Merge upstream 2.6.13-rc1-git1 into 'ieee80211' branch of netdev-2.6.

28 files changed, 482 insertions, 540 deletions

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index c8d94dcd8ef7..620f2ca94ed2 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -361,6 +361,11 @@ config NO_IDLE_HZ
 	  Alternatively, if you want dynamic tick automatically enabled
 	  during boot, pass "dyntick=enable" via the kernel command string.
 
+	  Please note that dynamic tick may affect the accuracy of
+	  timekeeping on some platforms depending on the implementation.
+	  Currently at least OMAP platform is known to have accurate
+	  timekeeping with dynamic tick.
+
 config ARCH_DISCONTIGMEM_ENABLE
 	bool
 	default (ARCH_LH7A40X && !LH7A40X_CONTIGMEM)
diff --git a/arch/arm/kernel/armksyms.c b/arch/arm/kernel/armksyms.c
index 4c38bd8bc298..b713c44c6fb4 100644
--- a/arch/arm/kernel/armksyms.c
+++ b/arch/arm/kernel/armksyms.c
@@ -30,9 +30,6 @@ extern void __lshrdi3(void);
 extern void __modsi3(void);
 extern void __muldi3(void);
 extern void __ucmpdi2(void);
-extern void __udivdi3(void);
-extern void __umoddi3(void);
-extern void __udivmoddi4(void);
 extern void __udivsi3(void);
 extern void __umodsi3(void);
 extern void __do_div64(void);
@@ -134,9 +131,6 @@ EXPORT_SYMBOL(__lshrdi3);
 EXPORT_SYMBOL(__modsi3);
 EXPORT_SYMBOL(__muldi3);
 EXPORT_SYMBOL(__ucmpdi2);
-EXPORT_SYMBOL(__udivdi3);
-EXPORT_SYMBOL(__umoddi3);
-EXPORT_SYMBOL(__udivmoddi4);
 EXPORT_SYMBOL(__udivsi3);
 EXPORT_SYMBOL(__umodsi3);
 EXPORT_SYMBOL(__do_div64);
diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c
index 8cf733daa800..35b7273cfdb4 100644
--- a/arch/arm/kernel/setup.c
+++ b/arch/arm/kernel/setup.c
@@ -359,7 +359,8 @@ void cpu_init(void)
 	      "I" (offsetof(struct stack, abt[0])),
 	      "I" (PSR_F_BIT | PSR_I_BIT | UND_MODE),
 	      "I" (offsetof(struct stack, und[0])),
-	      "I" (PSR_F_BIT | PSR_I_BIT | SVC_MODE));
+	      "I" (PSR_F_BIT | PSR_I_BIT | SVC_MODE)
+	    : "r14");
 }
 
 static struct machine_desc * __init setup_machine(unsigned int nr)
diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
index 34892758f098..a931409c8fe4 100644
--- a/arch/arm/kernel/smp.c
+++ b/arch/arm/kernel/smp.c
@@ -502,3 +502,126 @@ int __init setup_profiling_timer(unsigned int multiplier)
 {
 	return -EINVAL;
 }
+
+static int
+on_each_cpu_mask(void (*func)(void *), void *info, int retry, int wait,
+		 cpumask_t mask)
+{
+	int ret = 0;
+
+	preempt_disable();
+
+	ret = smp_call_function_on_cpu(func, info, retry, wait, mask);
+	if (cpu_isset(smp_processor_id(), mask))
+		func(info);
+
+	preempt_enable();
+
+	return ret;
+}
+
+/**********************************************************************/
+
+/*
+ * TLB operations
+ */
+struct tlb_args {
+	struct vm_area_struct *ta_vma;
+	unsigned long ta_start;
+	unsigned long ta_end;
+};
+
+static inline void ipi_flush_tlb_all(void *ignored)
+{
+	local_flush_tlb_all();
+}
+
+static inline void ipi_flush_tlb_mm(void *arg)
+{
+	struct mm_struct *mm = (struct mm_struct *)arg;
+
+	local_flush_tlb_mm(mm);
+}
+
+static inline void ipi_flush_tlb_page(void *arg)
+{
+	struct tlb_args *ta = (struct tlb_args *)arg;
+
+	local_flush_tlb_page(ta->ta_vma, ta->ta_start);
+}
+
+static inline void ipi_flush_tlb_kernel_page(void *arg)
+{
+	struct tlb_args *ta = (struct tlb_args *)arg;
+
+	local_flush_tlb_kernel_page(ta->ta_start);
+}
+
+static inline void ipi_flush_tlb_range(void *arg)
+{
+	struct tlb_args *ta = (struct tlb_args *)arg;
+
+	local_flush_tlb_range(ta->ta_vma, ta->ta_start, ta->ta_end);
+}
+
+static inline void ipi_flush_tlb_kernel_range(void *arg)
+{
+	struct tlb_args *ta = (struct tlb_args *)arg;
+
+	local_flush_tlb_kernel_range(ta->ta_start, ta->ta_end);
+}
+
+void flush_tlb_all(void)
+{
+	on_each_cpu(ipi_flush_tlb_all, NULL, 1, 1);
+}
+
+void flush_tlb_mm(struct mm_struct *mm)
+{
+	cpumask_t mask = mm->cpu_vm_mask;
+
+	on_each_cpu_mask(ipi_flush_tlb_mm, mm, 1, 1, mask);
+}
+
+void flush_tlb_page(struct vm_area_struct *vma, unsigned long uaddr)
+{
+	cpumask_t mask = vma->vm_mm->cpu_vm_mask;
+	struct tlb_args ta;
+
+	ta.ta_vma = vma;
+	ta.ta_start = uaddr;
+
+	on_each_cpu_mask(ipi_flush_tlb_page, &ta, 1, 1, mask);
+}
+
+void flush_tlb_kernel_page(unsigned long kaddr)
+{
+	struct tlb_args ta;
+
+	ta.ta_start = kaddr;
+
+	on_each_cpu(ipi_flush_tlb_kernel_page, &ta, 1, 1);
+}
+
+void flush_tlb_range(struct vm_area_struct *vma,
+                     unsigned long start, unsigned long end)
+{
+	cpumask_t mask = vma->vm_mm->cpu_vm_mask;
+	struct tlb_args ta;
+
+	ta.ta_vma = vma;
+	ta.ta_start = start;
+	ta.ta_end = end;
+
+	on_each_cpu_mask(ipi_flush_tlb_range, &ta, 1, 1, mask);
+}
+
+void flush_tlb_kernel_range(unsigned long start, unsigned long end)
+{
+	struct tlb_args ta;
+
+	ta.ta_start = start;
+	ta.ta_end = end;
+
+	on_each_cpu(ipi_flush_tlb_kernel_range, &ta, 1, 1);
+}
diff --git a/arch/arm/lib/Makefile b/arch/arm/lib/Makefile
index c0e65833ffc4..8725d63e4219 100644
--- a/arch/arm/lib/Makefile
+++ b/arch/arm/lib/Makefile
@@ -11,7 +11,7 @@ lib-y		:= backtrace.o changebit.o csumipv6.o csumpartial.o   \
 		   strnlen_user.o strchr.o strrchr.o testchangebit.o  \
 		   testclearbit.o testsetbit.o uaccess.o getuser.o    \
 		   putuser.o ashldi3.o ashrdi3.o lshrdi3.o muldi3.o   \
-		   ucmpdi2.o udivdi3.o lib1funcs.o div64.o	      \
+		   ucmpdi2.o lib1funcs.o div64.o	              \
 		   io-readsb.o io-writesb.o io-readsl.o io-writesl.o
 
 ifeq ($(CONFIG_CPU_32v3),y)
diff --git a/arch/arm/lib/longlong.h b/arch/arm/lib/longlong.h
deleted file mode 100644
index 90ae647e4d76..000000000000
--- a/arch/arm/lib/longlong.h
+++ /dev/null
@@ -1,183 +0,0 @@
-/* longlong.h -- based on code from gcc-2.95.3
-
-   definitions for mixed size 32/64 bit arithmetic.
-   Copyright (C) 1991, 92, 94, 95, 96, 1997, 1998 Free Software Foundation, Inc.
-
-   This definition 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.
-
-   This definition 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; if not, write to the Free Software
-   Foundation, Inc., 59 Temple Place - Suite 330,
-   Boston, MA 02111-1307, USA.  */
-
-/* Borrowed from GCC 2.95.3, I Molton 29/07/01 */
-
-#ifndef SI_TYPE_SIZE
-#define SI_TYPE_SIZE 32
-#endif
-
-#define __BITS4 (SI_TYPE_SIZE / 4)
-#define __ll_B (1L << (SI_TYPE_SIZE / 2))
-#define __ll_lowpart(t) ((u32) (t) % __ll_B)
-#define __ll_highpart(t) ((u32) (t) / __ll_B)
-
-/* Define auxiliary asm macros.
-
-   1) umul_ppmm(high_prod, low_prod, multipler, multiplicand)
-   multiplies two u32 integers MULTIPLER and MULTIPLICAND,
-   and generates a two-part u32 product in HIGH_PROD and
-   LOW_PROD.
-
-   2) __umulsidi3(a,b) multiplies two u32 integers A and B,
-   and returns a u64 product.  This is just a variant of umul_ppmm.
-
-   3) udiv_qrnnd(quotient, remainder, high_numerator, low_numerator,
-   denominator) divides a two-word unsigned integer, composed by the
-   integers HIGH_NUMERATOR and LOW_NUMERATOR, by DENOMINATOR and
-   places the quotient in QUOTIENT and the remainder in REMAINDER.
-   HIGH_NUMERATOR must be less than DENOMINATOR for correct operation.
-   If, in addition, the most significant bit of DENOMINATOR must be 1,
-   then the pre-processor symbol UDIV_NEEDS_NORMALIZATION is defined to 1.
-
-   4) sdiv_qrnnd(quotient, remainder, high_numerator, low_numerator,
-   denominator).  Like udiv_qrnnd but the numbers are signed.  The
-   quotient is rounded towards 0.
-
-   5) count_leading_zeros(count, x) counts the number of zero-bits from
-   the msb to the first non-zero bit.  This is the number of steps X
-   needs to be shifted left to set the msb.  Undefined for X == 0.
-
-   6) add_ssaaaa(high_sum, low_sum, high_addend_1, low_addend_1,
-   high_addend_2, low_addend_2) adds two two-word unsigned integers,
-   composed by HIGH_ADDEND_1 and LOW_ADDEND_1, and HIGH_ADDEND_2 and
-   LOW_ADDEND_2 respectively.  The result is placed in HIGH_SUM and
-   LOW_SUM.  Overflow (i.e. carry out) is not stored anywhere, and is
-   lost.
-
-   7) sub_ddmmss(high_difference, low_difference, high_minuend,
-   low_minuend, high_subtrahend, low_subtrahend) subtracts two
-   two-word unsigned integers, composed by HIGH_MINUEND_1 and
-   LOW_MINUEND_1, and HIGH_SUBTRAHEND_2 and LOW_SUBTRAHEND_2
-   respectively.  The result is placed in HIGH_DIFFERENCE and
-   LOW_DIFFERENCE.  Overflow (i.e. carry out) is not stored anywhere,
-   and is lost.
-
-   If any of these macros are left undefined for a particular CPU,
-   C macros are used.  */
-
-#if defined (__arm__)
-#define add_ssaaaa(sh, sl, ah, al, bh, bl) \
-  __asm__ ("adds	%1, %4, %5					\n\
-	adc	%0, %2, %3"						\
-	   : "=r" ((u32) (sh)),					\
-	     "=&r" ((u32) (sl))					\
-	   : "%r" ((u32) (ah)),					\
-	     "rI" ((u32) (bh)),					\
-	     "%r" ((u32) (al)),					\
-	     "rI" ((u32) (bl)))
-#define sub_ddmmss(sh, sl, ah, al, bh, bl) \
-  __asm__ ("subs	%1, %4, %5					\n\
-	sbc	%0, %2, %3"						\
-	   : "=r" ((u32) (sh)),					\
-	     "=&r" ((u32) (sl))					\
-	   : "r" ((u32) (ah)),					\
-	     "rI" ((u32) (bh)),					\
-	     "r" ((u32) (al)),					\
-	     "rI" ((u32) (bl)))
-#define umul_ppmm(xh, xl, a, b) \
-{register u32 __t0, __t1, __t2;					\
-  __asm__ ("%@ Inlined umul_ppmm					\n\
-	mov	%2, %5, lsr #16						\n\
-	mov	%0, %6, lsr #16						\n\
-	bic	%3, %5, %2, lsl #16					\n\
-	bic	%4, %6, %0, lsl #16					\n\
-	mul	%1, %3, %4						\n\
-	mul	%4, %2, %4						\n\
-	mul	%3, %0, %3						\n\
-	mul	%0, %2, %0						\n\
-	adds	%3, %4, %3						\n\
-	addcs	%0, %0, #65536						\n\
-	adds	%1, %1, %3, lsl #16					\n\
-	adc	%0, %0, %3, lsr #16"					\
-	   : "=&r" ((u32) (xh)),					\
-	     "=r" ((u32) (xl)),					\
-	     "=&r" (__t0), "=&r" (__t1), "=r" (__t2)			\
-	   : "r" ((u32) (a)),					\
-	     "r" ((u32) (b)));}
-#define UMUL_TIME 20
-#define UDIV_TIME 100
-#endif				/* __arm__ */
-
-#define __umulsidi3(u, v) \
-  ({DIunion __w;							\
-    umul_ppmm (__w.s.high, __w.s.low, u, v);				\
-    __w.ll; })
-
-#define __udiv_qrnnd_c(q, r, n1, n0, d) \
-  do {									\
-    u32 __d1, __d0, __q1, __q0;					\
-    u32 __r1, __r0, __m;						\
-    __d1 = __ll_highpart (d);						\
-    __d0 = __ll_lowpart (d);						\
-									\
-    __r1 = (n1) % __d1;							\
-    __q1 = (n1) / __d1;							\
-    __m = (u32) __q1 * __d0;					\
-    __r1 = __r1 * __ll_B | __ll_highpart (n0);				\
-    if (__r1 < __m)							\
-      {									\
-	__q1--, __r1 += (d);						\
-	if (__r1 >= (d)) /* i.e. we didn't get carry when adding to __r1 */\
-	  if (__r1 < __m)						\
-	    __q1--, __r1 += (d);					\
-      }									\
-    __r1 -= __m;							\
-									\
-    __r0 = __r1 % __d1;							\
-    __q0 = __r1 / __d1;							\
-    __m = (u32) __q0 * __d0;					\
-    __r0 = __r0 * __ll_B | __ll_lowpart (n0);				\
-    if (__r0 < __m)							\
-      {									\
-	__q0--, __r0 += (d);						\
-	if (__r0 >= (d))						\
-	  if (__r0 < __m)						\
-	    __q0--, __r0 += (d);					\
-      }									\
-    __r0 -= __m;							\
-									\
-    (q) = (u32) __q1 * __ll_B | __q0;				\
-    (r) = __r0;								\
-  } while (0)
-
-#define UDIV_NEEDS_NORMALIZATION 1
-#define udiv_qrnnd __udiv_qrnnd_c
-
-#define count_leading_zeros(count, x) \
-  do {									\
-    u32 __xr = (x);							\
-    u32 __a;							\
-									\
-    if (SI_TYPE_SIZE <= 32)						\
-      {									\
-	__a = __xr < ((u32)1<<2*__BITS4)				\
-	  ? (__xr < ((u32)1<<__BITS4) ? 0 : __BITS4)		\
-	  : (__xr < ((u32)1<<3*__BITS4) ?  2*__BITS4 : 3*__BITS4);	\
-      }									\
-    else								\
-      {									\
-	for (__a = SI_TYPE_SIZE - 8; __a > 0; __a -= 8)			\
-	  if (((__xr >> __a) & 0xff) != 0)				\
-	    break;							\
-      }									\
-									\
-    (count) = SI_TYPE_SIZE - (__clz_tab[__xr >> __a] + __a);		\
-  } while (0)
diff --git a/arch/arm/lib/udivdi3.c b/arch/arm/lib/udivdi3.c
deleted file mode 100644
index e343be4c6642..000000000000
--- a/arch/arm/lib/udivdi3.c
+++ /dev/null
@@ -1,222 +0,0 @@
-/* More subroutines needed by GCC output code on some machines.  */
-/* Compile this one with gcc.  */
-/* Copyright (C) 1989, 92-98, 1999 Free Software Foundation, Inc.
-
-This file is part of GNU CC.
-
-GNU CC 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.
-
-GNU CC 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 GNU CC; see the file COPYING.  If not, write to
-the Free Software Foundation, 59 Temple Place - Suite 330,
-Boston, MA 02111-1307, USA.  */
-
-/* As a special exception, if you link this library with other files,
-   some of which are compiled with GCC, to produce an executable,
-   this library does not by itself cause the resulting executable
-   to be covered by the GNU General Public License.
-   This exception does not however invalidate any other reasons why
-   the executable file might be covered by the GNU General Public License.
- */
-/* support functions required by the kernel. based on code from gcc-2.95.3 */
-/* I Molton     29/07/01 */
-
-#include "gcclib.h"
-#include "longlong.h"
-
-static const u8 __clz_tab[] = {
-	0, 1, 2, 2, 3, 3, 3, 3, 4, 4, 4, 4, 4, 4, 4, 4, 5, 5, 5, 5, 5, 5, 5, 5,
-	    5, 5, 5, 5, 5, 5, 5, 5,
-	6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6,
-	    6, 6, 6, 6, 6, 6, 6, 6,
-	7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
-	    7, 7, 7, 7, 7, 7, 7, 7,
-	7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
-	    7, 7, 7, 7, 7, 7, 7, 7,
-	8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8,
-	    8, 8, 8, 8, 8, 8, 8, 8,
-	8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8,
-	    8, 8, 8, 8, 8, 8, 8, 8,
-	8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8,
-	    8, 8, 8, 8, 8, 8, 8, 8,
-	8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8,
-	    8, 8, 8, 8, 8, 8, 8, 8,
-};
-
-u64 __udivmoddi4(u64 n, u64 d, u64 * rp)
-{
-	DIunion ww;
-	DIunion nn, dd;
-	DIunion rr;
-	u32 d0, d1, n0, n1, n2;
-	u32 q0, q1;
-	u32 b, bm;
-
-	nn.ll = n;
-	dd.ll = d;
-
-	d0 = dd.s.low;
-	d1 = dd.s.high;
-	n0 = nn.s.low;
-	n1 = nn.s.high;
-
-	if (d1 == 0) {
-		if (d0 > n1) {
-			/* 0q = nn / 0D */
-
-			count_leading_zeros(bm, d0);
-
-			if (bm != 0) {
-				/* Normalize, i.e. make the most significant bit of the
-				   denominator set.  */
-
-				d0 = d0 << bm;
-				n1 = (n1 << bm) | (n0 >> (SI_TYPE_SIZE - bm));
-				n0 = n0 << bm;
-			}
-
-			udiv_qrnnd(q0, n0, n1, n0, d0);
-			q1 = 0;
-
-			/* Remainder in n0 >> bm.  */
-		} else {
-			/* qq = NN / 0d */
-
-			if (d0 == 0)
-				d0 = 1 / d0;	/* Divide intentionally by zero.  */
-
-			count_leading_zeros(bm, d0);
-
-			if (bm == 0) {
-				/* From (n1 >= d0) /\ (the most significant bit of d0 is set),
-				   conclude (the most significant bit of n1 is set) /\ (the
-				   leading quotient digit q1 = 1).
-
-				   This special case is necessary, not an optimization.
-				   (Shifts counts of SI_TYPE_SIZE are undefined.)  */
-
-				n1 -= d0;
-				q1 = 1;
-			} else {
-				/* Normalize.  */
-
-				b = SI_TYPE_SIZE - bm;
-
-				d0 = d0 << bm;
-				n2 = n1 >> b;
-				n1 = (n1 << bm) | (n0 >> b);
-				n0 = n0 << bm;
-
-				udiv_qrnnd(q1, n1, n2, n1, d0);
-			}
-
-			/* n1 != d0...  */
-
-			udiv_qrnnd(q0, n0, n1, n0, d0);
-
-			/* Remainder in n0 >> bm.  */
-		}
-
-		if (rp != 0) {
-			rr.s.low = n0 >> bm;
-			rr.s.high = 0;
-			*rp = rr.ll;
-		}
-	} else {
-		if (d1 > n1) {
-			/* 00 = nn / DD */
-
-			q0 = 0;
-			q1 = 0;
-
-			/* Remainder in n1n0.  */
-			if (rp != 0) {
-				rr.s.low = n0;
-				rr.s.high = n1;
-				*rp = rr.ll;
-			}
-		} else {
-			/* 0q = NN / dd */
-
-			count_leading_zeros(bm, d1);
-			if (bm == 0) {
-				/* From (n1 >= d1) /\ (the most significant bit of d1 is set),
-				   conclude (the most significant bit of n1 is set) /\ (the
-				   quotient digit q0 = 0 or 1).
-
-				   This special case is necessary, not an optimization.  */
-
-				/* The condition on the next line takes advantage of that
-				   n1 >= d1 (true due to program flow).  */
-				if (n1 > d1 || n0 >= d0) {
-					q0 = 1;
-					sub_ddmmss(n1, n0, n1, n0, d1, d0);
-				} else
-					q0 = 0;
-
-				q1 = 0;
-
-				if (rp != 0) {
-					rr.s.low = n0;
-					rr.s.high = n1;
-					*rp = rr.ll;
-				}
-			} else {
-				u32 m1, m0;
-				/* Normalize.  */
-
-				b = SI_TYPE_SIZE - bm;
-
-				d1 = (d1 << bm) | (d0 >> b);
-				d0 = d0 << bm;
-				n2 = n1 >> b;
-				n1 = (n1 << bm) | (n0 >> b);
-				n0 = n0 << bm;
-
-				udiv_qrnnd(q0, n1, n2, n1, d1);
-				umul_ppmm(m1, m0, q0, d0);
-
-				if (m1 > n1 || (m1 == n1 && m0 > n0)) {
-					q0--;
-					sub_ddmmss(m1, m0, m1, m0, d1, d0);
-				}
-
-				q1 = 0;
-
-				/* Remainder in (n1n0 - m1m0) >> bm.  */
-				if (rp != 0) {
-					sub_ddmmss(n1, n0, n1, n0, m1, m0);
-					rr.s.low = (n1 << b) | (n0 >> bm);
-					rr.s.high = n1 >> bm;
-					*rp = rr.ll;
-				}
-			}
-		}
-	}
-
-	ww.s.low = q0;
-	ww.s.high = q1;
-	return ww.ll;
-}
-
-u64 __udivdi3(u64 n, u64 d)
-{
-	return __udivmoddi4(n, d, (u64 *) 0);
-}
-
-u64 __umoddi3(u64 u, u64 v)
-{
-	u64 w;
-
-	(void)__udivmoddi4(u, v, &w);
-
-	return w;
-}
diff --git a/arch/arm/mach-integrator/core.c b/arch/arm/mach-integrator/core.c
index 9222e57bd872..dacbf504dae2 100644
--- a/arch/arm/mach-integrator/core.c
+++ b/arch/arm/mach-integrator/core.c
@@ -20,6 +20,7 @@
 #include <asm/irq.h>
 #include <asm/io.h>
 #include <asm/hardware/amba.h>
+#include <asm/hardware/arm_timer.h>
 #include <asm/arch/cm.h>
 #include <asm/system.h>
 #include <asm/leds.h>
@@ -156,16 +157,6 @@ EXPORT_SYMBOL(cm_control);
 #define TICKS2USECS(x)	((x) / TICKS_PER_uSEC)
 #endif
 
-/*
- * What does it look like?
- */
-typedef struct TimerStruct {
-	unsigned long TimerLoad;
-	unsigned long TimerValue;
-	unsigned long TimerControl;
-	unsigned long TimerClear;
-} TimerStruct_t;
-
 static unsigned long timer_reload;
 
 /*
@@ -174,7 +165,6 @@ static unsigned long timer_reload;
  */
 unsigned long integrator_gettimeoffset(void)
 {
-	volatile TimerStruct_t *timer1 = (TimerStruct_t *)TIMER1_VA_BASE;
 	unsigned long ticks1, ticks2, status;
 
 	/*
@@ -183,11 +173,11 @@ unsigned long integrator_gettimeoffset(void)
 	 * an interrupt.  We get around this by ensuring that the
 	 * counter has not reloaded between our two reads.
 	 */
-	ticks2 = timer1->TimerValue & 0xffff;
+	ticks2 = readl(TIMER1_VA_BASE + TIMER_VALUE) & 0xffff;
 	do {
 		ticks1 = ticks2;
 		status = __raw_readl(VA_IC_BASE + IRQ_RAW_STATUS);
-		ticks2 = timer1->TimerValue & 0xffff;
+		ticks2 = readl(TIMER1_VA_BASE + TIMER_VALUE) & 0xffff;
 	} while (ticks2 > ticks1);
 
 	/*
@@ -213,14 +203,12 @@ unsigned long integrator_gettimeoffset(void)
 static irqreturn_t
 integrator_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 {
-	volatile TimerStruct_t *timer1 = (volatile TimerStruct_t *)TIMER1_VA_BASE;
-
 	write_seqlock(&xtime_lock);
 
 	/*
 	 * clear the interrupt
 	 */
-	timer1->TimerClear = 1;
+	writel(1, TIMER1_VA_BASE + TIMER_INTCLR);
 
 	/*
 	 * the clock tick routines are only processed on the
@@ -256,32 +244,29 @@ static struct irqaction integrator_timer_irq = {
  */
 void __init integrator_time_init(unsigned long reload, unsigned int ctrl)
 {
-	volatile TimerStruct_t *timer0 = (volatile TimerStruct_t *)TIMER0_VA_BASE;
-	volatile TimerStruct_t *timer1 = (volatile TimerStruct_t *)TIMER1_VA_BASE;
-	volatile TimerStruct_t *timer2 = (volatile TimerStruct_t *)TIMER2_VA_BASE;
-	unsigned int timer_ctrl = 0x80 | 0x40;	/* periodic */
+	unsigned int timer_ctrl = TIMER_CTRL_ENABLE | TIMER_CTRL_PERIODIC;
 
 	timer_reload = reload;
 	timer_ctrl |= ctrl;
 
 	if (timer_reload > 0x100000) {
 		timer_reload >>= 8;
-		timer_ctrl |= 0x08; /* /256 */
+		timer_ctrl |= TIMER_CTRL_DIV256;
 	} else if (timer_reload > 0x010000) {
 		timer_reload >>= 4;
-		timer_ctrl |= 0x04; /* /16 */
+		timer_ctrl |= TIMER_CTRL_DIV16;
 	}
 
 	/*
 	 * Initialise to a known state (all timers off)
 	 */
-	timer0->TimerControl = 0;
-	timer1->TimerControl = 0;
-	timer2->TimerControl = 0;
+	writel(0, TIMER0_VA_BASE + TIMER_CTRL);
+	writel(0, TIMER1_VA_BASE + TIMER_CTRL);
+	writel(0, TIMER2_VA_BASE + TIMER_CTRL);
 
-	timer1->TimerLoad    = timer_reload;
-	timer1->TimerValue   = timer_reload;
-	timer1->TimerControl = timer_ctrl;
+	writel(timer_reload, TIMER1_VA_BASE + TIMER_LOAD);
+	writel(timer_reload, TIMER1_VA_BASE + TIMER_VALUE);
+	writel(timer_ctrl, TIMER1_VA_BASE + TIMER_CTRL);
 
 	/*
 	 * Make irqs happen for the system timer
diff --git a/arch/arm/mach-omap/pm.c b/arch/arm/mach-omap/pm.c
index 00fac155df2a..6b03ccdc1e92 100644
--- a/arch/arm/mach-omap/pm.c
+++ b/arch/arm/mach-omap/pm.c
@@ -41,7 +41,9 @@
 #include <linux/pm.h>
 
 #include <asm/io.h>
+#include <asm/mach/time.h>
 #include <asm/mach-types.h>
+
 #include <asm/arch/omap16xx.h>
 #include <asm/arch/pm.h>
 #include <asm/arch/mux.h>
@@ -80,13 +82,13 @@ void omap_pm_idle(void)
 		return;
 	}
 	mask32 = omap_readl(ARM_SYSST);
-	local_fiq_enable();
-	local_irq_enable();
 
-#if defined(CONFIG_OMAP_32K_TIMER) && defined(CONFIG_NO_IDLE_HZ)
-	/* Override timer to use VST for the next cycle */
-	omap_32k_timer_next_vst_interrupt();
-#endif
+	/*
+	 * Since an interrupt may set up a timer, we don't want to
+	 * reprogram the hardware timer with interrupts enabled.
+	 * Re-enable interrupts only after returning from idle.
+	 */
+	timer_dyn_reprogram();
 
 	if ((mask32 & DSP_IDLE) == 0) {
 		__asm__ volatile ("mcr	p15, 0, r0, c7, c0, 4");
@@ -102,6 +104,8 @@ void omap_pm_idle(void)
 
 		func_ptr();
 	}
+	local_fiq_enable();
+	local_irq_enable();
 }
 
 /*
diff --git a/arch/arm/mach-omap/time.c b/arch/arm/mach-omap/time.c
index 589e8b2740dd..dd34e9f4c413 100644
--- a/arch/arm/mach-omap/time.c
+++ b/arch/arm/mach-omap/time.c
@@ -4,7 +4,7 @@
  * OMAP Timers
  *
  * Copyright (C) 2004 Nokia Corporation
- * Partial timer rewrite and additional VST timer support by
+ * Partial timer rewrite and additional dynamic tick timer support by
  * Tony Lindgen <tony@atomide.com> and
  * Tuukka Tikkanen <tuukka.tikkanen@elektrobit.com>
  *
@@ -261,7 +261,6 @@ unsigned long long sched_clock(void)
  * so with HZ = 100, TVR = 327.68.
  */
 #define OMAP_32K_TIMER_TICK_PERIOD	((32768 / HZ) - 1)
-#define MAX_SKIP_JIFFIES		25
 #define TIMER_32K_SYNCHRONIZED		0xfffbc410
 
 #define JIFFIES_TO_HW_TICKS(nr_jiffies, clock_rate)			\
@@ -347,6 +346,42 @@ static irqreturn_t omap_32k_timer_interrupt(int irq, void *dev_id,
 	return IRQ_HANDLED;
 }
 
+#ifdef CONFIG_NO_IDLE_HZ
+/*
+ * Programs the next timer interrupt needed. Called when dynamic tick is
+ * enabled, and to reprogram the ticks to skip from pm_idle. Note that
+ * we can keep the timer continuous, and don't need to set it to run in
+ * one-shot mode. This is because the timer will get reprogrammed again
+ * after next interrupt.
+ */
+void omap_32k_timer_reprogram(unsigned long next_tick)
+{
+	omap_32k_timer_start(JIFFIES_TO_HW_TICKS(next_tick, 32768) + 1);
+}
+
+static struct irqaction omap_32k_timer_irq;
+extern struct timer_update_handler timer_update;
+
+static int omap_32k_timer_enable_dyn_tick(void)
+{
+	/* No need to reprogram timer, just use the next interrupt */
+	return 0;
+}
+
+static int omap_32k_timer_disable_dyn_tick(void)
+{
+	omap_32k_timer_start(OMAP_32K_TIMER_TICK_PERIOD);
+	return 0;
+}
+
+static struct dyn_tick_timer omap_dyn_tick_timer = {
+	.enable		= omap_32k_timer_enable_dyn_tick,
+	.disable	= omap_32k_timer_disable_dyn_tick,
+	.reprogram	= omap_32k_timer_reprogram,
+	.handler	= omap_32k_timer_interrupt,
+};
+#endif	/* CONFIG_NO_IDLE_HZ */
+
 static struct irqaction omap_32k_timer_irq = {
 	.name		= "32KHz timer",
 	.flags		= SA_INTERRUPT | SA_TIMER,
@@ -355,6 +390,11 @@ static struct irqaction omap_32k_timer_irq = {
 
 static __init void omap_init_32k_timer(void)
 {
+
+#ifdef CONFIG_NO_IDLE_HZ
+	omap_timer.dyn_tick = &omap_dyn_tick_timer;
+#endif
+
 	setup_irq(INT_OS_TIMER, &omap_32k_timer_irq);
 	omap_timer.offset  = omap_32k_timer_gettimeoffset;
 	omap_32k_last_tick = omap_32k_sync_timer_read();
diff --git a/arch/arm/mach-s3c2410/Kconfig b/arch/arm/mach-s3c2410/Kconfig
index 534df0c6c770..d4d03d0daaec 100644
--- a/arch/arm/mach-s3c2410/Kconfig
+++ b/arch/arm/mach-s3c2410/Kconfig
@@ -154,6 +154,11 @@ config S3C2410_PM_CHECK_CHUNKSIZE
 	  the CRC data block will take more memory, but wil identify any
 	  faults with better precision.
 
+config PM_SIMTEC
+	bool
+	depends on PM && (ARCH_BAST || MACH_VR1000)
+	default y
+
 config S3C2410_LOWLEVEL_UART_PORT
 	int "S3C2410 UART to use for low-level messages"
 	default 0
diff --git a/arch/arm/mach-s3c2410/Makefile b/arch/arm/mach-s3c2410/Makefile
index 7c379aad5d62..f99b689e4392 100644
--- a/arch/arm/mach-s3c2410/Makefile
+++ b/arch/arm/mach-s3c2410/Makefile
@@ -18,6 +18,7 @@ obj-$(CONFIG_S3C2410_DMA)  += dma.o
 # Power Management support
 
 obj-$(CONFIG_PM)	   += pm.o sleep.o
+obj-$(CONFIG_PM_SIMTEC)	   += pm-simtec.o
 
 # S3C2440 support
 
diff --git a/arch/arm/mach-s3c2410/devs.c b/arch/arm/mach-s3c2410/devs.c
index 64792f678668..4664bd11adc1 100644
--- a/arch/arm/mach-s3c2410/devs.c
+++ b/arch/arm/mach-s3c2410/devs.c
@@ -96,8 +96,8 @@ struct platform_device s3c_device_lcd = {
 	.num_resources	  = ARRAY_SIZE(s3c_lcd_resource),
 	.resource	  = s3c_lcd_resource,
 	.dev              = {
-		.dma_mask = &s3c_device_lcd_dmamask,
-		.coherent_dma_mask = 0xffffffffUL
+		.dma_mask		= &s3c_device_lcd_dmamask,
+		.coherent_dma_mask	= 0xffffffffUL
 	}
 };
 
diff --git a/arch/arm/mach-s3c2410/irq.c b/arch/arm/mach-s3c2410/irq.c
index b668c48f4399..cf9f46d88061 100644
--- a/arch/arm/mach-s3c2410/irq.c
+++ b/arch/arm/mach-s3c2410/irq.c
@@ -40,8 +40,11 @@
  *   04-Nov-2004  Ben Dooks
  *		  Fix standard IRQ wake for EINT0..4 and RTC
  *
- *   22-Feb-2004  Ben Dooks
+ *   22-Feb-2005  Ben Dooks
  *		  Fixed edge-triggering on ADC IRQ
+ *
+ *   28-Jun-2005  Ben Dooks
+ *		  Mark IRQ_LCD valid
 */
 
 #include <linux/init.h>
@@ -366,7 +369,6 @@ static struct irqchip s3c_irq_eint0t4 = {
 #define INTMSK_UART1	 (1UL << (IRQ_UART1 - IRQ_EINT0))
 #define INTMSK_UART2	 (1UL << (IRQ_UART2 - IRQ_EINT0))
 #define INTMSK_ADCPARENT (1UL << (IRQ_ADCPARENT - IRQ_EINT0))
-#define INTMSK_LCD	 (1UL << (IRQ_LCD - IRQ_EINT0))
 
 static inline void
 s3c_irqsub_mask(unsigned int irqno, unsigned int parentbit,
@@ -716,7 +718,6 @@ void __init s3c24xx_init_irq(void)
 		case IRQ_UART0:
 		case IRQ_UART1:
 		case IRQ_UART2:
-		case IRQ_LCD:
 		case IRQ_ADCPARENT:
 			set_irq_chip(irqno, &s3c_irq_level_chip);
 			set_irq_handler(irqno, do_level_IRQ);
diff --git a/arch/arm/mach-s3c2410/mach-bast.c b/arch/arm/mach-s3c2410/mach-bast.c
index f3e970039b65..549bcb1f32c0 100644
--- a/arch/arm/mach-s3c2410/mach-bast.c
+++ b/arch/arm/mach-s3c2410/mach-bast.c
@@ -27,6 +27,7 @@
  *     10-Mar-2005 LCVR Changed S3C2410_VA to S3C24XX_VA
  *     14-Mar-2006 BJD  Updated for __iomem changes
  *     22-Jun-2006 BJD  Added DM9000 platform information
+ *     28-Jun-2006 BJD  Moved pm functionality out to common code
 */
 
 #include <linux/kernel.h>
@@ -67,7 +68,6 @@
 #include "devs.h"
 #include "cpu.h"
 #include "usb-simtec.h"
-#include "pm.h"
 
 #define COPYRIGHT ", (c) 2004-2005 Simtec Electronics"
 
@@ -405,44 +405,13 @@ void __init bast_map_io(void)
 	usb_simtec_init();
 }
 
-void __init bast_init_irq(void)
-{
-	s3c24xx_init_irq();
-}
-
-#ifdef CONFIG_PM
-
-/* bast_init_machine
- *
- * enable the power management functions for the EB2410ITX
-*/
-
-static __init void bast_init_machine(void)
-{
-	unsigned long gstatus4;
-
-	printk(KERN_INFO "BAST Power Manangement" COPYRIGHT "\n");
-
-	gstatus4  = (__raw_readl(S3C2410_BANKCON7) & 0x3) << 30;
-	gstatus4 |= (__raw_readl(S3C2410_BANKCON6) & 0x3) << 28;
-	gstatus4 |= (__raw_readl(S3C2410_BANKSIZE) & S3C2410_BANKSIZE_MASK);
-
-	__raw_writel(gstatus4, S3C2410_GSTATUS4);
-
-	s3c2410_pm_init();
-}
-
-#else
-#define bast_init_machine NULL
-#endif
-
 
 MACHINE_START(BAST, "Simtec-BAST")
      MAINTAINER("Ben Dooks <ben@simtec.co.uk>")
      BOOT_MEM(S3C2410_SDRAM_PA, S3C2410_PA_UART, (u32)S3C24XX_VA_UART)
      BOOT_PARAMS(S3C2410_SDRAM_PA + 0x100)
-     MAPIO(bast_map_io)
-     INITIRQ(bast_init_irq)
-	.init_machine	= bast_init_machine,
+
+	.map_io		= bast_map_io,
+	.init_irq	= s3c24xx_init_irq,
 	.timer		= &s3c24xx_timer,
 MACHINE_END
diff --git a/arch/arm/mach-s3c2410/mach-vr1000.c b/arch/arm/mach-s3c2410/mach-vr1000.c
index 76be074944a0..1db2855e3e56 100644
--- a/arch/arm/mach-s3c2410/mach-vr1000.c
+++ b/arch/arm/mach-s3c2410/mach-vr1000.c
@@ -371,16 +371,12 @@ void __init vr1000_map_io(void)
 	usb_simtec_init();
 }
 
-void __init vr1000_init_irq(void)
-{
-	s3c24xx_init_irq();
-}
 
 MACHINE_START(VR1000, "Thorcom-VR1000")
      MAINTAINER("Ben Dooks <ben@simtec.co.uk>")
      BOOT_MEM(S3C2410_SDRAM_PA, S3C2410_PA_UART, (u32)S3C24XX_VA_UART)
      BOOT_PARAMS(S3C2410_SDRAM_PA + 0x100)
-     MAPIO(vr1000_map_io)
-     INITIRQ(vr1000_init_irq)
+	.map_io		= vr1000_map_io,
+	.init_irq	= s3c24xx_init_irq,
 	.timer		= &s3c24xx_timer,
 MACHINE_END
diff --git a/arch/arm/mach-s3c2410/pm-simtec.c b/arch/arm/mach-s3c2410/pm-simtec.c
new file mode 100644
index 000000000000..2cb798832223
--- /dev/null
+++ b/arch/arm/mach-s3c2410/pm-simtec.c
@@ -0,0 +1,65 @@
+/* linux/arch/arm/mach-s3c2410/pm-simtec.c
+ *
+ * Copyright (c) 2004 Simtec Electronics
+ *	Ben Dooks <ben@simtec.co.uk>
+ *
+ * http://armlinux.simtec.co.uk/
+ *
+ * Power Management helpers for Simtec S3C24XX implementations
+ *
+ * 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/types.h>
+#include <linux/interrupt.h>
+#include <linux/list.h>
+#include <linux/timer.h>
+#include <linux/init.h>
+#include <linux/device.h>
+
+#include <asm/mach/arch.h>
+#include <asm/mach/map.h>
+
+#include <asm/hardware.h>
+#include <asm/io.h>
+
+#include <asm/arch/map.h>
+#include <asm/arch/regs-serial.h>
+#include <asm/arch/regs-gpio.h>
+#include <asm/arch/regs-mem.h>
+
+#include <asm/mach-types.h>
+
+#include "pm.h"
+
+#define COPYRIGHT ", (c) 2005 Simtec Electronics"
+
+/* pm_simtec_init
+ *
+ * enable the power management functions
+*/
+
+static __init int pm_simtec_init(void)
+{
+	unsigned long gstatus4;
+
+	/* check which machine we are running on */
+
+	if (!machine_is_bast() && !machine_is_vr1000())
+		return 0;
+
+	printk(KERN_INFO "Simtec Board Power Manangement" COPYRIGHT "\n");
+
+	gstatus4  = (__raw_readl(S3C2410_BANKCON7) & 0x3) << 30;
+	gstatus4 |= (__raw_readl(S3C2410_BANKCON6) & 0x3) << 28;
+	gstatus4 |= (__raw_readl(S3C2410_BANKSIZE) & S3C2410_BANKSIZE_MASK);
+
+	__raw_writel(gstatus4, S3C2410_GSTATUS4);
+
+	return s3c2410_pm_init();
+}
+
+arch_initcall(pm_simtec_init);
diff --git a/arch/arm/mach-versatile/core.c b/arch/arm/mach-versatile/core.c
index 9d1f2253e987..f01c0f8a2bb3 100644
--- a/arch/arm/mach-versatile/core.c
+++ b/arch/arm/mach-versatile/core.c
@@ -33,6 +33,7 @@
 #include <asm/mach-types.h>
 #include <asm/hardware/amba.h>
 #include <asm/hardware/amba_clcd.h>
+#include <asm/hardware/arm_timer.h>
 #include <asm/hardware/icst307.h>
 
 #include <asm/mach/arch.h>
@@ -788,38 +789,25 @@ void __init versatile_init(void)
  */
 #define TIMER_INTERVAL	(TICKS_PER_uSEC * mSEC_10)
 #if TIMER_INTERVAL >= 0x100000
-#define TIMER_RELOAD	(TIMER_INTERVAL >> 8)		/* Divide by 256 */
-#define TIMER_CTRL	0x88				/* Enable, Clock / 256 */
+#define TIMER_RELOAD	(TIMER_INTERVAL >> 8)
+#define TIMER_DIVISOR	(TIMER_CTRL_DIV256)
 #define TICKS2USECS(x)	(256 * (x) / TICKS_PER_uSEC)
 #elif TIMER_INTERVAL >= 0x10000
 #define TIMER_RELOAD	(TIMER_INTERVAL >> 4)		/* Divide by 16 */
-#define TIMER_CTRL	0x84				/* Enable, Clock / 16 */
+#define TIMER_DIVISOR	(TIMER_CTRL_DIV16)
 #define TICKS2USECS(x)	(16 * (x) / TICKS_PER_uSEC)
 #else
 #define TIMER_RELOAD	(TIMER_INTERVAL)
-#define TIMER_CTRL	0x80				/* Enable */
+#define TIMER_DIVISOR	(TIMER_CTRL_DIV1)
 #define TICKS2USECS(x)	((x) / TICKS_PER_uSEC)
 #endif
 
-#define TIMER_CTRL_IE	(1 << 5)	/* Interrupt Enable */
-
-/*
- * What does it look like?
- */
-typedef struct TimerStruct {
-	unsigned long TimerLoad;
-	unsigned long TimerValue;
-	unsigned long TimerControl;
-	unsigned long TimerClear;
-} TimerStruct_t;
-
 /*
  * Returns number of ms since last clock interrupt.  Note that interrupts
  * will have been disabled by do_gettimeoffset()
  */
 static unsigned long versatile_gettimeoffset(void)
 {
-	volatile TimerStruct_t *timer0 = (TimerStruct_t *)TIMER0_VA_BASE;
 	unsigned long ticks1, ticks2, status;
 
 	/*
@@ -828,11 +816,11 @@ static unsigned long versatile_gettimeoffset(void)
 	 * an interrupt.  We get around this by ensuring that the
 	 * counter has not reloaded between our two reads.
 	 */
-	ticks2 = timer0->TimerValue & 0xffff;
+	ticks2 = readl(TIMER0_VA_BASE + TIMER_VALUE) & 0xffff;
 	do {
 		ticks1 = ticks2;
 		status = __raw_readl(VA_IC_BASE + VIC_IRQ_RAW_STATUS);
-		ticks2 = timer0->TimerValue & 0xffff;
+		ticks2 = readl(TIMER0_VA_BASE + TIMER_VALUE) & 0xffff;
 	} while (ticks2 > ticks1);
 
 	/*
@@ -859,12 +847,10 @@ static unsigned long versatile_gettimeoffset(void)
  */
 static irqreturn_t versatile_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 {
-	volatile TimerStruct_t *timer0 = (volatile TimerStruct_t *)TIMER0_VA_BASE;
-
 	write_seqlock(&xtime_lock);
 
 	// ...clear the interrupt
-	timer0->TimerClear = 1;
+	writel(1, TIMER0_VA_BASE + TIMER_INTCLR);
 
 	timer_tick(regs);
 
@@ -884,31 +870,32 @@ static struct irqaction versatile_timer_irq = {
  */
 static void __init versatile_timer_init(void)
 {
-	volatile TimerStruct_t *timer0 = (volatile TimerStruct_t *)TIMER0_VA_BASE;
-	volatile TimerStruct_t *timer1 = (volatile TimerStruct_t *)TIMER1_VA_BASE;
-	volatile TimerStruct_t *timer2 = (volatile TimerStruct_t *)TIMER2_VA_BASE;
-	volatile TimerStruct_t *timer3 = (volatile TimerStruct_t *)TIMER3_VA_BASE;
+	u32 val;
 
 	/* 
 	 * set clock frequency: 
 	 *	VERSATILE_REFCLK is 32KHz
 	 *	VERSATILE_TIMCLK is 1MHz
 	 */
-	*(volatile unsigned int *)IO_ADDRESS(VERSATILE_SCTL_BASE) |= 
-	  ((VERSATILE_TIMCLK << VERSATILE_TIMER1_EnSel) | (VERSATILE_TIMCLK << VERSATILE_TIMER2_EnSel) | 
-	   (VERSATILE_TIMCLK << VERSATILE_TIMER3_EnSel) | (VERSATILE_TIMCLK << VERSATILE_TIMER4_EnSel));
+	val = readl(IO_ADDRESS(VERSATILE_SCTL_BASE));
+	writel((VERSATILE_TIMCLK << VERSATILE_TIMER1_EnSel) |
+	       (VERSATILE_TIMCLK << VERSATILE_TIMER2_EnSel) | 
+	       (VERSATILE_TIMCLK << VERSATILE_TIMER3_EnSel) |
+	       (VERSATILE_TIMCLK << VERSATILE_TIMER4_EnSel) | val,
+	       IO_ADDRESS(VERSATILE_SCTL_BASE));
 
 	/*
 	 * Initialise to a known state (all timers off)
 	 */
-	timer0->TimerControl = 0;
-	timer1->TimerControl = 0;
-	timer2->TimerControl = 0;
-	timer3->TimerControl = 0;
-
-	timer0->TimerLoad    = TIMER_RELOAD;
-	timer0->TimerValue   = TIMER_RELOAD;
-	timer0->TimerControl = TIMER_CTRL | 0x40 | TIMER_CTRL_IE;  /* periodic + IE */
+	writel(0, TIMER0_VA_BASE + TIMER_CTRL);
+	writel(0, TIMER1_VA_BASE + TIMER_CTRL);
+	writel(0, TIMER2_VA_BASE + TIMER_CTRL);
+	writel(0, TIMER3_VA_BASE + TIMER_CTRL);
+
+	writel(TIMER_RELOAD, TIMER0_VA_BASE + TIMER_LOAD);
+	writel(TIMER_RELOAD, TIMER0_VA_BASE + TIMER_VALUE);
+	writel(TIMER_DIVISOR | TIMER_CTRL_ENABLE | TIMER_CTRL_PERIODIC |
+	       TIMER_CTRL_IE, TIMER0_VA_BASE + TIMER_CTRL);
 
 	/* 
 	 * Make irqs happen for the system timer
diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c
index 6dcb23d64bf5..edffa47a4b2a 100644
--- a/arch/arm/mm/init.c
+++ b/arch/arm/mm/init.c
@@ -437,7 +437,7 @@ void __init paging_init(struct meminfo *mi, struct machine_desc *mdesc)
 	memtable_init(mi);
 	if (mdesc->map_io)
 		mdesc->map_io();
-	flush_tlb_all();
+	local_flush_tlb_all();
 
 	/*
 	 * initialise the zones within each node
diff --git a/arch/arm/mm/mm-armv.c b/arch/arm/mm/mm-armv.c
index 052ab443ec4e..c3bd503b43a2 100644
--- a/arch/arm/mm/mm-armv.c
+++ b/arch/arm/mm/mm-armv.c
@@ -682,7 +682,7 @@ void __init memtable_init(struct meminfo *mi)
 	}
 
 	flush_cache_all();
-	flush_tlb_all();
+	local_flush_tlb_all();
 
 	top_pmd = pmd_off_k(0xffff0000);
 }
diff --git a/arch/arm/oprofile/Makefile b/arch/arm/oprofile/Makefile
index ba1a6e9f2b28..8ffb523e6c77 100644
--- a/arch/arm/oprofile/Makefile
+++ b/arch/arm/oprofile/Makefile
@@ -6,6 +6,6 @@ DRIVER_OBJS = $(addprefix ../../../drivers/oprofile/, \
 		oprofilefs.o oprofile_stats.o \
 		timer_int.o )
 
-oprofile-y				:= $(DRIVER_OBJS) init.o
+oprofile-y				:= $(DRIVER_OBJS) init.o backtrace.o
 oprofile-$(CONFIG_CPU_XSCALE)		+= common.o op_model_xscale.o
 
diff --git a/arch/arm/oprofile/backtrace.c b/arch/arm/oprofile/backtrace.c
new file mode 100644
index 000000000000..ec58d3e2eb8b
--- /dev/null
+++ b/arch/arm/oprofile/backtrace.c
@@ -0,0 +1,144 @@
+/*
+ * Arm specific backtracing code for oprofile
+ *
+ * Copyright 2005 Openedhand Ltd.
+ *
+ * Author: Richard Purdie <rpurdie@openedhand.com>
+ *
+ * Based on i386 oprofile backtrace code by John Levon, David Smith
+ *
+ * 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/oprofile.h>
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <asm/ptrace.h>
+#include <asm/uaccess.h>
+
+
+/*
+ * The registers we're interested in are at the end of the variable
+ * length saved register structure. The fp points at the end of this
+ * structure so the address of this struct is:
+ * (struct frame_tail *)(xxx->fp)-1
+ */
+struct frame_tail {
+	struct frame_tail *fp;
+	unsigned long sp;
+	unsigned long lr;
+} __attribute__((packed));
+
+
+#ifdef CONFIG_FRAME_POINTER
+static struct frame_tail* kernel_backtrace(struct frame_tail *tail)
+{
+	oprofile_add_trace(tail->lr);
+
+	/* frame pointers should strictly progress back up the stack
+	 * (towards higher addresses) */
+	if (tail >= tail->fp)
+		return NULL;
+
+	return tail->fp-1;
+}
+#endif
+
+static struct frame_tail* user_backtrace(struct frame_tail *tail)
+{
+	struct frame_tail buftail;
+
+	/* hardware pte might not be valid due to dirty/accessed bit emulation
+	 * so we use copy_from_user and benefit from exception fixups */
+	if (copy_from_user(&buftail, tail, sizeof(struct frame_tail)))
+		return NULL;
+
+	oprofile_add_trace(buftail.lr);
+
+	/* frame pointers should strictly progress back up the stack
+	 * (towards higher addresses) */
+	if (tail >= buftail.fp)
+		return NULL;
+
+	return buftail.fp-1;
+}
+
+/* Compare two addresses and see if they're on the same page */
+#define CMP_ADDR_EQUAL(x,y,offset) ((((unsigned long) x) >> PAGE_SHIFT) \
+	== ((((unsigned long) y) + offset) >> PAGE_SHIFT))
+
+/* check that the page(s) containing the frame tail are present */
+static int pages_present(struct frame_tail *tail)
+{
+	struct mm_struct * mm = current->mm;
+
+	if (!check_user_page_readable(mm, (unsigned long)tail))
+		return 0;
+
+	if (CMP_ADDR_EQUAL(tail, tail, 8))
+		return 1;
+
+	if (!check_user_page_readable(mm, ((unsigned long)tail) + 8))
+		return 0;
+
+	return 1;
+}
+
+/*
+ * |             | /\ Higher addresses
+ * |             |
+ * --------------- stack base (address of current_thread_info)
+ * | thread info |
+ * .             .
+ * |    stack    |
+ * --------------- saved regs->ARM_fp value if valid (frame_tail address)
+ * .             .
+ * --------------- struct pt_regs stored on stack (struct pt_regs *)
+ * |             |
+ * .             .
+ * |             |
+ * --------------- %esp
+ * |             |
+ * |             | \/ Lower addresses
+ *
+ * Thus, &pt_regs <-> stack base restricts the valid(ish) fp values
+ */
+static int valid_kernel_stack(struct frame_tail *tail, struct pt_regs *regs)
+{
+	unsigned long tailaddr = (unsigned long)tail;
+	unsigned long stack = (unsigned long)regs;
+	unsigned long stack_base = (stack & ~(THREAD_SIZE - 1)) + THREAD_SIZE;
+
+	return (tailaddr > stack) && (tailaddr < stack_base);
+}
+
+void arm_backtrace(struct pt_regs const *regs, unsigned int depth)
+{
+	struct frame_tail *tail;
+	unsigned long last_address = 0;
+
+	tail = ((struct frame_tail *) regs->ARM_fp) - 1;
+
+	if (!user_mode(regs)) {
+
+#ifdef CONFIG_FRAME_POINTER
+		while (depth-- && tail && valid_kernel_stack(tail, regs)) {
+			tail = kernel_backtrace(tail);
+		}
+#endif
+		return;
+	}
+
+	while (depth-- && tail && !((unsigned long) tail & 3)) {
+		if ((!CMP_ADDR_EQUAL(last_address, tail, 0)
+			|| !CMP_ADDR_EQUAL(last_address, tail, 8))
+				&& !pages_present(tail))
+			return;
+		last_address = (unsigned long) tail;
+		tail = user_backtrace(tail);
+	}
+}
+
diff --git a/arch/arm/oprofile/init.c b/arch/arm/oprofile/init.c
index cce3d3015eb7..d315a3a86c86 100644
--- a/arch/arm/oprofile/init.c
+++ b/arch/arm/oprofile/init.c
@@ -20,6 +20,8 @@ int __init oprofile_arch_init(struct oprofile_operations *ops)
 	ret = pmu_init(ops, &op_xscale_spec);
 #endif
 
+	ops->backtrace = arm_backtrace;
+
 	return ret;
 }
 
diff --git a/arch/arm/oprofile/op_arm_model.h b/arch/arm/oprofile/op_arm_model.h
index 2d4caf4781ad..2148d07484b7 100644
--- a/arch/arm/oprofile/op_arm_model.h
+++ b/arch/arm/oprofile/op_arm_model.h
@@ -24,6 +24,8 @@ struct op_arm_model_spec {
 extern struct op_arm_model_spec op_xscale_spec;
 #endif
 
+extern void arm_backtrace(struct pt_regs * const regs, unsigned int depth);
+
 extern int __init pmu_init(struct oprofile_operations *ops, struct op_arm_model_spec *spec);
 extern void pmu_exit(void);
 #endif /* OP_ARM_MODEL_H */
diff --git a/arch/arm/vfp/vfp.h b/arch/arm/vfp/vfp.h
index 55a02bc994a3..4b97950984e9 100644
--- a/arch/arm/vfp/vfp.h
+++ b/arch/arm/vfp/vfp.h
@@ -117,7 +117,13 @@ static inline u64 vfp_estimate_div128to64(u64 nh, u64 nl, u64 m)
 	if (nh >= m)
 		return ~0ULL;
 	mh = m >> 32;
-	z = (mh << 32 <= nh) ? 0xffffffff00000000ULL : (nh / mh) << 32;
+	if (mh << 32 <= nh) {
+		z = 0xffffffff00000000ULL;
+	} else {
+		z = nh;
+		do_div(z, mh);
+		z <<= 32;
+	}
 	mul64to128(&termh, &terml, m, z);
 	sub128(&remh, &reml, nh, nl, termh, terml);
 	ml = m << 32;
@@ -126,7 +132,12 @@ static inline u64 vfp_estimate_div128to64(u64 nh, u64 nl, u64 m)
 		add128(&remh, &reml, remh, reml, mh, ml);
 	}
 	remh = (remh << 32) | (reml >> 32);
-	z |= (mh << 32 <= remh) ? 0xffffffff : remh / mh;
+	if (mh << 32 <= remh) {
+		z |= 0xffffffff;
+	} else {
+		do_div(remh, mh);
+		z |= remh;
+	}
 	return z;
 }
 
diff --git a/arch/arm/vfp/vfpdouble.c b/arch/arm/vfp/vfpdouble.c
index fa3053e84db5..b801cd66b6ea 100644
--- a/arch/arm/vfp/vfpdouble.c
+++ b/arch/arm/vfp/vfpdouble.c
@@ -32,6 +32,8 @@
  */
 #include <linux/kernel.h>
 #include <linux/bitops.h>
+
+#include <asm/div64.h>
 #include <asm/ptrace.h>
 #include <asm/vfp.h>
 
diff --git a/arch/arm/vfp/vfpmodule.c b/arch/arm/vfp/vfpmodule.c
index 3aeedd2afc70..22f3da4e0829 100644
--- a/arch/arm/vfp/vfpmodule.c
+++ b/arch/arm/vfp/vfpmodule.c
@@ -89,7 +89,7 @@ void vfp_raise_sigfpe(unsigned int sicode, struct pt_regs *regs)
 	current->thread.error_code = 0;
 	current->thread.trap_no = 6;
 
-	force_sig_info(SIGFPE, &info, current);
+	send_sig_info(SIGFPE, &info, current);
 }
 
 static void vfp_panic(char *reason)
diff --git a/arch/arm/vfp/vfpsingle.c b/arch/arm/vfp/vfpsingle.c
index 6849fe35cb2e..14dd696ddeb1 100644
--- a/arch/arm/vfp/vfpsingle.c
+++ b/arch/arm/vfp/vfpsingle.c
@@ -32,6 +32,8 @@
  */
 #include <linux/kernel.h>
 #include <linux/bitops.h>
+
+#include <asm/div64.h>
 #include <asm/ptrace.h>
 #include <asm/vfp.h>
 
@@ -303,7 +305,11 @@ u32 vfp_estimate_sqrt_significand(u32 exponent, u32 significand)
 		if (z <= a)
 			return (s32)a >> 1;
 	}
-	return (u32)(((u64)a << 31) / z) + (z >> 1);
+	{
+		u64 v = (u64)a << 31;
+		do_div(v, z);
+		return v + (z >> 1);
+	}
 }
 
 static u32 vfp_single_fsqrt(int sd, int unused, s32 m, u32 fpscr)
@@ -1107,7 +1113,11 @@ static u32 vfp_single_fdiv(int sd, int sn, s32 m, u32 fpscr)
 		vsn.significand >>= 1;
 		vsd.exponent++;
 	}
-	vsd.significand = ((u64)vsn.significand << 32) / vsm.significand;
+	{
+		u64 significand = (u64)vsn.significand << 32;
+		do_div(significand, vsm.significand);
+		vsd.significand = significand;
+	}
 	if ((vsd.significand & 0x3f) == 0)
 		vsd.significand |= ((u64)vsm.significand * vsd.significand != (u64)vsn.significand << 32);