Merge branch 'master'

59 files changed, 494 insertions, 560 deletions

diff --git a/MAINTAINERS b/MAINTAINERS
index abf7f7a17ae..767fb610963 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -1618,6 +1618,13 @@ M:	vandrove@vc.cvut.cz
 L:	linux-fbdev-devel@lists.sourceforge.net
 S:	Maintained
 
+MEGARAID SCSI DRIVERS
+P:     Neela Syam Kolli
+M:     Neela.Kolli@engenio.com
+S:     linux-scsi@vger.kernel.org
+W:     http://megaraid.lsilogic.com
+S:     Maintained
+
 MEMORY TECHNOLOGY DEVICES
 P:	David Woodhouse
 M:	dwmw2@infradead.org
diff --git a/Makefile b/Makefile
index 504ba3ceb29..be33d758c0f 100644
--- a/Makefile
+++ b/Makefile
@@ -660,8 +660,10 @@ quiet_cmd_sysmap = SYSMAP
 # Link of vmlinux
 # If CONFIG_KALLSYMS is set .version is already updated
 # Generate System.map and verify that the content is consistent
-
+# Use + in front of the vmlinux_version rule to silent warning with make -j2
+# First command is ':' to allow us to use + in front of the rule
 define rule_vmlinux__
+	:
 	$(if $(CONFIG_KALLSYMS),,+$(call cmd,vmlinux_version))
 
 	$(call cmd,vmlinux__)
diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S
index 08e58ecd44b..0d5db5279c5 100644
--- a/arch/arm/kernel/vmlinux.lds.S
+++ b/arch/arm/kernel/vmlinux.lds.S
@@ -89,13 +89,6 @@ SECTIONS
 		*(.got)			/* Global offset table		*/
 	}
 
-	. = ALIGN(16);
-	__ex_table : {			/* Exception table		*/
-		__start___ex_table = .;
-			*(__ex_table)
-		__stop___ex_table = .;
-	}
-
 	RODATA
 
 	_etext = .;			/* End of text and rodata section */
@@ -138,6 +131,14 @@ SECTIONS
 		*(.data.cacheline_aligned)
 
 		/*
+		 * The exception fixup table (might need resorting at runtime)
+		 */
+		. = ALIGN(32);
+		__start___ex_table = .;
+		*(__ex_table)
+		__stop___ex_table = .;
+
+		/*
 		 * and the usual data section
 		 */
 		*(.data)
diff --git a/arch/arm/mach-l7200/core.c b/arch/arm/mach-l7200/core.c
index 5fd8c9f97f9..03ed742ae2b 100644
--- a/arch/arm/mach-l7200/core.c
+++ b/arch/arm/mach-l7200/core.c
@@ -7,11 +7,17 @@
  */
 #include <linux/kernel.h>
 #include <linux/init.h>
+#include <linux/device.h>
 
+#include <asm/types.h>
+#include <asm/irq.h>
+#include <asm/mach-types.h>
 #include <asm/hardware.h>
 #include <asm/page.h>
 
+#include <asm/mach/arch.h>
 #include <asm/mach/map.h>
+#include <asm/mach/irq.h>
 
 /*
  * IRQ base register
@@ -47,6 +53,12 @@ static void l7200_unmask_irq(unsigned int irq)
 {
 	IRQ_ENABLE = 1 << irq;
 }
+
+static struct irqchip l7200_irq_chip = {
+	.ack		= l7200_mask_irq,
+	.mask		= l7200_mask_irq,
+	.unmask		= l7200_unmask_irq
+};
  
 static void __init l7200_init_irq(void)
 {
@@ -56,11 +68,9 @@ static void __init l7200_init_irq(void)
 	FIQ_ENABLECLEAR = 0xffffffff;	/* clear all fast interrupt enables */
 
 	for (irq = 0; irq < NR_IRQS; irq++) {
-		irq_desc[irq].valid	= 1;
-		irq_desc[irq].probe_ok	= 1;
-		irq_desc[irq].mask_ack	= l7200_mask_irq;
-		irq_desc[irq].mask	= l7200_mask_irq;
-		irq_desc[irq].unmask	= l7200_unmask_irq;
+		set_irq_chip(irq, &l7200_irq_chip);
+		set_irq_flags(irq, IRQF_VALID);
+		set_irq_handler(irq, do_level_IRQ);
 	}
 
 	init_FIQ();
diff --git a/arch/arm/mach-pxa/corgi_lcd.c b/arch/arm/mach-pxa/corgi_lcd.c
index c02ef7c0f7e..850538fadec 100644
--- a/arch/arm/mach-pxa/corgi_lcd.c
+++ b/arch/arm/mach-pxa/corgi_lcd.c
@@ -467,6 +467,7 @@ void corgi_put_hsync(void)
 {
 	if (get_hsync_time)
 		symbol_put(w100fb_get_hsynclen);
+	get_hsync_time = NULL;
 }
 
 void corgi_wait_hsync(void)
@@ -476,20 +477,37 @@ void corgi_wait_hsync(void)
 #endif
 
 #ifdef CONFIG_PXA_SHARP_Cxx00
+static struct device *spitz_pxafb_dev;
+
+static int is_pxafb_device(struct device * dev, void * data)
+{
+	struct platform_device *pdev = container_of(dev, struct platform_device, dev);
+
+	return (strncmp(pdev->name, "pxa2xx-fb", 9) == 0);
+}
+
 unsigned long spitz_get_hsync_len(void)
 {
+	if (!spitz_pxafb_dev) {
+		spitz_pxafb_dev = bus_find_device(&platform_bus_type, NULL, NULL, is_pxafb_device);
+		if (!spitz_pxafb_dev)
+			return 0;
+	}
 	if (!get_hsync_time)
 		get_hsync_time = symbol_get(pxafb_get_hsync_time);
 	if (!get_hsync_time)
 		return 0;
 
-	return pxafb_get_hsync_time(&pxafb_device.dev);
+	return pxafb_get_hsync_time(spitz_pxafb_dev);
 }
 
 void spitz_put_hsync(void)
 {
+	put_device(spitz_pxafb_dev);
 	if (get_hsync_time)
 		symbol_put(pxafb_get_hsync_time);
+	spitz_pxafb_dev = NULL;
+	get_hsync_time = NULL;
 }
 
 void spitz_wait_hsync(void)
diff --git a/arch/arm/mach-pxa/generic.c b/arch/arm/mach-pxa/generic.c
index d0660a8c4b7..d327c127edd 100644
--- a/arch/arm/mach-pxa/generic.c
+++ b/arch/arm/mach-pxa/generic.c
@@ -208,6 +208,11 @@ static struct platform_device pxafb_device = {
 	.resource	= pxafb_resources,
 };
 
+void __init set_pxa_fb_parent(struct device *parent_dev)
+{
+	pxafb_device.dev.parent = parent_dev;
+}
+
 static struct platform_device ffuart_device = {
 	.name		= "pxa2xx-uart",
 	.id		= 0,
diff --git a/arch/arm/mach-pxa/spitz.c b/arch/arm/mach-pxa/spitz.c
index 568afe3d6e1..d0ab428c2d7 100644
--- a/arch/arm/mach-pxa/spitz.c
+++ b/arch/arm/mach-pxa/spitz.c
@@ -36,7 +36,6 @@
 #include <asm/arch/irq.h>
 #include <asm/arch/mmc.h>
 #include <asm/arch/udc.h>
-#include <asm/arch/ohci.h>
 #include <asm/arch/pxafb.h>
 #include <asm/arch/akita.h>
 #include <asm/arch/spitz.h>
@@ -304,7 +303,6 @@ static struct platform_device *devices[] __initdata = {
 	&spitzkbd_device,
 	&spitzts_device,
 	&spitzbl_device,
-	&spitzbattery_device,
 };
 
 static void __init common_init(void)
@@ -328,7 +326,7 @@ static void __init common_init(void)
 
 	platform_add_devices(devices, ARRAY_SIZE(devices));
 	pxa_set_mci_info(&spitz_mci_platform_data);
-	pxafb_device.dev.parent = &spitzssp_device.dev;
+	set_pxa_fb_parent(&spitzssp_device.dev);
 	set_pxa_fb_info(&spitz_pxafb_info);
 }
 
diff --git a/arch/arm/mach-s3c2410/Kconfig b/arch/arm/mach-s3c2410/Kconfig
index 06807c6ee68..c796bcdd615 100644
--- a/arch/arm/mach-s3c2410/Kconfig
+++ b/arch/arm/mach-s3c2410/Kconfig
@@ -12,6 +12,7 @@ config MACH_ANUBIS
 config ARCH_BAST
 	bool "Simtec Electronics BAST (EB2410ITX)"
 	select CPU_S3C2410
+	select ISA
 	help
 	  Say Y here if you are using the Simtec Electronics EB2410ITX
 	  development board (also known as BAST)
diff --git a/arch/m32r/kernel/smp.c b/arch/m32r/kernel/smp.c
index a4576ac7e87..8b1f6eb7687 100644
--- a/arch/m32r/kernel/smp.c
+++ b/arch/m32r/kernel/smp.c
@@ -275,12 +275,14 @@ static void flush_tlb_all_ipi(void *info)
  *==========================================================================*/
 void smp_flush_tlb_mm(struct mm_struct *mm)
 {
-	int cpu_id = smp_processor_id();
+	int cpu_id;
 	cpumask_t cpu_mask;
-	unsigned long *mmc = &mm->context[cpu_id];
+	unsigned long *mmc;
 	unsigned long flags;
 
 	preempt_disable();
+	cpu_id = smp_processor_id();
+	mmc = &mm->context[cpu_id];
 	cpu_mask = mm->cpu_vm_mask;
 	cpu_clear(cpu_id, cpu_mask);
 
@@ -343,12 +345,14 @@ void smp_flush_tlb_range(struct vm_area_struct *vma, unsigned long start,
 void smp_flush_tlb_page(struct vm_area_struct *vma, unsigned long va)
 {
 	struct mm_struct *mm = vma->vm_mm;
-	int cpu_id = smp_processor_id();
+	int cpu_id;
 	cpumask_t cpu_mask;
-	unsigned long *mmc = &mm->context[cpu_id];
+	unsigned long *mmc;
 	unsigned long flags;
 
 	preempt_disable();
+	cpu_id = smp_processor_id();
+	mmc = &mm->context[cpu_id];
 	cpu_mask = mm->cpu_vm_mask;
 	cpu_clear(cpu_id, cpu_mask);
 
diff --git a/arch/mips/pci/fixup-tb0226.c b/arch/mips/pci/fixup-tb0226.c
index 61513d5d97d..b5d42b12de1 100644
--- a/arch/mips/pci/fixup-tb0226.c
+++ b/arch/mips/pci/fixup-tb0226.c
@@ -1,7 +1,7 @@
 /*
  *  fixup-tb0226.c, The TANBAC TB0226 specific PCI fixups.
  *
- *  Copyright (C) 2002-2004  Yoichi Yuasa <yuasa@hh.iij4u.or.jp>
+ *  Copyright (C) 2002-2005  Yoichi Yuasa <yuasa@hh.iij4u.or.jp>
  *
  *  This program is free software; you can redistribute it and/or modify
  *  it under the terms of the GNU General Public License as published by
@@ -20,6 +20,7 @@
 #include <linux/init.h>
 #include <linux/pci.h>
 
+#include <asm/vr41xx/giu.h>
 #include <asm/vr41xx/tb0226.h>
 
 int __init pcibios_map_irq(struct pci_dev *dev, u8 slot, u8 pin)
@@ -29,42 +30,42 @@ int __init pcibios_map_irq(struct pci_dev *dev, u8 slot, u8 pin)
 	switch (slot) {
 	case 12:
 		vr41xx_set_irq_trigger(GD82559_1_PIN,
-				       TRIGGER_LEVEL,
-				       SIGNAL_THROUGH);
-		vr41xx_set_irq_level(GD82559_1_PIN, LEVEL_LOW);
+				       IRQ_TRIGGER_LEVEL,
+				       IRQ_SIGNAL_THROUGH);
+		vr41xx_set_irq_level(GD82559_1_PIN, IRQ_LEVEL_LOW);
 		irq = GD82559_1_IRQ;
 		break;
 	case 13:
 		vr41xx_set_irq_trigger(GD82559_2_PIN,
-				       TRIGGER_LEVEL,
-				       SIGNAL_THROUGH);
-		vr41xx_set_irq_level(GD82559_2_PIN, LEVEL_LOW);
+				       IRQ_TRIGGER_LEVEL,
+				       IRQ_SIGNAL_THROUGH);
+		vr41xx_set_irq_level(GD82559_2_PIN, IRQ_LEVEL_LOW);
 		irq = GD82559_2_IRQ;
 		break;
 	case 14:
 		switch (pin) {
 		case 1:
 			vr41xx_set_irq_trigger(UPD720100_INTA_PIN,
-					       TRIGGER_LEVEL,
-					       SIGNAL_THROUGH);
+					       IRQ_TRIGGER_LEVEL,
+					       IRQ_SIGNAL_THROUGH);
 			vr41xx_set_irq_level(UPD720100_INTA_PIN,
-					     LEVEL_LOW);
+					     IRQ_LEVEL_LOW);
 			irq = UPD720100_INTA_IRQ;
 			break;
 		case 2:
 			vr41xx_set_irq_trigger(UPD720100_INTB_PIN,
-					       TRIGGER_LEVEL,
-					       SIGNAL_THROUGH);
+					       IRQ_TRIGGER_LEVEL,
+					       IRQ_SIGNAL_THROUGH);
 			vr41xx_set_irq_level(UPD720100_INTB_PIN,
-					     LEVEL_LOW);
+					     IRQ_LEVEL_LOW);
 			irq = UPD720100_INTB_IRQ;
 			break;
 		case 3:
 			vr41xx_set_irq_trigger(UPD720100_INTC_PIN,
-					       TRIGGER_LEVEL,
-					       SIGNAL_THROUGH);
+					       IRQ_TRIGGER_LEVEL,
+					       IRQ_SIGNAL_THROUGH);
 			vr41xx_set_irq_level(UPD720100_INTC_PIN,
-					     LEVEL_LOW);
+					     IRQ_LEVEL_LOW);
 			irq = UPD720100_INTC_IRQ;
 			break;
 		default:
diff --git a/arch/ppc64/kernel/pmac_setup.c b/arch/ppc64/kernel/pmac_setup.c
index 25755252067..fa8121d53b8 100644
--- a/arch/ppc64/kernel/pmac_setup.c
+++ b/arch/ppc64/kernel/pmac_setup.c
@@ -115,7 +115,7 @@ static void __pmac pmac_show_cpuinfo(struct seq_file *m)
 	
 	/* find motherboard type */
 	seq_printf(m, "machine\t\t: ");
-	np = find_devices("device-tree");
+	np = of_find_node_by_path("/");
 	if (np != NULL) {
 		pp = (char *) get_property(np, "model", NULL);
 		if (pp != NULL)
@@ -133,6 +133,7 @@ static void __pmac pmac_show_cpuinfo(struct seq_file *m)
 			}
 			seq_printf(m, "\n");
 		}
+		of_node_put(np);
 	} else
 		seq_printf(m, "PowerMac\n");
 
diff --git a/arch/sparc64/kernel/pci_iommu.c b/arch/sparc64/kernel/pci_iommu.c
index 425c60cfea1..a11910be101 100644
--- a/arch/sparc64/kernel/pci_iommu.c
+++ b/arch/sparc64/kernel/pci_iommu.c
@@ -49,12 +49,6 @@ static void __iommu_flushall(struct pci_iommu *iommu)
 
 	/* Ensure completion of previous PIO writes. */
 	(void) pci_iommu_read(iommu->write_complete_reg);
-
-	/* Now update everyone's flush point. */
-	for (entry = 0; entry < PBM_NCLUSTERS; entry++) {
-		iommu->alloc_info[entry].flush =
-			iommu->alloc_info[entry].next;
-	}
 }
 
 #define IOPTE_CONSISTENT(CTX) \
@@ -80,120 +74,117 @@ static void inline iopte_make_dummy(struct pci_iommu *iommu, iopte_t *iopte)
 	iopte_val(*iopte) = val;
 }
 
-void pci_iommu_table_init(struct pci_iommu *iommu, int tsbsize)
+/* Based largely upon the ppc64 iommu allocator.  */
+static long pci_arena_alloc(struct pci_iommu *iommu, unsigned long npages)
 {
-	int i;
-
-	tsbsize /= sizeof(iopte_t);
-
-	for (i = 0; i < tsbsize; i++)
-		iopte_make_dummy(iommu, &iommu->page_table[i]);
-}
-
-static iopte_t *alloc_streaming_cluster(struct pci_iommu *iommu, unsigned long npages)
-{
-	iopte_t *iopte, *limit, *first;
-	unsigned long cnum, ent, flush_point;
-
-	cnum = 0;
-	while ((1UL << cnum) < npages)
-		cnum++;
-	iopte  = (iommu->page_table +
-		  (cnum << (iommu->page_table_sz_bits - PBM_LOGCLUSTERS)));
-
-	if (cnum == 0)
-		limit = (iommu->page_table +
-			 iommu->lowest_consistent_map);
-	else
-		limit = (iopte +
-			 (1 << (iommu->page_table_sz_bits - PBM_LOGCLUSTERS)));
-
-	iopte += ((ent = iommu->alloc_info[cnum].next) << cnum);
-	flush_point = iommu->alloc_info[cnum].flush;
-	
-	first = iopte;
-	for (;;) {
-		if (IOPTE_IS_DUMMY(iommu, iopte)) {
-			if ((iopte + (1 << cnum)) >= limit)
-				ent = 0;
-			else
-				ent = ent + 1;
-			iommu->alloc_info[cnum].next = ent;
-			if (ent == flush_point)
-				__iommu_flushall(iommu);
-			break;
+	struct pci_iommu_arena *arena = &iommu->arena;
+	unsigned long n, i, start, end, limit;
+	int pass;
+
+	limit = arena->limit;
+	start = arena->hint;
+	pass = 0;
+
+again:
+	n = find_next_zero_bit(arena->map, limit, start);
+	end = n + npages;
+	if (unlikely(end >= limit)) {
+		if (likely(pass < 1)) {
+			limit = start;
+			start = 0;
+			__iommu_flushall(iommu);
+			pass++;
+			goto again;
+		} else {
+			/* Scanned the whole thing, give up. */
+			return -1;
 		}
-		iopte += (1 << cnum);
-		ent++;
-		if (iopte >= limit) {
-			iopte = (iommu->page_table +
-				 (cnum <<
-				  (iommu->page_table_sz_bits - PBM_LOGCLUSTERS)));
-			ent = 0;
+	}
+
+	for (i = n; i < end; i++) {
+		if (test_bit(i, arena->map)) {
+			start = i + 1;
+			goto again;
 		}
-		if (ent == flush_point)
-			__iommu_flushall(iommu);
-		if (iopte == first)
-			goto bad;
 	}
 
-	/* I've got your streaming cluster right here buddy boy... */
-	return iopte;
+	for (i = n; i < end; i++)
+		__set_bit(i, arena->map);
 
-bad:
-	printk(KERN_EMERG "pci_iommu: alloc_streaming_cluster of npages(%ld) failed!\n",
-	       npages);
-	return NULL;
+	arena->hint = end;
+
+	return n;
 }
 
-static void free_streaming_cluster(struct pci_iommu *iommu, dma_addr_t base,
-				   unsigned long npages, unsigned long ctx)
+static void pci_arena_free(struct pci_iommu_arena *arena, unsigned long base, unsigned long npages)
 {
-	unsigned long cnum, ent;
+	unsigned long i;
 
-	cnum = 0;
-	while ((1UL << cnum) < npages)
-		cnum++;
+	for (i = base; i < (base + npages); i++)
+		__clear_bit(i, arena->map);
+}
 
-	ent = (base << (32 - IO_PAGE_SHIFT + PBM_LOGCLUSTERS - iommu->page_table_sz_bits))
-		>> (32 + PBM_LOGCLUSTERS + cnum - iommu->page_table_sz_bits);
+void pci_iommu_table_init(struct pci_iommu *iommu, int tsbsize, u32 dma_offset, u32 dma_addr_mask)
+{
+	unsigned long i, tsbbase, order, sz, num_tsb_entries;
+
+	num_tsb_entries = tsbsize / sizeof(iopte_t);
+
+	/* Setup initial software IOMMU state. */
+	spin_lock_init(&iommu->lock);
+	iommu->ctx_lowest_free = 1;
+	iommu->page_table_map_base = dma_offset;
+	iommu->dma_addr_mask = dma_addr_mask;
+
+	/* Allocate and initialize the free area map.  */
+	sz = num_tsb_entries / 8;
+	sz = (sz + 7UL) & ~7UL;
+	iommu->arena.map = kmalloc(sz, GFP_KERNEL);
+	if (!iommu->arena.map) {
+		prom_printf("PCI_IOMMU: Error, kmalloc(arena.map) failed.\n");
+		prom_halt();
+	}
+	memset(iommu->arena.map, 0, sz);
+	iommu->arena.limit = num_tsb_entries;
 
-	/* If the global flush might not have caught this entry,
-	 * adjust the flush point such that we will flush before
-	 * ever trying to reuse it.
+	/* Allocate and initialize the dummy page which we
+	 * set inactive IO PTEs to point to.
 	 */
-#define between(X,Y,Z)	(((Z) - (Y)) >= ((X) - (Y)))
-	if (between(ent, iommu->alloc_info[cnum].next, iommu->alloc_info[cnum].flush))
-		iommu->alloc_info[cnum].flush = ent;
-#undef between
+	iommu->dummy_page = __get_free_pages(GFP_KERNEL, 0);
+	if (!iommu->dummy_page) {
+		prom_printf("PCI_IOMMU: Error, gfp(dummy_page) failed.\n");
+		prom_halt();
+	}
+	memset((void *)iommu->dummy_page, 0, PAGE_SIZE);
+	iommu->dummy_page_pa = (unsigned long) __pa(iommu->dummy_page);
+
+	/* Now allocate and setup the IOMMU page table itself.  */
+	order = get_order(tsbsize);
+	tsbbase = __get_free_pages(GFP_KERNEL, order);
+	if (!tsbbase) {
+		prom_printf("PCI_IOMMU: Error, gfp(tsb) failed.\n");
+		prom_halt();
+	}
+	iommu->page_table = (iopte_t *)tsbbase;
+
+	for (i = 0; i < num_tsb_entries; i++)
+		iopte_make_dummy(iommu, &iommu->page_table[i]);
 }
 
-/* We allocate consistent mappings from the end of cluster zero. */
-static iopte_t *alloc_consistent_cluster(struct pci_iommu *iommu, unsigned long npages)
+static inline iopte_t *alloc_npages(struct pci_iommu *iommu, unsigned long npages)
 {
-	iopte_t *iopte;
+	long entry;
 
-	iopte = iommu->page_table + (1 << (iommu->page_table_sz_bits - PBM_LOGCLUSTERS));
-	while (iopte > iommu->page_table) {
-		iopte--;
-		if (IOPTE_IS_DUMMY(iommu, iopte)) {
-			unsigned long tmp = npages;
+	entry = pci_arena_alloc(iommu, npages);
+	if (unlikely(entry < 0))
+		return NULL;
 
-			while (--tmp) {
-				iopte--;
-				if (!IOPTE_IS_DUMMY(iommu, iopte))
-					break;
-			}
-			if (tmp == 0) {
-				u32 entry = (iopte - iommu->page_table);
+	return iommu->page_table + entry;
+}
 
-				if (entry < iommu->lowest_consistent_map)
-					iommu->lowest_consistent_map = entry;
-				return iopte;
-			}
-		}
-	}
-	return NULL;
+static inline void free_npages(struct pci_iommu *iommu, dma_addr_t base, unsigned long npages)
+{
+	pci_arena_free(&iommu->arena, base >> IO_PAGE_SHIFT, npages);
 }
 
 static int iommu_alloc_ctx(struct pci_iommu *iommu)
@@ -233,7 +224,7 @@ void *pci_alloc_consistent(struct pci_dev *pdev, size_t size, dma_addr_t *dma_ad
 	struct pcidev_cookie *pcp;
 	struct pci_iommu *iommu;
 	iopte_t *iopte;
-	unsigned long flags, order, first_page, ctx;
+	unsigned long flags, order, first_page;
 	void *ret;
 	int npages;
 
@@ -251,9 +242,10 @@ void *pci_alloc_consistent(struct pci_dev *pdev, size_t size, dma_addr_t *dma_ad
 	iommu = pcp->pbm->iommu;
 
 	spin_lock_irqsave(&iommu->lock, flags);
-	iopte = alloc_consistent_cluster(iommu, size >> IO_PAGE_SHIFT);
-	if (iopte == NULL) {
-		spin_unlock_irqrestore(&iommu->lock, flags);
+	iopte = alloc_npages(iommu, size >> IO_PAGE_SHIFT);
+	spin_unlock_irqrestore(&iommu->lock, flags);
+
+	if (unlikely(iopte == NULL)) {
 		free_pages(first_page, order);
 		return NULL;
 	}
@@ -262,31 +254,15 @@ void *pci_alloc_consistent(struct pci_dev *pdev, size_t size, dma_addr_t *dma_ad
 		      ((iopte - iommu->page_table) << IO_PAGE_SHIFT));
 	ret = (void *) first_page;
 	npages = size >> IO_PAGE_SHIFT;
-	ctx = 0;
-	if (iommu->iommu_ctxflush)
-		ctx = iommu_alloc_ctx(iommu);
 	first_page = __pa(first_page);
 	while (npages--) {
-		iopte_val(*iopte) = (IOPTE_CONSISTENT(ctx) |
+		iopte_val(*iopte) = (IOPTE_CONSISTENT(0UL) |
 				     IOPTE_WRITE |
 				     (first_page & IOPTE_PAGE));
 		iopte++;
 		first_page += IO_PAGE_SIZE;
 	}
 
-	{
-		int i;
-		u32 daddr = *dma_addrp;
-
-		npages = size >> IO_PAGE_SHIFT;
-		for (i = 0; i < npages; i++) {
-			pci_iommu_write(iommu->iommu_flush, daddr);
-			daddr += IO_PAGE_SIZE;
-		}
-	}
-
-	spin_unlock_irqrestore(&iommu->lock, flags);
-
 	return ret;
 }
 
@@ -296,7 +272,7 @@ void pci_free_consistent(struct pci_dev *pdev, size_t size, void *cpu, dma_addr_
 	struct pcidev_cookie *pcp;
 	struct pci_iommu *iommu;
 	iopte_t *iopte;
-	unsigned long flags, order, npages, i, ctx;
+	unsigned long flags, order, npages;
 
 	npages = IO_PAGE_ALIGN(size) >> IO_PAGE_SHIFT;
 	pcp = pdev->sysdata;
@@ -306,46 +282,7 @@ void pci_free_consistent(struct pci_dev *pdev, size_t size, void *cpu, dma_addr_
 
 	spin_lock_irqsave(&iommu->lock, flags);
 
-	if ((iopte - iommu->page_table) ==
-	    iommu->lowest_consistent_map) {
-		iopte_t *walk = iopte + npages;
-		iopte_t *limit;
-
-		limit = (iommu->page_table +
-			 (1 << (iommu->page_table_sz_bits - PBM_LOGCLUSTERS)));
-		while (walk < limit) {
-			if (!IOPTE_IS_DUMMY(iommu, walk))
-				break;
-			walk++;
-		}
-		iommu->lowest_consistent_map =
-			(walk - iommu->page_table);
-	}
-
-	/* Data for consistent mappings cannot enter the streaming
-	 * buffers, so we only need to update the TSB.  We flush
-	 * the IOMMU here as well to prevent conflicts with the
-	 * streaming mapping deferred tlb flush scheme.
-	 */
-
-	ctx = 0;
-	if (iommu->iommu_ctxflush)
-		ctx = (iopte_val(*iopte) & IOPTE_CONTEXT) >> 47UL;
-
-	for (i = 0; i < npages; i++, iopte++)
-		iopte_make_dummy(iommu, iopte);
-
-	if (iommu->iommu_ctxflush) {
-		pci_iommu_write(iommu->iommu_ctxflush, ctx);
-	} else {
-		for (i = 0; i < npages; i++) {
-			u32 daddr = dvma + (i << IO_PAGE_SHIFT);
-
-			pci_iommu_write(iommu->iommu_flush, daddr);
-		}
-	}
-
-	iommu_free_ctx(iommu, ctx);
+	free_npages(iommu, dvma, npages);
 
 	spin_unlock_irqrestore(&iommu->lock, flags);
 
@@ -372,25 +309,27 @@ dma_addr_t pci_map_single(struct pci_dev *pdev, void *ptr, size_t sz, int direct
 	iommu = pcp->pbm->iommu;
 	strbuf = &pcp->pbm->stc;
 
-	if (direction == PCI_DMA_NONE)
-		BUG();
+	if (unlikely(direction == PCI_DMA_NONE))
+		goto bad_no_ctx;
 
 	oaddr = (unsigned long)ptr;
 	npages = IO_PAGE_ALIGN(oaddr + sz) - (oaddr & IO_PAGE_MASK);
 	npages >>= IO_PAGE_SHIFT;
 
 	spin_lock_irqsave(&iommu->lock, flags);
+	base = alloc_npages(iommu, npages);
+	ctx = 0;
+	if (iommu->iommu_ctxflush)
+		ctx = iommu_alloc_ctx(iommu);
+	spin_unlock_irqrestore(&iommu->lock, flags);
 
-	base = alloc_streaming_cluster(iommu, npages);
-	if (base == NULL)
+	if (unlikely(!base))
 		goto bad;
+
 	bus_addr = (iommu->page_table_map_base +
 		    ((base - iommu->page_table) << IO_PAGE_SHIFT));
 	ret = bus_addr | (oaddr & ~IO_PAGE_MASK);
 	base_paddr = __pa(oaddr & IO_PAGE_MASK);
-	ctx = 0;
-	if (iommu->iommu_ctxflush)
-		ctx = iommu_alloc_ctx(iommu);
 	if (strbuf->strbuf_enabled)
 		iopte_protection = IOPTE_STREAMING(ctx);
 	else
@@ -401,12 +340,13 @@ dma_addr_t pci_map_single(struct pci_dev *pdev, void *ptr, size_t sz, int direct
 	for (i = 0; i < npages; i++, base++, base_paddr += IO_PAGE_SIZE)
 		iopte_val(*base) = iopte_protection | base_paddr;
 
-	spin_unlock_irqrestore(&iommu->lock, flags);
-
 	return ret;
 
 bad:
-	spin_unlock_irqrestore(&iommu->lock, flags);
+	iommu_free_ctx(iommu, ctx);
+bad_no_ctx:
+	if (printk_ratelimit())
+		WARN_ON(1);
 	return PCI_DMA_ERROR_CODE;
 }
 
@@ -481,10 +421,13 @@ void pci_unmap_single(struct pci_dev *pdev, dma_addr_t bus_addr, size_t sz, int
 	struct pci_iommu *iommu;
 	struct pci_strbuf *strbuf;
 	iopte_t *base;
-	unsigned long flags, npages, ctx;
+	unsigned long flags, npages, ctx, i;
 
-	if (direction == PCI_DMA_NONE)
-		BUG();
+	if (unlikely(direction == PCI_DMA_NONE)) {
+		if (printk_ratelimit())
+			WARN_ON(1);
+		return;
+	}
 
 	pcp = pdev->sysdata;
 	iommu = pcp->pbm->iommu;
@@ -510,13 +453,14 @@ void pci_unmap_single(struct pci_dev *pdev, dma_addr_t bus_addr, size_t sz, int
 
 	/* Step 1: Kick data out of streaming buffers if necessary. */
 	if (strbuf->strbuf_enabled)
-		pci_strbuf_flush(strbuf, iommu, bus_addr, ctx, npages, direction);
+		pci_strbuf_flush(strbuf, iommu, bus_addr, ctx,
+				 npages, direction);
 
-	/* Step 2: Clear out first TSB entry. */
-	iopte_make_dummy(iommu, base);
+	/* Step 2: Clear out TSB entries. */
+	for (i = 0; i < npages; i++)
+		iopte_make_dummy(iommu, base + i);
 
-	free_streaming_cluster(iommu, bus_addr - iommu->page_table_map_base,
-			       npages, ctx);
+	free_npages(iommu, bus_addr - iommu->page_table_map_base, npages);
 
 	iommu_free_ctx(iommu, ctx);
 
@@ -621,6 +565,8 @@ int pci_map_sg(struct pci_dev *pdev, struct scatterlist *sglist, int nelems, int
 			pci_map_single(pdev,
 				       (page_address(sglist->page) + sglist->offset),
 				       sglist->length, direction);
+		if (unlikely(sglist->dma_address == PCI_DMA_ERROR_CODE))
+			return 0;
 		sglist->dma_length = sglist->length;
 		return 1;
 	}
@@ -629,21 +575,29 @@ int pci_map_sg(struct pci_dev *pdev, struct scatterlist *sglist, int nelems, int
 	iommu = pcp->pbm->iommu;
 	strbuf = &pcp->pbm->stc;
 	
-	if (direction == PCI_DMA_NONE)
-		BUG();
+	if (unlikely(direction == PCI_DMA_NONE))
+		goto bad_no_ctx;
 
 	/* Step 1: Prepare scatter list. */
 
 	npages = prepare_sg(sglist, nelems);
 
-	/* Step 2: Allocate a cluster. */
+	/* Step 2: Allocate a cluster and context, if necessary. */
 
 	spin_lock_irqsave(&iommu->lock, flags);
 
-	base = alloc_streaming_cluster(iommu, npages);
+	base = alloc_npages(iommu, npages);
+	ctx = 0;
+	if (iommu->iommu_ctxflush)
+		ctx = iommu_alloc_ctx(iommu);
+
+	spin_unlock_irqrestore(&iommu->lock, flags);
+
 	if (base == NULL)
 		goto bad;
-	dma_base = iommu->page_table_map_base + ((base - iommu->page_table) << IO_PAGE_SHIFT);
+
+	dma_base = iommu->page_table_map_base +
+		((base - iommu->page_table) << IO_PAGE_SHIFT);
 
 	/* Step 3: Normalize DMA addresses. */
 	used = nelems;
@@ -656,30 +610,28 @@ int pci_map_sg(struct pci_dev *pdev, struct scatterlist *sglist, int nelems, int
 	}
 	used = nelems - used;
 
-	/* Step 4: Choose a context if necessary. */
-	ctx = 0;
-	if (iommu->iommu_ctxflush)
-		ctx = iommu_alloc_ctx(iommu);
-
-	/* Step 5: Create the mappings. */
+	/* Step 4: Create the mappings. */
 	if (strbuf->strbuf_enabled)
 		iopte_protection = IOPTE_STREAMING(ctx);
 	else
 		iopte_protection = IOPTE_CONSISTENT(ctx);
 	if (direction != PCI_DMA_TODEVICE)
 		iopte_protection |= IOPTE_WRITE;
-	fill_sg (base, sglist, used, nelems, iopte_protection);
+
+	fill_sg(base, sglist, used, nelems, iopte_protection);
+
 #ifdef VERIFY_SG
 	verify_sglist(sglist, nelems, base, npages);
 #endif
 
-	spin_unlock_irqrestore(&iommu->lock, flags);
-
 	return used;
 
 bad:
-	spin_unlock_irqrestore(&iommu->lock, flags);
-	return PCI_DMA_ERROR_CODE;
+	iommu_free_ctx(iommu, ctx);
+bad_no_ctx:
+	if (printk_ratelimit())
+		WARN_ON(1);
+	return 0;
 }
 
 /* Unmap a set of streaming mode DMA translations. */
@@ -692,8 +644,10 @@ void pci_unmap_sg(struct pci_dev *pdev, struct scatterlist *sglist, int nelems,
 	unsigned long flags, ctx, i, npages;
 	u32 bus_addr;
 
-	if (direction == PCI_DMA_NONE)
-		BUG();
+	if (unlikely(direction == PCI_DMA_NONE)) {
+		if (printk_ratelimit())
+			WARN_ON(1);
+	}
 
 	pcp = pdev->sysdata;
 	iommu = pcp->pbm->iommu;
@@ -705,7 +659,8 @@ void pci_unmap_sg(struct pci_dev *pdev, struct scatterlist *sglist, int nelems,
 		if (sglist[i].dma_length == 0)
 			break;
 	i--;
-	npages = (IO_PAGE_ALIGN(sglist[i].dma_address + sglist[i].dma_length) - bus_addr) >> IO_PAGE_SHIFT;
+	npages = (IO_PAGE_ALIGN(sglist[i].dma_address + sglist[i].dma_length) -
+		  bus_addr) >> IO_PAGE_SHIFT;
 
 	base = iommu->page_table +
 		((bus_addr - iommu->page_table_map_base) >> IO_PAGE_SHIFT);
@@ -726,11 +681,11 @@ void pci_unmap_sg(struct pci_dev *pdev, struct scatterlist *sglist, int nelems,
 	if (strbuf->strbuf_enabled)
 		pci_strbuf_flush(strbuf, iommu, bus_addr, ctx, npages, direction);
 
-	/* Step 2: Clear out first TSB entry. */
-	iopte_make_dummy(iommu, base);
+	/* Step 2: Clear out the TSB entries. */
+	for (i = 0; i < npages; i++)
+		iopte_make_dummy(iommu, base + i);
 
-	free_streaming_cluster(iommu, bus_addr - iommu->page_table_map_base,
-			       npages, ctx);
+	free_npages(iommu, bus_addr - iommu->page_table_map_base, npages);
 
 	iommu_free_ctx(iommu, ctx);
 
diff --git a/arch/sparc64/kernel/pci_psycho.c b/arch/sparc64/kernel/pci_psycho.c
index 6ed1ef25e0a..c03ed5f49d3 100644
--- a/arch/sparc64/kernel/pci_psycho.c
+++ b/arch/sparc64/kernel/pci_psycho.c
@@ -1207,13 +1207,9 @@ static void psycho_scan_bus(struct pci_controller_info *p)
 static void psycho_iommu_init(struct pci_controller_info *p)
 {
 	struct pci_iommu *iommu = p->pbm_A.iommu;
-	unsigned long tsbbase, i;
+	unsigned long i;
 	u64 control;
 
-	/* Setup initial software IOMMU state. */
-	spin_lock_init(&iommu->lock);
-	iommu->ctx_lowest_free = 1;
-
 	/* Register addresses. */
 	iommu->iommu_control  = p->pbm_A.controller_regs + PSYCHO_IOMMU_CONTROL;
 	iommu->iommu_tsbbase  = p->pbm_A.controller_regs + PSYCHO_IOMMU_TSBBASE;
@@ -1240,40 +1236,10 @@ static void psycho_iommu_init(struct pci_controller_info *p)
 	/* Leave diag mode enabled for full-flushing done
 	 * in pci_iommu.c
 	 */
+	pci_iommu_table_init(iommu, IO_TSB_SIZE, 0xc0000000, 0xffffffff);
 
-	iommu->dummy_page = __get_free_pages(GFP_KERNEL, 0);
-	if (!iommu->dummy_page) {
-		prom_printf("PSYCHO_IOMMU: Error, gfp(dummy_page) failed.\n");
-		prom_halt();
-	}
-	memset((void *)iommu->dummy_page, 0, PAGE_SIZE);
-	iommu->dummy_page_pa = (unsigned long) __pa(iommu->dummy_page);
-
-	/* Using assumed page size 8K with 128K entries we need 1MB iommu page
-	 * table (128K ioptes * 8 bytes per iopte).  This is
-	 * page order 7 on UltraSparc.
-	 */
-	tsbbase = __get_free_pages(GFP_KERNEL, get_order(IO_TSB_SIZE));
-	if (!tsbbase) {
-		prom_printf("PSYCHO_IOMMU: Error, gfp(tsb) failed.\n");
-		prom_halt();
-	}
-	iommu->page_table = (iopte_t *)tsbbase;
-	iommu->page_table_sz_bits = 17;
-	iommu->page_table_map_base = 0xc0000000;
-	iommu->dma_addr_mask = 0xffffffff;
-	pci_iommu_table_init(iommu, IO_TSB_SIZE);
-
-	/* We start with no consistent mappings. */
-	iommu->lowest_consistent_map =
-		1 << (iommu->page_table_sz_bits - PBM_LOGCLUSTERS);
-
-	for (i = 0; i < PBM_NCLUSTERS; i++) {
-		iommu->alloc_info[i].flush = 0;
-		iommu->alloc_info[i].next = 0;
-	}
-
-	psycho_write(p->pbm_A.controller_regs + PSYCHO_IOMMU_TSBBASE, __pa(tsbbase));
+	psycho_write(p->pbm_A.controller_regs + PSYCHO_IOMMU_TSBBASE,
+		     __pa(iommu->page_table));
 
 	control = psycho_read(p->pbm_A.controller_regs + PSYCHO_IOMMU_CONTROL);
 	control &= ~(PSYCHO_IOMMU_CTRL_TSBSZ | PSYCHO_IOMMU_CTRL_TBWSZ);
@@ -1281,7 +1247,7 @@ static void psycho_iommu_init(struct pci_controller_info *p)
 	psycho_write(p->pbm_A.controller_regs + PSYCHO_IOMMU_CONTROL, control);
 
 	/* If necessary, hook us up for starfire IRQ translations. */
-	if(this_is_starfire)
+	if (this_is_starfire)
 		p->starfire_cookie = starfire_hookup(p->pbm_A.portid);
 	else
 		p->starfire_cookie = NULL;
diff --git a/arch/sparc64/kernel/pci_sabre.c b/arch/sparc64/kernel/pci_sabre.c
index 0ee6bd5b9ac..da8e1364194 100644
--- a/arch/sparc64/kernel/pci_sabre.c
+++ b/arch/sparc64/kernel/pci_sabre.c
@@ -1267,13 +1267,9 @@ static void sabre_iommu_init(struct pci_controller_info *p,
 			     u32 dma_mask)
 {
 	struct pci_iommu *iommu = p->pbm_A.iommu;
-	unsigned long tsbbase, i, order;
+	unsigned long i;
 	u64 control;
 
-	/* Setup initial software IOMMU state. */
-	spin_lock_init(&iommu->lock);
-	iommu->ctx_lowest_free = 1;
-
 	/* Register addresses. */
 	iommu->iommu_control  = p->pbm_A.controller_regs + SABRE_IOMMU_CONTROL;
 	iommu->iommu_tsbbase  = p->pbm_A.controller_regs + SABRE_IOMMU_TSBBASE;
@@ -1295,26 +1291,10 @@ static void sabre_iommu_init(struct pci_controller_info *p,
 	/* Leave diag mode enabled for full-flushing done
 	 * in pci_iommu.c
 	 */
+	pci_iommu_table_init(iommu, tsbsize * 1024 * 8, dvma_offset, dma_mask);
 
-	iommu->dummy_page = __get_free_pages(GFP_KERNEL, 0);
-	if (!iommu->dummy_page) {
-		prom_printf("PSYCHO_IOMMU: Error, gfp(dummy_page) failed.\n");
-		prom_halt();
-	}
-	memset((void *)iommu->dummy_page, 0, PAGE_SIZE);
-	iommu->dummy_page_pa = (unsigned long) __pa(iommu->dummy_page);
-
-	tsbbase = __get_free_pages(GFP_KERNEL, order = get_order(tsbsize * 1024 * 8));
-	if (!tsbbase) {
-		prom_printf("SABRE_IOMMU: Error, gfp(tsb) failed.\n");
-		prom_halt();
-	}
-	iommu->page_table = (iopte_t *)tsbbase;
-	iommu->page_table_map_base = dvma_offset;
-	iommu->dma_addr_mask = dma_mask;
-	pci_iommu_table_init(iommu, PAGE_SIZE << order);
-
-	sabre_write(p->pbm_A.controller_regs + SABRE_IOMMU_TSBBASE, __pa(tsbbase));
+	sabre_write(p->pbm_A.controller_regs + SABRE_IOMMU_TSBBASE,
+		    __pa(iommu->page_table));
 
 	control = sabre_read(p->pbm_A.controller_regs + SABRE_IOMMU_CONTROL);
 	control &= ~(SABRE_IOMMUCTRL_TSBSZ | SABRE_IOMMUCTRL_TBWSZ);
@@ -1322,11 +1302,9 @@ static void sabre_iommu_init(struct pci_controller_info *p,
 	switch(tsbsize) {
 	case 64:
 		control |= SABRE_IOMMU_TSBSZ_64K;
-		iommu->page_table_sz_bits = 16;
 		break;
 	case 128:
 		control |= SABRE_IOMMU_TSBSZ_128K;
-		iommu->page_table_sz_bits = 17;
 		break;
 	default:
 		prom_printf("iommu_init: Illegal TSB size %d\n", tsbsize);
@@ -1334,15 +1312,6 @@ static void sabre_iommu_init(struct pci_controller_info *p,
 		break;
 	}
 	sabre_write(p->pbm_A.controller_regs + SABRE_IOMMU_CONTROL, control);
-
-	/* We start with no consistent mappings. */
-	iommu->lowest_consistent_map =
-		1 << (iommu->page_table_sz_bits - PBM_LOGCLUSTERS);
-
-	for (i = 0; i < PBM_NCLUSTERS; i++) {
-		iommu->alloc_info[i].flush = 0;
-		iommu->alloc_info[i].next = 0;
-	}
 }
 
 static void pbm_register_toplevel_resources(struct pci_controller_info *p,
diff --git a/arch/sparc64/kernel/pci_schizo.c b/arch/sparc64/kernel/pci_schizo.c
index cae5b61fe2f..d8c4e0919b4 100644
--- a/arch/sparc64/kernel/pci_schizo.c
+++ b/arch/sparc64/kernel/pci_schizo.c
@@ -1765,7 +1765,7 @@ static void schizo_pbm_strbuf_init(struct pci_pbm_info *pbm)
 static void schizo_pbm_iommu_init(struct pci_pbm_info *pbm)
 {
 	struct pci_iommu *iommu = pbm->iommu;
-	unsigned long tsbbase, i, tagbase, database, order;
+	unsigned long i, tagbase, database;
 	u32 vdma[2], dma_mask;
 	u64 control;
 	int err, tsbsize;
@@ -1800,10 +1800,6 @@ static void schizo_pbm_iommu_init(struct pci_pbm_info *pbm)
 			prom_halt();
 	};
 
-	/* Setup initial software IOMMU state. */
-	spin_lock_init(&iommu->lock);
-	iommu->ctx_lowest_free = 1;
-
 	/* Register addresses, SCHIZO has iommu ctx flushing. */
 	iommu->iommu_control  = pbm->pbm_regs + SCHIZO_IOMMU_CONTROL;
 	iommu->iommu_tsbbase  = pbm->pbm_regs + SCHIZO_IOMMU_TSBBASE;
@@ -1832,56 +1828,9 @@ static void schizo_pbm_iommu_init(struct pci_pbm_info *pbm)
 	/* Leave diag mode enabled for full-flushing done
 	 * in pci_iommu.c
 	 */
+	pci_iommu_table_init(iommu, tsbsize * 8 * 1024, vdma[0], dma_mask);
 
-	iommu->dummy_page = __get_free_pages(GFP_KERNEL, 0);
-	if (!iommu->dummy_page) {
-		prom_printf("PSYCHO_IOMMU: Error, gfp(dummy_page) failed.\n");
-		prom_halt();
-	}
-	memset((void *)iommu->dummy_page, 0, PAGE_SIZE);
-	iommu->dummy_page_pa = (unsigned long) __pa(iommu->dummy_page);
-
-	/* Using assumed page size 8K with 128K entries we need 1MB iommu page
-	 * table (128K ioptes * 8 bytes per iopte).  This is
-	 * page order 7 on UltraSparc.
-	 */
-	order = get_order(tsbsize * 8 * 1024);
-	tsbbase = __get_free_pages(GFP_KERNEL, order);
-	if (!tsbbase) {
-		prom_printf("%s: Error, gfp(tsb) failed.\n", pbm->name);
-		prom_halt();
-	}
-
-	iommu->page_table = (iopte_t *)tsbbase;
-	iommu->page_table_map_base = vdma[0];
-	iommu->dma_addr_mask = dma_mask;
-	pci_iommu_table_init(iommu, PAGE_SIZE << order);
-
-	switch (tsbsize) {
-	case 64:
-		iommu->page_table_sz_bits = 16;
-		break;
-
-	case 128:
-		iommu->page_table_sz_bits = 17;
-		break;
-
-	default:
-		prom_printf("iommu_init: Illegal TSB size %d\n", tsbsize);
-		prom_halt();
-		break;
-	};
-
-	/* We start with no consistent mappings. */
-	iommu->lowest_consistent_map =
-		1 << (iommu->page_table_sz_bits - PBM_LOGCLUSTERS);
-
-	for (i = 0; i < PBM_NCLUSTERS; i++) {
-		iommu->alloc_info[i].flush = 0;
-		iommu->alloc_info[i].next = 0;
-	}
-
-	schizo_write(iommu->iommu_tsbbase, __pa(tsbbase));
+	schizo_write(iommu->iommu_tsbbase, __pa(iommu->page_table));
 
 	control = schizo_read(iommu->iommu_control);
 	control &= ~(SCHIZO_IOMMU_CTRL_TSBSZ | SCHIZO_IOMMU_CTRL_TBWSZ);
diff --git a/arch/sparc64/kernel/smp.c b/arch/sparc64/kernel/smp.c
index 590df5a16f5..b137fd63f5e 100644
--- a/arch/sparc64/kernel/smp.c
+++ b/arch/sparc64/kernel/smp.c
@@ -1001,13 +1001,6 @@ void smp_penguin_jailcell(int irq, struct pt_regs *regs)
 	preempt_enable();
 }
 
-extern unsigned long xcall_promstop;
-
-void smp_promstop_others(void)
-{
-	smp_cross_call(&xcall_promstop, 0, 0, 0);
-}
-
 #define prof_multiplier(__cpu)		cpu_data(__cpu).multiplier
 #define prof_counter(__cpu)		cpu_data(__cpu).counter
 
diff --git a/arch/sparc64/mm/ultra.S b/arch/sparc64/mm/ultra.S
index 058b8126c1a..e4c9151fa11 100644
--- a/arch/sparc64/mm/ultra.S
+++ b/arch/sparc64/mm/ultra.S
@@ -453,22 +453,6 @@ xcall_flush_dcache_page_spitfire: /* %g1 == physical page address
 	nop
 	nop
 
-	.globl		xcall_promstop
-xcall_promstop:
-	rdpr		%pstate, %g2
-	wrpr		%g2, PSTATE_IG | PSTATE_AG, %pstate
-	rdpr		%pil, %g2
-	wrpr		%g0, 15, %pil
-	sethi		%hi(109f), %g7
-	b,pt		%xcc, etrap_irq
-109:	 or		%g7, %lo(109b), %g7
-	flushw
-	call		prom_stopself
-	 nop
-	/* We should not return, just spin if we do... */
-1:	b,a,pt		%xcc, 1b
-	nop
-
 	.data
 
 errata32_hwbug:
diff --git a/arch/sparc64/prom/misc.c b/arch/sparc64/prom/misc.c
index 9b895faf077..87f5cfce23b 100644
--- a/arch/sparc64/prom/misc.c
+++ b/arch/sparc64/prom/misc.c
@@ -68,19 +68,11 @@ void prom_cmdline(void)
 	local_irq_restore(flags);
 }
 
-#ifdef CONFIG_SMP
-extern void smp_promstop_others(void);
-#endif
-
 /* Drop into the prom, but completely terminate the program.
  * No chance of continuing.
  */
 void prom_halt(void)
 {
-#ifdef CONFIG_SMP
-	smp_promstop_others();
-	udelay(8000);
-#endif
 again:
 	p1275_cmd("exit", P1275_INOUT(0, 0));
 	goto again; /* PROM is out to get me -DaveM */
@@ -88,10 +80,6 @@ again:
 
 void prom_halt_power_off(void)
 {
-#ifdef CONFIG_SMP
-	smp_promstop_others();
-	udelay(8000);
-#endif
 	p1275_cmd("SUNW,power-off", P1275_INOUT(0, 0));
 
 	/* if nothing else helps, we just halt */
diff --git a/drivers/acpi/event.c b/drivers/acpi/event.c
index bfa8b76de40..2dbb1b0f11d 100644
--- a/drivers/acpi/event.c
+++ b/drivers/acpi/event.c
@@ -58,9 +58,8 @@ acpi_system_read_event(struct file *file, char __user * buffer, size_t count,
 			return_VALUE(-EAGAIN);
 
 		result = acpi_bus_receive_event(&event);
-		if (result) {
-			return_VALUE(-EIO);
-		}
+		if (result)
+			return_VALUE(result);
 
 		chars_remaining = sprintf(str, "%s %s %08x %08x\n",
 					  event.device_class ? event.
diff --git a/drivers/char/mbcs.c b/drivers/char/mbcs.c
index 3fa64c63110..c268ee04b2a 100644
--- a/drivers/char/mbcs.c
+++ b/drivers/char/mbcs.c
@@ -830,6 +830,9 @@ static int __init mbcs_init(void)
 {
 	int rv;
 
+	if (!ia64_platform_is("sn2"))
+		return -ENODEV;
+
 	// Put driver into chrdevs[].  Get major number.
 	rv = register_chrdev(mbcs_major, DEVICE_NAME, &mbcs_ops);
 	if (rv < 0) {
diff --git a/drivers/char/n_r3964.c b/drivers/char/n_r3964.c
index 97d6dc24b80..853c98cee64 100644
--- a/drivers/char/n_r3964.c
+++ b/drivers/char/n_r3964.c
@@ -695,7 +695,7 @@ static void receive_char(struct r3964_info *pInfo, const unsigned char c)
             {
                TRACE_PE("IDLE - got STX but no space in rx_queue!");
                pInfo->state=R3964_WAIT_FOR_RX_BUF;
-	       mod_timer(&pInfo->tmr, R3964_TO_NO_BUF);
+	       mod_timer(&pInfo->tmr, jiffies + R3964_TO_NO_BUF);
                break;
             }
 start_receiving:
@@ -705,7 +705,7 @@ start_receiving:
             pInfo->last_rx = 0;
             pInfo->flags &= ~R3964_ERROR;
             pInfo->state=R3964_RECEIVING;
-	    mod_timer(&pInfo->tmr, R3964_TO_ZVZ);
+	    mod_timer(&pInfo->tmr, jiffies + R3964_TO_ZVZ);
 	    pInfo->nRetry = 0;
             put_char(pInfo, DLE);
             flush(pInfo);
@@ -732,7 +732,7 @@ start_receiving:
                if(pInfo->flags & R3964_BCC)
                {
                   pInfo->state = R3964_WAIT_FOR_BCC;
-		  mod_timer(&pInfo->tmr, R3964_TO_ZVZ);
+		  mod_timer(&pInfo->tmr, jiffies + R3964_TO_ZVZ);
                }
                else 
                {
@@ -744,7 +744,7 @@ start_receiving:
                pInfo->last_rx = c;
 char_to_buf:
                pInfo->rx_buf[pInfo->rx_position++] = c;
-	       mod_timer(&pInfo->tmr, R3964_TO_ZVZ);
+	       mod_timer(&pInfo->tmr, jiffies + R3964_TO_ZVZ);
             }
          }
         /* else: overflow-msg? BUF_SIZE>MTU; should not happen? */ 
diff --git a/drivers/input/keyboard/Kconfig b/drivers/input/keyboard/Kconfig
index 444f7756fee..571a68691a4 100644
--- a/drivers/input/keyboard/Kconfig
+++ b/drivers/input/keyboard/Kconfig
@@ -93,7 +93,7 @@ config KEYBOARD_LKKBD
 
 config KEYBOARD_LOCOMO
 	tristate "LoCoMo Keyboard Support"
-	depends on SHARP_LOCOMO
+	depends on SHARP_LOCOMO && INPUT_KEYBOARD
 	help
 	  Say Y here if you are running Linux on a Sharp Zaurus Collie or Poodle based PDA
 
diff --git a/drivers/input/keyboard/spitzkbd.c b/drivers/input/keyboard/spitzkbd.c
index 1714045a182..344f4600540 100644
--- a/drivers/input/keyboard/spitzkbd.c
+++ b/drivers/input/keyboard/spitzkbd.c
@@ -53,7 +53,7 @@ static unsigned char spitzkbd_keycode[NR_SCANCODES] = {
 	KEY_LEFTCTRL, KEY_1, KEY_3, KEY_5, KEY_6, KEY_7, KEY_9, KEY_0, KEY_BACKSPACE, SPITZ_KEY_EXOK, SPITZ_KEY_EXCANCEL, 0, 0, 0, 0, 0,  /* 1-16 */
 	0, KEY_2, KEY_4, KEY_R, KEY_Y, KEY_8, KEY_I, KEY_O, KEY_P, SPITZ_KEY_EXJOGDOWN, SPITZ_KEY_EXJOGUP, 0, 0, 0, 0, 0, /* 17-32 */
 	KEY_TAB, KEY_Q, KEY_E, KEY_T, KEY_G, KEY_U, KEY_J, KEY_K, 0, 0, 0, 0, 0, 0, 0, 0,                                 /* 33-48 */
-	SPITZ_KEY_CALENDER, KEY_W, KEY_S, KEY_F, KEY_V, KEY_H, KEY_M, KEY_L, 0, 0, KEY_RIGHTSHIFT, 0, 0, 0, 0, 0,         /* 49-64 */
+	SPITZ_KEY_CALENDER, KEY_W, KEY_S, KEY_F, KEY_V, KEY_H, KEY_M, KEY_L, 0, KEY_RIGHTSHIFT, 0, 0, 0, 0, 0, 0,         /* 49-64 */
 	SPITZ_KEY_ADDRESS, KEY_A, KEY_D, KEY_C, KEY_B, KEY_N, KEY_DOT, 0, KEY_ENTER, KEY_LEFTSHIFT, 0, 0, 0, 0, 0, 0,	  /* 65-80 */
 	SPITZ_KEY_MAIL, KEY_Z, KEY_X, KEY_MINUS, KEY_SPACE, KEY_COMMA, 0, KEY_UP, 0, 0, SPITZ_KEY_FN, 0, 0, 0, 0, 0,      /* 81-96 */
 	KEY_SYSRQ, SPITZ_KEY_JAP1, SPITZ_KEY_JAP2, SPITZ_KEY_CANCEL, SPITZ_KEY_OK, SPITZ_KEY_MENU, KEY_LEFT, KEY_DOWN, KEY_RIGHT, 0, 0, 0, 0, 0, 0, 0  /* 97-112 */
diff --git a/drivers/input/misc/uinput.c b/drivers/input/misc/uinput.c
index d5c5b32045a..4015a91f4b6 100644
--- a/drivers/input/misc/uinput.c
+++ b/drivers/input/misc/uinput.c
@@ -90,11 +90,11 @@ static inline int uinput_request_reserve_slot(struct uinput_device *udev, struct
 
 static void uinput_request_done(struct uinput_device *udev, struct uinput_request *request)
 {
-	complete(&request->done);
-
 	/* Mark slot as available */
 	udev->requests[request->id] = NULL;
 	wake_up_interruptible(&udev->requests_waitq);
+
+	complete(&request->done);
 }
 
 static int uinput_request_submit(struct input_dev *dev, struct uinput_request *request)
diff --git a/drivers/media/radio/radio-cadet.c b/drivers/media/radio/radio-cadet.c
index 022913da8c5..9b0406318f2 100644
--- a/drivers/media/radio/radio-cadet.c
+++ b/drivers/media/radio/radio-cadet.c
@@ -543,7 +543,7 @@ static int cadet_probe(void)
 
 	for(i=0;i<8;i++) {
 	        io=iovals[i];
-	        if(request_region(io,2, "cadet-probe")>=0) {
+		if (request_region(io, 2, "cadet-probe")) {
 		        cadet_setfreq(1410);
 			if(cadet_getfreq()==1410) {
 				release_region(io, 2);
diff --git a/drivers/media/video/vpx3220.c b/drivers/media/video/vpx3220.c
index 4437bdebe24..137b58f2c66 100644
--- a/drivers/media/video/vpx3220.c
+++ b/drivers/media/video/vpx3220.c
@@ -203,7 +203,7 @@ static const unsigned short init_ntsc[] = {
 	0x8c, 640,		/* Horizontal length */
 	0x8d, 640,		/* Number of pixels */
 	0x8f, 0xc00,		/* Disable window 2 */
-	0xf0, 0x173,		/* 13.5 MHz transport, Forced
+	0xf0, 0x73,		/* 13.5 MHz transport, Forced
 				 * mode, latch windows */
 	0xf2, 0x13,		/* NTSC M, composite input */
 	0xe7, 0x1e1,		/* Enable vertical standard
@@ -212,38 +212,36 @@ static const unsigned short init_ntsc[] = {
 
 static const unsigned short init_pal[] = {
 	0x88, 23,		/* Window 1 vertical begin */
-	0x89, 288 + 16,		/* Vertical lines in (16 lines
+	0x89, 288,		/* Vertical lines in (16 lines
 				 * skipped by the VFE) */
-	0x8a, 288 + 16,		/* Vertical lines out (16 lines
+	0x8a, 288,		/* Vertical lines out (16 lines
 				 * skipped by the VFE) */
 	0x8b, 16,		/* Horizontal begin */
 	0x8c, 768,		/* Horizontal length */
 	0x8d, 784, 		/* Number of pixels
 				 * Must be >= Horizontal begin + Horizontal length */
 	0x8f, 0xc00,		/* Disable window 2 */
-	0xf0, 0x177,		/* 13.5 MHz transport, Forced
+	0xf0, 0x77,		/* 13.5 MHz transport, Forced
 				 * mode, latch windows */
 	0xf2, 0x3d1,		/* PAL B,G,H,I, composite input */
-	0xe7, 0x261,		/* PAL/SECAM set to 288 + 16 lines 
-				 * change to 0x241 for 288 lines */
+	0xe7, 0x241,		/* PAL/SECAM set to 288 lines */
 };
 
 static const unsigned short init_secam[] = {
-	0x88, 23  - 16,		/* Window 1 vertical begin */
-	0x89, 288 + 16,		/* Vertical lines in (16 lines
+	0x88, 23,		/* Window 1 vertical begin */
+	0x89, 288,		/* Vertical lines in (16 lines
 				 * skipped by the VFE) */
-	0x8a, 288 + 16,		/* Vertical lines out (16 lines
+	0x8a, 288,		/* Vertical lines out (16 lines
 				 * skipped by the VFE) */
 	0x8b, 16,		/* Horizontal begin */
 	0x8c, 768,		/* Horizontal length */
 	0x8d, 784,		/* Number of pixels
 				 * Must be >= Horizontal begin + Horizontal length */
 	0x8f, 0xc00,		/* Disable window 2 */
-	0xf0, 0x177,		/* 13.5 MHz transport, Forced
+	0xf0, 0x77,		/* 13.5 MHz transport, Forced
 				 * mode, latch windows */
 	0xf2, 0x3d5,		/* SECAM, composite input */
-	0xe7, 0x261,		/* PAL/SECAM set to 288 + 16 lines 
-				 * change to 0x241 for 288 lines */
+	0xe7, 0x241,		/* PAL/SECAM set to 288 lines */
 };
 
 static const unsigned char init_common[] = {
@@ -410,6 +408,12 @@ vpx3220_command (struct i2c_client *client,
 	case DECODER_SET_NORM:
 	{
 		int *iarg = arg, data;
+		int temp_input;
+
+		/* Here we back up the input selection because it gets
+		   overwritten when we fill the registers with the
+                   choosen video norm */
+		temp_input = vpx3220_fp_read(client, 0xf2);
 
 		dprintk(1, KERN_DEBUG "%s: DECODER_SET_NORM %d\n",
 			I2C_NAME(client), *iarg);
@@ -449,6 +453,10 @@ vpx3220_command (struct i2c_client *client,
 
 		}
 		decoder->norm = *iarg;
+
+		/* And here we set the backed up video input again */
+		vpx3220_fp_write(client, 0xf2, temp_input | 0x0010);
+		udelay(10);
 	}
 		break;
 
diff --git a/drivers/net/wireless/Kconfig b/drivers/net/wireless/Kconfig
index 00a07f32a81..7187958e40c 100644
--- a/drivers/net/wireless/Kconfig
+++ b/drivers/net/wireless/Kconfig
@@ -243,7 +243,7 @@ config IPW_DEBUG
 
 config AIRO
 	tristate "Cisco/Aironet 34X/35X/4500/4800 ISA and PCI cards"
-	depends on NET_RADIO && ISA && (PCI || BROKEN)
+	depends on NET_RADIO && ISA_DMA_API && (PCI || BROKEN)
 	---help---
 	  This is the standard Linux driver to support Cisco/Aironet ISA and
 	  PCI 802.11 wireless cards.
diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
index 11ca44387cb..a6a630a950d 100644
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -1233,7 +1233,7 @@ static void __init quirk_alder_ioapic(struct pci_dev *pdev)
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL,	PCI_DEVICE_ID_INTEL_EESSC,	quirk_alder_ioapic );
 #endif
 
-#ifdef CONFIG_SCSI_SATA
+#ifdef CONFIG_SCSI_SATA_INTEL_COMBINED
 static void __devinit quirk_intel_ide_combined(struct pci_dev *pdev)
 {
 	u8 prog, comb, tmp;
@@ -1310,7 +1310,7 @@ static void __devinit quirk_intel_ide_combined(struct pci_dev *pdev)
 		request_region(0x170, 8, "libata");	/* port 1 */
 }
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL,    PCI_ANY_ID,	  quirk_intel_ide_combined );
-#endif /* CONFIG_SCSI_SATA */
+#endif /* CONFIG_SCSI_SATA_INTEL_COMBINED */
 
 
 int pcie_mch_quirk;
diff --git a/drivers/pcmcia/soc_common.c b/drivers/pcmcia/soc_common.c
index 888b70e6a48..9e7ccd8a432 100644
--- a/drivers/pcmcia/soc_common.c
+++ b/drivers/pcmcia/soc_common.c
@@ -66,7 +66,7 @@ void soc_pcmcia_debug(struct soc_pcmcia_socket *skt, const char *func,
 	if (pc_debug > lvl) {
 		printk(KERN_DEBUG "skt%u: %s: ", skt->nr, func);
 		va_start(args, fmt);
-		printk(fmt, args);
+		vprintk(fmt, args);
 		va_end(args);
 	}
 }
@@ -321,8 +321,6 @@ soc_common_pcmcia_get_socket(struct pcmcia_socket *sock, socket_state_t *state)
  * less punt all of this work and let the kernel handle the details
  * of power configuration, reset, &c. We also record the value of
  * `state' in order to regurgitate it to the PCMCIA core later.
- *
- * Returns: 0
  */
 static int
 soc_common_pcmcia_set_socket(struct pcmcia_socket *sock, socket_state_t *state)
@@ -407,7 +405,7 @@ soc_common_pcmcia_set_io_map(struct pcmcia_socket *sock, struct pccard_io_map *m
  * the map speed as requested, but override the address ranges
  * supplied by Card Services.
  *
- * Returns: 0 on success, -1 on error
+ * Returns: 0 on success, -ERRNO on error
  */
 static int
 soc_common_pcmcia_set_mem_map(struct pcmcia_socket *sock, struct pccard_mem_map *map)
@@ -655,8 +653,8 @@ static void soc_pcmcia_cpufreq_unregister(void)
 }
 
 #else
-#define soc_pcmcia_cpufreq_register()
-#define soc_pcmcia_cpufreq_unregister()
+static int soc_pcmcia_cpufreq_register(void) { return 0; }
+static void soc_pcmcia_cpufreq_unregister(void) {}
 #endif
 
 int soc_common_drv_pcmcia_probe(struct device *dev, struct pcmcia_low_level *ops, int first, int nr)
@@ -738,7 +736,7 @@ int soc_common_drv_pcmcia_probe(struct device *dev, struct pcmcia_low_level *ops
 			goto out_err_5;
 		}
 
-		if ( list_empty(&soc_pcmcia_sockets) )
+		if (list_empty(&soc_pcmcia_sockets))
 			soc_pcmcia_cpufreq_register();
 
 		list_add(&skt->node, &soc_pcmcia_sockets);
@@ -839,7 +837,7 @@ int soc_common_drv_pcmcia_remove(struct device *dev)
 		release_resource(&skt->res_io);
 		release_resource(&skt->res_skt);
 	}
-	if ( list_empty(&soc_pcmcia_sockets) )
+	if (list_empty(&soc_pcmcia_sockets))
 		soc_pcmcia_cpufreq_unregister();
 
 	up(&soc_pcmcia_sockets_lock);
diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig
index e2845b44277..9c9f162bd6e 100644
--- a/drivers/scsi/Kconfig
+++ b/drivers/scsi/Kconfig
@@ -569,6 +569,11 @@ config SCSI_SATA_VITESSE
 
 	  If unsure, say N.
 
+config SCSI_SATA_INTEL_COMBINED
+	bool
+	depends on IDE=y && !BLK_DEV_IDE_SATA && (SCSI_SATA_AHCI || SCSI_ATA_PIIX)
+	default y
+
 config SCSI_BUSLOGIC
 	tristate "BusLogic SCSI support"
 	depends on (PCI || ISA || MCA) && SCSI && ISA_DMA_API
diff --git a/drivers/scsi/aacraid/linit.c b/drivers/scsi/aacraid/linit.c
index de8490a9283..a1f9ceef0ac 100644
--- a/drivers/scsi/aacraid/linit.c
+++ b/drivers/scsi/aacraid/linit.c
@@ -453,9 +453,9 @@ static int aac_eh_reset(struct scsi_cmnd* cmd)
 		/*
 		 * We can exit If all the commands are complete
 		 */
+		spin_unlock_irq(host->host_lock);
 		if (active == 0)
 			return SUCCESS;
-		spin_unlock_irq(host->host_lock);
 		ssleep(1);
 		spin_lock_irq(host->host_lock);
 	}
diff --git a/drivers/scsi/qlogicpti.c b/drivers/scsi/qlogicpti.c
index a917ab7475a..1fd5fc6d0fe 100644
--- a/drivers/scsi/qlogicpti.c
+++ b/drivers/scsi/qlogicpti.c
@@ -1119,6 +1119,36 @@ static inline void update_can_queue(struct Scsi_Host *host, u_int in_ptr, u_int
 	host->sg_tablesize = QLOGICPTI_MAX_SG(num_free);
 }
 
+static unsigned int scsi_rbuf_get(struct scsi_cmnd *cmd, unsigned char **buf_out)
+{
+	unsigned char *buf;
+	unsigned int buflen;
+
+	if (cmd->use_sg) {
+		struct scatterlist *sg;
+
+		sg = (struct scatterlist *) cmd->request_buffer;
+		buf = kmap_atomic(sg->page, KM_IRQ0) + sg->offset;
+		buflen = sg->length;
+	} else {
+		buf = cmd->request_buffer;
+		buflen = cmd->request_bufflen;
+	}
+
+	*buf_out = buf;
+	return buflen;
+}
+
+static void scsi_rbuf_put(struct scsi_cmnd *cmd, unsigned char *buf)
+{
+	if (cmd->use_sg) {
+		struct scatterlist *sg;
+
+		sg = (struct scatterlist *) cmd->request_buffer;
+		kunmap_atomic(buf - sg->offset, KM_IRQ0);
+	}
+}
+
 /*
  * Until we scan the entire bus with inquiries, go throught this fella...
  */
@@ -1145,11 +1175,9 @@ static void ourdone(struct scsi_cmnd *Cmnd)
 		int ok = host_byte(Cmnd->result) == DID_OK;
 		if (Cmnd->cmnd[0] == 0x12 && ok) {
 			unsigned char *iqd;
+			unsigned int iqd_len;
 
-			if (Cmnd->use_sg != 0)
-				BUG();
-
-			iqd = ((unsigned char *)Cmnd->buffer);
+			iqd_len = scsi_rbuf_get(Cmnd, &iqd);
 
 			/* tags handled in midlayer */
 			/* enable sync mode? */
@@ -1163,6 +1191,9 @@ static void ourdone(struct scsi_cmnd *Cmnd)
 			if (iqd[7] & 0x20) {
 				qpti->dev_param[tgt].device_flags |= 0x20;
 			}
+
+			scsi_rbuf_put(Cmnd, iqd);
+
 			qpti->sbits |= (1 << tgt);
 		} else if (!ok) {
 			qpti->sbits |= (1 << tgt);
diff --git a/drivers/serial/8250_pnp.c b/drivers/serial/8250_pnp.c
index 6b321e82caf..c2786fc41cc 100644
--- a/drivers/serial/8250_pnp.c
+++ b/drivers/serial/8250_pnp.c
@@ -272,6 +272,8 @@ static const struct pnp_device_id pnp_dev_table[] = {
 	{	"SUP1421",		0	},
 	/* SupraExpress 33.6 Data/Fax PnP modem */
 	{	"SUP1590",		0	},
+	/* SupraExpress 336i Sp ASVD */
+	{	"SUP1620",		0	},
 	/* SupraExpress 33.6 Data/Fax PnP modem */
 	{	"SUP1760",		0	},
 	/* Phoebe Micro */
diff --git a/drivers/serial/sh-sci.c b/drivers/serial/sh-sci.c
index 51226630786..430754ebac8 100644
--- a/drivers/serial/sh-sci.c
+++ b/drivers/serial/sh-sci.c
@@ -967,7 +967,7 @@ static int sci_startup(struct uart_port *port)
 #endif
 
 	sci_request_irq(s);
-	sci_start_tx(port, 1);
+	sci_start_tx(port);
 	sci_start_rx(port, 1);
 
 	return 0;
diff --git a/drivers/usb/host/isp116x-hcd.c b/drivers/usb/host/isp116x-hcd.c
index 41bbae83fc7..e142056b0d2 100644
--- a/drivers/usb/host/isp116x-hcd.c
+++ b/drivers/usb/host/isp116x-hcd.c
@@ -326,7 +326,8 @@ static void postproc_atl_queue(struct isp116x *isp116x)
 					usb_settoggle(udev, ep->epnum,
 						      ep->nextpid ==
 						      USB_PID_OUT,
-						      PTD_GET_TOGGLE(ptd) ^ 1);
+						      PTD_GET_TOGGLE(ptd));
+				urb->actual_length += PTD_GET_COUNT(ptd);
 				urb->status = cc_to_error[TD_DATAUNDERRUN];
 				spin_unlock(&urb->lock);
 				continue;
diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c
index a99865c689c..41f92b92476 100644
--- a/drivers/usb/input/hid-core.c
+++ b/drivers/usb/input/hid-core.c
@@ -1702,10 +1702,7 @@ static struct hid_device *usb_hid_configure(struct usb_interface *intf)
 		if ((endpoint->bmAttributes & 3) != 3)		/* Not an interrupt endpoint */
 			continue;
 
-		/* handle potential highspeed HID correctly */
 		interval = endpoint->bInterval;
-		if (dev->speed == USB_SPEED_HIGH)
-			interval = 1 << (interval - 1);
 
 		/* Change the polling interval of mice. */
 		if (hid->collection->usage == HID_GD_MOUSE && hid_mousepoll_interval > 0)
diff --git a/drivers/usb/serial/generic.c b/drivers/usb/serial/generic.c
index ddde5fb13f6..5f7d3193d35 100644
--- a/drivers/usb/serial/generic.c
+++ b/drivers/usb/serial/generic.c
@@ -223,7 +223,7 @@ int usb_serial_generic_write_room (struct usb_serial_port *port)
 	dbg("%s - port %d", __FUNCTION__, port->number);
 
 	if (serial->num_bulk_out) {
-		if (port->write_urb_busy)
+		if (!(port->write_urb_busy))
 			room = port->bulk_out_size;
 	}
 
diff --git a/drivers/video/console/vgacon.c b/drivers/video/console/vgacon.c
index 6ef6f7760e4..809fee2140a 100644
--- a/drivers/video/console/vgacon.c
+++ b/drivers/video/console/vgacon.c
@@ -565,7 +565,11 @@ static int vgacon_switch(struct vc_data *c)
 		scr_memcpyw((u16 *) c->vc_origin, (u16 *) c->vc_screenbuf,
 			    c->vc_screenbuf_size > vga_vram_size ?
 				vga_vram_size : c->vc_screenbuf_size);
-		vgacon_doresize(c, c->vc_cols, c->vc_rows);
+		if (!(vga_video_num_columns % 2) &&
+		    vga_video_num_columns <= ORIG_VIDEO_COLS &&
+		    vga_video_num_lines <= (ORIG_VIDEO_LINES *
+			vga_default_font_height) / c->vc_font.height)
+			vgacon_doresize(c, c->vc_cols, c->vc_rows);
 	}
 
 	return 0;		/* Redrawing not needed */
@@ -1023,7 +1027,8 @@ static int vgacon_resize(struct vc_data *c, unsigned int width,
 	if (width % 2 || width > ORIG_VIDEO_COLS ||
 	    height > (ORIG_VIDEO_LINES * vga_default_font_height)/
 	    c->vc_font.height)
-		return -EINVAL;
+		/* let svgatextmode tinker with video timings */
+		return 0;
 
 	if (CON_IS_VISIBLE(c) && !vga_is_gfx) /* who knows */
 		vgacon_doresize(c, width, height);
diff --git a/drivers/video/sa1100fb.c b/drivers/video/sa1100fb.c
index beeec7b5142..8000890e427 100644
--- a/drivers/video/sa1100fb.c
+++ b/drivers/video/sa1100fb.c
@@ -592,6 +592,7 @@ sa1100fb_setcolreg(u_int regno, u_int red, u_int green, u_int blue,
 	return ret;
 }
 
+#ifdef CONFIG_CPU_FREQ
 /*
  *  sa1100fb_display_dma_period()
  *    Calculate the minimum period (in picoseconds) between two DMA
@@ -606,6 +607,7 @@ static inline unsigned int sa1100fb_display_dma_period(struct fb_var_screeninfo
 	 */
 	return var->pixclock * 8 * 16 / var->bits_per_pixel;
 }
+#endif
 
 /*
  *  sa1100fb_check_var():
diff --git a/drivers/w1/w1.c b/drivers/w1/w1.c
index 1b6b74c116a..14016b1cd94 100644
--- a/drivers/w1/w1.c
+++ b/drivers/w1/w1.c
@@ -77,8 +77,7 @@ static void w1_master_release(struct device *dev)
 
 	dev_dbg(dev, "%s: Releasing %s.\n", __func__, md->name);
 
-	if (md->nls && md->nls->sk_socket)
-		sock_release(md->nls->sk_socket);
+	dev_fini_netlink(md);
 	memset(md, 0, sizeof(struct w1_master) + sizeof(struct w1_bus_master));
 	kfree(md);
 }
diff --git a/fs/aio.c b/fs/aio.c
index d6b1551342b..9fe7216457d 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -398,7 +398,7 @@ static struct kiocb fastcall *__aio_get_req(struct kioctx *ctx)
 	if (unlikely(!req))
 		return NULL;
 
-	req->ki_flags = 1 << KIF_LOCKED;
+	req->ki_flags = 0;
 	req->ki_users = 2;
 	req->ki_key = 0;
 	req->ki_ctx = ctx;
@@ -547,25 +547,6 @@ struct kioctx *lookup_ioctx(unsigned long ctx_id)
 	return ioctx;
 }
 
-static int lock_kiocb_action(void *param)
-{
-	schedule();
-	return 0;
-}
-
-static inline void lock_kiocb(struct kiocb *iocb)
-{
-	wait_on_bit_lock(&iocb->ki_flags, KIF_LOCKED, lock_kiocb_action,
-			 TASK_UNINTERRUPTIBLE);
-}
-
-static inline void unlock_kiocb(struct kiocb *iocb)
-{
-	kiocbClearLocked(iocb);
-	smp_mb__after_clear_bit();
-	wake_up_bit(&iocb->ki_flags, KIF_LOCKED);
-}
-
 /*
  * use_mm
  *	Makes the calling kernel thread take on the specified
@@ -796,9 +777,7 @@ static int __aio_run_iocbs(struct kioctx *ctx)
 		 * Hold an extra reference while retrying i/o.
 		 */
 		iocb->ki_users++;       /* grab extra reference */
-		lock_kiocb(iocb);
 		aio_run_iocb(iocb);
-		unlock_kiocb(iocb);
 		if (__aio_put_req(ctx, iocb))  /* drop extra ref */
 			put_ioctx(ctx);
  	}
@@ -1542,7 +1521,6 @@ int fastcall io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb,
 
 	spin_lock_irq(&ctx->ctx_lock);
 	aio_run_iocb(req);
-	unlock_kiocb(req);
 	if (!list_empty(&ctx->run_list)) {
 		/* drain the run list */
 		while (__aio_run_iocbs(ctx))
@@ -1674,7 +1652,6 @@ asmlinkage long sys_io_cancel(aio_context_t ctx_id, struct iocb __user *iocb,
 	if (NULL != cancel) {
 		struct io_event tmp;
 		pr_debug("calling cancel\n");
-		lock_kiocb(kiocb);
 		memset(&tmp, 0, sizeof(tmp));
 		tmp.obj = (u64)(unsigned long)kiocb->ki_obj.user;
 		tmp.data = kiocb->ki_user_data;
@@ -1686,7 +1663,6 @@ asmlinkage long sys_io_cancel(aio_context_t ctx_id, struct iocb __user *iocb,
 			if (copy_to_user(result, &tmp, sizeof(tmp)))
 				ret = -EFAULT;
 		}
-		unlock_kiocb(kiocb);
 	} else
 		ret = -EINVAL;
 
diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c
index d7f7eb669d0..4a36839f0bb 100644
--- a/fs/nfs/delegation.c
+++ b/fs/nfs/delegation.c
@@ -85,6 +85,10 @@ int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct
 	struct nfs_delegation *delegation;
 	int status = 0;
 
+	/* Ensure we first revalidate the attributes and page cache! */
+	if ((nfsi->cache_validity & (NFS_INO_REVAL_PAGECACHE|NFS_INO_INVALID_ATTR)))
+		__nfs_revalidate_inode(NFS_SERVER(inode), inode);
+
 	delegation = nfs_alloc_delegation();
 	if (delegation == NULL)
 		return -ENOMEM;
diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index f6b9eda925c..6bdcfa95de9 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -137,7 +137,8 @@ static int nfs_revalidate_file(struct inode *inode, struct file *filp)
 	struct nfs_inode *nfsi = NFS_I(inode);
 	int retval = 0;
 
-	if ((nfsi->cache_validity & NFS_INO_REVAL_PAGECACHE) || nfs_attribute_timeout(inode))
+	if ((nfsi->cache_validity & (NFS_INO_REVAL_PAGECACHE|NFS_INO_INVALID_ATTR))
+			|| nfs_attribute_timeout(inode))
 		retval = __nfs_revalidate_inode(NFS_SERVER(inode), inode);
 	nfs_revalidate_mapping(inode, filp->f_mapping);
 	return 0;
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 6922469d6fc..d4eadeea128 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -877,12 +877,10 @@ static int nfs_wait_on_inode(struct inode *inode)
 	sigset_t oldmask;
 	int error;
 
-	atomic_inc(&inode->i_count);
 	rpc_clnt_sigmask(clnt, &oldmask);
 	error = wait_on_bit_lock(&nfsi->flags, NFS_INO_REVALIDATING,
 					nfs_wait_schedule, TASK_INTERRUPTIBLE);
 	rpc_clnt_sigunmask(clnt, &oldmask);
-	iput(inode);
 
 	return error;
 }
@@ -1226,10 +1224,6 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr)
 	loff_t cur_size, new_isize;
 	int data_unstable;
 
-	/* Do we hold a delegation? */
-	if (nfs_have_delegation(inode, FMODE_READ))
-		return 0;
-
 	spin_lock(&inode->i_lock);
 
 	/* Are we in the process of updating data on the server? */
@@ -1350,7 +1344,8 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsign
 	nfsi->read_cache_jiffies = fattr->timestamp;
 
 	/* Are we racing with known updates of the metadata on the server? */
-	data_unstable = ! nfs_verify_change_attribute(inode, verifier);
+	data_unstable = ! (nfs_verify_change_attribute(inode, verifier) ||
+		(nfsi->cache_validity & NFS_INO_REVAL_PAGECACHE));
 
 	/* Check if our cached file size is stale */
  	new_isize = nfs_size_to_loff_t(fattr->size);
diff --git a/fs/proc/base.c b/fs/proc/base.c
index 3b33f94020d..a170450aadb 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -103,7 +103,9 @@ enum pid_directory_inos {
 	PROC_TGID_NUMA_MAPS,
 	PROC_TGID_MOUNTS,
 	PROC_TGID_WCHAN,
+#ifdef CONFIG_MMU
 	PROC_TGID_SMAPS,
+#endif
 #ifdef CONFIG_SCHEDSTATS
 	PROC_TGID_SCHEDSTAT,
 #endif
@@ -141,7 +143,9 @@ enum pid_directory_inos {
 	PROC_TID_NUMA_MAPS,
 	PROC_TID_MOUNTS,
 	PROC_TID_WCHAN,
+#ifdef CONFIG_MMU
 	PROC_TID_SMAPS,
+#endif
 #ifdef CONFIG_SCHEDSTATS
 	PROC_TID_SCHEDSTAT,
 #endif
@@ -195,7 +199,9 @@ static struct pid_entry tgid_base_stuff[] = {
 	E(PROC_TGID_ROOT,      "root",    S_IFLNK|S_IRWXUGO),
 	E(PROC_TGID_EXE,       "exe",     S_IFLNK|S_IRWXUGO),
 	E(PROC_TGID_MOUNTS,    "mounts",  S_IFREG|S_IRUGO),
+#ifdef CONFIG_MMU
 	E(PROC_TGID_SMAPS,     "smaps",   S_IFREG|S_IRUGO),
+#endif
 #ifdef CONFIG_SECURITY
 	E(PROC_TGID_ATTR,      "attr",    S_IFDIR|S_IRUGO|S_IXUGO),
 #endif
@@ -235,7 +241,9 @@ static struct pid_entry tid_base_stuff[] = {
 	E(PROC_TID_ROOT,       "root",    S_IFLNK|S_IRWXUGO),
 	E(PROC_TID_EXE,        "exe",     S_IFLNK|S_IRWXUGO),
 	E(PROC_TID_MOUNTS,     "mounts",  S_IFREG|S_IRUGO),
+#ifdef CONFIG_MMU
 	E(PROC_TID_SMAPS,      "smaps",   S_IFREG|S_IRUGO),
+#endif
 #ifdef CONFIG_SECURITY
 	E(PROC_TID_ATTR,       "attr",    S_IFDIR|S_IRUGO|S_IXUGO),
 #endif
@@ -630,6 +638,7 @@ static struct file_operations proc_numa_maps_operations = {
 };
 #endif
 
+#ifdef CONFIG_MMU
 extern struct seq_operations proc_pid_smaps_op;
 static int smaps_open(struct inode *inode, struct file *file)
 {
@@ -648,6 +657,7 @@ static struct file_operations proc_smaps_operations = {
 	.llseek		= seq_lseek,
 	.release	= seq_release,
 };
+#endif
 
 extern struct seq_operations mounts_op;
 static int mounts_open(struct inode *inode, struct file *file)
@@ -1681,10 +1691,12 @@ static struct dentry *proc_pident_lookup(struct inode *dir,
 		case PROC_TGID_MOUNTS:
 			inode->i_fop = &proc_mounts_operations;
 			break;
+#ifdef CONFIG_MMU
 		case PROC_TID_SMAPS:
 		case PROC_TGID_SMAPS:
 			inode->i_fop = &proc_smaps_operations;
 			break;
+#endif
 #ifdef CONFIG_SECURITY
 		case PROC_TID_ATTR:
 			inode->i_nlink = 2;
diff --git a/fs/proc/nommu.c b/fs/proc/nommu.c
index f3bf016d5ee..cff10ab1af6 100644
--- a/fs/proc/nommu.c
+++ b/fs/proc/nommu.c
@@ -91,6 +91,7 @@ static void *nommu_vma_list_start(struct seq_file *m, loff_t *_pos)
 			next = _rb;
 			break;
 		}
+		pos--;
 	}
 
 	return next;
diff --git a/include/asm-arm/arch-pxa/pxafb.h b/include/asm-arm/arch-pxa/pxafb.h
index 21c0e16dce5..aba9b30f424 100644
--- a/include/asm-arm/arch-pxa/pxafb.h
+++ b/include/asm-arm/arch-pxa/pxafb.h
@@ -66,4 +66,5 @@ struct pxafb_mach_info {
 
 };
 void set_pxa_fb_info(struct pxafb_mach_info *hard_pxa_fb_info);
+void set_pxa_fb_parent(struct device *parent_dev);
 unsigned long pxafb_get_hsync_time(struct device *dev);
diff --git a/include/asm-arm/arch-s3c2410/io.h b/include/asm-arm/arch-s3c2410/io.h
index 418233a7ee6..4bf272ed9ad 100644
--- a/include/asm-arm/arch-s3c2410/io.h
+++ b/include/asm-arm/arch-s3c2410/io.h
@@ -9,7 +9,7 @@
  *  06-Dec-1997	RMK	Created.
  *  02-Sep-2003 BJD	Modified for S3C2410
  *  10-Mar-2005 LCVR	Changed S3C2410_VA to S3C24XX_VA
- *
+ *  13-Oct-2005 BJD	Fixed problems with LDRH/STRH offset range
  */
 
 #ifndef __ASM_ARM_ARCH_IO_H
@@ -97,7 +97,7 @@ DECLARE_IO(int,l,"")
 	else								\
 		__asm__ __volatile__(					\
 		"strb	%0, [%1, #0]	@ outbc"			\
-		: : "r" (value), "r" ((port)));		\
+		: : "r" (value), "r" ((port)));				\
 })
 
 #define __inbc(port)							\
@@ -110,35 +110,61 @@ DECLARE_IO(int,l,"")
 	else								\
 		__asm__ __volatile__(					\
 		"ldrb	%0, [%1, #0]	@ inbc"				\
-		: "=r" (result) : "r" ((port)));	\
+		: "=r" (result) : "r" ((port)));			\
 	result;								\
 })
 
 #define __outwc(value,port)						\
 ({									\
 	unsigned long v = value;					\
-	if (__PORT_PCIO((port)))					\
-		__asm__ __volatile__(					\
-		"strh	%0, [%1, %2]	@ outwc"			\
-		: : "r" (v), "r" (PCIO_BASE), "Jr" ((port)));	\
-	else								\
+	if (__PORT_PCIO((port))) {					\
+		if ((port) < 256 && (port) > -256)			\
+			__asm__ __volatile__(				\
+			"strh	%0, [%1, %2]	@ outwc"		\
+			: : "r" (v), "r" (PCIO_BASE), "Jr" ((port)));	\
+		else if ((port) > 0)					\
+			__asm__ __volatile__(				\
+			"strh	%0, [%1, %2]	@ outwc"		\
+			: : "r" (v),					\
+			    "r" (PCIO_BASE + ((port) & ~0xff)),		\
+			     "Jr" (((port) & 0xff)));			\
+		else							\
+			__asm__ __volatile__(				\
+			"strh	%0, [%1, #0]	@ outwc"		\
+			: : "r" (v),					\
+			    "r" (PCIO_BASE + (port)));			\
+	} else								\
 		__asm__ __volatile__(					\
 		"strh	%0, [%1, #0]	@ outwc"			\
-		: : "r" (v), "r" ((port)));	\
+		: : "r" (v), "r" ((port)));				\
 })
 
 #define __inwc(port)							\
 ({									\
 	unsigned short result;						\
-	if (__PORT_PCIO((port)))					\
-		__asm__ __volatile__(					\
-		"ldrh	%0, [%1, %2]	@ inwc"				\
-		: "=r" (result) : "r" (PCIO_BASE), "Jr" ((port)));	\
-	else								\
+	if (__PORT_PCIO((port))) {					\
+		if ((port) < 256 && (port) > -256 )			\
+			__asm__ __volatile__(				\
+			"ldrh	%0, [%1, %2]	@ inwc"			\
+			: "=r" (result)					\
+			: "r" (PCIO_BASE),				\
+			  "Jr" ((port)));				\
+		else if ((port) > 0)					\
+			__asm__ __volatile__(				\
+			"ldrh	%0, [%1, %2]	@ inwc"			\
+			: "=r" (result)					\
+			: "r" (PCIO_BASE + ((port) & ~0xff)),		\
+			  "Jr" (((port) & 0xff)));			\
+		else							\
+			__asm__ __volatile__(				\
+			"ldrh	%0, [%1, #0]	@ inwc"			\
+			: "=r" (result)					\
+			: "r" (PCIO_BASE + ((port))));			\
+	} else								\
 		__asm__ __volatile__(					\
 		"ldrh	%0, [%1, #0]	@ inwc"				\
-		: "=r" (result) : "r" ((port)));		\
-	result;						\
+		: "=r" (result) : "r" ((port)));			\
+	result;								\
 })
 
 #define __outlc(value,port)						\
diff --git a/include/asm-sparc64/pbm.h b/include/asm-sparc64/pbm.h
index 38bbbccb406..dd35a2c7798 100644
--- a/include/asm-sparc64/pbm.h
+++ b/include/asm-sparc64/pbm.h
@@ -27,23 +27,27 @@
  * PCI bus.
  */
 
-#define PBM_LOGCLUSTERS 3
-#define PBM_NCLUSTERS (1 << PBM_LOGCLUSTERS)
-
 struct pci_controller_info;
 
 /* This contains the software state necessary to drive a PCI
  * controller's IOMMU.
  */
+struct pci_iommu_arena {
+	unsigned long	*map;
+	unsigned int	hint;
+	unsigned int	limit;
+};
+
 struct pci_iommu {
 	/* This protects the controller's IOMMU and all
 	 * streaming buffers underneath.
 	 */
 	spinlock_t	lock;
 
+	struct pci_iommu_arena arena;
+
 	/* IOMMU page table, a linear array of ioptes. */
 	iopte_t		*page_table;		/* The page table itself. */
-	int		page_table_sz_bits;	/* log2 of ow many pages does it map? */
 
 	/* Base PCI memory space address where IOMMU mappings
 	 * begin.
@@ -62,12 +66,6 @@ struct pci_iommu {
 	 */
 	unsigned long	write_complete_reg;
 
-	/* The lowest used consistent mapping entry.  Since
-	 * we allocate consistent maps out of cluster 0 this
-	 * is relative to the beginning of closter 0.
-	 */
-	u32		lowest_consistent_map;
-
 	/* In order to deal with some buggy third-party PCI bridges that
 	 * do wrong prefetching, we never mark valid mappings as invalid.
 	 * Instead we point them at this dummy page.
@@ -75,16 +73,6 @@ struct pci_iommu {
 	unsigned long	dummy_page;
 	unsigned long	dummy_page_pa;
 
-	/* If PBM_NCLUSTERS is ever decreased to 4 or lower,
-	 * or if largest supported page_table_sz * 8K goes above
-	 * 2GB, you must increase the size of the type of
-	 * these counters.  You have been duly warned. -DaveM
-	 */
-	struct {
-		u16	next;
-		u16	flush;
-	} alloc_info[PBM_NCLUSTERS];
-
 	/* CTX allocation. */
 	unsigned long ctx_lowest_free;
 	unsigned long ctx_bitmap[IOMMU_NUM_CTXS / (sizeof(unsigned long) * 8)];
@@ -102,7 +90,7 @@ struct pci_iommu {
 	u32 dma_addr_mask;
 };
 
-extern void pci_iommu_table_init(struct pci_iommu *, int);
+extern void pci_iommu_table_init(struct pci_iommu *iommu, int tsbsize, u32 dma_offset, u32 dma_addr_mask);
 
 /* This describes a PCI bus module's streaming buffer. */
 struct pci_strbuf {
diff --git a/include/linux/acct.h b/include/linux/acct.h
index 1993a369176..19f70462b3b 100644
--- a/include/linux/acct.h
+++ b/include/linux/acct.h
@@ -162,13 +162,13 @@ typedef struct acct acct_t;
 #ifdef __KERNEL__
 /*
  * Yet another set of HZ to *HZ helper functions.
- * See <linux/times.h> for the original.
+ * See <linux/jiffies.h> for the original.
  */
 
 static inline u32 jiffies_to_AHZ(unsigned long x)
 {
 #if (TICK_NSEC % (NSEC_PER_SEC / AHZ)) == 0
-	return x / (HZ / USER_HZ);
+	return x / (HZ / AHZ);
 #else
         u64 tmp = (u64)x * TICK_NSEC;
         do_div(tmp, (NSEC_PER_SEC / AHZ));
diff --git a/include/linux/aio.h b/include/linux/aio.h
index 60def658b24..0decf66117c 100644
--- a/include/linux/aio.h
+++ b/include/linux/aio.h
@@ -24,7 +24,12 @@ struct kioctx;
 #define KIOCB_SYNC_KEY		(~0U)
 
 /* ki_flags bits */
-#define KIF_LOCKED		0
+/*
+ * This may be used for cancel/retry serialization in the future, but
+ * for now it's unused and we probably don't want modules to even
+ * think they can use it.
+ */
+/* #define KIF_LOCKED		0 */
 #define KIF_KICKED		1
 #define KIF_CANCELLED		2
 
diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index fe9778301d0..9bdba8169b4 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -393,15 +393,13 @@ extern cpumask_t cpu_present_map;
 #define for_each_present_cpu(cpu) for_each_cpu_mask((cpu), cpu_present_map)
 
 /* Find the highest possible smp_processor_id() */
-static inline unsigned int highest_possible_processor_id(void)
-{
-	unsigned int cpu, highest = 0;
-
-	for_each_cpu_mask(cpu, cpu_possible_map)
-		highest = cpu;
-
-	return highest;
-}
+#define highest_possible_processor_id() \
+({ \
+	unsigned int cpu, highest = 0; \
+	for_each_cpu_mask(cpu, cpu_possible_map) \
+		highest = cpu; \
+	highest; \
+})
 
 
 #endif /* __LINUX_CPUMASK_H */
diff --git a/include/linux/list.h b/include/linux/list.h
index e6ec5968227..084971f333f 100644
--- a/include/linux/list.h
+++ b/include/linux/list.h
@@ -442,12 +442,14 @@ static inline void list_splice_init(struct list_head *list,
  * as long as the traversal is guarded by rcu_read_lock().
  */
 #define list_for_each_rcu(pos, head) \
-	for (pos = (head)->next; prefetch(pos->next), pos != (head); \
-        	pos = rcu_dereference(pos->next))
+	for (pos = (head)->next; \
+		prefetch(rcu_dereference(pos)->next), pos != (head); \
+        	pos = pos->next)
 
 #define __list_for_each_rcu(pos, head) \
-	for (pos = (head)->next; pos != (head); \
-        	pos = rcu_dereference(pos->next))
+	for (pos = (head)->next; \
+		rcu_dereference(pos) != (head); \
+        	pos = pos->next)
 
 /**
  * list_for_each_safe_rcu	-	iterate over an rcu-protected list safe
@@ -461,8 +463,9 @@ static inline void list_splice_init(struct list_head *list,
  * as long as the traversal is guarded by rcu_read_lock().
  */
 #define list_for_each_safe_rcu(pos, n, head) \
-	for (pos = (head)->next, n = pos->next; pos != (head); \
-		pos = rcu_dereference(n), n = pos->next)
+	for (pos = (head)->next; \
+		n = rcu_dereference(pos)->next, pos != (head); \
+		pos = n)
 
 /**
  * list_for_each_entry_rcu	-	iterate over rcu list of given type
@@ -474,11 +477,11 @@ static inline void list_splice_init(struct list_head *list,
  * the _rcu list-mutation primitives such as list_add_rcu()
  * as long as the traversal is guarded by rcu_read_lock().
  */
-#define list_for_each_entry_rcu(pos, head, member)			\
-	for (pos = list_entry((head)->next, typeof(*pos), member);	\
-	     prefetch(pos->member.next), &pos->member != (head); 	\
-	     pos = rcu_dereference(list_entry(pos->member.next, 	\
-					typeof(*pos), member)))
+#define list_for_each_entry_rcu(pos, head, member) \
+	for (pos = list_entry((head)->next, typeof(*pos), member); \
+		prefetch(rcu_dereference(pos)->member.next), \
+			&pos->member != (head); \
+		pos = list_entry(pos->member.next, typeof(*pos), member))
 
 
 /**
@@ -492,8 +495,9 @@ static inline void list_splice_init(struct list_head *list,
  * as long as the traversal is guarded by rcu_read_lock().
  */
 #define list_for_each_continue_rcu(pos, head) \
-	for ((pos) = (pos)->next; prefetch((pos)->next), (pos) != (head); \
-        	(pos) = rcu_dereference((pos)->next))
+	for ((pos) = (pos)->next; \
+		prefetch(rcu_dereference((pos))->next), (pos) != (head); \
+        	(pos) = (pos)->next)
 
 /*
  * Double linked lists with a single pointer list head.
@@ -696,8 +700,9 @@ static inline void hlist_add_after_rcu(struct hlist_node *prev,
 	     pos = n)
 
 #define hlist_for_each_rcu(pos, head) \
-	for ((pos) = (head)->first; pos && ({ prefetch((pos)->next); 1; }); \
-		(pos) = rcu_dereference((pos)->next))
+	for ((pos) = (head)->first; \
+		rcu_dereference((pos)) && ({ prefetch((pos)->next); 1; }); \
+		(pos) = (pos)->next)
 
 /**
  * hlist_for_each_entry	- iterate over list of given type
@@ -762,9 +767,9 @@ static inline void hlist_add_after_rcu(struct hlist_node *prev,
  */
 #define hlist_for_each_entry_rcu(tpos, pos, head, member)		 \
 	for (pos = (head)->first;					 \
-	     pos && ({ prefetch(pos->next); 1;}) &&			 \
+	     rcu_dereference(pos) && ({ prefetch(pos->next); 1;}) &&	 \
 		({ tpos = hlist_entry(pos, typeof(*tpos), member); 1;}); \
-	     pos = rcu_dereference(pos->next))
+	     pos = pos->next)
 
 #else
 #warning "don't include kernel headers in userspace"
diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 4e65eb44adf..70191a5a148 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -94,6 +94,7 @@ struct rcu_data {
 	long  	       	batch;           /* Batch # for current RCU batch */
 	struct rcu_head *nxtlist;
 	struct rcu_head **nxttail;
+	long            count; /* # of queued items */
 	struct rcu_head *curlist;
 	struct rcu_head **curtail;
 	struct rcu_head *donelist;
diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c
index ad85d3f0dcc..7a51a5597c3 100644
--- a/kernel/posix-cpu-timers.c
+++ b/kernel/posix-cpu-timers.c
@@ -424,6 +424,7 @@ static void cleanup_timers(struct list_head *head,
 	cputime_t ptime = cputime_add(utime, stime);
 
 	list_for_each_entry_safe(timer, next, head, entry) {
+		put_task_struct(timer->task);
 		timer->task = NULL;
 		list_del_init(&timer->entry);
 		if (cputime_lt(timer->expires.cpu, ptime)) {
@@ -436,6 +437,7 @@ static void cleanup_timers(struct list_head *head,
 
 	++head;
 	list_for_each_entry_safe(timer, next, head, entry) {
+		put_task_struct(timer->task);
 		timer->task = NULL;
 		list_del_init(&timer->entry);
 		if (cputime_lt(timer->expires.cpu, utime)) {
@@ -448,6 +450,7 @@ static void cleanup_timers(struct list_head *head,
 
 	++head;
 	list_for_each_entry_safe(timer, next, head, entry) {
+		put_task_struct(timer->task);
 		timer->task = NULL;
 		list_del_init(&timer->entry);
 		if (timer->expires.sched < sched_time) {
diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c
index bef3b6901b7..2559d4b8f23 100644
--- a/kernel/rcupdate.c
+++ b/kernel/rcupdate.c
@@ -71,7 +71,7 @@ DEFINE_PER_CPU(struct rcu_data, rcu_bh_data) = { 0L };
 
 /* Fake initialization required by compiler */
 static DEFINE_PER_CPU(struct tasklet_struct, rcu_tasklet) = {NULL};
-static int maxbatch = 10;
+static int maxbatch = 10000;
 
 #ifndef __HAVE_ARCH_CMPXCHG
 /*
@@ -109,6 +109,10 @@ void fastcall call_rcu(struct rcu_head *head,
 	rdp = &__get_cpu_var(rcu_data);
 	*rdp->nxttail = head;
 	rdp->nxttail = &head->next;
+
+	if (unlikely(++rdp->count > 10000))
+		set_need_resched();
+
 	local_irq_restore(flags);
 }
 
@@ -140,6 +144,12 @@ void fastcall call_rcu_bh(struct rcu_head *head,
 	rdp = &__get_cpu_var(rcu_bh_data);
 	*rdp->nxttail = head;
 	rdp->nxttail = &head->next;
+	rdp->count++;
+/*
+ *  Should we directly call rcu_do_batch() here ?
+ *  if (unlikely(rdp->count > 10000))
+ *      rcu_do_batch(rdp);
+ */
 	local_irq_restore(flags);
 }
 
@@ -157,6 +167,7 @@ static void rcu_do_batch(struct rcu_data *rdp)
 		next = rdp->donelist = list->next;
 		list->func(list);
 		list = next;
+		rdp->count--;
 		if (++count >= maxbatch)
 			break;
 	}
diff --git a/kernel/time.c b/kernel/time.c
index dd5ae1162a8..40c2410ac99 100644
--- a/kernel/time.c
+++ b/kernel/time.c
@@ -570,6 +570,7 @@ void getnstimeofday(struct timespec *tv)
 	tv->tv_sec = x.tv_sec;
 	tv->tv_nsec = x.tv_usec * NSEC_PER_USEC;
 }
+EXPORT_SYMBOL_GPL(getnstimeofday);
 #endif
 
 #if (BITS_PER_LONG < 64)
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 0ea71e887bb..64f9570cff5 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -511,10 +511,11 @@ static int shrink_list(struct list_head *page_list, struct scan_control *sc)
 		 * PageDirty _after_ making sure that the page is freeable and
 		 * not in use by anybody. 	(pagecache + us == 2)
 		 */
-		if (page_count(page) != 2 || PageDirty(page)) {
-			write_unlock_irq(&mapping->tree_lock);
-			goto keep_locked;
-		}
+		if (unlikely(page_count(page) != 2))
+			goto cannot_free;
+		smp_rmb();
+		if (unlikely(PageDirty(page)))
+			goto cannot_free;
 
 #ifdef CONFIG_SWAP
 		if (PageSwapCache(page)) {
@@ -538,6 +539,10 @@ free_it:
 			__pagevec_release_nonlru(&freed_pvec);
 		continue;
 
+cannot_free:
+		write_unlock_irq(&mapping->tree_lock);
+		goto keep_locked;
+
 activate_locked:
 		SetPageActive(page);
 		pgactivate++;
diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index b03e90649eb..21deec25a12 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -975,7 +975,6 @@ replace_table(struct ip6t_table *table,
 		struct ip6t_entry *table_base;
 		unsigned int i;
 
-		for (i = 0; i < num_possible_cpus(); i++) {
 		for_each_cpu(i) {
 			table_base =
 				(void *)newinfo->entries