Merge branch 'master'

18 files changed, 308 insertions, 58 deletions

diff --git a/arch/x86_64/kernel/acpi/Makefile b/arch/x86_64/kernel/acpi/Makefile
index 7da9ace890b..4fe97071f29 100644
--- a/arch/x86_64/kernel/acpi/Makefile
+++ b/arch/x86_64/kernel/acpi/Makefile
@@ -1,3 +1,8 @@
 obj-y			:= boot.o
 boot-y			:= ../../../i386/kernel/acpi/boot.o
 obj-$(CONFIG_ACPI_SLEEP)	+= sleep.o wakeup.o
+
+ifneq ($(CONFIG_ACPI_PROCESSOR),)
+obj-y			+= processor.o
+endif
+
diff --git a/arch/x86_64/kernel/acpi/processor.c b/arch/x86_64/kernel/acpi/processor.c
new file mode 100644
index 00000000000..3bdc2baa5bb
--- /dev/null
+++ b/arch/x86_64/kernel/acpi/processor.c
@@ -0,0 +1,72 @@
+/*
+ * arch/x86_64/kernel/acpi/processor.c
+ *
+ * Copyright (C) 2005 Intel Corporation
+ * 	Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
+ * 	- Added _PDC for platforms with Intel CPUs
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/acpi.h>
+
+#include <acpi/processor.h>
+#include <asm/acpi.h>
+
+static void init_intel_pdc(struct acpi_processor *pr, struct cpuinfo_x86 *c)
+{
+	struct acpi_object_list *obj_list;
+	union acpi_object *obj;
+	u32 *buf;
+
+	/* allocate and initialize pdc. It will be used later. */
+	obj_list = kmalloc(sizeof(struct acpi_object_list), GFP_KERNEL);
+	if (!obj_list) {
+		printk(KERN_ERR "Memory allocation error\n");
+		return;
+	}
+
+	obj = kmalloc(sizeof(union acpi_object), GFP_KERNEL);
+	if (!obj) {
+		printk(KERN_ERR "Memory allocation error\n");
+		kfree(obj_list);
+		return;
+	}
+
+	buf = kmalloc(12, GFP_KERNEL);
+	if (!buf) {
+		printk(KERN_ERR "Memory allocation error\n");
+		kfree(obj);
+		kfree(obj_list);
+		return;
+	}
+
+	buf[0] = ACPI_PDC_REVISION_ID;
+	buf[1] = 1;
+	buf[2] = ACPI_PDC_EST_CAPABILITY_SMP;
+
+	obj->type = ACPI_TYPE_BUFFER;
+	obj->buffer.length = 12;
+	obj->buffer.pointer = (u8 *) buf;
+	obj_list->count = 1;
+	obj_list->pointer = obj;
+	pr->pdc = obj_list;
+
+	return;
+}
+
+/* Initialize _PDC data based on the CPU vendor */
+void arch_acpi_processor_init_pdc(struct acpi_processor *pr)
+{
+	unsigned int cpu = pr->id;
+	struct cpuinfo_x86 *c = cpu_data + cpu;
+
+	pr->pdc = NULL;
+	if (c->x86_vendor == X86_VENDOR_INTEL && cpu_has(c, X86_FEATURE_EST))
+		init_intel_pdc(pr, c);
+
+	return;
+}
+
+EXPORT_SYMBOL(arch_acpi_processor_init_pdc);
diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index 5d3c5b07b8d..6147770b434 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -35,8 +35,12 @@
 #include <asm/mach_apic.h>
 #include <asm/nmi.h>
 #include <asm/idle.h>
+#include <asm/proto.h>
+#include <asm/timex.h>
 
 int apic_verbosity;
+int apic_runs_main_timer;
+int apic_calibrate_pmtmr __initdata;
 
 int disable_apic_timer __initdata;
 
@@ -68,6 +72,26 @@ int get_maxlvt(void)
 	return maxlvt;
 }
 
+/*
+ * 'what should we do if we get a hw irq event on an illegal vector'.
+ * each architecture has to answer this themselves.
+ */
+void ack_bad_irq(unsigned int irq)
+{
+	printk("unexpected IRQ trap at vector %02x\n", irq);
+	/*
+	 * Currently unexpected vectors happen only on SMP and APIC.
+	 * We _must_ ack these because every local APIC has only N
+	 * irq slots per priority level, and a 'hanging, unacked' IRQ
+	 * holds up an irq slot - in excessive cases (when multiple
+	 * unexpected vectors occur) that might lock up the APIC
+	 * completely.
+  	 * But don't ack when the APIC is disabled. -AK
+	 */
+	if (!disable_apic)
+		ack_APIC_irq();
+}
+
 void clear_local_APIC(void)
 {
 	int maxlvt;
@@ -702,9 +726,17 @@ static void setup_APIC_timer(unsigned int clocks)
 			c2 |= inb_p(0x40) << 8;
 		} while (c2 - c1 < 300);
 	}
-
 	__setup_APIC_LVTT(clocks);
-
+	/* Turn off PIT interrupt if we use APIC timer as main timer.
+	   Only works with the PM timer right now
+	   TBD fix it for HPET too. */
+	if (vxtime.mode == VXTIME_PMTMR &&
+		smp_processor_id() == boot_cpu_id &&
+		apic_runs_main_timer == 1 &&
+		!cpu_isset(boot_cpu_id, timer_interrupt_broadcast_ipi_mask)) {
+		stop_timer_interrupt();
+		apic_runs_main_timer++;
+	}
 	local_irq_restore(flags);
 }
 
@@ -735,14 +767,27 @@ static int __init calibrate_APIC_clock(void)
 	__setup_APIC_LVTT(1000000000);
 
 	apic_start = apic_read(APIC_TMCCT);
-	rdtscl(tsc_start);
-
-	do {
+#ifdef CONFIG_X86_PM_TIMER
+	if (apic_calibrate_pmtmr && pmtmr_ioport) {
+		pmtimer_wait(5000);  /* 5ms wait */
 		apic = apic_read(APIC_TMCCT);
-		rdtscl(tsc);
-	} while ((tsc - tsc_start) < TICK_COUNT && (apic - apic_start) < TICK_COUNT);
+		result = (apic_start - apic) * 1000L / 5;
+	} else
+#endif
+	{
+		rdtscl(tsc_start);
+
+		do {
+			apic = apic_read(APIC_TMCCT);
+			rdtscl(tsc);
+		} while ((tsc - tsc_start) < TICK_COUNT &&
+				(apic - apic_start) < TICK_COUNT);
+
+		result = (apic_start - apic) * 1000L * cpu_khz /
+					(tsc - tsc_start);
+	}
+	printk("result %d\n", result);
 
-	result = (apic_start - apic) * 1000L * cpu_khz / (tsc - tsc_start);
 
 	printk(KERN_INFO "Detected %d.%03d MHz APIC timer.\n",
 		result / 1000 / 1000, result / 1000 % 1000);
@@ -872,6 +917,8 @@ void smp_local_timer_interrupt(struct pt_regs *regs)
 #ifdef CONFIG_SMP
 	update_process_times(user_mode(regs));
 #endif
+	if (apic_runs_main_timer > 1 && smp_processor_id() == boot_cpu_id)
+		main_timer_handler(regs);
 	/*
 	 * We take the 'long' return path, and there every subsystem
 	 * grabs the appropriate locks (kernel lock/ irq lock).
@@ -924,7 +971,7 @@ void smp_apic_timer_interrupt(struct pt_regs *regs)
  * multi-chassis. Use available data to take a good guess.
  * If in doubt, go HPET.
  */
-__init int oem_force_hpet_timer(void)
+__cpuinit int oem_force_hpet_timer(void)
 {
 	int i, clusters, zeros;
 	unsigned id;
@@ -1081,10 +1128,34 @@ static __init int setup_nolapic(char *str)
 
 static __init int setup_noapictimer(char *str) 
 { 
+	if (str[0] != ' ' && str[0] != 0)
+		return -1;
 	disable_apic_timer = 1;
 	return 0;
 } 
 
+static __init int setup_apicmaintimer(char *str)
+{
+	apic_runs_main_timer = 1;
+	nohpet = 1;
+	return 0;
+}
+__setup("apicmaintimer", setup_apicmaintimer);
+
+static __init int setup_noapicmaintimer(char *str)
+{
+	apic_runs_main_timer = -1;
+	return 0;
+}
+__setup("noapicmaintimer", setup_noapicmaintimer);
+
+static __init int setup_apicpmtimer(char *s)
+{
+	apic_calibrate_pmtmr = 1;
+	return setup_apicmaintimer(NULL);
+}
+__setup("apicpmtimer", setup_apicpmtimer);
+
 /* dummy parsing: see setup.c */
 
 __setup("disableapic", setup_disableapic); 
diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S
index dbdba56e8fa..b150c87a08c 100644
--- a/arch/x86_64/kernel/entry.S
+++ b/arch/x86_64/kernel/entry.S
@@ -499,7 +499,9 @@ ENTRY(stub_rt_sigreturn)
 	movq %gs:pda_irqstackptr,%rax
 	cmoveq %rax,%rsp /*todo This needs CFI annotation! */
 	pushq %rdi			# save old stack	
+#ifndef CONFIG_DEBUG_INFO
 	CFI_ADJUST_CFA_OFFSET	8
+#endif
 	call \func
 	.endm
 
@@ -509,7 +511,9 @@ ENTRY(common_interrupt)
 	/* 0(%rsp): oldrsp-ARGOFFSET */
 ret_from_intr:
 	popq  %rdi
+#ifndef CONFIG_DEBUG_INFO
 	CFI_ADJUST_CFA_OFFSET	-8
+#endif
 	cli	
 	decl %gs:pda_irqcount
 #ifdef CONFIG_DEBUG_INFO
@@ -922,7 +926,7 @@ KPROBE_ENTRY(debug)
 	.previous .text
 
 	/* runs on exception stack */	
-ENTRY(nmi)
+KPROBE_ENTRY(nmi)
 	INTR_FRAME
 	pushq $-1
 	CFI_ADJUST_CFA_OFFSET 8
@@ -969,6 +973,7 @@ paranoid_schedule:
 	cli
 	jmp paranoid_userspace
 	CFI_ENDPROC
+	.previous .text
 
 KPROBE_ENTRY(int3)
  	INTR_FRAME
diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c
index e8cf44ef877..4282d72b2a2 100644
--- a/arch/x86_64/kernel/io_apic.c
+++ b/arch/x86_64/kernel/io_apic.c
@@ -304,6 +304,14 @@ void __init check_ioapic(void)
 #endif
 					/* RED-PEN skip them on mptables too? */
 					return;
+				case PCI_VENDOR_ID_ATI:
+					if (apic_runs_main_timer != 0)
+						break;
+					printk(KERN_INFO
+	     "ATI board detected. Using APIC/PM timer.\n");
+					apic_runs_main_timer = 1;
+					nohpet = 1;
+					return;
 				} 
 
 				/* No multi-function device? */
@@ -2027,7 +2035,7 @@ int __init io_apic_get_redir_entries (int ioapic)
 }
 
 
-int io_apic_set_pci_routing (int ioapic, int pin, int irq, int edge_level, int active_high_low)
+int io_apic_set_pci_routing (int ioapic, int pin, int irq, int triggering, int polarity)
 {
 	struct IO_APIC_route_entry entry;
 	unsigned long flags;
@@ -2049,8 +2057,8 @@ int io_apic_set_pci_routing (int ioapic, int pin, int irq, int edge_level, int a
 	entry.delivery_mode = INT_DELIVERY_MODE;
 	entry.dest_mode = INT_DEST_MODE;
 	entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS);
-	entry.trigger = edge_level;
-	entry.polarity = active_high_low;
+	entry.trigger = triggering;
+	entry.polarity = polarity;
 	entry.mask = 1;					 /* Disabled (masked) */
 
 	irq = gsi_irq_sharing(irq);
@@ -2065,9 +2073,9 @@ int io_apic_set_pci_routing (int ioapic, int pin, int irq, int edge_level, int a
 	apic_printk(APIC_VERBOSE,KERN_DEBUG "IOAPIC[%d]: Set PCI routing entry (%d-%d -> 0x%x -> "
 		"IRQ %d Mode:%i Active:%i)\n", ioapic, 
 	       mp_ioapics[ioapic].mpc_apicid, pin, entry.vector, irq,
-	       edge_level, active_high_low);
+	       triggering, polarity);
 
-	ioapic_register_intr(irq, entry.vector, edge_level);
+	ioapic_register_intr(irq, entry.vector, triggering);
 
 	if (!ioapic && (irq < 16))
 		disable_8259A_irq(irq);
diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c
index 13a2eada6c9..b8b9529fa89 100644
--- a/arch/x86_64/kernel/mce.c
+++ b/arch/x86_64/kernel/mce.c
@@ -380,7 +380,7 @@ static void __cpuinit mce_cpu_features(struct cpuinfo_x86 *c)
  */
 void __cpuinit mcheck_init(struct cpuinfo_x86 *c)
 {
-	static cpumask_t mce_cpus __initdata = CPU_MASK_NONE;
+	static cpumask_t mce_cpus = CPU_MASK_NONE;
 
 	mce_cpu_quirks(c); 
 
diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c
index 1105250bf02..dc49bfb6db0 100644
--- a/arch/x86_64/kernel/mpparse.c
+++ b/arch/x86_64/kernel/mpparse.c
@@ -915,7 +915,7 @@ void __init mp_config_acpi_legacy_irqs (void)
 
 #define MAX_GSI_NUM	4096
 
-int mp_register_gsi(u32 gsi, int edge_level, int active_high_low)
+int mp_register_gsi(u32 gsi, int triggering, int polarity)
 {
 	int			ioapic = -1;
 	int			ioapic_pin = 0;
@@ -964,7 +964,7 @@ int mp_register_gsi(u32 gsi, int edge_level, int active_high_low)
 
 	mp_ioapic_routing[ioapic].pin_programmed[idx] |= (1<<bit);
 
-	if (edge_level) {
+	if (triggering == ACPI_LEVEL_SENSITIVE) {
 		/*
 		 * For PCI devices assign IRQs in order, avoiding gaps
 		 * due to unused I/O APIC pins.
@@ -986,8 +986,8 @@ int mp_register_gsi(u32 gsi, int edge_level, int active_high_low)
 	}
 
 	io_apic_set_pci_routing(ioapic, ioapic_pin, gsi,
-		edge_level == ACPI_EDGE_SENSITIVE ? 0 : 1,
-		active_high_low == ACPI_ACTIVE_HIGH ? 0 : 1);
+		triggering == ACPI_EDGE_SENSITIVE ? 0 : 1,
+		polarity == ACPI_ACTIVE_HIGH ? 0 : 1);
 	return gsi;
 }
 
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index 5fae6f0cd99..8be407a1f62 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -24,6 +24,7 @@
 #include <linux/sysdev.h>
 #include <linux/nmi.h>
 #include <linux/sysctl.h>
+#include <linux/kprobes.h>
 
 #include <asm/smp.h>
 #include <asm/mtrr.h>
@@ -468,7 +469,7 @@ void touch_nmi_watchdog (void)
  	touch_softlockup_watchdog();
 }
 
-void nmi_watchdog_tick (struct pt_regs * regs, unsigned reason)
+void __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason)
 {
 	int sum;
 	int touched = 0;
@@ -512,14 +513,14 @@ void nmi_watchdog_tick (struct pt_regs * regs, unsigned reason)
 	}
 }
 
-static int dummy_nmi_callback(struct pt_regs * regs, int cpu)
+static __kprobes int dummy_nmi_callback(struct pt_regs * regs, int cpu)
 {
 	return 0;
 }
  
 static nmi_callback_t nmi_callback = dummy_nmi_callback;
  
-asmlinkage void do_nmi(struct pt_regs * regs, long error_code)
+asmlinkage __kprobes void do_nmi(struct pt_regs * regs, long error_code)
 {
 	int cpu = safe_smp_processor_id();
 
diff --git a/arch/x86_64/kernel/pci-dma.c b/arch/x86_64/kernel/pci-dma.c
index 2f5d8328e2b..4ed391edd47 100644
--- a/arch/x86_64/kernel/pci-dma.c
+++ b/arch/x86_64/kernel/pci-dma.c
@@ -107,6 +107,9 @@ dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle,
 				goto again;
 			}
 
+			/* Let low level make its own zone decisions */
+			gfp &= ~(GFP_DMA32|GFP_DMA);
+
 			if (dma_ops->alloc_coherent)
 				return dma_ops->alloc_coherent(dev, size,
 							   dma_handle, gfp);
diff --git a/arch/x86_64/kernel/pci-gart.c b/arch/x86_64/kernel/pci-gart.c
index c37fc7726ba..2fe23a6c361 100644
--- a/arch/x86_64/kernel/pci-gart.c
+++ b/arch/x86_64/kernel/pci-gart.c
@@ -457,9 +457,12 @@ int gart_map_sg(struct device *dev, struct scatterlist *sg, int nents, int dir)
 error:
 	flush_gart(NULL);
 	gart_unmap_sg(dev, sg, nents, dir);
-	/* When it was forced try again unforced */
-	if (force_iommu) 
-		return dma_map_sg_nonforce(dev, sg, nents, dir);
+	/* When it was forced or merged try again in a dumb way */
+	if (force_iommu || iommu_merge) {
+		out = dma_map_sg_nonforce(dev, sg, nents, dir);
+		if (out > 0)
+			return out;
+	}
 	if (panic_on_overflow)
 		panic("dma_map_sg: overflow on %lu pages\n", pages);
 	iommu_full(dev, pages << PAGE_SHIFT, dir);
@@ -642,9 +645,18 @@ static int __init pci_iommu_init(void)
 	    (no_agp && init_k8_gatt(&info) < 0)) {
 		no_iommu = 1;
 		no_iommu_init();
+		printk(KERN_INFO "PCI-DMA: Disabling IOMMU.\n");
+		if (end_pfn > MAX_DMA32_PFN) {
+			printk(KERN_ERR "WARNING more than 4GB of memory "
+					"but IOMMU not compiled in.\n"
+			       KERN_ERR "WARNING 32bit PCI may malfunction.\n"
+			       KERN_ERR "You might want to enable "
+					"CONFIG_GART_IOMMU\n");
+		}
 		return -1;
 	}
 
+	printk(KERN_INFO "PCI-DMA: using GART IOMMU.\n");
 	aper_size = info.aper_size * 1024 * 1024;	
 	iommu_size = check_iommu_size(info.aper_base, aper_size); 
 	iommu_pages = iommu_size >> PAGE_SHIFT; 
@@ -718,7 +730,6 @@ static int __init pci_iommu_init(void)
 		     
 	flush_gart(NULL);
 
-	printk(KERN_INFO "PCI-DMA: using GART IOMMU.\n");
 	dma_ops = &gart_dma_ops;
 
 	return 0;
diff --git a/arch/x86_64/kernel/pci-nommu.c b/arch/x86_64/kernel/pci-nommu.c
index e4156497519..44adcc2d5e5 100644
--- a/arch/x86_64/kernel/pci-nommu.c
+++ b/arch/x86_64/kernel/pci-nommu.c
@@ -88,12 +88,5 @@ void __init no_iommu_init(void)
 {
 	if (dma_ops)
 		return;
-	printk(KERN_INFO "PCI-DMA: Disabling IOMMU.\n");
 	dma_ops = &nommu_dma_ops;
-	if (end_pfn > MAX_DMA32_PFN) {
-		printk(KERN_ERR
-		       "WARNING more than 4GB of memory but IOMMU not compiled in.\n"
-		       KERN_ERR "WARNING 32bit PCI may malfunction.\n"
-		       KERN_ERR "You might want to enable CONFIG_GART_IOMMU\n");
-	}
 }
diff --git a/arch/x86_64/kernel/pci-swiotlb.c b/arch/x86_64/kernel/pci-swiotlb.c
index 3569a25ad7f..990ed67896f 100644
--- a/arch/x86_64/kernel/pci-swiotlb.c
+++ b/arch/x86_64/kernel/pci-swiotlb.c
@@ -35,8 +35,8 @@ void pci_swiotlb_init(void)
 	    (end_pfn > MAX_DMA32_PFN || force_iommu))
 	       swiotlb = 1;
 	if (swiotlb) {
-		swiotlb_init();
 		printk(KERN_INFO "PCI-DMA: Using software bounce buffering for IO (SWIOTLB)\n");
+		swiotlb_init();
 		dma_ops = &swiotlb_dma_ops;
 	}
 }
diff --git a/arch/x86_64/kernel/pmtimer.c b/arch/x86_64/kernel/pmtimer.c
index feb5f108dd2..5c51d10408a 100644
--- a/arch/x86_64/kernel/pmtimer.c
+++ b/arch/x86_64/kernel/pmtimer.c
@@ -80,6 +80,31 @@ int pmtimer_mark_offset(void)
 	return lost - 1;
 }
 
+static unsigned pmtimer_wait_tick(void)
+{
+	u32 a, b;
+	for (a = b = inl(pmtmr_ioport) & ACPI_PM_MASK;
+	     a == b;
+	     b = inl(pmtmr_ioport) & ACPI_PM_MASK)
+		;
+	return b;
+}
+
+/* note: wait time is rounded up to one tick */
+void pmtimer_wait(unsigned us)
+{
+	u32 a, b;
+	a = pmtimer_wait_tick();
+	do {
+		b = inl(pmtmr_ioport);
+	} while (cyc2us(b - a) < us);
+}
+
+void pmtimer_resume(void)
+{
+	last_pmtmr_tick = inl(pmtmr_ioport);
+}
+
 unsigned int do_gettimeoffset_pm(void)
 {
 	u32 now, offset, delta = 0;
diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c
index 8ded407e4a9..22a05dec81a 100644
--- a/arch/x86_64/kernel/process.c
+++ b/arch/x86_64/kernel/process.c
@@ -516,8 +516,10 @@ out:
  * This could still be optimized: 
  * - fold all the options into a flag word and test it with a single test.
  * - could test fs/gs bitsliced
+ *
+ * Kprobes not supported here. Set the probe on schedule instead.
  */
-struct task_struct *
+__kprobes struct task_struct *
 __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 {
 	struct thread_struct *prev = &prev_p->thread,
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 28895c03cb1..363db5a003d 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -741,7 +741,7 @@ void __init setup_arch(char **cmdline_p)
 	e820_setup_gap();
 
 #ifdef CONFIG_GART_IOMMU
-       iommu_hole_init();
+	iommu_hole_init();
 #endif
 
 #ifdef CONFIG_VT
@@ -877,6 +877,7 @@ static void __init amd_detect_cmp(struct cpuinfo_x86 *c)
 static int __init init_amd(struct cpuinfo_x86 *c)
 {
 	int r;
+	unsigned level;
 
 #ifdef CONFIG_SMP
 	unsigned long value;
@@ -899,6 +900,11 @@ static int __init init_amd(struct cpuinfo_x86 *c)
 	   3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway */
 	clear_bit(0*32+31, &c->x86_capability);
 	
+	/* On C+ stepping K8 rep microcode works well for copy/memset */
+	level = cpuid_eax(1);
+	if (c->x86 == 15 && ((level >= 0x0f48 && level < 0x0f50) || level >= 0x0f58))
+		set_bit(X86_FEATURE_REP_GOOD, &c->x86_capability);
+
 	r = get_model_name(c);
 	if (!r) { 
 		switch (c->x86) { 
diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index f8c47c68844..c0844bffbf8 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -51,7 +51,7 @@ extern int using_apic_timer;
 DEFINE_SPINLOCK(rtc_lock);
 DEFINE_SPINLOCK(i8253_lock);
 
-static int nohpet __initdata = 0;
+int nohpet __initdata = 0;
 static int notsc __initdata = 0;
 
 #undef HPET_HACK_ENABLE_DANGEROUS
@@ -345,7 +345,7 @@ static noinline void handle_lost_ticks(int lost, struct pt_regs *regs)
 #endif
 }
 
-static irqreturn_t timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
+void main_timer_handler(struct pt_regs *regs)
 {
 	static unsigned long rtc_update = 0;
 	unsigned long tsc;
@@ -458,12 +458,17 @@ static irqreturn_t timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 	}
  
 	write_sequnlock(&xtime_lock);
+}
 
+static irqreturn_t timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
+{
+	if (apic_runs_main_timer > 1)
+		return IRQ_HANDLED;
+	main_timer_handler(regs);
 #ifdef CONFIG_X86_LOCAL_APIC
 	if (using_apic_timer)
 		smp_send_timer_broadcast_ipi();
 #endif
-
 	return IRQ_HANDLED;
 }
 
@@ -843,17 +848,43 @@ static int hpet_reenable(void)
 	return hpet_timer_stop_set_go(hpet_tick);
 }
 
-void __init pit_init(void)
+#define PIT_MODE 0x43
+#define PIT_CH0  0x40
+
+static void __init __pit_init(int val, u8 mode)
 {
 	unsigned long flags;
 
 	spin_lock_irqsave(&i8253_lock, flags);
-	outb_p(0x34, 0x43);		/* binary, mode 2, LSB/MSB, ch 0 */
-	outb_p(LATCH & 0xff, 0x40);	/* LSB */
-	outb_p(LATCH >> 8, 0x40);	/* MSB */
+	outb_p(mode, PIT_MODE);
+	outb_p(val & 0xff, PIT_CH0);	/* LSB */
+	outb_p(val >> 8, PIT_CH0);	/* MSB */
 	spin_unlock_irqrestore(&i8253_lock, flags);
 }
 
+void __init pit_init(void)
+{
+	__pit_init(LATCH, 0x34); /* binary, mode 2, LSB/MSB, ch 0 */
+}
+
+void __init pit_stop_interrupt(void)
+{
+	__pit_init(0, 0x30); /* mode 0 */
+}
+
+void __init stop_timer_interrupt(void)
+{
+	char *name;
+	if (vxtime.hpet_address) {
+		name = "HPET";
+		hpet_timer_stop_set_go(0);
+	} else {
+		name = "PIT";
+		pit_stop_interrupt();
+	}
+	printk(KERN_INFO "timer: %s interrupt stopped.\n", name);
+}
+
 int __init time_setup(char *str)
 {
 	report_lost_ticks = 1;
@@ -932,7 +963,7 @@ void __init time_init(void)
  * Make an educated guess if the TSC is trustworthy and synchronized
  * over all CPUs.
  */
-__init int unsynchronized_tsc(void)
+__cpuinit int unsynchronized_tsc(void)
 {
 #ifdef CONFIG_SMP
 	if (oem_force_hpet_timer())
@@ -1016,9 +1047,21 @@ static int timer_resume(struct sys_device *dev)
 	write_seqlock_irqsave(&xtime_lock,flags);
 	xtime.tv_sec = sec;
 	xtime.tv_nsec = 0;
+	if (vxtime.mode == VXTIME_HPET) {
+		if (hpet_use_timer)
+			vxtime.last = hpet_readl(HPET_T0_CMP) - hpet_tick;
+		else
+			vxtime.last = hpet_readl(HPET_COUNTER);
+#ifdef CONFIG_X86_PM_TIMER
+	} else if (vxtime.mode == VXTIME_PMTMR) {
+		pmtimer_resume();
+#endif
+	} else
+		vxtime.last_tsc = get_cycles_sync();
 	write_sequnlock_irqrestore(&xtime_lock,flags);
 	jiffies += sleep_length;
 	wall_jiffies += sleep_length;
+	monotonic_base += sleep_length * (NSEC_PER_SEC/HZ);
 	touch_softlockup_watchdog();
 	return 0;
 }
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index 8bb0aeda78b..ee1b2da9e5e 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -372,7 +372,7 @@ void out_of_line_bug(void)
 static DEFINE_SPINLOCK(die_lock);
 static int die_owner = -1;
 
-unsigned long oops_begin(void)
+unsigned __kprobes long oops_begin(void)
 {
 	int cpu = safe_smp_processor_id();
 	unsigned long flags;
@@ -391,7 +391,7 @@ unsigned long oops_begin(void)
 	return flags;
 }
 
-void oops_end(unsigned long flags)
+void __kprobes oops_end(unsigned long flags)
 { 
 	die_owner = -1;
 	bust_spinlocks(0);
@@ -400,7 +400,7 @@ void oops_end(unsigned long flags)
 		panic("Oops");
 }
 
-void __die(const char * str, struct pt_regs * regs, long err)
+void __kprobes __die(const char * str, struct pt_regs * regs, long err)
 {
 	static int die_counter;
 	printk(KERN_EMERG "%s: %04lx [%u] ", str, err & 0xffff,++die_counter);
@@ -432,7 +432,7 @@ void die(const char * str, struct pt_regs * regs, long err)
 	do_exit(SIGSEGV); 
 }
 
-void die_nmi(char *str, struct pt_regs *regs)
+void __kprobes die_nmi(char *str, struct pt_regs *regs)
 {
 	unsigned long flags = oops_begin();
 
@@ -575,7 +575,8 @@ asmlinkage void __kprobes do_general_protection(struct pt_regs * regs,
 	}
 }
 
-static void mem_parity_error(unsigned char reason, struct pt_regs * regs)
+static __kprobes void
+mem_parity_error(unsigned char reason, struct pt_regs * regs)
 {
 	printk("Uhhuh. NMI received. Dazed and confused, but trying to continue\n");
 	printk("You probably have a hardware problem with your RAM chips\n");
@@ -585,7 +586,8 @@ static void mem_parity_error(unsigned char reason, struct pt_regs * regs)
 	outb(reason, 0x61);
 }
 
-static void io_check_error(unsigned char reason, struct pt_regs * regs)
+static __kprobes void
+io_check_error(unsigned char reason, struct pt_regs * regs)
 {
 	printk("NMI: IOCK error (debug interrupt?)\n");
 	show_registers(regs);
@@ -598,7 +600,8 @@ static void io_check_error(unsigned char reason, struct pt_regs * regs)
 	outb(reason, 0x61);
 }
 
-static void unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
+static __kprobes void
+unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
 {	printk("Uhhuh. NMI received for unknown reason %02x.\n", reason);
 	printk("Dazed and confused, but trying to continue\n");
 	printk("Do you have a strange power saving mode enabled?\n");
@@ -606,7 +609,7 @@ static void unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
 
 /* Runs on IST stack. This code must keep interrupts off all the time.
    Nested NMIs are prevented by the CPU. */
-asmlinkage void default_do_nmi(struct pt_regs *regs)
+asmlinkage __kprobes void default_do_nmi(struct pt_regs *regs)
 {
 	unsigned char reason = 0;
 	int cpu;
@@ -658,7 +661,7 @@ asmlinkage void __kprobes do_int3(struct pt_regs * regs, long error_code)
 /* Help handler running on IST stack to switch back to user stack
    for scheduling or signal handling. The actual stack switch is done in
    entry.S */
-asmlinkage struct pt_regs *sync_regs(struct pt_regs *eregs)
+asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs)
 {
 	struct pt_regs *regs = eregs;
 	/* Did already sync */
diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S
index b0eed1faf74..74db0062d4a 100644
--- a/arch/x86_64/kernel/vmlinux.lds.S
+++ b/arch/x86_64/kernel/vmlinux.lds.S
@@ -172,13 +172,15 @@ SECTIONS
   . = ALIGN(4096);
   __initramfs_start = .;
   .init.ramfs : AT(ADDR(.init.ramfs) - LOAD_OFFSET) { *(.init.ramfs) }
-  __initramfs_end = .;	
-  . = ALIGN(32);
+  __initramfs_end = .;
+  /* temporary here to work around NR_CPUS. If you see this comment in 2.6.17+
+   complain */
+  . = ALIGN(4096);	
+  __init_end = .;	
+  . = ALIGN(CONFIG_X86_L1_CACHE_BYTES);
   __per_cpu_start = .;
   .data.percpu  : AT(ADDR(.data.percpu) - LOAD_OFFSET) { *(.data.percpu) }
   __per_cpu_end = .;
-  . = ALIGN(4096);
-  __init_end = .;
 
   . = ALIGN(4096);
   __nosave_begin = .;