34 files changed, 372 insertions, 545 deletions

diff --git a/include/asm-arm/arch-ixp4xx/io.h b/include/asm-arm/arch-ixp4xx/io.h
index 688f7f90d93e..942b622455bc 100644
--- a/include/asm-arm/arch-ixp4xx/io.h
+++ b/include/asm-arm/arch-ixp4xx/io.h
@@ -59,11 +59,10 @@ extern int ixp4xx_pci_write(u32 addr, u32 cmd, u32 data);
  * fallback to the default.
  */
 static inline void __iomem *
-__ixp4xx_ioremap(unsigned long addr, size_t size, unsigned long flags, unsigned long align)
+__ixp4xx_ioremap(unsigned long addr, size_t size, unsigned long flags)
 {
-	extern void __iomem * __ioremap(unsigned long, size_t, unsigned long, unsigned long);
 	if((addr < 0x48000000) || (addr > 0x4fffffff))
-		return __ioremap(addr, size, flags, align);
+		return __ioremap(addr, size, flags);
 
 	return (void *)addr;
 }
@@ -71,13 +70,11 @@ __ixp4xx_ioremap(unsigned long addr, size_t size, unsigned long flags, unsigned
 static inline void
 __ixp4xx_iounmap(void __iomem *addr)
 {
-	extern void __iounmap(void __iomem *addr);
-
 	if ((u32)addr >= VMALLOC_START)
 		__iounmap(addr);
 }
 
-#define __arch_ioremap(a, s, f, x)	__ixp4xx_ioremap(a, s, f, x)
+#define __arch_ioremap(a, s, f)		__ixp4xx_ioremap(a, s, f)
 #define	__arch_iounmap(a)		__ixp4xx_iounmap(a)
 
 #define	writeb(v, p)			__ixp4xx_writeb(v, p)
diff --git a/include/asm-arm/io.h b/include/asm-arm/io.h
index 2e6799632f12..ae69db4a1010 100644
--- a/include/asm-arm/io.h
+++ b/include/asm-arm/io.h
@@ -55,6 +55,12 @@ extern void __raw_readsl(void __iomem *addr, void *data, int longlen);
 #define __raw_readl(a)		(__chk_io_ptr(a), *(volatile unsigned int __force   *)(a))
 
 /*
+ * Architecture ioremap implementation.
+ */
+extern void __iomem * __ioremap(unsigned long, size_t, unsigned long);
+extern void __iounmap(void __iomem *addr);
+
+/*
  * Bad read/write accesses...
  */
 extern void __readwrite_bug(const char *fn);
@@ -256,18 +262,15 @@ out:
  * ioremap takes a PCI memory address, as specified in
  * Documentation/IO-mapping.txt.
  */
-extern void __iomem * __ioremap(unsigned long, size_t, unsigned long, unsigned long);
-extern void __iounmap(void __iomem *addr);
-
 #ifndef __arch_ioremap
-#define ioremap(cookie,size)		__ioremap(cookie,size,0,1)
-#define ioremap_nocache(cookie,size)	__ioremap(cookie,size,0,1)
-#define ioremap_cached(cookie,size)	__ioremap(cookie,size,L_PTE_CACHEABLE,1)
+#define ioremap(cookie,size)		__ioremap(cookie,size,0)
+#define ioremap_nocache(cookie,size)	__ioremap(cookie,size,0)
+#define ioremap_cached(cookie,size)	__ioremap(cookie,size,L_PTE_CACHEABLE)
 #define iounmap(cookie)			__iounmap(cookie)
 #else
-#define ioremap(cookie,size)		__arch_ioremap((cookie),(size),0,1)
-#define ioremap_nocache(cookie,size)	__arch_ioremap((cookie),(size),0,1)
-#define ioremap_cached(cookie,size)	__arch_ioremap((cookie),(size),L_PTE_CACHEABLE,1)
+#define ioremap(cookie,size)		__arch_ioremap((cookie),(size),0)
+#define ioremap_nocache(cookie,size)	__arch_ioremap((cookie),(size),0)
+#define ioremap_cached(cookie,size)	__arch_ioremap((cookie),(size),L_PTE_CACHEABLE)
 #define iounmap(cookie)			__arch_iounmap(cookie)
 #endif
 
diff --git a/include/asm-arm/uaccess.h b/include/asm-arm/uaccess.h
index a2fdad0138b3..064f0f5e8e2b 100644
--- a/include/asm-arm/uaccess.h
+++ b/include/asm-arm/uaccess.h
@@ -100,7 +100,6 @@ static inline void set_fs (mm_segment_t fs)
 extern int __get_user_1(void *);
 extern int __get_user_2(void *);
 extern int __get_user_4(void *);
-extern int __get_user_8(void *);
 extern int __get_user_bad(void);
 
 #define __get_user_x(__r2,__p,__e,__s,__i...)				\
@@ -114,7 +113,7 @@ extern int __get_user_bad(void);
 #define get_user(x,p)							\
 	({								\
 		const register typeof(*(p)) __user *__p asm("r0") = (p);\
-		register typeof(*(p)) __r2 asm("r2");			\
+		register unsigned int __r2 asm("r2");			\
 		register int __e asm("r0");				\
 		switch (sizeof(*(__p))) {				\
 		case 1:							\
@@ -126,12 +125,9 @@ extern int __get_user_bad(void);
 		case 4:							\
 	       		__get_user_x(__r2, __p, __e, 4, "lr");		\
 			break;						\
-		case 8:							\
-			__get_user_x(__r2, __p, __e, 8, "lr");		\
-	       		break;						\
 		default: __e = __get_user_bad(); break;			\
 		}							\
-		x = __r2;						\
+		x = (typeof(*(p))) __r2;				\
 		__e;							\
 	})
 
diff --git a/include/asm-parisc/irq.h b/include/asm-parisc/irq.h
index f876bdf22056..b0a30e2c9813 100644
--- a/include/asm-parisc/irq.h
+++ b/include/asm-parisc/irq.h
@@ -8,6 +8,7 @@
 #define _ASM_PARISC_IRQ_H
 
 #include <linux/config.h>
+#include <linux/cpumask.h>
 #include <asm/types.h>
 
 #define NO_IRQ		(-1)
@@ -49,10 +50,10 @@ extern int txn_alloc_irq(unsigned int nbits);
 extern int txn_claim_irq(int);
 extern unsigned int txn_alloc_data(unsigned int);
 extern unsigned long txn_alloc_addr(unsigned int);
+extern unsigned long txn_affinity_addr(unsigned int irq, int cpu);
 
 extern int cpu_claim_irq(unsigned int irq, struct hw_interrupt_type *, void *);
-
-extern int cpu_claim_irq(unsigned int irq, struct hw_interrupt_type *, void *);
+extern int cpu_check_affinity(unsigned int irq, cpumask_t *dest);
 
 /* soft power switch support (power.c) */
 extern struct tasklet_struct power_tasklet;
diff --git a/include/asm-parisc/smp.h b/include/asm-parisc/smp.h
index 9413f67a540b..dbdbd2e9fdf9 100644
--- a/include/asm-parisc/smp.h
+++ b/include/asm-parisc/smp.h
@@ -29,6 +29,7 @@ extern cpumask_t cpu_online_map;
 #define cpu_logical_map(cpu)	(cpu)
 
 extern void smp_send_reschedule(int cpu);
+extern void smp_send_all_nop(void);
 
 #endif /* !ASSEMBLY */
 
@@ -53,7 +54,11 @@ extern unsigned long cpu_present_mask;
 
 #define raw_smp_processor_id()	(current_thread_info()->cpu)
 
-#endif /* CONFIG_SMP */
+#else /* CONFIG_SMP */
+
+static inline void smp_send_all_nop(void) { return; }
+
+#endif
 
 #define NO_PROC_ID		0xFF		/* No processor magic marker */
 #define ANY_PROC_ID		0xFF		/* Any processor magic marker */
diff --git a/include/asm-parisc/spinlock.h b/include/asm-parisc/spinlock.h
index 7c3f406a746a..16c2ac075fc5 100644
--- a/include/asm-parisc/spinlock.h
+++ b/include/asm-parisc/spinlock.h
@@ -11,18 +11,25 @@ static inline int __raw_spin_is_locked(raw_spinlock_t *x)
 	return *a == 0;
 }
 
-#define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock)
+#define __raw_spin_lock(lock) __raw_spin_lock_flags(lock, 0)
 #define __raw_spin_unlock_wait(x) \
 		do { cpu_relax(); } while (__raw_spin_is_locked(x))
 
-static inline void __raw_spin_lock(raw_spinlock_t *x)
+static inline void __raw_spin_lock_flags(raw_spinlock_t *x,
+					 unsigned long flags)
 {
 	volatile unsigned int *a;
 
 	mb();
 	a = __ldcw_align(x);
 	while (__ldcw(a) == 0)
-		while (*a == 0);
+		while (*a == 0)
+			if (flags & PSW_SM_I) {
+				local_irq_enable();
+				cpu_relax();
+				local_irq_disable();
+			} else
+				cpu_relax();
 	mb();
 }
 
@@ -60,26 +67,20 @@ static inline int __raw_spin_trylock(raw_spinlock_t *x)
 
 static  __inline__ void __raw_read_lock(raw_rwlock_t *rw)
 {
-	unsigned long flags;
-	local_irq_save(flags);
 	__raw_spin_lock(&rw->lock);
 
 	rw->counter++;
 
 	__raw_spin_unlock(&rw->lock);
-	local_irq_restore(flags);
 }
 
 static  __inline__ void __raw_read_unlock(raw_rwlock_t *rw)
 {
-	unsigned long flags;
-	local_irq_save(flags);
 	__raw_spin_lock(&rw->lock);
 
 	rw->counter--;
 
 	__raw_spin_unlock(&rw->lock);
-	local_irq_restore(flags);
 }
 
 /* write_lock is less trivial.  We optimistically grab the lock and check
diff --git a/include/asm-parisc/tlbflush.h b/include/asm-parisc/tlbflush.h
index e97aa8d1eff5..c9ec39c6fc6c 100644
--- a/include/asm-parisc/tlbflush.h
+++ b/include/asm-parisc/tlbflush.h
@@ -12,21 +12,15 @@
  * N class systems, only one PxTLB inter processor broadcast can be
  * active at any one time on the Merced bus.  This tlb purge
  * synchronisation is fairly lightweight and harmless so we activate
- * it on all SMP systems not just the N class. */
-#ifdef CONFIG_SMP
+ * it on all SMP systems not just the N class.  We also need to have
+ * preemption disabled on uniprocessor machines, and spin_lock does that
+ * nicely.
+ */
 extern spinlock_t pa_tlb_lock;
 
 #define purge_tlb_start(x) spin_lock(&pa_tlb_lock)
 #define purge_tlb_end(x) spin_unlock(&pa_tlb_lock)
 
-#else
-
-#define purge_tlb_start(x) do { } while(0)
-#define purge_tlb_end(x) do { } while (0)
-
-#endif
-
-
 extern void flush_tlb_all(void);
 
 /*
@@ -88,7 +82,6 @@ static inline void flush_tlb_range(struct vm_area_struct *vma,
 	if (npages >= 512)  /* 2MB of space: arbitrary, should be tuned */
 		flush_tlb_all();
 	else {
-		preempt_disable();
 		mtsp(vma->vm_mm->context,1);
 		purge_tlb_start();
 		if (split_tlb) {
@@ -102,7 +95,6 @@ static inline void flush_tlb_range(struct vm_area_struct *vma,
 				pdtlb(start);
 				start += PAGE_SIZE;
 			}
-		preempt_enable();
 		}
 		purge_tlb_end();
 	}
diff --git a/include/asm-powerpc/cputable.h b/include/asm-powerpc/cputable.h
index 04e2726002cf..d1cfa3f515ea 100644
--- a/include/asm-powerpc/cputable.h
+++ b/include/asm-powerpc/cputable.h
@@ -90,6 +90,7 @@ extern void do_cpu_ftr_fixups(unsigned long offset);
 #define CPU_FTR_NEED_COHERENT		ASM_CONST(0x0000000000020000)
 #define CPU_FTR_NO_BTIC			ASM_CONST(0x0000000000040000)
 #define CPU_FTR_BIG_PHYS		ASM_CONST(0x0000000000080000)
+#define CPU_FTR_NODSISRALIGN  		ASM_CONST(0x0000000000100000)
 
 #ifdef __powerpc64__
 /* Add the 64b processor unique features in the top half of the word */
@@ -97,7 +98,6 @@ extern void do_cpu_ftr_fixups(unsigned long offset);
 #define CPU_FTR_16M_PAGE      		ASM_CONST(0x0000000200000000)
 #define CPU_FTR_TLBIEL         		ASM_CONST(0x0000000400000000)
 #define CPU_FTR_NOEXECUTE     		ASM_CONST(0x0000000800000000)
-#define CPU_FTR_NODSISRALIGN  		ASM_CONST(0x0000001000000000)
 #define CPU_FTR_IABR  			ASM_CONST(0x0000002000000000)
 #define CPU_FTR_MMCRA  			ASM_CONST(0x0000004000000000)
 #define CPU_FTR_CTRL			ASM_CONST(0x0000008000000000)
@@ -113,7 +113,6 @@ extern void do_cpu_ftr_fixups(unsigned long offset);
 #define CPU_FTR_16M_PAGE      		ASM_CONST(0x0)
 #define CPU_FTR_TLBIEL         		ASM_CONST(0x0)
 #define CPU_FTR_NOEXECUTE     		ASM_CONST(0x0)
-#define CPU_FTR_NODSISRALIGN  		ASM_CONST(0x0)
 #define CPU_FTR_IABR  			ASM_CONST(0x0)
 #define CPU_FTR_MMCRA  			ASM_CONST(0x0)
 #define CPU_FTR_CTRL			ASM_CONST(0x0)
@@ -273,18 +272,21 @@ enum {
 	CPU_FTRS_POWER3_32 = CPU_FTR_COMMON | CPU_FTR_SPLIT_ID_CACHE |
 	    CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE,
 	CPU_FTRS_POWER4_32 = CPU_FTR_COMMON | CPU_FTR_SPLIT_ID_CACHE |
-	    CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE,
+	    CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | CPU_FTR_NODSISRALIGN,
 	CPU_FTRS_970_32 = CPU_FTR_COMMON | CPU_FTR_SPLIT_ID_CACHE |
 	    CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | CPU_FTR_ALTIVEC_COMP |
-	    CPU_FTR_MAYBE_CAN_NAP,
+	    CPU_FTR_MAYBE_CAN_NAP | CPU_FTR_NODSISRALIGN,
 	CPU_FTRS_8XX = CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB,
-	CPU_FTRS_40X = CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB,
-	CPU_FTRS_44X = CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB,
-	CPU_FTRS_E200 = CPU_FTR_USE_TB,
-	CPU_FTRS_E500 = CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB,
+	CPU_FTRS_40X = CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB |
+	    CPU_FTR_NODSISRALIGN,
+	CPU_FTRS_44X = CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB |
+	    CPU_FTR_NODSISRALIGN,
+	CPU_FTRS_E200 = CPU_FTR_USE_TB | CPU_FTR_NODSISRALIGN,
+	CPU_FTRS_E500 = CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB |
+	    CPU_FTR_NODSISRALIGN,
 	CPU_FTRS_E500_2 = CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB |
-	    CPU_FTR_BIG_PHYS,
-	CPU_FTRS_GENERIC_32 = CPU_FTR_COMMON,
+	    CPU_FTR_BIG_PHYS | CPU_FTR_NODSISRALIGN,
+	CPU_FTRS_GENERIC_32 = CPU_FTR_COMMON | CPU_FTR_NODSISRALIGN,
 #ifdef __powerpc64__
 	CPU_FTRS_POWER3 = CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB |
 	    CPU_FTR_HPTE_TABLE | CPU_FTR_IABR,
diff --git a/include/asm-powerpc/delay.h b/include/asm-powerpc/delay.h
index 1492aa9ab716..54fe1f4f8fd0 100644
--- a/include/asm-powerpc/delay.h
+++ b/include/asm-powerpc/delay.h
@@ -13,43 +13,7 @@
  * Anton Blanchard.
  */
 
-extern unsigned long tb_ticks_per_usec;
-
-#ifdef CONFIG_PPC64
-/* define these here to prevent circular dependencies */
-/* these instructions control the thread priority on multi-threaded cpus */
-#define __HMT_low()	asm volatile("or 1,1,1")
-#define __HMT_medium()	asm volatile("or 2,2,2")
-#else
-#define __HMT_low()
-#define __HMT_medium()
-#endif
-
-#define __barrier()	asm volatile("" ::: "memory")
-
-static inline unsigned long __get_tb(void)
-{
-	unsigned long rval;
-
-	asm volatile("mftb %0" : "=r" (rval));
-	return rval;
-}
-
-static inline void __delay(unsigned long loops)
-{
-	unsigned long start = __get_tb();
-
-	while((__get_tb() - start) < loops)
-		__HMT_low();
-	__HMT_medium();
-	__barrier();
-}
-
-static inline void udelay(unsigned long usecs)
-{
-	unsigned long loops = tb_ticks_per_usec * usecs;
-
-	__delay(loops);
-}
+extern void __delay(unsigned long loops);
+extern void udelay(unsigned long usecs);
 
 #endif /* _ASM_POWERPC_DELAY_H */
diff --git a/include/asm-ppc/dma-mapping.h b/include/asm-powerpc/dma-mapping.h
index 6e9635114433..59a80163f75f 100644
--- a/include/asm-ppc/dma-mapping.h
+++ b/include/asm-powerpc/dma-mapping.h
@@ -1,15 +1,22 @@
 /*
- * This is based on both include/asm-sh/dma-mapping.h and
- * include/asm-ppc/pci.h
+ * Copyright (C) 2004 IBM
+ *
+ * Implements the generic device dma API for powerpc.
+ * the pci and vio busses
  */
-#ifndef __ASM_PPC_DMA_MAPPING_H
-#define __ASM_PPC_DMA_MAPPING_H
+#ifndef _ASM_DMA_MAPPING_H
+#define _ASM_DMA_MAPPING_H
 
 #include <linux/config.h>
+#include <linux/types.h>
+#include <linux/cache.h>
 /* need struct page definitions */
 #include <linux/mm.h>
 #include <asm/scatterlist.h>
 #include <asm/io.h>
+#include <asm/bug.h>
+
+#define DMA_ERROR_CODE		(~(dma_addr_t)0x0)
 
 #ifdef CONFIG_NOT_COHERENT_CACHE
 /*
@@ -24,22 +31,12 @@ extern void __dma_free_coherent(size_t size, void *vaddr);
 extern void __dma_sync(void *vaddr, size_t size, int direction);
 extern void __dma_sync_page(struct page *page, unsigned long offset,
 				 size_t size, int direction);
-#define dma_cache_inv(_start,_size) \
-	invalidate_dcache_range(_start, (_start + _size))
-#define dma_cache_wback(_start,_size) \
-	clean_dcache_range(_start, (_start + _size))
-#define dma_cache_wback_inv(_start,_size) \
-	flush_dcache_range(_start, (_start + _size))
 
 #else /* ! CONFIG_NOT_COHERENT_CACHE */
 /*
  * Cache coherent cores.
  */
 
-#define dma_cache_inv(_start,_size)		do { } while (0)
-#define dma_cache_wback(_start,_size)		do { } while (0)
-#define dma_cache_wback_inv(_start,_size)	do { } while (0)
-
 #define __dma_alloc_coherent(gfp, size, handle)	NULL
 #define __dma_free_coherent(size, addr)		do { } while (0)
 #define __dma_sync(addr, size, rw)		do { } while (0)
@@ -47,6 +44,30 @@ extern void __dma_sync_page(struct page *page, unsigned long offset,
 
 #endif /* ! CONFIG_NOT_COHERENT_CACHE */
 
+#ifdef CONFIG_PPC64
+
+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 nhwentries, enum dma_data_direction direction);
+
+#else /* CONFIG_PPC64 */
+
 #define dma_supported(dev, mask)	(1)
 
 static inline int dma_set_mask(struct device *dev, u64 dma_mask)
@@ -144,29 +165,27 @@ dma_map_sg(struct device *dev, struct scatterlist *sg, int nents,
 /* We don't do anything here. */
 #define dma_unmap_sg(dev, sg, nents, dir)	do { } while (0)
 
-static inline void
-dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle,
-			size_t size,
-			enum dma_data_direction direction)
+#endif /* CONFIG_PPC64 */
+
+static inline void dma_sync_single_for_cpu(struct device *dev,
+		dma_addr_t dma_handle, size_t size,
+		enum dma_data_direction direction)
 {
 	BUG_ON(direction == DMA_NONE);
-
 	__dma_sync(bus_to_virt(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)
+static inline void dma_sync_single_for_device(struct device *dev,
+		dma_addr_t dma_handle, size_t size,
+		enum dma_data_direction direction)
 {
 	BUG_ON(direction == DMA_NONE);
-
 	__dma_sync(bus_to_virt(dma_handle), size, direction);
 }
 
-static inline void
-dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nents,
-		    enum dma_data_direction direction)
+static inline void dma_sync_sg_for_cpu(struct device *dev,
+		struct scatterlist *sg, int nents,
+		enum dma_data_direction direction)
 {
 	int i;
 
@@ -176,9 +195,9 @@ dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nents,
 		__dma_sync_page(sg->page, sg->offset, sg->length, direction);
 }
 
-static inline void
-dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nents,
-		    enum dma_data_direction direction)
+static inline void dma_sync_sg_for_device(struct device *dev,
+		struct scatterlist *sg, int nents,
+		enum dma_data_direction direction)
 {
 	int i;
 
@@ -188,6 +207,15 @@ dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nents,
 		__dma_sync_page(sg->page, sg->offset, sg->length, direction);
 }
 
+static inline int dma_mapping_error(dma_addr_t dma_addr)
+{
+#ifdef CONFIG_PPC64
+	return (dma_addr == DMA_ERROR_CODE);
+#else
+	return 0;
+#endif
+}
+
 #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)
 #ifdef CONFIG_NOT_COHERENT_CACHE
@@ -198,40 +226,60 @@ dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nents,
 
 static inline int dma_get_cache_alignment(void)
 {
+#ifdef CONFIG_PPC64
+	/* no easy way to get cache size on all processors, so return
+	 * the maximum possible, to be safe */
+	return (1 << L1_CACHE_SHIFT_MAX);
+#else
 	/*
 	 * Each processor family will define its own L1_CACHE_SHIFT,
 	 * L1_CACHE_BYTES wraps to this, so this is always safe.
 	 */
 	return L1_CACHE_BYTES;
+#endif
 }
 
-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)
+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)
 {
 	/* just sync everything for now */
 	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)
+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)
 {
 	/* just sync everything for now */
 	dma_sync_single_for_device(dev, dma_handle, offset + size, direction);
 }
 
 static inline void dma_cache_sync(void *vaddr, size_t size,
-				  enum dma_data_direction direction)
+		enum dma_data_direction direction)
 {
+	BUG_ON(direction == DMA_NONE);
 	__dma_sync(vaddr, size, (int)direction);
 }
 
-static inline int dma_mapping_error(dma_addr_t dma_addr)
-{
-	return 0;
-}
-
-#endif				/* __ASM_PPC_DMA_MAPPING_H */
+/*
+ * DMA operations are abstracted for G5 vs. i/pSeries, PCI vs. VIO
+ */
+struct dma_mapping_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 *vaddr, dma_addr_t dma_handle);
+	dma_addr_t	(*map_single)(struct device *dev, void *ptr,
+				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 nents, enum dma_data_direction direction);
+	int		(*dma_supported)(struct device *dev, u64 mask);
+	int		(*dac_dma_supported)(struct device *dev, u64 mask);
+};
+
+#endif	/* _ASM_DMA_MAPPING_H */
diff --git a/include/asm-powerpc/eeh.h b/include/asm-powerpc/eeh.h
index 89f26ab31908..f8633aafe4ba 100644
--- a/include/asm-powerpc/eeh.h
+++ b/include/asm-powerpc/eeh.h
@@ -30,6 +30,8 @@ struct device_node;
 
 #ifdef CONFIG_EEH
 
+extern int eeh_subsystem_enabled;
+
 /* Values for eeh_mode bits in device_node */
 #define EEH_MODE_SUPPORTED	(1<<0)
 #define EEH_MODE_NOCHECK	(1<<1)
@@ -75,7 +77,7 @@ void eeh_remove_device(struct pci_dev *);
  * If this macro yields TRUE, the caller relays to eeh_check_failure()
  * which does further tests out of line.
  */
-#define EEH_POSSIBLE_ERROR(val, type)	((val) == (type)~0)
+#define EEH_POSSIBLE_ERROR(val, type)	((val) == (type)~0 && eeh_subsystem_enabled)
 
 /*
  * Reads from a device which has been isolated by EEH will return
diff --git a/include/asm-ppc64/io.h b/include/asm-powerpc/io.h
index 77fc07c3c6bd..48938d84d055 100644
--- a/include/asm-ppc64/io.h
+++ b/include/asm-powerpc/io.h
@@ -1,5 +1,5 @@
-#ifndef _PPC64_IO_H
-#define _PPC64_IO_H
+#ifndef _ASM_POWERPC_IO_H
+#define _ASM_POWERPC_IO_H
 
 /* 
  * This program is free software; you can redistribute it and/or
@@ -8,7 +8,10 @@
  * 2 of the License, or (at your option) any later version.
  */
 
-#include <linux/config.h>
+#ifndef CONFIG_PPC64
+#include <asm-ppc/io.h>
+#else
+
 #include <linux/compiler.h>
 #include <asm/page.h>
 #include <asm/byteorder.h>
@@ -455,4 +458,5 @@ extern int check_legacy_ioport(unsigned long base_port);
 
 #endif /* __KERNEL__ */
 
-#endif /* _PPC64_IO_H */
+#endif /* CONFIG_PPC64 */
+#endif /* _ASM_POWERPC_IO_H */
diff --git a/include/asm-ppc64/mmu.h b/include/asm-powerpc/mmu.h
index 1a7e0afa2dc6..c1b4bbabbe97 100644
--- a/include/asm-ppc64/mmu.h
+++ b/include/asm-powerpc/mmu.h
@@ -1,3 +1,10 @@
+#ifndef _ASM_POWERPC_MMU_H_
+#define _ASM_POWERPC_MMU_H_
+
+#ifndef CONFIG_PPC64
+#include <asm-ppc/mmu.h>
+#else
+
 /*
  * PowerPC memory management structures
  *
@@ -10,10 +17,6 @@
  * 2 of the License, or (at your option) any later version.
  */
 
-#ifndef _PPC64_MMU_H_
-#define _PPC64_MMU_H_
-
-#include <linux/config.h>
 #include <asm/asm-compat.h>
 #include <asm/page.h>
 
@@ -392,4 +395,5 @@ static inline unsigned long get_vsid(unsigned long context, unsigned long ea)
 
 #endif /* __ASSEMBLY */
 
-#endif /* _PPC64_MMU_H_ */
+#endif /* CONFIG_PPC64 */
+#endif /* _ASM_POWERPC_MMU_H_ */
diff --git a/include/asm-ppc64/mmu_context.h b/include/asm-powerpc/mmu_context.h
index 4f512e9fa6b8..ea6798c7d5fc 100644
--- a/include/asm-ppc64/mmu_context.h
+++ b/include/asm-powerpc/mmu_context.h
@@ -1,7 +1,10 @@
-#ifndef __PPC64_MMU_CONTEXT_H
-#define __PPC64_MMU_CONTEXT_H
+#ifndef __ASM_POWERPC_MMU_CONTEXT_H
+#define __ASM_POWERPC_MMU_CONTEXT_H
+
+#ifndef CONFIG_PPC64
+#include <asm-ppc/mmu_context.h>
+#else
 
-#include <linux/config.h>
 #include <linux/kernel.h>	
 #include <linux/mm.h>	
 #include <asm/mmu.h>	
@@ -82,4 +85,5 @@ static inline void activate_mm(struct mm_struct *prev, struct mm_struct *next)
 	local_irq_restore(flags);
 }
 
-#endif /* __PPC64_MMU_CONTEXT_H */
+#endif /* CONFIG_PPC64 */
+#endif /* __ASM_POWERPC_MMU_CONTEXT_H */
diff --git a/include/asm-ppc64/mmzone.h b/include/asm-powerpc/mmzone.h
index 54958d6cae04..54958d6cae04 100644
--- a/include/asm-ppc64/mmzone.h
+++ b/include/asm-powerpc/mmzone.h
diff --git a/include/asm-powerpc/page_64.h b/include/asm-powerpc/page_64.h
index c16f106b5373..1e6e7846824f 100644
--- a/include/asm-powerpc/page_64.h
+++ b/include/asm-powerpc/page_64.h
@@ -86,7 +86,11 @@ static inline void copy_page(void *to, void *from)
 extern u64 ppc64_pft_size;
 
 /* Large pages size */
+#ifdef CONFIG_HUGETLB_PAGE
 extern unsigned int HPAGE_SHIFT;
+#else
+#define HPAGE_SHIFT PAGE_SHIFT
+#endif
 #define HPAGE_SIZE		((1UL) << HPAGE_SHIFT)
 #define HPAGE_MASK		(~(HPAGE_SIZE - 1))
 #define HUGETLB_PAGE_ORDER	(HPAGE_SHIFT - PAGE_SHIFT)
diff --git a/include/asm-ppc64/pci-bridge.h b/include/asm-powerpc/pci-bridge.h
index cf04327a597a..223ec7bd81da 100644
--- a/include/asm-ppc64/pci-bridge.h
+++ b/include/asm-powerpc/pci-bridge.h
@@ -1,8 +1,10 @@
-#ifdef __KERNEL__
-#ifndef _ASM_PCI_BRIDGE_H
-#define _ASM_PCI_BRIDGE_H
+#ifndef _ASM_POWERPC_PCI_BRIDGE_H
+#define _ASM_POWERPC_PCI_BRIDGE_H
+
+#ifndef CONFIG_PPC64
+#include <asm-ppc/pci-bridge.h>
+#else
 
-#include <linux/config.h>
 #include <linux/pci.h>
 #include <linux/list.h>
 
@@ -147,5 +149,5 @@ extern void pcibios_free_controller(struct pci_controller *phb);
 #define PCI_PROBE_NORMAL	0	/* Do normal PCI probing */
 #define PCI_PROBE_DEVTREE	1	/* Instantiate from device tree */
 
+#endif /* CONFIG_PPC64 */
 #endif
-#endif /* __KERNEL__ */
diff --git a/include/asm-ppc64/pci.h b/include/asm-powerpc/pci.h
index fafdf885a3cc..d5934a076bd0 100644
--- a/include/asm-ppc64/pci.h
+++ b/include/asm-powerpc/pci.h
@@ -1,5 +1,5 @@
-#ifndef __PPC64_PCI_H
-#define __PPC64_PCI_H
+#ifndef __ASM_POWERPC_PCI_H
+#define __ASM_POWERPC_PCI_H
 #ifdef __KERNEL__
 
 /*
@@ -18,6 +18,7 @@
 #include <asm/scatterlist.h>
 #include <asm/io.h>
 #include <asm/prom.h>
+#include <asm/pci-bridge.h>
 
 #include <asm-generic/pci-dma-compat.h>
 
@@ -26,11 +27,21 @@
 
 struct pci_dev;
 
-#ifdef CONFIG_PPC_ISERIES
+/* Values for the `which' argument to sys_pciconfig_iobase syscall.  */
+#define IOBASE_BRIDGE_NUMBER	0
+#define IOBASE_MEMORY		1
+#define IOBASE_IO		2
+#define IOBASE_ISA_IO		3
+#define IOBASE_ISA_MEM		4
+
+/*
+ * Set this to 1 if you want the kernel to re-assign all PCI
+ * bus numbers
+ */
+extern int pci_assign_all_buses;
+#define pcibios_assign_all_busses()	(pci_assign_all_buses)
+
 #define pcibios_scan_all_fns(a, b)	0
-#else
-extern int pcibios_scan_all_fns(struct pci_bus *bus, int devfn);
-#endif
 
 static inline void pcibios_set_master(struct pci_dev *dev)
 {
@@ -50,6 +61,7 @@ static inline int pci_get_legacy_ide_irq(struct pci_dev *dev, int channel)
 	return channel ? 15 : 14;
 }
 
+#ifdef CONFIG_PPC64
 #define HAVE_ARCH_PCI_MWI 1
 static inline int pcibios_prep_mwi(struct pci_dev *dev)
 {
@@ -64,12 +76,10 @@ static inline int pcibios_prep_mwi(struct pci_dev *dev)
 	return 0;
 }
 
-extern unsigned int pcibios_assign_all_busses(void);
-
 extern struct dma_mapping_ops pci_dma_ops;
 
 /* For DAC DMA, we currently don't support it by default, but
- * we let the platform override this
+ * we let 64-bit platforms override this.
  */
 static inline int pci_dac_dma_supported(struct pci_dev *hwdev,u64 mask)
 {
@@ -102,6 +112,35 @@ extern int pci_domain_nr(struct pci_bus *bus);
 /* Decide whether to display the domain number in /proc */
 extern int pci_proc_domain(struct pci_bus *bus);
 
+#else /* 32-bit */
+
+#ifdef CONFIG_PCI
+static inline void pci_dma_burst_advice(struct pci_dev *pdev,
+					enum pci_dma_burst_strategy *strat,
+					unsigned long *strategy_parameter)
+{
+	*strat = PCI_DMA_BURST_INFINITY;
+	*strategy_parameter = ~0UL;
+}
+#endif
+
+/*
+ * At present there are very few 32-bit PPC machines that can have
+ * memory above the 4GB point, and we don't support that.
+ */
+#define pci_dac_dma_supported(pci_dev, mask)	(0)
+
+/* Return the index of the PCI controller for device PDEV. */
+#define pci_domain_nr(bus) ((struct pci_controller *)(bus)->sysdata)->index
+
+/* Set the name of the bus as it appears in /proc/bus/pci */
+static inline int pci_proc_domain(struct pci_bus *bus)
+{
+	return 0;
+}
+
+#endif /* CONFIG_PPC64 */
+
 struct vm_area_struct;
 /* Map a range of PCI memory or I/O space for a device into user space */
 int pci_mmap_page_range(struct pci_dev *pdev, struct vm_area_struct *vma,
@@ -110,6 +149,7 @@ int pci_mmap_page_range(struct pci_dev *pdev, struct vm_area_struct *vma,
 /* Tell drivers/pci/proc.c that we have pci_mmap_page_range() */
 #define HAVE_PCI_MMAP	1
 
+#ifdef CONFIG_PPC64
 /* pci_unmap_{single,page} is not a nop, thus... */
 #define DECLARE_PCI_UNMAP_ADDR(ADDR_NAME)	\
 	dma_addr_t ADDR_NAME;
@@ -124,22 +164,40 @@ int pci_mmap_page_range(struct pci_dev *pdev, struct vm_area_struct *vma,
 #define pci_unmap_len_set(PTR, LEN_NAME, VAL)		\
 	(((PTR)->LEN_NAME) = (VAL))
 
-/* The PCI address space does equal the physical memory
- * address space.  The networking and block device layers use
+/* The PCI address space does not equal the physical memory address
+ * space (we have an IOMMU).  The IDE and SCSI device layers use
  * this boolean for bounce buffer decisions.
  */
 #define PCI_DMA_BUS_IS_PHYS	(0)
+
+#else /* 32-bit */
+
+/* The PCI address space does equal the physical memory
+ * address space (no IOMMU).  The IDE and SCSI device layers use
+ * this boolean for bounce buffer decisions.
+ */
+#define PCI_DMA_BUS_IS_PHYS     (1)
+
+/* pci_unmap_{page,single} is a nop so... */
+#define DECLARE_PCI_UNMAP_ADDR(ADDR_NAME)
+#define DECLARE_PCI_UNMAP_LEN(LEN_NAME)
+#define pci_unmap_addr(PTR, ADDR_NAME)		(0)
+#define pci_unmap_addr_set(PTR, ADDR_NAME, VAL)	do { } while (0)
+#define pci_unmap_len(PTR, LEN_NAME)		(0)
+#define pci_unmap_len_set(PTR, LEN_NAME, VAL)	do { } while (0)
+
+#endif /* CONFIG_PPC64 */
 	
-extern void
-pcibios_resource_to_bus(struct pci_dev *dev, struct pci_bus_region *region,
+extern void pcibios_resource_to_bus(struct pci_dev *dev,
+			struct pci_bus_region *region,
 			struct resource *res);
 
-extern void
-pcibios_bus_to_resource(struct pci_dev *dev, struct resource *res,
+extern void pcibios_bus_to_resource(struct pci_dev *dev,
+			struct resource *res,
 			struct pci_bus_region *region);
 
-static inline struct resource *
-pcibios_select_root(struct pci_dev *pdev, struct resource *res)
+static inline struct resource *pcibios_select_root(struct pci_dev *pdev,
+			struct resource *res)
 {
 	struct resource *root = NULL;
 
@@ -151,14 +209,12 @@ pcibios_select_root(struct pci_dev *pdev, struct resource *res)
 	return root;
 }
 
-extern int
-unmap_bus_range(struct pci_bus *bus);
+extern int unmap_bus_range(struct pci_bus *bus);
 
-extern int
-remap_bus_range(struct pci_bus *bus);
+extern int remap_bus_range(struct pci_bus *bus);
 
-extern void
-pcibios_fixup_device_resources(struct pci_dev *dev, struct pci_bus *bus);
+extern void pcibios_fixup_device_resources(struct pci_dev *dev,
+			struct pci_bus *bus);
 
 extern struct pci_controller *init_phb_dynamic(struct device_node *dn);
 
@@ -180,14 +236,12 @@ extern pgprot_t	pci_phys_mem_access_prot(struct file *file,
 					 unsigned long size,
 					 pgprot_t prot);
 
-#ifdef CONFIG_PPC_MULTIPLATFORM
+#if defined(CONFIG_PPC_MULTIPLATFORM) || defined(CONFIG_PPC32)
 #define HAVE_ARCH_PCI_RESOURCE_TO_USER
 extern void pci_resource_to_user(const struct pci_dev *dev, int bar,
 				 const struct resource *rsrc,
 				 u64 *start, u64 *end);
-#endif /* CONFIG_PPC_MULTIPLATFORM */
-
+#endif /* CONFIG_PPC_MULTIPLATFORM || CONFIG_PPC32 */
 
 #endif	/* __KERNEL__ */
-
-#endif /* __PPC64_PCI_H */
+#endif /* __ASM_POWERPC_PCI_H */
diff --git a/include/asm-ppc64/pgalloc.h b/include/asm-powerpc/pgalloc.h
index dcf3622d1946..bfc2113b3630 100644
--- a/include/asm-ppc64/pgalloc.h
+++ b/include/asm-powerpc/pgalloc.h
@@ -1,5 +1,9 @@
-#ifndef _PPC64_PGALLOC_H
-#define _PPC64_PGALLOC_H
+#ifndef _ASM_POWERPC_PGALLOC_H
+#define _ASM_POWERPC_PGALLOC_H
+
+#ifndef CONFIG_PPC64
+#include <asm-ppc/pgalloc.h>
+#else
 
 #include <linux/mm.h>
 #include <linux/slab.h>
@@ -148,4 +152,5 @@ extern void pgtable_free_tlb(struct mmu_gather *tlb, pgtable_free_t pgf);
 
 #define check_pgt_cache()	do { } while (0)
 
-#endif /* _PPC64_PGALLOC_H */
+#endif /* CONFIG_PPC64 */
+#endif /* _ASM_POWERPC_PGALLOC_H */
diff --git a/include/asm-ppc64/pgtable-4k.h b/include/asm-powerpc/pgtable-4k.h
index e9590c06ad92..e9590c06ad92 100644
--- a/include/asm-ppc64/pgtable-4k.h
+++ b/include/asm-powerpc/pgtable-4k.h
diff --git a/include/asm-ppc64/pgtable-64k.h b/include/asm-powerpc/pgtable-64k.h
index 154f1840ece4..154f1840ece4 100644
--- a/include/asm-ppc64/pgtable-64k.h
+++ b/include/asm-powerpc/pgtable-64k.h
diff --git a/include/asm-ppc64/pgtable.h b/include/asm-powerpc/pgtable.h
index a9783ba7fe98..0303f57366c1 100644
--- a/include/asm-ppc64/pgtable.h
+++ b/include/asm-powerpc/pgtable.h
@@ -1,5 +1,9 @@
-#ifndef _PPC64_PGTABLE_H
-#define _PPC64_PGTABLE_H
+#ifndef _ASM_POWERPC_PGTABLE_H
+#define _ASM_POWERPC_PGTABLE_H
+
+#ifndef CONFIG_PPC64
+#include <asm-ppc/pgtable.h>
+#else
 
 /*
  * This file contains the functions and defines necessary to modify and use
@@ -47,6 +51,13 @@ struct mm_struct;
 #define VMALLOC_END   (VMALLOC_START + VMALLOC_SIZE)
 
 /*
+ * Define the address range of the imalloc VM area.
+ */
+#define PHBS_IO_BASE	VMALLOC_END
+#define IMALLOC_BASE	(PHBS_IO_BASE + 0x80000000ul)	/* Reserve 2 gigs for PHBs */
+#define IMALLOC_END	(VMALLOC_START + PGTABLE_RANGE)
+
+/*
  * Common bits in a linux-style PTE.  These match the bits in the
  * (hardware-defined) PowerPC PTE as closely as possible. Additional
  * bits may be defined in pgtable-*.h
@@ -69,7 +80,7 @@ struct mm_struct;
 
 #define _PAGE_WRENABLE	(_PAGE_RW | _PAGE_DIRTY)
 
-/* __pgprot defined in asm-ppc64/page.h */
+/* __pgprot defined in asm-powerpc/page.h */
 #define PAGE_NONE	__pgprot(_PAGE_PRESENT | _PAGE_ACCESSED)
 
 #define PAGE_SHARED	__pgprot(_PAGE_BASE | _PAGE_RW | _PAGE_USER)
@@ -509,4 +520,5 @@ void pgtable_cache_init(void);
 
 #endif /* __ASSEMBLY__ */
 
-#endif /* _PPC64_PGTABLE_H */
+#endif /* CONFIG_PPC64 */
+#endif /* _ASM_POWERPC_PGTABLE_H */
diff --git a/include/asm-powerpc/ppc-pci.h b/include/asm-powerpc/ppc-pci.h
index 2e36e5a7f4f3..36cdc869e580 100644
--- a/include/asm-powerpc/ppc-pci.h
+++ b/include/asm-powerpc/ppc-pci.h
@@ -48,8 +48,6 @@ extern void pSeries_final_fixup(void);
 extern void pSeries_irq_bus_setup(struct pci_bus *bus);
 
 extern unsigned long pci_probe_only;
-extern unsigned long pci_assign_all_buses;
-extern int pci_read_irq_line(struct pci_dev *pci_dev);
 
 /* ---- EEH internal-use-only related routines ---- */
 #ifdef CONFIG_EEH
diff --git a/include/asm-ppc64/spinlock.h b/include/asm-powerpc/spinlock.h
index 7d84fb5e39f1..caa4b14e0e94 100644
--- a/include/asm-ppc64/spinlock.h
+++ b/include/asm-powerpc/spinlock.h
@@ -18,31 +18,41 @@
  *
  * (the type definitions are in asm/spinlock_types.h)
  */
-#include <linux/config.h>
+#ifdef CONFIG_PPC64
 #include <asm/paca.h>
 #include <asm/hvcall.h>
 #include <asm/iseries/hv_call.h>
+#endif
+#include <asm/asm-compat.h>
+#include <asm/synch.h>
 
 #define __raw_spin_is_locked(x)		((x)->slock != 0)
 
+#ifdef CONFIG_PPC64
+/* use 0x800000yy when locked, where yy == CPU number */
+#define LOCK_TOKEN	(*(u32 *)(&get_paca()->lock_token))
+#else
+#define LOCK_TOKEN	1
+#endif
+
 /*
  * This returns the old value in the lock, so we succeeded
  * in getting the lock if the return value is 0.
  */
 static __inline__ unsigned long __spin_trylock(raw_spinlock_t *lock)
 {
-	unsigned long tmp, tmp2;
+	unsigned long tmp, token;
 
+	token = LOCK_TOKEN;
 	__asm__ __volatile__(
-"	lwz		%1,%3(13)		# __spin_trylock\n\
-1:	lwarx		%0,0,%2\n\
+"1:	lwarx		%0,0,%2		# __spin_trylock\n\
 	cmpwi		0,%0,0\n\
 	bne-		2f\n\
 	stwcx.		%1,0,%2\n\
 	bne-		1b\n\
 	isync\n\
-2:"	: "=&r" (tmp), "=&r" (tmp2)
-	: "r" (&lock->slock), "i" (offsetof(struct paca_struct, lock_token))
+2:"	: "=&r" (tmp)
+	: "r" (token), "r" (&lock->slock)
 	: "cr0", "memory");
 
 	return tmp;
@@ -113,11 +123,17 @@ static void __inline__ __raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long
 
 static __inline__ void __raw_spin_unlock(raw_spinlock_t *lock)
 {
-	__asm__ __volatile__("lwsync	# __raw_spin_unlock": : :"memory");
+	__asm__ __volatile__(SYNC_ON_SMP"	# __raw_spin_unlock"
+			     : : :"memory");
 	lock->slock = 0;
 }
 
+#ifdef CONFIG_PPC64
 extern void __raw_spin_unlock_wait(raw_spinlock_t *lock);
+#else
+#define __raw_spin_unlock_wait(lock) \
+	do { while (__raw_spin_is_locked(lock)) cpu_relax(); } while (0)
+#endif
 
 /*
  * Read-write spinlocks, allowing multiple readers
@@ -133,6 +149,14 @@ extern void __raw_spin_unlock_wait(raw_spinlock_t *lock);
 #define __raw_read_can_lock(rw)		((rw)->lock >= 0)
 #define __raw_write_can_lock(rw)	(!(rw)->lock)
 
+#ifdef CONFIG_PPC64
+#define __DO_SIGN_EXTEND	"extsw	%0,%0\n"
+#define WRLOCK_TOKEN		LOCK_TOKEN	/* it's negative */
+#else
+#define __DO_SIGN_EXTEND
+#define WRLOCK_TOKEN		(-1)
+#endif
+
 /*
  * This returns the old value in the lock + 1,
  * so we got a read lock if the return value is > 0.
@@ -142,11 +166,12 @@ static long __inline__ __read_trylock(raw_rwlock_t *rw)
 	long tmp;
 
 	__asm__ __volatile__(
-"1:	lwarx		%0,0,%1		# read_trylock\n\
-	extsw		%0,%0\n\
-	addic.		%0,%0,1\n\
-	ble-		2f\n\
-	stwcx.		%0,0,%1\n\
+"1:	lwarx		%0,0,%1		# read_trylock\n"
+	__DO_SIGN_EXTEND
+"	addic.		%0,%0,1\n\
+	ble-		2f\n"
+	PPC405_ERR77(0,%1)
+"	stwcx.		%0,0,%1\n\
 	bne-		1b\n\
 	isync\n\
 2:"	: "=&r" (tmp)
@@ -162,18 +187,19 @@ static long __inline__ __read_trylock(raw_rwlock_t *rw)
  */
 static __inline__ long __write_trylock(raw_rwlock_t *rw)
 {
-	long tmp, tmp2;
+	long tmp, token;
 
+	token = WRLOCK_TOKEN;
 	__asm__ __volatile__(
-"	lwz		%1,%3(13)	# write_trylock\n\
-1:	lwarx		%0,0,%2\n\
+"1:	lwarx		%0,0,%2	# write_trylock\n\
 	cmpwi		0,%0,0\n\
-	bne-		2f\n\
-	stwcx.		%1,0,%2\n\
+	bne-		2f\n"
+	PPC405_ERR77(0,%1)
+"	stwcx.		%1,0,%2\n\
 	bne-		1b\n\
 	isync\n\
-2:"	: "=&r" (tmp), "=&r" (tmp2)
-	: "r" (&rw->lock), "i" (offsetof(struct paca_struct, lock_token))
+2:"	: "=&r" (tmp)
+	: "r" (token), "r" (&rw->lock)
 	: "cr0", "memory");
 
 	return tmp;
@@ -224,8 +250,9 @@ static void __inline__ __raw_read_unlock(raw_rwlock_t *rw)
 	__asm__ __volatile__(
 	"eieio				# read_unlock\n\
 1:	lwarx		%0,0,%1\n\
-	addic		%0,%0,-1\n\
-	stwcx.		%0,0,%1\n\
+	addic		%0,%0,-1\n"
+	PPC405_ERR77(0,%1)
+"	stwcx.		%0,0,%1\n\
 	bne-		1b"
 	: "=&r"(tmp)
 	: "r"(&rw->lock)
@@ -234,7 +261,8 @@ static void __inline__ __raw_read_unlock(raw_rwlock_t *rw)
 
 static __inline__ void __raw_write_unlock(raw_rwlock_t *rw)
 {
-	__asm__ __volatile__("lwsync	# write_unlock": : :"memory");
+	__asm__ __volatile__(SYNC_ON_SMP"	# write_unlock"
+			     : : :"memory");
 	rw->lock = 0;
 }
 
diff --git a/include/asm-powerpc/topology.h b/include/asm-powerpc/topology.h
index 015d28746e1b..db8095cbe09b 100644
--- a/include/asm-powerpc/topology.h
+++ b/include/asm-powerpc/topology.h
@@ -41,6 +41,10 @@ static inline int node_to_first_cpu(int node)
 	.cache_hot_time		= (10*1000000),		\
 	.cache_nice_tries	= 1,			\
 	.per_cpu_gain		= 100,			\
+	.busy_idx		= 3,			\
+	.idle_idx		= 1,			\
+	.newidle_idx		= 2,			\
+	.wake_idx		= 1,			\
 	.flags			= SD_LOAD_BALANCE	\
 				| SD_BALANCE_EXEC	\
 				| SD_BALANCE_NEWIDLE	\
diff --git a/include/asm-ppc/cpm2.h b/include/asm-ppc/cpm2.h
index 43d2ebbc7748..b638b87cebe3 100644
--- a/include/asm-ppc/cpm2.h
+++ b/include/asm-ppc/cpm2.h
@@ -1091,5 +1091,7 @@ typedef struct im_idma {
 #define CPM_IMMR_OFFSET	0x101a8
 #endif
 
+#define FCC_PSMR_RMII	((uint)0x00020000)	/* Use RMII interface */
+
 #endif /* __CPM2__ */
 #endif /* __KERNEL__ */
diff --git a/include/asm-ppc/io.h b/include/asm-ppc/io.h
index 2bfdf9c98459..84ac6e258eef 100644
--- a/include/asm-ppc/io.h
+++ b/include/asm-ppc/io.h
@@ -545,6 +545,23 @@ extern void pci_iounmap(struct pci_dev *dev, void __iomem *);
 #include <asm/mpc8260_pci9.h>
 #endif
 
+#ifdef CONFIG_NOT_COHERENT_CACHE
+
+#define dma_cache_inv(_start,_size) \
+	invalidate_dcache_range(_start, (_start + _size))
+#define dma_cache_wback(_start,_size) \
+	clean_dcache_range(_start, (_start + _size))
+#define dma_cache_wback_inv(_start,_size) \
+	flush_dcache_range(_start, (_start + _size))
+
+#else
+
+#define dma_cache_inv(_start,_size)		do { } while (0)
+#define dma_cache_wback(_start,_size)		do { } while (0)
+#define dma_cache_wback_inv(_start,_size)	do { } while (0)
+
+#endif
+
 /*
  * Convert a physical pointer to a virtual kernel pointer for /dev/mem
  * access
diff --git a/include/asm-ppc64/dma-mapping.h b/include/asm-ppc64/dma-mapping.h
deleted file mode 100644
index fb68fa23bea8..000000000000
--- a/include/asm-ppc64/dma-mapping.h
+++ /dev/null
@@ -1,136 +0,0 @@
-/* Copyright (C) 2004 IBM
- *
- * Implements the generic device dma API for ppc64. Handles
- * the pci and vio busses
- */
-
-#ifndef _ASM_DMA_MAPPING_H
-#define _ASM_DMA_MAPPING_H
-
-#include <linux/types.h>
-#include <linux/cache.h>
-/* need struct page definitions */
-#include <linux/mm.h>
-#include <asm/scatterlist.h>
-#include <asm/bug.h>
-
-#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 nhwentries, enum dma_data_direction 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)
-{
-	BUG_ON(direction == DMA_NONE);
-	/* nothing to do */
-}
-
-static inline void
-dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle, size_t size,
-			   enum dma_data_direction direction)
-{
-	BUG_ON(direction == DMA_NONE);
-	/* nothing to do */
-}
-
-static inline void
-dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nelems,
-		    enum dma_data_direction direction)
-{
-	BUG_ON(direction == DMA_NONE);
-	/* nothing to do */
-}
-
-static inline void
-dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nelems,
-		       enum dma_data_direction direction)
-{
-	BUG_ON(direction == DMA_NONE);
-	/* nothing to do */
-}
-
-static inline int dma_mapping_error(dma_addr_t dma_addr)
-{
-	return (dma_addr == DMA_ERROR_CODE);
-}
-
-/* Now for the API extensions over the pci_ one */
-
-#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)	(1)
-
-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 << L1_CACHE_SHIFT_MAX);
-}
-
-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)
-{
-	BUG_ON(direction == DMA_NONE);
-	/* nothing to do */
-}
-
-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)
-{
-	BUG_ON(direction == DMA_NONE);
-	/* nothing to do */
-}
-
-static inline void
-dma_cache_sync(void *vaddr, size_t size,
-	       enum dma_data_direction direction)
-{
-	BUG_ON(direction == DMA_NONE);
-	/* nothing to do */
-}
-
-/*
- * DMA operations are abstracted for G5 vs. i/pSeries, PCI vs. VIO
- */
-struct dma_mapping_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 *vaddr, dma_addr_t dma_handle);
-	dma_addr_t	(*map_single)(struct device *dev, void *ptr,
-				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 nents, enum dma_data_direction direction);
-	int		(*dma_supported)(struct device *dev, u64 mask);
-	int		(*dac_dma_supported)(struct device *dev, u64 mask);
-};
-
-#endif	/* _ASM_DMA_MAPPING_H */
diff --git a/include/asm-ppc64/imalloc.h b/include/asm-ppc64/imalloc.h
deleted file mode 100644
index 42adf7033a81..000000000000
--- a/include/asm-ppc64/imalloc.h
+++ /dev/null
@@ -1,26 +0,0 @@
-#ifndef _PPC64_IMALLOC_H
-#define _PPC64_IMALLOC_H
-
-/*
- * Define the address range of the imalloc VM area.
- */
-#define PHBS_IO_BASE  	  VMALLOC_END
-#define IMALLOC_BASE      (PHBS_IO_BASE + 0x80000000ul)	/* Reserve 2 gigs for PHBs */
-#define IMALLOC_END       (VMALLOC_START + PGTABLE_RANGE)
-
-
-/* imalloc region types */
-#define IM_REGION_UNUSED	0x1
-#define IM_REGION_SUBSET	0x2
-#define IM_REGION_EXISTS	0x4
-#define IM_REGION_OVERLAP	0x8
-#define IM_REGION_SUPERSET	0x10
-
-extern struct vm_struct * im_get_free_area(unsigned long size);
-extern struct vm_struct * im_get_area(unsigned long v_addr, unsigned long size,
-				      int region_type);
-extern void im_free(void *addr);
-
-extern unsigned long ioremap_bot;
-
-#endif /* _PPC64_IMALLOC_H */
diff --git a/include/asm-ppc64/ptrace-common.h b/include/asm-ppc64/ptrace-common.h
deleted file mode 100644
index b1babb729673..000000000000
--- a/include/asm-ppc64/ptrace-common.h
+++ /dev/null
@@ -1,164 +0,0 @@
-/*
- *  linux/arch/ppc64/kernel/ptrace-common.h
- *
- *    Copyright (c) 2002 Stephen Rothwell, IBM Coproration
- *    Extracted from ptrace.c and ptrace32.c
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file README.legal in the main directory of
- * this archive for more details.
- */
-
-#ifndef _PPC64_PTRACE_COMMON_H
-#define _PPC64_PTRACE_COMMON_H
-
-#include <linux/config.h>
-#include <asm/system.h>
-
-/*
- * Set of msr bits that gdb can change on behalf of a process.
- */
-#define MSR_DEBUGCHANGE	(MSR_FE0 | MSR_SE | MSR_BE | MSR_FE1)
-
-/*
- * Get contents of register REGNO in task TASK.
- */
-static inline unsigned long get_reg(struct task_struct *task, int regno)
-{
-	unsigned long tmp = 0;
-
-	/*
-	 * Put the correct FP bits in, they might be wrong as a result
-	 * of our lazy FP restore.
-	 */
-	if (regno == PT_MSR) {
-		tmp = ((unsigned long *)task->thread.regs)[PT_MSR];
-		tmp |= task->thread.fpexc_mode;
-	} else if (regno < (sizeof(struct pt_regs) / sizeof(unsigned long))) {
-		tmp = ((unsigned long *)task->thread.regs)[regno];
-	}
-
-	return tmp;
-}
-
-/*
- * Write contents of register REGNO in task TASK.
- */
-static inline int put_reg(struct task_struct *task, int regno,
-			  unsigned long data)
-{
-	if (regno < PT_SOFTE) {
-		if (regno == PT_MSR)
-			data = (data & MSR_DEBUGCHANGE)
-				| (task->thread.regs->msr & ~MSR_DEBUGCHANGE);
-		((unsigned long *)task->thread.regs)[regno] = data;
-		return 0;
-	}
-	return -EIO;
-}
-
-static inline void set_single_step(struct task_struct *task)
-{
-	struct pt_regs *regs = task->thread.regs;
-	if (regs != NULL)
-		regs->msr |= MSR_SE;
-	set_ti_thread_flag(task->thread_info, TIF_SINGLESTEP);
-}
-
-static inline void clear_single_step(struct task_struct *task)
-{
-	struct pt_regs *regs = task->thread.regs;
-	if (regs != NULL)
-		regs->msr &= ~MSR_SE;
-	clear_ti_thread_flag(task->thread_info, TIF_SINGLESTEP);
-}
-
-#ifdef CONFIG_ALTIVEC
-/*
- * Get/set all the altivec registers vr0..vr31, vscr, vrsave, in one go.
- * The transfer totals 34 quadword.  Quadwords 0-31 contain the
- * corresponding vector registers.  Quadword 32 contains the vscr as the
- * last word (offset 12) within that quadword.  Quadword 33 contains the
- * vrsave as the first word (offset 0) within the quadword.
- *
- * This definition of the VMX state is compatible with the current PPC32
- * ptrace interface.  This allows signal handling and ptrace to use the
- * same structures.  This also simplifies the implementation of a bi-arch
- * (combined (32- and 64-bit) gdb.
- */
-
-/*
- * Get contents of AltiVec register state in task TASK
- */
-static inline int get_vrregs(unsigned long __user *data,
-			     struct task_struct *task)
-{
-	unsigned long regsize;
-
-	/* copy AltiVec registers VR[0] .. VR[31] */
-	regsize = 32 * sizeof(vector128);
-	if (copy_to_user(data, task->thread.vr, regsize))
-		return -EFAULT;
-	data += (regsize / sizeof(unsigned long));
-
-	/* copy VSCR */
-	regsize = 1 * sizeof(vector128);
-	if (copy_to_user(data, &task->thread.vscr, regsize))
-		return -EFAULT;
-	data += (regsize / sizeof(unsigned long));
-
-	/* copy VRSAVE */
-	if (put_user(task->thread.vrsave, (u32 __user *)data))
-		return -EFAULT;
-
-	return 0;
-}
-
-/*
- * Write contents of AltiVec register state into task TASK.
- */
-static inline int set_vrregs(struct task_struct *task,
-			     unsigned long __user *data)
-{
-	unsigned long regsize;
-
-	/* copy AltiVec registers VR[0] .. VR[31] */
-	regsize = 32 * sizeof(vector128);
-	if (copy_from_user(task->thread.vr, data, regsize))
-		return -EFAULT;
-	data += (regsize / sizeof(unsigned long));
-
-	/* copy VSCR */
-	regsize = 1 * sizeof(vector128);
-	if (copy_from_user(&task->thread.vscr, data, regsize))
-		return -EFAULT;
-	data += (regsize / sizeof(unsigned long));
-
-	/* copy VRSAVE */
-	if (get_user(task->thread.vrsave, (u32 __user *)data))
-		return -EFAULT;
-
-	return 0;
-}
-#endif
-
-static inline int ptrace_set_debugreg(struct task_struct *task,
-				      unsigned long addr, unsigned long data)
-{
-	/* We only support one DABR and no IABRS at the moment */
-	if (addr > 0)
-		return -EINVAL;
-
-	/* The bottom 3 bits are flags */
-	if ((data & ~0x7UL) >= TASK_SIZE)
-		return -EIO;
-
-	/* Ensure translation is on */
-	if (data && !(data & DABR_TRANSLATION))
-		return -EIO;
-
-	task->thread.dabr = data;
-	return 0;
-}
-
-#endif /* _PPC64_PTRACE_COMMON_H */
diff --git a/include/linux/cciss_ioctl.h b/include/linux/cciss_ioctl.h
index 424d5e622b43..6e27f42e3a57 100644
--- a/include/linux/cciss_ioctl.h
+++ b/include/linux/cciss_ioctl.h
@@ -10,8 +10,8 @@
 typedef struct _cciss_pci_info_struct
 {
 	unsigned char 	bus;
-	unsigned short	domain;
 	unsigned char 	dev_fn;
+	unsigned short	domain;
 	__u32 		board_id;
 } cciss_pci_info_struct; 
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index ac8b25fa6506..e99019057ba6 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1089,9 +1089,11 @@ enum {
 
 /*
  * Subdrivers support.
+ *
+ * The gendriver.owner field should be set to the module owner of this driver.
+ * The gendriver.name field should be set to the name of this driver
  */
 typedef struct ide_driver_s {
-	struct module			*owner;
 	const char			*version;
 	u8				media;
 	unsigned supports_dsc_overlap	: 1;
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 1013a42d10b1..0986d19be0b7 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -940,7 +940,9 @@ unsigned long max_sane_readahead(unsigned long nr);
 
 /* Do stack extension */
 extern int expand_stack(struct vm_area_struct *vma, unsigned long address);
+#ifdef CONFIG_IA64
 extern int expand_upwards(struct vm_area_struct *vma, unsigned long address);
+#endif
 
 /* Look up the first VMA which satisfies  addr < vm_end,  NULL if none. */
 extern struct vm_area_struct * find_vma(struct mm_struct * mm, unsigned long addr);
diff --git a/include/net/ieee80211.h b/include/net/ieee80211.h
index b93fd8c1d884..cde2f4f4f501 100644
--- a/include/net/ieee80211.h
+++ b/include/net/ieee80211.h
@@ -1042,7 +1042,7 @@ static inline u8 *ieee80211_get_payload(struct ieee80211_hdr *hdr)
 	case IEEE80211_4ADDR_LEN:
 		return ((struct ieee80211_hdr_4addr *)hdr)->payload;
 	}
-
+	return NULL;
 }
 
 static inline int ieee80211_is_ofdm_rate(u8 rate)