1 files changed, 7 insertions, 0 deletions
diff --git a/arch/ia64/include/uapi/asm/resource.h b/arch/ia64/include/uapi/asm/resource.h
new file mode 100644
index 00000000000..ba2272a87fc
--- /dev/null
+++ b/arch/ia64/include/uapi/asm/resource.h
@@ -0,0 +1,7 @@
+#ifndef _ASM_IA64_RESOURCE_H
+#define _ASM_IA64_RESOURCE_H
+
+#include <asm/ustack.h>
+#include <asm-generic/resource.h>
+
+#endif /* _ASM_IA64_RESOURCE_H */
diff --git a/arch/sparc/include/asm/Kbuild b/arch/sparc/include/asm/Kbuild
index 2d2769d766e..a4582181800 100644
--- a/arch/sparc/include/asm/Kbuild
+++ b/arch/sparc/include/asm/Kbuild
@@ -1,41 +1,21 @@
 # User exported sparc header files
-include include/asm-generic/Kbuild.asm
 
-header-y += ipcbuf_32.h
-header-y += ipcbuf_64.h
-header-y += posix_types_32.h
-header-y += posix_types_64.h
-header-y += ptrace_32.h
-header-y += ptrace_64.h
-header-y += sigcontext_32.h
-header-y += sigcontext_64.h
-header-y += siginfo_32.h
-header-y += siginfo_64.h
-header-y += signal_32.h
-header-y += signal_64.h
-header-y += stat_32.h
-header-y += stat_64.h
-header-y += unistd_32.h
-header-y += unistd_64.h
 
-header-y += apc.h
-header-y += asi.h
-header-y += display7seg.h
-header-y += envctrl.h
-header-y += fbio.h
-header-y += jsflash.h
-header-y += openprom.h
-header-y += openprom_32.h
-header-y += openprom_64.h
-header-y += openpromio.h
-header-y += perfctr.h
-header-y += psrcompat.h
-header-y += psr.h
-header-y += pstate.h
-header-y += reg.h
-header-y += reg_32.h
-header-y += reg_64.h
-header-y += traps.h
-header-y += uctx.h
-header-y += utrap.h
-header-y += watchdog.h
+generic-y += clkdev.h
+generic-y += cputime.h
+generic-y += div64.h
+generic-y += emergency-restart.h
+generic-y += exec.h
+generic-y += hash.h
+generic-y += irq_regs.h
+generic-y += linkage.h
+generic-y += local.h
+generic-y += local64.h
+generic-y += mcs_spinlock.h
+generic-y += module.h
+generic-y += mutex.h
+generic-y += preempt.h
+generic-y += serial.h
+generic-y += trace_clock.h
+generic-y += types.h
+generic-y += word-at-a-time.h
diff --git a/arch/sparc/include/asm/agp.h b/arch/sparc/include/asm/agp.h
index c2456870b05..70f52c1661b 100644
--- a/arch/sparc/include/asm/agp.h
+++ b/arch/sparc/include/asm/agp.h
@@ -7,10 +7,6 @@
 #define unmap_page_from_agp(page)
 #define flush_agp_cache() mb()
 
-/* Convert a physical address to an address suitable for the GART. */
-#define phys_to_gart(x) (x)
-#define gart_to_phys(x) (x)
-
 /* GATT allocation. Returns/accepts GATT kernel virtual address. */
 #define alloc_gatt_pages(order)		\
 	((char *)__get_free_pages(GFP_KERNEL, (order)))
diff --git a/arch/sparc/include/asm/asm-offsets.h b/arch/sparc/include/asm/asm-offsets.h
new file mode 100644
index 00000000000..d370ee36a18
--- /dev/null
+++ b/arch/sparc/include/asm/asm-offsets.h
@@ -0,0 +1 @@
+#include <generated/asm-offsets.h>
diff --git a/arch/sparc/include/asm/asm.h b/arch/sparc/include/asm/asm.h
new file mode 100644
index 00000000000..e8e1d94b4cc
--- /dev/null
+++ b/arch/sparc/include/asm/asm.h
@@ -0,0 +1,40 @@
+#ifndef _SPARC_ASM_H
+#define _SPARC_ASM_H
+
+/* Macros to assist the sharing of assembler code between 32-bit and
+ * 64-bit sparc.
+ */
+
+#ifdef CONFIG_SPARC64
+#define BRANCH32(TYPE, PREDICT, DEST) \
+	TYPE,PREDICT	%icc, DEST
+#define BRANCH32_ANNUL(TYPE, PREDICT, DEST) \
+	TYPE,a,PREDICT	%icc, DEST
+#define BRANCH_REG_ZERO(PREDICT, REG, DEST) \
+	brz,PREDICT	REG, DEST
+#define BRANCH_REG_ZERO_ANNUL(PREDICT, REG, DEST) \
+	brz,a,PREDICT	REG, DEST
+#define BRANCH_REG_NOT_ZERO(PREDICT, REG, DEST) \
+	brnz,PREDICT	REG, DEST
+#define BRANCH_REG_NOT_ZERO_ANNUL(PREDICT, REG, DEST) \
+	brnz,a,PREDICT	REG, DEST
+#else
+#define BRANCH32(TYPE, PREDICT, DEST) \
+	TYPE		DEST
+#define BRANCH32_ANNUL(TYPE, PREDICT, DEST) \
+	TYPE,a		DEST
+#define BRANCH_REG_ZERO(PREDICT, REG, DEST) \
+	cmp		REG, 0; \
+	be		DEST
+#define BRANCH_REG_ZERO_ANNUL(PREDICT, REG, DEST) \
+	cmp		REG, 0; \
+	be,a		DEST
+#define BRANCH_REG_NOT_ZERO(PREDICT, REG, DEST) \
+	cmp		REG, 0; \
+	bne		DEST
+#define BRANCH_REG_NOT_ZERO_ANNUL(PREDICT, REG, DEST) \
+	cmp		REG, 0; \
+	bne,a		DEST
+#endif
+
+#endif /* _SPARC_ASM_H */
diff --git a/arch/sparc/include/asm/asmmacro.h b/arch/sparc/include/asm/asmmacro.h
index a995bf8aba3..a0e28ef0255 100644
--- a/arch/sparc/include/asm/asmmacro.h
+++ b/arch/sparc/include/asm/asmmacro.h
@@ -6,17 +6,6 @@
 #ifndef _SPARC_ASMMACRO_H
 #define _SPARC_ASMMACRO_H
 
-#include <asm/btfixup.h>
-#include <asm/asi.h>
-
-#define GET_PROCESSOR4M_ID(reg) \
-	rd	%tbr, %reg; \
-	srl	%reg, 12, %reg; \
-	and	%reg, 3, %reg;
-
-#define GET_PROCESSOR4D_ID(reg) \
-	lda	[%g0] ASI_M_VIKING_TMP1, %reg;
-
 /* All trap entry points _must_ begin with this macro or else you
  * lose.  It makes sure the kernel has a proper window so that
  * c-code can be called.
@@ -31,10 +20,26 @@
 /* All traps low-level code here must end with this macro. */
 #define RESTORE_ALL b ret_trap_entry; clr %l6;
 
-/* sun4 probably wants half word accesses to ASI_SEGMAP, while sun4c+
-   likes byte accesses. These are to avoid ifdef mania. */
-
-#define lduXa	lduba
-#define stXa	stba
+/* Support for run-time patching of single instructions.
+ * This is used to handle the differences in the ASI for
+ * MMUREGS for LEON and SUN.
+ *
+ * Sample:
+ * LEON_PI(lda [%g0] ASI_LEON_MMUREGS, %o0
+ * SUN_PI_(lda [%g0] ASI_M_MMUREGS, %o0
+ * PI == Patch Instruction
+ *
+ * For LEON we will use the first variant,
+ * and for all other we will use the SUN variant.
+ * The order is important.
+ */
+#define LEON_PI(...)				\
+662:	__VA_ARGS__
+
+#define SUN_PI_(...)				\
+	.section .leon_1insn_patch, "ax";	\
+	.word 662b;				\
+	__VA_ARGS__;				\
+	.previous
 
 #endif /* !(_SPARC_ASMMACRO_H) */
diff --git a/arch/sparc/include/asm/atomic_32.h b/arch/sparc/include/asm/atomic_32.h
index 5c944b5a804..7aed2be45b4 100644
--- a/arch/sparc/include/asm/atomic_32.h
+++ b/arch/sparc/include/asm/atomic_32.h
@@ -13,19 +13,20 @@
 
 #include <linux/types.h>
 
-typedef struct { volatile int counter; } atomic_t;
+#include <asm/cmpxchg.h>
+#include <asm/barrier.h>
+#include <asm-generic/atomic64.h>
 
-#ifdef __KERNEL__
 
 #define ATOMIC_INIT(i)  { (i) }
 
-extern int __atomic_add_return(int, atomic_t *);
-extern int atomic_cmpxchg(atomic_t *, int, int);
+int __atomic_add_return(int, atomic_t *);
+int atomic_cmpxchg(atomic_t *, int, int);
 #define atomic_xchg(v, new) (xchg(&((v)->counter), new))
-extern int atomic_add_unless(atomic_t *, int, int);
-extern void atomic_set(atomic_t *, int);
+int __atomic_add_unless(atomic_t *, int, int);
+void atomic_set(atomic_t *, int);
 
-#define atomic_read(v)          ((v)->counter)
+#define atomic_read(v)          (*(volatile int *)&(v)->counter)
 
 #define atomic_add(i, v)	((void)__atomic_add_return( (int)(i), (v)))
 #define atomic_sub(i, v)	((void)__atomic_add_return(-(int)(i), (v)))
@@ -52,114 +53,4 @@ extern void atomic_set(atomic_t *, int);
 #define atomic_dec_and_test(v) (atomic_dec_return(v) == 0)
 #define atomic_sub_and_test(i, v) (atomic_sub_return(i, v) == 0)
 
-#define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
-
-/* This is the old 24-bit implementation.  It's still used internally
- * by some sparc-specific code, notably the semaphore implementation.
- */
-typedef struct { volatile int counter; } atomic24_t;
-
-#ifndef CONFIG_SMP
-
-#define ATOMIC24_INIT(i)  { (i) }
-#define atomic24_read(v)          ((v)->counter)
-#define atomic24_set(v, i)        (((v)->counter) = i)
-
-#else
-/* We do the bulk of the actual work out of line in two common
- * routines in assembler, see arch/sparc/lib/atomic.S for the
- * "fun" details.
- *
- * For SMP the trick is you embed the spin lock byte within
- * the word, use the low byte so signedness is easily retained
- * via a quick arithmetic shift.  It looks like this:
- *
- *	----------------------------------------
- *	| signed 24-bit counter value |  lock  |  atomic_t
- *	----------------------------------------
- *	 31                          8 7      0
- */
-
-#define ATOMIC24_INIT(i)	{ ((i) << 8) }
-
-static inline int atomic24_read(const atomic24_t *v)
-{
-	int ret = v->counter;
-
-	while(ret & 0xff)
-		ret = v->counter;
-
-	return ret >> 8;
-}
-
-#define atomic24_set(v, i)	(((v)->counter) = ((i) << 8))
-#endif
-
-static inline int __atomic24_add(int i, atomic24_t *v)
-{
-	register volatile int *ptr asm("g1");
-	register int increment asm("g2");
-	register int tmp1 asm("g3");
-	register int tmp2 asm("g4");
-	register int tmp3 asm("g7");
-
-	ptr = &v->counter;
-	increment = i;
-
-	__asm__ __volatile__(
-	"mov	%%o7, %%g4\n\t"
-	"call	___atomic24_add\n\t"
-	" add	%%o7, 8, %%o7\n"
-	: "=&r" (increment), "=r" (tmp1), "=r" (tmp2), "=r" (tmp3)
-	: "0" (increment), "r" (ptr)
-	: "memory", "cc");
-
-	return increment;
-}
-
-static inline int __atomic24_sub(int i, atomic24_t *v)
-{
-	register volatile int *ptr asm("g1");
-	register int increment asm("g2");
-	register int tmp1 asm("g3");
-	register int tmp2 asm("g4");
-	register int tmp3 asm("g7");
-
-	ptr = &v->counter;
-	increment = i;
-
-	__asm__ __volatile__(
-	"mov	%%o7, %%g4\n\t"
-	"call	___atomic24_sub\n\t"
-	" add	%%o7, 8, %%o7\n"
-	: "=&r" (increment), "=r" (tmp1), "=r" (tmp2), "=r" (tmp3)
-	: "0" (increment), "r" (ptr)
-	: "memory", "cc");
-
-	return increment;
-}
-
-#define atomic24_add(i, v) ((void)__atomic24_add((i), (v)))
-#define atomic24_sub(i, v) ((void)__atomic24_sub((i), (v)))
-
-#define atomic24_dec_return(v) __atomic24_sub(1, (v))
-#define atomic24_inc_return(v) __atomic24_add(1, (v))
-
-#define atomic24_sub_and_test(i, v) (__atomic24_sub((i), (v)) == 0)
-#define atomic24_dec_and_test(v) (__atomic24_sub(1, (v)) == 0)
-
-#define atomic24_inc(v) ((void)__atomic24_add(1, (v)))
-#define atomic24_dec(v) ((void)__atomic24_sub(1, (v)))
-
-#define atomic24_add_negative(i, v) (__atomic24_add((i), (v)) < 0)
-
-/* Atomic operations are already serializing */
-#define smp_mb__before_atomic_dec()	barrier()
-#define smp_mb__after_atomic_dec()	barrier()
-#define smp_mb__before_atomic_inc()	barrier()
-#define smp_mb__after_atomic_inc()	barrier()
-
-#endif /* !(__KERNEL__) */
-
-#include <asm-generic/atomic.h>
 #endif /* !(__ARCH_SPARC_ATOMIC__) */
diff --git a/arch/sparc/include/asm/atomic_64.h b/arch/sparc/include/asm/atomic_64.h
index 2c71ec4a3b1..bb894c8bec5 100644
--- a/arch/sparc/include/asm/atomic_64.h
+++ b/arch/sparc/include/asm/atomic_64.h
@@ -1,36 +1,34 @@
 /* atomic.h: Thankfully the V9 is at least reasonable for this
  *           stuff.
  *
- * Copyright (C) 1996, 1997, 2000 David S. Miller (davem@redhat.com)
+ * Copyright (C) 1996, 1997, 2000, 2012 David S. Miller (davem@redhat.com)
  */
 
 #ifndef __ARCH_SPARC64_ATOMIC__
 #define __ARCH_SPARC64_ATOMIC__
 
 #include <linux/types.h>
-#include <asm/system.h>
-
-typedef struct { volatile int counter; } atomic_t;
-typedef struct { volatile __s64 counter; } atomic64_t;
+#include <asm/cmpxchg.h>
+#include <asm/barrier.h>
 
 #define ATOMIC_INIT(i)		{ (i) }
 #define ATOMIC64_INIT(i)	{ (i) }
 
-#define atomic_read(v)		((v)->counter)
-#define atomic64_read(v)	((v)->counter)
+#define atomic_read(v)		(*(volatile int *)&(v)->counter)
+#define atomic64_read(v)	(*(volatile long *)&(v)->counter)
 
 #define atomic_set(v, i)	(((v)->counter) = i)
 #define atomic64_set(v, i)	(((v)->counter) = i)
 
-extern void atomic_add(int, atomic_t *);
-extern void atomic64_add(int, atomic64_t *);
-extern void atomic_sub(int, atomic_t *);
-extern void atomic64_sub(int, atomic64_t *);
+void atomic_add(int, atomic_t *);
+void atomic64_add(long, atomic64_t *);
+void atomic_sub(int, atomic_t *);
+void atomic64_sub(long, atomic64_t *);
 
-extern int atomic_add_ret(int, atomic_t *);
-extern int atomic64_add_ret(int, atomic64_t *);
-extern int atomic_sub_ret(int, atomic_t *);
-extern int atomic64_sub_ret(int, atomic64_t *);
+int atomic_add_ret(int, atomic_t *);
+long atomic64_add_ret(long, atomic64_t *);
+int atomic_sub_ret(int, atomic_t *);
+long atomic64_sub_ret(long, atomic64_t *);
 
 #define atomic_dec_return(v) atomic_sub_ret(1, v)
 #define atomic64_dec_return(v) atomic64_sub_ret(1, v)
@@ -73,7 +71,7 @@ extern int atomic64_sub_ret(int, atomic64_t *);
 #define atomic_cmpxchg(v, o, n) (cmpxchg(&((v)->counter), (o), (n)))
 #define atomic_xchg(v, new) (xchg(&((v)->counter), new))
 
-static inline int atomic_add_unless(atomic_t *v, int a, int u)
+static inline int __atomic_add_unless(atomic_t *v, int a, int u)
 {
 	int c, old;
 	c = atomic_read(v);
@@ -85,16 +83,14 @@ static inline int atomic_add_unless(atomic_t *v, int a, int u)
 			break;
 		c = old;
 	}
-	return c != (u);
+	return c;
 }
 
-#define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
-
 #define atomic64_cmpxchg(v, o, n) \
 	((__typeof__((v)->counter))cmpxchg(&((v)->counter), (o), (n)))
 #define atomic64_xchg(v, new) (xchg(&((v)->counter), new))
 
-static inline int atomic64_add_unless(atomic64_t *v, long a, long u)
+static inline long atomic64_add_unless(atomic64_t *v, long a, long u)
 {
 	long c, old;
 	c = atomic64_read(v);
@@ -111,18 +107,6 @@ static inline int atomic64_add_unless(atomic64_t *v, long a, long u)
 
 #define atomic64_inc_not_zero(v) atomic64_add_unless((v), 1, 0)
 
-/* Atomic operations are already serializing */
-#ifdef CONFIG_SMP
-#define smp_mb__before_atomic_dec()	membar_storeload_loadload();
-#define smp_mb__after_atomic_dec()	membar_storeload_storestore();
-#define smp_mb__before_atomic_inc()	membar_storeload_loadload();
-#define smp_mb__after_atomic_inc()	membar_storeload_storestore();
-#else
-#define smp_mb__before_atomic_dec()	barrier()
-#define smp_mb__after_atomic_dec()	barrier()
-#define smp_mb__before_atomic_inc()	barrier()
-#define smp_mb__after_atomic_inc()	barrier()
-#endif
-
-#include <asm-generic/atomic.h>
+long atomic64_dec_if_positive(atomic64_t *v);
+
 #endif /* !(__ARCH_SPARC64_ATOMIC__) */
diff --git a/arch/sparc/include/asm/auxio.h b/arch/sparc/include/asm/auxio.h
index 13dc67f0301..3e09a07b77e 100644
--- a/arch/sparc/include/asm/auxio.h
+++ b/arch/sparc/include/asm/auxio.h
@@ -1,5 +1,12 @@
 #ifndef ___ASM_SPARC_AUXIO_H
 #define ___ASM_SPARC_AUXIO_H
+
+#ifndef __ASSEMBLY__
+
+extern void __iomem *auxio_register;
+
+#endif /* ifndef __ASSEMBLY__ */
+
 #if defined(__sparc__) && defined(__arch64__)
 #include <asm/auxio_64.h>
 #else
diff --git a/arch/sparc/include/asm/auxio_32.h b/arch/sparc/include/asm/auxio_32.h
index e03e088be95..5d685df427b 100644
--- a/arch/sparc/include/asm/auxio_32.h
+++ b/arch/sparc/include/asm/auxio_32.h
@@ -6,7 +6,6 @@
 #ifndef _SPARC_AUXIO_H
 #define _SPARC_AUXIO_H
 
-#include <asm/system.h>
 #include <asm/vaddrs.h>
 
 /* This register is an unsigned char in IO space.  It does two things.
@@ -35,8 +34,8 @@
  * NOTE: these routines are implementation dependent--
  * understand the hardware you are querying!
  */
-extern void set_auxio(unsigned char bits_on, unsigned char bits_off);
-extern unsigned char get_auxio(void); /* .../asm/floppy.h */
+void set_auxio(unsigned char bits_on, unsigned char bits_off);
+unsigned char get_auxio(void); /* .../asm/floppy.h */
 
 /*
  * The following routines are provided for driver-compatibility
@@ -79,7 +78,7 @@ do { \
 
 
 /* AUXIO2 (Power Off Control) */
-extern __volatile__ unsigned char * auxio_power_register;
+extern volatile u8 __iomem *auxio_power_register;
 
 #define	AUXIO_POWER_DETECT_FAILURE	32
 #define	AUXIO_POWER_CLEAR_FAILURE	2
diff --git a/arch/sparc/include/asm/auxio_64.h b/arch/sparc/include/asm/auxio_64.h
index f61cd1e3e39..6079e59a7ad 100644
--- a/arch/sparc/include/asm/auxio_64.h
+++ b/arch/sparc/include/asm/auxio_64.h
@@ -75,8 +75,6 @@
 
 #ifndef __ASSEMBLY__
 
-extern void __iomem *auxio_register;
-
 #define AUXIO_LTE_ON	1
 #define AUXIO_LTE_OFF	0
 
@@ -84,7 +82,7 @@ extern void __iomem *auxio_register;
  *
  * on - AUXIO_LTE_ON or AUXIO_LTE_OFF
  */
-extern void auxio_set_lte(int on);
+void auxio_set_lte(int on);
 
 #define AUXIO_LED_ON	1
 #define AUXIO_LED_OFF	0
@@ -93,7 +91,7 @@ extern void auxio_set_lte(int on);
  *
  * on - AUXIO_LED_ON or AUXIO_LED_OFF
  */
-extern void auxio_set_led(int on);
+void auxio_set_led(int on);
 
 #endif /* ifndef __ASSEMBLY__ */
 
diff --git a/arch/sparc/include/asm/backoff.h b/arch/sparc/include/asm/backoff.h
index fa1fdf67e35..4e02086b839 100644
--- a/arch/sparc/include/asm/backoff.h
+++ b/arch/sparc/include/asm/backoff.h
@@ -1,6 +1,46 @@
 #ifndef _SPARC64_BACKOFF_H
 #define _SPARC64_BACKOFF_H
 
+/* The macros in this file implement an exponential backoff facility
+ * for atomic operations.
+ *
+ * When multiple threads compete on an atomic operation, it is
+ * possible for one thread to be continually denied a successful
+ * completion of the compare-and-swap instruction.  Heavily
+ * threaded cpu implementations like Niagara can compound this
+ * problem even further.
+ *
+ * When an atomic operation fails and needs to be retried, we spin a
+ * certain number of times.  At each subsequent failure of the same
+ * operation we double the spin count, realizing an exponential
+ * backoff.
+ *
+ * When we spin, we try to use an operation that will cause the
+ * current cpu strand to block, and therefore make the core fully
+ * available to any other other runnable strands.  There are two
+ * options, based upon cpu capabilities.
+ *
+ * On all cpus prior to SPARC-T4 we do three dummy reads of the
+ * condition code register.  Each read blocks the strand for something
+ * between 40 and 50 cpu cycles.
+ *
+ * For SPARC-T4 and later we have a special "pause" instruction
+ * available.  This is implemented using writes to register %asr27.
+ * The cpu will block the number of cycles written into the register,
+ * unless a disrupting trap happens first.  SPARC-T4 specifically
+ * implements pause with a granularity of 8 cycles.  Each strand has
+ * an internal pause counter which decrements every 8 cycles.  So the
+ * chip shifts the %asr27 value down by 3 bits, and writes the result
+ * into the pause counter.  If a value smaller than 8 is written, the
+ * chip blocks for 1 cycle.
+ *
+ * To achieve the same amount of backoff as the three %ccr reads give
+ * on earlier chips, we shift the backoff value up by 7 bits.  (Three
+ * %ccr reads block for about 128 cycles, 1 << 7 == 128) We write the
+ * whole amount we want to block into the pause register, rather than
+ * loop writing 128 each time.
+ */
+
 #define BACKOFF_LIMIT	(4 * 1024)
 
 #ifdef CONFIG_SMP
@@ -8,23 +48,37 @@
 #define BACKOFF_SETUP(reg)	\
 	mov	1, reg
 
-#define BACKOFF_SPIN(reg, tmp, label)	\
-	mov	reg, tmp; \
-88:	brnz,pt	tmp, 88b; \
-	 sub	tmp, 1, tmp; \
-	set	BACKOFF_LIMIT, tmp; \
-	cmp	reg, tmp; \
-	bg,pn	%xcc, label; \
-	 nop; \
-	ba,pt	%xcc, label; \
-	 sllx	reg, 1, reg;
+#define BACKOFF_LABEL(spin_label, continue_label) \
+	spin_label
+
+#define BACKOFF_SPIN(reg, tmp, label)		\
+	mov		reg, tmp;		\
+88:	rd		%ccr, %g0;		\
+	rd		%ccr, %g0;		\
+	rd		%ccr, %g0;		\
+	.section	.pause_3insn_patch,"ax";\
+	.word		88b;			\
+	sllx		tmp, 7, tmp;		\
+	wr		tmp, 0, %asr27;		\
+	clr		tmp;			\
+	.previous;				\
+	brnz,pt		tmp, 88b;		\
+	 sub		tmp, 1, tmp;		\
+	set		BACKOFF_LIMIT, tmp;	\
+	cmp		reg, tmp;		\
+	bg,pn		%xcc, label;		\
+	 nop;					\
+	ba,pt		%xcc, label;		\
+	 sllx		reg, 1, reg;
 
 #else
 
 #define BACKOFF_SETUP(reg)
-#define BACKOFF_SPIN(reg, tmp, label) \
-	ba,pt	%xcc, label; \
-	 nop;
+
+#define BACKOFF_LABEL(spin_label, continue_label) \
+	continue_label
+
+#define BACKOFF_SPIN(reg, tmp, label)
 
 #endif
 
diff --git a/arch/sparc/include/asm/barrier.h b/arch/sparc/include/asm/barrier.h
new file mode 100644
index 00000000000..b25f02a029e
--- /dev/null
+++ b/arch/sparc/include/asm/barrier.h
@@ -0,0 +1,8 @@
+#ifndef ___ASM_SPARC_BARRIER_H
+#define ___ASM_SPARC_BARRIER_H
+#if defined(__sparc__) && defined(__arch64__)
+#include <asm/barrier_64.h>
+#else
+#include <asm/barrier_32.h>
+#endif
+#endif
diff --git a/arch/sparc/include/asm/barrier_32.h b/arch/sparc/include/asm/barrier_32.h
new file mode 100644
index 00000000000..ae69eda288f
--- /dev/null
+++ b/arch/sparc/include/asm/barrier_32.h
@@ -0,0 +1,7 @@
+#ifndef __SPARC_BARRIER_H
+#define __SPARC_BARRIER_H
+
+#include <asm/processor.h> /* for nop() */
+#include <asm-generic/barrier.h>
+
+#endif /* !(__SPARC_BARRIER_H) */
diff --git a/arch/sparc/include/asm/barrier_64.h b/arch/sparc/include/asm/barrier_64.h
new file mode 100644
index 00000000000..305dcc3dc72
--- /dev/null
+++ b/arch/sparc/include/asm/barrier_64.h
@@ -0,0 +1,74 @@
+#ifndef __SPARC64_BARRIER_H
+#define __SPARC64_BARRIER_H
+
+/* These are here in an effort to more fully work around Spitfire Errata
+ * #51.  Essentially, if a memory barrier occurs soon after a mispredicted
+ * branch, the chip can stop executing instructions until a trap occurs.
+ * Therefore, if interrupts are disabled, the chip can hang forever.
+ *
+ * It used to be believed that the memory barrier had to be right in the
+ * delay slot, but a case has been traced recently wherein the memory barrier
+ * was one instruction after the branch delay slot and the chip still hung.
+ * The offending sequence was the following in sym_wakeup_done() of the
+ * sym53c8xx_2 driver:
+ *
+ *	call	sym_ccb_from_dsa, 0
+ *	 movge	%icc, 0, %l0
+ *	brz,pn	%o0, .LL1303
+ *	 mov	%o0, %l2
+ *	membar	#LoadLoad
+ *
+ * The branch has to be mispredicted for the bug to occur.  Therefore, we put
+ * the memory barrier explicitly into a "branch always, predicted taken"
+ * delay slot to avoid the problem case.
+ */
+#define membar_safe(type) \
+do {	__asm__ __volatile__("ba,pt	%%xcc, 1f\n\t" \
+			     " membar	" type "\n" \
+			     "1:\n" \
+			     : : : "memory"); \
+} while (0)
+
+/* The kernel always executes in TSO memory model these days,
+ * and furthermore most sparc64 chips implement more stringent
+ * memory ordering than required by the specifications.
+ */
+#define mb()	membar_safe("#StoreLoad")
+#define rmb()	__asm__ __volatile__("":::"memory")
+#define wmb()	__asm__ __volatile__("":::"memory")
+
+#define read_barrier_depends()		do { } while(0)
+#define set_mb(__var, __value) \
+	do { __var = __value; membar_safe("#StoreLoad"); } while(0)
+
+#ifdef CONFIG_SMP
+#define smp_mb()	mb()
+#define smp_rmb()	rmb()
+#define smp_wmb()	wmb()
+#else
+#define smp_mb()	__asm__ __volatile__("":::"memory")
+#define smp_rmb()	__asm__ __volatile__("":::"memory")
+#define smp_wmb()	__asm__ __volatile__("":::"memory")
+#endif
+
+#define smp_read_barrier_depends()	do { } while(0)
+
+#define smp_store_release(p, v)						\
+do {									\
+	compiletime_assert_atomic_type(*p);				\
+	barrier();							\
+	ACCESS_ONCE(*p) = (v);						\
+} while (0)
+
+#define smp_load_acquire(p)						\
+({									\
+	typeof(*p) ___p1 = ACCESS_ONCE(*p);				\
+	compiletime_assert_atomic_type(*p);				\
+	barrier();							\
+	___p1;								\
+})
+
+#define smp_mb__before_atomic()	barrier()
+#define smp_mb__after_atomic()	barrier()
+
+#endif /* !(__SPARC64_BARRIER_H) */
diff --git a/arch/sparc/include/asm/bitext.h b/arch/sparc/include/asm/bitext.h
index 297b2f2fcb4..9c988bf3adb 100644
--- a/arch/sparc/include/asm/bitext.h
+++ b/arch/sparc/include/asm/bitext.h
@@ -20,8 +20,8 @@ struct bit_map {
 	int num_colors;
 };
 
-extern int bit_map_string_get(struct bit_map *t, int len, int align);
-extern void bit_map_clear(struct bit_map *t, int offset, int len);
-extern void bit_map_init(struct bit_map *t, unsigned long *map, int size);
+int bit_map_string_get(struct bit_map *t, int len, int align);
+void bit_map_clear(struct bit_map *t, int offset, int len);
+void bit_map_init(struct bit_map *t, unsigned long *map, int size);
 
 #endif /* defined(_SPARC_BITEXT_H) */
diff --git a/arch/sparc/include/asm/bitops_32.h b/arch/sparc/include/asm/bitops_32.h
index 9cf4ae0cd7b..600ed1d9c8c 100644
--- a/arch/sparc/include/asm/bitops_32.h
+++ b/arch/sparc/include/asm/bitops_32.h
@@ -18,9 +18,9 @@
 #error only <linux/bitops.h> can be included directly
 #endif
 
-extern unsigned long ___set_bit(unsigned long *addr, unsigned long mask);
-extern unsigned long ___clear_bit(unsigned long *addr, unsigned long mask);
-extern unsigned long ___change_bit(unsigned long *addr, unsigned long mask);
+unsigned long ___set_bit(unsigned long *addr, unsigned long mask);
+unsigned long ___clear_bit(unsigned long *addr, unsigned long mask);
+unsigned long ___change_bit(unsigned long *addr, unsigned long mask);
 
 /*
  * Set bit 'nr' in 32-bit quantity at address 'addr' where bit '0'
@@ -90,9 +90,6 @@ static inline void change_bit(unsigned long nr, volatile unsigned long *addr)
 
 #include <asm-generic/bitops/non-atomic.h>
 
-#define smp_mb__before_clear_bit()	do { } while(0)
-#define smp_mb__after_clear_bit()	do { } while(0)
-
 #include <asm-generic/bitops/ffz.h>
 #include <asm-generic/bitops/__ffs.h>
 #include <asm-generic/bitops/sched.h>
@@ -103,9 +100,8 @@ static inline void change_bit(unsigned long nr, volatile unsigned long *addr)
 #include <asm-generic/bitops/hweight.h>
 #include <asm-generic/bitops/lock.h>
 #include <asm-generic/bitops/find.h>
-#include <asm-generic/bitops/ext2-non-atomic.h>
+#include <asm-generic/bitops/le.h>
 #include <asm-generic/bitops/ext2-atomic.h>
-#include <asm-generic/bitops/minix.h>
 
 #endif /* __KERNEL__ */
 
diff --git a/arch/sparc/include/asm/bitops_64.h b/arch/sparc/include/asm/bitops_64.h
index bb87b808022..2d522402a93 100644
--- a/arch/sparc/include/asm/bitops_64.h
+++ b/arch/sparc/include/asm/bitops_64.h
@@ -13,79 +13,40 @@
 
 #include <linux/compiler.h>
 #include <asm/byteorder.h>
+#include <asm/barrier.h>
 
-extern int test_and_set_bit(unsigned long nr, volatile unsigned long *addr);
-extern int test_and_clear_bit(unsigned long nr, volatile unsigned long *addr);
-extern int test_and_change_bit(unsigned long nr, volatile unsigned long *addr);
-extern void set_bit(unsigned long nr, volatile unsigned long *addr);
-extern void clear_bit(unsigned long nr, volatile unsigned long *addr);
-extern void change_bit(unsigned long nr, volatile unsigned long *addr);
+int test_and_set_bit(unsigned long nr, volatile unsigned long *addr);
+int test_and_clear_bit(unsigned long nr, volatile unsigned long *addr);
+int test_and_change_bit(unsigned long nr, volatile unsigned long *addr);
+void set_bit(unsigned long nr, volatile unsigned long *addr);
+void clear_bit(unsigned long nr, volatile unsigned long *addr);
+void change_bit(unsigned long nr, volatile unsigned long *addr);
 
 #include <asm-generic/bitops/non-atomic.h>
 
-#ifdef CONFIG_SMP
-#define smp_mb__before_clear_bit()	membar_storeload_loadload()
-#define smp_mb__after_clear_bit()	membar_storeload_storestore()
-#else
-#define smp_mb__before_clear_bit()	barrier()
-#define smp_mb__after_clear_bit()	barrier()
-#endif
-
-#include <asm-generic/bitops/ffz.h>
-#include <asm-generic/bitops/__ffs.h>
 #include <asm-generic/bitops/fls.h>
 #include <asm-generic/bitops/__fls.h>
 #include <asm-generic/bitops/fls64.h>
 
 #ifdef __KERNEL__
 
+int ffs(int x);
+unsigned long __ffs(unsigned long);
+
+#include <asm-generic/bitops/ffz.h>
 #include <asm-generic/bitops/sched.h>
-#include <asm-generic/bitops/ffs.h>
 
 /*
  * hweightN: returns the hamming weight (i.e. the number
  * of bits set) of a N-bit word
  */
 
-#ifdef ULTRA_HAS_POPULATION_COUNT
-
-static inline unsigned int hweight64(unsigned long w)
-{
-	unsigned int res;
-
-	__asm__ ("popc %1,%0" : "=r" (res) : "r" (w));
-	return res;
-}
-
-static inline unsigned int hweight32(unsigned int w)
-{
-	unsigned int res;
-
-	__asm__ ("popc %1,%0" : "=r" (res) : "r" (w & 0xffffffff));
-	return res;
-}
+unsigned long __arch_hweight64(__u64 w);
+unsigned int __arch_hweight32(unsigned int w);
+unsigned int __arch_hweight16(unsigned int w);
+unsigned int __arch_hweight8(unsigned int w);
 
-static inline unsigned int hweight16(unsigned int w)
-{
-	unsigned int res;
-
-	__asm__ ("popc %1,%0" : "=r" (res) : "r" (w & 0xffff));
-	return res;
-}
-
-static inline unsigned int hweight8(unsigned int w)
-{
-	unsigned int res;
-
-	__asm__ ("popc %1,%0" : "=r" (res) : "r" (w & 0xff));
-	return res;
-}
-
-#else
-
-#include <asm-generic/bitops/hweight.h>
-
-#endif
+#include <asm-generic/bitops/const_hweight.h>
 #include <asm-generic/bitops/lock.h>
 #endif /* __KERNEL__ */
 
@@ -93,14 +54,9 @@ static inline unsigned int hweight8(unsigned int w)
 
 #ifdef __KERNEL__
 
-#include <asm-generic/bitops/ext2-non-atomic.h>
-
-#define ext2_set_bit_atomic(lock,nr,addr) \
-	test_and_set_bit((nr) ^ 0x38,(unsigned long *)(addr))
-#define ext2_clear_bit_atomic(lock,nr,addr) \
-	test_and_clear_bit((nr) ^ 0x38,(unsigned long *)(addr))
+#include <asm-generic/bitops/le.h>
 
-#include <asm-generic/bitops/minix.h>
+#include <asm-generic/bitops/ext2-atomic-setbit.h>
 
 #endif /* __KERNEL__ */
 
diff --git a/arch/sparc/include/asm/btext.h b/arch/sparc/include/asm/btext.h
new file mode 100644
index 00000000000..75a32b109e1
--- /dev/null
+++ b/arch/sparc/include/asm/btext.h
@@ -0,0 +1,6 @@
+#ifndef _SPARC_BTEXT_H
+#define _SPARC_BTEXT_H
+
+int btext_find_display(void);
+
+#endif /* _SPARC_BTEXT_H */
diff --git a/arch/sparc/include/asm/btfixup.h b/arch/sparc/include/asm/btfixup.h
deleted file mode 100644
index 797722cf69f..00000000000
--- a/arch/sparc/include/asm/btfixup.h
+++ /dev/null
@@ -1,208 +0,0 @@
-/*
- *  asm/btfixup.h:    Macros for boot time linking.
- *
- *  Copyright (C) 1998 Jakub Jelinek (jj@sunsite.mff.cuni.cz)
- */
- 
-#ifndef _SPARC_BTFIXUP_H
-#define _SPARC_BTFIXUP_H
-
-#include <linux/init.h>
-
-#ifndef __ASSEMBLY__
-
-#ifdef MODULE
-extern unsigned int ___illegal_use_of_BTFIXUP_SIMM13_in_module(void);
-extern unsigned int ___illegal_use_of_BTFIXUP_SETHI_in_module(void);
-extern unsigned int ___illegal_use_of_BTFIXUP_HALF_in_module(void);
-extern unsigned int ___illegal_use_of_BTFIXUP_INT_in_module(void);
-
-#define BTFIXUP_SIMM13(__name) ___illegal_use_of_BTFIXUP_SIMM13_in_module()
-#define BTFIXUP_HALF(__name) ___illegal_use_of_BTFIXUP_HALF_in_module()
-#define BTFIXUP_SETHI(__name) ___illegal_use_of_BTFIXUP_SETHI_in_module()
-#define BTFIXUP_INT(__name) ___illegal_use_of_BTFIXUP_INT_in_module()
-#define BTFIXUP_BLACKBOX(__name) ___illegal_use_of_BTFIXUP_BLACKBOX_in_module
-
-#else
-
-#define BTFIXUP_SIMM13(__name) ___sf_##__name()
-#define BTFIXUP_HALF(__name) ___af_##__name()
-#define BTFIXUP_SETHI(__name) ___hf_##__name()
-#define BTFIXUP_INT(__name) ((unsigned int)&___i_##__name)
-/* This must be written in assembly and present in a sethi */
-#define BTFIXUP_BLACKBOX(__name) ___b_##__name
-#endif /* MODULE */
-
-/* Fixup call xx */
-
-#define BTFIXUPDEF_CALL(__type, __name, __args...) 					\
-	extern __type ___f_##__name(__args);						\
-	extern unsigned ___fs_##__name[3];
-#define BTFIXUPDEF_CALL_CONST(__type, __name, __args...) 				\
-	extern __type ___f_##__name(__args) __attribute_const__;			\
-	extern unsigned ___fs_##__name[3];
-#define BTFIXUP_CALL(__name) ___f_##__name
-
-#define BTFIXUPDEF_BLACKBOX(__name)							\
-	extern unsigned ___bs_##__name[2];
-
-/* Put bottom 13bits into some register variable */
-
-#define BTFIXUPDEF_SIMM13(__name)							\
-	static inline unsigned int ___sf_##__name(void) __attribute_const__;		\
-	extern unsigned ___ss_##__name[2];						\
-	static inline unsigned int ___sf_##__name(void) {				\
-		unsigned int ret;							\
-		__asm__ ("or %%g0, ___s_" #__name ", %0" : "=r"(ret));			\
-		return ret;								\
-	}
-#define BTFIXUPDEF_SIMM13_INIT(__name,__val)						\
-	static inline unsigned int ___sf_##__name(void) __attribute_const__;		\
-	extern unsigned ___ss_##__name[2];						\
-	static inline unsigned int ___sf_##__name(void) {				\
-		unsigned int ret;							\
-		__asm__ ("or %%g0, ___s_" #__name "__btset_" #__val ", %0" : "=r"(ret));\
-		return ret;								\
-	}
-
-/* Put either bottom 13 bits, or upper 22 bits into some register variable
- * (depending on the value, this will lead into sethi FIX, reg; or
- * mov FIX, reg; )
- */
-
-#define BTFIXUPDEF_HALF(__name)								\
-	static inline unsigned int ___af_##__name(void) __attribute_const__;		\
-	extern unsigned ___as_##__name[2];						\
-	static inline unsigned int ___af_##__name(void) {				\
-		unsigned int ret;							\
-		__asm__ ("or %%g0, ___a_" #__name ", %0" : "=r"(ret));			\
-		return ret;								\
-	}
-#define BTFIXUPDEF_HALF_INIT(__name,__val)						\
-	static inline unsigned int ___af_##__name(void) __attribute_const__;		\
-	extern unsigned ___as_##__name[2];						\
-	static inline unsigned int ___af_##__name(void) {				\
-		unsigned int ret;							\
-		__asm__ ("or %%g0, ___a_" #__name "__btset_" #__val ", %0" : "=r"(ret));\
-		return ret;								\
-	}
-
-/* Put upper 22 bits into some register variable */
-
-#define BTFIXUPDEF_SETHI(__name)							\
-	static inline unsigned int ___hf_##__name(void) __attribute_const__;		\
-	extern unsigned ___hs_##__name[2];						\
-	static inline unsigned int ___hf_##__name(void) {				\
-		unsigned int ret;							\
-		__asm__ ("sethi %%hi(___h_" #__name "), %0" : "=r"(ret));		\
-		return ret;								\
-	}
-#define BTFIXUPDEF_SETHI_INIT(__name,__val)						\
-	static inline unsigned int ___hf_##__name(void) __attribute_const__;		\
-	extern unsigned ___hs_##__name[2];						\
-	static inline unsigned int ___hf_##__name(void) {				\
-		unsigned int ret;							\
-		__asm__ ("sethi %%hi(___h_" #__name "__btset_" #__val "), %0" : 	\
-			 "=r"(ret));							\
-		return ret;								\
-	}
-
-/* Put a full 32bit integer into some register variable */
-
-#define BTFIXUPDEF_INT(__name)								\
-	extern unsigned char ___i_##__name;						\
-	extern unsigned ___is_##__name[2];
-
-#define BTFIXUPCALL_NORM	0x00000000			/* Always call */
-#define BTFIXUPCALL_NOP		0x01000000			/* Possibly optimize to nop */
-#define BTFIXUPCALL_RETINT(i)	(0x90102000|((i) & 0x1fff))	/* Possibly optimize to mov i, %o0 */
-#define BTFIXUPCALL_ORINT(i)	(0x90122000|((i) & 0x1fff))	/* Possibly optimize to or %o0, i, %o0 */
-#define BTFIXUPCALL_RETO0	0x01000000			/* Return first parameter, actually a nop */
-#define BTFIXUPCALL_ANDNINT(i)	(0x902a2000|((i) & 0x1fff))	/* Possibly optimize to andn %o0, i, %o0 */
-#define BTFIXUPCALL_SWAPO0O1	0xd27a0000			/* Possibly optimize to swap [%o0],%o1 */
-#define BTFIXUPCALL_SWAPO0G0	0xc07a0000			/* Possibly optimize to swap [%o0],%g0 */
-#define BTFIXUPCALL_SWAPG1G2	0xc4784000			/* Possibly optimize to swap [%g1],%g2 */
-#define BTFIXUPCALL_STG0O0	0xc0220000			/* Possibly optimize to st %g0,[%o0] */
-#define BTFIXUPCALL_STO1O0	0xd2220000			/* Possibly optimize to st %o1,[%o0] */
-
-#define BTFIXUPSET_CALL(__name, __addr, __insn)						\
-	do {										\
-		___fs_##__name[0] |= 1;							\
-		___fs_##__name[1] = (unsigned long)__addr;				\
-		___fs_##__name[2] = __insn;						\
-	} while (0)
-	
-#define BTFIXUPSET_BLACKBOX(__name, __func)						\
-	do {										\
-		___bs_##__name[0] |= 1;							\
-		___bs_##__name[1] = (unsigned long)__func;				\
-	} while (0)
-	
-#define BTFIXUPCOPY_CALL(__name, __from)						\
-	do {										\
-		___fs_##__name[0] |= 1;							\
-		___fs_##__name[1] = ___fs_##__from[1];					\
-		___fs_##__name[2] = ___fs_##__from[2];					\
-	} while (0)
-		
-#define BTFIXUPSET_SIMM13(__name, __val)						\
-	do {										\
-		___ss_##__name[0] |= 1;							\
-		___ss_##__name[1] = (unsigned)__val;					\
-	} while (0)
-	
-#define BTFIXUPCOPY_SIMM13(__name, __from)						\
-	do {										\
-		___ss_##__name[0] |= 1;							\
-		___ss_##__name[1] = ___ss_##__from[1];					\
-	} while (0)
-		
-#define BTFIXUPSET_HALF(__name, __val)							\
-	do {										\
-		___as_##__name[0] |= 1;							\
-		___as_##__name[1] = (unsigned)__val;					\
-	} while (0)
-	
-#define BTFIXUPCOPY_HALF(__name, __from)						\
-	do {										\
-		___as_##__name[0] |= 1;							\
-		___as_##__name[1] = ___as_##__from[1];					\
-	} while (0)
-		
-#define BTFIXUPSET_SETHI(__name, __val)							\
-	do {										\
-		___hs_##__name[0] |= 1;							\
-		___hs_##__name[1] = (unsigned)__val;					\
-	} while (0)
-	
-#define BTFIXUPCOPY_SETHI(__name, __from)						\
-	do {										\
-		___hs_##__name[0] |= 1;							\
-		___hs_##__name[1] = ___hs_##__from[1];					\
-	} while (0)
-		
-#define BTFIXUPSET_INT(__name, __val)							\
-	do {										\
-		___is_##__name[0] |= 1;							\
-		___is_##__name[1] = (unsigned)__val;					\
-	} while (0)
-	
-#define BTFIXUPCOPY_INT(__name, __from)							\
-	do {										\
-		___is_##__name[0] |= 1;							\
-		___is_##__name[1] = ___is_##__from[1];					\
-	} while (0)
-	
-#define BTFIXUPVAL_CALL(__name)								\
-	((unsigned long)___fs_##__name[1])
-	
-extern void btfixup(void);
-
-#else /* __ASSEMBLY__ */
-
-#define BTFIXUP_SETHI(__name)			%hi(___h_ ## __name)
-#define BTFIXUP_SETHI_INIT(__name,__val)	%hi(___h_ ## __name ## __btset_ ## __val)
-
-#endif /* __ASSEMBLY__ */
-	
-#endif /* !(_SPARC_BTFIXUP_H) */
diff --git a/arch/sparc/include/asm/bug.h b/arch/sparc/include/asm/bug.h
index 8a59e5a8c21..eaa8f8d3812 100644
--- a/arch/sparc/include/asm/bug.h
+++ b/arch/sparc/include/asm/bug.h
@@ -5,7 +5,7 @@
 #include <linux/compiler.h>
 
 #ifdef CONFIG_DEBUG_BUGVERBOSE
-extern void do_BUG(const char *file, int line);
+void do_BUG(const char *file, int line);
 #define BUG() do {					\
 	do_BUG(__FILE__, __LINE__);			\
 	__builtin_trap();				\
@@ -19,4 +19,7 @@ extern void do_BUG(const char *file, int line);
 
 #include <asm-generic/bug.h>
 
+struct pt_regs;
+void __noreturn die_if_kernel(char *str, struct pt_regs *regs);
+
 #endif
diff --git a/arch/sparc/include/asm/cache.h b/arch/sparc/include/asm/cache.h
index 41f85ae4bd4..5bb6991b485 100644
--- a/arch/sparc/include/asm/cache.h
+++ b/arch/sparc/include/asm/cache.h
@@ -7,9 +7,10 @@
 #ifndef _SPARC_CACHE_H
 #define _SPARC_CACHE_H
 
+#define ARCH_SLAB_MINALIGN	__alignof__(unsigned long long)
+
 #define L1_CACHE_SHIFT 5
 #define L1_CACHE_BYTES 32
-#define L1_CACHE_ALIGN(x) ((((x)+(L1_CACHE_BYTES-1))&~(L1_CACHE_BYTES-1)))
 
 #ifdef CONFIG_SPARC32
 #define SMP_CACHE_BYTES_SHIFT 5
@@ -19,120 +20,6 @@
 
 #define SMP_CACHE_BYTES (1 << SMP_CACHE_BYTES_SHIFT)
 
-#define __read_mostly __attribute__((__section__(".data.read_mostly")))
-
-#ifdef CONFIG_SPARC32
-#include <asm/asi.h>
-
-/* Direct access to the instruction cache is provided through and
- * alternate address space.  The IDC bit must be off in the ICCR on
- * HyperSparcs for these accesses to work.  The code below does not do
- * any checking, the caller must do so.  These routines are for
- * diagnostics only, but could end up being useful.  Use with care.
- * Also, you are asking for trouble if you execute these in one of the
- * three instructions following a %asr/%psr access or modification.
- */
-
-/* First, cache-tag access. */
-static inline unsigned int get_icache_tag(int setnum, int tagnum)
-{
-	unsigned int vaddr, retval;
-
-	vaddr = ((setnum&1) << 12) | ((tagnum&0x7f) << 5);
-	__asm__ __volatile__("lda [%1] %2, %0\n\t" :
-			     "=r" (retval) :
-			     "r" (vaddr), "i" (ASI_M_TXTC_TAG));
-	return retval;
-}
-
-static inline void put_icache_tag(int setnum, int tagnum, unsigned int entry)
-{
-	unsigned int vaddr;
-
-	vaddr = ((setnum&1) << 12) | ((tagnum&0x7f) << 5);
-	__asm__ __volatile__("sta %0, [%1] %2\n\t" : :
-			     "r" (entry), "r" (vaddr), "i" (ASI_M_TXTC_TAG) :
-			     "memory");
-}
-
-/* Second cache-data access.  The data is returned two-32bit quantities
- * at a time.
- */
-static inline void get_icache_data(int setnum, int tagnum, int subblock,
-				       unsigned int *data)
-{
-	unsigned int value1, value2, vaddr;
-
-	vaddr = ((setnum&0x1) << 12) | ((tagnum&0x7f) << 5) |
-		((subblock&0x3) << 3);
-	__asm__ __volatile__("ldda [%2] %3, %%g2\n\t"
-			     "or %%g0, %%g2, %0\n\t"
-			     "or %%g0, %%g3, %1\n\t" :
-			     "=r" (value1), "=r" (value2) :
-			     "r" (vaddr), "i" (ASI_M_TXTC_DATA) :
-			     "g2", "g3");
-	data[0] = value1; data[1] = value2;
-}
-
-static inline void put_icache_data(int setnum, int tagnum, int subblock,
-				       unsigned int *data)
-{
-	unsigned int value1, value2, vaddr;
-
-	vaddr = ((setnum&0x1) << 12) | ((tagnum&0x7f) << 5) |
-		((subblock&0x3) << 3);
-	value1 = data[0]; value2 = data[1];
-	__asm__ __volatile__("or %%g0, %0, %%g2\n\t"
-			     "or %%g0, %1, %%g3\n\t"
-			     "stda %%g2, [%2] %3\n\t" : :
-			     "r" (value1), "r" (value2), 
-			     "r" (vaddr), "i" (ASI_M_TXTC_DATA) :
-			     "g2", "g3", "memory" /* no joke */);
-}
-
-/* Different types of flushes with the ICACHE.  Some of the flushes
- * affect both the ICACHE and the external cache.  Others only clear
- * the ICACHE entries on the cpu itself.  V8's (most) allow
- * granularity of flushes on the packet (element in line), whole line,
- * and entire cache (ie. all lines) level.  The ICACHE only flushes are
- * ROSS HyperSparc specific and are in ross.h
- */
-
-/* Flushes which clear out both the on-chip and external caches */
-static inline void flush_ei_page(unsigned int addr)
-{
-	__asm__ __volatile__("sta %%g0, [%0] %1\n\t" : :
-			     "r" (addr), "i" (ASI_M_FLUSH_PAGE) :
-			     "memory");
-}
-
-static inline void flush_ei_seg(unsigned int addr)
-{
-	__asm__ __volatile__("sta %%g0, [%0] %1\n\t" : :
-			     "r" (addr), "i" (ASI_M_FLUSH_SEG) :
-			     "memory");
-}
-
-static inline void flush_ei_region(unsigned int addr)
-{
-	__asm__ __volatile__("sta %%g0, [%0] %1\n\t" : :
-			     "r" (addr), "i" (ASI_M_FLUSH_REGION) :
-			     "memory");
-}
-
-static inline void flush_ei_ctx(unsigned int addr)
-{
-	__asm__ __volatile__("sta %%g0, [%0] %1\n\t" : :
-			     "r" (addr), "i" (ASI_M_FLUSH_CTX) :
-			     "memory");
-}
-
-static inline void flush_ei_user(unsigned int addr)
-{
-	__asm__ __volatile__("sta %%g0, [%0] %1\n\t" : :
-			     "r" (addr), "i" (ASI_M_FLUSH_USER) :
-			     "memory");
-}
-#endif /* CONFIG_SPARC32 */
+#define __read_mostly __attribute__((__section__(".data..read_mostly")))
 
 #endif /* !(_SPARC_CACHE_H) */
diff --git a/arch/sparc/include/asm/cacheflush.h b/arch/sparc/include/asm/cacheflush.h
index 049168087b1..f6c4839b838 100644
--- a/arch/sparc/include/asm/cacheflush.h
+++ b/arch/sparc/include/asm/cacheflush.h
@@ -1,5 +1,9 @@
 #ifndef ___ASM_SPARC_CACHEFLUSH_H
 #define ___ASM_SPARC_CACHEFLUSH_H
+
+/* flush addr - to allow use of self-modifying code */
+#define flushi(addr)	__asm__ __volatile__ ("flush %0" : : "r" (addr) : "memory")
+
 #if defined(__sparc__) && defined(__arch64__)
 #include <asm/cacheflush_64.h>
 #else
diff --git a/arch/sparc/include/asm/cacheflush_32.h b/arch/sparc/include/asm/cacheflush_32.h
index 68ac1091027..12164006181 100644
--- a/arch/sparc/include/asm/cacheflush_32.h
+++ b/arch/sparc/include/asm/cacheflush_32.h
@@ -1,56 +1,18 @@
 #ifndef _SPARC_CACHEFLUSH_H
 #define _SPARC_CACHEFLUSH_H
 
-#include <linux/mm.h>		/* Common for other includes */
-// #include <linux/kernel.h> from pgalloc.h
-// #include <linux/sched.h>  from pgalloc.h
-
-// #include <asm/page.h>
-#include <asm/btfixup.h>
-
-/*
- * Fine grained cache flushing.
- */
-#ifdef CONFIG_SMP
-
-BTFIXUPDEF_CALL(void, local_flush_cache_all, void)
-BTFIXUPDEF_CALL(void, local_flush_cache_mm, struct mm_struct *)
-BTFIXUPDEF_CALL(void, local_flush_cache_range, struct vm_area_struct *, unsigned long, unsigned long)
-BTFIXUPDEF_CALL(void, local_flush_cache_page, struct vm_area_struct *, unsigned long)
-
-#define local_flush_cache_all() BTFIXUP_CALL(local_flush_cache_all)()
-#define local_flush_cache_mm(mm) BTFIXUP_CALL(local_flush_cache_mm)(mm)
-#define local_flush_cache_range(vma,start,end) BTFIXUP_CALL(local_flush_cache_range)(vma,start,end)
-#define local_flush_cache_page(vma,addr) BTFIXUP_CALL(local_flush_cache_page)(vma,addr)
-
-BTFIXUPDEF_CALL(void, local_flush_page_to_ram, unsigned long)
-BTFIXUPDEF_CALL(void, local_flush_sig_insns, struct mm_struct *, unsigned long)
-
-#define local_flush_page_to_ram(addr) BTFIXUP_CALL(local_flush_page_to_ram)(addr)
-#define local_flush_sig_insns(mm,insn_addr) BTFIXUP_CALL(local_flush_sig_insns)(mm,insn_addr)
-
-extern void smp_flush_cache_all(void);
-extern void smp_flush_cache_mm(struct mm_struct *mm);
-extern void smp_flush_cache_range(struct vm_area_struct *vma,
-				  unsigned long start,
-				  unsigned long end);
-extern void smp_flush_cache_page(struct vm_area_struct *vma, unsigned long page);
-
-extern void smp_flush_page_to_ram(unsigned long page);
-extern void smp_flush_sig_insns(struct mm_struct *mm, unsigned long insn_addr);
-
-#endif /* CONFIG_SMP */
-
-BTFIXUPDEF_CALL(void, flush_cache_all, void)
-BTFIXUPDEF_CALL(void, flush_cache_mm, struct mm_struct *)
-BTFIXUPDEF_CALL(void, flush_cache_range, struct vm_area_struct *, unsigned long, unsigned long)
-BTFIXUPDEF_CALL(void, flush_cache_page, struct vm_area_struct *, unsigned long)
-
-#define flush_cache_all() BTFIXUP_CALL(flush_cache_all)()
-#define flush_cache_mm(mm) BTFIXUP_CALL(flush_cache_mm)(mm)
-#define flush_cache_dup_mm(mm) BTFIXUP_CALL(flush_cache_mm)(mm)
-#define flush_cache_range(vma,start,end) BTFIXUP_CALL(flush_cache_range)(vma,start,end)
-#define flush_cache_page(vma,addr,pfn) BTFIXUP_CALL(flush_cache_page)(vma,addr)
+#include <asm/cachetlb_32.h>
+
+#define flush_cache_all() \
+	sparc32_cachetlb_ops->cache_all()
+#define flush_cache_mm(mm) \
+	sparc32_cachetlb_ops->cache_mm(mm)
+#define flush_cache_dup_mm(mm) \
+	sparc32_cachetlb_ops->cache_mm(mm)
+#define flush_cache_range(vma,start,end) \
+	sparc32_cachetlb_ops->cache_range(vma, start, end)
+#define flush_cache_page(vma,addr,pfn) \
+	sparc32_cachetlb_ops->cache_page(vma, addr)
 #define flush_icache_range(start, end)		do { } while (0)
 #define flush_icache_page(vma, pg)		do { } while (0)
 
@@ -67,14 +29,16 @@ BTFIXUPDEF_CALL(void, flush_cache_page, struct vm_area_struct *, unsigned long)
 		memcpy(dst, src, len);				\
 	} while (0)
 
-BTFIXUPDEF_CALL(void, __flush_page_to_ram, unsigned long)
-BTFIXUPDEF_CALL(void, flush_sig_insns, struct mm_struct *, unsigned long)
+#define __flush_page_to_ram(addr) \
+	sparc32_cachetlb_ops->page_to_ram(addr)
+#define flush_sig_insns(mm,insn_addr) \
+	sparc32_cachetlb_ops->sig_insns(mm, insn_addr)
+#define flush_page_for_dma(addr) \
+	sparc32_cachetlb_ops->page_for_dma(addr)
 
-#define __flush_page_to_ram(addr) BTFIXUP_CALL(__flush_page_to_ram)(addr)
-#define flush_sig_insns(mm,insn_addr) BTFIXUP_CALL(flush_sig_insns)(mm,insn_addr)
-
-extern void sparc_flush_page_to_ram(struct page *page);
+void sparc_flush_page_to_ram(struct page *page);
 
+#define ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE 1
 #define flush_dcache_page(page)			sparc_flush_page_to_ram(page)
 #define flush_dcache_mmap_lock(mapping)		do { } while (0)
 #define flush_dcache_mmap_unlock(mapping)	do { } while (0)
@@ -82,4 +46,13 @@ extern void sparc_flush_page_to_ram(struct page *page);
 #define flush_cache_vmap(start, end)		flush_cache_all()
 #define flush_cache_vunmap(start, end)		flush_cache_all()
 
+/* When a context switch happens we must flush all user windows so that
+ * the windows of the current process are flushed onto its stack. This
+ * way the windows are all clean for the next process and the stack
+ * frames are up to date.
+ */
+void flush_user_windows(void);
+void kill_user_windows(void);
+void flushw_all(void);
+
 #endif /* _SPARC_CACHEFLUSH_H */
diff --git a/arch/sparc/include/asm/cacheflush_64.h b/arch/sparc/include/asm/cacheflush_64.h
index c43321729b3..38965379e35 100644
--- a/arch/sparc/include/asm/cacheflush_64.h
+++ b/arch/sparc/include/asm/cacheflush_64.h
@@ -8,6 +8,13 @@
 #include <linux/mm.h>
 
 /* Cache flush operations. */
+#define flushw_all()	__asm__ __volatile__("flushw")
+
+void __flushw_user(void);
+#define flushw_user() __flushw_user()
+
+#define flush_user_windows flushw_user
+#define flush_register_windows flushw_all
 
 /* These are the same regardless of whether this is an SMP kernel or not. */
 #define flush_cache_mm(__mm) \
@@ -23,28 +30,29 @@
  * use block commit stores (which invalidate icache lines) during
  * module load, so we need this.
  */
-extern void flush_icache_range(unsigned long start, unsigned long end);
-extern void __flush_icache_page(unsigned long);
+void flush_icache_range(unsigned long start, unsigned long end);
+void __flush_icache_page(unsigned long);
 
-extern void __flush_dcache_page(void *addr, int flush_icache);
-extern void flush_dcache_page_impl(struct page *page);
+void __flush_dcache_page(void *addr, int flush_icache);
+void flush_dcache_page_impl(struct page *page);
 #ifdef CONFIG_SMP
-extern void smp_flush_dcache_page_impl(struct page *page, int cpu);
-extern void flush_dcache_page_all(struct mm_struct *mm, struct page *page);
+void smp_flush_dcache_page_impl(struct page *page, int cpu);
+void flush_dcache_page_all(struct mm_struct *mm, struct page *page);
 #else
 #define smp_flush_dcache_page_impl(page,cpu) flush_dcache_page_impl(page)
 #define flush_dcache_page_all(mm,page) flush_dcache_page_impl(page)
 #endif
 
-extern void __flush_dcache_range(unsigned long start, unsigned long end);
-extern void flush_dcache_page(struct page *page);
+void __flush_dcache_range(unsigned long start, unsigned long end);
+#define ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE 1
+void flush_dcache_page(struct page *page);
 
 #define flush_icache_page(vma, pg)	do { } while(0)
 #define flush_icache_user_range(vma,pg,adr,len)	do { } while (0)
 
-extern void flush_ptrace_access(struct vm_area_struct *, struct page *,
-				unsigned long uaddr, void *kaddr,
-				unsigned long len, int write);
+void flush_ptrace_access(struct vm_area_struct *, struct page *,
+			 unsigned long uaddr, void *kaddr,
+			 unsigned long len, int write);
 
 #define copy_to_user_page(vma, page, vaddr, dst, src, len)		\
 	do {								\
diff --git a/arch/sparc/include/asm/cachetlb_32.h b/arch/sparc/include/asm/cachetlb_32.h
new file mode 100644
index 00000000000..efb19889a08
--- /dev/null
+++ b/arch/sparc/include/asm/cachetlb_32.h
@@ -0,0 +1,29 @@
+#ifndef _SPARC_CACHETLB_H
+#define _SPARC_CACHETLB_H
+
+struct mm_struct;
+struct vm_area_struct;
+
+struct sparc32_cachetlb_ops {
+	void (*cache_all)(void);
+	void (*cache_mm)(struct mm_struct *);
+	void (*cache_range)(struct vm_area_struct *, unsigned long,
+			    unsigned long);
+	void (*cache_page)(struct vm_area_struct *, unsigned long);
+
+	void (*tlb_all)(void);
+	void (*tlb_mm)(struct mm_struct *);
+	void (*tlb_range)(struct vm_area_struct *, unsigned long,
+			  unsigned long);
+	void (*tlb_page)(struct vm_area_struct *, unsigned long);
+
+	void (*page_to_ram)(unsigned long);
+	void (*sig_insns)(struct mm_struct *, unsigned long);
+	void (*page_for_dma)(unsigned long);
+};
+extern const struct sparc32_cachetlb_ops *sparc32_cachetlb_ops;
+#ifdef CONFIG_SMP
+extern const struct sparc32_cachetlb_ops *local_ops;
+#endif
+
+#endif /* SPARC_CACHETLB_H */
diff --git a/arch/sparc/include/asm/checksum_32.h b/arch/sparc/include/asm/checksum_32.h
index bdbda1453aa..426b2389a1c 100644
--- a/arch/sparc/include/asm/checksum_32.h
+++ b/arch/sparc/include/asm/checksum_32.h
@@ -29,7 +29,7 @@
  *
  * it's best to have buff aligned on a 32-bit boundary
  */
-extern __wsum csum_partial(const void *buff, int len, __wsum sum);
+__wsum csum_partial(const void *buff, int len, __wsum sum);
 
 /* the same as csum_partial, but copies from fs:src while it
  * checksums
@@ -38,7 +38,7 @@ extern __wsum csum_partial(const void *buff, int len, __wsum sum);
  * better 64-bit) boundary
  */
 
-extern unsigned int __csum_partial_copy_sparc_generic (const unsigned char *, unsigned char *);
+unsigned int __csum_partial_copy_sparc_generic (const unsigned char *, unsigned char *);
 
 static inline __wsum
 csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum)
@@ -238,4 +238,16 @@ static inline __sum16 ip_compute_csum(const void *buff, int len)
 	return csum_fold(csum_partial(buff, len, 0));
 }
 
+#define HAVE_ARCH_CSUM_ADD
+static inline __wsum csum_add(__wsum csum, __wsum addend)
+{
+	__asm__ __volatile__(
+		"addcc   %0, %1, %0\n"
+		"addx    %0, %%g0, %0"
+		: "=r" (csum)
+		: "r" (addend), "0" (csum));
+
+	return csum;
+}
+
 #endif /* !(__SPARC_CHECKSUM_H) */
diff --git a/arch/sparc/include/asm/checksum_64.h b/arch/sparc/include/asm/checksum_64.h
index 019b9615e43..b8779a6a591 100644
--- a/arch/sparc/include/asm/checksum_64.h
+++ b/arch/sparc/include/asm/checksum_64.h
@@ -29,7 +29,7 @@
  *
  * it's best to have buff aligned on a 32-bit boundary
  */
-extern __wsum csum_partial(const void * buff, int len, __wsum sum);
+__wsum csum_partial(const void * buff, int len, __wsum sum);
 
 /* the same as csum_partial, but copies from user space while it
  * checksums
@@ -37,12 +37,12 @@ extern __wsum csum_partial(const void * buff, int len, __wsum sum);
  * here even more important to align src and dst on a 32-bit (or even
  * better 64-bit) boundary
  */
-extern __wsum csum_partial_copy_nocheck(const void *src, void *dst,
-					      int len, __wsum sum);
+__wsum csum_partial_copy_nocheck(const void *src, void *dst,
+				 int len, __wsum sum);
 
-extern long __csum_partial_copy_from_user(const void __user *src,
-					  void *dst, int len,
-					  __wsum sum);
+long __csum_partial_copy_from_user(const void __user *src,
+				   void *dst, int len,
+				   __wsum sum);
 
 static inline __wsum
 csum_partial_copy_from_user(const void __user *src,
@@ -59,9 +59,9 @@ csum_partial_copy_from_user(const void __user *src,
  *	Copy and checksum to user
  */
 #define HAVE_CSUM_COPY_USER
-extern long __csum_partial_copy_to_user(const void *src,
-					void __user *dst, int len,
-					  __wsum sum);
+long __csum_partial_copy_to_user(const void *src,
+				 void __user *dst, int len,
+				 __wsum sum);
 
 static inline __wsum
 csum_and_copy_to_user(const void *src,
@@ -77,7 +77,7 @@ csum_and_copy_to_user(const void *src,
 /* ihl is always 5 or greater, almost always is 5, and iph is word aligned
  * the majority of the time.
  */
-extern __sum16 ip_fast_csum(const void *iph, unsigned int ihl);
+__sum16 ip_fast_csum(const void *iph, unsigned int ihl);
 
 /* Fold a partial checksum without adding pseudo headers. */
 static inline __sum16 csum_fold(__wsum sum)
@@ -96,9 +96,9 @@ static inline __sum16 csum_fold(__wsum sum)
 }
 
 static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
-					       unsigned int len,
-					       unsigned short proto,
-					       __wsum sum)
+					unsigned int len,
+					unsigned short proto,
+					__wsum sum)
 {
 	__asm__ __volatile__(
 "	addcc		%1, %0, %0\n"
@@ -116,9 +116,9 @@ static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
  * returns a 16-bit checksum, already complemented
  */
 static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr,
-						   unsigned short len,
-						   unsigned short proto,
-						   __wsum sum)
+					unsigned short len,
+					unsigned short proto,
+					__wsum sum)
 {
 	return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum));
 }
@@ -164,4 +164,16 @@ static inline __sum16 ip_compute_csum(const void *buff, int len)
 	return csum_fold(csum_partial(buff, len, 0));
 }
 
+#define HAVE_ARCH_CSUM_ADD
+static inline __wsum csum_add(__wsum csum, __wsum addend)
+{
+	__asm__ __volatile__(
+		"addcc   %0, %1, %0\n"
+		"addx    %0, %%g0, %0"
+		: "=r" (csum)
+		: "r" (addend), "0" (csum));
+
+	return csum;
+}
+
 #endif /* !(__SPARC64_CHECKSUM_H) */
diff --git a/arch/sparc/include/asm/cmpxchg.h b/arch/sparc/include/asm/cmpxchg.h
new file mode 100644
index 00000000000..9355893efa5
--- /dev/null
+++ b/arch/sparc/include/asm/cmpxchg.h
@@ -0,0 +1,8 @@
+#ifndef ___ASM_SPARC_CMPXCHG_H
+#define ___ASM_SPARC_CMPXCHG_H
+#if defined(__sparc__) && defined(__arch64__)
+#include <asm/cmpxchg_64.h>
+#else
+#include <asm/cmpxchg_32.h>
+#endif
+#endif
diff --git a/arch/sparc/include/asm/cmpxchg_32.h b/arch/sparc/include/asm/cmpxchg_32.h
new file mode 100644
index 00000000000..32c29a133f9
--- /dev/null
+++ b/arch/sparc/include/asm/cmpxchg_32.h
@@ -0,0 +1,85 @@
+/* 32-bit atomic xchg() and cmpxchg() definitions.
+ *
+ * Copyright (C) 1996 David S. Miller (davem@davemloft.net)
+ * Copyright (C) 2000 Anton Blanchard (anton@linuxcare.com.au)
+ * Copyright (C) 2007 Kyle McMartin (kyle@parisc-linux.org)
+ *
+ * Additions by Keith M Wesolowski (wesolows@foobazco.org) based
+ * on asm-parisc/atomic.h Copyright (C) 2000 Philipp Rumpf <prumpf@tux.org>.
+ */
+
+#ifndef __ARCH_SPARC_CMPXCHG__
+#define __ARCH_SPARC_CMPXCHG__
+
+static inline unsigned long xchg_u32(__volatile__ unsigned long *m, unsigned long val)
+{
+	__asm__ __volatile__("swap [%2], %0"
+			     : "=&r" (val)
+			     : "0" (val), "r" (m)
+			     : "memory");
+	return val;
+}
+
+void __xchg_called_with_bad_pointer(void);
+
+static inline unsigned long __xchg(unsigned long x, __volatile__ void * ptr, int size)
+{
+	switch (size) {
+	case 4:
+		return xchg_u32(ptr, x);
+	}
+	__xchg_called_with_bad_pointer();
+	return x;
+}
+
+#define xchg(ptr,x) ((__typeof__(*(ptr)))__xchg((unsigned long)(x),(ptr),sizeof(*(ptr))))
+
+/* Emulate cmpxchg() the same way we emulate atomics,
+ * by hashing the object address and indexing into an array
+ * of spinlocks to get a bit of performance...
+ *
+ * See arch/sparc/lib/atomic32.c for implementation.
+ *
+ * Cribbed from <asm-parisc/atomic.h>
+ */
+#define __HAVE_ARCH_CMPXCHG	1
+
+/* bug catcher for when unsupported size is used - won't link */
+void __cmpxchg_called_with_bad_pointer(void);
+/* we only need to support cmpxchg of a u32 on sparc */
+unsigned long __cmpxchg_u32(volatile u32 *m, u32 old, u32 new_);
+
+/* don't worry...optimizer will get rid of most of this */
+static inline unsigned long
+__cmpxchg(volatile void *ptr, unsigned long old, unsigned long new_, int size)
+{
+	switch (size) {
+	case 4:
+		return __cmpxchg_u32((u32 *)ptr, (u32)old, (u32)new_);
+	default:
+		__cmpxchg_called_with_bad_pointer();
+		break;
+	}
+	return old;
+}
+
+#define cmpxchg(ptr, o, n)						\
+({									\
+	__typeof__(*(ptr)) _o_ = (o);					\
+	__typeof__(*(ptr)) _n_ = (n);					\
+	(__typeof__(*(ptr))) __cmpxchg((ptr), (unsigned long)_o_,	\
+			(unsigned long)_n_, sizeof(*(ptr)));		\
+})
+
+#include <asm-generic/cmpxchg-local.h>
+
+/*
+ * cmpxchg_local and cmpxchg64_local are atomic wrt current CPU. Always make
+ * them available.
+ */
+#define cmpxchg_local(ptr, o, n)				  	       \
+	((__typeof__(*(ptr)))__cmpxchg_local_generic((ptr), (unsigned long)(o),\
+			(unsigned long)(n), sizeof(*(ptr))))
+#define cmpxchg64_local(ptr, o, n) __cmpxchg64_local_generic((ptr), (o), (n))
+
+#endif /* __ARCH_SPARC_CMPXCHG__ */
diff --git a/arch/sparc/include/asm/cmpxchg_64.h b/arch/sparc/include/asm/cmpxchg_64.h
new file mode 100644
index 00000000000..0e1ed6cfbf6
--- /dev/null
+++ b/arch/sparc/include/asm/cmpxchg_64.h
@@ -0,0 +1,146 @@
+/* 64-bit atomic xchg() and cmpxchg() definitions.
+ *
+ * Copyright (C) 1996, 1997, 2000 David S. Miller (davem@redhat.com)
+ */
+
+#ifndef __ARCH_SPARC64_CMPXCHG__
+#define __ARCH_SPARC64_CMPXCHG__
+
+static inline unsigned long xchg32(__volatile__ unsigned int *m, unsigned int val)
+{
+	unsigned long tmp1, tmp2;
+
+	__asm__ __volatile__(
+"	mov		%0, %1\n"
+"1:	lduw		[%4], %2\n"
+"	cas		[%4], %2, %0\n"
+"	cmp		%2, %0\n"
+"	bne,a,pn	%%icc, 1b\n"
+"	 mov		%1, %0\n"
+	: "=&r" (val), "=&r" (tmp1), "=&r" (tmp2)
+	: "0" (val), "r" (m)
+	: "cc", "memory");
+	return val;
+}
+
+static inline unsigned long xchg64(__volatile__ unsigned long *m, unsigned long val)
+{
+	unsigned long tmp1, tmp2;
+
+	__asm__ __volatile__(
+"	mov		%0, %1\n"
+"1:	ldx		[%4], %2\n"
+"	casx		[%4], %2, %0\n"
+"	cmp		%2, %0\n"
+"	bne,a,pn	%%xcc, 1b\n"
+"	 mov		%1, %0\n"
+	: "=&r" (val), "=&r" (tmp1), "=&r" (tmp2)
+	: "0" (val), "r" (m)
+	: "cc", "memory");
+	return val;
+}
+
+#define xchg(ptr,x) ((__typeof__(*(ptr)))__xchg((unsigned long)(x),(ptr),sizeof(*(ptr))))
+
+void __xchg_called_with_bad_pointer(void);
+
+static inline unsigned long __xchg(unsigned long x, __volatile__ void * ptr,
+				       int size)
+{
+	switch (size) {
+	case 4:
+		return xchg32(ptr, x);
+	case 8:
+		return xchg64(ptr, x);
+	}
+	__xchg_called_with_bad_pointer();
+	return x;
+}
+
+/*
+ * Atomic compare and exchange.  Compare OLD with MEM, if identical,
+ * store NEW in MEM.  Return the initial value in MEM.  Success is
+ * indicated by comparing RETURN with OLD.
+ */
+
+#include <asm-generic/cmpxchg-local.h>
+
+#define __HAVE_ARCH_CMPXCHG 1
+
+static inline unsigned long
+__cmpxchg_u32(volatile int *m, int old, int new)
+{
+	__asm__ __volatile__("cas [%2], %3, %0"
+			     : "=&r" (new)
+			     : "0" (new), "r" (m), "r" (old)
+			     : "memory");
+
+	return new;
+}
+
+static inline unsigned long
+__cmpxchg_u64(volatile long *m, unsigned long old, unsigned long new)
+{
+	__asm__ __volatile__("casx [%2], %3, %0"
+			     : "=&r" (new)
+			     : "0" (new), "r" (m), "r" (old)
+			     : "memory");
+
+	return new;
+}
+
+/* This function doesn't exist, so you'll get a linker error
+   if something tries to do an invalid cmpxchg().  */
+void __cmpxchg_called_with_bad_pointer(void);
+
+static inline unsigned long
+__cmpxchg(volatile void *ptr, unsigned long old, unsigned long new, int size)
+{
+	switch (size) {
+		case 4:
+			return __cmpxchg_u32(ptr, old, new);
+		case 8:
+			return __cmpxchg_u64(ptr, old, new);
+	}
+	__cmpxchg_called_with_bad_pointer();
+	return old;
+}
+
+#define cmpxchg(ptr,o,n)						 \
+  ({									 \
+     __typeof__(*(ptr)) _o_ = (o);					 \
+     __typeof__(*(ptr)) _n_ = (n);					 \
+     (__typeof__(*(ptr))) __cmpxchg((ptr), (unsigned long)_o_,		 \
+				    (unsigned long)_n_, sizeof(*(ptr))); \
+  })
+
+/*
+ * cmpxchg_local and cmpxchg64_local are atomic wrt current CPU. Always make
+ * them available.
+ */
+
+static inline unsigned long __cmpxchg_local(volatile void *ptr,
+				      unsigned long old,
+				      unsigned long new, int size)
+{
+	switch (size) {
+	case 4:
+	case 8:	return __cmpxchg(ptr, old, new, size);
+	default:
+		return __cmpxchg_local_generic(ptr, old, new, size);
+	}
+
+	return old;
+}
+
+#define cmpxchg_local(ptr, o, n)				  	\
+	((__typeof__(*(ptr)))__cmpxchg_local((ptr), (unsigned long)(o),	\
+			(unsigned long)(n), sizeof(*(ptr))))
+#define cmpxchg64_local(ptr, o, n)					\
+  ({									\
+	BUILD_BUG_ON(sizeof(*(ptr)) != 8);				\
+	cmpxchg_local((ptr), (o), (n));					\
+  })
+#define cmpxchg64(ptr, o, n)	cmpxchg64_local((ptr), (o), (n))
+
+#endif /* __ARCH_SPARC64_CMPXCHG__ */
diff --git a/arch/sparc/include/asm/cmt.h b/arch/sparc/include/asm/cmt.h
deleted file mode 100644
index 870db592857..00000000000
--- a/arch/sparc/include/asm/cmt.h
+++ /dev/null
@@ -1,59 +0,0 @@
-#ifndef _SPARC64_CMT_H
-#define _SPARC64_CMT_H
-
-/* cmt.h: Chip Multi-Threading register definitions
- *
- * Copyright (C) 2004 David S. Miller (davem@redhat.com)
- */
-
-/* ASI_CORE_ID - private */
-#define LP_ID		0x0000000000000010UL
-#define  LP_ID_MAX	0x00000000003f0000UL
-#define  LP_ID_ID	0x000000000000003fUL
-
-/* ASI_INTR_ID - private */
-#define LP_INTR_ID	0x0000000000000000UL
-#define  LP_INTR_ID_ID	0x00000000000003ffUL
-
-/* ASI_CESR_ID - private */
-#define CESR_ID		0x0000000000000040UL
-#define  CESR_ID_ID	0x00000000000000ffUL
-
-/* ASI_CORE_AVAILABLE - shared */
-#define LP_AVAIL	0x0000000000000000UL
-#define  LP_AVAIL_1	0x0000000000000002UL
-#define  LP_AVAIL_0	0x0000000000000001UL
-
-/* ASI_CORE_ENABLE_STATUS - shared */
-#define LP_ENAB_STAT	0x0000000000000010UL
-#define  LP_ENAB_STAT_1	0x0000000000000002UL
-#define  LP_ENAB_STAT_0	0x0000000000000001UL
-
-/* ASI_CORE_ENABLE - shared */
-#define LP_ENAB		0x0000000000000020UL
-#define  LP_ENAB_1	0x0000000000000002UL
-#define  LP_ENAB_0	0x0000000000000001UL
-
-/* ASI_CORE_RUNNING - shared */
-#define LP_RUNNING_RW	0x0000000000000050UL
-#define LP_RUNNING_W1S	0x0000000000000060UL
-#define LP_RUNNING_W1C	0x0000000000000068UL
-#define  LP_RUNNING_1	0x0000000000000002UL
-#define  LP_RUNNING_0	0x0000000000000001UL
-
-/* ASI_CORE_RUNNING_STAT - shared */
-#define LP_RUN_STAT	0x0000000000000058UL
-#define  LP_RUN_STAT_1	0x0000000000000002UL
-#define  LP_RUN_STAT_0	0x0000000000000001UL
-
-/* ASI_XIR_STEERING - shared */
-#define LP_XIR_STEER	0x0000000000000030UL
-#define  LP_XIR_STEER_1	0x0000000000000002UL
-#define  LP_XIR_STEER_0	0x0000000000000001UL
-
-/* ASI_CMT_ERROR_STEERING - shared */
-#define CMT_ER_STEER	0x0000000000000040UL
-#define  CMT_ER_STEER_1	0x0000000000000002UL
-#define  CMT_ER_STEER_0	0x0000000000000001UL
-
-#endif /* _SPARC64_CMT_H */
diff --git a/arch/sparc/include/asm/compat.h b/arch/sparc/include/asm/compat.h
index f260b58f5ce..830502fe62b 100644
--- a/arch/sparc/include/asm/compat.h
+++ b/arch/sparc/include/asm/compat.h
@@ -5,7 +5,8 @@
  */
 #include <linux/types.h>
 
-#define COMPAT_USER_HZ	100
+#define COMPAT_USER_HZ		100
+#define COMPAT_UTS_MACHINE	"sparc\0\0"
 
 typedef u32		compat_size_t;
 typedef s32		compat_ssize_t;
@@ -35,6 +36,7 @@ typedef s64		compat_s64;
 typedef u32		compat_uint_t;
 typedef u32		compat_ulong_t;
 typedef u64		compat_u64;
+typedef u32		compat_uptr_t;
 
 struct compat_timespec {
 	compat_time_t	tv_sec;
@@ -133,7 +135,8 @@ struct compat_statfs {
 	compat_fsid_t	f_fsid;
 	int		f_namelen;	/* SunOS ignores this field. */
 	int		f_frsize;
-	int		f_spare[5];
+	int		f_flags;
+	int		f_spare[4];
 };
 
 #define COMPAT_RLIM_INFINITY 0x7fffffff
@@ -145,6 +148,65 @@ typedef u32		compat_old_sigset_t;
 
 typedef u32		compat_sigset_word;
 
+typedef union compat_sigval {
+	compat_int_t	sival_int;
+	compat_uptr_t	sival_ptr;
+} compat_sigval_t;
+
+#define SI_PAD_SIZE32	(128/sizeof(int) - 3)
+
+typedef struct compat_siginfo {
+	int si_signo;
+	int si_errno;
+	int si_code;
+
+	union {
+		int _pad[SI_PAD_SIZE32];
+
+		/* kill() */
+		struct {
+			compat_pid_t _pid;		/* sender's pid */
+			unsigned int _uid;		/* sender's uid */
+		} _kill;
+
+		/* POSIX.1b timers */
+		struct {
+			compat_timer_t _tid;		/* timer id */
+			int _overrun;			/* overrun count */
+			compat_sigval_t _sigval;	/* same as below */
+			int _sys_private;	/* not to be passed to user */
+		} _timer;
+
+		/* POSIX.1b signals */
+		struct {
+			compat_pid_t _pid;		/* sender's pid */
+			unsigned int _uid;		/* sender's uid */
+			compat_sigval_t _sigval;
+		} _rt;
+
+		/* SIGCHLD */
+		struct {
+			compat_pid_t _pid;		/* which child */
+			unsigned int _uid;		/* sender's uid */
+			int _status;			/* exit code */
+			compat_clock_t _utime;
+			compat_clock_t _stime;
+		} _sigchld;
+
+		/* SIGILL, SIGFPE, SIGSEGV, SIGBUS, SIGEMT */
+		struct {
+			u32 _addr; /* faulting insn/memory ref. */
+			int _trapno;
+		} _sigfault;
+
+		/* SIGPOLL */
+		struct {
+			int _band;	/* POLL_IN, POLL_OUT, POLL_MSG */
+			int _fd;
+		} _sigpoll;
+	} _sifields;
+} compat_siginfo_t;
+
 #define COMPAT_OFF_T_MAX	0x7fffffff
 #define COMPAT_LOFF_T_MAX	0x7fffffffffffffffL
 
@@ -154,7 +216,6 @@ typedef u32		compat_sigset_word;
  * as pointers because the syscall entry code will have
  * appropriately converted them already.
  */
-typedef	u32		compat_uptr_t;
 
 static inline void __user *compat_ptr(compat_uptr_t uptr)
 {
@@ -166,14 +227,15 @@ static inline compat_uptr_t ptr_to_compat(void __user *uptr)
 	return (u32)(unsigned long)uptr;
 }
 
-static inline void __user *compat_alloc_user_space(long len)
+static inline void __user *arch_compat_alloc_user_space(long len)
 {
 	struct pt_regs *regs = current_thread_info()->kregs;
 	unsigned long usp = regs->u_regs[UREG_I6];
 
-	if (!(test_thread_flag(TIF_32BIT)))
+	if (test_thread_64bit_stack(usp))
 		usp += STACK_BIAS;
-	else
+
+	if (test_thread_flag(TIF_32BIT))
 		usp &= 0xffffffffUL;
 
 	usp -= len;
@@ -240,4 +302,9 @@ struct compat_shmid64_ds {
 	unsigned int	__unused2;
 };
 
+static inline int is_compat_task(void)
+{
+	return test_thread_flag(TIF_32BIT);
+}
+
 #endif /* _ASM_SPARC64_COMPAT_H */
diff --git a/arch/sparc/include/asm/compat_signal.h b/arch/sparc/include/asm/compat_signal.h
index b759eab9b51..9ed1f128b4d 100644
--- a/arch/sparc/include/asm/compat_signal.h
+++ b/arch/sparc/include/asm/compat_signal.h
@@ -18,12 +18,6 @@ struct __old_sigaction32 {
 	unsigned int    	sa_flags;
 	unsigned		sa_restorer;     /* not used by Linux/SPARC yet */
 };
-
-typedef struct sigaltstack32 {
-	u32			ss_sp;
-	int			ss_flags;
-	compat_size_t		ss_size;
-} stack_t32;
 #endif
 
 #endif /* !(_COMPAT_SIGNAL_H) */
diff --git a/arch/sparc/include/asm/contregs.h b/arch/sparc/include/asm/contregs.h
index 48fa8a4ef35..b8abdfcf555 100644
--- a/arch/sparc/include/asm/contregs.h
+++ b/arch/sparc/include/asm/contregs.h
@@ -7,28 +7,6 @@
  * Copyright (C) 1995 David S. Miller (davem@caip.rutgers.edu)
  */
 
-/* 3=sun3
-   4=sun4 (as in sun4 sysmaint student book)
-   c=sun4c (according to davem) */
-
-#define AC_IDPROM     0x00000000    /* 34  ID PROM, R/O, byte, 32 bytes      */
-#define AC_PAGEMAP    0x10000000    /* 3   Pagemap R/W, long                 */
-#define AC_SEGMAP     0x20000000    /* 3   Segment map, byte                 */
-#define AC_CONTEXT    0x30000000    /* 34c current mmu-context               */
-#define AC_SENABLE    0x40000000    /* 34c system dvma/cache/reset enable reg*/
-#define AC_UDVMA_ENB  0x50000000    /* 34  Not used on Sun boards, byte      */
-#define AC_BUS_ERROR  0x60000000    /* 34  Not cleared on read, byte.        */
-#define AC_SYNC_ERR   0x60000000    /*  c fault type                         */
-#define AC_SYNC_VA    0x60000004    /*  c fault virtual address              */
-#define AC_ASYNC_ERR  0x60000008    /*  c asynchronous fault type            */
-#define AC_ASYNC_VA   0x6000000c    /*  c async fault virtual address        */
-#define AC_LEDS       0x70000000    /* 34  Zero turns on LEDs, byte          */
-#define AC_CACHETAGS  0x80000000    /* 34c direct access to the VAC tags     */
-#define AC_CACHEDDATA 0x90000000    /* 3 c direct access to the VAC data     */
-#define AC_UDVMA_MAP  0xD0000000    /* 4  Not used on Sun boards, byte       */
-#define AC_VME_VECTOR 0xE0000000    /* 4  For non-Autovector VME, byte       */
-#define AC_BOOT_SCC   0xF0000000    /* 34  bypass to access Zilog 8530. byte.*/
-
 /* s=Swift, h=Ross_HyperSPARC, v=TI_Viking, t=Tsunami, r=Ross_Cypress        */
 #define AC_M_PCR      0x0000        /* shv Processor Control Reg             */
 #define AC_M_CTPR     0x0100        /* shv Context Table Pointer Reg         */
diff --git a/arch/sparc/include/asm/cpu_type.h b/arch/sparc/include/asm/cpu_type.h
new file mode 100644
index 00000000000..84d7d83b808
--- /dev/null
+++ b/arch/sparc/include/asm/cpu_type.h
@@ -0,0 +1,28 @@
+#ifndef __ASM_CPU_TYPE_H
+#define __ASM_CPU_TYPE_H
+
+/*
+ * Sparc (general) CPU types
+ */
+enum sparc_cpu {
+  sun4m       = 0x00,
+  sun4d       = 0x01,
+  sun4e       = 0x02,
+  sun4u       = 0x03, /* V8 ploos ploos */
+  sun_unknown = 0x04,
+  ap1000      = 0x05, /* almost a sun4m */
+  sparc_leon  = 0x06, /* Leon SoC */
+};
+
+#ifdef CONFIG_SPARC32
+extern enum sparc_cpu sparc_cpu_model;
+
+#define SUN4M_NCPUS            4              /* Architectural limit of sun4m. */
+
+#else
+
+#define sparc_cpu_model sun4u
+
+#endif
+
+#endif /* __ASM_CPU_TYPE_H */
diff --git a/arch/sparc/include/asm/cpudata.h b/arch/sparc/include/asm/cpudata.h
index b5976de7cac..128b56b0867 100644
--- a/arch/sparc/include/asm/cpudata.h
+++ b/arch/sparc/include/asm/cpudata.h
@@ -1,5 +1,15 @@
 #ifndef ___ASM_SPARC_CPUDATA_H
 #define ___ASM_SPARC_CPUDATA_H
+
+#ifndef __ASSEMBLY__
+
+#include <linux/threads.h>
+#include <linux/percpu.h>
+
+extern const struct seq_operations cpuinfo_op;
+
+#endif /* !(__ASSEMBLY__) */
+
 #if defined(__sparc__) && defined(__arch64__)
 #include <asm/cpudata_64.h>
 #else
diff --git a/arch/sparc/include/asm/cpudata_32.h b/arch/sparc/include/asm/cpudata_32.h
index 31d48a0e32c..0300d94c25b 100644
--- a/arch/sparc/include/asm/cpudata_32.h
+++ b/arch/sparc/include/asm/cpudata_32.h
@@ -14,8 +14,11 @@
 typedef struct {
 	unsigned long udelay_val;
 	unsigned long clock_tick;
-	unsigned int multiplier;
 	unsigned int counter;
+#ifdef CONFIG_SMP
+	unsigned int irq_resched_count;
+	unsigned int irq_call_count;
+#endif
 	int prom_node;
 	int mid;
 	int next;
@@ -23,5 +26,6 @@ typedef struct {
 
 DECLARE_PER_CPU(cpuinfo_sparc, __cpu_data);
 #define cpu_data(__cpu) per_cpu(__cpu_data, (__cpu))
+#define local_cpu_data() __get_cpu_var(__cpu_data)
 
 #endif /* _SPARC_CPUDATA_H */
diff --git a/arch/sparc/include/asm/cpudata_64.h b/arch/sparc/include/asm/cpudata_64.h
index 7da7c13d23c..0e594076912 100644
--- a/arch/sparc/include/asm/cpudata_64.h
+++ b/arch/sparc/include/asm/cpudata_64.h
@@ -6,21 +6,15 @@
 #ifndef _SPARC64_CPUDATA_H
 #define _SPARC64_CPUDATA_H
 
-#include <asm/hypervisor.h>
-#include <asm/asi.h>
-
 #ifndef __ASSEMBLY__
 
-#include <linux/percpu.h>
-#include <linux/threads.h>
-
 typedef struct {
 	/* Dcache line 1 */
 	unsigned int	__softirq_pending; /* must be 1st, see rtrap.S */
-	unsigned int	__pad0;
+	unsigned int	__nmi_count;
 	unsigned long	clock_tick;	/* %tick's per second */
 	unsigned long	__pad;
-	unsigned int	__pad1;
+	unsigned int	irq0_irqs;
 	unsigned int	__pad2;
 
 	/* Dcache line 2, rarely used */
@@ -38,202 +32,8 @@ DECLARE_PER_CPU(cpuinfo_sparc, __cpu_data);
 #define cpu_data(__cpu)		per_cpu(__cpu_data, (__cpu))
 #define local_cpu_data()	__get_cpu_var(__cpu_data)
 
-/* Trap handling code needs to get at a few critical values upon
- * trap entry and to process TSB misses.  These cannot be in the
- * per_cpu() area as we really need to lock them into the TLB and
- * thus make them part of the main kernel image.  As a result we
- * try to make this as small as possible.
- *
- * This is padded out and aligned to 64-bytes to avoid false sharing
- * on SMP.
- */
-
-/* If you modify the size of this structure, please update
- * TRAP_BLOCK_SZ_SHIFT below.
- */
-struct thread_info;
-struct trap_per_cpu {
-/* D-cache line 1: Basic thread information, cpu and device mondo queues */
-	struct thread_info	*thread;
-	unsigned long		pgd_paddr;
-	unsigned long		cpu_mondo_pa;
-	unsigned long		dev_mondo_pa;
-
-/* D-cache line 2: Error Mondo Queue and kernel buffer pointers */
-	unsigned long		resum_mondo_pa;
-	unsigned long		resum_kernel_buf_pa;
-	unsigned long		nonresum_mondo_pa;
-	unsigned long		nonresum_kernel_buf_pa;
-
-/* Dcache lines 3, 4, 5, and 6: Hypervisor Fault Status */
-	struct hv_fault_status	fault_info;
-
-/* Dcache line 7: Physical addresses of CPU send mondo block and CPU list.  */
-	unsigned long		cpu_mondo_block_pa;
-	unsigned long		cpu_list_pa;
-	unsigned long		tsb_huge;
-	unsigned long		tsb_huge_temp;
-
-/* Dcache line 8: IRQ work list, and keep trap_block a power-of-2 in size.  */
-	unsigned long		irq_worklist_pa;
-	unsigned int		cpu_mondo_qmask;
-	unsigned int		dev_mondo_qmask;
-	unsigned int		resum_qmask;
-	unsigned int		nonresum_qmask;
-	void			*hdesc;
-} __attribute__((aligned(64)));
-extern struct trap_per_cpu trap_block[NR_CPUS];
-extern void init_cur_cpu_trap(struct thread_info *);
-extern void setup_tba(void);
-extern int ncpus_probed;
-extern const struct seq_operations cpuinfo_op;
-
-extern unsigned long real_hard_smp_processor_id(void);
-
-struct cpuid_patch_entry {
-	unsigned int	addr;
-	unsigned int	cheetah_safari[4];
-	unsigned int	cheetah_jbus[4];
-	unsigned int	starfire[4];
-	unsigned int	sun4v[4];
-};
-extern struct cpuid_patch_entry __cpuid_patch, __cpuid_patch_end;
-
-struct sun4v_1insn_patch_entry {
-	unsigned int	addr;
-	unsigned int	insn;
-};
-extern struct sun4v_1insn_patch_entry __sun4v_1insn_patch,
-	__sun4v_1insn_patch_end;
-
-struct sun4v_2insn_patch_entry {
-	unsigned int	addr;
-	unsigned int	insns[2];
-};
-extern struct sun4v_2insn_patch_entry __sun4v_2insn_patch,
-	__sun4v_2insn_patch_end;
-
 #endif /* !(__ASSEMBLY__) */
 
-#define TRAP_PER_CPU_THREAD		0x00
-#define TRAP_PER_CPU_PGD_PADDR		0x08
-#define TRAP_PER_CPU_CPU_MONDO_PA	0x10
-#define TRAP_PER_CPU_DEV_MONDO_PA	0x18
-#define TRAP_PER_CPU_RESUM_MONDO_PA	0x20
-#define TRAP_PER_CPU_RESUM_KBUF_PA	0x28
-#define TRAP_PER_CPU_NONRESUM_MONDO_PA	0x30
-#define TRAP_PER_CPU_NONRESUM_KBUF_PA	0x38
-#define TRAP_PER_CPU_FAULT_INFO		0x40
-#define TRAP_PER_CPU_CPU_MONDO_BLOCK_PA	0xc0
-#define TRAP_PER_CPU_CPU_LIST_PA	0xc8
-#define TRAP_PER_CPU_TSB_HUGE		0xd0
-#define TRAP_PER_CPU_TSB_HUGE_TEMP	0xd8
-#define TRAP_PER_CPU_IRQ_WORKLIST_PA	0xe0
-#define TRAP_PER_CPU_CPU_MONDO_QMASK	0xe8
-#define TRAP_PER_CPU_DEV_MONDO_QMASK	0xec
-#define TRAP_PER_CPU_RESUM_QMASK	0xf0
-#define TRAP_PER_CPU_NONRESUM_QMASK	0xf4
-
-#define TRAP_BLOCK_SZ_SHIFT		8
-
-#include <asm/scratchpad.h>
-
-#define __GET_CPUID(REG)				\
-	/* Spitfire implementation (default). */	\
-661:	ldxa		[%g0] ASI_UPA_CONFIG, REG;	\
-	srlx		REG, 17, REG;			\
-	 and		REG, 0x1f, REG;			\
-	nop;						\
-	.section	.cpuid_patch, "ax";		\
-	/* Instruction location. */			\
-	.word		661b;				\
-	/* Cheetah Safari implementation. */		\
-	ldxa		[%g0] ASI_SAFARI_CONFIG, REG;	\
-	srlx		REG, 17, REG;			\
-	and		REG, 0x3ff, REG;		\
-	nop;						\
-	/* Cheetah JBUS implementation. */		\
-	ldxa		[%g0] ASI_JBUS_CONFIG, REG;	\
-	srlx		REG, 17, REG;			\
-	and		REG, 0x1f, REG;			\
-	nop;						\
-	/* Starfire implementation. */			\
-	sethi		%hi(0x1fff40000d0 >> 9), REG;	\
-	sllx		REG, 9, REG;			\
-	or		REG, 0xd0, REG;			\
-	lduwa		[REG] ASI_PHYS_BYPASS_EC_E, REG;\
-	/* sun4v implementation. */			\
-	mov		SCRATCHPAD_CPUID, REG;		\
-	ldxa		[REG] ASI_SCRATCHPAD, REG;	\
-	nop;						\
-	nop;						\
-	.previous;
-
-#ifdef CONFIG_SMP
-
-#define TRAP_LOAD_TRAP_BLOCK(DEST, TMP)		\
-	__GET_CPUID(TMP)			\
-	sethi	%hi(trap_block), DEST;		\
-	sllx	TMP, TRAP_BLOCK_SZ_SHIFT, TMP;	\
-	or	DEST, %lo(trap_block), DEST;	\
-	add	DEST, TMP, DEST;		\
-
-/* Clobbers TMP, current address space PGD phys address into DEST.  */
-#define TRAP_LOAD_PGD_PHYS(DEST, TMP)		\
-	TRAP_LOAD_TRAP_BLOCK(DEST, TMP)		\
-	ldx	[DEST + TRAP_PER_CPU_PGD_PADDR], DEST;
-
-/* Clobbers TMP, loads local processor's IRQ work area into DEST.  */
-#define TRAP_LOAD_IRQ_WORK_PA(DEST, TMP)	\
-	TRAP_LOAD_TRAP_BLOCK(DEST, TMP)		\
-	add	DEST, TRAP_PER_CPU_IRQ_WORKLIST_PA, DEST;
-
-/* Clobbers TMP, loads DEST with current thread info pointer.  */
-#define TRAP_LOAD_THREAD_REG(DEST, TMP)		\
-	TRAP_LOAD_TRAP_BLOCK(DEST, TMP)		\
-	ldx	[DEST + TRAP_PER_CPU_THREAD], DEST;
-
-/* Given the current thread info pointer in THR, load the per-cpu
- * area base of the current processor into DEST.  REG1, REG2, and REG3 are
- * clobbered.
- *
- * You absolutely cannot use DEST as a temporary in this code.  The
- * reason is that traps can happen during execution, and return from
- * trap will load the fully resolved DEST per-cpu base.  This can corrupt
- * the calculations done by the macro mid-stream.
- */
-#define LOAD_PER_CPU_BASE(DEST, THR, REG1, REG2, REG3)	\
-	lduh	[THR + TI_CPU], REG1;			\
-	sethi	%hi(__per_cpu_shift), REG3;		\
-	sethi	%hi(__per_cpu_base), REG2;		\
-	ldx	[REG3 + %lo(__per_cpu_shift)], REG3;	\
-	ldx	[REG2 + %lo(__per_cpu_base)], REG2;	\
-	sllx	REG1, REG3, REG3;			\
-	add	REG3, REG2, DEST;
-
-#else
-
-#define TRAP_LOAD_TRAP_BLOCK(DEST, TMP)		\
-	sethi	%hi(trap_block), DEST;		\
-	or	DEST, %lo(trap_block), DEST;	\
-
-/* Uniprocessor versions, we know the cpuid is zero.  */
-#define TRAP_LOAD_PGD_PHYS(DEST, TMP)		\
-	TRAP_LOAD_TRAP_BLOCK(DEST, TMP)		\
-	ldx	[DEST + TRAP_PER_CPU_PGD_PADDR], DEST;
-
-/* Clobbers TMP, loads local processor's IRQ work area into DEST.  */
-#define TRAP_LOAD_IRQ_WORK_PA(DEST, TMP)	\
-	TRAP_LOAD_TRAP_BLOCK(DEST, TMP)		\
-	add	DEST, TRAP_PER_CPU_IRQ_WORKLIST_PA, DEST;
-
-#define TRAP_LOAD_THREAD_REG(DEST, TMP)		\
-	TRAP_LOAD_TRAP_BLOCK(DEST, TMP)		\
-	ldx	[DEST + TRAP_PER_CPU_THREAD], DEST;
-
-/* No per-cpu areas on uniprocessor, so no need to load DEST.  */
-#define LOAD_PER_CPU_BASE(DEST, THR, REG1, REG2, REG3)
-
-#endif /* !(CONFIG_SMP) */
+#include <asm/trap_block.h>
 
 #endif /* _SPARC64_CPUDATA_H */
diff --git a/arch/sparc/include/asm/cputime.h b/arch/sparc/include/asm/cputime.h
deleted file mode 100644
index 1a642b81e01..00000000000
--- a/arch/sparc/include/asm/cputime.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __SPARC_CPUTIME_H
-#define __SPARC_CPUTIME_H
-
-#include <asm-generic/cputime.h>
-
-#endif /* __SPARC_CPUTIME_H */
diff --git a/arch/sparc/include/asm/cypress.h b/arch/sparc/include/asm/cypress.h
deleted file mode 100644
index 95e9772ea39..00000000000
--- a/arch/sparc/include/asm/cypress.h
+++ /dev/null
@@ -1,79 +0,0 @@
-/*
- * cypress.h: Cypress module specific definitions and defines.
- *
- * Copyright (C) 1995 David S. Miller (davem@caip.rutgers.edu)
- */
-
-#ifndef _SPARC_CYPRESS_H
-#define _SPARC_CYPRESS_H
-
-/* Cypress chips have %psr 'impl' of '0001' and 'vers' of '0001'. */
-
-/* The MMU control register fields on the Sparc Cypress 604/605 MMU's.
- *
- * ---------------------------------------------------------------
- * |implvers| MCA | MCM |MV| MID |BM| C|RSV|MR|CM|CL|CE|RSV|NF|ME|
- * ---------------------------------------------------------------
- *  31    24 23-22 21-20 19 18-15 14 13  12 11 10  9  8 7-2  1  0
- *
- * MCA: MultiChip Access -- Used for configuration of multiple
- *      CY7C604/605 cache units.
- * MCM: MultiChip Mask -- Again, for multiple cache unit config.
- * MV: MultiChip Valid -- Indicates MCM and MCA have valid settings.
- * MID: ModuleID -- Unique processor ID for MBus transactions. (605 only)
- * BM: Boot Mode -- 0 = not in boot mode, 1 = in boot mode
- * C: Cacheable -- Indicates whether accesses are cacheable while
- *    the MMU is off.  0=no 1=yes
- * MR: MemoryReflection -- Indicates whether the bus attached to the
- *     MBus supports memory reflection. 0=no 1=yes (605 only)
- * CM: CacheMode -- Indicates whether the cache is operating in write
- *     through or copy-back mode. 0=write-through 1=copy-back
- * CL: CacheLock -- Indicates if the entire cache is locked or not.
- *     0=not-locked 1=locked  (604 only)
- * CE: CacheEnable -- Is the virtual cache on? 0=no 1=yes
- * NF: NoFault -- Do faults generate traps? 0=yes 1=no
- * ME: MmuEnable -- Is the MMU doing translations? 0=no 1=yes
- */
-
-#define CYPRESS_MCA       0x00c00000
-#define CYPRESS_MCM       0x00300000
-#define CYPRESS_MVALID    0x00080000
-#define CYPRESS_MIDMASK   0x00078000   /* Only on 605 */
-#define CYPRESS_BMODE     0x00004000
-#define CYPRESS_ACENABLE  0x00002000
-#define CYPRESS_MRFLCT    0x00000800   /* Only on 605 */
-#define CYPRESS_CMODE     0x00000400
-#define CYPRESS_CLOCK     0x00000200   /* Only on 604 */
-#define CYPRESS_CENABLE   0x00000100
-#define CYPRESS_NFAULT    0x00000002
-#define CYPRESS_MENABLE   0x00000001
-
-static inline void cypress_flush_page(unsigned long page)
-{
-	__asm__ __volatile__("sta %%g0, [%0] %1\n\t" : :
-			     "r" (page), "i" (ASI_M_FLUSH_PAGE));
-}
-
-static inline void cypress_flush_segment(unsigned long addr)
-{
-	__asm__ __volatile__("sta %%g0, [%0] %1\n\t" : :
-			     "r" (addr), "i" (ASI_M_FLUSH_SEG));
-}
-
-static inline void cypress_flush_region(unsigned long addr)
-{
-	__asm__ __volatile__("sta %%g0, [%0] %1\n\t" : :
-			     "r" (addr), "i" (ASI_M_FLUSH_REGION));
-}
-
-static inline void cypress_flush_context(void)
-{
-	__asm__ __volatile__("sta %%g0, [%%g0] %0\n\t" : :
-			     "i" (ASI_M_FLUSH_CTX));
-}
-
-/* XXX Displacement flushes for buggy chips and initial testing
- * XXX go here.
- */
-
-#endif /* !(_SPARC_CYPRESS_H) */
diff --git a/arch/sparc/include/asm/delay_32.h b/arch/sparc/include/asm/delay_32.h
index bc9aba2bead..3fb8ca144b4 100644
--- a/arch/sparc/include/asm/delay_32.h
+++ b/arch/sparc/include/asm/delay_32.h
@@ -20,8 +20,8 @@ static inline void __delay(unsigned long loops)
 }
 
 /* This is too messy with inline asm on the Sparc. */
-extern void __udelay(unsigned long usecs, unsigned long lpj);
-extern void __ndelay(unsigned long nsecs, unsigned long lpj);
+void __udelay(unsigned long usecs, unsigned long lpj);
+void __ndelay(unsigned long nsecs, unsigned long lpj);
 
 #ifdef CONFIG_SMP
 #define __udelay_val	cpu_data(smp_processor_id()).udelay_val
diff --git a/arch/sparc/include/asm/delay_64.h b/arch/sparc/include/asm/delay_64.h
index a77aa622d76..0ba5424856d 100644
--- a/arch/sparc/include/asm/delay_64.h
+++ b/arch/sparc/include/asm/delay_64.h
@@ -8,8 +8,8 @@
 
 #ifndef __ASSEMBLY__
 
-extern void __delay(unsigned long loops);
-extern void udelay(unsigned long usecs);
+void __delay(unsigned long loops);
+void udelay(unsigned long usecs);
 #define mdelay(n)	udelay((n) * 1000)
 
 #endif /* !__ASSEMBLY__ */
diff --git a/arch/sparc/include/asm/device.h b/arch/sparc/include/asm/device.h
index 19790eb99cc..bb3f0b0c675 100644
--- a/arch/sparc/include/asm/device.h
+++ b/arch/sparc/include/asm/device.h
@@ -6,18 +6,25 @@
 #ifndef _ASM_SPARC_DEVICE_H
 #define _ASM_SPARC_DEVICE_H
 
+#include <asm/openprom.h>
+
 struct device_node;
-struct of_device;
+struct platform_device;
 
 struct dev_archdata {
 	void			*iommu;
 	void			*stc;
 	void			*host_controller;
+	struct platform_device	*op;
+	int			numa_node;
+};
 
-	struct device_node	*prom_node;
-	struct of_device	*op;
+void of_propagate_archdata(struct platform_device *bus);
 
-	int			numa_node;
+struct pdev_archdata {
+	struct resource		resource[PROMREG_MAX];
+	unsigned int		irqs[PROMINTR_MAX];
+	int			num_irqs;
 };
 
 #endif /* _ASM_SPARC_DEVICE_H */
diff --git a/arch/sparc/include/asm/div64.h b/arch/sparc/include/asm/div64.h
deleted file mode 100644
index 6cd978cefb2..00000000000
--- a/arch/sparc/include/asm/div64.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/div64.h>
diff --git a/arch/sparc/include/asm/dma-mapping.h b/arch/sparc/include/asm/dma-mapping.h
index 0f4150e2661..1ee02710b2d 100644
--- a/arch/sparc/include/asm/dma-mapping.h
+++ b/arch/sparc/include/asm/dma-mapping.h
@@ -1,8 +1,79 @@
 #ifndef ___ASM_SPARC_DMA_MAPPING_H
 #define ___ASM_SPARC_DMA_MAPPING_H
-#if defined(__sparc__) && defined(__arch64__)
-#include <asm/dma-mapping_64.h>
-#else
-#include <asm/dma-mapping_32.h>
+
+#include <linux/scatterlist.h>
+#include <linux/mm.h>
+#include <linux/dma-debug.h>
+
+#define DMA_ERROR_CODE	(~(dma_addr_t)0x0)
+
+int dma_supported(struct device *dev, u64 mask);
+
+#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
+#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
+
+extern struct dma_map_ops *dma_ops;
+extern struct dma_map_ops *leon_dma_ops;
+extern struct dma_map_ops pci32_dma_ops;
+
+extern struct bus_type pci_bus_type;
+
+static inline struct dma_map_ops *get_dma_ops(struct device *dev)
+{
+#if defined(CONFIG_SPARC32) && defined(CONFIG_PCI)
+	if (sparc_cpu_model == sparc_leon)
+		return leon_dma_ops;
+	else if (dev->bus == &pci_bus_type)
+		return &pci32_dma_ops;
 #endif
+	return dma_ops;
+}
+
+#include <asm-generic/dma-mapping-common.h>
+
+#define dma_alloc_coherent(d,s,h,f)	dma_alloc_attrs(d,s,h,f,NULL)
+
+static inline void *dma_alloc_attrs(struct device *dev, size_t size,
+				    dma_addr_t *dma_handle, gfp_t flag,
+				    struct dma_attrs *attrs)
+{
+	struct dma_map_ops *ops = get_dma_ops(dev);
+	void *cpu_addr;
+
+	cpu_addr = ops->alloc(dev, size, dma_handle, flag, attrs);
+	debug_dma_alloc_coherent(dev, size, *dma_handle, cpu_addr);
+	return cpu_addr;
+}
+
+#define dma_free_coherent(d,s,c,h) dma_free_attrs(d,s,c,h,NULL)
+
+static inline void dma_free_attrs(struct device *dev, size_t size,
+				  void *cpu_addr, dma_addr_t dma_handle,
+				  struct dma_attrs *attrs)
+{
+	struct dma_map_ops *ops = get_dma_ops(dev);
+
+	debug_dma_free_coherent(dev, size, cpu_addr, dma_handle);
+	ops->free(dev, size, cpu_addr, dma_handle, attrs);
+}
+
+static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
+{
+	debug_dma_mapping_error(dev, dma_addr);
+	return (dma_addr == DMA_ERROR_CODE);
+}
+
+static inline int dma_set_mask(struct device *dev, u64 mask)
+{
+#ifdef CONFIG_PCI
+	if (dev->bus == &pci_bus_type) {
+		if (!dev->dma_mask || !dma_supported(dev, mask))
+			return -EINVAL;
+		*dev->dma_mask = mask;
+		return 0;
+	}
+#endif
+	return -EINVAL;
+}
+
 #endif
diff --git a/arch/sparc/include/asm/dma-mapping_32.h b/arch/sparc/include/asm/dma-mapping_32.h
deleted file mode 100644
index 8a57ea0573e..00000000000
--- a/arch/sparc/include/asm/dma-mapping_32.h
+++ /dev/null
@@ -1,60 +0,0 @@
-#ifndef _ASM_SPARC_DMA_MAPPING_H
-#define _ASM_SPARC_DMA_MAPPING_H
-
-#include <linux/types.h>
-
-struct device;
-struct scatterlist;
-struct page;
-
-#define DMA_ERROR_CODE	(~(dma_addr_t)0x0)
-
-extern int dma_supported(struct device *dev, u64 mask);
-extern int dma_set_mask(struct device *dev, u64 dma_mask);
-extern void *dma_alloc_coherent(struct device *dev, size_t size,
-				dma_addr_t *dma_handle, gfp_t flag);
-extern void dma_free_coherent(struct device *dev, size_t size,
-			      void *cpu_addr, dma_addr_t dma_handle);
-extern dma_addr_t dma_map_single(struct device *dev, void *cpu_addr,
-				 size_t size,
-				 enum dma_data_direction direction);
-extern void dma_unmap_single(struct device *dev, dma_addr_t dma_addr,
-			     size_t size,
-			     enum dma_data_direction direction);
-extern dma_addr_t dma_map_page(struct device *dev, struct page *page,
-			       unsigned long offset, size_t size,
-			       enum dma_data_direction direction);
-extern void dma_unmap_page(struct device *dev, dma_addr_t dma_address,
-			   size_t size, enum dma_data_direction direction);
-extern int dma_map_sg(struct device *dev, struct scatterlist *sg,
-		      int nents, enum dma_data_direction direction);
-extern void dma_unmap_sg(struct device *dev, struct scatterlist *sg,
-			 int nents, enum dma_data_direction direction);
-extern void dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle,
-				    size_t size,
-				    enum dma_data_direction direction);
-extern void dma_sync_single_for_device(struct device *dev,
-				       dma_addr_t dma_handle,
-				       size_t size,
-				       enum dma_data_direction direction);
-extern void dma_sync_single_range_for_cpu(struct device *dev,
-					  dma_addr_t dma_handle,
-					  unsigned long offset,
-					  size_t size,
-					  enum dma_data_direction direction);
-extern void dma_sync_single_range_for_device(struct device *dev,
-					     dma_addr_t dma_handle,
-					     unsigned long offset, size_t size,
-					     enum dma_data_direction direction);
-extern void dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg,
-				int nelems, enum dma_data_direction direction);
-extern void dma_sync_sg_for_device(struct device *dev,
-				   struct scatterlist *sg, int nelems,
-				   enum dma_data_direction direction);
-extern int dma_mapping_error(struct device *dev, dma_addr_t dma_addr);
-extern int dma_get_cache_alignment(void);
-
-#define dma_alloc_noncoherent	dma_alloc_coherent
-#define dma_free_noncoherent	dma_free_coherent
-
-#endif /* _ASM_SPARC_DMA_MAPPING_H */
diff --git a/arch/sparc/include/asm/dma-mapping_64.h b/arch/sparc/include/asm/dma-mapping_64.h
deleted file mode 100644
index bfa64f9702d..00000000000
--- a/arch/sparc/include/asm/dma-mapping_64.h
+++ /dev/null
@@ -1,154 +0,0 @@
-#ifndef _ASM_SPARC64_DMA_MAPPING_H
-#define _ASM_SPARC64_DMA_MAPPING_H
-
-#include <linux/scatterlist.h>
-#include <linux/mm.h>
-
-#define DMA_ERROR_CODE	(~(dma_addr_t)0x0)
-
-struct dma_ops {
-	void *(*alloc_coherent)(struct device *dev, size_t size,
-				dma_addr_t *dma_handle, gfp_t flag);
-	void (*free_coherent)(struct device *dev, size_t size,
-			      void *cpu_addr, dma_addr_t dma_handle);
-	dma_addr_t (*map_single)(struct device *dev, void *cpu_addr,
-				 size_t size,
-				 enum dma_data_direction direction);
-	void (*unmap_single)(struct device *dev, dma_addr_t dma_addr,
-			     size_t size,
-			     enum dma_data_direction direction);
-	int (*map_sg)(struct device *dev, struct scatterlist *sg, int nents,
-		      enum dma_data_direction direction);
-	void (*unmap_sg)(struct device *dev, struct scatterlist *sg,
-			 int nhwentries,
-			 enum dma_data_direction direction);
-	void (*sync_single_for_cpu)(struct device *dev,
-				    dma_addr_t dma_handle, size_t size,
-				    enum dma_data_direction direction);
-	void (*sync_sg_for_cpu)(struct device *dev, struct scatterlist *sg,
-				int nelems,
-				enum dma_data_direction direction);
-};
-extern const struct dma_ops *dma_ops;
-
-extern int dma_supported(struct device *dev, u64 mask);
-extern int dma_set_mask(struct device *dev, u64 dma_mask);
-
-static inline void *dma_alloc_coherent(struct device *dev, size_t size,
-				       dma_addr_t *dma_handle, gfp_t flag)
-{
-	return dma_ops->alloc_coherent(dev, size, dma_handle, flag);
-}
-
-static inline void dma_free_coherent(struct device *dev, size_t size,
-				     void *cpu_addr, dma_addr_t dma_handle)
-{
-	dma_ops->free_coherent(dev, size, cpu_addr, dma_handle);
-}
-
-static inline dma_addr_t dma_map_single(struct device *dev, void *cpu_addr,
-					size_t size,
-					enum dma_data_direction direction)
-{
-	return dma_ops->map_single(dev, cpu_addr, size, direction);
-}
-
-static inline void dma_unmap_single(struct device *dev, dma_addr_t dma_addr,
-				    size_t size,
-				    enum dma_data_direction direction)
-{
-	dma_ops->unmap_single(dev, dma_addr, size, direction);
-}
-
-static inline dma_addr_t dma_map_page(struct device *dev, struct page *page,
-				      unsigned long offset, size_t size,
-				      enum dma_data_direction direction)
-{
-	return dma_ops->map_single(dev, page_address(page) + offset,
-				   size, direction);
-}
-
-static inline void dma_unmap_page(struct device *dev, dma_addr_t dma_address,
-				  size_t size,
-				  enum dma_data_direction direction)
-{
-	dma_ops->unmap_single(dev, dma_address, size, direction);
-}
-
-static inline int dma_map_sg(struct device *dev, struct scatterlist *sg,
-			     int nents, enum dma_data_direction direction)
-{
-	return dma_ops->map_sg(dev, sg, nents, direction);
-}
-
-static inline void dma_unmap_sg(struct device *dev, struct scatterlist *sg,
-				int nents, enum dma_data_direction direction)
-{
-	dma_ops->unmap_sg(dev, sg, nents, direction);
-}
-
-static inline void dma_sync_single_for_cpu(struct device *dev,
-					   dma_addr_t dma_handle, size_t size,
-					   enum dma_data_direction direction)
-{
-	dma_ops->sync_single_for_cpu(dev, dma_handle, size, direction);
-}
-
-static inline void dma_sync_single_for_device(struct device *dev,
-					      dma_addr_t dma_handle,
-					      size_t size,
-					      enum dma_data_direction direction)
-{
-	/* No flushing needed to sync cpu writes to the device.  */
-}
-
-static inline void dma_sync_single_range_for_cpu(struct device *dev,
-						 dma_addr_t dma_handle,
-						 unsigned long offset,
-						 size_t size,
-						 enum dma_data_direction direction)
-{
-	dma_sync_single_for_cpu(dev, dma_handle+offset, size, direction);
-}
-
-static inline void dma_sync_single_range_for_device(struct device *dev,
-						    dma_addr_t dma_handle,
-						    unsigned long offset,
-						    size_t size,
-						    enum dma_data_direction direction)
-{
-	/* No flushing needed to sync cpu writes to the device.  */
-}
-
-
-static inline void dma_sync_sg_for_cpu(struct device *dev,
-				       struct scatterlist *sg, int nelems,
-				       enum dma_data_direction direction)
-{
-	dma_ops->sync_sg_for_cpu(dev, sg, nelems, direction);
-}
-
-static inline void dma_sync_sg_for_device(struct device *dev,
-					  struct scatterlist *sg, int nelems,
-					  enum dma_data_direction direction)
-{
-	/* No flushing needed to sync cpu writes to the device.  */
-}
-
-static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
-{
-	return (dma_addr == DMA_ERROR_CODE);
-}
-
-static inline int dma_get_cache_alignment(void)
-{
-	/* no easy way to get cache size on all processors, so return
-	 * the maximum possible, to be safe */
-	return (1 << INTERNODE_CACHE_SHIFT);
-}
-
-#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
-#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
-#define dma_is_consistent(d, h)	(1)
-
-#endif /* _ASM_SPARC64_DMA_MAPPING_H */
diff --git a/arch/sparc/include/asm/dma.h b/arch/sparc/include/asm/dma.h
index b554927bbaf..3d434ef5eae 100644
--- a/arch/sparc/include/asm/dma.h
+++ b/arch/sparc/include/asm/dma.h
@@ -92,27 +92,31 @@ extern int isa_dma_bridge_buggy;
 #ifdef CONFIG_SPARC32
 
 /* Routines for data transfer buffers. */
-BTFIXUPDEF_CALL(char *, mmu_lockarea, char *, unsigned long)
-BTFIXUPDEF_CALL(void,   mmu_unlockarea, char *, unsigned long)
-
-#define mmu_lockarea(vaddr,len) BTFIXUP_CALL(mmu_lockarea)(vaddr,len)
-#define mmu_unlockarea(vaddr,len) BTFIXUP_CALL(mmu_unlockarea)(vaddr,len)
-
-struct page;
 struct device;
 struct scatterlist;
 
-/* These are implementations for sbus_map_sg/sbus_unmap_sg... collapse later */
-BTFIXUPDEF_CALL(__u32, mmu_get_scsi_one, struct device *, char *, unsigned long)
-BTFIXUPDEF_CALL(void,  mmu_get_scsi_sgl, struct device *, struct scatterlist *, int)
-BTFIXUPDEF_CALL(void,  mmu_release_scsi_one, struct device *, __u32, unsigned long)
-BTFIXUPDEF_CALL(void,  mmu_release_scsi_sgl, struct device *, struct scatterlist *, int)
-
-#define mmu_get_scsi_one(dev,vaddr,len) BTFIXUP_CALL(mmu_get_scsi_one)(dev,vaddr,len)
-#define mmu_get_scsi_sgl(dev,sg,sz) BTFIXUP_CALL(mmu_get_scsi_sgl)(dev,sg,sz)
-#define mmu_release_scsi_one(dev,vaddr,len) BTFIXUP_CALL(mmu_release_scsi_one)(dev,vaddr,len)
-#define mmu_release_scsi_sgl(dev,sg,sz) BTFIXUP_CALL(mmu_release_scsi_sgl)(dev,sg,sz)
-
+struct sparc32_dma_ops {
+	__u32 (*get_scsi_one)(struct device *, char *, unsigned long);
+	void (*get_scsi_sgl)(struct device *, struct scatterlist *, int);
+	void (*release_scsi_one)(struct device *, __u32, unsigned long);
+	void (*release_scsi_sgl)(struct device *, struct scatterlist *,int);
+#ifdef CONFIG_SBUS
+	int (*map_dma_area)(struct device *, dma_addr_t *, unsigned long, unsigned long, int);
+	void (*unmap_dma_area)(struct device *, unsigned long, int);
+#endif
+};
+extern const struct sparc32_dma_ops *sparc32_dma_ops;
+
+#define mmu_get_scsi_one(dev,vaddr,len) \
+	sparc32_dma_ops->get_scsi_one(dev, vaddr, len)
+#define mmu_get_scsi_sgl(dev,sg,sz) \
+	sparc32_dma_ops->get_scsi_sgl(dev, sg, sz)
+#define mmu_release_scsi_one(dev,vaddr,len) \
+	sparc32_dma_ops->release_scsi_one(dev, vaddr,len)
+#define mmu_release_scsi_sgl(dev,sg,sz) \
+	sparc32_dma_ops->release_scsi_sgl(dev, sg, sz)
+
+#ifdef CONFIG_SBUS
 /*
  * mmu_map/unmap are provided by iommu/iounit; Invalid to call on IIep.
  *
@@ -123,17 +127,17 @@ BTFIXUPDEF_CALL(void,  mmu_release_scsi_sgl, struct device *, struct scatterlist
  * Second mapping is for device visible address, or "bus" address.
  * The bus address is returned at '*pba'.
  *
- * These functions seem distinct, but are hard to split. On sun4c,
- * at least for now, 'a' is equal to bus address, and retured in *pba.
+ * These functions seem distinct, but are hard to split.
  * On sun4m, page attributes depend on the CPU type, so we have to
  * know if we are mapping RAM or I/O, so it has to be an additional argument
  * to a separate mapping function for CPU visible mappings.
  */
-BTFIXUPDEF_CALL(int, mmu_map_dma_area, struct device *, dma_addr_t *, unsigned long, unsigned long, int len)
-BTFIXUPDEF_CALL(void, mmu_unmap_dma_area, struct device *, unsigned long busa, int len)
+#define sbus_map_dma_area(dev,pba,va,a,len) \
+	sparc32_dma_ops->map_dma_area(dev, pba, va, a, len)
+#define sbus_unmap_dma_area(dev,ba,len) \
+	sparc32_dma_ops->unmap_dma_area(dev, ba, len)
+#endif /* CONFIG_SBUS */
 
-#define mmu_map_dma_area(dev,pba,va,a,len) BTFIXUP_CALL(mmu_map_dma_area)(dev,pba,va,a,len)
-#define mmu_unmap_dma_area(dev,ba,len) BTFIXUP_CALL(mmu_unmap_dma_area)(dev,ba,len)
 #endif
 
 #endif /* !(_ASM_SPARC_DMA_H) */
diff --git a/arch/sparc/include/asm/ebus_dma.h b/arch/sparc/include/asm/ebus_dma.h
index f07a5b541c9..fcfb4948147 100644
--- a/arch/sparc/include/asm/ebus_dma.h
+++ b/arch/sparc/include/asm/ebus_dma.h
@@ -22,14 +22,14 @@ struct ebus_dma_info {
 	unsigned char	name[64];
 };
 
-extern int ebus_dma_register(struct ebus_dma_info *p);
-extern int ebus_dma_irq_enable(struct ebus_dma_info *p, int on);
-extern void ebus_dma_unregister(struct ebus_dma_info *p);
-extern int ebus_dma_request(struct ebus_dma_info *p, dma_addr_t bus_addr,
+int ebus_dma_register(struct ebus_dma_info *p);
+int ebus_dma_irq_enable(struct ebus_dma_info *p, int on);
+void ebus_dma_unregister(struct ebus_dma_info *p);
+int ebus_dma_request(struct ebus_dma_info *p, dma_addr_t bus_addr,
 			    size_t len);
-extern void ebus_dma_prepare(struct ebus_dma_info *p, int write);
-extern unsigned int ebus_dma_residue(struct ebus_dma_info *p);
-extern unsigned int ebus_dma_addr(struct ebus_dma_info *p);
-extern void ebus_dma_enable(struct ebus_dma_info *p, int on);
+void ebus_dma_prepare(struct ebus_dma_info *p, int write);
+unsigned int ebus_dma_residue(struct ebus_dma_info *p);
+unsigned int ebus_dma_addr(struct ebus_dma_info *p);
+void ebus_dma_enable(struct ebus_dma_info *p, int on);
 
 #endif /* __ASM_SPARC_EBUS_DMA_H */
diff --git a/arch/sparc/include/asm/elf_32.h b/arch/sparc/include/asm/elf_32.h
index 381a1b5256d..a24e41fcdde 100644
--- a/arch/sparc/include/asm/elf_32.h
+++ b/arch/sparc/include/asm/elf_32.h
@@ -104,8 +104,6 @@ typedef struct {
 #define ELF_CLASS	ELFCLASS32
 #define ELF_DATA	ELFDATA2MSB
 
-#define USE_ELF_CORE_DUMP
-
 #define ELF_EXEC_PAGESIZE	4096
 
 
@@ -120,16 +118,9 @@ typedef struct {
    instruction set this cpu supports.  This can NOT be done in userspace
    on Sparc.  */
 
-/* Sun4c has none of the capabilities, most sun4m's have them all.
- * XXX This is gross, set some global variable at boot time. -DaveM
- */
-#define ELF_HWCAP	((ARCH_SUN4C) ? 0 : \
-			 (HWCAP_SPARC_FLUSH | HWCAP_SPARC_STBAR | \
-			  HWCAP_SPARC_SWAP | \
-			  ((srmmu_modtype != Cypress && \
-			    srmmu_modtype != Cypress_vE && \
-			    srmmu_modtype != Cypress_vD) ? \
-			   HWCAP_SPARC_MULDIV : 0)))
+/* Most sun4m's have them all.  */
+#define ELF_HWCAP	(HWCAP_SPARC_FLUSH | HWCAP_SPARC_STBAR | \
+			 HWCAP_SPARC_SWAP | HWCAP_SPARC_MULDIV)
 
 /* This yields a string that ld.so will use to load implementation
    specific libraries for optimization.  This is more specific in
@@ -137,6 +128,4 @@ typedef struct {
 
 #define ELF_PLATFORM	(NULL)
 
-#define SET_PERSONALITY(ex) set_personality(PER_LINUX)
-
 #endif /* !(__ASMSPARC_ELF_H) */
diff --git a/arch/sparc/include/asm/elf_64.h b/arch/sparc/include/asm/elf_64.h
index 425c2f9be6d..370ca1e71ff 100644
--- a/arch/sparc/include/asm/elf_64.h
+++ b/arch/sparc/include/asm/elf_64.h
@@ -59,15 +59,42 @@
 #define R_SPARC_6		45
 
 /* Bits present in AT_HWCAP, primarily for Sparc32.  */
-
-#define HWCAP_SPARC_FLUSH       1    /* CPU supports flush instruction. */
-#define HWCAP_SPARC_STBAR       2
-#define HWCAP_SPARC_SWAP        4
-#define HWCAP_SPARC_MULDIV      8
-#define HWCAP_SPARC_V9		16
-#define HWCAP_SPARC_ULTRA3	32
-#define HWCAP_SPARC_BLKINIT	64
-#define HWCAP_SPARC_N2		128
+#define HWCAP_SPARC_FLUSH       0x00000001
+#define HWCAP_SPARC_STBAR       0x00000002
+#define HWCAP_SPARC_SWAP        0x00000004
+#define HWCAP_SPARC_MULDIV      0x00000008
+#define HWCAP_SPARC_V9		0x00000010
+#define HWCAP_SPARC_ULTRA3	0x00000020
+#define HWCAP_SPARC_BLKINIT	0x00000040
+#define HWCAP_SPARC_N2		0x00000080
+
+/* Solaris compatible AT_HWCAP bits. */
+#define AV_SPARC_MUL32		0x00000100 /* 32x32 multiply is efficient */
+#define AV_SPARC_DIV32		0x00000200 /* 32x32 divide is efficient */
+#define AV_SPARC_FSMULD		0x00000400 /* 'fsmuld' is efficient */
+#define AV_SPARC_V8PLUS		0x00000800 /* v9 insn available to 32bit */
+#define AV_SPARC_POPC		0x00001000 /* 'popc' is efficient */
+#define AV_SPARC_VIS		0x00002000 /* VIS insns available */
+#define AV_SPARC_VIS2		0x00004000 /* VIS2 insns available */
+#define AV_SPARC_ASI_BLK_INIT	0x00008000 /* block init ASIs available */
+#define AV_SPARC_FMAF		0x00010000 /* fused multiply-add */
+#define AV_SPARC_VIS3		0x00020000 /* VIS3 insns available */
+#define AV_SPARC_HPC		0x00040000 /* HPC insns available */
+#define AV_SPARC_RANDOM		0x00080000 /* 'random' insn available */
+#define AV_SPARC_TRANS		0x00100000 /* transaction insns available */
+#define AV_SPARC_FJFMAU		0x00200000 /* unfused multiply-add */
+#define AV_SPARC_IMA		0x00400000 /* integer multiply-add */
+#define AV_SPARC_ASI_CACHE_SPARING \
+				0x00800000 /* cache sparing ASIs available */
+#define AV_SPARC_PAUSE		0x01000000 /* PAUSE available */
+#define AV_SPARC_CBCOND		0x02000000 /* CBCOND insns available */
+
+/* Solaris decided to enumerate every single crypto instruction type
+ * in the AT_HWCAP bits.  This is wasteful, since if crypto is present,
+ * you still need to look in the CFR register to see if the opcode is
+ * really available.  So we simply advertise only "crypto" support.
+ */
+#define HWCAP_SPARC_CRYPTO	0x04000000 /* CRYPTO insns available */
 
 #define CORE_DUMP_USE_REGSET
 
@@ -152,7 +179,6 @@ typedef struct {
 					 (x)->e_machine == EM_SPARC32PLUS)
 #define compat_start_thread		start_thread32
 
-#define USE_ELF_CORE_DUMP
 #define ELF_EXEC_PAGESIZE	PAGE_SIZE
 
 /* This is the location that an ET_DYN program is loaded if exec'ed.  Typical
@@ -163,31 +189,8 @@ typedef struct {
 #define ELF_ET_DYN_BASE		0x0000010000000000UL
 #define COMPAT_ELF_ET_DYN_BASE	0x0000000070000000UL
 
-
-/* This yields a mask that user programs can use to figure out what
-   instruction set this cpu supports.  */
-
-/* On Ultra, we support all of the v8 capabilities. */
-static inline unsigned int sparc64_elf_hwcap(void)
-{
-	unsigned int cap = (HWCAP_SPARC_FLUSH | HWCAP_SPARC_STBAR |
-			    HWCAP_SPARC_SWAP | HWCAP_SPARC_MULDIV |
-			    HWCAP_SPARC_V9);
-
-	if (tlb_type == cheetah || tlb_type == cheetah_plus)
-		cap |= HWCAP_SPARC_ULTRA3;
-	else if (tlb_type == hypervisor) {
-		if (sun4v_chip_type == SUN4V_CHIP_NIAGARA1 ||
-		    sun4v_chip_type == SUN4V_CHIP_NIAGARA2)
-			cap |= HWCAP_SPARC_BLKINIT;
-		if (sun4v_chip_type == SUN4V_CHIP_NIAGARA2)
-			cap |= HWCAP_SPARC_N2;
-	}
-
-	return cap;
-}
-
-#define ELF_HWCAP	sparc64_elf_hwcap();
+extern unsigned long sparc64_elf_hwcap;
+#define ELF_HWCAP	sparc64_elf_hwcap
 
 /* This yields a string that ld.so will use to load implementation
    specific libraries for optimization.  This is more specific in
@@ -196,20 +199,14 @@ static inline unsigned int sparc64_elf_hwcap(void)
 #define ELF_PLATFORM	(NULL)
 
 #define SET_PERSONALITY(ex)				\
-do {	unsigned long new_flags = current_thread_info()->flags; \
-	new_flags &= _TIF_32BIT;			\
-	if ((ex).e_ident[EI_CLASS] == ELFCLASS32)	\
-		new_flags |= _TIF_32BIT;		\
-	else						\
-		new_flags &= ~_TIF_32BIT;		\
-	if ((current_thread_info()->flags & _TIF_32BIT) \
-	    != new_flags)				\
-		set_thread_flag(TIF_ABI_PENDING);	\
+do {	if ((ex).e_ident[EI_CLASS] == ELFCLASS32)	\
+		set_thread_flag(TIF_32BIT);		\
 	else						\
-		clear_thread_flag(TIF_ABI_PENDING);	\
+		clear_thread_flag(TIF_32BIT);		\
 	/* flush_thread will update pgd cache */	\
-	if (current->personality != PER_LINUX32)	\
-		set_personality(PER_LINUX);		\
+	if (personality(current->personality) != PER_LINUX32)	\
+		set_personality(PER_LINUX |		\
+			(current->personality & (~PER_MASK)));	\
 } while (0)
 
 #endif /* !(__ASM_SPARC64_ELF_H) */
diff --git a/arch/sparc/include/asm/emergency-restart.h b/arch/sparc/include/asm/emergency-restart.h
deleted file mode 100644
index 108d8c48e42..00000000000
--- a/arch/sparc/include/asm/emergency-restart.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef _ASM_EMERGENCY_RESTART_H
-#define _ASM_EMERGENCY_RESTART_H
-
-#include <asm-generic/emergency-restart.h>
-
-#endif /* _ASM_EMERGENCY_RESTART_H */
diff --git a/arch/sparc/include/asm/fb.h b/arch/sparc/include/asm/fb.h
index b83e4472965..2173432ad7f 100644
--- a/arch/sparc/include/asm/fb.h
+++ b/arch/sparc/include/asm/fb.h
@@ -1,5 +1,6 @@
 #ifndef _SPARC_FB_H_
 #define _SPARC_FB_H_
+#include <linux/console.h>
 #include <linux/fb.h>
 #include <linux/fs.h>
 #include <asm/page.h>
@@ -18,7 +19,10 @@ static inline int fb_is_primary_device(struct fb_info *info)
 	struct device *dev = info->device;
 	struct device_node *node;
 
-	node = dev->archdata.prom_node;
+	if (console_set_on_cmdline)
+		return 0;
+
+	node = dev->of_node;
 	if (node &&
 	    node == of_console_device)
 		return 1;
diff --git a/arch/sparc/include/asm/fbio.h b/arch/sparc/include/asm/fbio.h
index b9215a0907d..1d9afe277e9 100644
--- a/arch/sparc/include/asm/fbio.h
+++ b/arch/sparc/include/asm/fbio.h
@@ -1,225 +1,10 @@
 #ifndef __LINUX_FBIO_H
 #define __LINUX_FBIO_H
 
-#include <linux/compiler.h>
-#include <linux/types.h>
+#include <uapi/asm/fbio.h>
 
-/* Constants used for fbio SunOS compatibility */
-/* (C) 1996 Miguel de Icaza */
-
-/* Frame buffer types */
-#define FBTYPE_NOTYPE           -1
-#define FBTYPE_SUN1BW           0   /* mono */
-#define FBTYPE_SUN1COLOR        1 
-#define FBTYPE_SUN2BW           2 
-#define FBTYPE_SUN2COLOR        3 
-#define FBTYPE_SUN2GP           4 
-#define FBTYPE_SUN5COLOR        5 
-#define FBTYPE_SUN3COLOR        6 
-#define FBTYPE_MEMCOLOR         7 
-#define FBTYPE_SUN4COLOR        8 
- 
-#define FBTYPE_NOTSUN1          9 
-#define FBTYPE_NOTSUN2          10
-#define FBTYPE_NOTSUN3          11
- 
-#define FBTYPE_SUNFAST_COLOR    12  /* cg6 */
-#define FBTYPE_SUNROP_COLOR     13
-#define FBTYPE_SUNFB_VIDEO      14
-#define FBTYPE_SUNGIFB          15
-#define FBTYPE_SUNGPLAS         16
-#define FBTYPE_SUNGP3           17
-#define FBTYPE_SUNGT            18
-#define FBTYPE_SUNLEO           19      /* zx Leo card */
-#define FBTYPE_MDICOLOR         20      /* cg14 */
-#define FBTYPE_TCXCOLOR		21	/* SUNW,tcx card */
-
-#define FBTYPE_LASTPLUSONE      21	/* This is not last + 1 in fact... */
-
-/* Does not seem to be listed in the Sun file either */
-#define FBTYPE_CREATOR          22
-#define FBTYPE_PCI_IGA1682	23
-#define FBTYPE_P9100COLOR	24
-
-#define FBTYPE_PCI_GENERIC	1000
-#define FBTYPE_PCI_MACH64	1001
-
-/* fbio ioctls */
-/* Returned by FBIOGTYPE */
-struct  fbtype {
-        int     fb_type;        /* fb type, see above */
-        int     fb_height;      /* pixels */
-        int     fb_width;       /* pixels */
-        int     fb_depth;
-        int     fb_cmsize;      /* color map entries */
-        int     fb_size;        /* fb size in bytes */
-};
-#define FBIOGTYPE _IOR('F', 0, struct fbtype)
-
-struct  fbcmap {
-        int             index;          /* first element (0 origin) */
-        int             count;
-        unsigned char   __user *red;
-        unsigned char   __user *green;
-        unsigned char   __user *blue;
-};
-
-#ifdef __KERNEL__
 #define FBIOPUTCMAP_SPARC _IOW('F', 3, struct fbcmap)
 #define FBIOGETCMAP_SPARC _IOW('F', 4, struct fbcmap)
-#else
-#define FBIOPUTCMAP _IOW('F', 3, struct fbcmap)
-#define FBIOGETCMAP _IOW('F', 4, struct fbcmap)
-#endif
-
-/* # of device specific values */
-#define FB_ATTR_NDEVSPECIFIC    8
-/* # of possible emulations */
-#define FB_ATTR_NEMUTYPES       4
- 
-struct fbsattr {
-        int     flags;
-        int     emu_type;	/* -1 if none */
-        int     dev_specific[FB_ATTR_NDEVSPECIFIC];
-};
- 
-struct fbgattr {
-        int     real_type;	/* real frame buffer type */
-        int     owner;		/* unknown */
-        struct fbtype fbtype;	/* real frame buffer fbtype */
-        struct fbsattr sattr;   
-        int     emu_types[FB_ATTR_NEMUTYPES]; /* supported emulations */
-};
-#define FBIOSATTR  _IOW('F', 5, struct fbgattr) /* Unsupported: */
-#define FBIOGATTR  _IOR('F', 6, struct fbgattr)	/* supported */
-
-#define FBIOSVIDEO _IOW('F', 7, int)
-#define FBIOGVIDEO _IOR('F', 8, int)
-
-struct fbcursor {
-        short set;              /* what to set, choose from the list above */
-        short enable;           /* cursor on/off */
-        struct fbcurpos pos;    /* cursor position */
-        struct fbcurpos hot;    /* cursor hot spot */
-        struct fbcmap cmap;     /* color map info */
-        struct fbcurpos size;   /* cursor bit map size */
-        char __user *image;     /* cursor image bits */
-        char __user *mask;      /* cursor mask bits */
-};
-
-/* set/get cursor attributes/shape */
-#define FBIOSCURSOR     _IOW('F', 24, struct fbcursor)
-#define FBIOGCURSOR     _IOWR('F', 25, struct fbcursor)
- 
-/* set/get cursor position */
-#define FBIOSCURPOS     _IOW('F', 26, struct fbcurpos)
-#define FBIOGCURPOS     _IOW('F', 27, struct fbcurpos)
- 
-/* get max cursor size */
-#define FBIOGCURMAX     _IOR('F', 28, struct fbcurpos)
-
-/* wid manipulation */
-struct fb_wid_alloc {
-#define FB_WID_SHARED_8		0
-#define FB_WID_SHARED_24	1
-#define FB_WID_DBL_8		2
-#define FB_WID_DBL_24		3
-	__u32	wa_type;
-	__s32	wa_index;	/* Set on return */
-	__u32	wa_count;	
-};
-struct fb_wid_item {
-	__u32	wi_type;
-	__s32	wi_index;
-	__u32	wi_attrs;
-	__u32	wi_values[32];
-};
-struct fb_wid_list {
-	__u32	wl_flags;
-	__u32	wl_count;
-	struct fb_wid_item	*wl_list;
-};
-
-#define FBIO_WID_ALLOC	_IOWR('F', 30, struct fb_wid_alloc)
-#define FBIO_WID_FREE	_IOW('F', 31, struct fb_wid_alloc)
-#define FBIO_WID_PUT	_IOW('F', 32, struct fb_wid_list)
-#define FBIO_WID_GET	_IOWR('F', 33, struct fb_wid_list)
-
-/* Creator ioctls */
-#define FFB_IOCTL	('F'<<8)
-#define FFB_SYS_INFO		(FFB_IOCTL|80)
-#define FFB_CLUTREAD		(FFB_IOCTL|81)
-#define FFB_CLUTPOST		(FFB_IOCTL|82)
-#define FFB_SETDIAGMODE		(FFB_IOCTL|83)
-#define FFB_GETMONITORID	(FFB_IOCTL|84)
-#define FFB_GETVIDEOMODE	(FFB_IOCTL|85)
-#define FFB_SETVIDEOMODE	(FFB_IOCTL|86)
-#define FFB_SETSERVER		(FFB_IOCTL|87)
-#define FFB_SETOVCTL		(FFB_IOCTL|88)
-#define FFB_GETOVCTL		(FFB_IOCTL|89)
-#define FFB_GETSAXNUM		(FFB_IOCTL|90)
-#define FFB_FBDEBUG		(FFB_IOCTL|91)
-
-/* Cg14 ioctls */
-#define MDI_IOCTL          ('M'<<8)
-#define MDI_RESET          (MDI_IOCTL|1)
-#define MDI_GET_CFGINFO    (MDI_IOCTL|2)
-#define MDI_SET_PIXELMODE  (MDI_IOCTL|3)
-#    define MDI_32_PIX     32
-#    define MDI_16_PIX     16
-#    define MDI_8_PIX      8
-
-struct mdi_cfginfo {
-	int     mdi_ncluts;     /* Number of implemented CLUTs in this MDI */
-        int     mdi_type;       /* FBTYPE name */
-        int     mdi_height;     /* height */
-        int     mdi_width;      /* widht */
-        int     mdi_size;       /* available ram */
-        int     mdi_mode;       /* 8bpp, 16bpp or 32bpp */
-        int     mdi_pixfreq;    /* pixel clock (from PROM) */
-};
-
-/* SparcLinux specific ioctl for the MDI, should be replaced for
- * the SET_XLUT/SET_CLUTn ioctls instead
- */
-#define MDI_CLEAR_XLUT       (MDI_IOCTL|9)
-
-/* leo & ffb ioctls */
-struct fb_clut_alloc {
-	__u32	clutid;	/* Set on return */
- 	__u32	flag;
- 	__u32	index;
-};
-
-struct fb_clut {
-#define FB_CLUT_WAIT	0x00000001	/* Not yet implemented */
- 	__u32	flag;
- 	__u32	clutid;
- 	__u32	offset;
- 	__u32	count;
- 	char *	red;
- 	char *	green;
- 	char *	blue;
-};
-
-struct fb_clut32 {
- 	__u32	flag;
- 	__u32	clutid;
- 	__u32	offset;
- 	__u32	count;
- 	__u32	red;
- 	__u32	green;
- 	__u32	blue;
-};
-
-#define LEO_CLUTALLOC	_IOWR('L', 53, struct fb_clut_alloc)
-#define LEO_CLUTFREE	_IOW('L', 54, struct fb_clut_alloc)
-#define LEO_CLUTREAD	_IOW('L', 55, struct fb_clut)
-#define LEO_CLUTPOST	_IOW('L', 56, struct fb_clut)
-#define LEO_SETGAMMA	_IOW('L', 68, int) /* Not yet implemented */
-#define LEO_GETGAMMA	_IOR('L', 69, int) /* Not yet implemented */
-
-#ifdef __KERNEL__
 /* Addresses on the fd of a cgsix that are mappable */
 #define CG6_FBC    0x70000000
 #define CG6_TEC    0x70001000
@@ -260,47 +45,6 @@ struct fb_clut32 {
 #define CG14_CLUT3       0x6000  /* Color Look Up Table */
 #define CG14_AUTO	 0xf000
 
-#endif /* KERNEL */
-
-/* These are exported to userland for applications to use */
-/* Mappable offsets for the cg14: control registers */
-#define MDI_DIRECT_MAP 0x10000000
-#define MDI_CTLREG_MAP 0x20000000
-#define MDI_CURSOR_MAP 0x30000000
-#define MDI_SHDW_VRT_MAP 0x40000000
-
-/* Mappable offsets for the cg14: frame buffer resolutions */
-/* 32 bits */
-#define MDI_CHUNKY_XBGR_MAP 0x50000000
-#define MDI_CHUNKY_BGR_MAP 0x60000000
-
-/* 16 bits */
-#define MDI_PLANAR_X16_MAP 0x70000000
-#define MDI_PLANAR_C16_MAP 0x80000000
-
-/* 8 bit is done as CG3 MMAP offset */
-/* 32 bits, planar */
-#define MDI_PLANAR_X32_MAP 0x90000000
-#define MDI_PLANAR_B32_MAP 0xa0000000
-#define MDI_PLANAR_G32_MAP 0xb0000000
-#define MDI_PLANAR_R32_MAP 0xc0000000
-
-/* Mappable offsets on leo */
-#define LEO_SS0_MAP            0x00000000
-#define LEO_LC_SS0_USR_MAP     0x00800000
-#define LEO_LD_SS0_MAP         0x00801000
-#define LEO_LX_CURSOR_MAP      0x00802000
-#define LEO_SS1_MAP            0x00803000
-#define LEO_LC_SS1_USR_MAP     0x01003000
-#define LEO_LD_SS1_MAP         0x01004000
-#define LEO_UNK_MAP            0x01005000
-#define LEO_LX_KRN_MAP         0x01006000
-#define LEO_LC_SS0_KRN_MAP     0x01007000
-#define LEO_LC_SS1_KRN_MAP     0x01008000
-#define LEO_LD_GBL_MAP         0x01009000
-#define LEO_UNK2_MAP           0x0100a000
-
-#ifdef __KERNEL__
 struct  fbcmap32 {
 	int             index;          /* first element (0 origin) */
 	int             count;
@@ -325,6 +69,4 @@ struct fbcursor32 {
 
 #define FBIOSCURSOR32	_IOW('F', 24, struct fbcursor32)
 #define FBIOGCURSOR32	_IOW('F', 25, struct fbcursor32)
-#endif
-
 #endif /* __LINUX_FBIO_H */
diff --git a/arch/sparc/include/asm/fixmap.h b/arch/sparc/include/asm/fixmap.h
deleted file mode 100644
index f18fc0755ad..00000000000
--- a/arch/sparc/include/asm/fixmap.h
+++ /dev/null
@@ -1,110 +0,0 @@
-/*
- * fixmap.h: compile-time virtual memory allocation
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 1998 Ingo Molnar
- *
- * Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999
- */
-
-#ifndef _ASM_FIXMAP_H
-#define _ASM_FIXMAP_H
-
-#include <linux/kernel.h>
-#include <asm/page.h>
-#ifdef CONFIG_HIGHMEM
-#include <linux/threads.h>
-#include <asm/kmap_types.h>
-#endif
-
-/*
- * Here we define all the compile-time 'special' virtual
- * addresses. The point is to have a constant address at
- * compile time, but to set the physical address only
- * in the boot process. We allocate these special  addresses
- * from the top of unused virtual memory (0xfd000000 - 1 page) backwards.
- * Also this lets us do fail-safe vmalloc(), we
- * can guarantee that these special addresses and
- * vmalloc()-ed addresses never overlap.
- *
- * these 'compile-time allocated' memory buffers are
- * fixed-size 4k pages. (or larger if used with an increment
- * highger than 1) use fixmap_set(idx,phys) to associate
- * physical memory with fixmap indices.
- *
- * TLB entries of such buffers will not be flushed across
- * task switches.
- */
-
-/*
- * on UP currently we will have no trace of the fixmap mechanism,
- * no page table allocations, etc. This might change in the
- * future, say framebuffers for the console driver(s) could be
- * fix-mapped?
- */
-enum fixed_addresses {
-	FIX_HOLE,
-#ifdef CONFIG_HIGHMEM
-	FIX_KMAP_BEGIN,
-	FIX_KMAP_END = FIX_KMAP_BEGIN+(KM_TYPE_NR*NR_CPUS)-1,
-#endif
-	__end_of_fixed_addresses
-};
-
-extern void __set_fixmap (enum fixed_addresses idx,
-					unsigned long phys, pgprot_t flags);
-
-#define set_fixmap(idx, phys) \
-		__set_fixmap(idx, phys, PAGE_KERNEL)
-/*
- * Some hardware wants to get fixmapped without caching.
- */
-#define set_fixmap_nocache(idx, phys) \
-		__set_fixmap(idx, phys, PAGE_KERNEL_NOCACHE)
-/*
- * used by vmalloc.c.
- *
- * Leave one empty page between IO pages at 0xfd000000 and
- * the start of the fixmap.
- */
-#define FIXADDR_TOP	(0xfcfff000UL)
-#define FIXADDR_SIZE	((__end_of_fixed_addresses) << PAGE_SHIFT)
-#define FIXADDR_START	(FIXADDR_TOP - FIXADDR_SIZE)
-
-#define __fix_to_virt(x)	(FIXADDR_TOP - ((x) << PAGE_SHIFT))
-#define __virt_to_fix(x)	((FIXADDR_TOP - ((x)&PAGE_MASK)) >> PAGE_SHIFT)
-
-extern void __this_fixmap_does_not_exist(void);
-
-/*
- * 'index to address' translation. If anyone tries to use the idx
- * directly without tranlation, we catch the bug with a NULL-deference
- * kernel oops. Illegal ranges of incoming indices are caught too.
- */
-static inline unsigned long fix_to_virt(const unsigned int idx)
-{
-	/*
-	 * this branch gets completely eliminated after inlining,
-	 * except when someone tries to use fixaddr indices in an
-	 * illegal way. (such as mixing up address types or using
-	 * out-of-range indices).
-	 *
-	 * If it doesn't get removed, the linker will complain
-	 * loudly with a reasonably clear error message..
-	 */
-	if (idx >= __end_of_fixed_addresses)
-		__this_fixmap_does_not_exist();
-
-        return __fix_to_virt(idx);
-}
-
-static inline unsigned long virt_to_fix(const unsigned long vaddr)
-{
-	BUG_ON(vaddr >= FIXADDR_TOP || vaddr < FIXADDR_START);
-	return __virt_to_fix(vaddr);
-}
-
-#endif
diff --git a/arch/sparc/include/asm/floppy_32.h b/arch/sparc/include/asm/floppy_32.h
index c792830636d..071b83e52f1 100644
--- a/arch/sparc/include/asm/floppy_32.h
+++ b/arch/sparc/include/asm/floppy_32.h
@@ -9,13 +9,12 @@
 #include <linux/of.h>
 #include <linux/of_device.h>
 
-#include <asm/page.h>
 #include <asm/pgtable.h>
-#include <asm/system.h>
 #include <asm/idprom.h>
-#include <asm/machines.h>
 #include <asm/oplib.h>
 #include <asm/auxio.h>
+#include <asm/setup.h>
+#include <asm/page.h>
 #include <asm/irq.h>
 
 /* We don't need no stinkin' I/O port allocation crap. */
@@ -51,7 +50,6 @@ struct sun_flpy_controller {
 
 /* You'll only ever find one controller on a SparcStation anyways. */
 static struct sun_flpy_controller *sun_fdc = NULL;
-extern volatile unsigned char *fdc_status;
 
 struct sun_floppy_ops {
 	unsigned char (*fd_inb)(int port);
@@ -104,25 +102,13 @@ static struct sun_floppy_ops sun_fdops;
 /* Routines unique to each controller type on a Sun. */
 static void sun_set_dor(unsigned char value, int fdc_82077)
 {
-	if (sparc_cpu_model == sun4c) {
-		unsigned int bits = 0;
-		if (value & 0x10)
-			bits |= AUXIO_FLPY_DSEL;
-		if ((value & 0x80) == 0)
-			bits |= AUXIO_FLPY_EJCT;
-		set_auxio(bits, (~bits) & (AUXIO_FLPY_DSEL|AUXIO_FLPY_EJCT));
-	}
-	if (fdc_82077) {
+	if (fdc_82077)
 		sun_fdc->dor_82077 = value;
-	}
 }
 
 static unsigned char sun_read_dir(void)
 {
-	if (sparc_cpu_model == sun4c)
-		return (get_auxio() & AUXIO_FLPY_DCHG) ? 0x80 : 0;
-	else
-		return sun_fdc->dir_82077;
+	return sun_fdc->dir_82077;
 }
 
 static unsigned char sun_82072_fd_inb(int port)
@@ -138,7 +124,7 @@ static unsigned char sun_82072_fd_inb(int port)
 		return sun_fdc->data_82072;
 	case 7: /* FD_DIR */
 		return sun_read_dir();
-	};
+	}
 	panic("sun_82072_fd_inb: How did I get here?");
 }
 
@@ -161,7 +147,7 @@ static void sun_82072_fd_outb(unsigned char value, int port)
 	case 4: /* FD_STATUS */
 		sun_fdc->status_82072 = value;
 		break;
-	};
+	}
 	return;
 }
 
@@ -186,7 +172,7 @@ static unsigned char sun_82077_fd_inb(int port)
 		return sun_fdc->data_82077;
 	case 7: /* FD_DIR */
 		return sun_read_dir();
-	};
+	}
 	panic("sun_82077_fd_inb: How did I get here?");
 }
 
@@ -212,7 +198,7 @@ static void sun_82077_fd_outb(unsigned char value, int port)
 	case 3: /* FD_TDR */
 		sun_fdc->tapectl_82077 = value;
 		break;
-	};
+	}
 	return;
 }
 
@@ -226,13 +212,6 @@ static void sun_82077_fd_outb(unsigned char value, int port)
  * underruns.  If non-zero, doing_pdma encodes the direction of
  * the transfer for debugging.  1=read 2=write
  */
-extern char *pdma_vaddr;
-extern unsigned long pdma_size;
-extern volatile int doing_pdma;
-
-/* This is software state */
-extern char *pdma_base;
-extern unsigned long pdma_areasize;
 
 /* Common routines to all controller types on the Sparc. */
 static inline void virtual_dma_init(void)
@@ -243,10 +222,7 @@ static inline void virtual_dma_init(void)
 static inline void sun_fd_disable_dma(void)
 {
 	doing_pdma = 0;
-	if (pdma_base) {
-		mmu_unlockarea(pdma_base, pdma_areasize);
-		pdma_base = NULL;
-	}
+	pdma_base = NULL;
 }
 
 static inline void sun_fd_set_dma_mode(int mode)
@@ -276,65 +252,60 @@ static inline void sun_fd_set_dma_count(int length)
 
 static inline void sun_fd_enable_dma(void)
 {
-	pdma_vaddr = mmu_lockarea(pdma_vaddr, pdma_size);
 	pdma_base = pdma_vaddr;
 	pdma_areasize = pdma_size;
 }
 
-/* Our low-level entry point in arch/sparc/kernel/entry.S */
-extern int sparc_floppy_request_irq(int irq, unsigned long flags,
-				    irq_handler_t irq_handler);
+int sparc_floppy_request_irq(unsigned int irq, irq_handler_t irq_handler);
 
 static int sun_fd_request_irq(void)
 {
 	static int once = 0;
-	int error;
 
-	if(!once) {
+	if (!once) {
 		once = 1;
-		error = sparc_floppy_request_irq(FLOPPY_IRQ,
-						 IRQF_DISABLED,
-						 floppy_interrupt);
-		return ((error == 0) ? 0 : -1);
-	} else return 0;
+		return sparc_floppy_request_irq(FLOPPY_IRQ, floppy_interrupt);
+	} else {
+		return 0;
+	}
 }
 
 static struct linux_prom_registers fd_regs[2];
 
 static int sun_floppy_init(void)
 {
-	char state[128];
-	int tnode, fd_node, num_regs;
+	struct platform_device *op;
+	struct device_node *dp;
 	struct resource r;
+	char state[128];
+	phandle fd_node;
+	phandle tnode;
+	int num_regs;
 
 	use_virtual_dma = 1;
 
-	FLOPPY_IRQ = 11;
 	/* Forget it if we aren't on a machine that could possibly
 	 * ever have a floppy drive.
 	 */
-	if((sparc_cpu_model != sun4c && sparc_cpu_model != sun4m) ||
-	   ((idprom->id_machtype == (SM_SUN4C | SM_4C_SLC)) ||
-	    (idprom->id_machtype == (SM_SUN4C | SM_4C_ELC)))) {
+	if (sparc_cpu_model != sun4m) {
 		/* We certainly don't have a floppy controller. */
 		goto no_sun_fdc;
 	}
 	/* Well, try to find one. */
 	tnode = prom_getchild(prom_root_node);
 	fd_node = prom_searchsiblings(tnode, "obio");
-	if(fd_node != 0) {
+	if (fd_node != 0) {
 		tnode = prom_getchild(fd_node);
 		fd_node = prom_searchsiblings(tnode, "SUNW,fdtwo");
 	} else {
 		fd_node = prom_searchsiblings(tnode, "fd");
 	}
-	if(fd_node == 0) {
+	if (fd_node == 0) {
 		goto no_sun_fdc;
 	}
 
 	/* The sun4m lets us know if the controller is actually usable. */
-	if(sparc_cpu_model == sun4m &&
-	   prom_getproperty(fd_node, "status", state, sizeof(state)) != -1) {
+	if (prom_getproperty(fd_node, "status", state, sizeof(state)) != -1) {
 		if(!strcmp(state, "disabled")) {
 			goto no_sun_fdc;
 		}
@@ -345,11 +316,31 @@ static int sun_floppy_init(void)
 	memset(&r, 0, sizeof(r));
 	r.flags = fd_regs[0].which_io;
 	r.start = fd_regs[0].phys_addr;
-	sun_fdc = (struct sun_flpy_controller *)
-	    of_ioremap(&r, 0, fd_regs[0].reg_size, "floppy");
+	sun_fdc = of_ioremap(&r, 0, fd_regs[0].reg_size, "floppy");
+
+	/* Look up irq in platform_device.
+	 * We try "SUNW,fdtwo" and "fd"
+	 */
+	op = NULL;
+	for_each_node_by_name(dp, "SUNW,fdtwo") {
+		op = of_find_device_by_node(dp);
+		if (op)
+			break;
+	}
+	if (!op) {
+		for_each_node_by_name(dp, "fd") {
+			op = of_find_device_by_node(dp);
+			if (op)
+				break;
+		}
+	}
+	if (!op)
+		goto no_sun_fdc;
+
+	FLOPPY_IRQ = op->archdata.irqs[0];
 
 	/* Last minute sanity check... */
-	if(sun_fdc->status_82072 == 0xff) {
+	if (sun_fdc->status_82072 == 0xff) {
 		sun_fdc = NULL;
 		goto no_sun_fdc;
 	}
diff --git a/arch/sparc/include/asm/floppy_64.h b/arch/sparc/include/asm/floppy_64.h
index 36439d67ad7..625756406a7 100644
--- a/arch/sparc/include/asm/floppy_64.h
+++ b/arch/sparc/include/asm/floppy_64.h
@@ -43,7 +43,7 @@ struct sun_flpy_controller {
 /* You'll only ever find one controller on an Ultra anyways. */
 static struct sun_flpy_controller *sun_fdc = (struct sun_flpy_controller *)-1;
 unsigned long fdc_status;
-static struct of_device *floppy_op = NULL;
+static struct platform_device *floppy_op = NULL;
 
 struct sun_floppy_ops {
 	unsigned char	(*fd_inb) (unsigned long port);
@@ -111,7 +111,7 @@ static unsigned char sun_82077_fd_inb(unsigned long port)
 	case 7: /* FD_DIR */
 		/* XXX: Is DCL on 0x80 in sun4m? */
 		return sbus_readb(&sun_fdc->dir_82077);
-	};
+	}
 	panic("sun_82072_fd_inb: How did I get here?");
 }
 
@@ -135,7 +135,7 @@ static void sun_82077_fd_outb(unsigned char value, unsigned long port)
 	case 4: /* FD_STATUS */
 		sbus_writeb(value, &sun_fdc->status_82077);
 		break;
-	};
+	}
 	return;
 }
 
@@ -161,10 +161,7 @@ unsigned long pdma_areasize;
 static void sun_fd_disable_dma(void)
 {
 	doing_pdma = 0;
-	if (pdma_base) {
-		mmu_unlockarea(pdma_base, pdma_areasize);
-		pdma_base = NULL;
-	}
+	pdma_base = NULL;
 }
 
 static void sun_fd_set_dma_mode(int mode)
@@ -194,7 +191,6 @@ static void sun_fd_set_dma_count(int length)
 
 static void sun_fd_enable_dma(void)
 {
-	pdma_vaddr = mmu_lockarea(pdma_vaddr, pdma_size);
 	pdma_base = pdma_vaddr;
 	pdma_areasize = pdma_size;
 }
@@ -258,7 +254,7 @@ static int sun_fd_request_irq(void)
 		once = 1;
 
 		error = request_irq(FLOPPY_IRQ, sparc_floppy_irq,
-				    IRQF_DISABLED, "floppy", NULL);
+				    0, "floppy", NULL);
 
 		return ((error == 0) ? 0 : -1);
 	}
@@ -300,7 +296,7 @@ struct sun_pci_dma_op {
 static struct sun_pci_dma_op sun_pci_dma_current = { -1U, 0, 0, NULL};
 static struct sun_pci_dma_op sun_pci_dma_pending = { -1U, 0, 0, NULL};
 
-extern irqreturn_t floppy_interrupt(int irq, void *dev_id);
+irqreturn_t floppy_interrupt(int irq, void *dev_id);
 
 static unsigned char sun_pci_fd_inb(unsigned long port)
 {
@@ -548,7 +544,7 @@ static unsigned long __init sun_floppy_init(void)
 {
 	static int initialized = 0;
 	struct device_node *dp;
-	struct of_device *op;
+	struct platform_device *op;
 	const char *prop;
 	char state[128];
 
@@ -567,7 +563,7 @@ static unsigned long __init sun_floppy_init(void)
 	}
 	if (op) {
 		floppy_op = op;
-		FLOPPY_IRQ = op->irqs[0];
+		FLOPPY_IRQ = op->archdata.irqs[0];
 	} else {
 		struct device_node *ebus_dp;
 		void __iomem *auxio_reg;
@@ -589,11 +585,11 @@ static unsigned long __init sun_floppy_init(void)
 		if (!op)
 			return 0;
 
-		state_prop = of_get_property(op->node, "status", NULL);
+		state_prop = of_get_property(op->dev.of_node, "status", NULL);
 		if (state_prop && !strncmp(state_prop, "disabled", 8))
 			return 0;
 
-		FLOPPY_IRQ = op->irqs[0];
+		FLOPPY_IRQ = op->archdata.irqs[0];
 
 		/* Make sure the high density bit is set, some systems
 		 * (most notably Ultra5/Ultra10) come up with it clear.
@@ -661,7 +657,7 @@ static unsigned long __init sun_floppy_init(void)
 		config = 0;
 		for (dp = ebus_dp->child; dp; dp = dp->sibling) {
 			if (!strcmp(dp->name, "ecpp")) {
-				struct of_device *ecpp_op;
+				struct platform_device *ecpp_op;
 
 				ecpp_op = of_find_device_by_node(dp);
 				if (ecpp_op)
@@ -716,7 +712,7 @@ static unsigned long __init sun_floppy_init(void)
 
 		return sun_floppy_types[0];
 	}
-	prop = of_get_property(op->node, "status", NULL);
+	prop = of_get_property(op->dev.of_node, "status", NULL);
 	if (prop && !strncmp(state, "disabled", 8))
 		return 0;
 
diff --git a/arch/sparc/include/asm/ftrace.h b/arch/sparc/include/asm/ftrace.h
index d27716cd38c..9ec94ad116f 100644
--- a/arch/sparc/include/asm/ftrace.h
+++ b/arch/sparc/include/asm/ftrace.h
@@ -6,9 +6,24 @@
 #define MCOUNT_INSN_SIZE	4 /* sizeof mcount call */
 
 #ifndef __ASSEMBLY__
-extern void _mcount(void);
+void _mcount(void);
 #endif
 
 #endif
 
+#ifdef CONFIG_DYNAMIC_FTRACE
+/* reloction of mcount call site is the same as the address */
+static inline unsigned long ftrace_call_adjust(unsigned long addr)
+{
+	return addr;
+}
+
+struct dyn_arch_ftrace {
+};
+#endif /*  CONFIG_DYNAMIC_FTRACE */
+
+unsigned long prepare_ftrace_return(unsigned long parent,
+				    unsigned long self_addr,
+				    unsigned long frame_pointer);
+
 #endif /* _ASM_SPARC64_FTRACE */
diff --git a/arch/sparc/include/asm/futex_64.h b/arch/sparc/include/asm/futex_64.h
index 47f95839dc6..4e899b0dabf 100644
--- a/arch/sparc/include/asm/futex_64.h
+++ b/arch/sparc/include/asm/futex_64.h
@@ -4,7 +4,6 @@
 #include <linux/futex.h>
 #include <linux/uaccess.h>
 #include <asm/errno.h>
-#include <asm/system.h>
 
 #define __futex_cas_op(insn, ret, oldval, uaddr, oparg)	\
 	__asm__ __volatile__(				\
@@ -30,7 +29,7 @@
 	: "r" (uaddr), "r" (oparg), "i" (-EFAULT)	\
 	: "memory")
 
-static inline int futex_atomic_op_inuser(int encoded_op, int __user *uaddr)
+static inline int futex_atomic_op_inuser(int encoded_op, u32 __user *uaddr)
 {
 	int op = (encoded_op >> 28) & 7;
 	int cmp = (encoded_op >> 24) & 15;
@@ -38,7 +37,7 @@ static inline int futex_atomic_op_inuser(int encoded_op, int __user *uaddr)
 	int cmparg = (encoded_op << 20) >> 20;
 	int oldval = 0, ret, tem;
 
-	if (unlikely(!access_ok(VERIFY_WRITE, uaddr, sizeof(int))))
+	if (unlikely(!access_ok(VERIFY_WRITE, uaddr, sizeof(u32))))
 		return -EFAULT;
 	if (unlikely((((unsigned long) uaddr) & 0x3UL)))
 		return -EINVAL;
@@ -85,26 +84,30 @@ static inline int futex_atomic_op_inuser(int encoded_op, int __user *uaddr)
 }
 
 static inline int
-futex_atomic_cmpxchg_inatomic(int __user *uaddr, int oldval, int newval)
+futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr,
+			      u32 oldval, u32 newval)
 {
+	int ret = 0;
+
 	__asm__ __volatile__(
-	"\n1:	casa	[%3] %%asi, %2, %0\n"
+	"\n1:	casa	[%4] %%asi, %3, %1\n"
 	"2:\n"
 	"	.section .fixup,#alloc,#execinstr\n"
 	"	.align	4\n"
 	"3:	sethi	%%hi(2b), %0\n"
 	"	jmpl	%0 + %%lo(2b), %%g0\n"
-	"	 mov	%4, %0\n"
+	"	mov	%5, %0\n"
 	"	.previous\n"
 	"	.section __ex_table,\"a\"\n"
 	"	.align	4\n"
 	"	.word	1b, 3b\n"
 	"	.previous\n"
-	: "=r" (newval)
-	: "0" (newval), "r" (oldval), "r" (uaddr), "i" (-EFAULT)
+	: "+r" (ret), "=r" (newval)
+	: "1" (newval), "r" (oldval), "r" (uaddr), "i" (-EFAULT)
 	: "memory");
 
-	return newval;
+	*uval = newval;
+	return ret;
 }
 
 #endif /* !(_SPARC64_FUTEX_H) */
diff --git a/arch/sparc/include/asm/gpio.h b/arch/sparc/include/asm/gpio.h
index a0e3ac0af59..b3799d88ffc 100644
--- a/arch/sparc/include/asm/gpio.h
+++ b/arch/sparc/include/asm/gpio.h
@@ -1,36 +1,4 @@
-#ifndef __ASM_SPARC_GPIO_H
-#define __ASM_SPARC_GPIO_H
-
-#include <linux/errno.h>
-#include <asm-generic/gpio.h>
-
-#ifdef CONFIG_GPIOLIB
-
-static inline int gpio_get_value(unsigned int gpio)
-{
-	return __gpio_get_value(gpio);
-}
-
-static inline void gpio_set_value(unsigned int gpio, int value)
-{
-	__gpio_set_value(gpio, value);
-}
-
-static inline int gpio_cansleep(unsigned int gpio)
-{
-	return __gpio_cansleep(gpio);
-}
-
-static inline int gpio_to_irq(unsigned int gpio)
-{
-	return -ENOSYS;
-}
-
-static inline int irq_to_gpio(unsigned int irq)
-{
-	return -EINVAL;
-}
-
-#endif /* CONFIG_GPIOLIB */
-
-#endif /* __ASM_SPARC_GPIO_H */
+#ifndef __LINUX_GPIO_H
+#warning Include linux/gpio.h instead of asm/gpio.h
+#include <linux/gpio.h>
+#endif
diff --git a/arch/sparc/include/asm/hardirq_32.h b/arch/sparc/include/asm/hardirq_32.h
index 4f63ed8df55..ee93923b7f8 100644
--- a/arch/sparc/include/asm/hardirq_32.h
+++ b/arch/sparc/include/asm/hardirq_32.h
@@ -7,17 +7,6 @@
 #ifndef __SPARC_HARDIRQ_H
 #define __SPARC_HARDIRQ_H
 
-#include <linux/threads.h>
-#include <linux/spinlock.h>
-#include <linux/cache.h>
-
-/* entry.S is sensitive to the offsets of these fields */ /* XXX P3 Is it? */
-typedef struct {
-	unsigned int __softirq_pending;
-} ____cacheline_aligned irq_cpustat_t;
-
-#include <linux/irq_cpustat.h>	/* Standard mappings for irq_cpustat_t above */
-
-#define HARDIRQ_BITS    8
+#include <asm-generic/hardirq.h>
 
 #endif /* __SPARC_HARDIRQ_H */
diff --git a/arch/sparc/include/asm/hardirq_64.h b/arch/sparc/include/asm/hardirq_64.h
index 7c29fd1a87a..f478ff1ddd0 100644
--- a/arch/sparc/include/asm/hardirq_64.h
+++ b/arch/sparc/include/asm/hardirq_64.h
@@ -14,6 +14,4 @@
 
 void ack_bad_irq(unsigned int irq);
 
-#define HARDIRQ_BITS	8
-
 #endif /* !(__SPARC64_HARDIRQ_H) */
diff --git a/arch/sparc/include/asm/head_32.h b/arch/sparc/include/asm/head_32.h
index 7c35491a8b5..5f1dbe315bc 100644
--- a/arch/sparc/include/asm/head_32.h
+++ b/arch/sparc/include/asm/head_32.h
@@ -2,15 +2,8 @@
 #define __SPARC_HEAD_H
 
 #define KERNBASE        0xf0000000  /* First address the kernel will eventually be */
-#define LOAD_ADDR       0x4000      /* prom jumps to us here unless this is elf /boot */
-#define SUN4C_SEGSZ     (1 << 18)
-#define SRMMU_L1_KBASE_OFFSET ((KERNBASE>>24)<<2)  /* Used in boot remapping. */
-#define INTS_ENAB        0x01           /* entry.S uses this. */
-
-#define SUN4_PROM_VECTOR 0xFFE81000     /* SUN4 PROM needs to be hardwired */
 
 #define WRITE_PAUSE      nop; nop; nop; /* Have to do this after %wim/%psr chg */
-#define NOP_INSN         0x01000000     /* Used to patch sparc_save_state */
 
 /* Here are some trap goodies */
 
@@ -18,9 +11,7 @@
 #define TRAP_ENTRY(type, label) \
 	rd %psr, %l0; b label; rd %wim, %l3; nop;
 
-/* Data/text faults. Defaults to sun4c version at boot time. */
-#define SPARC_TFAULT rd %psr, %l0; rd %wim, %l3; b sun4c_fault; mov 1, %l7;
-#define SPARC_DFAULT rd %psr, %l0; rd %wim, %l3; b sun4c_fault; mov 0, %l7;
+/* Data/text faults */
 #define SRMMU_TFAULT rd %psr, %l0; rd %wim, %l3; b srmmu_fault; mov 1, %l7;
 #define SRMMU_DFAULT rd %psr, %l0; rd %wim, %l3; b srmmu_fault; mov 0, %l7;
 
@@ -64,15 +55,15 @@
 
 /* The Get Condition Codes software trap for userland. */
 #define GETCC_TRAP \
-        b getcc_trap_handler; mov %psr, %l0; nop; nop;
+        b getcc_trap_handler; rd %psr, %l0; nop; nop;
 
 /* The Set Condition Codes software trap for userland. */
 #define SETCC_TRAP \
-        b setcc_trap_handler; mov %psr, %l0; nop; nop;
+        b setcc_trap_handler; rd %psr, %l0; nop; nop;
 
 /* The Get PSR software trap for userland. */
 #define GETPSR_TRAP \
-	mov %psr, %i0; jmp %l2; rett %l2 + 4; nop;
+	rd %psr, %i0; jmp %l2; rett %l2 + 4; nop;
 
 /* This is for hard interrupts from level 1-14, 15 is non-maskable (nmi) and
  * gets handled with another macro.
@@ -80,16 +71,6 @@
 #define TRAP_ENTRY_INTERRUPT(int_level) \
         mov int_level, %l7; rd %psr, %l0; b real_irq_entry; rd %wim, %l3;
 
-/* NMI's (Non Maskable Interrupts) are special, you can't keep them
- * from coming in, and basically if you get one, the shows over. ;(
- * On the sun4c they are usually asynchronous memory errors, on the
- * the sun4m they could be either due to mem errors or a software
- * initiated interrupt from the prom/kern on an SMP box saying "I
- * command you to do CPU tricks, read your mailbox for more info."
- */
-#define NMI_TRAP \
-        rd %wim, %l3; b linux_trap_nmi_sun4c; mov %psr, %l0; nop;
-
 /* Window overflows/underflows are special and we need to try to be as
  * efficient as possible here....
  */
diff --git a/arch/sparc/include/asm/hibernate.h b/arch/sparc/include/asm/hibernate.h
new file mode 100644
index 00000000000..2ec34f84224
--- /dev/null
+++ b/arch/sparc/include/asm/hibernate.h
@@ -0,0 +1,23 @@
+/*
+ * hibernate.h:  Hibernaton support specific for sparc64.
+ *
+ * Copyright (C) 2013 Kirill V Tkhai (tkhai@yandex.ru)
+ */
+
+#ifndef ___SPARC_HIBERNATE_H
+#define ___SPARC_HIBERNATE_H
+
+struct saved_context {
+	unsigned long fp;
+	unsigned long cwp;
+	unsigned long wstate;
+
+	unsigned long tick;
+	unsigned long pstate;
+
+	unsigned long g4;
+	unsigned long g5;
+	unsigned long g6;
+};
+
+#endif
diff --git a/arch/sparc/include/asm/highmem.h b/arch/sparc/include/asm/highmem.h
index 3de42e77627..92ded294a4e 100644
--- a/arch/sparc/include/asm/highmem.h
+++ b/arch/sparc/include/asm/highmem.h
@@ -21,7 +21,6 @@
 #ifdef __KERNEL__
 
 #include <linux/interrupt.h>
-#include <asm/fixmap.h>
 #include <asm/vaddrs.h>
 #include <asm/kmap_types.h>
 #include <asm/pgtable.h>
@@ -29,11 +28,10 @@
 /* declarations for highmem.c */
 extern unsigned long highstart_pfn, highend_pfn;
 
-extern pte_t *kmap_pte;
 extern pgprot_t kmap_prot;
 extern pte_t *pkmap_page_table;
 
-extern void kmap_init(void) __init;
+void kmap_init(void) __init;
 
 /*
  * Right now we initialize only a single pte table. It can be extended
@@ -51,8 +49,8 @@ extern void kmap_init(void) __init;
 
 #define PKMAP_END (PKMAP_ADDR(LAST_PKMAP))
 
-extern void *kmap_high(struct page *page);
-extern void kunmap_high(struct page *page);
+void *kmap_high(struct page *page);
+void kunmap_high(struct page *page);
 
 static inline void *kmap(struct page *page)
 {
@@ -70,9 +68,8 @@ static inline void kunmap(struct page *page)
 	kunmap_high(page);
 }
 
-extern void *kmap_atomic(struct page *page, enum km_type type);
-extern void kunmap_atomic(void *kvaddr, enum km_type type);
-extern struct page *kmap_atomic_to_page(void *vaddr);
+void *kmap_atomic(struct page *page);
+void __kunmap_atomic(void *kvaddr);
 
 #define flush_cache_kmaps()	flush_cache_all()
 
diff --git a/arch/sparc/include/asm/hugetlb.h b/arch/sparc/include/asm/hugetlb.h
index 177061064ee..e4cab465b81 100644
--- a/arch/sparc/include/asm/hugetlb.h
+++ b/arch/sparc/include/asm/hugetlb.h
@@ -2,6 +2,7 @@
 #define _ASM_SPARC64_HUGETLB_H
 
 #include <asm/page.h>
+#include <asm-generic/hugetlb.h>
 
 
 void set_huge_pte_at(struct mm_struct *mm, unsigned long addr,
@@ -10,7 +11,9 @@ void set_huge_pte_at(struct mm_struct *mm, unsigned long addr,
 pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr,
 			      pte_t *ptep);
 
-void hugetlb_prefault_arch_hook(struct mm_struct *mm);
+static inline void hugetlb_prefault_arch_hook(struct mm_struct *mm)
+{
+}
 
 static inline int is_hugepage_only_range(struct mm_struct *mm,
 					 unsigned long addr,
@@ -58,14 +61,20 @@ static inline pte_t huge_pte_wrprotect(pte_t pte)
 static inline void huge_ptep_set_wrprotect(struct mm_struct *mm,
 					   unsigned long addr, pte_t *ptep)
 {
-	ptep_set_wrprotect(mm, addr, ptep);
+	pte_t old_pte = *ptep;
+	set_huge_pte_at(mm, addr, ptep, pte_wrprotect(old_pte));
 }
 
 static inline int huge_ptep_set_access_flags(struct vm_area_struct *vma,
 					     unsigned long addr, pte_t *ptep,
 					     pte_t pte, int dirty)
 {
-	return ptep_set_access_flags(vma, addr, ptep, pte, dirty);
+	int changed = !pte_same(*ptep, pte);
+	if (changed) {
+		set_huge_pte_at(vma->vm_mm, addr, ptep, pte);
+		flush_tlb_page(vma, addr);
+	}
+	return changed;
 }
 
 static inline pte_t huge_ptep_get(pte_t *ptep)
@@ -82,4 +91,8 @@ static inline void arch_release_hugepage(struct page *page)
 {
 }
 
+static inline void arch_clear_hugepage_flags(struct page *page)
+{
+}
+
 #endif /* _ASM_SPARC64_HUGETLB_H */
diff --git a/arch/sparc/include/asm/hvtramp.h b/arch/sparc/include/asm/hvtramp.h
index b2b9b947b3a..04b56f862bb 100644
--- a/arch/sparc/include/asm/hvtramp.h
+++ b/arch/sparc/include/asm/hvtramp.h
@@ -19,7 +19,7 @@ struct hvtramp_descr {
 	struct hvtramp_mapping	maps[1];
 };
 
-extern void hv_cpu_startup(unsigned long hvdescr_pa);
+void hv_cpu_startup(unsigned long hvdescr_pa);
 
 #endif
 
diff --git a/arch/sparc/include/asm/hypervisor.h b/arch/sparc/include/asm/hypervisor.h
index 109ae24ba24..94b39caea3e 100644
--- a/arch/sparc/include/asm/hypervisor.h
+++ b/arch/sparc/include/asm/hypervisor.h
@@ -98,7 +98,7 @@
 #define HV_FAST_MACH_EXIT		0x00
 
 #ifndef __ASSEMBLY__
-extern void sun4v_mach_exit(unsigned long exit_code);
+void sun4v_mach_exit(unsigned long exit_code);
 #endif
 
 /* Domain services.  */
@@ -127,9 +127,9 @@ extern void sun4v_mach_exit(unsigned long exit_code);
 #define HV_FAST_MACH_DESC		0x01
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_mach_desc(unsigned long buffer_pa,
-				     unsigned long buf_len,
-				     unsigned long *real_buf_len);
+unsigned long sun4v_mach_desc(unsigned long buffer_pa,
+			      unsigned long buf_len,
+			      unsigned long *real_buf_len);
 #endif
 
 /* mach_sir()
@@ -148,7 +148,7 @@ extern unsigned long sun4v_mach_desc(unsigned long buffer_pa,
 #define HV_FAST_MACH_SIR		0x02
 
 #ifndef __ASSEMBLY__
-extern void sun4v_mach_sir(void);
+void sun4v_mach_sir(void);
 #endif
 
 /* mach_set_watchdog()
@@ -204,8 +204,8 @@ extern void sun4v_mach_sir(void);
 #define HV_FAST_MACH_SET_WATCHDOG	0x05
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_mach_set_watchdog(unsigned long timeout,
-					     unsigned long *orig_timeout);
+unsigned long sun4v_mach_set_watchdog(unsigned long timeout,
+				      unsigned long *orig_timeout);
 #endif
 
 /* CPU services.
@@ -250,10 +250,10 @@ extern unsigned long sun4v_mach_set_watchdog(unsigned long timeout,
 #define HV_FAST_CPU_START		0x10
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_cpu_start(unsigned long cpuid,
-				     unsigned long pc,
-				     unsigned long rtba,
-				     unsigned long arg0);
+unsigned long sun4v_cpu_start(unsigned long cpuid,
+			      unsigned long pc,
+			      unsigned long rtba,
+			      unsigned long arg0);
 #endif
 
 /* cpu_stop()
@@ -278,7 +278,7 @@ extern unsigned long sun4v_cpu_start(unsigned long cpuid,
 #define HV_FAST_CPU_STOP		0x11
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_cpu_stop(unsigned long cpuid);
+unsigned long sun4v_cpu_stop(unsigned long cpuid);
 #endif
 
 /* cpu_yield()
@@ -295,7 +295,7 @@ extern unsigned long sun4v_cpu_stop(unsigned long cpuid);
 #define HV_FAST_CPU_YIELD		0x12
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_cpu_yield(void);
+unsigned long sun4v_cpu_yield(void);
 #endif
 
 /* cpu_qconf()
@@ -341,9 +341,9 @@ extern unsigned long sun4v_cpu_yield(void);
 #define  HV_CPU_QUEUE_NONRES_ERROR	 0x3f
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_cpu_qconf(unsigned long type,
-				     unsigned long queue_paddr,
-				     unsigned long num_queue_entries);
+unsigned long sun4v_cpu_qconf(unsigned long type,
+			      unsigned long queue_paddr,
+			      unsigned long num_queue_entries);
 #endif
 
 /* cpu_qinfo()
@@ -394,7 +394,9 @@ extern unsigned long sun4v_cpu_qconf(unsigned long type,
 #define HV_FAST_CPU_MONDO_SEND		0x42
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_cpu_mondo_send(unsigned long cpu_count, unsigned long cpu_list_pa, unsigned long mondo_block_pa);
+unsigned long sun4v_cpu_mondo_send(unsigned long cpu_count,
+				   unsigned long cpu_list_pa,
+				   unsigned long mondo_block_pa);
 #endif
 
 /* cpu_myid()
@@ -425,7 +427,7 @@ extern unsigned long sun4v_cpu_mondo_send(unsigned long cpu_count, unsigned long
 #define  HV_CPU_STATE_ERROR		 0x03
 
 #ifndef __ASSEMBLY__
-extern long sun4v_cpu_state(unsigned long cpuid);
+long sun4v_cpu_state(unsigned long cpuid);
 #endif
 
 /* cpu_set_rtba()
@@ -625,8 +627,8 @@ struct hv_fault_status {
 #define HV_FAST_MMU_TSB_CTX0		0x20
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_mmu_tsb_ctx0(unsigned long num_descriptions,
-					unsigned long tsb_desc_ra);
+unsigned long sun4v_mmu_tsb_ctx0(unsigned long num_descriptions,
+				 unsigned long tsb_desc_ra);
 #endif
 
 /* mmu_tsb_ctxnon0()
@@ -654,7 +656,7 @@ extern unsigned long sun4v_mmu_tsb_ctx0(unsigned long num_descriptions,
  * ARG3:	mmu context
  * ARG4:	flags (HV_MMU_{IMMU,DMMU})
  * RET0:	status
- * ERRORS:	EINVAL			Invalid virutal address, context, or
+ * ERRORS:	EINVAL			Invalid virtual address, context, or
  *					flags value
  *		ENOTSUPPORTED		ARG0 or ARG1 is non-zero
  *
@@ -710,7 +712,7 @@ extern unsigned long sun4v_mmu_tsb_ctx0(unsigned long num_descriptions,
 #define HV_FAST_MMU_DEMAP_ALL		0x24
 
 #ifndef __ASSEMBLY__
-extern void sun4v_mmu_demap_all(void);
+void sun4v_mmu_demap_all(void);
 #endif
 
 /* mmu_map_perm_addr()
@@ -721,7 +723,7 @@ extern void sun4v_mmu_demap_all(void);
  * ARG2:	TTE
  * ARG3:	flags (HV_MMU_{IMMU,DMMU})
  * RET0:	status
- * ERRORS:	EINVAL			Invalid virutal address or flags value
+ * ERRORS:	EINVAL			Invalid virtual address or flags value
  *		EBADPGSZ		Invalid page size value
  *		ENORADDR		Invalid real address in TTE
  *		ETOOMANY		Too many mappings (max of 8 reached)
@@ -740,10 +742,10 @@ extern void sun4v_mmu_demap_all(void);
 #define HV_FAST_MMU_MAP_PERM_ADDR	0x25
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_mmu_map_perm_addr(unsigned long vaddr,
-					     unsigned long set_to_zero,
-					     unsigned long tte,
-					     unsigned long flags);
+unsigned long sun4v_mmu_map_perm_addr(unsigned long vaddr,
+				      unsigned long set_to_zero,
+				      unsigned long tte,
+				      unsigned long flags);
 #endif
 
 /* mmu_fault_area_conf()
@@ -800,7 +802,7 @@ extern unsigned long sun4v_mmu_map_perm_addr(unsigned long vaddr,
  * ARG1:	reserved, must be zero
  * ARG2:	flags (HV_MMU_{IMMU,DMMU})
  * RET0:	status
- * ERRORS:	EINVAL			Invalid virutal address or flags value
+ * ERRORS:	EINVAL			Invalid virtual address or flags value
  *		ENOMAP			Specified mapping was not found
  *
  * Demaps any permanent page mapping (established via
@@ -945,7 +947,7 @@ extern unsigned long sun4v_mmu_map_perm_addr(unsigned long vaddr,
 #define HV_FAST_TOD_GET			0x50
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_tod_get(unsigned long *time);
+unsigned long sun4v_tod_get(unsigned long *time);
 #endif
 
 /* tod_set()
@@ -962,7 +964,7 @@ extern unsigned long sun4v_tod_get(unsigned long *time);
 #define HV_FAST_TOD_SET			0x51
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_tod_set(unsigned long time);
+unsigned long sun4v_tod_set(unsigned long time);
 #endif
 
 /* Console services */
@@ -1038,14 +1040,14 @@ extern unsigned long sun4v_tod_set(unsigned long time);
 #define HV_FAST_CONS_WRITE		0x63
 
 #ifndef __ASSEMBLY__
-extern long sun4v_con_getchar(long *status);
-extern long sun4v_con_putchar(long c);
-extern long sun4v_con_read(unsigned long buffer,
-			   unsigned long size,
-			   unsigned long *bytes_read);
-extern unsigned long sun4v_con_write(unsigned long buffer,
-				     unsigned long size,
-				     unsigned long *bytes_written);
+long sun4v_con_getchar(long *status);
+long sun4v_con_putchar(long c);
+long sun4v_con_read(unsigned long buffer,
+		    unsigned long size,
+		    unsigned long *bytes_read);
+unsigned long sun4v_con_write(unsigned long buffer,
+			      unsigned long size,
+			      unsigned long *bytes_written);
 #endif
 
 /* mach_set_soft_state()
@@ -1080,8 +1082,8 @@ extern unsigned long sun4v_con_write(unsigned long buffer,
 #define  HV_SOFT_STATE_TRANSITION	 0x02
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_mach_set_soft_state(unsigned long soft_state,
-					       unsigned long msg_string_ra);
+unsigned long sun4v_mach_set_soft_state(unsigned long soft_state,
+				        unsigned long msg_string_ra);
 #endif
 
 /* mach_get_soft_state()
@@ -1159,20 +1161,20 @@ extern unsigned long sun4v_mach_set_soft_state(unsigned long soft_state,
 #define HV_FAST_SVC_CLRSTATUS		0x84
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_svc_send(unsigned long svc_id,
-				    unsigned long buffer,
-				    unsigned long buffer_size,
-				    unsigned long *sent_bytes);
-extern unsigned long sun4v_svc_recv(unsigned long svc_id,
-				    unsigned long buffer,
-				    unsigned long buffer_size,
-				    unsigned long *recv_bytes);
-extern unsigned long sun4v_svc_getstatus(unsigned long svc_id,
-					 unsigned long *status_bits);
-extern unsigned long sun4v_svc_setstatus(unsigned long svc_id,
-					 unsigned long status_bits);
-extern unsigned long sun4v_svc_clrstatus(unsigned long svc_id,
-					 unsigned long status_bits);
+unsigned long sun4v_svc_send(unsigned long svc_id,
+			     unsigned long buffer,
+			     unsigned long buffer_size,
+			     unsigned long *sent_bytes);
+unsigned long sun4v_svc_recv(unsigned long svc_id,
+			     unsigned long buffer,
+			     unsigned long buffer_size,
+			     unsigned long *recv_bytes);
+unsigned long sun4v_svc_getstatus(unsigned long svc_id,
+				  unsigned long *status_bits);
+unsigned long sun4v_svc_setstatus(unsigned long svc_id,
+				  unsigned long status_bits);
+unsigned long sun4v_svc_clrstatus(unsigned long svc_id,
+				  unsigned long status_bits);
 #endif
 
 /* Trap trace services.
@@ -1205,7 +1207,7 @@ struct hv_trap_trace_control {
  * structure contents.  Attempts to do so will result in undefined
  * behavior for the guest.
  *
- * Each trap trace buffer entry is layed out as follows:
+ * Each trap trace buffer entry is laid out as follows:
  */
 #ifndef __ASSEMBLY__
 struct hv_trap_trace_entry {
@@ -1300,7 +1302,7 @@ struct hv_trap_trace_entry {
  * state in RET1.  Future systems may define various flags for the
  * enable argument (ARG0), for the moment a guest should pass
  * "(uint64_t) -1" to enable, and "(uint64_t) 0" to disable all
- * tracing - which will ensure future compatability.
+ * tracing - which will ensure future compatibility.
  */
 #define HV_FAST_TTRACE_ENABLE		0x92
 
@@ -1458,8 +1460,8 @@ struct hv_trap_trace_entry {
 #define HV_FAST_INTR_DEVINO2SYSINO	0xa0
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_devino_to_sysino(unsigned long devhandle,
-					    unsigned long devino);
+unsigned long sun4v_devino_to_sysino(unsigned long devhandle,
+				     unsigned long devino);
 #endif
 
 /* intr_getenabled()
@@ -1476,7 +1478,7 @@ extern unsigned long sun4v_devino_to_sysino(unsigned long devhandle,
 #define HV_FAST_INTR_GETENABLED		0xa1
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_intr_getenabled(unsigned long sysino);
+unsigned long sun4v_intr_getenabled(unsigned long sysino);
 #endif
 
 /* intr_setenabled()
@@ -1492,7 +1494,8 @@ extern unsigned long sun4v_intr_getenabled(unsigned long sysino);
 #define HV_FAST_INTR_SETENABLED		0xa2
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_intr_setenabled(unsigned long sysino, unsigned long intr_enabled);
+unsigned long sun4v_intr_setenabled(unsigned long sysino,
+				    unsigned long intr_enabled);
 #endif
 
 /* intr_getstate()
@@ -1508,7 +1511,7 @@ extern unsigned long sun4v_intr_setenabled(unsigned long sysino, unsigned long i
 #define HV_FAST_INTR_GETSTATE		0xa3
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_intr_getstate(unsigned long sysino);
+unsigned long sun4v_intr_getstate(unsigned long sysino);
 #endif
 
 /* intr_setstate()
@@ -1528,7 +1531,7 @@ extern unsigned long sun4v_intr_getstate(unsigned long sysino);
 #define HV_FAST_INTR_SETSTATE		0xa4
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_intr_setstate(unsigned long sysino, unsigned long intr_state);
+unsigned long sun4v_intr_setstate(unsigned long sysino, unsigned long intr_state);
 #endif
 
 /* intr_gettarget()
@@ -1546,7 +1549,7 @@ extern unsigned long sun4v_intr_setstate(unsigned long sysino, unsigned long int
 #define HV_FAST_INTR_GETTARGET		0xa5
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_intr_gettarget(unsigned long sysino);
+unsigned long sun4v_intr_gettarget(unsigned long sysino);
 #endif
 
 /* intr_settarget()
@@ -1563,7 +1566,7 @@ extern unsigned long sun4v_intr_gettarget(unsigned long sysino);
 #define HV_FAST_INTR_SETTARGET		0xa6
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_intr_settarget(unsigned long sysino, unsigned long cpuid);
+unsigned long sun4v_intr_settarget(unsigned long sysino, unsigned long cpuid);
 #endif
 
 /* vintr_get_cookie()
@@ -1647,30 +1650,30 @@ extern unsigned long sun4v_intr_settarget(unsigned long sysino, unsigned long cp
 #define HV_FAST_VINTR_SET_TARGET	0xae
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_vintr_get_cookie(unsigned long dev_handle,
-					    unsigned long dev_ino,
-					    unsigned long *cookie);
-extern unsigned long sun4v_vintr_set_cookie(unsigned long dev_handle,
-					    unsigned long dev_ino,
-					    unsigned long cookie);
-extern unsigned long sun4v_vintr_get_valid(unsigned long dev_handle,
-					   unsigned long dev_ino,
-					   unsigned long *valid);
-extern unsigned long sun4v_vintr_set_valid(unsigned long dev_handle,
-					   unsigned long dev_ino,
-					   unsigned long valid);
-extern unsigned long sun4v_vintr_get_state(unsigned long dev_handle,
-					   unsigned long dev_ino,
-					   unsigned long *state);
-extern unsigned long sun4v_vintr_set_state(unsigned long dev_handle,
-					   unsigned long dev_ino,
-					   unsigned long state);
-extern unsigned long sun4v_vintr_get_target(unsigned long dev_handle,
-					    unsigned long dev_ino,
-					    unsigned long *cpuid);
-extern unsigned long sun4v_vintr_set_target(unsigned long dev_handle,
-					    unsigned long dev_ino,
-					    unsigned long cpuid);
+unsigned long sun4v_vintr_get_cookie(unsigned long dev_handle,
+				     unsigned long dev_ino,
+				     unsigned long *cookie);
+unsigned long sun4v_vintr_set_cookie(unsigned long dev_handle,
+				     unsigned long dev_ino,
+				     unsigned long cookie);
+unsigned long sun4v_vintr_get_valid(unsigned long dev_handle,
+				    unsigned long dev_ino,
+				    unsigned long *valid);
+unsigned long sun4v_vintr_set_valid(unsigned long dev_handle,
+				    unsigned long dev_ino,
+				    unsigned long valid);
+unsigned long sun4v_vintr_get_state(unsigned long dev_handle,
+				    unsigned long dev_ino,
+				    unsigned long *state);
+unsigned long sun4v_vintr_set_state(unsigned long dev_handle,
+				    unsigned long dev_ino,
+				    unsigned long state);
+unsigned long sun4v_vintr_get_target(unsigned long dev_handle,
+				     unsigned long dev_ino,
+				     unsigned long *cpuid);
+unsigned long sun4v_vintr_set_target(unsigned long dev_handle,
+				     unsigned long dev_ino,
+				     unsigned long cpuid);
 #endif
 
 /* PCI IO services.
@@ -1880,7 +1883,7 @@ extern unsigned long sun4v_vintr_set_target(unsigned long dev_handle,
  * pci_device, at pci_config_offset from the beginning of the device's
  * configuration space.  If there was no error, RET1 is set to zero and
  * RET2 is set to the data read.  Insignificant bits in RET2 are not
- * guarenteed to have any specific value and therefore must be ignored.
+ * guaranteed to have any specific value and therefore must be ignored.
  *
  * The data returned in RET2 is size based byte swapped.
  *
@@ -1941,9 +1944,9 @@ extern unsigned long sun4v_vintr_set_target(unsigned long dev_handle,
  * and return the actual data read in RET2.  The data returned is size based
  * byte swapped.
  *
- * Non-significant bits in RET2 are not guarenteed to have any specific value
+ * Non-significant bits in RET2 are not guaranteed to have any specific value
  * and therefore must be ignored.  If RET1 is returned as non-zero, the data
- * value is not guarenteed to have any specific value and should be ignored.
+ * value is not guaranteed to have any specific value and should be ignored.
  *
  * The caller must have permission to read from the given devhandle, real
  * address, which must be an IO address.  The argument real address must be a
@@ -2456,9 +2459,9 @@ extern unsigned long sun4v_vintr_set_target(unsigned long dev_handle,
  *
  * As receive queue configuration causes a reset of the queue's head and
  * tail pointers there is no way for a gues to determine how many entries
- * have been received between a preceeding ldc_get_rx_state() API call
+ * have been received between a preceding ldc_get_rx_state() API call
  * and the completion of the configuration operation.  It should be noted
- * that datagram delivery is not guarenteed via domain channels anyway,
+ * that datagram delivery is not guaranteed via domain channels anyway,
  * and therefore any higher protocol should be resilient to datagram
  * loss if necessary.  However, to overcome this specific race potential
  * it is recommended, for example, that a higher level protocol be employed
@@ -2627,50 +2630,50 @@ struct ldc_mtable_entry {
 #define HV_FAST_LDC_REVOKE		0xef
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_ldc_tx_qconf(unsigned long channel,
-					unsigned long ra,
-					unsigned long num_entries);
-extern unsigned long sun4v_ldc_tx_qinfo(unsigned long channel,
-					unsigned long *ra,
-					unsigned long *num_entries);
-extern unsigned long sun4v_ldc_tx_get_state(unsigned long channel,
-					    unsigned long *head_off,
-					    unsigned long *tail_off,
-					    unsigned long *chan_state);
-extern unsigned long sun4v_ldc_tx_set_qtail(unsigned long channel,
-					    unsigned long tail_off);
-extern unsigned long sun4v_ldc_rx_qconf(unsigned long channel,
-					unsigned long ra,
-					unsigned long num_entries);
-extern unsigned long sun4v_ldc_rx_qinfo(unsigned long channel,
-					unsigned long *ra,
-					unsigned long *num_entries);
-extern unsigned long sun4v_ldc_rx_get_state(unsigned long channel,
-					    unsigned long *head_off,
-					    unsigned long *tail_off,
-					    unsigned long *chan_state);
-extern unsigned long sun4v_ldc_rx_set_qhead(unsigned long channel,
-					    unsigned long head_off);
-extern unsigned long sun4v_ldc_set_map_table(unsigned long channel,
-					     unsigned long ra,
-					     unsigned long num_entries);
-extern unsigned long sun4v_ldc_get_map_table(unsigned long channel,
-					     unsigned long *ra,
-					     unsigned long *num_entries);
-extern unsigned long sun4v_ldc_copy(unsigned long channel,
-				    unsigned long dir_code,
-				    unsigned long tgt_raddr,
-				    unsigned long lcl_raddr,
-				    unsigned long len,
-				    unsigned long *actual_len);
-extern unsigned long sun4v_ldc_mapin(unsigned long channel,
-				     unsigned long cookie,
-				     unsigned long *ra,
-				     unsigned long *perm);
-extern unsigned long sun4v_ldc_unmap(unsigned long ra);
-extern unsigned long sun4v_ldc_revoke(unsigned long channel,
-				      unsigned long cookie,
-				      unsigned long mte_cookie);
+unsigned long sun4v_ldc_tx_qconf(unsigned long channel,
+				 unsigned long ra,
+				 unsigned long num_entries);
+unsigned long sun4v_ldc_tx_qinfo(unsigned long channel,
+				 unsigned long *ra,
+				 unsigned long *num_entries);
+unsigned long sun4v_ldc_tx_get_state(unsigned long channel,
+				     unsigned long *head_off,
+				     unsigned long *tail_off,
+				     unsigned long *chan_state);
+unsigned long sun4v_ldc_tx_set_qtail(unsigned long channel,
+				     unsigned long tail_off);
+unsigned long sun4v_ldc_rx_qconf(unsigned long channel,
+				 unsigned long ra,
+				 unsigned long num_entries);
+unsigned long sun4v_ldc_rx_qinfo(unsigned long channel,
+				 unsigned long *ra,
+				 unsigned long *num_entries);
+unsigned long sun4v_ldc_rx_get_state(unsigned long channel,
+				     unsigned long *head_off,
+				     unsigned long *tail_off,
+				     unsigned long *chan_state);
+unsigned long sun4v_ldc_rx_set_qhead(unsigned long channel,
+				     unsigned long head_off);
+unsigned long sun4v_ldc_set_map_table(unsigned long channel,
+				      unsigned long ra,
+				      unsigned long num_entries);
+unsigned long sun4v_ldc_get_map_table(unsigned long channel,
+				      unsigned long *ra,
+				      unsigned long *num_entries);
+unsigned long sun4v_ldc_copy(unsigned long channel,
+			     unsigned long dir_code,
+			     unsigned long tgt_raddr,
+			     unsigned long lcl_raddr,
+			     unsigned long len,
+			     unsigned long *actual_len);
+unsigned long sun4v_ldc_mapin(unsigned long channel,
+			      unsigned long cookie,
+			      unsigned long *ra,
+			      unsigned long *perm);
+unsigned long sun4v_ldc_unmap(unsigned long ra);
+unsigned long sun4v_ldc_revoke(unsigned long channel,
+			       unsigned long cookie,
+			       unsigned long mte_cookie);
 #endif
 
 /* Performance counter services.  */
@@ -2713,6 +2716,30 @@ extern unsigned long sun4v_ldc_revoke(unsigned long channel,
  */
 #define HV_FAST_SET_PERFREG		0x101
 
+#define HV_N2_PERF_SPARC_CTL		0x0
+#define HV_N2_PERF_DRAM_CTL0		0x1
+#define HV_N2_PERF_DRAM_CNT0		0x2
+#define HV_N2_PERF_DRAM_CTL1		0x3
+#define HV_N2_PERF_DRAM_CNT1		0x4
+#define HV_N2_PERF_DRAM_CTL2		0x5
+#define HV_N2_PERF_DRAM_CNT2		0x6
+#define HV_N2_PERF_DRAM_CTL3		0x7
+#define HV_N2_PERF_DRAM_CNT3		0x8
+
+#define HV_FAST_N2_GET_PERFREG		0x104
+#define HV_FAST_N2_SET_PERFREG		0x105
+
+#ifndef __ASSEMBLY__
+unsigned long sun4v_niagara_getperf(unsigned long reg,
+				    unsigned long *val);
+unsigned long sun4v_niagara_setperf(unsigned long reg,
+				    unsigned long val);
+unsigned long sun4v_niagara2_getperf(unsigned long reg,
+				     unsigned long *val);
+unsigned long sun4v_niagara2_setperf(unsigned long reg,
+				     unsigned long val);
+#endif
+
 /* MMU statistics services.
  *
  * The hypervisor maintains MMU statistics and privileged code provides
@@ -2805,8 +2832,8 @@ struct hv_mmu_statistics {
 #define HV_FAST_MMUSTAT_INFO		0x103
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_mmustat_conf(unsigned long ra, unsigned long *orig_ra);
-extern unsigned long sun4v_mmustat_info(unsigned long *ra);
+unsigned long sun4v_mmustat_conf(unsigned long ra, unsigned long *orig_ra);
+unsigned long sun4v_mmustat_info(unsigned long *ra);
 #endif
 
 /* NCS crypto services  */
@@ -2895,14 +2922,31 @@ struct hv_ncs_qtail_update_arg {
 #define HV_FAST_NCS_REQUEST		0x110
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_ncs_request(unsigned long request,
-				       unsigned long arg_ra,
-				       unsigned long arg_size);
+unsigned long sun4v_ncs_request(unsigned long request,
+			        unsigned long arg_ra,
+			        unsigned long arg_size);
 #endif
 
 #define HV_FAST_FIRE_GET_PERFREG	0x120
 #define HV_FAST_FIRE_SET_PERFREG	0x121
 
+#define HV_FAST_REBOOT_DATA_SET		0x172
+
+#ifndef __ASSEMBLY__
+unsigned long sun4v_reboot_data_set(unsigned long ra,
+				    unsigned long len);
+#endif
+
+#define HV_FAST_VT_GET_PERFREG		0x184
+#define HV_FAST_VT_SET_PERFREG		0x185
+
+#ifndef __ASSEMBLY__
+unsigned long sun4v_vt_get_perfreg(unsigned long reg_num,
+				   unsigned long *reg_val);
+unsigned long sun4v_vt_set_perfreg(unsigned long reg_num,
+				   unsigned long reg_val);
+#endif
+
 /* Function numbers for HV_CORE_TRAP.  */
 #define HV_CORE_SET_VER			0x00
 #define HV_CORE_PUTCHAR			0x01
@@ -2916,34 +2960,42 @@ extern unsigned long sun4v_ncs_request(unsigned long request,
 #define HV_GRP_CORE			0x0001
 #define HV_GRP_INTR			0x0002
 #define HV_GRP_SOFT_STATE		0x0003
+#define HV_GRP_TM			0x0080
 #define HV_GRP_PCI			0x0100
 #define HV_GRP_LDOM			0x0101
 #define HV_GRP_SVC_CHAN			0x0102
 #define HV_GRP_NCS			0x0103
 #define HV_GRP_RNG			0x0104
+#define HV_GRP_PBOOT			0x0105
+#define HV_GRP_TPM			0x0107
+#define HV_GRP_SDIO			0x0108
+#define HV_GRP_SDIO_ERR			0x0109
+#define HV_GRP_REBOOT_DATA		0x0110
 #define HV_GRP_NIAG_PERF		0x0200
 #define HV_GRP_FIRE_PERF		0x0201
 #define HV_GRP_N2_CPU			0x0202
 #define HV_GRP_NIU			0x0204
 #define HV_GRP_VF_CPU			0x0205
+#define HV_GRP_KT_CPU			0x0209
+#define HV_GRP_VT_CPU			0x020c
 #define HV_GRP_DIAG			0x0300
 
 #ifndef __ASSEMBLY__
-extern unsigned long sun4v_get_version(unsigned long group,
-				       unsigned long *major,
-				       unsigned long *minor);
-extern unsigned long sun4v_set_version(unsigned long group,
-				       unsigned long major,
-				       unsigned long minor,
-				       unsigned long *actual_minor);
-
-extern int sun4v_hvapi_register(unsigned long group, unsigned long major,
-				unsigned long *minor);
-extern void sun4v_hvapi_unregister(unsigned long group);
-extern int sun4v_hvapi_get(unsigned long group,
-			   unsigned long *major,
-			   unsigned long *minor);
-extern void sun4v_hvapi_init(void);
+unsigned long sun4v_get_version(unsigned long group,
+			        unsigned long *major,
+			        unsigned long *minor);
+unsigned long sun4v_set_version(unsigned long group,
+			        unsigned long major,
+			        unsigned long minor,
+			        unsigned long *actual_minor);
+
+int sun4v_hvapi_register(unsigned long group, unsigned long major,
+			 unsigned long *minor);
+void sun4v_hvapi_unregister(unsigned long group);
+int sun4v_hvapi_get(unsigned long group,
+		    unsigned long *major,
+		    unsigned long *minor);
+void sun4v_hvapi_init(void);
 #endif
 
 #endif /* !(_SPARC64_HYPERVISOR_H) */
diff --git a/arch/sparc/include/asm/idprom.h b/arch/sparc/include/asm/idprom.h