ata: libata-scsi: fix ata_dsm_trim_pages() kernel-docgrafted

make htmldocs fails with:

  Documentation/driver-api/libata:604: ./drivers/ata/libata-scsi.c:2301:
  ERROR: Unexpected indentation. [docutils]

The Return: section of ata_dsm_trim_pages() ends its first sentence with
a colon and continues with an indented bullet list. reStructuredText
requires a blank line before an indented block, so docutils chokes on
the list.

Simply adding the missing blank line does not work either: Return: is a
kernel-doc "special section", which is terminated by the first blank
line, so the bullet list would end up in the Description section,
detached from the sentence introducing it.

Spell the two bounds out as prose instead, so that the Return: section
stays self-contained. Documentation-only change.

Fixes: e64e6b5dc867 ("ata: libata-scsi: scale DSM TRIM payload by MAX PAGES PER DSM COMMAND")
Reported-by: Thomas Huth <thuth@redhat.com>
Closes: https://lore.kernel.org/linux-ide/b0a0b8e8-cc4f-4c7b-8bc5-fee0712d405a@redhat.com/
Reviewed-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Link: https://lore.kernel.org/r/20260916130031.29990-2-cassel@kernel.org
Signed-off-by: Niklas Cassel <cassel@kernel.org>

28 files changed, 3309 insertions, 0 deletions

diff --git a/arch/arm/boot/compressed/.gitignore b/arch/arm/boot/compressed/.gitignore
new file mode 100644
index 000000000..d32f41778
--- /dev/null
+++ b/arch/arm/boot/compressed/.gitignore
@@ -0,0 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
+piggy_data
+vmlinux
+vmlinux.lds
diff --git a/arch/arm/boot/compressed/Makefile b/arch/arm/boot/compressed/Makefile
new file mode 100644
index 000000000..e3f550d62
--- /dev/null
+++ b/arch/arm/boot/compressed/Makefile
@@ -0,0 +1,160 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# linux/arch/arm/boot/compressed/Makefile
+#
+# create a compressed vmlinuz image from the original vmlinux
+#
+
+OBJS		=
+
+HEAD	= head.o
+OBJS	+= misc.o decompress.o
+ifeq ($(CONFIG_DEBUG_UNCOMPRESS),y)
+OBJS	+= debug.o
+AFLAGS_head.o += -DDEBUG
+endif
+
+# string library code (-Os is enforced to keep it much smaller)
+OBJS		+= string.o
+CFLAGS_string.o	:= -Os
+
+ifeq ($(CONFIG_ARM_VIRT_EXT),y)
+OBJS		+= hyp-stub.o
+endif
+
+#
+# Architecture dependencies
+#
+ifeq ($(CONFIG_ARCH_ACORN),y)
+OBJS		+= ll_char_wr.o font.o
+endif
+
+ifeq ($(CONFIG_ARCH_SA1100),y)
+OBJS		+= head-sa1100.o
+endif
+
+ifeq ($(CONFIG_CPU_XSCALE),y)
+OBJS		+= head-xscale.o
+endif
+
+ifeq ($(CONFIG_PXA_SHARPSL_DETECT_MACH_ID),y)
+OBJS		+= head-sharpsl.o
+endif
+
+ifeq ($(CONFIG_CPU_ENDIAN_BE32),y)
+ifeq ($(CONFIG_CPU_CP15),y)
+OBJS		+= big-endian.o
+else
+# The endian should be set by h/w design.
+endif
+endif
+
+#
+# We now have a PIC decompressor implementation.  Decompressors running
+# from RAM should not define ZTEXTADDR.  Decompressors running directly
+# from ROM or Flash must define ZTEXTADDR (preferably via the config)
+# FIXME: Previous assignment to ztextaddr-y is lost here. See SHARK
+ifeq ($(CONFIG_ZBOOT_ROM),y)
+ZTEXTADDR	:= $(CONFIG_ZBOOT_ROM_TEXT)
+ZBSSADDR	:= $(CONFIG_ZBOOT_ROM_BSS)
+else
+ZTEXTADDR	:= 0
+ZBSSADDR	:= ALIGN(8)
+endif
+
+MALLOC_SIZE	:= 65536
+
+AFLAGS_head.o += -DTEXT_OFFSET=$(TEXT_OFFSET) -DMALLOC_SIZE=$(MALLOC_SIZE)
+CPPFLAGS_vmlinux.lds := -DTEXT_START="$(ZTEXTADDR)" -DBSS_START="$(ZBSSADDR)"
+CPPFLAGS_vmlinux.lds += -DTEXT_OFFSET="$(TEXT_OFFSET)"
+CPPFLAGS_vmlinux.lds += -DMALLOC_SIZE="$(MALLOC_SIZE)"
+
+compress-$(CONFIG_KERNEL_GZIP) = gzip
+compress-$(CONFIG_KERNEL_LZO)  = lzo_with_size
+compress-$(CONFIG_KERNEL_LZMA) = lzma_with_size
+compress-$(CONFIG_KERNEL_XZ)   = xzkern_with_size
+compress-$(CONFIG_KERNEL_LZ4)  = lz4_with_size
+
+libfdt_objs := fdt_rw.o fdt_ro.o fdt_wip.o fdt.o
+
+ifeq ($(CONFIG_ARM_ATAG_DTB_COMPAT),y)
+CFLAGS_REMOVE_atags_to_fdt.o += -Wframe-larger-than=${CONFIG_FRAME_WARN}
+CFLAGS_atags_to_fdt.o += -Wframe-larger-than=1280
+OBJS	+= $(libfdt_objs) atags_to_fdt.o
+endif
+ifeq ($(CONFIG_USE_OF),y)
+OBJS	+= $(libfdt_objs) fdt_check_mem_start.o
+endif
+
+OBJS	+= lib1funcs.o ashldi3.o bswapsdi2.o
+
+targets       := vmlinux vmlinux.lds piggy_data piggy.o \
+		 head.o $(OBJS)
+
+KBUILD_CFLAGS += -DDISABLE_BRANCH_PROFILING
+
+ccflags-y := -fpic $(call cc-option,-mno-single-pic-base,) -fno-builtin \
+	     -I$(srctree)/scripts/dtc/libfdt -fno-stack-protector \
+	     $(DISABLE_KSTACK_ERASE) \
+	     -I$(obj)
+ccflags-remove-$(CONFIG_FUNCTION_TRACER) += -pg
+asflags-y := -DZIMAGE
+
+# Supply kernel BSS size to the decompressor via a linker symbol.
+KBSS_SZ = $(shell echo $$(($$($(NM) vmlinux | \
+		sed -n -e 's/^\([^ ]*\) [ABD] __bss_start$$/-0x\1/p' \
+		       -e 's/^\([^ ]*\) [ABD] __bss_stop$$/+0x\1/p') )) )
+LDFLAGS_vmlinux = --defsym _kernel_bss_size=$(KBSS_SZ)
+# Supply ZRELADDR to the decompressor via a linker symbol.
+ifneq ($(CONFIG_AUTO_ZRELADDR),y)
+LDFLAGS_vmlinux += --defsym zreladdr=$(ZRELADDR)
+endif
+ifeq ($(CONFIG_CPU_ENDIAN_BE8),y)
+LDFLAGS_vmlinux += --be8
+endif
+# Report unresolved symbol references
+LDFLAGS_vmlinux += --no-undefined
+# Delete all temporary local symbols
+LDFLAGS_vmlinux += -X
+# Report orphan sections
+ifdef CONFIG_LD_ORPHAN_WARN
+LDFLAGS_vmlinux += --orphan-handling=$(CONFIG_LD_ORPHAN_WARN_LEVEL)
+endif
+# Next argument is a linker script
+LDFLAGS_vmlinux += -T
+
+# We need to prevent any GOTOFF relocs being used with references
+# to symbols in the .bss section since we cannot relocate them
+# independently from the rest at run time.  This can be achieved by
+# ensuring that no private .bss symbols exist, as global symbols
+# always have a GOT entry which is what we need.
+# The .data section is already discarded by the linker script so no need
+# to bother about it here.
+check_for_bad_syms = \
+bad_syms=$$($(NM) $@ | sed -n 's/^.\{8\} [bc] \(.*\)/\1/p') && \
+[ -z "$$bad_syms" ] || \
+  ( echo "following symbols must have non local/private scope:" >&2; \
+    echo "$$bad_syms" >&2; false )
+
+check_for_multiple_zreladdr = \
+if [ $(words $(ZRELADDR)) -gt 1 -a "$(CONFIG_AUTO_ZRELADDR)" = "" ]; then \
+	echo 'multiple zreladdrs: $(ZRELADDR)'; \
+	echo 'This needs CONFIG_AUTO_ZRELADDR to be set'; \
+	false; \
+fi
+
+efi-obj-$(CONFIG_EFI_STUB) := $(objtree)/drivers/firmware/efi/libstub/lib.a
+
+$(obj)/vmlinux: $(obj)/vmlinux.lds $(obj)/$(HEAD) $(obj)/piggy.o \
+		$(addprefix $(obj)/, $(OBJS)) \
+		$(efi-obj-y) FORCE
+	@$(check_for_multiple_zreladdr)
+	$(call if_changed,ld)
+	@$(check_for_bad_syms)
+
+$(obj)/piggy_data: $(obj)/../Image FORCE
+	$(call if_changed,$(compress-y))
+
+$(obj)/piggy.o: $(obj)/piggy_data
+
+CFLAGS_font.o := -DBOOTLOADER
diff --git a/arch/arm/boot/compressed/ashldi3.S b/arch/arm/boot/compressed/ashldi3.S
new file mode 100644
index 000000000..216f82eda
--- /dev/null
+++ b/arch/arm/boot/compressed/ashldi3.S
@@ -0,0 +1,3 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* For __aeabi_llsl */
+#include "../../lib/ashldi3.S"
diff --git a/arch/arm/boot/compressed/atags_to_fdt.c b/arch/arm/boot/compressed/atags_to_fdt.c
new file mode 100644
index 000000000..627752f18
--- /dev/null
+++ b/arch/arm/boot/compressed/atags_to_fdt.c
@@ -0,0 +1,218 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/libfdt_env.h>
+#include <asm/setup.h>
+#include <libfdt.h>
+#include "misc.h"
+
+#if defined(CONFIG_ARM_ATAG_DTB_COMPAT_CMDLINE_EXTEND)
+#define do_extend_cmdline 1
+#else
+#define do_extend_cmdline 0
+#endif
+
+#define NR_BANKS 16
+
+static int node_offset(void *fdt, const char *node_path)
+{
+	int offset = fdt_path_offset(fdt, node_path);
+	if (offset == -FDT_ERR_NOTFOUND)
+		/* Add the node to root if not found, dropping the leading '/' */
+		offset = fdt_add_subnode(fdt, 0, node_path + 1);
+	return offset;
+}
+
+static int setprop(void *fdt, const char *node_path, const char *property,
+		   void *val_array, int size)
+{
+	int offset = node_offset(fdt, node_path);
+	if (offset < 0)
+		return offset;
+	return fdt_setprop(fdt, offset, property, val_array, size);
+}
+
+static int setprop_string(void *fdt, const char *node_path,
+			  const char *property, const char *string)
+{
+	int offset = node_offset(fdt, node_path);
+	if (offset < 0)
+		return offset;
+	return fdt_setprop_string(fdt, offset, property, string);
+}
+
+static int setprop_cell(void *fdt, const char *node_path,
+			const char *property, uint32_t val)
+{
+	int offset = node_offset(fdt, node_path);
+	if (offset < 0)
+		return offset;
+	return fdt_setprop_cell(fdt, offset, property, val);
+}
+
+static const void *getprop(const void *fdt, const char *node_path,
+			   const char *property, int *len)
+{
+	int offset = fdt_path_offset(fdt, node_path);
+
+	if (offset == -FDT_ERR_NOTFOUND)
+		return NULL;
+
+	return fdt_getprop(fdt, offset, property, len);
+}
+
+static uint32_t get_cell_size(const void *fdt)
+{
+	int len;
+	uint32_t cell_size = 1;
+	const __be32 *size_len =  getprop(fdt, "/", "#size-cells", &len);
+
+	if (size_len)
+		cell_size = fdt32_to_cpu(*size_len);
+	return cell_size;
+}
+
+static void merge_fdt_bootargs(void *fdt, const char *fdt_cmdline)
+{
+	char cmdline[COMMAND_LINE_SIZE];
+	const char *fdt_bootargs;
+	char *ptr = cmdline;
+	int len = 0;
+
+	/* copy the fdt command line into the buffer */
+	fdt_bootargs = getprop(fdt, "/chosen", "bootargs", &len);
+	if (fdt_bootargs)
+		if (len < COMMAND_LINE_SIZE) {
+			memcpy(ptr, fdt_bootargs, len);
+			/* len is the length of the string
+			 * including the NULL terminator */
+			ptr += len - 1;
+		}
+
+	/* and append the ATAG_CMDLINE */
+	if (fdt_cmdline) {
+		len = strlen(fdt_cmdline);
+		if (ptr - cmdline + len + 2 < COMMAND_LINE_SIZE) {
+			*ptr++ = ' ';
+			memcpy(ptr, fdt_cmdline, len);
+			ptr += len;
+		}
+	}
+	*ptr = '\0';
+
+	setprop_string(fdt, "/chosen", "bootargs", cmdline);
+}
+
+static void hex_str(char *out, uint32_t value)
+{
+	uint32_t digit;
+	int idx;
+
+	for (idx = 7; idx >= 0; idx--) {
+		digit = value >> 28;
+		value <<= 4;
+		digit &= 0xf;
+		if (digit < 10)
+			digit += '0';
+		else
+			digit += 'A'-10;
+		*out++ = digit;
+	}
+	*out = '\0';
+}
+
+/*
+ * Convert and fold provided ATAGs into the provided FDT.
+ *
+ * Return values:
+ *    = 0 -> pretend success
+ *    = 1 -> bad ATAG (may retry with another possible ATAG pointer)
+ *    < 0 -> error from libfdt
+ */
+int atags_to_fdt(void *atag_list, void *fdt, int total_space)
+{
+	struct tag *atag = atag_list;
+	/* In the case of 64 bits memory size, need to reserve 2 cells for
+	 * address and size for each bank */
+	__be32 mem_reg_property[2 * 2 * NR_BANKS];
+	int memcount = 0;
+	int ret, memsize;
+
+	/* make sure we've got an aligned pointer */
+	if ((u32)atag_list & 0x3)
+		return 1;
+
+	/* if we get a DTB here we're done already */
+	if (*(__be32 *)atag_list == cpu_to_fdt32(FDT_MAGIC))
+	       return 0;
+
+	/* validate the ATAG */
+	if (atag->hdr.tag != ATAG_CORE ||
+	    (atag->hdr.size != tag_size(tag_core) &&
+	     atag->hdr.size != 2))
+		return 1;
+
+	/* let's give it all the room it could need */
+	ret = fdt_open_into(fdt, fdt, total_space);
+	if (ret < 0)
+		return ret;
+
+	for_each_tag(atag, atag_list) {
+		if (atag->hdr.tag == ATAG_CMDLINE) {
+			/* Append the ATAGS command line to the device tree
+			 * command line.
+			 * NB: This means that if the same parameter is set in
+			 * the device tree and in the tags, the one from the
+			 * tags will be chosen.
+			 */
+			if (do_extend_cmdline)
+				merge_fdt_bootargs(fdt,
+						   atag->u.cmdline.cmdline);
+			else
+				setprop_string(fdt, "/chosen", "bootargs",
+					       atag->u.cmdline.cmdline);
+		} else if (atag->hdr.tag == ATAG_MEM) {
+			if (memcount >= sizeof(mem_reg_property)/4)
+				continue;
+			if (!atag->u.mem.size)
+				continue;
+			memsize = get_cell_size(fdt);
+
+			if (memsize == 2) {
+				/* if memsize is 2, that means that
+				 * each data needs 2 cells of 32 bits,
+				 * so the data are 64 bits */
+				__be64 *mem_reg_prop64 =
+					(__be64 *)mem_reg_property;
+				mem_reg_prop64[memcount++] =
+					cpu_to_fdt64(atag->u.mem.start);
+				mem_reg_prop64[memcount++] =
+					cpu_to_fdt64(atag->u.mem.size);
+			} else {
+				mem_reg_property[memcount++] =
+					cpu_to_fdt32(atag->u.mem.start);
+				mem_reg_property[memcount++] =
+					cpu_to_fdt32(atag->u.mem.size);
+			}
+
+		} else if (atag->hdr.tag == ATAG_INITRD2) {
+			uint32_t initrd_start, initrd_size;
+			initrd_start = atag->u.initrd.start;
+			initrd_size = atag->u.initrd.size;
+			setprop_cell(fdt, "/chosen", "linux,initrd-start",
+					initrd_start);
+			setprop_cell(fdt, "/chosen", "linux,initrd-end",
+					initrd_start + initrd_size);
+		} else if (atag->hdr.tag == ATAG_SERIAL) {
+			char serno[16+2];
+			hex_str(serno, atag->u.serialnr.high);
+			hex_str(serno+8, atag->u.serialnr.low);
+			setprop_string(fdt, "/", "serial-number", serno);
+		}
+	}
+
+	if (memcount) {
+		setprop(fdt, "/memory", "reg", mem_reg_property,
+			4 * memcount * memsize);
+	}
+
+	return fdt_pack(fdt);
+}
diff --git a/arch/arm/boot/compressed/big-endian.S b/arch/arm/boot/compressed/big-endian.S
new file mode 100644
index 000000000..0e092c36d
--- /dev/null
+++ b/arch/arm/boot/compressed/big-endian.S
@@ -0,0 +1,14 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ *  linux/arch/arm/boot/compressed/big-endian.S
+ *
+ *  Switch CPU into big endian mode.
+ *  Author: Nicolas Pitre
+ */
+
+	.section ".start", "ax"
+
+	mrc	p15, 0, r0, c1, c0, 0	@ read control reg
+	orr	r0, r0, #(1 << 7)	@ enable big endian mode
+	mcr	p15, 0, r0, c1, c0, 0	@ write control reg
+
diff --git a/arch/arm/boot/compressed/bswapsdi2.S b/arch/arm/boot/compressed/bswapsdi2.S
new file mode 100644
index 000000000..b2156b378
--- /dev/null
+++ b/arch/arm/boot/compressed/bswapsdi2.S
@@ -0,0 +1,3 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* For __bswapsi2, __bswapdi2 */
+#include "../../lib/bswapsdi2.S"
diff --git a/arch/arm/boot/compressed/debug.S b/arch/arm/boot/compressed/debug.S
new file mode 100644
index 000000000..fac40a717
--- /dev/null
+++ b/arch/arm/boot/compressed/debug.S
@@ -0,0 +1,48 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+#ifndef CONFIG_DEBUG_SEMIHOSTING
+
+#include CONFIG_DEBUG_LL_INCLUDE
+
+ENTRY(putc)
+	addruart r1, r2, r3
+#ifdef CONFIG_DEBUG_UART_FLOW_CONTROL
+	waituartcts r3, r1
+#endif
+	waituarttxrdy r3, r1
+	senduart r0, r1
+	busyuart r3, r1
+	mov	 pc, lr
+ENDPROC(putc)
+
+#else
+
+ENTRY(putc)
+	adr	r1, 1f
+	ldmia	r1, {r2, r3}
+	add	r2, r2, r1
+	ldr	r1, [r2, r3]
+	strb	r0, [r1]
+	mov	r0, #0x03		@ SYS_WRITEC
+   ARM(	svc	#0x123456	)
+#ifdef CONFIG_CPU_V7M
+ THUMB(	bkpt	#0xab		)
+#else
+ THUMB(	svc	#0xab		)
+#endif
+	mov	pc, lr
+	.align	2
+1:	.word	_GLOBAL_OFFSET_TABLE_ - .
+	.word	semi_writec_buf(GOT)
+ENDPROC(putc)
+
+	.bss
+	.global	semi_writec_buf
+	.type   semi_writec_buf, %object
+semi_writec_buf:
+	.space	4
+	.size	semi_writec_buf, 4
+
+#endif
diff --git a/arch/arm/boot/compressed/decompress.c b/arch/arm/boot/compressed/decompress.c
new file mode 100644
index 000000000..066985139
--- /dev/null
+++ b/arch/arm/boot/compressed/decompress.c
@@ -0,0 +1,66 @@
+// SPDX-License-Identifier: GPL-2.0
+#define _LINUX_STRING_H_
+
+#include <linux/compiler.h>	/* for inline */
+#include <linux/types.h>	/* for size_t */
+#include <linux/stddef.h>	/* for NULL */
+#include <linux/linkage.h>
+#include <asm/string.h>
+#include "misc.h"
+
+#define STATIC static
+#define STATIC_RW_DATA	/* non-static please */
+
+/* Diagnostic functions */
+#ifdef DEBUG
+#  define Assert(cond,msg) {if(!(cond)) error(msg);}
+#  define Trace(x) fprintf x
+#  define Tracev(x) {if (verbose) fprintf x ;}
+#  define Tracevv(x) {if (verbose>1) fprintf x ;}
+#  define Tracec(c,x) {if (verbose && (c)) fprintf x ;}
+#  define Tracecv(c,x) {if (verbose>1 && (c)) fprintf x ;}
+#else
+#  define Assert(cond,msg)
+#  define Trace(x)
+#  define Tracev(x)
+#  define Tracevv(x)
+#  define Tracec(c,x)
+#  define Tracecv(c,x)
+#endif
+
+/* Not needed, but used in some headers pulled in by decompressors */
+extern char * strstr(const char * s1, const char *s2);
+extern size_t strlen(const char *s);
+extern int strcmp(const char *cs, const char *ct);
+extern int memcmp(const void *cs, const void *ct, size_t count);
+extern char * strchrnul(const char *, int);
+
+#ifdef CONFIG_KERNEL_GZIP
+#include "../../../../lib/decompress_inflate.c"
+#endif
+
+#ifdef CONFIG_KERNEL_LZO
+#include "../../../../lib/decompress_unlzo.c"
+#endif
+
+#ifdef CONFIG_KERNEL_LZMA
+#include "../../../../lib/decompress_unlzma.c"
+#endif
+
+#ifdef CONFIG_KERNEL_XZ
+/* Prevent KASAN override of string helpers in decompressor */
+#undef memmove
+#define memmove memmove
+#undef memcpy
+#define memcpy memcpy
+#include "../../../../lib/decompress_unxz.c"
+#endif
+
+#ifdef CONFIG_KERNEL_LZ4
+#include "../../../../lib/decompress_unlz4.c"
+#endif
+
+int do_decompress(u8 *input, int len, u8 *output, void (*error)(char *x))
+{
+	return __decompress(input, len, NULL, NULL, output, 0, NULL, error);
+}
diff --git a/arch/arm/boot/compressed/efi-header.S b/arch/arm/boot/compressed/efi-header.S
new file mode 100644
index 000000000..65a3025c0
--- /dev/null
+++ b/arch/arm/boot/compressed/efi-header.S
@@ -0,0 +1,136 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (C) 2013-2017 Linaro Ltd
+ * Authors: Roy Franz <roy.franz@linaro.org>
+ *          Ard Biesheuvel <ard.biesheuvel@linaro.org>
+ */
+
+#include <linux/pe.h>
+#include <linux/sizes.h>
+
+		.macro	__nop
+ AR_CLASS(	mov	r0, r0		)
+  M_CLASS(	nop.w			)
+		.endm
+
+		.macro __initial_nops
+#ifdef CONFIG_EFI_STUB
+		@ This is a two-instruction NOP, which happens to bear the
+		@ PE/COFF signature "MZ" in the first two bytes, so the kernel
+		@ is accepted as an EFI binary. Booting via the UEFI stub
+		@ will not execute those instructions, but the ARM/Linux
+		@ boot protocol does, so we need some NOPs here.
+		.inst	IMAGE_DOS_SIGNATURE | (0xe225 << 16) @ eor r5, r5, 0x4d000
+		eor	r5, r5, 0x4d000			@ undo previous insn
+#else
+		__nop
+		__nop
+#endif
+		.endm
+
+		.macro	__EFI_HEADER
+#ifdef CONFIG_EFI_STUB
+		.set	start_offset, __efi_start - start
+		.org	start + 0x3c
+		@
+		@ The PE header can be anywhere in the file, but for
+		@ simplicity we keep it together with the MSDOS header
+		@ The offset to the PE/COFF header needs to be at offset
+		@ 0x3C in the MSDOS header.
+		@ The only 2 fields of the MSDOS header that are used are this
+		@ PE/COFF offset, and the "MZ" bytes at offset 0x0.
+		@
+		.long	pe_header - start		@ Offset to the PE header.
+
+pe_header:
+		.long	IMAGE_NT_SIGNATURE
+
+coff_header:
+		.short	IMAGE_FILE_MACHINE_THUMB	@ Machine
+		.short	section_count			@ NumberOfSections
+		.long	0 				@ TimeDateStamp
+		.long	0				@ PointerToSymbolTable
+		.long	0				@ NumberOfSymbols
+		.short	section_table - optional_header	@ SizeOfOptionalHeader
+		.short	IMAGE_FILE_32BIT_MACHINE | \
+			IMAGE_FILE_DEBUG_STRIPPED | \
+			IMAGE_FILE_EXECUTABLE_IMAGE | \
+			IMAGE_FILE_LINE_NUMS_STRIPPED	@ Characteristics
+
+#define __pecoff_code_size (__pecoff_data_start - __efi_start)
+
+optional_header:
+		.short	IMAGE_NT_OPTIONAL_HDR32_MAGIC	@ PE32 format
+		.byte	0x02				@ MajorLinkerVersion
+		.byte	0x14				@ MinorLinkerVersion
+		.long	__pecoff_code_size		@ SizeOfCode
+		.long	__pecoff_data_size		@ SizeOfInitializedData
+		.long	0				@ SizeOfUninitializedData
+		.long	efi_pe_entry - start		@ AddressOfEntryPoint
+		.long	start_offset			@ BaseOfCode
+		.long	__pecoff_data_start - start	@ BaseOfData
+
+extra_header_fields:
+		.long	0				@ ImageBase
+		.long	SZ_4K				@ SectionAlignment
+		.long	SZ_512				@ FileAlignment
+		.short	0				@ MajorOsVersion
+		.short	0				@ MinorOsVersion
+		.short	LINUX_EFISTUB_MAJOR_VERSION	@ MajorImageVersion
+		.short	LINUX_EFISTUB_MINOR_VERSION	@ MinorImageVersion
+		.short	0				@ MajorSubsystemVersion
+		.short	0				@ MinorSubsystemVersion
+		.long	0				@ Win32VersionValue
+
+		.long	__pecoff_end - start		@ SizeOfImage
+		.long	start_offset			@ SizeOfHeaders
+		.long	0				@ CheckSum
+		.short	IMAGE_SUBSYSTEM_EFI_APPLICATION	@ Subsystem
+		.short	0				@ DllCharacteristics
+		.long	0				@ SizeOfStackReserve
+		.long	0				@ SizeOfStackCommit
+		.long	0				@ SizeOfHeapReserve
+		.long	0				@ SizeOfHeapCommit
+		.long	0				@ LoaderFlags
+		.long	(section_table - .) / 8		@ NumberOfRvaAndSizes
+
+		.quad	0				@ ExportTable
+		.quad	0				@ ImportTable
+		.quad	0				@ ResourceTable
+		.quad	0				@ ExceptionTable
+		.quad	0				@ CertificationTable
+		.quad	0				@ BaseRelocationTable
+
+section_table:
+		.ascii	".text\0\0\0"
+		.long	__pecoff_code_size		@ VirtualSize
+		.long	__efi_start			@ VirtualAddress
+		.long	__pecoff_code_size		@ SizeOfRawData
+		.long	__efi_start			@ PointerToRawData
+		.long	0				@ PointerToRelocations
+		.long	0				@ PointerToLineNumbers
+		.short	0				@ NumberOfRelocations
+		.short	0				@ NumberOfLineNumbers
+		.long	IMAGE_SCN_CNT_CODE | \
+			IMAGE_SCN_MEM_READ | \
+			IMAGE_SCN_MEM_EXECUTE		@ Characteristics
+
+		.ascii	".data\0\0\0"
+		.long	__pecoff_data_size		@ VirtualSize
+		.long	__pecoff_data_start - start	@ VirtualAddress
+		.long	__pecoff_data_rawsize		@ SizeOfRawData
+		.long	__pecoff_data_start - start	@ PointerToRawData
+		.long	0				@ PointerToRelocations
+		.long	0				@ PointerToLineNumbers
+		.short	0				@ NumberOfRelocations
+		.short	0				@ NumberOfLineNumbers
+		.long	IMAGE_SCN_CNT_INITIALIZED_DATA | \
+			IMAGE_SCN_MEM_READ | \
+			IMAGE_SCN_MEM_WRITE		@ Characteristics
+
+		.set	section_count, (. - section_table) / 40
+
+		.align	12
+__efi_start:
+#endif
+		.endm
diff --git a/arch/arm/boot/compressed/fdt.c b/arch/arm/boot/compressed/fdt.c
new file mode 100644
index 000000000..f8ea7a201
--- /dev/null
+++ b/arch/arm/boot/compressed/fdt.c
@@ -0,0 +1,2 @@
+// SPDX-License-Identifier: GPL-2.0-only
+#include "../../../../lib/fdt.c"
diff --git a/arch/arm/boot/compressed/fdt_check_mem_start.c b/arch/arm/boot/compressed/fdt_check_mem_start.c
new file mode 100644
index 000000000..aa856567f
--- /dev/null
+++ b/arch/arm/boot/compressed/fdt_check_mem_start.c
@@ -0,0 +1,168 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/kernel.h>
+#include <linux/libfdt.h>
+#include <linux/sizes.h>
+#include "misc.h"
+
+static const void *get_prop(const void *fdt, const char *node_path,
+			    const char *property, int minlen)
+{
+	const void *prop;
+	int offset, len;
+
+	offset = fdt_path_offset(fdt, node_path);
+	if (offset < 0)
+		return NULL;
+
+	prop = fdt_getprop(fdt, offset, property, &len);
+	if (!prop || len < minlen)
+		return NULL;
+
+	return prop;
+}
+
+static uint32_t get_cells(const void *fdt, const char *name)
+{
+	const fdt32_t *prop = get_prop(fdt, "/", name, sizeof(fdt32_t));
+
+	if (!prop) {
+		/* default */
+		return 1;
+	}
+
+	return fdt32_ld(prop);
+}
+
+static uint64_t get_val(const fdt32_t *cells, uint32_t ncells)
+{
+	uint64_t r;
+
+	r = fdt32_ld(cells);
+	if (ncells > 1)
+		r = (r << 32) | fdt32_ld(cells + 1);
+
+	return r;
+}
+
+/*
+ * Check the start of physical memory
+ *
+ * Traditionally, the start address of physical memory is obtained by masking
+ * the program counter.  However, this does require that this address is a
+ * multiple of 128 MiB, precluding booting Linux on platforms where this
+ * requirement is not fulfilled.
+ * Hence validate the calculated address against the memory information in the
+ * DTB, and, if out-of-range, replace it by the real start address.
+ * To preserve backwards compatibility (systems reserving a block of memory
+ * at the start of physical memory, kdump, ...), the traditional method is
+ * used if it yields a valid address, unless the "linux,usable-memory-range"
+ * property is present.
+ *
+ * Return value: start address of physical memory to use
+ */
+uint32_t fdt_check_mem_start(uint32_t mem_start, const void *fdt)
+{
+	uint32_t addr_cells, size_cells, usable_base, base;
+	uint32_t fdt_mem_start = 0xffffffff;
+	const fdt32_t *usable, *reg, *endp;
+	uint64_t size, usable_end, end;
+	const char *type;
+	int offset, len;
+
+	if (!fdt)
+		return mem_start;
+
+	if (fdt_magic(fdt) != FDT_MAGIC)
+		return mem_start;
+
+	/* There may be multiple cells on LPAE platforms */
+	addr_cells = get_cells(fdt, "#address-cells");
+	size_cells = get_cells(fdt, "#size-cells");
+	if (addr_cells > 2 || size_cells > 2)
+		return mem_start;
+
+	/*
+	 * Usable memory in case of a crash dump kernel
+	 * This property describes a limitation: memory within this range is
+	 * only valid when also described through another mechanism
+	 */
+	usable = get_prop(fdt, "/chosen", "linux,usable-memory-range",
+			  (addr_cells + size_cells) * sizeof(fdt32_t));
+	if (usable) {
+		size = get_val(usable + addr_cells, size_cells);
+		if (!size)
+			return mem_start;
+
+		if (addr_cells > 1 && fdt32_ld(usable)) {
+			/* Outside 32-bit address space */
+			return mem_start;
+		}
+
+		usable_base = fdt32_ld(usable + addr_cells - 1);
+		usable_end = usable_base + size;
+	}
+
+	/* Walk all memory nodes and regions */
+	for (offset = fdt_next_node(fdt, -1, NULL); offset >= 0;
+	     offset = fdt_next_node(fdt, offset, NULL)) {
+		type = fdt_getprop(fdt, offset, "device_type", NULL);
+		if (!type || strcmp(type, "memory"))
+			continue;
+
+		reg = fdt_getprop(fdt, offset, "linux,usable-memory", &len);
+		if (!reg)
+			reg = fdt_getprop(fdt, offset, "reg", &len);
+		if (!reg)
+			continue;
+
+		for (endp = reg + (len / sizeof(fdt32_t));
+		     endp - reg >= addr_cells + size_cells;
+		     reg += addr_cells + size_cells) {
+			size = get_val(reg + addr_cells, size_cells);
+			if (!size)
+				continue;
+
+			if (addr_cells > 1 && fdt32_ld(reg)) {
+				/* Outside 32-bit address space, skipping */
+				continue;
+			}
+
+			base = fdt32_ld(reg + addr_cells - 1);
+			end = base + size;
+			if (usable) {
+				/*
+				 * Clip to usable range, which takes precedence
+				 * over mem_start
+				 */
+				if (base < usable_base)
+					base = usable_base;
+
+				if (end > usable_end)
+					end = usable_end;
+
+				if (end <= base)
+					continue;
+			} else if (mem_start >= base && mem_start < end) {
+				/* Calculated address is valid, use it */
+				return mem_start;
+			}
+
+			if (base < fdt_mem_start)
+				fdt_mem_start = base;
+		}
+	}
+
+	if (fdt_mem_start == 0xffffffff) {
+		/* No usable memory found, falling back to default */
+		return mem_start;
+	}
+
+	/*
+	 * The calculated address is not usable, or was overridden by the
+	 * "linux,usable-memory-range" property.
+	 * Use the lowest usable physical memory address from the DTB instead,
+	 * and make sure this is a multiple of 2 MiB for phys/virt patching.
+	 */
+	return round_up(fdt_mem_start, SZ_2M);
+}
diff --git a/arch/arm/boot/compressed/fdt_ro.c b/arch/arm/boot/compressed/fdt_ro.c
new file mode 100644
index 000000000..93970a4ad
--- /dev/null
+++ b/arch/arm/boot/compressed/fdt_ro.c
@@ -0,0 +1,2 @@
+// SPDX-License-Identifier: GPL-2.0-only
+#include "../../../../lib/fdt_ro.c"
diff --git a/arch/arm/boot/compressed/fdt_rw.c b/arch/arm/boot/compressed/fdt_rw.c
new file mode 100644
index 000000000..f7c6b8b7e
--- /dev/null
+++ b/arch/arm/boot/compressed/fdt_rw.c
@@ -0,0 +1,2 @@
+// SPDX-License-Identifier: GPL-2.0-only
+#include "../../../../lib/fdt_rw.c"
diff --git a/arch/arm/boot/compressed/fdt_wip.c b/arch/arm/boot/compressed/fdt_wip.c
new file mode 100644
index 000000000..048d2c7a0
--- /dev/null
+++ b/arch/arm/boot/compressed/fdt_wip.c
@@ -0,0 +1,2 @@
+// SPDX-License-Identifier: GPL-2.0-only
+#include "../../../../lib/fdt_wip.c"
diff --git a/arch/arm/boot/compressed/font.c b/arch/arm/boot/compressed/font.c
new file mode 100644
index 000000000..46a677649
--- /dev/null
+++ b/arch/arm/boot/compressed/font.c
@@ -0,0 +1,2 @@
+// SPDX-License-Identifier: GPL-2.0-only
+#include "../../../../lib/fonts/font_acorn_8x8.c"
diff --git a/arch/arm/boot/compressed/head-sa1100.S b/arch/arm/boot/compressed/head-sa1100.S
new file mode 100644
index 000000000..23eae1a65
--- /dev/null
+++ b/arch/arm/boot/compressed/head-sa1100.S
@@ -0,0 +1,45 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* 
+ * linux/arch/arm/boot/compressed/head-sa1100.S
+ * 
+ * Copyright (C) 1999 Nicolas Pitre <nico@fluxnic.net>
+ * 
+ * SA1100 specific tweaks.  This is merged into head.S by the linker.
+ *
+ */
+
+#include <linux/linkage.h>
+#include <asm/mach-types.h>
+
+		.section        ".start", "ax"
+		.arch	armv4
+
+__SA1100_start:
+
+		@ Preserve r8/r7 i.e. kernel entry values
+#ifdef CONFIG_SA1100_COLLIE
+		mov	r7, #MACH_TYPE_COLLIE
+#endif
+		mrc	p15, 0, r0, c1, c0, 0	@ read control reg
+		ands	r0, r0, #0x0d
+		beq	99f
+
+		@ Data cache might be active.
+		@ Be sure to flush kernel binary out of the cache,
+		@ whatever state it is, before it is turned off.
+		@ This is done by fetching through currently executed
+		@ memory to be sure we hit the same cache.
+		bic	r2, pc, #0x1f
+		add	r3, r2, #0x4000		@ 16 kb is quite enough...
+1:		ldr	r0, [r2], #32
+		teq	r2, r3
+		bne	1b
+		mcr	p15, 0, r0, c7, c10, 4	@ drain WB
+		mcr	p15, 0, r0, c7, c7, 0	@ flush I & D caches
+
+		@ disabling MMU and caches
+		mrc	p15, 0, r0, c1, c0, 0	@ read control reg
+		bic	r0, r0, #0x0d		@ clear WB, DC, MMU
+		bic	r0, r0, #0x1000		@ clear Icache
+		mcr	p15, 0, r0, c1, c0, 0
+99:
diff --git a/arch/arm/boot/compressed/head-sharpsl.S b/arch/arm/boot/compressed/head-sharpsl.S
new file mode 100644
index 000000000..992e78450
--- /dev/null
+++ b/arch/arm/boot/compressed/head-sharpsl.S
@@ -0,0 +1,151 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * linux/arch/arm/boot/compressed/head-sharpsl.S
+ *
+ * Copyright (C) 2004-2005 Richard Purdie <rpurdie@rpsys.net>
+ *
+ * Sharp's bootloader doesn't pass any kind of machine ID
+ * so we have to figure out the machine for ourselves...
+ *
+ * Support for Poodle, Corgi (SL-C700), Shepherd (SL-C750)
+ * Husky (SL-C760), Tosa (SL-C6000), Spitz (SL-C3000),
+ * Akita (SL-C1000) and Borzoi (SL-C3100).
+ *
+ */
+
+#include <linux/linkage.h>
+#include <asm/mach-types.h>
+
+#ifndef CONFIG_PXA_SHARPSL
+#error What am I doing here...
+#endif
+
+		.section        ".start", "ax"
+
+__SharpSL_start:
+
+/* Check for TC6393 - if found we have a Tosa */
+	ldr	r7, .TOSAID
+	mov	r1, #0x10000000		@ Base address of TC6393 chip
+	mov 	r6, #0x03
+	ldrh	r3, [r1, #8]		@ Load TC6393XB Revison: This is 0x0003
+	cmp	r6, r3
+	beq	.SHARPEND		@ Success -> tosa
+
+/* Check for pxa270 - if found, branch */
+	mrc p15, 0, r4, c0, c0		@ Get Processor ID
+	and	r4, r4, #0xffffff00
+	ldr	r3, .PXA270ID
+	cmp	r4, r3
+	beq	.PXA270
+
+/* Check for w100 - if not found we have a Poodle */
+	ldr	r1, .W100ADDR		@ Base address of w100 chip + regs offset
+
+	mov r6, #0x31			@ Load Magic Init value
+	str	r6, [r1, #0x280]	@ to SCRATCH_UMSK
+	mov r5, #0x3000
+.W100LOOP:
+	subs r5, r5, #1
+	bne .W100LOOP
+	mov r6, #0x30			@ Load 2nd Magic Init value
+	str	r6, [r1, #0x280]	@ to SCRATCH_UMSK
+
+	ldr	r6, [r1, #0]		@ Load Chip ID
+	ldr	r3, .W100ID
+	ldr	r7, .POODLEID
+	cmp	r6, r3
+	bne	.SHARPEND			@ We have no w100 - Poodle
+
+/* Check for pxa250 - if found we have a Corgi */
+	ldr	r7, .CORGIID
+	ldr	r3, .PXA255ID
+	cmp	r4, r3
+	blo	.SHARPEND			@ We have a PXA250 - Corgi
+
+/* Check for 64MiB flash - if found we have a Shepherd */
+	bl	get_flash_ids
+	ldr	r7, .SHEPHERDID
+	cmp	r3, #0x76			@ 64MiB flash
+	beq	.SHARPEND			@ We have Shepherd
+
+/* Must be a Husky */
+	ldr	r7, .HUSKYID		@ Must be Husky
+	b .SHARPEND
+
+.PXA270:
+/* Check for 16MiB flash - if found we have Spitz */
+	bl	get_flash_ids
+	ldr	r7, .SPITZID
+	cmp	r3, #0x73			@ 16MiB flash
+	beq	.SHARPEND			@ We have Spitz
+
+/* Check for a second SCOOP chip - if found we have Borzoi */
+	ldr	r1, .SCOOP2ADDR
+	ldr	r7, .BORZOIID
+	mov 	r6, #0x0140
+	strh	r6, [r1]
+	ldrh	r6, [r1]
+	cmp	r6, #0x0140
+	beq	.SHARPEND			@ We have Borzoi
+
+/* Must be Akita */
+	ldr	r7, .AKITAID
+	b	.SHARPEND			@ We have Borzoi
+
+.PXA255ID:
+	.word	0x69052d00		@ PXA255 Processor ID
+.PXA270ID:
+	.word	0x69054100		@ PXA270 Processor ID
+.W100ID:
+	.word	0x57411002		@ w100 Chip ID
+.W100ADDR:
+	.word 	0x08010000		@ w100 Chip ID Reg Address
+.SCOOP2ADDR:
+	.word	0x08800040
+.POODLEID:
+	.word	MACH_TYPE_POODLE
+.CORGIID:
+	.word	MACH_TYPE_CORGI
+.SHEPHERDID:
+	.word	MACH_TYPE_SHEPHERD
+.HUSKYID:
+	.word	MACH_TYPE_HUSKY
+.TOSAID:
+	.word	MACH_TYPE_TOSA
+.SPITZID:
+	.word	MACH_TYPE_SPITZ
+.AKITAID:
+	.word	MACH_TYPE_AKITA
+.BORZOIID:
+	.word	MACH_TYPE_BORZOI
+
+/*
+ * Return: r2 - NAND Manufacturer ID
+ *         r3 - NAND Chip ID
+ * Corrupts: r1
+ */
+get_flash_ids:
+	mov	r1, #0x0c000000		@ Base address of NAND chip
+	ldrb	r3, [r1, #24]		@ Load FLASHCTL
+	bic	r3, r3, #0x11		@ SET NCE
+	orr	r3, r3, #0x0a		@ SET CLR + FLWP
+	strb	r3, [r1, #24]		@ Save to FLASHCTL
+	mov 	r2, #0x90		@ Command "readid"
+	strb	r2, [r1, #20]		@ Save to FLASHIO
+	bic	r3, r3, #2		@ CLR CLE
+	orr	r3, r3, #4		@ SET ALE
+	strb	r3, [r1, #24]		@ Save to FLASHCTL
+	mov	r2, #0			@ Address 0x00
+	strb	r2, [r1, #20]		@ Save to FLASHIO
+	bic	r3, r3, #4		@ CLR ALE
+	strb	r3, [r1, #24]		@ Save to FLASHCTL
+.fids1:
+	ldrb	r3, [r1, #24]		@ Load FLASHCTL
+	tst	r3, #32			@ Is chip ready?
+	beq	.fids1
+	ldrb	r2, [r1, #20]		@ NAND Manufacturer ID
+	ldrb	r3, [r1, #20]		@ NAND Chip ID
+	mov	pc, lr
+
+.SHARPEND:
diff --git a/arch/arm/boot/compressed/head-xscale.S b/arch/arm/boot/compressed/head-xscale.S
new file mode 100644
index 000000000..20fa44d59
--- /dev/null
+++ b/arch/arm/boot/compressed/head-xscale.S
@@ -0,0 +1,35 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * linux/arch/arm/boot/compressed/head-xscale.S
+ *
+ * XScale specific tweaks.  This is merged into head.S by the linker.
+ *
+ */
+
+#include <linux/linkage.h>
+
+		.section        ".start", "ax"
+
+__XScale_start:
+
+		@ Preserve r8/r7 i.e. kernel entry values
+
+		@ Data cache might be active.
+		@ Be sure to flush kernel binary out of the cache,
+		@ whatever state it is, before it is turned off.
+		@ This is done by fetching through currently executed
+		@ memory to be sure we hit the same cache.
+		bic	r2, pc, #0x1f
+		add	r3, r2, #0x10000	@ 64 kb is quite enough...
+1:		ldr	r0, [r2], #32
+		teq	r2, r3
+		bne	1b
+		mcr	p15, 0, r0, c7, c10, 4	@ drain WB
+		mcr	p15, 0, r0, c7, c7, 0	@ flush I & D caches
+
+		@ disabling MMU and caches
+		mrc	p15, 0, r0, c1, c0, 0	@ read control reg
+		bic	r0, r0, #0x05		@ clear DC, MMU
+		bic	r0, r0, #0x1000		@ clear Icache
+		mcr	p15, 0, r0, c1, c0, 0
+
diff --git a/arch/arm/boot/compressed/head.S b/arch/arm/boot/compressed/head.S
new file mode 100644
index 000000000..9f406e9c0
--- /dev/null
+++ b/arch/arm/boot/compressed/head.S
@@ -0,0 +1,1531 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ *  linux/arch/arm/boot/compressed/head.S
+ *
+ *  Copyright (C) 1996-2002 Russell King
+ *  Copyright (C) 2004 Hyok S. Choi (MPU support)
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include <asm/v7m.h>
+
+#include "efi-header.S"
+
+#ifdef __ARMEB__
+#define OF_DT_MAGIC 0xd00dfeed
+#else
+#define OF_DT_MAGIC 0xedfe0dd0
+#endif
+
+ AR_CLASS(	.arch	armv7-a	)
+ M_CLASS(	.arch	armv7-m	)
+
+/*
+ * Debugging stuff
+ *
+ * Note that these macros must not contain any code which is not
+ * 100% relocatable.  Any attempt to do so will result in a crash.
+ * Please select one of the following when turning on debugging.
+ */
+#ifdef DEBUG
+
+#if defined(CONFIG_DEBUG_ICEDCC)
+
+#if defined(CONFIG_CPU_V6) || defined(CONFIG_CPU_V6K) || defined(CONFIG_CPU_V7)
+		.macro	loadsp, rb, tmp1, tmp2
+		.endm
+		.macro	writeb, ch, rb, tmp
+		mcr	p14, 0, \ch, c0, c5, 0
+		.endm
+#elif defined(CONFIG_CPU_XSCALE)
+		.macro	loadsp, rb, tmp1, tmp2
+		.endm
+		.macro	writeb, ch, rb, tmp
+		mcr	p14, 0, \ch, c8, c0, 0
+		.endm
+#else
+		.macro	loadsp, rb, tmp1, tmp2
+		.endm
+		.macro	writeb, ch, rb, tmp
+		mcr	p14, 0, \ch, c1, c0, 0
+		.endm
+#endif
+
+#else
+
+#include CONFIG_DEBUG_LL_INCLUDE
+
+		.macro	writeb,	ch, rb, tmp
+#ifdef CONFIG_DEBUG_UART_FLOW_CONTROL
+		waituartcts \tmp, \rb
+#endif
+		waituarttxrdy \tmp, \rb
+		senduart \ch, \rb
+		busyuart \tmp, \rb
+		.endm
+
+#if defined(CONFIG_ARCH_SA1100)
+		.macro	loadsp, rb, tmp1, tmp2
+		mov	\rb, #0x80000000	@ physical base address
+		add	\rb, \rb, #0x00010000	@ Ser1
+		.endm
+#else
+		.macro	loadsp,	rb, tmp1, tmp2
+		addruart \rb, \tmp1, \tmp2
+		.endm
+#endif
+#endif
+#endif
+
+		.macro	kputc,val
+		mov	r0, \val
+		bl	putc
+		.endm
+
+		.macro	kphex,val,len
+		mov	r0, \val
+		mov	r1, #\len
+		bl	phex
+		.endm
+
+		/*
+		 * Debug kernel copy by printing the memory addresses involved
+		 */
+		.macro dbgkc, begin, end, cbegin, cend
+#ifdef DEBUG
+		kputc   #'C'
+		kputc   #':'
+		kputc   #'0'
+		kputc   #'x'
+		kphex   \begin, 8	/* Start of compressed kernel */
+		kputc	#'-'
+		kputc	#'0'
+		kputc	#'x'
+		kphex	\end, 8		/* End of compressed kernel */
+		kputc	#'-'
+		kputc	#'>'
+		kputc   #'0'
+		kputc   #'x'
+		kphex   \cbegin, 8	/* Start of kernel copy */
+		kputc	#'-'
+		kputc	#'0'
+		kputc	#'x'
+		kphex	\cend, 8	/* End of kernel copy */
+		kputc	#'\n'
+#endif
+		.endm
+
+		/*
+		 * Debug print of the final appended DTB location
+		 */
+		.macro dbgadtb, begin, size
+#ifdef DEBUG
+		kputc   #'D'
+		kputc   #'T'
+		kputc   #'B'
+		kputc   #':'
+		kputc   #'0'
+		kputc   #'x'
+		kphex   \begin, 8	/* Start of appended DTB */
+		kputc	#' '
+		kputc	#'('
+		kputc	#'0'
+		kputc	#'x'
+		kphex	\size, 8	/* Size of appended DTB */
+		kputc	#')'
+		kputc	#'\n'
+#endif
+		.endm
+
+		.macro	enable_cp15_barriers, reg
+		mrc	p15, 0, \reg, c1, c0, 0	@ read SCTLR
+		tst	\reg, #(1 << 5)		@ CP15BEN bit set?
+		bne	.L_\@
+		orr	\reg, \reg, #(1 << 5)	@ CP15 barrier instructions
+		mcr	p15, 0, \reg, c1, c0, 0	@ write SCTLR
+ ARM(		.inst   0xf57ff06f		@ v7+ isb	)
+ THUMB(		isb						)
+.L_\@:
+		.endm
+
+		/*
+		 * The kernel build system appends the size of the
+		 * decompressed kernel at the end of the compressed data
+		 * in little-endian form.
+		 */
+		.macro	get_inflated_image_size, res:req, tmp1:req, tmp2:req
+		adr	\res, .Linflated_image_size_offset
+		ldr	\tmp1, [\res]
+		add	\tmp1, \tmp1, \res	@ address of inflated image size
+
+		ldrb	\res, [\tmp1]		@ get_unaligned_le32
+		ldrb	\tmp2, [\tmp1, #1]
+		orr	\res, \res, \tmp2, lsl #8
+		ldrb	\tmp2, [\tmp1, #2]
+		ldrb	\tmp1, [\tmp1, #3]
+		orr	\res, \res, \tmp2, lsl #16
+		orr	\res, \res, \tmp1, lsl #24
+		.endm
+
+		.macro	be32tocpu, val, tmp
+#ifndef __ARMEB__
+		/* convert to little endian */
+		rev_l	\val, \tmp
+#endif
+		.endm
+
+		.section ".start", "ax"
+/*
+ * sort out different calling conventions
+ */
+		.align
+		/*
+		 * Always enter in ARM state for CPUs that support the ARM ISA.
+		 * As of today (2014) that's exactly the members of the A and R
+		 * classes.
+		 */
+ AR_CLASS(	.arm	)
+start:
+		.type	start,#function
+		/*
+		 * These 7 nops along with the 1 nop immediately below for
+		 * !THUMB2 form 8 nops that make the compressed kernel bootable
+		 * on legacy ARM systems that were assuming the kernel in a.out
+		 * binary format. The boot loaders on these systems would
+		 * jump 32 bytes into the image to skip the a.out header.
+		 * with these 8 nops filling exactly 32 bytes, things still
+		 * work as expected on these legacy systems. Thumb2 mode keeps
+		 * 7 of the nops as it turns out that some boot loaders
+		 * were patching the initial instructions of the kernel, i.e
+		 * had started to exploit this "patch area".
+		 */
+		__initial_nops
+		.rept	5
+		__nop
+		.endr
+#ifndef CONFIG_THUMB2_KERNEL
+		__nop
+#else
+ AR_CLASS(	sub	pc, pc, #3	)	@ A/R: switch to Thumb2 mode
+  M_CLASS(	nop.w			)	@ M: already in Thumb2 mode
+		.thumb
+#endif
+		W(b)	1f
+
+		.word	_magic_sig	@ Magic numbers to help the loader
+		.word	_magic_start	@ absolute load/run zImage address
+		.word	_magic_end	@ zImage end address
+		.word	0x04030201	@ endianness flag
+		.word	0x45454545	@ another magic number to indicate
+		.word	_magic_table	@ additional data table
+
+		__EFI_HEADER
+1:
+ ARM_BE8(	setend	be		)	@ go BE8 if compiled for BE8
+ AR_CLASS(	mrs	r9, cpsr	)
+#ifdef CONFIG_ARM_VIRT_EXT
+		bl	__hyp_stub_install	@ get into SVC mode, reversibly
+#endif
+		mov	r7, r1			@ save architecture ID
+		mov	r8, r2			@ save atags pointer
+
+#ifndef CONFIG_CPU_V7M
+		/*
+		 * Booting from Angel - need to enter SVC mode and disable
+		 * FIQs/IRQs (numeric definitions from angel arm.h source).
+		 * We only do this if we were in user mode on entry.
+		 */
+		mrs	r2, cpsr		@ get current mode
+		tst	r2, #3			@ not user?
+		bne	not_angel
+		mov	r0, #0x17		@ angel_SWIreason_EnterSVC
+ ARM(		swi	0x123456	)	@ angel_SWI_ARM
+ THUMB(		svc	0xab		)	@ angel_SWI_THUMB
+not_angel:
+		safe_svcmode_maskall r0
+		msr	spsr_cxsf, r9		@ Save the CPU boot mode in
+						@ SPSR
+#endif
+		/*
+		 * Note that some cache flushing and other stuff may
+		 * be needed here - is there an Angel SWI call for this?
+		 */
+
+		/*
+		 * some architecture specific code can be inserted
+		 * by the linker here, but it should preserve r7, r8, and r9.
+		 */
+
+		.text
+
+#ifdef CONFIG_AUTO_ZRELADDR
+		/*
+		 * Find the start of physical memory.  As we are executing
+		 * without the MMU on, we are in the physical address space.
+		 * We just need to get rid of any offset by aligning the
+		 * address.
+		 *
+		 * This alignment is a balance between the requirements of
+		 * different platforms - we have chosen 128MB to allow
+		 * platforms which align the start of their physical memory
+		 * to 128MB to use this feature, while allowing the zImage
+		 * to be placed within the first 128MB of memory on other
+		 * platforms.  Increasing the alignment means we place
+		 * stricter alignment requirements on the start of physical
+		 * memory, but relaxing it means that we break people who
+		 * are already placing their zImage in (eg) the top 64MB
+		 * of this range.
+		 */
+		mov	r0, pc
+		and	r0, r0, #0xf8000000
+#ifdef CONFIG_USE_OF
+		adr	r1, LC1
+#ifdef CONFIG_ARM_APPENDED_DTB
+		/*
+		 * Look for an appended DTB.  If found, we cannot use it to
+		 * validate the calculated start of physical memory, as its
+		 * memory nodes may need to be augmented by ATAGS stored at
+		 * an offset from the same start of physical memory.
+		 */
+		ldr	r2, [r1, #4]	@ get &_edata
+		add	r2, r2, r1	@ relocate it
+		ldr	r2, [r2]	@ get DTB signature
+		ldr	r3, =OF_DT_MAGIC
+		cmp	r2, r3		@ do we have a DTB there?
+		beq	1f		@ if yes, skip validation
+#endif /* CONFIG_ARM_APPENDED_DTB */
+
+		/*
+		 * Make sure we have some stack before calling C code.
+		 * No GOT fixup has occurred yet, but none of the code we're
+		 * about to call uses any global variables.
+		 */
+		ldr	sp, [r1]	@ get stack location
+		add	sp, sp, r1	@ apply relocation
+
+		/* Validate calculated start against passed DTB */
+		mov	r1, r8
+		bl	fdt_check_mem_start
+1:
+#endif /* CONFIG_USE_OF */
+		/* Determine final kernel image address. */
+		add	r4, r0, #TEXT_OFFSET
+#else
+		ldr	r4, =zreladdr
+#endif
+
+		/*
+		 * Set up a page table only if it won't overwrite ourself.
+		 * That means r4 < pc || r4 - 16k page directory > &_end.
+		 * Given that r4 > &_end is most unfrequent, we add a rough
+		 * additional 1MB of room for a possible appended DTB.
+		 */
+		mov	r0, pc
+		cmp	r0, r4
+		ldrcc	r0, .Lheadroom
+		addcc	r0, r0, pc
+		cmpcc	r4, r0
+		orrcc	r4, r4, #1		@ remember we skipped cache_on
+		blcs	cache_on
+
+restart:	adr	r0, LC1
+		ldr	sp, [r0]
+		ldr	r6, [r0, #4]
+		add	sp, sp, r0
+		add	r6, r6, r0
+
+		get_inflated_image_size	r9, r10, lr
+
+#ifndef CONFIG_ZBOOT_ROM
+		/* malloc space is above the relocated stack (64k max) */
+		add	r10, sp, #MALLOC_SIZE
+#else
+		/*
+		 * With ZBOOT_ROM the bss/stack is non relocatable,
+		 * but someone could still run this code from RAM,
+		 * in which case our reference is _edata.
+		 */
+		mov	r10, r6
+#endif
+
+		mov	r5, #0			@ init dtb size to 0
+#ifdef CONFIG_ARM_APPENDED_DTB
+/*
+ *   r4  = final kernel address (possibly with LSB set)
+ *   r5  = appended dtb size (still unknown)
+ *   r6  = _edata
+ *   r7  = architecture ID
+ *   r8  = atags/device tree pointer
+ *   r9  = size of decompressed image
+ *   r10 = end of this image, including  bss/stack/malloc space if non XIP
+ *   sp  = stack pointer
+ *
+ * if there are device trees (dtb) appended to zImage, advance r10 so that the
+ * dtb data will get relocated along with the kernel if necessary.
+ */
+
+		ldr	lr, [r6, #0]
+		ldr	r1, =OF_DT_MAGIC
+		cmp	lr, r1
+		bne	dtb_check_done		@ not found
+
+#ifdef CONFIG_ARM_ATAG_DTB_COMPAT
+		/*
+		 * OK... Let's do some funky business here.
+		 * If we do have a DTB appended to zImage, and we do have
+		 * an ATAG list around, we want the later to be translated
+		 * and folded into the former here. No GOT fixup has occurred
+		 * yet, but none of the code we're about to call uses any
+		 * global variable.
+		*/
+
+		/* Get the initial DTB size */
+		ldr	r5, [r6, #4]
+		be32tocpu r5, r1
+		dbgadtb	r6, r5
+		/* 50% DTB growth should be good enough */
+		add	r5, r5, r5, lsr #1
+		/* preserve 64-bit alignment */
+		add	r5, r5, #7
+		bic	r5, r5, #7
+		/* clamp to 32KB min and 1MB max */
+		cmp	r5, #(1 << 15)
+		movlo	r5, #(1 << 15)
+		cmp	r5, #(1 << 20)
+		movhi	r5, #(1 << 20)
+		/* temporarily relocate the stack past the DTB work space */
+		add	sp, sp, r5
+
+		mov	r0, r8
+		mov	r1, r6
+		mov	r2, r5
+		bl	atags_to_fdt
+
+		/*
+		 * If returned value is 1, there is no ATAG at the location
+		 * pointed by r8.  Try the typical 0x100 offset from start
+		 * of RAM and hope for the best.
+		 */
+		cmp	r0, #1
+		sub	r0, r4, #TEXT_OFFSET
+		bic	r0, r0, #1
+		add	r0, r0, #0x100
+		mov	r1, r6
+		mov	r2, r5
+		bleq	atags_to_fdt
+
+		sub	sp, sp, r5
+#endif
+
+		mov	r8, r6			@ use the appended device tree
+
+		/*
+		 * Make sure that the DTB doesn't end up in the final
+		 * kernel's .bss area. To do so, we adjust the decompressed
+		 * kernel size to compensate if that .bss size is larger
+		 * than the relocated code.
+		 */
+		ldr	r5, =_kernel_bss_size
+		adr	r1, wont_overwrite
+		sub	r1, r6, r1
+		subs	r1, r5, r1
+		addhi	r9, r9, r1
+
+		/* Get the current DTB size */
+		ldr	r5, [r6, #4]
+		be32tocpu r5, r1
+
+		/* preserve 64-bit alignment */
+		add	r5, r5, #7
+		bic	r5, r5, #7
+
+		/* relocate some pointers past the appended dtb */
+		add	r6, r6, r5
+		add	r10, r10, r5
+		add	sp, sp, r5
+dtb_check_done:
+#endif
+
+/*
+ * Check to see if we will overwrite ourselves.
+ *   r4  = final kernel address (possibly with LSB set)
+ *   r9  = size of decompressed image
+ *   r10 = end of this image, including  bss/stack/malloc space if non XIP
+ * We basically want:
+ *   r4 - 16k page directory >= r10 -> OK
+ *   r4 + image length <= address of wont_overwrite -> OK
+ * Note: the possible LSB in r4 is harmless here.
+ */
+		add	r10, r10, #16384
+		cmp	r4, r10
+		bhs	wont_overwrite
+		add	r10, r4, r9
+		adr	r9, wont_overwrite
+		cmp	r10, r9
+		bls	wont_overwrite
+
+/*
+ * Relocate ourselves past the end of the decompressed kernel.
+ *   r6  = _edata
+ *   r10 = end of the decompressed kernel
+ * Because we always copy ahead, we need to do it from the end and go
+ * backward in case the source and destination overlap.
+ */
+		/*
+		 * Bump to the next 256-byte boundary with the size of
+		 * the relocation code added. This avoids overwriting
+		 * ourself when the offset is small.
+		 */
+		add	r10, r10, #((reloc_code_end - restart + 256) & ~255)
+		bic	r10, r10, #255
+
+		/* Get start of code we want to copy and align it down. */
+		adr	r5, restart
+		bic	r5, r5, #31
+
+/* Relocate the hyp vector base if necessary */
+#ifdef CONFIG_ARM_VIRT_EXT
+		mrs	r0, spsr
+		and	r0, r0, #MODE_MASK
+		cmp	r0, #HYP_MODE
+		bne	1f
+
+		/*
+		 * Compute the address of the hyp vectors after relocation.
+		 * Call __hyp_set_vectors with the new address so that we
+		 * can HVC again after the copy.
+		 */
+		adr_l	r0, __hyp_stub_vectors
+		sub	r0, r0, r5
+		add	r0, r0, r10
+		bl	__hyp_set_vectors
+1:
+#endif
+
+		sub	r9, r6, r5		@ size to copy
+		add	r9, r9, #31		@ rounded up to a multiple
+		bic	r9, r9, #31		@ ... of 32 bytes
+		add	r6, r9, r5
+		add	r9, r9, r10
+
+#ifdef DEBUG
+		sub     r10, r6, r5
+		sub     r10, r9, r10
+		/*
+		 * We are about to copy the kernel to a new memory area.
+		 * The boundaries of the new memory area can be found in
+		 * r10 and r9, whilst r5 and r6 contain the boundaries
+		 * of the memory we are going to copy.
+		 * Calling dbgkc will help with the printing of this
+		 * information.
+		 */
+		dbgkc	r5, r6, r10, r9
+#endif
+
+1:		ldmdb	r6!, {r0 - r3, r10 - r12, lr}
+		cmp	r6, r5
+		stmdb	r9!, {r0 - r3, r10 - r12, lr}
+		bhi	1b
+
+		/* Preserve offset to relocated code. */
+		sub	r6, r9, r6
+
+		mov	r0, r9			@ start of relocated zImage
+		add	r1, sp, r6		@ end of relocated zImage
+		bl	cache_clean_flush
+
+		badr	r0, restart
+		add	r0, r0, r6
+		mov	pc, r0
+
+wont_overwrite:
+		adr	r0, LC0
+		ldmia	r0, {r1, r2, r3, r11, r12}
+		sub	r0, r0, r1		@ calculate the delta offset
+
+/*
+ * If delta is zero, we are running at the address we were linked at.
+ *   r0  = delta
+ *   r2  = BSS start
+ *   r3  = BSS end
+ *   r4  = kernel execution address (possibly with LSB set)
+ *   r5  = appended dtb size (0 if not present)
+ *   r7  = architecture ID
+ *   r8  = atags pointer
+ *   r11 = GOT start
+ *   r12 = GOT end
+ *   sp  = stack pointer
+ */
+		orrs	r1, r0, r5
+		beq	not_relocated
+
+		add	r11, r11, r0
+		add	r12, r12, r0
+
+#ifndef CONFIG_ZBOOT_ROM
+		/*
+		 * If we're running fully PIC === CONFIG_ZBOOT_ROM = n,
+		 * we need to fix up pointers into the BSS region.
+		 * Note that the stack pointer has already been fixed up.
+		 */
+		add	r2, r2, r0
+		add	r3, r3, r0
+
+		/*
+		 * Relocate all entries in the GOT table.
+		 * Bump bss entries to _edata + dtb size
+		 */
+1:		ldr	r1, [r11, #0]		@ relocate entries in the GOT
+		add	r1, r1, r0		@ This fixes up C references
+		cmp	r1, r2			@ if entry >= bss_start &&
+		cmphs	r3, r1			@       bss_end > entry
+		addhi	r1, r1, r5		@    entry += dtb size
+		str	r1, [r11], #4		@ next entry
+		cmp	r11, r12
+		blo	1b
+
+		/* bump our bss pointers too */
+		add	r2, r2, r5
+		add	r3, r3, r5
+
+#else
+
+		/*
+		 * Relocate entries in the GOT table.  We only relocate
+		 * the entries that are outside the (relocated) BSS region.
+		 */
+1:		ldr	r1, [r11, #0]		@ relocate entries in the GOT
+		cmp	r1, r2			@ entry < bss_start ||
+		cmphs	r3, r1			@ _end < entry
+		addlo	r1, r1, r0		@ table.  This fixes up the
+		str	r1, [r11], #4		@ C references.
+		cmp	r11, r12
+		blo	1b
+#endif
+
+not_relocated:	mov	r0, #0
+1:		str	r0, [r2], #4		@ clear bss
+		str	r0, [r2], #4
+		str	r0, [r2], #4
+		str	r0, [r2], #4
+		cmp	r2, r3
+		blo	1b
+
+		/*
+		 * Did we skip the cache setup earlier?
+		 * That is indicated by the LSB in r4.
+		 * Do it now if so.
+		 */
+		tst	r4, #1
+		bic	r4, r4, #1
+		blne	cache_on
+
+/*
+ * The C runtime environment should now be setup sufficiently.
+ * Set up some pointers, and start decompressing.
+ *   r4  = kernel execution address
+ *   r7  = architecture ID
+ *   r8  = atags pointer
+ */
+		mov	r0, r4
+		mov	r1, sp			@ malloc space above stack
+		add	r2, sp, #MALLOC_SIZE	@ 64k max
+		mov	r3, r7
+		bl	decompress_kernel
+
+		get_inflated_image_size	r1, r2, r3
+
+		mov	r0, r4			@ start of inflated image
+		add	r1, r1, r0		@ end of inflated image
+		bl	cache_clean_flush
+		bl	cache_off
+
+#ifdef CONFIG_ARM_VIRT_EXT
+		mrs	r0, spsr		@ Get saved CPU boot mode
+		and	r0, r0, #MODE_MASK
+		cmp	r0, #HYP_MODE		@ if not booted in HYP mode...
+		bne	__enter_kernel		@ boot kernel directly
+
+		adr_l	r0, __hyp_reentry_vectors
+		bl	__hyp_set_vectors
+		__HVC(0)			@ otherwise bounce to hyp mode
+
+		b	.			@ should never be reached
+#else
+		b	__enter_kernel
+#endif
+
+		.align	2
+		.type	LC0, #object
+LC0:		.word	LC0			@ r1
+		.word	__bss_start		@ r2
+		.word	_end			@ r3
+		.word	_got_start		@ r11
+		.word	_got_end		@ ip
+		.size	LC0, . - LC0
+
+		.type	LC1, #object
+LC1:		.word	.L_user_stack_end - LC1	@ sp
+		.word	_edata - LC1		@ r6
+		.size	LC1, . - LC1
+
+.Lheadroom:
+		.word	_end - restart + 16384 + 1024*1024
+
+.Linflated_image_size_offset:
+		.long	(input_data_end - 4) - .
+
+#ifdef CONFIG_ARCH_RPC
+		.globl	params
+params:		ldr	r0, =0x10000100		@ params_phys for RPC
+		mov	pc, lr
+		.ltorg
+		.align
+#endif
+
+/*
+ * dcache_line_size - get the minimum D-cache line size from the CTR register
+ * on ARMv7.
+ */
+		.macro	dcache_line_size, reg, tmp
+#ifdef CONFIG_CPU_V7M
+		movw	\tmp, #:lower16:BASEADDR_V7M_SCB + V7M_SCB_CTR
+		movt	\tmp, #:upper16:BASEADDR_V7M_SCB + V7M_SCB_CTR
+		ldr	\tmp, [\tmp]
+#else
+		mrc	p15, 0, \tmp, c0, c0, 1		@ read ctr
+#endif
+		lsr	\tmp, \tmp, #16
+		and	\tmp, \tmp, #0xf		@ cache line size encoding
+		mov	\reg, #4			@ bytes per word
+		mov	\reg, \reg, lsl \tmp		@ actual cache line size
+		.endm
+
+/*
+ * Turn on the cache.  We need to setup some page tables so that we
+ * can have both the I and D caches on.
+ *
+ * We place the page tables 16k down from the kernel execution address,
+ * and we hope that nothing else is using it.  If we're using it, we
+ * will go pop!
+ *
+ * On entry,
+ *  r4 = kernel execution address
+ *  r7 = architecture number
+ *  r8 = atags pointer
+ * On exit,
+ *  r0, r1, r2, r3, r9, r10, r12 corrupted
+ * This routine must preserve:
+ *  r4, r7, r8
+ */
+		.align	5
+cache_on:	mov	r3, #8			@ cache_on function
+		b	call_cache_fn
+
+/*
+ * Initialize the highest priority protection region, PR7
+ * to cover all 32bit address and cacheable and bufferable.
+ */
+__armv4_mpu_cache_on:
+		mov	r0, #0x3f		@ 4G, the whole
+		mcr	p15, 0, r0, c6, c7, 0	@ PR7 Area Setting
+		mcr 	p15, 0, r0, c6, c7, 1
+
+		mov	r0, #0x80		@ PR7
+		mcr	p15, 0, r0, c2, c0, 0	@ D-cache on
+		mcr	p15, 0, r0, c2, c0, 1	@ I-cache on
+		mcr	p15, 0, r0, c3, c0, 0	@ write-buffer on
+
+		mov	r0, #0xc000
+		mcr	p15, 0, r0, c5, c0, 1	@ I-access permission
+		mcr	p15, 0, r0, c5, c0, 0	@ D-access permission
+
+		mov	r0, #0
+		mcr	p15, 0, r0, c7, c10, 4	@ drain write buffer
+		mcr	p15, 0, r0, c7, c5, 0	@ flush(inval) I-Cache
+		mcr	p15, 0, r0, c7, c6, 0	@ flush(inval) D-Cache
+		mrc	p15, 0, r0, c1, c0, 0	@ read control reg
+						@ ...I .... ..D. WC.M
+		orr	r0, r0, #0x002d		@ .... .... ..1. 11.1
+		orr	r0, r0, #0x1000		@ ...1 .... .... ....
+
+		mcr	p15, 0, r0, c1, c0, 0	@ write control reg
+
+		mov	r0, #0
+		mcr	p15, 0, r0, c7, c5, 0	@ flush(inval) I-Cache
+		mcr	p15, 0, r0, c7, c6, 0	@ flush(inval) D-Cache
+		mov	pc, lr
+
+__armv3_mpu_cache_on:
+		mov	r0, #0x3f		@ 4G, the whole
+		mcr	p15, 0, r0, c6, c7, 0	@ PR7 Area Setting
+
+		mov	r0, #0x80		@ PR7
+		mcr	p15, 0, r0, c2, c0, 0	@ cache on
+		mcr	p15, 0, r0, c3, c0, 0	@ write-buffer on
+
+		mov	r0, #0xc000
+		mcr	p15, 0, r0, c5, c0, 0	@ access permission
+
+		mov	r0, #0
+		mcr	p15, 0, r0, c7, c0, 0	@ invalidate whole cache v3
+		/*
+		 * ?? ARMv3 MMU does not allow reading the control register,
+		 * does this really work on ARMv3 MPU?
+		 */
+		mrc	p15, 0, r0, c1, c0, 0	@ read control reg
+						@ .... .... .... WC.M
+		orr	r0, r0, #0x000d		@ .... .... .... 11.1
+		/* ?? this overwrites the value constructed above? */
+		mov	r0, #0
+		mcr	p15, 0, r0, c1, c0, 0	@ write control reg
+
+		/* ?? invalidate for the second time? */
+		mcr	p15, 0, r0, c7, c0, 0	@ invalidate whole cache v3
+		mov	pc, lr
+
+#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
+#define CB_BITS 0x08
+#else
+#define CB_BITS 0x0c
+#endif
+
+__setup_mmu:	sub	r3, r4, #16384		@ Page directory size
+		bic	r3, r3, #0xff		@ Align the pointer
+		bic	r3, r3, #0x3f00
+/*
+ * Initialise the page tables, turning on the cacheable and bufferable
+ * bits for the RAM area only.
+ */
+		mov	r0, r3
+		mov	r9, r0, lsr #18
+		mov	r9, r9, lsl #18		@ start of RAM
+		add	r10, r9, #0x10000000	@ a reasonable RAM size
+		mov	r1, #0x12		@ XN|U + section mapping
+		orr	r1, r1, #3 << 10	@ AP=11
+		add	r2, r3, #16384
+1:		cmp	r1, r9			@ if virt > start of RAM
+		cmphs	r10, r1			@   && end of RAM > virt
+		bic	r1, r1, #0x1c		@ clear XN|U + C + B
+		orrlo	r1, r1, #0x10		@ Set XN|U for non-RAM
+		orrhs	r1, r1, r6		@ set RAM section settings
+		str	r1, [r0], #4		@ 1:1 mapping
+		add	r1, r1, #1048576
+		teq	r0, r2
+		bne	1b
+/*
+ * If ever we are running from Flash, then we surely want the cache
+ * to be enabled also for our execution instance...  We map 2MB of it
+ * so there is no map overlap problem for up to 1 MB compressed kernel.
+ * If the execution is in RAM then we would only be duplicating the above.
+ */
+		orr	r1, r6, #0x04		@ ensure B is set for this
+		orr	r1, r1, #3 << 10
+		mov	r2, pc
+		mov	r2, r2, lsr #20
+		orr	r1, r1, r2, lsl #20
+		add	r0, r3, r2, lsl #2
+		str	r1, [r0], #4
+		add	r1, r1, #1048576
+		str	r1, [r0]
+		mov	pc, lr
+ENDPROC(__setup_mmu)
+
+@ Enable unaligned access on v6, to allow better code generation
+@ for the decompressor C code:
+__armv6_mmu_cache_on:
+		mrc	p15, 0, r0, c1, c0, 0	@ read SCTLR
+		bic	r0, r0, #2		@ A (no unaligned access fault)
+		orr	r0, r0, #1 << 22	@ U (v6 unaligned access model)
+		mcr	p15, 0, r0, c1, c0, 0	@ write SCTLR
+		b	__armv4_mmu_cache_on
+
+__arm926ejs_mmu_cache_on:
+#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
+		mov	r0, #4			@ put dcache in WT mode
+		mcr	p15, 7, r0, c15, c0, 0
+#endif
+
+__armv4_mmu_cache_on:
+		mov	r12, lr
+#ifdef CONFIG_MMU
+		mov	r6, #CB_BITS | 0x12	@ U
+		bl	__setup_mmu
+		mov	r0, #0
+		mcr	p15, 0, r0, c7, c10, 4	@ drain write buffer
+		mcr	p15, 0, r0, c8, c7, 0	@ flush I,D TLBs
+		mrc	p15, 0, r0, c1, c0, 0	@ read control reg
+		orr	r0, r0, #0x5000		@ I-cache enable, RR cache replacement
+		orr	r0, r0, #0x0030
+ ARM_BE8(	orr	r0, r0, #1 << 25 )	@ big-endian page tables
+		bl	__common_mmu_cache_on
+		mov	r0, #0
+		mcr	p15, 0, r0, c8, c7, 0	@ flush I,D TLBs
+#endif
+		mov	pc, r12
+
+__armv7_mmu_cache_on:
+		enable_cp15_barriers	r11
+		mov	r12, lr
+#ifdef CONFIG_MMU
+		mrc	p15, 0, r11, c0, c1, 4	@ read ID_MMFR0
+		tst	r11, #0xf		@ VMSA
+		movne	r6, #CB_BITS | 0x02	@ !XN
+		blne	__setup_mmu
+		mov	r0, #0
+		mcr	p15, 0, r0, c7, c10, 4	@ drain write buffer
+		tst	r11, #0xf		@ VMSA
+		mcrne	p15, 0, r0, c8, c7, 0	@ flush I,D TLBs
+#endif
+		mrc	p15, 0, r0, c1, c0, 0	@ read control reg
+		bic	r0, r0, #1 << 28	@ clear SCTLR.TRE
+		orr	r0, r0, #0x5000		@ I-cache enable, RR cache replacement
+		orr	r0, r0, #0x003c		@ write buffer
+		bic	r0, r0, #2		@ A (no unaligned access fault)
+		orr	r0, r0, #1 << 22	@ U (v6 unaligned access model)
+						@ (needed for ARM1176)
+#ifdef CONFIG_MMU
+ ARM_BE8(	orr	r0, r0, #1 << 25 )	@ big-endian page tables
+		mrcne   p15, 0, r6, c2, c0, 2   @ read ttb control reg
+		orrne	r0, r0, #1		@ MMU enabled
+		movne	r1, #0xfffffffd		@ domain 0 = client
+		bic     r6, r6, #1 << 31        @ 32-bit translation system
+		bic     r6, r6, #(7 << 0) | (1 << 4)	@ use only ttbr0
+		mcrne	p15, 0, r3, c2, c0, 0	@ load page table pointer
+		mcrne	p15, 0, r1, c3, c0, 0	@ load domain access control
+		mcrne   p15, 0, r6, c2, c0, 2   @ load ttb control
+#endif
+		mcr	p15, 0, r0, c7, c5, 4	@ ISB
+		mcr	p15, 0, r0, c1, c0, 0	@ load control register
+		mrc	p15, 0, r0, c1, c0, 0	@ and read it back
+		mov	r0, #0
+		mcr	p15, 0, r0, c7, c5, 4	@ ISB
+		mov	pc, r12
+
+__fa526_cache_on:
+		mov	r12, lr
+		mov	r6, #CB_BITS | 0x12	@ U
+		bl	__setup_mmu
+		mov	r0, #0
+		mcr	p15, 0, r0, c7, c7, 0	@ Invalidate whole cache
+		mcr	p15, 0, r0, c7, c10, 4	@ drain write buffer
+		mcr	p15, 0, r0, c8, c7, 0	@ flush UTLB
+		mrc	p15, 0, r0, c1, c0, 0	@ read control reg
+		orr	r0, r0, #0x1000		@ I-cache enable
+		bl	__common_mmu_cache_on
+		mov	r0, #0
+		mcr	p15, 0, r0, c8, c7, 0	@ flush UTLB
+		mov	pc, r12
+
+__common_mmu_cache_on:
+#ifndef CONFIG_THUMB2_KERNEL
+#ifndef DEBUG
+		orr	r0, r0, #0x000d		@ Write buffer, mmu
+#endif
+		mov	r1, #-1
+		mcr	p15, 0, r3, c2, c0, 0	@ load page table pointer
+		mcr	p15, 0, r1, c3, c0, 0	@ load domain access control
+		b	1f
+		.align	5			@ cache line aligned
+1:		mcr	p15, 0, r0, c1, c0, 0	@ load control register
+		mrc	p15, 0, r0, c1, c0, 0	@ and read it back to
+		sub	pc, lr, r0, lsr #32	@ properly flush pipeline
+#endif
+
+#define PROC_ENTRY_SIZE (4*5)
+
+/*
+ * Here follow the relocatable cache support functions for the
+ * various processors.  This is a generic hook for locating an
+ * entry and jumping to an instruction at the specified offset
+ * from the start of the block.  Please note this is all position
+ * independent code.
+ *
+ *  r1  = corrupted
+ *  r2  = corrupted
+ *  r3  = block offset
+ *  r9  = corrupted
+ *  r12 = corrupted
+ */
+
+call_cache_fn:	adr	r12, proc_types
+#ifdef CONFIG_CPU_CP15
+		mrc	p15, 0, r9, c0, c0	@ get processor ID
+#elif defined(CONFIG_CPU_V7M)
+		/*
+		 * On v7-M the processor id is located in the V7M_SCB_CPUID
+		 * register, but as cache handling is IMPLEMENTATION DEFINED on
+		 * v7-M (if existant at all) we just return early here.
+		 * If V7M_SCB_CPUID were used the cpu ID functions (i.e.
+		 * __armv7_mmu_cache_{on,off,flush}) would be selected which
+		 * use cp15 registers that are not implemented on v7-M.
+		 */
+		bx	lr
+#else
+		ldr	r9, =CONFIG_PROCESSOR_ID
+#endif
+1:		ldr	r1, [r12, #0]		@ get value
+		ldr	r2, [r12, #4]		@ get mask
+		eor	r1, r1, r9		@ (real ^ match)
+		tst	r1, r2			@       & mask
+ ARM(		addeq	pc, r12, r3		) @ call cache function
+ THUMB(		addeq	r12, r3			)
+ THUMB(		moveq	pc, r12			) @ call cache function
+		add	r12, r12, #PROC_ENTRY_SIZE
+		b	1b
+
+/*
+ * Table for cache operations.  This is basically:
+ *   - CPU ID match
+ *   - CPU ID mask
+ *   - 'cache on' method instruction
+ *   - 'cache off' method instruction
+ *   - 'cache flush' method instruction
+ *
+ * We match an entry using: ((real_id ^ match) & mask) == 0
+ *
+ * Writethrough caches generally only need 'on' and 'off'
+ * methods.  Writeback caches _must_ have the flush method
+ * defined.
+ */
+		.align	2
+		.type	proc_types,#object
+proc_types:
+		.word	0x41000000		@ old ARM ID
+		.word	0xff00f000
+		mov	pc, lr
+ THUMB(		nop				)
+		mov	pc, lr
+ THUMB(		nop				)
+		mov	pc, lr
+ THUMB(		nop				)
+
+		.word	0x41007000		@ ARM7/710
+		.word	0xfff8fe00
+		mov	pc, lr
+ THUMB(		nop				)
+		mov	pc, lr
+ THUMB(		nop				)
+		mov	pc, lr
+ THUMB(		nop				)
+
+		.word	0x41807200		@ ARM720T (writethrough)
+		.word	0xffffff00
+		W(b)	__armv4_mmu_cache_on
+		W(b)	__armv4_mmu_cache_off
+		mov	pc, lr
+ THUMB(		nop				)
+
+		.word	0x41007400		@ ARM74x
+		.word	0xff00ff00
+		W(b)	__armv3_mpu_cache_on
+		W(b)	__armv3_mpu_cache_off
+		W(b)	__armv3_mpu_cache_flush
+		
+		.word	0x41009400		@ ARM94x
+		.word	0xff00ff00
+		W(b)	__armv4_mpu_cache_on
+		W(b)	__armv4_mpu_cache_off
+		W(b)	__armv4_mpu_cache_flush
+
+		.word	0x41069260		@ ARM926EJ-S (v5TEJ)
+		.word	0xff0ffff0
+		W(b)	__arm926ejs_mmu_cache_on
+		W(b)	__armv4_mmu_cache_off
+		W(b)	__armv5tej_mmu_cache_flush
+
+		.word	0x00007000		@ ARM7 IDs
+		.word	0x0000f000
+		mov	pc, lr
+ THUMB(		nop				)
+		mov	pc, lr
+ THUMB(		nop				)
+		mov	pc, lr
+ THUMB(		nop				)
+
+		@ Everything from here on will be the new ID system.
+
+		.word	0x4401a100		@ sa110 / sa1100
+		.word	0xffffffe0
+		W(b)	__armv4_mmu_cache_on
+		W(b)	__armv4_mmu_cache_off
+		W(b)	__armv4_mmu_cache_flush
+
+		.word	0x6901b110		@ sa1110
+		.word	0xfffffff0
+		W(b)	__armv4_mmu_cache_on
+		W(b)	__armv4_mmu_cache_off
+		W(b)	__armv4_mmu_cache_flush
+
+		.word	0x56056900
+		.word	0xffffff00		@ PXA9xx
+		W(b)	__armv4_mmu_cache_on
+		W(b)	__armv4_mmu_cache_off
+		W(b)	__armv4_mmu_cache_flush
+
+		.word	0x56158000		@ PXA168
+		.word	0xfffff000
+		W(b)	__armv4_mmu_cache_on
+		W(b)	__armv4_mmu_cache_off
+		W(b)	__armv5tej_mmu_cache_flush
+
+		.word	0x56050000		@ Feroceon
+		.word	0xff0f0000
+		W(b)	__armv4_mmu_cache_on
+		W(b)	__armv4_mmu_cache_off
+		W(b)	__armv5tej_mmu_cache_flush
+
+#ifdef CONFIG_CPU_FEROCEON_OLD_ID
+		/* this conflicts with the standard ARMv5TE entry */
+		.long	0x41009260		@ Old Feroceon
+		.long	0xff00fff0
+		b	__armv4_mmu_cache_on
+		b	__armv4_mmu_cache_off
+		b	__armv5tej_mmu_cache_flush
+#endif
+
+		.word	0x66015261		@ FA526
+		.word	0xff01fff1
+		W(b)	__fa526_cache_on
+		W(b)	__armv4_mmu_cache_off
+		W(b)	__fa526_cache_flush
+
+		@ These match on the architecture ID
+
+		.word	0x00020000		@ ARMv4T
+		.word	0x000f0000
+		W(b)	__armv4_mmu_cache_on
+		W(b)	__armv4_mmu_cache_off
+		W(b)	__armv4_mmu_cache_flush
+
+		.word	0x00050000		@ ARMv5TE
+		.word	0x000f0000
+		W(b)	__armv4_mmu_cache_on
+		W(b)	__armv4_mmu_cache_off
+		W(b)	__armv4_mmu_cache_flush
+
+		.word	0x00060000		@ ARMv5TEJ
+		.word	0x000f0000
+		W(b)	__armv4_mmu_cache_on
+		W(b)	__armv4_mmu_cache_off
+		W(b)	__armv5tej_mmu_cache_flush
+
+		.word	0x0007b000		@ ARMv6
+		.word	0x000ff000
+		W(b)	__armv6_mmu_cache_on
+		W(b)	__armv4_mmu_cache_off
+		W(b)	__armv6_mmu_cache_flush
+
+		.word	0x000f0000		@ new CPU Id
+		.word	0x000f0000
+		W(b)	__armv7_mmu_cache_on
+		W(b)	__armv7_mmu_cache_off
+		W(b)	__armv7_mmu_cache_flush
+
+		.word	0			@ unrecognised type
+		.word	0
+		mov	pc, lr
+ THUMB(		nop				)
+		mov	pc, lr
+ THUMB(		nop				)
+		mov	pc, lr
+ THUMB(		nop				)
+
+		.size	proc_types, . - proc_types
+
+		/*
+		 * If you get a "non-constant expression in ".if" statement"
+		 * error from the assembler on this line, check that you have
+		 * not accidentally written a "b" instruction where you should
+		 * have written W(b).
+		 */
+		.if (. - proc_types) % PROC_ENTRY_SIZE != 0
+		.error "The size of one or more proc_types entries is wrong."
+		.endif
+
+/*
+ * Turn off the Cache and MMU.  ARMv3 does not support
+ * reading the control register, but ARMv4 does.
+ *
+ * On exit,
+ *  r0, r1, r2, r3, r9, r12 corrupted
+ * This routine must preserve:
+ *  r4, r7, r8
+ */
+		.align	5
+cache_off:	mov	r3, #12			@ cache_off function
+		b	call_cache_fn
+
+__armv4_mpu_cache_off:
+		mrc	p15, 0, r0, c1, c0
+		bic	r0, r0, #0x000d
+		mcr	p15, 0, r0, c1, c0	@ turn MPU and cache off
+		mov	r0, #0
+		mcr	p15, 0, r0, c7, c10, 4	@ drain write buffer
+		mcr	p15, 0, r0, c7, c6, 0	@ flush D-Cache
+		mcr	p15, 0, r0, c7, c5, 0	@ flush I-Cache
+		mov	pc, lr
+
+__armv3_mpu_cache_off:
+		mrc	p15, 0, r0, c1, c0
+		bic	r0, r0, #0x000d
+		mcr	p15, 0, r0, c1, c0, 0	@ turn MPU and cache off
+		mov	r0, #0
+		mcr	p15, 0, r0, c7, c0, 0	@ invalidate whole cache v3
+		mov	pc, lr
+
+__armv4_mmu_cache_off:
+#ifdef CONFIG_MMU
+		mrc	p15, 0, r0, c1, c0
+		bic	r0, r0, #0x000d
+		mcr	p15, 0, r0, c1, c0	@ turn MMU and cache off
+		mov	r0, #0
+		mcr	p15, 0, r0, c7, c7	@ invalidate whole cache v4
+		mcr	p15, 0, r0, c8, c7	@ invalidate whole TLB v4
+#endif
+		mov	pc, lr
+
+__armv7_mmu_cache_off:
+		mrc	p15, 0, r0, c1, c0
+#ifdef CONFIG_MMU
+		bic	r0, r0, #0x0005
+#else
+		bic	r0, r0, #0x0004
+#endif
+		mcr	p15, 0, r0, c1, c0	@ turn MMU and cache off
+		mov	r0, #0
+#ifdef CONFIG_MMU
+		mcr	p15, 0, r0, c8, c7, 0	@ invalidate whole TLB
+#endif
+		mcr	p15, 0, r0, c7, c5, 6	@ invalidate BTC
+		mcr	p15, 0, r0, c7, c10, 4	@ DSB
+		mcr	p15, 0, r0, c7, c5, 4	@ ISB
+		mov	pc, lr
+
+/*
+ * Clean and flush the cache to maintain consistency.
+ *
+ * On entry,
+ *  r0 = start address
+ *  r1 = end address (exclusive)
+ * On exit,
+ *  r1, r2, r3, r9, r10, r11, r12 corrupted
+ * This routine must preserve:
+ *  r4, r6, r7, r8
+ */
+		.align	5
+cache_clean_flush:
+		mov	r3, #16
+		mov	r11, r1
+		b	call_cache_fn
+
+__armv4_mpu_cache_flush:
+		tst	r4, #1
+		movne	pc, lr
+		mov	r2, #1
+		mov	r3, #0
+		mcr	p15, 0, ip, c7, c6, 0	@ invalidate D cache
+		mov	r1, #7 << 5		@ 8 segments
+1:		orr	r3, r1, #63 << 26	@ 64 entries
+2:		mcr	p15, 0, r3, c7, c14, 2	@ clean & invalidate D index
+		subs	r3, r3, #1 << 26
+		bcs	2b			@ entries 63 to 0
+		subs 	r1, r1, #1 << 5
+		bcs	1b			@ segments 7 to 0
+
+		teq	r2, #0
+		mcrne	p15, 0, ip, c7, c5, 0	@ invalidate I cache
+		mcr	p15, 0, ip, c7, c10, 4	@ drain WB
+		mov	pc, lr
+		
+__fa526_cache_flush:
+		tst	r4, #1
+		movne	pc, lr
+		mov	r1, #0
+		mcr	p15, 0, r1, c7, c14, 0	@ clean and invalidate D cache
+		mcr	p15, 0, r1, c7, c5, 0	@ flush I cache
+		mcr	p15, 0, r1, c7, c10, 4	@ drain WB
+		mov	pc, lr
+
+__armv6_mmu_cache_flush:
+		mov	r1, #0
+		tst	r4, #1
+		mcreq	p15, 0, r1, c7, c14, 0	@ clean+invalidate D
+		mcr	p15, 0, r1, c7, c5, 0	@ invalidate I+BTB
+		mcreq	p15, 0, r1, c7, c15, 0	@ clean+invalidate unified
+		mcr	p15, 0, r1, c7, c10, 4	@ drain WB
+		mov	pc, lr
+
+__armv7_mmu_cache_flush:
+		enable_cp15_barriers	r10
+		tst	r4, #1
+		bne	iflush
+		mrc	p15, 0, r10, c0, c1, 5	@ read ID_MMFR1
+		tst	r10, #0xf << 16		@ hierarchical cache (ARMv7)
+		mov	r10, #0
+		beq	hierarchical
+		mcr	p15, 0, r10, c7, c14, 0	@ clean+invalidate D
+		b	iflush
+hierarchical:
+		dcache_line_size r1, r2		@ r1 := dcache min line size
+		sub	r2, r1, #1		@ r2 := line size mask
+		bic	r0, r0, r2		@ round down start to line size
+		sub	r11, r11, #1		@ end address is exclusive
+		bic	r11, r11, r2		@ round down end to line size
+0:		cmp	r0, r11			@ finished?
+		bgt	iflush
+		mcr	p15, 0, r0, c7, c14, 1	@ Dcache clean/invalidate by VA
+		add	r0, r0, r1
+		b	0b
+iflush:
+		mcr	p15, 0, r10, c7, c10, 4	@ DSB
+		mcr	p15, 0, r10, c7, c5, 0	@ invalidate I+BTB
+		mcr	p15, 0, r10, c7, c10, 4	@ DSB
+		mcr	p15, 0, r10, c7, c5, 4	@ ISB
+		mov	pc, lr
+
+__armv5tej_mmu_cache_flush:
+		tst	r4, #1
+		movne	pc, lr
+1:		mrc	p15, 0, APSR_nzcv, c7, c14, 3	@ test,clean,invalidate D cache
+		bne	1b
+		mcr	p15, 0, r0, c7, c5, 0	@ flush I cache
+		mcr	p15, 0, r0, c7, c10, 4	@ drain WB
+		mov	pc, lr
+
+__armv4_mmu_cache_flush:
+		tst	r4, #1
+		movne	pc, lr
+		mov	r2, #64*1024		@ default: 32K dcache size (*2)
+		mov	r11, #32		@ default: 32 byte line size
+		mrc	p15, 0, r3, c0, c0, 1	@ read cache type
+		teq	r3, r9			@ cache ID register present?
+		beq	no_cache_id
+		mov	r1, r3, lsr #18
+		and	r1, r1, #7
+		mov	r2, #1024
+		mov	r2, r2, lsl r1		@ base dcache size *2
+		tst	r3, #1 << 14		@ test M bit
+		addne	r2, r2, r2, lsr #1	@ +1/2 size if M == 1
+		mov	r3, r3, lsr #12
+		and	r3, r3, #3
+		mov	r11, #8
+		mov	r11, r11, lsl r3	@ cache line size in bytes
+no_cache_id:
+		mov	r1, pc
+		bic	r1, r1, #63		@ align to longest cache line
+		add	r2, r1, r2
+1:
+ ARM(		ldr	r3, [r1], r11		) @ s/w flush D cache
+ THUMB(		ldr     r3, [r1]		) @ s/w flush D cache
+ THUMB(		add     r1, r1, r11		)
+		teq	r1, r2
+		bne	1b
+
+		mcr	p15, 0, r1, c7, c5, 0	@ flush I cache
+		mcr	p15, 0, r1, c7, c6, 0	@ flush D cache
+		mcr	p15, 0, r1, c7, c10, 4	@ drain WB
+		mov	pc, lr
+
+__armv3_mmu_cache_flush:
+__armv3_mpu_cache_flush:
+		tst	r4, #1
+		movne	pc, lr
+		mov	r1, #0
+		mcr	p15, 0, r1, c7, c0, 0	@ invalidate whole cache v3
+		mov	pc, lr
+
+/*
+ * Various debugging routines for printing hex characters and
+ * memory, which again must be relocatable.
+ */
+#ifdef DEBUG
+		.align	2
+		.type	phexbuf,#object
+phexbuf:	.space	12
+		.size	phexbuf, . - phexbuf
+
+@ phex corrupts {r0, r1, r2, r3}
+phex:		adr	r3, phexbuf
+		mov	r2, #0
+		strb	r2, [r3, r1]
+1:		subs	r1, r1, #1
+		movmi	r0, r3
+		bmi	puts
+		and	r2, r0, #15
+		mov	r0, r0, lsr #4
+		cmp	r2, #10
+		addge	r2, r2, #7
+		add	r2, r2, #'0'
+		strb	r2, [r3, r1]
+		b	1b
+
+@ puts corrupts {r0, r1, r2, r3}
+puts:		loadsp	r3, r2, r1
+1:		ldrb	r2, [r0], #1
+		teq	r2, #0
+		moveq	pc, lr
+2:		writeb	r2, r3, r1
+		mov	r1, #0x00020000
+3:		subs	r1, r1, #1
+		bne	3b
+		teq	r2, #'\n'
+		moveq	r2, #'\r'
+		beq	2b
+		teq	r0, #0
+		bne	1b
+		mov	pc, lr
+@ putc corrupts {r0, r1, r2, r3}
+putc:
+		mov	r2, r0
+		loadsp	r3, r1, r0
+		mov	r0, #0
+		b	2b
+
+@ memdump corrupts {r0, r1, r2, r3, r10, r11, r12, lr}
+memdump:	mov	r12, r0
+		mov	r10, lr
+		mov	r11, #0
+2:		mov	r0, r11, lsl #2
+		add	r0, r0, r12
+		mov	r1, #8
+		bl	phex
+		mov	r0, #':'
+		bl	putc
+1:		mov	r0, #' '
+		bl	putc
+		ldr	r0, [r12, r11, lsl #2]
+		mov	r1, #8
+		bl	phex
+		and	r0, r11, #7
+		teq	r0, #3
+		moveq	r0, #' '
+		bleq	putc
+		and	r0, r11, #7
+		add	r11, r11, #1
+		teq	r0, #7
+		bne	1b
+		mov	r0, #'\n'
+		bl	putc
+		cmp	r11, #64
+		blt	2b
+		mov	pc, r10
+#endif
+
+		.ltorg
+
+#ifdef CONFIG_ARM_VIRT_EXT
+.align 5
+__hyp_reentry_vectors:
+		W(b)	.			@ reset
+		W(b)	.			@ undef
+#ifdef CONFIG_EFI_STUB
+		W(b)	__enter_kernel_from_hyp	@ hvc from HYP
+#else
+		W(b)	.			@ svc
+#endif
+		W(b)	.			@ pabort
+		W(b)	.			@ dabort
+		W(b)	__enter_kernel		@ hyp
+		W(b)	.			@ irq
+		W(b)	.			@ fiq
+#endif /* CONFIG_ARM_VIRT_EXT */
+
+__enter_kernel:
+		mov	r0, #0			@ must be 0
+		mov	r1, r7			@ restore architecture number
+		mov	r2, r8			@ restore atags pointer
+ ARM(		mov	pc, r4		)	@ call kernel
+ M_CLASS(	add	r4, r4, #1	)	@ enter in Thumb mode for M class
+ THUMB(		bx	r4		)	@ entry point is always ARM for A/R classes
+
+reloc_code_end:
+
+#ifdef CONFIG_EFI_STUB
+__enter_kernel_from_hyp:
+		mrc	p15, 4, r0, c1, c0, 0	@ read HSCTLR
+		bic	r0, r0, #0x5		@ disable MMU and caches
+		mcr	p15, 4, r0, c1, c0, 0	@ write HSCTLR
+		isb
+		b	__enter_kernel
+
+ENTRY(efi_enter_kernel)
+		mov	r4, r0			@ preserve image base
+		mov	r8, r1			@ preserve DT pointer
+
+		adr_l	r0, call_cache_fn
+		adr	r1, 0f			@ clean the region of code we
+		bl	cache_clean_flush	@ may run with the MMU off
+
+#ifdef CONFIG_ARM_VIRT_EXT
+		@
+		@ The EFI spec does not support booting on ARM in HYP mode,
+		@ since it mandates that the MMU and caches are on, with all
+		@ 32-bit addressable DRAM mapped 1:1 using short descriptors.
+		@
+		@ While the EDK2 reference implementation adheres to this,
+		@ U-Boot might decide to enter the EFI stub in HYP mode
+		@ anyway, with the MMU and caches either on or off.
+		@
+		mrs	r0, cpsr		@ get the current mode
+		msr	spsr_cxsf, r0		@ record boot mode
+		and	r0, r0, #MODE_MASK	@ are we running in HYP mode?
+		cmp	r0, #HYP_MODE
+		bne	.Lefi_svc
+
+		mrc	p15, 4, r1, c1, c0, 0	@ read HSCTLR
+		tst	r1, #0x1		@ MMU enabled at HYP?
+		beq	1f
+
+		@
+		@ When running in HYP mode with the caches on, we're better
+		@ off just carrying on using the cached 1:1 mapping that the
+		@ firmware provided. Set up the HYP vectors so HVC instructions
+		@ issued from HYP mode take us to the correct handler code. We
+		@ will disable the MMU before jumping to the kernel proper.
+		@
+ ARM(		bic	r1, r1, #(1 << 30)	) @ clear HSCTLR.TE
+ THUMB(		orr	r1, r1, #(1 << 30)	) @ set HSCTLR.TE
+		mcr	p15, 4, r1, c1, c0, 0
+		adr	r0, __hyp_reentry_vectors
+		mcr	p15, 4, r0, c12, c0, 0	@ set HYP vector base (HVBAR)
+		isb
+		b	.Lefi_hyp
+
+		@
+		@ When running in HYP mode with the caches off, we need to drop
+		@ into SVC mode now, and let the decompressor set up its cached
+		@ 1:1 mapping as usual.
+		@
+1:		mov	r9, r4			@ preserve image base
+		bl	__hyp_stub_install	@ install HYP stub vectors
+		safe_svcmode_maskall	r1	@ drop to SVC mode
+		msr	spsr_cxsf, r0		@ record boot mode
+		orr	r4, r9, #1		@ restore image base and set LSB
+		b	.Lefi_hyp
+.Lefi_svc:
+#endif
+		mrc	p15, 0, r0, c1, c0, 0	@ read SCTLR
+		tst	r0, #0x1		@ MMU enabled?
+		orreq	r4, r4, #1		@ set LSB if not
+
+.Lefi_hyp:
+		mov	r0, r8			@ DT start
+		add	r1, r8, r2		@ DT end
+		bl	cache_clean_flush
+
+		adr	r0, 0f			@ switch to our stack
+		ldr	sp, [r0]
+		add	sp, sp, r0
+
+		mov	r5, #0			@ appended DTB size
+		mov	r7, #0xFFFFFFFF		@ machine ID
+		b	wont_overwrite
+ENDPROC(efi_enter_kernel)
+0:		.long	.L_user_stack_end - .
+#endif
+
+		.align
+		.section ".stack", "aw", %nobits
+.L_user_stack:	.space	4096
+.L_user_stack_end:
diff --git a/arch/arm/boot/compressed/hyp-stub.S b/arch/arm/boot/compressed/hyp-stub.S
new file mode 100644
index 000000000..a703eaa86
--- /dev/null
+++ b/arch/arm/boot/compressed/hyp-stub.S
@@ -0,0 +1,2 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#include "../../kernel/hyp-stub.S"
diff --git a/arch/arm/boot/compressed/lib1funcs.S b/arch/arm/boot/compressed/lib1funcs.S
new file mode 100644
index 000000000..815dec73b
--- /dev/null
+++ b/arch/arm/boot/compressed/lib1funcs.S
@@ -0,0 +1,3 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* For __aeabi_uidivmod */
+#include "../../lib/lib1funcs.S"
diff --git a/arch/arm/boot/compressed/ll_char_wr.S b/arch/arm/boot/compressed/ll_char_wr.S
new file mode 100644
index 000000000..1ec8cb289
--- /dev/null
+++ b/arch/arm/boot/compressed/ll_char_wr.S
@@ -0,0 +1,131 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ *  linux/arch/arm/lib/ll_char_wr.S
+ *
+ *  Copyright (C) 1995, 1996 Russell King.
+ *
+ *  Speedups & 1bpp code (C) 1996 Philip Blundell & Russell King.
+ *
+ *  10-04-96	RMK	Various cleanups & reduced register usage.
+ *  08-04-98	RMK	Shifts re-ordered
+ */
+
+@ Regs: [] = corruptible
+@       {} = used
+@       () = do not use
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+	.text
+
+LC0:	.word	LC0
+	.word	bytes_per_char_h
+	.word	video_size_row
+	.word	acorndata_8x8
+	.word	con_charconvtable
+
+/*
+ * r0 = ptr
+ * r1 = char
+ * r2 = white
+ */
+ENTRY(ll_write_char)
+	stmfd	sp!, {r4 - r7, lr}
+@
+@ Smashable regs: {r0 - r3}, [r4 - r7], (r8 - fp), [ip], (sp), [lr], (pc)
+@
+	/*
+	 * calculate offset into character table
+	 */
+	mov	r1, r1, lsl #3
+	/*
+	 * calculate offset required for each row.
+	 */
+	adr	ip, LC0
+	ldmia	ip, {r3, r4, r5, r6, lr}
+	sub	ip, ip, r3
+	add	r6, r6, ip
+	add	lr, lr, ip
+	ldr	r4, [r4, ip]
+	ldr	r5, [r5, ip]
+	/*
+	 * Go to resolution-dependent routine...
+	 */
+	cmp	r4, #4
+	blt	Lrow1bpp
+	add	r0, r0, r5, lsl #3		@ Move to bottom of character
+	orr	r1, r1, #7
+	ldrb	r7, [r6, r1]
+	teq	r4, #8
+	beq	Lrow8bpplp
+@
+@ Smashable regs: {r0 - r3}, [r4], {r5 - r7}, (r8 - fp), [ip], (sp), {lr}, (pc)
+@
+Lrow4bpplp:
+	ldr	r7, [lr, r7, lsl #2]
+	mul	r7, r2, r7
+	sub	r1, r1, #1			@ avoid using r7 directly after
+	str	r7, [r0, -r5]!
+	ldrb	r7, [r6, r1]
+	ldr	r7, [lr, r7, lsl #2]
+	mul	r7, r2, r7
+	tst	r1, #7				@ avoid using r7 directly after
+	str	r7, [r0, -r5]!
+	subne	r1, r1, #1
+	ldrbne	r7, [r6, r1]
+	bne	Lrow4bpplp
+	ldmfd	sp!, {r4 - r7, pc}
+
+@
+@ Smashable regs: {r0 - r3}, [r4], {r5 - r7}, (r8 - fp), [ip], (sp), {lr}, (pc)
+@
+Lrow8bpplp:
+	mov	ip, r7, lsr #4
+	ldr	ip, [lr, ip, lsl #2]
+	mul	r4, r2, ip
+	and	ip, r7, #15			@ avoid r4
+	ldr	ip, [lr, ip, lsl #2]		@ avoid r4
+	mul	ip, r2, ip			@ avoid r4
+	sub	r1, r1, #1			@ avoid ip
+	sub	r0, r0, r5			@ avoid ip
+	stmia	r0, {r4, ip}
+	ldrb	r7, [r6, r1]
+	mov	ip, r7, lsr #4
+	ldr	ip, [lr, ip, lsl #2]
+	mul	r4, r2, ip
+	and	ip, r7, #15			@ avoid r4
+	ldr	ip, [lr, ip, lsl #2]		@ avoid r4
+	mul	ip, r2, ip			@ avoid r4
+	tst	r1, #7				@ avoid ip
+	sub	r0, r0, r5			@ avoid ip
+	stmia	r0, {r4, ip}
+	subne	r1, r1, #1
+	ldrbne	r7, [r6, r1]
+	bne	Lrow8bpplp
+	ldmfd	sp!, {r4 - r7, pc}
+
+@
+@ Smashable regs: {r0 - r3}, [r4], {r5, r6}, [r7], (r8 - fp), [ip], (sp), [lr], (pc)
+@
+Lrow1bpp:
+	add	r6, r6, r1
+	ldmia	r6, {r4, r7}
+	strb	r4, [r0], r5
+	mov	r4, r4, lsr #8
+	strb	r4, [r0], r5
+	mov	r4, r4, lsr #8
+	strb	r4, [r0], r5
+	mov	r4, r4, lsr #8
+	strb	r4, [r0], r5
+	strb	r7, [r0], r5
+	mov	r7, r7, lsr #8
+	strb	r7, [r0], r5
+	mov	r7, r7, lsr #8
+	strb	r7, [r0], r5
+	mov	r7, r7, lsr #8
+	strb	r7, [r0], r5
+	ldmfd	sp!, {r4 - r7, pc}
+
+	.bss
+ENTRY(con_charconvtable)
+	.space	1024
diff --git a/arch/arm/boot/compressed/misc-ep93xx.h b/arch/arm/boot/compressed/misc-ep93xx.h
new file mode 100644
index 000000000..b0a1b42aa
--- /dev/null
+++ b/arch/arm/boot/compressed/misc-ep93xx.h
@@ -0,0 +1,87 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Copyright (C) 2006 Lennert Buytenhek <buytenh@wantstofly.org>
+ */
+
+/*
+ * These machine IDs are no longer used by the kernel since EP93xx was converted
+ * to DT booting, but they are still passed in by bootloaders, so we use our own
+ * local definitions of the relevant macros.
+ */
+#define machine_is_bk3()		(__machine_arch_type == 1880)
+#define machine_is_edb9301()		(__machine_arch_type == 462)
+#define machine_is_edb9302a()		(__machine_arch_type == 1127)
+#define machine_is_edb9302()		(__machine_arch_type == 538)
+#define machine_is_edb9307a()		(__machine_arch_type == 1128)
+#define machine_is_edb9307()		(__machine_arch_type == 607)
+#define machine_is_edb9312()		(__machine_arch_type == 451)
+#define machine_is_edb9315a()		(__machine_arch_type == 772)
+#define machine_is_edb9315()		(__machine_arch_type == 463)
+#define machine_is_ts72xx()		(__machine_arch_type == 673)
+#define machine_is_vision_ep9307()	(__machine_arch_type == 1578)
+
+static inline unsigned int __raw_readl(unsigned int ptr)
+{
+	return *((volatile unsigned int *)ptr);
+}
+
+static inline void __raw_writeb(unsigned char value, unsigned int ptr)
+{
+	*((volatile unsigned char *)ptr) = value;
+}
+
+static inline void __raw_writel(unsigned int value, unsigned int ptr)
+{
+	*((volatile unsigned int *)ptr) = value;
+}
+
+/*
+ * Some bootloaders don't turn off DMA from the ethernet MAC before
+ * jumping to linux, which means that we might end up with bits of RX
+ * status and packet data scribbled over the uncompressed kernel image.
+ * Work around this by resetting the ethernet MAC before we uncompress.
+ */
+#define PHYS_ETH_SELF_CTL		0x80010020
+#define ETH_SELF_CTL_RESET		0x00000001
+
+static inline void ep93xx_ethernet_reset(void)
+{
+	unsigned int v;
+
+	/* Reset the ethernet MAC.  */
+	v = __raw_readl(PHYS_ETH_SELF_CTL);
+	__raw_writel(v | ETH_SELF_CTL_RESET, PHYS_ETH_SELF_CTL);
+
+	/* Wait for reset to finish.  */
+	while (__raw_readl(PHYS_ETH_SELF_CTL) & ETH_SELF_CTL_RESET)
+		;
+}
+
+#define TS72XX_WDT_CONTROL_PHYS_BASE	0x23800000
+#define TS72XX_WDT_FEED_PHYS_BASE	0x23c00000
+#define TS72XX_WDT_FEED_VAL		0x05
+
+static inline void __maybe_unused ts72xx_watchdog_disable(void)
+{
+	__raw_writeb(TS72XX_WDT_FEED_VAL, TS72XX_WDT_FEED_PHYS_BASE);
+	__raw_writeb(0, TS72XX_WDT_CONTROL_PHYS_BASE);
+}
+
+static inline void ep93xx_decomp_setup(void)
+{
+	if (machine_is_ts72xx())
+		ts72xx_watchdog_disable();
+
+	if (machine_is_edb9301() ||
+	    machine_is_edb9302() ||
+	    machine_is_edb9302a() ||
+	    machine_is_edb9307() ||
+	    machine_is_edb9307a() ||
+	    machine_is_edb9312() ||
+	    machine_is_edb9315() ||
+	    machine_is_edb9315a() ||
+	    machine_is_ts72xx() ||
+	    machine_is_bk3() ||
+	    machine_is_vision_ep9307())
+		ep93xx_ethernet_reset();
+}
diff --git a/arch/arm/boot/compressed/misc.c b/arch/arm/boot/compressed/misc.c
new file mode 100644
index 000000000..6c41b2705
--- /dev/null
+++ b/arch/arm/boot/compressed/misc.c
@@ -0,0 +1,160 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * misc.c
+ * 
+ * This is a collection of several routines from gzip-1.0.3 
+ * adapted for Linux.
+ *
+ * malloc by Hannu Savolainen 1993 and Matthias Urlichs 1994
+ *
+ * Modified for ARM Linux by Russell King
+ *
+ * Nicolas Pitre <nico@visuaide.com>  1999/04/14 :
+ *  For this code to run directly from Flash, all constant variables must
+ *  be marked with 'const' and all other variables initialized at run-time 
+ *  only.  This way all non constant variables will end up in the bss segment,
+ *  which should point to addresses in RAM and cleared to 0 on start.
+ *  This allows for a much quicker boot time.
+ */
+
+unsigned int __machine_arch_type;
+
+#include <linux/compiler.h>	/* for inline */
+#include <linux/types.h>
+#include <linux/linkage.h>
+#include "misc.h"
+#ifdef CONFIG_ARCH_EP93XX
+#include "misc-ep93xx.h"
+#endif
+
+static void putstr(const char *ptr);
+
+#include CONFIG_UNCOMPRESS_INCLUDE
+
+#ifdef CONFIG_DEBUG_ICEDCC
+
+#if defined(CONFIG_CPU_V6) || defined(CONFIG_CPU_V6K) || defined(CONFIG_CPU_V7)
+
+static void icedcc_putc(int ch)
+{
+	int status, i = 0x4000000;
+
+	do {
+		if (--i < 0)
+			return;
+
+		asm volatile ("mrc p14, 0, %0, c0, c1, 0" : "=r" (status));
+	} while (status & (1 << 29));
+
+	asm("mcr p14, 0, %0, c0, c5, 0" : : "r" (ch));
+}
+
+
+#elif defined(CONFIG_CPU_XSCALE)
+
+static void icedcc_putc(int ch)
+{
+	int status, i = 0x4000000;
+
+	do {
+		if (--i < 0)
+			return;
+
+		asm volatile ("mrc p14, 0, %0, c14, c0, 0" : "=r" (status));
+	} while (status & (1 << 28));
+
+	asm("mcr p14, 0, %0, c8, c0, 0" : : "r" (ch));
+}
+
+#else
+
+static void icedcc_putc(int ch)
+{
+	int status, i = 0x4000000;
+
+	do {
+		if (--i < 0)
+			return;
+
+		asm volatile ("mrc p14, 0, %0, c0, c0, 0" : "=r" (status));
+	} while (status & 2);
+
+	asm("mcr p14, 0, %0, c1, c0, 0" : : "r" (ch));
+}
+
+#endif
+
+#define putc(ch)	icedcc_putc(ch)
+#endif
+
+static void putstr(const char *ptr)
+{
+	char c;
+
+	while ((c = *ptr++) != '\0') {
+		if (c == '\n')
+			putc('\r');
+		putc(c);
+	}
+
+	flush();
+}
+
+/*
+ * gzip declarations
+ */
+unsigned char *output_data;
+
+unsigned long free_mem_ptr;
+unsigned long free_mem_end_ptr;
+
+#ifndef arch_error
+#define arch_error(x)
+#endif
+
+void error(char *x)
+{
+	arch_error(x);
+
+	putstr("\n\n");
+	putstr(x);
+	putstr("\n\n -- System halted");
+
+	while(1);	/* Halt */
+}
+
+asmlinkage void __div0(void)
+{
+	error("Attempting division by 0!");
+}
+
+void
+decompress_kernel(unsigned long output_start, unsigned long free_mem_ptr_p,
+		unsigned long free_mem_ptr_end_p,
+		int arch_id)
+{
+	int ret;
+
+	output_data		= (unsigned char *)output_start;
+	free_mem_ptr		= free_mem_ptr_p;
+	free_mem_end_ptr	= free_mem_ptr_end_p;
+	__machine_arch_type	= arch_id;
+
+#ifdef CONFIG_ARCH_EP93XX
+	ep93xx_decomp_setup();
+#endif
+	arch_decomp_setup();
+
+	putstr("Uncompressing Linux...");
+	ret = do_decompress(input_data, input_data_end - input_data,
+			    output_data, error);
+	if (ret)
+		error("decompressor returned an error");
+	else
+		putstr(" done, booting the kernel.\n");
+}
+
+void __fortify_panic(const u8 reason, size_t avail, size_t size)
+{
+	error("detected buffer overflow");
+}
diff --git a/arch/arm/boot/compressed/misc.h b/arch/arm/boot/compressed/misc.h
new file mode 100644
index 000000000..8c73940b5
--- /dev/null
+++ b/arch/arm/boot/compressed/misc.h
@@ -0,0 +1,21 @@
+#ifndef MISC_H
+#define MISC_H
+
+#include <linux/compiler.h>
+
+void error(char *x) __noreturn;
+extern unsigned long free_mem_ptr;
+extern unsigned long free_mem_end_ptr;
+void __div0(void);
+void
+decompress_kernel(unsigned long output_start, unsigned long free_mem_ptr_p,
+		  unsigned long free_mem_ptr_end_p, int arch_id);
+void __fortify_panic(const u8 reason, size_t avail, size_t size);
+int atags_to_fdt(void *atag_list, void *fdt, int total_space);
+uint32_t fdt_check_mem_start(uint32_t mem_start, const void *fdt);
+int do_decompress(u8 *input, int len, u8 *output, void (*error)(char *x));
+
+extern char input_data[];
+extern char input_data_end[];
+
+#endif
diff --git a/arch/arm/boot/compressed/piggy.S b/arch/arm/boot/compressed/piggy.S
new file mode 100644
index 000000000..27577644e
--- /dev/null
+++ b/arch/arm/boot/compressed/piggy.S
@@ -0,0 +1,7 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+	.section .piggydata, "a"
+	.globl	input_data
+input_data:
+	.incbin	"arch/arm/boot/compressed/piggy_data"
+	.globl	input_data_end
+input_data_end:
diff --git a/arch/arm/boot/compressed/string.c b/arch/arm/boot/compressed/string.c
new file mode 100644
index 000000000..fcc678fce
--- /dev/null
+++ b/arch/arm/boot/compressed/string.c
@@ -0,0 +1,162 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * arch/arm/boot/compressed/string.c
+ *
+ * Small subset of simple string routines
+ */
+
+#define __NO_FORTIFY
+#include <linux/string.h>
+
+/*
+ * The decompressor is built without KASan but uses the same redirects as the
+ * rest of the kernel when CONFIG_KASAN is enabled, defining e.g. memcpy()
+ * to __memcpy() but since we are not linking with the main kernel string
+ * library in the decompressor, that will lead to link failures.
+ *
+ * Undefine KASan's versions, define the wrapped functions and alias them to
+ * the right names so that when e.g. __memcpy() appear in the code, it will
+ * still be linked to this local version of memcpy().
+ */
+#ifdef CONFIG_KASAN
+#undef memcpy
+#undef memmove
+#undef memset
+void *__memcpy(void *__dest, __const void *__src, size_t __n) __alias(memcpy);
+void *__memmove(void *__dest, __const void *__src, size_t count) __alias(memmove);
+void *__memset(void *s, int c, size_t count) __alias(memset);
+#endif
+
+void *memcpy(void *__dest, __const void *__src, size_t __n)
+{
+	int i = 0;
+	unsigned char *d = (unsigned char *)__dest, *s = (unsigned char *)__src;
+
+	for (i = __n >> 3; i > 0; i--) {
+		*d++ = *s++;
+		*d++ = *s++;
+		*d++ = *s++;
+		*d++ = *s++;
+		*d++ = *s++;
+		*d++ = *s++;
+		*d++ = *s++;
+		*d++ = *s++;
+	}
+
+	if (__n & 1 << 2) {
+		*d++ = *s++;
+		*d++ = *s++;
+		*d++ = *s++;
+		*d++ = *s++;
+	}
+
+	if (__n & 1 << 1) {
+		*d++ = *s++;
+		*d++ = *s++;
+	}
+
+	if (__n & 1)
+		*d++ = *s++;
+
+	return __dest;
+}
+
+void *memmove(void *__dest, __const void *__src, size_t count)
+{
+	unsigned char *d = __dest;
+	const unsigned char *s = __src;
+
+	if (__dest == __src)
+		return __dest;
+
+	if (__dest < __src)
+		return memcpy(__dest, __src, count);
+
+	while (count--)
+		d[count] = s[count];
+	return __dest;
+}
+
+size_t strlen(const char *s)
+{
+	const char *sc = s;
+
+	while (*sc != '\0')
+		sc++;
+	return sc - s;
+}
+
+size_t strnlen(const char *s, size_t count)
+{
+	const char *sc;
+
+	for (sc = s; count-- && *sc != '\0'; ++sc)
+		/* nothing */;
+	return sc - s;
+}
+
+int memcmp(const void *cs, const void *ct, size_t count)
+{
+	const unsigned char *su1 = cs, *su2 = ct, *end = su1 + count;
+	int res = 0;
+
+	while (su1 < end) {
+		res = *su1++ - *su2++;
+		if (res)
+			break;
+	}
+	return res;
+}
+
+int strcmp(const char *cs, const char *ct)
+{
+	unsigned char c1, c2;
+	int res = 0;
+
+	do {
+		c1 = *cs++;
+		c2 = *ct++;
+		res = c1 - c2;
+		if (res)
+			break;
+	} while (c1);
+	return res;
+}
+
+void *memchr(const void *s, int c, size_t count)
+{
+	const unsigned char *p = s;
+
+	while (count--)
+		if ((unsigned char)c == *p++)
+			return (void *)(p - 1);
+	return NULL;
+}
+
+char *strchr(const char *s, int c)
+{
+	while (*s != (char)c)
+		if (*s++ == '\0')
+			return NULL;
+	return (char *)s;
+}
+
+char *strrchr(const char *s, int c)
+{
+	const char *last = NULL;
+	do {
+		if (*s == (char)c)
+			last = s;
+	} while (*s++);
+	return (char *)last;
+}
+
+#undef memset
+
+void *memset(void *s, int c, size_t count)
+{
+	char *xs = s;
+	while (count--)
+		*xs++ = c;
+	return s;
+}
diff --git a/arch/arm/boot/compressed/vmlinux.lds.S b/arch/arm/boot/compressed/vmlinux.lds.S
new file mode 100644
index 000000000..2d916647d
--- /dev/null
+++ b/arch/arm/boot/compressed/vmlinux.lds.S
@@ -0,0 +1,144 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ *  Copyright (C) 2000 Russell King
+ */
+#include <asm/vmlinux.lds.h>
+
+#ifdef CONFIG_CPU_ENDIAN_BE8
+#define ZIMAGE_MAGIC(x) ( (((x) >> 24) & 0x000000ff) | \
+			  (((x) >>  8) & 0x0000ff00) | \
+			  (((x) <<  8) & 0x00ff0000) | \
+			  (((x) << 24) & 0xff000000) )
+#else
+#define ZIMAGE_MAGIC(x) (x)
+#endif
+
+OUTPUT_ARCH(arm)
+ENTRY(_start)
+SECTIONS
+{
+  /DISCARD/ : {
+    COMMON_DISCARDS
+    *(.ARM.exidx*)
+    *(.ARM.extab*)
+    *(.modinfo)
+    *(.note.*)
+    *(.rel.*)
+    *(.printk_index)
+    /*
+     * Discard any r/w data - this produces a link error if we have any,
+     * which is required for PIC decompression.  Local data generates
+     * GOTOFF relocations, which prevents it being relocated independently
+     * of the text/got segments.
+     */
+    *(.data)
+  }
+
+  . = TEXT_START;
+  _text = .;
+
+  .text : {
+    _start = .;
+    *(.start)
+    *(.text)
+    *(.text.*)
+    ARM_STUBS_TEXT
+  }
+  .table : ALIGN(4) {
+    _table_start = .;
+    LONG(ZIMAGE_MAGIC(6))
+    LONG(ZIMAGE_MAGIC(0x5a534c4b))
+    LONG(ZIMAGE_MAGIC(__piggy_size_addr - _start))
+    LONG(ZIMAGE_MAGIC(_kernel_bss_size))
+    LONG(ZIMAGE_MAGIC(TEXT_OFFSET))
+    LONG(ZIMAGE_MAGIC(MALLOC_SIZE))
+    LONG(0)
+    _table_end = .;
+  }
+  .rodata : {
+    *(.rodata)
+    *(.rodata.*)
+    *(.data.rel.ro)
+    *(.data.rel.ro.*)
+  }
+  .piggydata : {
+    *(.piggydata)
+    __piggy_size_addr = . - 4;
+  }
+
+  . = ALIGN(4);
+  _etext = .;
+
+  .got.plt		: { *(.got.plt) }
+#ifndef CONFIG_EFI_STUB
+  _got_start = .;
+  .got			: { *(.got) }
+  _got_end = .;
+#endif
+
+  /* ensure the zImage file size is always a multiple of 64 bits */
+  /* (without a dummy byte, ld just ignores the empty section) */
+  .pad			: { BYTE(0); . = ALIGN(8); }
+
+#ifdef CONFIG_EFI_STUB
+  .data : ALIGN(4096) {
+    __pecoff_data_start = .;
+    _got_start = .;
+    *(.got)
+    _got_end = .;
+    /*
+     * The EFI stub always executes from RAM, and runs strictly before the
+     * decompressor, so we can make an exception for its r/w data, and keep it
+     */
+    *(.data.efistub .bss.efistub)
+    __pecoff_data_end = .;
+
+    /*
+     * PE/COFF mandates a file size which is a multiple of 512 bytes if the
+     * section size equals or exceeds 4 KB
+     */
+    . = ALIGN(512);
+  }
+  __pecoff_data_rawsize = . - ADDR(.data);
+#endif
+
+  _edata = .;
+
+  /*
+   * The image_end section appears after any additional loadable sections
+   * that the linker may decide to insert in the binary image.  Having
+   * this symbol allows further debug in the near future.
+   */
+  .image_end (NOLOAD) : {
+    /*
+     * EFI requires that the image is aligned to 512 bytes, and appended
+     * DTB requires that we know where the end of the image is.  Ensure
+     * that both are satisfied by ensuring that there are no additional
+     * sections emitted into the decompressor image.
+     */
+    _edata_real = .;
+  }
+
+  _magic_sig = ZIMAGE_MAGIC(0x016f2818);
+  _magic_start = ZIMAGE_MAGIC(_start);
+  _magic_end = ZIMAGE_MAGIC(_edata);
+  _magic_table = ZIMAGE_MAGIC(_table_start - _start);
+
+  . = BSS_START;
+  __bss_start = .;
+  .bss			: { *(.bss .bss.*) }
+  _end = .;
+
+  . = ALIGN(8);		/* the stack must be 64-bit aligned */
+  .stack		: { *(.stack) }
+
+  PROVIDE(__pecoff_data_size = ALIGN(512) - ADDR(.data));
+  PROVIDE(__pecoff_end = ALIGN(512));
+
+  STABS_DEBUG
+  DWARF_DEBUG
+  ARM_DETAILS
+
+  ARM_ASSERTS
+}
+ASSERT(_edata_real == _edata, "error: zImage file size is incorrect");