addik	r3, r5, 0
+	blti	r6, return_here
+	bri	$lcheckr6
+	rsub	r18, r5, r6 /* microblazecmp */
+	bgti	r18, return_here
+
+/* if r6 [bit 31] is set, then return result as r5-r6 */
+$lcheckr6:
+	bgtid	r6, div0
+	addik	r3, r0, 0
+	addik	r18, r0, 0x7fffffff
+	and	r5, r5, r18
+	and	r6, r6, r18
+	brid	return_here
+	rsub	r3, r6, r5
+/* first part: try to find the first '1' in the r5 */
+div0:
+	blti	r5, div2
+div1:
+	add	r5, r5, r5 /* left shift logical r5 */
+	bgeid	r5, div1
+	addik	r29, r29, -1
+div2:
+	/* left shift logical r5 get the '1' into the carry */
+	add	r5, r5, r5
+	addc	r3, r3, r3 /* move that bit into the mod register */
+	rsub	r31, r6, r3 /* try to subtract (r3 a r6) */
+	blti	r31, mod_too_small
+	/* move the r31 to mod since the result was positive */
+	or	r3, r0, r31
+	addik	r30, r30, 1
+mod_too_small:
+	addik	r29, r29, -1
+	beqi	r29, loop_end
+	add	r30, r30, r30 /* shift in the '1' into div */
+	bri	div2 /* div2 */
+loop_end:
+	bri	return_here
+div_by_zero:
+result_is_zero:
+	or	r3, r0, r0 /* set result to 0 */
+return_here:
+/* restore values of csrs and that of r3 and the divisor and the dividend */
+	lwi	r29, r1, 0
+	lwi	r30, r1, 4
+	lwi	r31, r1, 8
+	rtsd	r15, 8
+	addik	r1, r1, 12
+
+.size __umodsi3, . - __umodsi3
+.end __umodsi3
diff --git a/arch/microblaze/mm/Makefile b/arch/microblaze/mm/Makefile
new file mode 100644
index 0000000..7313bd8
--- /dev/null
+++ b/arch/microblaze/mm/Makefile
@@ -0,0 +1,8 @@
+#
+# Makefile
+#
+
+obj-y := consistent.o init.o
+
+obj-$(CONFIG_MMU) += pgtable.o mmu_context.o fault.o
+obj-$(CONFIG_HIGHMEM) += highmem.o
diff --git a/arch/microblaze/mm/consistent.c b/arch/microblaze/mm/consistent.c
new file mode 100644
index 0000000..b06c3a7
--- /dev/null
+++ b/arch/microblaze/mm/consistent.c
@@ -0,0 +1,266 @@
+/*
+ * Microblaze support for cache consistent memory.
+ * Copyright (C) 2010 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2010 PetaLogix
+ * Copyright (C) 2005 John Williams <jwilliams@itee.uq.edu.au>
+ *
+ * Based on PowerPC version derived from arch/arm/mm/consistent.c
+ * Copyright (C) 2001 Dan Malek (dmalek@jlc.net)
+ * Copyright (C) 2000 Russell King
+ *
+ * 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/signal.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/ptrace.h>
+#include <linux/mman.h>
+#include <linux/mm.h>
+#include <linux/swap.h>
+#include <linux/stddef.h>
+#include <linux/vmalloc.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/bootmem.h>
+#include <linux/highmem.h>
+#include <linux/pci.h>
+#include <linux/interrupt.h>
+#include <linux/gfp.h>
+
+#include <asm/pgalloc.h>
+#include <linux/io.h>
+#include <linux/hardirq.h>
+#include <linux/mmu_context.h>
+#include <asm/mmu.h>
+#include <linux/uaccess.h>
+#include <asm/pgtable.h>
+#include <asm/cpuinfo.h>
+#include <asm/tlbflush.h>
+
+#ifndef CONFIG_MMU
+/* I have to use dcache values because I can't relate on ram size */
+# define UNCACHED_SHADOW_MASK (cpuinfo.dcache_high - cpuinfo.dcache_base + 1)
+#endif
+
+/*
+ * Consistent memory allocators. Used for DMA devices that want to
+ * share uncached memory with the processor core.
+ * My crufty no-MMU approach is simple. In the HW platform we can optionally
+ * mirror the DDR up above the processor cacheable region.  So, memory accessed
+ * in this mirror region will not be cached.  It's alloced from the same
+ * pool as normal memory, but the handle we return is shifted up into the
+ * uncached region.  This will no doubt cause big problems if memory allocated
+ * here is not also freed properly. -- JW
+ */
+void *consistent_alloc(gfp_t gfp, size_t size, dma_addr_t *dma_handle)
+{
+	unsigned long order, vaddr;
+	void *ret;
+	unsigned int i, err = 0;
+	struct page *page, *end;
+
+#ifdef CONFIG_MMU
+	phys_addr_t pa;
+	struct vm_struct *area;
+	unsigned long va;
+#endif
+
+	if (in_interrupt())
+		BUG();
+
+	/* Only allocate page size areas. */
+	size = PAGE_ALIGN(size);
+	order = get_order(size);
+
+	vaddr = __get_free_pages(gfp, order);
+	if (!vaddr)
+		return NULL;
+
+	/*
+	 * we need to ensure that there are no cachelines in use,
+	 * or worse dirty in this area.
+	 */
+	flush_dcache_range(virt_to_phys((void *)vaddr),
+					virt_to_phys((void *)vaddr) + size);
+
+#ifndef CONFIG_MMU
+	ret = (void *)vaddr;
+	/*
+	 * Here's the magic!  Note if the uncached shadow is not implemented,
+	 * it's up to the calling code to also test that condition and make
+	 * other arranegments, such as manually flushing the cache and so on.
+	 */
+# ifdef CONFIG_XILINX_UNCACHED_SHADOW
+	ret = (void *)((unsigned) ret | UNCACHED_SHADOW_MASK);
+# endif
+	if ((unsigned int)ret > cpuinfo.dcache_base &&
+				(unsigned int)ret < cpuinfo.dcache_high)
+		pr_warn("ERROR: Your cache coherent area is CACHED!!!\n");
+
+	/* dma_handle is same as physical (shadowed) address */
+	*dma_handle = (dma_addr_t)ret;
+#else
+	/* Allocate some common virtual space to map the new pages. */
+	area = get_vm_area(size, VM_ALLOC);
+	if (!area) {
+		free_pages(vaddr, order);
+		return NULL;
+	}
+	va = (unsigned long) area->addr;
+	ret = (void *)va;
+
+	/* This gives us the real physical address of the first page. */
+	*dma_handle = pa = __virt_to_phys(vaddr);
+#endif
+
+	/*
+	 * free wasted pages.  We skip the first page since we know
+	 * that it will have count = 1 and won't require freeing.
+	 * We also mark the pages in use as reserved so that
+	 * remap_page_range works.
+	 */
+	page = virt_to_page(vaddr);
+	end = page + (1 << order);
+
+	split_page(page, order);
+
+	for (i = 0; i < size && err == 0; i += PAGE_SIZE) {
+#ifdef CONFIG_MMU
+		/* MS: This is the whole magic - use cache inhibit pages */
+		err = map_page(va + i, pa + i, _PAGE_KERNEL | _PAGE_NO_CACHE);
+#endif
+
+		SetPageReserved(page);
+		page++;
+	}
+
+	/* Free the otherwise unused pages. */
+	while (page < end) {
+		__free_page(page);
+		page++;
+	}
+
+	if (err) {
+		free_pages(vaddr, order);
+		return NULL;
+	}
+
+	return ret;
+}
+EXPORT_SYMBOL(consistent_alloc);
+
+#ifdef CONFIG_MMU
+static pte_t *consistent_virt_to_pte(void *vaddr)
+{
+	unsigned long addr = (unsigned long)vaddr;
+
+	return pte_offset_kernel(pmd_offset(pgd_offset_k(addr), addr), addr);
+}
+
+unsigned long consistent_virt_to_pfn(void *vaddr)
+{
+	pte_t *ptep = consistent_virt_to_pte(vaddr);
+
+	if (pte_none(*ptep) || !pte_present(*ptep))
+		return 0;
+
+	return pte_pfn(*ptep);
+}
+#endif
+
+/*
+ * free page(s) as defined by the above mapping.
+ */
+void consistent_free(size_t size, void *vaddr)
+{
+	struct page *page;
+
+	if (in_interrupt())
+		BUG();
+
+	size = PAGE_ALIGN(size);
+
+#ifndef CONFIG_MMU
+	/* Clear SHADOW_MASK bit in address, and free as per usual */
+# ifdef CONFIG_XILINX_UNCACHED_SHADOW
+	vaddr = (void *)((unsigned)vaddr & ~UNCACHED_SHADOW_MASK);
+# endif
+	page = virt_to_page(vaddr);
+
+	do {
+		__free_reserved_page(page);
+		page++;
+	} while (size -= PAGE_SIZE);
+#else
+	do {
+		pte_t *ptep = consistent_virt_to_pte(vaddr);
+		unsigned long pfn;
+
+		if (!pte_none(*ptep) && pte_present(*ptep)) {
+			pfn = pte_pfn(*ptep);
+			pte_clear(&init_mm, (unsigned int)vaddr, ptep);
+			if (pfn_valid(pfn)) {
+				page = pfn_to_page(pfn);
+				__free_reserved_page(page);
+			}
+		}
+		vaddr += PAGE_SIZE;
+	} while (size -= PAGE_SIZE);
+
+	/* flush tlb */
+	flush_tlb_all();
+#endif
+}
+EXPORT_SYMBOL(consistent_free);
+
+/*
+ * make an area consistent.
+ */
+void consistent_sync(void *vaddr, size_t size, int direction)
+{
+	unsigned long start;
+	unsigned long end;
+
+	start = (unsigned long)vaddr;
+
+	/* Convert start address back down to unshadowed memory region */
+#ifdef CONFIG_XILINX_UNCACHED_SHADOW
+	start &= ~UNCACHED_SHADOW_MASK;
+#endif
+	end = start + size;
+
+	switch (direction) {
+	case PCI_DMA_NONE:
+		BUG();
+	case PCI_DMA_FROMDEVICE:	/* invalidate only */
+		invalidate_dcache_range(start, end);
+		break;
+	case PCI_DMA_TODEVICE:		/* writeback only */
+		flush_dcache_range(start, end);
+		break;
+	case PCI_DMA_BIDIRECTIONAL:	/* writeback and invalidate */
+		flush_dcache_range(start, end);
+		break;
+	}
+}
+EXPORT_SYMBOL(consistent_sync);
+
+/*
+ * consistent_sync_page makes memory consistent. identical
+ * to consistent_sync, but takes a struct page instead of a
+ * virtual address
+ */
+void consistent_sync_page(struct page *page, unsigned long offset,
+	size_t size, int direction)
+{
+	unsigned long start = (unsigned long)page_address(page) + offset;
+	consistent_sync((void *)start, size, direction);
+}
+EXPORT_SYMBOL(consistent_sync_page);
diff --git a/arch/microblaze/mm/fault.c b/arch/microblaze/mm/fault.c
new file mode 100644
index 0000000..177dfc0
--- /dev/null
+++ b/arch/microblaze/mm/fault.c
@@ -0,0 +1,306 @@
+/*
+ *  arch/microblaze/mm/fault.c
+ *
+ *    Copyright (C) 2007 Xilinx, Inc.  All rights reserved.
+ *
+ *  Derived from "arch/ppc/mm/fault.c"
+ *    Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org)
+ *
+ *  Derived from "arch/i386/mm/fault.c"
+ *    Copyright (C) 1991, 1992, 1993, 1994  Linus Torvalds
+ *
+ *  Modified by Cort Dougan and Paul Mackerras.
+ *
+ * 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/module.h>
+#include <linux/signal.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/ptrace.h>
+#include <linux/mman.h>
+#include <linux/mm.h>
+#include <linux/interrupt.h>
+
+#include <asm/page.h>
+#include <asm/pgtable.h>
+#include <asm/mmu.h>
+#include <linux/mmu_context.h>
+#include <linux/uaccess.h>
+#include <asm/exceptions.h>
+
+static unsigned long pte_misses;	/* updated by do_page_fault() */
+static unsigned long pte_errors;	/* updated by do_page_fault() */
+
+/*
+ * Check whether the instruction at regs->pc is a store using
+ * an update addressing form which will update r1.
+ */
+static int store_updates_sp(struct pt_regs *regs)
+{
+	unsigned int inst;
+
+	if (get_user(inst, (unsigned int __user *)regs->pc))
+		return 0;
+	/* check for 1 in the rD field */
+	if (((inst >> 21) & 0x1f) != 1)
+		return 0;
+	/* check for store opcodes */
+	if ((inst & 0xd0000000) == 0xd0000000)
+		return 1;
+	return 0;
+}
+
+
+/*
+ * bad_page_fault is called when we have a bad access from the kernel.
+ * It is called from do_page_fault above and from some of the procedures
+ * in traps.c.
+ */
+void bad_page_fault(struct pt_regs *regs, unsigned long address, int sig)
+{
+	const struct exception_table_entry *fixup;
+/* MS: no context */
+	/* Are we prepared to handle this fault?  */
+	fixup = search_exception_tables(regs->pc);
+	if (fixup) {
+		regs->pc = fixup->fixup;
+		return;
+	}
+
+	/* kernel has accessed a bad area */
+	die("kernel access of bad area", regs, sig);
+}
+
+/*
+ * The error_code parameter is ESR for a data fault,
+ * 0 for an instruction fault.
+ */
+void do_page_fault(struct pt_regs *regs, unsigned long address,
+		   unsigned long error_code)
+{
+	struct vm_area_struct *vma;
+	struct mm_struct *mm = current->mm;
+	siginfo_t info;
+	int code = SEGV_MAPERR;
+	int is_write = error_code & ESR_S;
+	int fault;
+	unsigned int flags = FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_KILLABLE;
+
+	regs->ear = address;
+	regs->esr = error_code;
+
+	/* On a kernel SLB miss we can only check for a valid exception entry */
+	if (unlikely(kernel_mode(regs) && (address >= TASK_SIZE))) {
+		pr_warn("kernel task_size exceed");
+		_exception(SIGSEGV, regs, code, address);
+	}
+
+	/* for instr TLB miss and instr storage exception ESR_S is undefined */
+	if ((error_code & 0x13) == 0x13 || (error_code & 0x11) == 0x11)
+		is_write = 0;
+
+	if (unlikely(faulthandler_disabled() || !mm)) {
+		if (kernel_mode(regs))
+			goto bad_area_nosemaphore;
+
+		/* faulthandler_disabled() in user mode is really bad,
+		   as is current->mm == NULL. */
+		pr_emerg("Page fault in user mode with faulthandler_disabled(), mm = %p\n",
+			 mm);
+		pr_emerg("r15 = %lx  MSR = %lx\n",
+		       regs->r15, regs->msr);
+		die("Weird page fault", regs, SIGSEGV);
+	}
+
+	if (user_mode(regs))
+		flags |= FAULT_FLAG_USER;
+
+	/* When running in the kernel we expect faults to occur only to
+	 * addresses in user space.  All other faults represent errors in the
+	 * kernel and should generate an OOPS.  Unfortunately, in the case of an
+	 * erroneous fault occurring in a code path which already holds mmap_sem
+	 * we will deadlock attempting to validate the fault against the
+	 * address space.  Luckily the kernel only validly references user
+	 * space from well defined areas of code, which are listed in the
+	 * exceptions table.
+	 *
+	 * As the vast majority of faults will be valid we will only perform
+	 * the source reference check when there is a possibility of a deadlock.
+	 * Attempt to lock the address space, if we cannot we then validate the
+	 * source.  If this is invalid we can skip the address space check,
+	 * thus avoiding the deadlock.
+	 */
+	if (unlikely(!down_read_trylock(&mm->mmap_sem))) {
+		if (kernel_mode(regs) && !search_exception_tables(regs->pc))
+			goto bad_area_nosemaphore;
+
+retry:
+		down_read(&mm->mmap_sem);
+	}
+
+	vma = find_vma(mm, address);
+	if (unlikely(!vma))
+		goto bad_area;
+
+	if (vma->vm_start <= address)
+		goto good_area;
+
+	if (unlikely(!(vma->vm_flags & VM_GROWSDOWN)))
+		goto bad_area;
+
+	if (unlikely(!is_write))
+		goto bad_area;
+
+	/*
+	 * N.B. The ABI allows programs to access up to
+	 * a few hundred bytes below the stack pointer (TBD).
+	 * The kernel signal delivery code writes up to about 1.5kB
+	 * below the stack pointer (r1) before decrementing it.
+	 * The exec code can write slightly over 640kB to the stack
+	 * before setting the user r1.  Thus we allow the stack to
+	 * expand to 1MB without further checks.
+	 */
+	if (unlikely(address + 0x100000 < vma->vm_end)) {
+
+		/* get user regs even if this fault is in kernel mode */
+		struct pt_regs *uregs = current->thread.regs;
+		if (uregs == NULL)
+			goto bad_area;
+
+		/*
+		 * A user-mode access to an address a long way below
+		 * the stack pointer is only valid if the instruction
+		 * is one which would update the stack pointer to the
+		 * address accessed if the instruction completed,
+		 * i.e. either stwu rs,n(r1) or stwux rs,r1,rb
+		 * (or the byte, halfword, float or double forms).
+		 *
+		 * If we don't check this then any write to the area
+		 * between the last mapped region and the stack will
+		 * expand the stack rather than segfaulting.
+		 */
+		if (address + 2048 < uregs->r1
+			&& (kernel_mode(regs) || !store_updates_sp(regs)))
+				goto bad_area;
+	}
+	if (expand_stack(vma, address))
+		goto bad_area;
+
+good_area:
+	code = SEGV_ACCERR;
+
+	/* a write */
+	if (unlikely(is_write)) {
+		if (unlikely(!(vma->vm_flags & VM_WRITE)))
+			goto bad_area;
+		flags |= FAULT_FLAG_WRITE;
+	/* a read */
+	} else {
+		/* protection fault */
+		if (unlikely(error_code & 0x08000000))
+			goto bad_area;
+		if (unlikely(!(vma->vm_flags & (VM_READ | VM_EXEC))))
+			goto bad_area;
+	}
+
+	/*
+	 * If for any reason at all we couldn't handle the fault,
+	 * make sure we exit gracefully rather than endlessly redo
+	 * the fault.
+	 */
+	fault = handle_mm_fault(mm, vma, address, flags);
+
+	if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current))
+		return;
+
+	if (unlikely(fault & VM_FAULT_ERROR)) {
+		if (fault & VM_FAULT_OOM)
+			goto out_of_memory;
+		else if (fault & VM_FAULT_SIGSEGV)
+			goto bad_area;
+		else if (fault & VM_FAULT_SIGBUS)
+			goto do_sigbus;
+		BUG();
+	}
+
+	if (flags & FAULT_FLAG_ALLOW_RETRY) {
+		if (unlikely(fault & VM_FAULT_MAJOR))
+			current->maj_flt++;
+		else
+			current->min_flt++;
+		if (fault & VM_FAULT_RETRY) {
+			flags &= ~FAULT_FLAG_ALLOW_RETRY;
+			flags |= FAULT_FLAG_TRIED;
+
+			/*
+			 * No need to up_read(&mm->mmap_sem) as we would
+			 * have already released it in __lock_page_or_retry
+			 * in mm/filemap.c.
+			 */
+
+			goto retry;
+		}
+	}
+
+	up_read(&mm->mmap_sem);
+
+	/*
+	 * keep track of tlb+htab misses that are good addrs but
+	 * just need pte's created via handle_mm_fault()
+	 * -- Cort
+	 */
+	pte_misses++;
+	return;
+
+bad_area:
+	up_read(&mm->mmap_sem);
+
+bad_area_nosemaphore:
+	pte_errors++;
+
+	/* User mode accesses cause a SIGSEGV */
+	if (user_mode(regs)) {
+		_exception(SIGSEGV, regs, code, address);
+/*		info.si_signo = SIGSEGV;
+		info.si_errno = 0;
+		info.si_code = code;
+		info.si_addr = (void *) address;
+		force_sig_info(SIGSEGV, &info, current);*/
+		return;
+	}
+
+	bad_page_fault(regs, address, SIGSEGV);
+	return;
+
+/*
+ * We ran out of memory, or some other thing happened to us that made
+ * us unable to handle the page fault gracefully.
+ */
+out_of_memory:
+	up_read(&mm->mmap_sem);
+	if (!user_mode(regs))
+		bad_page_fault(regs, address, SIGKILL);
+	else
+		pagefault_out_of_memory();
+	return;
+
+do_sigbus:
+	up_read(&mm->mmap_sem);
+	if (user_mode(regs)) {
+		info.si_signo = SIGBUS;
+		info.si_errno = 0;
+		info.si_code = BUS_ADRERR;
+		info.si_addr = (void __user *)address;
+		force_sig_info(SIGBUS, &info, current);
+		return;
+	}
+	bad_page_fault(regs, address, SIGBUS);
+}
diff --git a/arch/microblaze/mm/highmem.c b/arch/microblaze/mm/highmem.c
new file mode 100644
index 0000000..2fcc5a5
--- /dev/null
+++ b/arch/microblaze/mm/highmem.c
@@ -0,0 +1,90 @@
+/*
+ * highmem.c: virtual kernel memory mappings for high memory
+ *
+ * PowerPC version, stolen from the i386 version.
+ *
+ * Used in CONFIG_HIGHMEM systems for memory pages which
+ * are not addressable by direct kernel virtual addresses.
+ *
+ * Copyright (C) 1999 Gerhard Wichert, Siemens AG
+ *		      Gerhard.Wichert@pdb.siemens.de
+ *
+ *
+ * Redesigned the x86 32-bit VM architecture to deal with
+ * up to 16 Terrabyte physical memory. With current x86 CPUs
+ * we now support up to 64 Gigabytes physical RAM.
+ *
+ * Copyright (C) 1999 Ingo Molnar <mingo@redhat.com>
+ *
+ * Reworked for PowerPC by various contributors. Moved from
+ * highmem.h by Benjamin Herrenschmidt (c) 2009 IBM Corp.
+ */
+
+#include <linux/export.h>
+#include <linux/highmem.h>
+
+/*
+ * The use of kmap_atomic/kunmap_atomic is discouraged - kmap/kunmap
+ * gives a more generic (and caching) interface. But kmap_atomic can
+ * be used in IRQ contexts, so in some (very limited) cases we need
+ * it.
+ */
+#include <asm/tlbflush.h>
+
+void *kmap_atomic_prot(struct page *page, pgprot_t prot)
+{
+
+	unsigned long vaddr;
+	int idx, type;
+
+	preempt_disable();
+	pagefault_disable();
+	if (!PageHighMem(page))
+		return page_address(page);
+
+
+	type = kmap_atomic_idx_push();
+	idx = type + KM_TYPE_NR*smp_processor_id();
+	vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
+#ifdef CONFIG_DEBUG_HIGHMEM
+	BUG_ON(!pte_none(*(kmap_pte-idx)));
+#endif
+	set_pte_at(&init_mm, vaddr, kmap_pte-idx, mk_pte(page, prot));
+	local_flush_tlb_page(NULL, vaddr);
+
+	return (void *) vaddr;
+}
+EXPORT_SYMBOL(kmap_atomic_prot);
+
+void __kunmap_atomic(void *kvaddr)
+{
+	unsigned long vaddr = (unsigned long) kvaddr & PAGE_MASK;
+	int type;
+
+	if (vaddr < __fix_to_virt(FIX_KMAP_END)) {
+		pagefault_enable();
+		preempt_enable();
+		return;
+	}
+
+	type = kmap_atomic_idx();
+#ifdef CONFIG_DEBUG_HIGHMEM
+	{
+		unsigned int idx;
+
+		idx = type + KM_TYPE_NR * smp_processor_id();
+		BUG_ON(vaddr != __fix_to_virt(FIX_KMAP_BEGIN + idx));
+
+		/*
+		 * force other mappings to Oops if they'll try to access
+		 * this pte without first remap it
+		 */
+		pte_clear(&init_mm, vaddr, kmap_pte-idx);
+		local_flush_tlb_page(NULL, vaddr);
+	}
+#endif
+	kmap_atomic_idx_pop();
+	pagefault_enable();
+	preempt_enable();
+}
+EXPORT_SYMBOL(__kunmap_atomic);
diff --git a/arch/microblaze/mm/init.c b/arch/microblaze/mm/init.c
new file mode 100644
index 0000000..77bc7c7
--- /dev/null
+++ b/arch/microblaze/mm/init.c
@@ -0,0 +1,437 @@
+/*
+ * Copyright (C) 2007-2008 Michal Simek <monstr@monstr.eu>
+ * 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/bootmem.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/memblock.h>
+#include <linux/mm.h> /* mem_init */
+#include <linux/initrd.h>
+#include <linux/pagemap.h>
+#include <linux/pfn.h>
+#include <linux/slab.h>
+#include <linux/swap.h>
+#include <linux/export.h>
+
+#include <asm/page.h>
+#include <asm/mmu_context.h>
+#include <asm/pgalloc.h>
+#include <asm/sections.h>
+#include <asm/tlb.h>
+#include <asm/fixmap.h>
+
+/* Use for MMU and noMMU because of PCI generic code */
+int mem_init_done;
+
+#ifndef CONFIG_MMU
+unsigned int __page_offset;
+EXPORT_SYMBOL(__page_offset);
+
+#else
+static int init_bootmem_done;
+#endif /* CONFIG_MMU */
+
+char *klimit = _end;
+
+/*
+ * Initialize the bootmem system and give it all the memory we
+ * have available.
+ */
+unsigned long memory_start;
+EXPORT_SYMBOL(memory_start);
+unsigned long memory_size;
+EXPORT_SYMBOL(memory_size);
+unsigned long lowmem_size;
+
+#ifdef CONFIG_HIGHMEM
+pte_t *kmap_pte;
+EXPORT_SYMBOL(kmap_pte);
+pgprot_t kmap_prot;
+EXPORT_SYMBOL(kmap_prot);
+
+static inline pte_t *virt_to_kpte(unsigned long vaddr)
+{
+	return pte_offset_kernel(pmd_offset(pgd_offset_k(vaddr),
+			vaddr), vaddr);
+}
+
+static void __init highmem_init(void)
+{
+	pr_debug("%x\n", (u32)PKMAP_BASE);
+	map_page(PKMAP_BASE, 0, 0);	/* XXX gross */
+	pkmap_page_table = virt_to_kpte(PKMAP_BASE);
+
+	kmap_pte = virt_to_kpte(__fix_to_virt(FIX_KMAP_BEGIN));
+	kmap_prot = PAGE_KERNEL;
+}
+
+static void highmem_setup(void)
+{
+	unsigned long pfn;
+
+	for (pfn = max_low_pfn; pfn < max_pfn; ++pfn) {
+		struct page *page = pfn_to_page(pfn);
+
+		/* FIXME not sure about */
+		if (!memblock_is_reserved(pfn << PAGE_SHIFT))
+			free_highmem_page(page);
+	}
+}
+#endif /* CONFIG_HIGHMEM */
+
+/*
+ * paging_init() sets up the page tables - in fact we've already done this.
+ */
+static void __init paging_init(void)
+{
+	unsigned long zones_size[MAX_NR_ZONES];
+#ifdef CONFIG_MMU
+	int idx;
+
+	/* Setup fixmaps */
+	for (idx = 0; idx < __end_of_fixed_addresses; idx++)
+		clear_fixmap(idx);
+#endif
+
+	/* Clean every zones */
+	memset(zones_size, 0, sizeof(zones_size));
+
+#ifdef CONFIG_HIGHMEM
+	highmem_init();
+
+	zones_size[ZONE_DMA] = max_low_pfn;
+	zones_size[ZONE_HIGHMEM] = max_pfn;
+#else
+	zones_size[ZONE_DMA] = max_pfn;
+#endif
+
+	/* We don't have holes in memory map */
+	free_area_init_nodes(zones_size);
+}
+
+void __init setup_memory(void)
+{
+	unsigned long map_size;
+	struct memblock_region *reg;
+
+#ifndef CONFIG_MMU
+	u32 kernel_align_start, kernel_align_size;
+
+	/* Find main memory where is the kernel */
+	for_each_memblock(memory, reg) {
+		memory_start = (u32)reg->base;
+		lowmem_size = reg->size;
+		if ((memory_start <= (u32)_text) &&
+			((u32)_text <= (memory_start + lowmem_size - 1))) {
+			memory_size = lowmem_size;
+			PAGE_OFFSET = memory_start;
+			pr_info("%s: Main mem: 0x%x, size 0x%08x\n",
+				__func__, (u32) memory_start,
+					(u32) memory_size);
+			break;
+		}
+	}
+
+	if (!memory_start || !memory_size) {
+		panic("%s: Missing memory setting 0x%08x, size=0x%08x\n",
+			__func__, (u32) memory_start, (u32) memory_size);
+	}
+
+	/* reservation of region where is the kernel */
+	kernel_align_start = PAGE_DOWN((u32)_text);
+	/* ALIGN can be remove because _end in vmlinux.lds.S is align */
+	kernel_align_size = PAGE_UP((u32)klimit) - kernel_align_start;
+	pr_info("%s: kernel addr:0x%08x-0x%08x size=0x%08x\n",
+		__func__, kernel_align_start, kernel_align_start
+			+ kernel_align_size, kernel_align_size);
+	memblock_reserve(kernel_align_start, kernel_align_size);
+#endif
+	/*
+	 * Kernel:
+	 * start: base phys address of kernel - page align
+	 * end: base phys address of kernel - page align
+	 *
+	 * min_low_pfn - the first page (mm/bootmem.c - node_boot_start)
+	 * max_low_pfn
+	 * max_mapnr - the first unused page (mm/bootmem.c - node_low_pfn)
+	 */
+
+	/* memory start is from the kernel end (aligned) to higher addr */
+	min_low_pfn = memory_start >> PAGE_SHIFT; /* minimum for allocation */
+	/* RAM is assumed contiguous */
+	max_mapnr = memory_size >> PAGE_SHIFT;
+	max_low_pfn = ((u64)memory_start + (u64)lowmem_size) >> PAGE_SHIFT;
+	max_pfn = ((u64)memory_start + (u64)memory_size) >> PAGE_SHIFT;
+
+	pr_info("%s: max_mapnr: %#lx\n", __func__, max_mapnr);
+	pr_info("%s: min_low_pfn: %#lx\n", __func__, min_low_pfn);
+	pr_info("%s: max_low_pfn: %#lx\n", __func__, max_low_pfn);
+	pr_info("%s: max_pfn: %#lx\n", __func__, max_pfn);
+
+	/*
+	 * Find an area to use for the bootmem bitmap.
+	 * We look for the first area which is at least
+	 * 128kB in length (128kB is enough for a bitmap
+	 * for 4GB of memory, using 4kB pages), plus 1 page
+	 * (in case the address isn't page-aligned).
+	 */
+	map_size = init_bootmem_node(NODE_DATA(0),
+		PFN_UP(TOPHYS((u32)klimit)), min_low_pfn, max_low_pfn);
+	memblock_reserve(PFN_UP(TOPHYS((u32)klimit)) << PAGE_SHIFT, map_size);
+
+	/* Add active regions with valid PFNs */
+	for_each_memblock(memory, reg) {
+		unsigned long start_pfn, end_pfn;
+
+		start_pfn = memblock_region_memory_base_pfn(reg);
+		end_pfn = memblock_region_memory_end_pfn(reg);
+		memblock_set_node(start_pfn << PAGE_SHIFT,
+				  (end_pfn - start_pfn) << PAGE_SHIFT,
+				  &memblock.memory, 0);
+	}
+
+	/* free bootmem is whole main memory */
+	free_bootmem_with_active_regions(0, max_low_pfn);
+
+	/* reserve allocate blocks */
+	for_each_memblock(reserved, reg) {
+		unsigned long top = reg->base + reg->size - 1;
+
+		pr_debug("reserved - 0x%08x-0x%08x, %lx, %lx\n",
+			 (u32) reg->base, (u32) reg->size, top,
+						memory_start + lowmem_size - 1);
+
+		if (top <= (memory_start + lowmem_size - 1)) {
+			reserve_bootmem(reg->base, reg->size, BOOTMEM_DEFAULT);
+		} else if (reg->base < (memory_start + lowmem_size - 1)) {
+			unsigned long trunc_size = memory_start + lowmem_size -
+								reg->base;
+			reserve_bootmem(reg->base, trunc_size, BOOTMEM_DEFAULT);
+		}
+	}
+
+	/* XXX need to clip this if using highmem? */
+	sparse_memory_present_with_active_regions(0);
+
+#ifdef CONFIG_MMU
+	init_bootmem_done = 1;
+#endif
+	paging_init();
+}
+
+#ifdef CONFIG_BLK_DEV_INITRD
+void free_initrd_mem(unsigned long start, unsigned long end)
+{
+	free_reserved_area((void *)start, (void *)end, -1, "initrd");
+}
+#endif
+
+void free_initmem(void)
+{
+	free_initmem_default(-1);
+}
+
+void __init mem_init(void)
+{
+	high_memory = (void *)__va(memory_start + lowmem_size - 1);
+
+	/* this will put all memory onto the freelists */
+	free_all_bootmem();
+#ifdef CONFIG_HIGHMEM
+	highmem_setup();
+#endif
+
+	mem_init_print_info(NULL);
+#ifdef CONFIG_MMU
+	pr_info("Kernel virtual memory layout:\n");
+	pr_info("  * 0x%08lx..0x%08lx  : fixmap\n", FIXADDR_START, FIXADDR_TOP);
+#ifdef CONFIG_HIGHMEM
+	pr_info("  * 0x%08lx..0x%08lx  : highmem PTEs\n",
+		PKMAP_BASE, PKMAP_ADDR(LAST_PKMAP));
+#endif /* CONFIG_HIGHMEM */
+	pr_info("  * 0x%08lx..0x%08lx  : early ioremap\n",
+		ioremap_bot, ioremap_base);
+	pr_info("  * 0x%08lx..0x%08lx  : vmalloc & ioremap\n",
+		(unsigned long)VMALLOC_START, VMALLOC_END);
+#endif
+	mem_init_done = 1;
+}
+
+#ifndef CONFIG_MMU
+int page_is_ram(unsigned long pfn)
+{
+	return __range_ok(pfn, 0);
+}
+#else
+int page_is_ram(unsigned long pfn)
+{
+	return pfn < max_low_pfn;
+}
+
+/*
+ * Check for command-line options that affect what MMU_init will do.
+ */
+static void mm_cmdline_setup(void)
+{
+	unsigned long maxmem = 0;
+	char *p = cmd_line;
+
+	/* Look for mem= option on command line */
+	p = strstr(cmd_line, "mem=");
+	if (p) {
+		p += 4;
+		maxmem = memparse(p, &p);
+		if (maxmem && memory_size > maxmem) {
+			memory_size = maxmem;
+			memblock.memory.regions[0].size = memory_size;
+		}
+	}
+}
+
+/*
+ * MMU_init_hw does the chip-specific initialization of the MMU hardware.
+ */
+static void __init mmu_init_hw(void)
+{
+	/*
+	 * The Zone Protection Register (ZPR) defines how protection will
+	 * be applied to every page which is a member of a given zone. At
+	 * present, we utilize only two of the zones.
+	 * The zone index bits (of ZSEL) in the PTE are used for software
+	 * indicators, except the LSB.  For user access, zone 1 is used,
+	 * for kernel access, zone 0 is used.  We set all but zone 1
+	 * to zero, allowing only kernel access as indicated in the PTE.
+	 * For zone 1, we set a 01 binary (a value of 10 will not work)
+	 * to allow user access as indicated in the PTE.  This also allows
+	 * kernel access as indicated in the PTE.
+	 */
+	__asm__ __volatile__ ("ori r11, r0, 0x10000000;" \
+			"mts rzpr, r11;"
+			: : : "r11");
+}
+
+/*
+ * MMU_init sets up the basic memory mappings for the kernel,
+ * including both RAM and possibly some I/O regions,
+ * and sets up the page tables and the MMU hardware ready to go.
+ */
+
+/* called from head.S */
+asmlinkage void __init mmu_init(void)
+{
+	unsigned int kstart, ksize;
+
+	if (!memblock.reserved.cnt) {
+		pr_emerg("Error memory count\n");
+		machine_restart(NULL);
+	}
+
+	if ((u32) memblock.memory.regions[0].size < 0x400000) {
+		pr_emerg("Memory must be greater than 4MB\n");
+		machine_restart(NULL);
+	}
+
+	if ((u32) memblock.memory.regions[0].size < kernel_tlb) {
+		pr_emerg("Kernel size is greater than memory node\n");
+		machine_restart(NULL);
+	}
+
+	/* Find main memory where the kernel is */
+	memory_start = (u32) memblock.memory.regions[0].base;
+	lowmem_size = memory_size = (u32) memblock.memory.regions[0].size;
+
+	if (lowmem_size > CONFIG_LOWMEM_SIZE) {
+		lowmem_size = CONFIG_LOWMEM_SIZE;
+#ifndef CONFIG_HIGHMEM
+		memory_size = lowmem_size;
+#endif
+	}
+
+	mm_cmdline_setup(); /* FIXME parse args from command line - not used */
+
+	/*
+	 * Map out the kernel text/data/bss from the available physical
+	 * memory.
+	 */
+	kstart = __pa(CONFIG_KERNEL_START); /* kernel start */
+	/* kernel size */
+	ksize = PAGE_ALIGN(((u32)_end - (u32)CONFIG_KERNEL_START));
+	memblock_reserve(kstart, ksize);
+
+#if defined(CONFIG_BLK_DEV_INITRD)
+	/* Remove the init RAM disk from the available memory. */
+	if (initrd_start) {
+		unsigned long size;
+		size = initrd_end - initrd_start;
+		memblock_reserve(__virt_to_phys(initrd_start), size);
+	}
+#endif /* CONFIG_BLK_DEV_INITRD */
+
+	/* Initialize the MMU hardware */
+	mmu_init_hw();
+
+	/* Map in all of RAM starting at CONFIG_KERNEL_START */
+	mapin_ram();
+
+	/* Extend vmalloc and ioremap area as big as possible */
+#ifdef CONFIG_HIGHMEM
+	ioremap_base = ioremap_bot = PKMAP_BASE;
+#else
+	ioremap_base = ioremap_bot = FIXADDR_START;
+#endif
+
+	/* Initialize the context management stuff */
+	mmu_context_init();
+
+	/* Shortly after that, the entire linear mapping will be available */
+	/* This will also cause that unflatten device tree will be allocated
+	 * inside 768MB limit */
+	memblock_set_current_limit(memory_start + lowmem_size - 1);
+}
+
+/* This is only called until mem_init is done. */
+void __init *early_get_page(void)
+{
+	void *p;
+	if (init_bootmem_done) {
+		p = alloc_bootmem_pages(PAGE_SIZE);
+	} else {
+		/*
+		 * Mem start + kernel_tlb -> here is limit
+		 * because of mem mapping from head.S
+		 */
+		p = __va(memblock_alloc_base(PAGE_SIZE, PAGE_SIZE,
+					memory_start + kernel_tlb));
+	}
+	return p;
+}
+
+#endif /* CONFIG_MMU */
+
+void * __init_refok alloc_maybe_bootmem(size_t size, gfp_t mask)
+{
+	if (mem_init_done)
+		return kmalloc(size, mask);
+	else
+		return alloc_bootmem(size);
+}
+
+void * __init_refok zalloc_maybe_bootmem(size_t size, gfp_t mask)
+{
+	void *p;
+
+	if (mem_init_done)
+		p = kzalloc(size, mask);
+	else {
+		p = alloc_bootmem(size);
+		if (p)
+			memset(p, 0, size);
+	}
+	return p;
+}
diff --git a/arch/microblaze/mm/mmu_context.c b/arch/microblaze/mm/mmu_context.c
new file mode 100644
index 0000000..26ff82f
--- /dev/null
+++ b/arch/microblaze/mm/mmu_context.c
@@ -0,0 +1,70 @@
+/*
+ * This file contains the routines for handling the MMU.
+ *
+ *    Copyright (C) 2007 Xilinx, Inc.  All rights reserved.
+ *
+ *  Derived from arch/ppc/mm/4xx_mmu.c:
+ *  -- paulus
+ *
+ *  Derived from arch/ppc/mm/init.c:
+ *    Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org)
+ *
+ *  Modifications by Paul Mackerras (PowerMac) (paulus@cs.anu.edu.au)
+ *  and Cort Dougan (PReP) (cort@cs.nmt.edu)
+ *    Copyright (C) 1996 Paul Mackerras
+ *  Amiga/APUS changes by Jesper Skov (jskov@cygnus.co.uk).
+ *
+ *  Derived from "arch/i386/mm/init.c"
+ *    Copyright (C) 1991, 1992, 1993, 1994  Linus Torvalds
+ *
+ *  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 <linux/mm.h>
+#include <linux/init.h>
+
+#include <asm/tlbflush.h>
+#include <asm/mmu_context.h>
+
+mm_context_t next_mmu_context;
+unsigned long context_map[LAST_CONTEXT / BITS_PER_LONG + 1];
+atomic_t nr_free_contexts;
+struct mm_struct *context_mm[LAST_CONTEXT+1];
+
+/*
+ * Initialize the context management stuff.
+ */
+void __init mmu_context_init(void)
+{
+	/*
+	 * The use of context zero is reserved for the kernel.
+	 * This code assumes FIRST_CONTEXT < 32.
+	 */
+	context_map[0] = (1 << FIRST_CONTEXT) - 1;
+	next_mmu_context = FIRST_CONTEXT;
+	atomic_set(&nr_free_contexts, LAST_CONTEXT - FIRST_CONTEXT + 1);
+}
+
+/*
+ * Steal a context from a task that has one at the moment.
+ *
+ * This isn't an LRU system, it just frees up each context in
+ * turn (sort-of pseudo-random replacement :).  This would be the
+ * place to implement an LRU scheme if anyone were motivated to do it.
+ */
+void steal_context(void)
+{
+	struct mm_struct *mm;
+
+	/* free up context `next_mmu_context' */
+	/* if we shouldn't free context 0, don't... */
+	if (next_mmu_context < FIRST_CONTEXT)
+		next_mmu_context = FIRST_CONTEXT;
+	mm = context_mm[next_mmu_context];
+	flush_tlb_mm(mm);
+	destroy_context(mm);
+}
diff --git a/arch/microblaze/mm/pgtable.c b/arch/microblaze/mm/pgtable.c
new file mode 100644
index 0000000..4f4520e
--- /dev/null
+++ b/arch/microblaze/mm/pgtable.c
@@ -0,0 +1,260 @@
+/*
+ *  This file contains the routines setting up the linux page tables.
+ *
+ * Copyright (C) 2008 Michal Simek
+ * Copyright (C) 2008 PetaLogix
+ *
+ *    Copyright (C) 2007 Xilinx, Inc.  All rights reserved.
+ *
+ *  Derived from arch/ppc/mm/pgtable.c:
+ *    -- paulus
+ *
+ *  Derived from arch/ppc/mm/init.c:
+ *    Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org)
+ *
+ *  Modifications by Paul Mackerras (PowerMac) (paulus@cs.anu.edu.au)
+ *  and Cort Dougan (PReP) (cort@cs.nmt.edu)
+ *    Copyright (C) 1996 Paul Mackerras
+ *  Amiga/APUS changes by Jesper Skov (jskov@cygnus.co.uk).
+ *
+ *  Derived from "arch/i386/mm/init.c"
+ *    Copyright (C) 1991, 1992, 1993, 1994  Linus Torvalds
+ *
+ *  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/types.h>
+#include <linux/vmalloc.h>
+#include <linux/init.h>
+
+#include <asm/pgtable.h>
+#include <asm/pgalloc.h>
+#include <linux/io.h>
+#include <asm/mmu.h>
+#include <asm/sections.h>
+#include <asm/fixmap.h>
+
+unsigned long ioremap_base;
+unsigned long ioremap_bot;
+EXPORT_SYMBOL(ioremap_bot);
+
+#ifndef CONFIG_SMP
+struct pgtable_cache_struct quicklists;
+#endif
+
+static void __iomem *__ioremap(phys_addr_t addr, unsigned long size,
+		unsigned long flags)
+{
+	unsigned long v, i;
+	phys_addr_t p;
+	int err;
+
+	/*
+	 * Choose an address to map it to.
+	 * Once the vmalloc system is running, we use it.
+	 * Before then, we use space going down from ioremap_base
+	 * (ioremap_bot records where we're up to).
+	 */
+	p = addr & PAGE_MASK;
+	size = PAGE_ALIGN(addr + size) - p;
+
+	/*
+	 * Don't allow anybody to remap normal RAM that we're using.
+	 * mem_init() sets high_memory so only do the check after that.
+	 *
+	 * However, allow remap of rootfs: TBD
+	 */
+
+	if (mem_init_done &&
+		p >= memory_start && p < virt_to_phys(high_memory) &&
+		!(p >= __virt_to_phys((phys_addr_t)__bss_stop) &&
+		p < __virt_to_phys((phys_addr_t)__bss_stop))) {
+		pr_warn("__ioremap(): phys addr "PTE_FMT" is RAM lr %pf\n",
+			(unsigned long)p, __builtin_return_address(0));
+		return NULL;
+	}
+
+	if (size == 0)
+		return NULL;
+
+	/*
+	 * Is it already mapped? If the whole area is mapped then we're
+	 * done, otherwise remap it since we want to keep the virt addrs for
+	 * each request contiguous.
+	 *
+	 * We make the assumption here that if the bottom and top
+	 * of the range we want are mapped then it's mapped to the
+	 * same virt address (and this is contiguous).
+	 *  -- Cort
+	 */
+
+	if (mem_init_done) {
+		struct vm_struct *area;
+		area = get_vm_area(size, VM_IOREMAP);
+		if (area == NULL)
+			return NULL;
+		v = (unsigned long) area->addr;
+	} else {
+		v = (ioremap_bot -= size);
+	}
+
+	if ((flags & _PAGE_PRESENT) == 0)
+		flags |= _PAGE_KERNEL;
+	if (flags & _PAGE_NO_CACHE)
+		flags |= _PAGE_GUARDED;
+
+	err = 0;
+	for (i = 0; i < size && err == 0; i += PAGE_SIZE)
+		err = map_page(v + i, p + i, flags);
+	if (err) {
+		if (mem_init_done)
+			vfree((void *)v);
+		return NULL;
+	}
+
+	return (void __iomem *) (v + ((unsigned long)addr & ~PAGE_MASK));
+}
+
+void __iomem *ioremap(phys_addr_t addr, unsigned long size)
+{
+	return __ioremap(addr, size, _PAGE_NO_CACHE);
+}
+EXPORT_SYMBOL(ioremap);
+
+void iounmap(void __iomem *addr)
+{
+	if ((__force void *)addr > high_memory &&
+					(unsigned long) addr < ioremap_bot)
+		vfree((void *) (PAGE_MASK & (unsigned long) addr));
+}
+EXPORT_SYMBOL(iounmap);
+
+
+int map_page(unsigned long va, phys_addr_t pa, int flags)
+{
+	pmd_t *pd;
+	pte_t *pg;
+	int err = -ENOMEM;
+	/* Use upper 10 bits of VA to index the first level map */
+	pd = pmd_offset(pgd_offset_k(va), va);
+	/* Use middle 10 bits of VA to index the second-level map */
+	pg = pte_alloc_kernel(pd, va); /* from powerpc - pgtable.c */
+	/* pg = pte_alloc_kernel(&init_mm, pd, va); */
+
+	if (pg != NULL) {
+		err = 0;
+		set_pte_at(&init_mm, va, pg, pfn_pte(pa >> PAGE_SHIFT,
+				__pgprot(flags)));
+		if (unlikely(mem_init_done))
+			_tlbie(va);
+	}
+	return err;
+}
+
+/*
+ * Map in all of physical memory starting at CONFIG_KERNEL_START.
+ */
+void __init mapin_ram(void)
+{
+	unsigned long v, p, s, f;
+
+	v = CONFIG_KERNEL_START;
+	p = memory_start;
+	for (s = 0; s < lowmem_size; s += PAGE_SIZE) {
+		f = _PAGE_PRESENT | _PAGE_ACCESSED |
+				_PAGE_SHARED | _PAGE_HWEXEC;
+		if ((char *) v < _stext || (char *) v >= _etext)
+			f |= _PAGE_WRENABLE;
+		else
+			/* On the MicroBlaze, no user access
+			   forces R/W kernel access */
+			f |= _PAGE_USER;
+		map_page(v, p, f);
+		v += PAGE_SIZE;
+		p += PAGE_SIZE;
+	}
+}
+
+/* is x a power of 2? */
+#define is_power_of_2(x)	((x) != 0 && (((x) & ((x) - 1)) == 0))
+
+/* Scan the real Linux page tables and return a PTE pointer for
+ * a virtual address in a context.
+ * Returns true (1) if PTE was found, zero otherwise.  The pointer to
+ * the PTE pointer is unmodified if PTE is not found.
+ */
+static int get_pteptr(struct mm_struct *mm, unsigned long addr, pte_t **ptep)
+{
+	pgd_t	*pgd;
+	pmd_t	*pmd;
+	pte_t	*pte;
+	int     retval = 0;
+
+	pgd = pgd_offset(mm, addr & PAGE_MASK);
+	if (pgd) {
+		pmd = pmd_offset(pgd, addr & PAGE_MASK);
+		if (pmd_present(*pmd)) {
+			pte = pte_offset_kernel(pmd, addr & PAGE_MASK);
+			if (pte) {
+				retval = 1;
+				*ptep = pte;
+			}
+		}
+	}
+	return retval;
+}
+
+/* Find physical address for this virtual address.  Normally used by
+ * I/O functions, but anyone can call it.
+ */
+unsigned long iopa(unsigned long addr)
+{
+	unsigned long pa;
+
+	pte_t *pte;
+	struct mm_struct *mm;
+
+	/* Allow mapping of user addresses (within the thread)
+	 * for DMA if necessary.
+	 */
+	if (addr < TASK_SIZE)
+		mm = current->mm;
+	else
+		mm = &init_mm;
+
+	pa = 0;
+	if (get_pteptr(mm, addr, &pte))
+		pa = (pte_val(*pte) & PAGE_MASK) | (addr & ~PAGE_MASK);
+
+	return pa;
+}
+
+__init_refok pte_t *pte_alloc_one_kernel(struct mm_struct *mm,
+		unsigned long address)
+{
+	pte_t *pte;
+	if (mem_init_done) {
+		pte = (pte_t *)__get_free_page(GFP_KERNEL |
+					__GFP_REPEAT | __GFP_ZERO);
+	} else {
+		pte = (pte_t *)early_get_page();
+		if (pte)
+			clear_page(pte);
+	}
+	return pte;
+}
+
+void __set_fixmap(enum fixed_addresses idx, phys_addr_t phys, pgprot_t flags)
+{
+	unsigned long address = __fix_to_virt(idx);
+
+	if (idx >= __end_of_fixed_addresses)
+		BUG();
+
+	map_page(address, phys, pgprot_val(flags));
+}
diff --git a/arch/microblaze/oprofile/Makefile b/arch/microblaze/oprofile/Makefile
new file mode 100644
index 0000000..0d0348c
--- /dev/null
+++ b/arch/microblaze/oprofile/Makefile
@@ -0,0 +1,13 @@
+#
+# arch/microblaze/oprofile/Makefile
+#
+
+obj-$(CONFIG_OPROFILE) += oprofile.o
+
+DRIVER_OBJS := $(addprefix ../../../drivers/oprofile/, \
+		oprof.o cpu_buffer.o buffer_sync.o \
+		event_buffer.o oprofile_files.o \
+		oprofilefs.o oprofile_stats.o \
+		timer_int.o )
+
+oprofile-y := $(DRIVER_OBJS) microblaze_oprofile.o
diff --git a/arch/microblaze/oprofile/microblaze_oprofile.c b/arch/microblaze/oprofile/microblaze_oprofile.c
new file mode 100644
index 0000000..def17e5
--- /dev/null
+++ b/arch/microblaze/oprofile/microblaze_oprofile.c
@@ -0,0 +1,22 @@
+/*
+ * Microblaze oprofile code
+ *
+ * Copyright (C) 2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2009 PetaLogix
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/oprofile.h>
+#include <linux/init.h>
+
+int __init oprofile_arch_init(struct oprofile_operations *ops)
+{
+	return -1;
+}
+
+void oprofile_arch_exit(void)
+{
+}
diff --git a/arch/microblaze/pci/Makefile b/arch/microblaze/pci/Makefile
new file mode 100644
index 0000000..d1114fb
--- /dev/null
+++ b/arch/microblaze/pci/Makefile
@@ -0,0 +1,6 @@
+#
+# Makefile
+#
+
+obj-$(CONFIG_PCI)		+= pci-common.o indirect_pci.o iomap.o
+obj-$(CONFIG_PCI_XILINX)	+= xilinx_pci.o
diff --git a/arch/microblaze/pci/indirect_pci.c b/arch/microblaze/pci/indirect_pci.c
new file mode 100644
index 0000000..ae4fca4
--- /dev/null
+++ b/arch/microblaze/pci/indirect_pci.c
@@ -0,0 +1,163 @@
+/*
+ * Support for indirect PCI bridges.
+ *
+ * Copyright (C) 1998 Gabriel Paubert.
+ *
+ * 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 <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include <linux/string.h>
+#include <linux/init.h>
+
+#include <linux/io.h>
+#include <asm/prom.h>
+#include <asm/pci-bridge.h>
+
+static int
+indirect_read_config(struct pci_bus *bus, unsigned int devfn, int offset,
+		     int len, u32 *val)
+{
+	struct pci_controller *hose = pci_bus_to_host(bus);
+	volatile void __iomem *cfg_data;
+	u8 cfg_type = 0;
+	u32 bus_no, reg;
+
+	if (hose->indirect_type & INDIRECT_TYPE_NO_PCIE_LINK) {
+		if (bus->number != hose->first_busno)
+			return PCIBIOS_DEVICE_NOT_FOUND;
+		if (devfn != 0)
+			return PCIBIOS_DEVICE_NOT_FOUND;
+	}
+
+	if (hose->indirect_type & INDIRECT_TYPE_SET_CFG_TYPE)
+		if (bus->number != hose->first_busno)
+			cfg_type = 1;
+
+	bus_no = (bus->number == hose->first_busno) ?
+			hose->self_busno : bus->number;
+
+	if (hose->indirect_type & INDIRECT_TYPE_EXT_REG)
+		reg = ((offset & 0xf00) << 16) | (offset & 0xfc);
+	else
+		reg = offset & 0xfc; /* Only 3 bits for function */
+
+	if (hose->indirect_type & INDIRECT_TYPE_BIG_ENDIAN)
+		out_be32(hose->cfg_addr, (0x80000000 | (bus_no << 16) |
+			 (devfn << 8) | reg | cfg_type));
+	else
+		out_le32(hose->cfg_addr, (0x80000000 | (bus_no << 16) |
+			 (devfn << 8) | reg | cfg_type));
+
+	/*
+	 * Note: the caller has already checked that offset is
+	 * suitably aligned and that len is 1, 2 or 4.
+	 */
+	cfg_data = hose->cfg_data + (offset & 3); /* Only 3 bits for function */
+	switch (len) {
+	case 1:
+		*val = in_8(cfg_data);
+		break;
+	case 2:
+		*val = in_le16(cfg_data);
+		break;
+	default:
+		*val = in_le32(cfg_data);
+		break;
+	}
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static int
+indirect_write_config(struct pci_bus *bus, unsigned int devfn, int offset,
+		      int len, u32 val)
+{
+	struct pci_controller *hose = pci_bus_to_host(bus);
+	volatile void __iomem *cfg_data;
+	u8 cfg_type = 0;
+	u32 bus_no, reg;
+
+	if (hose->indirect_type & INDIRECT_TYPE_NO_PCIE_LINK) {
+		if (bus->number != hose->first_busno)
+			return PCIBIOS_DEVICE_NOT_FOUND;
+		if (devfn != 0)
+			return PCIBIOS_DEVICE_NOT_FOUND;
+	}
+
+	if (hose->indirect_type & INDIRECT_TYPE_SET_CFG_TYPE)
+		if (bus->number != hose->first_busno)
+			cfg_type = 1;
+
+	bus_no = (bus->number == hose->first_busno) ?
+			hose->self_busno : bus->number;
+
+	if (hose->indirect_type & INDIRECT_TYPE_EXT_REG)
+		reg = ((offset & 0xf00) << 16) | (offset & 0xfc);
+	else
+		reg = offset & 0xfc;
+
+	if (hose->indirect_type & INDIRECT_TYPE_BIG_ENDIAN)
+		out_be32(hose->cfg_addr, (0x80000000 | (bus_no << 16) |
+			 (devfn << 8) | reg | cfg_type));
+	else
+		out_le32(hose->cfg_addr, (0x80000000 | (bus_no << 16) |
+			 (devfn << 8) | reg | cfg_type));
+
+	/* suppress setting of PCI_PRIMARY_BUS */
+	if (hose->indirect_type & INDIRECT_TYPE_SURPRESS_PRIMARY_BUS)
+		if ((offset == PCI_PRIMARY_BUS) &&
+			(bus->number == hose->first_busno))
+			val &= 0xffffff00;
+
+	/* Workaround for PCI_28 Errata in 440EPx/GRx */
+	if ((hose->indirect_type & INDIRECT_TYPE_BROKEN_MRM) &&
+			offset == PCI_CACHE_LINE_SIZE) {
+		val = 0;
+	}
+
+	/*
+	 * Note: the caller has already checked that offset is
+	 * suitably aligned and that len is 1, 2 or 4.
+	 */
+	cfg_data = hose->cfg_data + (offset & 3);
+	switch (len) {
+	case 1:
+		out_8(cfg_data, val);
+		break;
+	case 2:
+		out_le16(cfg_data, val);
+		break;
+	default:
+		out_le32(cfg_data, val);
+		break;
+	}
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static struct pci_ops indirect_pci_ops = {
+	.read = indirect_read_config,
+	.write = indirect_write_config,
+};
+
+void __init
+setup_indirect_pci(struct pci_controller *hose,
+		   resource_size_t cfg_addr,
+		   resource_size_t cfg_data, u32 flags)
+{
+	resource_size_t base = cfg_addr & PAGE_MASK;
+	void __iomem *mbase;
+
+	mbase = ioremap(base, PAGE_SIZE);
+	hose->cfg_addr = mbase + (cfg_addr & ~PAGE_MASK);
+	if ((cfg_data & PAGE_MASK) != base)
+		mbase = ioremap(cfg_data & PAGE_MASK, PAGE_SIZE);
+	hose->cfg_data = mbase + (cfg_data & ~PAGE_MASK);
+	hose->ops = &indirect_pci_ops;
+	hose->indirect_type = flags;
+}
diff --git a/arch/microblaze/pci/iomap.c b/arch/microblaze/pci/iomap.c
new file mode 100644
index 0000000..94149f5
--- /dev/null
+++ b/arch/microblaze/pci/iomap.c
@@ -0,0 +1,21 @@
+/*
+ * ppc64 "iomap" interface implementation.
+ *
+ * (C) Copyright 2004 Linus Torvalds
+ */
+#include <linux/init.h>
+#include <linux/pci.h>
+#include <linux/mm.h>
+#include <linux/export.h>
+#include <linux/io.h>
+#include <asm/pci-bridge.h>
+
+void pci_iounmap(struct pci_dev *dev, void __iomem *addr)
+{
+	if (isa_vaddr_is_ioport(addr))
+		return;
+	if (pcibios_vaddr_is_ioport(addr))
+		return;
+	iounmap(addr);
+}
+EXPORT_SYMBOL(pci_iounmap);
diff --git a/arch/microblaze/pci/pci-common.c b/arch/microblaze/pci/pci-common.c
new file mode 100644
index 0000000..ae838ed
--- /dev/null
+++ b/arch/microblaze/pci/pci-common.c
@@ -0,0 +1,1504 @@
+/*
+ * Contains common pci routines for ALL ppc platform
+ * (based on pci_32.c and pci_64.c)
+ *
+ * Port for PPC64 David Engebretsen, IBM Corp.
+ * Contains common pci routines for ppc64 platform, pSeries and iSeries brands.
+ *
+ * Copyright (C) 2003 Anton Blanchard <anton@au.ibm.com>, IBM
+ *   Rework, based on alpha PCI code.
+ *
+ * Common pmac/prep/chrp pci routines. -- Cort
+ *
+ * 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 <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/string.h>
+#include <linux/init.h>
+#include <linux/bootmem.h>
+#include <linux/mm.h>
+#include <linux/list.h>
+#include <linux/syscalls.h>
+#include <linux/irq.h>
+#include <linux/vmalloc.h>
+#include <linux/slab.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/of_irq.h>
+#include <linux/of_pci.h>
+#include <linux/export.h>
+
+#include <asm/processor.h>
+#include <linux/io.h>
+#include <asm/pci-bridge.h>
+#include <asm/byteorder.h>
+
+static DEFINE_SPINLOCK(hose_spinlock);
+LIST_HEAD(hose_list);
+
+/* XXX kill that some day ... */
+static int global_phb_number;		/* Global phb counter */
+
+/* ISA Memory physical address */
+resource_size_t isa_mem_base;
+
+unsigned long isa_io_base;
+static int pci_bus_count;
+
+struct pci_controller *pcibios_alloc_controller(struct device_node *dev)
+{
+	struct pci_controller *phb;
+
+	phb = zalloc_maybe_bootmem(sizeof(struct pci_controller), GFP_KERNEL);
+	if (!phb)
+		return NULL;
+	spin_lock(&hose_spinlock);
+	phb->global_number = global_phb_number++;
+	list_add_tail(&phb->list_node, &hose_list);
+	spin_unlock(&hose_spinlock);
+	phb->dn = dev;
+	phb->is_dynamic = mem_init_done;
+	return phb;
+}
+
+void pcibios_free_controller(struct pci_controller *phb)
+{
+	spin_lock(&hose_spinlock);
+	list_del(&phb->list_node);
+	spin_unlock(&hose_spinlock);
+
+	if (phb->is_dynamic)
+		kfree(phb);
+}
+
+static resource_size_t pcibios_io_size(const struct pci_controller *hose)
+{
+	return resource_size(&hose->io_resource);
+}
+
+int pcibios_vaddr_is_ioport(void __iomem *address)
+{
+	int ret = 0;
+	struct pci_controller *hose;
+	resource_size_t size;
+
+	spin_lock(&hose_spinlock);
+	list_for_each_entry(hose, &hose_list, list_node) {
+		size = pcibios_io_size(hose);
+		if (address >= hose->io_base_virt &&
+		    address < (hose->io_base_virt + size)) {
+			ret = 1;
+			break;
+		}
+	}
+	spin_unlock(&hose_spinlock);
+	return ret;
+}
+
+unsigned long pci_address_to_pio(phys_addr_t address)
+{
+	struct pci_controller *hose;
+	resource_size_t size;
+	unsigned long ret = ~0;
+
+	spin_lock(&hose_spinlock);
+	list_for_each_entry(hose, &hose_list, list_node) {
+		size = pcibios_io_size(hose);
+		if (address >= hose->io_base_phys &&
+		    address < (hose->io_base_phys + size)) {
+			unsigned long base =
+				(unsigned long)hose->io_base_virt - _IO_BASE;
+			ret = base + (address - hose->io_base_phys);
+			break;
+		}
+	}
+	spin_unlock(&hose_spinlock);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(pci_address_to_pio);
+
+/*
+ * Return the domain number for this bus.
+ */
+int pci_domain_nr(struct pci_bus *bus)
+{
+	struct pci_controller *hose = pci_bus_to_host(bus);
+
+	return hose->global_number;
+}
+EXPORT_SYMBOL(pci_domain_nr);
+
+/* This routine is meant to be used early during boot, when the
+ * PCI bus numbers have not yet been assigned, and you need to
+ * issue PCI config cycles to an OF device.
+ * It could also be used to "fix" RTAS config cycles if you want
+ * to set pci_assign_all_buses to 1 and still use RTAS for PCI
+ * config cycles.
+ */
+struct pci_controller *pci_find_hose_for_OF_device(struct device_node *node)
+{
+	while (node) {
+		struct pci_controller *hose, *tmp;
+		list_for_each_entry_safe(hose, tmp, &hose_list, list_node)
+			if (hose->dn == node)
+				return hose;
+		node = node->parent;
+	}
+	return NULL;
+}
+
+void pcibios_set_master(struct pci_dev *dev)
+{
+	/* No special bus mastering setup handling */
+}
+
+/*
+ * Platform support for /proc/bus/pci/X/Y mmap()s,
+ * modelled on the sparc64 implementation by Dave Miller.
+ *  -- paulus.
+ */
+
+/*
+ * Adjust vm_pgoff of VMA such that it is the physical page offset
+ * corresponding to the 32-bit pci bus offset for DEV requested by the user.
+ *
+ * Basically, the user finds the base address for his device which he wishes
+ * to mmap.  They read the 32-bit value from the config space base register,
+ * add whatever PAGE_SIZE multiple offset they wish, and feed this into the
+ * offset parameter of mmap on /proc/bus/pci/XXX for that device.
+ *
+ * Returns negative error code on failure, zero on success.
+ */
+static struct resource *__pci_mmap_make_offset(struct pci_dev *dev,
+					       resource_size_t *offset,
+					       enum pci_mmap_state mmap_state)
+{
+	struct pci_controller *hose = pci_bus_to_host(dev->bus);
+	unsigned long io_offset = 0;
+	int i, res_bit;
+
+	if (!hose)
+		return NULL;		/* should never happen */
+
+	/* If memory, add on the PCI bridge address offset */
+	if (mmap_state == pci_mmap_mem) {
+#if 0 /* See comment in pci_resource_to_user() for why this is disabled */
+		*offset += hose->pci_mem_offset;
+#endif
+		res_bit = IORESOURCE_MEM;
+	} else {
+		io_offset = (unsigned long)hose->io_base_virt - _IO_BASE;
+		*offset += io_offset;
+		res_bit = IORESOURCE_IO;
+	}
+
+	/*
+	 * Check that the offset requested corresponds to one of the
+	 * resources of the device.
+	 */
+	for (i = 0; i <= PCI_ROM_RESOURCE; i++) {
+		struct resource *rp = &dev->resource[i];
+		int flags = rp->flags;
+
+		/* treat ROM as memory (should be already) */
+		if (i == PCI_ROM_RESOURCE)
+			flags |= IORESOURCE_MEM;
+
+		/* Active and same type? */
+		if ((flags & res_bit) == 0)
+			continue;
+
+		/* In the range of this resource? */
+		if (*offset < (rp->start & PAGE_MASK) || *offset > rp->end)
+			continue;
+
+		/* found it! construct the final physical address */
+		if (mmap_state == pci_mmap_io)
+			*offset += hose->io_base_phys - io_offset;
+		return rp;
+	}
+
+	return NULL;
+}
+
+/*
+ * Set vm_page_prot of VMA, as appropriate for this architecture, for a pci
+ * device mapping.
+ */
+static pgprot_t __pci_mmap_set_pgprot(struct pci_dev *dev, struct resource *rp,
+				      pgprot_t protection,
+				      enum pci_mmap_state mmap_state,
+				      int write_combine)
+{
+	pgprot_t prot = protection;
+
+	/* Write combine is always 0 on non-memory space mappings. On
+	 * memory space, if the user didn't pass 1, we check for a
+	 * "prefetchable" resource. This is a bit hackish, but we use
+	 * this to workaround the inability of /sysfs to provide a write
+	 * combine bit
+	 */
+	if (mmap_state != pci_mmap_mem)
+		write_combine = 0;
+	else if (write_combine == 0) {
+		if (rp->flags & IORESOURCE_PREFETCH)
+			write_combine = 1;
+	}
+
+	return pgprot_noncached(prot);
+}
+
+/*
+ * This one is used by /dev/mem and fbdev who have no clue about the
+ * PCI device, it tries to find the PCI device first and calls the
+ * above routine
+ */
+pgprot_t pci_phys_mem_access_prot(struct file *file,
+				  unsigned long pfn,
+				  unsigned long size,
+				  pgprot_t prot)
+{
+	struct pci_dev *pdev = NULL;
+	struct resource *found = NULL;
+	resource_size_t offset = ((resource_size_t)pfn) << PAGE_SHIFT;
+	int i;
+
+	if (page_is_ram(pfn))
+		return prot;
+
+	prot = pgprot_noncached(prot);
+	for_each_pci_dev(pdev) {
+		for (i = 0; i <= PCI_ROM_RESOURCE; i++) {
+			struct resource *rp = &pdev->resource[i];
+			int flags = rp->flags;
+
+			/* Active and same type? */
+			if ((flags & IORESOURCE_MEM) == 0)
+				continue;
+			/* In the range of this resource? */
+			if (offset < (rp->start & PAGE_MASK) ||
+			    offset > rp->end)
+				continue;
+			found = rp;
+			break;
+		}
+		if (found)
+			break;
+	}
+	if (found) {
+		if (found->flags & IORESOURCE_PREFETCH)
+			prot = pgprot_noncached_wc(prot);
+		pci_dev_put(pdev);
+	}
+
+	pr_debug("PCI: Non-PCI map for %llx, prot: %lx\n",
+		 (unsigned long long)offset, pgprot_val(prot));
+
+	return prot;
+}
+
+/*
+ * Perform the actual remap of the pages for a PCI device mapping, as
+ * appropriate for this architecture.  The region in the process to map
+ * is described by vm_start and vm_end members of VMA, the base physical
+ * address is found in vm_pgoff.
+ * The pci device structure is provided so that architectures may make mapping
+ * decisions on a per-device or per-bus basis.
+ *
+ * Returns a negative error code on failure, zero on success.
+ */
+int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma,
+			enum pci_mmap_state mmap_state, int write_combine)
+{
+	resource_size_t offset =
+		((resource_size_t)vma->vm_pgoff) << PAGE_SHIFT;
+	struct resource *rp;
+	int ret;
+
+	rp = __pci_mmap_make_offset(dev, &offset, mmap_state);
+	if (rp == NULL)
+		return -EINVAL;
+
+	vma->vm_pgoff = offset >> PAGE_SHIFT;
+	vma->vm_page_prot = __pci_mmap_set_pgprot(dev, rp,
+						  vma->vm_page_prot,
+						  mmap_state, write_combine);
+
+	ret = remap_pfn_range(vma, vma->vm_start, vma->vm_pgoff,
+			       vma->vm_end - vma->vm_start, vma->vm_page_prot);
+
+	return ret;
+}
+
+/* This provides legacy IO read access on a bus */
+int pci_legacy_read(struct pci_bus *bus, loff_t port, u32 *val, size_t size)
+{
+	unsigned long offset;
+	struct pci_controller *hose = pci_bus_to_host(bus);
+	struct resource *rp = &hose->io_resource;
+	void __iomem *addr;
+
+	/* Check if port can be supported by that bus. We only check
+	 * the ranges of the PHB though, not the bus itself as the rules
+	 * for forwarding legacy cycles down bridges are not our problem
+	 * here. So if the host bridge supports it, we do it.
+	 */
+	offset = (unsigned long)hose->io_base_virt - _IO_BASE;
+	offset += port;
+
+	if (!(rp->flags & IORESOURCE_IO))
+		return -ENXIO;
+	if (offset < rp->start || (offset + size) > rp->end)
+		return -ENXIO;
+	addr = hose->io_base_virt + port;
+
+	switch (size) {
+	case 1:
+		*((u8 *)val) = in_8(addr);
+		return 1;
+	case 2:
+		if (port & 1)
+			return -EINVAL;
+		*((u16 *)val) = in_le16(addr);
+		return 2;
+	case 4:
+		if (port & 3)
+			return -EINVAL;
+		*((u32 *)val) = in_le32(addr);
+		return 4;
+	}
+	return -EINVAL;
+}
+
+/* This provides legacy IO write access on a bus */
+int pci_legacy_write(struct pci_bus *bus, loff_t port, u32 val, size_t size)
+{
+	unsigned long offset;
+	struct pci_controller *hose = pci_bus_to_host(bus);
+	struct resource *rp = &hose->io_resource;
+	void __iomem *addr;
+
+	/* Check if port can be supported by that bus. We only check
+	 * the ranges of the PHB though, not the bus itself as the rules
+	 * for forwarding legacy cycles down bridges are not our problem
+	 * here. So if the host bridge supports it, we do it.
+	 */
+	offset = (unsigned long)hose->io_base_virt - _IO_BASE;
+	offset += port;
+
+	if (!(rp->flags & IORESOURCE_IO))
+		return -ENXIO;
+	if (offset < rp->start || (offset + size) > rp->end)
+		return -ENXIO;
+	addr = hose->io_base_virt + port;
+
+	/* WARNING: The generic code is idiotic. It gets passed a pointer
+	 * to what can be a 1, 2 or 4 byte quantity and always reads that
+	 * as a u32, which means that we have to correct the location of
+	 * the data read within those 32 bits for size 1 and 2
+	 */
+	switch (size) {
+	case 1:
+		out_8(addr, val >> 24);
+		return 1;
+	case 2:
+		if (port & 1)
+			return -EINVAL;
+		out_le16(addr, val >> 16);
+		return 2;
+	case 4:
+		if (port & 3)
+			return -EINVAL;
+		out_le32(addr, val);
+		return 4;
+	}
+	return -EINVAL;
+}
+
+/* This provides legacy IO or memory mmap access on a bus */
+int pci_mmap_legacy_page_range(struct pci_bus *bus,
+			       struct vm_area_struct *vma,
+			       enum pci_mmap_state mmap_state)
+{
+	struct pci_controller *hose = pci_bus_to_host(bus);
+	resource_size_t offset =
+		((resource_size_t)vma->vm_pgoff) << PAGE_SHIFT;
+	resource_size_t size = vma->vm_end - vma->vm_start;
+	struct resource *rp;
+
+	pr_debug("pci_mmap_legacy_page_range(%04x:%02x, %s @%llx..%llx)\n",
+		 pci_domain_nr(bus), bus->number,
+		 mmap_state == pci_mmap_mem ? "MEM" : "IO",
+		 (unsigned long long)offset,
+		 (unsigned long long)(offset + size - 1));
+
+	if (mmap_state == pci_mmap_mem) {
+		/* Hack alert !
+		 *
+		 * Because X is lame and can fail starting if it gets an error
+		 * trying to mmap legacy_mem (instead of just moving on without
+		 * legacy memory access) we fake it here by giving it anonymous
+		 * memory, effectively behaving just like /dev/zero
+		 */
+		if ((offset + size) > hose->isa_mem_size) {
+#ifdef CONFIG_MMU
+			pr_debug("Process %s (pid:%d) mapped non-existing PCI",
+				current->comm, current->pid);
+			pr_debug("legacy memory for 0%04x:%02x\n",
+				pci_domain_nr(bus), bus->number);
+#endif
+			if (vma->vm_flags & VM_SHARED)
+				return shmem_zero_setup(vma);
+			return 0;
+		}
+		offset += hose->isa_mem_phys;
+	} else {
+		unsigned long io_offset = (unsigned long)hose->io_base_virt -
+								_IO_BASE;
+		unsigned long roffset = offset + io_offset;
+		rp = &hose->io_resource;
+		if (!(rp->flags & IORESOURCE_IO))
+			return -ENXIO;
+		if (roffset < rp->start || (roffset + size) > rp->end)
+			return -ENXIO;
+		offset += hose->io_base_phys;
+	}
+	pr_debug(" -> mapping phys %llx\n", (unsigned long long)offset);
+
+	vma->vm_pgoff = offset >> PAGE_SHIFT;
+	vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
+	return remap_pfn_range(vma, vma->vm_start, vma->vm_pgoff,
+			       vma->vm_end - vma->vm_start,
+			       vma->vm_page_prot);
+}
+
+void pci_resource_to_user(const struct pci_dev *dev, int bar,
+			  const struct resource *rsrc,
+			  resource_size_t *start, resource_size_t *end)
+{
+	struct pci_controller *hose = pci_bus_to_host(dev->bus);
+	resource_size_t offset = 0;
+
+	if (hose == NULL)
+		return;
+
+	if (rsrc->flags & IORESOURCE_IO)
+		offset = (unsigned long)hose->io_base_virt - _IO_BASE;
+
+	/* We pass a fully fixed up address to userland for MMIO instead of
+	 * a BAR value because X is lame and expects to be able to use that
+	 * to pass to /dev/mem !
+	 *
+	 * That means that we'll have potentially 64 bits values where some
+	 * userland apps only expect 32 (like X itself since it thinks only
+	 * Sparc has 64 bits MMIO) but if we don't do that, we break it on
+	 * 32 bits CHRPs :-(
+	 *
+	 * Hopefully, the sysfs insterface is immune to that gunk. Once X
+	 * has been fixed (and the fix spread enough), we can re-enable the
+	 * 2 lines below and pass down a BAR value to userland. In that case
+	 * we'll also have to re-enable the matching code in
+	 * __pci_mmap_make_offset().
+	 *
+	 * BenH.
+	 */
+#if 0
+	else if (rsrc->flags & IORESOURCE_MEM)
+		offset = hose->pci_mem_offset;
+#endif
+
+	*start = rsrc->start - offset;
+	*end = rsrc->end - offset;
+}
+
+/**
+ * pci_process_bridge_OF_ranges - Parse PCI bridge resources from device tree
+ * @hose: newly allocated pci_controller to be setup
+ * @dev: device node of the host bridge
+ * @primary: set if primary bus (32 bits only, soon to be deprecated)
+ *
+ * This function will parse the "ranges" property of a PCI host bridge device
+ * node and setup the resource mapping of a pci controller based on its
+ * content.
+ *
+ * Life would be boring if it wasn't for a few issues that we have to deal
+ * with here:
+ *
+ *   - We can only cope with one IO space range and up to 3 Memory space
+ *     ranges. However, some machines (thanks Apple !) tend to split their
+ *     space into lots of small contiguous ranges. So we have to coalesce.
+ *
+ *   - We can only cope with all memory ranges having the same offset
+ *     between CPU addresses and PCI addresses. Unfortunately, some bridges
+ *     are setup for a large 1:1 mapping along with a small "window" which
+ *     maps PCI address 0 to some arbitrary high address of the CPU space in
+ *     order to give access to the ISA memory hole.
+ *     The way out of here that I've chosen for now is to always set the
+ *     offset based on the first resource found, then override it if we
+ *     have a different offset and the previous was set by an ISA hole.
+ *
+ *   - Some busses have IO space not starting at 0, which causes trouble with
+ *     the way we do our IO resource renumbering. The code somewhat deals with
+ *     it for 64 bits but I would expect problems on 32 bits.
+ *
+ *   - Some 32 bits platforms such as 4xx can have physical space larger than
+ *     32 bits so we need to use 64 bits values for the parsing
+ */
+void pci_process_bridge_OF_ranges(struct pci_controller *hose,
+				  struct device_node *dev, int primary)
+{
+	int memno = 0, isa_hole = -1;
+	unsigned long long isa_mb = 0;
+	struct resource *res;
+	struct of_pci_range range;
+	struct of_pci_range_parser parser;
+
+	pr_info("PCI host bridge %s %s ranges:\n",
+	       dev->full_name, primary ? "(primary)" : "");
+
+	/* Check for ranges property */
+	if (of_pci_range_parser_init(&parser, dev))
+		return;
+
+	pr_debug("Parsing ranges property...\n");
+	for_each_of_pci_range(&parser, &range) {
+		/* Read next ranges element */
+		pr_debug("pci_space: 0x%08x pci_addr:0x%016llx ",
+				range.pci_space, range.pci_addr);
+		pr_debug("cpu_addr:0x%016llx size:0x%016llx\n",
+					range.cpu_addr, range.size);
+
+		/* If we failed translation or got a zero-sized region
+		 * (some FW try to feed us with non sensical zero sized regions
+		 * such as power3 which look like some kind of attempt
+		 * at exposing the VGA memory hole)
+		 */
+		if (range.cpu_addr == OF_BAD_ADDR || range.size == 0)
+			continue;
+
+		/* Act based on address space type */
+		res = NULL;
+		switch (range.flags & IORESOURCE_TYPE_BITS) {
+		case IORESOURCE_IO:
+			pr_info("  IO 0x%016llx..0x%016llx -> 0x%016llx\n",
+				range.cpu_addr, range.cpu_addr + range.size - 1,
+				range.pci_addr);
+
+			/* We support only one IO range */
+			if (hose->pci_io_size) {
+				pr_info(" \\--> Skipped (too many) !\n");
+				continue;
+			}
+			/* On 32 bits, limit I/O space to 16MB */
+			if (range.size > 0x01000000)
+				range.size = 0x01000000;
+
+			/* 32 bits needs to map IOs here */
+			hose->io_base_virt = ioremap(range.cpu_addr,
+						range.size);
+
+			/* Expect trouble if pci_addr is not 0 */
+			if (primary)
+				isa_io_base =
+					(unsigned long)hose->io_base_virt;
+			/* pci_io_size and io_base_phys always represent IO
+			 * space starting at 0 so we factor in pci_addr
+			 */
+			hose->pci_io_size = range.pci_addr + range.size;
+			hose->io_base_phys = range.cpu_addr - range.pci_addr;
+
+			/* Build resource */
+			res = &hose->io_resource;
+			range.cpu_addr = range.pci_addr;
+
+			break;
+		case IORESOURCE_MEM:
+			pr_info(" MEM 0x%016llx..0x%016llx -> 0x%016llx %s\n",
+				range.cpu_addr, range.cpu_addr + range.size - 1,
+				range.pci_addr,
+				(range.pci_space & 0x40000000) ?
+				"Prefetch" : "");
+
+			/* We support only 3 memory ranges */
+			if (memno >= 3) {
+				pr_info(" \\--> Skipped (too many) !\n");
+				continue;
+			}
+			/* Handles ISA memory hole space here */
+			if (range.pci_addr == 0) {
+				isa_mb = range.cpu_addr;
+				isa_hole = memno;
+				if (primary || isa_mem_base == 0)
+					isa_mem_base = range.cpu_addr;
+				hose->isa_mem_phys = range.cpu_addr;
+				hose->isa_mem_size = range.size;
+			}
+
+			/* We get the PCI/Mem offset from the first range or
+			 * the, current one if the offset came from an ISA
+			 * hole. If they don't match, bugger.
+			 */
+			if (memno == 0 ||
+			    (isa_hole >= 0 && range.pci_addr != 0 &&
+			     hose->pci_mem_offset == isa_mb))
+				hose->pci_mem_offset = range.cpu_addr -
+							range.pci_addr;
+			else if (range.pci_addr != 0 &&
+				 hose->pci_mem_offset != range.cpu_addr -
+							range.pci_addr) {
+				pr_info(" \\--> Skipped (offset mismatch) !\n");
+				continue;
+			}
+
+			/* Build resource */
+			res = &hose->mem_resources[memno++];
+			break;
+		}
+		if (res != NULL) {
+			res->name = dev->full_name;
+			res->flags = range.flags;
+			res->start = range.cpu_addr;
+			res->end = range.cpu_addr + range.size - 1;
+			res->parent = res->child = res->sibling = NULL;
+		}
+	}
+
+	/* If there's an ISA hole and the pci_mem_offset is -not- matching
+	 * the ISA hole offset, then we need to remove the ISA hole from
+	 * the resource list for that brige
+	 */
+	if (isa_hole >= 0 && hose->pci_mem_offset != isa_mb) {
+		unsigned int next = isa_hole + 1;
+		pr_info(" Removing ISA hole at 0x%016llx\n", isa_mb);
+		if (next < memno)
+			memmove(&hose->mem_resources[isa_hole],
+				&hose->mem_resources[next],
+				sizeof(struct resource) * (memno - next));
+		hose->mem_resources[--memno].flags = 0;
+	}
+}
+
+/* Decide whether to display the domain number in /proc */
+int pci_proc_domain(struct pci_bus *bus)
+{
+	return 0;
+}
+
+/* This header fixup will do the resource fixup for all devices as they are
+ * probed, but not for bridge ranges
+ */
+static void pcibios_fixup_resources(struct pci_dev *dev)
+{
+	struct pci_controller *hose = pci_bus_to_host(dev->bus);
+	int i;
+
+	if (!hose) {
+		pr_err("No host bridge for PCI dev %s !\n",
+		       pci_name(dev));
+		return;
+	}
+	for (i = 0; i < DEVICE_COUNT_RESOURCE; i++) {
+		struct resource *res = dev->resource + i;
+		if (!res->flags)
+			continue;
+		if (res->start == 0) {
+			pr_debug("PCI:%s Resource %d %016llx-%016llx [%x]",
+				 pci_name(dev), i,
+				 (unsigned long long)res->start,
+				 (unsigned long long)res->end,
+				 (unsigned int)res->flags);
+			pr_debug("is unassigned\n");
+			res->end -= res->start;
+			res->start = 0;
+			res->flags |= IORESOURCE_UNSET;
+			continue;
+		}
+
+		pr_debug("PCI:%s Resource %d %016llx-%016llx [%x]\n",
+			 pci_name(dev), i,
+			 (unsigned long long)res->start,
+			 (unsigned long long)res->end,
+			 (unsigned int)res->flags);
+	}
+}
+DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pcibios_fixup_resources);
+
+/* This function tries to figure out if a bridge resource has been initialized
+ * by the firmware or not. It doesn't have to be absolutely bullet proof, but
+ * things go more smoothly when it gets it right. It should covers cases such
+ * as Apple "closed" bridge resources and bare-metal pSeries unassigned bridges
+ */
+static int pcibios_uninitialized_bridge_resource(struct pci_bus *bus,
+						 struct resource *res)
+{
+	struct pci_controller *hose = pci_bus_to_host(bus);
+	struct pci_dev *dev = bus->self;
+	resource_size_t offset;
+	u16 command;
+	int i;
+
+	/* Job is a bit different between memory and IO */
+	if (res->flags & IORESOURCE_MEM) {
+		/* If the BAR is non-0 (res != pci_mem_offset) then it's
+		 * probably been initialized by somebody
+		 */
+		if (res->start != hose->pci_mem_offset)
+			return 0;
+
+		/* The BAR is 0, let's check if memory decoding is enabled on
+		 * the bridge. If not, we consider it unassigned
+		 */
+		pci_read_config_word(dev, PCI_COMMAND, &command);
+		if ((command & PCI_COMMAND_MEMORY) == 0)
+			return 1;
+
+		/* Memory decoding is enabled and the BAR is 0. If any of
+		 * the bridge resources covers that starting address (0 then
+		 * it's good enough for us for memory
+		 */
+		for (i = 0; i < 3; i++) {
+			if ((hose->mem_resources[i].flags & IORESOURCE_MEM) &&
+			   hose->mem_resources[i].start == hose->pci_mem_offset)
+				return 0;
+		}
+
+		/* Well, it starts at 0 and we know it will collide so we may as
+		 * well consider it as unassigned. That covers the Apple case.
+		 */
+		return 1;
+	} else {
+		/* If the BAR is non-0, then we consider it assigned */
+		offset = (unsigned long)hose->io_base_virt - _IO_BASE;
+		if (((res->start - offset) & 0xfffffffful) != 0)
+			return 0;
+
+		/* Here, we are a bit different than memory as typically IO
+		 * space starting at low addresses -is- valid. What we do
+		 * instead if that we consider as unassigned anything that
+		 * doesn't have IO enabled in the PCI command register,
+		 * and that's it.
+		 */
+		pci_read_config_word(dev, PCI_COMMAND, &command);
+		if (command & PCI_COMMAND_IO)
+			return 0;
+
+		/* It's starting at 0 and IO is disabled in the bridge, consider
+		 * it unassigned
+		 */
+		return 1;
+	}
+}
+
+/* Fixup resources of a PCI<->PCI bridge */
+static void pcibios_fixup_bridge(struct pci_bus *bus)
+{
+	struct resource *res;
+	int i;
+
+	struct pci_dev *dev = bus->self;
+
+	pci_bus_for_each_resource(bus, res, i) {
+		if (!res)
+			continue;
+		if (!res->flags)
+			continue;
+		if (i >= 3 && bus->self->transparent)
+			continue;
+
+		pr_debug("PCI:%s Bus rsrc %d %016llx-%016llx [%x] fixup...\n",
+			 pci_name(dev), i,
+			 (unsigned long long)res->start,
+			 (unsigned long long)res->end,
+			 (unsigned int)res->flags);
+
+		/* Try to detect uninitialized P2P bridge resources,
+		 * and clear them out so they get re-assigned later
+		 */
+		if (pcibios_uninitialized_bridge_resource(bus, res)) {
+			res->flags = 0;
+			pr_debug("PCI:%s            (unassigned)\n",
+								pci_name(dev));
+		} else {
+			pr_debug("PCI:%s            %016llx-%016llx\n",
+				 pci_name(dev),
+				 (unsigned long long)res->start,
+				 (unsigned long long)res->end);
+		}
+	}
+}
+
+void pcibios_setup_bus_self(struct pci_bus *bus)
+{
+	/* Fix up the bus resources for P2P bridges */
+	if (bus->self != NULL)
+		pcibios_fixup_bridge(bus);
+}
+
+void pcibios_setup_bus_devices(struct pci_bus *bus)
+{
+	struct pci_dev *dev;
+
+	pr_debug("PCI: Fixup bus devices %d (%s)\n",
+		 bus->number, bus->self ? pci_name(bus->self) : "PHB");
+
+	list_for_each_entry(dev, &bus->devices, bus_list) {
+		/* Setup OF node pointer in archdata */
+		dev->dev.of_node = pci_device_to_OF_node(dev);
+
+		/* Fixup NUMA node as it may not be setup yet by the generic
+		 * code and is needed by the DMA init
+		 */
+		set_dev_node(&dev->dev, pcibus_to_node(dev->bus));
+
+		/* Read default IRQs and fixup if necessary */
+		dev->irq = of_irq_parse_and_map_pci(dev, 0, 0);
+	}
+}
+
+void pcibios_fixup_bus(struct pci_bus *bus)
+{
+	/* When called from the generic PCI probe, read PCI<->PCI bridge
+	 * bases. This is -not- called when generating the PCI tree from
+	 * the OF device-tree.
+	 */
+	if (bus->self != NULL)
+		pci_read_bridge_bases(bus);
+
+	/* Now fixup the bus bus */
+	pcibios_setup_bus_self(bus);
+
+	/* Now fixup devices on that bus */
+	pcibios_setup_bus_devices(bus);
+}
+EXPORT_SYMBOL(pcibios_fixup_bus);
+
+static int skip_isa_ioresource_align(struct pci_dev *dev)
+{
+	return 0;
+}
+
+/*
+ * We need to avoid collisions with `mirrored' VGA ports
+ * and other strange ISA hardware, so we always want the
+ * addresses to be allocated in the 0x000-0x0ff region
+ * modulo 0x400.
+ *
+ * Why? Because some silly external IO cards only decode
+ * the low 10 bits of the IO address. The 0x00-0xff region
+ * is reserved for motherboard devices that decode all 16
+ * bits, so it's ok to allocate at, say, 0x2800-0x28ff,
+ * but we want to try to avoid allocating at 0x2900-0x2bff
+ * which might have be mirrored at 0x0100-0x03ff..
+ */
+resource_size_t pcibios_align_resource(void *data, const struct resource *res,
+				resource_size_t size, resource_size_t align)
+{
+	struct pci_dev *dev = data;
+	resource_size_t start = res->start;
+
+	if (res->flags & IORESOURCE_IO) {
+		if (skip_isa_ioresource_align(dev))
+			return start;
+		if (start & 0x300)
+			start = (start + 0x3ff) & ~0x3ff;
+	}
+
+	return start;
+}
+EXPORT_SYMBOL(pcibios_align_resource);
+
+/*
+ * Reparent resource children of pr that conflict with res
+ * under res, and make res replace those children.
+ */
+static int __init reparent_resources(struct resource *parent,
+				     struct resource *res)
+{
+	struct resource *p, **pp;
+	struct resource **firstpp = NULL;
+
+	for (pp = &parent->child; (p = *pp) != NULL; pp = &p->sibling) {
+		if (p->end < res->start)
+			continue;
+		if (res->end < p->start)
+			break;
+		if (p->start < res->start || p->end > res->end)
+			return -1;	/* not completely contained */
+		if (firstpp == NULL)
+			firstpp = pp;
+	}
+	if (firstpp == NULL)
+		return -1;	/* didn't find any conflicting entries? */
+	res->parent = parent;
+	res->child = *firstpp;
+	res->sibling = *pp;
+	*firstpp = res;
+	*pp = NULL;
+	for (p = res->child; p != NULL; p = p->sibling) {
+		p->parent = res;
+		pr_debug("PCI: Reparented %s [%llx..%llx] under %s\n",
+			 p->name,
+			 (unsigned long long)p->start,
+			 (unsigned long long)p->end, res->name);
+	}
+	return 0;
+}
+
+/*
+ *  Handle resources of PCI devices.  If the world were perfect, we could
+ *  just allocate all the resource regions and do nothing more.  It isn't.
+ *  On the other hand, we cannot just re-allocate all devices, as it would
+ *  require us to know lots of host bridge internals.  So we attempt to
+ *  keep as much of the original configuration as possible, but tweak it
+ *  when it's found to be wrong.
+ *
+ *  Known BIOS problems we have to work around:
+ *	- I/O or memory regions not configured
+ *	- regions configured, but not enabled in the command register
+ *	- bogus I/O addresses above 64K used
+ *	- expansion ROMs left enabled (this may sound harmless, but given
+ *	  the fact the PCI specs explicitly allow address decoders to be
+ *	  shared between expansion ROMs and other resource regions, it's
+ *	  at least dangerous)
+ *
+ *  Our solution:
+ *	(1) Allocate resources for all buses behind PCI-to-PCI bridges.
+ *	    This gives us fixed barriers on where we can allocate.
+ *	(2) Allocate resources for all enabled devices.  If there is
+ *	    a collision, just mark the resource as unallocated. Also
+ *	    disable expansion ROMs during this step.
+ *	(3) Try to allocate resources for disabled devices.  If the
+ *	    resources were assigned correctly, everything goes well,
+ *	    if they weren't, they won't disturb allocation of other
+ *	    resources.
+ *	(4) Assign new addresses to resources which were either
+ *	    not configured at all or misconfigured.  If explicitly
+ *	    requested by the user, configure expansion ROM address
+ *	    as well.
+ */
+
+static void pcibios_allocate_bus_resources(struct pci_bus *bus)
+{
+	struct pci_bus *b;
+	int i;
+	struct resource *res, *pr;
+
+	pr_debug("PCI: Allocating bus resources for %04x:%02x...\n",
+		 pci_domain_nr(bus), bus->number);
+
+	pci_bus_for_each_resource(bus, res, i) {
+		if (!res || !res->flags
+		    || res->start > res->end || res->parent)
+			continue;
+		if (bus->parent == NULL)
+			pr = (res->flags & IORESOURCE_IO) ?
+				&ioport_resource : &iomem_resource;
+		else {
+			/* Don't bother with non-root busses when
+			 * re-assigning all resources. We clear the
+			 * resource flags as if they were colliding
+			 * and as such ensure proper re-allocation
+			 * later.
+			 */
+			pr = pci_find_parent_resource(bus->self, res);
+			if (pr == res) {
+				/* this happens when the generic PCI
+				 * code (wrongly) decides that this
+				 * bridge is transparent  -- paulus
+				 */
+				continue;
+			}
+		}
+
+		pr_debug("PCI: %s (bus %d) bridge rsrc %d: %016llx-%016llx ",
+			 bus->self ? pci_name(bus->self) : "PHB",
+			 bus->number, i,
+			 (unsigned long long)res->start,
+			 (unsigned long long)res->end);
+		pr_debug("[0x%x], parent %p (%s)\n",
+			 (unsigned int)res->flags,
+			 pr, (pr && pr->name) ? pr->name : "nil");
+
+		if (pr && !(pr->flags & IORESOURCE_UNSET)) {
+			struct pci_dev *dev = bus->self;
+
+			if (request_resource(pr, res) == 0)
+				continue;
+			/*
+			 * Must be a conflict with an existing entry.
+			 * Move that entry (or entries) under the
+			 * bridge resource and try again.
+			 */
+			if (reparent_resources(pr, res) == 0)
+				continue;
+
+			if (dev && i < PCI_BRIDGE_RESOURCE_NUM &&
+			    pci_claim_bridge_resource(dev,
+						 i + PCI_BRIDGE_RESOURCES) == 0)
+				continue;
+
+		}
+		pr_warn("PCI: Cannot allocate resource region ");
+		pr_cont("%d of PCI bridge %d, will remap\n", i, bus->number);
+		res->start = res->end = 0;
+		res->flags = 0;
+	}
+
+	list_for_each_entry(b, &bus->children, node)
+		pcibios_allocate_bus_resources(b);
+}
+
+static inline void alloc_resource(struct pci_dev *dev, int idx)
+{
+	struct resource *pr, *r = &dev->resource[idx];
+
+	pr_debug("PCI: Allocating %s: Resource %d: %016llx..%016llx [%x]\n",
+		 pci_name(dev), idx,
+		 (unsigned long long)r->start,
+		 (unsigned long long)r->end,
+		 (unsigned int)r->flags);
+
+	pr = pci_find_parent_resource(dev, r);
+	if (!pr || (pr->flags & IORESOURCE_UNSET) ||
+	    request_resource(pr, r) < 0) {
+		pr_warn("PCI: Cannot allocate resource region %d ", idx);
+		pr_cont("of device %s, will remap\n", pci_name(dev));
+		if (pr)
+			pr_debug("PCI:  parent is %p: %016llx-%016llx [%x]\n",
+				 pr,
+				 (unsigned long long)pr->start,
+				 (unsigned long long)pr->end,
+				 (unsigned int)pr->flags);
+		/* We'll assign a new address later */
+		r->flags |= IORESOURCE_UNSET;
+		r->end -= r->start;
+		r->start = 0;
+	}
+}
+
+static void __init pcibios_allocate_resources(int pass)
+{
+	struct pci_dev *dev = NULL;
+	int idx, disabled;
+	u16 command;
+	struct resource *r;
+
+	for_each_pci_dev(dev) {
+		pci_read_config_word(dev, PCI_COMMAND, &command);
+		for (idx = 0; idx <= PCI_ROM_RESOURCE; idx++) {
+			r = &dev->resource[idx];
+			if (r->parent)		/* Already allocated */
+				continue;
+			if (!r->flags || (r->flags & IORESOURCE_UNSET))
+				continue;	/* Not assigned at all */
+			/* We only allocate ROMs on pass 1 just in case they
+			 * have been screwed up by firmware
+			 */
+			if (idx == PCI_ROM_RESOURCE)
+				disabled = 1;
+			if (r->flags & IORESOURCE_IO)
+				disabled = !(command & PCI_COMMAND_IO);
+			else
+				disabled = !(command & PCI_COMMAND_MEMORY);
+			if (pass == disabled)
+				alloc_resource(dev, idx);
+		}
+		if (pass)
+			continue;
+		r = &dev->resource[PCI_ROM_RESOURCE];
+		if (r->flags) {
+			/* Turn the ROM off, leave the resource region,
+			 * but keep it unregistered.
+			 */
+			u32 reg;
+			pci_read_config_dword(dev, dev->rom_base_reg, &reg);
+			if (reg & PCI_ROM_ADDRESS_ENABLE) {
+				pr_debug("PCI: Switching off ROM of %s\n",
+					 pci_name(dev));
+				r->flags &= ~IORESOURCE_ROM_ENABLE;
+				pci_write_config_dword(dev, dev->rom_base_reg,
+						reg & ~PCI_ROM_ADDRESS_ENABLE);
+			}
+		}
+	}
+}
+
+static void __init pcibios_reserve_legacy_regions(struct pci_bus *bus)
+{
+	struct pci_controller *hose = pci_bus_to_host(bus);
+	resource_size_t	offset;
+	struct resource *res, *pres;
+	int i;
+
+	pr_debug("Reserving legacy ranges for domain %04x\n",
+							pci_domain_nr(bus));
+
+	/* Check for IO */
+	if (!(hose->io_resource.flags & IORESOURCE_IO))
+		goto no_io;
+	offset = (unsigned long)hose->io_base_virt - _IO_BASE;
+	res = kzalloc(sizeof(struct resource), GFP_KERNEL);
+	BUG_ON(res == NULL);
+	res->name = "Legacy IO";
+	res->flags = IORESOURCE_IO;
+	res->start = offset;
+	res->end = (offset + 0xfff) & 0xfffffffful;
+	pr_debug("Candidate legacy IO: %pR\n", res);
+	if (request_resource(&hose->io_resource, res)) {
+		pr_debug("PCI %04x:%02x Cannot reserve Legacy IO %pR\n",
+		       pci_domain_nr(bus), bus->number, res);
+		kfree(res);
+	}
+
+ no_io:
+	/* Check for memory */
+	offset = hose->pci_mem_offset;
+	pr_debug("hose mem offset: %016llx\n", (unsigned long long)offset);
+	for (i = 0; i < 3; i++) {
+		pres = &hose->mem_resources[i];
+		if (!(pres->flags & IORESOURCE_MEM))
+			continue;
+		pr_debug("hose mem res: %pR\n", pres);
+		if ((pres->start - offset) <= 0xa0000 &&
+		    (pres->end - offset) >= 0xbffff)
+			break;
+	}
+	if (i >= 3)
+		return;
+	res = kzalloc(sizeof(struct resource), GFP_KERNEL);
+	BUG_ON(res == NULL);
+	res->name = "Legacy VGA memory";
+	res->flags = IORESOURCE_MEM;
+	res->start = 0xa0000 + offset;
+	res->end = 0xbffff + offset;
+	pr_debug("Candidate VGA memory: %pR\n", res);
+	if (request_resource(pres, res)) {
+		pr_debug("PCI %04x:%02x Cannot reserve VGA memory %pR\n",
+		       pci_domain_nr(bus), bus->number, res);
+		kfree(res);
+	}
+}
+
+void __init pcibios_resource_survey(void)
+{
+	struct pci_bus *b;
+
+	/* Allocate and assign resources. If we re-assign everything, then
+	 * we skip the allocate phase
+	 */
+	list_for_each_entry(b, &pci_root_buses, node)
+		pcibios_allocate_bus_resources(b);
+
+	pcibios_allocate_resources(0);
+	pcibios_allocate_resources(1);
+
+	/* Before we start assigning unassigned resource, we try to reserve
+	 * the low IO area and the VGA memory area if they intersect the
+	 * bus available resources to avoid allocating things on top of them
+	 */
+	list_for_each_entry(b, &pci_root_buses, node)
+		pcibios_reserve_legacy_regions(b);
+
+	/* Now proceed to assigning things that were left unassigned */
+	pr_debug("PCI: Assigning unassigned resources...\n");
+	pci_assign_unassigned_resources();
+}
+
+/* This is used by the PCI hotplug driver to allocate resource
+ * of newly plugged busses. We can try to consolidate with the
+ * rest of the code later, for now, keep it as-is as our main
+ * resource allocation function doesn't deal with sub-trees yet.
+ */
+void pcibios_claim_one_bus(struct pci_bus *bus)
+{
+	struct pci_dev *dev;
+	struct pci_bus *child_bus;
+
+	list_for_each_entry(dev, &bus->devices, bus_list) {
+		int i;
+
+		for (i = 0; i < PCI_NUM_RESOURCES; i++) {
+			struct resource *r = &dev->resource[i];
+
+			if (r->parent || !r->start || !r->flags)
+				continue;
+
+			pr_debug("PCI: Claiming %s: ", pci_name(dev));
+			pr_debug("Resource %d: %016llx..%016llx [%x]\n",
+				 i, (unsigned long long)r->start,
+				 (unsigned long long)r->end,
+				 (unsigned int)r->flags);
+
+			if (pci_claim_resource(dev, i) == 0)
+				continue;
+
+			pci_claim_bridge_resource(dev, i);
+		}
+	}
+
+	list_for_each_entry(child_bus, &bus->children, node)
+		pcibios_claim_one_bus(child_bus);
+}
+EXPORT_SYMBOL_GPL(pcibios_claim_one_bus);
+
+
+/* pcibios_finish_adding_to_bus
+ *
+ * This is to be called by the hotplug code after devices have been
+ * added to a bus, this include calling it for a PHB that is just
+ * being added
+ */
+void pcibios_finish_adding_to_bus(struct pci_bus *bus)
+{
+	pr_debug("PCI: Finishing adding to hotplug bus %04x:%02x\n",
+		 pci_domain_nr(bus), bus->number);
+
+	/* Allocate bus and devices resources */
+	pcibios_allocate_bus_resources(bus);
+	pcibios_claim_one_bus(bus);
+
+	/* Add new devices to global lists.  Register in proc, sysfs. */
+	pci_bus_add_devices(bus);
+
+	/* Fixup EEH */
+	/* eeh_add_device_tree_late(bus); */
+}
+EXPORT_SYMBOL_GPL(pcibios_finish_adding_to_bus);
+
+static void pcibios_setup_phb_resources(struct pci_controller *hose,
+					struct list_head *resources)
+{
+	unsigned long io_offset;
+	struct resource *res;
+	int i;
+
+	/* Hookup PHB IO resource */
+	res = &hose->io_resource;
+
+	/* Fixup IO space offset */
+	io_offset = (unsigned long)hose->io_base_virt - isa_io_base;
+	res->start = (res->start + io_offset) & 0xffffffffu;
+	res->end = (res->end + io_offset) & 0xffffffffu;
+
+	if (!res->flags) {
+		pr_warn("PCI: I/O resource not set for host ");
+		pr_cont("bridge %s (domain %d)\n",
+			hose->dn->full_name, hose->global_number);
+		/* Workaround for lack of IO resource only on 32-bit */
+		res->start = (unsigned long)hose->io_base_virt - isa_io_base;
+		res->end = res->start + IO_SPACE_LIMIT;
+		res->flags = IORESOURCE_IO;
+	}
+	pci_add_resource_offset(resources, res,
+		(__force resource_size_t)(hose->io_base_virt - _IO_BASE));
+
+	pr_debug("PCI: PHB IO resource    = %016llx-%016llx [%lx]\n",
+		 (unsigned long long)res->start,
+		 (unsigned long long)res->end,
+		 (unsigned long)res->flags);
+
+	/* Hookup PHB Memory resources */
+	for (i = 0; i < 3; ++i) {
+		res = &hose->mem_resources[i];
+		if (!res->flags) {
+			if (i > 0)
+				continue;
+			pr_err("PCI: Memory resource 0 not set for ");
+			pr_cont("host bridge %s (domain %d)\n",
+				hose->dn->full_name, hose->global_number);
+
+			/* Workaround for lack of MEM resource only on 32-bit */
+			res->start = hose->pci_mem_offset;
+			res->end = (resource_size_t)-1LL;
+			res->flags = IORESOURCE_MEM;
+
+		}
+		pci_add_resource_offset(resources, res, hose->pci_mem_offset);
+
+		pr_debug("PCI: PHB MEM resource %d = %016llx-%016llx [%lx]\n",
+			i, (unsigned long long)res->start,
+			(unsigned long long)res->end,
+			(unsigned long)res->flags);
+	}
+
+	pr_debug("PCI: PHB MEM offset     = %016llx\n",
+		 (unsigned long long)hose->pci_mem_offset);
+	pr_debug("PCI: PHB IO  offset     = %08lx\n",
+		 (unsigned long)hose->io_base_virt - _IO_BASE);
+}
+
+struct device_node *pcibios_get_phb_of_node(struct pci_bus *bus)
+{
+	struct pci_controller *hose = bus->sysdata;
+
+	return of_node_get(hose->dn);
+}
+
+static void pcibios_scan_phb(struct pci_controller *hose)
+{
+	LIST_HEAD(resources);
+	struct pci_bus *bus;
+	struct device_node *node = hose->dn;
+
+	pr_debug("PCI: Scanning PHB %s\n", of_node_full_name(node));
+
+	pcibios_setup_phb_resources(hose, &resources);
+
+	bus = pci_scan_root_bus(hose->parent, hose->first_busno,
+				hose->ops, hose, &resources);
+	if (bus == NULL) {
+		pr_err("Failed to create bus for PCI domain %04x\n",
+		       hose->global_number);
+		pci_free_resource_list(&resources);
+		return;
+	}
+	bus->busn_res.start = hose->first_busno;
+	hose->bus = bus;
+
+	hose->last_busno = bus->busn_res.end;
+}
+
+static int __init pcibios_init(void)
+{
+	struct pci_controller *hose, *tmp;
+	int next_busno = 0;
+
+	pr_info("PCI: Probing PCI hardware\n");
+
+	/* Scan all of the recorded PCI controllers.  */
+	list_for_each_entry_safe(hose, tmp, &hose_list, list_node) {
+		hose->last_busno = 0xff;
+		pcibios_scan_phb(hose);
+		if (next_busno <= hose->last_busno)
+			next_busno = hose->last_busno + 1;
+	}
+	pci_bus_count = next_busno;
+
+	/* Call common code to handle resource allocation */
+	pcibios_resource_survey();
+	list_for_each_entry_safe(hose, tmp, &hose_list, list_node) {
+		if (hose->bus)
+			pci_bus_add_devices(hose->bus);
+	}
+
+	return 0;
+}
+
+subsys_initcall(pcibios_init);
+
+static struct pci_controller *pci_bus_to_hose(int bus)
+{
+	struct pci_controller *hose, *tmp;
+
+	list_for_each_entry_safe(hose, tmp, &hose_list, list_node)
+		if (bus >= hose->first_busno && bus <= hose->last_busno)
+			return hose;
+	return NULL;
+}
+
+/* Provide information on locations of various I/O regions in physical
+ * memory.  Do this on a per-card basis so that we choose the right
+ * root bridge.
+ * Note that the returned IO or memory base is a physical address
+ */
+
+long sys_pciconfig_iobase(long which, unsigned long bus, unsigned long devfn)
+{
+	struct pci_controller *hose;
+	long result = -EOPNOTSUPP;
+
+	hose = pci_bus_to_hose(bus);
+	if (!hose)
+		return -ENODEV;
+
+	switch (which) {
+	case IOBASE_BRIDGE_NUMBER:
+		return (long)hose->first_busno;
+	case IOBASE_MEMORY:
+		return (long)hose->pci_mem_offset;
+	case IOBASE_IO:
+		return (long)hose->io_base_phys;
+	case IOBASE_ISA_IO:
+		return (long)isa_io_base;
+	case IOBASE_ISA_MEM:
+		return (long)isa_mem_base;
+	}
+
+	return result;
+}
+
+/*
+ * Null PCI config access functions, for the case when we can't
+ * find a hose.
+ */
+#define NULL_PCI_OP(rw, size, type)					\
+static int								\
+null_##rw##_config_##size(struct pci_dev *dev, int offset, type val)	\
+{									\
+	return PCIBIOS_DEVICE_NOT_FOUND;				\
+}
+
+static int
+null_read_config(struct pci_bus *bus, unsigned int devfn, int offset,
+		 int len, u32 *val)
+{
+	return PCIBIOS_DEVICE_NOT_FOUND;
+}
+
+static int
+null_write_config(struct pci_bus *bus, unsigned int devfn, int offset,
+		  int len, u32 val)
+{
+	return PCIBIOS_DEVICE_NOT_FOUND;
+}
+
+static struct pci_ops null_pci_ops = {
+	.read = null_read_config,
+	.write = null_write_config,
+};
+
+/*
+ * These functions are used early on before PCI scanning is done
+ * and all of the pci_dev and pci_bus structures have been created.
+ */
+static struct pci_bus *
+fake_pci_bus(struct pci_controller *hose, int busnr)
+{
+	static struct pci_bus bus;
+
+	if (!hose)
+		pr_err("Can't find hose for PCI bus %d!\n", busnr);
+
+	bus.number = busnr;
+	bus.sysdata = hose;
+	bus.ops = hose ? hose->ops : &null_pci_ops;
+	return &bus;
+}
+
+#define EARLY_PCI_OP(rw, size, type)					\
+int early_##rw##_config_##size(struct pci_controller *hose, int bus,	\
+			       int devfn, int offset, type value)	\
+{									\
+	return pci_bus_##rw##_config_##size(fake_pci_bus(hose, bus),	\
+					    devfn, offset, value);	\
+}
+
+EARLY_PCI_OP(read, byte, u8 *)
+EARLY_PCI_OP(read, word, u16 *)
+EARLY_PCI_OP(read, dword, u32 *)
+EARLY_PCI_OP(write, byte, u8)
+EARLY_PCI_OP(write, word, u16)
+EARLY_PCI_OP(write, dword, u32)
+
+int early_find_capability(struct pci_controller *hose, int bus, int devfn,
+			  int cap)
+{
+	return pci_bus_find_capability(fake_pci_bus(hose, bus), devfn, cap);
+}
+
diff --git a/arch/microblaze/pci/xilinx_pci.c b/arch/microblaze/pci/xilinx_pci.c
new file mode 100644
index 0000000..14c7da5
--- /dev/null
+++ b/arch/microblaze/pci/xilinx_pci.c
@@ -0,0 +1,169 @@
+/*
+ * PCI support for Xilinx plbv46_pci soft-core which can be used on
+ * Xilinx Virtex ML410 / ML510 boards.
+ *
+ * Copyright 2009 Roderick Colenbrander
+ * Copyright 2009 Secret Lab Technologies Ltd.
+ *
+ * The pci bridge fixup code was copied from ppc4xx_pci.c and was written
+ * by Benjamin Herrenschmidt.
+ * Copyright 2007 Ben. Herrenschmidt <benh@kernel.crashing.org>, IBM Corp.
+ *
+ * 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/ioport.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/pci.h>
+#include <linux/io.h>
+
+#define XPLB_PCI_ADDR 0x10c
+#define XPLB_PCI_DATA 0x110
+#define XPLB_PCI_BUS  0x114
+
+#define PCI_HOST_ENABLE_CMD (PCI_COMMAND_SERR | PCI_COMMAND_PARITY | \
+				PCI_COMMAND_MASTER | PCI_COMMAND_MEMORY)
+
+static struct of_device_id xilinx_pci_match[] = {
+	{ .compatible = "xlnx,plbv46-pci-1.03.a", },
+	{}
+};
+
+/**
+ * xilinx_pci_fixup_bridge - Block Xilinx PHB configuration.
+ */
+static void xilinx_pci_fixup_bridge(struct pci_dev *dev)
+{
+	struct pci_controller *hose;
+	int i;
+
+	if (dev->devfn || dev->bus->self)
+		return;
+
+	hose = pci_bus_to_host(dev->bus);
+	if (!hose)
+		return;
+
+	if (!of_match_node(xilinx_pci_match, hose->dn))
+		return;
+
+	/* Hide the PCI host BARs from the kernel as their content doesn't
+	 * fit well in the resource management
+	 */
+	for (i = 0; i < DEVICE_COUNT_RESOURCE; i++) {
+		dev->resource[i].start = 0;
+		dev->resource[i].end = 0;
+		dev->resource[i].flags = 0;
+	}
+
+	dev_info(&dev->dev, "Hiding Xilinx plb-pci host bridge resources %s\n",
+		 pci_name(dev));
+}
+DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, xilinx_pci_fixup_bridge);
+
+#ifdef DEBUG
+/**
+ * xilinx_pci_exclude_device - Don't do config access for non-root bus
+ *
+ * This is a hack.  Config access to any bus other than bus 0 does not
+ * currently work on the ML510 so we prevent it here.
+ */
+static int
+xilinx_pci_exclude_device(struct pci_controller *hose, u_char bus, u8 devfn)
+{
+	return (bus != 0);
+}
+
+/**
+ * xilinx_early_pci_scan - List pci config space for available devices
+ *
+ * List pci devices in very early phase.
+ */
+static void __init xilinx_early_pci_scan(struct pci_controller *hose)
+{
+	u32 bus = 0;
+	u32 val, dev, func, offset;
+
+	/* Currently we have only 2 device connected - up-to 32 devices */
+	for (dev = 0; dev < 2; dev++) {
+		/* List only first function number - up-to 8 functions */
+		for (func = 0; func < 1; func++) {
+			pr_info("%02x:%02x:%02x", bus, dev, func);
+			/* read the first 64 standardized bytes */
+			/* Up-to 192 bytes can be list of capabilities */
+			for (offset = 0; offset < 64; offset += 4) {
+				early_read_config_dword(hose, bus,
+					PCI_DEVFN(dev, func), offset, &val);
+				if (offset == 0 && val == 0xFFFFFFFF) {
+					pr_cont("\nABSENT");
+					break;
+				}
+				if (!(offset % 0x10))
+					pr_cont("\n%04x:    ", offset);
+
+				pr_cont("%08x  ", val);
+			}
+			pr_info("\n");
+		}
+	}
+}
+#else
+static void __init xilinx_early_pci_scan(struct pci_controller *hose)
+{
+}
+#endif
+
+/**
+ * xilinx_pci_init - Find and register a Xilinx PCI host bridge
+ */
+void __init xilinx_pci_init(void)
+{
+	struct pci_controller *hose;
+	struct resource r;
+	void __iomem *pci_reg;
+	struct device_node *pci_node;
+
+	pci_node = of_find_matching_node(NULL, xilinx_pci_match);
+	if (!pci_node)
+		return;
+
+	if (of_address_to_resource(pci_node, 0, &r)) {
+		pr_err("xilinx-pci: cannot resolve base address\n");
+		return;
+	}
+
+	hose = pcibios_alloc_controller(pci_node);
+	if (!hose) {
+		pr_err("xilinx-pci: pcibios_alloc_controller() failed\n");
+		return;
+	}
+
+	/* Setup config space */
+	setup_indirect_pci(hose, r.start + XPLB_PCI_ADDR,
+			   r.start + XPLB_PCI_DATA,
+			   INDIRECT_TYPE_SET_CFG_TYPE);
+
+	/* According to the xilinx plbv46_pci documentation the soft-core starts
+	 * a self-init when the bus master enable bit is set. Without this bit
+	 * set the pci bus can't be scanned.
+	 */
+	early_write_config_word(hose, 0, 0, PCI_COMMAND, PCI_HOST_ENABLE_CMD);
+
+	/* Set the max latency timer to 255 */
+	early_write_config_byte(hose, 0, 0, PCI_LATENCY_TIMER, 0xff);
+
+	/* Set the max bus number to 255, and bus/subbus no's to 0 */
+	pci_reg = of_iomap(pci_node, 0);
+	out_be32(pci_reg + XPLB_PCI_BUS, 0x000000ff);
+	iounmap(pci_reg);
+
+	/* Register the host bridge with the linux kernel! */
+	pci_process_bridge_OF_ranges(hose, pci_node,
+					INDIRECT_TYPE_SET_CFG_TYPE);
+
+	pr_info("xilinx-pci: Registered PCI host bridge\n");
+	xilinx_early_pci_scan(hose);
+}