Merge branch 'next' into for-linus

36 files changed, 245 insertions, 159 deletions

diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
index dd4698c67cc..baf638487a2 100644
--- a/arch/arm/mm/Kconfig
+++ b/arch/arm/mm/Kconfig
@@ -1,9 +1,5 @@
 comment "Processor Type"
 
-config CPU_32
-	bool
-	default y
-
 # Select CPU types depending on the architecture selected.  This selects
 # which CPUs we support in the kernel image, and the compiler instruction
 # optimiser behaviour.
diff --git a/arch/arm/mm/Makefile b/arch/arm/mm/Makefile
index 827e238e5d4..e8d34a80851 100644
--- a/arch/arm/mm/Makefile
+++ b/arch/arm/mm/Makefile
@@ -27,6 +27,9 @@ obj-$(CONFIG_CPU_ABRT_EV5TJ)	+= abort-ev5tj.o
 obj-$(CONFIG_CPU_ABRT_EV6)	+= abort-ev6.o
 obj-$(CONFIG_CPU_ABRT_EV7)	+= abort-ev7.o
 
+AFLAGS_abort-ev6.o	:=-Wa,-march=armv6k
+AFLAGS_abort-ev7.o	:=-Wa,-march=armv7-a
+
 obj-$(CONFIG_CPU_PABRT_LEGACY)	+= pabort-legacy.o
 obj-$(CONFIG_CPU_PABRT_V6)	+= pabort-v6.o
 obj-$(CONFIG_CPU_PABRT_V7)	+= pabort-v7.o
@@ -39,6 +42,9 @@ obj-$(CONFIG_CPU_CACHE_V6)	+= cache-v6.o
 obj-$(CONFIG_CPU_CACHE_V7)	+= cache-v7.o
 obj-$(CONFIG_CPU_CACHE_FA)	+= cache-fa.o
 
+AFLAGS_cache-v6.o	:=-Wa,-march=armv6
+AFLAGS_cache-v7.o	:=-Wa,-march=armv7-a
+
 obj-$(CONFIG_CPU_COPY_V3)	+= copypage-v3.o
 obj-$(CONFIG_CPU_COPY_V4WT)	+= copypage-v4wt.o
 obj-$(CONFIG_CPU_COPY_V4WB)	+= copypage-v4wb.o
@@ -58,6 +64,9 @@ obj-$(CONFIG_CPU_TLB_V6)	+= tlb-v6.o
 obj-$(CONFIG_CPU_TLB_V7)	+= tlb-v7.o
 obj-$(CONFIG_CPU_TLB_FA)	+= tlb-fa.o
 
+AFLAGS_tlb-v6.o		:=-Wa,-march=armv6
+AFLAGS_tlb-v7.o		:=-Wa,-march=armv7-a
+
 obj-$(CONFIG_CPU_ARM610)	+= proc-arm6_7.o
 obj-$(CONFIG_CPU_ARM710)	+= proc-arm6_7.o
 obj-$(CONFIG_CPU_ARM7TDMI)	+= proc-arm7tdmi.o
@@ -84,6 +93,9 @@ obj-$(CONFIG_CPU_FEROCEON)	+= proc-feroceon.o
 obj-$(CONFIG_CPU_V6)		+= proc-v6.o
 obj-$(CONFIG_CPU_V7)		+= proc-v7.o
 
+AFLAGS_proc-v6.o	:=-Wa,-march=armv6
+AFLAGS_proc-v7.o	:=-Wa,-march=armv7-a
+
 obj-$(CONFIG_CACHE_FEROCEON_L2)	+= cache-feroceon-l2.o
 obj-$(CONFIG_CACHE_L2X0)	+= cache-l2x0.o
 obj-$(CONFIG_CACHE_XSC3L2)	+= cache-xsc3l2.o
diff --git a/arch/arm/mm/cache-fa.S b/arch/arm/mm/cache-fa.S
index b63a8f7b95c..a89444a3c01 100644
--- a/arch/arm/mm/cache-fa.S
+++ b/arch/arm/mm/cache-fa.S
@@ -127,15 +127,16 @@ ENTRY(fa_coherent_user_range)
 	mov	pc, lr
 
 /*
- *	flush_kern_dcache_page(kaddr)
+ *	flush_kern_dcache_area(void *addr, size_t size)
  *
  *	Ensure that the data held in the page kaddr is written back
  *	to the page in question.
  *
- *	- kaddr   - kernel address (guaranteed to be page aligned)
+ *	- addr	- kernel address
+ *	- size	- size of region
  */
-ENTRY(fa_flush_kern_dcache_page)
-	add	r1, r0, #PAGE_SZ
+ENTRY(fa_flush_kern_dcache_area)
+	add	r1, r0, r1
 1:	mcr	p15, 0, r0, c7, c14, 1		@ clean & invalidate D line
 	add	r0, r0, #CACHE_DLINESIZE
 	cmp	r0, r1
@@ -213,7 +214,7 @@ ENTRY(fa_cache_fns)
 	.long	fa_flush_user_cache_range
 	.long	fa_coherent_kern_range
 	.long	fa_coherent_user_range
-	.long	fa_flush_kern_dcache_page
+	.long	fa_flush_kern_dcache_area
 	.long	fa_dma_inv_range
 	.long	fa_dma_clean_range
 	.long	fa_dma_flush_range
diff --git a/arch/arm/mm/cache-l2x0.c b/arch/arm/mm/cache-l2x0.c
index 747f9a9021b..cb8fc6573b1 100644
--- a/arch/arm/mm/cache-l2x0.c
+++ b/arch/arm/mm/cache-l2x0.c
@@ -28,69 +28,120 @@
 static void __iomem *l2x0_base;
 static DEFINE_SPINLOCK(l2x0_lock);
 
-static inline void sync_writel(unsigned long val, unsigned long reg,
-			       unsigned long complete_mask)
+static inline void cache_wait(void __iomem *reg, unsigned long mask)
 {
-	unsigned long flags;
-
-	spin_lock_irqsave(&l2x0_lock, flags);
-	writel(val, l2x0_base + reg);
 	/* wait for the operation to complete */
-	while (readl(l2x0_base + reg) & complete_mask)
+	while (readl(reg) & mask)
 		;
-	spin_unlock_irqrestore(&l2x0_lock, flags);
 }
 
 static inline void cache_sync(void)
 {
-	sync_writel(0, L2X0_CACHE_SYNC, 1);
+	void __iomem *base = l2x0_base;
+	writel(0, base + L2X0_CACHE_SYNC);
+	cache_wait(base + L2X0_CACHE_SYNC, 1);
 }
 
 static inline void l2x0_inv_all(void)
 {
+	unsigned long flags;
+
 	/* invalidate all ways */
-	sync_writel(0xff, L2X0_INV_WAY, 0xff);
+	spin_lock_irqsave(&l2x0_lock, flags);
+	writel(0xff, l2x0_base + L2X0_INV_WAY);
+	cache_wait(l2x0_base + L2X0_INV_WAY, 0xff);
 	cache_sync();
+	spin_unlock_irqrestore(&l2x0_lock, flags);
 }
 
 static void l2x0_inv_range(unsigned long start, unsigned long end)
 {
-	unsigned long addr;
+	void __iomem *base = l2x0_base;
+	unsigned long flags;
 
+	spin_lock_irqsave(&l2x0_lock, flags);
 	if (start & (CACHE_LINE_SIZE - 1)) {
 		start &= ~(CACHE_LINE_SIZE - 1);
-		sync_writel(start, L2X0_CLEAN_INV_LINE_PA, 1);
+		cache_wait(base + L2X0_CLEAN_INV_LINE_PA, 1);
+		writel(start, base + L2X0_CLEAN_INV_LINE_PA);
 		start += CACHE_LINE_SIZE;
 	}
 
 	if (end & (CACHE_LINE_SIZE - 1)) {
 		end &= ~(CACHE_LINE_SIZE - 1);
-		sync_writel(end, L2X0_CLEAN_INV_LINE_PA, 1);
+		cache_wait(base + L2X0_CLEAN_INV_LINE_PA, 1);
+		writel(end, base + L2X0_CLEAN_INV_LINE_PA);
 	}
 
-	for (addr = start; addr < end; addr += CACHE_LINE_SIZE)
-		sync_writel(addr, L2X0_INV_LINE_PA, 1);
+	while (start < end) {
+		unsigned long blk_end = start + min(end - start, 4096UL);
+
+		while (start < blk_end) {
+			cache_wait(base + L2X0_INV_LINE_PA, 1);
+			writel(start, base + L2X0_INV_LINE_PA);
+			start += CACHE_LINE_SIZE;
+		}
+
+		if (blk_end < end) {
+			spin_unlock_irqrestore(&l2x0_lock, flags);
+			spin_lock_irqsave(&l2x0_lock, flags);
+		}
+	}
+	cache_wait(base + L2X0_INV_LINE_PA, 1);
 	cache_sync();
+	spin_unlock_irqrestore(&l2x0_lock, flags);
 }
 
 static void l2x0_clean_range(unsigned long start, unsigned long end)
 {
-	unsigned long addr;
+	void __iomem *base = l2x0_base;
+	unsigned long flags;
 
+	spin_lock_irqsave(&l2x0_lock, flags);
 	start &= ~(CACHE_LINE_SIZE - 1);
-	for (addr = start; addr < end; addr += CACHE_LINE_SIZE)
-		sync_writel(addr, L2X0_CLEAN_LINE_PA, 1);
+	while (start < end) {
+		unsigned long blk_end = start + min(end - start, 4096UL);
+
+		while (start < blk_end) {
+			cache_wait(base + L2X0_CLEAN_LINE_PA, 1);
+			writel(start, base + L2X0_CLEAN_LINE_PA);
+			start += CACHE_LINE_SIZE;
+		}
+
+		if (blk_end < end) {
+			spin_unlock_irqrestore(&l2x0_lock, flags);
+			spin_lock_irqsave(&l2x0_lock, flags);
+		}
+	}
+	cache_wait(base + L2X0_CLEAN_LINE_PA, 1);
 	cache_sync();
+	spin_unlock_irqrestore(&l2x0_lock, flags);
 }
 
 static void l2x0_flush_range(unsigned long start, unsigned long end)
 {
-	unsigned long addr;
+	void __iomem *base = l2x0_base;
+	unsigned long flags;
 
+	spin_lock_irqsave(&l2x0_lock, flags);
 	start &= ~(CACHE_LINE_SIZE - 1);
-	for (addr = start; addr < end; addr += CACHE_LINE_SIZE)
-		sync_writel(addr, L2X0_CLEAN_INV_LINE_PA, 1);
+	while (start < end) {
+		unsigned long blk_end = start + min(end - start, 4096UL);
+
+		while (start < blk_end) {
+			cache_wait(base + L2X0_CLEAN_INV_LINE_PA, 1);
+			writel(start, base + L2X0_CLEAN_INV_LINE_PA);
+			start += CACHE_LINE_SIZE;
+		}
+
+		if (blk_end < end) {
+			spin_unlock_irqrestore(&l2x0_lock, flags);
+			spin_lock_irqsave(&l2x0_lock, flags);
+		}
+	}
+	cache_wait(base + L2X0_CLEAN_INV_LINE_PA, 1);
 	cache_sync();
+	spin_unlock_irqrestore(&l2x0_lock, flags);
 }
 
 void __init l2x0_init(void __iomem *base, __u32 aux_val, __u32 aux_mask)
diff --git a/arch/arm/mm/cache-v3.S b/arch/arm/mm/cache-v3.S
index 8a4abebc478..2a482731ea3 100644
--- a/arch/arm/mm/cache-v3.S
+++ b/arch/arm/mm/cache-v3.S
@@ -72,14 +72,15 @@ ENTRY(v3_coherent_user_range)
 	mov	pc, lr
 
 /*
- *	flush_kern_dcache_page(void *page)
+ *	flush_kern_dcache_area(void *page, size_t size)
  *
  *	Ensure no D cache aliasing occurs, either with itself or
  *	the I cache
  *
- *	- addr	- page aligned address
+ *	- addr	- kernel address
+ *	- size	- region size
  */
-ENTRY(v3_flush_kern_dcache_page)
+ENTRY(v3_flush_kern_dcache_area)
 	/* FALLTHROUGH */
 
 /*
@@ -129,7 +130,7 @@ ENTRY(v3_cache_fns)
 	.long	v3_flush_user_cache_range
 	.long	v3_coherent_kern_range
 	.long	v3_coherent_user_range
-	.long	v3_flush_kern_dcache_page
+	.long	v3_flush_kern_dcache_area
 	.long	v3_dma_inv_range
 	.long	v3_dma_clean_range
 	.long	v3_dma_flush_range
diff --git a/arch/arm/mm/cache-v4.S b/arch/arm/mm/cache-v4.S
index 3668611cb40..5c7da3e372e 100644
--- a/arch/arm/mm/cache-v4.S
+++ b/arch/arm/mm/cache-v4.S
@@ -82,14 +82,15 @@ ENTRY(v4_coherent_user_range)
 	mov	pc, lr
 
 /*
- *	flush_kern_dcache_page(void *page)
+ *	flush_kern_dcache_area(void *addr, size_t size)
  *
  *	Ensure no D cache aliasing occurs, either with itself or
  *	the I cache
  *
- *	- addr	- page aligned address
+ *	- addr	- kernel address
+ *	- size	- region size
  */
-ENTRY(v4_flush_kern_dcache_page)
+ENTRY(v4_flush_kern_dcache_area)
 	/* FALLTHROUGH */
 
 /*
@@ -141,7 +142,7 @@ ENTRY(v4_cache_fns)
 	.long	v4_flush_user_cache_range
 	.long	v4_coherent_kern_range
 	.long	v4_coherent_user_range
-	.long	v4_flush_kern_dcache_page
+	.long	v4_flush_kern_dcache_area
 	.long	v4_dma_inv_range
 	.long	v4_dma_clean_range
 	.long	v4_dma_flush_range
diff --git a/arch/arm/mm/cache-v4wb.S b/arch/arm/mm/cache-v4wb.S
index 2ebc1b3bf85..3dbedf1ec0e 100644
--- a/arch/arm/mm/cache-v4wb.S
+++ b/arch/arm/mm/cache-v4wb.S
@@ -114,15 +114,16 @@ ENTRY(v4wb_flush_user_cache_range)
 	mov	pc, lr
 
 /*
- *	flush_kern_dcache_page(void *page)
+ *	flush_kern_dcache_area(void *addr, size_t size)
  *
  *	Ensure no D cache aliasing occurs, either with itself or
  *	the I cache
  *
- *	- addr	- page aligned address
+ *	- addr	- kernel address
+ *	- size	- region size
  */
-ENTRY(v4wb_flush_kern_dcache_page)
-	add	r1, r0, #PAGE_SZ
+ENTRY(v4wb_flush_kern_dcache_area)
+	add	r1, r0, r1
 	/* fall through */
 
 /*
@@ -224,7 +225,7 @@ ENTRY(v4wb_cache_fns)
 	.long	v4wb_flush_user_cache_range
 	.long	v4wb_coherent_kern_range
 	.long	v4wb_coherent_user_range
-	.long	v4wb_flush_kern_dcache_page
+	.long	v4wb_flush_kern_dcache_area
 	.long	v4wb_dma_inv_range
 	.long	v4wb_dma_clean_range
 	.long	v4wb_dma_flush_range
diff --git a/arch/arm/mm/cache-v4wt.S b/arch/arm/mm/cache-v4wt.S
index c54fa2cc40e..b3b7410270b 100644
--- a/arch/arm/mm/cache-v4wt.S
+++ b/arch/arm/mm/cache-v4wt.S
@@ -117,17 +117,18 @@ ENTRY(v4wt_coherent_user_range)
 	mov	pc, lr
 
 /*
- *	flush_kern_dcache_page(void *page)
+ *	flush_kern_dcache_area(void *addr, size_t size)
  *
  *	Ensure no D cache aliasing occurs, either with itself or
  *	the I cache
  *
- *	- addr	- page aligned address
+ *	- addr	- kernel address
+ *	- size	- region size
  */
-ENTRY(v4wt_flush_kern_dcache_page)
+ENTRY(v4wt_flush_kern_dcache_area)
 	mov	r2, #0
 	mcr	p15, 0, r2, c7, c5, 0		@ invalidate I cache
-	add	r1, r0, #PAGE_SZ
+	add	r1, r0, r1
 	/* fallthrough */
 
 /*
@@ -180,7 +181,7 @@ ENTRY(v4wt_cache_fns)
 	.long	v4wt_flush_user_cache_range
 	.long	v4wt_coherent_kern_range
 	.long	v4wt_coherent_user_range
-	.long	v4wt_flush_kern_dcache_page
+	.long	v4wt_flush_kern_dcache_area
 	.long	v4wt_dma_inv_range
 	.long	v4wt_dma_clean_range
 	.long	v4wt_dma_flush_range
diff --git a/arch/arm/mm/cache-v6.S b/arch/arm/mm/cache-v6.S
index 295e25dd638..4ba0a24ce6f 100644
--- a/arch/arm/mm/cache-v6.S
+++ b/arch/arm/mm/cache-v6.S
@@ -159,15 +159,16 @@ ENDPROC(v6_coherent_user_range)
 ENDPROC(v6_coherent_kern_range)
 
 /*
- *	v6_flush_kern_dcache_page(kaddr)
+ *	v6_flush_kern_dcache_area(void *addr, size_t size)
  *
  *	Ensure that the data held in the page kaddr is written back
  *	to the page in question.
  *
- *	- kaddr   - kernel address (guaranteed to be page aligned)
+ *	- addr	- kernel address
+ *	- size	- region size
  */
-ENTRY(v6_flush_kern_dcache_page)
-	add	r1, r0, #PAGE_SZ
+ENTRY(v6_flush_kern_dcache_area)
+	add	r1, r0, r1
 1:
 #ifdef HARVARD_CACHE
 	mcr	p15, 0, r0, c7, c14, 1		@ clean & invalidate D line
@@ -271,7 +272,7 @@ ENTRY(v6_cache_fns)
 	.long	v6_flush_user_cache_range
 	.long	v6_coherent_kern_range
 	.long	v6_coherent_user_range
-	.long	v6_flush_kern_dcache_page
+	.long	v6_flush_kern_dcache_area
 	.long	v6_dma_inv_range
 	.long	v6_dma_clean_range
 	.long	v6_dma_flush_range
diff --git a/arch/arm/mm/cache-v7.S b/arch/arm/mm/cache-v7.S
index e1bd9759617..9073db849fb 100644
--- a/arch/arm/mm/cache-v7.S
+++ b/arch/arm/mm/cache-v7.S
@@ -186,16 +186,17 @@ ENDPROC(v7_coherent_kern_range)
 ENDPROC(v7_coherent_user_range)
 
 /*
- *	v7_flush_kern_dcache_page(kaddr)
+ *	v7_flush_kern_dcache_area(void *addr, size_t size)
  *
  *	Ensure that the data held in the page kaddr is written back
  *	to the page in question.
  *
- *	- kaddr   - kernel address (guaranteed to be page aligned)
+ *	- addr	- kernel address
+ *	- size	- region size
  */
-ENTRY(v7_flush_kern_dcache_page)
+ENTRY(v7_flush_kern_dcache_area)
 	dcache_line_size r2, r3
-	add	r1, r0, #PAGE_SZ
+	add	r1, r0, r1
 1:
 	mcr	p15, 0, r0, c7, c14, 1		@ clean & invalidate D line / unified line
 	add	r0, r0, r2
@@ -203,7 +204,7 @@ ENTRY(v7_flush_kern_dcache_page)
 	blo	1b
 	dsb
 	mov	pc, lr
-ENDPROC(v7_flush_kern_dcache_page)
+ENDPROC(v7_flush_kern_dcache_area)
 
 /*
  *	v7_dma_inv_range(start,end)
@@ -279,7 +280,7 @@ ENTRY(v7_cache_fns)
 	.long	v7_flush_user_cache_range
 	.long	v7_coherent_kern_range
 	.long	v7_coherent_user_range
-	.long	v7_flush_kern_dcache_page
+	.long	v7_flush_kern_dcache_area
 	.long	v7_dma_inv_range
 	.long	v7_dma_clean_range
 	.long	v7_dma_flush_range
diff --git a/arch/arm/mm/cache-xsc3l2.c b/arch/arm/mm/cache-xsc3l2.c
index 5d180cb0bd9..c3154928bcc 100644
--- a/arch/arm/mm/cache-xsc3l2.c
+++ b/arch/arm/mm/cache-xsc3l2.c
@@ -221,15 +221,14 @@ static int __init xsc3_l2_init(void)
 	if (!cpu_is_xsc3() || !xsc3_l2_present())
 		return 0;
 
-	if (!(get_cr() & CR_L2)) {
+	if (get_cr() & CR_L2) {
 		pr_info("XScale3 L2 cache enabled.\n");
-		adjust_cr(CR_L2, CR_L2);
 		xsc3_l2_inv_all();
-	}
 
-	outer_cache.inv_range = xsc3_l2_inv_range;
-	outer_cache.clean_range = xsc3_l2_clean_range;
-	outer_cache.flush_range = xsc3_l2_flush_range;
+		outer_cache.inv_range = xsc3_l2_inv_range;
+		outer_cache.clean_range = xsc3_l2_clean_range;
+		outer_cache.flush_range = xsc3_l2_flush_range;
+	}
 
 	return 0;
 }
diff --git a/arch/arm/mm/copypage-v6.c b/arch/arm/mm/copypage-v6.c
index 841f355319b..0fa1319273d 100644
--- a/arch/arm/mm/copypage-v6.c
+++ b/arch/arm/mm/copypage-v6.c
@@ -48,7 +48,7 @@ static void v6_copy_user_highpage_nonaliasing(struct page *to,
 	 */
 	if (page_address(to) != NULL)
 #endif
-		__cpuc_flush_dcache_page(kto);
+		__cpuc_flush_dcache_area(kto, PAGE_SIZE);
 	kunmap_atomic(kto, KM_USER1);
 	kunmap_atomic(kfrom, KM_USER0);
 }
diff --git a/arch/arm/mm/fault-armv.c b/arch/arm/mm/fault-armv.c
index 72960229195..56ee15321b0 100644
--- a/arch/arm/mm/fault-armv.c
+++ b/arch/arm/mm/fault-armv.c
@@ -207,9 +207,8 @@ void __init check_writebuffer_bugs(void)
 	page = alloc_page(GFP_KERNEL);
 	if (page) {
 		unsigned long *p1, *p2;
-		pgprot_t prot = __pgprot(L_PTE_PRESENT|L_PTE_YOUNG|
-					 L_PTE_DIRTY|L_PTE_WRITE|
-					 L_PTE_MT_BUFFERABLE);
+		pgprot_t prot = __pgprot_modify(PAGE_KERNEL,
+					L_PTE_MT_MASK, L_PTE_MT_BUFFERABLE);
 
 		p1 = vmap(&page, 1, VM_IOREMAP, prot);
 		p2 = vmap(&page, 1, VM_IOREMAP, prot);
diff --git a/arch/arm/mm/flush.c b/arch/arm/mm/flush.c
index 329594e760c..6f3a4b7a3b8 100644
--- a/arch/arm/mm/flush.c
+++ b/arch/arm/mm/flush.c
@@ -131,7 +131,7 @@ void __flush_dcache_page(struct address_space *mapping, struct page *page)
 	 */
 	if (addr)
 #endif
-		__cpuc_flush_dcache_page(addr);
+		__cpuc_flush_dcache_area(addr, PAGE_SIZE);
 
 	/*
 	 * If this is a page cache page, and we have an aliasing VIPT cache,
@@ -258,5 +258,5 @@ void __flush_anon_page(struct vm_area_struct *vma, struct page *page, unsigned l
 	 * in this mapping of the page.  FIXME: this is overkill
 	 * since we actually ask for a write-back and invalidate.
 	 */
-	__cpuc_flush_dcache_page(page_address(page));
+	__cpuc_flush_dcache_area(page_address(page), PAGE_SIZE);
 }
diff --git a/arch/arm/mm/highmem.c b/arch/arm/mm/highmem.c
index 30f82fb5918..2be1ec7c1b4 100644
--- a/arch/arm/mm/highmem.c
+++ b/arch/arm/mm/highmem.c
@@ -79,7 +79,7 @@ void kunmap_atomic(void *kvaddr, enum km_type type)
 	unsigned int idx = type + KM_TYPE_NR * smp_processor_id();
 
 	if (kvaddr >= (void *)FIXADDR_START) {
-		__cpuc_flush_dcache_page((void *)vaddr);
+		__cpuc_flush_dcache_area((void *)vaddr, PAGE_SIZE);
 #ifdef CONFIG_DEBUG_HIGHMEM
 		BUG_ON(vaddr != __fix_to_virt(FIX_KMAP_BEGIN + idx));
 		set_pte_ext(TOP_PTE(vaddr), __pte(0), 0);
diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c
index 52c40d15567..a04ffbbbe25 100644
--- a/arch/arm/mm/init.c
+++ b/arch/arm/mm/init.c
@@ -616,7 +616,7 @@ void __init mem_init(void)
 		"%dK data, %dK init, %luK highmem)\n",
 		nr_free_pages() << (PAGE_SHIFT-10), codesize >> 10,
 		datasize >> 10, initsize >> 10,
-		(unsigned long) (totalhigh_pages << (PAGE_SHIFT-10)));
+		totalhigh_pages << (PAGE_SHIFT-10));
 
 	if (PAGE_SIZE >= 16384 && num_physpages <= 128) {
 		extern int sysctl_overcommit_memory;
diff --git a/arch/arm/mm/mmu.c b/arch/arm/mm/mmu.c
index 8c7fbd19a4b..761ffede6a2 100644
--- a/arch/arm/mm/mmu.c
+++ b/arch/arm/mm/mmu.c
@@ -453,8 +453,7 @@ static void __init build_mem_type_table(void)
 
 	pgprot_user   = __pgprot(L_PTE_PRESENT | L_PTE_YOUNG | user_pgprot);
 	pgprot_kernel = __pgprot(L_PTE_PRESENT | L_PTE_YOUNG |
-				 L_PTE_DIRTY | L_PTE_WRITE |
-				 L_PTE_EXEC | kern_pgprot);
+				 L_PTE_DIRTY | L_PTE_WRITE | kern_pgprot);
 
 	mem_types[MT_LOW_VECTORS].prot_l1 |= ecc_mask;
 	mem_types[MT_HIGH_VECTORS].prot_l1 |= ecc_mask;
@@ -1068,4 +1067,6 @@ void setup_mm_for_reboot(char mode)
 		pmd[1] = __pmd(pmdval + (1 << (PGDIR_SHIFT - 1)));
 		flush_pmd_entry(pmd);
 	}
+
+	local_flush_tlb_all();
 }
diff --git a/arch/arm/mm/nommu.c b/arch/arm/mm/nommu.c
index 900811cc913..374a8311bc8 100644
--- a/arch/arm/mm/nommu.c
+++ b/arch/arm/mm/nommu.c
@@ -61,7 +61,7 @@ void setup_mm_for_reboot(char mode)
 
 void flush_dcache_page(struct page *page)
 {
-	__cpuc_flush_dcache_page(page_address(page));
+	__cpuc_flush_dcache_area(page_address(page), PAGE_SIZE);
 }
 EXPORT_SYMBOL(flush_dcache_page);
 
diff --git a/arch/arm/mm/proc-arm1020.S b/arch/arm/mm/proc-arm1020.S
index d9fb4b98c49..8012e24282b 100644
--- a/arch/arm/mm/proc-arm1020.S
+++ b/arch/arm/mm/proc-arm1020.S
@@ -231,17 +231,18 @@ ENTRY(arm1020_coherent_user_range)
 	mov	pc, lr
 
 /*
- *	flush_kern_dcache_page(void *page)
+ *	flush_kern_dcache_area(void *addr, size_t size)
  *
  *	Ensure no D cache aliasing occurs, either with itself or
  *	the I cache
  *
- *	- page	- page aligned address
+ *	- addr	- kernel address
+ *	- size	- region size
  */
-ENTRY(arm1020_flush_kern_dcache_page)
+ENTRY(arm1020_flush_kern_dcache_area)
 	mov	ip, #0
 #ifndef CONFIG_CPU_DCACHE_DISABLE
-	add	r1, r0, #PAGE_SZ
+	add	r1, r0, r1
 1:	mcr	p15, 0, r0, c7, c14, 1		@ clean+invalidate D entry
 	mcr	p15, 0, ip, c7, c10, 4		@ drain WB
 	add	r0, r0, #CACHE_DLINESIZE
@@ -335,7 +336,7 @@ ENTRY(arm1020_cache_fns)
 	.long	arm1020_flush_user_cache_range
 	.long	arm1020_coherent_kern_range
 	.long	arm1020_coherent_user_range
-	.long	arm1020_flush_kern_dcache_page
+	.long	arm1020_flush_kern_dcache_area
 	.long	arm1020_dma_inv_range
 	.long	arm1020_dma_clean_range
 	.long	arm1020_dma_flush_range
diff --git a/arch/arm/mm/proc-arm1020e.S b/arch/arm/mm/proc-arm1020e.S
index 7453b75dcea..41fe25d234f 100644
--- a/arch/arm/mm/proc-arm1020e.S
+++ b/arch/arm/mm/proc-arm1020e.S
@@ -225,17 +225,18 @@ ENTRY(arm1020e_coherent_user_range)
 	mov	pc, lr
 
 /*
- *	flush_kern_dcache_page(void *page)
+ *	flush_kern_dcache_area(void *addr, size_t size)
  *
  *	Ensure no D cache aliasing occurs, either with itself or
  *	the I cache
  *
- *	- page	- page aligned address
+ *	- addr	- kernel address
+ *	- size	- region size
  */
-ENTRY(arm1020e_flush_kern_dcache_page)
+ENTRY(arm1020e_flush_kern_dcache_area)
 	mov	ip, #0
 #ifndef CONFIG_CPU_DCACHE_DISABLE
-	add	r1, r0, #PAGE_SZ
+	add	r1, r0, r1
 1:	mcr	p15, 0, r0, c7, c14, 1		@ clean+invalidate D entry
 	add	r0, r0, #CACHE_DLINESIZE
 	cmp	r0, r1
@@ -321,7 +322,7 @@ ENTRY(arm1020e_cache_fns)
 	.long	arm1020e_flush_user_cache_range
 	.long	arm1020e_coherent_kern_range
 	.long	arm1020e_coherent_user_range
-	.long	arm1020e_flush_kern_dcache_page
+	.long	arm1020e_flush_kern_dcache_area
 	.long	arm1020e_dma_inv_range
 	.long	arm1020e_dma_clean_range
 	.long	arm1020e_dma_flush_range
diff --git a/arch/arm/mm/proc-arm1022.S b/arch/arm/mm/proc-arm1022.S
index 8eb72d75a8b..20a5b1b31a7 100644
--- a/arch/arm/mm/proc-arm1022.S
+++ b/arch/arm/mm/proc-arm1022.S
@@ -214,17 +214,18 @@ ENTRY(arm1022_coherent_user_range)
 	mov	pc, lr
 
 /*
- *	flush_kern_dcache_page(void *page)
+ *	flush_kern_dcache_area(void *addr, size_t size)
  *
  *	Ensure no D cache aliasing occurs, either with itself or
  *	the I cache
  *
- *	- page	- page aligned address
+ *	- addr	- kernel address
+ *	- size	- region size
  */
-ENTRY(arm1022_flush_kern_dcache_page)
+ENTRY(arm1022_flush_kern_dcache_area)
 	mov	ip, #0
 #ifndef CONFIG_CPU_DCACHE_DISABLE
-	add	r1, r0, #PAGE_SZ
+	add	r1, r0, r1
 1:	mcr	p15, 0, r0, c7, c14, 1		@ clean+invalidate D entry
 	add	r0, r0, #CACHE_DLINESIZE
 	cmp	r0, r1
@@ -310,7 +311,7 @@ ENTRY(arm1022_cache_fns)
 	.long	arm1022_flush_user_cache_range
 	.long	arm1022_coherent_kern_range
 	.long	arm1022_coherent_user_range
-	.long	arm1022_flush_kern_dcache_page
+	.long	arm1022_flush_kern_dcache_area
 	.long	arm1022_dma_inv_range
 	.long	arm1022_dma_clean_range
 	.long	arm1022_dma_flush_range
diff --git a/arch/arm/mm/proc-arm1026.S b/arch/arm/mm/proc-arm1026.S
index 3b59f0d6713..96aedb10fcc 100644
--- a/arch/arm/mm/proc-arm1026.S
+++ b/arch/arm/mm/proc-arm1026.S
@@ -208,17 +208,18 @@ ENTRY(arm1026_coherent_user_range)
 	mov	pc, lr
 
 /*
- *	flush_kern_dcache_page(void *page)
+ *	flush_kern_dcache_area(void *addr, size_t size)
  *
  *	Ensure no D cache aliasing occurs, either with itself or
  *	the I cache
  *
- *	- page	- page aligned address
+ *	- addr	- kernel address
+ *	- size	- region size
  */
-ENTRY(arm1026_flush_kern_dcache_page)
+ENTRY(arm1026_flush_kern_dcache_area)
 	mov	ip, #0
 #ifndef CONFIG_CPU_DCACHE_DISABLE
-	add	r1, r0, #PAGE_SZ
+	add	r1, r0, r1
 1:	mcr	p15, 0, r0, c7, c14, 1		@ clean+invalidate D entry
 	add	r0, r0, #CACHE_DLINESIZE
 	cmp	r0, r1
@@ -304,7 +305,7 @@ ENTRY(arm1026_cache_fns)
 	.long	arm1026_flush_user_cache_range
 	.long	arm1026_coherent_kern_range
 	.long	arm1026_coherent_user_range
-	.long	arm1026_flush_kern_dcache_page
+	.long	arm1026_flush_kern_dcache_area
 	.long	arm1026_dma_inv_range
 	.long	arm1026_dma_clean_range
 	.long	arm1026_dma_flush_range
diff --git a/arch/arm/mm/proc-arm920.S b/arch/arm/mm/proc-arm920.S
index 2b7c197cc58..471669e2d7c 100644
--- a/arch/arm/mm/proc-arm920.S
+++ b/arch/arm/mm/proc-arm920.S
@@ -207,15 +207,16 @@ ENTRY(arm920_coherent_user_range)
 	mov	pc, lr
 
 /*
- *	flush_kern_dcache_page(void *page)
+ *	flush_kern_dcache_area(void *addr, size_t size)
  *
  *	Ensure no D cache aliasing occurs, either with itself or
  *	the I cache
  *
- *	- addr	- page aligned address
+ *	- addr	- kernel address
+ *	- size	- region size
  */
-ENTRY(arm920_flush_kern_dcache_page)
-	add	r1, r0, #PAGE_SZ
+ENTRY(arm920_flush_kern_dcache_area)
+	add	r1, r0, r1
 1:	mcr	p15, 0, r0, c7, c14, 1		@ clean+invalidate D entry
 	add	r0, r0, #CACHE_DLINESIZE
 	cmp	r0, r1
@@ -293,7 +294,7 @@ ENTRY(arm920_cache_fns)
 	.long	arm920_flush_user_cache_range
 	.long	arm920_coherent_kern_range
 	.long	arm920_coherent_user_range
-	.long	arm920_flush_kern_dcache_page
+	.long	arm920_flush_kern_dcache_area
 	.long	arm920_dma_inv_range
 	.long	arm920_dma_clean_range
 	.long	arm920_dma_flush_range
diff --git a/arch/arm/mm/proc-arm922.S b/arch/arm/mm/proc-arm922.S
index 06a1aa4e339..ee111b00fa4 100644
--- a/arch/arm/mm/proc-arm922.S
+++ b/arch/arm/mm/proc-arm922.S
@@ -209,15 +209,16 @@ ENTRY(arm922_coherent_user_range)
 	mov	pc, lr
 
 /*
- *	flush_kern_dcache_page(void *page)
+ *	flush_kern_dcache_area(void *addr, size_t size)
  *
  *	Ensure no D cache aliasing occurs, either with itself or
  *	the I cache
  *
- *	- addr	- page aligned address
+ *	- addr	- kernel address
+ *	- size	- region size
  */
-ENTRY(arm922_flush_kern_dcache_page)
-	add	r1, r0, #PAGE_SZ
+ENTRY(arm922_flush_kern_dcache_area)
+	add	r1, r0, r1
 1:	mcr	p15, 0, r0, c7, c14, 1		@ clean+invalidate D entry
 	add	r0, r0, #CACHE_DLINESIZE
 	cmp	r0, r1
@@ -295,7 +296,7 @@ ENTRY(arm922_cache_fns)
 	.long	arm922_flush_user_cache_range
 	.long	arm922_coherent_kern_range
 	.long	arm922_coherent_user_range
-	.long	arm922_flush_kern_dcache_page
+	.long	arm922_flush_kern_dcache_area
 	.long	arm922_dma_inv_range
 	.long	arm922_dma_clean_range
 	.long	arm922_dma_flush_range
diff --git a/arch/arm/mm/proc-arm925.S b/arch/arm/mm/proc-arm925.S
index cb53435a85a..8deb5bde58e 100644
--- a/arch/arm/mm/proc-arm925.S
+++ b/arch/arm/mm/proc-arm925.S
@@ -251,15 +251,16 @@ ENTRY(arm925_coherent_user_range)
 	mov	pc, lr
 
 /*
- *	flush_kern_dcache_page(void *page)
+ *	flush_kern_dcache_area(void *addr, size_t size)
  *
  *	Ensure no D cache aliasing occurs, either with itself or
  *	the I cache
  *
- *	- addr	- page aligned address
+ *	- addr	- kernel address
+ *	- size	- region size
  */
-ENTRY(arm925_flush_kern_dcache_page)
-	add	r1, r0, #PAGE_SZ
+ENTRY(arm925_flush_kern_dcache_area)
+	add	r1, r0, r1
 1:	mcr	p15, 0, r0, c7, c14, 1		@ clean+invalidate D entry
 	add	r0, r0, #CACHE_DLINESIZE
 	cmp	r0, r1
@@ -346,7 +347,7 @@ ENTRY(arm925_cache_fns)
 	.long	arm925_flush_user_cache_range
 	.long	arm925_coherent_kern_range
 	.long	arm925_coherent_user_range
-	.long	arm925_flush_kern_dcache_page
+	.long	arm925_flush_kern_dcache_area
 	.long	arm925_dma_inv_range
 	.long	arm925_dma_clean_range
 	.long	arm925_dma_flush_range
diff --git a/arch/arm/mm/proc-arm926.S b/arch/arm/mm/proc-arm926.S
index 1c4848704bb..64db6e275a4 100644
--- a/arch/arm/mm/proc-arm926.S
+++ b/arch/arm/mm/proc-arm926.S
@@ -214,15 +214,16 @@ ENTRY(arm926_coherent_user_range)
 	mov	pc, lr
 
 /*
- *	flush_kern_dcache_page(void *page)
+ *	flush_kern_dcache_area(void *addr, size_t size)
  *
  *	Ensure no D cache aliasing occurs, either with itself or
  *	the I cache
  *
- *	- addr	- page aligned address
+ *	- addr	- kernel address
+ *	- size	- region size
  */
-ENTRY(arm926_flush_kern_dcache_page)
-	add	r1, r0, #PAGE_SZ
+ENTRY(arm926_flush_kern_dcache_area)
+	add	r1, r0, r1
 1:	mcr	p15, 0, r0, c7, c14, 1		@ clean+invalidate D entry
 	add	r0, r0, #CACHE_DLINESIZE
 	cmp	r0, r1
@@ -309,7 +310,7 @@ ENTRY(arm926_cache_fns)
 	.long	arm926_flush_user_cache_range
 	.long	arm926_coherent_kern_range
 	.long	arm926_coherent_user_range
-	.long	arm926_flush_kern_dcache_page
+	.long	arm926_flush_kern_dcache_area
 	.long	arm926_dma_inv_range
 	.long	arm926_dma_clean_range
 	.long	arm926_dma_flush_range
diff --git a/arch/arm/mm/proc-arm940.S b/arch/arm/mm/proc-arm940.S
index 5b0f8464c8f..8196b9f401f 100644
--- a/arch/arm/mm/proc-arm940.S
+++ b/arch/arm/mm/proc-arm940.S
@@ -141,14 +141,15 @@ ENTRY(arm940_coherent_user_range)
 	/* FALLTHROUGH */
 
 /*
- *	flush_kern_dcache_page(void *page)
+ *	flush_kern_dcache_area(void *addr, size_t size)
  *
  *	Ensure no D cache aliasing occurs, either with itself or
  *	the I cache
  *
- *	- addr	- page aligned address
+ *	- addr	- kernel address
+ *	- size	- region size
  */
-ENTRY(arm940_flush_kern_dcache_page)
+ENTRY(arm940_flush_kern_dcache_area)
 	mov	ip, #0
 	mov	r1, #(CACHE_DSEGMENTS - 1) << 4	@ 4 segments
 1:	orr	r3, r1, #(CACHE_DENTRIES - 1) << 26 @ 64 entries
@@ -238,7 +239,7 @@ ENTRY(arm940_cache_fns)
 	.long	arm940_flush_user_cache_range
 	.long	arm940_coherent_kern_range
 	.long	arm940_coherent_user_range
-	.long	arm940_flush_kern_dcache_page
+	.long	arm940_flush_kern_dcache_area
 	.long	arm940_dma_inv_range
 	.long	arm940_dma_clean_range
 	.long	arm940_dma_flush_range
diff --git a/arch/arm/mm/proc-arm946.S b/arch/arm/mm/proc-arm946.S
index 40c0449a139..9a951239c86 100644
--- a/arch/arm/mm/proc-arm946.S
+++ b/arch/arm/mm/proc-arm946.S
@@ -183,16 +183,17 @@ ENTRY(arm946_coherent_user_range)
 	mov	pc, lr
 
 /*
- *	flush_kern_dcache_page(void *page)
+ *	flush_kern_dcache_area(void *addr, size_t size)
  *
  *	Ensure no D cache aliasing occurs, either with itself or
  *	the I cache
  *
- *	- addr	- page aligned address
+ *	- addr	- kernel address
+ *	- size	- region size
  * (same as arm926)
  */
-ENTRY(arm946_flush_kern_dcache_page)
-	add	r1, r0, #PAGE_SZ
+ENTRY(arm946_flush_kern_dcache_area)
+	add	r1, r0, r1
 1:	mcr	p15, 0, r0, c7, c14, 1		@ clean+invalidate D entry
 	add	r0, r0, #CACHE_DLINESIZE
 	cmp	r0, r1
@@ -280,7 +281,7 @@ ENTRY(arm946_cache_fns)
 	.long	arm946_flush_user_cache_range
 	.long	arm946_coherent_kern_range
 	.long	arm946_coherent_user_range
-	.long	arm946_flush_kern_dcache_page
+	.long	arm946_flush_kern_dcache_area
 	.long	arm946_dma_inv_range
 	.long	arm946_dma_clean_range
 	.long	arm946_dma_flush_range
diff --git a/arch/arm/mm/proc-feroceon.S b/arch/arm/mm/proc-feroceon.S
index d0d7795200f..dbc39383e66 100644
--- a/arch/arm/mm/proc-feroceon.S
+++ b/arch/arm/mm/proc-feroceon.S
@@ -226,16 +226,17 @@ ENTRY(feroceon_coherent_user_range)
 	mov	pc, lr
 
 /*
- *	flush_kern_dcache_page(void *page)
+ *	flush_kern_dcache_area(void *addr, size_t size)
  *
  *	Ensure no D cache aliasing occurs, either with itself or
  *	the I cache
  *
- *	- addr	- page aligned address
+ *	- addr	- kernel address
+ *	- size	- region size
  */
 	.align	5
-ENTRY(feroceon_flush_kern_dcache_page)
-	add	r1, r0, #PAGE_SZ
+ENTRY(feroceon_flush_kern_dcache_area)
+	add	r1, r0, r1
 1:	mcr	p15, 0, r0, c7, c14, 1		@ clean+invalidate D entry
 	add	r0, r0, #CACHE_DLINESIZE
 	cmp	r0, r1
@@ -246,7 +247,7 @@ ENTRY(feroceon_flush_kern_dcache_page)
 	mov	pc, lr
 
 	.align	5
-ENTRY(feroceon_range_flush_kern_dcache_page)
+ENTRY(feroceon_range_flush_kern_dcache_area)
 	mrs	r2, cpsr
 	add	r1, r0, #PAGE_SZ - CACHE_DLINESIZE	@ top addr is inclusive
 	orr	r3, r2, #PSR_I_BIT
@@ -372,7 +373,7 @@ ENTRY(feroceon_cache_fns)
 	.long	feroceon_flush_user_cache_range
 	.long	feroceon_coherent_kern_range
 	.long	feroceon_coherent_user_range
-	.long	feroceon_flush_kern_dcache_page
+	.long	feroceon_flush_kern_dcache_area
 	.long	feroceon_dma_inv_range
 	.long	feroceon_dma_clean_range
 	.long	feroceon_dma_flush_range
@@ -383,7 +384,7 @@ ENTRY(feroceon_range_cache_fns)
 	.long	feroceon_flush_user_cache_range
 	.long	feroceon_coherent_kern_range
 	.long	feroceon_coherent_user_range
-	.long	feroceon_range_flush_kern_dcache_page
+	.long	feroceon_range_flush_kern_dcache_area
 	.long	feroceon_range_dma_inv_range
 	.long	feroceon_range_dma_clean_range
 	.long	feroceon_range_dma_flush_range
diff --git a/arch/arm/mm/proc-mohawk.S b/arch/arm/mm/proc-mohawk.S
index 52b5fd74fbb..9674d36cc97 100644
--- a/arch/arm/mm/proc-mohawk.S
+++ b/arch/arm/mm/proc-mohawk.S
@@ -186,15 +186,16 @@ ENTRY(mohawk_coherent_user_range)
 	mov	pc, lr
 
 /*
- *	flush_kern_dcache_page(void *page)
+ *	flush_kern_dcache_area(void *addr, size_t size)
  *
  *	Ensure no D cache aliasing occurs, either with itself or
  *	the I cache
  *
- *	- addr	- page aligned address
+ *	- addr	- kernel address
+ *	- size	- region size
  */
-ENTRY(mohawk_flush_kern_dcache_page)
-	add	r1, r0, #PAGE_SZ
+ENTRY(mohawk_flush_kern_dcache_area)
+	add	r1, r0, r1
 1:	mcr	p15, 0, r0, c7, c14, 1		@ clean+invalidate D entry
 	add	r0, r0, #CACHE_DLINESIZE
 	cmp	r0, r1
@@ -273,7 +274,7 @@ ENTRY(mohawk_cache_fns)
 	.long	mohawk_flush_user_cache_range
 	.long	mohawk_coherent_kern_range
 	.long	mohawk_coherent_user_range
-	.long	mohawk_flush_kern_dcache_page
+	.long	mohawk_flush_kern_dcache_area
 	.long	mohawk_dma_inv_range
 	.long	mohawk_dma_clean_range
 	.long	mohawk_dma_flush_range
diff --git a/arch/arm/mm/proc-syms.c b/arch/arm/mm/proc-syms.c
index ac5c80062b7..3e6210b4d6d 100644
--- a/arch/arm/mm/proc-syms.c
+++ b/arch/arm/mm/proc-syms.c
@@ -27,8 +27,7 @@ EXPORT_SYMBOL(__cpuc_flush_kern_all);
 EXPORT_SYMBOL(__cpuc_flush_user_all);
 EXPORT_SYMBOL(__cpuc_flush_user_range);
 EXPORT_SYMBOL(__cpuc_coherent_kern_range);
-EXPORT_SYMBOL(__cpuc_flush_dcache_page);
-EXPORT_SYMBOL(dmac_inv_range);  /* because of flush_ioremap_region() */
+EXPORT_SYMBOL(__cpuc_flush_dcache_area);
 #else
 EXPORT_SYMBOL(cpu_cache);
 #endif
diff --git a/arch/arm/mm/proc-v6.S b/arch/arm/mm/proc-v6.S
index 5485c821101..7a5337ed7d6 100644
--- a/arch/arm/mm/proc-v6.S
+++ b/arch/arm/mm/proc-v6.S
@@ -59,8 +59,6 @@ ENTRY(cpu_v6_proc_fin)
  *	to what would be the reset vector.
  *
  *	- loc   - location to jump to for soft reset
- *
- *	It is assumed that:
  */
 	.align	5
 ENTRY(cpu_v6_reset)
@@ -254,10 +252,9 @@ __pj4_v6_proc_info:
 	.long	0x560f5810
 	.long	0xff0ffff0
 	.long   PMD_TYPE_SECT | \
-		PMD_SECT_BUFFERABLE | \
-		PMD_SECT_CACHEABLE | \
 		PMD_SECT_AP_WRITE | \
-		PMD_SECT_AP_READ
+		PMD_SECT_AP_READ | \
+		PMD_FLAGS
 	.long   PMD_TYPE_SECT | \
 		PMD_SECT_XN | \
 		PMD_SECT_AP_WRITE | \
diff --git a/arch/arm/mm/proc-v7.S b/arch/arm/mm/proc-v7.S
index 3a285218fd1..7aaf88a3b7a 100644
--- a/arch/arm/mm/proc-v7.S
+++ b/arch/arm/mm/proc-v7.S
@@ -45,7 +45,14 @@ ENTRY(cpu_v7_proc_init)
 ENDPROC(cpu_v7_proc_init)
 
 ENTRY(cpu_v7_proc_fin)
-	mov	pc, lr
+	stmfd	sp!, {lr}
+	cpsid	if				@ disable interrupts
+	bl	v7_flush_kern_cache_all
+	mrc	p15, 0, r0, c1, c0, 0		@ ctrl register
+	bic	r0, r0, #0x1000			@ ...i............
+	bic	r0, r0, #0x0006			@ .............ca.
+	mcr	p15, 0, r0, c1, c0, 0		@ disable caches
+	ldmfd	sp!, {pc}
 ENDPROC(cpu_v7_proc_fin)
 
 /*
@@ -56,8 +63,6 @@ ENDPROC(cpu_v7_proc_fin)
  *	to what would be the reset vector.
  *
  *	- loc   - location to jump to for soft reset
- *
- *	It is assumed that:
  */
 	.align	5
 ENTRY(cpu_v7_reset)
diff --git a/arch/arm/mm/proc-xsc3.S b/arch/arm/mm/proc-xsc3.S
index fab134e2982..8e4f6dca899 100644
--- a/arch/arm/mm/proc-xsc3.S
+++ b/arch/arm/mm/proc-xsc3.S
@@ -226,15 +226,16 @@ ENTRY(xsc3_coherent_user_range)
 	mov	pc, lr
 
 /*
- *	flush_kern_dcache_page(void *page)
+ *	flush_kern_dcache_area(void *addr, size_t size)
  *
  *	Ensure no D cache aliasing occurs, either with itself or
  *	the I cache.
  *
- *	- addr	- page aligned address
+ *	- addr	- kernel address
+ *	- size	- region size
  */
-ENTRY(xsc3_flush_kern_dcache_page)
-	add	r1, r0, #PAGE_SZ
+ENTRY(xsc3_flush_kern_dcache_area)
+	add	r1, r0, r1
 1:	mcr	p15, 0, r0, c7, c14, 1		@ clean/invalidate L1 D line
 	add	r0, r0, #CACHELINESIZE
 	cmp	r0, r1
@@ -309,7 +310,7 @@ ENTRY(xsc3_cache_fns)
 	.long	xsc3_flush_user_cache_range
 	.long	xsc3_coherent_kern_range
 	.long	xsc3_coherent_user_range
-	.long	xsc3_flush_kern_dcache_page
+	.long	xsc3_flush_kern_dcache_area
 	.long	xsc3_dma_inv_range
 	.long	xsc3_dma_clean_range
 	.long	xsc3_dma_flush_range
@@ -406,6 +407,13 @@ __xsc3_setup:
 
 	adr	r5, xsc3_crval
 	ldmia	r5, {r5, r6}
+
+#ifdef CONFIG_CACHE_XSC3L2
+	mrc	p15, 1, r0, c0, c0, 1		@ get L2 present information
+	ands	r0, r0, #0xf8
+	orrne	r6, r6, #(1 << 26)		@ enable L2 if present
+#endif
+
 	mrc	p15, 0, r0, c1, c0, 0		@ get control register
 	bic	r0, r0, r5			@ ..V. ..R. .... ..A.
 	orr	r0, r0, r6			@ ..VI Z..S .... .C.M (mmu)
diff --git a/arch/arm/mm/proc-xscale.S b/arch/arm/mm/proc-xscale.S
index f056c283682..93df47265f2 100644
--- a/arch/arm/mm/proc-xscale.S
+++ b/arch/arm/mm/proc-xscale.S
@@ -284,15 +284,16 @@ ENTRY(xscale_coherent_user_range)
 	mov	pc, lr
 
 /*
- *	flush_kern_dcache_page(void *page)
+ *	flush_kern_dcache_area(void *addr, size_t size)
  *
  *	Ensure no D cache aliasing occurs, either with itself or
  *	the I cache
  *
- *	- addr	- page aligned address
+ *	- addr	- kernel address
+ *	- size	- region size
  */
-ENTRY(xscale_flush_kern_dcache_page)
-	add	r1, r0, #PAGE_SZ
+ENTRY(xscale_flush_kern_dcache_area)
+	add	r1, r0, r1
 1:	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
 	mcr	p15, 0, r0, c7, c6, 1		@ invalidate D entry
 	add	r0, r0, #CACHELINESIZE
@@ -368,7 +369,7 @@ ENTRY(xscale_cache_fns)
 	.long	xscale_flush_user_cache_range
 	.long	xscale_coherent_kern_range
 	.long	xscale_coherent_user_range
-	.long	xscale_flush_kern_dcache_page
+	.long	xscale_flush_kern_dcache_area
 	.long	xscale_dma_inv_range
 	.long	xscale_dma_clean_range
 	.long	xscale_dma_flush_range
@@ -392,7 +393,7 @@ ENTRY(xscale_80200_A0_A1_cache_fns)
 	.long	xscale_flush_user_cache_range
 	.long	xscale_coherent_kern_range
 	.long	xscale_coherent_user_range
-	.long	xscale_flush_kern_dcache_page
+	.long	xscale_flush_kern_dcache_area
 	.long	xscale_dma_flush_range
 	.long	xscale_dma_clean_range
 	.long	xscale_dma_flush_range
diff --git a/arch/arm/mm/tlb-v7.S b/arch/arm/mm/tlb-v7.S
index a26a605b73b..0cb1848bd87 100644
--- a/arch/arm/mm/tlb-v7.S
+++ b/arch/arm/mm/tlb-v7.S
@@ -40,7 +40,6 @@ ENTRY(v7wbi_flush_user_tlb_range)
 	asid	r3, r3				@ mask ASID
 	orr	r0, r3, r0, lsl #PAGE_SHIFT	@ Create initial MVA
 	mov	r1, r1, lsl #PAGE_SHIFT
-	vma_vm_flags r2, r2			@ get vma->vm_flags
 1:
 #ifdef CONFIG_SMP
 	mcr	p15, 0, r0, c8, c3, 1		@ TLB invalidate U MVA (shareable)