x86/kprobes: Fix crash when probing CS CALL instructionsgrafted

When using eBPF to probe CS CALL instructions within a function,
a crash can be triggered.

The eBPF tool probes offset 257 of the __hrtimer_run_queues()
function:

<__hrtimer_run_queues+249>:  nopl   0x0(%rax,%rax,1)
<__hrtimer_run_queues+254>:  mov    %r14,%rdi
<__hrtimer_run_queues+257>:  cs call <__x86_indirect_thunk_r12>
<__hrtimer_run_queues+263>:  mov    %eax,%r12d
<__hrtimer_run_queues+266>:  xchg   %ax,%ax
<__hrtimer_run_queues+268>:  mov    %r13,%rdi

Which triggers this crash:

  BUG: unable to handle page fault for address: 00000000000f41c9
  #PF: supervisor write access in kernel mode
  #PF: error_code(0x0002) - not-present page
  PGD 0 P4D 0
  Oops: 0002 [#1] SMP NOPTI
  CPU: 1 PID: 0 Comm: swapper/1 Kdump: loaded Tainted: P
  RIP: 0010:__hrtimer_run_queues+0x106/0x230

Note that __hrtimer_run_queues+0x106 is __hrtimer_run_queues+262, which is
at the 6th byte of the above CS CALL instruction. Since the CS CALL
instruction occupies 6 bytes, the exception occurred in the middle of that
call instruction.

The root cause is that when using eBPF tools to probe in the middle of a
function, a kprobe with INT3 is used as the underlying implementation.

During single-step emulation of the original CALL instruction,
int3_emulate_call() assumes that the probed CALL instruction is 5 bytes
long. However, the actual CS-prefixed CALL instruction occupies 6 bytes,
so it constructs an incorrect exception return address. When the CPU
returns from the kprobe handler, the next instruction to be executed is at
the address of the last byte of that CS CALL instruction. Coincidentally,
starting from that address, the CPU fetches and decodes a completely
different instruction, which ultimately triggers a kernel crash.

Fix the issue by using the actual instruction length obtained from
the instruction decoder when constructing the exception return
address, rather than relying on the hardcoded CALL_INSN_SIZE macro.

[ mingo: Refined the changelog ]

Fixes: 6256e668b7af ("x86/kprobes: Use int3 instead of debug trap for single-step")
Suggested-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Jinke Han <jinkehan@didiglobal.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Acked-by: Yafang Shao <laoar.shao@gmail.com>
Acked-by: Borislav Petkov <bp@alien8.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://patch.msgid.link/20260908073742.GA10517@didi-ThinkCentre-M920t-N000

133 files changed, 15007 insertions, 0 deletions

diff --git a/arch/microblaze/Kbuild b/arch/microblaze/Kbuild
new file mode 100644
index 000000000..077a0b8e9
--- /dev/null
+++ b/arch/microblaze/Kbuild
@@ -0,0 +1,8 @@
+# SPDX-License-Identifier: GPL-2.0-only
+obj-y			+= kernel/
+obj-y			+= mm/
+obj-$(CONFIG_PCI)	+= pci/
+obj-y			+= boot/dts/
+
+# for cleaning
+subdir- += boot
diff --git a/arch/microblaze/Kconfig b/arch/microblaze/Kconfig
new file mode 100644
index 000000000..484ebb3ba
--- /dev/null
+++ b/arch/microblaze/Kconfig
@@ -0,0 +1,218 @@
+# SPDX-License-Identifier: GPL-2.0-only
+config MICROBLAZE
+	def_bool y
+	select ARCH_32BIT_OFF_T
+	select ARCH_NO_SWAP
+	select ARCH_HAS_DMA_PREP_COHERENT
+	select ARCH_HAS_GCOV_PROFILE_ALL
+	select ARCH_HAS_SYNC_DMA_FOR_CPU
+	select ARCH_HAS_SYNC_DMA_FOR_DEVICE
+	select ARCH_MIGHT_HAVE_PC_PARPORT
+	select ARCH_WANT_IPC_PARSE_VERSION
+	select BUILDTIME_TABLE_SORT
+	select TIMER_OF
+	select CLONE_BACKWARDS3
+	select COMMON_CLK
+	select DMA_DIRECT_REMAP
+	select GENERIC_ATOMIC64
+	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 HAS_IOPORT if PCI
+	select HAVE_ARCH_HASH
+	select HAVE_ARCH_KGDB
+	select HAVE_ARCH_SECCOMP
+	select HAVE_DEBUG_KMEMLEAK
+	select HAVE_DMA_CONTIGUOUS
+	select HAVE_DYNAMIC_FTRACE
+	select HAVE_FUNCTION_GRAPH_TRACER
+	select HAVE_FUNCTION_TRACER
+	select HAVE_PAGE_SIZE_4KB
+	select HAVE_PCI
+	select IRQ_DOMAIN
+	select XILINX_INTC
+	select MODULES_USE_ELF_RELA
+	select OF
+	select OF_EARLY_FLATTREE
+	select PCI_DOMAINS_GENERIC if PCI
+	select PCI_SYSCALL if PCI
+	select CPU_NO_EFFICIENT_FFS
+	select MMU_GATHER_NO_RANGE
+	select SPARSE_IRQ
+	select ZONE_DMA
+	select TRACE_IRQFLAGS_SUPPORT
+	select GENERIC_IRQ_MULTI_HANDLER
+
+# Endianness selection
+choice
+	prompt "Endianness selection"
+	default CPU_LITTLE_ENDIAN
+	help
+	  microblaze architectures can be configured for either little or
+	  big endian formats. Be sure to select the appropriate mode.
+
+config CPU_BIG_ENDIAN
+	bool "Big endian"
+
+config CPU_LITTLE_ENDIAN
+	bool "Little endian"
+
+endchoice
+
+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
+
+source "arch/microblaze/Kconfig.platform"
+
+menu "Processor type and features"
+
+source "kernel/Kconfig.hz"
+
+config MMU
+	def_bool y
+
+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.
+
+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 HIGHMEM
+	bool "High memory support"
+	select KMAP_LOCAL
+	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
+	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"
+
+config TASK_SIZE_BOOL
+	bool "Set custom user task size"
+	depends on ADVANCED_OPTIONS
+	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"
+
+config MB_MANAGER
+	bool "Support for Microblaze Manager"
+	depends on ADVANCED_OPTIONS
+	help
+	  This option enables API for configuring the MicroBlaze manager
+	  control register, which is consumed by the break handler to
+	  block the break.
+
+	  Say N here unless you know what you are doing.
+
+endmenu
diff --git a/arch/microblaze/Kconfig.debug b/arch/microblaze/Kconfig.debug
new file mode 100644
index 000000000..a4e40e534
--- /dev/null
+++ b/arch/microblaze/Kconfig.debug
@@ -0,0 +1 @@
+# SPDX-License-Identifier: GPL-2.0-only
diff --git a/arch/microblaze/Kconfig.platform b/arch/microblaze/Kconfig.platform
new file mode 100644
index 000000000..9cf9007ed
--- /dev/null
+++ b/arch/microblaze/Kconfig.platform
@@ -0,0 +1,71 @@
+# SPDX-License-Identifier: GPL-2.0-only
+# For a description of the syntax of this configuration file,
+# see Documentation/kbuild/kconfig-language.rst.
+#
+# Platform selection Kconfig menu for MicroBlaze targets
+#
+
+menu "Platform options"
+
+config OPT_LIB_FUNCTION
+	bool "Optimized lib function"
+	default y
+	help
+	  Turns on optimized library functions (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 "Optimized lib function ASM"
+	depends on OPT_LIB_FUNCTION && (XILINX_MICROBLAZE0_USE_BARREL = 1)
+	depends on CPU_BIG_ENDIAN
+	default n
+	help
+	  Turns on optimized library functions (memcpy and memmove).
+	  They are written in assembly.
+
+# 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 000000000..02e6be9c5
--- /dev/null
+++ b/arch/microblaze/Makefile
@@ -0,0 +1,93 @@
+# SPDX-License-Identifier: GPL-2.0
+KBUILD_DEFCONFIG := mmu_defconfig
+
+UTS_SYSNAME = -DUTS_SYSNAME=\"Linux\"
+
+# What CPU version are we building for, and crack it open
+# as major.minor.rev
+CPU_VER   := $(CONFIG_XILINX_MICROBLAZE0_HW_VER)
+CPU_MAJOR := $(word 1, $(subst ., , $(CPU_VER)))
+CPU_MINOR := $(word 2, $(subst ., , $(CPU_VER)))
+CPU_REV   := $(word 3, $(subst ., , $(CPU_VER)))
+
+export CPU_VER CPU_MAJOR CPU_MINOR CPU_REV
+
+# Use cpu-related CONFIG_ vars to set compile options.
+# The various CONFIG_XILINX cpu features options are integers 0/1/2...
+# rather than bools y/n
+
+# Work out HW multipler support. This is tricky.
+# 1. Spartan2 has no HW multipliers.
+# 2. MicroBlaze v3.x always uses them, except in Spartan 2
+# 3. All other FPGa/CPU ver combos, we can trust the CONFIG_ settings
+ifeq (,$(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
+
+ifdef CONFIG_CPU_BIG_ENDIAN
+KBUILD_CFLAGS += -mbig-endian
+KBUILD_AFLAGS += -mbig-endian
+KBUILD_LDFLAGS += -EB
+else
+KBUILD_CFLAGS += -mlittle-endian
+KBUILD_AFLAGS += -mlittle-endian
+KBUILD_LDFLAGS += -EL
+endif
+
+CPUFLAGS-1 += $(call cc-option,-mcpu=v$(CPU_VER))
+
+# r31 holds current when in kernel mode
+KBUILD_CFLAGS += -ffixed-r31 $(CPUFLAGS-y) $(CPUFLAGS-1) $(CPUFLAGS-2)
+
+libs-y += arch/microblaze/lib/
+
+boot := arch/microblaze/boot
+
+# Are we making a simpleImage.<boardname> target? If so, crack out the boardname
+DTB:=$(subst simpleImage.,,$(filter simpleImage.%, $(MAKECMDGOALS)))
+
+export DTB
+
+all: linux.bin
+
+archheaders:
+	$(Q)$(MAKE) $(build)=arch/microblaze/kernel/syscalls all
+
+PHONY += linux.bin linux.bin.gz linux.bin.ub
+linux.bin.ub linux.bin.gz: linux.bin
+linux.bin: vmlinux
+linux.bin linux.bin.gz linux.bin.ub:
+	$(Q)$(MAKE) $(build)=$(boot) $(boot)/$@
+	@echo 'Kernel: $(boot)/$@ is ready' ' (#'$(or $(KBUILD_BUILD_VERSION),`cat .version`)')'
+
+PHONY += simpleImage.$(DTB)
+simpleImage.$(DTB): vmlinux
+	$(Q)$(MAKE) $(build)=$(boot) $(addprefix $(boot)/$@., ub unstrip strip)
+	@echo 'Kernel: $(boot)/$@ is ready' ' (#'$(or $(KBUILD_BUILD_VERSION),`cat .version`)')'
+
+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> - Create the following images with <dt>.dtb linked in'
+  echo '                    simpleImage.<dt>        : raw image'
+  echo '                    simpleImage.<dt>.ub     : raw image with U-Boot header'
+  echo '                    simpleImage.<dt>.unstrip: ELF (identical to vmlinux)'
+  echo '                    simpleImage.<dt>.strip  : stripped ELF'
+  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 += $(boot)/simpleImage.*
diff --git a/arch/microblaze/boot/.gitignore b/arch/microblaze/boot/.gitignore
new file mode 100644
index 000000000..11a9e229f
--- /dev/null
+++ b/arch/microblaze/boot/.gitignore
@@ -0,0 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0-only
+linux.bin*
+simpleImage.*
diff --git a/arch/microblaze/boot/Makefile b/arch/microblaze/boot/Makefile
new file mode 100644
index 000000000..2b42c370d
--- /dev/null
+++ b/arch/microblaze/boot/Makefile
@@ -0,0 +1,35 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# arch/microblaze/boot/Makefile
+#
+
+targets := linux.bin linux.bin.gz linux.bin.ub simpleImage.*
+
+OBJCOPYFLAGS := -R .note -R .comment -R .note.gnu.build-id -O binary
+
+$(obj)/linux.bin: vmlinux FORCE
+	$(call if_changed,objcopy)
+
+$(obj)/linux.bin.ub: $(obj)/linux.bin FORCE
+	$(call if_changed,uimage)
+
+$(obj)/linux.bin.gz: $(obj)/linux.bin FORCE
+	$(call if_changed,gzip)
+
+quiet_cmd_strip = STRIP   $< $@$2
+	cmd_strip = $(STRIP) -K microblaze_start -K _end -K __log_buf \
+				-K _fdt_start $< -o $@$2
+
+UIMAGE_LOADADDR = $(CONFIG_KERNEL_BASE_ADDR)
+
+$(obj)/simpleImage.$(DTB): vmlinux FORCE
+	$(call if_changed,objcopy)
+
+$(obj)/simpleImage.$(DTB).ub: $(obj)/simpleImage.$(DTB) FORCE
+	$(call if_changed,uimage)
+
+$(obj)/simpleImage.$(DTB).unstrip: vmlinux FORCE
+	$(call if_changed,copy)
+
+$(obj)/simpleImage.$(DTB).strip: vmlinux FORCE
+	$(call if_changed,strip)
diff --git a/arch/microblaze/boot/dts/Makefile b/arch/microblaze/boot/dts/Makefile
new file mode 100644
index 000000000..b84e2cbb2
--- /dev/null
+++ b/arch/microblaze/boot/dts/Makefile
@@ -0,0 +1,20 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+
+dtb-y := system.dtb
+
+ifneq ($(DTB),)
+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,copy)
+endif
+endif
+
+# 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 000000000..23345af37
--- /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 000000000..22252451e
--- /dev/null
+++ b/arch/microblaze/boot/dts/system.dts
@@ -0,0 +1,353 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Device Tree Generator version: 1.1
+ *
+ * (C) Copyright 2007-2008 Xilinx, Inc.
+ * (C) Copyright 2007-2009 Michal Simek
+ *
+ * Michal SIMEK <monstr@monstr.eu>
+ *
+ * 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";
+	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";
+		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>;
+			};
+		} ;
+
+		gpio-restart {
+			compatible = "gpio-restart";
+			/*
+			 * FIXME: is this active low or active high?
+			 * the current flag (1) indicates active low.
+			 * delay measures are templates, should be adjusted
+			 * to datasheet or trial-and-error with real hardware.
+			 */
+			gpios = <&LEDs_8Bit 2 1>;
+			active-delay = <100>;
+			inactive-delay = <10>;
+			wait-delay = <100>;
+		};
+
+		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>;
+		} ;
+		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,one-timer-only = <0x0>;
+		} ;
+	} ;
+}  ;
diff --git a/arch/microblaze/configs/mmu_defconfig b/arch/microblaze/configs/mmu_defconfig
new file mode 100644
index 000000000..fbbdcb394
--- /dev/null
+++ b/arch/microblaze/configs/mmu_defconfig
@@ -0,0 +1,91 @@
+CONFIG_SYSVIPC=y
+CONFIG_POSIX_MQUEUE=y
+CONFIG_AUDIT=y
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
+CONFIG_EXPERT=y
+CONFIG_BASE_SMALL=y
+CONFIG_KALLSYMS_ALL=y
+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_HIGHMEM=y
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+CONFIG_PARTITION_ADVANCED=y
+# CONFIG_EFI_PARTITION is not set
+CONFIG_CMA=y
+CONFIG_NET=y
+CONFIG_PACKET=y
+CONFIG_UNIX=y
+CONFIG_INET=y
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+CONFIG_IP_PNP_BOOTP=y
+CONFIG_IP_PNP_RARP=y
+# CONFIG_IPV6 is not set
+CONFIG_BRIDGE=m
+CONFIG_PCI=y
+CONFIG_DEVTMPFS=y
+CONFIG_DEVTMPFS_MOUNT=y
+CONFIG_MTD=y
+CONFIG_MTD_CFI=y
+CONFIG_MTD_CFI_INTELEXT=y
+CONFIG_MTD_CFI_AMDSTD=y
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_SIZE=8192
+CONFIG_NETDEVICES=y
+CONFIG_XILINX_EMACLITE=y
+CONFIG_XILINX_AXI_EMAC=y
+CONFIG_XILINX_LL_TEMAC=y
+CONFIG_MARVELL_PHY=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_OF_PLATFORM=y
+CONFIG_SERIAL_UARTLITE=y
+CONFIG_SERIAL_UARTLITE_CONSOLE=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_POWER_RESET=y
+CONFIG_POWER_RESET_GPIO_RESTART=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_GENIRQ=y
+CONFIG_UIO_DMEM_GENIRQ=y
+CONFIG_EXT4_FS=y
+# CONFIG_DNOTIFY is not set
+CONFIG_TMPFS=y
+CONFIG_CRAMFS=y
+CONFIG_ROMFS_FS=y
+CONFIG_NFS_FS=y
+CONFIG_ROOT_NFS=y
+CONFIG_CIFS=y
+CONFIG_ENCRYPTED_KEYS=y
+CONFIG_DMA_CMA=y
+CONFIG_DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT=y
+CONFIG_KGDB=y
+CONFIG_KGDB_TESTS=y
+CONFIG_KGDB_KDB=y
+CONFIG_DETECT_HUNG_TASK=y
+CONFIG_DEBUG_SPINLOCK=y
diff --git a/arch/microblaze/include/asm/Kbuild b/arch/microblaze/include/asm/Kbuild
new file mode 100644
index 000000000..0030309b4
--- /dev/null
+++ b/arch/microblaze/include/asm/Kbuild
@@ -0,0 +1,12 @@
+# SPDX-License-Identifier: GPL-2.0
+generated-y += syscall_table.h
+generic-y += cmpxchg.h
+generic-y += extable.h
+generic-y += kvm_para.h
+generic-y += mcs_spinlock.h
+generic-y += parport.h
+generic-y += ring_buffer.h
+generic-y += syscalls.h
+generic-y += tlb.h
+generic-y += user.h
+generic-y += text-patching.h
diff --git a/arch/microblaze/include/asm/asm-compat.h b/arch/microblaze/include/asm/asm-compat.h
new file mode 100644
index 000000000..9f0461476
--- /dev/null
+++ b/arch/microblaze/include/asm/asm-compat.h
@@ -0,0 +1,18 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _ASM_MICROBLAZE_ASM_COMPAT_H
+#define _ASM_MICROBLAZE_ASM_COMPAT_H
+
+#include <asm/types.h>
+
+#ifdef __ASSEMBLER__
+#  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 000000000..d370ee36a
--- /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/barrier.h b/arch/microblaze/include/asm/barrier.h
new file mode 100644
index 000000000..70b0a0177
--- /dev/null
+++ b/arch/microblaze/include/asm/barrier.h
@@ -0,0 +1,13 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (c) 2015 - 2020 Xilinx, Inc. All rights reserved.
+ */
+
+#ifndef _ASM_MICROBLAZE_BARRIER_H
+#define _ASM_MICROBLAZE_BARRIER_H
+
+#define mb()	__asm__ __volatile__ ("mbar 1" : : : "memory")
+
+#include <asm-generic/barrier.h>
+
+#endif /* _ASM_MICROBLAZE_BARRIER_H */
diff --git a/arch/microblaze/include/asm/cache.h b/arch/microblaze/include/asm/cache.h
new file mode 100644
index 000000000..1903988b9
--- /dev/null
+++ b/arch/microblaze/include/asm/cache.h
@@ -0,0 +1,26 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * 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>
+ */
+
+#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
+
+/* MS be sure that SLAB allocates aligned objects */
+#define ARCH_DMA_MINALIGN	L1_CACHE_BYTES
+
+#define ARCH_SLAB_MINALIGN	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 000000000..ffa2cf389
--- /dev/null
+++ b/arch/microblaze/include/asm/cacheflush.h
@@ -0,0 +1,103 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * 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>
+ */
+
+#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/core-api/cachetlb.rst */
+
+/*
+ * 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 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);
+
+static inline void flush_dcache_folio(struct folio *folio)
+{
+	unsigned long addr = folio_pfn(folio) << PAGE_SHIFT;
+
+	flush_dcache_range(addr, addr + folio_size(folio));
+}
+#define flush_dcache_folio flush_dcache_folio
+
+#define flush_cache_page(vma, vmaddr, pfn) \
+	flush_dcache_range(pfn << PAGE_SHIFT, (pfn << PAGE_SHIFT) + PAGE_SIZE);
+
+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);
+	}
+}
+#define copy_to_user_page copy_to_user_page
+
+#include <asm-generic/cacheflush.h>
+
+#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 000000000..2e5ebd599
--- /dev/null
+++ b/arch/microblaze/include/asm/checksum.h
@@ -0,0 +1,36 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2008 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+
+#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, __u32 len,
+		   __u8 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/cpuinfo.h b/arch/microblaze/include/asm/cpuinfo.h
new file mode 100644
index 000000000..786ffa669
--- /dev/null
+++ b/arch/microblaze/include/asm/cpuinfo.h
@@ -0,0 +1,105 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * 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>
+ */
+
+#ifndef _ASM_MICROBLAZE_CPUINFO_H
+#define _ASM_MICROBLAZE_CPUINFO_H
+
+#include <linux/of.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/current.h b/arch/microblaze/include/asm/current.h
new file mode 100644
index 000000000..099e69f32
--- /dev/null
+++ b/arch/microblaze/include/asm/current.h
@@ -0,0 +1,26 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+
+#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 __ASSEMBLER__
+/*
+ * Dedicate r31 to keeping the current task pointer
+ */
+register struct task_struct *current asm("r31");
+
+# define get_current()	current
+# endif /* __ASSEMBLER__ */
+
+#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 000000000..05fe9e3e0
--- /dev/null
+++ b/arch/microblaze/include/asm/delay.h
@@ -0,0 +1,85 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * 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/dma.h b/arch/microblaze/include/asm/dma.h
new file mode 100644
index 000000000..7484c9eb6
--- /dev/null
+++ b/arch/microblaze/include/asm/dma.h
@@ -0,0 +1,12 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+
+#ifndef _ASM_MICROBLAZE_DMA_H
+#define _ASM_MICROBLAZE_DMA_H
+
+/* Virtual address corresponding to last available physical memory address.  */
+#define MAX_DMA_ADDRESS (CONFIG_KERNEL_START + memory_size - 1)
+
+#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 000000000..5331a8473
--- /dev/null
+++ b/arch/microblaze/include/asm/elf.h
@@ -0,0 +1,27 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+#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/entry.h b/arch/microblaze/include/asm/entry.h
new file mode 100644
index 000000000..9efadf123
--- /dev/null
+++ b/arch/microblaze/include/asm/entry.h
@@ -0,0 +1,34 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * 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>
+ */
+
+#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 __ASSEMBLER__
+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 /* __ASSEMBLER__ */
+
+#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 000000000..c4591e4f7
--- /dev/null
+++ b/arch/microblaze/include/asm/exceptions.h
@@ -0,0 +1,69 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * 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>
+ */
+
+#ifndef _ASM_MICROBLAZE_EXCEPTIONS_H
+#define _ASM_MICROBLAZE_EXCEPTIONS_H
+
+#ifdef __KERNEL__
+#ifndef __ASSEMBLER__
+
+/* 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 /*__ASSEMBLER__ */
+#endif /* __KERNEL__ */
+#endif /* _ASM_MICROBLAZE_EXCEPTIONS_H */
diff --git a/arch/microblaze/include/asm/fixmap.h b/arch/microblaze/include/asm/fixmap.h
new file mode 100644
index 000000000..f9797849e
--- /dev/null
+++ b/arch/microblaze/include/asm/fixmap.h
@@ -0,0 +1,66 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * fixmap.h: compile-time virtual memory allocation
+ *
+ * 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 __ASSEMBLER__
+#include <linux/kernel.h>
+#include <asm/page.h>
+#ifdef CONFIG_HIGHMEM
+#include <linux/threads.h>
+#include <asm/kmap_size.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_MAX_IDX * 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 /* !__ASSEMBLER__ */
+#endif
diff --git a/arch/microblaze/include/asm/flat.h b/arch/microblaze/include/asm/flat.h
new file mode 100644
index 000000000..edff4306f
--- /dev/null
+++ b/arch/microblaze/include/asm/flat.h
@@ -0,0 +1,81 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * uClinux flat-format executables
+ *
+ * Copyright (C) 2005 John Williams <jwilliams@itee.uq.edu.au>
+ */
+
+#ifndef _ASM_MICROBLAZE_FLAT_H
+#define _ASM_MICROBLAZE_FLAT_H
+
+#include <linux/unaligned.h>
+
+/*
+ * 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 int flat_get_addr_from_rp(u32 __user *rp, u32 relval, u32 flags,
+					u32 *addr)
+{
+	u32 *p = (__force u32 *)rp;
+
+	/* Is it a split 64/32 reference? */
+	if (relval & 0x80000000) {
+		/* Grab the two halves of the reference */
+		u32 val_hi, val_lo;
+
+		val_hi = get_unaligned(p);
+		val_lo = get_unaligned(p+1);
+
+		/* Crack the address out */
+		*addr = ((val_hi & 0xffff) << 16) + (val_lo & 0xffff);
+	} else {
+		/* Get the address straight out */
+		*addr = get_unaligned(p);
+	}
+
+	return 0;
+}
+
+/*
+ * Insert an address into the symbol reference at rp. rp is potentially
+ * unaligned.
+ */
+
+static inline int
+flat_put_addr_at_rp(u32 __user *rp, u32 addr, u32 relval)
+{
+	u32 *p = (__force u32 *)rp;
+	/* Is this a split 64/32 reloc? */
+	if (relval & 0x80000000) {
+		/* Get the two "halves" */
+		unsigned long val_hi = get_unaligned(p);
+		unsigned long val_lo = get_unaligned(p + 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, p);
+		put_unaligned(val_lo, p+1);
+	} else {
+		/* Put it straight in, no messing around */
+		put_unaligned(addr, p);
+	}
+	return 0;
+}
+
+#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 000000000..27c1bafb6
--- /dev/null
+++ b/arch/microblaze/include/asm/ftrace.h
@@ -0,0 +1,28 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#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 __ASSEMBLER__
+extern void _mcount(void);
+extern void ftrace_call_graph(void);
+void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr);
+#endif
+
+#ifdef CONFIG_DYNAMIC_FTRACE
+/* relocation 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 000000000..86131ed84
--- /dev/null
+++ b/arch/microblaze/include/asm/futex.h
@@ -0,0 +1,99 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#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
+arch_futex_atomic_op_inuser(int op, int oparg, int *oval, u32 __user *uaddr)
+{
+	int oldval = 0, ret;
+
+	if (!access_ok(uaddr, sizeof(u32)))
+		return -EFAULT;
+
+	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;
+	}
+
+	if (!ret)
+		*oval = oldval;
+
+	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(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/hash.h b/arch/microblaze/include/asm/hash.h
new file mode 100644
index 000000000..ef4741a40
--- /dev/null
+++ b/arch/microblaze/include/asm/hash.h
@@ -0,0 +1,82 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _ASM_HASH_H
+#define _ASM_HASH_H
+
+/*
+ * Fortunately, most people who want to run Linux on Microblaze enable
+ * both multiplier and barrel shifter, but omitting them is technically
+ * a supported configuration.
+ *
+ * With just a barrel shifter, we can implement an efficient constant
+ * multiply using shifts and adds.  GCC can find a 9-step solution, but
+ * this 6-step solution was found by Yevgen Voronenko's implementation
+ * of the Hcub algorithm at http://spiral.ece.cmu.edu/mcm/gen.html.
+ *
+ * That software is really not designed for a single multiplier this large,
+ * but if you run it enough times with different seeds, it'll find several
+ * 6-shift, 6-add sequences for computing x * 0x61C88647.  They are all
+ *	c = (x << 19) + x;
+ *	a = (x <<  9) + c;
+ *	b = (x << 23) + a;
+ *	return (a<<11) + (b<<6) + (c<<3) - b;
+ * with variations on the order of the final add.
+ *
+ * Without even a shifter, it's hopless; any hash function will suck.
+ */
+
+#if CONFIG_XILINX_MICROBLAZE0_USE_HW_MUL == 0
+
+#define HAVE_ARCH__HASH_32 1
+
+/* Multiply by GOLDEN_RATIO_32 = 0x61C88647 */
+static inline u32 __attribute_const__ __hash_32(u32 a)
+{
+#if CONFIG_XILINX_MICROBLAZE0_USE_BARREL
+	unsigned int b, c;
+
+	/* Phase 1: Compute three intermediate values */
+	b =  a << 23;
+	c = (a << 19) + a;
+	a = (a <<  9) + c;
+	b += a;
+
+	/* Phase 2: Compute (a << 11) + (b << 6) + (c << 3) - b */
+	a <<= 5;
+	a += b;		/* (a << 5) + b */
+	a <<= 3;
+	a += c;		/* (a << 8) + (b << 3) + c */
+	a <<= 3;
+	return a - b;	/* (a << 11) + (b << 6) + (c << 3) - b */
+#else
+	/*
+	 * "This is really going to hurt."
+	 *
+	 * Without a barrel shifter, left shifts are implemented as
+	 * repeated additions, and the best we can do is an optimal
+	 * addition-subtraction chain.  This one is not known to be
+	 * optimal, but at 37 steps, it's decent for a 31-bit multiplier.
+	 *
+	 * Question: given its size (37*4 = 148 bytes per instance),
+	 * and slowness, is this worth having inline?
+	 */
+	unsigned int b, c, d;
+
+	b = a << 4;	/* 4    */
+	c = b << 1;	/* 1  5 */
+	b += a;		/* 1  6 */
+	c += b;		/* 1  7 */
+	c <<= 3;	/* 3 10 */
+	c -= a;		/* 1 11 */
+	d = c << 7;	/* 7 18 */
+	d += b;		/* 1 19 */
+	d <<= 8;	/* 8 27 */
+	d += a;		/* 1 28 */
+	d <<= 1;	/* 1 29 */
+	d += b;		/* 1 30 */
+	d <<= 6;	/* 6 36 */
+	return d + c;	/* 1 37 total instructions*/
+#endif
+}
+
+#endif /* !CONFIG_XILINX_MICROBLAZE0_USE_HW_MUL */
+#endif /* _ASM_HASH_H */
diff --git a/arch/microblaze/include/asm/highmem.h b/arch/microblaze/include/asm/highmem.h
new file mode 100644
index 000000000..4418633fb
--- /dev/null
+++ b/arch/microblaze/include/asm/highmem.h
@@ -0,0 +1,61 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * 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 *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))
+
+#define flush_cache_kmaps()	{ flush_icache(); flush_dcache(); }
+
+#define arch_kmap_local_post_map(vaddr, pteval)	\
+	local_flush_tlb_page(NULL, vaddr);
+#define arch_kmap_local_post_unmap(vaddr)	\
+	local_flush_tlb_page(NULL, vaddr);
+
+#endif /* __KERNEL__ */
+
+#endif /* _ASM_HIGHMEM_H */
diff --git a/arch/microblaze/include/asm/io.h b/arch/microblaze/include/asm/io.h
new file mode 100644
index 000000000..c1d78b897
--- /dev/null
+++ b/arch/microblaze/include/asm/io.h
@@ -0,0 +1,60 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2007-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+
+#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)
+
+extern void iounmap(volatile void __iomem *addr);
+
+extern void __iomem *ioremap(phys_addr_t address, unsigned long size);
+
+/* 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 000000000..cb6ab55d1
--- /dev/null
+++ b/arch/microblaze/include/asm/irq.h
@@ -0,0 +1,14 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+
+#ifndef _ASM_MICROBLAZE_IRQ_H
+#define _ASM_MICROBLAZE_IRQ_H
+
+#include <asm-generic/irq.h>
+
+struct pt_regs;
+extern void do_IRQ(struct pt_regs *regs);
+
+#endif /* _ASM_MICROBLAZE_IRQ_H */
diff --git a/arch/microblaze/include/asm/irqflags.h b/arch/microblaze/include/asm/irqflags.h
new file mode 100644
index 000000000..818c6c9f5
--- /dev/null
+++ b/arch/microblaze/include/asm/irqflags.h
@@ -0,0 +1,119 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+
+#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/kgdb.h b/arch/microblaze/include/asm/kgdb.h
new file mode 100644
index 000000000..321c3c8bf
--- /dev/null
+++ b/arch/microblaze/include/asm/kgdb.h
@@ -0,0 +1,32 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifdef __KERNEL__
+#ifndef __MICROBLAZE_KGDB_H__
+#define __MICROBLAZE_KGDB_H__
+
+#ifndef __ASSEMBLER__
+
+#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 /* __ASSEMBLER__ */
+#endif /* __MICROBLAZE_KGDB_H__ */
+#endif /* __KERNEL__ */
diff --git a/arch/microblaze/include/asm/mmu.h b/arch/microblaze/include/asm/mmu.h
new file mode 100644
index 000000000..7262dc4da
--- /dev/null
+++ b/arch/microblaze/include/asm/mmu.h
@@ -0,0 +1,119 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+
+#ifndef _ASM_MICROBLAZE_MMU_H
+#define _ASM_MICROBLAZE_MMU_H
+
+#  ifdef __KERNEL__
+#   ifndef __ASSEMBLER__
+
+/* 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 /* __ASSEMBLER__ */
+
+/*
+ * 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 /* _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 000000000..866e52da5
--- /dev/null
+++ b/arch/microblaze/include/asm/mmu_context.h
@@ -0,0 +1,2 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+# include <asm/mmu_context_mm.h>
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 000000000..c2c77f708
--- /dev/null
+++ b/arch/microblaze/include/asm/mmu_context_mm.h
@@ -0,0 +1,140 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+
+#ifndef _ASM_MICROBLAZE_MMU_CONTEXT_H
+#define _ASM_MICROBLAZE_MMU_CONTEXT_H
+
+#include <linux/atomic.h>
+#include <linux/mm_types.h>
+#include <linux/sched.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.
+ */
+
+# 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.
+ */
+#define destroy_context destroy_context
+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.
+ */
+#define activate_mm activate_mm
+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);
+
+#include <asm-generic/mmu_context.h>
+
+# 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 000000000..eda1c183b
--- /dev/null
+++ b/arch/microblaze/include/asm/module.h
@@ -0,0 +1,28 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+
+#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/page.h b/arch/microblaze/include/asm/page.h
new file mode 100644
index 000000000..e1e396367
--- /dev/null
+++ b/arch/microblaze/include/asm/page.h
@@ -0,0 +1,141 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * 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.
+ */
+
+#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__
+
+#include <vdso/page.h>
+
+#define LOAD_OFFSET	ASM_CONST((CONFIG_KERNEL_START-CONFIG_KERNEL_BASE_ADDR))
+
+#define PTE_SHIFT	(PAGE_SHIFT - 2)	/* 1024 ptes per page */
+
+#ifndef __ASSEMBLER__
+
+/*
+ * 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"
+
+# define copy_page(to, from)			memcpy((to), (from), PAGE_SIZE)
+# define clear_page(pgaddr)			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 */
+typedef struct { unsigned long pgd; } pgd_t;
+
+# define pte_val(x)	((x).pte)
+# define pgprot_val(x)	((x).pgprot)
+
+#   define pgd_val(x)      ((x).pgd)
+
+# define __pte(x)	((pte_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_page(kaddr)	(pfn_to_page(__pa(kaddr) >> PAGE_SHIFT))
+#  define page_to_virt(page)   __va(page_to_pfn(page) << PAGE_SHIFT)
+
+#  define ARCH_PFN_OFFSET	(memory_start >> PAGE_SHIFT)
+# endif /* __ASSEMBLER__ */
+
+/* Convert between virtual and physical address for MMU. */
+/* Handle MicroBlaze processor with virtual memory. */
+#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)
+
+#ifndef __ASSEMBLER__
+
+# define __pa(x)	__virt_to_phys((unsigned long)(x))
+# define __va(x)	((void *)__phys_to_virt((unsigned long)(x)))
+
+static inline unsigned long virt_to_pfn(const void *vaddr)
+{
+	return phys_to_pfn(__pa(vaddr));
+}
+
+static inline const void *pfn_to_virt(unsigned long pfn)
+{
+	return __va(pfn_to_phys((pfn)));
+}
+
+#define	virt_addr_valid(vaddr)	(pfn_valid(virt_to_pfn(vaddr)))
+
+#endif /* __ASSEMBLER__ */
+
+#define TOPHYS(addr)  __virt_to_phys(addr)
+
+#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/pci-bridge.h b/arch/microblaze/include/asm/pci-bridge.h
new file mode 100644
index 000000000..be5f504be
--- /dev/null
+++ b/arch/microblaze/include/asm/pci-bridge.h
@@ -0,0 +1,49 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+#ifndef _ASM_MICROBLAZE_PCI_BRIDGE_H
+#define _ASM_MICROBLAZE_PCI_BRIDGE_H
+#ifdef __KERNEL__
+/*
+ */
+#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;
+	struct list_head list_node;
+
+	void __iomem *io_base_virt;
+
+	/* 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;
+};
+
+#ifdef CONFIG_PCI
+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 */
+
+#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 000000000..91f1f71c2
--- /dev/null
+++ b/arch/microblaze/include/asm/pci.h
@@ -0,0 +1,44 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ *
+ * 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/pci-bridge.h>
+
+#define PCIBIOS_MIN_IO		0x1000
+#define PCIBIOS_MIN_MEM		0x10000000
+
+/*
+ * 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);
+
+/* Tell PCI code what kind of PCI resource mappings we support */
+#define HAVE_PCI_MMAP			1
+#define ARCH_GENERIC_PCI_MMAP_RESOURCE	1
+
+struct file;
+
+static inline void __init xilinx_pci_init(void) { return; }
+
+#endif	/* __KERNEL__ */
+#endif /* __ASM_MICROBLAZE_PCI_H */
diff --git a/arch/microblaze/include/asm/pgalloc.h b/arch/microblaze/include/asm/pgalloc.h
new file mode 100644
index 000000000..084a8a0dc
--- /dev/null
+++ b/arch/microblaze/include/asm/pgalloc.h
@@ -0,0 +1,36 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+
+#ifndef _ASM_MICROBLAZE_PGALLOC_H
+#define _ASM_MICROBLAZE_PGALLOC_H
+
+#include <linux/kernel.h>	/* For min/max macros */
+#include <linux/highmem.h>
+#include <linux/pgtable.h>
+#include <asm/setup.h>
+#include <asm/io.h>
+#include <asm/page.h>
+#include <asm/cache.h>
+
+#define __HAVE_ARCH_PTE_ALLOC_ONE_KERNEL
+#include <asm-generic/pgalloc.h>
+
+extern void __bad_pte(pmd_t *pmd);
+
+#define pgd_alloc(mm)		__pgd_alloc(mm, 0)
+
+extern pte_t *pte_alloc_one_kernel(struct mm_struct *mm);
+
+#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))
+
+#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 000000000..7678c040a
--- /dev/null
+++ b/arch/microblaze/include/asm/pgtable.h
@@ -0,0 +1,437 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+
+#ifndef _ASM_MICROBLAZE_PGTABLE_H
+#define _ASM_MICROBLAZE_PGTABLE_H
+
+#include <asm/setup.h>
+
+#ifndef __ASSEMBLER__
+extern int mem_init_done;
+#endif
+
+#include <asm-generic/pgtable-nopmd.h>
+
+#ifdef __KERNEL__
+#ifndef __ASSEMBLER__
+
+#include <linux/sched.h>
+#include <linux/threads.h>
+#include <asm/processor.h>		/* For TASK_SIZE */
+#include <asm/mmu.h>
+#include <asm/page.h>
+
+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..
+ */
+
+/* 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 /* __ASSEMBLER__ */
+
+/*
+ * 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
+ *
+ */
+
+/* PGDIR_SHIFT determines what a top-level page table entry can map */
+#define PGDIR_SHIFT	(PAGE_SHIFT + PTE_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 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 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 PTEs 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 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
+
+/* We borrow bit 24 to store the exclusive marker in swap PTEs. */
+#define _PAGE_SWP_EXCLUSIVE	_PAGE_DIRTY
+
+/*
+ * 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 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_PTE_SHIFT		PAGE_SHIFT
+
+#define pte_pfn(x)		(pte_val(x) >> PFN_PTE_SHIFT)
+
+#define pfn_pte(pfn, prot) \
+	__pte(((pte_basic_t)(pfn) << PFN_PTE_SHIFT) | pgprot_val(prot))
+
+#ifndef __ASSEMBLER__
+/*
+ * 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_novma(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;
+}
+
+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(pte_t *ptep, pte_t pte)
+{
+	*ptep = pte;
+}
+
+#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
+struct vm_area_struct;
+static inline bool 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*/
+static inline unsigned long pmd_page_vaddr(pmd_t pmd)
+{
+	return ((unsigned long) (pmd_val(pmd) & PAGE_MASK));
+}
+
+/* returns pfn of the pmd entry*/
+#define pmd_pfn(pmd)	(__pa(pmd_val(pmd)) >> PAGE_SHIFT)
+
+/* returns struct *page of the pmd entry*/
+#define pmd_page(pmd)	(pfn_to_page(__pa(pmd_val(pmd)) >> PAGE_SHIFT))
+
+/* Find an entry in the third-level page table.. */
+
+extern pgd_t swapper_pg_dir[PTRS_PER_PGD];
+
+/*
+ * Encode/decode swap entries and swap PTEs. Swap PTEs are all PTEs that
+ * are !pte_none() && !pte_present().
+ *
+ *                         1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 3 3
+ *   0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+ *   <------------------ offset -------------------> E < type -> 0 0
+ *
+ *   E is the exclusive marker that is not stored in swap entries.
+ */
+#define __swp_type(entry)	((entry).val & 0x1f)
+#define __swp_offset(entry)	((entry).val >> 6)
+#define __swp_entry(type, offset) \
+		((swp_entry_t) { ((type) & 0x1f) | ((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 })
+
+static inline bool pte_swp_exclusive(pte_t pte)
+{
+	return pte_val(pte) & _PAGE_SWP_EXCLUSIVE;
+}
+
+static inline pte_t pte_swp_mkexclusive(pte_t pte)
+{
+	pte_val(pte) |= _PAGE_SWP_EXCLUSIVE;
+	return pte;
+}
+
+static inline pte_t pte_swp_clear_exclusive(pte_t pte)
+{
+	pte_val(pte) &= ~_PAGE_SWP_EXCLUSIVE;
+	return pte;
+}
+
+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
+
+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);
+
+#endif /* __ASSEMBLER__ */
+#endif /* __KERNEL__ */
+
+#ifndef __ASSEMBLER__
+extern unsigned long ioremap_bot, ioremap_base;
+
+void setup_memory(void);
+#endif /* __ASSEMBLER__ */
+
+#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 000000000..d59bdfffc
--- /dev/null
+++ b/arch/microblaze/include/asm/processor.h
@@ -0,0 +1,92 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+
+#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 __ASSEMBLER__
+/* from kernel/cpu/mb.c */
+extern const struct seq_operations cpuinfo_op;
+
+# define cpu_relax()		barrier()
+
+#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 /* __ASSEMBLER__ */
+
+/*
+ * 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 __ASSEMBLER__
+
+/* 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, \
+}
+
+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))
+
+#  define STACK_TOP	TASK_SIZE
+#  define STACK_TOP_MAX	STACK_TOP
+
+#ifdef CONFIG_DEBUG_FS
+extern struct dentry *of_debugfs_root;
+#endif
+
+#  endif /* __ASSEMBLER__ */
+#endif /* _ASM_MICROBLAZE_PROCESSOR_H */
diff --git a/arch/microblaze/include/asm/ptrace.h b/arch/microblaze/include/asm/ptrace.h
new file mode 100644
index 000000000..17982292a
--- /dev/null
+++ b/arch/microblaze/include/asm/ptrace.h
@@ -0,0 +1,24 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+#ifndef _ASM_MICROBLAZE_PTRACE_H
+#define _ASM_MICROBLAZE_PTRACE_H
+
+#include <uapi/asm/ptrace.h>
+
+#ifndef __ASSEMBLER__
+#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 /* __ASSEMBLER__ */
+#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 000000000..186ee8c3c
--- /dev/null
+++ b/arch/microblaze/include/asm/pvr.h
@@ -0,0 +1,224 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * 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
+ */
+
+#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 000000000..6b36693fc
--- /dev/null
+++ b/arch/microblaze/include/asm/registers.h
@@ -0,0 +1,45 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+
+#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 */
+
+/* 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 /* _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 000000000..95cdcabee
--- /dev/null
+++ b/arch/microblaze/include/asm/seccomp.h
@@ -0,0 +1,11 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#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 000000000..f5008f5e7
--- /dev/null
+++ b/arch/microblaze/include/asm/sections.h
@@ -0,0 +1,20 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+
+#ifndef _ASM_MICROBLAZE_SECTIONS_H
+#define _ASM_MICROBLAZE_SECTIONS_H
+
+#include <asm-generic/sections.h>
+
+# ifndef __ASSEMBLER__
+extern char _ssbss[], _esbss[];
+extern unsigned long __ivt_start[], __ivt_end[];
+
+extern u32 _fdt_start[], _fdt_end[];
+
+# endif /* !__ASSEMBLER__ */
+#endif /* _ASM_MICROBLAZE_SECTIONS_H */
diff --git a/arch/microblaze/include/asm/setup.h b/arch/microblaze/include/asm/setup.h
new file mode 100644
index 000000000..837ed0bba
--- /dev/null
+++ b/arch/microblaze/include/asm/setup.h
@@ -0,0 +1,29 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2007-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+#ifndef _ASM_MICROBLAZE_SETUP_H
+#define _ASM_MICROBLAZE_SETUP_H
+
+#include <uapi/asm/setup.h>
+
+# ifndef __ASSEMBLER__
+extern char cmd_line[COMMAND_LINE_SIZE];
+
+extern char *klimit;
+
+extern void mmu_reset(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);
+
+# endif /* __ASSEMBLER__ */
+#endif /* _ASM_MICROBLAZE_SETUP_H */
diff --git a/arch/microblaze/include/asm/string.h b/arch/microblaze/include/asm/string.h
new file mode 100644
index 000000000..8798ad2c1
--- /dev/null
+++ b/arch/microblaze/include/asm/string.h
@@ -0,0 +1,23 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+
+#ifndef _ASM_MICROBLAZE_STRING_H
+#define _ASM_MICROBLAZE_STRING_H
+
+#ifdef __KERNEL__
+
+#ifdef CONFIG_OPT_LIB_FUNCTION
+#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
+
+#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 000000000..5afd6d997
--- /dev/null
+++ b/arch/microblaze/include/asm/switch_to.h
@@ -0,0 +1,21 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+
+#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 000000000..b5b6b91fa
--- /dev/null
+++ b/arch/microblaze/include/asm/syscall.h
@@ -0,0 +1,86 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#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_set_nr(struct task_struct *task,
+				  struct pt_regs *regs,
+				  int nr)
+{
+	regs->r12 = nr;
+}
+
+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 syscall_get_arguments(struct task_struct *task,
+					 struct pt_regs *regs,
+					 unsigned long *args)
+{
+	unsigned int i = 0;
+	unsigned int n = 6;
+
+	while (n--)
+		*args++ = microblaze_get_syscall_arg(regs, i++);
+}
+
+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(struct task_struct *task)
+{
+	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 000000000..0153f7c27
--- /dev/null
+++ b/arch/microblaze/include/asm/thread_info.h
@@ -0,0 +1,143 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+
+#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 __ASSEMBLER__
+# 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;
+};
+
+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*/
+
+	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,	\
+}
+
+/* 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 /* __ASSEMBLER__ */
+
+/*
+ * 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_NOTIFY_SIGNAL	5	/* signal notifications exist */
+#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_NOTIFY_SIGNAL	(1 << TIF_NOTIFY_SIGNAL)
+#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
+
+#endif /* __KERNEL__ */
+#endif /* _ASM_MICROBLAZE_THREAD_INFO_H */
diff --git a/arch/microblaze/include/asm/tlbflush.h b/arch/microblaze/include/asm/tlbflush.h
new file mode 100644
index 000000000..a31ae9d44
--- /dev/null
+++ b/arch/microblaze/include/asm/tlbflush.h
@@ -0,0 +1,53 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+
+#ifndef _ASM_MICROBLAZE_TLBFLUSH_H
+#define _ASM_MICROBLAZE_TLBFLUSH_H
+
+#include <linux/sched.h>
+#include <linux/threads.h>
+#include <asm/processor.h>	/* For TASK_SIZE */
+#include <asm/mmu.h>
+#include <asm/page.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_range(vmf, vma, addr, ptep, nr) do { } while (0)
+#define update_mmu_cache(vma, addr, pte) \
+	update_mmu_cache_range(NULL, vma, addr, ptep, 1)
+
+#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) { }
+
+#endif /* _ASM_MICROBLAZE_TLBFLUSH_H */
diff --git a/arch/microblaze/include/asm/uaccess.h b/arch/microblaze/include/asm/uaccess.h
new file mode 100644
index 000000000..afa0dd8d0
--- /dev/null
+++ b/arch/microblaze/include/asm/uaccess.h
@@ -0,0 +1,268 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+
+#ifndef _ASM_MICROBLAZE_UACCESS_H
+#define _ASM_MICROBLAZE_UACCESS_H
+
+#include <linux/kernel.h>
+
+#include <asm/mmu.h>
+#include <asm/page.h>
+#include <linux/pgtable.h>
+#include <asm/extable.h>
+#include <linux/string.h>
+#include <asm-generic/access_ok.h>
+
+# define __FIXUP_SECTION	".section .fixup,\"ax\"\n"
+# define __EX_TABLE_SECTION	".section __ex_table,\"a\"\n"
+
+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(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) ({				\
+	const typeof(*(ptr)) __user *__gu_ptr = (ptr);	\
+	access_ok(__gu_ptr, sizeof(*__gu_ptr)) ?	\
+		__get_user(x, __gu_ptr) : -EFAULT;	\
+})
+
+#define __get_user(x, ptr)						\
+({									\
+	long __gu_err;							\
+	switch (sizeof(*(ptr))) {					\
+	case 1:								\
+		__get_user_asm("lbu", (ptr), x, __gu_err);		\
+		break;							\
+	case 2:								\
+		__get_user_asm("lhu", (ptr), x, __gu_err);		\
+		break;							\
+	case 4:								\
+		__get_user_asm("lw", (ptr), x, __gu_err);		\
+		break;							\
+	case 8: {							\
+		__u64 __x = 0;						\
+		__gu_err = raw_copy_from_user(&__x, ptr, 8) ?		\
+							-EFAULT : 0;	\
+		(x) = (typeof(x))(typeof((x) - (x)))__x;		\
+		break;							\
+	}								\
+	default:							\
+		/* __gu_val = 0; __gu_err = -EINVAL;*/ __gu_err = __user_bad();\
+	}								\
+	__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(__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;							\
+})
+
+static inline unsigned long
+raw_copy_from_user(void *to, const void __user *from, unsigned long n)
+{
+	return __copy_tofrom_user((__force void __user *)to, from, n);
+}
+
+static inline unsigned long
+raw_copy_to_user(void __user *to, const void *from, unsigned long n)
+{
+	return __copy_tofrom_user(to, (__force const void __user *)from, n);
+}
+#define INLINE_COPY_USER
+
+/*
+ * Copy a null terminated string from userspace.
+ */
+__must_check long strncpy_from_user(char *dst, const char __user *src,
+				    long count);
+
+/*
+ * Return the size of a string (including the ending 0)
+ *
+ * Return 0 on exception, a value greater than N if too long
+ */
+__must_check long strnlen_user(const char __user *sstr, long len);
+
+#endif /* _ASM_MICROBLAZE_UACCESS_H */
diff --git a/arch/microblaze/include/asm/unistd.h b/arch/microblaze/include/asm/unistd.h
new file mode 100644
index 000000000..fedda9908
--- /dev/null
+++ b/arch/microblaze/include/asm/unistd.h
@@ -0,0 +1,38 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2007-2008 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2006 Atmark Techno, Inc.
+ */
+#ifndef _ASM_MICROBLAZE_UNISTD_H
+#define _ASM_MICROBLAZE_UNISTD_H
+
+#include <uapi/asm/unistd.h>
+
+#ifndef __ASSEMBLER__
+
+/* #define __ARCH_WANT_OLD_READDIR */
+/* #define __ARCH_WANT_OLD_STAT */
+#define __ARCH_WANT_NEW_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_TIME32
+#define __ARCH_WANT_SYS_UTIME32
+#define __ARCH_WANT_SYS_WAITPID
+#define __ARCH_WANT_SYS_SOCKETCALL
+#define __ARCH_WANT_SYS_FADVISE64
+#define __ARCH_WANT_SYS_GETPGRP
+#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 /* __ASSEMBLER__ */
+
+#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 000000000..3db81777a
--- /dev/null
+++ b/arch/microblaze/include/asm/unwind.h
@@ -0,0 +1,27 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Backtrace support for Microblaze
+ *
+ * Copyright (C) 2010  Digital Design Corporation
+ */
+
+#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,
+		       const char *loglvl);
+
+#endif	/* __MICROBLAZE_UNWIND_H */
+
diff --git a/arch/microblaze/include/asm/vmalloc.h b/arch/microblaze/include/asm/vmalloc.h
new file mode 100644
index 000000000..04013a42b
--- /dev/null
+++ b/arch/microblaze/include/asm/vmalloc.h
@@ -0,0 +1,4 @@
+#ifndef _ASM_MICROBLAZE_VMALLOC_H
+#define _ASM_MICROBLAZE_VMALLOC_H
+
+#endif /* _ASM_MICROBLAZE_VMALLOC_H */
diff --git a/arch/microblaze/include/asm/xilinx_mb_manager.h b/arch/microblaze/include/asm/xilinx_mb_manager.h
new file mode 100644
index 000000000..121a32248
--- /dev/null
+++ b/arch/microblaze/include/asm/xilinx_mb_manager.h
@@ -0,0 +1,29 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2022 Xilinx, Inc.
+ */
+#ifndef _XILINX_MB_MANAGER_H
+#define _XILINX_MB_MANAGER_H
+
+# ifndef __ASSEMBLER__
+
+#include <linux/of_address.h>
+
+/*
+ * When the break vector gets asserted because of error injection, the break
+ * signal must be blocked before exiting from the break handler, Below api
+ * updates the manager address and control register and error counter callback
+ * arguments, which will be used by the break handler to block the break and
+ * call the callback function.
+ */
+void xmb_manager_register(uintptr_t phys_baseaddr, u32 cr_val,
+			  void (*callback)(void *data),
+			  void *priv, void (*reset_callback)(void *data));
+asmlinkage void xmb_inject_err(void);
+
+# endif /* __ASSEMBLER__ */
+
+/* Error injection offset */
+#define XMB_INJECT_ERR_OFFSET	0x200
+
+#endif /* _XILINX_MB_MANAGER_H */
diff --git a/arch/microblaze/include/uapi/asm/Kbuild b/arch/microblaze/include/uapi/asm/Kbuild
new file mode 100644
index 000000000..da95c5a03
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/Kbuild
@@ -0,0 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0
+generated-y += unistd_32.h
+generic-y += ucontext.h
diff --git a/arch/microblaze/include/uapi/asm/auxvec.h b/arch/microblaze/include/uapi/asm/auxvec.h
new file mode 100644
index 000000000..93dd07bd0
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/auxvec.h
@@ -0,0 +1,2 @@
+/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
+
diff --git a/arch/microblaze/include/uapi/asm/byteorder.h b/arch/microblaze/include/uapi/asm/byteorder.h
new file mode 100644
index 000000000..763660169
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/byteorder.h
@@ -0,0 +1,11 @@
+/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
+#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 000000000..6b656de29
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/elf.h
@@ -0,0 +1,123 @@
+/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
+/*
+ * 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/posix_types.h b/arch/microblaze/include/uapi/asm/posix_types.h
new file mode 100644
index 000000000..f3249da69
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/posix_types.h
@@ -0,0 +1,10 @@
+/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
+#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 000000000..8039957a1
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/ptrace.h
@@ -0,0 +1,73 @@
+/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
+/*
+ * 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 __ASSEMBLER__
+
+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 /* __ASSEMBLER__ */
+
+#endif /* _UAPI_ASM_MICROBLAZE_PTRACE_H */
diff --git a/arch/microblaze/include/uapi/asm/setup.h b/arch/microblaze/include/uapi/asm/setup.h
new file mode 100644
index 000000000..16c56807f
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/setup.h
@@ -0,0 +1,17 @@
+/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
+/*
+ * 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
+
+#endif /* _UAPI_ASM_MICROBLAZE_SETUP_H */
diff --git a/arch/microblaze/include/uapi/asm/sigcontext.h b/arch/microblaze/include/uapi/asm/sigcontext.h
new file mode 100644
index 000000000..47eb2e211
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/sigcontext.h
@@ -0,0 +1,21 @@
+/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
+/*
+ * 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/unistd.h b/arch/microblaze/include/uapi/asm/unistd.h
new file mode 100644
index 000000000..3f2d7cb68
--- /dev/null
+++ b/arch/microblaze/include/uapi/asm/unistd.h
@@ -0,0 +1,16 @@
+/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
+/*
+ * 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
+
+#include <asm/unistd_32.h>
+
+#endif /* _UAPI_ASM_MICROBLAZE_UNISTD_H */
diff --git a/arch/microblaze/kernel/.gitignore b/arch/microblaze/kernel/.gitignore
new file mode 100644
index 000000000..bbb90f92d
--- /dev/null
+++ b/arch/microblaze/kernel/.gitignore
@@ -0,0 +1,2 @@
+# SPDX-License-Identifier: GPL-2.0-only
+vmlinux.lds
diff --git a/arch/microblaze/kernel/Makefile b/arch/microblaze/kernel/Makefile
new file mode 100644
index 000000000..241e466e7
--- /dev/null
+++ b/arch/microblaze/kernel/Makefile
@@ -0,0 +1,29 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# 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_ftrace.o = -pg
+CFLAGS_REMOVE_process.o = -pg
+endif
+
+always-$(KBUILD_BUILTIN) := vmlinux.lds
+
+obj-y += head.o dma.o exceptions.o \
+	hw_exception_handler.o irq.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-$(CONFIG_MODULES)		+= microblaze_ksyms.o module.o
+obj-y				+= misc.o
+obj-$(CONFIG_STACKTRACE)	+= stacktrace.o
+obj-$(CONFIG_FUNCTION_TRACER)	+= ftrace.o mcount.o
+obj-$(CONFIG_KGDB)		+= kgdb.o
+
+obj-y	+= entry.o
diff --git a/arch/microblaze/kernel/asm-offsets.c b/arch/microblaze/kernel/asm-offsets.c
new file mode 100644
index 000000000..b4b67d58e
--- /dev/null
+++ b/arch/microblaze/kernel/asm-offsets.c
@@ -0,0 +1,132 @@
+/*
+ * 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.
+ */
+#define COMPILE_OFFSETS
+
+#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));
+	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();
+
+	/* struct thread_info */
+	DEFINE(TI_TASK, offsetof(struct thread_info, task));
+	DEFINE(TI_FLAGS, offsetof(struct thread_info, flags));
+	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();
+
+	/* struct cpuinfo */
+	DEFINE(CI_DCS, offsetof(struct cpuinfo, dcache_size));
+	DEFINE(CI_DCL, offsetof(struct cpuinfo, dcache_line_length));
+	DEFINE(CI_ICS, offsetof(struct cpuinfo, icache_size));
+	DEFINE(CI_ICL, offsetof(struct cpuinfo, icache_line_length));
+	BLANK();
+
+	return 0;
+}
diff --git a/arch/microblaze/kernel/cpu/Makefile b/arch/microblaze/kernel/cpu/Makefile
new file mode 100644
index 000000000..059afc75a
--- /dev/null
+++ b/arch/microblaze/kernel/cpu/Makefile
@@ -0,0 +1,13 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# 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 000000000..dcba53803
--- /dev/null
+++ b/arch/microblaze/kernel/cpu/cache.c
@@ -0,0 +1,656 @@
+/*
+ * 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);				\
+	if (start <  UINT_MAX - cache_size)				\
+		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 000000000..c7ee51b09
--- /dev/null
+++ b/arch/microblaze/kernel/cpu/cpuinfo-pvr-full.c
@@ -0,0 +1,110 @@
+/*
+ * 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); }
+
+#define err_printk(x) \
+	pr_err("ERROR: Microblaze " x "-different for PVR and DTS\n");
+
+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 000000000..03da36dc6
--- /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) \
+	pr_err("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 000000000..cd9b44507
--- /dev/null
+++ b/arch/microblaze/kernel/cpu/cpuinfo.c
@@ -0,0 +1,142 @@
+/*
+ * 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},
+	{"9.6", 0x23},
+	{"10.0", 0x24},
+	{"11.0", 0x25},
+	{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},
+	{"UltraScale+ Zynq", 0x15},
+	{"UltraScale+ Virtex", 0x16},
+	{"UltraScale+ Kintex", 0x17},
+	{"Spartan7", 0x18},
+	{NULL, 0},
+};
+
+struct cpuinfo cpuinfo;
+static struct device_node *cpu;
+
+void __init setup_cpuinfo(void)
+{
+	cpu = of_get_cpu_node(0, NULL);
+	if (!cpu)
+		pr_err("You don't have cpu or are missing cpu reg property!!!\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__);
+
+	of_node_put(cpu);
+}
+
+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 000000000..37cb28982
--- /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",
+		   str_yes_no(cpuinfo.use_instr & PVR0_USE_BARREL_MASK),
+		   str_yes_no(cpuinfo.use_instr & PVR2_USE_MSR_INSTR),
+		   str_yes_no(cpuinfo.use_instr & PVR2_USE_PCMP_INSTR),
+		   str_yes_no(cpuinfo.use_instr & PVR0_USE_DIV_MASK));
+
+	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",
+		   str_yes_no(cpuinfo.hw_debug));
+
+	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 000000000..f139052a3
--- /dev/null
+++ b/arch/microblaze/kernel/cpu/pvr.c
@@ -0,0 +1,81 @@
+/*
+ * 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>
+#include <linux/irqflags.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 000000000..04d091ade
--- /dev/null
+++ b/arch/microblaze/kernel/dma.c
@@ -0,0 +1,43 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * 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-map-ops.h>
+#include <linux/gfp.h>
+#include <linux/export.h>
+#include <linux/bug.h>
+#include <asm/cacheflush.h>
+
+static void __dma_sync(phys_addr_t 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();
+	}
+}
+
+void arch_sync_dma_for_device(phys_addr_t paddr, size_t size,
+		enum dma_data_direction dir)
+{
+	__dma_sync(paddr, size, dir);
+}
+
+void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size,
+		enum dma_data_direction dir)
+{
+	__dma_sync(paddr, size, dir);
+}
diff --git a/arch/microblaze/kernel/entry.S b/arch/microblaze/kernel/entry.S
new file mode 100644
index 000000000..582d7256d
--- /dev/null
+++ b/arch/microblaze/kernel/entry.S
@@ -0,0 +1,1311 @@
+/*
+ * 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 <asm/xilinx_mb_manager.h>
+
+#include <linux/errno.h>
+#include <asm/signal.h>
+#include <asm/mmu.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_GP \
+	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 RESTORE_REGS \
+	lwi	r11, r1, PT_MSR;					\
+	mts	rmsr , r11;						\
+	RESTORE_REGS_GP
+
+#define RESTORE_REGS_RTBD \
+	lwi	r11, r1, PT_MSR;					\
+	andni	r11, r11, MSR_EIP;          /* clear EIP */             \
+	ori	r11, r11, MSR_EE | MSR_BIP; /* set EE and BIP */        \
+	mts	rmsr , r11;						\
+	RESTORE_REGS_GP
+
+#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
+
+.extern cpuinfo
+
+C_ENTRY(mb_flush_dcache):
+	addik	r1, r1, -PT_SIZE
+	SAVE_REGS
+
+	addik	r3, r0, cpuinfo
+	lwi	r7, r3, CI_DCS
+	lwi	r8, r3, CI_DCL
+	sub	r9, r7, r8
+1:
+	wdc.flush r9, r0
+	bgtid	r9, 1b
+	addk	r9, r9, r8
+
+	RESTORE_REGS
+	addik	r1, r1, PT_SIZE
+	rtsd	r15, 8
+	nop
+
+C_ENTRY(mb_invalidate_icache):
+	addik	r1, r1, -PT_SIZE
+	SAVE_REGS
+
+	addik	r3, r0, cpuinfo
+	lwi	r7, r3, CI_ICS
+	lwi	r8, r3, CI_ICL
+	sub	r9, r7, r8
+1:
+	wic 	r9, r0
+	bgtid	r9, 1b
+	addk	r9, r9, r8
+
+	RESTORE_REGS
+	addik	r1, r1, PT_SIZE
+	rtsd	r15, 8
+	nop
+
+/*
+ * 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_RTBD;
+	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_RTBD;
+	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 like the child is in the middle of a
+   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_RTBD;
+	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_RTBD;
+	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_PREEMPTION
+	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 */
+
+	/* interrupts are off that's why I am calling preempt_chedule_irq */
+	bralid	r15, preempt_schedule_irq
+	nop
+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
+
+#ifdef CONFIG_MB_MANAGER
+
+#define	PT_PID		PT_SIZE
+#define	PT_TLBI		PT_SIZE + 4
+#define	PT_ZPR		PT_SIZE	+ 8
+#define	PT_TLBL0	PT_SIZE + 12
+#define	PT_TLBH0	PT_SIZE + 16
+
+C_ENTRY(_xtmr_manager_reset):
+	lwi	r1, r0, xmb_manager_stackpointer
+
+	/* Restore MSR */
+	lwi	r2, r1, PT_MSR
+	mts	rmsr, r2
+	bri	4
+
+	/* restore Special purpose registers */
+	lwi	r2, r1, PT_PID
+	mts	rpid, r2
+
+	lwi	r2, r1, PT_TLBI
+	mts	rtlbx, r2
+
+	lwi	r2, r1, PT_ZPR
+	mts	rzpr, r2
+
+#if CONFIG_XILINX_MICROBLAZE0_USE_FPU
+	lwi	r2, r1, PT_FSR
+	mts	rfsr, r2
+#endif
+
+	/* restore all the tlb's */
+	addik	r3, r0, TOPHYS(tlb_skip)
+	addik	r6, r0, PT_TLBL0
+	addik	r7, r0, PT_TLBH0
+restore_tlb:
+	add	r6, r6, r1
+	add	r7, r7, r1
+	lwi	r2, r6, 0
+	mts 	rtlblo, r2
+	lwi	r2, r7, 0
+	mts	rtlbhi, r2
+	addik	r6, r6, 4
+	addik	r7, r7, 4
+	bgtid	r3, restore_tlb
+	addik	r3, r3, -1
+
+	lwi  	r5, r0, TOPHYS(xmb_manager_dev)
+	lwi	r8, r0, TOPHYS(xmb_manager_reset_callback)
+	set_vms
+	/* return from reset need -8 to adjust for rtsd r15, 8 */
+	addik   r15, r0, ret_from_reset - 8
+	rtbd	r8, 0
+	nop
+
+ret_from_reset:
+	set_bip /* Ints masked for state restore */
+	VM_OFF
+	/* MS: Restore all regs */
+	RESTORE_REGS
+	lwi	r14, r1, PT_R14
+	lwi	r16, r1, PT_PC
+	addik	r1, r1, PT_SIZE + 36
+	rtbd	r16, 0
+	nop
+
+/*
+ * Break handler for MB Manager. Enter to _xmb_manager_break by
+ * injecting fault in one of the TMR Microblaze core.
+ * FIXME: This break handler supports getting
+ * called from kernel space only.
+ */
+C_ENTRY(_xmb_manager_break):
+	/*
+	 * Reserve memory in the stack for context store/restore
+	 * (which includes memory for storing tlbs (max two tlbs))
+	 */
+	addik	r1, r1, -PT_SIZE - 36
+	swi	r1, r0, xmb_manager_stackpointer
+	SAVE_REGS
+	swi	r14, r1, PT_R14	/* rewrite saved R14 value */
+	swi	r16, r1, PT_PC; /* PC and r16 are the same */
+
+	lwi	r6, r0, TOPHYS(xmb_manager_baseaddr)
+	lwi	r7, r0, TOPHYS(xmb_manager_crval)
+	/*
+	 * When the break vector gets asserted because of error injection,
+	 * the break signal must be blocked before exiting from the
+	 * break handler, below code configures the tmr manager
+	 * control register to block break signal.
+	 */
+	swi	r7, r6, 0
+
+	/* Save the special purpose registers  */
+	mfs	r2, rpid
+	swi	r2, r1, PT_PID
+
+	mfs	r2, rtlbx
+	swi	r2, r1, PT_TLBI
+
+	mfs	r2, rzpr
+	swi	r2, r1, PT_ZPR
+
+#if CONFIG_XILINX_MICROBLAZE0_USE_FPU
+	mfs	r2, rfsr
+	swi	r2, r1, PT_FSR
+#endif
+	mfs	r2, rmsr
+	swi	r2, r1, PT_MSR
+
+	/* Save all the tlb's */
+	addik	r3, r0, TOPHYS(tlb_skip)
+	addik	r6, r0, PT_TLBL0
+	addik	r7, r0, PT_TLBH0
+save_tlb:
+	add	r6, r6, r1
+	add	r7, r7, r1
+	mfs	r2, rtlblo
+	swi	r2, r6, 0
+	mfs	r2, rtlbhi
+	swi	r2, r7, 0
+	addik	r6, r6, 4
+	addik	r7, r7, 4
+	bgtid	r3, save_tlb
+	addik	r3, r3, -1
+
+	lwi  	r5, r0, TOPHYS(xmb_manager_dev)
+	lwi	r8, r0, TOPHYS(xmb_manager_callback)
+	/* return from break need -8 to adjust for rtsd r15, 8 */
+	addik   r15, r0, ret_from_break - 8
+	rtbd	r8, 0
+	nop
+
+ret_from_break:
+	/* flush the d-cache */
+	bralid	r15, mb_flush_dcache
+	nop
+
+	/*
+	 * To make sure microblaze i-cache is in a proper state
+	 * invalidate the i-cache.
+	 */
+	bralid	r15, mb_invalidate_icache
+	nop
+
+	set_bip; /* Ints masked for state restore */
+	VM_OFF;
+	mbar	1
+	mbar	2
+	bri	4
+	suspend
+	nop
+#endif
+
+/*
+ * 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_RTBD
+	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_RTBD
+	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
+
+#ifdef CONFIG_MB_MANAGER
+.global xmb_inject_err
+.section .text
+.align 2
+.ent xmb_inject_err
+.type xmb_inject_err, @function
+xmb_inject_err:
+	addik	r1, r1, -PT_SIZE
+	SAVE_REGS
+
+	/* Switch to real mode */
+	VM_OFF;
+	set_bip;
+	mbar	1
+	mbar	2
+	bralid	r15, XMB_INJECT_ERR_OFFSET
+	nop;
+
+	/* enable virtual mode */
+	set_vms;
+	/* barrier for instructions and data accesses */
+	mbar	1
+	mbar	2
+	/*
+	 * Enable Interrupts, Virtual Protected Mode, equalize
+	 * initial state for all possible entries.
+	 */
+	rtbd    r0, 1f
+	nop;
+1:
+	RESTORE_REGS
+	addik	r1, r1, PT_SIZE
+	rtsd	r15, 8;
+	nop;
+.end xmb_inject_err
+
+.section .data
+.global xmb_manager_dev
+.global xmb_manager_baseaddr
+.global xmb_manager_crval
+.global xmb_manager_callback
+.global xmb_manager_reset_callback
+.global xmb_manager_stackpointer
+.align 4
+xmb_manager_dev:
+	.long 0
+xmb_manager_baseaddr:
+	.long 0
+xmb_manager_crval:
+	.long 0
+xmb_manager_callback:
+	.long 0
+xmb_manager_reset_callback:
+	.long 0
+xmb_manager_stackpointer:
+	.long 0
+
+/*
+ * When the break vector gets asserted because of error injection,
+ * the break signal must be blocked before exiting from the
+ * break handler, Below api updates the manager address and
+ * control register and error count callback arguments,
+ * which will be used by the break handler to block the
+ * break and call the callback function.
+ */
+.global xmb_manager_register
+.section .text
+.align 2
+.ent xmb_manager_register
+.type xmb_manager_register, @function
+xmb_manager_register:
+	swi	r5, r0, xmb_manager_baseaddr
+	swi	r6, r0, xmb_manager_crval
+	swi	r7, r0, xmb_manager_callback
+	swi	r8, r0, xmb_manager_dev
+	swi	r9, r0, xmb_manager_reset_callback
+
+	rtsd	r15, 8;
+	nop;
+.end xmb_manager_register
+#endif
+
+ENTRY(_reset)
+	VM_OFF
+	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 && !defined(CONFIG_MB_MANAGER)
+	.org	0x0
+	brai	CONFIG_MANUAL_RESET_VECTOR
+#elif defined(CONFIG_MB_MANAGER)
+	.org	0x0
+	brai	TOPHYS(_xtmr_manager_reset);
+#endif
+	.org	0x8
+	brai	TOPHYS(_user_exception); /* syscall handler */
+	.org	0x10
+	brai	TOPHYS(_interrupt);	/* Interrupt handler */
+#ifdef CONFIG_MB_MANAGER
+	.org	0x18
+	brai	TOPHYS(_xmb_manager_break);	/* microblaze manager break handler */
+#else
+	.org	0x18
+	brai	TOPHYS(_debug_exception);	/* debug trap handler */
+#endif
+	.org	0x20
+	brai	TOPHYS(_hw_exception_handler);	/* HW exception handler */
+
+#ifdef CONFIG_MB_MANAGER
+	/*
+	 * For TMR Inject API which injects the error should
+	 * be executed from LMB.
+	 * TMR Inject is programmed with address of 0x200 so that
+	 * when program counter matches with this address error will
+	 * be injected. 0x200 is expected to be next available bram
+	 * offset, hence used for this api.
+	 */
+	.org	XMB_INJECT_ERR_OFFSET
+xmb_inject_error:
+	nop
+	rtsd	r15, 8
+	nop
+#endif
+
+.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 000000000..fd153d5fa
--- /dev/null
+++ b/arch/microblaze/kernel/exceptions.c
@@ -0,0 +1,144 @@
+/*
+ * 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/sched/debug.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);
+	/* make_task_dead() should take care of panic'ing from an interrupt
+	 * context so we don't handle it here
+	 */
+	make_task_dead(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)
+{
+	if (kernel_mode(regs))
+		die("Exception in kernel mode", regs, signr);
+
+	force_sig_fault(signr, code, (void __user *)addr);
+}
+
+asmlinkage void full_exception(struct pt_regs *regs, unsigned int type,
+							int fsr, int addr)
+{
+	addr = regs->pc;
+
+#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;
+	case MICROBLAZE_PRIVILEGED_EXCEPTION:
+		pr_debug("Privileged exception\n");
+		_exception(SIGILL, regs, ILL_PRVOPC, addr);
+		break;
+	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 000000000..188749d62
--- /dev/null
+++ b/arch/microblaze/kernel/ftrace.c
@@ -0,0 +1,217 @@
+/*
+ * 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;
+	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;
+	}
+
+	if (function_graph_enter(old, self_addr, 0, NULL))
+		*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 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 000000000..9bd3e513c
--- /dev/null
+++ b/arch/microblaze/kernel/head.S
@@ -0,0 +1,372 @@
+/*
+ * 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 */
+
+#include <asm/setup.h> /* COMMAND_LINE_SIZE */
+#include <asm/mmu.h>
+#include <asm/processor.h>
+
+.section .data
+/* The MMU requires a page aligned page directory. */
+.align 12
+.global swapper_pg_dir
+swapper_pg_dir:
+	.space	PAGE_SIZE
+
+.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, (0x10000 - 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:
+
+#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+r11 - r5 contain pointer to command line */
+	lbu	r2, r5, r11
+	beqid	r2, skip		/* Skip if no data */
+	sb	r2, r4, r11		/* addr[r4+r11]= 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 + r11 */
+	sw	r7, r4, r11		/* addr[r4 + r11] = 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:
+
+	/* 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
+
+	addik	r11, r0, machine_early_init
+	brald	r15, r11
+	nop
+
+	/*
+	 * 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
diff --git a/arch/microblaze/kernel/hw_exception_handler.S b/arch/microblaze/kernel/hw_exception_handler.S
new file mode 100644
index 000000000..07ea23965
--- /dev/null
+++ b/arch/microblaze/kernel/hw_exception_handler.S
@@ -0,0 +1,1094 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * 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.
+ */
+
+/*
+ * 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 <linux/pgtable.h>
+
+#include <asm/mmu.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
+
+	#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;
+
+#define LWREG_NOP			\
+	bri	ex_handler_unhandled;	\
+	nop;
+
+#define SWREG_NOP			\
+	bri	ex_handler_unhandled;	\
+	nop;
+
+/* 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;
+
+	#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
+
+
+.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 */
+.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)
+
+.global _hw_exception_handler
+.section .text
+.align 4
+.ent _hw_exception_handler
+_hw_exception_handler:
+	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);
+	swi	r3, r1, PT_R3
+	swi	r4, r1, PT_R4
+	swi	r5, r1, PT_R5
+	swi	r6, r1, PT_R6
+
+	swi	r11, r1, PT_R11
+	swi	r31, r1, PT_R31
+	lwi	r31, r0, TOPHYS(PER_CPU(CURRENT_SAVE)) /* get saved current */
+
+	mfs	r5, rmsr;
+	nop
+	swi	r5, r1, 0;
+	mfs	r4, resr
+	nop
+	mfs	r3, rear;
+	nop
+
+	andi	r5, r4, 0x1F;		/* Extract ESR[EXC] */
+
+	/* 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
+
+/* 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
+	 */
+	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
+
+	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:
+	RESTORE_STATE;
+	rted	r17, 0
+	nop
+
+	/* 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
+		ori	r3, r3, TLB_VALID | TLB_PAGESZ(PAGESZ_4K)
+		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
+
+.end _hw_exception_handler
+
+/* 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
+
+.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);
+lw_r31: 	R3_TO_LWREG_V	(31);
+
+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);
+sw_r31:		SWREG_TO_R3_V	(31);
+
+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);
+
+/* 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/irq.c b/arch/microblaze/kernel/irq.c
new file mode 100644
index 000000000..1f8cb4c4f
--- /dev/null
+++ b/arch/microblaze/kernel/irq.c
@@ -0,0 +1,39 @@
+/*
+ * 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>
+
+void __irq_entry do_IRQ(struct pt_regs *regs)
+{
+	struct pt_regs *old_regs = set_irq_regs(regs);
+	trace_hardirqs_off();
+
+	irq_enter();
+	handle_arch_irq(regs);
+	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 000000000..df4b9d011
--- /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 unchangeable */
+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
+ */
+const 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 000000000..fed9da5de
--- /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 000000000..ad7596d7b
--- /dev/null
+++ b/arch/microblaze/kernel/microblaze_ksyms.c
@@ -0,0 +1,56 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2008-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2008-2009 PetaLogix
+ */
+
+#include <linux/export.h>
+#include <linux/string.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>
+#include <asm/xilinx_mb_manager.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);
+
+#ifdef CONFIG_OPT_LIB_ASM
+EXPORT_SYMBOL(memcpy);
+EXPORT_SYMBOL(memmove);
+#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);
+
+#ifdef CONFIG_MB_MANAGER
+extern void xmb_manager_register(uintptr_t phys_baseaddr, u32 cr_val,
+				 void (*callback)(void *data),
+				 void *priv, void (*reset_callback)(void *data));
+EXPORT_SYMBOL(xmb_manager_register);
+extern asmlinkage void xmb_inject_err(void);
+EXPORT_SYMBOL(xmb_inject_err);
+#endif
diff --git a/arch/microblaze/kernel/misc.S b/arch/microblaze/kernel/misc.S
new file mode 100644
index 000000000..1228a09d8
--- /dev/null
+++ b/arch/microblaze/kernel/misc.S
@@ -0,0 +1,66 @@
+/*
+ * 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
+	mbar	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
+	mbar	1 /* sync */
+_tlbie_1:
+	rtsd	r15, 8
+	nop
+
+	.size  _tlbie, . - _tlbie
diff --git a/arch/microblaze/kernel/module.c b/arch/microblaze/kernel/module.c
new file mode 100644
index 000000000..e5db3a57b
--- /dev/null
+++ b/arch/microblaze/kernel/module.c
@@ -0,0 +1,96 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2007-2009 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2007-2009 PetaLogix
+ */
+
+#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 <linux/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;
+
+	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:
+			*location = value;
+			break;
+
+		case R_MICROBLAZE_64:
+			location[0] = (location[0] & 0xFFFF0000) |
+					(value >> 16);
+			location[1] = (location[1] & 0xFFFF0000) |
+					(value & 0xFFFF);
+			break;
+
+		case R_MICROBLAZE_64_PCREL:
+			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/process.c b/arch/microblaze/kernel/process.c
new file mode 100644
index 000000000..6cbf642d7
--- /dev/null
+++ b/arch/microblaze/kernel/process.c
@@ -0,0 +1,143 @@
+/*
+ * 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/sched/debug.h>
+#include <linux/sched/task.h>
+#include <linux/sched/task_stack.h>
+#include <linux/pm.h>
+#include <linux/tick.h>
+#include <linux/bitops.h>
+#include <linux/ptrace.h>
+#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(struct task_struct *p, const struct kernel_clone_args *args)
+{
+	u64 clone_flags = args->flags;
+	unsigned long usp = args->stack;
+	unsigned long tls = args->tls;
+	struct pt_regs *childregs = task_pt_regs(p);
+	struct thread_info *ti = task_thread_info(p);
+
+	if (unlikely(args->fn)) {
+		/* 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)args->fn;
+		ti->cpu_context.r19 = (unsigned long)args->fn_arg;
+		childregs->pt_mode = 1;
+		local_save_flags(childregs->msr);
+		ti->cpu_context.msr = childregs->msr & ~MSR_IE;
+		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;
+	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;
+	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 = tls;
+
+	return 0;
+}
+
+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;
+	regs->msr |= MSR_UMS;
+	regs->msr &= ~MSR_VM;
+}
+
+#include <linux/elfcore.h>
+/*
+ * Set up a thread for executing a new program
+ */
+int elf_core_copy_task_fpregs(struct task_struct *t, elf_fpregset_t *fpu)
+{
+	return 0; /* MicroBlaze has no separate FPU registers */
+}
+
+void arch_cpu_idle(void)
+{
+}
diff --git a/arch/microblaze/kernel/prom.c b/arch/microblaze/kernel/prom.c
new file mode 100644
index 000000000..76ac4cfdf
--- /dev/null
+++ b/arch/microblaze/kernel/prom.c
@@ -0,0 +1,30 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * 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
+ */
+
+#include <linux/kernel.h>
+#include <linux/string.h>
+#include <linux/memblock.h>
+#include <linux/of_fdt.h>
+
+void __init early_init_devtree(void *params)
+{
+	pr_debug(" -> early_init_devtree(%p)\n", params);
+
+	early_init_dt_scan(params, __pa(params));
+	if (!strlen(boot_command_line))
+		strscpy(boot_command_line, cmd_line, COMMAND_LINE_SIZE);
+
+	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 000000000..236264e93
--- /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/sched/task_stack.h>
+#include <linux/ptrace.h>
+#include <linux/signal.h>
+#include <linux/elf.h>
+#include <linux/audit.h>
+#include <linux/seccomp.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) &&
+	    !ptrace_report_syscall_permit_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))
+		ptrace_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 000000000..2f66c7963
--- /dev/null
+++ b/arch/microblaze/kernel/reset.c
@@ -0,0 +1,42 @@
+/*
+ * 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/delay.h>
+#include <linux/reboot.h>
+
+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)
+		;
+}
+
+void machine_restart(char *cmd)
+{
+	do_kernel_restart(cmd);
+	/* Give the restart hook 1 s to take us down */
+	mdelay(1000);
+	pr_emerg("Reboot failed -- System halted\n");
+	while (1);
+}
diff --git a/arch/microblaze/kernel/setup.c b/arch/microblaze/kernel/setup.c
new file mode 100644
index 000000000..f417333ec
--- /dev/null
+++ b/arch/microblaze/kernel/setup.c
@@ -0,0 +1,199 @@
+/*
+ * 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/of_clk.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 <linux/pgtable.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.h>
+#include <linux/dma-mapping.h>
+#include <asm/cacheflush.h>
+#include <asm/entry.h>
+#include <asm/cpuinfo.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 */
+
+/*
+ * 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] __section(".data");
+
+void __init setup_arch(char **cmdline_p)
+{
+	*cmdline_p = boot_command_line;
+
+	setup_memory();
+
+	console_verbose();
+
+	unflatten_device_tree();
+
+	setup_cpuinfo();
+
+	microblaze_cache_init();
+
+	xilinx_pci_init();
+}
+
+#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);
+
+/* initialize device tree for usage in early_printk */
+	early_init_devtree(_fdt_start);
+
+	/* 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();
+	timer_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 0;
+}
+arch_initcall(microblaze_debugfs_init);
+
+static int __init debugfs_tlb(void)
+{
+	debugfs_create_u32("tlb_skip", S_IRUGO, of_debugfs_root, &tlb_skip);
+	return 0;
+}
+device_initcall(debugfs_tlb);
+#endif
diff --git a/arch/microblaze/kernel/signal.c b/arch/microblaze/kernel/signal.c
new file mode 100644
index 000000000..c78a0ff48
--- /dev/null
+++ b/arch/microblaze/kernel/signal.c
@@ -0,0 +1,316 @@
+/*
+ * 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 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/resume_user_mode.h>
+#include <asm/entry.h>
+#include <asm/ucontext.h>
+#include <linux/uaccess.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(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);
+	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;
+	pmd_t *pmdp;
+	pte_t *ptep;
+
+	frame = get_sigframe(ksig, regs, sizeof(*frame));
+
+	if (!access_ok(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);
+	pmdp = pmd_off(current->mm, address);
+
+	preempt_disable();
+	ptep = pte_offset_map(pmdp, address);
+	if (ptep && 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);
+	}
+	if (ptep)
+		pte_unmap(ptep);
+	preempt_enable();
+	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, read_thread_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) ||
+	    test_thread_flag(TIF_NOTIFY_SIGNAL))
+		do_signal(regs, in_syscall);
+
+	if (test_thread_flag(TIF_NOTIFY_RESUME))
+		resume_user_mode_work(regs);
+}
diff --git a/arch/microblaze/kernel/stacktrace.c b/arch/microblaze/kernel/stacktrace.c
new file mode 100644
index 000000000..b266c4d6e
--- /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 000000000..0850b099f
--- /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,
+		unsigned long, pgoff)
+{
+	if (pgoff & ~PAGE_MASK)
+		return -EINVAL;
+
+	return ksys_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 ksys_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 000000000..3bc60a2b1
--- /dev/null
+++ b/arch/microblaze/kernel/syscall_table.S
@@ -0,0 +1,5 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#define __SYSCALL(nr, entry) .long entry
+ENTRY(sys_call_table)
+#include <asm/syscall_table.h>
diff --git a/arch/microblaze/kernel/syscalls/Makefile b/arch/microblaze/kernel/syscalls/Makefile
new file mode 100644
index 000000000..b265e4bc1
--- /dev/null
+++ b/arch/microblaze/kernel/syscalls/Makefile
@@ -0,0 +1,32 @@
+# SPDX-License-Identifier: GPL-2.0
+kapi := arch/$(SRCARCH)/include/generated/asm
+uapi := arch/$(SRCARCH)/include/generated/uapi/asm
+
+$(shell mkdir -p $(uapi) $(kapi))
+
+syscall := $(src)/syscall.tbl
+syshdr := $(srctree)/scripts/syscallhdr.sh
+systbl := $(srctree)/scripts/syscalltbl.sh
+
+quiet_cmd_syshdr = SYSHDR  $@
+      cmd_syshdr = $(CONFIG_SHELL) $(syshdr) --emit-nr $< $@
+
+quiet_cmd_systbl = SYSTBL  $@
+      cmd_systbl = $(CONFIG_SHELL) $(systbl) $< $@
+
+$(uapi)/unistd_32.h: $(syscall) $(syshdr) FORCE
+	$(call if_changed,syshdr)
+
+$(kapi)/syscall_table.h: $(syscall) $(systbl) FORCE
+	$(call if_changed,systbl)
+
+uapisyshdr-y		+= unistd_32.h
+kapisyshdr-y		+= syscall_table.h
+
+uapisyshdr-y	:= $(addprefix $(uapi)/, $(uapisyshdr-y))
+kapisyshdr-y	:= $(addprefix $(kapi)/, $(kapisyshdr-y))
+targets		+= $(addprefix ../../../../, $(uapisyshdr-y) $(kapisyshdr-y))
+
+PHONY += all
+all: $(uapisyshdr-y) $(kapisyshdr-y)
+	@:
diff --git a/arch/microblaze/kernel/syscalls/syscall.tbl b/arch/microblaze/kernel/syscalls/syscall.tbl
new file mode 100644
index 000000000..c55a5c96f
--- /dev/null
+++ b/arch/microblaze/kernel/syscalls/syscall.tbl
@@ -0,0 +1,480 @@
+# SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note
+#
+# system call numbers and entry vectors for microblaze
+#
+# The format is:
+# <number> <abi> <name> <entry point>
+#
+# The <abi> is always "common" for this file
+#
+0	common	restart_syscall			sys_restart_syscall
+1	common	exit				sys_exit
+2	common	fork				sys_fork
+3	common	read				sys_read
+4	common	write				sys_write
+5	common	open				sys_open
+6	common	close				sys_close
+7	common	waitpid				sys_waitpid
+8	common	creat				sys_creat
+9	common	link				sys_link
+10	common	unlink				sys_unlink
+11	common	execve				sys_execve
+12	common	chdir				sys_chdir
+13	common	time				sys_time32
+14	common	mknod				sys_mknod
+15	common	chmod				sys_chmod
+16	common	lchown				sys_lchown
+17	common	break				sys_ni_syscall
+18	common	oldstat				sys_ni_syscall
+19	common	lseek				sys_lseek
+20	common	getpid				sys_getpid
+21	common	mount				sys_mount
+22	common	umount				sys_oldumount
+23	common	setuid				sys_setuid
+24	common	getuid				sys_getuid
+25	common	stime				sys_stime32
+26	common	ptrace				sys_ptrace
+27	common	alarm				sys_alarm
+28	common	oldfstat			sys_ni_syscall
+29	common	pause				sys_pause
+30	common	utime				sys_utime32
+31	common	stty				sys_ni_syscall
+32	common	gtty				sys_ni_syscall
+33	common	access				sys_access
+34	common	nice				sys_nice
+35	common	ftime				sys_ni_syscall
+36	common	sync				sys_sync
+37	common	kill				sys_kill
+38	common	rename				sys_rename
+39	common	mkdir				sys_mkdir
+40	common	rmdir				sys_rmdir
+41	common	dup				sys_dup
+42	common	pipe				sys_pipe
+43	common	times				sys_times
+44	common	prof				sys_ni_syscall
+45	common	brk				sys_brk
+46	common	setgid				sys_setgid
+47	common	getgid				sys_getgid
+48	common	signal				sys_signal
+49	common	geteuid				sys_geteuid
+50	common	getegid				sys_getegid
+51	common	acct				sys_acct
+52	common	umount2				sys_umount
+53	common	lock				sys_ni_syscall
+54	common	ioctl				sys_ioctl
+55	common	fcntl				sys_fcntl
+56	common	mpx				sys_ni_syscall
+57	common	setpgid				sys_setpgid
+58	common	ulimit				sys_ni_syscall
+59	common	oldolduname			sys_ni_syscall
+60	common	umask				sys_umask
+61	common	chroot				sys_chroot
+62	common	ustat				sys_ustat
+63	common	dup2				sys_dup2
+64	common	getppid				sys_getppid
+65	common	getpgrp				sys_getpgrp
+66	common	setsid				sys_setsid
+67	common	sigaction			sys_ni_syscall
+68	common	sgetmask			sys_sgetmask
+69	common	ssetmask			sys_ssetmask
+70	common	setreuid			sys_setreuid
+71	common	setregid			sys_setregid
+72	common	sigsuspend			sys_ni_syscall
+73	common	sigpending			sys_sigpending
+74	common	sethostname			sys_sethostname
+75	common	setrlimit			sys_setrlimit
+76	common	getrlimit			sys_ni_syscall
+77	common	getrusage			sys_getrusage
+78	common	gettimeofday			sys_gettimeofday
+79	common	settimeofday			sys_settimeofday
+80	common	getgroups			sys_getgroups
+81	common	setgroups			sys_setgroups
+82	common	select				sys_ni_syscall
+83	common	symlink				sys_symlink
+84	common	oldlstat			sys_ni_syscall
+85	common	readlink			sys_readlink
+86	common	uselib				sys_uselib
+87	common	swapon				sys_swapon
+88	common	reboot				sys_reboot
+89	common	readdir				sys_ni_syscall
+90	common	mmap				sys_mmap
+91	common	munmap				sys_munmap
+92	common	truncate			sys_truncate
+93	common	ftruncate			sys_ftruncate
+94	common	fchmod				sys_fchmod
+95	common	fchown				sys_fchown
+96	common	getpriority			sys_getpriority
+97	common	setpriority			sys_setpriority
+98	common	profil				sys_ni_syscall
+99	common	statfs				sys_statfs
+100	common	fstatfs				sys_fstatfs
+101	common	ioperm				sys_ni_syscall
+102	common	socketcall			sys_socketcall
+103	common	syslog				sys_syslog
+104	common	setitimer			sys_setitimer
+105	common	getitimer			sys_getitimer
+106	common	stat				sys_newstat
+107	common	lstat				sys_newlstat
+108	common	fstat				sys_newfstat
+109	common	olduname			sys_ni_syscall
+110	common	iopl				sys_ni_syscall
+111	common	vhangup				sys_vhangup
+112	common	idle				sys_ni_syscall
+113	common	vm86old				sys_ni_syscall
+114	common	wait4				sys_wait4
+115	common	swapoff				sys_swapoff
+116	common	sysinfo				sys_sysinfo
+117	common	ipc				sys_ni_syscall
+118	common	fsync				sys_fsync
+119	common	sigreturn			sys_ni_syscall
+120	common	clone				sys_clone
+121	common	setdomainname			sys_setdomainname
+122	common	uname				sys_newuname
+123	common	modify_ldt			sys_ni_syscall
+124	common	adjtimex			sys_adjtimex_time32
+125	common	mprotect			sys_mprotect
+126	common	sigprocmask			sys_sigprocmask
+127	common	create_module			sys_ni_syscall
+128	common	init_module			sys_init_module
+129	common	delete_module			sys_delete_module
+130	common	get_kernel_syms			sys_ni_syscall
+131	common	quotactl			sys_quotactl
+132	common	getpgid				sys_getpgid
+133	common	fchdir				sys_fchdir
+134	common	bdflush				sys_ni_syscall
+135	common	sysfs				sys_sysfs
+136	common	personality			sys_personality
+137	common	afs_syscall			sys_ni_syscall
+138	common	setfsuid			sys_setfsuid
+139	common	setfsgid			sys_setfsgid
+140	common	_llseek				sys_llseek
+141	common	getdents			sys_getdents
+142	common	_newselect			sys_select
+143	common	flock				sys_flock
+144	common	msync				sys_msync
+145	common	readv				sys_readv
+146	common	writev				sys_writev
+147	common	getsid				sys_getsid
+148	common	fdatasync			sys_fdatasync
+149	common	_sysctl				sys_ni_syscall
+150	common	mlock				sys_mlock
+151	common	munlock				sys_munlock
+152	common	mlockall			sys_mlockall
+153	common	munlockall			sys_munlockall
+154	common	sched_setparam			sys_sched_setparam
+155	common	sched_getparam			sys_sched_getparam
+156	common	sched_setscheduler		sys_sched_setscheduler
+157	common	sched_getscheduler		sys_sched_getscheduler
+158	common	sched_yield			sys_sched_yield
+159	common	sched_get_priority_max		sys_sched_get_priority_max
+160	common	sched_get_priority_min		sys_sched_get_priority_min
+161	common	sched_rr_get_interval		sys_sched_rr_get_interval_time32
+162	common	nanosleep			sys_nanosleep_time32
+163	common	mremap				sys_mremap
+164	common	setresuid			sys_setresuid
+165	common	getresuid			sys_getresuid
+166	common	vm86				sys_ni_syscall
+167	common	query_module			sys_ni_syscall
+168	common	poll				sys_poll
+169	common	nfsservctl			sys_ni_syscall
+170	common	setresgid			sys_setresgid
+171	common	getresgid			sys_getresgid
+172	common	prctl				sys_prctl
+173	common	rt_sigreturn			sys_rt_sigreturn_wrapper
+174	common	rt_sigaction			sys_rt_sigaction
+175	common	rt_sigprocmask			sys_rt_sigprocmask
+176	common	rt_sigpending			sys_rt_sigpending
+177	common	rt_sigtimedwait			sys_rt_sigtimedwait_time32
+178	common	rt_sigqueueinfo			sys_rt_sigqueueinfo
+179	common	rt_sigsuspend			sys_rt_sigsuspend
+180	common	pread64				sys_pread64
+181	common	pwrite64			sys_pwrite64
+182	common	chown				sys_chown
+183	common	getcwd				sys_getcwd
+184	common	capget				sys_capget
+185	common	capset				sys_capset
+186	common	sigaltstack			sys_ni_syscall
+187	common	sendfile			sys_sendfile
+188	common	getpmsg				sys_ni_syscall
+189	common	putpmsg				sys_ni_syscall
+190	common	vfork				sys_vfork
+191	common	ugetrlimit			sys_getrlimit
+192	common	mmap2				sys_mmap2
+193	common	truncate64			sys_truncate64
+194	common	ftruncate64			sys_ftruncate64
+195	common	stat64				sys_stat64
+196	common	lstat64				sys_lstat64
+197	common	fstat64				sys_fstat64
+198	common	lchown32			sys_lchown
+199	common	getuid32			sys_getuid
+200	common	getgid32			sys_getgid
+201	common	geteuid32			sys_geteuid
+202	common	getegid32			sys_getegid
+203	common	setreuid32			sys_setreuid
+204	common	setregid32			sys_setregid
+205	common	getgroups32			sys_getgroups
+206	common	setgroups32			sys_setgroups
+207	common	fchown32			sys_fchown
+208	common	setresuid32			sys_setresuid
+209	common	getresuid32			sys_getresuid
+210	common	setresgid32			sys_setresgid
+211	common	getresgid32			sys_getresgid
+212	common	chown32				sys_chown
+213	common	setuid32			sys_setuid
+214	common	setgid32			sys_setgid
+215	common	setfsuid32			sys_setfsuid
+216	common	setfsgid32			sys_setfsgid
+217	common	pivot_root			sys_pivot_root
+218	common	mincore				sys_mincore
+219	common	madvise				sys_madvise
+220	common	getdents64			sys_getdents64
+221	common	fcntl64				sys_fcntl64
+# 222 is reserved for TUX
+# 223 is unused
+224	common	gettid				sys_gettid
+225	common	readahead			sys_readahead
+226	common	setxattr			sys_setxattr
+227	common	lsetxattr			sys_lsetxattr
+228	common	fsetxattr			sys_fsetxattr
+229	common	getxattr			sys_getxattr
+230	common	lgetxattr			sys_lgetxattr
+231	common	fgetxattr			sys_fgetxattr
+232	common	listxattr			sys_listxattr
+233	common	llistxattr			sys_llistxattr
+234	common	flistxattr			sys_flistxattr
+235	common	removexattr			sys_removexattr
+236	common	lremovexattr			sys_lremovexattr
+237	common	fremovexattr			sys_fremovexattr
+238	common	tkill				sys_tkill
+239	common	sendfile64			sys_sendfile64
+240	common	futex				sys_futex_time32
+241	common	sched_setaffinity		sys_sched_setaffinity
+242	common	sched_getaffinity		sys_sched_getaffinity
+243	common	set_thread_area			sys_ni_syscall
+244	common	get_thread_area			sys_ni_syscall
+245	common	io_setup			sys_io_setup
+246	common	io_destroy			sys_io_destroy
+247	common	io_getevents			sys_io_getevents_time32
+248	common	io_submit			sys_io_submit
+249	common	io_cancel			sys_io_cancel
+250	common	fadvise64			sys_fadvise64
+# 251 is available for reuse (was briefly sys_set_zone_reclaim)
+252	common	exit_group			sys_exit_group
+253	common	lookup_dcookie			sys_ni_syscall
+254	common	epoll_create			sys_epoll_create
+255	common	epoll_ctl			sys_epoll_ctl
+256	common	epoll_wait			sys_epoll_wait
+257	common	remap_file_pages		sys_remap_file_pages
+258	common	set_tid_address			sys_set_tid_address
+259	common	timer_create			sys_timer_create
+260	common	timer_settime			sys_timer_settime32
+261	common	timer_gettime			sys_timer_gettime32
+262	common	timer_getoverrun		sys_timer_getoverrun
+263	common	timer_delete			sys_timer_delete
+264	common	clock_settime			sys_clock_settime32
+265	common	clock_gettime			sys_clock_gettime32
+266	common	clock_getres			sys_clock_getres_time32
+267	common	clock_nanosleep			sys_clock_nanosleep_time32
+268	common	statfs64			sys_statfs64
+269	common	fstatfs64			sys_fstatfs64
+270	common	tgkill				sys_tgkill
+271	common	utimes				sys_utimes_time32
+272	common	fadvise64_64			sys_fadvise64_64
+273	common	vserver				sys_ni_syscall
+274	common	mbind				sys_mbind
+275	common	get_mempolicy			sys_get_mempolicy
+276	common	set_mempolicy			sys_set_mempolicy
+277	common	mq_open				sys_mq_open
+278	common	mq_unlink			sys_mq_unlink
+279	common	mq_timedsend			sys_mq_timedsend_time32
+280	common	mq_timedreceive			sys_mq_timedreceive_time32
+281	common	mq_notify			sys_mq_notify
+282	common	mq_getsetattr			sys_mq_getsetattr
+283	common	kexec_load			sys_kexec_load
+284	common	waitid				sys_waitid
+# 285 was setaltroot
+286	common	add_key				sys_add_key
+287	common	request_key			sys_request_key
+288	common	keyctl				sys_keyctl
+289	common	ioprio_set			sys_ioprio_set
+290	common	ioprio_get			sys_ioprio_get
+291	common	inotify_init			sys_inotify_init
+292	common	inotify_add_watch		sys_inotify_add_watch
+293	common	inotify_rm_watch		sys_inotify_rm_watch
+294	common	migrate_pages			sys_ni_syscall
+295	common	openat				sys_openat
+296	common	mkdirat				sys_mkdirat
+297	common	mknodat				sys_mknodat
+298	common	fchownat			sys_fchownat
+299	common	futimesat			sys_futimesat_time32
+300	common	fstatat64			sys_fstatat64
+301	common	unlinkat			sys_unlinkat
+302	common	renameat			sys_renameat
+303	common	linkat				sys_linkat
+304	common	symlinkat			sys_symlinkat
+305	common	readlinkat			sys_readlinkat
+306	common	fchmodat			sys_fchmodat
+307	common	faccessat			sys_faccessat
+308	common	pselect6			sys_pselect6_time32
+309	common	ppoll				sys_ppoll_time32
+310	common	unshare				sys_unshare
+311	common	set_robust_list			sys_set_robust_list
+312	common	get_robust_list			sys_get_robust_list
+313	common	splice				sys_splice
+314	common	sync_file_range			sys_sync_file_range
+315	common	tee				sys_tee
+316	common	vmsplice			sys_vmsplice
+317	common	move_pages			sys_move_pages
+318	common	getcpu				sys_getcpu
+319	common	epoll_pwait			sys_epoll_pwait
+320	common	utimensat			sys_utimensat_time32
+321	common	signalfd			sys_signalfd
+322	common	timerfd_create			sys_timerfd_create
+323	common	eventfd				sys_eventfd
+324	common	fallocate			sys_fallocate
+325	common	semtimedop			sys_semtimedop_time32
+326	common	timerfd_settime			sys_timerfd_settime32
+327	common	timerfd_gettime			sys_timerfd_gettime32
+328	common	semctl				sys_old_semctl
+329	common	semget				sys_semget
+330	common	semop				sys_semop
+331	common	msgctl				sys_old_msgctl
+332	common	msgget				sys_msgget
+333	common	msgrcv				sys_msgrcv
+334	common	msgsnd				sys_msgsnd
+335	common	shmat				sys_shmat
+336	common	shmctl				sys_old_shmctl
+337	common	shmdt				sys_shmdt
+338	common	shmget				sys_shmget
+339	common	signalfd4			sys_signalfd4
+340	common	eventfd2			sys_eventfd2
+341	common	epoll_create1			sys_epoll_create1
+342	common	dup3				sys_dup3
+343	common	pipe2				sys_pipe2
+344	common	inotify_init1			sys_inotify_init1
+345	common	socket				sys_socket
+346	common	socketpair			sys_socketpair
+347	common	bind				sys_bind
+348	common	listen				sys_listen
+349	common	accept				sys_accept
+350	common	connect				sys_connect
+351	common	getsockname			sys_getsockname
+352	common	getpeername			sys_getpeername
+353	common	sendto				sys_sendto
+354	common	send				sys_send
+355	common	recvfrom			sys_recvfrom
+356	common	recv				sys_recv
+357	common	setsockopt			sys_setsockopt
+358	common	getsockopt			sys_getsockopt
+359	common	shutdown			sys_shutdown
+360	common	sendmsg				sys_sendmsg
+361	common	recvmsg				sys_recvmsg
+362	common	accept4				sys_accept4
+363	common	preadv				sys_preadv
+364	common	pwritev				sys_pwritev
+365	common	rt_tgsigqueueinfo		sys_rt_tgsigqueueinfo
+366	common	perf_event_open			sys_perf_event_open
+367	common	recvmmsg			sys_recvmmsg_time32
+368	common	fanotify_init			sys_fanotify_init
+369	common	fanotify_mark			sys_fanotify_mark
+370	common	prlimit64			sys_prlimit64
+371	common	name_to_handle_at		sys_name_to_handle_at
+372	common	open_by_handle_at		sys_open_by_handle_at
+373	common	clock_adjtime			sys_clock_adjtime32
+374	common	syncfs				sys_syncfs
+375	common	setns				sys_setns
+376	common	sendmmsg			sys_sendmmsg
+377	common	process_vm_readv		sys_process_vm_readv
+378	common	process_vm_writev		sys_process_vm_writev
+379	common	kcmp				sys_kcmp
+380	common	finit_module			sys_finit_module
+381	common	sched_setattr			sys_sched_setattr
+382	common	sched_getattr			sys_sched_getattr
+383	common	renameat2			sys_renameat2
+384	common	seccomp				sys_seccomp
+385	common	getrandom			sys_getrandom
+386	common	memfd_create			sys_memfd_create
+387	common	bpf				sys_bpf
+388	common	execveat			sys_execveat
+389	common	userfaultfd			sys_userfaultfd
+390	common	membarrier			sys_membarrier
+391	common	mlock2				sys_mlock2
+392	common	copy_file_range			sys_copy_file_range
+393	common	preadv2				sys_preadv2
+394	common	pwritev2			sys_pwritev2
+395	common	pkey_mprotect			sys_pkey_mprotect
+396	common	pkey_alloc			sys_pkey_alloc
+397	common	pkey_free			sys_pkey_free
+398	common	statx				sys_statx
+399	common	io_pgetevents			sys_io_pgetevents_time32
+400	common	rseq				sys_rseq
+# 401 and 402 are unused
+403	common	clock_gettime64			sys_clock_gettime
+404	common	clock_settime64			sys_clock_settime
+405	common	clock_adjtime64			sys_clock_adjtime
+406	common	clock_getres_time64		sys_clock_getres
+407	common	clock_nanosleep_time64		sys_clock_nanosleep
+408	common	timer_gettime64			sys_timer_gettime
+409	common	timer_settime64			sys_timer_settime
+410	common	timerfd_gettime64		sys_timerfd_gettime
+411	common	timerfd_settime64		sys_timerfd_settime
+412	common	utimensat_time64		sys_utimensat
+413	common	pselect6_time64			sys_pselect6
+414	common	ppoll_time64			sys_ppoll
+416	common	io_pgetevents_time64		sys_io_pgetevents
+417	common	recvmmsg_time64			sys_recvmmsg
+418	common	mq_timedsend_time64		sys_mq_timedsend
+419	common	mq_timedreceive_time64		sys_mq_timedreceive
+420	common	semtimedop_time64		sys_semtimedop
+421	common	rt_sigtimedwait_time64		sys_rt_sigtimedwait
+422	common	futex_time64			sys_futex
+423	common	sched_rr_get_interval_time64	sys_sched_rr_get_interval
+424	common	pidfd_send_signal		sys_pidfd_send_signal
+425	common	io_uring_setup			sys_io_uring_setup
+426	common	io_uring_enter			sys_io_uring_enter
+427	common	io_uring_register		sys_io_uring_register
+428	common	open_tree			sys_open_tree
+429	common	move_mount			sys_move_mount
+430	common	fsopen				sys_fsopen
+431	common	fsconfig			sys_fsconfig
+432	common	fsmount				sys_fsmount
+433	common	fspick				sys_fspick
+434	common	pidfd_open			sys_pidfd_open
+435	common	clone3				sys_clone3
+436	common	close_range			sys_close_range
+437	common	openat2				sys_openat2
+438	common	pidfd_getfd			sys_pidfd_getfd
+439	common	faccessat2			sys_faccessat2
+440	common	process_madvise			sys_process_madvise
+441	common	epoll_pwait2			sys_epoll_pwait2
+442	common	mount_setattr			sys_mount_setattr
+443	common	quotactl_fd			sys_quotactl_fd
+444	common	landlock_create_ruleset		sys_landlock_create_ruleset
+445	common	landlock_add_rule		sys_landlock_add_rule
+446	common	landlock_restrict_self		sys_landlock_restrict_self
+# 447 reserved for memfd_secret
+448	common	process_mrelease		sys_process_mrelease
+449	common  futex_waitv                     sys_futex_waitv
+450	common	set_mempolicy_home_node		sys_set_mempolicy_home_node
+451	common	cachestat			sys_cachestat
+452	common	fchmodat2			sys_fchmodat2
+453	common	map_shadow_stack		sys_map_shadow_stack
+454	common	futex_wake			sys_futex_wake
+455	common	futex_wait			sys_futex_wait
+456	common	futex_requeue			sys_futex_requeue
+457	common	statmount			sys_statmount
+458	common	listmount			sys_listmount
+459	common	lsm_get_self_attr		sys_lsm_get_self_attr
+460	common	lsm_set_self_attr		sys_lsm_set_self_attr
+461	common	lsm_list_modules		sys_lsm_list_modules
+462	common	mseal				sys_mseal
+463	common	setxattrat			sys_setxattrat
+464	common	getxattrat			sys_getxattrat
+465	common	listxattrat			sys_listxattrat
+466	common	removexattrat			sys_removexattrat
+467	common	open_tree_attr			sys_open_tree_attr
+468	common	file_getattr			sys_file_getattr
+469	common	file_setattr			sys_file_setattr
+470	common	listns				sys_listns
+471	common	rseq_slice_yield		sys_rseq_slice_yield
+472	common	fchroot				sys_fchroot
diff --git a/arch/microblaze/kernel/timer.c b/arch/microblaze/kernel/timer.c
new file mode 100644
index 000000000..a2ab67b74
--- /dev/null
+++ b/arch/microblaze/kernel/timer.c
@@ -0,0 +1,330 @@
+/*
+ * 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/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;
+	timer_ack();
+	evt->event_handler(evt);
+	return IRQ_HANDLED;
+}
+
+static __init int 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.max_delta_ticks = (u32)~0;
+	clockevent_xilinx_timer.min_delta_ns =
+		clockevent_delta2ns(1, &clockevent_xilinx_timer);
+	clockevent_xilinx_timer.min_delta_ticks = 1;
+	clockevent_xilinx_timer.cpumask = cpumask_of(0);
+	clockevents_register_device(&clockevent_xilinx_timer);
+
+	return 0;
+}
+
+static u64 xilinx_clock_read(void)
+{
+	return read_fn(timer_baseaddr + TCR1);
+}
+
+static u64 xilinx_read(struct clocksource *cs)
+{
+	/* reading actual value of timer 1 */
+	return (u64)xilinx_clock_read();
+}
+
+static struct timecounter xilinx_tc = {
+	.cc = NULL,
+};
+
+static u64 xilinx_cc_read(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)
+{
+	int ret;
+
+	ret = clocksource_register_hz(&clocksource_microblaze,
+				      timer_clock_freq);
+	if (ret) {
+		pr_err("failed to register clocksource");
+		return ret;
+	}
+
+	/* 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 */
+	return init_xilinx_timecounter();
+}
+
+static int __init xilinx_timer_init(struct device_node *timer)
+{
+	struct clk *clk;
+	static int initialized;
+	u32 irq;
+	u32 timer_num = 1;
+	int ret;
+
+	/* If this property is present, the device is a PWM and not a timer */
+	if (of_property_present(timer, "#pwm-cells"))
+		return 0;
+
+	if (initialized)
+		return -EINVAL;
+
+	initialized = 1;
+
+	timer_baseaddr = of_iomap(timer, 0);
+	if (!timer_baseaddr) {
+		pr_err("ERROR: invalid timer base address\n");
+		return -ENXIO;
+	}
+
+	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);
+	if (irq <= 0) {
+		pr_err("Failed to parse and map irq");
+		return -EINVAL;
+	}
+
+	of_property_read_u32(timer, "xlnx,one-timer-only", &timer_num);
+	if (timer_num) {
+		pr_err("Please enable two timers in HW\n");
+		return -EINVAL;
+	}
+
+	pr_info("%pOF: irq=%d\n", timer, 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;
+
+	ret = request_irq(irq, timer_interrupt, IRQF_TIMER, "timer",
+			  &clockevent_xilinx_timer);
+	if (ret) {
+		pr_err("Failed to setup IRQ");
+		return ret;
+	}
+
+	ret = xilinx_clocksource_init();
+	if (ret)
+		return ret;
+
+	ret = xilinx_clockevent_init();
+	if (ret)
+		return ret;
+
+	sched_clock_register(xilinx_clock_read, 32, timer_clock_freq);
+
+	return 0;
+}
+
+TIMER_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 000000000..080aa7692
--- /dev/null
+++ b/arch/microblaze/kernel/traps.c
@@ -0,0 +1,79 @@
+/*
+ * 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/cpu.h>
+#include <linux/export.h>
+#include <linux/kernel.h>
+#include <linux/kallsyms.h>
+#include <linux/sched.h>
+#include <linux/sched/debug.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, const char *loglvl)
+{
+	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;
+
+	printk("%sKernel Stack:\n", loglvl);
+
+	/*
+	 * 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(loglvl, "", DUMP_PREFIX_ADDRESS, 32, 4, (void *)fp,
+		       words_to_show << 2, 0);
+	printk("%s\n\nCall Trace:\n", loglvl);
+	microblaze_unwind(task, NULL, loglvl);
+	printk("%s\n", loglvl);
+
+	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 000000000..a530a7a6b
--- /dev/null
+++ b/arch/microblaze/kernel/unwind.c
@@ -0,0 +1,309 @@
+/*
+ * 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/sched/task_stack.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,
+				    const char *loglvl);
+
+/**
+ * unwind_trap - Unwind through a system trap, that stored previous state
+ *		 on the stack.
+ */
+static inline void unwind_trap(struct task_struct *task, unsigned long pc,
+				unsigned long fp, struct stack_trace *trace,
+				const char *loglvl)
+{
+	/* To be implemented */
+}
+
+/**
+ * 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
+ * @loglvl : Used for printk log level if (trace == NULL).
+ */
+static void microblaze_unwind_inner(struct task_struct *task,
+			     unsigned long pc, unsigned long fp,
+			     unsigned long leaf_return,
+			     struct stack_trace *trace,
+			     const char *loglvl)
+{
+	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()
+			 */
+			printk("%sHW EXCEPTION\n", loglvl);
+			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)
+					printk("%s%s\n", loglvl, handler->trap_name);
+				unwind_trap(task, pc, fp, trace, loglvl);
+				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)) {
+				printk("%s[<%p>] PID %lu [%s]\n",
+					loglvl, (void *) pc,
+					(unsigned long) task->pid,
+					task->comm);
+				break;
+			} else
+				print_ip_sym(loglvl, 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
+ * @loglvl : Used for printk log level if (trace == NULL).
+ */
+void microblaze_unwind(struct task_struct *task, struct stack_trace *trace,
+		       const char *loglvl)
+{
+	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, loglvl);
+		} 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, loglvl);
+		}
+	} 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, loglvl);
+	}
+}
+
diff --git a/arch/microblaze/kernel/vmlinux.lds.S b/arch/microblaze/kernel/vmlinux.lds.S
new file mode 100644
index 000000000..ae50d3d04
--- /dev/null
+++ b/arch/microblaze/kernel/vmlinux.lds.S
@@ -0,0 +1,142 @@
+/*
+ * 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)
+
+#define RO_EXCEPTION_TABLE_ALIGN	16
+
+#include <asm/cache.h>
+#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
+		SOFTIRQENTRY_TEXT
+		. = ALIGN (4) ;
+		_etext = . ;
+	}
+
+	. = ALIGN (8) ;
+	__fdt_blob : AT(ADDR(__fdt_blob) - LOAD_OFFSET) {
+		_fdt_start = . ;		/* place for fdt blob */
+		*(__fdt_blob) ;			/* Any link-placed DTB */
+	        . = _fdt_start + 0x10000;	/* Pad up to 64kbyte */
+		_fdt_end = . ;
+	}
+
+	. = ALIGN(16);
+	RO_DATA(4096)
+
+	/*
+	 * 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(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) ;
+	}
+
+	PERCPU_SECTION(L1_CACHE_BYTES)
+
+	. = 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
+	}
+
+	__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 000000000..9fe7ab688
--- /dev/null
+++ b/arch/microblaze/lib/Makefile
@@ -0,0 +1,24 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# 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 000000000..4d0f9481b
--- /dev/null
+++ b/arch/microblaze/lib/ashldi3.c
@@ -0,0 +1,29 @@
+// SPDX-License-Identifier: GPL-2.0
+#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 000000000..268098a96
--- /dev/null
+++ b/arch/microblaze/lib/ashrdi3.c
@@ -0,0 +1,31 @@
+// SPDX-License-Identifier: GPL-2.0
+#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 000000000..b3f896385
--- /dev/null
+++ b/arch/microblaze/lib/cmpdi2.c
@@ -0,0 +1,27 @@
+// SPDX-License-Identifier: GPL-2.0
+#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 000000000..919fb69f8
--- /dev/null
+++ b/arch/microblaze/lib/divsi3.S
@@ -0,0 +1,74 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#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 000000000..fdc48bb06
--- /dev/null
+++ b/arch/microblaze/lib/fastcopy.S
@@ -0,0 +1,666 @@
+/*
+ * 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
+ */
+
+#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 000000000..a909366c2
--- /dev/null
+++ b/arch/microblaze/lib/libgcc.h
@@ -0,0 +1,33 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#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 000000000..00eb82682
--- /dev/null
+++ b/arch/microblaze/lib/lshrdi3.c
@@ -0,0 +1,29 @@
+// SPDX-License-Identifier: GPL-2.0
+#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 000000000..9966dce55
--- /dev/null
+++ b/arch/microblaze/lib/memcpy.c
@@ -0,0 +1,179 @@
+/*
+ * 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 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;
+			fallthrough;
+		case 2:
+			*dst++ = *src++;
+			--c;
+			fallthrough;
+		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++;
+		fallthrough;
+	case 2:
+		*dst++ = *src++;
+		fallthrough;
+	case 1:
+		*dst++ = *src++;
+	}
+
+	return v_dst;
+}
+EXPORT_SYMBOL(memcpy);
+#endif /* CONFIG_OPT_LIB_FUNCTION */
diff --git a/arch/microblaze/lib/memmove.c b/arch/microblaze/lib/memmove.c
new file mode 100644
index 000000000..c1f08c484
--- /dev/null
+++ b/arch/microblaze/lib/memmove.c
@@ -0,0 +1,195 @@
+/*
+ * 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 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;
+			fallthrough;
+		case 2:
+			*--dst = *--src;
+			--c;
+			fallthrough;
+		case 1:
+			*--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 + 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;
+		fallthrough;
+	case 3:
+		*--dst = *--src;
+		fallthrough;
+	case 2:
+		*--dst = *--src;
+		fallthrough;
+	case 1:
+		*--dst = *--src;
+	}
+	return v_dst;
+}
+EXPORT_SYMBOL(memmove);
+#endif /* CONFIG_OPT_LIB_FUNCTION */
diff --git a/arch/microblaze/lib/memset.c b/arch/microblaze/lib/memset.c
new file mode 100644
index 000000000..7c2352d56
--- /dev/null
+++ b/arch/microblaze/lib/memset.c
@@ -0,0 +1,94 @@
+/*
+ * 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 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;
+			fallthrough;
+		case 2:
+			*src++ = c;
+			--n;
+			fallthrough;
+		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. */
+	switch (n) {
+	case 3:
+		*src++ = c;
+		fallthrough;
+	case 2:
+		*src++ = c;
+		fallthrough;
+	case 1:
+		*src++ = c;
+		break;
+	default:
+		break;
+	}
+
+	return v_src;
+}
+EXPORT_SYMBOL(memset);
+#endif /* CONFIG_OPT_LIB_FUNCTION */
diff --git a/arch/microblaze/lib/modsi3.S b/arch/microblaze/lib/modsi3.S
new file mode 100644
index 000000000..64154865c
--- /dev/null
+++ b/arch/microblaze/lib/modsi3.S
@@ -0,0 +1,74 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#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 000000000..8e6bc17e2
--- /dev/null
+++ b/arch/microblaze/lib/muldi3.c
@@ -0,0 +1,58 @@
+// SPDX-License-Identifier: GPL-2.0
+#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 000000000..d6703a4d9
--- /dev/null
+++ b/arch/microblaze/lib/mulsi3.S
@@ -0,0 +1,47 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#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 000000000..dd5f3bfbc
--- /dev/null
+++ b/arch/microblaze/lib/uaccess_old.S
@@ -0,0 +1,176 @@
+/*
+ * 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>
+
+/* 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 storing 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 000000000..c60e068ff
--- /dev/null
+++ b/arch/microblaze/lib/ucmpdi2.c
@@ -0,0 +1,21 @@
+// SPDX-License-Identifier: GPL-2.0
+#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 000000000..acdc66723
--- /dev/null
+++ b/arch/microblaze/lib/udivsi3.S
@@ -0,0 +1,85 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#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 000000000..f4b814e88
--- /dev/null
+++ b/arch/microblaze/lib/umodsi3.S
@@ -0,0 +1,87 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#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 000000000..75edfc110
--- /dev/null
+++ b/arch/microblaze/mm/Makefile
@@ -0,0 +1,6 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+# Makefile
+#
+
+obj-y := consistent.o init.o pgtable.o mmu_context.o fault.o
diff --git a/arch/microblaze/mm/consistent.c b/arch/microblaze/mm/consistent.c
new file mode 100644
index 000000000..b7ad4a986
--- /dev/null
+++ b/arch/microblaze/mm/consistent.c
@@ -0,0 +1,23 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * 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>
+ */
+
+#include <linux/kernel.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/dma-map-ops.h>
+#include <asm/cpuinfo.h>
+#include <asm/cacheflush.h>
+
+void arch_dma_prep_coherent(struct page *page, size_t size)
+{
+	phys_addr_t paddr = page_to_phys(page);
+
+	flush_dcache_range(paddr, paddr + size);
+}
diff --git a/arch/microblaze/mm/fault.c b/arch/microblaze/mm/fault.c
new file mode 100644
index 000000000..d3c3c33b7
--- /dev/null
+++ b/arch/microblaze/mm/fault.c
@@ -0,0 +1,299 @@
+/*
+ *  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/extable.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 <linux/perf_event.h>
+
+#include <asm/page.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;
+	int code = SEGV_MAPERR;
+	int is_write = error_code & ESR_S;
+	vm_fault_t fault;
+	unsigned int flags = FAULT_FLAG_DEFAULT;
+
+	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;
+
+	perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, address);
+
+	/* 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_lock
+	 * 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(!mmap_read_trylock(mm))) {
+		if (kernel_mode(regs) && !search_exception_tables(regs->pc))
+			goto bad_area_nosemaphore;
+
+retry:
+		mmap_read_lock(mm);
+	}
+
+	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;
+	}
+	vma = expand_stack(mm, address);
+	if (!vma)
+		goto bad_area_nosemaphore;
+
+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(vma, address, flags, regs);
+
+	if (fault_signal_pending(fault, regs)) {
+		if (!user_mode(regs))
+			bad_page_fault(regs, address, SIGBUS);
+		return;
+	}
+
+	/* The fault is fully completed (including releasing mmap lock) */
+	if (fault & VM_FAULT_COMPLETED)
+		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 (fault & VM_FAULT_RETRY) {
+		flags |= FAULT_FLAG_TRIED;
+
+		/*
+		 * No need to mmap_read_unlock(mm) as we would
+		 * have already released it in __lock_page_or_retry
+		 * in mm/filemap.c.
+		 */
+
+		goto retry;
+	}
+
+	mmap_read_unlock(mm);
+
+	/*
+	 * 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:
+	mmap_read_unlock(mm);
+
+bad_area_nosemaphore:
+	pte_errors++;
+
+	/* User mode accesses cause a SIGSEGV */
+	if (user_mode(regs)) {
+		_exception(SIGSEGV, regs, code, address);
+		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:
+	mmap_read_unlock(mm);
+	if (!user_mode(regs))
+		bad_page_fault(regs, address, SIGKILL);
+	else
+		pagefault_out_of_memory();
+	return;
+
+do_sigbus:
+	mmap_read_unlock(mm);
+	if (user_mode(regs)) {
+		force_sig_fault(SIGBUS, BUS_ADRERR, (void __user *)address);
+		return;
+	}
+	bad_page_fault(regs, address, SIGBUS);
+}
diff --git a/arch/microblaze/mm/init.c b/arch/microblaze/mm/init.c
new file mode 100644
index 000000000..848cdee13
--- /dev/null
+++ b/arch/microblaze/mm/init.c
@@ -0,0 +1,259 @@
+/*
+ * 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/dma-map-ops.h>
+#include <linux/memblock.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/mm.h> /* mem_init */
+#include <linux/initrd.h>
+#include <linux/of_fdt.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;
+
+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;
+
+EXPORT_SYMBOL(min_low_pfn);
+EXPORT_SYMBOL(max_low_pfn);
+
+#ifdef CONFIG_HIGHMEM
+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);
+}
+#endif /* CONFIG_HIGHMEM */
+
+void __init arch_zone_limits_init(unsigned long *max_zone_pfns)
+{
+#ifdef CONFIG_HIGHMEM
+	max_zone_pfns[ZONE_DMA] = max_low_pfn;
+	max_zone_pfns[ZONE_HIGHMEM] = max_pfn;
+#else
+	max_zone_pfns[ZONE_DMA] = max_pfn;
+#endif
+}
+
+/*
+ * paging_init() sets up the page tables - in fact we've already done this.
+ */
+static void __init paging_init(void)
+{
+	int idx;
+
+	/* Setup fixmaps */
+	for (idx = 0; idx < __end_of_fixed_addresses; idx++)
+		clear_fixmap(idx);
+
+#ifdef CONFIG_HIGHMEM
+	highmem_init();
+#endif
+}
+
+void __init setup_memory(void)
+{
+	/*
+	 * 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
+	 */
+
+	/* memory start is from the kernel end (aligned) to higher addr */
+	min_low_pfn = memory_start >> PAGE_SHIFT; /* minimum for allocation */
+	max_low_pfn = ((u64)memory_start + (u64)lowmem_size) >> PAGE_SHIFT;
+	max_pfn = ((u64)memory_start + (u64)memory_size) >> PAGE_SHIFT;
+
+	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);
+
+	paging_init();
+}
+
+void __init mem_init(void)
+{
+	mem_init_done = 1;
+}
+
+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 __init 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 ((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);
+
+	parse_early_param();
+
+	early_init_fdt_scan_reserved_mem();
+
+	/* CMA initialization */
+	dma_contiguous_reserve(memory_start + lowmem_size - 1);
+
+	memblock_dump_all();
+}
+
+static const pgprot_t protection_map[16] = {
+	[VM_NONE]					= PAGE_NONE,
+	[VM_READ]					= PAGE_READONLY_X,
+	[VM_WRITE]					= PAGE_COPY,
+	[VM_WRITE | VM_READ]				= PAGE_COPY_X,
+	[VM_EXEC]					= PAGE_READONLY,
+	[VM_EXEC | VM_READ]				= PAGE_READONLY_X,
+	[VM_EXEC | VM_WRITE]				= PAGE_COPY,
+	[VM_EXEC | VM_WRITE | VM_READ]			= PAGE_COPY_X,
+	[VM_SHARED]					= PAGE_NONE,
+	[VM_SHARED | VM_READ]				= PAGE_READONLY_X,
+	[VM_SHARED | VM_WRITE]				= PAGE_SHARED,
+	[VM_SHARED | VM_WRITE | VM_READ]		= PAGE_SHARED_X,
+	[VM_SHARED | VM_EXEC]				= PAGE_READONLY,
+	[VM_SHARED | VM_EXEC | VM_READ]			= PAGE_READONLY_X,
+	[VM_SHARED | VM_EXEC | VM_WRITE]		= PAGE_SHARED,
+	[VM_SHARED | VM_EXEC | VM_WRITE | VM_READ]	= PAGE_SHARED_X
+};
+DECLARE_VM_GET_PAGE_PROT
diff --git a/arch/microblaze/mm/mmu_context.c b/arch/microblaze/mm/mmu_context.c
new file mode 100644
index 000000000..cbc234816
--- /dev/null
+++ b/arch/microblaze/mm/mmu_context.c
@@ -0,0 +1,65 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * 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
+ */
+
+#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 000000000..e96dd1b7a
--- /dev/null
+++ b/arch/microblaze/mm/pgtable.c
@@ -0,0 +1,264 @@
+/*
+ *  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 <linux/mm_types.h>
+#include <linux/pgtable.h>
+#include <linux/memblock.h>
+#include <linux/kallsyms.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);
+
+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 %ps\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(volatile 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)
+{
+	p4d_t *p4d;
+	pud_t *pud;
+	pmd_t *pd;
+	pte_t *pg;
+	int err = -ENOMEM;
+
+	/* Use upper 10 bits of VA to index the first level map */
+	p4d = p4d_offset(pgd_offset_k(va), va);
+	pud = pud_offset(p4d, va);
+	pd = pmd_offset(pud, 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 (!is_kernel_text(v))
+			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;
+	p4d_t	*p4d;
+	pud_t	*pud;
+	pmd_t	*pmd;
+	pte_t	*pte;
+	int     retval = 0;
+
+	pgd = pgd_offset(mm, addr & PAGE_MASK);
+	if (pgd) {
+		p4d = p4d_offset(pgd, addr & PAGE_MASK);
+		pud = pud_offset(p4d, addr & PAGE_MASK);
+		pmd = pmd_offset(pud, 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;
+}
+
+__ref pte_t *pte_alloc_one_kernel(struct mm_struct *mm)
+{
+	if (mem_init_done)
+		return __pte_alloc_one_kernel(mm);
+	else
+		return memblock_alloc_try_nid(PAGE_SIZE, PAGE_SIZE,
+					      MEMBLOCK_LOW_LIMIT,
+					      memory_start + kernel_tlb,
+					      NUMA_NO_NODE);
+}
+
+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/pci/Makefile b/arch/microblaze/pci/Makefile
new file mode 100644
index 000000000..f8267d20e
--- /dev/null
+++ b/arch/microblaze/pci/Makefile
@@ -0,0 +1,6 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+# Makefile
+#
+
+obj-$(CONFIG_PCI)		+= iomap.o
diff --git a/arch/microblaze/pci/iomap.c b/arch/microblaze/pci/iomap.c
new file mode 100644
index 000000000..b2ee8ace9
--- /dev/null
+++ b/arch/microblaze/pci/iomap.c
@@ -0,0 +1,58 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * 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>
+
+static DEFINE_SPINLOCK(hose_spinlock);
+LIST_HEAD(hose_list);
+
+unsigned long isa_io_base;
+EXPORT_SYMBOL(isa_io_base);
+
+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;
+}
+
+/* Display the domain number in /proc */
+int pci_proc_domain(struct pci_bus *bus)
+{
+	return pci_domain_nr(bus);
+}
+
+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);