Merge branch 'linus' into x86/urgent

Merge needed to go past commit 7ca43e756 (mm: use debug_kmap_atomic)
and fix it.

Signed-off-by: Ingo Molnar <mingo@elte.hu>

38 files changed, 834 insertions, 796 deletions

diff --git a/arch/ia64/dig/Makefile b/arch/ia64/dig/Makefile
index 5c0283830bd..2f7caddf093 100644
--- a/arch/ia64/dig/Makefile
+++ b/arch/ia64/dig/Makefile
@@ -7,8 +7,8 @@
 
 obj-y := setup.o
 ifeq ($(CONFIG_DMAR), y)
-obj-$(CONFIG_IA64_GENERIC) += machvec.o machvec_vtd.o dig_vtd_iommu.o
+obj-$(CONFIG_IA64_GENERIC) += machvec.o machvec_vtd.o
 else
 obj-$(CONFIG_IA64_GENERIC) += machvec.o
 endif
-obj-$(CONFIG_IA64_DIG_VTD) += dig_vtd_iommu.o
+
diff --git a/arch/ia64/dig/dig_vtd_iommu.c b/arch/ia64/dig/dig_vtd_iommu.c
deleted file mode 100644
index 1c8a079017a..00000000000
--- a/arch/ia64/dig/dig_vtd_iommu.c
+++ /dev/null
@@ -1,59 +0,0 @@
-#include <linux/types.h>
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/intel-iommu.h>
-
-void *
-vtd_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle,
-		 gfp_t flags)
-{
-	return intel_alloc_coherent(dev, size, dma_handle, flags);
-}
-EXPORT_SYMBOL_GPL(vtd_alloc_coherent);
-
-void
-vtd_free_coherent(struct device *dev, size_t size, void *vaddr,
-		 dma_addr_t dma_handle)
-{
-	intel_free_coherent(dev, size, vaddr, dma_handle);
-}
-EXPORT_SYMBOL_GPL(vtd_free_coherent);
-
-dma_addr_t
-vtd_map_single_attrs(struct device *dev, void *addr, size_t size,
-		     int dir, struct dma_attrs *attrs)
-{
-	return intel_map_single(dev, (phys_addr_t)addr, size, dir);
-}
-EXPORT_SYMBOL_GPL(vtd_map_single_attrs);
-
-void
-vtd_unmap_single_attrs(struct device *dev, dma_addr_t iova, size_t size,
-		       int dir, struct dma_attrs *attrs)
-{
-	intel_unmap_single(dev, iova, size, dir);
-}
-EXPORT_SYMBOL_GPL(vtd_unmap_single_attrs);
-
-int
-vtd_map_sg_attrs(struct device *dev, struct scatterlist *sglist, int nents,
-		 int dir, struct dma_attrs *attrs)
-{
-	return intel_map_sg(dev, sglist, nents, dir);
-}
-EXPORT_SYMBOL_GPL(vtd_map_sg_attrs);
-
-void
-vtd_unmap_sg_attrs(struct device *dev, struct scatterlist *sglist,
-		   int nents, int dir, struct dma_attrs *attrs)
-{
-	intel_unmap_sg(dev, sglist, nents, dir);
-}
-EXPORT_SYMBOL_GPL(vtd_unmap_sg_attrs);
-
-int
-vtd_dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
-{
-	return 0;
-}
-EXPORT_SYMBOL_GPL(vtd_dma_mapping_error);
diff --git a/arch/ia64/hp/common/hwsw_iommu.c b/arch/ia64/hp/common/hwsw_iommu.c
index 2769dbfd03b..e4a80d82e3d 100644
--- a/arch/ia64/hp/common/hwsw_iommu.c
+++ b/arch/ia64/hp/common/hwsw_iommu.c
@@ -13,49 +13,34 @@
  */
 
 #include <linux/device.h>
+#include <linux/dma-mapping.h>
 #include <linux/swiotlb.h>
-
 #include <asm/machvec.h>
 
+extern struct dma_map_ops sba_dma_ops, swiotlb_dma_ops;
+
 /* swiotlb declarations & definitions: */
 extern int swiotlb_late_init_with_default_size (size_t size);
 
-/* hwiommu declarations & definitions: */
-
-extern ia64_mv_dma_alloc_coherent	sba_alloc_coherent;
-extern ia64_mv_dma_free_coherent	sba_free_coherent;
-extern ia64_mv_dma_map_single_attrs	sba_map_single_attrs;
-extern ia64_mv_dma_unmap_single_attrs	sba_unmap_single_attrs;
-extern ia64_mv_dma_map_sg_attrs		sba_map_sg_attrs;
-extern ia64_mv_dma_unmap_sg_attrs	sba_unmap_sg_attrs;
-extern ia64_mv_dma_supported		sba_dma_supported;
-extern ia64_mv_dma_mapping_error	sba_dma_mapping_error;
-
-#define hwiommu_alloc_coherent		sba_alloc_coherent
-#define hwiommu_free_coherent		sba_free_coherent
-#define hwiommu_map_single_attrs	sba_map_single_attrs
-#define hwiommu_unmap_single_attrs	sba_unmap_single_attrs
-#define hwiommu_map_sg_attrs		sba_map_sg_attrs
-#define hwiommu_unmap_sg_attrs		sba_unmap_sg_attrs
-#define hwiommu_dma_supported		sba_dma_supported
-#define hwiommu_dma_mapping_error	sba_dma_mapping_error
-#define hwiommu_sync_single_for_cpu	machvec_dma_sync_single
-#define hwiommu_sync_sg_for_cpu		machvec_dma_sync_sg
-#define hwiommu_sync_single_for_device	machvec_dma_sync_single
-#define hwiommu_sync_sg_for_device	machvec_dma_sync_sg
-
-
 /*
  * Note: we need to make the determination of whether or not to use
  * the sw I/O TLB based purely on the device structure.  Anything else
  * would be unreliable or would be too intrusive.
  */
-static inline int
-use_swiotlb (struct device *dev)
+static inline int use_swiotlb(struct device *dev)
 {
-	return dev && dev->dma_mask && !hwiommu_dma_supported(dev, *dev->dma_mask);
+	return dev && dev->dma_mask &&
+		!sba_dma_ops.dma_supported(dev, *dev->dma_mask);
 }
 
+struct dma_map_ops *hwsw_dma_get_ops(struct device *dev)
+{
+	if (use_swiotlb(dev))
+		return &swiotlb_dma_ops;
+	return &sba_dma_ops;
+}
+EXPORT_SYMBOL(hwsw_dma_get_ops);
+
 void __init
 hwsw_init (void)
 {
@@ -71,125 +56,3 @@ hwsw_init (void)
 #endif
 	}
 }
-
-void *
-hwsw_alloc_coherent (struct device *dev, size_t size, dma_addr_t *dma_handle, gfp_t flags)
-{
-	if (use_swiotlb(dev))
-		return swiotlb_alloc_coherent(dev, size, dma_handle, flags);
-	else
-		return hwiommu_alloc_coherent(dev, size, dma_handle, flags);
-}
-
-void
-hwsw_free_coherent (struct device *dev, size_t size, void *vaddr, dma_addr_t dma_handle)
-{
-	if (use_swiotlb(dev))
-		swiotlb_free_coherent(dev, size, vaddr, dma_handle);
-	else
-		hwiommu_free_coherent(dev, size, vaddr, dma_handle);
-}
-
-dma_addr_t
-hwsw_map_single_attrs(struct device *dev, void *addr, size_t size, int dir,
-		       struct dma_attrs *attrs)
-{
-	if (use_swiotlb(dev))
-		return swiotlb_map_single_attrs(dev, addr, size, dir, attrs);
-	else
-		return hwiommu_map_single_attrs(dev, addr, size, dir, attrs);
-}
-EXPORT_SYMBOL(hwsw_map_single_attrs);
-
-void
-hwsw_unmap_single_attrs(struct device *dev, dma_addr_t iova, size_t size,
-			 int dir, struct dma_attrs *attrs)
-{
-	if (use_swiotlb(dev))
-		return swiotlb_unmap_single_attrs(dev, iova, size, dir, attrs);
-	else
-		return hwiommu_unmap_single_attrs(dev, iova, size, dir, attrs);
-}
-EXPORT_SYMBOL(hwsw_unmap_single_attrs);
-
-int
-hwsw_map_sg_attrs(struct device *dev, struct scatterlist *sglist, int nents,
-		   int dir, struct dma_attrs *attrs)
-{
-	if (use_swiotlb(dev))
-		return swiotlb_map_sg_attrs(dev, sglist, nents, dir, attrs);
-	else
-		return hwiommu_map_sg_attrs(dev, sglist, nents, dir, attrs);
-}
-EXPORT_SYMBOL(hwsw_map_sg_attrs);
-
-void
-hwsw_unmap_sg_attrs(struct device *dev, struct scatterlist *sglist, int nents,
-		     int dir, struct dma_attrs *attrs)
-{
-	if (use_swiotlb(dev))
-		return swiotlb_unmap_sg_attrs(dev, sglist, nents, dir, attrs);
-	else
-		return hwiommu_unmap_sg_attrs(dev, sglist, nents, dir, attrs);
-}
-EXPORT_SYMBOL(hwsw_unmap_sg_attrs);
-
-void
-hwsw_sync_single_for_cpu (struct device *dev, dma_addr_t addr, size_t size, int dir)
-{
-	if (use_swiotlb(dev))
-		swiotlb_sync_single_for_cpu(dev, addr, size, dir);
-	else
-		hwiommu_sync_single_for_cpu(dev, addr, size, dir);
-}
-
-void
-hwsw_sync_sg_for_cpu (struct device *dev, struct scatterlist *sg, int nelems, int dir)
-{
-	if (use_swiotlb(dev))
-		swiotlb_sync_sg_for_cpu(dev, sg, nelems, dir);
-	else
-		hwiommu_sync_sg_for_cpu(dev, sg, nelems, dir);
-}
-
-void
-hwsw_sync_single_for_device (struct device *dev, dma_addr_t addr, size_t size, int dir)
-{
-	if (use_swiotlb(dev))
-		swiotlb_sync_single_for_device(dev, addr, size, dir);
-	else
-		hwiommu_sync_single_for_device(dev, addr, size, dir);
-}
-
-void
-hwsw_sync_sg_for_device (struct device *dev, struct scatterlist *sg, int nelems, int dir)
-{
-	if (use_swiotlb(dev))
-		swiotlb_sync_sg_for_device(dev, sg, nelems, dir);
-	else
-		hwiommu_sync_sg_for_device(dev, sg, nelems, dir);
-}
-
-int
-hwsw_dma_supported (struct device *dev, u64 mask)
-{
-	if (hwiommu_dma_supported(dev, mask))
-		return 1;
-	return swiotlb_dma_supported(dev, mask);
-}
-
-int
-hwsw_dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
-{
-	return hwiommu_dma_mapping_error(dev, dma_addr) ||
-		swiotlb_dma_mapping_error(dev, dma_addr);
-}
-
-EXPORT_SYMBOL(hwsw_dma_mapping_error);
-EXPORT_SYMBOL(hwsw_dma_supported);
-EXPORT_SYMBOL(hwsw_alloc_coherent);
-EXPORT_SYMBOL(hwsw_free_coherent);
-EXPORT_SYMBOL(hwsw_sync_single_for_cpu);
-EXPORT_SYMBOL(hwsw_sync_single_for_device);
-EXPORT_SYMBOL(hwsw_sync_sg_for_cpu);
-EXPORT_SYMBOL(hwsw_sync_sg_for_device);
diff --git a/arch/ia64/hp/common/sba_iommu.c b/arch/ia64/hp/common/sba_iommu.c
index 6d5e6c5630e..56ceb68eb99 100644
--- a/arch/ia64/hp/common/sba_iommu.c
+++ b/arch/ia64/hp/common/sba_iommu.c
@@ -36,6 +36,7 @@
 #include <linux/bitops.h>         /* hweight64() */
 #include <linux/crash_dump.h>
 #include <linux/iommu-helper.h>
+#include <linux/dma-mapping.h>
 
 #include <asm/delay.h>		/* ia64_get_itc() */
 #include <asm/io.h>
@@ -908,11 +909,13 @@ sba_mark_invalid(struct ioc *ioc, dma_addr_t iova, size_t byte_cnt)
  *
  * See Documentation/PCI/PCI-DMA-mapping.txt
  */
-dma_addr_t
-sba_map_single_attrs(struct device *dev, void *addr, size_t size, int dir,
-		     struct dma_attrs *attrs)
+static dma_addr_t sba_map_page(struct device *dev, struct page *page,
+			       unsigned long poff, size_t size,
+			       enum dma_data_direction dir,
+			       struct dma_attrs *attrs)
 {
 	struct ioc *ioc;
+	void *addr = page_address(page) + poff;
 	dma_addr_t iovp;
 	dma_addr_t offset;
 	u64 *pdir_start;
@@ -990,7 +993,14 @@ sba_map_single_attrs(struct device *dev, void *addr, size_t size, int dir,
 #endif
 	return SBA_IOVA(ioc, iovp, offset);
 }
-EXPORT_SYMBOL(sba_map_single_attrs);
+
+static dma_addr_t sba_map_single_attrs(struct device *dev, void *addr,
+				       size_t size, enum dma_data_direction dir,
+				       struct dma_attrs *attrs)
+{
+	return sba_map_page(dev, virt_to_page(addr),
+			    (unsigned long)addr & ~PAGE_MASK, size, dir, attrs);
+}
 
 #ifdef ENABLE_MARK_CLEAN
 static SBA_INLINE void
@@ -1026,8 +1036,8 @@ sba_mark_clean(struct ioc *ioc, dma_addr_t iova, size_t size)
  *
  * See Documentation/PCI/PCI-DMA-mapping.txt
  */
-void sba_unmap_single_attrs(struct device *dev, dma_addr_t iova, size_t size,
-			    int dir, struct dma_attrs *attrs)
+static void sba_unmap_page(struct device *dev, dma_addr_t iova, size_t size,
+			   enum dma_data_direction dir, struct dma_attrs *attrs)
 {
 	struct ioc *ioc;
 #if DELAYED_RESOURCE_CNT > 0
@@ -1094,7 +1104,12 @@ void sba_unmap_single_attrs(struct device *dev, dma_addr_t iova, size_t size,
 	spin_unlock_irqrestore(&ioc->res_lock, flags);
 #endif /* DELAYED_RESOURCE_CNT == 0 */
 }
-EXPORT_SYMBOL(sba_unmap_single_attrs);
+
+void sba_unmap_single_attrs(struct device *dev, dma_addr_t iova, size_t size,
+			    enum dma_data_direction dir, struct dma_attrs *attrs)
+{
+	sba_unmap_page(dev, iova, size, dir, attrs);
+}
 
 /**
  * sba_alloc_coherent - allocate/map shared mem for DMA
@@ -1104,7 +1119,7 @@ EXPORT_SYMBOL(sba_unmap_single_attrs);
  *
  * See Documentation/PCI/PCI-DMA-mapping.txt
  */
-void *
+static void *
 sba_alloc_coherent (struct device *dev, size_t size, dma_addr_t *dma_handle, gfp_t flags)
 {
 	struct ioc *ioc;
@@ -1167,7 +1182,8 @@ sba_alloc_coherent (struct device *dev, size_t size, dma_addr_t *dma_handle, gfp
  *
  * See Documentation/PCI/PCI-DMA-mapping.txt
  */
-void sba_free_coherent (struct device *dev, size_t size, void *vaddr, dma_addr_t dma_handle)
+static void sba_free_coherent (struct device *dev, size_t size, void *vaddr,
+			       dma_addr_t dma_handle)
 {
 	sba_unmap_single_attrs(dev, dma_handle, size, 0, NULL);
 	free_pages((unsigned long) vaddr, get_order(size));
@@ -1422,8 +1438,9 @@ sba_coalesce_chunks(struct ioc *ioc, struct device *dev,
  *
  * See Documentation/PCI/PCI-DMA-mapping.txt
  */
-int sba_map_sg_attrs(struct device *dev, struct scatterlist *sglist, int nents,
-		     int dir, struct dma_attrs *attrs)
+static int sba_map_sg_attrs(struct device *dev, struct scatterlist *sglist,
+			    int nents, enum dma_data_direction dir,
+			    struct dma_attrs *attrs)
 {
 	struct ioc *ioc;
 	int coalesced, filled = 0;
@@ -1502,7 +1519,6 @@ int sba_map_sg_attrs(struct device *dev, struct scatterlist *sglist, int nents,
 
 	return filled;
 }
-EXPORT_SYMBOL(sba_map_sg_attrs);
 
 /**
  * sba_unmap_sg_attrs - unmap Scatter/Gather list
@@ -1514,8 +1530,9 @@ EXPORT_SYMBOL(sba_map_sg_attrs);
  *
  * See Documentation/PCI/PCI-DMA-mapping.txt
  */
-void sba_unmap_sg_attrs(struct device *dev, struct scatterlist *sglist,
-			int nents, int dir, struct dma_attrs *attrs)
+static void sba_unmap_sg_attrs(struct device *dev, struct scatterlist *sglist,
+			       int nents, enum dma_data_direction dir,
+			       struct dma_attrs *attrs)
 {
 #ifdef ASSERT_PDIR_SANITY
 	struct ioc *ioc;
@@ -1551,7 +1568,6 @@ void sba_unmap_sg_attrs(struct device *dev, struct scatterlist *sglist,
 #endif
 
 }
-EXPORT_SYMBOL(sba_unmap_sg_attrs);
 
 /**************************************************************
 *
@@ -2064,6 +2080,8 @@ static struct acpi_driver acpi_sba_ioc_driver = {
 	},
 };
 
+extern struct dma_map_ops swiotlb_dma_ops;
+
 static int __init
 sba_init(void)
 {
@@ -2077,6 +2095,7 @@ sba_init(void)
 	 * a successful kdump kernel boot is to use the swiotlb.
 	 */
 	if (is_kdump_kernel()) {
+		dma_ops = &swiotlb_dma_ops;
 		if (swiotlb_late_init_with_default_size(64 * (1<<20)) != 0)
 			panic("Unable to initialize software I/O TLB:"
 				  " Try machvec=dig boot option");
@@ -2092,6 +2111,7 @@ sba_init(void)
 		 * If we didn't find something sba_iommu can claim, we
 		 * need to setup the swiotlb and switch to the dig machvec.
 		 */
+		dma_ops = &swiotlb_dma_ops;
 		if (swiotlb_late_init_with_default_size(64 * (1<<20)) != 0)
 			panic("Unable to find SBA IOMMU or initialize "
 			      "software I/O TLB: Try machvec=dig boot option");
@@ -2138,15 +2158,13 @@ nosbagart(char *str)
 	return 1;
 }
 
-int
-sba_dma_supported (struct device *dev, u64 mask)
+static int sba_dma_supported (struct device *dev, u64 mask)
 {
 	/* make sure it's at least 32bit capable */
 	return ((mask & 0xFFFFFFFFUL) == 0xFFFFFFFFUL);
 }
 
-int
-sba_dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
+static int sba_dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
 {
 	return 0;
 }
@@ -2176,7 +2194,22 @@ sba_page_override(char *str)
 
 __setup("sbapagesize=",sba_page_override);
 
-EXPORT_SYMBOL(sba_dma_mapping_error);
-EXPORT_SYMBOL(sba_dma_supported);
-EXPORT_SYMBOL(sba_alloc_coherent);
-EXPORT_SYMBOL(sba_free_coherent);
+struct dma_map_ops sba_dma_ops = {
+	.alloc_coherent		= sba_alloc_coherent,
+	.free_coherent		= sba_free_coherent,
+	.map_page		= sba_map_page,
+	.unmap_page		= sba_unmap_page,
+	.map_sg			= sba_map_sg_attrs,
+	.unmap_sg		= sba_unmap_sg_attrs,
+	.sync_single_for_cpu	= machvec_dma_sync_single,
+	.sync_sg_for_cpu	= machvec_dma_sync_sg,
+	.sync_single_for_device	= machvec_dma_sync_single,
+	.sync_sg_for_device	= machvec_dma_sync_sg,
+	.dma_supported		= sba_dma_supported,
+	.mapping_error		= sba_dma_mapping_error,
+};
+
+void sba_dma_init(void)
+{
+	dma_ops = &sba_dma_ops;
+}
diff --git a/arch/ia64/hp/sim/simserial.c b/arch/ia64/hp/sim/simserial.c
index 24b1ad5334c..2bef5261d96 100644
--- a/arch/ia64/hp/sim/simserial.c
+++ b/arch/ia64/hp/sim/simserial.c
@@ -24,6 +24,7 @@
 #include <linux/major.h>
 #include <linux/fcntl.h>
 #include <linux/mm.h>
+#include <linux/seq_file.h>
 #include <linux/slab.h>
 #include <linux/capability.h>
 #include <linux/console.h>
@@ -848,38 +849,36 @@ static int rs_open(struct tty_struct *tty, struct file * filp)
  * /proc fs routines....
  */
 
-static inline int line_info(char *buf, struct serial_state *state)
+static inline void line_info(struct seq_file *m, struct serial_state *state)
 {
-	return sprintf(buf, "%d: uart:%s port:%lX irq:%d\n",
+	seq_printf(m, "%d: uart:%s port:%lX irq:%d\n",
 		       state->line, uart_config[state->type].name,
 		       state->port, state->irq);
 }
 
-static int rs_read_proc(char *page, char **start, off_t off, int count,
-		 int *eof, void *data)
+static int rs_proc_show(struct seq_file *m, void *v)
 {
-	int i, len = 0, l;
-	off_t	begin = 0;
-
-	len += sprintf(page, "simserinfo:1.0 driver:%s\n", serial_version);
-	for (i = 0; i < NR_PORTS && len < 4000; i++) {
-		l = line_info(page + len, &rs_table[i]);
-		len += l;
-		if (len+begin > off+count)
-			goto done;
-		if (len+begin < off) {
-			begin += len;
-			len = 0;
-		}
-	}
-	*eof = 1;
-done:
-	if (off >= len+begin)
-		return 0;
-	*start = page + (begin-off);
-	return ((count < begin+len-off) ? count : begin+len-off);
+	int i;
+
+	seq_printf(m, "simserinfo:1.0 driver:%s\n", serial_version);
+	for (i = 0; i < NR_PORTS; i++)
+		line_info(m, &rs_table[i]);
+	return 0;
 }
 
+static int rs_proc_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, rs_proc_show, NULL);
+}
+
+static const struct file_operations rs_proc_fops = {
+	.owner		= THIS_MODULE,
+	.open		= rs_proc_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
 /*
  * ---------------------------------------------------------------------
  * rs_init() and friends
@@ -917,7 +916,7 @@ static const struct tty_operations hp_ops = {
 	.start = rs_start,
 	.hangup = rs_hangup,
 	.wait_until_sent = rs_wait_until_sent,
-	.read_proc = rs_read_proc,
+	.proc_fops = &rs_proc_fops,
 };
 
 /*
diff --git a/arch/ia64/ia32/ia32_entry.S b/arch/ia64/ia32/ia32_entry.S
index a46f8395e9a..af9405cd70e 100644
--- a/arch/ia64/ia32/ia32_entry.S
+++ b/arch/ia64/ia32/ia32_entry.S
@@ -240,7 +240,7 @@ ia32_syscall_table:
 	data8 sys_ni_syscall
 	data8 sys_umask		  /* 60 */
 	data8 sys_chroot
-	data8 sys_ustat
+	data8 compat_sys_ustat
 	data8 sys_dup2
 	data8 sys_getppid
 	data8 sys_getpgrp	  /* 65 */
diff --git a/arch/ia64/include/asm/dma-mapping.h b/arch/ia64/include/asm/dma-mapping.h
index 1f912d92758..36c0009dbec 100644
--- a/arch/ia64/include/asm/dma-mapping.h
+++ b/arch/ia64/include/asm/dma-mapping.h
@@ -11,99 +11,128 @@
 
 #define ARCH_HAS_DMA_GET_REQUIRED_MASK
 
-struct dma_mapping_ops {
-	int             (*mapping_error)(struct device *dev,
-					 dma_addr_t dma_addr);
-	void*           (*alloc_coherent)(struct device *dev, size_t size,
-				dma_addr_t *dma_handle, gfp_t gfp);
-	void            (*free_coherent)(struct device *dev, size_t size,
-				void *vaddr, dma_addr_t dma_handle);
-	dma_addr_t      (*map_single)(struct device *hwdev, unsigned long ptr,
-				size_t size, int direction);
-	void            (*unmap_single)(struct device *dev, dma_addr_t addr,
-				size_t size, int direction);
-	void            (*sync_single_for_cpu)(struct device *hwdev,
-				dma_addr_t dma_handle, size_t size,
-				int direction);
-	void            (*sync_single_for_device)(struct device *hwdev,
-				dma_addr_t dma_handle, size_t size,
-				int direction);
-	void            (*sync_single_range_for_cpu)(struct device *hwdev,
-				dma_addr_t dma_handle, unsigned long offset,
-				size_t size, int direction);
-	void            (*sync_single_range_for_device)(struct device *hwdev,
-				dma_addr_t dma_handle, unsigned long offset,
-				size_t size, int direction);
-	void            (*sync_sg_for_cpu)(struct device *hwdev,
-				struct scatterlist *sg, int nelems,
-				int direction);
-	void            (*sync_sg_for_device)(struct device *hwdev,
-				struct scatterlist *sg, int nelems,
-				int direction);
-	int             (*map_sg)(struct device *hwdev, struct scatterlist *sg,
-				int nents, int direction);
-	void            (*unmap_sg)(struct device *hwdev,
-				struct scatterlist *sg, int nents,
-				int direction);
-	int             (*dma_supported_op)(struct device *hwdev, u64 mask);
-	int		is_phys;
-};
-
-extern struct dma_mapping_ops *dma_ops;
+extern struct dma_map_ops *dma_ops;
 extern struct ia64_machine_vector ia64_mv;
 extern void set_iommu_machvec(void);
 
-#define dma_alloc_coherent(dev, size, handle, gfp)	\
-	platform_dma_alloc_coherent(dev, size, handle, (gfp) | GFP_DMA)
+extern void machvec_dma_sync_single(struct device *, dma_addr_t, size_t,
+				    enum dma_data_direction);
+extern void machvec_dma_sync_sg(struct device *, struct scatterlist *, int,
+				enum dma_data_direction);
 
-/* coherent mem. is cheap */
-static inline void *
-dma_alloc_noncoherent(struct device *dev, size_t size, dma_addr_t *dma_handle,
-		      gfp_t flag)
+static inline void *dma_alloc_coherent(struct device *dev, size_t size,
+				       dma_addr_t *daddr, gfp_t gfp)
 {
-	return dma_alloc_coherent(dev, size, dma_handle, flag);
+	struct dma_map_ops *ops = platform_dma_get_ops(dev);
+	return ops->alloc_coherent(dev, size, daddr, gfp);
 }
-#define dma_free_coherent	platform_dma_free_coherent
-static inline void
-dma_free_noncoherent(struct device *dev, size_t size, void *cpu_addr,
-		     dma_addr_t dma_handle)
+
+static inline void dma_free_coherent(struct device *dev, size_t size,
+				     void *caddr, dma_addr_t daddr)
+{
+	struct dma_map_ops *ops = platform_dma_get_ops(dev);
+	ops->free_coherent(dev, size, caddr, daddr);
+}
+
+#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
+#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
+
+static inline dma_addr_t dma_map_single_attrs(struct device *dev,
+					      void *caddr, size_t size,
+					      enum dma_data_direction dir,
+					      struct dma_attrs *attrs)
+{
+	struct dma_map_ops *ops = platform_dma_get_ops(dev);
+	return ops->map_page(dev, virt_to_page(caddr),
+			     (unsigned long)caddr & ~PAGE_MASK, size,
+			     dir, attrs);
+}
+
+static inline void dma_unmap_single_attrs(struct device *dev, dma_addr_t daddr,
+					  size_t size,
+					  enum dma_data_direction dir,
+					  struct dma_attrs *attrs)
+{
+	struct dma_map_ops *ops = platform_dma_get_ops(dev);
+	ops->unmap_page(dev, daddr, size, dir, attrs);
+}
+
+#define dma_map_single(d, a, s, r) dma_map_single_attrs(d, a, s, r, NULL)
+#define dma_unmap_single(d, a, s, r) dma_unmap_single_attrs(d, a, s, r, NULL)
+
+static inline int dma_map_sg_attrs(struct device *dev, struct scatterlist *sgl,
+				   int nents, enum dma_data_direction dir,
+				   struct dma_attrs *attrs)
+{
+	struct dma_map_ops *ops = platform_dma_get_ops(dev);
+	return ops->map_sg(dev, sgl, nents, dir, attrs);
+}
+
+static inline void dma_unmap_sg_attrs(struct device *dev,
+				      struct scatterlist *sgl, int nents,
+				      enum dma_data_direction dir,
+				      struct dma_attrs *attrs)
+{
+	struct dma_map_ops *ops = platform_dma_get_ops(dev);
+	ops->unmap_sg(dev, sgl, nents, dir, attrs);
+}
+
+#define dma_map_sg(d, s, n, r) dma_map_sg_attrs(d, s, n, r, NULL)
+#define dma_unmap_sg(d, s, n, r) dma_unmap_sg_attrs(d, s, n, r, NULL)
+
+static inline void dma_sync_single_for_cpu(struct device *dev, dma_addr_t daddr,
+					   size_t size,
+					   enum dma_data_direction dir)
 {
-	dma_free_coherent(dev, size, cpu_addr, dma_handle);
+	struct dma_map_ops *ops = platform_dma_get_ops(dev);
+	ops->sync_single_for_cpu(dev, daddr, size, dir);
 }
-#define dma_map_single_attrs	platform_dma_map_single_attrs
-static inline dma_addr_t dma_map_single(struct device *dev, void *cpu_addr,
-					size_t size, int dir)
+
+static inline void dma_sync_sg_for_cpu(struct device *dev,
+				       struct scatterlist *sgl,
+				       int nents, enum dma_data_direction dir)
 {
-	return dma_map_single_attrs(dev, cpu_addr, size, dir, NULL);
+	struct dma_map_ops *ops = platform_dma_get_ops(dev);
+	ops->sync_sg_for_cpu(dev, sgl, nents, dir);
 }
-#define dma_map_sg_attrs	platform_dma_map_sg_attrs
-static inline int dma_map_sg(struct device *dev, struct scatterlist *sgl,
-			     int nents, int dir)
+
+static inline void dma_sync_single_for_device(struct device *dev,
+					      dma_addr_t daddr,
+					      size_t size,
+					      enum dma_data_direction dir)
 {
-	return dma_map_sg_attrs(dev, sgl, nents, dir, NULL);
+	struct dma_map_ops *ops = platform_dma_get_ops(dev);
+	ops->sync_single_for_device(dev, daddr, size, dir);
 }
-#define dma_unmap_single_attrs	platform_dma_unmap_single_attrs
-static inline void dma_unmap_single(struct device *dev, dma_addr_t cpu_addr,
-				    size_t size, int dir)
+
+static inline void dma_sync_sg_for_device(struct device *dev,
+					  struct scatterlist *sgl,
+					  int nents,
+					  enum dma_data_direction dir)
 {
-	return dma_unmap_single_attrs(dev, cpu_addr, size, dir, NULL);
+	struct dma_map_ops *ops = platform_dma_get_ops(dev);
+	ops->sync_sg_for_device(dev, sgl, nents, dir);
 }
-#define dma_unmap_sg_attrs	platform_dma_unmap_sg_attrs
-static inline void dma_unmap_sg(struct device *dev, struct scatterlist *sgl,
-				int nents, int dir)
+
+static inline int dma_mapping_error(struct device *dev, dma_addr_t daddr)
+{
+	struct dma_map_ops *ops = platform_dma_get_ops(dev);
+	return ops->mapping_error(dev, daddr);
+}
+
+static inline dma_addr_t dma_map_page(struct device *dev, struct page *page,
+				      size_t offset, size_t size,
+				      enum dma_data_direction dir)
 {
-	return dma_unmap_sg_attrs(dev, sgl, nents, dir, NULL);
+	struct dma_map_ops *ops = platform_dma_get_ops(dev);
+	return ops->map_page(dev, page, offset, size, dir, NULL);
 }
-#define dma_sync_single_for_cpu	platform_dma_sync_single_for_cpu
-#define dma_sync_sg_for_cpu	platform_dma_sync_sg_for_cpu
-#define dma_sync_single_for_device platform_dma_sync_single_for_device
-#define dma_sync_sg_for_device	platform_dma_sync_sg_for_device
-#define dma_mapping_error	platform_dma_mapping_error
 
-#define dma_map_page(dev, pg, off, size, dir)				\
-	dma_map_single(dev, page_address(pg) + (off), (size), (dir))
-#define dma_unmap_page(dev, dma_addr, size, dir)			\
-	dma_unmap_single(dev, dma_addr, size, dir)
+static inline void dma_unmap_page(struct device *dev, dma_addr_t addr,
+				  size_t size, enum dma_data_direction dir)
+{
+	dma_unmap_single(dev, addr, size, dir);
+}
 
 /*
  * Rest of this file is part of the "Advanced DMA API".  Use at your own risk.
@@ -115,7 +144,11 @@ static inline void dma_unmap_sg(struct device *dev, struct scatterlist *sgl,
 #define dma_sync_single_range_for_device(dev, dma_handle, offset, size, dir)	\
 	dma_sync_single_for_device(dev, dma_handle, size, dir)
 
-#define dma_supported		platform_dma_supported
+static inline int dma_supported(struct device *dev, u64 mask)
+{
+	struct dma_map_ops *ops = platform_dma_get_ops(dev);
+	return ops->dma_supported(dev, mask);
+}
 
 static inline int
 dma_set_mask (struct device *dev, u64 mask)
@@ -141,11 +174,4 @@ dma_cache_sync (struct device *dev, void *vaddr, size_t size,
 
 #define dma_is_consistent(d, h)	(1)	/* all we do is coherent memory... */
 
-static inline struct dma_mapping_ops *get_dma_ops(struct device *dev)
-{
-	return dma_ops;
-}
-
-
-
 #endif /* _ASM_IA64_DMA_MAPPING_H */
diff --git a/arch/ia64/include/asm/kvm.h b/arch/ia64/include/asm/kvm.h
index 2b0a38e8470..18a7e49abbc 100644
--- a/arch/ia64/include/asm/kvm.h
+++ b/arch/ia64/include/asm/kvm.h
@@ -165,7 +165,40 @@ struct saved_vpd {
 	unsigned long  vcpuid[5];
 	unsigned long  vpsr;
 	unsigned long  vpr;
-	unsigned long  vcr[128];
+	union {
+		unsigned long  vcr[128];
+		struct {
+			unsigned long dcr;
+			unsigned long itm;
+			unsigned long iva;
+			unsigned long rsv1[5];
+			unsigned long pta;
+			unsigned long rsv2[7];
+			unsigned long ipsr;
+			unsigned long isr;
+			unsigned long rsv3;
+			unsigned long iip;
+			unsigned long ifa;
+			unsigned long itir;
+			unsigned long iipa;
+			unsigned long ifs;
+			unsigned long iim;
+			unsigned long iha;
+			unsigned long rsv4[38];
+			unsigned long lid;
+			unsigned long ivr;
+			unsigned long tpr;
+			unsigned long eoi;
+			unsigned long irr[4];
+			unsigned long itv;
+			unsigned long pmv;
+			unsigned long cmcv;
+			unsigned long rsv5[5];
+			unsigned long lrr0;
+			unsigned long lrr1;
+			unsigned long rsv6[46];
+		};
+	};
 };
 
 struct kvm_regs {
@@ -213,4 +246,18 @@ struct kvm_sregs {
 struct kvm_fpu {
 };
 
+#define KVM_IA64_VCPU_STACK_SHIFT	16
+#define KVM_IA64_VCPU_STACK_SIZE	(1UL << KVM_IA64_VCPU_STACK_SHIFT)
+
+struct kvm_ia64_vcpu_stack {
+	unsigned char stack[KVM_IA64_VCPU_STACK_SIZE];
+};
+
+struct kvm_debug_exit_arch {
+};
+
+/* for KVM_SET_GUEST_DEBUG */
+struct kvm_guest_debug_arch {
+};
+
 #endif
diff --git a/arch/ia64/include/asm/kvm_host.h b/arch/ia64/include/asm/kvm_host.h
index 34866366165..4542651e6ac 100644
--- a/arch/ia64/include/asm/kvm_host.h
+++ b/arch/ia64/include/asm/kvm_host.h
@@ -112,7 +112,11 @@
 #define VCPU_STRUCT_SHIFT	16
 #define VCPU_STRUCT_SIZE	(__IA64_UL_CONST(1) << VCPU_STRUCT_SHIFT)
 
-#define KVM_STK_OFFSET		VCPU_STRUCT_SIZE
+/*
+ * This must match KVM_IA64_VCPU_STACK_{SHIFT,SIZE} arch/ia64/include/asm/kvm.h
+ */
+#define KVM_STK_SHIFT		16
+#define KVM_STK_OFFSET		(__IA64_UL_CONST(1)<< KVM_STK_SHIFT)
 
 #define KVM_VM_STRUCT_SHIFT	19
 #define KVM_VM_STRUCT_SIZE	(__IA64_UL_CONST(1) << KVM_VM_STRUCT_SHIFT)
@@ -153,10 +157,10 @@ struct kvm_vm_data {
 	struct kvm_vcpu_data vcpu_data[KVM_MAX_VCPUS];
 };
 
-#define VCPU_BASE(n)	KVM_VM_DATA_BASE + \
-				offsetof(struct kvm_vm_data, vcpu_data[n])
-#define VM_BASE		KVM_VM_DATA_BASE + \
-				offsetof(struct kvm_vm_data, kvm_vm_struct)
+#define VCPU_BASE(n)	(KVM_VM_DATA_BASE + \
+				offsetof(struct kvm_vm_data, vcpu_data[n]))
+#define KVM_VM_BASE	(KVM_VM_DATA_BASE + \
+				offsetof(struct kvm_vm_data, kvm_vm_struct))
 #define KVM_MEM_DIRTY_LOG_BASE	KVM_VM_DATA_BASE + \
 				offsetof(struct kvm_vm_data, kvm_mem_dirty_log)
 
@@ -235,8 +239,6 @@ struct kvm_vm_data {
 
 struct kvm;
 struct kvm_vcpu;
-struct kvm_guest_debug{
-};
 
 struct kvm_mmio_req {
 	uint64_t addr;          /*  physical address		*/
@@ -462,6 +464,8 @@ struct kvm_arch {
 	unsigned long	metaphysical_rr4;
 	unsigned long	vmm_init_rr;
 
+	int		online_vcpus;
+
 	struct kvm_ioapic *vioapic;
 	struct kvm_vm_stat stat;
 	struct kvm_sal_data rdv_sal_data;
diff --git a/arch/ia64/include/asm/machvec.h b/arch/ia64/include/asm/machvec.h
index fe87b212170..367d299d993 100644
--- a/arch/ia64/include/asm/machvec.h
+++ b/arch/ia64/include/asm/machvec.h
@@ -11,7 +11,6 @@
 #define _ASM_IA64_MACHVEC_H
 
 #include <linux/types.h>
-#include <linux/swiotlb.h>
 
 /* forward declarations: */
 struct device;
@@ -45,24 +44,8 @@ typedef void ia64_mv_kernel_launch_event_t(void);
 
 /* DMA-mapping interface: */
 typedef void ia64_mv_dma_init (void);
-typedef void *ia64_mv_dma_alloc_coherent (struct device *, size_t, dma_addr_t *, gfp_t);
-typedef void ia64_mv_dma_free_coherent (struct device *, size_t, void *, dma_addr_t);
-typedef dma_addr_t ia64_mv_dma_map_single (struct device *, void *, size_t, int);
-typedef void ia64_mv_dma_unmap_single (struct device *, dma_addr_t, size_t, int);
-typedef int ia64_mv_dma_map_sg (struct device *, struct scatterlist *, int, int);
-typedef void ia64_mv_dma_unmap_sg (struct device *, struct scatterlist *, int, int);
-typedef void ia64_mv_dma_sync_single_for_cpu (struct device *, dma_addr_t, size_t, int);
-typedef void ia64_mv_dma_sync_sg_for_cpu (struct device *, struct scatterlist *, int, int);
-typedef void ia64_mv_dma_sync_single_for_device (struct device *, dma_addr_t, size_t, int);
-typedef void ia64_mv_dma_sync_sg_for_device (struct device *, struct scatterlist *, int, int);
-typedef int ia64_mv_dma_mapping_error(struct device *, dma_addr_t dma_addr);
-typedef int ia64_mv_dma_supported (struct device *, u64);
-
-typedef dma_addr_t ia64_mv_dma_map_single_attrs (struct device *, void *, size_t, int, struct dma_attrs *);
-typedef void ia64_mv_dma_unmap_single_attrs (struct device *, dma_addr_t, size_t, int, struct dma_attrs *);
-typedef int ia64_mv_dma_map_sg_attrs (struct device *, struct scatterlist *, int, int, struct dma_attrs *);
-typedef void ia64_mv_dma_unmap_sg_attrs (struct device *, struct scatterlist *, int, int, struct dma_attrs *);
 typedef u64 ia64_mv_dma_get_required_mask (struct device *);
+typedef struct dma_map_ops *ia64_mv_dma_get_ops(struct device *);
 
 /*
  * WARNING: The legacy I/O space is _architected_.  Platforms are
@@ -114,8 +97,6 @@ machvec_noop_bus (struct pci_bus *bus)
 
 extern void machvec_setup (char **);
 extern void machvec_timer_interrupt (int, void *);
-extern void machvec_dma_sync_single (struct device *, dma_addr_t, size_t, int);
-extern void machvec_dma_sync_sg (struct device *, struct scatterlist *, int, int);
 extern void machvec_tlb_migrate_finish (struct mm_struct *);
 
 # if defined (CONFIG_IA64_HP_SIM)
@@ -148,19 +129,8 @@ extern void machvec_tlb_migrate_finish (struct mm_struct *);
 #  define platform_global_tlb_purge	ia64_mv.global_tlb_purge
 #  define platform_tlb_migrate_finish	ia64_mv.tlb_migrate_finish
 #  define platform_dma_init		ia64_mv.dma_init
-#  define platform_dma_alloc_coherent	ia64_mv.dma_alloc_coherent
-#  define platform_dma_free_coherent	ia64_mv.dma_free_coherent
-#  define platform_dma_map_single_attrs	ia64_mv.dma_map_single_attrs
-#  define platform_dma_unmap_single_attrs	ia64_mv.dma_unmap_single_attrs
-#  define platform_dma_map_sg_attrs	ia64_mv.dma_map_sg_attrs
-#  define platform_dma_unmap_sg_attrs	ia64_mv.dma_unmap_sg_attrs
-#  define platform_dma_sync_single_for_cpu ia64_mv.dma_sync_single_for_cpu
-#  define platform_dma_sync_sg_for_cpu	ia64_mv.dma_sync_sg_for_cpu
-#  define platform_dma_sync_single_for_device ia64_mv.dma_sync_single_for_device
-#  define platform_dma_sync_sg_for_device ia64_mv.dma_sync_sg_for_device
-#  define platform_dma_mapping_error		ia64_mv.dma_mapping_error
-#  define platform_dma_supported	ia64_mv.dma_supported
 #  define platform_dma_get_required_mask ia64_mv.dma_get_required_mask
+#  define platform_dma_get_ops		ia64_mv.dma_get_ops
 #  define platform_irq_to_vector	ia64_mv.irq_to_vector
 #  define platform_local_vector_to_irq	ia64_mv.local_vector_to_irq
 #  define platform_pci_get_legacy_mem	ia64_mv.pci_get_legacy_mem
@@ -203,19 +173,8 @@ struct ia64_machine_vector {
 	ia64_mv_global_tlb_purge_t *global_tlb_purge;
 	ia64_mv_tlb_migrate_finish_t *tlb_migrate_finish;
 	ia64_mv_dma_init *dma_init;
-	ia64_mv_dma_alloc_coherent *dma_alloc_coherent;
-	ia64_mv_dma_free_coherent *dma_free_coherent;
-	ia64_mv_dma_map_single_attrs *dma_map_single_attrs;
-	ia64_mv_dma_unmap_single_attrs *dma_unmap_single_attrs;
-	ia64_mv_dma_map_sg_attrs *dma_map_sg_attrs;
-	ia64_mv_dma_unmap_sg_attrs *dma_unmap_sg_attrs;
-	ia64_mv_dma_sync_single_for_cpu *dma_sync_single_for_cpu;
-	ia64_mv_dma_sync_sg_for_cpu *dma_sync_sg_for_cpu;
-	ia64_mv_dma_sync_single_for_device *dma_sync_single_for_device;
-	ia64_mv_dma_sync_sg_for_device *dma_sync_sg_for_device;
-	ia64_mv_dma_mapping_error *dma_mapping_error;
-	ia64_mv_dma_supported *dma_supported;
 	ia64_mv_dma_get_required_mask *dma_get_required_mask;
+	ia64_mv_dma_get_ops *dma_get_ops;
 	ia64_mv_irq_to_vector *irq_to_vector;
 	ia64_mv_local_vector_to_irq *local_vector_to_irq;
 	ia64_mv_pci_get_legacy_mem_t *pci_get_legacy_mem;
@@ -254,19 +213,8 @@ struct ia64_machine_vector {
 	platform_global_tlb_purge,		\
 	platform_tlb_migrate_finish,		\
 	platform_dma_init,			\
-	platform_dma_alloc_coherent,		\
-	platform_dma_free_coherent,		\
-	platform_dma_map_single_attrs,		\
-	platform_dma_unmap_single_attrs,	\
-	platform_dma_map_sg_attrs,		\
-	platform_dma_unmap_sg_attrs,		\
-	platform_dma_sync_single_for_cpu,	\
-	platform_dma_sync_sg_for_cpu,		\
-	platform_dma_sync_single_for_device,	\
-	platform_dma_sync_sg_for_device,	\
-	platform_dma_mapping_error,			\
-	platform_dma_supported,			\
 	platform_dma_get_required_mask,		\
+	platform_dma_get_ops,			\
 	platform_irq_to_vector,			\
 	platform_local_vector_to_irq,		\
 	platform_pci_get_legacy_mem,		\
@@ -302,6 +250,9 @@ extern void machvec_init_from_cmdline(const char *cmdline);
 #  error Unknown configuration.  Update arch/ia64/include/asm/machvec.h.
 # endif /* CONFIG_IA64_GENERIC */
 
+extern void swiotlb_dma_init(void);
+extern struct dma_map_ops *dma_get_ops(struct device *);
+
 /*
  * Define default versions so we can extend machvec for new platforms without having
  * to update the machvec files for all existing platforms.
@@ -332,43 +283,10 @@ extern void machvec_init_from_cmdline(const char *cmdline);
 # define platform_kernel_launch_event	machvec_noop
 #endif
 #ifndef platform_dma_init
-# define platform_dma_init		swiotlb_init
-#endif
-#ifndef platform_dma_alloc_coherent
-# define platform_dma_alloc_coherent	swiotlb_alloc_coherent
-#endif
-#ifndef platform_dma_free_coherent
-# define platform_dma_free_coherent	swiotlb_free_coherent
-#endif
-#ifndef platform_dma_map_single_attrs
-# define platform_dma_map_single_attrs	swiotlb_map_single_attrs
-#endif
-#ifndef platform_dma_unmap_single_attrs
-# define platform_dma_unmap_single_attrs	swiotlb_unmap_single_attrs
-#endif
-#ifndef platform_dma_map_sg_attrs
-# define platform_dma_map_sg_attrs	swiotlb_map_sg_attrs
-#endif
-#ifndef platform_dma_unmap_sg_attrs
-# define platform_dma_unmap_sg_attrs	swiotlb_unmap_sg_attrs
-#endif
-#ifndef platform_dma_sync_single_for_cpu
-# define platform_dma_sync_single_for_cpu	swiotlb_sync_single_for_cpu
-#endif
-#ifndef platform_dma_sync_sg_for_cpu
-# define platform_dma_sync_sg_for_cpu		swiotlb_sync_sg_for_cpu
-#endif
-#ifndef platform_dma_sync_single_for_device
-# define platform_dma_sync_single_for_device	swiotlb_sync_single_for_device
-#endif
-#ifndef platform_dma_sync_sg_for_device
-# define platform_dma_sync_sg_for_device	swiotlb_sync_sg_for_device
-#endif
-#ifndef platform_dma_mapping_error
-# define platform_dma_mapping_error		swiotlb_dma_mapping_error
+# define platform_dma_init		swiotlb_dma_init
 #endif
-#ifndef platform_dma_supported
-# define  platform_dma_supported	swiotlb_dma_supported
+#ifndef platform_dma_get_ops
+# define platform_dma_get_ops		dma_get_ops
 #endif
 #ifndef platform_dma_get_required_mask
 # define  platform_dma_get_required_mask	ia64_dma_get_required_mask
diff --git a/arch/ia64/include/asm/machvec_dig_vtd.h b/arch/ia64/include/asm/machvec_dig_vtd.h
index 3400b561e71..6ab1de5c45e 100644
--- a/arch/ia64/include/asm/machvec_dig_vtd.h
+++ b/arch/ia64/include/asm/machvec_dig_vtd.h
@@ -2,14 +2,6 @@
 #define _ASM_IA64_MACHVEC_DIG_VTD_h
 
 extern ia64_mv_setup_t			dig_setup;
-extern ia64_mv_dma_alloc_coherent	vtd_alloc_coherent;
-extern ia64_mv_dma_free_coherent	vtd_free_coherent;
-extern ia64_mv_dma_map_single_attrs	vtd_map_single_attrs;
-extern ia64_mv_dma_unmap_single_attrs	vtd_unmap_single_attrs;
-extern ia64_mv_dma_map_sg_attrs		vtd_map_sg_attrs;
-extern ia64_mv_dma_unmap_sg_attrs	vtd_unmap_sg_attrs;
-extern ia64_mv_dma_supported		iommu_dma_supported;
-extern ia64_mv_dma_mapping_error	vtd_dma_mapping_error;
 extern ia64_mv_dma_init			pci_iommu_alloc;
 
 /*
@@ -22,17 +14,5 @@ extern ia64_mv_dma_init			pci_iommu_alloc;
 #define platform_name				"dig_vtd"
 #define platform_setup				dig_setup
 #define platform_dma_init			pci_iommu_alloc
-#define platform_dma_alloc_coherent		vtd_alloc_coherent
-#define platform_dma_free_coherent		vtd_free_coherent
-#define platform_dma_map_single_attrs		vtd_map_single_attrs
-#define platform_dma_unmap_single_attrs		vtd_unmap_single_attrs
-#define platform_dma_map_sg_attrs		vtd_map_sg_attrs
-#define platform_dma_unmap_sg_attrs		vtd_unmap_sg_attrs
-#define platform_dma_sync_single_for_cpu	machvec_dma_sync_single
-#define platform_dma_sync_sg_for_cpu		machvec_dma_sync_sg
-#define platform_dma_sync_single_for_device	machvec_dma_sync_single
-#define platform_dma_sync_sg_for_device		machvec_dma_sync_sg
-#define platform_dma_supported			iommu_dma_supported
-#define platform_dma_mapping_error		vtd_dma_mapping_error
 
 #endif /* _ASM_IA64_MACHVEC_DIG_VTD_h */
diff --git a/arch/ia64/include/asm/machvec_hpzx1.h b/arch/ia64/include/asm/machvec_hpzx1.h
index 2f57f5144b9..3bd83d78a41 100644
--- a/arch/ia64/include/asm/machvec_hpzx1.h
+++ b/arch/ia64/include/asm/machvec_hpzx1.h
@@ -2,14 +2,7 @@
 #define _ASM_IA64_MACHVEC_HPZX1_h
 
 extern ia64_mv_setup_t			dig_setup;
-extern ia64_mv_dma_alloc_coherent	sba_alloc_coherent;
-extern ia64_mv_dma_free_coherent	sba_free_coherent;
-extern ia64_mv_dma_map_single_attrs	sba_map_single_attrs;
-extern ia64_mv_dma_unmap_single_attrs	sba_unmap_single_attrs;
-extern ia64_mv_dma_map_sg_attrs		sba_map_sg_attrs;
-extern ia64_mv_dma_unmap_sg_attrs	sba_unmap_sg_attrs;
-extern ia64_mv_dma_supported		sba_dma_supported;
-extern ia64_mv_dma_mapping_error	sba_dma_mapping_error;
+extern ia64_mv_dma_init			sba_dma_init;
 
 /*
  * This stuff has dual use!
@@ -20,18 +13,6 @@ extern ia64_mv_dma_mapping_error	sba_dma_mapping_error;
  */
 #define platform_name				"hpzx1"
 #define platform_setup				dig_setup
-#define platform_dma_init			machvec_noop
-#define platform_dma_alloc_coherent		sba_alloc_coherent
-#define platform_dma_free_coherent		sba_free_coherent
-#define platform_dma_map_single_attrs		sba_map_single_attrs
-#define platform_dma_unmap_single_attrs		sba_unmap_single_attrs
-#define platform_dma_map_sg_attrs		sba_map_sg_attrs
-#define platform_dma_unmap_sg_attrs		sba_unmap_sg_attrs
-#define platform_dma_sync_single_for_cpu	machvec_dma_sync_single
-#define platform_dma_sync_sg_for_cpu		machvec_dma_sync_sg
-#define platform_dma_sync_single_for_device	machvec_dma_sync_single
-#define platform_dma_sync_sg_for_device		machvec_dma_sync_sg
-#define platform_dma_supported			sba_dma_supported
-#define platform_dma_mapping_error		sba_dma_mapping_error
+#define platform_dma_init			sba_dma_init
 
 #endif /* _ASM_IA64_MACHVEC_HPZX1_h */
diff --git a/arch/ia64/include/asm/machvec_hpzx1_swiotlb.h b/arch/ia64/include/asm/machvec_hpzx1_swiotlb.h
index a842cdda827..1091ac39740 100644
--- a/arch/ia64/include/asm/machvec_hpzx1_swiotlb.h
+++ b/arch/ia64/include/asm/machvec_hpzx1_swiotlb.h
@@ -2,18 +2,7 @@
 #define _ASM_IA64_MACHVEC_HPZX1_SWIOTLB_h
 
 extern ia64_mv_setup_t				dig_setup;
-extern ia64_mv_dma_alloc_coherent		hwsw_alloc_coherent;
-extern ia64_mv_dma_free_coherent		hwsw_free_coherent;
-extern ia64_mv_dma_map_single_attrs		hwsw_map_single_attrs;
-extern ia64_mv_dma_unmap_single_attrs		hwsw_unmap_single_attrs;
-extern ia64_mv_dma_map_sg_attrs			hwsw_map_sg_attrs;
-extern ia64_mv_dma_unmap_sg_attrs		hwsw_unmap_sg_attrs;
-extern ia64_mv_dma_supported			hwsw_dma_supported;
-extern ia64_mv_dma_mapping_error		hwsw_dma_mapping_error;
-extern ia64_mv_dma_sync_single_for_cpu		hwsw_sync_single_for_cpu;
-extern ia64_mv_dma_sync_sg_for_cpu		hwsw_sync_sg_for_cpu;
-extern ia64_mv_dma_sync_single_for_device	hwsw_sync_single_for_device;
-extern ia64_mv_dma_sync_sg_for_device		hwsw_sync_sg_for_device;
+extern ia64_mv_dma_get_ops			hwsw_dma_get_ops;
 
 /*
  * This stuff has dual use!
@@ -23,20 +12,8 @@ extern ia64_mv_dma_sync_sg_for_device		hwsw_sync_sg_for_device;
  * the macros are used directly.
  */
 #define platform_name				"hpzx1_swiotlb"
-
 #define platform_setup				dig_setup
 #define platform_dma_init			machvec_noop
-#define platform_dma_alloc_coherent		hwsw_alloc_coherent
-#define platform_dma_free_coherent		hwsw_free_coherent
-#define platform_dma_map_single_attrs		hwsw_map_single_attrs
-#define platform_dma_unmap_single_attrs		hwsw_unmap_single_attrs
-#define platform_dma_map_sg_attrs		hwsw_map_sg_attrs
-#define platform_dma_unmap_sg_attrs		hwsw_unmap_sg_attrs
-#define platform_dma_supported			hwsw_dma_supported
-#define platform_dma_mapping_error		hwsw_dma_mapping_error
-#define platform_dma_sync_single_for_cpu	hwsw_sync_single_for_cpu
-#define platform_dma_sync_sg_for_cpu		hwsw_sync_sg_for_cpu
-#define platform_dma_sync_single_for_device	hwsw_sync_single_for_device
-#define platform_dma_sync_sg_for_device		hwsw_sync_sg_for_device
+#define platform_dma_get_ops			hwsw_dma_get_ops
 
 #endif /* _ASM_IA64_MACHVEC_HPZX1_SWIOTLB_h */
diff --git a/arch/ia64/include/asm/machvec_sn2.h b/arch/ia64/include/asm/machvec_sn2.h
index f1a6e0d6dfa..f061a30aac4 100644
--- a/arch/ia64/include/asm/machvec_sn2.h
+++ b/arch/ia64/include/asm/machvec_sn2.h
@@ -55,19 +55,8 @@ extern ia64_mv_readb_t __sn_readb_relaxed;
 extern ia64_mv_readw_t __sn_readw_relaxed;
 extern ia64_mv_readl_t __sn_readl_relaxed;
 extern ia64_mv_readq_t __sn_readq_relaxed;
-extern ia64_mv_dma_alloc_coherent	sn_dma_alloc_coherent;
-extern ia64_mv_dma_free_coherent	sn_dma_free_coherent;
-extern ia64_mv_dma_map_single_attrs	sn_dma_map_single_attrs;
-extern ia64_mv_dma_unmap_single_attrs	sn_dma_unmap_single_attrs;
-extern ia64_mv_dma_map_sg_attrs		sn_dma_map_sg_attrs;
-extern ia64_mv_dma_unmap_sg_attrs	sn_dma_unmap_sg_attrs;
-extern ia64_mv_dma_sync_single_for_cpu	sn_dma_sync_single_for_cpu;
-extern ia64_mv_dma_sync_sg_for_cpu	sn_dma_sync_sg_for_cpu;
-extern ia64_mv_dma_sync_single_for_device sn_dma_sync_single_for_device;
-extern ia64_mv_dma_sync_sg_for_device	sn_dma_sync_sg_for_device;
-extern ia64_mv_dma_mapping_error	sn_dma_mapping_error;
-extern ia64_mv_dma_supported		sn_dma_supported;
 extern ia64_mv_dma_get_required_mask	sn_dma_get_required_mask;
+extern ia64_mv_dma_init			sn_dma_init;
 extern ia64_mv_migrate_t		sn_migrate;
 extern ia64_mv_kernel_launch_event_t	sn_kernel_launch_event;
 extern ia64_mv_setup_msi_irq_t		sn_setup_msi_irq;
@@ -111,20 +100,8 @@ extern ia64_mv_pci_fixup_bus_t		sn_pci_fixup_bus;
 #define platform_pci_get_legacy_mem	sn_pci_get_legacy_mem
 #define platform_pci_legacy_read	sn_pci_legacy_read
 #define platform_pci_legacy_write	sn_pci_legacy_write
-#define platform_dma_init		machvec_noop
-#define platform_dma_alloc_coherent	sn_dma_alloc_coherent
-#define platform_dma_free_coherent	sn_dma_free_coherent
-#define platform_dma_map_single_attrs	sn_dma_map_single_attrs
-#define platform_dma_unmap_single_attrs	sn_dma_unmap_single_attrs
-#define platform_dma_map_sg_attrs	sn_dma_map_sg_attrs
-#define platform_dma_unmap_sg_attrs	sn_dma_unmap_sg_attrs
-#define platform_dma_sync_single_for_cpu sn_dma_sync_single_for_cpu
-#define platform_dma_sync_sg_for_cpu	sn_dma_sync_sg_for_cpu
-#define platform_dma_sync_single_for_device sn_dma_sync_single_for_device
-#define platform_dma_sync_sg_for_device	sn_dma_sync_sg_for_device
-#define platform_dma_mapping_error		sn_dma_mapping_error
-#define platform_dma_supported		sn_dma_supported
 #define platform_dma_get_required_mask	sn_dma_get_required_mask
+#define platform_dma_init		sn_dma_init
 #define platform_migrate		sn_migrate
 #define platform_kernel_launch_event    sn_kernel_launch_event
 #ifdef CONFIG_PCI_MSI
diff --git a/arch/ia64/include/asm/msidef.h b/arch/ia64/include/asm/msidef.h
new file mode 100644
index 00000000000..592c1047a0c
--- /dev/null
+++ b/arch/ia64/include/asm/msidef.h
@@ -0,0 +1,42 @@
+#ifndef _IA64_MSI_DEF_H
+#define _IA64_MSI_DEF_H
+
+/*
+ * Shifts for APIC-based data
+ */
+
+#define     MSI_DATA_VECTOR_SHIFT	0
+#define	    MSI_DATA_VECTOR(v)		(((u8)v) << MSI_DATA_VECTOR_SHIFT)
+#define     MSI_DATA_VECTOR_MASK	0xffffff00
+
+#define     MSI_DATA_DELIVERY_MODE_SHIFT	8
+#define     MSI_DATA_DELIVERY_FIXED	(0 << MSI_DATA_DELIVERY_MODE_SHIFT)
+#define     MSI_DATA_DELIVERY_LOWPRI	(1 << MSI_DATA_DELIVERY_MODE_SHIFT)
+
+#define     MSI_DATA_LEVEL_SHIFT	14
+#define     MSI_DATA_LEVEL_DEASSERT	(0 << MSI_DATA_LEVEL_SHIFT)
+#define     MSI_DATA_LEVEL_ASSERT	(1 << MSI_DATA_LEVEL_SHIFT)
+
+#define     MSI_DATA_TRIGGER_SHIFT	15
+#define     MSI_DATA_TRIGGER_EDGE	(0 << MSI_DATA_TRIGGER_SHIFT)
+#define     MSI_DATA_TRIGGER_LEVEL	(1 << MSI_DATA_TRIGGER_SHIFT)
+
+/*
+ * Shift/mask fields for APIC-based bus address
+ */
+
+#define     MSI_ADDR_DEST_ID_SHIFT	4
+#define     MSI_ADDR_HEADER		0xfee00000
+
+#define     MSI_ADDR_DEST_ID_MASK	0xfff0000f
+#define     MSI_ADDR_DEST_ID_CPU(cpu)	((cpu) << MSI_ADDR_DEST_ID_SHIFT)
+
+#define     MSI_ADDR_DEST_MODE_SHIFT	2
+#define     MSI_ADDR_DEST_MODE_PHYS	(0 << MSI_ADDR_DEST_MODE_SHIFT)
+#define	    MSI_ADDR_DEST_MODE_LOGIC	(1 << MSI_ADDR_DEST_MODE_SHIFT)
+
+#define     MSI_ADDR_REDIRECTION_SHIFT	3
+#define     MSI_ADDR_REDIRECTION_CPU	(0 << MSI_ADDR_REDIRECTION_SHIFT)
+#define     MSI_ADDR_REDIRECTION_LOWPRI	(1 << MSI_ADDR_REDIRECTION_SHIFT)
+
+#endif/* _IA64_MSI_DEF_H */
diff --git a/arch/ia64/include/asm/socket.h b/arch/ia64/include/asm/socket.h
index d5ef0aa3e31..745421225ec 100644
--- a/arch/ia64/include/asm/socket.h
+++ b/arch/ia64/include/asm/socket.h
@@ -63,4 +63,7 @@
 
 #define SO_MARK			36
 
+#define SO_TIMESTAMPING		37
+#define SCM_TIMESTAMPING	SO_TIMESTAMPING
+
 #endif /* _ASM_IA64_SOCKET_H */
diff --git a/arch/ia64/include/asm/topology.h b/arch/ia64/include/asm/topology.h
index 3193f4417e1..f260dcf2151 100644
--- a/arch/ia64/include/asm/topology.h
+++ b/arch/ia64/include/asm/topology.h
@@ -44,11 +44,6 @@
 #define parent_node(nid) (nid)
 
 /*
- * Returns the number of the first CPU on Node 'node'.
- */
-#define node_to_first_cpu(node) (cpumask_first(cpumask_of_node(node)))
-
-/*
  * Determines the node for a given pci bus
  */
 #define pcibus_to_node(bus) PCI_CONTROLLER(bus)->node
diff --git a/arch/ia64/kernel/Makefile b/arch/ia64/kernel/Makefile
index c381ea95489..f2778f2c4fd 100644
--- a/arch/ia64/kernel/Makefile
+++ b/arch/ia64/kernel/Makefile
@@ -7,7 +7,7 @@ extra-y	:= head.o init_task.o vmlinux.lds
 obj-y := acpi.o entry.o efi.o efi_stub.o gate-data.o fsys.o ia64_ksyms.o irq.o irq_ia64.o	\
 	 irq_lsapic.o ivt.o machvec.o pal.o patch.o process.o perfmon.o ptrace.o sal.o		\
 	 salinfo.o setup.o signal.o sys_ia64.o time.o traps.o unaligned.o \
-	 unwind.o mca.o mca_asm.o topology.o
+	 unwind.o mca.o mca_asm.o topology.o dma-mapping.o
 
 obj-$(CONFIG_IA64_BRL_EMU)	+= brl_emu.o
 obj-$(CONFIG_IA64_GENERIC)	+= acpi-ext.o
@@ -43,9 +43,7 @@ ifneq ($(CONFIG_IA64_ESI),)
 obj-y				+= esi_stub.o	# must be in kernel proper
 endif
 obj-$(CONFIG_DMAR)		+= pci-dma.o
-ifeq ($(CONFIG_DMAR), y)
 obj-$(CONFIG_SWIOTLB)		+= pci-swiotlb.o
-endif
 
 # The gate DSO image is built using a special linker script.
 targets += gate.so gate-syms.o
diff --git a/arch/ia64/kernel/dma-mapping.c b/arch/ia64/kernel/dma-mapping.c
new file mode 100644
index 00000000000..086a2aeb040
--- /dev/null
+++ b/arch/ia64/kernel/dma-mapping.c
@@ -0,0 +1,13 @@
+#include <linux/dma-mapping.h>
+
+/* Set this to 1 if there is a HW IOMMU in the system */
+int iommu_detected __read_mostly;
+
+struct dma_map_ops *dma_ops;
+EXPORT_SYMBOL(dma_ops);
+
+struct dma_map_ops *dma_get_ops(struct device *dev)
+{
+	return dma_ops;
+}
+EXPORT_SYMBOL(dma_get_ops);
diff --git a/arch/ia64/kernel/irq.c b/arch/ia64/kernel/irq.c
index 226233a6fa1..7429752ef5a 100644
--- a/arch/ia64/kernel/irq.c
+++ b/arch/ia64/kernel/irq.c
@@ -80,7 +80,7 @@ int show_interrupts(struct seq_file *p, void *v)
 		seq_printf(p, "%10u ", kstat_irqs(i));
 #else
 		for_each_online_cpu(j) {
-			seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
+			seq_printf(p, "%10u ", kstat_irqs_cpu(i, j));
 		}
 #endif
 		seq_printf(p, " %14s", irq_desc[i].chip->name);
diff --git a/arch/ia64/kernel/irq_ia64.c b/arch/ia64/kernel/irq_ia64.c
index 927ad027820..acc4d19ae62 100644
--- a/arch/ia64/kernel/irq_ia64.c
+++ b/arch/ia64/kernel/irq_ia64.c
@@ -493,16 +493,15 @@ ia64_handle_irq (ia64_vector vector, struct pt_regs *regs)
 	saved_tpr = ia64_getreg(_IA64_REG_CR_TPR);
 	ia64_srlz_d();
 	while (vector != IA64_SPURIOUS_INT_VECTOR) {
-		struct irq_desc *desc = irq_to_desc(vector);
+		int irq = local_vector_to_irq(vector);
+		struct irq_desc *desc = irq_to_desc(irq);
 
 		if (unlikely(IS_LOCAL_TLB_FLUSH(vector))) {
 			smp_local_flush_tlb();
-			kstat_incr_irqs_this_cpu(vector, desc);
-		} else if (unlikely(IS_RESCHEDULE(vector)))
-			kstat_incr_irqs_this_cpu(vector, desc);
-		else {
-			int irq = local_vector_to_irq(vector);
-
+			kstat_incr_irqs_this_cpu(irq, desc);
+		} else if (unlikely(IS_RESCHEDULE(vector))) {
+			kstat_incr_irqs_this_cpu(irq, desc);
+		} else {
 			ia64_setreg(_IA64_REG_CR_TPR, vector);
 			ia64_srlz_d();
 
@@ -545,24 +544,24 @@ void ia64_process_pending_intr(void)
 
 	vector = ia64_get_ivr();
 
-	 irq_enter();
-	 saved_tpr = ia64_getreg(_IA64_REG_CR_TPR);
-	 ia64_srlz_d();
+	irq_enter();
+	saved_tpr = ia64_getreg(_IA64_REG_CR_TPR);
+	ia64_srlz_d();
 
 	 /*
 	  * Perform normal interrupt style processing
 	  */
 	while (vector != IA64_SPURIOUS_INT_VECTOR) {
-		struct irq_desc *desc = irq_to_desc(vector);
+		int irq = local_vector_to_irq(vector);
+		struct irq_desc *desc = irq_to_desc(irq);
 
 		if (unlikely(IS_LOCAL_TLB_FLUSH(vector))) {
 			smp_local_flush_tlb();
-			kstat_incr_irqs_this_cpu(vector, desc);
-		} else if (unlikely(IS_RESCHEDULE(vector)))
-			kstat_incr_irqs_this_cpu(vector, desc);
-		else {
+			kstat_incr_irqs_this_cpu(irq, desc);
+		} else if (unlikely(IS_RESCHEDULE(vector))) {
+			kstat_incr_irqs_this_cpu(irq, desc);
+		} else {
 			struct pt_regs *old_regs = set_irq_regs(NULL);
-			int irq = local_vector_to_irq(vector);
 
 			ia64_setreg(_IA64_REG_CR_TPR, vector);
 			ia64_srlz_d();
diff --git a/arch/ia64/kernel/machvec.c b/arch/ia64/kernel/machvec.c
index 7ccb228ceed..d41a40ef80c 100644
--- a/arch/ia64/kernel/machvec.c
+++ b/arch/ia64/kernel/machvec.c
@@ -1,5 +1,5 @@
 #include <linux/module.h>
-
+#include <linux/dma-mapping.h>
 #include <asm/machvec.h>
 #include <asm/system.h>
 
@@ -75,14 +75,16 @@ machvec_timer_interrupt (int irq, void *dev_id)
 EXPORT_SYMBOL(machvec_timer_interrupt);
 
 void
-machvec_dma_sync_single (struct device *hwdev, dma_addr_t dma_handle, size_t size, int dir)
+machvec_dma_sync_single(struct device *hwdev, dma_addr_t dma_handle, size_t size,
+			enum dma_data_direction dir)
 {
 	mb();
 }
 EXPORT_SYMBOL(machvec_dma_sync_single);
 
 void
-machvec_dma_sync_sg (struct device *hwdev, struct scatterlist *sg, int n, int dir)
+machvec_dma_sync_sg(struct device *hwdev, struct scatterlist *sg, int n,
+		    enum dma_data_direction dir)
 {
 	mb();
 }
diff --git a/arch/ia64/kernel/msi_ia64.c b/arch/ia64/kernel/msi_ia64.c
index dcb6b7c51ea..2b15e233f7f 100644
--- a/arch/ia64/kernel/msi_ia64.c
+++ b/arch/ia64/kernel/msi_ia64.c
@@ -7,44 +7,7 @@
 #include <linux/msi.h>
 #include <linux/dmar.h>
 #include <asm/smp.h>
-
-/*
- * Shifts for APIC-based data
- */
-
-#define MSI_DATA_VECTOR_SHIFT		0
-#define	    MSI_DATA_VECTOR(v)		(((u8)v) << MSI_DATA_VECTOR_SHIFT)
-#define MSI_DATA_VECTOR_MASK		0xffffff00
-
-#define MSI_DATA_DELIVERY_SHIFT		8
-#define     MSI_DATA_DELIVERY_FIXED	(0 << MSI_DATA_DELIVERY_SHIFT)
-#define     MSI_DATA_DELIVERY_LOWPRI	(1 << MSI_DATA_DELIVERY_SHIFT)
-
-#define MSI_DATA_LEVEL_SHIFT		14
-#define     MSI_DATA_LEVEL_DEASSERT	(0 << MSI_DATA_LEVEL_SHIFT)
-#define     MSI_DATA_LEVEL_ASSERT	(1 << MSI_DATA_LEVEL_SHIFT)
-
-#define MSI_DATA_TRIGGER_SHIFT		15
-#define     MSI_DATA_TRIGGER_EDGE	(0 << MSI_DATA_TRIGGER_SHIFT)
-#define     MSI_DATA_TRIGGER_LEVEL	(1 << MSI_DATA_TRIGGER_SHIFT)
-
-/*
- * Shift/mask fields for APIC-based bus address
- */
-
-#define MSI_TARGET_CPU_SHIFT		4
-#define MSI_ADDR_HEADER			0xfee00000
-
-#define MSI_ADDR_DESTID_MASK		0xfff0000f
-#define     MSI_ADDR_DESTID_CPU(cpu)	((cpu) << MSI_TARGET_CPU_SHIFT)
-
-#define MSI_ADDR_DESTMODE_SHIFT		2
-#define     MSI_ADDR_DESTMODE_PHYS	(0 << MSI_ADDR_DESTMODE_SHIFT)
-#define	    MSI_ADDR_DESTMODE_LOGIC	(1 << MSI_ADDR_DESTMODE_SHIFT)
-
-#define MSI_ADDR_REDIRECTION_SHIFT	3
-#define     MSI_ADDR_REDIRECTION_CPU	(0 << MSI_ADDR_REDIRECTION_SHIFT)
-#define     MSI_ADDR_REDIRECTION_LOWPRI	(1 << MSI_ADDR_REDIRECTION_SHIFT)
+#include <asm/msidef.h>
 
 static struct irq_chip	ia64_msi_chip;
 
@@ -65,8 +28,8 @@ static void ia64_set_msi_irq_affinity(unsigned int irq,
 	read_msi_msg(irq, &msg);
 
 	addr = msg.address_lo;
-	addr &= MSI_ADDR_DESTID_MASK;
-	addr |= MSI_ADDR_DESTID_CPU(cpu_physical_id(cpu));
+	addr &= MSI_ADDR_DEST_ID_MASK;
+	addr |= MSI_ADDR_DEST_ID_CPU(cpu_physical_id(cpu));
 	msg.address_lo = addr;
 
 	data = msg.data;
@@ -98,9 +61,9 @@ int ia64_setup_msi_irq(struct pci_dev *pdev, struct msi_desc *desc)
 	msg.address_hi = 0;
 	msg.address_lo =
 		MSI_ADDR_HEADER |
-		MSI_ADDR_DESTMODE_PHYS |
+		MSI_ADDR_DEST_MODE_PHYS |
 		MSI_ADDR_REDIRECTION_CPU |
-		MSI_ADDR_DESTID_CPU(dest_phys_id);
+		MSI_ADDR_DEST_ID_CPU(dest_phys_id);
 
 	msg.data =
 		MSI_DATA_TRIGGER_EDGE |
@@ -183,8 +146,8 @@ static void dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
 
 	msg.data &= ~MSI_DATA_VECTOR_MASK;
 	msg.data |= MSI_DATA_VECTOR(cfg->vector);
-	msg.address_lo &= ~MSI_ADDR_DESTID_MASK;
-	msg.address_lo |= MSI_ADDR_DESTID_CPU(cpu_physical_id(cpu));
+	msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK;
+	msg.address_lo |= MSI_ADDR_DEST_ID_CPU(cpu_physical_id(cpu));
 
 	dmar_msi_write(irq, &msg);
 	cpumask_copy(irq_desc[irq].affinity, mask);
@@ -215,9 +178,9 @@ msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_msg *msg)
 	msg->address_hi = 0;
 	msg->address_lo =
 		MSI_ADDR_HEADER |
-		MSI_ADDR_DESTMODE_PHYS |
+		MSI_ADDR_DEST_MODE_PHYS |
 		MSI_ADDR_REDIRECTION_CPU |
-		MSI_ADDR_DESTID_CPU(dest);
+		MSI_ADDR_DEST_ID_CPU(dest);
 
 	msg->data =
 		MSI_DATA_TRIGGER_EDGE |
diff --git a/arch/ia64/kernel/palinfo.c b/arch/ia64/kernel/palinfo.c
index e5c57f413ca..a4f19c70aad 100644
--- a/arch/ia64/kernel/palinfo.c
+++ b/arch/ia64/kernel/palinfo.c
@@ -1002,8 +1002,6 @@ create_palinfo_proc_entries(unsigned int cpu)
 		*pdir = create_proc_read_entry(
 				palinfo_entries[j].name, 0, cpu_dir,
 				palinfo_read_entry, (void *)f.value);
-		if (*pdir)
-			(*pdir)->owner = THIS_MODULE;
 		pdir++;
 	}
 }
diff --git a/arch/ia64/kernel/pci-dma.c b/arch/ia64/kernel/pci-dma.c
index d0ada067a4a..e4cb443bb98 100644
--- a/arch/ia64/kernel/pci-dma.c
+++ b/arch/ia64/kernel/pci-dma.c
@@ -32,9 +32,6 @@ int force_iommu __read_mostly = 1;
 int force_iommu __read_mostly;
 #endif
 
-/* Set this to 1 if there is a HW IOMMU in the system */
-int iommu_detected __read_mostly;
-
 /* Dummy device used for NULL arguments (normally ISA). Better would
    be probably a smaller DMA mask, but this is bug-to-bug compatible
    to i386. */
@@ -44,18 +41,7 @@ struct device fallback_dev = {
 	.dma_mask = &fallback_dev.coherent_dma_mask,
 };
 
-void __init pci_iommu_alloc(void)
-{
-	/*
-	 * The order of these functions is important for
-	 * fall-back/fail-over reasons
-	 */
-	detect_intel_iommu();
-
-#ifdef CONFIG_SWIOTLB
-	pci_swiotlb_init();
-#endif
-}
+extern struct dma_map_ops intel_dma_ops;
 
 static int __init pci_iommu_init(void)
 {
@@ -79,15 +65,12 @@ iommu_dma_init(void)
 	return;
 }
 
-struct dma_mapping_ops *dma_ops;
-EXPORT_SYMBOL(dma_ops);
-
 int iommu_dma_supported(struct device *dev, u64 mask)
 {
-	struct dma_mapping_ops *ops = get_dma_ops(dev);
+	struct dma_map_ops *ops = platform_dma_get_ops(dev);
 
-	if (ops->dma_supported_op)
-		return ops->dma_supported_op(dev, mask);
+	if (ops->dma_supported)
+		return ops->dma_supported(dev, mask);
 
 	/* Copied from i386. Doesn't make much sense, because it will
 	   only work for pci_alloc_coherent.
@@ -116,4 +99,25 @@ int iommu_dma_supported(struct device *dev, u64 mask)
 }
 EXPORT_SYMBOL(iommu_dma_supported);
 
+void __init pci_iommu_alloc(void)
+{
+	dma_ops = &intel_dma_ops;
+
+	dma_ops->sync_single_for_cpu = machvec_dma_sync_single;
+	dma_ops->sync_sg_for_cpu = machvec_dma_sync_sg;
+	dma_ops->sync_single_for_device = machvec_dma_sync_single;
+	dma_ops->sync_sg_for_device = machvec_dma_sync_sg;
+	dma_ops->dma_supported = iommu_dma_supported;
+
+	/*
+	 * The order of these functions is important for
+	 * fall-back/fail-over reasons
+	 */
+	detect_intel_iommu();
+
+#ifdef CONFIG_SWIOTLB
+	pci_swiotlb_init();
+#endif
+}
+
 #endif
diff --git a/arch/ia64/kernel/pci-swiotlb.c b/arch/ia64/kernel/pci-swiotlb.c
index 16c50516dbc..573f02c39a0 100644
--- a/arch/ia64/kernel/pci-swiotlb.c
+++ b/arch/ia64/kernel/pci-swiotlb.c
@@ -13,23 +13,37 @@
 int swiotlb __read_mostly;
 EXPORT_SYMBOL(swiotlb);
 
-struct dma_mapping_ops swiotlb_dma_ops = {
-	.mapping_error = swiotlb_dma_mapping_error,
-	.alloc_coherent = swiotlb_alloc_coherent,
+static void *ia64_swiotlb_alloc_coherent(struct device *dev, size_t size,
+					 dma_addr_t *dma_handle, gfp_t gfp)
+{
+	if (dev->coherent_dma_mask != DMA_64BIT_MASK)
+		gfp |= GFP_DMA;
+	return swiotlb_alloc_coherent(dev, size, dma_handle, gfp);
+}
+
+struct dma_map_ops swiotlb_dma_ops = {
+	.alloc_coherent = ia64_swiotlb_alloc_coherent,
 	.free_coherent = swiotlb_free_coherent,
-	.map_single = swiotlb_map_single,
-	.unmap_single = swiotlb_unmap_single,
+	.map_page = swiotlb_map_page,
+	.unmap_page = swiotlb_unmap_page,
+	.map_sg = swiotlb_map_sg_attrs,
+	.unmap_sg = swiotlb_unmap_sg_attrs,
 	.sync_single_for_cpu = swiotlb_sync_single_for_cpu,
 	.sync_single_for_device = swiotlb_sync_single_for_device,
 	.sync_single_range_for_cpu = swiotlb_sync_single_range_for_cpu,
 	.sync_single_range_for_device = swiotlb_sync_single_range_for_device,
 	.sync_sg_for_cpu = swiotlb_sync_sg_for_cpu,
 	.sync_sg_for_device = swiotlb_sync_sg_for_device,
-	.map_sg = swiotlb_map_sg,
-	.unmap_sg = swiotlb_unmap_sg,
-	.dma_supported_op = swiotlb_dma_supported,
+	.dma_supported = swiotlb_dma_supported,
+	.mapping_error = swiotlb_dma_mapping_error,
 };
 
+void __init swiotlb_dma_init(void)
+{
+	dma_ops = &swiotlb_dma_ops;
+	swiotlb_init();
+}
+
 void __init pci_swiotlb_init(void)
 {
 	if (!iommu_detected) {
diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c
index 0e499757309..5c0f408cfd7 100644
--- a/arch/ia64/kernel/perfmon.c
+++ b/arch/ia64/kernel/perfmon.c
@@ -2196,7 +2196,7 @@ pfmfs_delete_dentry(struct dentry *dentry)
 	return 1;
 }
 
-static struct dentry_operations pfmfs_dentry_operations = {
+static const struct dentry_operations pfmfs_dentry_operations = {
 	.d_delete = pfmfs_delete_dentry,
 };
 
diff --git a/arch/ia64/kernel/time.c b/arch/ia64/kernel/time.c
index f0ebb342409..d6747bae52d 100644
--- a/arch/ia64/kernel/time.c
+++ b/arch/ia64/kernel/time.c
@@ -20,6 +20,7 @@
 #include <linux/efi.h>
 #include <linux/timex.h>
 #include <linux/clocksource.h>
+#include <linux/platform_device.h>
 
 #include <asm/machvec.h>
 #include <asm/delay.h>
@@ -405,6 +406,21 @@ static struct irqaction timer_irqaction = {
 	.name =		"timer"
 };
 
+static struct platform_device rtc_efi_dev = {
+	.name = "rtc-efi",
+	.id = -1,
+};
+
+static int __init rtc_init(void)
+{
+	if (platform_device_register(&rtc_efi_dev) < 0)
+		printk(KERN_ERR "unable to register rtc device...\n");
+
+	/* not necessarily an error */
+	return 0;
+}
+module_init(rtc_init);
+
 void __init
 time_init (void)
 {
diff --git a/arch/ia64/kvm/Kconfig b/arch/ia64/kvm/Kconfig
index f833a0b4188..0a2d6b86075 100644
--- a/arch/ia64/kvm/Kconfig
+++ b/arch/ia64/kvm/Kconfig
@@ -4,6 +4,10 @@
 config HAVE_KVM
 	bool
 
+config HAVE_KVM_IRQCHIP
+       bool
+       default y
+
 menuconfig VIRTUALIZATION
 	bool "Virtualization"
 	depends on HAVE_KVM || IA64
diff --git a/arch/ia64/kvm/irq.h b/arch/ia64/kvm/irq.h
index c6786e8b1bf..c0785a72827 100644
--- a/arch/ia64/kvm/irq.h
+++ b/arch/ia64/kvm/irq.h
@@ -23,6 +23,8 @@
 #ifndef __IRQ_H
 #define __IRQ_H
 
+#include "lapic.h"
+
 static inline int irqchip_in_kernel(struct kvm *kvm)
 {
 	return 1;
diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c
index 28f982045f2..076b00d1dbf 100644
--- a/arch/ia64/kvm/kvm-ia64.c
+++ b/arch/ia64/kvm/kvm-ia64.c
@@ -182,7 +182,7 @@ int kvm_dev_ioctl_check_extension(long ext)
 	switch (ext) {
 	case KVM_CAP_IRQCHIP:
 	case KVM_CAP_MP_STATE:
-
+	case KVM_CAP_IRQ_INJECT_STATUS:
 		r = 1;
 		break;
 	case KVM_CAP_COALESCED_MMIO:
@@ -314,7 +314,7 @@ static struct kvm_vcpu *lid_to_vcpu(struct kvm *kvm, unsigned long id,
 	union ia64_lid lid;
 	int i;
 
-	for (i = 0; i < KVM_MAX_VCPUS; i++) {
+	for (i = 0; i < kvm->arch.online_vcpus; i++) {
 		if (kvm->vcpus[i]) {
 			lid.val = VCPU_LID(kvm->vcpus[i]);
 			if (lid.id == id && lid.eid == eid)
@@ -388,7 +388,7 @@ static int handle_global_purge(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 
 	call_data.ptc_g_data = p->u.ptc_g_data;
 
-	for (i = 0; i < KVM_MAX_VCPUS; i++) {
+	for (i = 0; i < kvm->arch.online_vcpus; i++) {
 		if (!kvm->vcpus[i] || kvm->vcpus[i]->arch.mp_state ==
 						KVM_MP_STATE_UNINITIALIZED ||
 					vcpu == kvm->vcpus[i])
@@ -788,6 +788,8 @@ struct  kvm *kvm_arch_create_vm(void)
 		return ERR_PTR(-ENOMEM);
 	kvm_init_vm(kvm);
 
+	kvm->arch.online_vcpus = 0;
+
 	return kvm;
 
 }
@@ -919,7 +921,13 @@ long kvm_arch_vm_ioctl(struct file *filp,
 		r = kvm_ioapic_init(kvm);
 		if (r)
 			goto out;
+		r = kvm_setup_default_irq_routing(kvm);
+		if (r) {
+			kfree(kvm->arch.vioapic);
+			goto out;
+		}
 		break;
+	case KVM_IRQ_LINE_STATUS:
 	case KVM_IRQ_LINE: {
 		struct kvm_irq_level irq_event;
 
@@ -927,10 +935,17 @@ long kvm_arch_vm_ioctl(struct file *filp,
 		if (copy_from_user(&irq_event, argp, sizeof irq_event))
 			goto out;
 		if (irqchip_in_kernel(kvm)) {
+			__s32 status;
 			mutex_lock(&kvm->lock);
-			kvm_set_irq(kvm, KVM_USERSPACE_IRQ_SOURCE_ID,
+			status = kvm_set_irq(kvm, KVM_USERSPACE_IRQ_SOURCE_ID,
 				    irq_event.irq, irq_event.level);
 			mutex_unlock(&kvm->lock);
+			if (ioctl == KVM_IRQ_LINE_STATUS) {
+				irq_event.status = status;
+				if (copy_to_user(argp, &irq_event,
+							sizeof irq_event))
+					goto out;
+			}
 			r = 0;
 		}
 		break;
@@ -1149,7 +1164,7 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
 
 		/*Initialize itc offset for vcpus*/
 		itc_offset = 0UL - ia64_getreg(_IA64_REG_AR_ITC);
-		for (i = 0; i < KVM_MAX_VCPUS; i++) {
+		for (i = 0; i < kvm->arch.online_vcpus; i++) {
 			v = (struct kvm_vcpu *)((char *)vcpu +
 					sizeof(struct kvm_vcpu_data) * i);
 			v->arch.itc_offset = itc_offset;
@@ -1283,6 +1298,8 @@ struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm,
 		goto fail;
 	}
 
+	kvm->arch.online_vcpus++;
+
 	return vcpu;
 fail:
 	return ERR_PTR(r);
@@ -1303,8 +1320,8 @@ int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
 	return -EINVAL;
 }
 
-int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu,
-		struct kvm_debug_guest *dbg)
+int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
+					struct kvm_guest_debug *dbg)
 {
 	return -EINVAL;
 }
@@ -1421,6 +1438,23 @@ int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
 	return 0;
 }
 
+int kvm_arch_vcpu_ioctl_get_stack(struct kvm_vcpu *vcpu,
+				  struct kvm_ia64_vcpu_stack *stack)
+{
+	memcpy(stack, vcpu, sizeof(struct kvm_ia64_vcpu_stack));
+	return 0;
+}
+
+int kvm_arch_vcpu_ioctl_set_stack(struct kvm_vcpu *vcpu,
+				  struct kvm_ia64_vcpu_stack *stack)
+{
+	memcpy(vcpu + 1, &stack->stack[0] + sizeof(struct kvm_vcpu),
+	       sizeof(struct kvm_ia64_vcpu_stack) - sizeof(struct kvm_vcpu));
+
+	vcpu->arch.exit_data = ((struct kvm_vcpu *)stack)->arch.exit_data;
+	return 0;
+}
+
 void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
 {
 
@@ -1430,9 +1464,78 @@ void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
 
 
 long kvm_arch_vcpu_ioctl(struct file *filp,
-		unsigned int ioctl, unsigned long arg)
+			 unsigned int ioctl, unsigned long arg)
 {
-	return -EINVAL;
+	struct kvm_vcpu *vcpu = filp->private_data;
+	void __user *argp = (void __user *)arg;
+	struct kvm_ia64_vcpu_stack *stack = NULL;
+	long r;
+
+	switch (ioctl) {
+	case KVM_IA64_VCPU_GET_STACK: {
+		struct kvm_ia64_vcpu_stack __user *user_stack;
+	        void __user *first_p = argp;
+
+		r = -EFAULT;
+		if (copy_from_user(&user_stack, first_p, sizeof(void *)))
+			goto out;
+
+		if (!access_ok(VERIFY_WRITE, user_stack,
+			       sizeof(struct kvm_ia64_vcpu_stack))) {
+			printk(KERN_INFO "KVM_IA64_VCPU_GET_STACK: "
+			       "Illegal user destination address for stack\n");
+			goto out;
+		}
+		stack = kzalloc(sizeof(struct kvm_ia64_vcpu_stack), GFP_KERNEL);
+		if (!stack) {
+			r = -ENOMEM;
+			goto out;
+		}
+
+		r = kvm_arch_vcpu_ioctl_get_stack(vcpu, stack);
+		if (r)
+			goto out;
+
+		if (copy_to_user(user_stack, stack,
+				 sizeof(struct kvm_ia64_vcpu_stack)))
+			goto out;
+
+		break;
+	}
+	case KVM_IA64_VCPU_SET_STACK: {
+		struct kvm_ia64_vcpu_stack __user *user_stack;
+	        void __user *first_p = argp;
+
+		r = -EFAULT;
+		if (copy_from_user(&user_stack, first_p, sizeof(void *)))
+			goto out;
+
+		if (!access_ok(VERIFY_READ, user_stack,
+			    sizeof(struct kvm_ia64_vcpu_stack))) {
+			printk(KERN_INFO "KVM_IA64_VCPU_SET_STACK: "
+			       "Illegal user address for stack\n");
+			goto out;
+		}
+		stack = kmalloc(sizeof(struct kvm_ia64_vcpu_stack), GFP_KERNEL);
+		if (!stack) {
+			r = -ENOMEM;
+			goto out;
+		}
+		if (copy_from_user(stack, user_stack,
+				   sizeof(struct kvm_ia64_vcpu_stack)))
+			goto out;
+
+		r = kvm_arch_vcpu_ioctl_set_stack(vcpu, stack);
+		break;
+	}
+
+	default:
+		r = -EINVAL;
+	}
+
+out:
+	kfree(stack);
+	return r;
 }
 
 int kvm_arch_set_memory_region(struct kvm *kvm,
@@ -1472,7 +1575,7 @@ void kvm_arch_flush_shadow(struct kvm *kvm)
 }
 
 long kvm_arch_dev_ioctl(struct file *filp,
-		unsigned int ioctl, unsigned long arg)
+			unsigned int ioctl, unsigned long arg)
 {
 	return -EINVAL;
 }
@@ -1737,7 +1840,7 @@ struct kvm_vcpu *kvm_get_lowest_prio_vcpu(struct kvm *kvm, u8 vector,
 	struct kvm_vcpu *lvcpu = kvm->vcpus[0];
 	int i;
 
-	for (i = 1; i < KVM_MAX_VCPUS; i++) {
+	for (i = 1; i < kvm->arch.online_vcpus; i++) {
 		if (!kvm->vcpus[i])
 			continue;
 		if (lvcpu->arch.xtp > kvm->vcpus[i]->arch.xtp)
diff --git a/arch/ia64/kvm/kvm_fw.c b/arch/ia64/kvm/kvm_fw.c
index cb7600bdff9..a8ae52ed563 100644
--- a/arch/ia64/kvm/kvm_fw.c
+++ b/arch/ia64/kvm/kvm_fw.c
@@ -227,6 +227,18 @@ static struct ia64_pal_retval pal_proc_get_features(struct kvm_vcpu *vcpu)
 	return result;
 }
 
+static struct ia64_pal_retval pal_register_info(struct kvm_vcpu *vcpu)
+{
+
+	struct ia64_pal_retval result = {0, 0, 0, 0};
+	long in0, in1, in2, in3;
+
+	kvm_get_pal_call_data(vcpu, &in0, &in1, &in2, &in3);
+	result.status = ia64_pal_register_info(in1, &result.v1, &result.v2);
+
+	return result;
+}
+
 static struct ia64_pal_retval pal_cache_info(struct kvm_vcpu *vcpu)
 {
 
@@ -268,8 +280,12 @@ static struct ia64_pal_retval pal_vm_summary(struct kvm_vcpu *vcpu)
 static struct ia64_pal_retval pal_vm_info(struct kvm_vcpu *vcpu)
 {
 	struct ia64_pal_retval result;
+	unsigned long in0, in1, in2, in3;
 
-	INIT_PAL_STATUS_UNIMPLEMENTED(result);
+	kvm_get_pal_call_data(vcpu, &in0, &in1, &in2, &in3);
+
+	result.status = ia64_pal_vm_info(in1, in2,
+			(pal_tc_info_u_t *)&result.v1, &result.v2);
 
 	return result;
 }
@@ -292,6 +308,108 @@ static void prepare_for_halt(struct kvm_vcpu *vcpu)
 	vcpu->arch.timer_fired = 0;
 }
 
+static struct ia64_pal_retval pal_perf_mon_info(struct kvm_vcpu *vcpu)
+{
+	long status;
+	unsigned long in0, in1, in2, in3, r9;
+	unsigned long pm_buffer[16];
+
+	kvm_get_pal_call_data(vcpu, &in0, &in1, &in2, &in3);
+	status = ia64_pal_perf_mon_info(pm_buffer,
+				(pal_perf_mon_info_u_t *) &r9);
+	if (status != 0) {
+		printk(KERN_DEBUG"PAL_PERF_MON_INFO fails ret=%ld\n", status);
+	} else {
+		if (in1)
+			memcpy((void *)in1, pm_buffer, sizeof(pm_buffer));
+		else {
+			status = PAL_STATUS_EINVAL;
+			printk(KERN_WARNING"Invalid parameters "
+						"for PAL call:0x%lx!\n", in0);
+		}
+	}
+	return (struct ia64_pal_retval){status, r9, 0, 0};
+}
+
+static struct ia64_pal_retval pal_halt_info(struct kvm_vcpu *vcpu)
+{
+	unsigned long in0, in1, in2, in3;
+	long status;
+	unsigned long res = 1000UL | (1000UL << 16) | (10UL << 32)
+					| (1UL << 61) | (1UL << 60);
+
+	kvm_get_pal_call_data(vcpu, &in0, &in1, &in2, &in3);
+	if (in1) {
+		memcpy((void *)in1, &res, sizeof(res));
+		status = 0;
+	} else{
+		status = PAL_STATUS_EINVAL;
+		printk(KERN_WARNING"Invalid parameters "
+					"for PAL call:0x%lx!\n", in0);
+	}
+
+	return (struct ia64_pal_retval){status, 0, 0, 0};
+}
+
+static struct ia64_pal_retval pal_mem_attrib(struct kvm_vcpu *vcpu)
+{
+	unsigned long r9;
+	long status;
+
+	status = ia64_pal_mem_attrib(&r9);
+
+	return (struct ia64_pal_retval){status, r9, 0, 0};
+}
+
+static void remote_pal_prefetch_visibility(void *v)
+{
+	s64 trans_type = (s64)v;
+	ia64_pal_prefetch_visibility(trans_type);
+}
+
+static struct ia64_pal_retval pal_prefetch_visibility(struct kvm_vcpu *vcpu)
+{
+	struct ia64_pal_retval result = {0, 0, 0, 0};
+	unsigned long in0, in1, in2, in3;
+	kvm_get_pal_call_data(vcpu, &in0, &in1, &in2, &in3);
+	result.status = ia64_pal_prefetch_visibility(in1);
+	if (result.status == 0) {
+		/* Must be performed on all remote processors
+		in the coherence domain. */
+		smp_call_function(remote_pal_prefetch_visibility,
+					(void *)in1, 1);
+		/* Unnecessary on remote processor for other vcpus!*/
+		result.status = 1;
+	}
+	return result;
+}
+
+static void remote_pal_mc_drain(void *v)
+{
+	ia64_pal_mc_drain();
+}
+
+static struct ia64_pal_retval pal_get_brand_info(struct kvm_vcpu *vcpu)
+{
+	struct ia64_pal_retval result = {0, 0, 0, 0};
+	unsigned long in0, in1, in2, in3;
+
+	kvm_get_pal_call_data(vcpu, &in0, &in1, &in2, &in3);
+
+	if (in1 == 0 && in2) {
+		char brand_info[128];
+		result.status = ia64_pal_get_brand_info(brand_info);
+		if (result.status == PAL_STATUS_SUCCESS)
+			memcpy((void *)in2, brand_info, 128);
+	} else {
+		result.status = PAL_STATUS_REQUIRES_MEMORY;
+		printk(KERN_WARNING"Invalid parameters for "
+					"PAL call:0x%lx!\n", in0);
+	}
+
+	return result;
+}
+
 int kvm_pal_emul(struct kvm_vcpu *vcpu, struct kvm_run *run)
 {
 
@@ -300,14 +418,22 @@ int kvm_pal_emul(struct kvm_vcpu *vcpu, struct kvm_run *run)
 	int ret = 1;
 
 	gr28 = kvm_get_pal_call_index(vcpu);
-	/*printk("pal_call index:%lx\n",gr28);*/
 	switch (gr28) {
 	case PAL_CACHE_FLUSH:
 		result = pal_cache_flush(vcpu);
 		break;
+	case PAL_MEM_ATTRIB:
+		result = pal_mem_attrib(vcpu);
+		break;
 	case PAL_CACHE_SUMMARY:
 		result = pal_cache_summary(vcpu);
 		break;
+	case PAL_PERF_MON_INFO:
+		result = pal_perf_mon_info(vcpu);
+		break;
+	case PAL_HALT_INFO:
+		result = pal_halt_info(vcpu);
+		break;
 	case PAL_HALT_LIGHT:
 	{
 		INIT_PAL_STATUS_SUCCESS(result);
@@ -317,6 +443,16 @@ int kvm_pal_emul(struct kvm_vcpu *vcpu, struct kvm_run *run)
 	}
 		break;
 
+	case PAL_PREFETCH_VISIBILITY:
+		result = pal_prefetch_visibility(vcpu);
+		break;
+	case PAL_MC_DRAIN:
+		result.status = ia64_pal_mc_drain();
+		/* FIXME: All vcpus likely call PAL_MC_DRAIN.
+		   That causes the congestion. */
+		smp_call_function(remote_pal_mc_drain, NULL, 1);
+		break;
+
 	case PAL_FREQ_RATIOS:
 		result = pal_freq_ratios(vcpu);
 		break;
@@ -346,6 +482,9 @@ int kvm_pal_emul(struct kvm_vcpu *vcpu, struct kvm_run *run)
 		INIT_PAL_STATUS_SUCCESS(result);
 		result.v1 = (1L << 32) | 1L;
 		break;
+	case PAL_REGISTER_INFO:
+		result = pal_register_info(vcpu);
+		break;
 	case PAL_VM_PAGE_SIZE:
 		result.status = ia64_pal_vm_page_size(&result.v0,
 							&result.v1);
@@ -365,12 +504,18 @@ int kvm_pal_emul(struct kvm_vcpu *vcpu, struct kvm_run *run)
 		result.status = ia64_pal_version(
 				(pal_version_u_t *)&result.v0,
 				(pal_version_u_t *)&result.v1);
-
 		break;
 	case PAL_FIXED_ADDR:
 		result.status = PAL_STATUS_SUCCESS;
 		result.v0 = vcpu->vcpu_id;
 		break;
+	case PAL_BRAND_INFO:
+		result = pal_get_brand_info(vcpu);
+		break;
+	case PAL_GET_PSTATE:
+	case PAL_CACHE_SHARED_INFO:
+		INIT_PAL_STATUS_UNIMPLEMENTED(result);
+		break;
 	default:
 		INIT_PAL_STATUS_UNIMPLEMENTED(result);
 		printk(KERN_WARNING"kvm: Unsupported pal call,"
diff --git a/arch/ia64/kvm/process.c b/arch/ia64/kvm/process.c
index 230eae482f3..b1dc80952d9 100644
--- a/arch/ia64/kvm/process.c
+++ b/arch/ia64/kvm/process.c
@@ -167,7 +167,6 @@ static u64 vcpu_get_itir_on_fault(struct kvm_vcpu *vcpu, u64 ifa)
 	return (rr1.val);
 }
 
-
 /*
  * Set vIFA & vITIR & vIHA, when vPSR.ic =1
  * Parameter:
@@ -222,8 +221,6 @@ void itlb_fault(struct kvm_vcpu *vcpu, u64 vadr)
 	inject_guest_interruption(vcpu, IA64_INST_TLB_VECTOR);
 }
 
-
-
 /*
  * Data Nested TLB Fault
  *  @ Data Nested TLB Vector
@@ -245,7 +242,6 @@ void alt_dtlb(struct kvm_vcpu *vcpu, u64 vadr)
 	inject_guest_interruption(vcpu, IA64_ALT_DATA_TLB_VECTOR);
 }
 
-
 /*
  * Data TLB Fault
  *  @ Data TLB vector
@@ -265,8 +261,6 @@ static void _vhpt_fault(struct kvm_vcpu *vcpu, u64 vadr)
 	/* If vPSR.ic, IFA, ITIR, IHA*/
 	set_ifa_itir_iha(vcpu, vadr, 1, 1, 1);
 	inject_guest_interruption(vcpu, IA64_VHPT_TRANS_VECTOR);
-
-
 }
 
 /*
@@ -279,7 +273,6 @@ void ivhpt_fault(struct kvm_vcpu *vcpu, u64 vadr)
 	_vhpt_fault(vcpu, vadr);
 }
 
-
 /*
  * VHPT Data Fault
  *  @ VHPT Translation vector
@@ -290,8 +283,6 @@ void dvhpt_fault(struct kvm_vcpu *vcpu, u64 vadr)
 	_vhpt_fault(vcpu, vadr);
 }
 
-
-
 /*
  * Deal with:
  *  General Exception vector
@@ -301,7 +292,6 @@ void _general_exception(struct kvm_vcpu *vcpu)
 	inject_guest_interruption(vcpu, IA64_GENEX_VECTOR);
 }
 
-
 /*
  * Illegal Operation Fault
  *  @ General Exception Vector
@@ -419,19 +409,16 @@ static void __page_not_present(struct kvm_vcpu *vcpu, u64 vadr)
 	inject_guest_interruption(vcpu, IA64_PAGE_NOT_PRESENT_VECTOR);
 }
 
-
 void data_page_not_present(struct kvm_vcpu *vcpu, u64 vadr)
 {
 	__page_not_present(vcpu, vadr);
 }
 
-
 void inst_page_not_present(struct kvm_vcpu *vcpu, u64 vadr)
 {
 	__page_not_present(vcpu, vadr);
 }
 
-
 /* Deal with
  *  Data access rights vector
  */
@@ -563,22 +550,64 @@ void reflect_interruption(u64 ifa, u64 isr, u64 iim,
 	inject_guest_interruption(vcpu, vector);
 }
 
+static unsigned long kvm_trans_pal_call_args(struct kvm_vcpu *vcpu,
+						unsigned long arg)
+{
+	struct thash_data *data;
+	unsigned long gpa, poff;
+
+	if (!is_physical_mode(vcpu)) {
+		/* Depends on caller to provide the DTR or DTC mapping.*/
+		data = vtlb_lookup(vcpu, arg, D_TLB);
+		if (data)
+			gpa = data->page_flags & _PAGE_PPN_MASK;
+		else {
+			data = vhpt_lookup(arg);
+			if (!data)
+				return 0;
+			gpa = data->gpaddr & _PAGE_PPN_MASK;
+		}
+
+		poff = arg & (PSIZE(data->ps) - 1);
+		arg = PAGEALIGN(gpa, data->ps) | poff;
+	}
+	arg = kvm_gpa_to_mpa(arg << 1 >> 1);
+
+	return (unsigned long)__va(arg);
+}
+
 static void set_pal_call_data(struct kvm_vcpu *vcpu)
 {
 	struct exit_ctl_data *p = &vcpu->arch.exit_data;
+	unsigned long gr28 = vcpu_get_gr(vcpu, 28);
+	unsigned long gr29 = vcpu_get_gr(vcpu, 29);
+	unsigned long gr30 = vcpu_get_gr(vcpu, 30);
 
 	/*FIXME:For static and stacked convention, firmware
 	 * has put the parameters in gr28-gr31 before
 	 * break to vmm  !!*/
 
-	p->u.pal_data.gr28 = vcpu_get_gr(vcpu, 28);
-	p->u.pal_data.gr29 = vcpu_get_gr(vcpu, 29);
-	p->u.pal_data.gr30 = vcpu_get_gr(vcpu, 30);
+	switch (gr28) {
+	case PAL_PERF_MON_INFO:
+	case PAL_HALT_INFO:
+		p->u.pal_data.gr29 =  kvm_trans_pal_call_args(vcpu, gr29);
+		p->u.pal_data.gr30 = vcpu_get_gr(vcpu, 30);
+		break;
+	case PAL_BRAND_INFO:
+		p->u.pal_data.gr29 = gr29;;
+		p->u.pal_data.gr30 = kvm_trans_pal_call_args(vcpu, gr30);
+		break;
+	default:
+		p->u.pal_data.gr29 = gr29;;
+		p->u.pal_data.gr30 = vcpu_get_gr(vcpu, 30);
+	}
+	p->u.pal_data.gr28 = gr28;
 	p->u.pal_data.gr31 = vcpu_get_gr(vcpu, 31);
+
 	p->exit_reason = EXIT_REASON_PAL_CALL;
 }
 
-static void set_pal_call_result(struct kvm_vcpu *vcpu)
+static void get_pal_call_result(struct kvm_vcpu *vcpu)
 {
 	struct exit_ctl_data *p = &vcpu->arch.exit_data;
 
@@ -606,7 +635,7 @@ static void set_sal_call_data(struct kvm_vcpu *vcpu)
 	p->exit_reason = EXIT_REASON_SAL_CALL;
 }
 
-static void set_sal_call_result(struct kvm_vcpu *vcpu)
+static void get_sal_call_result(struct kvm_vcpu *vcpu)
 {
 	struct exit_ctl_data *p = &vcpu->arch.exit_data;
 
@@ -629,13 +658,13 @@ void  kvm_ia64_handle_break(unsigned long ifa, struct kvm_pt_regs *regs,
 		if (iim == DOMN_PAL_REQUEST) {
 			set_pal_call_data(v);
 			vmm_transition(v);
-			set_pal_call_result(v);
+			get_pal_call_result(v);
 			vcpu_increment_iip(v);
 			return;
 		} else if (iim == DOMN_SAL_REQUEST) {
 			set_sal_call_data(v);
 			vmm_transition(v);
-			set_sal_call_result(v);
+			get_sal_call_result(v);
 			vcpu_increment_iip(v);
 			return;
 		}
@@ -703,7 +732,6 @@ void vhpi_detection(struct kvm_vcpu *vcpu)
 	}
 }
 
-
 void leave_hypervisor_tail(void)
 {
 	struct kvm_vcpu *v = current_vcpu;
@@ -737,7 +765,6 @@ void leave_hypervisor_tail(void)
 	}
 }
 
-
 static inline void handle_lds(struct kvm_pt_regs *regs)
 {
 	regs->cr_ipsr |= IA64_PSR_ED;
diff --git a/arch/ia64/kvm/vcpu.c b/arch/ia64/kvm/vcpu.c
index ecd526b5532..d4d28050587 100644
--- a/arch/ia64/kvm/vcpu.c
+++ b/arch/ia64/kvm/vcpu.c
@@ -112,7 +112,6 @@ void switch_to_physical_rid(struct kvm_vcpu *vcpu)
 	return;
 }
 
-
 void switch_to_virtual_rid(struct kvm_vcpu *vcpu)
 {
 	unsigned long psr;
@@ -166,8 +165,6 @@ void switch_mm_mode(struct kvm_vcpu *vcpu, struct ia64_psr old_psr,
 	return;
 }
 
-
-
 /*
  * In physical mode, insert tc/tr for region 0 and 4 uses
  * RID[0] and RID[4] which is for physical mode emulation.
@@ -269,7 +266,6 @@ static inline unsigned long fph_index(struct kvm_pt_regs *regs,
 	return rotate_reg(96, rrb_fr, (regnum - IA64_FIRST_ROTATING_FR));
 }
 
-
 /*
  * The inverse of the above: given bspstore and the number of
  * registers, calculate ar.bsp.
@@ -811,12 +807,15 @@ static inline void vcpu_set_itm(struct kvm_vcpu *vcpu, u64 val);
 static void vcpu_set_itc(struct kvm_vcpu *vcpu, u64 val)
 {
 	struct kvm_vcpu *v;
+	struct kvm *kvm;
 	int i;
 	long itc_offset = val - ia64_getreg(_IA64_REG_AR_ITC);
 	unsigned long vitv = VCPU(vcpu, itv);
 
+	kvm = (struct kvm *)KVM_VM_BASE;
+
 	if (vcpu->vcpu_id == 0) {
-		for (i = 0; i < KVM_MAX_VCPUS; i++) {
+		for (i = 0; i < kvm->arch.online_vcpus; i++) {
 			v = (struct kvm_vcpu *)((char *)vcpu +
 					sizeof(struct kvm_vcpu_data) * i);
 			VMX(v, itc_offset) = itc_offset;
@@ -1039,8 +1038,6 @@ u64 vcpu_tak(struct kvm_vcpu *vcpu, u64 vadr)
 	return key;
 }
 
-
-
 void kvm_thash(struct kvm_vcpu *vcpu, INST64 inst)
 {
 	unsigned long thash, vadr;
@@ -1050,7 +1047,6 @@ void kvm_thash(struct kvm_vcpu *vcpu, INST64 inst)
 	vcpu_set_gr(vcpu, inst.M46.r1, thash, 0);
 }
 
-
 void kvm_ttag(struct kvm_vcpu *vcpu, INST64 inst)
 {
 	unsigned long tag, vadr;
@@ -1131,7 +1127,6 @@ int vcpu_tpa(struct kvm_vcpu *vcpu, u64 vadr, u64 *padr)
 	return IA64_NO_FAULT;
 }
 
-
 int kvm_tpa(struct kvm_vcpu *vcpu, INST64 inst)
 {
 	unsigned long r1, r3;
@@ -1154,7 +1149,6 @@ void kvm_tak(struct kvm_vcpu *vcpu, INST64 inst)
 	vcpu_set_gr(vcpu, inst.M46.r1, r1, 0);
 }
 
-
 /************************************
  * Insert/Purge translation register/cache
  ************************************/
@@ -1385,7 +1379,6 @@ void kvm_mov_to_ar_reg(struct kvm_vcpu *vcpu, INST64 inst)
 	vcpu_set_itc(vcpu, r2);
 }
 
-
 void kvm_mov_from_ar_reg(struct kvm_vcpu *vcpu, INST64 inst)
 {
 	unsigned long r1;
@@ -1393,8 +1386,9 @@ void kvm_mov_from_ar_reg(struct kvm_vcpu *vcpu, INST64 inst)
 	r1 = vcpu_get_itc(vcpu);
 	vcpu_set_gr(vcpu, inst.M31.r1, r1, 0);
 }
+
 /**************************************************************************
-  struct kvm_vcpu*protection key register access routines
+  struct kvm_vcpu protection key register access routines
  **************************************************************************/
 
 unsigned long vcpu_get_pkr(struct kvm_vcpu *vcpu, unsigned long reg)
@@ -1407,20 +1401,6 @@ void vcpu_set_pkr(struct kvm_vcpu *vcpu, unsigned long reg, unsigned long val)
 	ia64_set_pkr(reg, val);
 }
 
-
-unsigned long vcpu_get_itir_on_fault(struct kvm_vcpu *vcpu, unsigned long ifa)
-{
-	union ia64_rr rr, rr1;
-
-	rr.val = vcpu_get_rr(vcpu, ifa);
-	rr1.val = 0;
-	rr1.ps = rr.ps;
-	rr1.rid = rr.rid;
-	return (rr1.val);
-}
-
-
-
 /********************************
  * Moves to privileged registers
  ********************************/
@@ -1464,8 +1444,6 @@ unsigned long vcpu_set_rr(struct kvm_vcpu *vcpu, unsigned long reg,
 	return (IA64_NO_FAULT);
 }
 
-
-
 void kvm_mov_to_rr(struct kvm_vcpu *vcpu, INST64 inst)
 {
 	unsigned long r3, r2;
@@ -1510,8 +1488,6 @@ void kvm_mov_to_pkr(struct kvm_vcpu *vcpu, INST64 inst)
 	vcpu_set_pkr(vcpu, r3, r2);
 }
 
-
-
 void kvm_mov_from_rr(struct kvm_vcpu *vcpu, INST64 inst)
 {
 	unsigned long r3, r1;
@@ -1557,7 +1533,6 @@ void kvm_mov_from_pmc(struct kvm_vcpu *vcpu, INST64 inst)
 	vcpu_set_gr(vcpu, inst.M43.r1, r1, 0);
 }
 
-
 unsigned long vcpu_get_cpuid(struct kvm_vcpu *vcpu, unsigned long reg)
 {
 	/* FIXME: This could get called as a result of a rsvd-reg fault */
@@ -1609,7 +1584,6 @@ unsigned long kvm_mov_to_cr(struct kvm_vcpu *vcpu, INST64 inst)
 	return 0;
 }
 
-
 unsigned long kvm_mov_from_cr(struct kvm_vcpu *vcpu, INST64 inst)
 {
 	unsigned long tgt = inst.M33.r1;
@@ -1633,8 +1607,6 @@ unsigned long kvm_mov_from_cr(struct kvm_vcpu *vcpu, INST64 inst)
 	return 0;
 }
 
-
-
 void vcpu_set_psr(struct kvm_vcpu *vcpu, unsigned long val)
 {
 
@@ -1776,9 +1748,6 @@ void vcpu_bsw1(struct kvm_vcpu *vcpu)
 	}
 }
 
-
-
-
 void vcpu_rfi(struct kvm_vcpu *vcpu)
 {
 	unsigned long ifs, psr;
@@ -1796,7 +1765,6 @@ void vcpu_rfi(struct kvm_vcpu *vcpu)
 	regs->cr_iip = VCPU(vcpu, iip);
 }
 
-
 /*
    VPSR can't keep track of below bits of guest PSR
    This function gets guest PSR
diff --git a/arch/ia64/kvm/vcpu.h b/arch/ia64/kvm/vcpu.h
index b2f12a562bd..042af92ced8 100644
--- a/arch/ia64/kvm/vcpu.h
+++ b/arch/ia64/kvm/vcpu.h
@@ -703,7 +703,7 @@ extern u64 guest_vhpt_lookup(u64 iha, u64 *pte);
 extern void thash_purge_entries(struct kvm_vcpu *v, u64 va, u64 ps);
 extern void thash_purge_entries_remote(struct kvm_vcpu *v, u64 va, u64 ps);
 extern u64 translate_phy_pte(u64 *pte, u64 itir, u64 va);
-extern int thash_purge_and_insert(struct kvm_vcpu *v, u64 pte,
+extern void thash_purge_and_insert(struct kvm_vcpu *v, u64 pte,
 		u64 itir, u64 ifa, int type);
 extern void thash_purge_all(struct kvm_vcpu *v);
 extern struct thash_data *vtlb_lookup(struct kvm_vcpu *v,
@@ -738,7 +738,7 @@ void kvm_init_vhpt(struct kvm_vcpu *v);
 void thash_init(struct thash_cb *hcb, u64 sz);
 
 void panic_vm(struct kvm_vcpu *v, const char *fmt, ...);
-
+u64 kvm_gpa_to_mpa(u64 gpa);
 extern u64 ia64_call_vsa(u64 proc, u64 arg1, u64 arg2, u64 arg3,
 		u64 arg4, u64 arg5, u64 arg6, u64 arg7);
 
diff --git a/arch/ia64/kvm/vtlb.c b/arch/ia64/kvm/vtlb.c
index 6b6307a3bd5..38232b37668 100644
--- a/arch/ia64/kvm/vtlb.c
+++ b/arch/ia64/kvm/vtlb.c
@@ -164,11 +164,11 @@ static void vhpt_insert(u64 pte, u64 itir, u64 ifa, u64 gpte)
 	unsigned long ps, gpaddr;
 
 	ps = itir_ps(itir);
+	rr.val = ia64_get_rr(ifa);
 
-	gpaddr = ((gpte & _PAGE_PPN_MASK) >> ps << ps) |
-		(ifa & ((1UL << ps) - 1));
+	 gpaddr = ((gpte & _PAGE_PPN_MASK) >> ps << ps) |
+					(ifa & ((1UL << ps) - 1));
 
-	rr.val = ia64_get_rr(ifa);
 	head = (struct thash_data *)ia64_thash(ifa);
 	head->etag = INVALID_TI_TAG;
 	ia64_mf();
@@ -412,16 +412,14 @@ u64 translate_phy_pte(u64 *pte, u64 itir, u64 va)
 
 /*
  * Purge overlap TCs and then insert the new entry to emulate itc ops.
- *    Notes: Only TC entry can purge and insert.
- *    1 indicates this is MMIO
+ * Notes: Only TC entry can purge and insert.
  */
-int thash_purge_and_insert(struct kvm_vcpu *v, u64 pte, u64 itir,
+void  thash_purge_and_insert(struct kvm_vcpu *v, u64 pte, u64 itir,
 						u64 ifa, int type)
 {
 	u64 ps;
 	u64 phy_pte, io_mask, index;
 	union ia64_rr vrr, mrr;
-	int ret = 0;
 
 	ps = itir_ps(itir);
 	vrr.val = vcpu_get_rr(v, ifa);
@@ -441,25 +439,19 @@ int thash_purge_and_insert(struct kvm_vcpu *v, u64 pte, u64 itir,
 		phy_pte &= ~_PAGE_MA_MASK;
 	}
 
-	if (pte & VTLB_PTE_IO)
-		ret = 1;
-
 	vtlb_purge(v, ifa, ps);
 	vhpt_purge(v, ifa, ps);
 
-	if (ps == mrr.ps) {
-		if (!(pte&VTLB_PTE_IO)) {
-			vhpt_insert(phy_pte, itir, ifa, pte);
-		} else {
-			vtlb_insert(v, pte, itir, ifa);
-			vcpu_quick_region_set(VMX(v, tc_regions), ifa);
-		}
-	} else if (ps > mrr.ps) {
+	if ((ps != mrr.ps) || (pte & VTLB_PTE_IO)) {
 		vtlb_insert(v, pte, itir, ifa);
 		vcpu_quick_region_set(VMX(v, tc_regions), ifa);
-		if (!(pte&VTLB_PTE_IO))
-			vhpt_insert(phy_pte, itir, ifa, pte);
-	} else {
+	}
+	if (pte & VTLB_PTE_IO)
+		return;
+
+	if (ps >= mrr.ps)
+		vhpt_insert(phy_pte, itir, ifa, pte);
+	else {
 		u64 psr;
 		phy_pte  &= ~PAGE_FLAGS_RV_MASK;
 		psr = ia64_clear_ic();
@@ -469,7 +461,6 @@ int thash_purge_and_insert(struct kvm_vcpu *v, u64 pte, u64 itir,
 	if (!(pte&VTLB_PTE_IO))
 		mark_pages_dirty(v, pte, ps);
 
-	return ret;
 }
 
 /*
@@ -509,7 +500,6 @@ void thash_purge_all(struct kvm_vcpu *v)
 	local_flush_tlb_all();
 }
 
-
 /*
  * Lookup the hash table and its collision chain to find an entry
  * covering this address rid:va or the entry.
@@ -517,7 +507,6 @@ void thash_purge_all(struct kvm_vcpu *v)
  * INPUT:
  *  in: TLB format for both VHPT & TLB.
  */
-
 struct thash_data *vtlb_lookup(struct kvm_vcpu *v, u64 va, int is_data)
 {
 	struct thash_data  *cch;
@@ -547,7 +536,6 @@ struct thash_data *vtlb_lookup(struct kvm_vcpu *v, u64 va, int is_data)
 	return NULL;
 }
 
-
 /*
  * Initialize internal control data before service.
  */
@@ -573,6 +561,10 @@ void thash_init(struct thash_cb *hcb, u64 sz)
 u64 kvm_get_mpt_entry(u64 gpfn)
 {
 	u64 *base = (u64 *) KVM_P2M_BASE;
+
+	if (gpfn >= (KVM_P2M_SIZE >> 3))
+		panic_vm(current_vcpu, "Invalid gpfn =%lx\n", gpfn);
+
 	return *(base + gpfn);
 }
 
@@ -589,7 +581,6 @@ u64 kvm_gpa_to_mpa(u64 gpa)
 	return (pte >> PAGE_SHIFT << PAGE_SHIFT) | (gpa & ~PAGE_MASK);
 }
 
-
 /*
  * Fetch guest bundle code.
  * INPUT:
@@ -631,7 +622,6 @@ int fetch_code(struct kvm_vcpu *vcpu, u64 gip, IA64_BUNDLE *pbundle)
 	return IA64_NO_FAULT;
 }
 
-
 void kvm_init_vhpt(struct kvm_vcpu *v)
 {
 	v->arch.vhpt.num = VHPT_NUM_ENTRIES;
diff --git a/arch/ia64/sn/kernel/sn2/prominfo_proc.c b/arch/ia64/sn/kernel/sn2/prominfo_proc.c
index 4dcce3d0e04..e6332881864 100644
--- a/arch/ia64/sn/kernel/sn2/prominfo_proc.c
+++ b/arch/ia64/sn/kernel/sn2/prominfo_proc.c
@@ -225,7 +225,6 @@ static struct proc_dir_entry *sgi_prominfo_entry;
 int __init prominfo_init(void)
 {
 	struct proc_dir_entry **entp;
-	struct proc_dir_entry *p;
 	cnodeid_t cnodeid;
 	unsigned long nasid;
 	int size;
@@ -246,14 +245,10 @@ int __init prominfo_init(void)
 		sprintf(name, "node%d", cnodeid);
 		*entp = proc_mkdir(name, sgi_prominfo_entry);
 		nasid = cnodeid_to_nasid(cnodeid);
-		p = create_proc_read_entry("fit", 0, *entp, read_fit_entry,
+		create_proc_read_entry("fit", 0, *entp, read_fit_entry,
 					   (void *)nasid);
-		if (p)
-			p->owner = THIS_MODULE;
-		p = create_proc_read_entry("version", 0, *entp,
+		create_proc_read_entry("version", 0, *entp,
 					   read_version_entry, (void *)nasid);
-		if (p)
-			p->owner = THIS_MODULE;
 		entp++;
 	}
 
diff --git a/arch/ia64/sn/pci/pci_dma.c b/arch/ia64/sn/pci/pci_dma.c
index 863f5017baa..8c130e8f00e 100644
--- a/arch/ia64/sn/pci/pci_dma.c
+++ b/arch/ia64/sn/pci/pci_dma.c
@@ -10,7 +10,7 @@
  */
 
 #include <linux/module.h>
-#include <linux/dma-attrs.h>
+#include <linux/dma-mapping.h>
 #include <asm/dma.h>
 #include <asm/sn/intr.h>
 #include <asm/sn/pcibus_provider_defs.h>
@@ -31,7 +31,7 @@
  * this function.  Of course, SN only supports devices that have 32 or more
  * address bits when using the PMU.
  */
-int sn_dma_supported(struct device *dev, u64 mask)
+static int sn_dma_supported(struct device *dev, u64 mask)
 {
 	BUG_ON(dev->bus != &pci_bus_type);
 
@@ -39,7 +39,6 @@ int sn_dma_supported(struct device *dev, u64 mask)
 		return 0;
 	return 1;
 }
-EXPORT_SYMBOL(sn_dma_supported);
 
 /**
  * sn_dma_set_mask - set the DMA mask
@@ -75,8 +74,8 @@ EXPORT_SYMBOL(sn_dma_set_mask);
  * queue for a SCSI controller).  See Documentation/DMA-API.txt for
  * more information.
  */
-void *sn_dma_alloc_coherent(struct device *dev, size_t size,
-			    dma_addr_t * dma_handle, gfp_t flags)
+static void *sn_dma_alloc_coherent(struct device *dev, size_t size,
+				   dma_addr_t * dma_handle, gfp_t flags)
 {
 	void *cpuaddr;
 	unsigned long phys_addr;
@@ -124,7 +123,6 @@ void *sn_dma_alloc_coherent(struct device *dev, size_t size,
 
 	return cpuaddr;
 }
-EXPORT_SYMBOL(sn_dma_alloc_coherent);
 
 /**
  * sn_pci_free_coherent - free memory associated with coherent DMAable region
@@ -136,8 +134,8 @@ EXPORT_SYMBOL(sn_dma_alloc_coherent);
  * Frees the memory allocated by dma_alloc_coherent(), potentially unmapping
  * any associated IOMMU mappings.
  */
-void sn_dma_free_coherent(struct device *dev, size_t size, void *cpu_addr,
-			  dma_addr_t dma_handle)
+static void sn_dma_free_coherent(struct device *dev, size_t size, void *cpu_addr,
+				 dma_addr_t dma_handle)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
 	struct sn_pcibus_provider *provider = SN_PCIDEV_BUSPROVIDER(pdev);
@@ -147,7 +145,6 @@ void sn_dma_free_coherent(struct device *dev, size_t size, void *cpu_addr,
 	provider->dma_unmap(pdev, dma_handle, 0);
 	free_pages((unsigned long)cpu_addr, get_order(size));
 }
-EXPORT_SYMBOL(sn_dma_free_coherent);
 
 /**
  * sn_dma_map_single_attrs - map a single page for DMA
@@ -173,10 +170,12 @@ EXPORT_SYMBOL(sn_dma_free_coherent);
  * TODO: simplify our interface;
  *       figure out how to save dmamap handle so can use two step.
  */
-dma_addr_t sn_dma_map_single_attrs(struct device *dev, void *cpu_addr,
-				   size_t size, int direction,
-				   struct dma_attrs *attrs)
+static dma_addr_t sn_dma_map_page(struct device *dev, struct page *page,
+				  unsigned long offset, size_t size,
+				  enum dma_data_direction dir,
+				  struct dma_attrs *attrs)
 {
+	void *cpu_addr = page_address(page) + offset;
 	dma_addr_t dma_addr;
 	unsigned long phys_addr;
 	struct pci_dev *pdev = to_pci_dev(dev);
@@ -201,7 +200,6 @@ dma_addr_t sn_dma_map_single_attrs(struct device *dev, void *cpu_addr,
 	}
 	return dma_addr;
 }
-EXPORT_SYMBOL(sn_dma_map_single_attrs);
 
 /**
  * sn_dma_unmap_single_attrs - unamp a DMA mapped page
@@ -215,21 +213,20 @@ EXPORT_SYMBOL(sn_dma_map_single_attrs);
  * by @dma_handle into the coherence domain.  On SN, we're always cache
  * coherent, so we just need to free any ATEs associated with this mapping.
  */
-void sn_dma_unmap_single_attrs(struct device *dev, dma_addr_t dma_addr,
-			       size_t size, int direction,
-			       struct dma_attrs *attrs)
+static void sn_dma_unmap_page(struct device *dev, dma_addr_t dma_addr,
+			      size_t size, enum dma_data_direction dir,
+			      struct dma_attrs *attrs)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
 	struct sn_pcibus_provider *provider = SN_PCIDEV_BUSPROVIDER(pdev);
 
 	BUG_ON(dev->bus != &pci_bus_type);
 
-	provider->dma_unmap(pdev, dma_addr, direction);
+	provider->dma_unmap(pdev, dma_addr, dir);
 }
-EXPORT_SYMBOL(sn_dma_unmap_single_attrs);
 
 /**
- * sn_dma_unmap_sg_attrs - unmap a DMA scatterlist
+ * sn_dma_unmap_sg - unmap a DMA scatterlist
  * @dev: device to unmap
  * @sg: scatterlist to unmap
  * @nhwentries: number of scatterlist entries
@@ -238,9 +235,9 @@ EXPORT_SYMBOL(sn_dma_unmap_single_attrs);
  *
  * Unmap a set of streaming mode DMA translations.
  */
-void sn_dma_unmap_sg_attrs(struct device *dev, struct scatterlist *sgl,
-			   int nhwentries, int direction,
-			   struct dma_attrs *attrs)
+static void sn_dma_unmap_sg(struct device *dev, struct scatterlist *sgl,
+			    int nhwentries, enum dma_data_direction dir,
+			    struct dma_attrs *attrs)
 {
 	int i;
 	struct pci_dev *pdev = to_pci_dev(dev);
@@ -250,15 +247,14 @@ void sn_dma_unmap_sg_attrs(struct device *dev, struct scatterlist *sgl,
 	BUG_ON(dev->bus != &pci_bus_type);
 
 	for_each_sg(sgl, sg, nhwentries, i) {
-		provider->dma_unmap(pdev, sg->dma_address, direction);
+		provider->dma_unmap(pdev, sg->dma_address, dir);
 		sg->dma_address = (dma_addr_t) NULL;
 		sg->dma_length = 0;
 	}
 }
-EXPORT_SYMBOL(sn_dma_unmap_sg_attrs);
 
 /**
- * sn_dma_map_sg_attrs - map a scatterlist for DMA
+ * sn_dma_map_sg - map a scatterlist for DMA
  * @dev: device to map for
  * @sg: scatterlist to map
  * @nhwentries: number of entries
@@ -272,8 +268,9 @@ EXPORT_SYMBOL(sn_dma_unmap_sg_attrs);
  *
  * Maps each entry of @sg for DMA.
  */
-int sn_dma_map_sg_attrs(struct device *dev, struct scatterlist *sgl,
-			int nhwentries, int direction, struct dma_attrs *attrs)
+static int sn_dma_map_sg(struct device *dev, struct scatterlist *sgl,
+			 int nhwentries, enum dma_data_direction dir,
+			 struct dma_attrs *attrs)
 {
 	unsigned long phys_addr;
 	struct scatterlist *saved_sg = sgl, *sg;
@@ -310,8 +307,7 @@ int sn_dma_map_sg_attrs(struct device *dev, struct scatterlist *sgl,
 			 * Free any successfully allocated entries.
 			 */
 			if (i > 0)
-				sn_dma_unmap_sg_attrs(dev, saved_sg, i,
-						      direction, attrs);
+				sn_dma_unmap_sg(dev, saved_sg, i, dir, attrs);
 			return 0;
 		}
 
@@ -320,41 +316,36 @@ int sn_dma_map_sg_attrs(struct device *dev, struct scatterlist *sgl,
 
 	return nhwentries;
 }
-EXPORT_SYMBOL(sn_dma_map_sg_attrs);
 
-void sn_dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle,
-				size_t size, int direction)
+static void sn_dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle,
+				       size_t size, enum dma_data_direction dir)
 {
 	BUG_ON(dev->bus != &pci_bus_type);
 }
-EXPORT_SYMBOL(sn_dma_sync_single_for_cpu);
 
-void sn_dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle,
-				   size_t size, int direction)
+static void sn_dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle,
+					  size_t size,
+					  enum dma_data_direction dir)
 {
 	BUG_ON(dev->bus != &pci_bus_type);
 }
-EXPORT_SYMBOL(sn_dma_sync_single_for_device);
 
-void sn_dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg,
-			    int nelems, int direction)
+static void sn_dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg,
+				   int nelems, enum dma_data_direction dir)
 {
 	BUG_ON(dev->bus != &pci_bus_type);
 }
-EXPORT_SYMBOL(sn_dma_sync_sg_for_cpu);
 
-void sn_dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg,
-			       int nelems, int direction)
+static void sn_dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg,
+				      int nelems, enum dma_data_direction dir)
 {
 	BUG_ON(dev->bus != &pci_bus_type);
 }
-EXPORT_SYMBOL(sn_dma_sync_sg_for_device);
 
-int sn_dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
+static int sn_dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
 {
 	return 0;
 }
-EXPORT_SYMBOL(sn_dma_mapping_error);
 
 u64 sn_dma_get_required_mask(struct device *dev)
 {
@@ -471,3 +462,23 @@ int sn_pci_legacy_write(struct pci_bus *bus, u16 port, u32 val, u8 size)
  out:
 	return ret;
 }
+
+static struct dma_map_ops sn_dma_ops = {
+	.alloc_coherent		= sn_dma_alloc_coherent,
+	.free_coherent		= sn_dma_free_coherent,
+	.map_page		= sn_dma_map_page,
+	.unmap_page		= sn_dma_unmap_page,
+	.map_sg			= sn_dma_map_sg,
+	.unmap_sg		= sn_dma_unmap_sg,
+	.sync_single_for_cpu 	= sn_dma_sync_single_for_cpu,
+	.sync_sg_for_cpu	= sn_dma_sync_sg_for_cpu,
+	.sync_single_for_device = sn_dma_sync_single_for_device,
+	.sync_sg_for_device	= sn_dma_sync_sg_for_device,
+	.mapping_error		= sn_dma_mapping_error,
+	.dma_supported		= sn_dma_supported,
+};
+
+void sn_dma_init(void)
+{
+	dma_ops = &sn_dma_ops;
+}