Merge linux-2.6 into linux-acpi-2.6 test

30 files changed, 2040 insertions, 237 deletions

diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index addf07393ef..e1c9ea03f31 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -357,6 +357,12 @@ source "kernel/power/Kconfig"
 
 source "drivers/acpi/Kconfig"
 
+if PM
+
+source "arch/ia64/kernel/cpufreq/Kconfig"
+
+endif
+
 endmenu
 
 if !IA64_HP_SIM
diff --git a/arch/ia64/hp/sim/boot/fw-emu.c b/arch/ia64/hp/sim/boot/fw-emu.c
index 5c46928e3dc..30fdfb1d0a5 100644
--- a/arch/ia64/hp/sim/boot/fw-emu.c
+++ b/arch/ia64/hp/sim/boot/fw-emu.c
@@ -237,17 +237,6 @@ sal_emulator (long index, unsigned long in1, unsigned long in2,
 	return ((struct sal_ret_values) {status, r9, r10, r11});
 }
 
-
-/*
- * This is here to work around a bug in egcs-1.1.1b that causes the
- * compiler to crash (seems like a bug in the new alias analysis code.
- */
-void *
-id (long addr)
-{
-	return (void *) addr;
-}
-
 struct ia64_boot_param *
 sys_fw_init (const char *args, int arglen)
 {
diff --git a/arch/ia64/ia32/ia32_signal.c b/arch/ia64/ia32/ia32_signal.c
index ebb89be2aa2..aa891c9bc9b 100644
--- a/arch/ia64/ia32/ia32_signal.c
+++ b/arch/ia64/ia32/ia32_signal.c
@@ -29,7 +29,6 @@
 #include <asm/uaccess.h>
 #include <asm/rse.h>
 #include <asm/sigcontext.h>
-#include <asm/segment.h>
 
 #include "ia32priv.h"
 
diff --git a/arch/ia64/kernel/Makefile b/arch/ia64/kernel/Makefile
index e1fb68ddec2..b242594be55 100644
--- a/arch/ia64/kernel/Makefile
+++ b/arch/ia64/kernel/Makefile
@@ -20,6 +20,7 @@ obj-$(CONFIG_SMP)		+= smp.o smpboot.o domain.o
 obj-$(CONFIG_NUMA)		+= numa.o
 obj-$(CONFIG_PERFMON)		+= perfmon_default_smpl.o
 obj-$(CONFIG_IA64_CYCLONE)	+= cyclone.o
+obj-$(CONFIG_CPU_FREQ)		+= cpufreq/
 obj-$(CONFIG_IA64_MCA_RECOVERY)	+= mca_recovery.o
 obj-$(CONFIG_KPROBES)		+= kprobes.o jprobes.o
 obj-$(CONFIG_IA64_UNCACHED_ALLOCATOR)	+= uncached.o
diff --git a/arch/ia64/kernel/cpufreq/Kconfig b/arch/ia64/kernel/cpufreq/Kconfig
new file mode 100644
index 00000000000..2d9d5279b98
--- /dev/null
+++ b/arch/ia64/kernel/cpufreq/Kconfig
@@ -0,0 +1,29 @@
+
+#
+# CPU Frequency scaling
+#
+
+menu "CPU Frequency scaling"
+
+source "drivers/cpufreq/Kconfig"
+
+if CPU_FREQ
+
+comment "CPUFreq processor drivers"
+
+config IA64_ACPI_CPUFREQ
+	tristate "ACPI Processor P-States driver"
+	select CPU_FREQ_TABLE
+	depends on ACPI_PROCESSOR
+	help
+	This driver adds a CPUFreq driver which utilizes the ACPI
+	Processor Performance States.
+
+	For details, take a look at <file:Documentation/cpu-freq/>.
+
+	If in doubt, say N.
+
+endif   # CPU_FREQ
+
+endmenu
+
diff --git a/arch/ia64/kernel/cpufreq/Makefile b/arch/ia64/kernel/cpufreq/Makefile
new file mode 100644
index 00000000000..f748d34c02f
--- /dev/null
+++ b/arch/ia64/kernel/cpufreq/Makefile
@@ -0,0 +1 @@
+obj-$(CONFIG_IA64_ACPI_CPUFREQ)		+= acpi-cpufreq.o
diff --git a/arch/ia64/kernel/cpufreq/acpi-cpufreq.c b/arch/ia64/kernel/cpufreq/acpi-cpufreq.c
new file mode 100644
index 00000000000..da4d5cf80a4
--- /dev/null
+++ b/arch/ia64/kernel/cpufreq/acpi-cpufreq.c
@@ -0,0 +1,499 @@
+/*
+ * arch/ia64/kernel/cpufreq/acpi-cpufreq.c
+ * This file provides the ACPI based P-state support. This
+ * module works with generic cpufreq infrastructure. Most of
+ * the code is based on i386 version
+ * (arch/i386/kernel/cpu/cpufreq/acpi-cpufreq.c)
+ *
+ * Copyright (C) 2005 Intel Corp
+ *      Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
+ */
+
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/cpufreq.h>
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+#include <asm/io.h>
+#include <asm/uaccess.h>
+#include <asm/pal.h>
+
+#include <linux/acpi.h>
+#include <acpi/processor.h>
+
+#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, "acpi-cpufreq", msg)
+
+MODULE_AUTHOR("Venkatesh Pallipadi");
+MODULE_DESCRIPTION("ACPI Processor P-States Driver");
+MODULE_LICENSE("GPL");
+
+
+struct cpufreq_acpi_io {
+	struct acpi_processor_performance	acpi_data;
+	struct cpufreq_frequency_table		*freq_table;
+	unsigned int				resume;
+};
+
+static struct cpufreq_acpi_io	*acpi_io_data[NR_CPUS];
+
+static struct cpufreq_driver acpi_cpufreq_driver;
+
+
+static int
+processor_set_pstate (
+	u32	value)
+{
+	s64 retval;
+
+	dprintk("processor_set_pstate\n");
+
+	retval = ia64_pal_set_pstate((u64)value);
+
+	if (retval) {
+		dprintk("Failed to set freq to 0x%x, with error 0x%x\n",
+		        value, retval);
+		return -ENODEV;
+	}
+	return (int)retval;
+}
+
+
+static int
+processor_get_pstate (
+	u32	*value)
+{
+	u64	pstate_index = 0;
+	s64 	retval;
+
+	dprintk("processor_get_pstate\n");
+
+	retval = ia64_pal_get_pstate(&pstate_index);
+	*value = (u32) pstate_index;
+
+	if (retval)
+		dprintk("Failed to get current freq with "
+		        "error 0x%x, idx 0x%x\n", retval, *value);
+
+	return (int)retval;
+}
+
+
+/* To be used only after data->acpi_data is initialized */
+static unsigned
+extract_clock (
+	struct cpufreq_acpi_io *data,
+	unsigned value,
+	unsigned int cpu)
+{
+	unsigned long i;
+
+	dprintk("extract_clock\n");
+
+	for (i = 0; i < data->acpi_data.state_count; i++) {
+		if (value >= data->acpi_data.states[i].control)
+			return data->acpi_data.states[i].core_frequency;
+	}
+	return data->acpi_data.states[i-1].core_frequency;
+}
+
+
+static unsigned int
+processor_get_freq (
+	struct cpufreq_acpi_io	*data,
+	unsigned int		cpu)
+{
+	int			ret = 0;
+	u32			value = 0;
+	cpumask_t		saved_mask;
+	unsigned long 		clock_freq;
+
+	dprintk("processor_get_freq\n");
+
+	saved_mask = current->cpus_allowed;
+	set_cpus_allowed(current, cpumask_of_cpu(cpu));
+	if (smp_processor_id() != cpu) {
+		ret = -EAGAIN;
+		goto migrate_end;
+	}
+
+	/*
+	 * processor_get_pstate gets the average frequency since the
+	 * last get. So, do two PAL_get_freq()...
+	 */
+	ret = processor_get_pstate(&value);
+	ret = processor_get_pstate(&value);
+
+	if (ret) {
+		set_cpus_allowed(current, saved_mask);
+		printk(KERN_WARNING "get performance failed with error %d\n",
+		       ret);
+		ret = -EAGAIN;
+		goto migrate_end;
+	}
+	clock_freq = extract_clock(data, value, cpu);
+	ret = (clock_freq*1000);
+
+migrate_end:
+	set_cpus_allowed(current, saved_mask);
+	return ret;
+}
+
+
+static int
+processor_set_freq (
+	struct cpufreq_acpi_io	*data,
+	unsigned int		cpu,
+	int			state)
+{
+	int			ret = 0;
+	u32			value = 0;
+	struct cpufreq_freqs    cpufreq_freqs;
+	cpumask_t		saved_mask;
+	int			retval;
+
+	dprintk("processor_set_freq\n");
+
+	saved_mask = current->cpus_allowed;
+	set_cpus_allowed(current, cpumask_of_cpu(cpu));
+	if (smp_processor_id() != cpu) {
+		retval = -EAGAIN;
+		goto migrate_end;
+	}
+
+	if (state == data->acpi_data.state) {
+		if (unlikely(data->resume)) {
+			dprintk("Called after resume, resetting to P%d\n", state);
+			data->resume = 0;
+		} else {
+			dprintk("Already at target state (P%d)\n", state);
+			retval = 0;
+			goto migrate_end;
+		}
+	}
+
+	dprintk("Transitioning from P%d to P%d\n",
+		data->acpi_data.state, state);
+
+	/* cpufreq frequency struct */
+	cpufreq_freqs.cpu = cpu;
+	cpufreq_freqs.old = data->freq_table[data->acpi_data.state].frequency;
+	cpufreq_freqs.new = data->freq_table[state].frequency;
+
+	/* notify cpufreq */
+	cpufreq_notify_transition(&cpufreq_freqs, CPUFREQ_PRECHANGE);
+
+	/*
+	 * First we write the target state's 'control' value to the
+	 * control_register.
+	 */
+
+	value = (u32) data->acpi_data.states[state].control;
+
+	dprintk("Transitioning to state: 0x%08x\n", value);
+
+	ret = processor_set_pstate(value);
+	if (ret) {
+		unsigned int tmp = cpufreq_freqs.new;
+		cpufreq_notify_transition(&cpufreq_freqs, CPUFREQ_POSTCHANGE);
+		cpufreq_freqs.new = cpufreq_freqs.old;
+		cpufreq_freqs.old = tmp;
+		cpufreq_notify_transition(&cpufreq_freqs, CPUFREQ_PRECHANGE);
+		cpufreq_notify_transition(&cpufreq_freqs, CPUFREQ_POSTCHANGE);
+		printk(KERN_WARNING "Transition failed with error %d\n", ret);
+		retval = -ENODEV;
+		goto migrate_end;
+	}
+
+	cpufreq_notify_transition(&cpufreq_freqs, CPUFREQ_POSTCHANGE);
+
+	data->acpi_data.state = state;
+
+	retval = 0;
+
+migrate_end:
+	set_cpus_allowed(current, saved_mask);
+	return (retval);
+}
+
+
+static unsigned int
+acpi_cpufreq_get (
+	unsigned int		cpu)
+{
+	struct cpufreq_acpi_io *data = acpi_io_data[cpu];
+
+	dprintk("acpi_cpufreq_get\n");
+
+	return processor_get_freq(data, cpu);
+}
+
+
+static int
+acpi_cpufreq_target (
+	struct cpufreq_policy   *policy,
+	unsigned int target_freq,
+	unsigned int relation)
+{
+	struct cpufreq_acpi_io *data = acpi_io_data[policy->cpu];
+	unsigned int next_state = 0;
+	unsigned int result = 0;
+
+	dprintk("acpi_cpufreq_setpolicy\n");
+
+	result = cpufreq_frequency_table_target(policy,
+			data->freq_table, target_freq, relation, &next_state);
+	if (result)
+		return (result);
+
+	result = processor_set_freq(data, policy->cpu, next_state);
+
+	return (result);
+}
+
+
+static int
+acpi_cpufreq_verify (
+	struct cpufreq_policy   *policy)
+{
+	unsigned int result = 0;
+	struct cpufreq_acpi_io *data = acpi_io_data[policy->cpu];
+
+	dprintk("acpi_cpufreq_verify\n");
+
+	result = cpufreq_frequency_table_verify(policy,
+			data->freq_table);
+
+	return (result);
+}
+
+
+/*
+ * processor_init_pdc - let BIOS know about the SMP capabilities
+ * of this driver
+ * @perf: processor-specific acpi_io_data struct
+ * @cpu: CPU being initialized
+ *
+ * To avoid issues with legacy OSes, some BIOSes require to be informed of
+ * the SMP capabilities of OS P-state driver. Here we set the bits in _PDC
+ * accordingly. Actual call to _PDC is done in driver/acpi/processor.c
+ */
+static void
+processor_init_pdc (
+		struct acpi_processor_performance *perf,
+		unsigned int cpu,
+		struct acpi_object_list *obj_list
+		)
+{
+	union acpi_object *obj;
+	u32 *buf;
+
+	dprintk("processor_init_pdc\n");
+
+	perf->pdc = NULL;
+	/* Initialize pdc. It will be used later. */
+	if (!obj_list)
+		return;
+
+	if (!(obj_list->count && obj_list->pointer))
+		return;
+
+	obj = obj_list->pointer;
+	if ((obj->buffer.length == 12) && obj->buffer.pointer) {
+		buf = (u32 *)obj->buffer.pointer;
+       		buf[0] = ACPI_PDC_REVISION_ID;
+       		buf[1] = 1;
+       		buf[2] = ACPI_PDC_EST_CAPABILITY_SMP;
+		perf->pdc = obj_list;
+	}
+	return;
+}
+
+
+static int
+acpi_cpufreq_cpu_init (
+	struct cpufreq_policy   *policy)
+{
+	unsigned int		i;
+	unsigned int		cpu = policy->cpu;
+	struct cpufreq_acpi_io	*data;
+	unsigned int		result = 0;
+
+	union acpi_object		arg0 = {ACPI_TYPE_BUFFER};
+	u32				arg0_buf[3];
+	struct acpi_object_list 	arg_list = {1, &arg0};
+
+	dprintk("acpi_cpufreq_cpu_init\n");
+	/* setup arg_list for _PDC settings */
+        arg0.buffer.length = 12;
+        arg0.buffer.pointer = (u8 *) arg0_buf;
+
+	data = kmalloc(sizeof(struct cpufreq_acpi_io), GFP_KERNEL);
+	if (!data)
+		return (-ENOMEM);
+
+	memset(data, 0, sizeof(struct cpufreq_acpi_io));
+
+	acpi_io_data[cpu] = data;
+
+	processor_init_pdc(&data->acpi_data, cpu, &arg_list);
+	result = acpi_processor_register_performance(&data->acpi_data, cpu);
+	data->acpi_data.pdc = NULL;
+
+	if (result)
+		goto err_free;
+
+	/* capability check */
+	if (data->acpi_data.state_count <= 1) {
+		dprintk("No P-States\n");
+		result = -ENODEV;
+		goto err_unreg;
+	}
+
+	if ((data->acpi_data.control_register.space_id !=
+					ACPI_ADR_SPACE_FIXED_HARDWARE) ||
+	    (data->acpi_data.status_register.space_id !=
+					ACPI_ADR_SPACE_FIXED_HARDWARE)) {
+		dprintk("Unsupported address space [%d, %d]\n",
+			(u32) (data->acpi_data.control_register.space_id),
+			(u32) (data->acpi_data.status_register.space_id));
+		result = -ENODEV;
+		goto err_unreg;
+	}
+
+	/* alloc freq_table */
+	data->freq_table = kmalloc(sizeof(struct cpufreq_frequency_table) *
+	                           (data->acpi_data.state_count + 1),
+	                           GFP_KERNEL);
+	if (!data->freq_table) {
+		result = -ENOMEM;
+		goto err_unreg;
+	}
+
+	/* detect transition latency */
+	policy->cpuinfo.transition_latency = 0;
+	for (i=0; i<data->acpi_data.state_count; i++) {
+		if ((data->acpi_data.states[i].transition_latency * 1000) >
+		    policy->cpuinfo.transition_latency) {
+			policy->cpuinfo.transition_latency =
+			    data->acpi_data.states[i].transition_latency * 1000;
+		}
+	}
+	policy->governor = CPUFREQ_DEFAULT_GOVERNOR;
+
+	policy->cur = processor_get_freq(data, policy->cpu);
+
+	/* table init */
+	for (i = 0; i <= data->acpi_data.state_count; i++)
+	{
+		data->freq_table[i].index = i;
+		if (i < data->acpi_data.state_count) {
+			data->freq_table[i].frequency =
+			      data->acpi_data.states[i].core_frequency * 1000;
+		} else {
+			data->freq_table[i].frequency = CPUFREQ_TABLE_END;
+		}
+	}
+
+	result = cpufreq_frequency_table_cpuinfo(policy, data->freq_table);
+	if (result) {
+		goto err_freqfree;
+	}
+
+	/* notify BIOS that we exist */
+	acpi_processor_notify_smm(THIS_MODULE);
+
+	printk(KERN_INFO "acpi-cpufreq: CPU%u - ACPI performance management "
+	       "activated.\n", cpu);
+
+	for (i = 0; i < data->acpi_data.state_count; i++)
+		dprintk("     %cP%d: %d MHz, %d mW, %d uS, %d uS, 0x%x 0x%x\n",
+			(i == data->acpi_data.state?'*':' '), i,
+			(u32) data->acpi_data.states[i].core_frequency,
+			(u32) data->acpi_data.states[i].power,
+			(u32) data->acpi_data.states[i].transition_latency,
+			(u32) data->acpi_data.states[i].bus_master_latency,
+			(u32) data->acpi_data.states[i].status,
+			(u32) data->acpi_data.states[i].control);
+
+	cpufreq_frequency_table_get_attr(data->freq_table, policy->cpu);
+
+	/* the first call to ->target() should result in us actually
+	 * writing something to the appropriate registers. */
+	data->resume = 1;
+
+	return (result);
+
+ err_freqfree:
+	kfree(data->freq_table);
+ err_unreg:
+	acpi_processor_unregister_performance(&data->acpi_data, cpu);
+ err_free:
+	kfree(data);
+	acpi_io_data[cpu] = NULL;
+
+	return (result);
+}
+
+
+static int
+acpi_cpufreq_cpu_exit (
+	struct cpufreq_policy   *policy)
+{
+	struct cpufreq_acpi_io *data = acpi_io_data[policy->cpu];
+
+	dprintk("acpi_cpufreq_cpu_exit\n");
+
+	if (data) {
+		cpufreq_frequency_table_put_attr(policy->cpu);
+		acpi_io_data[policy->cpu] = NULL;
+		acpi_processor_unregister_performance(&data->acpi_data,
+		                                      policy->cpu);
+		kfree(data);
+	}
+
+	return (0);
+}
+
+
+static struct freq_attr* acpi_cpufreq_attr[] = {
+	&cpufreq_freq_attr_scaling_available_freqs,
+	NULL,
+};
+
+
+static struct cpufreq_driver acpi_cpufreq_driver = {
+	.verify 	= acpi_cpufreq_verify,
+	.target 	= acpi_cpufreq_target,
+	.get 		= acpi_cpufreq_get,
+	.init		= acpi_cpufreq_cpu_init,
+	.exit		= acpi_cpufreq_cpu_exit,
+	.name		= "acpi-cpufreq",
+	.owner		= THIS_MODULE,
+	.attr           = acpi_cpufreq_attr,
+};
+
+
+static int __init
+acpi_cpufreq_init (void)
+{
+	dprintk("acpi_cpufreq_init\n");
+
+ 	return cpufreq_register_driver(&acpi_cpufreq_driver);
+}
+
+
+static void __exit
+acpi_cpufreq_exit (void)
+{
+	dprintk("acpi_cpufreq_exit\n");
+
+	cpufreq_unregister_driver(&acpi_cpufreq_driver);
+	return;
+}
+
+
+late_initcall(acpi_cpufreq_init);
+module_exit(acpi_cpufreq_exit);
+
diff --git a/arch/ia64/kernel/sys_ia64.c b/arch/ia64/kernel/sys_ia64.c
index 770fab37928..f2dbcd1db0d 100644
--- a/arch/ia64/kernel/sys_ia64.c
+++ b/arch/ia64/kernel/sys_ia64.c
@@ -35,7 +35,7 @@ arch_get_unmapped_area (struct file *filp, unsigned long addr, unsigned long len
 		return -ENOMEM;
 
 #ifdef CONFIG_HUGETLB_PAGE
-	if (REGION_NUMBER(addr) == REGION_HPAGE)
+	if (REGION_NUMBER(addr) == RGN_HPAGE)
 		addr = 0;
 #endif
 	if (!addr)
diff --git a/arch/ia64/kernel/uncached.c b/arch/ia64/kernel/uncached.c
index 490dfc9ab47..4e9d06c48a8 100644
--- a/arch/ia64/kernel/uncached.c
+++ b/arch/ia64/kernel/uncached.c
@@ -184,7 +184,7 @@ uncached_free_page(unsigned long maddr)
 {
 	int node;
 
-	node = nasid_to_cnodeid(NASID_GET(maddr));
+	node = paddr_to_nid(maddr - __IA64_UNCACHED_OFFSET);
 
 	dprintk(KERN_DEBUG "uncached_free_page(%lx) on node %i\n", maddr, node);
 
@@ -217,7 +217,7 @@ uncached_build_memmap(unsigned long start, unsigned long end, void *arg)
 
 	memset((char *)vstart, 0, length);
 
-	node = nasid_to_cnodeid(NASID_GET(start));
+	node = paddr_to_nid(start);
 
 	for (; vstart < vend ; vstart += PAGE_SIZE) {
 		dprintk(KERN_INFO "sticking %lx into the pool!\n", vstart);
diff --git a/arch/ia64/lib/Makefile b/arch/ia64/lib/Makefile
index 1902c3c2ef9..799407e7726 100644
--- a/arch/ia64/lib/Makefile
+++ b/arch/ia64/lib/Makefile
@@ -6,7 +6,7 @@ obj-y := io.o
 
 lib-y := __divsi3.o __udivsi3.o __modsi3.o __umodsi3.o			\
 	__divdi3.o __udivdi3.o __moddi3.o __umoddi3.o			\
-	bitop.o checksum.o clear_page.o csum_partial_copy.o copy_page.o	\
+	bitop.o checksum.o clear_page.o csum_partial_copy.o		\
 	clear_user.o strncpy_from_user.o strlen_user.o strnlen_user.o	\
 	flush.o ip_fast_csum.o do_csum.o				\
 	memset.o strlen.o swiotlb.o
diff --git a/arch/ia64/lib/swiotlb.c b/arch/ia64/lib/swiotlb.c
index ab7b3ad99a7..dbc0b3e449c 100644
--- a/arch/ia64/lib/swiotlb.c
+++ b/arch/ia64/lib/swiotlb.c
@@ -93,8 +93,7 @@ static int __init
 setup_io_tlb_npages(char *str)
 {
 	if (isdigit(*str)) {
-		io_tlb_nslabs = simple_strtoul(str, &str, 0) <<
-			(PAGE_SHIFT - IO_TLB_SHIFT);
+		io_tlb_nslabs = simple_strtoul(str, &str, 0);
 		/* avoid tail segment of size < IO_TLB_SEGSIZE */
 		io_tlb_nslabs = ALIGN(io_tlb_nslabs, IO_TLB_SEGSIZE);
 	}
@@ -117,7 +116,7 @@ swiotlb_init_with_default_size (size_t default_size)
 	unsigned long i;
 
 	if (!io_tlb_nslabs) {
-		io_tlb_nslabs = (default_size >> PAGE_SHIFT);
+		io_tlb_nslabs = (default_size >> IO_TLB_SHIFT);
 		io_tlb_nslabs = ALIGN(io_tlb_nslabs, IO_TLB_SEGSIZE);
 	}
 
diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c
index e0a776a3044..2d13889d0a9 100644
--- a/arch/ia64/mm/hugetlbpage.c
+++ b/arch/ia64/mm/hugetlbpage.c
@@ -76,7 +76,7 @@ int is_aligned_hugepage_range(unsigned long addr, unsigned long len)
 		return -EINVAL;
 	if (addr & ~HPAGE_MASK)
 		return -EINVAL;
-	if (REGION_NUMBER(addr) != REGION_HPAGE)
+	if (REGION_NUMBER(addr) != RGN_HPAGE)
 		return -EINVAL;
 
 	return 0;
@@ -87,7 +87,7 @@ struct page *follow_huge_addr(struct mm_struct *mm, unsigned long addr, int writ
 	struct page *page;
 	pte_t *ptep;
 
-	if (REGION_NUMBER(addr) != REGION_HPAGE)
+	if (REGION_NUMBER(addr) != RGN_HPAGE)
 		return ERR_PTR(-EINVAL);
 
 	ptep = huge_pte_offset(mm, addr);
@@ -142,8 +142,8 @@ unsigned long hugetlb_get_unmapped_area(struct file *file, unsigned long addr, u
 		return -ENOMEM;
 	if (len & ~HPAGE_MASK)
 		return -EINVAL;
-	/* This code assumes that REGION_HPAGE != 0. */
-	if ((REGION_NUMBER(addr) != REGION_HPAGE) || (addr & (HPAGE_SIZE - 1)))
+	/* This code assumes that RGN_HPAGE != 0. */
+	if ((REGION_NUMBER(addr) != RGN_HPAGE) || (addr & (HPAGE_SIZE - 1)))
 		addr = HPAGE_REGION_BASE;
 	else
 		addr = ALIGN(addr, HPAGE_SIZE);
diff --git a/arch/ia64/pci/pci.c b/arch/ia64/pci/pci.c
index f9472c50ab4..9977c122e9f 100644
--- a/arch/ia64/pci/pci.c
+++ b/arch/ia64/pci/pci.c
@@ -24,7 +24,6 @@
 
 #include <asm/machvec.h>
 #include <asm/page.h>
-#include <asm/segment.h>
 #include <asm/system.h>
 #include <asm/io.h>
 #include <asm/sal.h>
diff --git a/arch/ia64/sn/include/tio.h b/arch/ia64/sn/include/tio.h
index 0139124dd54..6b2e7b75eb1 100644
--- a/arch/ia64/sn/include/tio.h
+++ b/arch/ia64/sn/include/tio.h
@@ -3,7 +3,7 @@
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  *
- * Copyright (C) 2000-2004 Silicon Graphics, Inc. All rights reserved.
+ * Copyright (C) 2000-2005 Silicon Graphics, Inc. All rights reserved.
  */
 
 #ifndef _ASM_IA64_SN_TIO_H
@@ -26,6 +26,10 @@
 #define TIO_ITTE_VALID_MASK	0x1
 #define TIO_ITTE_VALID_SHIFT	16
 
+#define TIO_ITTE_WIDGET(itte) \
+	(((itte) >> TIO_ITTE_WIDGET_SHIFT) & TIO_ITTE_WIDGET_MASK)
+#define TIO_ITTE_VALID(itte) \
+	(((itte) >> TIO_ITTE_VALID_SHIFT) & TIO_ITTE_VALID_MASK)
 
 #define TIO_ITTE_PUT(nasid, bigwin, widget, addr, valid) \
         REMOTE_HUB_S((nasid), TIO_ITTE(bigwin), \
diff --git a/arch/ia64/sn/include/xtalk/hubdev.h b/arch/ia64/sn/include/xtalk/hubdev.h
index 580a1c0403a..71c2b271b4c 100644
--- a/arch/ia64/sn/include/xtalk/hubdev.h
+++ b/arch/ia64/sn/include/xtalk/hubdev.h
@@ -3,7 +3,7 @@
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  *
- * Copyright (C) 1992 - 1997, 2000-2004 Silicon Graphics, Inc. All rights reserved.
+ * Copyright (C) 1992 - 1997, 2000-2005 Silicon Graphics, Inc. All rights reserved.
  */
 #ifndef _ASM_IA64_SN_XTALK_HUBDEV_H
 #define _ASM_IA64_SN_XTALK_HUBDEV_H
@@ -16,6 +16,9 @@
 #define IIO_ITTE_WIDGET_MASK    ((1<<IIO_ITTE_WIDGET_BITS)-1)
 #define IIO_ITTE_WIDGET_SHIFT   8
 
+#define IIO_ITTE_WIDGET(itte)	\
+	(((itte) >> IIO_ITTE_WIDGET_SHIFT) & IIO_ITTE_WIDGET_MASK)
+
 /*
  * Use the top big window as a surrogate for the first small window
  */
@@ -34,7 +37,8 @@ struct sn_flush_device_list {
 	unsigned long sfdl_force_int_addr;
 	unsigned long sfdl_flush_value;
 	volatile unsigned long *sfdl_flush_addr;
-	uint64_t sfdl_persistent_busnum;
+	uint32_t sfdl_persistent_busnum;
+	uint32_t sfdl_persistent_segment;
 	struct pcibus_info *sfdl_pcibus_info;
 	spinlock_t sfdl_flush_lock;
 };
@@ -58,7 +62,8 @@ struct hubdev_info {
 
 	void				*hdi_nodepda;
 	void				*hdi_node_vertex;
-	void				*hdi_xtalk_vertex;
+	uint32_t			max_segment_number;
+	uint32_t			max_pcibus_number;
 };
 
 extern void hubdev_init_node(nodepda_t *, cnodeid_t);
diff --git a/arch/ia64/sn/kernel/bte.c b/arch/ia64/sn/kernel/bte.c
index 647deae9bfc..45854c637e9 100644
--- a/arch/ia64/sn/kernel/bte.c
+++ b/arch/ia64/sn/kernel/bte.c
@@ -29,16 +29,30 @@
 
 /* two interfaces on two btes */
 #define MAX_INTERFACES_TO_TRY		4
+#define MAX_NODES_TO_TRY		2
 
 static struct bteinfo_s *bte_if_on_node(nasid_t nasid, int interface)
 {
 	nodepda_t *tmp_nodepda;
 
+	if (nasid_to_cnodeid(nasid) == -1)
+		return (struct bteinfo_s *)NULL;;
+
 	tmp_nodepda = NODEPDA(nasid_to_cnodeid(nasid));
 	return &tmp_nodepda->bte_if[interface];
 
 }
 
+static inline void bte_start_transfer(struct bteinfo_s *bte, u64 len, u64 mode)
+{
+	if (is_shub2()) {
+		BTE_CTRL_STORE(bte, (IBLS_BUSY | ((len) | (mode) << 24)));
+	} else {
+		BTE_LNSTAT_STORE(bte, len);
+		BTE_CTRL_STORE(bte, mode);
+	}
+}
+
 /************************************************************************
  * Block Transfer Engine copy related functions.
  *
@@ -67,13 +81,15 @@ bte_result_t bte_copy(u64 src, u64 dest, u64 len, u64 mode, void *notification)
 {
 	u64 transfer_size;
 	u64 transfer_stat;
+	u64 notif_phys_addr;
 	struct bteinfo_s *bte;
 	bte_result_t bte_status;
 	unsigned long irq_flags;
 	unsigned long itc_end = 0;
-	struct bteinfo_s *btes_to_try[MAX_INTERFACES_TO_TRY];
-	int bte_if_index;
-	int bte_pri, bte_sec;
+	int nasid_to_try[MAX_NODES_TO_TRY];
+	int my_nasid = get_nasid();
+	int bte_if_index, nasid_index;
+	int bte_first, btes_per_node = BTES_PER_NODE;
 
 	BTE_PRINTK(("bte_copy(0x%lx, 0x%lx, 0x%lx, 0x%lx, 0x%p)\n",
 		    src, dest, len, mode, notification));
@@ -86,36 +102,26 @@ bte_result_t bte_copy(u64 src, u64 dest, u64 len, u64 mode, void *notification)
 		 (src & L1_CACHE_MASK) || (dest & L1_CACHE_MASK));
 	BUG_ON(!(len < ((BTE_LEN_MASK + 1) << L1_CACHE_SHIFT)));
 
-	/* CPU 0 (per node) tries bte0 first, CPU 1 try bte1 first */
-	if (cpuid_to_subnode(smp_processor_id()) == 0) {
-		bte_pri = 0;
-		bte_sec = 1;
-	} else {
-		bte_pri = 1;
-		bte_sec = 0;
-	}
+	/*
+	 * Start with interface corresponding to cpu number
+	 */
+	bte_first = raw_smp_processor_id() % btes_per_node;
 
 	if (mode & BTE_USE_DEST) {
 		/* try remote then local */
-		btes_to_try[0] = bte_if_on_node(NASID_GET(dest), bte_pri);
-		btes_to_try[1] = bte_if_on_node(NASID_GET(dest), bte_sec);
+		nasid_to_try[0] = NASID_GET(dest);
 		if (mode & BTE_USE_ANY) {
-			btes_to_try[2] = bte_if_on_node(get_nasid(), bte_pri);
-			btes_to_try[3] = bte_if_on_node(get_nasid(), bte_sec);
+			nasid_to_try[1] = my_nasid;
 		} else {
-			btes_to_try[2] = NULL;