Merge branch 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip

* 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  x86, mce: Clean up thermal init by introducing intel_thermal_supported()
  x86, mce: Thermal monitoring depends on APIC being enabled
  x86: Gart: fix breakage due to IOMMU initialization cleanup
  x86: Move swiotlb initialization before dma32_free_bootmem
  x86: Fix build warning in arch/x86/mm/mmio-mod.c
  x86: Remove usedac in feature-removal-schedule.txt
  x86: Fix duplicated UV BAU interrupt vector
  nvram: Fix write beyond end condition; prove to gcc copy is safe
  mm: Adjust do_pages_stat() so gcc can see copy_from_user() is safe
  x86: Limit the number of processor bootup messages
  x86: Remove enabling x2apic message for every CPU
  doc: Add documentation for bootloader_{type,version}
  x86, msr: Add support for non-contiguous cpumasks
  x86: Use find_e820() instead of hard coded trampoline address
  x86, AMD: Fix stale cpuid4_info shared_map data in shared_cpu_map cpumasks

Trivial percpu-naming-introduced conflicts in arch/x86/kernel/cpu/intel_cacheinfo.c

22 files changed, 133 insertions, 78 deletions

diff --git a/arch/x86/include/asm/irq_vectors.h b/arch/x86/include/asm/irq_vectors.h
index 6a635bd..4611f08 100644
--- a/arch/x86/include/asm/irq_vectors.h
+++ b/arch/x86/include/asm/irq_vectors.h
@@ -113,7 +113,7 @@
  */
 #define LOCAL_PENDING_VECTOR		0xec
 
-#define UV_BAU_MESSAGE			0xec
+#define UV_BAU_MESSAGE			0xea
 
 /*
  * Self IPI vector for machine checks
diff --git a/arch/x86/include/asm/msr.h b/arch/x86/include/asm/msr.h
index 5bef931..2d228fc 100644
--- a/arch/x86/include/asm/msr.h
+++ b/arch/x86/include/asm/msr.h
@@ -244,6 +244,9 @@ do {                                                            \
 
 #define write_rdtscp_aux(val) wrmsr(0xc0000103, (val), 0)
 
+struct msr *msrs_alloc(void);
+void msrs_free(struct msr *msrs);
+
 #ifdef CONFIG_SMP
 int rdmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 *l, u32 *h);
 int wrmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h);
diff --git a/arch/x86/include/asm/trampoline.h b/arch/x86/include/asm/trampoline.h
index 90f06c2..cb507bb 100644
--- a/arch/x86/include/asm/trampoline.h
+++ b/arch/x86/include/asm/trampoline.h
@@ -16,7 +16,6 @@ extern unsigned long initial_code;
 extern unsigned long initial_gs;
 
 #define TRAMPOLINE_SIZE roundup(trampoline_end - trampoline_data, PAGE_SIZE)
-#define TRAMPOLINE_BASE 0x6000
 
 extern unsigned long setup_trampoline(void);
 extern void __init reserve_trampoline_memory(void);
diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c
index e0dfb68..3704997 100644
--- a/arch/x86/kernel/aperture_64.c
+++ b/arch/x86/kernel/aperture_64.c
@@ -280,7 +280,8 @@ void __init early_gart_iommu_check(void)
 	 * or BIOS forget to put that in reserved.
 	 * try to update e820 to make that region as reserved.
 	 */
-	int i, fix, slot;
+	u32 agp_aper_base = 0, agp_aper_order = 0;
+	int i, fix, slot, valid_agp = 0;
 	u32 ctl;
 	u32 aper_size = 0, aper_order = 0, last_aper_order = 0;
 	u64 aper_base = 0, last_aper_base = 0;
@@ -290,6 +291,8 @@ void __init early_gart_iommu_check(void)
 		return;
 
 	/* This is mostly duplicate of iommu_hole_init */
+	agp_aper_base = search_agp_bridge(&agp_aper_order, &valid_agp);
+
 	fix = 0;
 	for (i = 0; i < ARRAY_SIZE(bus_dev_ranges); i++) {
 		int bus;
@@ -342,10 +345,10 @@ void __init early_gart_iommu_check(void)
 		}
 	}
 
-	if (!fix)
+	if (valid_agp)
 		return;
 
-	/* different nodes have different setting, disable them all at first*/
+	/* disable them all at first */
 	for (i = 0; i < ARRAY_SIZE(bus_dev_ranges); i++) {
 		int bus;
 		int dev_base, dev_limit;
@@ -458,8 +461,6 @@ out:
 
 	if (aper_alloc) {
 		/* Got the aperture from the AGP bridge */
-	} else if (!valid_agp) {
-		/* Do nothing */
 	} else if ((!no_iommu && max_pfn > MAX_DMA32_PFN) ||
 		   force_iommu ||
 		   valid_agp ||
diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c
index efb2b9c..aa57c07 100644
--- a/arch/x86/kernel/apic/apic.c
+++ b/arch/x86/kernel/apic/apic.c
@@ -1341,7 +1341,7 @@ void enable_x2apic(void)
 
 	rdmsr(MSR_IA32_APICBASE, msr, msr2);
 	if (!(msr & X2APIC_ENABLE)) {
-		pr_info("Enabling x2apic\n");
+		printk_once(KERN_INFO "Enabling x2apic\n");
 		wrmsr(MSR_IA32_APICBASE, msr | X2APIC_ENABLE, 0);
 	}
 }
diff --git a/arch/x86/kernel/cpu/addon_cpuid_features.c b/arch/x86/kernel/cpu/addon_cpuid_features.c
index c965e52..468489b 100644
--- a/arch/x86/kernel/cpu/addon_cpuid_features.c
+++ b/arch/x86/kernel/cpu/addon_cpuid_features.c
@@ -74,6 +74,7 @@ void __cpuinit detect_extended_topology(struct cpuinfo_x86 *c)
 	unsigned int eax, ebx, ecx, edx, sub_index;
 	unsigned int ht_mask_width, core_plus_mask_width;
 	unsigned int core_select_mask, core_level_siblings;
+	static bool printed;
 
 	if (c->cpuid_level < 0xb)
 		return;
@@ -127,12 +128,14 @@ void __cpuinit detect_extended_topology(struct cpuinfo_x86 *c)
 
 	c->x86_max_cores = (core_level_siblings / smp_num_siblings);
 
-
-	printk(KERN_INFO  "CPU: Physical Processor ID: %d\n",
-	       c->phys_proc_id);
-	if (c->x86_max_cores > 1)
-		printk(KERN_INFO  "CPU: Processor Core ID: %d\n",
-		       c->cpu_core_id);
+	if (!printed) {
+		printk(KERN_INFO  "CPU: Physical Processor ID: %d\n",
+		       c->phys_proc_id);
+		if (c->x86_max_cores > 1)
+			printk(KERN_INFO  "CPU: Processor Core ID: %d\n",
+			       c->cpu_core_id);
+		printed = 1;
+	}
 	return;
 #endif
 }
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index 7128b37..8dc3ea1 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -375,8 +375,6 @@ static void __cpuinit srat_detect_node(struct cpuinfo_x86 *c)
 			node = nearby_node(apicid);
 	}
 	numa_set_node(cpu, node);
-
-	printk(KERN_INFO "CPU %d/0x%x -> Node %d\n", cpu, apicid, node);
 #endif
 }
 
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 20399b7..4868e4a 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -427,6 +427,7 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c)
 #ifdef CONFIG_X86_HT
 	u32 eax, ebx, ecx, edx;
 	int index_msb, core_bits;
+	static bool printed;
 
 	if (!cpu_has(c, X86_FEATURE_HT))
 		return;
@@ -442,7 +443,7 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c)
 	smp_num_siblings = (ebx & 0xff0000) >> 16;
 
 	if (smp_num_siblings == 1) {
-		printk(KERN_INFO  "CPU: Hyper-Threading is disabled\n");
+		printk_once(KERN_INFO "CPU0: Hyper-Threading is disabled\n");
 		goto out;
 	}
 
@@ -469,11 +470,12 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c)
 				       ((1 << core_bits) - 1);
 
 out:
-	if ((c->x86_max_cores * smp_num_siblings) > 1) {
+	if (!printed && (c->x86_max_cores * smp_num_siblings) > 1) {
 		printk(KERN_INFO  "CPU: Physical Processor ID: %d\n",
 		       c->phys_proc_id);
 		printk(KERN_INFO  "CPU: Processor Core ID: %d\n",
 		       c->cpu_core_id);
+		printed = 1;
 	}
 #endif
 }
@@ -1115,7 +1117,7 @@ void __cpuinit cpu_init(void)
 	if (cpumask_test_and_set_cpu(cpu, cpu_initialized_mask))
 		panic("CPU#%d already initialized!\n", cpu);
 
-	printk(KERN_INFO "Initializing CPU#%d\n", cpu);
+	pr_debug("Initializing CPU#%d\n", cpu);
 
 	clear_in_cr4(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE);
 
diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c
index c900b73..9c31e8b 100644
--- a/arch/x86/kernel/cpu/intel.c
+++ b/arch/x86/kernel/cpu/intel.c
@@ -270,8 +270,6 @@ static void __cpuinit srat_detect_node(struct cpuinfo_x86 *c)
 		node = cpu_to_node(cpu);
 	}
 	numa_set_node(cpu, node);
-
-	printk(KERN_INFO "CPU %d/0x%x -> Node %d\n", cpu, apicid, node);
 #endif
 }
 
diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c
index 0c06bca..fc6c8ef 100644
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -507,18 +507,19 @@ static void __cpuinit cache_shared_cpu_map_setup(unsigned int cpu, int index)
 {
 	struct _cpuid4_info	*this_leaf, *sibling_leaf;
 	unsigned long num_threads_sharing;
-	int index_msb, i;
+	int index_msb, i, sibling;
 	struct cpuinfo_x86 *c = &cpu_data(cpu);
 
 	if ((index == 3) && (c->x86_vendor == X86_VENDOR_AMD)) {
-		struct cpuinfo_x86 *d;
-		for_each_online_cpu(i) {
+		for_each_cpu(i, c->llc_shared_map) {
 			if (!per_cpu(ici_cpuid4_info, i))
 				continue;
-			d = &cpu_data(i);
 			this_leaf = CPUID4_INFO_IDX(i, index);
-			cpumask_copy(to_cpumask(this_leaf->shared_cpu_map),
-				     d->llc_shared_map);
+			for_each_cpu(sibling, c->llc_shared_map) {
+				if (!cpu_online(sibling))
+					continue;
+				set_bit(sibling, this_leaf->shared_cpu_map);
+			}
 		}
 		return;
 	}
diff --git a/arch/x86/kernel/cpu/mcheck/therm_throt.c b/arch/x86/kernel/cpu/mcheck/therm_throt.c
index 4fef985..81c499e 100644
--- a/arch/x86/kernel/cpu/mcheck/therm_throt.c
+++ b/arch/x86/kernel/cpu/mcheck/therm_throt.c
@@ -256,6 +256,16 @@ asmlinkage void smp_thermal_interrupt(struct pt_regs *regs)
 	ack_APIC_irq();
 }
 
+/* Thermal monitoring depends on APIC, ACPI and clock modulation */
+static int intel_thermal_supported(struct cpuinfo_x86 *c)
+{
+	if (!cpu_has_apic)
+		return 0;
+	if (!cpu_has(c, X86_FEATURE_ACPI) || !cpu_has(c, X86_FEATURE_ACC))
+		return 0;
+	return 1;
+}
+
 void __init mcheck_intel_therm_init(void)
 {
 	/*
@@ -263,8 +273,7 @@ void __init mcheck_intel_therm_init(void)
 	 * LVT value on BSP and use that value to restore APs' thermal LVT
 	 * entry BIOS programmed later
 	 */
-	if (cpu_has(&boot_cpu_data, X86_FEATURE_ACPI) &&
-		cpu_has(&boot_cpu_data, X86_FEATURE_ACC))
+	if (intel_thermal_supported(&boot_cpu_data))
 		lvtthmr_init = apic_read(APIC_LVTTHMR);
 }
 
@@ -274,8 +283,7 @@ void intel_init_thermal(struct cpuinfo_x86 *c)
 	int tm2 = 0;
 	u32 l, h;
 
-	/* Thermal monitoring depends on ACPI and clock modulation*/
-	if (!cpu_has(c, X86_FEATURE_ACPI) || !cpu_has(c, X86_FEATURE_ACC))
+	if (!intel_thermal_supported(c))
 		return;
 
 	/*
@@ -339,8 +347,8 @@ void intel_init_thermal(struct cpuinfo_x86 *c)
 	l = apic_read(APIC_LVTTHMR);
 	apic_write(APIC_LVTTHMR, l & ~APIC_LVT_MASKED);
 
-	printk(KERN_INFO "CPU%d: Thermal monitoring enabled (%s)\n",
-	       cpu, tm2 ? "TM2" : "TM1");
+	printk_once(KERN_INFO "CPU0: Thermal monitoring enabled (%s)\n",
+		       tm2 ? "TM2" : "TM1");
 
 	/* enable thermal throttle processing */
 	atomic_set(&therm_throt_en, 1);
diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
index d17d482..f50447d 100644
--- a/arch/x86/kernel/e820.c
+++ b/arch/x86/kernel/e820.c
@@ -732,7 +732,16 @@ struct early_res {
 	char overlap_ok;
 };
 static struct early_res early_res[MAX_EARLY_RES] __initdata = {
-	{ 0, PAGE_SIZE, "BIOS data page" },	/* BIOS data page */
+	{ 0, PAGE_SIZE, "BIOS data page", 1 },	/* BIOS data page */
+#ifdef CONFIG_X86_32
+	/*
+	 * But first pinch a few for the stack/trampoline stuff
+	 * FIXME: Don't need the extra page at 4K, but need to fix
+	 * trampoline before removing it. (see the GDT stuff)
+	 */
+	{ PAGE_SIZE, PAGE_SIZE, "EX TRAMPOLINE", 1 },
+#endif
+
 	{}
 };
 
diff --git a/arch/x86/kernel/head32.c b/arch/x86/kernel/head32.c
index 4f8e250..5051b94 100644
--- a/arch/x86/kernel/head32.c
+++ b/arch/x86/kernel/head32.c
@@ -29,8 +29,6 @@ static void __init i386_default_early_setup(void)
 
 void __init i386_start_kernel(void)
 {
-	reserve_trampoline_memory();
-
 	reserve_early(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS");
 
 #ifdef CONFIG_BLK_DEV_INITRD
diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
index 0b06cd7..b5a9896 100644
--- a/arch/x86/kernel/head64.c
+++ b/arch/x86/kernel/head64.c
@@ -98,8 +98,6 @@ void __init x86_64_start_reservations(char *real_mode_data)
 {
 	copy_bootdata(__va(real_mode_data));
 
-	reserve_trampoline_memory();
-
 	reserve_early(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS");
 
 #ifdef CONFIG_BLK_DEV_INITRD
diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c
index 35a57c9..40b54ce 100644
--- a/arch/x86/kernel/mpparse.c
+++ b/arch/x86/kernel/mpparse.c
@@ -945,9 +945,6 @@ void __init early_reserve_e820_mpc_new(void)
 {
 	if (enable_update_mptable && alloc_mptable) {
 		u64 startt = 0;
-#ifdef CONFIG_X86_TRAMPOLINE
-		startt = TRAMPOLINE_BASE;
-#endif
 		mpc_new_phys = early_reserve_e820(startt, mpc_new_length, 4);
 	}
 }
diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c
index afcc58b..fcc2f2b 100644
--- a/arch/x86/kernel/pci-dma.c
+++ b/arch/x86/kernel/pci-dma.c
@@ -120,11 +120,14 @@ static void __init dma32_free_bootmem(void)
 
 void __init pci_iommu_alloc(void)
 {
+	int use_swiotlb;
+
+	use_swiotlb = pci_swiotlb_init();
 #ifdef CONFIG_X86_64
 	/* free the range so iommu could get some range less than 4G */
 	dma32_free_bootmem();
 #endif
-	if (pci_swiotlb_init())
+	if (use_swiotlb)
 		return;
 
 	gart_iommu_hole_init();
diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c
index e6a0d40..56c0e73 100644
--- a/arch/x86/kernel/pci-gart_64.c
+++ b/arch/x86/kernel/pci-gart_64.c
@@ -710,7 +710,8 @@ static void gart_iommu_shutdown(void)
 	struct pci_dev *dev;
 	int i;
 
-	if (no_agp)
+	/* don't shutdown it if there is AGP installed */
+	if (!no_agp)
 		return;
 
 	for (i = 0; i < num_k8_northbridges; i++) {
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 946a311..f7b8b98 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -73,6 +73,7 @@
 
 #include <asm/mtrr.h>
 #include <asm/apic.h>
+#include <asm/trampoline.h>
 #include <asm/e820.h>
 #include <asm/mpspec.h>
 #include <asm/setup.h>
@@ -875,6 +876,13 @@ void __init setup_arch(char **cmdline_p)
 
 	reserve_brk();
 
+	/*
+	 * Find and reserve possible boot-time SMP configuration:
+	 */
+	find_smp_config();
+
+	reserve_trampoline_memory();
+
 #ifdef CONFIG_ACPI_SLEEP
 	/*
 	 * Reserve low memory region for sleep support.
@@ -921,11 +929,6 @@ void __init setup_arch(char **cmdline_p)
 
 	early_acpi_boot_init();
 
-	/*
-	 * Find and reserve possible boot-time SMP configuration:
-	 */
-	find_smp_config();
-
 #ifdef CONFIG_ACPI_NUMA
 	/*
 	 * Parse SRAT to discover nodes.
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 29e6744..678d0b8 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -671,6 +671,26 @@ static void __cpuinit do_fork_idle(struct work_struct *work)
 	complete(&c_idle->done);
 }
 
+/* reduce the number of lines printed when booting a large cpu count system */
+static void __cpuinit announce_cpu(int cpu, int apicid)
+{
+	static int current_node = -1;
+	int node = cpu_to_node(cpu);
+
+	if (system_state == SYSTEM_BOOTING) {
+		if (node != current_node) {
+			if (current_node > (-1))
+				pr_cont(" Ok.\n");
+			current_node = node;
+			pr_info("Booting Node %3d, Processors ", node);
+		}
+		pr_cont(" #%d%s", cpu, cpu == (nr_cpu_ids - 1) ? " Ok.\n" : "");
+		return;
+	} else
+		pr_info("Booting Node %d Processor %d APIC 0x%x\n",
+			node, cpu, apicid);
+}
+
 /*
  * NOTE - on most systems this is a PHYSICAL apic ID, but on multiquad
  * (ie clustered apic addressing mode), this is a LOGICAL apic ID.
@@ -737,9 +757,8 @@ do_rest:
 	/* start_ip had better be page-aligned! */
 	start_ip = setup_trampoline();
 
-	/* So we see what's up   */
-	printk(KERN_INFO "Booting processor %d APIC 0x%x ip 0x%lx\n",
-			  cpu, apicid, start_ip);
+	/* So we see what's up */
+	announce_cpu(cpu, apicid);
 
 	/*
 	 * This grunge runs the startup process for
@@ -788,21 +807,17 @@ do_rest:
 			udelay(100);
 		}
 
-		if (cpumask_test_cpu(cpu, cpu_callin_mask)) {
-			/* number CPUs logically, starting from 1 (BSP is 0) */
-			pr_debug("OK.\n");
-			printk(KERN_INFO "CPU%d: ", cpu);
-			print_cpu_info(&cpu_data(cpu));
-			pr_debug("CPU has booted.\n");
-		} else {
+		if (cpumask_test_cpu(cpu, cpu_callin_mask))
+			pr_debug("CPU%d: has booted.\n", cpu);
+		else {
 			boot_error = 1;
 			if (*((volatile unsigned char *)trampoline_base)
 					== 0xA5)
 				/* trampoline started but...? */
-				printk(KERN_ERR "Stuck ??\n");
+				pr_err("CPU%d: Stuck ??\n", cpu);
 			else
 				/* trampoline code not run */
-				printk(KERN_ERR "Not responding.\n");
+				pr_err("CPU%d: Not responding.\n", cpu);
 			if (apic->inquire_remote_apic)
 				apic->inquire_remote_apic(apicid);
 		}
@@ -1293,14 +1308,16 @@ void native_cpu_die(unsigned int cpu)
 	for (i = 0; i < 10; i++) {
 		/* They ack this in play_dead by setting CPU_DEAD */
 		if (per_cpu(cpu_state, cpu) == CPU_DEAD) {
-			printk(KERN_INFO "CPU %d is now offline\n", cpu);
+			if (system_state == SYSTEM_RUNNING)
+				pr_info("CPU %u is now offline\n", cpu);
+
 			if (1 == num_online_cpus())
 				alternatives_smp_switch(0);
 			return;
 		}
 		msleep(100);
 	}
-	printk(KERN_ERR "CPU %u didn't die...\n", cpu);
+	pr_err("CPU %u didn't die...\n", cpu);
 }
 
 void play_dead_common(void)
diff --git a/arch/x86/kernel/trampoline.c b/arch/x86/kernel/trampoline.c
index cd02212..c652ef6 100644
--- a/arch/x86/kernel/trampoline.c
+++ b/arch/x86/kernel/trampoline.c
@@ -12,21 +12,19 @@
 #endif
 
 /* ready for x86_64 and x86 */
-unsigned char *__trampinitdata trampoline_base = __va(TRAMPOLINE_BASE);
+unsigned char *__trampinitdata trampoline_base;
 
 void __init reserve_trampoline_memory(void)
 {
-#ifdef CONFIG_X86_32
-	/*
-	 * But first pinch a few for the stack/trampoline stuff
-	 * FIXME: Don't need the extra page at 4K, but need to fix
-	 * trampoline before removing it. (see the GDT stuff)
-	 */
-	reserve_early(PAGE_SIZE, PAGE_SIZE + PAGE_SIZE, "EX TRAMPOLINE");
-#endif
+	unsigned long mem;
+
 	/* Has to be in very low memory so we can execute real-mode AP code. */
-	reserve_early(TRAMPOLINE_BASE, TRAMPOLINE_BASE + TRAMPOLINE_SIZE,
-			"TRAMPOLINE");
+	mem = find_e820_area(0, 1<<20, TRAMPOLINE_SIZE, PAGE_SIZE);
+	if (mem == -1L)
+		panic("Cannot allocate trampoline\n");
+
+	trampoline_base = __va(mem);
+	reserve_early(mem, mem + TRAMPOLINE_SIZE, "TRAMPOLINE");
 }
 
 /*
diff --git a/arch/x86/lib/msr.c b/arch/x86/lib/msr.c
index 41628b1..8728341 100644
--- a/arch/x86/lib/msr.c
+++ b/arch/x86/lib/msr.c
@@ -7,7 +7,6 @@ struct msr_info {
 	u32 msr_no;
 	struct msr reg;
 	struct msr *msrs;
-	int off;
 	int err;
 };
 
@@ -18,7 +17,7 @@ static void __rdmsr_on_cpu(void *info)
 	int this_cpu = raw_smp_processor_id();
 
 	if (rv->msrs)
-		reg = &rv->msrs[this_cpu - rv->off];
+		reg = per_cpu_ptr(rv->msrs, this_cpu);
 	else
 		reg = &rv->reg;
 
@@ -32,7 +31,7 @@ static void __wrmsr_on_cpu(void *info)
 	int this_cpu = raw_smp_processor_id();
 
 	if (rv->msrs)
-		reg = &rv->msrs[this_cpu - rv->off];
+		reg = per_cpu_ptr(rv->msrs, this_cpu);
 	else
 		reg = &rv->reg;
 
@@ -80,7 +79,6 @@ static void __rwmsr_on_cpus(const struct cpumask *mask, u32 msr_no,
 
 	memset(&rv, 0, sizeof(rv));
 
-	rv.off    = cpumask_first(mask);
 	rv.msrs	  = msrs;
 	rv.msr_no = msr_no;
 
@@ -120,6 +118,26 @@ void wrmsr_on_cpus(const struct cpumask *mask, u32 msr_no, struct msr *msrs)
 }
 EXPORT_SYMBOL(wrmsr_on_cpus);
 
+struct msr *msrs_alloc(void)
+{
+	struct msr *msrs = NULL;
+
+	msrs = alloc_percpu(struct msr);
+	if (!msrs) {
+		pr_warning("%s: error allocating msrs\n", __func__);
+		return NULL;
+	}
+
+	return msrs;
+}
+EXPORT_SYMBOL(msrs_alloc);
+
+void msrs_free(struct msr *msrs)
+{
+	free_percpu(msrs);
+}
+EXPORT_SYMBOL(msrs_free);
+
 /* These "safe" variants are slower and should be used when the target MSR
    may not actually exist. */
 static void __rdmsr_safe_on_cpu(void *info)
diff --git a/arch/x86/mm/mmio-mod.c b/arch/x86/mm/mmio-mod.c
index 4c765e9..34a3291 100644
--- a/arch/x86/mm/mmio-mod.c
+++ b/arch/x86/mm/mmio-mod.c
@@ -20,7 +20,7 @@
  * Derived from the read-mod example from relay-examples by Tom Zanussi.
  */
 
-#define pr_fmt(fmt) "mmiotrace: "
+#define pr_fmt(fmt) "mmiotrace: " fmt
 
 #define DEBUG 1