Add files via upload

arch/arm/lib/Makefile

@@ -0,0 +1,44 @@
+#
+# linux/arch/arm/lib/Makefile
+#
+# Copyright (C) 1995-2000 Russell King
+#
+
+lib-y		:= backtrace.o changebit.o csumipv6.o csumpartial.o   \
+		   csumpartialcopy.o csumpartialcopyuser.o clearbit.o \
+		   delay.o delay-loop.o findbit.o memchr.o memcpy.o   \
+		   memmove.o memset.o memzero.o setbit.o              \
+		   strchr.o strrchr.o                                 \
+		   testchangebit.o testclearbit.o testsetbit.o        \
+		   ashldi3.o ashrdi3.o lshrdi3.o muldi3.o             \
+		   ucmpdi2.o lib1funcs.o div64.o                      \
+		   io-readsb.o io-writesb.o io-readsl.o io-writesl.o  \
+		   call_with_stack.o bswapsdi2.o
+
+mmu-y		:= clear_user.o copy_page.o getuser.o putuser.o       \
+		   copy_from_user.o copy_to_user.o
+
+# using lib_ here won't override already available weak symbols
+obj-$(CONFIG_UACCESS_WITH_MEMCPY) += uaccess_with_memcpy.o
+
+lib-$(CONFIG_MMU) += $(mmu-y)
+
+ifeq ($(CONFIG_CPU_32v3),y)
+  lib-y	+= io-readsw-armv3.o io-writesw-armv3.o
+else
+  lib-y	+= io-readsw-armv4.o io-writesw-armv4.o
+endif
+
+ifeq ($(CONFIG_ARCH_RPC),y)
+  lib-y				+= ecard.o io-acorn.o floppydma.o
+  AFLAGS_delay-loop.o		+= -march=armv4
+endif
+
+$(obj)/csumpartialcopy.o:	$(obj)/csumpartialcopygeneric.S
+$(obj)/csumpartialcopyuser.o:	$(obj)/csumpartialcopygeneric.S
+
+ifeq ($(CONFIG_KERNEL_MODE_NEON),y)
+  NEON_FLAGS			:= -mfloat-abi=softfp -mfpu=neon
+  CFLAGS_xor-neon.o		+= $(NEON_FLAGS)
+  obj-$(CONFIG_XOR_BLOCKS)	+= xor-neon.o
+endif

arch/arm/lib/ashldi3.S

@@ -0,0 +1,54 @@
+/* Copyright 1995, 1996, 1998, 1999, 2000, 2003, 2004, 2005
+   Free Software Foundation, Inc.
+
+This file is free software; you can redistribute it and/or modify it
+under the terms of the GNU General Public License as published by the
+Free Software Foundation; either version 2, or (at your option) any
+later version.
+
+In addition to the permissions in the GNU General Public License, the
+Free Software Foundation gives you unlimited permission to link the
+compiled version of this file into combinations with other programs,
+and to distribute those combinations without any restriction coming
+from the use of this file.  (The General Public License restrictions
+do apply in other respects; for example, they cover modification of
+the file, and distribution when not linked into a combine
+executable.)
+
+This file is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with this program; see the file COPYING.  If not, write to
+the Free Software Foundation, 51 Franklin Street, Fifth Floor,
+Boston, MA 02110-1301, USA.  */
+
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+#ifdef __ARMEB__
+#define al r1
+#define ah r0
+#else
+#define al r0
+#define ah r1
+#endif
+
+ENTRY(__ashldi3)
+ENTRY(__aeabi_llsl)
+
+	subs	r3, r2, #32
+	rsb	ip, r2, #32
+	movmi	ah, ah, lsl r2
+	movpl	ah, al, lsl r3
+ ARM(	orrmi	ah, ah, al, lsr ip	)
+ THUMB(	lsrmi	r3, al, ip		)
+ THUMB(	orrmi	ah, ah, r3		)
+	mov	al, al, lsl r2
+	ret	lr
+
+ENDPROC(__ashldi3)
+ENDPROC(__aeabi_llsl)

arch/arm/lib/ashrdi3.S

@@ -0,0 +1,54 @@
+/* Copyright 1995, 1996, 1998, 1999, 2000, 2003, 2004, 2005
+   Free Software Foundation, Inc.
+
+This file is free software; you can redistribute it and/or modify it
+under the terms of the GNU General Public License as published by the
+Free Software Foundation; either version 2, or (at your option) any
+later version.
+
+In addition to the permissions in the GNU General Public License, the
+Free Software Foundation gives you unlimited permission to link the
+compiled version of this file into combinations with other programs,
+and to distribute those combinations without any restriction coming
+from the use of this file.  (The General Public License restrictions
+do apply in other respects; for example, they cover modification of
+the file, and distribution when not linked into a combine
+executable.)
+
+This file is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with this program; see the file COPYING.  If not, write to
+the Free Software Foundation, 51 Franklin Street, Fifth Floor,
+Boston, MA 02110-1301, USA.  */
+
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+#ifdef __ARMEB__
+#define al r1
+#define ah r0
+#else
+#define al r0
+#define ah r1
+#endif
+
+ENTRY(__ashrdi3)
+ENTRY(__aeabi_lasr)
+
+	subs	r3, r2, #32
+	rsb	ip, r2, #32
+	movmi	al, al, lsr r2
+	movpl	al, ah, asr r3
+ ARM(	orrmi	al, al, ah, lsl ip	)
+ THUMB(	lslmi	r3, ah, ip		)
+ THUMB(	orrmi	al, al, r3		)
+	mov	ah, ah, asr r2
+	ret	lr
+
+ENDPROC(__ashrdi3)
+ENDPROC(__aeabi_lasr)

arch/arm/lib/backtrace.S

@@ -0,0 +1,140 @@
+/*
+ *  linux/arch/arm/lib/backtrace.S
+ *
+ *  Copyright (C) 1995, 1996 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * 27/03/03 Ian Molton Clean up CONFIG_CPU
+ *
+ */
+#include <linux/kern_levels.h>
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+.extern _stext, _etext
+__stext:
+	.word _stext
+__etext:
+	.word _etext
+		.text
+
+@ fp is 0 or stack frame
+
+#define frame	r4
+#define sv_fp	r5
+#define sv_pc	r6
+#define mask	r7
+#define offset	r8
+
+ENTRY(c_backtrace)
+
+#if !defined(CONFIG_FRAME_POINTER) || !defined(CONFIG_PRINTK)
+		ret	lr
+ENDPROC(c_backtrace)
+#else
+		stmfd	sp!, {r4 - r8, lr}	@ Save an extra register so we have a location...
+		movs	frame, r0		@ if frame pointer is zero
+		beq	no_frame		@ we have no stack frames
+
+		tst	r1, #0x10		@ 26 or 32-bit mode?
+ ARM(		moveq	mask, #0xfc000003	)
+ THUMB(		moveq	mask, #0xfc000000	)
+ THUMB(		orreq	mask, #0x03		)
+		movne	mask, #0		@ mask for 32-bit
+
+1:		stmfd	sp!, {pc}		@ calculate offset of PC stored
+		ldr	r0, [sp], #4		@ by stmfd for this CPU
+		adr	r1, 1b
+		sub	offset, r0, r1
+
+/*
+ * Stack frame layout:
+ *             optionally saved caller registers (r4 - r10)
+ *             saved fp
+ *             saved sp
+ *             saved lr
+ *    frame => saved pc
+ *             optionally saved arguments (r0 - r3)
+ * saved sp => <next word>
+ *
+ * Functions start with the following code sequence:
+ *                  mov   ip, sp
+ *                  stmfd sp!, {r0 - r3} (optional)
+ * corrected pc =>  stmfd sp!, {..., fp, ip, lr, pc}
+ */
+for_each_frame:	tst	frame, mask		@ Check for address exceptions
+		bne	no_frame
+
+1001:		ldr	sv_pc, [frame, #0]	@ get saved pc
+1002:		ldr	sv_fp, [frame, #-12]	@ get saved fp
+
+		sub	sv_pc, sv_pc, offset	@ Correct PC for prefetching
+		bic	sv_pc, sv_pc, mask	@ mask PC/LR for the mode
+
+		ldr	r0, __stext
+		cmp	sv_pc, r0
+		blo	.badpc
+		ldr	r0, __etext
+		cmp	sv_pc, r0
+		bhs	.badpc
+
+1003:		ldr	r2, [sv_pc, #-4]	@ if stmfd sp!, {args} exists,
+		ldr	r3, .Ldsi+4		@ adjust saved 'pc' back one
+		teq	r3, r2, lsr #10		@ instruction
+		subne	r0, sv_pc, #4		@ allow for mov
+		subeq	r0, sv_pc, #8		@ allow for mov + stmia
+
+		ldr	r1, [frame, #-4]	@ get saved lr
+		mov	r2, frame
+		bic	r1, r1, mask		@ mask PC/LR for the mode
+		bl	dump_backtrace_entry
+
+		ldr	r1, [sv_pc, #-4]	@ if stmfd sp!, {args} exists,
+		ldr	r3, .Ldsi+4
+		teq	r3, r1, lsr #11
+		ldreq	r0, [frame, #-8]	@ get sp
+		subeq	r0, r0, #4		@ point at the last arg
+		bleq	dump_backtrace_stm	@ dump saved registers
+
+1004:		ldr	r1, [sv_pc, #0]		@ if stmfd sp!, {..., fp, ip, lr, pc}
+		ldr	r3, .Ldsi		@ instruction exists,
+		teq	r3, r1, lsr #11
+		subeq	r0, frame, #16
+		bleq	dump_backtrace_stm	@ dump saved registers
+
+		teq	sv_fp, #0		@ zero saved fp means
+		beq	no_frame		@ no further frames
+
+		cmp	sv_fp, frame		@ next frame must be
+		mov	frame, sv_fp		@ above the current frame
+		bhi	for_each_frame
+
+1006:		adr	r0, .Lbad
+		mov	r1, frame
+		bl	printk
+no_frame:	ldmfd	sp!, {r4 - r8, pc}
+ENDPROC(c_backtrace)
+		
+		.pushsection __ex_table,"a"
+		.align	3
+		.long	1001b, 1006b
+		.long	1002b, 1006b
+		.long	1003b, 1006b
+		.long	1004b, 1006b
+		.popsection
+
+.badpc:
+		adr	r0, .Lbadpc
+		mov	r1, sv_pc
+		bl	printk
+		b	no_frame
+.Lbad:		.asciz	"Backtrace aborted due to bad frame pointer <%p>\n"
+		.align
+.Lbadpc:	.asciz	"Backtrace aborted due to bad pc <%p>\n"
+		.align
+.Ldsi:		.word	0xe92dd800 >> 11	@ stmfd sp!, {... fp, ip, lr, pc}
+		.word	0xe92d0000 >> 11	@ stmfd sp!, {}
+
+#endif

arch/arm/lib/bitops.h

@@ -0,0 +1,106 @@
+#include <asm/assembler.h>
+#include <asm/unwind.h>
+
+#if __LINUX_ARM_ARCH__ >= 6
+	.macro	bitop, name, instr
+ENTRY(	\name		)
+UNWIND(	.fnstart	)
+	ands	ip, r1, #3
+	strneb	r1, [ip]		@ assert word-aligned
+	mov	r2, #1
+	and	r3, r0, #31		@ Get bit offset
+	mov	r0, r0, lsr #5
+	add	r1, r1, r0, lsl #2	@ Get word offset
+#if __LINUX_ARM_ARCH__ >= 7 && defined(CONFIG_SMP)
+	.arch_extension	mp
+	ALT_SMP(W(pldw)	[r1])
+	ALT_UP(W(nop))
+#endif
+	mov	r3, r2, lsl r3
+1:	ldrex	r2, [r1]
+	\instr	r2, r2, r3
+	strex	r0, r2, [r1]
+	cmp	r0, #0
+	bne	1b
+	bx	lr
+UNWIND(	.fnend		)
+ENDPROC(\name		)
+	.endm
+
+	.macro	testop, name, instr, store
+ENTRY(	\name		)
+UNWIND(	.fnstart	)
+	ands	ip, r1, #3
+	strneb	r1, [ip]		@ assert word-aligned
+	mov	r2, #1
+	and	r3, r0, #31		@ Get bit offset
+	mov	r0, r0, lsr #5
+	add	r1, r1, r0, lsl #2	@ Get word offset
+	mov	r3, r2, lsl r3		@ create mask
+	smp_dmb
+#if __LINUX_ARM_ARCH__ >= 7 && defined(CONFIG_SMP)
+	.arch_extension	mp
+	ALT_SMP(W(pldw)	[r1])
+	ALT_UP(W(nop))
+#endif
+1:	ldrex	r2, [r1]
+	ands	r0, r2, r3		@ save old value of bit
+	\instr	r2, r2, r3		@ toggle bit
+	strex	ip, r2, [r1]
+	cmp	ip, #0
+	bne	1b
+	smp_dmb
+	cmp	r0, #0
+	movne	r0, #1
+2:	bx	lr
+UNWIND(	.fnend		)
+ENDPROC(\name		)
+	.endm
+#else
+	.macro	bitop, name, instr
+ENTRY(	\name		)
+UNWIND(	.fnstart	)
+	ands	ip, r1, #3
+	strneb	r1, [ip]		@ assert word-aligned
+	and	r2, r0, #31
+	mov	r0, r0, lsr #5
+	mov	r3, #1
+	mov	r3, r3, lsl r2
+	save_and_disable_irqs ip
+	ldr	r2, [r1, r0, lsl #2]
+	\instr	r2, r2, r3
+	str	r2, [r1, r0, lsl #2]
+	restore_irqs ip
+	ret	lr
+UNWIND(	.fnend		)
+ENDPROC(\name		)
+	.endm
+
+/**
+ * testop - implement a test_and_xxx_bit operation.
+ * @instr: operational instruction
+ * @store: store instruction
+ *
+ * Note: we can trivially conditionalise the store instruction
+ * to avoid dirtying the data cache.
+ */
+	.macro	testop, name, instr, store
+ENTRY(	\name		)
+UNWIND(	.fnstart	)
+	ands	ip, r1, #3
+	strneb	r1, [ip]		@ assert word-aligned
+	and	r3, r0, #31
+	mov	r0, r0, lsr #5
+	save_and_disable_irqs ip
+	ldr	r2, [r1, r0, lsl #2]!
+	mov	r0, #1
+	tst	r2, r0, lsl r3
+	\instr	r2, r2, r0, lsl r3
+	\store	r2, [r1]
+	moveq	r0, #0
+	restore_irqs ip
+	ret	lr
+UNWIND(	.fnend		)
+ENDPROC(\name		)
+	.endm
+#endif

arch/arm/lib/bswapsdi2.S

@@ -0,0 +1,37 @@
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+#if __LINUX_ARM_ARCH__ >= 6
+ENTRY(__bswapsi2)
+	rev r0, r0
+	bx lr
+ENDPROC(__bswapsi2)
+
+ENTRY(__bswapdi2)
+	rev r3, r0
+	rev r0, r1
+	mov r1, r3
+	bx lr
+ENDPROC(__bswapdi2)
+#else
+ENTRY(__bswapsi2)
+	eor r3, r0, r0, ror #16
+	mov r3, r3, lsr #8
+	bic r3, r3, #0xff00
+	eor r0, r3, r0, ror #8
+	ret lr
+ENDPROC(__bswapsi2)
+
+ENTRY(__bswapdi2)
+	mov ip, r1
+	eor r3, ip, ip, ror #16
+	eor r1, r0, r0, ror #16
+	mov r1, r1, lsr #8
+	mov r3, r3, lsr #8
+	bic r3, r3, #0xff00
+	bic r1, r1, #0xff00
+	eor r1, r1, r0, ror #8
+	eor r0, r3, ip, ror #8
+	ret lr
+ENDPROC(__bswapdi2)
+#endif

arch/arm/lib/call_with_stack.S

@@ -0,0 +1,44 @@
+/*
+ * arch/arm/lib/call_with_stack.S
+ *
+ * Copyright (C) 2011 ARM Ltd.
+ * Written by Will Deacon <will.deacon@arm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+/*
+ * void call_with_stack(void (*fn)(void *), void *arg, void *sp)
+ *
+ * Change the stack to that pointed at by sp, then invoke fn(arg) with
+ * the new stack.
+ */
+ENTRY(call_with_stack)
+	str	sp, [r2, #-4]!
+	str	lr, [r2, #-4]!
+
+	mov	sp, r2
+	mov	r2, r0
+	mov	r0, r1
+
+	badr	lr, 1f
+	ret	r2
+
+1:	ldr	lr, [sp]
+	ldr	sp, [sp, #4]
+	ret	lr
+ENDPROC(call_with_stack)

arch/arm/lib/changebit.S

@@ -0,0 +1,15 @@
+/*
+ *  linux/arch/arm/lib/changebit.S
+ *
+ *  Copyright (C) 1995-1996 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include "bitops.h"
+                .text
+
+bitop	_change_bit, eor

arch/arm/lib/clear_user.S

@@ -0,0 +1,58 @@
+/*
+ *  linux/arch/arm/lib/clear_user.S
+ *
+ *  Copyright (C) 1995, 1996,1997,1998 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include <asm/unwind.h>
+
+		.text
+
+/* Prototype: unsigned long arm_clear_user(void *addr, size_t sz)
+ * Purpose  : clear some user memory
+ * Params   : addr - user memory address to clear
+ *          : sz   - number of bytes to clear
+ * Returns  : number of bytes NOT cleared
+ */
+ENTRY(__clear_user_std)
+WEAK(arm_clear_user)
+UNWIND(.fnstart)
+UNWIND(.save {r1, lr})
+		stmfd	sp!, {r1, lr}
+		mov	r2, #0
+		cmp	r1, #4
+		blt	2f
+		ands	ip, r0, #3
+		beq	1f
+		cmp	ip, #2
+		strusr	r2, r0, 1
+		strusr	r2, r0, 1, le
+		strusr	r2, r0, 1, lt
+		rsb	ip, ip, #4
+		sub	r1, r1, ip		@  7  6  5  4  3  2  1
+1:		subs	r1, r1, #8		@ -1 -2 -3 -4 -5 -6 -7
+		strusr	r2, r0, 4, pl, rept=2
+		bpl	1b
+		adds	r1, r1, #4		@  3  2  1  0 -1 -2 -3
+		strusr	r2, r0, 4, pl
+2:		tst	r1, #2			@ 1x 1x 0x 0x 1x 1x 0x
+		strusr	r2, r0, 1, ne, rept=2
+		tst	r1, #1			@ x1 x0 x1 x0 x1 x0 x1
+		it	ne			@ explicit IT needed for the label
+USER(		strnebt	r2, [r0])
+		mov	r0, #0
+		ldmfd	sp!, {r1, pc}
+UNWIND(.fnend)
+ENDPROC(arm_clear_user)
+ENDPROC(__clear_user_std)
+
+		.pushsection .text.fixup,"ax"
+		.align	0
+9001:		ldmfd	sp!, {r0, pc}
+		.popsection
+

arch/arm/lib/clearbit.S

@@ -0,0 +1,15 @@
+/*
+ *  linux/arch/arm/lib/clearbit.S
+ *
+ *  Copyright (C) 1995-1996 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include "bitops.h"
+                .text
+
+bitop	_clear_bit, bic

arch/arm/lib/copy_from_user.S

@@ -0,0 +1,106 @@
+/*
+ *  linux/arch/arm/lib/copy_from_user.S
+ *
+ *  Author:	Nicolas Pitre
+ *  Created:	Sep 29, 2005
+ *  Copyright:	MontaVista Software, Inc.
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include <asm/unwind.h>
+
+/*
+ * Prototype:
+ *
+ *	size_t arm_copy_from_user(void *to, const void *from, size_t n)
+ *
+ * Purpose:
+ *
+ *	copy a block to kernel memory from user memory
+ *
+ * Params:
+ *
+ *	to = kernel memory
+ *	from = user memory
+ *	n = number of bytes to copy
+ *
+ * Return value:
+ *
+ *	Number of bytes NOT copied.
+ */
+
+#ifndef CONFIG_THUMB2_KERNEL
+#define LDR1W_SHIFT	0
+#else
+#define LDR1W_SHIFT	1
+#endif
+#define STR1W_SHIFT	0
+
+	.macro ldr1w ptr reg abort
+	ldrusr	\reg, \ptr, 4, abort=\abort
+	.endm
+
+	.macro ldr4w ptr reg1 reg2 reg3 reg4 abort
+	ldr1w \ptr, \reg1, \abort
+	ldr1w \ptr, \reg2, \abort
+	ldr1w \ptr, \reg3, \abort
+	ldr1w \ptr, \reg4, \abort
+	.endm
+
+	.macro ldr8w ptr reg1 reg2 reg3 reg4 reg5 reg6 reg7 reg8 abort
+	ldr4w \ptr, \reg1, \reg2, \reg3, \reg4, \abort
+	ldr4w \ptr, \reg5, \reg6, \reg7, \reg8, \abort
+	.endm
+
+	.macro ldr1b ptr reg cond=al abort
+	ldrusr	\reg, \ptr, 1, \cond, abort=\abort
+	.endm
+
+	.macro str1w ptr reg abort
+	W(str) \reg, [\ptr], #4
+	.endm
+
+	.macro str8w ptr reg1 reg2 reg3 reg4 reg5 reg6 reg7 reg8 abort
+	stmia \ptr!, {\reg1, \reg2, \reg3, \reg4, \reg5, \reg6, \reg7, \reg8}
+	.endm
+
+	.macro str1b ptr reg cond=al abort
+	str\cond\()b \reg, [\ptr], #1
+	.endm
+
+	.macro enter reg1 reg2
+	mov	r3, #0
+	stmdb	sp!, {r0, r2, r3, \reg1, \reg2}
+	.endm
+
+	.macro usave reg1 reg2
+	UNWIND(	.save {r0, r2, r3, \reg1, \reg2}	)
+	.endm
+
+	.macro exit reg1 reg2
+	add	sp, sp, #8
+	ldmfd	sp!, {r0, \reg1, \reg2}
+	.endm
+
+	.text
+
+ENTRY(arm_copy_from_user)
+
+#include "copy_template.S"
+
+ENDPROC(arm_copy_from_user)
+
+	.pushsection .fixup,"ax"
+	.align 0
+	copy_abort_preamble
+	ldmfd	sp!, {r1, r2, r3}
+	sub	r0, r0, r1
+	rsb	r0, r0, r2
+	copy_abort_end
+	.popsection
+

arch/arm/lib/copy_page.S

@@ -0,0 +1,47 @@
+/*
+ *  linux/arch/arm/lib/copypage.S
+ *
+ *  Copyright (C) 1995-1999 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ *  ASM optimised string functions
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include <asm/asm-offsets.h>
+#include <asm/cache.h>
+
+#define COPY_COUNT (PAGE_SZ / (2 * L1_CACHE_BYTES) PLD( -1 ))
+
+		.text
+		.align	5
+/*
+ * StrongARM optimised copy_page routine
+ * now 1.78bytes/cycle, was 1.60 bytes/cycle (50MHz bus -> 89MB/s)
+ * Note that we probably achieve closer to the 100MB/s target with
+ * the core clock switching.
+ */
+ENTRY(copy_page)
+		stmfd	sp!, {r4, lr}			@	2
+	PLD(	pld	[r1, #0]		)
+	PLD(	pld	[r1, #L1_CACHE_BYTES]		)
+		mov	r2, #COPY_COUNT			@	1
+		ldmia	r1!, {r3, r4, ip, lr}		@	4+1
+1:	PLD(	pld	[r1, #2 * L1_CACHE_BYTES])
+	PLD(	pld	[r1, #3 * L1_CACHE_BYTES])
+2:
+	.rept	(2 * L1_CACHE_BYTES / 16 - 1)
+		stmia	r0!, {r3, r4, ip, lr}		@	4
+		ldmia	r1!, {r3, r4, ip, lr}		@	4
+	.endr
+		subs	r2, r2, #1			@	1
+		stmia	r0!, {r3, r4, ip, lr}		@	4
+		ldmgtia	r1!, {r3, r4, ip, lr}		@	4
+		bgt	1b				@	1
+	PLD(	ldmeqia r1!, {r3, r4, ip, lr}	)
+	PLD(	beq	2b			)
+		ldmfd	sp!, {r4, pc}			@	3
+ENDPROC(copy_page)

arch/arm/lib/copy_template.S

@@ -0,0 +1,297 @@
+/*
+ *  linux/arch/arm/lib/copy_template.s
+ *
+ *  Code template for optimized memory copy functions
+ *
+ *  Author:	Nicolas Pitre
+ *  Created:	Sep 28, 2005
+ *  Copyright:	MontaVista Software, Inc.
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+/*
+ * Theory of operation
+ * -------------------
+ *
+ * This file provides the core code for a forward memory copy used in
+ * the implementation of memcopy(), copy_to_user() and copy_from_user().
+ *
+ * The including file must define the following accessor macros
+ * according to the need of the given function:
+ *
+ * ldr1w ptr reg abort
+ *
+ *	This loads one word from 'ptr', stores it in 'reg' and increments
+ *	'ptr' to the next word. The 'abort' argument is used for fixup tables.
+ *
+ * ldr4w ptr reg1 reg2 reg3 reg4 abort
+ * ldr8w ptr, reg1 reg2 reg3 reg4 reg5 reg6 reg7 reg8 abort
+ *
+ *	This loads four or eight words starting from 'ptr', stores them
+ *	in provided registers and increments 'ptr' past those words.
+ *	The'abort' argument is used for fixup tables.
+ *
+ * ldr1b ptr reg cond abort
+ *
+ *	Similar to ldr1w, but it loads a byte and increments 'ptr' one byte.
+ *	It also must apply the condition code if provided, otherwise the
+ *	"al" condition is assumed by default.
+ *
+ * str1w ptr reg abort
+ * str8w ptr reg1 reg2 reg3 reg4 reg5 reg6 reg7 reg8 abort
+ * str1b ptr reg cond abort
+ *
+ *	Same as their ldr* counterparts, but data is stored to 'ptr' location
+ *	rather than being loaded.
+ *
+ * enter reg1 reg2
+ *
+ *	Preserve the provided registers on the stack plus any additional
+ *	data as needed by the implementation including this code. Called
+ *	upon code entry.
+ *
+ * usave reg1 reg2
+ *
+ *	Unwind annotation macro is corresponding for 'enter' macro.
+ *	It tell unwinder that preserved some provided registers on the stack
+ *	and additional data by a prior 'enter' macro.
+ *
+ * exit reg1 reg2
+ *
+ *	Restore registers with the values previously saved with the
+ *	'preserv' macro. Called upon code termination.
+ *
+ * LDR1W_SHIFT
+ * STR1W_SHIFT
+ *
+ *	Correction to be applied to the "ip" register when branching into
+ *	the ldr1w or str1w instructions (some of these macros may expand to
+ *	than one 32bit instruction in Thumb-2)
+ */
+
+
+	UNWIND(	.fnstart			)
+		enter	r4, lr
+	UNWIND(	.fnend				)
+
+	UNWIND(	.fnstart			)
+		usave	r4, lr			  @ in first stmdb block
+
+		subs	r2, r2, #4
+		blt	8f
+		ands	ip, r0, #3
+	PLD(	pld	[r1, #0]		)
+		bne	9f
+		ands	ip, r1, #3
+		bne	10f
+
+1:		subs	r2, r2, #(28)
+		stmfd	sp!, {r5 - r8}
+	UNWIND(	.fnend				)
+
+	UNWIND(	.fnstart			)
+		usave	r4, lr
+	UNWIND(	.save	{r5 - r8}		) @ in second stmfd block
+		blt	5f
+
+	CALGN(	ands	ip, r0, #31		)
+	CALGN(	rsb	r3, ip, #32		)
+	CALGN(	sbcnes	r4, r3, r2		)  @ C is always set here
+	CALGN(	bcs	2f			)
+	CALGN(	adr	r4, 6f			)
+	CALGN(	subs	r2, r2, r3		)  @ C gets set
+	CALGN(	add	pc, r4, ip		)
+
+	PLD(	pld	[r1, #0]		)
+2:	PLD(	subs	r2, r2, #96		)
+	PLD(	pld	[r1, #28]		)
+	PLD(	blt	4f			)
+	PLD(	pld	[r1, #60]		)
+	PLD(	pld	[r1, #92]		)
+
+3:	PLD(	pld	[r1, #124]		)
+4:		ldr8w	r1, r3, r4, r5, r6, r7, r8, ip, lr, abort=20f
+		subs	r2, r2, #32
+		str8w	r0, r3, r4, r5, r6, r7, r8, ip, lr, abort=20f
+		bge	3b
+	PLD(	cmn	r2, #96			)
+	PLD(	bge	4b			)
+
+5:		ands	ip, r2, #28
+		rsb	ip, ip, #32
+#if LDR1W_SHIFT > 0
+		lsl	ip, ip, #LDR1W_SHIFT
+#endif
+		addne	pc, pc, ip		@ C is always clear here
+		b	7f
+6:
+		.rept	(1 << LDR1W_SHIFT)
+		W(nop)
+		.endr
+		ldr1w	r1, r3, abort=20f
+		ldr1w	r1, r4, abort=20f
+		ldr1w	r1, r5, abort=20f
+		ldr1w	r1, r6, abort=20f
+		ldr1w	r1, r7, abort=20f
+		ldr1w	r1, r8, abort=20f
+		ldr1w	r1, lr, abort=20f
+
+#if LDR1W_SHIFT < STR1W_SHIFT
+		lsl	ip, ip, #STR1W_SHIFT - LDR1W_SHIFT
+#elif LDR1W_SHIFT > STR1W_SHIFT
+		lsr	ip, ip, #LDR1W_SHIFT - STR1W_SHIFT
+#endif
+		add	pc, pc, ip
+		nop
+		.rept	(1 << STR1W_SHIFT)
+		W(nop)
+		.endr
+		str1w	r0, r3, abort=20f
+		str1w	r0, r4, abort=20f
+		str1w	r0, r5, abort=20f
+		str1w	r0, r6, abort=20f
+		str1w	r0, r7, abort=20f
+		str1w	r0, r8, abort=20f
+		str1w	r0, lr, abort=20f
+
+	CALGN(	bcs	2b			)
+
+7:		ldmfd	sp!, {r5 - r8}
+	UNWIND(	.fnend				) @ end of second stmfd block
+
+	UNWIND(	.fnstart			)
+		usave	r4, lr			  @ still in first stmdb block
+8:		movs	r2, r2, lsl #31
+		ldr1b	r1, r3, ne, abort=21f
+		ldr1b	r1, r4, cs, abort=21f
+		ldr1b	r1, ip, cs, abort=21f
+		str1b	r0, r3, ne, abort=21f
+		str1b	r0, r4, cs, abort=21f
+		str1b	r0, ip, cs, abort=21f
+
+		exit	r4, pc
+
+9:		rsb	ip, ip, #4
+		cmp	ip, #2
+		ldr1b	r1, r3, gt, abort=21f
+		ldr1b	r1, r4, ge, abort=21f
+		ldr1b	r1, lr, abort=21f
+		str1b	r0, r3, gt, abort=21f
+		str1b	r0, r4, ge, abort=21f
+		subs	r2, r2, ip
+		str1b	r0, lr, abort=21f
+		blt	8b
+		ands	ip, r1, #3
+		beq	1b
+
+10:		bic	r1, r1, #3
+		cmp	ip, #2
+		ldr1w	r1, lr, abort=21f
+		beq	17f
+		bgt	18f
+	UNWIND(	.fnend				)
+
+
+		.macro	forward_copy_shift pull push
+
+	UNWIND(	.fnstart			)
+		usave	r4, lr			  @ still in first stmdb block
+		subs	r2, r2, #28
+		blt	14f
+
+	CALGN(	ands	ip, r0, #31		)
+	CALGN(	rsb	ip, ip, #32		)
+	CALGN(	sbcnes	r4, ip, r2		)  @ C is always set here
+	CALGN(	subcc	r2, r2, ip		)
+	CALGN(	bcc	15f			)
+
+11:		stmfd	sp!, {r5 - r9}
+	UNWIND(	.fnend				)
+
+	UNWIND(	.fnstart			)
+		usave	r4, lr
+	UNWIND(	.save	{r5 - r9}		) @ in new second stmfd block
+	PLD(	pld	[r1, #0]		)
+	PLD(	subs	r2, r2, #96		)
+	PLD(	pld	[r1, #28]		)
+	PLD(	blt	13f			)
+	PLD(	pld	[r1, #60]		)
+	PLD(	pld	[r1, #92]		)
+
+12:	PLD(	pld	[r1, #124]		)
+13:		ldr4w	r1, r4, r5, r6, r7, abort=19f
+		mov	r3, lr, lspull #\pull
+		subs	r2, r2, #32
+		ldr4w	r1, r8, r9, ip, lr, abort=19f
+		orr	r3, r3, r4, lspush #\push
+		mov	r4, r4, lspull #\pull
+		orr	r4, r4, r5, lspush #\push
+		mov	r5, r5, lspull #\pull
+		orr	r5, r5, r6, lspush #\push
+		mov	r6, r6, lspull #\pull
+		orr	r6, r6, r7, lspush #\push
+		mov	r7, r7, lspull #\pull
+		orr	r7, r7, r8, lspush #\push
+		mov	r8, r8, lspull #\pull
+		orr	r8, r8, r9, lspush #\push
+		mov	r9, r9, lspull #\pull
+		orr	r9, r9, ip, lspush #\push
+		mov	ip, ip, lspull #\pull
+		orr	ip, ip, lr, lspush #\push
+		str8w	r0, r3, r4, r5, r6, r7, r8, r9, ip, , abort=19f
+		bge	12b
+	PLD(	cmn	r2, #96			)
+	PLD(	bge	13b			)
+
+		ldmfd	sp!, {r5 - r9}
+	UNWIND(	.fnend				) @ end of the second stmfd block
+
+	UNWIND(	.fnstart			)
+		usave	r4, lr			  @ still in first stmdb block
+14:		ands	ip, r2, #28
+		beq	16f
+
+15:		mov	r3, lr, lspull #\pull
+		ldr1w	r1, lr, abort=21f
+		subs	ip, ip, #4
+		orr	r3, r3, lr, lspush #\push
+		str1w	r0, r3, abort=21f
+		bgt	15b
+	CALGN(	cmp	r2, #0			)
+	CALGN(	bge	11b			)
+
+16:		sub	r1, r1, #(\push / 8)
+		b	8b
+	UNWIND(	.fnend				)
+
+		.endm
+
+
+		forward_copy_shift	pull=8	push=24
+
+17:		forward_copy_shift	pull=16	push=16
+
+18:		forward_copy_shift	pull=24	push=8
+
+
+/*
+ * Abort preamble and completion macros.
+ * If a fixup handler is required then those macros must surround it.
+ * It is assumed that the fixup code will handle the private part of
+ * the exit macro.
+ */
+
+	.macro	copy_abort_preamble
+19:	ldmfd	sp!, {r5 - r9}
+	b	21f
+20:	ldmfd	sp!, {r5 - r8}
+21:
+	.endm
+
+	.macro	copy_abort_end
+	ldmfd	sp!, {r4, pc}
+	.endm
+

arch/arm/lib/copy_to_user.S

@@ -0,0 +1,111 @@
+/*
+ *  linux/arch/arm/lib/copy_to_user.S
+ *
+ *  Author:	Nicolas Pitre
+ *  Created:	Sep 29, 2005
+ *  Copyright:	MontaVista Software, Inc.
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include <asm/unwind.h>
+
+/*
+ * Prototype:
+ *
+ *	size_t arm_copy_to_user(void *to, const void *from, size_t n)
+ *
+ * Purpose:
+ *
+ *	copy a block to user memory from kernel memory
+ *
+ * Params:
+ *
+ *	to = user memory
+ *	from = kernel memory
+ *	n = number of bytes to copy
+ *
+ * Return value:
+ *
+ *	Number of bytes NOT copied.
+ */
+
+#define LDR1W_SHIFT	0
+#ifndef CONFIG_THUMB2_KERNEL
+#define STR1W_SHIFT	0
+#else
+#define STR1W_SHIFT	1
+#endif
+
+	.macro ldr1w ptr reg abort
+	W(ldr) \reg, [\ptr], #4
+	.endm
+
+	.macro ldr4w ptr reg1 reg2 reg3 reg4 abort
+	ldmia \ptr!, {\reg1, \reg2, \reg3, \reg4}
+	.endm
+
+	.macro ldr8w ptr reg1 reg2 reg3 reg4 reg5 reg6 reg7 reg8 abort
+	ldmia \ptr!, {\reg1, \reg2, \reg3, \reg4, \reg5, \reg6, \reg7, \reg8}
+	.endm
+
+	.macro ldr1b ptr reg cond=al abort
+	ldr\cond\()b \reg, [\ptr], #1
+	.endm
+
+	.macro str1w ptr reg abort
+	strusr	\reg, \ptr, 4, abort=\abort
+	.endm
+
+	.macro str8w ptr reg1 reg2 reg3 reg4 reg5 reg6 reg7 reg8 abort
+	str1w \ptr, \reg1, \abort
+	str1w \ptr, \reg2, \abort
+	str1w \ptr, \reg3, \abort
+	str1w \ptr, \reg4, \abort
+	str1w \ptr, \reg5, \abort
+	str1w \ptr, \reg6, \abort
+	str1w \ptr, \reg7, \abort
+	str1w \ptr, \reg8, \abort
+	.endm
+
+	.macro str1b ptr reg cond=al abort
+	strusr	\reg, \ptr, 1, \cond, abort=\abort
+	.endm
+
+	.macro enter reg1 reg2
+	mov	r3, #0
+	stmdb	sp!, {r0, r2, r3, \reg1, \reg2}
+	.endm
+
+	.macro usave reg1 reg2
+	UNWIND(	.save {r0, r2, r3, \reg1, \reg2}	)
+	.endm
+
+	.macro exit reg1 reg2
+	add	sp, sp, #8
+	ldmfd	sp!, {r0, \reg1, \reg2}
+	.endm
+
+	.text
+
+ENTRY(__copy_to_user_std)
+WEAK(arm_copy_to_user)
+
+#include "copy_template.S"
+
+ENDPROC(arm_copy_to_user)
+ENDPROC(__copy_to_user_std)
+
+	.pushsection .text.fixup,"ax"
+	.align 0
+	copy_abort_preamble
+	ldmfd	sp!, {r1, r2, r3}
+	sub	r0, r0, r1
+	rsb	r0, r0, r2
+	copy_abort_end
+	.popsection
+

arch/arm/lib/csumipv6.S

@@ -0,0 +1,33 @@
+/*
+ *  linux/arch/arm/lib/csumipv6.S
+ *
+ *  Copyright (C) 1995-1998 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+		.text
+
+ENTRY(__csum_ipv6_magic)
+		str	lr, [sp, #-4]!
+		adds	ip, r2, r3
+		ldmia	r1, {r1 - r3, lr}
+		adcs	ip, ip, r1
+		adcs	ip, ip, r2
+		adcs	ip, ip, r3
+		adcs	ip, ip, lr
+		ldmia	r0, {r0 - r3}
+		adcs	r0, ip, r0
+		adcs	r0, r0, r1
+		adcs	r0, r0, r2
+		ldr	r2, [sp, #4]
+		adcs	r0, r0, r3
+		adcs	r0, r0, r2
+		adcs	r0, r0, #0
+		ldmfd	sp!, {pc}
+ENDPROC(__csum_ipv6_magic)
+

arch/arm/lib/csumpartial.S

@@ -0,0 +1,142 @@
+/*
+ *  linux/arch/arm/lib/csumpartial.S
+ *
+ *  Copyright (C) 1995-1998 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+		.text
+
+/*
+ * Function: __u32 csum_partial(const char *src, int len, __u32 sum)
+ * Params  : r0 = buffer, r1 = len, r2 = checksum
+ * Returns : r0 = new checksum
+ */
+
+buf	.req	r0
+len	.req	r1
+sum	.req	r2
+td0	.req	r3
+td1	.req	r4	@ save before use
+td2	.req	r5	@ save before use
+td3	.req	lr
+
+.Lzero:		mov	r0, sum
+		add	sp, sp, #4
+		ldr	pc, [sp], #4
+
+		/*
+		 * Handle 0 to 7 bytes, with any alignment of source and
+		 * destination pointers.  Note that when we get here, C = 0
+		 */
+.Lless8:		teq	len, #0			@ check for zero count
+		beq	.Lzero
+
+		/* we must have at least one byte. */
+		tst	buf, #1			@ odd address?
+		movne	sum, sum, ror #8
+		ldrneb	td0, [buf], #1
+		subne	len, len, #1
+		adcnes	sum, sum, td0, put_byte_1
+
+.Lless4:		tst	len, #6
+		beq	.Lless8_byte
+
+		/* we are now half-word aligned */
+
+.Lless8_wordlp:
+#if __LINUX_ARM_ARCH__ >= 4
+		ldrh	td0, [buf], #2
+		sub	len, len, #2
+#else
+		ldrb	td0, [buf], #1
+		ldrb	td3, [buf], #1
+		sub	len, len, #2
+#ifndef __ARMEB__
+		orr	td0, td0, td3, lsl #8
+#else
+		orr	td0, td3, td0, lsl #8
+#endif
+#endif
+		adcs	sum, sum, td0
+		tst	len, #6
+		bne	.Lless8_wordlp
+
+.Lless8_byte:	tst	len, #1			@ odd number of bytes
+		ldrneb	td0, [buf], #1		@ include last byte
+		adcnes	sum, sum, td0, put_byte_0	@ update checksum
+
+.Ldone:		adc	r0, sum, #0		@ collect up the last carry
+		ldr	td0, [sp], #4
+		tst	td0, #1			@ check buffer alignment
+		movne	r0, r0, ror #8		@ rotate checksum by 8 bits
+		ldr	pc, [sp], #4		@ return
+
+.Lnot_aligned:	tst	buf, #1			@ odd address
+		ldrneb	td0, [buf], #1		@ make even
+		subne	len, len, #1
+		adcnes	sum, sum, td0, put_byte_1	@ update checksum
+
+		tst	buf, #2			@ 32-bit aligned?
+#if __LINUX_ARM_ARCH__ >= 4
+		ldrneh	td0, [buf], #2		@ make 32-bit aligned
+		subne	len, len, #2
+#else
+		ldrneb	td0, [buf], #1
+		ldrneb	ip, [buf], #1
+		subne	len, len, #2
+#ifndef __ARMEB__
+		orrne	td0, td0, ip, lsl #8
+#else
+		orrne	td0, ip, td0, lsl #8
+#endif
+#endif
+		adcnes	sum, sum, td0		@ update checksum
+		ret	lr
+
+ENTRY(csum_partial)
+		stmfd	sp!, {buf, lr}
+		cmp	len, #8			@ Ensure that we have at least
+		blo	.Lless8			@ 8 bytes to copy.
+
+		tst	buf, #1
+		movne	sum, sum, ror #8
+
+		adds	sum, sum, #0		@ C = 0
+		tst	buf, #3			@ Test destination alignment
+		blne	.Lnot_aligned		@ align destination, return here
+
+1:		bics	ip, len, #31
+		beq	3f
+
+		stmfd	sp!, {r4 - r5}
+2:		ldmia	buf!, {td0, td1, td2, td3}
+		adcs	sum, sum, td0
+		adcs	sum, sum, td1
+		adcs	sum, sum, td2
+		adcs	sum, sum, td3
+		ldmia	buf!, {td0, td1, td2, td3}
+		adcs	sum, sum, td0
+		adcs	sum, sum, td1
+		adcs	sum, sum, td2
+		adcs	sum, sum, td3
+		sub	ip, ip, #32
+		teq	ip, #0
+		bne	2b
+		ldmfd	sp!, {r4 - r5}
+
+3:		tst	len, #0x1c		@ should not change C
+		beq	.Lless4
+
+4:		ldr	td0, [buf], #4
+		sub	len, len, #4
+		adcs	sum, sum, td0
+		tst	len, #0x1c
+		bne	4b
+		b	.Lless4
+ENDPROC(csum_partial)

arch/arm/lib/csumpartialcopy.S

@@ -0,0 +1,53 @@
+/*
+ *  linux/arch/arm/lib/csumpartialcopy.S
+ *
+ *  Copyright (C) 1995-1998 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+		.text
+
+/* Function: __u32 csum_partial_copy_nocheck(const char *src, char *dst, int len, __u32 sum)
+ * Params  : r0 = src, r1 = dst, r2 = len, r3 = checksum
+ * Returns : r0 = new checksum
+ */
+
+		.macro	save_regs
+		stmfd	sp!, {r1, r4 - r8, lr}
+		.endm
+
+		.macro	load_regs
+		ldmfd	sp!, {r1, r4 - r8, pc}
+		.endm
+
+		.macro	load1b, reg1
+		ldrb	\reg1, [r0], #1
+		.endm
+
+		.macro	load2b, reg1, reg2
+		ldrb	\reg1, [r0], #1
+		ldrb	\reg2, [r0], #1
+		.endm
+
+		.macro	load1l, reg1
+		ldr	\reg1, [r0], #4
+		.endm
+
+		.macro	load2l, reg1, reg2
+		ldr	\reg1, [r0], #4
+		ldr	\reg2, [r0], #4
+		.endm
+
+		.macro	load4l, reg1, reg2, reg3, reg4
+		ldmia	r0!, {\reg1, \reg2, \reg3, \reg4}
+		.endm
+
+#define FN_ENTRY	ENTRY(csum_partial_copy_nocheck)
+#define FN_EXIT		ENDPROC(csum_partial_copy_nocheck)
+
+#include "csumpartialcopygeneric.S"

arch/arm/lib/csumpartialcopygeneric.S

@@ -0,0 +1,333 @@
+/*
+ *  linux/arch/arm/lib/csumpartialcopygeneric.S
+ *
+ *  Copyright (C) 1995-2001 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <asm/assembler.h>
+
+/*
+ * unsigned int
+ * csum_partial_copy_xxx(const char *src, char *dst, int len, int sum, )
+ *  r0 = src, r1 = dst, r2 = len, r3 = sum
+ *  Returns : r0 = checksum
+ *
+ * Note that 'tst' and 'teq' preserve the carry flag.
+ */
+
+src	.req	r0
+dst	.req	r1
+len	.req	r2
+sum	.req	r3
+
+.Lzero:		mov	r0, sum
+		load_regs
+
+		/*
+		 * Align an unaligned destination pointer.  We know that
+		 * we have >= 8 bytes here, so we don't need to check
+		 * the length.  Note that the source pointer hasn't been
+		 * aligned yet.
+		 */
+.Ldst_unaligned:
+		tst	dst, #1
+		beq	.Ldst_16bit
+
+		load1b	ip
+		sub	len, len, #1
+		adcs	sum, sum, ip, put_byte_1	@ update checksum
+		strb	ip, [dst], #1
+		tst	dst, #2
+		reteq	lr			@ dst is now 32bit aligned
+
+.Ldst_16bit:	load2b	r8, ip
+		sub	len, len, #2
+		adcs	sum, sum, r8, put_byte_0
+		strb	r8, [dst], #1
+		adcs	sum, sum, ip, put_byte_1
+		strb	ip, [dst], #1
+		ret	lr			@ dst is now 32bit aligned
+
+		/*
+		 * Handle 0 to 7 bytes, with any alignment of source and
+		 * destination pointers.  Note that when we get here, C = 0
+		 */
+.Lless8:	teq	len, #0			@ check for zero count
+		beq	.Lzero
+
+		/* we must have at least one byte. */
+		tst	dst, #1			@ dst 16-bit aligned
+		beq	.Lless8_aligned
+
+		/* Align dst */
+		load1b	ip
+		sub	len, len, #1
+		adcs	sum, sum, ip, put_byte_1	@ update checksum
+		strb	ip, [dst], #1
+		tst	len, #6
+		beq	.Lless8_byteonly
+
+1:		load2b	r8, ip
+		sub	len, len, #2
+		adcs	sum, sum, r8, put_byte_0
+		strb	r8, [dst], #1
+		adcs	sum, sum, ip, put_byte_1
+		strb	ip, [dst], #1
+.Lless8_aligned:
+		tst	len, #6
+		bne	1b
+.Lless8_byteonly:
+		tst	len, #1
+		beq	.Ldone
+		load1b	r8
+		adcs	sum, sum, r8, put_byte_0	@ update checksum
+		strb	r8, [dst], #1
+		b	.Ldone
+
+FN_ENTRY
+		save_regs
+
+		cmp	len, #8			@ Ensure that we have at least
+		blo	.Lless8			@ 8 bytes to copy.
+
+		adds	sum, sum, #0		@ C = 0
+		tst	dst, #3			@ Test destination alignment
+		blne	.Ldst_unaligned		@ align destination, return here
+
+		/*
+		 * Ok, the dst pointer is now 32bit aligned, and we know
+		 * that we must have more than 4 bytes to copy.  Note
+		 * that C contains the carry from the dst alignment above.
+		 */
+
+		tst	src, #3			@ Test source alignment
+		bne	.Lsrc_not_aligned
+
+		/* Routine for src & dst aligned */
+
+		bics	ip, len, #15
+		beq	2f
+
+1:		load4l	r4, r5, r6, r7
+		stmia	dst!, {r4, r5, r6, r7}
+		adcs	sum, sum, r4
+		adcs	sum, sum, r5
+		adcs	sum, sum, r6
+		adcs	sum, sum, r7
+		sub	ip, ip, #16
+		teq	ip, #0
+		bne	1b
+
+2:		ands	ip, len, #12
+		beq	4f
+		tst	ip, #8
+		beq	3f
+		load2l	r4, r5
+		stmia	dst!, {r4, r5}
+		adcs	sum, sum, r4
+		adcs	sum, sum, r5
+		tst	ip, #4
+		beq	4f
+
+3:		load1l	r4
+		str	r4, [dst], #4
+		adcs	sum, sum, r4
+
+4:		ands	len, len, #3
+		beq	.Ldone
+		load1l	r4
+		tst	len, #2
+		mov	r5, r4, get_byte_0
+		beq	.Lexit
+		adcs	sum, sum, r4, lspush #16
+		strb	r5, [dst], #1
+		mov	r5, r4, get_byte_1
+		strb	r5, [dst], #1
+		mov	r5, r4, get_byte_2
+.Lexit:		tst	len, #1
+		strneb	r5, [dst], #1
+		andne	r5, r5, #255
+		adcnes	sum, sum, r5, put_byte_0
+
+		/*
+		 * If the dst pointer was not 16-bit aligned, we
+		 * need to rotate the checksum here to get around
+		 * the inefficient byte manipulations in the
+		 * architecture independent code.
+		 */
+.Ldone:		adc	r0, sum, #0
+		ldr	sum, [sp, #0]		@ dst
+		tst	sum, #1
+		movne	r0, r0, ror #8
+		load_regs
+
+.Lsrc_not_aligned:
+		adc	sum, sum, #0		@ include C from dst alignment
+		and	ip, src, #3
+		bic	src, src, #3
+		load1l	r5
+		cmp	ip, #2
+		beq	.Lsrc2_aligned
+		bhi	.Lsrc3_aligned
+		mov	r4, r5, lspull #8		@ C = 0
+		bics	ip, len, #15
+		beq	2f
+1:		load4l	r5, r6, r7, r8
+		orr	r4, r4, r5, lspush #24
+		mov	r5, r5, lspull #8
+		orr	r5, r5, r6, lspush #24
+		mov	r6, r6, lspull #8
+		orr	r6, r6, r7, lspush #24
+		mov	r7, r7, lspull #8
+		orr	r7, r7, r8, lspush #24
+		stmia	dst!, {r4, r5, r6, r7}
+		adcs	sum, sum, r4
+		adcs	sum, sum, r5
+		adcs	sum, sum, r6
+		adcs	sum, sum, r7
+		mov	r4, r8, lspull #8
+		sub	ip, ip, #16
+		teq	ip, #0
+		bne	1b
+2:		ands	ip, len, #12
+		beq	4f
+		tst	ip, #8
+		beq	3f
+		load2l	r5, r6
+		orr	r4, r4, r5, lspush #24
+		mov	r5, r5, lspull #8
+		orr	r5, r5, r6, lspush #24
+		stmia	dst!, {r4, r5}
+		adcs	sum, sum, r4
+		adcs	sum, sum, r5
+		mov	r4, r6, lspull #8
+		tst	ip, #4
+		beq	4f
+3:		load1l	r5
+		orr	r4, r4, r5, lspush #24
+		str	r4, [dst], #4
+		adcs	sum, sum, r4
+		mov	r4, r5, lspull #8
+4:		ands	len, len, #3
+		beq	.Ldone
+		mov	r5, r4, get_byte_0
+		tst	len, #2
+		beq	.Lexit
+		adcs	sum, sum, r4, lspush #16
+		strb	r5, [dst], #1
+		mov	r5, r4, get_byte_1
+		strb	r5, [dst], #1
+		mov	r5, r4, get_byte_2
+		b	.Lexit
+
+.Lsrc2_aligned:	mov	r4, r5, lspull #16
+		adds	sum, sum, #0
+		bics	ip, len, #15
+		beq	2f
+1:		load4l	r5, r6, r7, r8
+		orr	r4, r4, r5, lspush #16
+		mov	r5, r5, lspull #16
+		orr	r5, r5, r6, lspush #16
+		mov	r6, r6, lspull #16
+		orr	r6, r6, r7, lspush #16
+		mov	r7, r7, lspull #16
+		orr	r7, r7, r8, lspush #16
+		stmia	dst!, {r4, r5, r6, r7}
+		adcs	sum, sum, r4
+		adcs	sum, sum, r5
+		adcs	sum, sum, r6
+		adcs	sum, sum, r7
+		mov	r4, r8, lspull #16
+		sub	ip, ip, #16
+		teq	ip, #0
+		bne	1b
+2:		ands	ip, len, #12
+		beq	4f
+		tst	ip, #8
+		beq	3f
+		load2l	r5, r6
+		orr	r4, r4, r5, lspush #16
+		mov	r5, r5, lspull #16
+		orr	r5, r5, r6, lspush #16
+		stmia	dst!, {r4, r5}
+		adcs	sum, sum, r4
+		adcs	sum, sum, r5
+		mov	r4, r6, lspull #16
+		tst	ip, #4
+		beq	4f
+3:		load1l	r5
+		orr	r4, r4, r5, lspush #16
+		str	r4, [dst], #4
+		adcs	sum, sum, r4
+		mov	r4, r5, lspull #16
+4:		ands	len, len, #3
+		beq	.Ldone
+		mov	r5, r4, get_byte_0
+		tst	len, #2
+		beq	.Lexit
+		adcs	sum, sum, r4
+		strb	r5, [dst], #1
+		mov	r5, r4, get_byte_1
+		strb	r5, [dst], #1
+		tst	len, #1
+		beq	.Ldone
+		load1b	r5
+		b	.Lexit
+
+.Lsrc3_aligned:	mov	r4, r5, lspull #24
+		adds	sum, sum, #0
+		bics	ip, len, #15
+		beq	2f
+1:		load4l	r5, r6, r7, r8
+		orr	r4, r4, r5, lspush #8
+		mov	r5, r5, lspull #24
+		orr	r5, r5, r6, lspush #8
+		mov	r6, r6, lspull #24
+		orr	r6, r6, r7, lspush #8
+		mov	r7, r7, lspull #24
+		orr	r7, r7, r8, lspush #8
+		stmia	dst!, {r4, r5, r6, r7}
+		adcs	sum, sum, r4
+		adcs	sum, sum, r5
+		adcs	sum, sum, r6
+		adcs	sum, sum, r7
+		mov	r4, r8, lspull #24
+		sub	ip, ip, #16
+		teq	ip, #0
+		bne	1b
+2:		ands	ip, len, #12
+		beq	4f
+		tst	ip, #8
+		beq	3f
+		load2l	r5, r6
+		orr	r4, r4, r5, lspush #8
+		mov	r5, r5, lspull #24
+		orr	r5, r5, r6, lspush #8
+		stmia	dst!, {r4, r5}
+		adcs	sum, sum, r4
+		adcs	sum, sum, r5
+		mov	r4, r6, lspull #24
+		tst	ip, #4
+		beq	4f
+3:		load1l	r5
+		orr	r4, r4, r5, lspush #8
+		str	r4, [dst], #4
+		adcs	sum, sum, r4
+		mov	r4, r5, lspull #24
+4:		ands	len, len, #3
+		beq	.Ldone
+		mov	r5, r4, get_byte_0
+		tst	len, #2
+		beq	.Lexit
+		strb	r5, [dst], #1
+		adcs	sum, sum, r4
+		load1l	r4
+		mov	r5, r4, get_byte_0
+		strb	r5, [dst], #1
+		adcs	sum, sum, r4, lspush #24
+		mov	r5, r4, get_byte_1
+		b	.Lexit
+FN_EXIT

arch/arm/lib/csumpartialcopyuser.S

@@ -0,0 +1,101 @@
+/*
+ *  linux/arch/arm/lib/csumpartialcopyuser.S
+ *
+ *  Copyright (C) 1995-1998 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * 27/03/03 Ian Molton Clean up CONFIG_CPU
+ *
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include <asm/errno.h>
+#include <asm/asm-offsets.h>
+
+		.text
+
+#ifdef CONFIG_CPU_SW_DOMAIN_PAN
+		.macro	save_regs
+		mrc	p15, 0, ip, c3, c0, 0
+		stmfd	sp!, {r1, r2, r4 - r8, ip, lr}
+		uaccess_enable ip
+		.endm
+
+		.macro	load_regs
+		ldmfd	sp!, {r1, r2, r4 - r8, ip, lr}
+		mcr	p15, 0, ip, c3, c0, 0
+		ret	lr
+		.endm
+#else
+		.macro	save_regs
+		stmfd	sp!, {r1, r2, r4 - r8, lr}
+		.endm
+
+		.macro	load_regs
+		ldmfd	sp!, {r1, r2, r4 - r8, pc}
+		.endm
+#endif
+
+		.macro	load1b,	reg1
+		ldrusr	\reg1, r0, 1
+		.endm
+
+		.macro	load2b, reg1, reg2
+		ldrusr	\reg1, r0, 1
+		ldrusr	\reg2, r0, 1
+		.endm
+
+		.macro	load1l, reg1
+		ldrusr	\reg1, r0, 4
+		.endm
+
+		.macro	load2l, reg1, reg2
+		ldrusr	\reg1, r0, 4
+		ldrusr	\reg2, r0, 4
+		.endm
+
+		.macro	load4l, reg1, reg2, reg3, reg4
+		ldrusr	\reg1, r0, 4
+		ldrusr	\reg2, r0, 4
+		ldrusr	\reg3, r0, 4
+		ldrusr	\reg4, r0, 4
+		.endm
+
+/*
+ * unsigned int
+ * csum_partial_copy_from_user(const char *src, char *dst, int len, int sum, int *err_ptr)
+ *  r0 = src, r1 = dst, r2 = len, r3 = sum, [sp] = *err_ptr
+ *  Returns : r0 = checksum, [[sp, #0], #0] = 0 or -EFAULT
+ */
+
+#define FN_ENTRY	ENTRY(csum_partial_copy_from_user)
+#define FN_EXIT		ENDPROC(csum_partial_copy_from_user)
+
+#include "csumpartialcopygeneric.S"
+
+/*
+ * FIXME: minor buglet here
+ * We don't return the checksum for the data present in the buffer.  To do
+ * so properly, we would have to add in whatever registers were loaded before
+ * the fault, which, with the current asm above is not predictable.
+ */
+		.pushsection .text.fixup,"ax"
+		.align	4
+9001:		mov	r4, #-EFAULT
+#ifdef CONFIG_CPU_SW_DOMAIN_PAN
+		ldr	r5, [sp, #9*4]		@ *err_ptr
+#else
+		ldr	r5, [sp, #8*4]		@ *err_ptr
+#endif
+		str	r4, [r5]
+		ldmia	sp, {r1, r2}		@ retrieve dst, len
+		add	r2, r2, r1
+		mov	r0, #0			@ zero the buffer
+9002:		teq	r2, r1
+		strneb	r0, [r1], #1
+		bne	9002b
+		load_regs
+		.popsection

arch/arm/lib/delay-loop.S

@@ -0,0 +1,63 @@
+/*
+ *  linux/arch/arm/lib/delay.S
+ *
+ *  Copyright (C) 1995, 1996 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include <asm/delay.h>
+
+		.text
+
+.LC0:		.word	loops_per_jiffy
+.LC1:		.word	UDELAY_MULT
+
+/*
+ * r0  <= 2000
+ * HZ  <= 1000
+ */
+
+ENTRY(__loop_udelay)
+		ldr	r2, .LC1
+		mul	r0, r2, r0
+ENTRY(__loop_const_udelay)			@ 0 <= r0 <= 0x7fffff06
+		ldr	r2, .LC0
+		ldr	r2, [r2]
+		umull	r1, r0, r2, r0
+		adds	r1, r1, #0xffffffff
+		adcs	r0, r0, r0
+		reteq	lr
+
+/*
+ * loops = r0 * HZ * loops_per_jiffy / 1000000
+ */
+		.align 3
+
+@ Delay routine
+ENTRY(__loop_delay)
+		subs	r0, r0, #1
+#if 0
+		retls	lr
+		subs	r0, r0, #1
+		retls	lr
+		subs	r0, r0, #1
+		retls	lr
+		subs	r0, r0, #1
+		retls	lr
+		subs	r0, r0, #1
+		retls	lr
+		subs	r0, r0, #1
+		retls	lr
+		subs	r0, r0, #1
+		retls	lr
+		subs	r0, r0, #1
+#endif
+		bhi	__loop_delay
+		ret	lr
+ENDPROC(__loop_udelay)
+ENDPROC(__loop_const_udelay)
+ENDPROC(__loop_delay)

arch/arm/lib/delay.c

@@ -0,0 +1,117 @@
+/*
+ * Delay loops based on the OpenRISC implementation.
+ *
+ * Copyright (C) 2012 ARM Limited
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * Author: Will Deacon <will.deacon@arm.com>
+ */
+
+#include <linux/clocksource.h>
+#include <linux/delay.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/timex.h>
+
+/*
+ * Default to the loop-based delay implementation.
+ */
+struct arm_delay_ops arm_delay_ops __ro_after_init = {
+	.delay		= __loop_delay,
+	.const_udelay	= __loop_const_udelay,
+	.udelay		= __loop_udelay,
+};
+
+static const struct delay_timer *delay_timer;
+static bool delay_calibrated;
+static u64 delay_res;
+
+int read_current_timer(unsigned long *timer_val)
+{
+	if (!delay_timer)
+		return -ENXIO;
+
+	*timer_val = delay_timer->read_current_timer();
+	return 0;
+}
+EXPORT_SYMBOL_GPL(read_current_timer);
+
+static inline u64 cyc_to_ns(u64 cyc, u32 mult, u32 shift)
+{
+	return (cyc * mult) >> shift;
+}
+
+static void __timer_delay(unsigned long cycles)
+{
+	cycles_t start = get_cycles();
+
+	while ((get_cycles() - start) < cycles)
+		cpu_relax();
+}
+
+static void __timer_const_udelay(unsigned long xloops)
+{
+	unsigned long long loops = xloops;
+	loops *= arm_delay_ops.ticks_per_jiffy;
+	__timer_delay(loops >> UDELAY_SHIFT);
+}
+
+static void __timer_udelay(unsigned long usecs)
+{
+	__timer_const_udelay(usecs * UDELAY_MULT);
+}
+
+void __init register_current_timer_delay(const struct delay_timer *timer)
+{
+	u32 new_mult, new_shift;
+	u64 res;
+
+	clocks_calc_mult_shift(&new_mult, &new_shift, timer->freq,
+			       NSEC_PER_SEC, 3600);
+	res = cyc_to_ns(1ULL, new_mult, new_shift);
+
+	if (res > 1000) {
+		pr_err("Ignoring delay timer %ps, which has insufficient resolution of %lluns\n",
+			timer, res);
+		return;
+	}
+
+	if (!delay_calibrated && (!delay_res || (res < delay_res))) {
+		pr_info("Switching to timer-based delay loop, resolution %lluns\n", res);
+		delay_timer			= timer;
+		lpj_fine			= timer->freq / HZ;
+		delay_res			= res;
+
+		/* cpufreq may scale loops_per_jiffy, so keep a private copy */
+		arm_delay_ops.ticks_per_jiffy	= lpj_fine;
+		arm_delay_ops.delay		= __timer_delay;
+		arm_delay_ops.const_udelay	= __timer_const_udelay;
+		arm_delay_ops.udelay		= __timer_udelay;
+	} else {
+		pr_info("Ignoring duplicate/late registration of read_current_timer delay\n");
+	}
+}
+
+unsigned long calibrate_delay_is_known(void)
+{
+	delay_calibrated = true;
+	return lpj_fine;
+}
+
+void calibration_delay_done(void)
+{
+	delay_calibrated = true;
+}

arch/arm/lib/div64.S

@@ -0,0 +1,212 @@
+/*
+ *  linux/arch/arm/lib/div64.S
+ *
+ *  Optimized computation of 64-bit dividend / 32-bit divisor
+ *
+ *  Author:	Nicolas Pitre
+ *  Created:	Oct 5, 2003
+ *  Copyright:	Monta Vista Software, Inc.
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include <asm/unwind.h>
+
+#ifdef __ARMEB__
+#define xh r0
+#define xl r1
+#define yh r2
+#define yl r3
+#else
+#define xl r0
+#define xh r1
+#define yl r2
+#define yh r3
+#endif
+
+/*
+ * __do_div64: perform a division with 64-bit dividend and 32-bit divisor.
+ *
+ * Note: Calling convention is totally non standard for optimal code.
+ *       This is meant to be used by do_div() from include/asm/div64.h only.
+ *
+ * Input parameters:
+ * 	xh-xl	= dividend (clobbered)
+ * 	r4	= divisor (preserved)
+ *
+ * Output values:
+ * 	yh-yl	= result
+ * 	xh	= remainder
+ *
+ * Clobbered regs: xl, ip
+ */
+
+ENTRY(__do_div64)
+UNWIND(.fnstart)
+
+	@ Test for easy paths first.
+	subs	ip, r4, #1
+	bls	9f			@ divisor is 0 or 1
+	tst	ip, r4
+	beq	8f			@ divisor is power of 2
+
+	@ See if we need to handle upper 32-bit result.
+	cmp	xh, r4
+	mov	yh, #0
+	blo	3f
+
+	@ Align divisor with upper part of dividend.
+	@ The aligned divisor is stored in yl preserving the original.
+	@ The bit position is stored in ip.
+
+#if __LINUX_ARM_ARCH__ >= 5
+
+	clz	yl, r4
+	clz	ip, xh
+	sub	yl, yl, ip
+	mov	ip, #1
+	mov	ip, ip, lsl yl
+	mov	yl, r4, lsl yl
+
+#else
+
+	mov	yl, r4
+	mov	ip, #1
+1:	cmp	yl, #0x80000000
+	cmpcc	yl, xh
+	movcc	yl, yl, lsl #1
+	movcc	ip, ip, lsl #1
+	bcc	1b
+
+#endif
+
+	@ The division loop for needed upper bit positions.
+ 	@ Break out early if dividend reaches 0.
+2:	cmp	xh, yl
+	orrcs	yh, yh, ip
+	subcss	xh, xh, yl
+	movnes	ip, ip, lsr #1
+	mov	yl, yl, lsr #1
+	bne	2b
+
+	@ See if we need to handle lower 32-bit result.
+3:	cmp	xh, #0
+	mov	yl, #0
+	cmpeq	xl, r4
+	movlo	xh, xl
+	retlo	lr
+
+	@ The division loop for lower bit positions.
+	@ Here we shift remainer bits leftwards rather than moving the
+	@ divisor for comparisons, considering the carry-out bit as well.
+	mov	ip, #0x80000000
+4:	movs	xl, xl, lsl #1
+	adcs	xh, xh, xh
+	beq	6f
+	cmpcc	xh, r4
+5:	orrcs	yl, yl, ip
+	subcs	xh, xh, r4
+	movs	ip, ip, lsr #1
+	bne	4b
+	ret	lr
+
+	@ The top part of remainder became zero.  If carry is set
+	@ (the 33th bit) this is a false positive so resume the loop.
+	@ Otherwise, if lower part is also null then we are done.
+6:	bcs	5b
+	cmp	xl, #0
+	reteq	lr
+
+	@ We still have remainer bits in the low part.  Bring them up.
+
+#if __LINUX_ARM_ARCH__ >= 5
+
+	clz	xh, xl			@ we know xh is zero here so...
+	add	xh, xh, #1
+	mov	xl, xl, lsl xh
+	mov	ip, ip, lsr xh
+
+#else
+
+7:	movs	xl, xl, lsl #1
+	mov	ip, ip, lsr #1
+	bcc	7b
+
+#endif
+
+	@ Current remainder is now 1.  It is worthless to compare with
+	@ divisor at this point since divisor can not be smaller than 3 here.
+	@ If possible, branch for another shift in the division loop.
+	@ If no bit position left then we are done.
+	movs	ip, ip, lsr #1
+	mov	xh, #1
+	bne	4b
+	ret	lr
+
+8:	@ Division by a power of 2: determine what that divisor order is
+	@ then simply shift values around
+
+#if __LINUX_ARM_ARCH__ >= 5
+
+	clz	ip, r4
+	rsb	ip, ip, #31
+
+#else
+
+	mov	yl, r4
+	cmp	r4, #(1 << 16)
+	mov	ip, #0
+	movhs	yl, yl, lsr #16
+	movhs	ip, #16
+
+	cmp	yl, #(1 << 8)
+	movhs	yl, yl, lsr #8
+	addhs	ip, ip, #8
+
+	cmp	yl, #(1 << 4)
+	movhs	yl, yl, lsr #4
+	addhs	ip, ip, #4
+
+	cmp	yl, #(1 << 2)
+	addhi	ip, ip, #3
+	addls	ip, ip, yl, lsr #1
+
+#endif
+
+	mov	yh, xh, lsr ip
+	mov	yl, xl, lsr ip
+	rsb	ip, ip, #32
+ ARM(	orr	yl, yl, xh, lsl ip	)
+ THUMB(	lsl	xh, xh, ip		)
+ THUMB(	orr	yl, yl, xh		)
+	mov	xh, xl, lsl ip
+	mov	xh, xh, lsr ip
+	ret	lr
+
+	@ eq -> division by 1: obvious enough...
+9:	moveq	yl, xl
+	moveq	yh, xh
+	moveq	xh, #0
+	reteq	lr
+UNWIND(.fnend)
+
+UNWIND(.fnstart)
+UNWIND(.pad #4)
+UNWIND(.save {lr})
+Ldiv0_64:
+	@ Division by 0:
+	str	lr, [sp, #-8]!
+	bl	__div0
+
+	@ as wrong as it could be...
+	mov	yl, #0
+	mov	yh, #0
+	mov	xh, #0
+	ldr	pc, [sp], #8
+
+UNWIND(.fnend)
+ENDPROC(__do_div64)

arch/arm/lib/ecard.S

@@ -0,0 +1,44 @@
+/*
+ *  linux/arch/arm/lib/ecard.S
+ *
+ *  Copyright (C) 1995, 1996 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * 27/03/03 Ian Molton Clean up CONFIG_CPU
+ *
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+#define CPSR2SPSR(rt) \
+		mrs	rt, cpsr; \
+		msr	spsr_cxsf, rt
+
+@ Purpose: call an expansion card loader to read bytes.
+@ Proto  : char read_loader(int offset, char *card_base, char *loader);
+@ Returns: byte read
+
+ENTRY(ecard_loader_read)
+		stmfd	sp!, {r4 - r12, lr}
+		mov	r11, r1
+		mov	r1, r0
+		CPSR2SPSR(r0)
+		mov	lr, pc
+		mov	pc, r2
+		ldmfd	sp!, {r4 - r12, pc}
+
+@ Purpose: call an expansion card loader to reset the card
+@ Proto  : void read_loader(int card_base, char *loader);
+@ Returns: byte read
+
+ENTRY(ecard_loader_reset)
+		stmfd	sp!, {r4 - r12, lr}
+		mov	r11, r0
+		CPSR2SPSR(r0)
+		mov	lr, pc
+		add	pc, r1, #8
+		ldmfd	sp!, {r4 - r12, pc}
+

arch/arm/lib/findbit.S

@@ -0,0 +1,196 @@
+/*
+ *  linux/arch/arm/lib/findbit.S
+ *
+ *  Copyright (C) 1995-2000 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * 16th March 2001 - John Ripley <jripley@sonicblue.com>
+ *   Fixed so that "size" is an exclusive not an inclusive quantity.
+ *   All users of these functions expect exclusive sizes, and may
+ *   also call with zero size.
+ * Reworked by rmk.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+                .text
+
+/*
+ * Purpose  : Find a 'zero' bit
+ * Prototype: int find_first_zero_bit(void *addr, unsigned int maxbit);
+ */
+ENTRY(_find_first_zero_bit_le)
+		teq	r1, #0	
+		beq	3f
+		mov	r2, #0
+1:
+ ARM(		ldrb	r3, [r0, r2, lsr #3]	)
+ THUMB(		lsr	r3, r2, #3		)
+ THUMB(		ldrb	r3, [r0, r3]		)
+		eors	r3, r3, #0xff		@ invert bits
+		bne	.L_found		@ any now set - found zero bit
+		add	r2, r2, #8		@ next bit pointer
+2:		cmp	r2, r1			@ any more?
+		blo	1b
+3:		mov	r0, r1			@ no free bits
+		ret	lr
+ENDPROC(_find_first_zero_bit_le)
+
+/*
+ * Purpose  : Find next 'zero' bit
+ * Prototype: int find_next_zero_bit(void *addr, unsigned int maxbit, int offset)
+ */
+ENTRY(_find_next_zero_bit_le)
+		teq	r1, #0
+		beq	3b
+		ands	ip, r2, #7
+		beq	1b			@ If new byte, goto old routine
+ ARM(		ldrb	r3, [r0, r2, lsr #3]	)
+ THUMB(		lsr	r3, r2, #3		)
+ THUMB(		ldrb	r3, [r0, r3]		)
+		eor	r3, r3, #0xff		@ now looking for a 1 bit
+		movs	r3, r3, lsr ip		@ shift off unused bits
+		bne	.L_found
+		orr	r2, r2, #7		@ if zero, then no bits here
+		add	r2, r2, #1		@ align bit pointer
+		b	2b			@ loop for next bit
+ENDPROC(_find_next_zero_bit_le)
+
+/*
+ * Purpose  : Find a 'one' bit
+ * Prototype: int find_first_bit(const unsigned long *addr, unsigned int maxbit);
+ */
+ENTRY(_find_first_bit_le)
+		teq	r1, #0	
+		beq	3f
+		mov	r2, #0
+1:
+ ARM(		ldrb	r3, [r0, r2, lsr #3]	)
+ THUMB(		lsr	r3, r2, #3		)
+ THUMB(		ldrb	r3, [r0, r3]		)
+		movs	r3, r3
+		bne	.L_found		@ any now set - found zero bit
+		add	r2, r2, #8		@ next bit pointer
+2:		cmp	r2, r1			@ any more?
+		blo	1b
+3:		mov	r0, r1			@ no free bits
+		ret	lr
+ENDPROC(_find_first_bit_le)
+
+/*
+ * Purpose  : Find next 'one' bit
+ * Prototype: int find_next_zero_bit(void *addr, unsigned int maxbit, int offset)
+ */
+ENTRY(_find_next_bit_le)
+		teq	r1, #0
+		beq	3b
+		ands	ip, r2, #7
+		beq	1b			@ If new byte, goto old routine
+ ARM(		ldrb	r3, [r0, r2, lsr #3]	)
+ THUMB(		lsr	r3, r2, #3		)
+ THUMB(		ldrb	r3, [r0, r3]		)
+		movs	r3, r3, lsr ip		@ shift off unused bits
+		bne	.L_found
+		orr	r2, r2, #7		@ if zero, then no bits here
+		add	r2, r2, #1		@ align bit pointer
+		b	2b			@ loop for next bit
+ENDPROC(_find_next_bit_le)
+
+#ifdef __ARMEB__
+
+ENTRY(_find_first_zero_bit_be)
+		teq	r1, #0
+		beq	3f
+		mov	r2, #0
+1:		eor	r3, r2, #0x18		@ big endian byte ordering
+ ARM(		ldrb	r3, [r0, r3, lsr #3]	)
+ THUMB(		lsr	r3, #3			)
+ THUMB(		ldrb	r3, [r0, r3]		)
+		eors	r3, r3, #0xff		@ invert bits
+		bne	.L_found		@ any now set - found zero bit
+		add	r2, r2, #8		@ next bit pointer
+2:		cmp	r2, r1			@ any more?
+		blo	1b
+3:		mov	r0, r1			@ no free bits
+		ret	lr
+ENDPROC(_find_first_zero_bit_be)
+
+ENTRY(_find_next_zero_bit_be)
+		teq	r1, #0
+		beq	3b
+		ands	ip, r2, #7
+		beq	1b			@ If new byte, goto old routine
+		eor	r3, r2, #0x18		@ big endian byte ordering
+ ARM(		ldrb	r3, [r0, r3, lsr #3]	)
+ THUMB(		lsr	r3, #3			)
+ THUMB(		ldrb	r3, [r0, r3]		)
+		eor	r3, r3, #0xff		@ now looking for a 1 bit
+		movs	r3, r3, lsr ip		@ shift off unused bits
+		bne	.L_found
+		orr	r2, r2, #7		@ if zero, then no bits here
+		add	r2, r2, #1		@ align bit pointer
+		b	2b			@ loop for next bit
+ENDPROC(_find_next_zero_bit_be)
+
+ENTRY(_find_first_bit_be)
+		teq	r1, #0
+		beq	3f
+		mov	r2, #0
+1:		eor	r3, r2, #0x18		@ big endian byte ordering
+ ARM(		ldrb	r3, [r0, r3, lsr #3]	)
+ THUMB(		lsr	r3, #3			)
+ THUMB(		ldrb	r3, [r0, r3]		)
+		movs	r3, r3
+		bne	.L_found		@ any now set - found zero bit
+		add	r2, r2, #8		@ next bit pointer
+2:		cmp	r2, r1			@ any more?
+		blo	1b
+3:		mov	r0, r1			@ no free bits
+		ret	lr
+ENDPROC(_find_first_bit_be)
+
+ENTRY(_find_next_bit_be)
+		teq	r1, #0
+		beq	3b
+		ands	ip, r2, #7
+		beq	1b			@ If new byte, goto old routine
+		eor	r3, r2, #0x18		@ big endian byte ordering
+ ARM(		ldrb	r3, [r0, r3, lsr #3]	)
+ THUMB(		lsr	r3, #3			)
+ THUMB(		ldrb	r3, [r0, r3]		)
+		movs	r3, r3, lsr ip		@ shift off unused bits
+		bne	.L_found
+		orr	r2, r2, #7		@ if zero, then no bits here
+		add	r2, r2, #1		@ align bit pointer
+		b	2b			@ loop for next bit
+ENDPROC(_find_next_bit_be)
+
+#endif
+
+/*
+ * One or more bits in the LSB of r3 are assumed to be set.
+ */
+.L_found:
+#if __LINUX_ARM_ARCH__ >= 5
+		rsb	r0, r3, #0
+		and	r3, r3, r0
+		clz	r3, r3
+		rsb	r3, r3, #31
+		add	r0, r2, r3
+#else
+		tst	r3, #0x0f
+		addeq	r2, r2, #4
+		movne	r3, r3, lsl #4
+		tst	r3, #0x30
+		addeq	r2, r2, #2
+		movne	r3, r3, lsl #2
+		tst	r3, #0x40
+		addeq	r2, r2, #1
+		mov	r0, r2
+#endif
+		cmp	r1, r0			@ Clamp to maxbit
+		movlo	r0, r1
+		ret	lr
+

arch/arm/lib/floppydma.S

@@ -0,0 +1,32 @@
+/*
+ *  linux/arch/arm/lib/floppydma.S
+ *
+ *  Copyright (C) 1995, 1996 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+		.text
+
+		.global	floppy_fiqin_end
+ENTRY(floppy_fiqin_start)
+		subs	r9, r9, #1
+		ldrgtb	r12, [r11, #-4]
+		ldrleb	r12, [r11], #0
+		strb	r12, [r10], #1
+		subs	pc, lr, #4
+floppy_fiqin_end:
+
+		.global	floppy_fiqout_end
+ENTRY(floppy_fiqout_start)
+		subs	r9, r9, #1
+		ldrgeb	r12, [r10], #1
+		movlt	r12, #0
+		strleb	r12, [r11], #0
+		subles	pc, lr, #4
+		strb	r12, [r11, #-4]
+		subs	pc, lr, #4
+floppy_fiqout_end:

arch/arm/lib/getuser.S

@@ -0,0 +1,159 @@
+/*
+ *  linux/arch/arm/lib/getuser.S
+ *
+ *  Copyright (C) 2001 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ *  Idea from x86 version, (C) Copyright 1998 Linus Torvalds
+ *
+ * These functions have a non-standard call interface to make them more
+ * efficient, especially as they return an error value in addition to
+ * the "real" return value.
+ *
+ * __get_user_X
+ *
+ * Inputs:	r0 contains the address
+ *		r1 contains the address limit, which must be preserved
+ * Outputs:	r0 is the error code
+ *		r2, r3 contains the zero-extended value
+ *		lr corrupted
+ *
+ * No other registers must be altered.  (see <asm/uaccess.h>
+ * for specific ASM register usage).
+ *
+ * Note that ADDR_LIMIT is either 0 or 0xc0000000.
+ * Note also that it is intended that __get_user_bad is not global.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include <asm/errno.h>
+#include <asm/domain.h>
+
+ENTRY(__get_user_1)
+	check_uaccess r0, 1, r1, r2, __get_user_bad
+1: TUSER(ldrb)	r2, [r0]
+	mov	r0, #0
+	ret	lr
+ENDPROC(__get_user_1)
+_ASM_NOKPROBE(__get_user_1)
+
+ENTRY(__get_user_2)
+	check_uaccess r0, 2, r1, r2, __get_user_bad
+#ifdef CONFIG_CPU_USE_DOMAINS
+rb	.req	ip
+2:	ldrbt	r2, [r0], #1
+3:	ldrbt	rb, [r0], #0
+#else
+rb	.req	r0
+2:	ldrb	r2, [r0]
+3:	ldrb	rb, [r0, #1]
+#endif
+#ifndef __ARMEB__
+	orr	r2, r2, rb, lsl #8
+#else
+	orr	r2, rb, r2, lsl #8
+#endif
+	mov	r0, #0
+	ret	lr
+ENDPROC(__get_user_2)
+_ASM_NOKPROBE(__get_user_2)
+
+ENTRY(__get_user_4)
+	check_uaccess r0, 4, r1, r2, __get_user_bad
+4: TUSER(ldr)	r2, [r0]
+	mov	r0, #0
+	ret	lr
+ENDPROC(__get_user_4)
+_ASM_NOKPROBE(__get_user_4)
+
+ENTRY(__get_user_8)
+	check_uaccess r0, 8, r1, r2, __get_user_bad8
+#ifdef CONFIG_THUMB2_KERNEL
+5: TUSER(ldr)	r2, [r0]
+6: TUSER(ldr)	r3, [r0, #4]
+#else
+5: TUSER(ldr)	r2, [r0], #4
+6: TUSER(ldr)	r3, [r0]
+#endif
+	mov	r0, #0
+	ret	lr
+ENDPROC(__get_user_8)
+_ASM_NOKPROBE(__get_user_8)
+
+#ifdef __ARMEB__
+ENTRY(__get_user_32t_8)
+	check_uaccess r0, 8, r1, r2, __get_user_bad
+#ifdef CONFIG_CPU_USE_DOMAINS
+	add	r0, r0, #4
+7:	ldrt	r2, [r0]
+#else
+7:	ldr	r2, [r0, #4]
+#endif
+	mov	r0, #0
+	ret	lr
+ENDPROC(__get_user_32t_8)
+_ASM_NOKPROBE(__get_user_32t_8)
+
+ENTRY(__get_user_64t_1)
+	check_uaccess r0, 1, r1, r2, __get_user_bad8
+8: TUSER(ldrb)	r3, [r0]
+	mov	r0, #0
+	ret	lr
+ENDPROC(__get_user_64t_1)
+_ASM_NOKPROBE(__get_user_64t_1)
+
+ENTRY(__get_user_64t_2)
+	check_uaccess r0, 2, r1, r2, __get_user_bad8
+#ifdef CONFIG_CPU_USE_DOMAINS
+rb	.req	ip
+9:	ldrbt	r3, [r0], #1
+10:	ldrbt	rb, [r0], #0
+#else
+rb	.req	r0
+9:	ldrb	r3, [r0]
+10:	ldrb	rb, [r0, #1]
+#endif
+	orr	r3, rb, r3, lsl #8
+	mov	r0, #0
+	ret	lr
+ENDPROC(__get_user_64t_2)
+_ASM_NOKPROBE(__get_user_64t_2)
+
+ENTRY(__get_user_64t_4)
+	check_uaccess r0, 4, r1, r2, __get_user_bad8
+11: TUSER(ldr)	r3, [r0]
+	mov	r0, #0
+	ret	lr
+ENDPROC(__get_user_64t_4)
+_ASM_NOKPROBE(__get_user_64t_4)
+#endif
+
+__get_user_bad8:
+	mov	r3, #0
+__get_user_bad:
+	mov	r2, #0
+	mov	r0, #-EFAULT
+	ret	lr
+ENDPROC(__get_user_bad)
+ENDPROC(__get_user_bad8)
+_ASM_NOKPROBE(__get_user_bad)
+_ASM_NOKPROBE(__get_user_bad8)
+
+.pushsection __ex_table, "a"
+	.long	1b, __get_user_bad
+	.long	2b, __get_user_bad
+	.long	3b, __get_user_bad
+	.long	4b, __get_user_bad
+	.long	5b, __get_user_bad8
+	.long	6b, __get_user_bad8
+#ifdef __ARMEB__
+	.long   7b, __get_user_bad
+	.long	8b, __get_user_bad8
+	.long	9b, __get_user_bad8
+	.long	10b, __get_user_bad8
+	.long	11b, __get_user_bad8
+#endif
+.popsection

arch/arm/lib/io-acorn.S

@@ -0,0 +1,32 @@
+/*
+ *  linux/arch/arm/lib/io-acorn.S
+ *
+ *  Copyright (C) 1995, 1996 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * 27/03/03 Ian Molton Clean up CONFIG_CPU
+ *
+ */
+#include <linux/linkage.h>
+#include <linux/kern_levels.h>
+#include <asm/assembler.h>
+
+		.text
+		.align
+
+.Liosl_warning:
+		.ascii	KERN_WARNING "insl/outsl not implemented, called from %08lX\0"
+		.align
+
+/*
+ * These make no sense on Acorn machines.
+ * Print a warning message.
+ */
+ENTRY(insl)
+ENTRY(outsl)
+		adr	r0, .Liosl_warning
+		mov	r1, lr
+		b	printk

arch/arm/lib/io-readsb.S

@@ -0,0 +1,123 @@
+/*
+ *  linux/arch/arm/lib/io-readsb.S
+ *
+ *  Copyright (C) 1995-2000 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+.Linsb_align:	rsb	ip, ip, #4
+		cmp	ip, r2
+		movgt	ip, r2
+		cmp	ip, #2
+		ldrb	r3, [r0]
+		strb	r3, [r1], #1
+		ldrgeb	r3, [r0]
+		strgeb	r3, [r1], #1
+		ldrgtb	r3, [r0]
+		strgtb	r3, [r1], #1
+		subs	r2, r2, ip
+		bne	.Linsb_aligned
+
+ENTRY(__raw_readsb)
+		teq	r2, #0		@ do we have to check for the zero len?
+		reteq	lr
+		ands	ip, r1, #3
+		bne	.Linsb_align
+
+.Linsb_aligned:	stmfd	sp!, {r4 - r6, lr}
+
+		subs	r2, r2, #16
+		bmi	.Linsb_no_16
+
+.Linsb_16_lp:	ldrb	r3, [r0]
+		ldrb	r4, [r0]
+		ldrb	r5, [r0]
+		mov	r3, r3,     put_byte_0
+		ldrb	r6, [r0]
+		orr	r3, r3, r4, put_byte_1
+		ldrb	r4, [r0]
+		orr	r3, r3, r5, put_byte_2
+		ldrb	r5, [r0]
+		orr	r3, r3, r6, put_byte_3
+		ldrb	r6, [r0]
+		mov	r4, r4,     put_byte_0
+		ldrb	ip, [r0]
+		orr	r4, r4, r5, put_byte_1
+		ldrb	r5, [r0]
+		orr	r4, r4, r6, put_byte_2
+		ldrb	r6, [r0]
+		orr	r4, r4, ip, put_byte_3
+		ldrb	ip, [r0]
+		mov	r5, r5,     put_byte_0
+		ldrb	lr, [r0]
+		orr	r5, r5, r6, put_byte_1
+		ldrb	r6, [r0]
+		orr	r5, r5, ip, put_byte_2
+		ldrb	ip, [r0]
+		orr	r5, r5, lr, put_byte_3
+		ldrb	lr, [r0]
+		mov	r6, r6,     put_byte_0
+		orr	r6, r6, ip, put_byte_1
+		ldrb	ip, [r0]
+		orr	r6, r6, lr, put_byte_2
+		orr	r6, r6, ip, put_byte_3
+		stmia	r1!, {r3 - r6}
+
+		subs	r2, r2, #16
+		bpl	.Linsb_16_lp
+
+		tst	r2, #15
+		ldmeqfd	sp!, {r4 - r6, pc}
+
+.Linsb_no_16:	tst	r2, #8
+		beq	.Linsb_no_8
+
+		ldrb	r3, [r0]
+		ldrb	r4, [r0]
+		ldrb	r5, [r0]
+		mov	r3, r3,     put_byte_0
+		ldrb	r6, [r0]
+		orr	r3, r3, r4, put_byte_1
+		ldrb	r4, [r0]
+		orr	r3, r3, r5, put_byte_2
+		ldrb	r5, [r0]
+		orr	r3, r3, r6, put_byte_3
+		ldrb	r6, [r0]
+		mov	r4, r4,     put_byte_0
+		ldrb	ip, [r0]
+		orr	r4, r4, r5, put_byte_1
+		orr	r4, r4, r6, put_byte_2
+		orr	r4, r4, ip, put_byte_3
+		stmia	r1!, {r3, r4}
+
+.Linsb_no_8:	tst	r2, #4
+		beq	.Linsb_no_4
+
+		ldrb	r3, [r0]
+		ldrb	r4, [r0]
+		ldrb	r5, [r0]
+		ldrb	r6, [r0]
+		mov	r3, r3,     put_byte_0
+		orr	r3, r3, r4, put_byte_1
+		orr	r3, r3, r5, put_byte_2
+		orr	r3, r3, r6, put_byte_3
+		str	r3, [r1], #4
+
+.Linsb_no_4:	ands	r2, r2, #3
+		ldmeqfd	sp!, {r4 - r6, pc}
+
+		cmp	r2, #2
+		ldrb	r3, [r0]
+		strb	r3, [r1], #1
+		ldrgeb	r3, [r0]
+		strgeb	r3, [r1], #1
+		ldrgtb	r3, [r0]
+		strgtb	r3, [r1]
+
+		ldmfd	sp!, {r4 - r6, pc}
+ENDPROC(__raw_readsb)

arch/arm/lib/io-readsl.S

@@ -0,0 +1,79 @@
+/*
+ *  linux/arch/arm/lib/io-readsl.S
+ *
+ *  Copyright (C) 1995-2000 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+ENTRY(__raw_readsl)
+		teq	r2, #0		@ do we have to check for the zero len?
+		reteq	lr
+		ands	ip, r1, #3
+		bne	3f
+
+		subs	r2, r2, #4
+		bmi	2f
+		stmfd	sp!, {r4, lr}
+1:		ldr	r3, [r0, #0]
+		ldr	r4, [r0, #0]
+		ldr	ip, [r0, #0]
+		ldr	lr, [r0, #0]
+		subs	r2, r2, #4
+		stmia	r1!, {r3, r4, ip, lr}
+		bpl	1b
+		ldmfd	sp!, {r4, lr}
+2:		movs	r2, r2, lsl #31
+		ldrcs	r3, [r0, #0]
+		ldrcs	ip, [r0, #0]
+		stmcsia	r1!, {r3, ip}
+		ldrne	r3, [r0, #0]
+		strne	r3, [r1, #0]
+		ret	lr
+
+3:		ldr	r3, [r0]
+		cmp	ip, #2
+		mov	ip, r3, get_byte_0
+		strb	ip, [r1], #1
+		bgt	6f
+		mov	ip, r3, get_byte_1
+		strb	ip, [r1], #1
+		beq	5f
+		mov	ip, r3, get_byte_2
+		strb	ip, [r1], #1
+
+4:		subs	r2, r2, #1
+		mov	ip, r3, lspull #24
+		ldrne	r3, [r0]
+		orrne	ip, ip, r3, lspush #8
+		strne	ip, [r1], #4
+		bne	4b
+		b	8f
+
+5:		subs	r2, r2, #1
+		mov	ip, r3, lspull #16
+		ldrne	r3, [r0]
+		orrne	ip, ip, r3, lspush #16
+		strne	ip, [r1], #4
+		bne	5b
+		b	7f
+
+6:		subs	r2, r2, #1
+		mov	ip, r3, lspull #8
+		ldrne	r3, [r0]
+		orrne	ip, ip, r3, lspush #24
+		strne	ip, [r1], #4
+		bne	6b
+
+		mov	r3, ip, get_byte_2
+		strb	r3, [r1, #2]
+7:		mov	r3, ip, get_byte_1
+		strb	r3, [r1, #1]
+8:		mov	r3, ip, get_byte_0
+		strb	r3, [r1, #0]
+		ret	lr
+ENDPROC(__raw_readsl)

arch/arm/lib/io-readsw-armv3.S

@@ -0,0 +1,106 @@
+/*
+ *  linux/arch/arm/lib/io-readsw-armv3.S
+ *
+ *  Copyright (C) 1995-2000 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+.Linsw_bad_alignment:
+		adr	r0, .Linsw_bad_align_msg
+		mov	r2, lr
+		b	panic
+.Linsw_bad_align_msg:
+		.asciz	"insw: bad buffer alignment (0x%p, lr=0x%08lX)\n"
+		.align
+
+.Linsw_align:	tst	r1, #1
+		bne	.Linsw_bad_alignment
+
+		ldr	r3, [r0]
+		strb	r3, [r1], #1
+		mov	r3, r3, lsr #8
+		strb	r3, [r1], #1
+
+		subs	r2, r2, #1
+		reteq	lr
+
+ENTRY(__raw_readsw)
+		teq	r2, #0		@ do we have to check for the zero len?
+		reteq	lr
+		tst	r1, #3
+		bne	.Linsw_align
+
+.Linsw_aligned:	mov	ip, #0xff
+		orr	ip, ip, ip, lsl #8
+		stmfd	sp!, {r4, r5, r6, lr}
+
+		subs	r2, r2, #8
+		bmi	.Lno_insw_8
+
+.Linsw_8_lp:	ldr	r3, [r0]
+		and	r3, r3, ip
+		ldr	r4, [r0]
+		orr	r3, r3, r4, lsl #16
+
+		ldr	r4, [r0]
+		and	r4, r4, ip
+		ldr	r5, [r0]
+		orr	r4, r4, r5, lsl #16
+
+		ldr	r5, [r0]
+		and	r5, r5, ip
+		ldr	r6, [r0]
+		orr	r5, r5, r6, lsl #16
+
+		ldr	r6, [r0]
+		and	r6, r6, ip
+		ldr	lr, [r0]
+		orr	r6, r6, lr, lsl #16
+
+		stmia	r1!, {r3 - r6}
+
+		subs	r2, r2, #8
+		bpl	.Linsw_8_lp
+
+		tst	r2, #7
+		ldmeqfd	sp!, {r4, r5, r6, pc}
+
+.Lno_insw_8:	tst	r2, #4
+		beq	.Lno_insw_4
+
+		ldr	r3, [r0]
+		and	r3, r3, ip
+		ldr	r4, [r0]
+		orr	r3, r3, r4, lsl #16
+
+		ldr	r4, [r0]
+		and	r4, r4, ip
+		ldr	r5, [r0]
+		orr	r4, r4, r5, lsl #16
+
+		stmia	r1!, {r3, r4}
+
+.Lno_insw_4:	tst	r2, #2
+		beq	.Lno_insw_2
+
+		ldr	r3, [r0]
+		and	r3, r3, ip
+		ldr	r4, [r0]
+		orr	r3, r3, r4, lsl #16
+
+		str	r3, [r1], #4
+
+.Lno_insw_2:	tst	r2, #1
+		ldrne	r3, [r0]
+		strneb	r3, [r1], #1
+		movne	r3, r3, lsr #8
+		strneb	r3, [r1]
+
+		ldmfd	sp!, {r4, r5, r6, pc}
+
+

arch/arm/lib/io-readsw-armv4.S

@@ -0,0 +1,131 @@
+/*
+ *  linux/arch/arm/lib/io-readsw-armv4.S
+ *
+ *  Copyright (C) 1995-2000 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+		.macro	pack, rd, hw1, hw2
+#ifndef __ARMEB__
+		orr	\rd, \hw1, \hw2, lsl #16
+#else
+		orr	\rd, \hw2, \hw1, lsl #16
+#endif
+		.endm
+
+.Linsw_align:	movs	ip, r1, lsl #31
+		bne	.Linsw_noalign
+		ldrh	ip, [r0]
+		sub	r2, r2, #1
+		strh	ip, [r1], #2
+
+ENTRY(__raw_readsw)
+		teq	r2, #0
+		reteq	lr
+		tst	r1, #3
+		bne	.Linsw_align
+
+		stmfd	sp!, {r4, r5, lr}
+
+		subs	r2, r2, #8
+		bmi	.Lno_insw_8
+
+.Linsw_8_lp:	ldrh	r3, [r0]
+		ldrh	r4, [r0]
+		pack	r3, r3, r4
+
+		ldrh	r4, [r0]
+		ldrh	r5, [r0]
+		pack	r4, r4, r5
+
+		ldrh	r5, [r0]
+		ldrh	ip, [r0]
+		pack	r5, r5, ip
+
+		ldrh	ip, [r0]
+		ldrh	lr, [r0]
+		pack	ip, ip, lr
+
+		subs	r2, r2, #8
+		stmia	r1!, {r3 - r5, ip}
+		bpl	.Linsw_8_lp
+
+.Lno_insw_8:	tst	r2, #4
+		beq	.Lno_insw_4
+
+		ldrh	r3, [r0]
+		ldrh	r4, [r0]
+		pack	r3, r3, r4
+
+		ldrh	r4, [r0]
+		ldrh	ip, [r0]
+		pack	r4, r4, ip
+
+		stmia	r1!, {r3, r4}
+
+.Lno_insw_4:	movs	r2, r2, lsl #31
+		bcc	.Lno_insw_2
+
+		ldrh	r3, [r0]
+		ldrh	ip, [r0]
+		pack	r3, r3, ip
+		str	r3, [r1], #4
+
+.Lno_insw_2:	ldrneh	r3, [r0]
+		strneh	r3, [r1]
+
+		ldmfd	sp!, {r4, r5, pc}
+
+#ifdef __ARMEB__
+#define _BE_ONLY_(code...)	code
+#define _LE_ONLY_(code...)
+#define push_hbyte0		lsr #8
+#define pull_hbyte1		lsl #24
+#else
+#define _BE_ONLY_(code...)
+#define _LE_ONLY_(code...) code
+#define push_hbyte0		lsl #24
+#define pull_hbyte1		lsr #8
+#endif
+
+.Linsw_noalign:	stmfd	sp!, {r4, lr}
+		ldrccb	ip, [r1, #-1]!
+		bcc	1f
+
+		ldrh	ip, [r0]
+		sub	r2, r2, #1
+   _BE_ONLY_(	mov	ip, ip, ror #8		)
+		strb	ip, [r1], #1
+   _LE_ONLY_(	mov	ip, ip, lsr #8		)
+   _BE_ONLY_(	mov	ip, ip, lsr #24		)
+
+1:		subs	r2, r2, #2
+		bmi	3f
+   _BE_ONLY_(	mov	ip, ip, lsl #24		)
+
+2:		ldrh	r3, [r0]
+		ldrh	r4, [r0]
+		subs	r2, r2, #2
+		orr	ip, ip, r3, lsl #8
+		orr	ip, ip, r4, push_hbyte0
+		str	ip, [r1], #4
+		mov	ip, r4, pull_hbyte1
+		bpl	2b
+
+   _BE_ONLY_(	mov	ip, ip, lsr #24		)
+
+3:		tst	r2, #1
+		strb	ip, [r1], #1
+		ldrneh	ip, [r0]
+   _BE_ONLY_(	movne	ip, ip, ror #8		)
+		strneb	ip, [r1], #1
+   _LE_ONLY_(	movne	ip, ip, lsr #8		)
+   _BE_ONLY_(	movne	ip, ip, lsr #24		)
+		strneb	ip, [r1]
+		ldmfd	sp!, {r4, pc}
+ENDPROC(__raw_readsw)

arch/arm/lib/io-writesb.S

@@ -0,0 +1,94 @@
+/*
+ *  linux/arch/arm/lib/io-writesb.S
+ *
+ *  Copyright (C) 1995-2000 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+		.macro	outword, rd
+#ifndef __ARMEB__
+		strb	\rd, [r0]
+		mov	\rd, \rd, lsr #8
+		strb	\rd, [r0]
+		mov	\rd, \rd, lsr #8
+		strb	\rd, [r0]
+		mov	\rd, \rd, lsr #8
+		strb	\rd, [r0]
+#else
+		mov	lr, \rd, lsr #24
+		strb	lr, [r0]
+		mov	lr, \rd, lsr #16
+		strb	lr, [r0]
+		mov	lr, \rd, lsr #8
+		strb	lr, [r0]
+		strb	\rd, [r0]
+#endif
+		.endm
+
+.Loutsb_align:	rsb	ip, ip, #4
+		cmp	ip, r2
+		movgt	ip, r2
+		cmp	ip, #2
+		ldrb	r3, [r1], #1
+		strb	r3, [r0]
+		ldrgeb	r3, [r1], #1
+		strgeb	r3, [r0]
+		ldrgtb	r3, [r1], #1
+		strgtb	r3, [r0]
+		subs	r2, r2, ip
+		bne	.Loutsb_aligned
+
+ENTRY(__raw_writesb)
+		teq	r2, #0		@ do we have to check for the zero len?
+		reteq	lr
+		ands	ip, r1, #3
+		bne	.Loutsb_align
+
+.Loutsb_aligned:
+		stmfd	sp!, {r4, r5, lr}
+
+		subs	r2, r2, #16
+		bmi	.Loutsb_no_16
+
+.Loutsb_16_lp:	ldmia	r1!, {r3, r4, r5, ip}
+		outword	r3
+		outword	r4
+		outword	r5
+		outword	ip
+		subs	r2, r2, #16
+		bpl	.Loutsb_16_lp
+
+		tst	r2, #15
+		ldmeqfd	sp!, {r4, r5, pc}
+
+.Loutsb_no_16:	tst	r2, #8
+		beq	.Loutsb_no_8
+
+		ldmia	r1!, {r3, r4}
+		outword	r3
+		outword	r4
+
+.Loutsb_no_8:	tst	r2, #4
+		beq	.Loutsb_no_4
+
+		ldr	r3, [r1], #4
+		outword	r3
+
+.Loutsb_no_4:	ands	r2, r2, #3
+		ldmeqfd	sp!, {r4, r5, pc}
+
+		cmp	r2, #2
+		ldrb	r3, [r1], #1
+		strb	r3, [r0]
+		ldrgeb	r3, [r1], #1
+		strgeb	r3, [r0]
+		ldrgtb	r3, [r1]
+		strgtb	r3, [r0]
+
+		ldmfd	sp!, {r4, r5, pc}
+ENDPROC(__raw_writesb)

arch/arm/lib/io-writesl.S

@@ -0,0 +1,67 @@
+/*
+ *  linux/arch/arm/lib/io-writesl.S
+ *
+ *  Copyright (C) 1995-2000 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+ENTRY(__raw_writesl)
+		teq	r2, #0		@ do we have to check for the zero len?
+		reteq	lr
+		ands	ip, r1, #3
+		bne	3f
+
+		subs	r2, r2, #4
+		bmi	2f
+		stmfd	sp!, {r4, lr}
+1:		ldmia	r1!, {r3, r4, ip, lr}
+		subs	r2, r2, #4
+		str	r3, [r0, #0]
+		str	r4, [r0, #0]
+		str	ip, [r0, #0]
+		str	lr, [r0, #0]
+		bpl	1b
+		ldmfd	sp!, {r4, lr}
+2:		movs	r2, r2, lsl #31
+		ldmcsia	r1!, {r3, ip}
+		strcs	r3, [r0, #0]
+		ldrne	r3, [r1, #0]
+		strcs	ip, [r0, #0]
+		strne	r3, [r0, #0]
+		ret	lr
+
+3:		bic	r1, r1, #3
+		ldr	r3, [r1], #4
+		cmp	ip, #2
+		blt	5f
+		bgt	6f
+
+4:		mov	ip, r3, lspull #16
+		ldr	r3, [r1], #4
+		subs	r2, r2, #1
+		orr	ip, ip, r3, lspush #16
+		str	ip, [r0]
+		bne	4b
+		ret	lr
+
+5:		mov	ip, r3, lspull #8
+		ldr	r3, [r1], #4
+		subs	r2, r2, #1
+		orr	ip, ip, r3, lspush #24
+		str	ip, [r0]
+		bne	5b
+		ret	lr
+
+6:		mov	ip, r3, lspull #24
+		ldr	r3, [r1], #4
+		subs	r2, r2, #1
+		orr	ip, ip, r3, lspush #8
+		str	ip, [r0]
+		bne	6b
+		ret	lr
+ENDPROC(__raw_writesl)

arch/arm/lib/io-writesw-armv3.S

@@ -0,0 +1,126 @@
+/*
+ *  linux/arch/arm/lib/io-writesw-armv3.S
+ *
+ *  Copyright (C) 1995-2000 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+.Loutsw_bad_alignment:
+		adr	r0, .Loutsw_bad_align_msg
+		mov	r2, lr
+		b	panic
+.Loutsw_bad_align_msg:
+		.asciz	"outsw: bad buffer alignment (0x%p, lr=0x%08lX)\n"
+		.align
+
+.Loutsw_align:	tst	r1, #1
+		bne	.Loutsw_bad_alignment
+
+		add	r1, r1, #2
+
+		ldr	r3, [r1, #-4]
+		mov	r3, r3, lsr #16
+		orr	r3, r3, r3, lsl #16
+		str	r3, [r0]
+		subs	r2, r2, #1
+		reteq	lr
+
+ENTRY(__raw_writesw)
+		teq	r2, #0		@ do we have to check for the zero len?
+		reteq	lr
+		tst	r1, #3
+		bne	.Loutsw_align
+
+		stmfd	sp!, {r4, r5, r6, lr}
+
+		subs	r2, r2, #8
+		bmi	.Lno_outsw_8
+
+.Loutsw_8_lp:	ldmia	r1!, {r3, r4, r5, r6}
+
+		mov	ip, r3, lsl #16
+		orr	ip, ip, ip, lsr #16
+		str	ip, [r0]
+
+		mov	ip, r3, lsr #16
+		orr	ip, ip, ip, lsl #16
+		str	ip, [r0]
+
+		mov	ip, r4, lsl #16
+		orr	ip, ip, ip, lsr #16
+		str	ip, [r0]
+
+		mov	ip, r4, lsr #16
+		orr	ip, ip, ip, lsl #16
+		str	ip, [r0]
+
+		mov	ip, r5, lsl #16
+		orr	ip, ip, ip, lsr #16
+		str	ip, [r0]
+
+		mov	ip, r5, lsr #16
+		orr	ip, ip, ip, lsl #16
+		str	ip, [r0]
+
+		mov	ip, r6, lsl #16
+		orr	ip, ip, ip, lsr #16
+		str	ip, [r0]
+
+		mov	ip, r6, lsr #16
+		orr	ip, ip, ip, lsl #16
+		str	ip, [r0]
+
+		subs	r2, r2, #8
+		bpl	.Loutsw_8_lp
+
+		tst	r2, #7
+		ldmeqfd	sp!, {r4, r5, r6, pc}
+
+.Lno_outsw_8:	tst	r2, #4
+		beq	.Lno_outsw_4
+
+		ldmia	r1!, {r3, r4}
+
+		mov	ip, r3, lsl #16
+		orr	ip, ip, ip, lsr #16
+		str	ip, [r0]
+
+		mov	ip, r3, lsr #16
+		orr	ip, ip, ip, lsl #16
+		str	ip, [r0]
+
+		mov	ip, r4, lsl #16
+		orr	ip, ip, ip, lsr #16
+		str	ip, [r0]
+
+		mov	ip, r4, lsr #16
+		orr	ip, ip, ip, lsl #16
+		str	ip, [r0]
+
+.Lno_outsw_4:	tst	r2, #2
+		beq	.Lno_outsw_2
+
+		ldr	r3, [r1], #4
+
+		mov	ip, r3, lsl #16
+		orr	ip, ip, ip, lsr #16
+		str	ip, [r0]
+
+		mov	ip, r3, lsr #16
+		orr	ip, ip, ip, lsl #16
+		str	ip, [r0]
+
+.Lno_outsw_2:	tst	r2, #1
+
+		ldrne	r3, [r1]
+
+		movne	ip, r3, lsl #16
+		orrne	ip, ip, ip, lsr #16
+		strne	ip, [r0]
+
+		ldmfd	sp!, {r4, r5, r6, pc}

arch/arm/lib/io-writesw-armv4.S

@@ -0,0 +1,100 @@
+/*
+ *  linux/arch/arm/lib/io-writesw-armv4.S
+ *
+ *  Copyright (C) 1995-2000 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+		.macro	outword, rd
+#ifndef __ARMEB__
+		strh	\rd, [r0]
+		mov	\rd, \rd, lsr #16
+		strh	\rd, [r0]
+#else
+		mov	lr, \rd, lsr #16
+		strh	lr, [r0]
+		strh	\rd, [r0]
+#endif
+		.endm
+
+.Loutsw_align:	movs	ip, r1, lsl #31
+		bne	.Loutsw_noalign
+
+		ldrh	r3, [r1], #2
+		sub	r2, r2, #1
+		strh	r3, [r0]
+
+ENTRY(__raw_writesw)
+		teq	r2, #0
+		reteq	lr
+		ands	r3, r1, #3
+		bne	.Loutsw_align
+
+		stmfd	sp!, {r4, r5, lr}
+
+		subs	r2, r2, #8
+		bmi	.Lno_outsw_8
+
+.Loutsw_8_lp:	ldmia	r1!, {r3, r4, r5, ip}
+		subs	r2, r2, #8
+		outword	r3
+		outword	r4
+		outword	r5
+		outword	ip
+		bpl	.Loutsw_8_lp
+
+.Lno_outsw_8:	tst	r2, #4
+		beq	.Lno_outsw_4
+
+		ldmia	r1!, {r3, ip}
+		outword	r3
+		outword	ip
+
+.Lno_outsw_4:	movs	r2, r2, lsl #31
+		bcc	.Lno_outsw_2
+
+		ldr	r3, [r1], #4
+		outword	r3
+
+.Lno_outsw_2:	ldrneh	r3, [r1]
+		strneh	r3, [r0]
+
+		ldmfd	sp!, {r4, r5, pc}
+
+#ifdef __ARMEB__
+#define pull_hbyte0	lsl #8
+#define push_hbyte1	lsr #24
+#else
+#define pull_hbyte0	lsr #24
+#define push_hbyte1	lsl #8
+#endif
+
+.Loutsw_noalign:
+ ARM(		ldr	r3, [r1, -r3]!	)
+ THUMB(		rsb	r3, r3, #0	)
+ THUMB(		ldr	r3, [r1, r3]	)
+ THUMB(		sub	r1, r3		)
+		subcs	r2, r2, #1
+		bcs	2f
+		subs	r2, r2, #2
+		bmi	3f
+
+1:		mov	ip, r3, lsr #8
+		strh	ip, [r0]
+2:		mov	ip, r3, pull_hbyte0
+		ldr	r3, [r1, #4]!
+		subs	r2, r2, #2
+		orr	ip, ip, r3, push_hbyte1
+		strh	ip, [r0]
+		bpl	1b
+
+		tst	r2, #1
+3:		movne	ip, r3, lsr #8
+		strneh	ip, [r0]
+		ret	lr
+ENDPROC(__raw_writesw)

arch/arm/lib/lib1funcs.S

@@ -0,0 +1,371 @@
+/*
+ * linux/arch/arm/lib/lib1funcs.S: Optimized ARM division routines
+ *
+ * Author: Nicolas Pitre <nico@fluxnic.net>
+ *   - contributed to gcc-3.4 on Sep 30, 2003
+ *   - adapted for the Linux kernel on Oct 2, 2003
+ */
+
+/* Copyright 1995, 1996, 1998, 1999, 2000, 2003 Free Software Foundation, Inc.
+
+This file is free software; you can redistribute it and/or modify it
+under the terms of the GNU General Public License as published by the
+Free Software Foundation; either version 2, or (at your option) any
+later version.
+
+In addition to the permissions in the GNU General Public License, the
+Free Software Foundation gives you unlimited permission to link the
+compiled version of this file into combinations with other programs,
+and to distribute those combinations without any restriction coming
+from the use of this file.  (The General Public License restrictions
+do apply in other respects; for example, they cover modification of
+the file, and distribution when not linked into a combine
+executable.)
+
+This file is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with this program; see the file COPYING.  If not, write to
+the Free Software Foundation, 59 Temple Place - Suite 330,
+Boston, MA 02111-1307, USA.  */
+
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include <asm/unwind.h>
+
+.macro ARM_DIV_BODY dividend, divisor, result, curbit
+
+#if __LINUX_ARM_ARCH__ >= 5
+
+	clz	\curbit, \divisor
+	clz	\result, \dividend
+	sub	\result, \curbit, \result
+	mov	\curbit, #1
+	mov	\divisor, \divisor, lsl \result
+	mov	\curbit, \curbit, lsl \result
+	mov	\result, #0
+	
+#else
+
+	@ Initially shift the divisor left 3 bits if possible,
+	@ set curbit accordingly.  This allows for curbit to be located
+	@ at the left end of each 4 bit nibbles in the division loop
+	@ to save one loop in most cases.
+	tst	\divisor, #0xe0000000
+	moveq	\divisor, \divisor, lsl #3
+	moveq	\curbit, #8
+	movne	\curbit, #1
+
+	@ Unless the divisor is very big, shift it up in multiples of
+	@ four bits, since this is the amount of unwinding in the main
+	@ division loop.  Continue shifting until the divisor is 
+	@ larger than the dividend.
+1:	cmp	\divisor, #0x10000000
+	cmplo	\divisor, \dividend
+	movlo	\divisor, \divisor, lsl #4
+	movlo	\curbit, \curbit, lsl #4
+	blo	1b
+
+	@ For very big divisors, we must shift it a bit at a time, or
+	@ we will be in danger of overflowing.
+1:	cmp	\divisor, #0x80000000
+	cmplo	\divisor, \dividend
+	movlo	\divisor, \divisor, lsl #1
+	movlo	\curbit, \curbit, lsl #1
+	blo	1b
+
+	mov	\result, #0
+
+#endif
+
+	@ Division loop
+1:	cmp	\dividend, \divisor
+	subhs	\dividend, \dividend, \divisor
+	orrhs	\result,   \result,   \curbit
+	cmp	\dividend, \divisor,  lsr #1
+	subhs	\dividend, \dividend, \divisor, lsr #1
+	orrhs	\result,   \result,   \curbit,  lsr #1
+	cmp	\dividend, \divisor,  lsr #2
+	subhs	\dividend, \dividend, \divisor, lsr #2
+	orrhs	\result,   \result,   \curbit,  lsr #2
+	cmp	\dividend, \divisor,  lsr #3
+	subhs	\dividend, \dividend, \divisor, lsr #3
+	orrhs	\result,   \result,   \curbit,  lsr #3
+	cmp	\dividend, #0			@ Early termination?
+	movnes	\curbit,   \curbit,  lsr #4	@ No, any more bits to do?
+	movne	\divisor,  \divisor, lsr #4
+	bne	1b
+
+.endm
+
+
+.macro ARM_DIV2_ORDER divisor, order
+
+#if __LINUX_ARM_ARCH__ >= 5
+
+	clz	\order, \divisor
+	rsb	\order, \order, #31
+
+#else
+
+	cmp	\divisor, #(1 << 16)
+	movhs	\divisor, \divisor, lsr #16
+	movhs	\order, #16
+	movlo	\order, #0
+
+	cmp	\divisor, #(1 << 8)
+	movhs	\divisor, \divisor, lsr #8
+	addhs	\order, \order, #8
+
+	cmp	\divisor, #(1 << 4)
+	movhs	\divisor, \divisor, lsr #4
+	addhs	\order, \order, #4
+
+	cmp	\divisor, #(1 << 2)
+	addhi	\order, \order, #3
+	addls	\order, \order, \divisor, lsr #1
+
+#endif
+
+.endm
+
+
+.macro ARM_MOD_BODY dividend, divisor, order, spare
+
+#if __LINUX_ARM_ARCH__ >= 5
+
+	clz	\order, \divisor
+	clz	\spare, \dividend
+	sub	\order, \order, \spare
+	mov	\divisor, \divisor, lsl \order
+
+#else
+
+	mov	\order, #0
+
+	@ Unless the divisor is very big, shift it up in multiples of
+	@ four bits, since this is the amount of unwinding in the main
+	@ division loop.  Continue shifting until the divisor is 
+	@ larger than the dividend.
+1:	cmp	\divisor, #0x10000000
+	cmplo	\divisor, \dividend
+	movlo	\divisor, \divisor, lsl #4
+	addlo	\order, \order, #4
+	blo	1b
+
+	@ For very big divisors, we must shift it a bit at a time, or
+	@ we will be in danger of overflowing.
+1:	cmp	\divisor, #0x80000000
+	cmplo	\divisor, \dividend
+	movlo	\divisor, \divisor, lsl #1
+	addlo	\order, \order, #1
+	blo	1b
+
+#endif
+
+	@ Perform all needed subtractions to keep only the reminder.
+	@ Do comparisons in batch of 4 first.
+	subs	\order, \order, #3		@ yes, 3 is intended here
+	blt	2f
+
+1:	cmp	\dividend, \divisor
+	subhs	\dividend, \dividend, \divisor
+	cmp	\dividend, \divisor,  lsr #1
+	subhs	\dividend, \dividend, \divisor, lsr #1
+	cmp	\dividend, \divisor,  lsr #2
+	subhs	\dividend, \dividend, \divisor, lsr #2
+	cmp	\dividend, \divisor,  lsr #3
+	subhs	\dividend, \dividend, \divisor, lsr #3
+	cmp	\dividend, #1
+	mov	\divisor, \divisor, lsr #4
+	subges	\order, \order, #4
+	bge	1b
+
+	tst	\order, #3
+	teqne	\dividend, #0
+	beq	5f
+
+	@ Either 1, 2 or 3 comparison/subtractions are left.
+2:	cmn	\order, #2
+	blt	4f
+	beq	3f
+	cmp	\dividend, \divisor
+	subhs	\dividend, \dividend, \divisor
+	mov	\divisor,  \divisor,  lsr #1
+3:	cmp	\dividend, \divisor
+	subhs	\dividend, \dividend, \divisor
+	mov	\divisor,  \divisor,  lsr #1
+4:	cmp	\dividend, \divisor
+	subhs	\dividend, \dividend, \divisor
+5:
+.endm
+
+
+#ifdef CONFIG_ARM_PATCH_IDIV
+	.align	3
+#endif
+
+ENTRY(__udivsi3)
+ENTRY(__aeabi_uidiv)
+UNWIND(.fnstart)
+
+	subs	r2, r1, #1
+	reteq	lr
+	bcc	Ldiv0
+	cmp	r0, r1
+	bls	11f
+	tst	r1, r2
+	beq	12f
+
+	ARM_DIV_BODY r0, r1, r2, r3
+
+	mov	r0, r2
+	ret	lr
+
+11:	moveq	r0, #1
+	movne	r0, #0
+	ret	lr
+
+12:	ARM_DIV2_ORDER r1, r2
+
+	mov	r0, r0, lsr r2
+	ret	lr
+
+UNWIND(.fnend)
+ENDPROC(__udivsi3)
+ENDPROC(__aeabi_uidiv)
+
+ENTRY(__umodsi3)
+UNWIND(.fnstart)
+
+	subs	r2, r1, #1			@ compare divisor with 1
+	bcc	Ldiv0
+	cmpne	r0, r1				@ compare dividend with divisor
+	moveq   r0, #0
+	tsthi	r1, r2				@ see if divisor is power of 2
+	andeq	r0, r0, r2
+	retls	lr
+
+	ARM_MOD_BODY r0, r1, r2, r3
+
+	ret	lr
+
+UNWIND(.fnend)
+ENDPROC(__umodsi3)
+
+#ifdef CONFIG_ARM_PATCH_IDIV
+	.align 3
+#endif
+
+ENTRY(__divsi3)
+ENTRY(__aeabi_idiv)
+UNWIND(.fnstart)
+
+	cmp	r1, #0
+	eor	ip, r0, r1			@ save the sign of the result.
+	beq	Ldiv0
+	rsbmi	r1, r1, #0			@ loops below use unsigned.
+	subs	r2, r1, #1			@ division by 1 or -1 ?
+	beq	10f
+	movs	r3, r0
+	rsbmi	r3, r0, #0			@ positive dividend value
+	cmp	r3, r1
+	bls	11f
+	tst	r1, r2				@ divisor is power of 2 ?
+	beq	12f
+
+	ARM_DIV_BODY r3, r1, r0, r2
+
+	cmp	ip, #0
+	rsbmi	r0, r0, #0
+	ret	lr
+
+10:	teq	ip, r0				@ same sign ?
+	rsbmi	r0, r0, #0
+	ret	lr
+
+11:	movlo	r0, #0
+	moveq	r0, ip, asr #31
+	orreq	r0, r0, #1
+	ret	lr
+
+12:	ARM_DIV2_ORDER r1, r2
+
+	cmp	ip, #0
+	mov	r0, r3, lsr r2
+	rsbmi	r0, r0, #0
+	ret	lr
+
+UNWIND(.fnend)
+ENDPROC(__divsi3)
+ENDPROC(__aeabi_idiv)
+
+ENTRY(__modsi3)
+UNWIND(.fnstart)
+
+	cmp	r1, #0
+	beq	Ldiv0
+	rsbmi	r1, r1, #0			@ loops below use unsigned.
+	movs	ip, r0				@ preserve sign of dividend
+	rsbmi	r0, r0, #0			@ if negative make positive
+	subs	r2, r1, #1			@ compare divisor with 1
+	cmpne	r0, r1				@ compare dividend with divisor
+	moveq	r0, #0
+	tsthi	r1, r2				@ see if divisor is power of 2
+	andeq	r0, r0, r2
+	bls	10f
+
+	ARM_MOD_BODY r0, r1, r2, r3
+
+10:	cmp	ip, #0
+	rsbmi	r0, r0, #0
+	ret	lr
+
+UNWIND(.fnend)
+ENDPROC(__modsi3)
+
+#ifdef CONFIG_AEABI
+
+ENTRY(__aeabi_uidivmod)
+UNWIND(.fnstart)
+UNWIND(.save {r0, r1, ip, lr}	)
+
+	stmfd	sp!, {r0, r1, ip, lr}
+	bl	__aeabi_uidiv
+	ldmfd	sp!, {r1, r2, ip, lr}
+	mul	r3, r0, r2
+	sub	r1, r1, r3
+	ret	lr
+
+UNWIND(.fnend)
+ENDPROC(__aeabi_uidivmod)
+
+ENTRY(__aeabi_idivmod)
+UNWIND(.fnstart)
+UNWIND(.save {r0, r1, ip, lr}	)
+	stmfd	sp!, {r0, r1, ip, lr}
+	bl	__aeabi_idiv
+	ldmfd	sp!, {r1, r2, ip, lr}
+	mul	r3, r0, r2
+	sub	r1, r1, r3
+	ret	lr
+
+UNWIND(.fnend)
+ENDPROC(__aeabi_idivmod)
+
+#endif
+
+Ldiv0:
+UNWIND(.fnstart)
+UNWIND(.pad #4)
+UNWIND(.save {lr})
+	str	lr, [sp, #-8]!
+	bl	__div0
+	mov	r0, #0			@ About as wrong as it could be.
+	ldr	pc, [sp], #8
+UNWIND(.fnend)
+ENDPROC(Ldiv0)

arch/arm/lib/lshrdi3.S

@@ -0,0 +1,54 @@
+/* Copyright 1995, 1996, 1998, 1999, 2000, 2003, 2004, 2005
+   Free Software Foundation, Inc.
+
+This file is free software; you can redistribute it and/or modify it
+under the terms of the GNU General Public License as published by the
+Free Software Foundation; either version 2, or (at your option) any
+later version.
+
+In addition to the permissions in the GNU General Public License, the
+Free Software Foundation gives you unlimited permission to link the
+compiled version of this file into combinations with other programs,
+and to distribute those combinations without any restriction coming
+from the use of this file.  (The General Public License restrictions
+do apply in other respects; for example, they cover modification of
+the file, and distribution when not linked into a combine
+executable.)
+
+This file is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with this program; see the file COPYING.  If not, write to
+the Free Software Foundation, 51 Franklin Street, Fifth Floor,
+Boston, MA 02110-1301, USA.  */
+
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+#ifdef __ARMEB__
+#define al r1
+#define ah r0
+#else
+#define al r0
+#define ah r1
+#endif
+
+ENTRY(__lshrdi3)
+ENTRY(__aeabi_llsr)
+
+	subs	r3, r2, #32
+	rsb	ip, r2, #32
+	movmi	al, al, lsr r2
+	movpl	al, ah, lsr r3
+ ARM(	orrmi	al, al, ah, lsl ip	)
+ THUMB(	lslmi	r3, ah, ip		)
+ THUMB(	orrmi	al, al, r3		)
+	mov	ah, ah, lsr r2
+	ret	lr
+
+ENDPROC(__lshrdi3)
+ENDPROC(__aeabi_llsr)

arch/arm/lib/memchr.S

@@ -0,0 +1,26 @@
+/*
+ *  linux/arch/arm/lib/memchr.S
+ *
+ *  Copyright (C) 1995-2000 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ *  ASM optimised string functions
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+	.text
+	.align	5
+ENTRY(memchr)
+1:	subs	r2, r2, #1
+	bmi	2f
+	ldrb	r3, [r0], #1
+	teq	r3, r1
+	bne	1b
+	sub	r0, r0, #1
+2:	movne	r0, #0
+	ret	lr
+ENDPROC(memchr)

arch/arm/lib/memcpy.S

@@ -0,0 +1,70 @@
+/*
+ *  linux/arch/arm/lib/memcpy.S
+ *
+ *  Author:	Nicolas Pitre
+ *  Created:	Sep 28, 2005
+ *  Copyright:	MontaVista Software, Inc.
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include <asm/unwind.h>
+
+#define LDR1W_SHIFT	0
+#define STR1W_SHIFT	0
+
+	.macro ldr1w ptr reg abort
+	W(ldr) \reg, [\ptr], #4
+	.endm
+
+	.macro ldr4w ptr reg1 reg2 reg3 reg4 abort
+	ldmia \ptr!, {\reg1, \reg2, \reg3, \reg4}
+	.endm
+
+	.macro ldr8w ptr reg1 reg2 reg3 reg4 reg5 reg6 reg7 reg8 abort
+	ldmia \ptr!, {\reg1, \reg2, \reg3, \reg4, \reg5, \reg6, \reg7, \reg8}
+	.endm
+
+	.macro ldr1b ptr reg cond=al abort
+	ldr\cond\()b \reg, [\ptr], #1
+	.endm
+
+	.macro str1w ptr reg abort
+	W(str) \reg, [\ptr], #4
+	.endm
+
+	.macro str8w ptr reg1 reg2 reg3 reg4 reg5 reg6 reg7 reg8 abort
+	stmia \ptr!, {\reg1, \reg2, \reg3, \reg4, \reg5, \reg6, \reg7, \reg8}
+	.endm
+
+	.macro str1b ptr reg cond=al abort
+	str\cond\()b \reg, [\ptr], #1
+	.endm
+
+	.macro enter reg1 reg2
+	stmdb sp!, {r0, \reg1, \reg2}
+	.endm
+
+	.macro usave reg1 reg2
+	UNWIND(	.save	{r0, \reg1, \reg2}	)
+	.endm
+
+	.macro exit reg1 reg2
+	ldmfd sp!, {r0, \reg1, \reg2}
+	.endm
+
+	.text
+
+/* Prototype: void *memcpy(void *dest, const void *src, size_t n); */
+
+ENTRY(mmiocpy)
+ENTRY(memcpy)
+
+#include "copy_template.S"
+
+ENDPROC(memcpy)
+ENDPROC(mmiocpy)

arch/arm/lib/memmove.S

@@ -0,0 +1,227 @@
+/*
+ *  linux/arch/arm/lib/memmove.S
+ *
+ *  Author:	Nicolas Pitre
+ *  Created:	Sep 28, 2005
+ *  Copyright:	(C) MontaVista Software Inc.
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include <asm/unwind.h>
+
+		.text
+
+/*
+ * Prototype: void *memmove(void *dest, const void *src, size_t n);
+ *
+ * Note:
+ *
+ * If the memory regions don't overlap, we simply branch to memcpy which is
+ * normally a bit faster. Otherwise the copy is done going downwards.  This
+ * is a transposition of the code from copy_template.S but with the copy
+ * occurring in the opposite direction.
+ */
+
+ENTRY(memmove)
+	UNWIND(	.fnstart			)
+
+		subs	ip, r0, r1
+		cmphi	r2, ip
+		bls	memcpy
+
+		stmfd	sp!, {r0, r4, lr}
+	UNWIND(	.fnend				)
+
+	UNWIND(	.fnstart			)
+	UNWIND(	.save	{r0, r4, lr}		) @ in first stmfd block
+		add	r1, r1, r2
+		add	r0, r0, r2
+		subs	r2, r2, #4
+		blt	8f
+		ands	ip, r0, #3
+	PLD(	pld	[r1, #-4]		)
+		bne	9f
+		ands	ip, r1, #3
+		bne	10f
+
+1:		subs	r2, r2, #(28)
+		stmfd	sp!, {r5 - r8}
+	UNWIND(	.fnend				)
+
+	UNWIND(	.fnstart			)
+	UNWIND(	.save	{r0, r4, lr}		)
+	UNWIND(	.save	{r5 - r8}		) @ in second stmfd block
+		blt	5f
+
+	CALGN(	ands	ip, r0, #31		)
+	CALGN(	sbcnes	r4, ip, r2		)  @ C is always set here
+	CALGN(	bcs	2f			)
+	CALGN(	adr	r4, 6f			)
+	CALGN(	subs	r2, r2, ip		)  @ C is set here
+	CALGN(	rsb	ip, ip, #32		)
+	CALGN(	add	pc, r4, ip		)
+
+	PLD(	pld	[r1, #-4]		)
+2:	PLD(	subs	r2, r2, #96		)
+	PLD(	pld	[r1, #-32]		)
+	PLD(	blt	4f			)
+	PLD(	pld	[r1, #-64]		)
+	PLD(	pld	[r1, #-96]		)
+
+3:	PLD(	pld	[r1, #-128]		)
+4:		ldmdb	r1!, {r3, r4, r5, r6, r7, r8, ip, lr}
+		subs	r2, r2, #32
+		stmdb	r0!, {r3, r4, r5, r6, r7, r8, ip, lr}
+		bge	3b
+	PLD(	cmn	r2, #96			)
+	PLD(	bge	4b			)
+
+5:		ands	ip, r2, #28
+		rsb	ip, ip, #32
+		addne	pc, pc, ip		@ C is always clear here
+		b	7f
+6:		W(nop)
+		W(ldr)	r3, [r1, #-4]!
+		W(ldr)	r4, [r1, #-4]!
+		W(ldr)	r5, [r1, #-4]!
+		W(ldr)	r6, [r1, #-4]!
+		W(ldr)	r7, [r1, #-4]!
+		W(ldr)	r8, [r1, #-4]!
+		W(ldr)	lr, [r1, #-4]!
+
+		add	pc, pc, ip
+		nop
+		W(nop)
+		W(str)	r3, [r0, #-4]!
+		W(str)	r4, [r0, #-4]!
+		W(str)	r5, [r0, #-4]!
+		W(str)	r6, [r0, #-4]!
+		W(str)	r7, [r0, #-4]!
+		W(str)	r8, [r0, #-4]!
+		W(str)	lr, [r0, #-4]!
+
+	CALGN(	bcs	2b			)
+
+7:		ldmfd	sp!, {r5 - r8}
+	UNWIND(	.fnend				) @ end of second stmfd block
+
+	UNWIND(	.fnstart			)
+	UNWIND(	.save	{r0, r4, lr}		) @ still in first stmfd block
+
+8:		movs	r2, r2, lsl #31
+		ldrneb	r3, [r1, #-1]!
+		ldrcsb	r4, [r1, #-1]!
+		ldrcsb	ip, [r1, #-1]
+		strneb	r3, [r0, #-1]!
+		strcsb	r4, [r0, #-1]!
+		strcsb	ip, [r0, #-1]
+		ldmfd	sp!, {r0, r4, pc}
+
+9:		cmp	ip, #2
+		ldrgtb	r3, [r1, #-1]!
+		ldrgeb	r4, [r1, #-1]!
+		ldrb	lr, [r1, #-1]!
+		strgtb	r3, [r0, #-1]!
+		strgeb	r4, [r0, #-1]!
+		subs	r2, r2, ip
+		strb	lr, [r0, #-1]!
+		blt	8b
+		ands	ip, r1, #3
+		beq	1b
+
+10:		bic	r1, r1, #3
+		cmp	ip, #2
+		ldr	r3, [r1, #0]
+		beq	17f
+		blt	18f
+	UNWIND(	.fnend				)
+
+
+		.macro	backward_copy_shift push pull
+
+	UNWIND(	.fnstart			)
+	UNWIND(	.save	{r0, r4, lr}		) @ still in first stmfd block
+		subs	r2, r2, #28
+		blt	14f
+
+	CALGN(	ands	ip, r0, #31		)
+	CALGN(	sbcnes	r4, ip, r2		)  @ C is always set here
+	CALGN(	subcc	r2, r2, ip		)
+	CALGN(	bcc	15f			)
+
+11:		stmfd	sp!, {r5 - r9}
+	UNWIND(	.fnend				)
+
+	UNWIND(	.fnstart			)
+	UNWIND(	.save	{r0, r4, lr}		)
+	UNWIND(	.save	{r5 - r9}		) @ in new second stmfd block
+
+	PLD(	pld	[r1, #-4]		)
+	PLD(	subs	r2, r2, #96		)
+	PLD(	pld	[r1, #-32]		)
+	PLD(	blt	13f			)
+	PLD(	pld	[r1, #-64]		)
+	PLD(	pld	[r1, #-96]		)
+
+12:	PLD(	pld	[r1, #-128]		)
+13:		ldmdb   r1!, {r7, r8, r9, ip}
+		mov     lr, r3, lspush #\push
+		subs    r2, r2, #32
+		ldmdb   r1!, {r3, r4, r5, r6}
+		orr     lr, lr, ip, lspull #\pull
+		mov     ip, ip, lspush #\push
+		orr     ip, ip, r9, lspull #\pull
+		mov     r9, r9, lspush #\push
+		orr     r9, r9, r8, lspull #\pull
+		mov     r8, r8, lspush #\push
+		orr     r8, r8, r7, lspull #\pull
+		mov     r7, r7, lspush #\push
+		orr     r7, r7, r6, lspull #\pull
+		mov     r6, r6, lspush #\push
+		orr     r6, r6, r5, lspull #\pull
+		mov     r5, r5, lspush #\push
+		orr     r5, r5, r4, lspull #\pull
+		mov     r4, r4, lspush #\push
+		orr     r4, r4, r3, lspull #\pull
+		stmdb   r0!, {r4 - r9, ip, lr}
+		bge	12b
+	PLD(	cmn	r2, #96			)
+	PLD(	bge	13b			)
+
+		ldmfd	sp!, {r5 - r9}
+	UNWIND(	.fnend				) @ end of the second stmfd block
+
+	UNWIND(	.fnstart			)
+	UNWIND(	.save {r0, r4, lr}		) @ still in first stmfd block
+
+14:		ands	ip, r2, #28
+		beq	16f
+
+15:		mov     lr, r3, lspush #\push
+		ldr	r3, [r1, #-4]!
+		subs	ip, ip, #4
+		orr	lr, lr, r3, lspull #\pull
+		str	lr, [r0, #-4]!
+		bgt	15b
+	CALGN(	cmp	r2, #0			)
+	CALGN(	bge	11b			)
+
+16:		add	r1, r1, #(\pull / 8)
+		b	8b
+	UNWIND(	.fnend				)
+
+		.endm
+
+
+		backward_copy_shift	push=8	pull=24
+
+17:		backward_copy_shift	push=16	pull=16
+
+18:		backward_copy_shift	push=24	pull=8
+
+ENDPROC(memmove)

arch/arm/lib/memset.S

@@ -0,0 +1,137 @@
+/*
+ *  linux/arch/arm/lib/memset.S
+ *
+ *  Copyright (C) 1995-2000 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ *  ASM optimised string functions
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include <asm/unwind.h>
+
+	.text
+	.align	5
+
+ENTRY(mmioset)
+ENTRY(memset)
+UNWIND( .fnstart         )
+	ands	r3, r0, #3		@ 1 unaligned?
+	mov	ip, r0			@ preserve r0 as return value
+	bne	6f			@ 1
+/*
+ * we know that the pointer in ip is aligned to a word boundary.
+ */
+1:	orr	r1, r1, r1, lsl #8
+	orr	r1, r1, r1, lsl #16
+	mov	r3, r1
+	cmp	r2, #16
+	blt	4f
+
+#if ! CALGN(1)+0
+
+/*
+ * We need 2 extra registers for this loop - use r8 and the LR
+ */
+	stmfd	sp!, {r8, lr}
+UNWIND( .fnend              )
+UNWIND( .fnstart            )
+UNWIND( .save {r8, lr}      )
+	mov	r8, r1
+	mov	lr, r1
+
+2:	subs	r2, r2, #64
+	stmgeia	ip!, {r1, r3, r8, lr}	@ 64 bytes at a time.
+	stmgeia	ip!, {r1, r3, r8, lr}
+	stmgeia	ip!, {r1, r3, r8, lr}
+	stmgeia	ip!, {r1, r3, r8, lr}
+	bgt	2b
+	ldmeqfd	sp!, {r8, pc}		@ Now <64 bytes to go.
+/*
+ * No need to correct the count; we're only testing bits from now on
+ */
+	tst	r2, #32
+	stmneia	ip!, {r1, r3, r8, lr}
+	stmneia	ip!, {r1, r3, r8, lr}
+	tst	r2, #16
+	stmneia	ip!, {r1, r3, r8, lr}
+	ldmfd	sp!, {r8, lr}
+UNWIND( .fnend              )
+
+#else
+
+/*
+ * This version aligns the destination pointer in order to write
+ * whole cache lines at once.
+ */
+
+	stmfd	sp!, {r4-r8, lr}
+UNWIND( .fnend                 )
+UNWIND( .fnstart               )
+UNWIND( .save {r4-r8, lr}      )
+	mov	r4, r1
+	mov	r5, r1
+	mov	r6, r1
+	mov	r7, r1
+	mov	r8, r1
+	mov	lr, r1
+
+	cmp	r2, #96
+	tstgt	ip, #31
+	ble	3f
+
+	and	r8, ip, #31
+	rsb	r8, r8, #32
+	sub	r2, r2, r8
+	movs	r8, r8, lsl #(32 - 4)
+	stmcsia	ip!, {r4, r5, r6, r7}
+	stmmiia	ip!, {r4, r5}
+	tst	r8, #(1 << 30)
+	mov	r8, r1
+	strne	r1, [ip], #4
+
+3:	subs	r2, r2, #64
+	stmgeia	ip!, {r1, r3-r8, lr}
+	stmgeia	ip!, {r1, r3-r8, lr}
+	bgt	3b
+	ldmeqfd	sp!, {r4-r8, pc}
+
+	tst	r2, #32
+	stmneia	ip!, {r1, r3-r8, lr}
+	tst	r2, #16
+	stmneia	ip!, {r4-r7}
+	ldmfd	sp!, {r4-r8, lr}
+UNWIND( .fnend                 )
+
+#endif
+
+UNWIND( .fnstart            )
+4:	tst	r2, #8
+	stmneia	ip!, {r1, r3}
+	tst	r2, #4
+	strne	r1, [ip], #4
+/*
+ * When we get here, we've got less than 4 bytes to zero.  We
+ * may have an unaligned pointer as well.
+ */
+5:	tst	r2, #2
+	strneb	r1, [ip], #1
+	strneb	r1, [ip], #1
+	tst	r2, #1
+	strneb	r1, [ip], #1
+	ret	lr
+
+6:	subs	r2, r2, #4		@ 1 do we have enough
+	blt	5b			@ 1 bytes to align with?
+	cmp	r3, #2			@ 1
+	strltb	r1, [ip], #1		@ 1
+	strleb	r1, [ip], #1		@ 1
+	strb	r1, [ip], #1		@ 1
+	add	r2, r2, r3		@ 1 (r2 = r2 - (4 - r3))
+	b	1b
+UNWIND( .fnend   )
+ENDPROC(memset)
+ENDPROC(mmioset)

arch/arm/lib/memzero.S

@@ -0,0 +1,137 @@
+/*
+ *  linux/arch/arm/lib/memzero.S
+ *
+ *  Copyright (C) 1995-2000 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include <asm/unwind.h>
+
+	.text
+	.align	5
+	.word	0
+/*
+ * Align the pointer in r0.  r3 contains the number of bytes that we are
+ * mis-aligned by, and r1 is the number of bytes.  If r1 < 4, then we
+ * don't bother; we use byte stores instead.
+ */
+UNWIND(	.fnstart			)
+1:	subs	r1, r1, #4		@ 1 do we have enough
+	blt	5f			@ 1 bytes to align with?
+	cmp	r3, #2			@ 1
+	strltb	r2, [r0], #1		@ 1
+	strleb	r2, [r0], #1		@ 1
+	strb	r2, [r0], #1		@ 1
+	add	r1, r1, r3		@ 1 (r1 = r1 - (4 - r3))
+/*
+ * The pointer is now aligned and the length is adjusted.  Try doing the
+ * memzero again.
+ */
+
+ENTRY(__memzero)
+	mov	r2, #0			@ 1
+	ands	r3, r0, #3		@ 1 unaligned?
+	bne	1b			@ 1
+/*
+ * r3 = 0, and we know that the pointer in r0 is aligned to a word boundary.
+ */
+	cmp	r1, #16			@ 1 we can skip this chunk if we
+	blt	4f			@ 1 have < 16 bytes
+
+#if ! CALGN(1)+0
+
+/*
+ * We need an extra register for this loop - save the return address and
+ * use the LR
+ */
+	str	lr, [sp, #-4]!		@ 1
+UNWIND(	.fnend				)
+UNWIND(	.fnstart			)
+UNWIND(	.save 	{lr}			)
+	mov	ip, r2			@ 1
+	mov	lr, r2			@ 1
+
+3:	subs	r1, r1, #64		@ 1 write 32 bytes out per loop
+	stmgeia	r0!, {r2, r3, ip, lr}	@ 4
+	stmgeia	r0!, {r2, r3, ip, lr}	@ 4
+	stmgeia	r0!, {r2, r3, ip, lr}	@ 4
+	stmgeia	r0!, {r2, r3, ip, lr}	@ 4
+	bgt	3b			@ 1
+	ldmeqfd	sp!, {pc}		@ 1/2 quick exit
+/*
+ * No need to correct the count; we're only testing bits from now on
+ */
+	tst	r1, #32			@ 1
+	stmneia	r0!, {r2, r3, ip, lr}	@ 4
+	stmneia	r0!, {r2, r3, ip, lr}	@ 4
+	tst	r1, #16			@ 1 16 bytes or more?
+	stmneia	r0!, {r2, r3, ip, lr}	@ 4
+	ldr	lr, [sp], #4		@ 1
+UNWIND(	.fnend				)
+
+#else
+
+/*
+ * This version aligns the destination pointer in order to write
+ * whole cache lines at once.
+ */
+
+	stmfd	sp!, {r4-r7, lr}
+UNWIND(	.fnend		       )
+UNWIND(	.fnstart	       )
+UNWIND(	.save 	{r4-r7, lr}    )
+	mov	r4, r2
+	mov	r5, r2
+	mov	r6, r2
+	mov	r7, r2
+	mov	ip, r2
+	mov	lr, r2
+
+	cmp	r1, #96
+	andgts	ip, r0, #31
+	ble	3f
+
+	rsb	ip, ip, #32
+	sub	r1, r1, ip
+	movs	ip, ip, lsl #(32 - 4)
+	stmcsia	r0!, {r4, r5, r6, r7}
+	stmmiia	r0!, {r4, r5}
+	movs	ip, ip, lsl #2
+	strcs	r2, [r0], #4
+
+3:	subs	r1, r1, #64
+	stmgeia	r0!, {r2-r7, ip, lr}
+	stmgeia	r0!, {r2-r7, ip, lr}
+	bgt	3b
+	ldmeqfd	sp!, {r4-r7, pc}
+
+	tst	r1, #32
+	stmneia	r0!, {r2-r7, ip, lr}
+	tst	r1, #16
+	stmneia	r0!, {r4-r7}
+	ldmfd	sp!, {r4-r7, lr}
+UNWIND(	.fnend		       )
+
+#endif
+
+UNWIND(	.fnstart			)
+4:	tst	r1, #8			@ 1 8 bytes or more?
+	stmneia	r0!, {r2, r3}		@ 2
+	tst	r1, #4			@ 1 4 bytes or more?
+	strne	r2, [r0], #4		@ 1
+/*
+ * When we get here, we've got less than 4 bytes to zero.  We
+ * may have an unaligned pointer as well.
+ */
+5:	tst	r1, #2			@ 1 2 bytes or more?
+	strneb	r2, [r0], #1		@ 1
+	strneb	r2, [r0], #1		@ 1
+	tst	r1, #1			@ 1 a byte left over
+	strneb	r2, [r0], #1		@ 1
+	ret	lr			@ 1
+UNWIND(	.fnend				)
+ENDPROC(__memzero)

arch/arm/lib/muldi3.S

@@ -0,0 +1,48 @@
+/*
+ *  linux/arch/arm/lib/muldi3.S
+ *
+ *  Author:     Nicolas Pitre
+ *  Created:    Oct 19, 2005
+ *  Copyright:  Monta Vista Software, Inc.
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+#ifdef __ARMEB__
+#define xh r0
+#define xl r1
+#define yh r2
+#define yl r3
+#else
+#define xl r0
+#define xh r1
+#define yl r2
+#define yh r3
+#endif
+
+ENTRY(__muldi3)
+ENTRY(__aeabi_lmul)
+
+	mul	xh, yl, xh
+	mla	xh, xl, yh, xh
+	mov	ip, xl, lsr #16
+	mov	yh, yl, lsr #16
+	bic	xl, xl, ip, lsl #16
+	bic	yl, yl, yh, lsl #16
+	mla	xh, yh, ip, xh
+	mul	yh, xl, yh
+	mul	xl, yl, xl
+	mul	ip, yl, ip
+	adds	xl, xl, yh, lsl #16
+	adc	xh, xh, yh, lsr #16
+	adds	xl, xl, ip, lsl #16
+	adc	xh, xh, ip, lsr #16
+	ret	lr
+
+ENDPROC(__muldi3)
+ENDPROC(__aeabi_lmul)

arch/arm/lib/putuser.S

@@ -0,0 +1,98 @@
+/*
+ *  linux/arch/arm/lib/putuser.S
+ *
+ *  Copyright (C) 2001 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ *  Idea from x86 version, (C) Copyright 1998 Linus Torvalds
+ *
+ * These functions have a non-standard call interface to make
+ * them more efficient, especially as they return an error
+ * value in addition to the "real" return value.
+ *
+ * __put_user_X
+ *
+ * Inputs:	r0 contains the address
+ *		r1 contains the address limit, which must be preserved
+ *		r2, r3 contains the value
+ * Outputs:	r0 is the error code
+ *		lr corrupted
+ *
+ * No other registers must be altered.  (see <asm/uaccess.h>
+ * for specific ASM register usage).
+ *
+ * Note that ADDR_LIMIT is either 0 or 0xc0000000
+ * Note also that it is intended that __put_user_bad is not global.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include <asm/errno.h>
+#include <asm/domain.h>
+
+ENTRY(__put_user_1)
+	check_uaccess r0, 1, r1, ip, __put_user_bad
+1: TUSER(strb)	r2, [r0]
+	mov	r0, #0
+	ret	lr
+ENDPROC(__put_user_1)
+
+ENTRY(__put_user_2)
+	check_uaccess r0, 2, r1, ip, __put_user_bad
+	mov	ip, r2, lsr #8
+#ifdef CONFIG_THUMB2_KERNEL
+#ifndef __ARMEB__
+2: TUSER(strb)	r2, [r0]
+3: TUSER(strb)	ip, [r0, #1]
+#else
+2: TUSER(strb)	ip, [r0]
+3: TUSER(strb)	r2, [r0, #1]
+#endif
+#else	/* !CONFIG_THUMB2_KERNEL */
+#ifndef __ARMEB__
+2: TUSER(strb)	r2, [r0], #1
+3: TUSER(strb)	ip, [r0]
+#else
+2: TUSER(strb)	ip, [r0], #1
+3: TUSER(strb)	r2, [r0]
+#endif
+#endif	/* CONFIG_THUMB2_KERNEL */
+	mov	r0, #0
+	ret	lr
+ENDPROC(__put_user_2)
+
+ENTRY(__put_user_4)
+	check_uaccess r0, 4, r1, ip, __put_user_bad
+4: TUSER(str)	r2, [r0]
+	mov	r0, #0
+	ret	lr
+ENDPROC(__put_user_4)
+
+ENTRY(__put_user_8)
+	check_uaccess r0, 8, r1, ip, __put_user_bad
+#ifdef CONFIG_THUMB2_KERNEL
+5: TUSER(str)	r2, [r0]
+6: TUSER(str)	r3, [r0, #4]
+#else
+5: TUSER(str)	r2, [r0], #4
+6: TUSER(str)	r3, [r0]
+#endif
+	mov	r0, #0
+	ret	lr
+ENDPROC(__put_user_8)
+
+__put_user_bad:
+	mov	r0, #-EFAULT
+	ret	lr
+ENDPROC(__put_user_bad)
+
+.pushsection __ex_table, "a"
+	.long	1b, __put_user_bad
+	.long	2b, __put_user_bad
+	.long	3b, __put_user_bad
+	.long	4b, __put_user_bad
+	.long	5b, __put_user_bad
+	.long	6b, __put_user_bad
+.popsection

arch/arm/lib/setbit.S

@@ -0,0 +1,15 @@
+/*
+ *  linux/arch/arm/lib/setbit.S
+ *
+ *  Copyright (C) 1995-1996 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include "bitops.h"
+		.text
+
+bitop	_set_bit, orr

arch/arm/lib/strchr.S

@@ -0,0 +1,27 @@
+/*
+ *  linux/arch/arm/lib/strchr.S
+ *
+ *  Copyright (C) 1995-2000 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ *  ASM optimised string functions
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+		.text
+		.align	5
+ENTRY(strchr)
+		and	r1, r1, #0xff
+1:		ldrb	r2, [r0], #1
+		teq	r2, r1
+		teqne	r2, #0
+		bne	1b
+		teq	r2, r1
+		movne	r0, #0
+		subeq	r0, r0, #1
+		ret	lr
+ENDPROC(strchr)

arch/arm/lib/strrchr.S

@@ -0,0 +1,26 @@
+/*
+ *  linux/arch/arm/lib/strrchr.S
+ *
+ *  Copyright (C) 1995-2000 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ *  ASM optimised string functions
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+		.text
+		.align	5
+ENTRY(strrchr)
+		mov	r3, #0
+1:		ldrb	r2, [r0], #1
+		teq	r2, r1
+		subeq	r3, r0, #1
+		teq	r2, #0
+		bne	1b
+		mov	r0, r3
+		ret	lr
+ENDPROC(strrchr)

arch/arm/lib/testchangebit.S

@@ -0,0 +1,15 @@
+/*
+ *  linux/arch/arm/lib/testchangebit.S
+ *
+ *  Copyright (C) 1995-1996 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include "bitops.h"
+                .text
+
+testop	_test_and_change_bit, eor, str

arch/arm/lib/testclearbit.S

@@ -0,0 +1,15 @@
+/*
+ *  linux/arch/arm/lib/testclearbit.S
+ *
+ *  Copyright (C) 1995-1996 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include "bitops.h"
+                .text
+
+testop	_test_and_clear_bit, bicne, strne

arch/arm/lib/testsetbit.S

@@ -0,0 +1,15 @@
+/*
+ *  linux/arch/arm/lib/testsetbit.S
+ *
+ *  Copyright (C) 1995-1996 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include "bitops.h"
+                .text
+
+testop	_test_and_set_bit, orreq, streq

arch/arm/lib/uaccess_with_memcpy.c

@@ -0,0 +1,286 @@
+/*
+ *  linux/arch/arm/lib/uaccess_with_memcpy.c
+ *
+ *  Written by: Lennert Buytenhek and Nicolas Pitre
+ *  Copyright (C) 2009 Marvell Semiconductor
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/ctype.h>
+#include <linux/uaccess.h>
+#include <linux/rwsem.h>
+#include <linux/mm.h>
+#include <linux/sched.h>
+#include <linux/hardirq.h> /* for in_atomic() */
+#include <linux/gfp.h>
+#include <linux/highmem.h>
+#include <linux/hugetlb.h>
+#include <asm/current.h>
+#include <asm/page.h>
+
+static int
+pin_page_for_write(const void __user *_addr, pte_t **ptep, spinlock_t **ptlp)
+{
+	unsigned long addr = (unsigned long)_addr;
+	pgd_t *pgd;
+	pmd_t *pmd;
+	pte_t *pte;
+	pud_t *pud;
+	spinlock_t *ptl;
+
+	pgd = pgd_offset(current->mm, addr);
+	if (unlikely(pgd_none(*pgd) || pgd_bad(*pgd)))
+		return 0;
+
+	pud = pud_offset(pgd, addr);
+	if (unlikely(pud_none(*pud) || pud_bad(*pud)))
+		return 0;
+
+	pmd = pmd_offset(pud, addr);
+	if (unlikely(pmd_none(*pmd)))
+		return 0;
+
+	/*
+	 * A pmd can be bad if it refers to a HugeTLB or THP page.
+	 *
+	 * Both THP and HugeTLB pages have the same pmd layout
+	 * and should not be manipulated by the pte functions.
+	 *
+	 * Lock the page table for the destination and check
+	 * to see that it's still huge and whether or not we will
+	 * need to fault on write.
+	 */
+	if (unlikely(pmd_thp_or_huge(*pmd))) {
+		ptl = &current->mm->page_table_lock;
+		spin_lock(ptl);
+		if (unlikely(!pmd_thp_or_huge(*pmd)
+			|| pmd_hugewillfault(*pmd))) {
+			spin_unlock(ptl);
+			return 0;
+		}
+
+		*ptep = NULL;
+		*ptlp = ptl;
+		return 1;
+	}
+
+	if (unlikely(pmd_bad(*pmd)))
+		return 0;
+
+	pte = pte_offset_map_lock(current->mm, pmd, addr, &ptl);
+	if (unlikely(!pte_present(*pte) || !pte_young(*pte) ||
+	    !pte_write(*pte) || !pte_dirty(*pte))) {
+		pte_unmap_unlock(pte, ptl);
+		return 0;
+	}
+
+	*ptep = pte;
+	*ptlp = ptl;
+
+	return 1;
+}
+
+static unsigned long noinline
+__copy_to_user_memcpy(void __user *to, const void *from, unsigned long n)
+{
+	unsigned long ua_flags;
+	int atomic;
+
+	if (unlikely(segment_eq(get_fs(), KERNEL_DS))) {
+		memcpy((void *)to, from, n);
+		return 0;
+	}
+
+	/* the mmap semaphore is taken only if not in an atomic context */
+	atomic = faulthandler_disabled();
+
+	if (!atomic)
+		down_read(&current->mm->mmap_sem);
+	while (n) {
+		pte_t *pte;
+		spinlock_t *ptl;
+		int tocopy;
+
+		while (!pin_page_for_write(to, &pte, &ptl)) {
+			if (!atomic)
+				up_read(&current->mm->mmap_sem);
+			if (__put_user(0, (char __user *)to))
+				goto out;
+			if (!atomic)
+				down_read(&current->mm->mmap_sem);
+		}
+
+		tocopy = (~(unsigned long)to & ~PAGE_MASK) + 1;
+		if (tocopy > n)
+			tocopy = n;
+
+		ua_flags = uaccess_save_and_enable();
+		memcpy((void *)to, from, tocopy);
+		uaccess_restore(ua_flags);
+		to += tocopy;
+		from += tocopy;
+		n -= tocopy;
+
+		if (pte)
+			pte_unmap_unlock(pte, ptl);
+		else
+			spin_unlock(ptl);
+	}
+	if (!atomic)
+		up_read(&current->mm->mmap_sem);
+
+out:
+	return n;
+}
+
+unsigned long
+arm_copy_to_user(void __user *to, const void *from, unsigned long n)
+{
+	/*
+	 * This test is stubbed out of the main function above to keep
+	 * the overhead for small copies low by avoiding a large
+	 * register dump on the stack just to reload them right away.
+	 * With frame pointer disabled, tail call optimization kicks in
+	 * as well making this test almost invisible.
+	 */
+	if (n < 64) {
+		unsigned long ua_flags = uaccess_save_and_enable();
+		n = __copy_to_user_std(to, from, n);
+		uaccess_restore(ua_flags);
+	} else {
+		n = __copy_to_user_memcpy(to, from, n);
+	}
+	return n;
+}
+	
+static unsigned long noinline
+__clear_user_memset(void __user *addr, unsigned long n)
+{
+	unsigned long ua_flags;
+
+	if (unlikely(segment_eq(get_fs(), KERNEL_DS))) {
+		memset((void *)addr, 0, n);
+		return 0;
+	}
+
+	down_read(&current->mm->mmap_sem);
+	while (n) {
+		pte_t *pte;
+		spinlock_t *ptl;
+		int tocopy;
+
+		while (!pin_page_for_write(addr, &pte, &ptl)) {
+			up_read(&current->mm->mmap_sem);
+			if (__put_user(0, (char __user *)addr))
+				goto out;
+			down_read(&current->mm->mmap_sem);
+		}
+
+		tocopy = (~(unsigned long)addr & ~PAGE_MASK) + 1;
+		if (tocopy > n)
+			tocopy = n;
+
+		ua_flags = uaccess_save_and_enable();
+		memset((void *)addr, 0, tocopy);
+		uaccess_restore(ua_flags);
+		addr += tocopy;
+		n -= tocopy;
+
+		if (pte)
+			pte_unmap_unlock(pte, ptl);
+		else
+			spin_unlock(ptl);
+	}
+	up_read(&current->mm->mmap_sem);
+
+out:
+	return n;
+}
+
+unsigned long arm_clear_user(void __user *addr, unsigned long n)
+{
+	/* See rational for this in __copy_to_user() above. */
+	if (n < 64) {
+		unsigned long ua_flags = uaccess_save_and_enable();
+		n = __clear_user_std(addr, n);
+		uaccess_restore(ua_flags);
+	} else {
+		n = __clear_user_memset(addr, n);
+	}
+	return n;
+}
+
+#if 0
+
+/*
+ * This code is disabled by default, but kept around in case the chosen
+ * thresholds need to be revalidated.  Some overhead (small but still)
+ * would be implied by a runtime determined variable threshold, and
+ * so far the measurement on concerned targets didn't show a worthwhile
+ * variation.
+ *
+ * Note that a fairly precise sched_clock() implementation is needed
+ * for results to make some sense.
+ */
+
+#include <linux/vmalloc.h>
+
+static int __init test_size_treshold(void)
+{
+	struct page *src_page, *dst_page;
+	void *user_ptr, *kernel_ptr;
+	unsigned long long t0, t1, t2;
+	int size, ret;
+
+	ret = -ENOMEM;
+	src_page = alloc_page(GFP_KERNEL);
+	if (!src_page)
+		goto no_src;
+	dst_page = alloc_page(GFP_KERNEL);
+	if (!dst_page)
+		goto no_dst;
+	kernel_ptr = page_address(src_page);
+	user_ptr = vmap(&dst_page, 1, VM_IOREMAP, __pgprot(__P010));
+	if (!user_ptr)
+		goto no_vmap;
+
+	/* warm up the src page dcache */
+	ret = __copy_to_user_memcpy(user_ptr, kernel_ptr, PAGE_SIZE);
+
+	for (size = PAGE_SIZE; size >= 4; size /= 2) {
+		t0 = sched_clock();
+		ret |= __copy_to_user_memcpy(user_ptr, kernel_ptr, size);
+		t1 = sched_clock();
+		ret |= __copy_to_user_std(user_ptr, kernel_ptr, size);
+		t2 = sched_clock();
+		printk("copy_to_user: %d %llu %llu\n", size, t1 - t0, t2 - t1);
+	}
+
+	for (size = PAGE_SIZE; size >= 4; size /= 2) {
+		t0 = sched_clock();
+		ret |= __clear_user_memset(user_ptr, size);
+		t1 = sched_clock();
+		ret |= __clear_user_std(user_ptr, size);
+		t2 = sched_clock();
+		printk("clear_user: %d %llu %llu\n", size, t1 - t0, t2 - t1);
+	}
+
+	if (ret)
+		ret = -EFAULT;
+
+	vunmap(user_ptr);
+no_vmap:
+	put_page(dst_page);
+no_dst:
+	put_page(src_page);
+no_src:
+	return ret;
+}
+
+subsys_initcall(test_size_treshold);
+
+#endif

arch/arm/lib/ucmpdi2.S

@@ -0,0 +1,53 @@
+/*
+ *  linux/arch/arm/lib/ucmpdi2.S
+ *
+ *  Author:	Nicolas Pitre
+ *  Created:	Oct 19, 2005
+ *  Copyright:	Monta Vista Software, Inc.
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+#ifdef __ARMEB__
+#define xh r0
+#define xl r1
+#define yh r2
+#define yl r3
+#else
+#define xl r0
+#define xh r1
+#define yl r2
+#define yh r3
+#endif
+
+ENTRY(__ucmpdi2)
+
+	cmp	xh, yh
+	cmpeq	xl, yl
+	movlo	r0, #0
+	moveq	r0, #1
+	movhi	r0, #2
+	ret	lr
+
+ENDPROC(__ucmpdi2)
+
+#ifdef CONFIG_AEABI
+
+ENTRY(__aeabi_ulcmp)
+
+	cmp	xh, yh
+	cmpeq	xl, yl
+	movlo	r0, #-1
+	moveq	r0, #0
+	movhi	r0, #1
+	ret	lr
+
+ENDPROC(__aeabi_ulcmp)
+
+#endif
+

arch/arm/lib/xor-neon.c

@@ -0,0 +1,46 @@
+/*
+ * linux/arch/arm/lib/xor-neon.c
+ *
+ * Copyright (C) 2013 Linaro Ltd <ard.biesheuvel@linaro.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/raid/xor.h>
+#include <linux/module.h>
+
+MODULE_LICENSE("GPL");
+
+#ifndef __ARM_NEON__
+#error You should compile this file with '-mfloat-abi=softfp -mfpu=neon'
+#endif
+
+/*
+ * Pull in the reference implementations while instructing GCC (through
+ * -ftree-vectorize) to attempt to exploit implicit parallelism and emit
+ * NEON instructions.
+ */
+#if __GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 6)
+#pragma GCC optimize "tree-vectorize"
+#else
+/*
+ * While older versions of GCC do not generate incorrect code, they fail to
+ * recognize the parallel nature of these functions, and emit plain ARM code,
+ * which is known to be slower than the optimized ARM code in asm-arm/xor.h.
+ */
+#warning This code requires at least version 4.6 of GCC
+#endif
+
+#pragma GCC diagnostic ignored "-Wunused-variable"
+#include <asm-generic/xor.h>
+
+struct xor_block_template const xor_block_neon_inner = {
+	.name	= "__inner_neon__",
+	.do_2	= xor_8regs_2,
+	.do_3	= xor_8regs_3,
+	.do_4	= xor_8regs_4,
+	.do_5	= xor_8regs_5,
+};
+EXPORT_SYMBOL(xor_block_neon_inner);

arch/arm/mach-at91/Kconfig

@@ -0,0 +1,131 @@
+menuconfig ARCH_AT91
+	bool "Atmel SoCs"
+	depends on ARCH_MULTI_V4T || ARCH_MULTI_V5 || ARCH_MULTI_V7
+	select COMMON_CLK_AT91
+	select GPIOLIB
+	select PINCTRL
+	select SOC_BUS
+
+if ARCH_AT91
+config SOC_SAMA5D2
+	bool "SAMA5D2 family"
+	depends on ARCH_MULTI_V7
+	select SOC_SAMA5
+	select CACHE_L2X0
+	select HAVE_FB_ATMEL
+	select HAVE_AT91_UTMI
+	select HAVE_AT91_USB_CLK
+	select HAVE_AT91_H32MX
+	select HAVE_AT91_GENERATED_CLK
+	select PINCTRL_AT91PIO4
+	help
+	  Select this if ou are using one of Atmel's SAMA5D2 family SoC.
+
+config SOC_SAMA5D3
+	bool "SAMA5D3 family"
+	depends on ARCH_MULTI_V7
+	select SOC_SAMA5
+	select HAVE_FB_ATMEL
+	select HAVE_AT91_UTMI
+	select HAVE_AT91_SMD
+	select HAVE_AT91_USB_CLK
+	select PINCTRL_AT91
+	help
+	  Select this if you are using one of Atmel's SAMA5D3 family SoC.
+	  This support covers SAMA5D31, SAMA5D33, SAMA5D34, SAMA5D35, SAMA5D36.
+
+config SOC_SAMA5D4
+	bool "SAMA5D4 family"
+	depends on ARCH_MULTI_V7
+	select SOC_SAMA5
+	select CACHE_L2X0
+	select HAVE_FB_ATMEL
+	select HAVE_AT91_UTMI
+	select HAVE_AT91_SMD
+	select HAVE_AT91_USB_CLK
+	select HAVE_AT91_H32MX
+	select PINCTRL_AT91
+	help
+	  Select this if you are using one of Atmel's SAMA5D4 family SoC.
+
+config SOC_AT91RM9200
+	bool "AT91RM9200"
+	depends on ARCH_MULTI_V4T
+	select ATMEL_AIC_IRQ
+	select ATMEL_ST
+	select CPU_ARM920T
+	select HAVE_AT91_USB_CLK
+	select PINCTRL_AT91
+	select SOC_SAM_V4_V5
+	select SRAM if PM
+	help
+	  Select this if you are using Atmel's AT91RM9200 SoC.
+
+config SOC_AT91SAM9
+	bool "AT91SAM9"
+	depends on ARCH_MULTI_V5
+	select ATMEL_AIC_IRQ
+	select ATMEL_SDRAMC
+	select CPU_ARM926T
+	select HAVE_AT91_SMD
+	select HAVE_AT91_USB_CLK
+	select HAVE_AT91_UTMI
+	select HAVE_FB_ATMEL
+	select MEMORY
+	select PINCTRL_AT91
+	select SOC_SAM_V4_V5
+	select SRAM if PM
+	help
+	  Select this if you are using one of those Atmel SoC:
+	    AT91SAM9260
+	    AT91SAM9261
+	    AT91SAM9263
+	    AT91SAM9G15
+	    AT91SAM9G20
+	    AT91SAM9G25
+	    AT91SAM9G35
+	    AT91SAM9G45
+	    AT91SAM9G46
+	    AT91SAM9M10
+	    AT91SAM9M11
+	    AT91SAM9N12
+	    AT91SAM9RL
+	    AT91SAM9X25
+	    AT91SAM9X35
+	    AT91SAM9XE
+
+config HAVE_AT91_UTMI
+	bool
+
+config HAVE_AT91_USB_CLK
+	bool
+
+config COMMON_CLK_AT91
+	bool
+	select COMMON_CLK
+	select MFD_SYSCON
+
+config HAVE_AT91_SMD
+	bool
+
+config HAVE_AT91_H32MX
+	bool
+
+config HAVE_AT91_GENERATED_CLK
+	bool
+
+config SOC_SAM_V4_V5
+	bool
+
+config SOC_SAM_V7
+	bool
+
+config SOC_SAMA5
+	bool
+	select ATMEL_AIC5_IRQ
+	select ATMEL_SDRAMC
+	select MEMORY
+	select SOC_SAM_V7
+	select SRAM if PM
+
+endif

arch/arm/mach-at91/Makefile

@@ -0,0 +1,20 @@
+#
+# Makefile for the linux kernel.
+#
+obj-y		:= soc.o
+
+# CPU-specific support
+obj-$(CONFIG_SOC_AT91RM9200)	+= at91rm9200.o
+obj-$(CONFIG_SOC_AT91SAM9)	+= at91sam9.o
+obj-$(CONFIG_SOC_SAMA5)		+= sama5.o
+
+# Power Management
+obj-$(CONFIG_PM)		+= pm.o
+obj-$(CONFIG_PM)		+= pm_suspend.o
+
+ifeq ($(CONFIG_CPU_V7),y)
+AFLAGS_pm_suspend.o := -march=armv7-a
+endif
+ifeq ($(CONFIG_PM_DEBUG),y)
+CFLAGS_pm.o += -DDEBUG
+endif

arch/arm/mach-at91/at91rm9200.c

@@ -0,0 +1,46 @@
+/*
+ *  Setup code for AT91RM9200
+ *
+ *  Copyright (C) 2011 Atmel,
+ *                2011 Nicolas Ferre <nicolas.ferre@atmel.com>
+ *                2012 Joachim Eastwood <manabian@gmail.com>
+ *
+ * Licensed under GPLv2 or later.
+ */
+
+#include <linux/of.h>
+#include <linux/of_platform.h>
+
+#include <asm/mach/arch.h>
+
+#include "generic.h"
+#include "soc.h"
+
+static const struct at91_soc rm9200_socs[] = {
+	AT91_SOC(AT91RM9200_CIDR_MATCH, 0, "at91rm9200 BGA", "at91rm9200"),
+	{ /* sentinel */ },
+};
+
+static void __init at91rm9200_dt_device_init(void)
+{
+	struct soc_device *soc;
+	struct device *soc_dev = NULL;
+
+	soc = at91_soc_init(rm9200_socs);
+	if (soc != NULL)
+		soc_dev = soc_device_to_device(soc);
+
+	of_platform_default_populate(NULL, NULL, soc_dev);
+
+	at91rm9200_pm_init();
+}
+
+static const char *const at91rm9200_dt_board_compat[] __initconst = {
+	"atmel,at91rm9200",
+	NULL
+};
+
+DT_MACHINE_START(at91rm9200_dt, "Atmel AT91RM9200")
+	.init_machine	= at91rm9200_dt_device_init,
+	.dt_compat	= at91rm9200_dt_board_compat,
+MACHINE_END

arch/arm/mach-at91/at91sam9.c

@@ -0,0 +1,117 @@
+/*
+ *  Setup code for AT91SAM9
+ *
+ *  Copyright (C) 2011 Atmel,
+ *                2011 Nicolas Ferre <nicolas.ferre@atmel.com>
+ *
+ * Licensed under GPLv2 or later.
+ */
+
+#include <linux/of.h>
+#include <linux/of_platform.h>
+
+#include <asm/mach/arch.h>
+#include <asm/system_misc.h>
+
+#include "generic.h"
+#include "soc.h"
+
+static const struct at91_soc at91sam9_socs[] = {
+	AT91_SOC(AT91SAM9260_CIDR_MATCH, 0, "at91sam9260", NULL),
+	AT91_SOC(AT91SAM9261_CIDR_MATCH, 0, "at91sam9261", NULL),
+	AT91_SOC(AT91SAM9263_CIDR_MATCH, 0, "at91sam9263", NULL),
+	AT91_SOC(AT91SAM9G20_CIDR_MATCH, 0, "at91sam9g20", NULL),
+	AT91_SOC(AT91SAM9RL64_CIDR_MATCH, 0, "at91sam9rl64", NULL),
+	AT91_SOC(AT91SAM9G45_CIDR_MATCH, AT91SAM9M11_EXID_MATCH,
+		 "at91sam9m11", "at91sam9g45"),
+	AT91_SOC(AT91SAM9G45_CIDR_MATCH, AT91SAM9M10_EXID_MATCH,
+		 "at91sam9m10", "at91sam9g45"),
+	AT91_SOC(AT91SAM9G45_CIDR_MATCH, AT91SAM9G46_EXID_MATCH,
+		 "at91sam9g46", "at91sam9g45"),
+	AT91_SOC(AT91SAM9G45_CIDR_MATCH, AT91SAM9G45_EXID_MATCH,
+		 "at91sam9g45", "at91sam9g45"),
+	AT91_SOC(AT91SAM9X5_CIDR_MATCH, AT91SAM9G15_EXID_MATCH,
+		 "at91sam9g15", "at91sam9x5"),
+	AT91_SOC(AT91SAM9X5_CIDR_MATCH, AT91SAM9G35_EXID_MATCH,
+		 "at91sam9g35", "at91sam9x5"),
+	AT91_SOC(AT91SAM9X5_CIDR_MATCH, AT91SAM9X35_EXID_MATCH,
+		 "at91sam9x35", "at91sam9x5"),
+	AT91_SOC(AT91SAM9X5_CIDR_MATCH, AT91SAM9G25_EXID_MATCH,
+		 "at91sam9g25", "at91sam9x5"),
+	AT91_SOC(AT91SAM9X5_CIDR_MATCH, AT91SAM9X25_EXID_MATCH,
+		 "at91sam9x25", "at91sam9x5"),
+	AT91_SOC(AT91SAM9N12_CIDR_MATCH, AT91SAM9CN12_EXID_MATCH,
+		 "at91sam9cn12", "at91sam9n12"),
+	AT91_SOC(AT91SAM9N12_CIDR_MATCH, AT91SAM9N12_EXID_MATCH,
+		 "at91sam9n12", "at91sam9n12"),
+	AT91_SOC(AT91SAM9N12_CIDR_MATCH, AT91SAM9CN11_EXID_MATCH,
+		 "at91sam9cn11", "at91sam9n12"),
+	AT91_SOC(AT91SAM9XE128_CIDR_MATCH, 0, "at91sam9xe128", "at91sam9xe128"),
+	AT91_SOC(AT91SAM9XE256_CIDR_MATCH, 0, "at91sam9xe256", "at91sam9xe256"),
+	AT91_SOC(AT91SAM9XE512_CIDR_MATCH, 0, "at91sam9xe512", "at91sam9xe512"),
+	{ /* sentinel */ },
+};
+
+static void __init at91sam9_common_init(void)
+{
+	struct soc_device *soc;
+	struct device *soc_dev = NULL;
+
+	soc = at91_soc_init(at91sam9_socs);
+	if (soc != NULL)
+		soc_dev = soc_device_to_device(soc);
+
+	of_platform_default_populate(NULL, NULL, soc_dev);
+}
+
+static void __init at91sam9_dt_device_init(void)
+{
+	at91sam9_common_init();
+	at91sam9260_pm_init();
+}
+
+static const char *const at91_dt_board_compat[] __initconst = {
+	"atmel,at91sam9",
+	NULL
+};
+
+DT_MACHINE_START(at91sam_dt, "Atmel AT91SAM9")
+	/* Maintainer: Atmel */
+	.init_machine	= at91sam9_dt_device_init,
+	.dt_compat	= at91_dt_board_compat,
+MACHINE_END
+
+static void __init at91sam9g45_dt_device_init(void)
+{
+	at91sam9_common_init();
+	at91sam9g45_pm_init();
+}
+
+static const char *const at91sam9g45_board_compat[] __initconst = {
+	"atmel,at91sam9g45",
+	NULL
+};
+
+DT_MACHINE_START(at91sam9g45_dt, "Atmel AT91SAM9G45")
+	/* Maintainer: Atmel */
+	.init_machine	= at91sam9g45_dt_device_init,
+	.dt_compat	= at91sam9g45_board_compat,
+MACHINE_END
+
+static void __init at91sam9x5_dt_device_init(void)
+{
+	at91sam9_common_init();
+	at91sam9x5_pm_init();
+}
+
+static const char *const at91sam9x5_board_compat[] __initconst = {
+	"atmel,at91sam9x5",
+	"atmel,at91sam9n12",
+	NULL
+};
+
+DT_MACHINE_START(at91sam9x5_dt, "Atmel AT91SAM9")
+	/* Maintainer: Atmel */
+	.init_machine	= at91sam9x5_dt_device_init,
+	.dt_compat	= at91sam9x5_board_compat,
+MACHINE_END

arch/arm/mach-at91/generic.h

@@ -0,0 +1,28 @@
+/*
+ * linux/arch/arm/mach-at91/generic.h
+ *
+ *  Copyright (C) 2005 David Brownell
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef _AT91_GENERIC_H
+#define _AT91_GENERIC_H
+
+#ifdef CONFIG_PM
+extern void __init at91rm9200_pm_init(void);
+extern void __init at91sam9260_pm_init(void);
+extern void __init at91sam9g45_pm_init(void);
+extern void __init at91sam9x5_pm_init(void);
+extern void __init sama5_pm_init(void);
+#else
+static inline void __init at91rm9200_pm_init(void) { }
+static inline void __init at91sam9260_pm_init(void) { }
+static inline void __init at91sam9g45_pm_init(void) { }
+static inline void __init at91sam9x5_pm_init(void) { }
+static inline void __init sama5_pm_init(void) { }
+#endif
+
+#endif /* _AT91_GENERIC_H */

arch/arm/mach-at91/pm.c

@@ -0,0 +1,517 @@
+/*
+ * arch/arm/mach-at91/pm.c
+ * AT91 Power Management
+ *
+ * Copyright (C) 2005 David Brownell
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/gpio.h>
+#include <linux/suspend.h>
+#include <linux/sched.h>
+#include <linux/proc_fs.h>
+#include <linux/genalloc.h>
+#include <linux/interrupt.h>
+#include <linux/sysfs.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_platform.h>
+#include <linux/of_address.h>
+#include <linux/platform_device.h>
+#include <linux/platform_data/atmel.h>
+#include <linux/io.h>
+#include <linux/clk/at91_pmc.h>
+
+#include <asm/irq.h>
+#include <linux/atomic.h>
+#include <asm/mach/time.h>
+#include <asm/mach/irq.h>
+#include <asm/fncpy.h>
+#include <asm/cacheflush.h>
+#include <asm/system_misc.h>
+
+#include "generic.h"
+#include "pm.h"
+
+static void __iomem *pmc;
+
+/*
+ * FIXME: this is needed to communicate between the pinctrl driver and
+ * the PM implementation in the machine. Possibly part of the PM
+ * implementation should be moved down into the pinctrl driver and get
+ * called as part of the generic suspend/resume path.
+ */
+#ifdef CONFIG_PINCTRL_AT91
+extern void at91_pinctrl_gpio_suspend(void);
+extern void at91_pinctrl_gpio_resume(void);
+#endif
+
+static struct {
+	unsigned long uhp_udp_mask;
+	int memctrl;
+} at91_pm_data;
+
+void __iomem *at91_ramc_base[2];
+
+static int at91_pm_valid_state(suspend_state_t state)
+{
+	switch (state) {
+		case PM_SUSPEND_ON:
+		case PM_SUSPEND_STANDBY:
+		case PM_SUSPEND_MEM:
+			return 1;
+
+		default:
+			return 0;
+	}
+}
+
+
+static suspend_state_t target_state;
+
+/*
+ * Called after processes are frozen, but before we shutdown devices.
+ */
+static int at91_pm_begin(suspend_state_t state)
+{
+	target_state = state;
+	return 0;
+}
+
+/*
+ * Verify that all the clocks are correct before entering
+ * slow-clock mode.
+ */
+static int at91_pm_verify_clocks(void)
+{
+	unsigned long scsr;
+	int i;
+
+	scsr = readl(pmc + AT91_PMC_SCSR);
+
+	/* USB must not be using PLLB */
+	if ((scsr & at91_pm_data.uhp_udp_mask) != 0) {
+		pr_err("AT91: PM - Suspend-to-RAM with USB still active\n");
+		return 0;
+	}
+
+	/* PCK0..PCK3 must be disabled, or configured to use clk32k */
+	for (i = 0; i < 4; i++) {
+		u32 css;
+
+		if ((scsr & (AT91_PMC_PCK0 << i)) == 0)
+			continue;
+		css = readl(pmc + AT91_PMC_PCKR(i)) & AT91_PMC_CSS;
+		if (css != AT91_PMC_CSS_SLOW) {
+			pr_err("AT91: PM - Suspend-to-RAM with PCK%d src %d\n", i, css);
+			return 0;
+		}
+	}
+
+	return 1;
+}
+
+/*
+ * Call this from platform driver suspend() to see how deeply to suspend.
+ * For example, some controllers (like OHCI) need one of the PLL clocks
+ * in order to act as a wakeup source, and those are not available when
+ * going into slow clock mode.
+ *
+ * REVISIT: generalize as clk_will_be_available(clk)?  Other platforms have
+ * the very same problem (but not using at91 main_clk), and it'd be better
+ * to add one generic API rather than lots of platform-specific ones.
+ */
+int at91_suspend_entering_slow_clock(void)
+{
+	return (target_state == PM_SUSPEND_MEM);
+}
+EXPORT_SYMBOL(at91_suspend_entering_slow_clock);
+
+static void (*at91_suspend_sram_fn)(void __iomem *pmc, void __iomem *ramc0,
+			  void __iomem *ramc1, int memctrl);
+
+extern void at91_pm_suspend_in_sram(void __iomem *pmc, void __iomem *ramc0,
+			    void __iomem *ramc1, int memctrl);
+extern u32 at91_pm_suspend_in_sram_sz;
+
+static void at91_pm_suspend(suspend_state_t state)
+{
+	unsigned int pm_data = at91_pm_data.memctrl;
+
+	pm_data |= (state == PM_SUSPEND_MEM) ?
+				AT91_PM_MODE(AT91_PM_SLOW_CLOCK) : 0;
+
+	flush_cache_all();
+	outer_disable();
+
+	at91_suspend_sram_fn(pmc, at91_ramc_base[0],
+			     at91_ramc_base[1], pm_data);
+
+	outer_resume();
+}
+
+static int at91_pm_enter(suspend_state_t state)
+{
+#ifdef CONFIG_PINCTRL_AT91
+	at91_pinctrl_gpio_suspend();
+#endif
+	switch (state) {
+	/*
+	 * Suspend-to-RAM is like STANDBY plus slow clock mode, so
+	 * drivers must suspend more deeply, the master clock switches
+	 * to the clk32k and turns off the main oscillator
+	 */
+	case PM_SUSPEND_MEM:
+		/*
+		 * Ensure that clocks are in a valid state.
+		 */
+		if (!at91_pm_verify_clocks())
+			goto error;
+
+		at91_pm_suspend(state);
+
+		break;
+
+	/*
+	 * STANDBY mode has *all* drivers suspended; ignores irqs not
+	 * marked as 'wakeup' event sources; and reduces DRAM power.
+	 * But otherwise it's identical to PM_SUSPEND_ON: cpu idle, and
+	 * nothing fancy done with main or cpu clocks.
+	 */
+	case PM_SUSPEND_STANDBY:
+		at91_pm_suspend(state);
+		break;
+
+	case PM_SUSPEND_ON:
+		cpu_do_idle();
+		break;
+
+	default:
+		pr_debug("AT91: PM - bogus suspend state %d\n", state);
+		goto error;
+	}
+
+error:
+	target_state = PM_SUSPEND_ON;
+
+#ifdef CONFIG_PINCTRL_AT91
+	at91_pinctrl_gpio_resume();
+#endif
+	return 0;
+}
+
+/*
+ * Called right prior to thawing processes.
+ */
+static void at91_pm_end(void)
+{
+	target_state = PM_SUSPEND_ON;
+}
+
+
+static const struct platform_suspend_ops at91_pm_ops = {
+	.valid	= at91_pm_valid_state,
+	.begin	= at91_pm_begin,
+	.enter	= at91_pm_enter,
+	.end	= at91_pm_end,
+};
+
+static struct platform_device at91_cpuidle_device = {
+	.name = "cpuidle-at91",
+};
+
+static void at91_pm_set_standby(void (*at91_standby)(void))
+{
+	if (at91_standby)
+		at91_cpuidle_device.dev.platform_data = at91_standby;
+}
+
+/*
+ * The AT91RM9200 goes into self-refresh mode with this command, and will
+ * terminate self-refresh automatically on the next SDRAM access.
+ *
+ * Self-refresh mode is exited as soon as a memory access is made, but we don't
+ * know for sure when that happens. However, we need to restore the low-power
+ * mode if it was enabled before going idle. Restoring low-power mode while
+ * still in self-refresh is "not recommended", but seems to work.
+ */
+static void at91rm9200_standby(void)
+{
+	u32 lpr = at91_ramc_read(0, AT91_MC_SDRAMC_LPR);
+
+	asm volatile(
+		"b    1f\n\t"
+		".align    5\n\t"
+		"1:  mcr    p15, 0, %0, c7, c10, 4\n\t"
+		"    str    %0, [%1, %2]\n\t"
+		"    str    %3, [%1, %4]\n\t"
+		"    mcr    p15, 0, %0, c7, c0, 4\n\t"
+		"    str    %5, [%1, %2]"
+		:
+		: "r" (0), "r" (at91_ramc_base[0]), "r" (AT91_MC_SDRAMC_LPR),
+		  "r" (1), "r" (AT91_MC_SDRAMC_SRR),
+		  "r" (lpr));
+}
+
+/* We manage both DDRAM/SDRAM controllers, we need more than one value to
+ * remember.
+ */
+static void at91_ddr_standby(void)
+{
+	/* Those two values allow us to delay self-refresh activation
+	 * to the maximum. */
+	u32 lpr0, lpr1 = 0;
+	u32 saved_lpr0, saved_lpr1 = 0;
+
+	if (at91_ramc_base[1]) {
+		saved_lpr1 = at91_ramc_read(1, AT91_DDRSDRC_LPR);
+		lpr1 = saved_lpr1 & ~AT91_DDRSDRC_LPCB;
+		lpr1 |= AT91_DDRSDRC_LPCB_SELF_REFRESH;
+	}
+
+	saved_lpr0 = at91_ramc_read(0, AT91_DDRSDRC_LPR);
+	lpr0 = saved_lpr0 & ~AT91_DDRSDRC_LPCB;
+	lpr0 |= AT91_DDRSDRC_LPCB_SELF_REFRESH;
+
+	/* self-refresh mode now */
+	at91_ramc_write(0, AT91_DDRSDRC_LPR, lpr0);
+	if (at91_ramc_base[1])
+		at91_ramc_write(1, AT91_DDRSDRC_LPR, lpr1);
+
+	cpu_do_idle();
+
+	at91_ramc_write(0, AT91_DDRSDRC_LPR, saved_lpr0);
+	if (at91_ramc_base[1])
+		at91_ramc_write(1, AT91_DDRSDRC_LPR, saved_lpr1);
+}
+
+static void sama5d3_ddr_standby(void)
+{
+	u32 lpr0;
+	u32 saved_lpr0;
+
+	saved_lpr0 = at91_ramc_read(0, AT91_DDRSDRC_LPR);
+	lpr0 = saved_lpr0 & ~AT91_DDRSDRC_LPCB;
+	lpr0 |= AT91_DDRSDRC_LPCB_POWER_DOWN;
+
+	at91_ramc_write(0, AT91_DDRSDRC_LPR, lpr0);
+
+	cpu_do_idle();
+
+	at91_ramc_write(0, AT91_DDRSDRC_LPR, saved_lpr0);
+}
+
+/* We manage both DDRAM/SDRAM controllers, we need more than one value to
+ * remember.
+ */
+static void at91sam9_sdram_standby(void)
+{
+	u32 lpr0, lpr1 = 0;
+	u32 saved_lpr0, saved_lpr1 = 0;
+
+	if (at91_ramc_base[1]) {
+		saved_lpr1 = at91_ramc_read(1, AT91_SDRAMC_LPR);
+		lpr1 = saved_lpr1 & ~AT91_SDRAMC_LPCB;
+		lpr1 |= AT91_SDRAMC_LPCB_SELF_REFRESH;
+	}
+
+	saved_lpr0 = at91_ramc_read(0, AT91_SDRAMC_LPR);
+	lpr0 = saved_lpr0 & ~AT91_SDRAMC_LPCB;
+	lpr0 |= AT91_SDRAMC_LPCB_SELF_REFRESH;
+
+	/* self-refresh mode now */
+	at91_ramc_write(0, AT91_SDRAMC_LPR, lpr0);
+	if (at91_ramc_base[1])
+		at91_ramc_write(1, AT91_SDRAMC_LPR, lpr1);
+
+	cpu_do_idle();
+
+	at91_ramc_write(0, AT91_SDRAMC_LPR, saved_lpr0);
+	if (at91_ramc_base[1])
+		at91_ramc_write(1, AT91_SDRAMC_LPR, saved_lpr1);
+}
+
+static const struct of_device_id ramc_ids[] __initconst = {
+	{ .compatible = "atmel,at91rm9200-sdramc", .data = at91rm9200_standby },
+	{ .compatible = "atmel,at91sam9260-sdramc", .data = at91sam9_sdram_standby },
+	{ .compatible = "atmel,at91sam9g45-ddramc", .data = at91_ddr_standby },
+	{ .compatible = "atmel,sama5d3-ddramc", .data = sama5d3_ddr_standby },
+	{ /*sentinel*/ }
+};
+
+static __init void at91_dt_ramc(void)
+{
+	struct device_node *np;
+	const struct of_device_id *of_id;
+	int idx = 0;
+	const void *standby = NULL;
+
+	for_each_matching_node_and_match(np, ramc_ids, &of_id) {
+		at91_ramc_base[idx] = of_iomap(np, 0);
+		if (!at91_ramc_base[idx])
+			panic(pr_fmt("unable to map ramc[%d] cpu registers\n"), idx);
+
+		if (!standby)
+			standby = of_id->data;
+
+		idx++;
+	}
+
+	if (!idx)
+		panic(pr_fmt("unable to find compatible ram controller node in dtb\n"));
+
+	if (!standby) {
+		pr_warn("ramc no standby function available\n");
+		return;
+	}
+
+	at91_pm_set_standby(standby);
+}
+
+static void at91rm9200_idle(void)
+{
+	/*
+	 * Disable the processor clock.  The processor will be automatically
+	 * re-enabled by an interrupt or by a reset.
+	 */
+	writel(AT91_PMC_PCK, pmc + AT91_PMC_SCDR);
+}
+
+static void at91sam9_idle(void)
+{
+	writel(AT91_PMC_PCK, pmc + AT91_PMC_SCDR);
+	cpu_do_idle();
+}
+
+static void __init at91_pm_sram_init(void)
+{
+	struct gen_pool *sram_pool;
+	phys_addr_t sram_pbase;
+	unsigned long sram_base;
+	struct device_node *node;
+	struct platform_device *pdev = NULL;
+
+	for_each_compatible_node(node, NULL, "mmio-sram") {
+		pdev = of_find_device_by_node(node);
+		if (pdev) {
+			of_node_put(node);
+			break;
+		}
+	}
+
+	if (!pdev) {
+		pr_warn("%s: failed to find sram device!\n", __func__);
+		return;
+	}
+
+	sram_pool = gen_pool_get(&pdev->dev, NULL);
+	if (!sram_pool) {
+		pr_warn("%s: sram pool unavailable!\n", __func__);
+		return;
+	}
+
+	sram_base = gen_pool_alloc(sram_pool, at91_pm_suspend_in_sram_sz);
+	if (!sram_base) {
+		pr_warn("%s: unable to alloc sram!\n", __func__);
+		return;
+	}
+
+	sram_pbase = gen_pool_virt_to_phys(sram_pool, sram_base);
+	at91_suspend_sram_fn = __arm_ioremap_exec(sram_pbase,
+					at91_pm_suspend_in_sram_sz, false);
+	if (!at91_suspend_sram_fn) {
+		pr_warn("SRAM: Could not map\n");
+		return;
+	}
+
+	/* Copy the pm suspend handler to SRAM */
+	at91_suspend_sram_fn = fncpy(at91_suspend_sram_fn,
+			&at91_pm_suspend_in_sram, at91_pm_suspend_in_sram_sz);
+}
+
+static const struct of_device_id atmel_pmc_ids[] __initconst = {
+	{ .compatible = "atmel,at91rm9200-pmc"  },
+	{ .compatible = "atmel,at91sam9260-pmc" },
+	{ .compatible = "atmel,at91sam9g45-pmc" },
+	{ .compatible = "atmel,at91sam9n12-pmc" },
+	{ .compatible = "atmel,at91sam9x5-pmc" },
+	{ .compatible = "atmel,sama5d3-pmc" },
+	{ .compatible = "atmel,sama5d2-pmc" },
+	{ /* sentinel */ },
+};
+
+static void __init at91_pm_init(void (*pm_idle)(void))
+{
+	struct device_node *pmc_np;
+
+	if (at91_cpuidle_device.dev.platform_data)
+		platform_device_register(&at91_cpuidle_device);
+
+	pmc_np = of_find_matching_node(NULL, atmel_pmc_ids);
+	pmc = of_iomap(pmc_np, 0);
+	if (!pmc) {
+		pr_err("AT91: PM not supported, PMC not found\n");
+		return;
+	}
+
+	if (pm_idle)
+		arm_pm_idle = pm_idle;
+
+	at91_pm_sram_init();
+
+	if (at91_suspend_sram_fn)
+		suspend_set_ops(&at91_pm_ops);
+	else
+		pr_info("AT91: PM not supported, due to no SRAM allocated\n");
+}
+
+void __init at91rm9200_pm_init(void)
+{
+	at91_dt_ramc();
+
+	/*
+	 * AT91RM9200 SDRAM low-power mode cannot be used with self-refresh.
+	 */
+	at91_ramc_write(0, AT91_MC_SDRAMC_LPR, 0);
+
+	at91_pm_data.uhp_udp_mask = AT91RM9200_PMC_UHP | AT91RM9200_PMC_UDP;
+	at91_pm_data.memctrl = AT91_MEMCTRL_MC;
+
+	at91_pm_init(at91rm9200_idle);
+}
+
+void __init at91sam9260_pm_init(void)
+{
+	at91_dt_ramc();
+	at91_pm_data.memctrl = AT91_MEMCTRL_SDRAMC;
+	at91_pm_data.uhp_udp_mask = AT91SAM926x_PMC_UHP | AT91SAM926x_PMC_UDP;
+	at91_pm_init(at91sam9_idle);
+}
+
+void __init at91sam9g45_pm_init(void)
+{
+	at91_dt_ramc();
+	at91_pm_data.uhp_udp_mask = AT91SAM926x_PMC_UHP;
+	at91_pm_data.memctrl = AT91_MEMCTRL_DDRSDR;
+	at91_pm_init(at91sam9_idle);
+}
+
+void __init at91sam9x5_pm_init(void)
+{
+	at91_dt_ramc();
+	at91_pm_data.uhp_udp_mask = AT91SAM926x_PMC_UHP | AT91SAM926x_PMC_UDP;
+	at91_pm_data.memctrl = AT91_MEMCTRL_DDRSDR;
+	at91_pm_init(at91sam9_idle);
+}
+
+void __init sama5_pm_init(void)
+{
+	at91_dt_ramc();
+	at91_pm_data.uhp_udp_mask = AT91SAM926x_PMC_UHP | AT91SAM926x_PMC_UDP;
+	at91_pm_data.memctrl = AT91_MEMCTRL_DDRSDR;
+	at91_pm_init(NULL);
+}

arch/arm/mach-at91/pm.h

@@ -0,0 +1,42 @@
+/*
+ * AT91 Power Management
+ *
+ * Copyright (C) 2005 David Brownell
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+#ifndef __ARCH_ARM_MACH_AT91_PM
+#define __ARCH_ARM_MACH_AT91_PM
+
+#include <asm/proc-fns.h>
+
+#include <linux/mfd/syscon/atmel-mc.h>
+#include <soc/at91/at91sam9_ddrsdr.h>
+#include <soc/at91/at91sam9_sdramc.h>
+
+#ifndef __ASSEMBLY__
+extern void __iomem *at91_ramc_base[];
+
+#define at91_ramc_read(id, field) \
+	__raw_readl(at91_ramc_base[id] + field)
+
+#define at91_ramc_write(id, field, value) \
+	__raw_writel(value, at91_ramc_base[id] + field)
+#endif
+
+#define AT91_MEMCTRL_MC		0
+#define AT91_MEMCTRL_SDRAMC	1
+#define AT91_MEMCTRL_DDRSDR	2
+
+#define	AT91_PM_MEMTYPE_MASK	0x0f
+
+#define	AT91_PM_MODE_OFFSET	4
+#define	AT91_PM_MODE_MASK	0x01
+#define	AT91_PM_MODE(x)		(((x) & AT91_PM_MODE_MASK) << AT91_PM_MODE_OFFSET)
+
+#define	AT91_PM_SLOW_CLOCK	0x01
+
+#endif

arch/arm/mach-at91/pm_suspend.S

@@ -0,0 +1,338 @@
+/*
+ * arch/arm/mach-at91/pm_slow_clock.S
+ *
+ *  Copyright (C) 2006 Savin Zlobec
+ *
+ * AT91SAM9 support:
+ *  Copyright (C) 2007 Anti Sullin <anti.sullin@artecdesign.ee
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+#include <linux/linkage.h>
+#include <linux/clk/at91_pmc.h>
+#include "pm.h"
+
+#define	SRAMC_SELF_FRESH_ACTIVE		0x01
+#define	SRAMC_SELF_FRESH_EXIT		0x00
+
+pmc	.req	r0
+tmp1	.req	r4
+tmp2	.req	r5
+
+/*
+ * Wait until master clock is ready (after switching master clock source)
+ */
+	.macro wait_mckrdy
+1:	ldr	tmp1, [pmc, #AT91_PMC_SR]
+	tst	tmp1, #AT91_PMC_MCKRDY
+	beq	1b
+	.endm
+
+/*
+ * Wait until master oscillator has stabilized.
+ */
+	.macro wait_moscrdy
+1:	ldr	tmp1, [pmc, #AT91_PMC_SR]
+	tst	tmp1, #AT91_PMC_MOSCS
+	beq	1b
+	.endm
+
+/*
+ * Wait until PLLA has locked.
+ */
+	.macro wait_pllalock
+1:	ldr	tmp1, [pmc, #AT91_PMC_SR]
+	tst	tmp1, #AT91_PMC_LOCKA
+	beq	1b
+	.endm
+
+/*
+ * Put the processor to enter the idle state
+ */
+	.macro at91_cpu_idle
+
+#if defined(CONFIG_CPU_V7)
+	mov	tmp1, #AT91_PMC_PCK
+	str	tmp1, [pmc, #AT91_PMC_SCDR]
+
+	dsb
+
+	wfi		@ Wait For Interrupt
+#else
+	mcr	p15, 0, tmp1, c7, c0, 4
+#endif
+
+	.endm
+
+	.text
+
+	.arm
+
+/*
+ * void at91_pm_suspend_in_sram(void __iomem *pmc, void __iomem *sdramc,
+ *			void __iomem *ramc1, int memctrl)
+ * @input param:
+ * 	@r0: base address of AT91_PMC
+ *  	@r1: base address of SDRAM Controller (SDRAM, DDRSDR, or AT91_SYS)
+ *	@r2: base address of second SDRAM Controller or 0 if not present
+ *	@r3: pm information
+ */
+/* at91_pm_suspend_in_sram must be 8-byte aligned per the requirements of fncpy() */
+	.align 3
+ENTRY(at91_pm_suspend_in_sram)
+	/* Save registers on stack */
+	stmfd	sp!, {r4 - r12, lr}
+
+	/* Drain write buffer */
+	mov	tmp1, #0
+	mcr	p15, 0, tmp1, c7, c10, 4
+
+	str	r0, .pmc_base
+	str	r1, .sramc_base
+	str	r2, .sramc1_base
+
+	and	r0, r3, #AT91_PM_MEMTYPE_MASK
+	str	r0, .memtype
+
+	lsr	r0, r3, #AT91_PM_MODE_OFFSET
+	and	r0, r0, #AT91_PM_MODE_MASK
+	str	r0, .pm_mode
+
+	/* Active the self-refresh mode */
+	mov	r0, #SRAMC_SELF_FRESH_ACTIVE
+	bl	at91_sramc_self_refresh
+
+	ldr	r0, .pm_mode
+	tst	r0, #AT91_PM_SLOW_CLOCK
+	beq	skip_disable_main_clock
+
+	ldr	pmc, .pmc_base
+
+	/* Save Master clock setting */
+	ldr	tmp1, [pmc, #AT91_PMC_MCKR]
+	str	tmp1, .saved_mckr
+
+	/*
+	 * Set the Master clock source to slow clock
+	 */
+	bic	tmp1, tmp1, #AT91_PMC_CSS
+	str	tmp1, [pmc, #AT91_PMC_MCKR]
+
+	wait_mckrdy
+
+	/* Save PLLA setting and disable it */
+	ldr	tmp1, [pmc, #AT91_CKGR_PLLAR]
+	str	tmp1, .saved_pllar
+
+	mov	tmp1, #AT91_PMC_PLLCOUNT
+	orr	tmp1, tmp1, #(1 << 29)		/* bit 29 always set */
+	str	tmp1, [pmc, #AT91_CKGR_PLLAR]
+
+	/* Turn off the main oscillator */
+	ldr	tmp1, [pmc, #AT91_CKGR_MOR]
+	bic	tmp1, tmp1, #AT91_PMC_MOSCEN
+	orr	tmp1, tmp1, #AT91_PMC_KEY
+	str	tmp1, [pmc, #AT91_CKGR_MOR]
+
+skip_disable_main_clock:
+	ldr	pmc, .pmc_base
+
+	/* Wait for interrupt */
+	at91_cpu_idle
+
+	ldr	r0, .pm_mode
+	tst	r0, #AT91_PM_SLOW_CLOCK
+	beq	skip_enable_main_clock
+
+	ldr	pmc, .pmc_base
+
+	/* Turn on the main oscillator */
+	ldr	tmp1, [pmc, #AT91_CKGR_MOR]
+	orr	tmp1, tmp1, #AT91_PMC_MOSCEN
+	orr	tmp1, tmp1, #AT91_PMC_KEY
+	str	tmp1, [pmc, #AT91_CKGR_MOR]
+
+	wait_moscrdy
+
+	/* Restore PLLA setting */
+	ldr	tmp1, .saved_pllar
+	str	tmp1, [pmc, #AT91_CKGR_PLLAR]
+
+	tst	tmp1, #(AT91_PMC_MUL &  0xff0000)
+	bne	3f
+	tst	tmp1, #(AT91_PMC_MUL & ~0xff0000)
+	beq	4f
+3:
+	wait_pllalock
+4:
+
+	/*
+	 * Restore master clock setting
+	 */
+	ldr	tmp1, .saved_mckr
+	str	tmp1, [pmc, #AT91_PMC_MCKR]
+
+	wait_mckrdy
+
+skip_enable_main_clock:
+	/* Exit the self-refresh mode */
+	mov	r0, #SRAMC_SELF_FRESH_EXIT
+	bl	at91_sramc_self_refresh
+
+	/* Restore registers, and return */
+	ldmfd	sp!, {r4 - r12, pc}
+ENDPROC(at91_pm_suspend_in_sram)
+
+/*
+ * void at91_sramc_self_refresh(unsigned int is_active)
+ *
+ * @input param:
+ *	@r0: 1 - active self-refresh mode
+ *	     0 - exit self-refresh mode
+ * register usage:
+ * 	@r1: memory type
+ *	@r2: base address of the sram controller
+ */
+
+ENTRY(at91_sramc_self_refresh)
+	ldr	r1, .memtype
+	ldr	r2, .sramc_base
+
+	cmp	r1, #AT91_MEMCTRL_MC
+	bne	ddrc_sf
+
+	/*
+	 * at91rm9200 Memory controller
+	 */
+
+	 /*
+	  * For exiting the self-refresh mode, do nothing,
+	  * automatically exit the self-refresh mode.
+	  */
+	tst	r0, #SRAMC_SELF_FRESH_ACTIVE
+	beq	exit_sramc_sf
+
+	/* Active SDRAM self-refresh mode */
+	mov	r3, #1
+	str	r3, [r2, #AT91_MC_SDRAMC_SRR]
+	b	exit_sramc_sf
+
+ddrc_sf:
+	cmp	r1, #AT91_MEMCTRL_DDRSDR
+	bne	sdramc_sf
+
+	/*
+	 * DDR Memory controller
+	 */
+	tst	r0, #SRAMC_SELF_FRESH_ACTIVE
+	beq	ddrc_exit_sf
+
+	/* LPDDR1 --> force DDR2 mode during self-refresh */
+	ldr	r3, [r2, #AT91_DDRSDRC_MDR]
+	str	r3, .saved_sam9_mdr
+	bic	r3, r3, #~AT91_DDRSDRC_MD
+	cmp	r3, #AT91_DDRSDRC_MD_LOW_POWER_DDR
+	ldreq	r3, [r2, #AT91_DDRSDRC_MDR]
+	biceq	r3, r3, #AT91_DDRSDRC_MD
+	orreq	r3, r3, #AT91_DDRSDRC_MD_DDR2
+	streq	r3, [r2, #AT91_DDRSDRC_MDR]
+
+	/* Active DDRC self-refresh mode */
+	ldr	r3, [r2, #AT91_DDRSDRC_LPR]
+	str	r3, .saved_sam9_lpr
+	bic	r3, r3, #AT91_DDRSDRC_LPCB
+	orr	r3, r3, #AT91_DDRSDRC_LPCB_SELF_REFRESH
+	str	r3, [r2, #AT91_DDRSDRC_LPR]
+
+	/* If using the 2nd ddr controller */
+	ldr	r2, .sramc1_base
+	cmp	r2, #0
+	beq	no_2nd_ddrc
+
+	ldr	r3, [r2, #AT91_DDRSDRC_MDR]
+	str	r3, .saved_sam9_mdr1
+	bic	r3, r3, #~AT91_DDRSDRC_MD
+	cmp	r3, #AT91_DDRSDRC_MD_LOW_POWER_DDR
+	ldreq	r3, [r2, #AT91_DDRSDRC_MDR]
+	biceq	r3, r3, #AT91_DDRSDRC_MD
+	orreq	r3, r3, #AT91_DDRSDRC_MD_DDR2
+	streq	r3, [r2, #AT91_DDRSDRC_MDR]
+
+	/* Active DDRC self-refresh mode */
+	ldr	r3, [r2, #AT91_DDRSDRC_LPR]
+	str	r3, .saved_sam9_lpr1
+	bic	r3, r3, #AT91_DDRSDRC_LPCB
+	orr	r3, r3, #AT91_DDRSDRC_LPCB_SELF_REFRESH
+	str	r3, [r2, #AT91_DDRSDRC_LPR]
+
+no_2nd_ddrc:
+	b	exit_sramc_sf
+
+ddrc_exit_sf:
+	/* Restore MDR in case of LPDDR1 */
+	ldr	r3, .saved_sam9_mdr
+	str	r3, [r2, #AT91_DDRSDRC_MDR]
+	/* Restore LPR on AT91 with DDRAM */
+	ldr	r3, .saved_sam9_lpr
+	str	r3, [r2, #AT91_DDRSDRC_LPR]
+
+	/* If using the 2nd ddr controller */
+	ldr	r2, .sramc1_base
+	cmp	r2, #0
+	ldrne	r3, .saved_sam9_mdr1
+	strne	r3, [r2, #AT91_DDRSDRC_MDR]
+	ldrne	r3, .saved_sam9_lpr1
+	strne	r3, [r2, #AT91_DDRSDRC_LPR]
+
+	b	exit_sramc_sf
+
+	/*
+	 * SDRAMC Memory controller
+	 */
+sdramc_sf:
+	tst	r0, #SRAMC_SELF_FRESH_ACTIVE
+	beq	sdramc_exit_sf
+
+	/* Active SDRAMC self-refresh mode */
+	ldr	r3, [r2, #AT91_SDRAMC_LPR]
+	str	r3, .saved_sam9_lpr
+	bic	r3, r3, #AT91_SDRAMC_LPCB
+	orr	r3, r3, #AT91_SDRAMC_LPCB_SELF_REFRESH
+	str	r3, [r2, #AT91_SDRAMC_LPR]
+
+sdramc_exit_sf:
+	ldr	r3, .saved_sam9_lpr
+	str	r3, [r2, #AT91_SDRAMC_LPR]
+
+exit_sramc_sf:
+	mov	pc, lr
+ENDPROC(at91_sramc_self_refresh)
+
+.pmc_base:
+	.word 0
+.sramc_base:
+	.word 0
+.sramc1_base:
+	.word 0
+.memtype:
+	.word 0
+.pm_mode:
+	.word 0
+.saved_mckr:
+	.word 0
+.saved_pllar:
+	.word 0
+.saved_sam9_lpr:
+	.word 0
+.saved_sam9_lpr1:
+	.word 0
+.saved_sam9_mdr:
+	.word 0
+.saved_sam9_mdr1:
+	.word 0
+
+ENTRY(at91_pm_suspend_in_sram_sz)
+	.word .-at91_pm_suspend_in_sram

arch/arm/mach-at91/sama5.c

@@ -0,0 +1,97 @@
+/*
+ *  Setup code for SAMA5
+ *
+ *  Copyright (C) 2013 Atmel,
+ *                2013 Ludovic Desroches <ludovic.desroches@atmel.com>
+ *
+ * Licensed under GPLv2 or later.
+ */
+
+#include <linux/of.h>
+#include <linux/of_platform.h>
+
+#include <asm/mach/arch.h>
+#include <asm/mach/map.h>
+#include <asm/system_misc.h>
+
+#include "generic.h"
+#include "soc.h"
+
+static const struct at91_soc sama5_socs[] = {
+	AT91_SOC(SAMA5D2_CIDR_MATCH, SAMA5D21CU_EXID_MATCH,
+		 "sama5d21", "sama5d2"),
+	AT91_SOC(SAMA5D2_CIDR_MATCH, SAMA5D22CU_EXID_MATCH,
+		 "sama5d22", "sama5d2"),
+	AT91_SOC(SAMA5D2_CIDR_MATCH, SAMA5D23CU_EXID_MATCH,
+		 "sama5d23", "sama5d2"),
+	AT91_SOC(SAMA5D2_CIDR_MATCH, SAMA5D24CX_EXID_MATCH,
+		 "sama5d24", "sama5d2"),
+	AT91_SOC(SAMA5D2_CIDR_MATCH, SAMA5D24CU_EXID_MATCH,
+		 "sama5d24", "sama5d2"),
+	AT91_SOC(SAMA5D2_CIDR_MATCH, SAMA5D26CU_EXID_MATCH,
+		 "sama5d26", "sama5d2"),
+	AT91_SOC(SAMA5D2_CIDR_MATCH, SAMA5D27CU_EXID_MATCH,
+		 "sama5d27", "sama5d2"),
+	AT91_SOC(SAMA5D2_CIDR_MATCH, SAMA5D27CN_EXID_MATCH,
+		 "sama5d27", "sama5d2"),
+	AT91_SOC(SAMA5D2_CIDR_MATCH, SAMA5D28CU_EXID_MATCH,
+		 "sama5d28", "sama5d2"),
+	AT91_SOC(SAMA5D2_CIDR_MATCH, SAMA5D28CN_EXID_MATCH,
+		 "sama5d28", "sama5d2"),
+	AT91_SOC(SAMA5D3_CIDR_MATCH, SAMA5D31_EXID_MATCH,
+		 "sama5d31", "sama5d3"),
+	AT91_SOC(SAMA5D3_CIDR_MATCH, SAMA5D33_EXID_MATCH,
+		 "sama5d33", "sama5d3"),
+	AT91_SOC(SAMA5D3_CIDR_MATCH, SAMA5D34_EXID_MATCH,
+		 "sama5d34", "sama5d3"),
+	AT91_SOC(SAMA5D3_CIDR_MATCH, SAMA5D35_EXID_MATCH,
+		 "sama5d35", "sama5d3"),
+	AT91_SOC(SAMA5D3_CIDR_MATCH, SAMA5D36_EXID_MATCH,
+		 "sama5d36", "sama5d3"),
+	AT91_SOC(SAMA5D4_CIDR_MATCH, SAMA5D41_EXID_MATCH,
+		 "sama5d41", "sama5d4"),
+	AT91_SOC(SAMA5D4_CIDR_MATCH, SAMA5D42_EXID_MATCH,
+		 "sama5d42", "sama5d4"),
+	AT91_SOC(SAMA5D4_CIDR_MATCH, SAMA5D43_EXID_MATCH,
+		 "sama5d43", "sama5d4"),
+	AT91_SOC(SAMA5D4_CIDR_MATCH, SAMA5D44_EXID_MATCH,
+		 "sama5d44", "sama5d4"),
+	{ /* sentinel */ },
+};
+
+static void __init sama5_dt_device_init(void)
+{
+	struct soc_device *soc;
+	struct device *soc_dev = NULL;
+
+	soc = at91_soc_init(sama5_socs);
+	if (soc != NULL)
+		soc_dev = soc_device_to_device(soc);
+
+	of_platform_default_populate(NULL, NULL, soc_dev);
+	sama5_pm_init();
+}
+
+static const char *const sama5_dt_board_compat[] __initconst = {
+	"atmel,sama5",
+	NULL
+};
+
+DT_MACHINE_START(sama5_dt, "Atmel SAMA5")
+	/* Maintainer: Atmel */
+	.init_machine	= sama5_dt_device_init,
+	.dt_compat	= sama5_dt_board_compat,
+MACHINE_END
+
+static const char *const sama5_alt_dt_board_compat[] __initconst = {
+	"atmel,sama5d2",
+	"atmel,sama5d4",
+	NULL
+};
+
+DT_MACHINE_START(sama5_alt_dt, "Atmel SAMA5")
+	/* Maintainer: Atmel */
+	.init_machine	= sama5_dt_device_init,
+	.dt_compat	= sama5_alt_dt_board_compat,
+	.l2c_aux_mask	= ~0UL,
+MACHINE_END

arch/arm/mach-at91/soc.c

@@ -0,0 +1,142 @@
+/*
+ * Copyright (C) 2015 Atmel
+ *
+ * Alexandre Belloni <alexandre.belloni@free-electrons.com
+ * Boris Brezillon <boris.brezillon@free-electrons.com
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ *
+ */
+
+#define pr_fmt(fmt)	"AT91: " fmt
+
+#include <linux/io.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/of_platform.h>
+#include <linux/slab.h>
+#include <linux/sys_soc.h>
+
+#include "soc.h"
+
+#define AT91_DBGU_CIDR			0x40
+#define AT91_DBGU_EXID			0x44
+#define AT91_CHIPID_CIDR		0x00
+#define AT91_CHIPID_EXID		0x04
+#define AT91_CIDR_VERSION(x)		((x) & 0x1f)
+#define AT91_CIDR_EXT			BIT(31)
+#define AT91_CIDR_MATCH_MASK		0x7fffffe0
+
+static int __init at91_get_cidr_exid_from_dbgu(u32 *cidr, u32 *exid)
+{
+	struct device_node *np;
+	void __iomem *regs;
+
+	np = of_find_compatible_node(NULL, NULL, "atmel,at91rm9200-dbgu");
+	if (!np)
+		np = of_find_compatible_node(NULL, NULL,
+					     "atmel,at91sam9260-dbgu");
+	if (!np)
+		return -ENODEV;
+
+	regs = of_iomap(np, 0);
+	of_node_put(np);
+
+	if (!regs) {
+		pr_warn("Could not map DBGU iomem range");
+		return -ENXIO;
+	}
+
+	*cidr = readl(regs + AT91_DBGU_CIDR);
+	*exid = readl(regs + AT91_DBGU_EXID);
+
+	iounmap(regs);
+
+	return 0;
+}
+
+static int __init at91_get_cidr_exid_from_chipid(u32 *cidr, u32 *exid)
+{
+	struct device_node *np;
+	void __iomem *regs;
+
+	np = of_find_compatible_node(NULL, NULL, "atmel,sama5d2-chipid");
+	if (!np)
+		return -ENODEV;
+
+	regs = of_iomap(np, 0);
+	of_node_put(np);
+
+	if (!regs) {
+		pr_warn("Could not map DBGU iomem range");
+		return -ENXIO;
+	}
+
+	*cidr = readl(regs + AT91_CHIPID_CIDR);
+	*exid = readl(regs + AT91_CHIPID_EXID);
+
+	iounmap(regs);
+
+	return 0;
+}
+
+struct soc_device * __init at91_soc_init(const struct at91_soc *socs)
+{
+	struct soc_device_attribute *soc_dev_attr;
+	const struct at91_soc *soc;
+	struct soc_device *soc_dev;
+	u32 cidr, exid;
+	int ret;
+
+	/*
+	 * With SAMA5D2 and later SoCs, CIDR and EXID registers are no more
+	 * in the dbgu device but in the chipid device whose purpose is only
+	 * to expose these two registers.
+	 */
+	ret = at91_get_cidr_exid_from_dbgu(&cidr, &exid);
+	if (ret)
+		ret = at91_get_cidr_exid_from_chipid(&cidr, &exid);
+	if (ret) {
+		if (ret == -ENODEV)
+			pr_warn("Could not find identification node");
+		return NULL;
+	}
+
+	for (soc = socs; soc->name; soc++) {
+		if (soc->cidr_match != (cidr & AT91_CIDR_MATCH_MASK))
+			continue;
+
+		if (!(cidr & AT91_CIDR_EXT) || soc->exid_match == exid)
+			break;
+	}
+
+	if (!soc->name) {
+		pr_warn("Could not find matching SoC description\n");
+		return NULL;
+	}
+
+	soc_dev_attr = kzalloc(sizeof(*soc_dev_attr), GFP_KERNEL);
+	if (!soc_dev_attr)
+		return NULL;
+
+	soc_dev_attr->family = soc->family;
+	soc_dev_attr->soc_id = soc->name;
+	soc_dev_attr->revision = kasprintf(GFP_KERNEL, "%X",
+					   AT91_CIDR_VERSION(cidr));
+	soc_dev = soc_device_register(soc_dev_attr);
+	if (IS_ERR(soc_dev)) {
+		kfree(soc_dev_attr->revision);
+		kfree(soc_dev_attr);
+		pr_warn("Could not register SoC device\n");
+		return NULL;
+	}
+
+	if (soc->family)
+		pr_info("Detected SoC family: %s\n", soc->family);
+	pr_info("Detected SoC: %s, revision %X\n", soc->name,
+		AT91_CIDR_VERSION(cidr));
+
+	return soc_dev;
+}

arch/arm/mach-at91/soc.h

@@ -0,0 +1,91 @@
+/*
+ * Copyright (C) 2015 Atmel
+ *
+ * Boris Brezillon <boris.brezillon@free-electrons.com
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ *
+ */
+
+#ifndef __AT91_SOC_H
+#define __AT91_SOC_H
+
+#include <linux/sys_soc.h>
+
+struct at91_soc {
+	u32 cidr_match;
+	u32 exid_match;
+	const char *name;
+	const char *family;
+};
+
+#define AT91_SOC(__cidr, __exid, __name, __family)		\
+	{							\
+		.cidr_match = (__cidr),				\
+		.exid_match = (__exid),				\
+		.name = (__name),				\
+		.family = (__family),				\
+	}
+
+struct soc_device * __init
+at91_soc_init(const struct at91_soc *socs);
+
+#define AT91RM9200_CIDR_MATCH		0x09290780
+
+#define AT91SAM9260_CIDR_MATCH		0x019803a0
+#define AT91SAM9261_CIDR_MATCH		0x019703a0
+#define AT91SAM9263_CIDR_MATCH		0x019607a0
+#define AT91SAM9G20_CIDR_MATCH		0x019905a0
+#define AT91SAM9RL64_CIDR_MATCH		0x019b03a0
+#define AT91SAM9G45_CIDR_MATCH		0x019b05a0
+#define AT91SAM9X5_CIDR_MATCH		0x019a05a0
+#define AT91SAM9N12_CIDR_MATCH		0x019a07a0
+
+#define AT91SAM9M11_EXID_MATCH		0x00000001
+#define AT91SAM9M10_EXID_MATCH		0x00000002
+#define AT91SAM9G46_EXID_MATCH		0x00000003
+#define AT91SAM9G45_EXID_MATCH		0x00000004
+
+#define AT91SAM9G15_EXID_MATCH		0x00000000
+#define AT91SAM9G35_EXID_MATCH		0x00000001
+#define AT91SAM9X35_EXID_MATCH		0x00000002
+#define AT91SAM9G25_EXID_MATCH		0x00000003
+#define AT91SAM9X25_EXID_MATCH		0x00000004
+
+#define AT91SAM9CN12_EXID_MATCH		0x00000005
+#define AT91SAM9N12_EXID_MATCH		0x00000006
+#define AT91SAM9CN11_EXID_MATCH		0x00000009
+
+#define AT91SAM9XE128_CIDR_MATCH	0x329973a0
+#define AT91SAM9XE256_CIDR_MATCH	0x329a93a0
+#define AT91SAM9XE512_CIDR_MATCH	0x329aa3a0
+
+#define SAMA5D2_CIDR_MATCH		0x0a5c08c0
+#define SAMA5D21CU_EXID_MATCH		0x0000005a
+#define SAMA5D22CU_EXID_MATCH		0x00000059
+#define SAMA5D22CN_EXID_MATCH		0x00000069
+#define SAMA5D23CU_EXID_MATCH		0x00000058
+#define SAMA5D24CX_EXID_MATCH		0x00000004
+#define SAMA5D24CU_EXID_MATCH		0x00000014
+#define SAMA5D26CU_EXID_MATCH		0x00000012
+#define SAMA5D27CU_EXID_MATCH		0x00000011
+#define SAMA5D27CN_EXID_MATCH		0x00000021
+#define SAMA5D28CU_EXID_MATCH		0x00000010
+#define SAMA5D28CN_EXID_MATCH		0x00000020
+
+#define SAMA5D3_CIDR_MATCH		0x0a5c07c0
+#define SAMA5D31_EXID_MATCH		0x00444300
+#define SAMA5D33_EXID_MATCH		0x00414300
+#define SAMA5D34_EXID_MATCH		0x00414301
+#define SAMA5D35_EXID_MATCH		0x00584300
+#define SAMA5D36_EXID_MATCH		0x00004301
+
+#define SAMA5D4_CIDR_MATCH		0x0a5c07c0
+#define SAMA5D41_EXID_MATCH		0x00000001
+#define SAMA5D42_EXID_MATCH		0x00000002
+#define SAMA5D43_EXID_MATCH		0x00000003
+#define SAMA5D44_EXID_MATCH		0x00000004
+
+#endif /* __AT91_SOC_H */

arch/arm/mach-axxia/Kconfig

@@ -0,0 +1,15 @@
+config ARCH_AXXIA
+	bool "LSI Axxia platforms"
+	depends on ARCH_MULTI_V7 && ARM_LPAE
+	select ARCH_DMA_ADDR_T_64BIT
+	select ARM_AMBA
+	select ARM_GIC
+	select ARM_TIMER_SP804
+	select HAVE_ARM_ARCH_TIMER
+	select MFD_SYSCON
+	select ZONE_DMA
+	help
+	  This enables support for the LSI Axxia devices.
+
+	  The LSI Axxia platforms require a Flattened Device Tree to be passed
+	  to the kernel.

arch/arm/mach-axxia/Makefile

@@ -0,0 +1,2 @@
+obj-y				+= axxia.o
+obj-$(CONFIG_SMP)		+= platsmp.o

arch/arm/mach-axxia/axxia.c

@@ -0,0 +1,28 @@
+/*
+ * Support for the LSI Axxia SoC devices based on ARM cores.
+ *
+ * Copyright (C) 2012 LSI
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+#include <linux/init.h>
+#include <asm/mach/arch.h>
+
+static const char *const axxia_dt_match[] __initconst = {
+	"lsi,axm5516",
+	"lsi,axm5516-sim",
+	"lsi,axm5516-emu",
+	NULL
+};
+
+DT_MACHINE_START(AXXIA_DT, "LSI Axxia AXM55XX")
+	.dt_compat = axxia_dt_match,
+MACHINE_END

arch/arm/mach-axxia/platsmp.c

@@ -0,0 +1,89 @@
+/*
+ * linux/arch/arm/mach-axxia/platsmp.c
+ *
+ * Copyright (C) 2012 LSI Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/init.h>
+#include <linux/io.h>
+#include <linux/smp.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <asm/cacheflush.h>
+
+/* Syscon register offsets for releasing cores from reset */
+#define SC_CRIT_WRITE_KEY	0x1000
+#define SC_RST_CPU_HOLD		0x1010
+
+/*
+ * Write the kernel entry point for secondary CPUs to the specified address
+ */
+static void write_release_addr(u32 release_phys)
+{
+	u32 *virt = (u32 *) phys_to_virt(release_phys);
+	writel_relaxed(virt_to_phys(secondary_startup), virt);
+	/* Make sure this store is visible to other CPUs */
+	smp_wmb();
+	__cpuc_flush_dcache_area(virt, sizeof(u32));
+}
+
+static int axxia_boot_secondary(unsigned int cpu, struct task_struct *idle)
+{
+	struct device_node *syscon_np;
+	void __iomem *syscon;
+	u32 tmp;
+
+	syscon_np = of_find_compatible_node(NULL, NULL, "lsi,axxia-syscon");
+	if (!syscon_np)
+		return -ENOENT;
+
+	syscon = of_iomap(syscon_np, 0);
+	if (!syscon)
+		return -ENOMEM;
+
+	tmp = readl(syscon + SC_RST_CPU_HOLD);
+	writel(0xab, syscon + SC_CRIT_WRITE_KEY);
+	tmp &= ~(1 << cpu);
+	writel(tmp, syscon + SC_RST_CPU_HOLD);
+
+	return 0;
+}
+
+static void __init axxia_smp_prepare_cpus(unsigned int max_cpus)
+{
+	int cpu_count = 0;
+	int cpu;
+
+	/*
+	 * Initialise the present map, which describes the set of CPUs actually
+	 * populated at the present time.
+	 */
+	for_each_possible_cpu(cpu) {
+		struct device_node *np;
+		u32 release_phys;
+
+		np = of_get_cpu_node(cpu, NULL);
+		if (!np)
+			continue;
+		if (of_property_read_u32(np, "cpu-release-addr", &release_phys))
+			continue;
+
+		if (cpu_count < max_cpus) {
+			set_cpu_present(cpu, true);
+			cpu_count++;
+		}
+
+		if (release_phys != 0)
+			write_release_addr(release_phys);
+	}
+}
+
+static const struct smp_operations axxia_smp_ops __initconst = {
+	.smp_prepare_cpus	= axxia_smp_prepare_cpus,
+	.smp_boot_secondary	= axxia_boot_secondary,
+};
+CPU_METHOD_OF_DECLARE(axxia_smp, "lsi,syscon-release", &axxia_smp_ops);

arch/arm/mach-bcm/Kconfig

@@ -0,0 +1,212 @@
+menuconfig ARCH_BCM
+	bool "Broadcom SoC Support"
+	depends on ARCH_MULTI_V6_V7
+	help
+	  This enables support for Broadcom ARM based SoC chips
+
+if ARCH_BCM
+
+comment "IPROC architected SoCs"
+
+config ARCH_BCM_IPROC
+	bool
+	select ARM_GIC
+	select CACHE_L2X0
+	select HAVE_ARM_SCU if SMP
+	select HAVE_ARM_TWD if SMP
+	select ARM_GLOBAL_TIMER
+	select CLKSRC_MMIO
+	select GPIOLIB
+	select ARM_AMBA
+	select PINCTRL
+	help
+	  This enables support for systems based on Broadcom IPROC architected SoCs.
+	  The IPROC complex contains one or more ARM CPUs along with common
+	  core periperals. Application specific SoCs are created by adding a
+	  uArchitecture containing peripherals outside of the IPROC complex.
+	  Currently supported SoCs are Cygnus.
+
+config ARCH_BCM_CYGNUS
+	bool "Broadcom Cygnus Support"
+	depends on ARCH_MULTI_V7
+	select ARCH_BCM_IPROC
+	help
+	  Enable support for the Cygnus family,
+	  which includes the following variants:
+	  BCM11300, BCM11320, BCM11350, BCM11360,
+	  BCM58300, BCM58302, BCM58303, BCM58305.
+
+config ARCH_BCM_NSP
+	bool "Broadcom Northstar Plus SoC Support"
+	depends on ARCH_MULTI_V7
+	select ARCH_BCM_IPROC
+	select ARM_ERRATA_754322
+	select ARM_ERRATA_775420
+	select ARM_ERRATA_764369 if SMP
+	select HAVE_SMP
+	help
+	  Support for Broadcom Northstar Plus SoC.
+	  Broadcom Northstar Plus family of SoCs are used for switching control
+	  and management applications as well as residential router/gateway
+	  applications. The SoC features dual core Cortex A9 ARM CPUs,
+	  integrating several peripheral interfaces including multiple Gigabit
+	  Ethernet PHYs, DDR3 memory, PCIE Gen-2, USB 2.0 and USB 3.0, serial and
+	  NAND flash, SATA and several other IO controllers.
+
+config ARCH_BCM_5301X
+	bool "Broadcom BCM470X / BCM5301X ARM SoC"
+	depends on ARCH_MULTI_V7
+	select ARCH_BCM_IPROC
+	select ARM_ERRATA_754322
+	select ARM_ERRATA_775420
+	select ARM_ERRATA_764369 if SMP
+	select HAVE_SMP
+
+	help
+	  Support for Broadcom BCM470X and BCM5301X SoCs with ARM CPU cores.
+
+	  This is a network SoC line mostly used in home routers and
+	  wifi access points, it's internal name is Northstar.
+	  This inclused the following SoC: BCM53010, BCM53011, BCM53012,
+	  BCM53014, BCM53015, BCM53016, BCM53017, BCM53018, BCM4707,
+	  BCM4708 and BCM4709.
+
+	  Do not confuse this with the BCM4760 which is a totally
+	  different SoC or with the older BCM47XX and BCM53XX based
+	  network SoC using a MIPS CPU, they are supported by arch/mips/bcm47xx
+
+comment "KONA architected SoCs"
+
+config ARCH_BCM_MOBILE
+	bool
+	select GPIOLIB
+	select ARM_ERRATA_754322
+	select ARM_ERRATA_775420
+	select ARM_GIC
+	select GPIO_BCM_KONA
+	select TICK_ONESHOT
+	select HAVE_ARM_ARCH_TIMER
+	select PINCTRL
+	select ARCH_BCM_MOBILE_SMP if SMP
+	select BCM_KONA_TIMER
+	help
+	  This enables support for systems based on Broadcom mobile SoCs.
+
+config ARCH_BCM_281XX
+	bool "Broadcom BCM281XX SoC family"
+	depends on ARCH_MULTI_V7
+	select ARCH_BCM_MOBILE
+	select HAVE_SMP
+	help
+	  Enable support for the BCM281XX family, which includes
+	  BCM11130, BCM11140, BCM11351, BCM28145 and BCM28155
+	  variants.
+
+config ARCH_BCM_21664
+	bool "Broadcom BCM21664 SoC family"
+	depends on ARCH_MULTI_V7
+	select ARCH_BCM_MOBILE
+	select HAVE_SMP
+	help
+	  Enable support for the BCM21664 family, which includes
+	  BCM21663 and BCM21664 variants.
+
+config ARCH_BCM_23550
+	bool "Broadcom BCM23550 SoC"
+	depends on ARCH_MULTI_V7
+	select ARCH_BCM_MOBILE
+	select HAVE_SMP
+	help
+	  Enable support for the BCM23550.
+
+config ARCH_BCM_MOBILE_L2_CACHE
+	bool "Broadcom mobile SoC level 2 cache support"
+	depends on ARCH_BCM_281XX || ARCH_BCM_21664
+	default y
+	select CACHE_L2X0
+	select ARCH_BCM_MOBILE_SMC
+
+config ARCH_BCM_MOBILE_SMC
+	bool
+	depends on ARCH_BCM_MOBILE
+
+config ARCH_BCM_MOBILE_SMP
+	bool
+	depends on ARCH_BCM_MOBILE
+	select HAVE_ARM_SCU
+	select ARM_ERRATA_764369
+	help
+	  SMP support for the BCM281XX, BCM21664 and BCM23550 SoC families.
+	  Provided as an option so SMP support for SoCs of this type
+	  can be disabled for an SMP-enabled kernel.
+
+comment "Other Architectures"
+
+config ARCH_BCM2835
+	bool "Broadcom BCM2835 family"
+	depends on ARCH_MULTI_V6 || ARCH_MULTI_V7
+	select GPIOLIB
+	select ARM_AMBA
+	select ARM_ERRATA_411920 if ARCH_MULTI_V6
+	select ARM_TIMER_SP804
+	select HAVE_ARM_ARCH_TIMER if ARCH_MULTI_V7
+	select CLKSRC_OF
+	select BCM2835_TIMER
+	select PINCTRL
+	select PINCTRL_BCM2835
+	help
+	  This enables support for the Broadcom BCM2835 and BCM2836 SoCs.
+	  This SoC is used in the Raspberry Pi and Roku 2 devices.
+
+config ARCH_BCM_53573
+	bool "Broadcom BCM53573 SoC series support"
+	depends on ARCH_MULTI_V7
+	select ARCH_BCM_IPROC
+	select HAVE_ARM_ARCH_TIMER
+	help
+	  BCM53573 series is set of SoCs using ARM Cortex-A7 CPUs with wireless
+	  embedded in the chipset.
+	  This SoC line is mostly used in home routers and is some cheaper
+	  alternative for Northstar family.
+
+	  The base chip is BCM53573 and there are some packaging modifications
+	  like BCM47189 and BCM47452.
+
+config ARCH_BCM_63XX
+	bool "Broadcom BCM63xx DSL SoC"
+	depends on ARCH_MULTI_V7
+	depends on MMU
+	select ARM_ERRATA_754322
+	select ARM_ERRATA_764369 if SMP
+	select ARM_GIC
+	select ARM_GLOBAL_TIMER
+	select CACHE_L2X0
+	select HAVE_ARM_ARCH_TIMER
+	select HAVE_ARM_TWD if SMP
+	select HAVE_ARM_SCU if SMP
+	select HAVE_SMP
+	help
+	  This enables support for systems based on Broadcom DSL SoCs.
+	  It currently supports the 'BCM63XX' ARM-based family, which includes
+	  the BCM63138 variant.
+
+config ARCH_BRCMSTB
+	bool "Broadcom BCM7XXX based boards"
+	depends on ARCH_MULTI_V7
+	select ARM_GIC
+	select ARM_ERRATA_798181 if SMP
+	select HAVE_ARM_ARCH_TIMER
+	select BRCMSTB_L2_IRQ
+	select BCM7120_L2_IRQ
+	select ARCH_DMA_ADDR_T_64BIT if ARM_LPAE
+	select ZONE_DMA if ARM_LPAE
+	select SOC_BRCMSTB
+	select SOC_BUS
+	help
+	  Say Y if you intend to run the kernel on a Broadcom ARM-based STB
+	  chipset.
+
+	  This enables support for Broadcom ARM-based set-top box chipsets,
+	  including the 7445 family of chips.
+
+endif

arch/arm/mach-bcm/Makefile

@@ -0,0 +1,63 @@
+#
+# Copyright (C) 2012-2015 Broadcom Corporation
+#
+# This program is free software; you can redistribute it and/or
+# modify it under the terms of the GNU General Public License as
+# published by the Free Software Foundation version 2.
+#
+# This program is distributed "as is" WITHOUT ANY WARRANTY of any
+# kind, whether express or implied; without even the implied warranty
+# of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+
+# Cygnus
+obj-$(CONFIG_ARCH_BCM_CYGNUS) +=  bcm_cygnus.o
+
+# Northstar Plus
+obj-$(CONFIG_ARCH_BCM_NSP)	+= bcm_nsp.o
+
+ifeq ($(CONFIG_ARCH_BCM_NSP),y)
+obj-$(CONFIG_SMP)		+= platsmp.o
+endif
+
+# BCM281XX
+obj-$(CONFIG_ARCH_BCM_281XX)	+= board_bcm281xx.o
+
+# BCM21664
+obj-$(CONFIG_ARCH_BCM_21664)	+= board_bcm21664.o
+
+# BCM23550
+obj-$(CONFIG_ARCH_BCM_23550)	+= board_bcm23550.o
+
+# BCM281XX, BCM21664 and BCM23550 SMP support
+obj-$(CONFIG_ARCH_BCM_MOBILE_SMP) += platsmp.o
+
+# BCM281XX and BCM21664 L2 cache control
+obj-$(CONFIG_ARCH_BCM_MOBILE_L2_CACHE) += kona_l2_cache.o
+
+# Support for secure monitor traps
+obj-$(CONFIG_ARCH_BCM_MOBILE_SMC) += bcm_kona_smc.o
+ifeq ($(call as-instr,.arch_extension sec,as_has_sec),as_has_sec)
+CFLAGS_bcm_kona_smc.o		+= -Wa,-march=armv7-a+sec -DREQUIRES_SEC
+endif
+
+# BCM2835
+obj-$(CONFIG_ARCH_BCM2835)	+= board_bcm2835.o
+
+# BCM5301X
+obj-$(CONFIG_ARCH_BCM_5301X)	+= bcm_5301x.o
+ifeq ($(CONFIG_ARCH_BCM_5301X),y)
+obj-$(CONFIG_SMP)		+= platsmp.o
+endif
+
+# BCM63XXx
+ifeq ($(CONFIG_ARCH_BCM_63XX),y)
+obj-y				+= bcm63xx.o
+obj-$(CONFIG_SMP)		+= bcm63xx_smp.o bcm63xx_pmb.o
+endif
+
+ifeq ($(CONFIG_ARCH_BRCMSTB),y)
+CFLAGS_platsmp-brcmstb.o	+= -march=armv7-a
+obj-y				+= brcmstb.o
+obj-$(CONFIG_SMP)		+= platsmp-brcmstb.o
+endif

arch/arm/mach-bcm/bcm63xx.c

@@ -0,0 +1,27 @@
+/*
+ * Copyright (C) 2014 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/of_platform.h>
+
+#include <asm/mach/arch.h>
+
+static const char * const bcm63xx_dt_compat[] = {
+	"brcm,bcm63138",
+	NULL
+};
+
+DT_MACHINE_START(BCM63XXX_DT, "BCM63xx DSL SoC")
+	.dt_compat	= bcm63xx_dt_compat,
+	.l2c_aux_val	= 0,
+	.l2c_aux_mask	= ~0,
+MACHINE_END

arch/arm/mach-bcm/bcm63xx_pmb.c

@@ -0,0 +1,219 @@
+/*
+ * Broadcom BCM63138 PMB initialization for secondary CPU(s)
+ *
+ * Copyright (C) 2015 Broadcom Corporation
+ * Author: Florian Fainelli <f.fainelli@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+#include <linux/kernel.h>
+#include <linux/io.h>
+#include <linux/spinlock.h>
+#include <linux/reset/bcm63xx_pmb.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+
+#include "bcm63xx_smp.h"
+
+/* ARM Control register definitions */
+#define CORE_PWR_CTRL_SHIFT	0
+#define CORE_PWR_CTRL_MASK	0x3
+#define PLL_PWR_ON		BIT(8)
+#define PLL_LDO_PWR_ON		BIT(9)
+#define PLL_CLAMP_ON		BIT(10)
+#define CPU_RESET_N(x)		BIT(13 + (x))
+#define NEON_RESET_N		BIT(15)
+#define PWR_CTRL_STATUS_SHIFT	28
+#define PWR_CTRL_STATUS_MASK	0x3
+#define PWR_DOWN_SHIFT		30
+#define PWR_DOWN_MASK		0x3
+
+/* CPU Power control register definitions */
+#define MEM_PWR_OK		BIT(0)
+#define MEM_PWR_ON		BIT(1)
+#define MEM_CLAMP_ON		BIT(2)
+#define MEM_PWR_OK_STATUS	BIT(4)
+#define MEM_PWR_ON_STATUS	BIT(5)
+#define MEM_PDA_SHIFT		8
+#define MEM_PDA_MASK		0xf
+#define  MEM_PDA_CPU_MASK	0x1
+#define  MEM_PDA_NEON_MASK	0xf
+#define CLAMP_ON		BIT(15)
+#define PWR_OK_SHIFT		16
+#define PWR_OK_MASK		0xf
+#define PWR_ON_SHIFT		20
+#define  PWR_CPU_MASK		0x03
+#define  PWR_NEON_MASK		0x01
+#define PWR_ON_MASK		0xf
+#define PWR_OK_STATUS_SHIFT	24
+#define PWR_OK_STATUS_MASK	0xf
+#define PWR_ON_STATUS_SHIFT	28
+#define PWR_ON_STATUS_MASK	0xf
+
+#define ARM_CONTROL		0x30
+#define ARM_PWR_CONTROL_BASE	0x34
+#define ARM_PWR_CONTROL(x)	(ARM_PWR_CONTROL_BASE + (x) * 0x4)
+#define ARM_NEON_L2		0x3c
+
+/* Perform a value write, then spin until the value shifted by
+ * shift is seen, masked with mask and is different from cond.
+ */
+static int bpcm_wr_rd_mask(void __iomem *master,
+			   unsigned int addr, u32 off, u32 *val,
+			   u32 shift, u32 mask, u32 cond)
+{
+	int ret;
+
+	ret = bpcm_wr(master, addr, off, *val);
+	if (ret)
+		return ret;
+
+	do {
+		ret = bpcm_rd(master, addr, off, val);
+		if (ret)
+			return ret;
+
+		cpu_relax();
+	} while (((*val >> shift) & mask) != cond);
+
+	return ret;
+}
+
+/* Global lock to serialize accesses to the PMB registers while we
+ * are bringing up the secondary CPU
+ */
+static DEFINE_SPINLOCK(pmb_lock);
+
+static int bcm63xx_pmb_get_resources(struct device_node *dn,
+				     void __iomem **base,
+				     unsigned int *cpu,
+				     unsigned int *addr)
+{
+	struct of_phandle_args args;
+	int ret;
+
+	ret = of_property_read_u32(dn, "reg", cpu);
+	if (ret) {
+		pr_err("CPU is missing a reg node\n");
+		return ret;
+	}
+
+	ret = of_parse_phandle_with_args(dn, "resets", "#reset-cells",
+					 0, &args);
+	if (ret) {
+		pr_err("CPU is missing a resets phandle\n");
+		return ret;
+	}
+
+	if (args.args_count != 2) {
+		pr_err("reset-controller does not conform to reset-cells\n");
+		return -EINVAL;
+	}
+
+	*base = of_iomap(args.np, 0);
+	if (!*base) {
+		pr_err("failed remapping PMB register\n");
+		return -ENOMEM;
+	}
+
+	/* We do not need the number of zones */
+	*addr = args.args[0];
+
+	return 0;
+}
+
+int bcm63xx_pmb_power_on_cpu(struct device_node *dn)
+{
+	void __iomem *base;
+	unsigned int cpu, addr;
+	unsigned long flags;
+	u32 val, ctrl;
+	int ret;
+
+	ret = bcm63xx_pmb_get_resources(dn, &base, &cpu, &addr);
+	if (ret)
+		return ret;
+
+	/* We would not know how to enable a third and greater CPU */
+	WARN_ON(cpu > 1);
+
+	spin_lock_irqsave(&pmb_lock, flags);
+
+	/* Check if the CPU is already on and save the ARM_CONTROL register
+	 * value since we will use it later for CPU de-assert once done with
+	 * the CPU-specific power sequence
+	 */
+	ret = bpcm_rd(base, addr, ARM_CONTROL, &ctrl);
+	if (ret)
+		goto out;
+
+	if (ctrl & CPU_RESET_N(cpu)) {
+		pr_info("PMB: CPU%d is already powered on\n", cpu);
+		ret = 0;
+		goto out;
+	}
+
+	/* Power on PLL */
+	ret = bpcm_rd(base, addr, ARM_PWR_CONTROL(cpu), &val);
+	if (ret)
+		goto out;
+
+	val |= (PWR_CPU_MASK << PWR_ON_SHIFT);
+
+	ret = bpcm_wr_rd_mask(base, addr, ARM_PWR_CONTROL(cpu), &val,
+			PWR_ON_STATUS_SHIFT, PWR_CPU_MASK, PWR_CPU_MASK);
+	if (ret)
+		goto out;
+
+	val |= (PWR_CPU_MASK << PWR_OK_SHIFT);
+
+	ret = bpcm_wr_rd_mask(base, addr, ARM_PWR_CONTROL(cpu), &val,
+			PWR_OK_STATUS_SHIFT, PWR_CPU_MASK, PWR_CPU_MASK);
+	if (ret)
+		goto out;
+
+	val &= ~CLAMP_ON;
+
+	ret = bpcm_wr(base, addr, ARM_PWR_CONTROL(cpu), val);
+	if (ret)
+		goto out;
+
+	/* Power on CPU<N> RAM */
+	val &= ~(MEM_PDA_MASK << MEM_PDA_SHIFT);
+
+	ret = bpcm_wr(base, addr, ARM_PWR_CONTROL(cpu), val);
+	if (ret)
+		goto out;
+
+	val |= MEM_PWR_ON;
+
+	ret = bpcm_wr_rd_mask(base, addr, ARM_PWR_CONTROL(cpu), &val,
+			0, MEM_PWR_ON_STATUS, MEM_PWR_ON_STATUS);
+	if (ret)
+		goto out;
+
+	val |= MEM_PWR_OK;
+
+	ret = bpcm_wr_rd_mask(base, addr, ARM_PWR_CONTROL(cpu), &val,
+			0, MEM_PWR_OK_STATUS, MEM_PWR_OK_STATUS);
+	if (ret)
+		goto out;
+
+	val &= ~MEM_CLAMP_ON;
+
+	ret = bpcm_wr(base, addr, ARM_PWR_CONTROL(cpu), val);
+	if (ret)
+		goto out;
+
+	/* De-assert CPU reset */
+	ctrl |= CPU_RESET_N(cpu);
+
+	ret = bpcm_wr(base, addr, ARM_CONTROL, ctrl);
+out:
+	spin_unlock_irqrestore(&pmb_lock, flags);
+	iounmap(base);
+	return ret;
+}

arch/arm/mach-bcm/bcm63xx_smp.c

@@ -0,0 +1,169 @@
+/*
+ * Broadcom BCM63138 DSL SoCs SMP support code
+ *
+ * Copyright (C) 2015, Broadcom Corporation
+ *
+ * Licensed under the terms of the GPLv2
+ */
+
+#include <linux/delay.h>
+#include <linux/init.h>
+#include <linux/smp.h>
+#include <linux/io.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+
+#include <asm/cacheflush.h>
+#include <asm/smp_scu.h>
+#include <asm/smp_plat.h>
+#include <asm/vfp.h>
+
+#include "bcm63xx_smp.h"
+
+/* Size of mapped Cortex A9 SCU address space */
+#define CORTEX_A9_SCU_SIZE	0x58
+
+/*
+ * Enable the Cortex A9 Snoop Control Unit
+ *
+ * By the time this is called we already know there are multiple
+ * cores present.  We assume we're running on a Cortex A9 processor,
+ * so any trouble getting the base address register or getting the
+ * SCU base is a problem.
+ *
+ * Return 0 if successful or an error code otherwise.
+ */
+static int __init scu_a9_enable(void)
+{
+	unsigned long config_base;
+	void __iomem *scu_base;
+	unsigned int i, ncores;
+
+	if (!scu_a9_has_base()) {
+		pr_err("no configuration base address register!\n");
+		return -ENXIO;
+	}
+
+	/* Config base address register value is zero for uniprocessor */
+	config_base = scu_a9_get_base();
+	if (!config_base) {
+		pr_err("hardware reports only one core\n");
+		return -ENOENT;
+	}
+
+	scu_base = ioremap((phys_addr_t)config_base, CORTEX_A9_SCU_SIZE);
+	if (!scu_base) {
+		pr_err("failed to remap config base (%lu/%u) for SCU\n",
+			config_base, CORTEX_A9_SCU_SIZE);
+		return -ENOMEM;
+	}
+
+	scu_enable(scu_base);
+
+	ncores = scu_base ? scu_get_core_count(scu_base) : 1;
+
+	if (ncores > nr_cpu_ids) {
+		pr_warn("SMP: %u cores greater than maximum (%u), clipping\n",
+				ncores, nr_cpu_ids);
+		ncores = nr_cpu_ids;
+	}
+
+	/* The BCM63138 SoC has two Cortex-A9 CPUs, CPU0 features a complete
+	 * and fully functional VFP unit that can be used, but CPU1 does not.
+	 * Since we will not be able to trap kernel-mode NEON to force
+	 * migration to CPU0, just do not advertise VFP support at all.
+	 *
+	 * This will make vfp_init bail out and do not attempt to use VFP at
+	 * all, for kernel-mode NEON, we do not want to introduce any
+	 * conditionals in hot-paths, so we just restrict the system to UP.
+	 */
+#ifdef CONFIG_VFP
+	if (ncores > 1) {
+		pr_warn("SMP: secondary CPUs lack VFP unit, disabling VFP\n");
+		vfp_disable();
+
+#ifdef CONFIG_KERNEL_MODE_NEON
+		WARN(1, "SMP: kernel-mode NEON enabled, restricting to UP\n");
+		ncores = 1;
+#endif
+	}
+#endif
+
+	for (i = 0; i < ncores; i++)
+		set_cpu_possible(i, true);
+
+	iounmap(scu_base);	/* That's the last we'll need of this */
+
+	return 0;
+}
+
+static const struct of_device_id bcm63138_bootlut_ids[] = {
+	{ .compatible = "brcm,bcm63138-bootlut", },
+	{ /* sentinel */ },
+};
+
+#define BOOTLUT_RESET_VECT	0x20
+
+static int bcm63138_smp_boot_secondary(unsigned int cpu,
+				       struct task_struct *idle)
+{
+	void __iomem *bootlut_base;
+	struct device_node *dn;
+	int ret = 0;
+	u32 val;
+
+	dn = of_find_matching_node(NULL, bcm63138_bootlut_ids);
+	if (!dn) {
+		pr_err("SMP: unable to find bcm63138 boot LUT node\n");
+		return -ENODEV;
+	}
+
+	bootlut_base = of_iomap(dn, 0);
+	of_node_put(dn);
+
+	if (!bootlut_base) {
+		pr_err("SMP: unable to remap boot LUT base register\n");
+		return -ENOMEM;
+	}
+
+	/* Locate the secondary CPU node */
+	dn = of_get_cpu_node(cpu, NULL);
+	if (!dn) {
+		pr_err("SMP: failed to locate secondary CPU%d node\n", cpu);
+		ret = -ENODEV;
+		goto out;
+	}
+
+	/* Write the secondary init routine to the BootLUT reset vector */
+	val = virt_to_phys(secondary_startup);
+	writel_relaxed(val, bootlut_base + BOOTLUT_RESET_VECT);
+
+	/* Power up the core, will jump straight to its reset vector when we
+	 * return
+	 */
+	ret = bcm63xx_pmb_power_on_cpu(dn);
+	if (ret)
+		goto out;
+out:
+	iounmap(bootlut_base);
+
+	return ret;
+}
+
+static void __init bcm63138_smp_prepare_cpus(unsigned int max_cpus)
+{
+	int ret;
+
+	ret = scu_a9_enable();
+	if (ret) {
+		pr_warn("SMP: Cortex-A9 SCU setup failed\n");
+		return;
+	}
+}
+
+static const struct smp_operations bcm63138_smp_ops __initconst = {
+	.smp_prepare_cpus	= bcm63138_smp_prepare_cpus,
+	.smp_boot_secondary	= bcm63138_smp_boot_secondary,
+};
+
+CPU_METHOD_OF_DECLARE(bcm63138_smp, "brcm,bcm63138", &bcm63138_smp_ops);

arch/arm/mach-bcm/bcm63xx_smp.h

@@ -0,0 +1,8 @@
+#ifndef __BCM63XX_SMP_H
+#define __BCM63XX_SMP_H
+
+struct device_node;
+
+extern int bcm63xx_pmb_power_on_cpu(struct device_node *dn);
+
+#endif /* __BCM63XX_SMP_H */

arch/arm/mach-bcm/bcm_5301x.c

@@ -0,0 +1,50 @@
+/*
+ * Broadcom BCM470X / BCM5301X ARM platform code.
+ *
+ * Copyright 2013 Hauke Mehrtens <hauke@hauke-m.de>
+ *
+ * Licensed under the GNU/GPL. See COPYING for details.
+ */
+#include <linux/of_platform.h>
+#include <asm/hardware/cache-l2x0.h>
+
+#include <asm/mach/arch.h>
+#include <asm/siginfo.h>
+#include <asm/signal.h>
+
+#define FSR_EXTERNAL		(1 << 12)
+#define FSR_READ		(0 << 10)
+#define FSR_IMPRECISE		0x0406
+
+static const char *const bcm5301x_dt_compat[] __initconst = {
+	"brcm,bcm4708",
+	NULL,
+};
+
+static int bcm5301x_abort_handler(unsigned long addr, unsigned int fsr,
+				  struct pt_regs *regs)
+{
+	/*
+	 * We want to ignore aborts forwarded from the PCIe bus that are
+	 * expected and shouldn't really be passed by the PCIe controller.
+	 * The biggest disadvantage is the same FSR code may be reported when
+	 * reading non-existing APB register and we shouldn't ignore that.
+	 */
+	if (fsr == (FSR_EXTERNAL | FSR_READ | FSR_IMPRECISE))
+		return 0;
+
+	return 1;
+}
+
+static void __init bcm5301x_init_early(void)
+{
+	hook_fault_code(16 + 6, bcm5301x_abort_handler, SIGBUS, BUS_OBJERR,
+			"imprecise external abort");
+}
+
+DT_MACHINE_START(BCM5301X, "BCM5301X")
+	.l2c_aux_val	= 0,
+	.l2c_aux_mask	= ~0,
+	.dt_compat	= bcm5301x_dt_compat,
+	.init_early	= bcm5301x_init_early,
+MACHINE_END

arch/arm/mach-bcm/bcm_cygnus.c

@@ -0,0 +1,25 @@
+/*
+ * Copyright (C) 2014 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <asm/mach/arch.h>
+
+static const char * const bcm_cygnus_dt_compat[] __initconst = {
+	"brcm,cygnus",
+	NULL,
+};
+
+DT_MACHINE_START(BCM_CYGNUS_DT, "Broadcom Cygnus SoC")
+	.l2c_aux_val	= 0,
+	.l2c_aux_mask	= ~0,
+	.dt_compat = bcm_cygnus_dt_compat,
+MACHINE_END

arch/arm/mach-bcm/bcm_kona_smc.c

@@ -0,0 +1,182 @@
+/*
+ * Copyright (C) 2013 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <stdarg.h>
+#include <linux/smp.h>
+#include <linux/io.h>
+#include <linux/ioport.h>
+
+#include <asm/cacheflush.h>
+#include <linux/of_address.h>
+
+#include "bcm_kona_smc.h"
+
+static u32		bcm_smc_buffer_phys;	/* physical address */
+static void __iomem	*bcm_smc_buffer;	/* virtual address */
+
+struct bcm_kona_smc_data {
+	unsigned service_id;
+	unsigned arg0;
+	unsigned arg1;
+	unsigned arg2;
+	unsigned arg3;
+	unsigned result;
+};
+
+static const struct of_device_id bcm_kona_smc_ids[] __initconst = {
+	{.compatible = "brcm,kona-smc"},
+	{.compatible = "bcm,kona-smc"}, /* deprecated name */
+	{},
+};
+
+/* Map in the args buffer area */
+int __init bcm_kona_smc_init(void)
+{
+	struct device_node *node;
+	const __be32 *prop_val;
+	u64 prop_size = 0;
+	unsigned long buffer_size;
+	u32 buffer_phys;
+
+	/* Read buffer addr and size from the device tree node */
+	node = of_find_matching_node(NULL, bcm_kona_smc_ids);
+	if (!node)
+		return -ENODEV;
+
+	prop_val = of_get_address(node, 0, &prop_size, NULL);
+	if (!prop_val)
+		return -EINVAL;
+
+	/* We assume space for four 32-bit arguments */
+	if (prop_size < 4 * sizeof(u32) || prop_size > (u64)ULONG_MAX)
+		return -EINVAL;
+	buffer_size = (unsigned long)prop_size;
+
+	buffer_phys = be32_to_cpup(prop_val);
+	if (!buffer_phys)
+		return -EINVAL;
+
+	bcm_smc_buffer = ioremap(buffer_phys, buffer_size);
+	if (!bcm_smc_buffer)
+		return -ENOMEM;
+	bcm_smc_buffer_phys = buffer_phys;
+
+	pr_info("Kona Secure API initialized\n");
+
+	return 0;
+}
+
+/*
+ * int bcm_kona_do_smc(u32 service_id, u32 buffer_addr)
+ *
+ * Only core 0 can run the secure monitor code.  If an "smc" request
+ * is initiated on a different core it must be redirected to core 0
+ * for execution.  We rely on the caller to handle this.
+ *
+ * Each "smc" request supplies a service id and the address of a
+ * buffer containing parameters related to the service to be
+ * performed.  A flags value defines the behavior of the level 2
+ * cache and interrupt handling while the secure monitor executes.
+ *
+ * Parameters to the "smc" request are passed in r4-r6 as follows:
+ *     r4	service id
+ *     r5	flags (SEC_ROM_*)
+ *     r6	physical address of buffer with other parameters
+ *
+ * Execution of an "smc" request produces two distinct results.
+ *
+ * First, the secure monitor call itself (regardless of the specific
+ * service request) can succeed, or can produce an error.  When an
+ * "smc" request completes this value is found in r12; it should
+ * always be SEC_EXIT_NORMAL.
+ *
+ * In addition, the particular service performed produces a result.
+ * The values that should be expected depend on the service.  We
+ * therefore return this value to the caller, so it can handle the
+ * request result appropriately.  This result value is found in r0
+ * when the "smc" request completes.
+ */
+static int bcm_kona_do_smc(u32 service_id, u32 buffer_phys)
+{
+	register u32 ip asm("ip");	/* Also called r12 */
+	register u32 r0 asm("r0");
+	register u32 r4 asm("r4");
+	register u32 r5 asm("r5");
+	register u32 r6 asm("r6");
+
+	r4 = service_id;
+	r5 = 0x3;		/* Keep IRQ and FIQ off in SM */
+	r6 = buffer_phys;
+
+	asm volatile (
+		/* Make sure we got the registers we want */
+		__asmeq("%0", "ip")
+		__asmeq("%1", "r0")
+		__asmeq("%2", "r4")
+		__asmeq("%3", "r5")
+		__asmeq("%4", "r6")
+#ifdef REQUIRES_SEC
+		".arch_extension sec\n"
+#endif
+		"	smc    #0\n"
+		: "=r" (ip), "=r" (r0)
+		: "r" (r4), "r" (r5), "r" (r6)
+		: "r1", "r2", "r3", "r7", "lr");
+
+	BUG_ON(ip != SEC_EXIT_NORMAL);
+
+	return r0;
+}
+
+/* __bcm_kona_smc() should only run on CPU 0, with pre-emption disabled */
+static void __bcm_kona_smc(void *info)
+{
+	struct bcm_kona_smc_data *data = info;
+	u32 *args = bcm_smc_buffer;
+
+	BUG_ON(smp_processor_id() != 0);
+	BUG_ON(!args);
+
+	/* Copy the four 32 bit argument values into the bounce area */
+	writel_relaxed(data->arg0, args++);
+	writel_relaxed(data->arg1, args++);
+	writel_relaxed(data->arg2, args++);
+	writel(data->arg3, args);
+
+	/* Flush caches for input data passed to Secure Monitor */
+	flush_cache_all();
+
+	/* Trap into Secure Monitor and record the request result */
+	data->result = bcm_kona_do_smc(data->service_id, bcm_smc_buffer_phys);
+}
+
+unsigned bcm_kona_smc(unsigned service_id, unsigned arg0, unsigned arg1,
+		  unsigned arg2, unsigned arg3)
+{
+	struct bcm_kona_smc_data data;
+
+	data.service_id = service_id;
+	data.arg0 = arg0;
+	data.arg1 = arg1;
+	data.arg2 = arg2;
+	data.arg3 = arg3;
+	data.result = 0;
+
+	/*
+	 * Due to a limitation of the secure monitor, we must use the SMP
+	 * infrastructure to forward all secure monitor calls to Core 0.
+	 */
+	smp_call_function_single(0, __bcm_kona_smc, &data, 1);
+
+	return data.result;
+}

arch/arm/mach-bcm/bcm_kona_smc.h

@@ -0,0 +1,32 @@
+/*
+ * Copyright (C) 2013 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#ifndef BCM_KONA_SMC_H
+#define BCM_KONA_SMC_H
+
+#include <linux/types.h>
+
+/* Broadcom Secure Service API service IDs, return codes, and exit codes */
+#define SSAPI_ENABLE_L2_CACHE		0x01000002
+#define SEC_ROM_RET_OK			0x00000001
+#define SEC_EXIT_NORMAL			0x1
+
+extern int __init bcm_kona_smc_init(void);
+
+extern unsigned bcm_kona_smc(unsigned service_id,
+			     unsigned arg0,
+			     unsigned arg1,
+			     unsigned arg2,
+			     unsigned arg3);
+
+#endif /* BCM_KONA_SMC_H */

arch/arm/mach-bcm/bcm_nsp.c

@@ -0,0 +1,25 @@
+/*
+ * Copyright (C) 2015 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <asm/mach/arch.h>
+
+static const char *const bcm_nsp_dt_compat[] __initconst = {
+	"brcm,nsp",
+	NULL,
+};
+
+DT_MACHINE_START(NSP_DT, "Broadcom Northstar Plus SoC")
+	.l2c_aux_val	= 0,
+	.l2c_aux_mask	= ~0,
+	.dt_compat = bcm_nsp_dt_compat,
+MACHINE_END

arch/arm/mach-bcm/board_bcm21664.c

@@ -0,0 +1,31 @@
+/*
+ * Copyright (C) 2014 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <asm/mach/arch.h>
+
+#include "kona_l2_cache.h"
+
+static void __init bcm21664_init(void)
+{
+	kona_l2_cache_init();
+}
+
+static const char * const bcm21664_dt_compat[] = {
+	"brcm,bcm21664",
+	NULL,
+};
+
+DT_MACHINE_START(BCM21664_DT, "BCM21664 Broadcom Application Processor")
+	.init_machine = bcm21664_init,
+	.dt_compat = bcm21664_dt_compat,
+MACHINE_END

arch/arm/mach-bcm/board_bcm23550.c

@@ -0,0 +1,25 @@
+/*
+ * Copyright (C) 2016 Broadcom
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/of_platform.h>
+
+#include <asm/mach/arch.h>
+
+static const char * const bcm23550_dt_compat[] = {
+	"brcm,bcm23550",
+	NULL,
+};
+
+DT_MACHINE_START(BCM23550_DT, "BCM23550 Broadcom Application Processor")
+	.dt_compat = bcm23550_dt_compat,
+MACHINE_END

arch/arm/mach-bcm/board_bcm281xx.c

@@ -0,0 +1,72 @@
+/*
+ * Copyright (C) 2012-2014 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/clocksource.h>
+#include <linux/of_address.h>
+
+#include <asm/mach/arch.h>
+
+#include "kona_l2_cache.h"
+
+#define SECWDOG_OFFSET			0x00000000
+#define SECWDOG_RESERVED_MASK		0xe2000000
+#define SECWDOG_WD_LOAD_FLAG_MASK	0x10000000
+#define SECWDOG_EN_MASK			0x08000000
+#define SECWDOG_SRSTEN_MASK		0x04000000
+#define SECWDOG_CLKS_SHIFT		20
+#define SECWDOG_COUNT_SHIFT		0
+
+static void bcm281xx_restart(enum reboot_mode mode, const char *cmd)
+{
+	uint32_t val;
+	void __iomem *base;
+	struct device_node *np_wdog;
+
+	np_wdog = of_find_compatible_node(NULL, NULL, "brcm,kona-wdt");
+	if (!np_wdog) {
+		pr_emerg("Couldn't find brcm,kona-wdt\n");
+		return;
+	}
+	base = of_iomap(np_wdog, 0);
+	if (!base) {
+		pr_emerg("Couldn't map brcm,kona-wdt\n");
+		return;
+	}
+
+	/* Enable watchdog with short timeout (244us). */
+	val = readl(base + SECWDOG_OFFSET);
+	val &= SECWDOG_RESERVED_MASK | SECWDOG_WD_LOAD_FLAG_MASK;
+	val |= SECWDOG_EN_MASK | SECWDOG_SRSTEN_MASK |
+		(0x15 << SECWDOG_CLKS_SHIFT) |
+		(0x8 << SECWDOG_COUNT_SHIFT);
+	writel(val, base + SECWDOG_OFFSET);
+
+	/* Wait for reset */
+	while (1);
+}
+
+static void __init bcm281xx_init(void)
+{
+	kona_l2_cache_init();
+}
+
+static const char * const bcm281xx_dt_compat[] = {
+	"brcm,bcm11351",	/* Have to use the first number upstreamed */
+	NULL,
+};
+
+DT_MACHINE_START(BCM281XX_DT, "BCM281xx Broadcom Application Processor")
+	.init_machine = bcm281xx_init,
+	.restart = bcm281xx_restart,
+	.dt_compat = bcm281xx_dt_compat,
+MACHINE_END

arch/arm/mach-bcm/board_bcm2835.c

@@ -0,0 +1,41 @@
+/*
+ * Copyright (C) 2010 Broadcom
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/init.h>
+#include <linux/irqchip.h>
+#include <linux/of_address.h>
+#include <linux/clk/bcm2835.h>
+
+#include <asm/mach/arch.h>
+#include <asm/mach/map.h>
+
+static void __init bcm2835_init(void)
+{
+	bcm2835_init_clocks();
+}
+
+static const char * const bcm2835_compat[] = {
+#ifdef CONFIG_ARCH_MULTI_V6
+	"brcm,bcm2835",
+#endif
+#ifdef CONFIG_ARCH_MULTI_V7
+	"brcm,bcm2836",
+#endif
+	NULL
+};
+
+DT_MACHINE_START(BCM2835, "BCM2835")
+	.init_machine = bcm2835_init,
+	.dt_compat = bcm2835_compat
+MACHINE_END

arch/arm/mach-bcm/brcmstb.c

@@ -0,0 +1,53 @@
+/*
+ * Copyright (C) 2013-2014 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/init.h>
+#include <linux/irqchip.h>
+#include <linux/of_platform.h>
+#include <linux/soc/brcmstb/brcmstb.h>
+
+#include <asm/mach-types.h>
+#include <asm/mach/arch.h>
+
+/*
+ * Storage for debug-macro.S's state.
+ *
+ * This must be in .data not .bss so that it gets initialized each time the
+ * kernel is loaded. The data is declared here rather than debug-macro.S so
+ * that multiple inclusions of debug-macro.S point at the same data.
+ */
+u32 brcmstb_uart_config[3] = {
+	/* Debug UART initialization required */
+	1,
+	/* Debug UART physical address */
+	0,
+	/* Debug UART virtual address */
+	0,
+};
+
+static void __init brcmstb_init_irq(void)
+{
+	irqchip_init();
+	brcmstb_biuctrl_init();
+}
+
+static const char *const brcmstb_match[] __initconst = {
+	"brcm,bcm7445",
+	"brcm,brcmstb",
+	NULL
+};
+
+DT_MACHINE_START(BRCMSTB, "Broadcom STB (Flattened Device Tree)")
+	.dt_compat	= brcmstb_match,
+	.init_irq	= brcmstb_init_irq,
+MACHINE_END

arch/arm/mach-bcm/kona_l2_cache.c

@@ -0,0 +1,48 @@
+/*
+ * Copyright (C) 2012-2014 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+
+#include <linux/init.h>
+#include <linux/printk.h>
+#include <asm/hardware/cache-l2x0.h>
+
+#include "bcm_kona_smc.h"
+#include "kona_l2_cache.h"
+
+void __init kona_l2_cache_init(void)
+{
+	unsigned int result;
+	int ret;
+
+	ret = bcm_kona_smc_init();
+	if (ret) {
+		pr_info("Secure API not available (%d). Skipping L2 init.\n",
+			ret);
+		return;
+	}
+
+	result = bcm_kona_smc(SSAPI_ENABLE_L2_CACHE, 0, 0, 0, 0);
+	if (result != SEC_ROM_RET_OK) {
+		pr_err("Secure Monitor call failed (%u)! Skipping L2 init.\n",
+			result);
+		return;
+	}
+
+	/*
+	 * The aux_val and aux_mask have no effect since L2 cache is already
+	 * enabled.  Pass 0s for aux_val and 1s for aux_mask for default value.
+	 */
+	ret = l2x0_of_init(0, ~0);
+	if (ret)
+		pr_err("Couldn't enable L2 cache: %d\n", ret);
+}

arch/arm/mach-bcm/kona_l2_cache.h

@@ -0,0 +1,18 @@
+/*
+ * Copyright (C) 2012-2014 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#ifdef CONFIG_ARCH_BCM_MOBILE_L2_CACHE
+void	kona_l2_cache_init(void);
+#else
+#define kona_l2_cache_init() ((void)0)
+#endif

arch/arm/mach-bcm/platsmp-brcmstb.c

@@ -0,0 +1,368 @@
+/*
+ * Broadcom STB CPU SMP and hotplug support for ARM
+ *
+ * Copyright (C) 2013-2014 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/delay.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/io.h>
+#include <linux/jiffies.h>
+#include <linux/of_address.h>
+#include <linux/of_platform.h>
+#include <linux/printk.h>
+#include <linux/regmap.h>
+#include <linux/smp.h>
+#include <linux/mfd/syscon.h>
+
+#include <asm/cacheflush.h>
+#include <asm/cp15.h>
+#include <asm/mach-types.h>
+#include <asm/smp_plat.h>
+
+enum {
+	ZONE_MAN_CLKEN_MASK		= BIT(0),
+	ZONE_MAN_RESET_CNTL_MASK	= BIT(1),
+	ZONE_MAN_MEM_PWR_MASK		= BIT(4),
+	ZONE_RESERVED_1_MASK		= BIT(5),
+	ZONE_MAN_ISO_CNTL_MASK		= BIT(6),
+	ZONE_MANUAL_CONTROL_MASK	= BIT(7),
+	ZONE_PWR_DN_REQ_MASK		= BIT(9),
+	ZONE_PWR_UP_REQ_MASK		= BIT(10),
+	ZONE_BLK_RST_ASSERT_MASK	= BIT(12),
+	ZONE_PWR_OFF_STATE_MASK		= BIT(25),
+	ZONE_PWR_ON_STATE_MASK		= BIT(26),
+	ZONE_DPG_PWR_STATE_MASK		= BIT(28),
+	ZONE_MEM_PWR_STATE_MASK		= BIT(29),
+	ZONE_RESET_STATE_MASK		= BIT(31),
+	CPU0_PWR_ZONE_CTRL_REG		= 1,
+	CPU_RESET_CONFIG_REG		= 2,
+};
+
+static void __iomem *cpubiuctrl_block;
+static void __iomem *hif_cont_block;
+static u32 cpu0_pwr_zone_ctrl_reg;
+static u32 cpu_rst_cfg_reg;
+static u32 hif_cont_reg;
+
+#ifdef CONFIG_HOTPLUG_CPU
+/*
+ * We must quiesce a dying CPU before it can be killed by the boot CPU. Because
+ * one or more cache may be disabled, we must flush to ensure coherency. We
+ * cannot use traditionl completion structures or spinlocks as they rely on
+ * coherency.
+ */
+static DEFINE_PER_CPU_ALIGNED(int, per_cpu_sw_state);
+
+static int per_cpu_sw_state_rd(u32 cpu)
+{
+	sync_cache_r(SHIFT_PERCPU_PTR(&per_cpu_sw_state, per_cpu_offset(cpu)));
+	return per_cpu(per_cpu_sw_state, cpu);
+}
+
+static void per_cpu_sw_state_wr(u32 cpu, int val)
+{
+	dmb();
+	per_cpu(per_cpu_sw_state, cpu) = val;
+	sync_cache_w(SHIFT_PERCPU_PTR(&per_cpu_sw_state, per_cpu_offset(cpu)));
+}
+#else
+static inline void per_cpu_sw_state_wr(u32 cpu, int val) { }
+#endif
+
+static void __iomem *pwr_ctrl_get_base(u32 cpu)
+{
+	void __iomem *base = cpubiuctrl_block + cpu0_pwr_zone_ctrl_reg;
+	base += (cpu_logical_map(cpu) * 4);
+	return base;
+}
+
+static u32 pwr_ctrl_rd(u32 cpu)
+{
+	void __iomem *base = pwr_ctrl_get_base(cpu);
+	return readl_relaxed(base);
+}
+
+static void pwr_ctrl_set(unsigned int cpu, u32 val, u32 mask)
+{
+	void __iomem *base = pwr_ctrl_get_base(cpu);
+	writel((readl(base) & mask) | val, base);
+}
+
+static void pwr_ctrl_clr(unsigned int cpu, u32 val, u32 mask)
+{
+	void __iomem *base = pwr_ctrl_get_base(cpu);
+	writel((readl(base) & mask) & ~val, base);
+}
+
+#define POLL_TMOUT_MS 500
+static int pwr_ctrl_wait_tmout(unsigned int cpu, u32 set, u32 mask)
+{
+	const unsigned long timeo = jiffies + msecs_to_jiffies(POLL_TMOUT_MS);
+	u32 tmp;
+
+	do {
+		tmp = pwr_ctrl_rd(cpu) & mask;
+		if (!set == !tmp)
+			return 0;
+	} while (time_before(jiffies, timeo));
+
+	tmp = pwr_ctrl_rd(cpu) & mask;
+	if (!set == !tmp)
+		return 0;
+
+	return -ETIMEDOUT;
+}
+
+static void cpu_rst_cfg_set(u32 cpu, int set)
+{
+	u32 val;
+	val = readl_relaxed(cpubiuctrl_block + cpu_rst_cfg_reg);
+	if (set)
+		val |= BIT(cpu_logical_map(cpu));
+	else
+		val &= ~BIT(cpu_logical_map(cpu));
+	writel_relaxed(val, cpubiuctrl_block + cpu_rst_cfg_reg);
+}
+
+static void cpu_set_boot_addr(u32 cpu, unsigned long boot_addr)
+{
+	const int reg_ofs = cpu_logical_map(cpu) * 8;
+	writel_relaxed(0, hif_cont_block + hif_cont_reg + reg_ofs);
+	writel_relaxed(boot_addr, hif_cont_block + hif_cont_reg + 4 + reg_ofs);
+}
+
+static void brcmstb_cpu_boot(u32 cpu)
+{
+	/* Mark this CPU as "up" */
+	per_cpu_sw_state_wr(cpu, 1);
+
+	/*
+	 * Set the reset vector to point to the secondary_startup
+	 * routine
+	 */
+	cpu_set_boot_addr(cpu, virt_to_phys(secondary_startup));
+
+	/* Unhalt the cpu */
+	cpu_rst_cfg_set(cpu, 0);
+}
+
+static void brcmstb_cpu_power_on(u32 cpu)
+{
+	/*
+	 * The secondary cores power was cut, so we must go through
+	 * power-on initialization.
+	 */
+	pwr_ctrl_set(cpu, ZONE_MAN_ISO_CNTL_MASK, 0xffffff00);
+	pwr_ctrl_set(cpu, ZONE_MANUAL_CONTROL_MASK, -1);
+	pwr_ctrl_set(cpu, ZONE_RESERVED_1_MASK, -1);
+
+	pwr_ctrl_set(cpu, ZONE_MAN_MEM_PWR_MASK, -1);
+
+	if (pwr_ctrl_wait_tmout(cpu, 1, ZONE_MEM_PWR_STATE_MASK))
+		panic("ZONE_MEM_PWR_STATE_MASK set timeout");
+
+	pwr_ctrl_set(cpu, ZONE_MAN_CLKEN_MASK, -1);
+
+	if (pwr_ctrl_wait_tmout(cpu, 1, ZONE_DPG_PWR_STATE_MASK))
+		panic("ZONE_DPG_PWR_STATE_MASK set timeout");
+
+	pwr_ctrl_clr(cpu, ZONE_MAN_ISO_CNTL_MASK, -1);
+	pwr_ctrl_set(cpu, ZONE_MAN_RESET_CNTL_MASK, -1);
+}
+
+static int brcmstb_cpu_get_power_state(u32 cpu)
+{
+	int tmp = pwr_ctrl_rd(cpu);
+	return (tmp & ZONE_RESET_STATE_MASK) ? 0 : 1;
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+
+static void brcmstb_cpu_die(u32 cpu)
+{
+	v7_exit_coherency_flush(all);
+
+	per_cpu_sw_state_wr(cpu, 0);
+
+	/* Sit and wait to die */
+	wfi();
+
+	/* We should never get here... */
+	while (1)
+		;
+}
+
+static int brcmstb_cpu_kill(u32 cpu)
+{
+	/*
+	 * Ordinarily, the hardware forbids power-down of CPU0 (which is good
+	 * because it is the boot CPU), but this is not true when using BPCM
+	 * manual mode.  Consequently, we must avoid turning off CPU0 here to
+	 * ensure that TI2C master reset will work.
+	 */
+	if (cpu == 0) {
+		pr_warn("SMP: refusing to power off CPU0\n");
+		return 1;
+	}
+
+	while (per_cpu_sw_state_rd(cpu))
+		;
+
+	pwr_ctrl_set(cpu, ZONE_MANUAL_CONTROL_MASK, -1);
+	pwr_ctrl_clr(cpu, ZONE_MAN_RESET_CNTL_MASK, -1);
+	pwr_ctrl_clr(cpu, ZONE_MAN_CLKEN_MASK, -1);
+	pwr_ctrl_set(cpu, ZONE_MAN_ISO_CNTL_MASK, -1);
+	pwr_ctrl_clr(cpu, ZONE_MAN_MEM_PWR_MASK, -1);
+
+	if (pwr_ctrl_wait_tmout(cpu, 0, ZONE_MEM_PWR_STATE_MASK))
+		panic("ZONE_MEM_PWR_STATE_MASK clear timeout");
+
+	pwr_ctrl_clr(cpu, ZONE_RESERVED_1_MASK, -1);
+
+	if (pwr_ctrl_wait_tmout(cpu, 0, ZONE_DPG_PWR_STATE_MASK))
+		panic("ZONE_DPG_PWR_STATE_MASK clear timeout");
+
+	/* Flush pipeline before resetting CPU */
+	mb();
+
+	/* Assert reset on the CPU */
+	cpu_rst_cfg_set(cpu, 1);
+
+	return 1;
+}
+
+#endif /* CONFIG_HOTPLUG_CPU */
+
+static int __init setup_hifcpubiuctrl_regs(struct device_node *np)
+{
+	int rc = 0;
+	char *name;
+	struct device_node *syscon_np = NULL;
+
+	name = "syscon-cpu";
+
+	syscon_np = of_parse_phandle(np, name, 0);
+	if (!syscon_np) {
+		pr_err("can't find phandle %s\n", name);
+		rc = -EINVAL;
+		goto cleanup;
+	}
+
+	cpubiuctrl_block = of_iomap(syscon_np, 0);
+	if (!cpubiuctrl_block) {
+		pr_err("iomap failed for cpubiuctrl_block\n");
+		rc = -EINVAL;
+		goto cleanup;
+	}
+
+	rc = of_property_read_u32_index(np, name, CPU0_PWR_ZONE_CTRL_REG,
+					&cpu0_pwr_zone_ctrl_reg);
+	if (rc) {
+		pr_err("failed to read 1st entry from %s property (%d)\n", name,
+			rc);
+		rc = -EINVAL;
+		goto cleanup;
+	}
+
+	rc = of_property_read_u32_index(np, name, CPU_RESET_CONFIG_REG,
+					&cpu_rst_cfg_reg);
+	if (rc) {
+		pr_err("failed to read 2nd entry from %s property (%d)\n", name,
+			rc);
+		rc = -EINVAL;
+		goto cleanup;
+	}
+
+cleanup:
+	of_node_put(syscon_np);
+	return rc;
+}
+
+static int __init setup_hifcont_regs(struct device_node *np)
+{
+	int rc = 0;
+	char *name;
+	struct device_node *syscon_np = NULL;
+
+	name = "syscon-cont";
+
+	syscon_np = of_parse_phandle(np, name, 0);
+	if (!syscon_np) {
+		pr_err("can't find phandle %s\n", name);
+		rc = -EINVAL;
+		goto cleanup;
+	}
+
+	hif_cont_block = of_iomap(syscon_np, 0);
+	if (!hif_cont_block) {
+		pr_err("iomap failed for hif_cont_block\n");
+		rc = -EINVAL;
+		goto cleanup;
+	}
+
+	/* Offset is at top of hif_cont_block */
+	hif_cont_reg = 0;
+
+cleanup:
+	of_node_put(syscon_np);
+	return rc;
+}
+
+static void __init brcmstb_cpu_ctrl_setup(unsigned int max_cpus)
+{
+	int rc;
+	struct device_node *np;
+	char *name;
+
+	name = "brcm,brcmstb-smpboot";
+	np = of_find_compatible_node(NULL, NULL, name);
+	if (!np) {
+		pr_err("can't find compatible node %s\n", name);
+		return;
+	}
+
+	rc = setup_hifcpubiuctrl_regs(np);
+	if (rc)
+		return;
+
+	rc = setup_hifcont_regs(np);
+	if (rc)
+		return;
+}
+
+static int brcmstb_boot_secondary(unsigned int cpu, struct task_struct *idle)
+{
+	/* Missing the brcm,brcmstb-smpboot DT node? */
+	if (!cpubiuctrl_block || !hif_cont_block)
+		return -ENODEV;
+
+	/* Bring up power to the core if necessary */
+	if (brcmstb_cpu_get_power_state(cpu) == 0)
+		brcmstb_cpu_power_on(cpu);
+
+	brcmstb_cpu_boot(cpu);
+
+	return 0;
+}
+
+static const struct smp_operations brcmstb_smp_ops __initconst = {
+	.smp_prepare_cpus	= brcmstb_cpu_ctrl_setup,
+	.smp_boot_secondary	= brcmstb_boot_secondary,
+#ifdef CONFIG_HOTPLUG_CPU
+	.cpu_kill		= brcmstb_cpu_kill,
+	.cpu_die		= brcmstb_cpu_die,
+#endif
+};
+
+CPU_METHOD_OF_DECLARE(brcmstb_smp, "brcm,brahma-b15", &brcmstb_smp_ops);

arch/arm/mach-bcm/platsmp.c

@@ -0,0 +1,306 @@
+/*
+ * Copyright (C) 2014-2015 Broadcom Corporation
+ * Copyright 2014 Linaro Limited
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/cpumask.h>
+#include <linux/delay.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/io.h>
+#include <linux/jiffies.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/sched.h>
+#include <linux/smp.h>
+
+#include <asm/cacheflush.h>
+#include <asm/smp.h>
+#include <asm/smp_plat.h>
+#include <asm/smp_scu.h>
+
+/* Size of mapped Cortex A9 SCU address space */
+#define CORTEX_A9_SCU_SIZE	0x58
+
+#define SECONDARY_TIMEOUT_NS	NSEC_PER_MSEC	/* 1 msec (in nanoseconds) */
+#define BOOT_ADDR_CPUID_MASK	0x3
+
+/* Name of device node property defining secondary boot register location */
+#define OF_SECONDARY_BOOT	"secondary-boot-reg"
+#define MPIDR_CPUID_BITMASK	0x3
+
+/*
+ * Enable the Cortex A9 Snoop Control Unit
+ *
+ * By the time this is called we already know there are multiple
+ * cores present.  We assume we're running on a Cortex A9 processor,
+ * so any trouble getting the base address register or getting the
+ * SCU base is a problem.
+ *
+ * Return 0 if successful or an error code otherwise.
+ */
+static int __init scu_a9_enable(void)
+{
+	unsigned long config_base;
+	void __iomem *scu_base;
+
+	if (!scu_a9_has_base()) {
+		pr_err("no configuration base address register!\n");
+		return -ENXIO;
+	}
+
+	/* Config base address register value is zero for uniprocessor */
+	config_base = scu_a9_get_base();
+	if (!config_base) {
+		pr_err("hardware reports only one core\n");
+		return -ENOENT;
+	}
+
+	scu_base = ioremap((phys_addr_t)config_base, CORTEX_A9_SCU_SIZE);
+	if (!scu_base) {
+		pr_err("failed to remap config base (%lu/%u) for SCU\n",
+			config_base, CORTEX_A9_SCU_SIZE);
+		return -ENOMEM;
+	}
+
+	scu_enable(scu_base);
+
+	iounmap(scu_base);	/* That's the last we'll need of this */
+
+	return 0;
+}
+
+static u32 secondary_boot_addr_for(unsigned int cpu)
+{
+	u32 secondary_boot_addr = 0;
+	struct device_node *cpu_node = of_get_cpu_node(cpu, NULL);
+
+        if (!cpu_node) {
+		pr_err("Failed to find device tree node for CPU%u\n", cpu);
+		return 0;
+	}
+
+	if (of_property_read_u32(cpu_node,
+				 OF_SECONDARY_BOOT,
+				 &secondary_boot_addr))
+		pr_err("required secondary boot register not specified for CPU%u\n",
+			cpu);
+
+	of_node_put(cpu_node);
+
+	return secondary_boot_addr;
+}
+
+static int nsp_write_lut(unsigned int cpu)
+{
+	void __iomem *sku_rom_lut;
+	phys_addr_t secondary_startup_phy;
+	const u32 secondary_boot_addr = secondary_boot_addr_for(cpu);
+
+	if (!secondary_boot_addr)
+		return -EINVAL;
+
+	sku_rom_lut = ioremap_nocache((phys_addr_t)secondary_boot_addr,
+				      sizeof(phys_addr_t));
+	if (!sku_rom_lut) {
+		pr_warn("unable to ioremap SKU-ROM LUT register for cpu %u\n", cpu);
+		return -ENOMEM;
+	}
+
+	secondary_startup_phy = virt_to_phys(secondary_startup);
+	BUG_ON(secondary_startup_phy > (phys_addr_t)U32_MAX);
+
+	writel_relaxed(secondary_startup_phy, sku_rom_lut);
+
+	/* Ensure the write is visible to the secondary core */
+	smp_wmb();
+
+	iounmap(sku_rom_lut);
+
+	return 0;
+}
+
+static void __init bcm_smp_prepare_cpus(unsigned int max_cpus)
+{
+	const cpumask_t only_cpu_0 = { CPU_BITS_CPU0 };
+
+	/* Enable the SCU on Cortex A9 based SoCs */
+	if (scu_a9_enable()) {
+		/* Update the CPU present map to reflect uniprocessor mode */
+		pr_warn("failed to enable A9 SCU - disabling SMP\n");
+		init_cpu_present(&only_cpu_0);
+	}
+}
+
+/*
+ * The ROM code has the secondary cores looping, waiting for an event.
+ * When an event occurs each core examines the bottom two bits of the
+ * secondary boot register.  When a core finds those bits contain its
+ * own core id, it performs initialization, including computing its boot
+ * address by clearing the boot register value's bottom two bits.  The
+ * core signals that it is beginning its execution by writing its boot
+ * address back to the secondary boot register, and finally jumps to
+ * that address.
+ *
+ * So to start a core executing we need to:
+ * - Encode the (hardware) CPU id with the bottom bits of the secondary
+ *   start address.
+ * - Write that value into the secondary boot register.
+ * - Generate an event to wake up the secondary CPU(s).
+ * - Wait for the secondary boot register to be re-written, which
+ *   indicates the secondary core has started.
+ */
+static int kona_boot_secondary(unsigned int cpu, struct task_struct *idle)
+{
+	void __iomem *boot_reg;
+	phys_addr_t boot_func;
+	u64 start_clock;
+	u32 cpu_id;
+	u32 boot_val;
+	bool timeout = false;
+	const u32 secondary_boot_addr = secondary_boot_addr_for(cpu);
+
+	cpu_id = cpu_logical_map(cpu);
+	if (cpu_id & ~BOOT_ADDR_CPUID_MASK) {
+		pr_err("bad cpu id (%u > %u)\n", cpu_id, BOOT_ADDR_CPUID_MASK);
+		return -EINVAL;
+	}
+
+	if (!secondary_boot_addr)
+		return -EINVAL;
+
+	boot_reg = ioremap_nocache((phys_addr_t)secondary_boot_addr,
+				   sizeof(phys_addr_t));
+	if (!boot_reg) {
+		pr_err("unable to map boot register for cpu %u\n", cpu_id);
+		return -ENOMEM;
+	}
+
+	/*
+	 * Secondary cores will start in secondary_startup(),
+	 * defined in "arch/arm/kernel/head.S"
+	 */
+	boot_func = virt_to_phys(secondary_startup);
+	BUG_ON(boot_func & BOOT_ADDR_CPUID_MASK);
+	BUG_ON(boot_func > (phys_addr_t)U32_MAX);
+
+	/* The core to start is encoded in the low bits */
+	boot_val = (u32)boot_func | cpu_id;
+	writel_relaxed(boot_val, boot_reg);
+
+	sev();
+
+	/* The low bits will be cleared once the core has started */
+	start_clock = local_clock();
+	while (!timeout && readl_relaxed(boot_reg) == boot_val)
+		timeout = local_clock() - start_clock > SECONDARY_TIMEOUT_NS;
+
+	iounmap(boot_reg);
+
+	if (!timeout)
+		return 0;
+
+	pr_err("timeout waiting for cpu %u to start\n", cpu_id);
+
+	return -ENXIO;
+}
+
+/* Cluster Dormant Control command to bring CPU into a running state */
+#define CDC_CMD			6
+#define CDC_CMD_OFFSET		0
+#define CDC_CMD_REG(cpu)	(CDC_CMD_OFFSET + 4*(cpu))
+
+/*
+ * BCM23550 has a Cluster Dormant Control block that keeps the core in
+ * idle state. A command needs to be sent to the block to bring the CPU
+ * into running state.
+ */
+static int bcm23550_boot_secondary(unsigned int cpu, struct task_struct *idle)
+{
+	void __iomem *cdc_base;
+	struct device_node *dn;
+	char *name;
+	int ret;
+
+	/* Make sure a CDC node exists before booting the
+	 * secondary core.
+	 */
+	name = "brcm,bcm23550-cdc";
+	dn = of_find_compatible_node(NULL, NULL, name);
+	if (!dn) {
+		pr_err("unable to find cdc node\n");
+		return -ENODEV;
+	}
+
+	cdc_base = of_iomap(dn, 0);
+	of_node_put(dn);
+
+	if (!cdc_base) {
+		pr_err("unable to remap cdc base register\n");
+		return -ENOMEM;
+	}
+
+	/* Boot the secondary core */
+	ret = kona_boot_secondary(cpu, idle);
+	if (ret)
+		goto out;
+
+	/* Bring this CPU to RUN state so that nIRQ nFIQ
+	 * signals are unblocked.
+	 */
+	writel_relaxed(CDC_CMD, cdc_base + CDC_CMD_REG(cpu));
+
+out:
+	iounmap(cdc_base);
+
+	return ret;
+}
+
+static int nsp_boot_secondary(unsigned int cpu, struct task_struct *idle)
+{
+	int ret;
+
+	/*
+	 * After wake up, secondary core branches to the startup
+	 * address programmed at SKU ROM LUT location.
+	 */
+	ret = nsp_write_lut(cpu);
+	if (ret) {
+		pr_err("unable to write startup addr to SKU ROM LUT\n");
+		goto out;
+	}
+
+	/* Send a CPU wakeup interrupt to the secondary core */
+	arch_send_wakeup_ipi_mask(cpumask_of(cpu));
+
+out:
+	return ret;
+}
+
+static const struct smp_operations kona_smp_ops __initconst = {
+	.smp_prepare_cpus	= bcm_smp_prepare_cpus,
+	.smp_boot_secondary	= kona_boot_secondary,
+};
+CPU_METHOD_OF_DECLARE(bcm_smp_bcm281xx, "brcm,bcm11351-cpu-method",
+			&kona_smp_ops);
+
+static const struct smp_operations bcm23550_smp_ops __initconst = {
+	.smp_boot_secondary	= bcm23550_boot_secondary,
+};
+CPU_METHOD_OF_DECLARE(bcm_smp_bcm23550, "brcm,bcm23550",
+			&bcm23550_smp_ops);
+
+static const struct smp_operations nsp_smp_ops __initconst = {
+	.smp_prepare_cpus	= bcm_smp_prepare_cpus,
+	.smp_boot_secondary	= nsp_boot_secondary,
+};
+CPU_METHOD_OF_DECLARE(bcm_smp_nsp, "brcm,bcm-nsp-smp", &nsp_smp_ops);

arch/arm/mach-berlin/Kconfig

@@ -0,0 +1,37 @@
+menuconfig ARCH_BERLIN
+	bool "Marvell Berlin SoCs"
+	depends on ARCH_MULTI_V7
+	select ARCH_HAS_RESET_CONTROLLER
+	select ARM_GIC
+	select DW_APB_ICTL
+	select DW_APB_TIMER_OF
+	select GENERIC_IRQ_CHIP
+	select GPIOLIB
+	select MFD_SYSCON
+	select PINCTRL
+
+if ARCH_BERLIN
+
+config MACH_BERLIN_BG2
+	bool "Marvell Armada 1500 (BG2)"
+	select CACHE_L2X0
+	select CPU_PJ4B
+	select HAVE_ARM_SCU if SMP
+	select HAVE_ARM_TWD if SMP
+	select HAVE_SMP
+	select PINCTRL_BERLIN_BG2
+
+config MACH_BERLIN_BG2CD
+	bool "Marvell Armada 1500-mini (BG2CD)"
+	select CACHE_L2X0
+	select HAVE_ARM_TWD if SMP
+	select PINCTRL_BERLIN_BG2CD
+
+config MACH_BERLIN_BG2Q
+	bool "Marvell Armada 1500 Pro (BG2-Q)"
+	select CACHE_L2X0
+	select HAVE_ARM_SCU if SMP
+	select HAVE_ARM_TWD if SMP
+	select PINCTRL_BERLIN_BG2Q
+
+endif

arch/arm/mach-berlin/Makefile

@@ -0,0 +1,2 @@
+obj-y			+= berlin.o
+obj-$(CONFIG_SMP)	+= headsmp.o platsmp.o

arch/arm/mach-berlin/berlin.c

@@ -0,0 +1,34 @@
+/*
+ * Device Tree support for Marvell Berlin SoCs.
+ *
+ * Sebastian Hesselbarth <sebastian.hesselbarth@gmail.com>
+ *
+ * based on GPL'ed 2.6 kernel sources
+ *  (c) Marvell International Ltd.
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <linux/init.h>
+#include <linux/io.h>
+#include <linux/kernel.h>
+#include <linux/of_platform.h>
+#include <asm/hardware/cache-l2x0.h>
+#include <asm/mach/arch.h>
+
+static const char * const berlin_dt_compat[] = {
+	"marvell,berlin",
+	NULL,
+};
+
+DT_MACHINE_START(BERLIN_DT, "Marvell Berlin")
+	.dt_compat	= berlin_dt_compat,
+	/*
+	 * with DT probing for L2CCs, berlin_init_machine can be removed.
+	 * Note: 88DE3005 (Armada 1500-mini) uses pl310 l2cc
+	 */
+	.l2c_aux_val	= 0x30c00000,
+	.l2c_aux_mask	= 0xfeffffff,
+MACHINE_END

arch/arm/mach-berlin/headsmp.S

@@ -0,0 +1,24 @@
+/*
+ * Copyright (C) 2014 Marvell Technology Group Ltd.
+ *
+ * Antoine Ténart <antoine.tenart@free-electrons.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/linkage.h>
+#include <linux/init.h>
+#include <asm/assembler.h>
+
+/*
+ * If the following instruction is set in the reset exception vector, CPUs
+ * will fetch the value of the software reset address vector when being
+ * reset.
+ */
+.global boot_inst
+boot_inst:
+	ldr	pc, [pc, #140]
+
+	.align

arch/arm/mach-berlin/platsmp.c

@@ -0,0 +1,130 @@
+/*
+ * Copyright (C) 2014 Marvell Technology Group Ltd.
+ *
+ * Antoine Ténart <antoine.tenart@free-electrons.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/io.h>
+#include <linux/delay.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+
+#include <asm/cacheflush.h>
+#include <asm/cp15.h>
+#include <asm/smp_plat.h>
+#include <asm/smp_scu.h>
+
+/*
+ * There are two reset registers, one with self-clearing (SC)
+ * reset and one with non-self-clearing reset (NON_SC).
+ */
+#define CPU_RESET_SC		0x00
+#define CPU_RESET_NON_SC	0x20
+
+#define RESET_VECT		0x00
+#define SW_RESET_ADDR		0x94
+
+extern u32 boot_inst;
+
+static void __iomem *cpu_ctrl;
+
+static inline void berlin_perform_reset_cpu(unsigned int cpu)
+{
+	u32 val;
+
+	val = readl(cpu_ctrl + CPU_RESET_NON_SC);
+	val &= ~BIT(cpu_logical_map(cpu));
+	writel(val, cpu_ctrl + CPU_RESET_NON_SC);
+	val |= BIT(cpu_logical_map(cpu));
+	writel(val, cpu_ctrl + CPU_RESET_NON_SC);
+}
+
+static int berlin_boot_secondary(unsigned int cpu, struct task_struct *idle)
+{
+	if (!cpu_ctrl)
+		return -EFAULT;
+
+	/*
+	 * Reset the CPU, making it to execute the instruction in the reset
+	 * exception vector.
+	 */
+	berlin_perform_reset_cpu(cpu);
+
+	return 0;
+}
+
+static void __init berlin_smp_prepare_cpus(unsigned int max_cpus)
+{
+	struct device_node *np;
+	void __iomem *scu_base;
+	void __iomem *vectors_base;
+
+	np = of_find_compatible_node(NULL, NULL, "arm,cortex-a9-scu");
+	scu_base = of_iomap(np, 0);
+	of_node_put(np);
+	if (!scu_base)
+		return;
+
+	np = of_find_compatible_node(NULL, NULL, "marvell,berlin-cpu-ctrl");
+	cpu_ctrl = of_iomap(np, 0);
+	of_node_put(np);
+	if (!cpu_ctrl)
+		goto unmap_scu;
+
+	vectors_base = ioremap(CONFIG_VECTORS_BASE, SZ_32K);
+	if (!vectors_base)
+		goto unmap_scu;
+
+	scu_enable(scu_base);
+	flush_cache_all();
+
+	/*
+	 * Write the first instruction the CPU will execute after being reset
+	 * in the reset exception vector.
+	 */
+	writel(boot_inst, vectors_base + RESET_VECT);
+
+	/*
+	 * Write the secondary startup address into the SW reset address
+	 * vector. This is used by boot_inst.
+	 */
+	writel(virt_to_phys(secondary_startup), vectors_base + SW_RESET_ADDR);
+
+	iounmap(vectors_base);
+unmap_scu:
+	iounmap(scu_base);
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+static void berlin_cpu_die(unsigned int cpu)
+{
+	v7_exit_coherency_flush(louis);
+	while (1)
+		cpu_do_idle();
+}
+
+static int berlin_cpu_kill(unsigned int cpu)
+{
+	u32 val;
+
+	val = readl(cpu_ctrl + CPU_RESET_NON_SC);
+	val &= ~BIT(cpu_logical_map(cpu));
+	writel(val, cpu_ctrl + CPU_RESET_NON_SC);
+
+	return 1;
+}
+#endif
+
+static const struct smp_operations berlin_smp_ops __initconst = {
+	.smp_prepare_cpus	= berlin_smp_prepare_cpus,
+	.smp_boot_secondary	= berlin_boot_secondary,
+#ifdef CONFIG_HOTPLUG_CPU
+	.cpu_die		= berlin_cpu_die,
+	.cpu_kill		= berlin_cpu_kill,
+#endif
+};
+CPU_METHOD_OF_DECLARE(berlin_smp, "marvell,berlin-smp", &berlin_smp_ops);

arch/arm/mach-clps711x/Kconfig

@@ -0,0 +1,15 @@
+menuconfig ARCH_CLPS711X
+	bool "Cirrus Logic EP721x/EP731x-based"
+	depends on ARCH_MULTI_V4T
+	select AUTO_ZRELADDR
+	select CLKSRC_OF
+	select CLPS711X_TIMER
+	select COMMON_CLK
+	select CPU_ARM720T
+	select GENERIC_CLOCKEVENTS
+	select GPIOLIB
+	select MFD_SYSCON
+	select OF_IRQ
+	select USE_OF
+	help
+	  Select this if you use ARMv4T Cirrus Logic chips.

arch/arm/mach-clps711x/Makefile

@@ -0,0 +1 @@
+obj-y += board-dt.o

arch/arm/mach-clps711x/board-dt.c

@@ -0,0 +1,82 @@
+/*
+ *  Author: Alexander Shiyan <shc_work@mail.ru>, 2016
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/io.h>
+#include <linux/of_fdt.h>
+#include <linux/platform_device.h>
+#include <linux/random.h>
+#include <linux/sizes.h>
+
+#include <linux/mfd/syscon/clps711x.h>
+
+#include <asm/system_info.h>
+#include <asm/system_misc.h>
+#include <asm/mach/arch.h>
+#include <asm/mach/map.h>
+
+#define CLPS711X_VIRT_BASE	IOMEM(0xfeff4000)
+#define CLPS711X_PHYS_BASE	(0x80000000)
+# define SYSFLG1		(0x0140)
+# define HALT			(0x0800)
+# define UNIQID			(0x2440)
+# define RANDID0		(0x2700)
+# define RANDID1		(0x2704)
+# define RANDID2		(0x2708)
+# define RANDID3		(0x270c)
+
+static struct map_desc clps711x_io_desc __initdata = {
+	.virtual	= (unsigned long)CLPS711X_VIRT_BASE,
+	.pfn		= __phys_to_pfn(CLPS711X_PHYS_BASE),
+	.length		= 48 * SZ_1K,
+	.type		= MT_DEVICE,
+};
+
+static void __init clps711x_map_io(void)
+{
+	iotable_init(&clps711x_io_desc, 1);
+}
+
+static const struct resource clps711x_cpuidle_res =
+	DEFINE_RES_MEM(CLPS711X_PHYS_BASE + HALT, SZ_128);
+
+static void __init clps711x_init(void)
+{
+	u32 id[5];
+
+	id[0] = readl(CLPS711X_VIRT_BASE + UNIQID);
+	id[1] = readl(CLPS711X_VIRT_BASE + RANDID0);
+	id[2] = readl(CLPS711X_VIRT_BASE + RANDID1);
+	id[3] = readl(CLPS711X_VIRT_BASE + RANDID2);
+	id[4] = readl(CLPS711X_VIRT_BASE + RANDID3);
+	system_rev = SYSFLG1_VERID(readl(CLPS711X_VIRT_BASE + SYSFLG1));
+
+	add_device_randomness(id, sizeof(id));
+
+	system_serial_low = id[0];
+
+	platform_device_register_simple("clps711x-cpuidle", PLATFORM_DEVID_NONE,
+					&clps711x_cpuidle_res, 1);
+}
+
+static void clps711x_restart(enum reboot_mode mode, const char *cmd)
+{
+	soft_restart(0);
+}
+
+static const char *clps711x_compat[] __initconst = {
+	"cirrus,ep7209",
+	NULL
+};
+
+DT_MACHINE_START(CLPS711X_DT, "Cirrus Logic CLPS711X (Device Tree Support)")
+	.dt_compat	= clps711x_compat,
+	.map_io		= clps711x_map_io,
+	.init_late	= clps711x_init,
+	.restart	= clps711x_restart,
+MACHINE_END