Merge branch 'linus' into x86/mce3

Conflicts:
	arch/x86/kernel/cpu/mcheck/mce_64.c
	arch/x86/kernel/irq.c

Merge reason: Resolve the conflicts above.

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

18 files changed, 1929 insertions, 461 deletions

diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile
index 4e242f9a06e4..3efcb2b96a15 100644
--- a/arch/x86/kernel/cpu/Makefile
+++ b/arch/x86/kernel/cpu/Makefile
@@ -1,5 +1,5 @@
 #
-# Makefile for x86-compatible CPU details and quirks
+# Makefile for x86-compatible CPU details, features and quirks
 #
 
 # Don't trace early stages of a secondary CPU boot
@@ -23,11 +23,13 @@ obj-$(CONFIG_CPU_SUP_CENTAUR)		+= centaur.o
 obj-$(CONFIG_CPU_SUP_TRANSMETA_32)	+= transmeta.o
 obj-$(CONFIG_CPU_SUP_UMC_32)		+= umc.o
 
-obj-$(CONFIG_X86_MCE)	+= mcheck/
-obj-$(CONFIG_MTRR)	+= mtrr/
-obj-$(CONFIG_CPU_FREQ)	+= cpufreq/
+obj-$(CONFIG_PERF_COUNTERS)		+= perf_counter.o
 
-obj-$(CONFIG_X86_LOCAL_APIC) += perfctr-watchdog.o
+obj-$(CONFIG_X86_MCE)			+= mcheck/
+obj-$(CONFIG_MTRR)			+= mtrr/
+obj-$(CONFIG_CPU_FREQ)			+= cpufreq/
+
+obj-$(CONFIG_X86_LOCAL_APIC)		+= perfctr-watchdog.o
 
 quiet_cmd_mkcapflags = MKCAP   $@
       cmd_mkcapflags = $(PERL) $(srctree)/$(src)/mkcapflags.pl $< $@
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index 728b3750a3e8..e5b27d8f1b47 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -6,6 +6,7 @@
 #include <asm/processor.h>
 #include <asm/apic.h>
 #include <asm/cpu.h>
+#include <asm/pci-direct.h>
 
 #ifdef CONFIG_X86_64
 # include <asm/numa_64.h>
@@ -351,6 +352,15 @@ static void __cpuinit early_init_amd(struct cpuinfo_x86 *c)
 		    (c->x86_model == 8 && c->x86_mask >= 8))
 			set_cpu_cap(c, X86_FEATURE_K6_MTRR);
 #endif
+#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_PCI)
+	/* check CPU config space for extended APIC ID */
+	if (c->x86 >= 0xf) {
+		unsigned int val;
+		val = read_pci_config(0, 24, 0, 0x68);
+		if ((val & ((1 << 17) | (1 << 18))) == ((1 << 17) | (1 << 18)))
+			set_cpu_cap(c, X86_FEATURE_EXTD_APICID);
+	}
+#endif
 }
 
 static void __cpuinit init_amd(struct cpuinfo_x86 *c)
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index b0517aa2bd3b..3ffdcfa9abdf 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -13,6 +13,7 @@
 #include <linux/io.h>
 
 #include <asm/stackprotector.h>
+#include <asm/perf_counter.h>
 #include <asm/mmu_context.h>
 #include <asm/hypervisor.h>
 #include <asm/processor.h>
@@ -874,6 +875,7 @@ void __init identify_boot_cpu(void)
 #else
 	vgetcpu_set_mode();
 #endif
+	init_hw_perf_counters();
 }
 
 void __cpuinit identify_secondary_cpu(struct cpuinfo_x86 *c)
diff --git a/arch/x86/kernel/cpu/cpu_debug.c b/arch/x86/kernel/cpu/cpu_debug.c
index 2fc4f6bb9ca5..6b2a52dd0403 100644
--- a/arch/x86/kernel/cpu/cpu_debug.c
+++ b/arch/x86/kernel/cpu/cpu_debug.c
@@ -32,9 +32,7 @@
 
 static DEFINE_PER_CPU(struct cpu_cpuX_base, cpu_arr[CPU_REG_ALL_BIT]);
 static DEFINE_PER_CPU(struct cpu_private *, priv_arr[MAX_CPU_FILES]);
-static DEFINE_PER_CPU(unsigned, cpu_modelflag);
 static DEFINE_PER_CPU(int, cpu_priv_count);
-static DEFINE_PER_CPU(unsigned, cpu_model);
 
 static DEFINE_MUTEX(cpu_debug_lock);
 
@@ -80,302 +78,102 @@ static struct cpu_file_base cpu_file[] = {
 	{ "value",	CPU_REG_ALL,	1	},
 };
 
-/* Intel Registers Range */
-static struct cpu_debug_range cpu_intel_range[] = {
-	{ 0x00000000, 0x00000001, CPU_MC,	CPU_INTEL_ALL		},
-	{ 0x00000006, 0x00000007, CPU_MONITOR,	CPU_CX_AT_XE		},
-	{ 0x00000010, 0x00000010, CPU_TIME,	CPU_INTEL_ALL		},
-	{ 0x00000011, 0x00000013, CPU_PMC,	CPU_INTEL_PENTIUM	},
-	{ 0x00000017, 0x00000017, CPU_PLATFORM,	CPU_PX_CX_AT_XE		},
-	{ 0x0000001B, 0x0000001B, CPU_APIC,	CPU_P6_CX_AT_XE		},
-
-	{ 0x0000002A, 0x0000002A, CPU_POWERON,	CPU_PX_CX_AT_XE		},
-	{ 0x0000002B, 0x0000002B, CPU_POWERON,	CPU_INTEL_XEON		},
-	{ 0x0000002C, 0x0000002C, CPU_FREQ,	CPU_INTEL_XEON		},
-	{ 0x0000003A, 0x0000003A, CPU_CONTROL,	CPU_CX_AT_XE		},
-
-	{ 0x00000040, 0x00000043, CPU_LBRANCH,	CPU_PM_CX_AT_XE		},
-	{ 0x00000044, 0x00000047, CPU_LBRANCH,	CPU_PM_CO_AT		},
-	{ 0x00000060, 0x00000063, CPU_LBRANCH,	CPU_C2_AT		},
-	{ 0x00000064, 0x00000067, CPU_LBRANCH,	CPU_INTEL_ATOM		},
-
-	{ 0x00000079, 0x00000079, CPU_BIOS,	CPU_P6_CX_AT_XE		},
-	{ 0x00000088, 0x0000008A, CPU_CACHE,	CPU_INTEL_P6		},
-	{ 0x0000008B, 0x0000008B, CPU_BIOS,	CPU_P6_CX_AT_XE		},
-	{ 0x0000009B, 0x0000009B, CPU_MONITOR,	CPU_INTEL_XEON		},
-
-	{ 0x000000C1, 0x000000C2, CPU_PMC,	CPU_P6_CX_AT		},
-	{ 0x000000CD, 0x000000CD, CPU_FREQ,	CPU_CX_AT		},
-	{ 0x000000E7, 0x000000E8, CPU_PERF,	CPU_CX_AT		},
-	{ 0x000000FE, 0x000000FE, CPU_MTRR,	CPU_P6_CX_XE		},
-
-	{ 0x00000116, 0x00000116, CPU_CACHE,	CPU_INTEL_P6		},
-	{ 0x00000118, 0x00000118, CPU_CACHE,	CPU_INTEL_P6		},
-	{ 0x00000119, 0x00000119, CPU_CACHE,	CPU_INTEL_PX		},
-	{ 0x0000011A, 0x0000011B, CPU_CACHE,	CPU_INTEL_P6		},
-	{ 0x0000011E, 0x0000011E, CPU_CACHE,	CPU_PX_CX_AT		},
-
-	{ 0x00000174, 0x00000176, CPU_SYSENTER,	CPU_P6_CX_AT_XE		},
-	{ 0x00000179, 0x0000017A, CPU_MC,	CPU_PX_CX_AT_XE		},
-	{ 0x0000017B, 0x0000017B, CPU_MC,	CPU_P6_XE		},
-	{ 0x00000186, 0x00000187, CPU_PMC,	CPU_P6_CX_AT		},
-	{ 0x00000198, 0x00000199, CPU_PERF,	CPU_PM_CX_AT_XE		},
-	{ 0x0000019A, 0x0000019A, CPU_TIME,	CPU_PM_CX_AT_XE		},
-	{ 0x0000019B, 0x0000019D, CPU_THERM,	CPU_PM_CX_AT_XE		},
-	{ 0x000001A0, 0x000001A0, CPU_MISC,	CPU_PM_CX_AT_XE		},
-
-	{ 0x000001C9, 0x000001C9, CPU_LBRANCH,	CPU_PM_CX_AT		},
-	{ 0x000001D7, 0x000001D8, CPU_LBRANCH,	CPU_INTEL_XEON		},
-	{ 0x000001D9, 0x000001D9, CPU_DEBUG,	CPU_CX_AT_XE		},
-	{ 0x000001DA, 0x000001DA, CPU_LBRANCH,	CPU_INTEL_XEON		},
-	{ 0x000001DB, 0x000001DB, CPU_LBRANCH,	CPU_P6_XE		},
-	{ 0x000001DC, 0x000001DC, CPU_LBRANCH,	CPU_INTEL_P6		},
-	{ 0x000001DD, 0x000001DE, CPU_LBRANCH,	CPU_PX_CX_AT_XE		},
-	{ 0x000001E0, 0x000001E0, CPU_LBRANCH,	CPU_INTEL_P6		},
-
-	{ 0x00000200, 0x0000020F, CPU_MTRR,	CPU_P6_CX_XE		},
-	{ 0x00000250, 0x00000250, CPU_MTRR,	CPU_P6_CX_XE		},
-	{ 0x00000258, 0x00000259, CPU_MTRR,	CPU_P6_CX_XE		},
-	{ 0x00000268, 0x0000026F, CPU_MTRR,	CPU_P6_CX_XE		},
-	{ 0x00000277, 0x00000277, CPU_PAT,	CPU_C2_AT_XE		},
-	{ 0x000002FF, 0x000002FF, CPU_MTRR,	CPU_P6_CX_XE		},
-
-	{ 0x00000300, 0x00000308, CPU_PMC,	CPU_INTEL_XEON		},
-	{ 0x00000309, 0x0000030B, CPU_PMC,	CPU_C2_AT_XE		},
-	{ 0x0000030C, 0x00000311, CPU_PMC,	CPU_INTEL_XEON		},
-	{ 0x00000345, 0x00000345, CPU_PMC,	CPU_C2_AT		},
-	{ 0x00000360, 0x00000371, CPU_PMC,	CPU_INTEL_XEON		},
-	{ 0x0000038D, 0x00000390, CPU_PMC,	CPU_C2_AT		},
-	{ 0x000003A0, 0x000003BE, CPU_PMC,	CPU_INTEL_XEON		},
-	{ 0x000003C0, 0x000003CD, CPU_PMC,	CPU_INTEL_XEON		},
-	{ 0x000003E0, 0x000003E1, CPU_PMC,	CPU_INTEL_XEON		},
-	{ 0x000003F0, 0x000003F0, CPU_PMC,	CPU_INTEL_XEON		},
-	{ 0x000003F1, 0x000003F1, CPU_PMC,	CPU_C2_AT_XE		},
-	{ 0x000003F2, 0x000003F2, CPU_PMC,	CPU_INTEL_XEON		},
-
-	{ 0x00000400, 0x00000402, CPU_MC,	CPU_PM_CX_AT_XE		},
-	{ 0x00000403, 0x00000403, CPU_MC,	CPU_INTEL_XEON		},
-	{ 0x00000404, 0x00000406, CPU_MC,	CPU_PM_CX_AT_XE		},
-	{ 0x00000407, 0x00000407, CPU_MC,	CPU_INTEL_XEON		},
-	{ 0x00000408, 0x0000040A, CPU_MC,	CPU_PM_CX_AT_XE		},
-	{ 0x0000040B, 0x0000040B, CPU_MC,	CPU_INTEL_XEON		},
-	{ 0x0000040C, 0x0000040E, CPU_MC,	CPU_PM_CX_XE		},
-	{ 0x0000040F, 0x0000040F, CPU_MC,	CPU_INTEL_XEON		},
-	{ 0x00000410, 0x00000412, CPU_MC,	CPU_PM_CX_AT_XE		},
-	{ 0x00000413, 0x00000417, CPU_MC,	CPU_CX_AT_XE		},
-	{ 0x00000480, 0x0000048B, CPU_VMX,	CPU_CX_AT_XE		},
-
-	{ 0x00000600, 0x00000600, CPU_DEBUG,	CPU_PM_CX_AT_XE		},
-	{ 0x00000680, 0x0000068F, CPU_LBRANCH,	CPU_INTEL_XEON		},
-	{ 0x000006C0, 0x000006CF, CPU_LBRANCH,	CPU_INTEL_XEON		},
-
-	{ 0x000107CC, 0x000107D3, CPU_PMC,	CPU_INTEL_XEON_MP	},
-
-	{ 0xC0000080, 0xC0000080, CPU_FEATURES,	CPU_INTEL_XEON		},
-	{ 0xC0000081, 0xC0000082, CPU_CALL,	CPU_INTEL_XEON		},
-	{ 0xC0000084, 0xC0000084, CPU_CALL,	CPU_INTEL_XEON		},
-	{ 0xC0000100, 0xC0000102, CPU_BASE,	CPU_INTEL_XEON		},
+/* CPU Registers Range */
+static struct cpu_debug_range cpu_reg_range[] = {
+	{ 0x00000000, 0x00000001, CPU_MC,	},
+	{ 0x00000006, 0x00000007, CPU_MONITOR,	},
+	{ 0x00000010, 0x00000010, CPU_TIME,	},
+	{ 0x00000011, 0x00000013, CPU_PMC,	},
+	{ 0x00000017, 0x00000017, CPU_PLATFORM,	},
+	{ 0x0000001B, 0x0000001B, CPU_APIC,	},
+	{ 0x0000002A, 0x0000002B, CPU_POWERON,	},
+	{ 0x0000002C, 0x0000002C, CPU_FREQ,	},
+	{ 0x0000003A, 0x0000003A, CPU_CONTROL,	},
+	{ 0x00000040, 0x00000047, CPU_LBRANCH,	},
+	{ 0x00000060, 0x00000067, CPU_LBRANCH,	},
+	{ 0x00000079, 0x00000079, CPU_BIOS,	},
+	{ 0x00000088, 0x0000008A, CPU_CACHE,	},
+	{ 0x0000008B, 0x0000008B, CPU_BIOS,	},
+	{ 0x0000009B, 0x0000009B, CPU_MONITOR,	},
+	{ 0x000000C1, 0x000000C4, CPU_PMC,	},
+	{ 0x000000CD, 0x000000CD, CPU_FREQ,	},
+	{ 0x000000E7, 0x000000E8, CPU_PERF,	},
+	{ 0x000000FE, 0x000000FE, CPU_MTRR,	},
+
+	{ 0x00000116, 0x0000011E, CPU_CACHE,	},
+	{ 0x00000174, 0x00000176, CPU_SYSENTER,	},
+	{ 0x00000179, 0x0000017B, CPU_MC,	},
+	{ 0x00000186, 0x00000189, CPU_PMC,	},
+	{ 0x00000198, 0x00000199, CPU_PERF,	},
+	{ 0x0000019A, 0x0000019A, CPU_TIME,	},
+	{ 0x0000019B, 0x0000019D, CPU_THERM,	},
+	{ 0x000001A0, 0x000001A0, CPU_MISC,	},
+	{ 0x000001C9, 0x000001C9, CPU_LBRANCH,	},
+	{ 0x000001D7, 0x000001D8, CPU_LBRANCH,	},
+	{ 0x000001D9, 0x000001D9, CPU_DEBUG,	},
+	{ 0x000001DA, 0x000001E0, CPU_LBRANCH,	},
+
+	{ 0x00000200, 0x0000020F, CPU_MTRR,	},
+	{ 0x00000250, 0x00000250, CPU_MTRR,	},
+	{ 0x00000258, 0x00000259, CPU_MTRR,	},
+	{ 0x00000268, 0x0000026F, CPU_MTRR,	},
+	{ 0x00000277, 0x00000277, CPU_PAT,	},
+	{ 0x000002FF, 0x000002FF, CPU_MTRR,	},
+
+	{ 0x00000300, 0x00000311, CPU_PMC,	},
+	{ 0x00000345, 0x00000345, CPU_PMC,	},
+	{ 0x00000360, 0x00000371, CPU_PMC,	},
+	{ 0x0000038D, 0x00000390, CPU_PMC,	},
+	{ 0x000003A0, 0x000003BE, CPU_PMC,	},
+	{ 0x000003C0, 0x000003CD, CPU_PMC,	},
+	{ 0x000003E0, 0x000003E1, CPU_PMC,	},
+	{ 0x000003F0, 0x000003F2, CPU_PMC,	},
+
+	{ 0x00000400, 0x00000417, CPU_MC,	},
+	{ 0x00000480, 0x0000048B, CPU_VMX,	},
+
+	{ 0x00000600, 0x00000600, CPU_DEBUG,	},
+	{ 0x00000680, 0x0000068F, CPU_LBRANCH,	},
+	{ 0x000006C0, 0x000006CF, CPU_LBRANCH,	},
+
+	{ 0x000107CC, 0x000107D3, CPU_PMC,	},
+
+	{ 0xC0000080, 0xC0000080, CPU_FEATURES,	},
+	{ 0xC0000081, 0xC0000084, CPU_CALL,	},
+	{ 0xC0000100, 0xC0000102, CPU_BASE,	},
+	{ 0xC0000103, 0xC0000103, CPU_TIME,	},
+
+	{ 0xC0010000, 0xC0010007, CPU_PMC,	},
+	{ 0xC0010010, 0xC0010010, CPU_CONF,	},
+	{ 0xC0010015, 0xC0010015, CPU_CONF,	},
+	{ 0xC0010016, 0xC001001A, CPU_MTRR,	},
+	{ 0xC001001D, 0xC001001D, CPU_MTRR,	},
+	{ 0xC001001F, 0xC001001F, CPU_CONF,	},
+	{ 0xC0010030, 0xC0010035, CPU_BIOS,	},
+	{ 0xC0010044, 0xC0010048, CPU_MC,	},
+	{ 0xC0010050, 0xC0010056, CPU_SMM,	},
+	{ 0xC0010058, 0xC0010058, CPU_CONF,	},
+	{ 0xC0010060, 0xC0010060, CPU_CACHE,	},
+	{ 0xC0010061, 0xC0010068, CPU_SMM,	},
+	{ 0xC0010069, 0xC001006B, CPU_SMM,	},
+	{ 0xC0010070, 0xC0010071, CPU_SMM,	},
+	{ 0xC0010111, 0xC0010113, CPU_SMM,	},
+	{ 0xC0010114, 0xC0010118, CPU_SVM,	},
+	{ 0xC0010140, 0xC0010141, CPU_OSVM,	},
+	{ 0xC0011022, 0xC0011023, CPU_CONF,	},
 };
 
-/* AMD Registers Range */
-static struct cpu_debug_range cpu_amd_range[] = {
-	{ 0x00000000, 0x00000001, CPU_MC,	CPU_K10_PLUS,		},
-	{ 0x00000010, 0x00000010, CPU_TIME,	CPU_K8_PLUS,		},
-	{ 0x0000001B, 0x0000001B, CPU_APIC,	CPU_K8_PLUS,		},
-	{ 0x0000002A, 0x0000002A, CPU_POWERON,	CPU_K7_PLUS		},
-	{ 0x0000008B, 0x0000008B, CPU_VER,	CPU_K8_PLUS		},
-	{ 0x000000FE, 0x000000FE, CPU_MTRR,	CPU_K8_PLUS,		},
-
-	{ 0x00000174, 0x00000176, CPU_SYSENTER,	CPU_K8_PLUS,		},
-	{ 0x00000179, 0x0000017B, CPU_MC,	CPU_K8_PLUS,		},
-	{ 0x000001D9, 0x000001D9, CPU_DEBUG,	CPU_K8_PLUS,		},
-	{ 0x000001DB, 0x000001DE, CPU_LBRANCH,	CPU_K8_PLUS,		},
-
-	{ 0x00000200, 0x0000020F, CPU_MTRR,	CPU_K8_PLUS,		},
-	{ 0x00000250, 0x00000250, CPU_MTRR,	CPU_K8_PLUS,		},
-	{ 0x00000258, 0x00000259, CPU_MTRR,	CPU_K8_PLUS,		},
-	{ 0x00000268, 0x0000026F, CPU_MTRR,	CPU_K8_PLUS,		},
-	{ 0x00000277, 0x00000277, CPU_PAT,	CPU_K8_PLUS,		},
-	{ 0x000002FF, 0x000002FF, CPU_MTRR,	CPU_K8_PLUS,		},
-
-	{ 0x00000400, 0x00000413, CPU_MC,	CPU_K8_PLUS,		},
-
-	{ 0xC0000080, 0xC0000080, CPU_FEATURES,	CPU_AMD_ALL,		},
-	{ 0xC0000081, 0xC0000084, CPU_CALL,	CPU_K8_PLUS,		},
-	{ 0xC0000100, 0xC0000102, CPU_BASE,	CPU_K8_PLUS,		},
-	{ 0xC0000103, 0xC0000103, CPU_TIME,	CPU_K10_PLUS,		},
-
-	{ 0xC0010000, 0xC0010007, CPU_PMC,	CPU_K8_PLUS,		},
-	{ 0xC0010010, 0xC0010010, CPU_CONF,	CPU_K7_PLUS,		},
-	{ 0xC0010015, 0xC0010015, CPU_CONF,	CPU_K7_PLUS,		},
-	{ 0xC0010016, 0xC001001A, CPU_MTRR,	CPU_K8_PLUS,		},
-	{ 0xC001001D, 0xC001001D, CPU_MTRR,	CPU_K8_PLUS,		},
-	{ 0xC001001F, 0xC001001F, CPU_CONF,	CPU_K8_PLUS,		},
-	{ 0xC0010030, 0xC0010035, CPU_BIOS,	CPU_K8_PLUS,		},
-	{ 0xC0010044, 0xC0010048, CPU_MC,	CPU_K8_PLUS,		},
-	{ 0xC0010050, 0xC0010056, CPU_SMM,	CPU_K0F_PLUS,		},
-	{ 0xC0010058, 0xC0010058, CPU_CONF,	CPU_K10_PLUS,		},
-	{ 0xC0010060, 0xC0010060, CPU_CACHE,	CPU_AMD_11,		},
-	{ 0xC0010061, 0xC0010068, CPU_SMM,	CPU_K10_PLUS,		},
-	{ 0xC0010069, 0xC001006B, CPU_SMM,	CPU_AMD_11,		},
-	{ 0xC0010070, 0xC0010071, CPU_SMM,	CPU_K10_PLUS,		},
-	{ 0xC0010111, 0xC0010113, CPU_SMM,	CPU_K8_PLUS,		},
-	{ 0xC0010114, 0xC0010118, CPU_SVM,	CPU_K10_PLUS,		},
-	{ 0xC0010140, 0xC0010141, CPU_OSVM,	CPU_K10_PLUS,		},
-	{ 0xC0011022, 0xC0011023, CPU_CONF,	CPU_K10_PLUS,		},
-};
-
-
-/* Intel */
-static int get_intel_modelflag(unsigned model)
-{
-	int flag;
-
-	switch (model) {
-	case 0x0501:
-	case 0x0502:
-	case 0x0504:
-		flag = CPU_INTEL_PENTIUM;
-		break;
-	case 0x0601:
-	case 0x0603:
-	case 0x0605:
-	case 0x0607:
-	case 0x0608:
-	case 0x060A:
-	case 0x060B:
-		flag = CPU_INTEL_P6;
-		break;
-	case 0x0609:
-	case 0x060D:
-		flag = CPU_INTEL_PENTIUM_M;
-		break;
-	case 0x060E:
-		flag = CPU_INTEL_CORE;
-		break;
-	case 0x060F:
-	case 0x0617:
-		flag = CPU_INTEL_CORE2;
-		break;
-	case 0x061C:
-		flag = CPU_INTEL_ATOM;
-		break;
-	case 0x0F00:
-	case 0x0F01:
-	case 0x0F02:
-	case 0x0F03:
-	case 0x0F04:
-		flag = CPU_INTEL_XEON_P4;
-		break;
-	case 0x0F06:
-		flag = CPU_INTEL_XEON_MP;
-		break;
-	default:
-		flag = CPU_NONE;
-		break;
-	}
-
-	return flag;
-}
-
-/* AMD */
-static int get_amd_modelflag(unsigned model)
-{
-	int flag;
-
-	switch (model >> 8) {
-	case 0x6:
-		flag = CPU_AMD_K6;
-		break;
-	case 0x7:
-		flag = CPU_AMD_K7;
-		break;
-	case 0x8:
-		flag = CPU_AMD_K8;
-		break;
-	case 0xf:
-		flag = CPU_AMD_0F;
-		break;
-	case 0x10:
-		flag = CPU_AMD_10;
-		break;
-	case 0x11:
-		flag = CPU_AMD_11;
-		break;
-	default:
-		flag = CPU_NONE;
-		break;
-	}
-
-	return flag;
-}
-
-static int get_cpu_modelflag(unsigned cpu)
-{
-	int flag;
-
-	flag = per_cpu(cpu_model, cpu);
-
-	switch (flag >> 16) {
-	case X86_VENDOR_INTEL:
-		flag = get_intel_modelflag(flag);
-		break;
-	case X86_VENDOR_AMD:
-		flag = get_amd_modelflag(flag & 0xffff);
-		break;
-	default:
-		flag = CPU_NONE;
-		break;
-	}
-
-	return flag;
-}
-
-static int get_cpu_range_count(unsigned cpu)
-{
-	int index;
-
-	switch (per_cpu(cpu_model, cpu) >> 16) {
-	case X86_VENDOR_INTEL:
-		index = ARRAY_SIZE(cpu_intel_range);
-		break;
-	case X86_VENDOR_AMD:
-		index = ARRAY_SIZE(cpu_amd_range);
-		break;
-	default:
-		index = 0;
-		break;
-	}
-
-	return index;
-}
-
 static int is_typeflag_valid(unsigned cpu, unsigned flag)
 {
-	unsigned vendor, modelflag;
-	int i, index;
+	int i;
 
 	/* Standard Registers should be always valid */
 	if (flag >= CPU_TSS)
 		return 1;
 
-	modelflag = per_cpu(cpu_modelflag, cpu);
-	vendor = per_cpu(cpu_model, cpu) >> 16;
-	index = get_cpu_range_count(cpu);
-
-	for (i = 0; i < index; i++) {
-		switch (vendor) {
-		case X86_VENDOR_INTEL:
-			if ((cpu_intel_range[i].model & modelflag) &&
-			    (cpu_intel_range[i].flag & flag))
-				return 1;
-			break;
-		case X86_VENDOR_AMD:
-			if ((cpu_amd_range[i].model & modelflag) &&
-			    (cpu_amd_range[i].flag & flag))
-				return 1;
-			break;
-		}
+	for (i = 0; i < ARRAY_SIZE(cpu_reg_range); i++) {
+		if (cpu_reg_range[i].flag == flag)
+			return 1;
 	}
 
 	/* Invalid */
@@ -385,26 +183,11 @@ static int is_typeflag_valid(unsigned cpu, unsigned flag)
 static unsigned get_cpu_range(unsigned cpu, unsigned *min, unsigned *max,
 			      int index, unsigned flag)
 {
-	unsigned modelflag;
-
-	modelflag = per_cpu(cpu_modelflag, cpu);
-	*max = 0;
-	switch (per_cpu(cpu_model, cpu) >> 16) {
-	case X86_VENDOR_INTEL:
-		if ((cpu_intel_range[index].model & modelflag) &&
-		    (cpu_intel_range[index].flag & flag)) {
-			*min = cpu_intel_range[index].min;
-			*max = cpu_intel_range[index].max;
-		}
-		break;
-	case X86_VENDOR_AMD:
-		if ((cpu_amd_range[index].model & modelflag) &&
-		    (cpu_amd_range[index].flag & flag)) {
-			*min = cpu_amd_range[index].min;
-			*max = cpu_amd_range[index].max;
-		}
-		break;
-	}
+	if (cpu_reg_range[index].flag == flag) {
+		*min = cpu_reg_range[index].min;
+		*max = cpu_reg_range[index].max;
+	} else
+		*max = 0;
 
 	return *max;
 }
@@ -434,7 +217,7 @@ static void print_msr(struct seq_file *seq, unsigned cpu, unsigned flag)
 	unsigned msr, msr_min, msr_max;
 	struct cpu_private *priv;
 	u32 low, high;
-	int i, range;
+	int i;
 
 	if (seq) {
 		priv = seq->private;
@@ -446,9 +229,7 @@ static void print_msr(struct seq_file *seq, unsigned cpu, unsigned flag)
 		}
 	}
 
-	range = get_cpu_range_count(cpu);
-
-	for (i = 0; i < range; i++) {
+	for (i = 0; i < ARRAY_SIZE(cpu_reg_range); i++) {
 		if (!get_cpu_range(cpu, &msr_min, &msr_max, i, flag))
 			continue;
 
@@ -800,13 +581,11 @@ static int cpu_init_msr(unsigned cpu, unsigned type, struct dentry *dentry)
 {
 	struct dentry *cpu_dentry = NULL;
 	unsigned reg, reg_min, reg_max;
-	int i, range, err = 0;
+	int i, err = 0;
 	char reg_dir[12];
 	u32 low, high;
 
-	range = get_cpu_range_count(cpu);
-
-	for (i = 0; i < range; i++) {
+	for (i = 0; i < ARRAY_SIZE(cpu_reg_range); i++) {
 		if (!get_cpu_range(cpu, &reg_min, &reg_max, i,
 				   cpu_base[type].flag))
 			continue;
@@ -862,10 +641,6 @@ static int cpu_init_cpu(void)
 		cpui = &cpu_data(cpu);
 		if (!cpu_has(cpui, X86_FEATURE_MSR))
 			continue;
-		per_cpu(cpu_model, cpu) = ((cpui->x86_vendor << 16) |
-					   (cpui->x86 << 8) |
-					   (cpui->x86_model));
-		per_cpu(cpu_modelflag, cpu) = get_cpu_modelflag(cpu);
 
 		sprintf(cpu_dir, "cpu%d", cpu);
 		cpu_dentry = debugfs_create_dir(cpu_dir, cpu_debugfs_dir);
diff --git a/arch/x86/kernel/cpu/cpufreq/Kconfig b/arch/x86/kernel/cpu/cpufreq/Kconfig
index 52c839875478..f138c6c389b9 100644
--- a/arch/x86/kernel/cpu/cpufreq/Kconfig
+++ b/arch/x86/kernel/cpu/cpufreq/Kconfig
@@ -220,11 +220,14 @@ config X86_LONGHAUL
 	  If in doubt, say N.
 
 config X86_E_POWERSAVER
-	tristate "VIA C7 Enhanced PowerSaver"
+	tristate "VIA C7 Enhanced PowerSaver (DANGEROUS)"
 	select CPU_FREQ_TABLE
-	depends on X86_32
+	depends on X86_32 && EXPERIMENTAL
 	help
-	  This adds the CPUFreq driver for VIA C7 processors.
+	  This adds the CPUFreq driver for VIA C7 processors.  However, this driver
+	  does not have any safeguards to prevent operating the CPU out of spec
+	  and is thus considered dangerous.  Please use the regular ACPI cpufreq
+	  driver, enabled by CONFIG_X86_ACPI_CPUFREQ.
 
 	  If in doubt, say N.
 
diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
index 54b6de2cd947..ae9b503220ca 100644
--- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
+++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
@@ -90,11 +90,7 @@ static int check_est_cpu(unsigned int cpuid)
 {
 	struct cpuinfo_x86 *cpu = &cpu_data(cpuid);
 
-	if (cpu->x86_vendor != X86_VENDOR_INTEL ||
-	    !cpu_has(cpu, X86_FEATURE_EST))
-		return 0;
-
-	return 1;
+	return cpu_has(cpu, X86_FEATURE_EST);
 }
 
 static unsigned extract_io(u32 value, struct acpi_cpufreq_data *data)
@@ -550,7 +546,7 @@ static int __init acpi_cpufreq_early_init(void)
 		return -ENOMEM;
 	}
 	for_each_possible_cpu(i) {
-		if (!alloc_cpumask_var_node(
+		if (!zalloc_cpumask_var_node(
 			&per_cpu_ptr(acpi_perf_data, i)->shared_cpu_map,
 			GFP_KERNEL, cpu_to_node(i))) {
 
diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k7.c b/arch/x86/kernel/cpu/cpufreq/powernow-k7.c
index a8363e5be4ef..d47c775eb0ab 100644
--- a/arch/x86/kernel/cpu/cpufreq/powernow-k7.c
+++ b/arch/x86/kernel/cpu/cpufreq/powernow-k7.c
@@ -322,7 +322,7 @@ static int powernow_acpi_init(void)
 		goto err0;
 	}
 
-	if (!alloc_cpumask_var(&acpi_processor_perf->shared_cpu_map,
+	if (!zalloc_cpumask_var(&acpi_processor_perf->shared_cpu_map,
 								GFP_KERNEL)) {
 		retval = -ENOMEM;
 		goto err05;
diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
index f6b32d112357..cf52215d9eb1 100644
--- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
+++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
@@ -835,7 +835,7 @@ static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data)
 {
 	struct cpufreq_frequency_table *powernow_table;
 	int ret_val = -ENODEV;
-	acpi_integer space_id;
+	acpi_integer control, status;
 
 	if (acpi_processor_register_performance(&data->acpi_data, data->cpu)) {
 		dprintk("register performance failed: bad ACPI data\n");
@@ -848,12 +848,13 @@ static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data)
 		goto err_out;
 	}
 
-	space_id = data->acpi_data.control_register.space_id;
-	if ((space_id != ACPI_ADR_SPACE_FIXED_HARDWARE) ||
-		(space_id != ACPI_ADR_SPACE_FIXED_HARDWARE)) {
+	control = data->acpi_data.control_register.space_id;
+	status = data->acpi_data.status_register.space_id;
+
+	if ((control != ACPI_ADR_SPACE_FIXED_HARDWARE) ||
+	    (status != ACPI_ADR_SPACE_FIXED_HARDWARE)) {
 		dprintk("Invalid control/status registers (%x - %x)\n",
-			data->acpi_data.control_register.space_id,
-			space_id);
+			control, status);
 		goto err_out;
 	}
 
@@ -886,7 +887,7 @@ static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data)
 	/* notify BIOS that we exist */
 	acpi_processor_notify_smm(THIS_MODULE);
 
-	if (!alloc_cpumask_var(&data->acpi_data.shared_cpu_map, GFP_KERNEL)) {
+	if (!zalloc_cpumask_var(&data->acpi_data.shared_cpu_map, GFP_KERNEL)) {
 		printk(KERN_ERR PFX
 				"unable to alloc powernow_k8_data cpumask\n");
 		ret_val = -ENOMEM;
diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
index c9f1fdc02830..55c831ed71ce 100644
--- a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
+++ b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
@@ -471,7 +471,7 @@ static int centrino_target (struct cpufreq_policy *policy,
 
 	if (unlikely(!alloc_cpumask_var(&saved_mask, GFP_KERNEL)))
 		return -ENOMEM;
-	if (unlikely(!alloc_cpumask_var(&covered_cpus, GFP_KERNEL))) {
+	if (unlikely(!zalloc_cpumask_var(&covered_cpus, GFP_KERNEL))) {
 		free_cpumask_var(saved_mask);
 		return -ENOMEM;
 	}
diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c
index 483eda96e102..789efe217e1a 100644
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -17,6 +17,7 @@
 
 #include <asm/processor.h>
 #include <asm/smp.h>
+#include <asm/k8.h>
 
 #define LVL_1_INST	1
 #define LVL_1_DATA	2
@@ -159,14 +160,6 @@ struct _cpuid4_info_regs {
 	unsigned long can_disable;
 };
 
-#if defined(CONFIG_PCI) && defined(CONFIG_SYSFS)
-static struct pci_device_id k8_nb_id[] = {
-	{ PCI_DEVICE(PCI_VENDOR_ID_AMD, 0x1103) },
-	{ PCI_DEVICE(PCI_VENDOR_ID_AMD, 0x1203) },
-	{}
-};
-#endif
-
 unsigned short			num_cache_leaves;
 
 /* AMD doesn't have CPUID4. Emulate it here to report the same
@@ -207,10 +200,17 @@ union l3_cache {
 };
 
 static const unsigned short __cpuinitconst assocs[] = {
-	[1] = 1, [2] = 2, [4] = 4, [6] = 8,
-	[8] = 16, [0xa] = 32, [0xb] = 48,
+	[1] = 1,
+	[2] = 2,
+	[4] = 4,
+	[6] = 8,
+	[8] = 16,
+	[0xa] = 32,
+	[0xb] = 48,
 	[0xc] = 64,
-	[0xf] = 0xffff // ??
+	[0xd] = 96,
+	[0xe] = 128,
+	[0xf] = 0xffff /* fully associative - no way to show this currently */
 };
 
 static const unsigned char __cpuinitconst levels[] = { 1, 1, 2, 3 };
@@ -271,7 +271,8 @@ amd_cpuid4(int leaf, union _cpuid4_leaf_eax *eax,
 	eax->split.type = types[leaf];
 	eax->split.level = levels[leaf];
 	if (leaf == 3)
-		eax->split.num_threads_sharing = current_cpu_data.x86_max_cores - 1;
+		eax->split.num_threads_sharing =
+			current_cpu_data.x86_max_cores - 1;
 	else
 		eax->split.num_threads_sharing = 0;
 	eax->split.num_cores_on_die = current_cpu_data.x86_max_cores - 1;
@@ -291,6 +292,14 @@ amd_check_l3_disable(int index, struct _cpuid4_info_regs *this_leaf)
 {
 	if (index < 3)
 		return;
+
+	if (boot_cpu_data.x86 == 0x11)
+		return;
+
+	/* see erratum #382 */
+	if ((boot_cpu_data.x86 == 0x10) && (boot_cpu_data.x86_model < 0x8))
+		return;
+
 	this_leaf->can_disable = 1;
 }
 
@@ -696,97 +705,75 @@ static ssize_t show_type(struct _cpuid4_info *this_leaf, char *buf)
 #define to_object(k)	container_of(k, struct _index_kobject, kobj)
 #define to_attr(a)	container_of(a, struct _cache_attr, attr)
 
-#ifdef CONFIG_PCI
-static struct pci_dev *get_k8_northbridge(int node)
-{
-	struct pci_dev *dev = NULL;
-	int i;
-
-	for (i = 0; i <= node; i++) {
-		do {
-			dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev);
-			if (!dev)
-				break;
-		} while (!pci_match_id(&k8_nb_id[0], dev));
-		if (!dev)
-			break;
-	}
-	return dev;
-}
-#else
-static struct pci_dev *get_k8_northbridge(int node)
-{
-	return NULL;
-}
-#endif
-
-static ssize_t show_cache_disable(struct _cpuid4_info *this_leaf, char *buf)
+static ssize_t show_cache_disable(struct _cpuid4_info *this_leaf, char *buf,
+				  unsigned int index)
 {
-	const struct cpumask *mask = to_cpumask(this_leaf->shared_cpu_map);
-	int node = cpu_to_node(cpumask_first(mask));
-	struct pci_dev *dev = NULL;
-	ssize_t ret = 0;
-	int i;
+	int cpu = cpumask_first(to_cpumask(this_leaf->shared_cpu_map));
+	int node = cpu_to_node(cpu);
+	struct pci_dev *dev = node_to_k8_nb_misc(node);
+	unsigned int reg = 0;
 
 	if (!this_leaf->can_disable)
-		return sprintf(buf, "Feature not enabled\n");
-
-	dev = get_k8_northbridge(node);
-	if (!dev) {
-		printk(KERN_ERR "Attempting AMD northbridge operation on a system with no northbridge\n");
 		return -EINVAL;
-	}
 
-	for (i = 0; i < 2; i++) {
-		unsigned int reg;
+	if (!dev)
+		return -EINVAL;
 
-		pci_read_config_dword(dev, 0x1BC + i * 4, &reg);
+	pci_read_config_dword(dev, 0x1BC + index * 4, &reg);
+	return sprintf(buf, "%x\n", reg);
+}
 
-		ret += sprintf(buf, "%sEntry: %d\n", buf, i);
-		ret += sprintf(buf, "%sReads:  %s\tNew Entries: %s\n",  
-			buf,
-			reg & 0x80000000 ? "Disabled" : "Allowed",
-			reg & 0x40000000 ? "Disabled" : "Allowed");
-		ret += sprintf(buf, "%sSubCache: %x\tIndex: %x\n",
-			buf, (reg & 0x30000) >> 16, reg & 0xfff);
-	}
-	return ret;
+#define SHOW_CACHE_DISABLE(index)					\
+static ssize_t								\
+show_cache_disable_##index(struct _cpuid4_info *this_leaf, char *buf)  	\
+{									\
+	return show_cache_disable(this_leaf, buf, index);		\
 }
+SHOW_CACHE_DISABLE(0)
+SHOW_CACHE_DISABLE(1)
 
-static ssize_t
-store_cache_disable(struct _cpuid4_info *this_leaf, const char *buf,
-		    size_t count)
+static ssize_t store_cache_disable(struct _cpuid4_info *this_leaf,
+	const char *buf, size_t count, unsigned int index)
 {
-	const struct cpumask *mask = to_cpumask(this_leaf->shared_cpu_map);
-	int node = cpu_to_node(cpumask_first(mask));
-	struct pci_dev *dev = NULL;
-	unsigned int ret, index, val;
+	int cpu = cpumask_first(to_cpumask(this_leaf->shared_cpu_map));
+	int node = cpu_to_node(cpu);
+	struct pci_dev *dev = node_to_k8_nb_misc(node);
+	unsigned long val = 0;
+	unsigned int scrubber = 0;
 
 	if (!this_leaf->can_disable)
-		return 0;
-
-	if (strlen(buf) > 15)
 		return -EINVAL;
 
-	ret = sscanf(buf, "%x %x", &index, &val);
-	if (ret != 2)
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	if (!dev)
 		return -EINVAL;
-	if (index > 1)
+
+	if (strict_strtoul(buf, 10, &val) < 0)
 		return -EINVAL;
 
 	val |= 0xc0000000;
-	dev = get_k8_northbridge(node);
-	if (!dev) {
-		printk(KERN_ERR "Attempting AMD northbridge operation on a system with no northbridge\n");
-		return -EINVAL;
-	}
+
+	pci_read_config_dword(dev, 0x58, &scrubber);
+	scrubber &= ~0x1f000000;
+	pci_write_config_dword(dev, 0x58, scrubber);
 
 	pci_write_config_dword(dev, 0x1BC + index * 4, val & ~0x40000000);
 	wbinvd();
 	pci_write_config_dword(dev, 0x1BC + index * 4, val);
+	return count;
+}
 
-	return 1;
+#define STORE_CACHE_DISABLE(index)					\
+static ssize_t								\
+store_cache_disable_##index(struct _cpuid4_info *this_leaf,	     	\
+			    const char *buf, size_t count)		\
+{									\
+	return store_cache_disable(this_leaf, buf, count, index);	\
 }
+STORE_CACHE_DISABLE(0)
+STORE_CACHE_DISABLE(1)
 
 struct _cache_attr {
 	struct attribute attr;
@@ -808,7 +795,10 @@ define_one_ro(size);
 define_one_ro(shared_cpu_map);
 define_one_ro(shared_cpu_list);
 
-static struct _cache_attr cache_disable = __ATTR(cache_disable, 0644, show_cache_disable, store_cache_disable);
+static struct _cache_attr cache_disable_0 = __ATTR(cache_disable_0, 0644,
+		show_cache_disable_0, store_cache_disable_0);
+static struct _cache_attr cache_disable_1 = __ATTR(cache_disable_1, 0644,
+		show_cache_disable_1, store_cache_disable_1);
 
 static struct attribute * default_attrs[] = {
 	&type.attr,
@@ -820,7 +810,8 @@ static struct attribute * default_attrs[] = {
 	&size.attr,
 	&shared_cpu_map.attr,
 	&shared_cpu_list.attr,
-	&cache_disable.attr,
+	&cache_disable_0.attr,
+	&cache_disable_1.attr,
 	NULL
 };
 
diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c b/arch/x86/kernel/cpu/mcheck/mce_intel_64.c
index 046087e9808f..f2ef6952c400 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_intel_64.c
@@ -15,7 +15,6 @@
 #include <asm/hw_irq.h>
 #include <asm/idle.h>
 #include <asm/therm_throt.h>
-#include <asm/apic.h>
 
 #include "mce.h"
 
diff --git a/arch/x86/kernel/cpu/mtrr/cleanup.c b/arch/x86/kernel/cpu/mtrr/cleanup.c
index ce0fe4b5c04f..1d584a18a50d 100644
--- a/arch/x86/kernel/cpu/mtrr/cleanup.c
+++ b/arch/x86/kernel/cpu/mtrr/cleanup.c
@@ -808,7 +808,7 @@ int __init mtrr_cleanup(unsigned address_bits)
 
 	if (!is_cpu(INTEL) || enable_mtrr_cleanup < 1)
 		return 0;
-	rdmsr(MTRRdefType_MSR, def, dummy);
+	rdmsr(MSR_MTRRdefType, def, dummy);
 	def &= 0xff;
 	if (def != MTRR_TYPE_UNCACHABLE)
 		return 0;
@@ -1003,7 +1003,7 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn)
 	 */
 	if (!is_cpu(INTEL) || disable_mtrr_trim)
 		return 0;
-	rdmsr(MTRRdefType_MSR, def, dummy);
+	rdmsr(MSR_MTRRdefType, def, dummy);
 	def &= 0xff;
 	if (def != MTRR_TYPE_UNCACHABLE)
 		return 0;
diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c
index d21d4fb161f7..0543f69f0b27 100644
--- a/arch/x86/kernel/cpu/mtrr/generic.c
+++ b/arch/x86/kernel/cpu/mtrr/generic.c
@@ -20,9 +20,9 @@ struct fixed_range_block {
 };
 
 static struct fixed_range_block fixed_range_blocks[] = {
-	{ MTRRfix64K_00000_MSR, 1 }, /* one  64k MTRR  */
-	{ MTRRfix16K_80000_MSR, 2 }, /* two  16k MTRRs */
-	{ MTRRfix4K_C0000_MSR,  8 }, /* eight 4k MTRRs */
+	{ MSR_MTRRfix64K_00000, 1 }, /* one  64k MTRR  */
+	{ MSR_MTRRfix16K_80000, 2 }, /* two  16k MTRRs */
+	{ MSR_MTRRfix4K_C0000,  8 }, /* eight 4k MTRRs */
 	{}
 };
 
@@ -194,12 +194,12 @@ get_fixed_ranges(mtrr_type * frs)
 
 	k8_check_syscfg_dram_mod_en();
 
-	rdmsr(MTRRfix64K_00000_MSR, p[0], p[1]);
+	rdmsr(MSR_MTRRfix64K_00000, p[0], p[1]);
 
 	for (i = 0; i < 2; i++)
-		rdmsr(MTRRfix16K_80000_MSR + i, p[2 + i * 2], p[3 + i * 2]);
+		rdmsr(MSR_MTRRfix16K_80000 + i, p[2 + i * 2], p[3 + i * 2]);
 	for (i = 0; i < 8; i++)
-		rdmsr(MTRRfix4K_C0000_MSR + i, p[6 + i * 2], p[7 + i * 2]);
+		rdmsr(MSR_MTRRfix4K_C0000 + i, p[6 + i * 2], p[7 + i * 2]);
 }
 
 void mtrr_save_fixed_ranges(void *info)
@@ -310,7 +310,7 @@ void __init get_mtrr_state(void)
 
 	vrs = mtrr_state.var_ranges;
 
-	rdmsr(MTRRcap_MSR, lo, dummy);
+	rdmsr(MSR_MTRRcap, lo, dummy);
 	mtrr_state.have_fixed = (lo >> 8) & 1;
 
 	for (i = 0; i < num_var_ranges; i++)
@@ -318,7 +318,7 @@ void __init get_mtrr_state(void)
 	if (mtrr_state.have_fixed)
 		get_fixed_ranges(mtrr_state.fixed_ranges);
 
-	rdmsr(MTRRdefType_MSR, lo, dummy);
+	rdmsr(MSR_MTRRdefType, lo, dummy);
 	mtrr_state.def_type = (lo & 0xff);
 	mtrr_state.enabled = (lo & 0xc00) >> 10;
 
@@ -583,10 +583,10 @@ static void prepare_set(void) __acquires(set_atomicity_lock)
 	__flush_tlb();
 
 	/*  Save MTRR state */
-	rdmsr(MTRRdefType_MSR, deftype_lo, deftype_hi);
+	rdmsr(MSR_MTRRdefType, deftype_lo, deftype_hi);
 
 	/*  Disable MTRRs, and set the default type to uncached  */
-	mtrr_wrmsr(MTRRdefType_MSR, deftype_lo & ~0xcff, deftype_hi);
+	mtrr_wrmsr(MSR_MTRRdefType, deftype_lo & ~0xcff, deftype_hi);
 }
 
 static void post_set(void) __releases(set_atomicity_lock)
@@ -595,7 +595,7 @@ static void post_set(void) __releases(set_atomicity_lock)
 	__flush_tlb();
 
 	/* Intel (P6) standard MTRRs */
-	mtrr_wrmsr(MTRRdefType_MSR, deftype_lo, deftype_hi);
+	mtrr_wrmsr(MSR_MTRRdefType, deftype_lo, deftype_hi);
 		
 	/*  Enable caches  */
 	write_cr0(read_cr0() & 0xbfffffff);
@@ -707,7 +707,7 @@ int generic_validate_add_page(unsigned long base, unsigned long size, unsigned i
 static int generic_have_wrcomb(void)
 {
 	unsigned long config, dummy;
-	rdmsr(MTRRcap_MSR, config, dummy);
+	rdmsr(MSR_MTRRcap, config, dummy);
 	return (config & (1 << 10));
 }
 
diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c
index 03cda01f57c7..8fc248b5aeaf 100644
--- a/arch/x86/kernel/cpu/mtrr/main.c
+++ b/arch/x86/kernel/cpu/mtrr/main.c
@@ -104,7 +104,7 @@ static void __init set_num_var_ranges(void)
 	unsigned long config = 0, dummy;
 
 	if (use_intel()) {
-		rdmsr(MTRRcap_MSR, config, dummy);
+		rdmsr(MSR_MTRRcap, config, dummy);
 	} else if (is_cpu(AMD))
 		config = 2;
 	else if (is_cpu(CYRIX) || is_cpu(CENTAUR))
diff --git a/arch/x86/kernel/cpu/mtrr/mtrr.h b/arch/x86/kernel/cpu/mtrr/mtrr.h
index 77f67f7b347a..7538b767f206 100644
--- a/arch/x86/kernel/cpu/mtrr/mtrr.h
+++ b/arch/x86/kernel/cpu/mtrr/mtrr.h
@@ -5,21 +5,6 @@
 #include <linux/types.h>
 #include <linux/stddef.h>
 
-#define MTRRcap_MSR     0x0fe
-#define MTRRdefType_MSR 0x2ff
-
-#define MTRRfix64K_00000_MSR 0x250
-#define MTRRfix16K_80000_MSR 0x258
-#define MTRRfix16K_A0000_MSR 0x259
-#define MTRRfix4K_C0000_MSR 0x268
-#define MTRRfix4K_C8000_MSR 0x269
-#define MTRRfix4K_D0000_MSR 0x26a
-#define MTRRfix4K_D8000_MSR 0x26b
-#define MTRRfix4K_E0000_MSR 0x26c
-#define MTRRfix4K_E8000_MSR 0x26d
-#define MTRRfix4K_F0000_MSR 0x26e
-#define MTRRfix4K_F8000_MSR 0x26f
-
 #define MTRR_CHANGE_MASK_FIXED     0x01
 #define MTRR_CHANGE_MASK_VARIABLE  0x02
 #define MTRR_CHANGE_MASK_DEFTYPE   0x04
diff --git a/arch/x86/kernel/cpu/mtrr/state.c b/arch/x86/kernel/cpu/mtrr/state.c
index 7f7e2753685b..1f5fb1588d1f 100644
--- a/arch/x86/kernel/cpu/mtrr/state.c
+++ b/arch/x86/kernel/cpu/mtrr/state.c
@@ -35,7 +35,7 @@ void set_mtrr_prepare_save(struct set_mtrr_context *ctxt)
 
 		if (use_intel())
 			/*  Save MTRR state */
-			rdmsr(MTRRdefType_MSR, ctxt->deftype_lo, ctxt->deftype_hi);
+			rdmsr(MSR_MTRRdefType, ctxt->deftype_lo, ctxt->deftype_hi);
 		else
 			/* Cyrix ARRs - everything else were excluded at the top */
 			ctxt->ccr3 = getCx86(CX86_CCR3);
@@ -46,7 +46,7 @@ void set_mtrr_cache_disable(struct set_mtrr_context *ctxt)
 {
 	if (use_intel())
 		/*  Disable MTRRs, and set the default type to uncached  */
-		mtrr_wrmsr(MTRRdefType_MSR, ctxt->deftype_lo & 0xf300UL,
+		mtrr_wrmsr(MSR_MTRRdefType, ctxt->deftype_lo & 0xf300UL,
 		      ctxt->deftype_hi);
 	else if (is_cpu(CYRIX))
 		/* Cyrix ARRs - everything else were excluded at the top */
@@ -64,7 +64,7 @@ void set_mtrr_done(struct set_mtrr_context *ctxt)
 		/*  Restore MTRRdefType  */
 		if (use_intel())
 			/* Intel (P6) standard MTRRs */
-			mtrr_wrmsr(MTRRdefType_MSR, ctxt->deftype_lo, ctxt->deftype_hi);
+			mtrr_wrmsr(MSR_MTRRdefType, ctxt->deftype_lo, ctxt->deftype_hi);
 		else
 			/* Cyrix ARRs - everything else was excluded at the top */
 			setCx86(CX86_CCR3, ctxt->ccr3);
diff --git a/arch/x86/kernel/cpu/perf_counter.c b/arch/x86/kernel/cpu/perf_counter.c
new file mode 100644
index 000000000000..895c82e78455
--- /dev/null
+++ b/arch/x86/kernel/cpu/perf_counter.c
@@ -0,0 +1,1704 @@
+/*
+ * Performance counter x86 architecture code
+ *
+ *  Copyright (C) 2008 Thomas Gleixner <tglx@linutronix.de>
+ *  Copyright (C) 2008-2009 Red Hat, Inc., Ingo Molnar
+ *  Copyright (C) 2009 Jaswinder Singh Rajput
+ *  Copyright (C) 2009 Advanced Micro Devices, Inc., Robert Richter
+ *  Copyright (C) 2008-2009 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com>
+ *
+ *  For licencing details see kernel-base/COPYING
+ */
+
+#include <linux/perf_counter.h>
+#include <linux/capability.h>
+#include <linux/notifier.h>
+#include <linux/hardirq.h>
+#include <linux/kprobes.h>
+#include <linux/module.h>
+#include <linux/kdebug.h>
+#include <linux/sched.h>
+#include <linux/uaccess.h>
+
+#include <asm/apic.h>
+#include <asm/stacktrace.h>
+#include <asm/nmi.h>
+
+static u64 perf_counter_mask __read_mostly;
+
+struct cpu_hw_counters {
+	struct perf_counter	*counters[X86_PMC_IDX_MAX];
+	unsigned long		used_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
+	unsigned long		active_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
+	unsigned long		interrupts;
+	int			enabled;
+};
+
+/*
+ * struct x86_pmu - generic x86 pmu
+ */
+struct x86_pmu {
+	const char	*name;
+	int		version;
+	int		(*handle_irq)(struct pt_regs *);
+	void		(*disable_all)(void);
+	void		(*enable_all)(void);
+	void		(*enable)(struct hw_perf_counter *, int);
+	void		(*disable)(struct hw_perf_counter *, int);
+	unsigned	eventsel;
+	unsigned	perfctr;
+	u64		(*event_map)(int);
+	u64		(*raw_event)(u64);
+	int		max_events;
+	int		num_counters;
+	int		num_counters_fixed;
+	int		counter_bits;
+	u64		counter_mask;
+	u64		max_period;
+	u64		intel_ctrl;
+};
+
+static struct x86_pmu x86_pmu __read_mostly;
+
+static DEFINE_PER_CPU(struct cpu_hw_counters, cpu_hw_counters) = {
+	.enabled = 1,
+};
+
+/*
+ * Intel PerfMon v3. Used on Core2 and later.
+ */
+static const u64 intel_perfmon_event_map[] =
+{
+  [PERF_COUNT_HW_CPU_CYCLES]		= 0x003c,
+  [PERF_COUNT_HW_INSTRUCTIONS]		= 0x00c0,
+  [PERF_COUNT_HW_CACHE_REFERENCES]	= 0x4f2e,
+  [PERF_COUNT_HW_CACHE_MISSES]		= 0x412e,
+  [PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= 0x00c4,
+  [PERF_COUNT_HW_BRANCH_MISSES]		= 0x00c5,
+  [PERF_COUNT_HW_BUS_CYCLES]		= 0x013c,
+};
+
+static u64 intel_pmu_event_map(int event)
+{
+	return intel_perfmon_event_map[event];
+}
+
+/*
+ * Generalized hw caching related event table, filled
+ * in on a per model basis. A value of 0 means
+ * 'not supported', -1 means 'event makes no sense on
+ * this CPU', any other value means the raw event
+ * ID.
+ */
+
+#define C(x) PERF_COUNT_HW_CACHE_##x
+
+static u64 __read_mostly hw_cache_event_ids
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX];
+
+static const u64 nehalem_hw_cache_event_ids
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] =
+{
+ [ C(L1D) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0f40, /* L1D_CACHE_LD.MESI            */
+		[ C(RESULT_MISS)   ] = 0x0140, /* L1D_CACHE_LD.I_STATE         */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0f41, /* L1D_CACHE_ST.MESI            */
+		[ C(RESULT_MISS)   ] = 0x0141, /* L1D_CACHE_ST.I_STATE         */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x014e, /* L1D_PREFETCH.REQUESTS        */
+		[ C(RESULT_MISS)   ] = 0x024e, /* L1D_PREFETCH.MISS            */
+	},
+ },
+ [ C(L1I ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0380, /* L1I.READS                    */
+		[ C(RESULT_MISS)   ] = 0x0280, /* L1I.MISSES                   */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(LL  ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0324, /* L2_RQSTS.LOADS               */
+		[ C(RESULT_MISS)   ] = 0x0224, /* L2_RQSTS.LD_MISS             */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0c24, /* L2_RQSTS.RFOS                */
+		[ C(RESULT_MISS)   ] = 0x0824, /* L2_RQSTS.RFO_MISS            */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x4f2e, /* LLC Reference                */
+		[ C(RESULT_MISS)   ] = 0x412e, /* LLC Misses                   */
+	},
+ },
+ [ C(DTLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0f40, /* L1D_CACHE_LD.MESI   (alias)  */
+		[ C(RESULT_MISS)   ] = 0x0108, /* DTLB_LOAD_MISSES.ANY         */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0f41, /* L1D_CACHE_ST.MESI   (alias)  */
+		[ C(RESULT_MISS)   ] = 0x010c, /* MEM_STORE_RETIRED.DTLB_MISS  */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(ITLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x01c0, /* INST_RETIRED.ANY_P           */
+		[ C(RESULT_MISS)   ] = 0x20c8, /* ITLB_MISS_RETIRED            */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+ [ C(BPU ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x00c4, /* BR_INST_RETIRED.ALL_BRANCHES */
+		[ C(RESULT_MISS)   ] = 0x03e8, /* BPU_CLEARS.ANY               */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+};
+
+static const u64 core2_hw_cache_event_ids
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] =
+{
+ [ C(L1D) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0f40, /* L1D_CACHE_LD.MESI          */
+		[ C(RESULT_MISS)   ] = 0x0140, /* L1D_CACHE_LD.I_STATE       */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0f41, /* L1D_CACHE_ST.MESI          */
+		[ C(RESULT_MISS)   ] = 0x0141, /* L1D_CACHE_ST.I_STATE       */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x104e, /* L1D_PREFETCH.REQUESTS      */
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(L1I ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0080, /* L1I.READS                  */
+		[ C(RESULT_MISS)   ] = 0x0081, /* L1I.MISSES                 */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(LL  ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x4f29, /* L2_LD.MESI                 */
+		[ C(RESULT_MISS)   ] = 0x4129, /* L2_LD.ISTATE               */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x4f2A, /* L2_ST.MESI                 */
+		[ C(RESULT_MISS)   ] = 0x412A, /* L2_ST.ISTATE               */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(DTLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0f40, /* L1D_CACHE_LD.MESI  (alias) */
+		[ C(RESULT_MISS)   ] = 0x0208, /* DTLB_MISSES.MISS_LD        */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0f41, /* L1D_CACHE_ST.MESI  (alias) */
+		[ C(RESULT_MISS)   ] = 0x0808, /* DTLB_MISSES.MISS_ST        */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(ITLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x00c0, /* INST_RETIRED.ANY_P         */
+		[ C(RESULT_MISS)   ] = 0x1282, /* ITLBMISSES                 */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+ [ C(BPU ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x00c4, /* BR_INST_RETIRED.ANY        */
+		[ C(RESULT_MISS)   ] = 0x00c5, /* BP_INST_RETIRED.MISPRED    */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+};
+
+static const u64 atom_hw_cache_event_ids
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] =
+{
+ [ C(L1D) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x2140, /* L1D_CACHE.LD               */
+		[ C(RESULT_MISS)   ] = 0,
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x2240, /* L1D_CACHE.ST               */
+		[ C(RESULT_MISS)   ] = 0,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(L1I ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0380, /* L1I.READS                  */
+		[ C(RESULT_MISS)   ] = 0x0280, /* L1I.MISSES                 */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(LL  ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x4f29, /* L2_LD.MESI                 */
+		[ C(RESULT_MISS)   ] = 0x4129, /* L2_LD.ISTATE               */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x4f2A, /* L2_ST.MESI                 */
+		[ C(RESULT_MISS)   ] = 0x412A, /* L2_ST.ISTATE               */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(DTLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x2140, /* L1D_CACHE_LD.MESI  (alias) */
+		[ C(RESULT_MISS)   ] = 0x0508, /* DTLB_MISSES.MISS_LD        */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x2240, /* L1D_CACHE_ST.MESI  (alias) */
+		[ C(RESULT_MISS)   ] = 0x0608, /* DTLB_MISSES.MISS_ST        */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(ITLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x00c0, /* INST_RETIRED.ANY_P         */
+		[ C(RESULT_MISS)   ] = 0x0282, /* ITLB.MISSES                */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+ [ C(BPU ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x00c4, /* BR_INST_RETIRED.ANY        */
+		[ C(RESULT_MISS)   ] = 0x00c5, /* BP_INST_RETIRED.MISPRED    */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+};
+
+static u64 intel_pmu_raw_event(u64 event)
+{
+#define CORE_EVNTSEL_EVENT_MASK		0x000000FFULL
+#define CORE_EVNTSEL_UNIT_MASK		0x0000FF00ULL
+#define CORE_EVNTSEL_EDGE_MASK		0x00040000ULL
+#define CORE_EVNTSEL_INV_MASK		0x00800000ULL
+#define CORE_EVNTSEL_COUNTER_MASK	0xFF000000ULL
+
+#define CORE_EVNTSEL_MASK		\
+	(CORE_EVNTSEL_EVENT_MASK |	\
+	 CORE_EVNTSEL_UNIT_MASK  |	\
+	 CORE_EVNTSEL_EDGE_MASK  |	\
+	 CORE_EVNTSEL_INV_MASK  |	\
+	 CORE_EVNTSEL_COUNTER_MASK)
+
+	return event & CORE_EVNTSEL_MASK;
+}
+
+static const u64 amd_0f_hw_cache_event_ids
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] =
+{
+ [ C(L1D) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(L1I ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0080, /* Instruction cache fetches  */
+		[ C(RESULT_MISS)   ] = 0x0081, /* Instruction cache misses   */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(LL  ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(DTLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(ITLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0080, /* Instruction fecthes        */
+		[ C(RESULT_MISS)   ] = 0x0085, /* Instr. fetch ITLB misses   */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+ [ C(BPU ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x00c2, /* Retired Branch Instr.      */
+		[ C(RESULT_MISS)   ] = 0x00c3, /* Retired Mispredicted BI    */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+};
+
+/*
+ * AMD Performance Monitor K7 and later.
+ */
+static const u64 amd_perfmon_event_map[] =
+{
+  [PERF_COUNT_HW_CPU_CYCLES]		= 0x0076,
+  [PERF_COUNT_HW_INSTRUCTIONS]		= 0x00c0,
+  [PERF_COUNT_HW_CACHE_REFERENCES]	= 0x0080,
+  [PERF_COUNT_HW_CACHE_MISSES]		= 0x0081,
+  [PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= 0x00c4,
+  [PERF_COUNT_HW_BRANCH_MISSES]		= 0x00c5,
+};
+
+static u64 amd_pmu_event_map(int event)
+{
+	return amd_perfmon_event_map[event];
+}
+
+static u64 amd_pmu_raw_event(u64 event)
+{
+#define K7_EVNTSEL_EVENT_MASK	0x7000000FFULL
+#define K7_EVNTSEL_UNIT_MASK	0x00000FF00ULL
+#define K7_EVNTSEL_EDGE_MASK	0x000040000ULL
+#define K7_EVNTSEL_INV_MASK	0x000800000ULL
+#define K7_EVNTSEL_COUNTER_MASK	0x0FF000000ULL
+
+#define K7_EVNTSEL_MASK			\
+	(K7_EVNTSEL_EVENT_MASK |	\
+	 K7_EVNTSEL_UNIT_MASK  |	\
+	 K7_EVNTSEL_EDGE_MASK  |	\
+	 K7_EVNTSEL_INV_MASK   |	\
+	 K7_EVNTSEL_COUNTER_MASK)
+
+	return event & K7_EVNTSEL_MASK;
+}
+
+/*
+ * Propagate counter elapsed time into the generic counter.
+ * Can only be executed on the CPU where the counter is active.
+ * Returns the delta events processed.
+ */
+static u64
+x86_perf_counter_update(struct perf_counter *counter,
+			struct hw_perf_counter *hwc, int idx)
+{
+	int shift = 64 - x86_pmu.counter_bits;
+	u64 prev_raw_count, new_raw_count;
+	s64 delta;
+
+	/*
+	 * Careful: an NMI might modify the previous counter value.
+	 *
+	 * Our tactic to handle this is to first atomically read and
+	 * exchange a new raw count - then add that new-prev delta
+	 * count to the generic counter atomically:
+	 */
+again:
+	prev_raw_count = atomic64_read(&hwc->prev_count);
+	rdmsrl(hwc->counter_base + idx, new_raw_count);
+
+	if (atomic64_cmpxchg(&hwc->prev_count, prev_raw_count,
+					new_raw_count) != prev_raw_count)
+		goto again;
+
+	/*
+	 * Now we have the new raw value and have updated the prev
+	 * timestamp already. We can now calculate the elapsed delta
+	 * (counter-)time and add that to the generic counter.
+	 *
+	 * Careful, not all hw sign-extends above the physical width
+	 * of the count.
+	 */
+	delta = (new_raw_count << shift) - (prev_raw_count << shift);
+	delta >>= shift;
+
+	atomic64_add(delta, &counter->count);
+	atomic64_sub(delta, &hwc->period_left);
+
+	return new_raw_count;
+}
+
+static atomic_t active_counters;
+static DEFINE_MUTEX(pmc_reserve_mutex);
+
+static bool reserve_pmc_hardware(void)
+{
+	int i;
+
+	if (nmi_watchdog == NMI_LOCAL_APIC)
+		disable_lapic_nmi_watchdog();
+
+	for (i = 0; i < x86_pmu.num_counters; i++) {
+		if (!reserve_perfctr_nmi(x86_pmu.perfctr + i))
+			goto perfctr_fail;
+	}
+
+	for (i = 0; i < x86_pmu.num_counters; i++) {
+		if (!reserve_evntsel_nmi(x86_pmu.eventsel + i))
+			goto eventsel_fail;
+	}
+
+	return true;
+
+eventsel_fail:
+	for (i--; i >= 0; i--)
+		release_evntsel_nmi(x86_pmu.eventsel + i);
+
+	i = x86_pmu.num_counters;
+
+perfctr_fail:
+	for (i--; i >= 0; i--)
+		release_perfctr_nmi(x86_pmu.perfctr + i);
+
+	if (nmi_watchdog == NMI_LOCAL_APIC)
+		enable_lapic_nmi_watchdog();
+
+	return false;
+}
+
+static void release_pmc_hardware(void)
+{
+	int i;
+
+	for (i = 0; i < x86_pmu.num_counters; i++) {
+		release_perfctr_nmi(x86_pmu.perfctr + i);
+		release_evntsel_nmi(x86_pmu.eventsel + i);
+	}
+
+	if (nmi_watchdog == NMI_LOCAL_APIC)
+		enable_lapic_nmi_watchdog();
+}
+
+static void hw_perf_counter_destroy(struct perf_counter *counter)
+{
+	if (atomic_dec_and_mutex_lock(&active_counters, &pmc_reserve_mutex)) {
+		release_pmc_hardware();
+		mutex_unlock(&pmc_reserve_mutex);
+	}
+}
+
+static inline int x86_pmu_initialized(void)
+{
+	return x86_pmu.handle_irq != NULL;
+}
+
+static inline int
+set_ext_hw_attr(struct hw_perf_counter *hwc, struct perf_counter_attr *attr)
+{
+	unsigned int cache_type, cache_op, cache_result;
+	u64 config, val;
+
+	config = attr->config;
+
+	cache_type = (config >>  0) & 0xff;
+	if (cache_type >= PERF_COUNT_HW_CACHE_MAX)
+		return -EINVAL;
+
+	cache_op = (config >>  8) & 0xff;
+	if (cache_op >= PERF_COUNT_HW_CACHE_OP_MAX)
+		return -EINVAL;
+
+	cache_result = (config >> 16) & 0xff;
+	if (cache_result >= PERF_COUNT_HW_CACHE_RESULT_MAX)
+		return -EINVAL;
+
+	val = hw_cache_event_ids[cache_type][cache_op][cache_result];
+
+	if (val == 0)
+		return -ENOENT;
+
+	if (val == -1)
+		return -EINVAL;
+
+	hwc->config |= val;
+
+	return 0;
+}
+
+/*
+ * Setup the hardware configuration for a given attr_type
+ */
+static int __hw_perf_counter_init(struct perf_counter *counter)
+{
+	struct perf_counter_attr *attr = &counter->attr;
+	struct hw_perf_counter *hwc = &counter->hw;
+	int err;
+
+	if (!x86_pmu_initialized())
+		return -ENODEV;
+
+	err = 0;
+	if (!atomic_inc_not_zero(&active_counters)) {
+		mutex_lock(&pmc_reserve_mutex);
+		if (atomic_read(&active_counters) == 0 && !reserve_pmc_hardware())
+			err = -EBUSY;
+		else
+			atomic_inc(&active_counters);
+		mutex_unlock(&pmc_reserve_mutex);
+	}
+	if (err)
+		return err;
+
+	/*
+	 * Generate PMC IRQs:
+	 * (keep 'enabled' bit clear for now)
+	 */
+	hwc->config = ARCH_PERFMON_EVENTSEL_INT;
+
+	/*
+	 * Count user and OS events unless requested not to.
+	 */
+	if (!attr->exclude_user)
+		hwc->config |= ARCH_PERFMON_EVENTSEL_USR;
+	if (!attr->exclude_kernel)
+		hwc->config |= ARCH_PERFMON_EVENTSEL_OS;
+
+	if (!hwc->sample_period) {
+		hwc->sample_period = x86_pmu.max_period;
+		hwc->last_period = hwc->sample_period;
+		atomic64_set(&hwc->period_left, hwc->sample_period);
+	}
+
+	counter->destroy = hw_perf_counter_destroy;
+
+	/*
+	 * Raw event type provide the config in the event structure
+	 */
+	if (attr->type == PERF_TYPE_RAW) {
+		hwc->config |= x86_pmu.raw_event(attr->config);
+		return 0;
+	}
+
+	if (attr->type == PERF_TYPE_HW_CACHE)
+		return set_ext_hw_attr(hwc, attr);
+
+	if (attr->config >= x86_pmu.max_events)
+		return -EINVAL;
+	/*
+	 * The generic map:
+	 */
+	hwc->config |= x86_pmu.event_map(attr->config);
+
+	return 0;
+}
+
+static void intel_pmu_disable_all(void)
+{
+	wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0);
+}
+
+static void amd_pmu_disable_all(void)
+{
+	struct cpu_hw_counters *cpuc = &__get_cpu_var(cpu_hw_counters);
+	int idx;
+
+	if (!cpuc->enabled)
+		return;
+
+	cpuc->enabled = 0;
+	/*
+	 * ensure we write the disable before we start disabling the
+	 * counters proper, so that amd_pmu_enable_counter() does the
+	 * right thing.
+	 */
+	barrier();
+
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
+		u64 val;
+
+		if (!test_bit(idx, cpuc->active_mask))
+			continue;
+		rdmsrl(MSR_K7_EVNTSEL0 + idx, val);
+		if (!(val & ARCH_PERFMON_EVENTSEL0_ENABLE))
+			continue;
+		val &= ~ARCH_PERFMON_EVENTSEL0_ENABLE;
+		wrmsrl(MSR_K7_EVNTSEL0 + idx, val);
+	}
+}
+
+void hw_perf_disable(void)
+{
+	if (!x86_pmu_initialized())
+		return;
+	return x86_pmu.disable_all();
+}
+
+static void intel_pmu_enable_all(void)
+{
+	wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, x86_pmu.intel_ctrl);
+}
+
+static void amd_pmu_enable_all(void)
+{
+	struct cpu_hw_counters *cpuc = &__get_cpu_var(cpu_hw_counters);
+	int idx;
+
+	if (cpuc->enabled)
+		return;
+
+	cpuc->enabled = 1;
+	barrier();
+
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
+		u64 val;
+
+		if (!test_bit(idx, cpuc->active_mask))
+			continue;
+		rdmsrl(MSR_K7_EVNTSEL0 + idx, val);
+		if (val & ARCH_PERFMON_EVENTSEL0_ENABLE)
+			continue;
+		val |= ARCH_PERFMON_EVENTSEL0_ENABLE;
+		wrmsrl(MSR_K7_EVNTSEL0 + idx, val);
+	}
+}
+
+void hw_perf_enable(void)
+{
+	if (!x86_pmu_initialized())
+		return;
+	x86_pmu.enable_all();
+}
+
+static inline u64 intel_pmu_get_status(void)
+{
+	u64 status;
+
+	rdmsrl(MSR_CORE_PERF_GLOBAL_STATUS, status);
+
+	return status;
+}
+
+static inline void intel_pmu_ack_status(u64 ack)
+{
+	wrmsrl(MSR_CORE_PERF_GLOBAL_OVF_CTRL, ack);
+}
+
+static inline void x86_pmu_enable_counter(struct hw_perf_counter *hwc, int idx)
+{
+	int err;
+	err = checking_wrmsrl(hwc->config_base + idx,
+			      hwc->config | ARCH_PERFMON_EVENTSEL0_ENABLE);
+}
+
+static inline void x86_pmu_disable_counter(struct hw_perf_counter *hwc, int idx)
+{
+	int err;
+	err = checking_wrmsrl(hwc->config_base + idx,
+			      hwc->config);
+}
+
+static inline void
+intel_pmu_disable_fixed(struct hw_perf_counter *hwc, int __idx)
+{
+	int idx = __idx - X86_PMC_IDX_FIXED;
+	u64 ctrl_val, mask;
+	int err;
+
+	mask = 0xfULL << (idx * 4);
+
+	rdmsrl(hwc->config_base, ctrl_val);
+	ctrl_val &= ~mask;
+	err = checking_wrmsrl(hwc->config_base, ctrl_val);
+}
+
+static inline void
+intel_pmu_disable_counter(struct hw_perf_counter *hwc, int idx)
+{
+	if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) {
+		intel_pmu_disable_fixed(hwc, idx);
+		return;
+	}
+
+	x86_pmu_disable_counter(hwc, idx);
+}
+
+static inline void
+amd_pmu_disable_counter(struct hw_perf_counter *hwc, int idx)
+{
+	x86_pmu_disable_counter(hwc, idx);
+}
+
+static DEFINE_PER_CPU(u64, prev_left[X86_PMC_IDX_MAX]);
+
+/*
+ * Set the next IRQ period, based on the hwc->period_left value.
+ * To be called with the counter disabled in hw:
+ */
+static int
+x86_perf_counter_set_period(struct perf_counter *counter,
+			     struct hw_perf_counter *hwc, int idx)
+{
+	s64 left = atomic64_read(&hwc->period_left);
+	s64 period = hwc->sample_period;
+	int err, ret = 0;
+
+	/*
+	 * If we are way outside a reasoable range then just skip forward:
+	 */
+	if (unlikely(left <= -period)) {
+		left = period;
+		atomic64_set(&hwc->period_left, left);
+		hwc->last_period = period;
+		ret = 1;
+	}
+
+	if (unlikely(left <= 0)) {
+		left += period;
+		atomic64_set(&hwc->period_left, left);
+		hwc->last_period = period;
+		ret = 1;
+	}
+	/*
+	 * Quirk: certain CPUs dont like it if just 1 event is left:
+	 */
+	if (unlikely(left < 2))
+		left = 2;
+
+	if (left > x86_pmu.max_period)
+		left = x86_pmu.max_period;
+
+	per_cpu(prev_left[idx], smp_processor_id()) = left;
+
+	/*
+	 * The hw counter starts counting from this counter offset,
+	 * mark it to be able to extra future deltas:
+	 */
+	atomic64_set(&hwc->prev_count, (u64)-left);
+
+	err = checking_wrmsrl(hwc->counter_base + idx,
+			     (u64)(-left) & x86_pmu.counter_mask);
+
+	return ret;
+}
+
+static inline void
+intel_pmu_enable_fixed(struct hw_perf_counter *hwc, int __idx)
+{
+	int idx = __idx - X86_PMC_IDX_FIXED;
+	u64 ctrl_val, bits, mask;
+	int err;
+
+	/*
+	 * Enable IRQ generation (0x8),
+	 * and enable ring-3 counting (0x2) and ring-0 counting (0x1)
+	 * if requested:
+	 */
+	bits = 0x8ULL;
+	if (hwc->config & ARCH_PERFMON_EVENTSEL_USR)
+		bits |= 0x2;
+	if (hwc->config & ARCH_PERFMON_EVENTSEL_OS)
+		bits |= 0x1;
+	bits <<= (idx * 4);
+	mask = 0xfULL << (idx * 4);
+
+	rdmsrl(hwc->config_base, ctrl_val);
+	ctrl_val &= ~mask;
+	ctrl_val |= bits;
+	err = checking_wrmsrl(hwc->config_base, ctrl_val);
+}
+
+static void intel_pmu_enable_counter(struct hw_perf_counter *hwc, int idx)
+{
+	if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) {
+		intel_pmu_enable_fixed(hwc, idx);
+		return;
+	}
+
+	x86_pmu_enable_counter(hwc, idx);
+}
+
+static void amd_pmu_enable_counter(struct hw_perf_counter *hwc, int idx)
+{
+	struct cpu_hw_counters *cpuc = &__get_cpu_var(cpu_hw_counters);
+
+	if (cpuc->enabled)
+		x86_pmu_enable_counter(hwc, idx);
+	else
+		x86_pmu_disable_counter(hwc, idx);
+}
+
+static int
+fixed_mode_idx(struct perf_counter *counter, struct hw_perf_counter *hwc)
+{
+	unsigned int event;
+
+	if (!x86_pmu.num_counters_fixed)
+		return -1;
+
+	event = hwc->config & ARCH_PERFMON_EVENT_MASK;
+
+	if (unlikely(event == x86_pmu.event_map(PERF_COUNT_HW_INSTRUCTIONS)))
+		return X86_PMC_IDX_FIXED_INSTRUCTIONS;
+	if (unlikely(event == x86_pmu.event_map(PERF_COUNT_HW_CPU_CYCLES)))
+		return X86_PMC_IDX_FIXED_CPU_CYCLES;
+	if (unlikely(event == x86_pmu.event_map(PERF_COUNT_HW_BUS_CYCLES)))
+		return X86_PMC_IDX_FIXED_BUS_CYCLES;
+
+	return -1;
+}
+
+/*
+ * Find a PMC slot for the freshly enabled / scheduled in counter:
+ */
+static int x86_pmu_enable(struct perf_counter *counter)
+{
+	struct cpu_hw_counters *cpuc = &__get_cpu_var(cpu_hw_counters);
+	struct hw_perf_counter *hwc = &counter->hw;
+	int idx;
+
+	idx = fixed_mode_idx(counter, hwc);
+	if (idx >= 0) {
+		/*
+		 * Try to get the fixed counter, if that is already taken
+		 * then try to get a generic counter:
+		 */
+		if (test_and_set_bit(idx, cpuc->used_mask))
+			goto try_generic;
+
+		hwc->config_base = MSR_ARCH_PERFMON_FIXED_CTR_CTRL;
+		/*
+		 * We set it so that counter_base + idx in wrmsr/rdmsr maps to
+		 * MSR_ARCH_PERFMON_FIXED_CTR0 ... CTR2:
+		 */
+		hwc->counter_base =
+			MSR_ARCH_PERFMON_FIXED_CTR0 - X86_PMC_IDX_FIXED;
+		hwc->idx = idx;
+	} else {
+		idx = hwc->idx;
+		/* Try to get the previous generic counter again */
+		if (test_and_set_bit(idx, cpuc->used_mask)) {
+try_generic:
+			idx = find_first_zero_bit(cpuc->used_mask,
+						  x86_pmu.num_counters);
+			if (idx == x86_pmu.num_counters)
+				return -EAGAIN;
+
+			set_bit(idx, cpuc->used_mask);
+			hwc->idx = idx;
+		}
+		hwc->config_base  = x86_pmu.eventsel;
+		hwc->counter_base = x86_pmu.perfctr;
+	}
+
+	perf_counters_lapic_init();
+
+	x86_pmu.disable(hwc, idx);
+
+	cpuc->counters[idx] = counter;
+	set_bit(idx, cpuc->active_mask);
+
+	x86_perf_counter_set_period(counter, hwc, idx);
+	x86_pmu.enable(hwc, idx);
+
+	return 0;
+}
+
+static void x86_pmu_unthrottle(struct perf_counter *counter)
+{
+	struct cpu_hw_counters *cpuc = &__get_cpu_var(cpu_hw_counters);
+	struct hw_perf_counter *hwc = &counter->hw;
+
+	if (WARN_ON_ONCE(hwc->idx >= X86_PMC_IDX_MAX ||
+				cpuc->counters[hwc->idx] != counter))
+		return;
+
+	x86_pmu.enable(hwc, hwc->idx);
+}
+
+void perf_counter_print_debug(void)
+{
+	u64 ctrl, status, overflow, pmc_ctrl, pmc_count, prev_left, fixed;
+	struct cpu_hw_counters *cpuc;
+	unsigned long flags;
+	int cpu, idx;
+
+	if (!x86_pmu.num_counters)
+		return;
+
+	local_irq_save(flags);
+
+	cpu = smp_processor_id();
+	cpuc = &per_cpu(cpu_hw_counters, cpu);
+
+	if (x86_pmu.version >= 2) {
+		rdmsrl(MSR_CORE_PERF_GLOBAL_CTRL, ctrl);
+		rdmsrl(MSR_CORE_PERF_GLOBAL_STATUS, status);
+		rdmsrl(MSR_CORE_PERF_GLOBAL_OVF_CTRL, overflow);
+		rdmsrl(MSR_ARCH_PERFMON_FIXED_CTR_CTRL, fixed);
+
+		pr_info("\n");
+		pr_info("CPU#%d: ctrl:       %016llx\n", cpu, ctrl);
+		pr_info("CPU#%d: status:     %016llx\n", cpu, status);
+		pr_info("CPU#%d: overflow:   %016llx\n", cpu, overflow);
+		pr_info("CPU#%d: fixed:      %016llx\n", cpu, fixed);
+	}
+	pr_info("CPU#%d: used:       %016llx\n", cpu, *(u64 *)cpuc->used_mask);
+
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
+		rdmsrl(x86_pmu.eventsel + idx, pmc_ctrl);
+		rdmsrl(x86_pmu.perfctr  + idx, pmc_count);
+
+		prev_left = per_cpu(prev_left[idx], cpu);
+
+		pr_info("CPU#%d:   gen-PMC%d ctrl:  %016llx\n",
+			cpu, idx, pmc_ctrl);
+		pr_info("CPU#%d:   gen-PMC%d count: %016llx\n",
+			cpu, idx, pmc_count);
+		pr_info("CPU#%d:   gen-PMC%d left:  %016llx\n",
+			cpu, idx, prev_left);
+	}
+	for (idx = 0; idx < x86_pmu.num_counters_fixed; idx++) {
+		rdmsrl(MSR_ARCH_PERFMON_FIXED_CTR0 + idx, pmc_count);
+
+		pr_info("CPU#%d: fixed-PMC%d count: %016llx\n",
+			cpu, idx, pmc_count);
+	}
+	local_irq_restore(flags);
+}
+
+static void x86_pmu_disable(struct perf_counter *counter)
+{
+	struct cpu_hw_counters *cpuc = &__get_cpu_var(cpu_hw_counters);
+	struct hw_perf_counter *hwc = &counter->hw;
+	int idx = hwc->idx;
+
+	/*
+	 * Must be done before we disable, otherwise the nmi handler
+	 * could reenable again:
+	 */
+	clear_bit(idx, cpuc->active_mask);
+	x86_pmu.disable(hwc, idx);
+
+	/*
+	 * Make sure the cleared pointer becomes visible before we
+	 * (potentially) free the counter:
+	 */
+	barrier();
+
+	/*
+	 * Drain the remaining delta count out of a counter
+	 * that we are disabling:
+	 */
+	x86_perf_counter_update(counter, hwc, idx);
+	cpuc->counters[idx] = NULL;
+	clear_bit(idx, cpuc->used_mask);
+}
+
+/*
+ * Save and restart an expired counter. Called by NMI contexts,
+ * so it has to be careful about preempting normal counter ops:
+ */
+static int intel_pmu_save_and_restart(struct perf_counter *counter)
+{
+	struct hw_perf_counter *hwc = &counter->hw;
+	int idx = hwc->idx;
+	int ret;
+
+	x86_perf_counter_update(counter, hwc, idx);
+	ret = x86_perf_counter_set_period(counter, hwc, idx);
+
+	if (counter->state == PERF_COUNTER_STATE_ACTIVE)
+		intel_pmu_enable_counter(hwc, idx);
+
+	return ret;
+}
+
+static void intel_pmu_reset(void)
+{
+	unsigned long flags;
+	int idx;
+
+	if (!x86_pmu.num_counters)
+		return;
+
+	local_irq_save(flags);
+
+	printk("clearing PMU state on CPU#%d\n", smp_processor_id());
+
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
+		checking_wrmsrl(x86_pmu.eventsel + idx, 0ull);
+		checking_wrmsrl(x86_pmu.perfctr  + idx, 0ull);
+	}
+	for (idx = 0; idx < x86_pmu.num_counters_fixed; idx++) {
+		checking_wrmsrl(MSR_ARCH_PERFMON_FIXED_CTR0 + idx, 0ull);
+	}
+
+	local_irq_restore(flags);
+}
+
+
+/*
+ * This handler is triggered by the local APIC, so the APIC IRQ handling
+ * rules apply:
+ */
+static int intel_pmu_handle_irq(struct pt_regs *regs)
+{
+	struct perf_sample_data data;
+	struct cpu_hw_counters *cpuc;
+	int bit, cpu, loops;
+	u64 ack, status;
+
+	data.regs = regs;
+	data.addr = 0;
+
+	cpu = smp_processor_id();
+	cpuc = &per_cpu(cpu_hw_counters, cpu);
+
+	perf_disable();
+	status = intel_pmu_get_status();
+	if (!status) {
+		perf_enable();
+		return 0;
+	}
+
+	loops = 0;
+again:
+	if (++loops > 100) {
+		WARN_ONCE(1, "perfcounters: irq loop stuck!\n");
+		perf_counter_print_debug();
+		intel_pmu_reset();
+		perf_enable();
+		return 1;
+	}
+
+	inc_irq_stat(apic_perf_irqs);
+	ack = status;
+	for_each_bit(bit, (unsigned long *)&status, X86_PMC_IDX_MAX) {
+		struct perf_counter *counter = cpuc->counters[bit];
+
+		clear_bit(bit, (unsigned long *) &status);
+		if (!test_bit(bit, cpuc->active_mask))
+			continue;
+
+		if (!intel_pmu_save_and_restart(counter))
+			continue;
+
+		if (perf_counter_overflow(counter, 1, &data))
+			intel_pmu_disable_counter(&counter->hw, bit);
+	}
+
+	intel_pmu_ack_status(ack);
+
+	/*
+	 * Repeat if there is more work to be done:
+	 */
+	status = intel_pmu_get_status();
+	if (status)
+		goto again;
+
+	perf_enable();
+
+	return 1;
+}
+
+static int amd_pmu_handle_irq(struct pt_regs *regs)
+{
+	struct perf_sample_data data;
+	struct cpu_hw_counters *cpuc;
+	struct perf_counter *counter;
+	struct hw_perf_counter *hwc;
+	int cpu, idx, handled = 0;
+	u64 val;
+
+	data.regs = regs;
+	data.addr = 0;
+
+	cpu = smp_processor_id();
+	cpuc = &per_cpu(cpu_hw_counters, cpu);
+
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
+		if (!test_bit(idx, cpuc->active_mask))
+			continue;
+
+		counter = cpuc->counters[idx];
+		hwc = &counter->hw;
+
+		val = x86_perf_counter_update(counter, hwc, idx);
+		if (val & (1ULL << (x86_pmu.counter_bits - 1)))
+			continue;
+
+		/*
+		 * counter overflow
+		 */
+		handled		= 1;
+		data.period	= counter->hw.last_period;
+
+		if (!x86_perf_counter_set_period(counter, hwc, idx))
+			continue;
+
+		if (perf_counter_overflow(counter, 1, &data))
+			amd_pmu_disable_counter(hwc, idx);
+	}
+
+	if (handled)
+		inc_irq_stat(apic_perf_irqs);
+
+	return handled;
+}
+
+void smp_perf_pending_interrupt(struct pt_regs *regs)
+{
+	irq_enter();
+	ack_APIC_irq();
+	inc_irq_stat(apic_pending_irqs);
+	perf_counter_do_pending();
+	irq_exit();
+}
+
+void set_perf_counter_pending(void)
+{
+	apic->send_IPI_self(LOCAL_PENDING_VECTOR);
+}
+
+void perf_counters_lapic_init(void)
+{
+	if (!x86_pmu_initialized())
+		return;
+
+	/*
+	 * Always use NMI for PMU
+	 */
+	apic_write(APIC_LVTPC, APIC_DM_NMI);
+}
+
+static int __kprobes
+perf_counter_nmi_handler(struct notifier_block *self,
+			 unsigned long cmd, void *__args)
+{
+	struct die_args *args = __args;
+	struct pt_regs *regs;
+
+	if (!atomic_read(&active_counters))
+		return NOTIFY_DONE;
+
+	switch (cmd) {
+	case DIE_NMI:
+	case DIE_NMI_IPI:
+		break;
+
+	default:
+		return NOTIFY_DONE;
+	}
+
+	regs = args->regs;
+
+	apic_write(APIC_LVTPC, APIC_DM_NMI);
+	/*
+	 * Can't rely on the handled return value to say it was our NMI, two
+	 * counters could trigger 'simultaneously' raising two back-to-back NMIs.
+	 *
+	 * If the first NMI handles both, the latter will be empty and daze
+	 * the CPU.
+	 */
+	x86_pmu.handle_irq(regs);
+
+	return NOTIFY_STOP;
+}
+
+static __read_mostly struct notifier_block perf_counter_nmi_notifier = {
+	.notifier_call		= perf_counter_nmi_handler,
+	.next			= NULL,
+	.priority		= 1
+};
+
+static struct x86_pmu intel_pmu = {
+	.name			= "Intel",
+	.handle_irq		= intel_pmu_handle_irq,
+	.disable_all		= intel_pmu_disable_all,
+	.enable_all		= intel_pmu_enable_all,
+	.enable			= intel_pmu_enable_counter,
+	.disable		= intel_pmu_disable_counter,
+	.eventsel		= MSR_ARCH_PERFMON_EVENTSEL0,
+	.perfctr		= MSR_ARCH_PERFMON_PERFCTR0,
+	.event_map		= intel_pmu_event_map,
+	.raw_event		= intel_pmu_raw_event,
+	.max_events		= ARRAY_SIZE(intel_perfmon_event_map),
+	/*
+	 * Intel PMCs cannot be accessed sanely above 32 bit width,
+	 * so we install an artificial 1<<31 period regardless of
+	 * the generic counter period:
+	 */
+	.max_period		= (1ULL << 31) - 1,
+};
+
+static struct x86_pmu amd_pmu = {
+	.name			= "AMD",
+	.handle_irq		= amd_pmu_handle_irq,
+	.disable_all		= amd_pmu_disable_all,
+	.enable_all		= amd_pmu_enable_all,
+	.enable			= amd_pmu_enable_counter,
+	.disable		= amd_pmu_disable_counter,
+	.eventsel		= MSR_K7_EVNTSEL0,
+	.perfctr		= MSR_K7_PERFCTR0,
+	.event_map		= amd_pmu_event_map,
+	.raw_event		= amd_pmu_raw_event,
+	.max_events		= ARRAY_SIZE(amd_perfmon_event_map),
+	.num_counters		= 4,
+	.counter_bits		= 48,
+	.counter_mask		= (1ULL << 48) - 1,
+	/* use highest bit to detect overflow */
+	.max_period		= (1ULL << 47) - 1,
+};
+
+static int intel_pmu_init(void)
+{
+	union cpuid10_edx edx;
+	union cpuid10_eax eax;
+	unsigned int unused;
+	unsigned int ebx;
+	int version;
+
+	if (!cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
+		return -ENODEV;
+
+	/*
+	 * Check whether the Architectural PerfMon supports
+	 * Branch Misses Retired Event or not.
+	 */
+	cpuid(10, &eax.full, &ebx, &unused, &edx.full);
+	if (eax.split.mask_length <= ARCH_PERFMON_BRANCH_MISSES_RETIRED)
+		return -ENODEV;
+
+	version = eax.split.version_id;
+	if (version < 2)
+		return -ENODEV;
+
+	x86_pmu				= intel_pmu;
+	x86_pmu.version			= version;
+	x86_pmu.num_counters		= eax.split.num_counters;
+	x86_pmu.counter_bits		= eax.split.bit_width;
+	x86_pmu.counter_mask		= (1ULL << eax.split.bit_width) - 1;
+
+	/*
+	 * Quirk: v2 perfmon does not report fixed-purpose counters, so
+	 * assume at least 3 counters:
+	 */
+	x86_pmu.num_counters_fixed	= max((int)edx.split.num_counters_fixed, 3);
+
+	rdmsrl(MSR_CORE_PERF_GLOBAL_CTRL, x86_pmu.intel_ctrl);
+
+	/*
+	 * Install the hw-cache-events table:
+	 */
+	switch (boot_cpu_data.x86_model) {
+	case 15: /* original 65 nm celeron/pentium/core2/xeon, "Merom"/"Conroe" */
+	case 22: /* single-core 65 nm celeron/core2solo "Merom-L"/"Conroe-L" */
+	case 23: /* current 45 nm celeron/core2/xeon "Penryn"/"Wolfdale" */
+	case 29: /* six-core 45 nm xeon "Dunnington" */
+		memcpy(hw_cache_event_ids, core2_hw_cache_event_ids,
+		       sizeof(hw_cache_event_ids));
+
+		pr_cont("Core2 events, ");
+		break;
+	default:
+	case 26:
+		memcpy(hw_cache_event_ids, nehalem_hw_cache_event_ids,
+		       sizeof(hw_cache_event_ids));
+
+		pr_cont("Nehalem/Corei7 events, ");
+		break;
+	case 28:
+		memcpy(hw_cache_event_ids, atom_hw_cache_event_ids,
+		       sizeof(hw_cache_event_ids));
+
+		pr_cont("Atom events, ");
+		break;
+	}
+	return 0;
+}
+
+static int amd_pmu_init(void)
+{
+	x86_pmu = amd_pmu;
+
+	switch (boot_cpu_data.x86) {
+	case 0x0f:
+	case 0x10:
+	case 0x11:
+		memcpy(hw_cache_event_ids, amd_0f_hw_cache_event_ids,
+		       sizeof(hw_cache_event_ids));
+
+		pr_cont("AMD Family 0f/10/11 events, ");
+		break;
+	}
+	return 0;
+}
+
+void __init init_hw_perf_counters(void)
+{
+	int err;
+
+	pr_info("Performance Counters: ");
+
+	switch (boot_cpu_data.x86_vendor) {
+	case X86_VENDOR_INTEL:
+		err = intel_pmu_init();
+		break;
+	case X86_VENDOR_AMD:
+		err = amd_pmu_init();
+		break;
+	default:
+		return;
+	}
+	if (err != 0) {
+		pr_cont("no PMU driver, software counters only.\n");
+		return;
+	}
+
+	pr_cont("%s PMU driver.\n", x86_pmu.name);
+
+	if (x86_pmu.num_counters > X86_PMC_MAX_GENERIC) {
+		x86_pmu.num_counters = X86_PMC_MAX_GENERIC;
+		WARN(1, KERN_ERR "hw perf counters %d > max(%d), clipping!",
+		     x86_pmu.num_counters, X86_PMC_MAX_GENERIC);
+	}
+	perf_counter_mask = (1 << x86_pmu.num_counters) - 1;
+	perf_max_counters = x86_pmu.num_counters;
+
+	if (x86_pmu.num_counters_fixed > X86_PMC_MAX_FIXED) {
+		x86_pmu.num_counters_fixed = X86_PMC_MAX_FIXED;
+		WARN(1, KERN_ERR "hw perf counters fixed %d > max(%d), clipping!",
+		     x86_pmu.num_counters_fixed, X86_PMC_MAX_FIXED);
+	}
+
+	perf_counter_mask |=
+		((1LL << x86_pmu.num_counters_fixed)-1) << X86_PMC_IDX_FIXED;
+
+	perf_counters_lapic_init();
+	register_die_notifier(&perf_counter_nmi_notifier);
+
+	pr_info("... version:                 %d\n",     x86_pmu.version);
+	pr_info("... bit width:               %d\n",     x86_pmu.counter_bits);
+	pr_info("... generic counters:        %d\n",     x86_pmu.num_counters);
+	pr_info("... value mask:              %016Lx\n", x86_pmu.counter_mask);
+	pr_info("... max period:              %016Lx\n", x86_pmu.max_period);
+	pr_info("... fixed-purpose counters:  %d\n",     x86_pmu.num_counters_fixed);
+	pr_info("... counter mask:            %016Lx\n", perf_counter_mask);
+}
+
+static inline void x86_pmu_read(struct perf_counter *counter)
+{
+	x86_perf_counter_update(counter, &counter->hw, counter->hw.idx);
+}
+
+static const struct pmu pmu = {
+	.enable		= x86_pmu_enable,
+	.disable	= x86_pmu_disable,
+	.read		= x86_pmu_read,
+	.unthrottle	= x86_pmu_unthrottle,
+};
+
+const struct pmu *hw_perf_counter_init(struct perf_counter *counter)
+{
+	int err;
+
+	err = __hw_perf_counter_init(counter);
+	if (err)
+		return ERR_PTR(err);
+
+	return &pmu;
+}
+
+/*
+ * callchain support
+ */
+
+static inline
+void callchain_store(struct perf_callchain_entry *entry, unsigned long ip)
+{
+	if (entry->nr < MAX_STACK_DEPTH)
+		entry->ip[entry->nr++] = ip;
+}
+
+static DEFINE_PER_CPU(struct perf_callchain_entry, irq_entry);
+static DEFINE_PER_CPU(struct perf_callchain_entry, nmi_entry);
+
+
+static void
+backtrace_warning_symbol(void *data, char *msg, unsigned long symbol)
+{
+	/* Ignore warnings */
+}
+
+static void backtrace_warning(void *data, char *msg)
+{
+	/* Ignore warnings */
+}
+
+static int backtrace_stack(void *data, char *name)
+{
+	/* Don't bother with IRQ stacks for now */
+	return -1;
+}
+
+static void backtrace_address(void *data, unsigned long addr, int reliable)
+{
+	struct perf_callchain_entry *entry = data;
+
+	if (reliable)
+		callchain_store(entry, addr);
+}
+
+static const struct stacktrace_ops backtrace_ops = {
+	.warning		= backtrace_warning,
+	.warning_symbol		= backtrace_warning_symbol,
+	.stack			= backtrace_stack,
+	.address		= backtrace_address,
+};
+
+static void
+perf_callchain_kernel(struct pt_regs *regs, struct perf_callchain_entry *entry)
+{
+	unsigned long bp;
+	char *stack;
+	int nr = entry->nr;
+
+	callchain_store(entry, instruction_pointer(regs));
+
+	stack = ((char *)regs + sizeof(struct pt_regs));
+#ifdef CONFIG_FRAME_POINTER
+	bp = frame_pointer(regs);
+#else
+	bp = 0;
+#endif
+
+	dump_trace(NULL, regs, (void *)stack, bp, &backtrace_ops, entry);
+
+	entry->kernel = entry->nr - nr;
+}
+
+
+struct stack_frame {
+	const void __user	*next_fp;
+	unsigned long		return_address;
+};
+
+static int copy_stack_frame(const void __user *fp, struct stack_frame *frame)
+{
+	int ret;
+
+	if (!access_ok(VERIFY_READ, fp, sizeof(*frame)))
+		return 0;
+
+	ret = 1;
+	pagefault_disable();
+	if (__copy_from_user_inatomic(frame, fp, sizeof(*frame)))
+		ret = 0;
+	pagefault_enable();
+
+	return ret;
+}
+
+static void
+perf_callchain_user(struct pt_regs *regs, struct perf_callchain_entry *entry)
+{
+	struct stack_frame frame;
+	const void __user *fp;
+	int nr = entry->nr;
+
+	regs = (struct pt_regs *)current->thread.sp0 - 1;
+	fp   = (void __user *)regs->bp;
+
+	callchain_store(entry, regs->ip);
+
+	while (entry->nr < MAX_STACK_DEPTH) {
+		frame.next_fp	     = NULL;
+		frame.return_address = 0;
+
+		if (!copy_stack_frame(fp, &frame))
+			break;
+
+		if ((unsigned long)fp < user_stack_pointer(regs))
+			break;
+
+		callchain_store(entry, frame.return_address);
+		fp = frame.next_fp;
+	}
+
+	entry->user = entry->nr - nr;
+}
+
+static void
+perf_do_callchain(struct pt_regs *regs, struct perf_callchain_entry *entry)
+{
+	int is_user;
+
+	if (!regs)
+		return;
+
+	is_user = user_mode(regs);
+
+	if (!current || current->pid == 0)
+		return;
+
+	if (is_user && current->state != TASK_RUNNING)
+		return;
+
+	if (!is_user)
+		perf_callchain_kernel(regs, entry);
+
+	if (current->mm)
+		perf_callchain_user(regs, entry);
+}
+
+struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
+{
+	struct perf_callchain_entry *entry;
+
+	if (in_nmi())
+		entry = &__get_cpu_var(nmi_entry);
+	else
+		entry = &__get_cpu_var(irq_entry);
+
+	entry->nr = 0;
+	entry->hv = 0;
+	entry->kernel = 0;
+	entry->user = 0;
+
+	perf_do_callchain(regs, entry);
+
+	return entry;
+}
diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c
index f6c70a164e32..d6f5b9fbde32 100644
--- a/arch/x86/kernel/cpu/perfctr-watchdog.c
+++ b/arch/x86/kernel/cpu/perfctr-watchdog.c
@@ -19,8 +19,8 @@
 #include <linux/nmi.h>
 #include <linux/kprobes.h>
 
-#include <asm/genapic.h>
-#include <asm/intel_arch_perfmon.h>
+#include <asm/apic.h>
+#include <asm/perf_counter.h>
 
 struct nmi_watchdog_ctlblk {
 	unsigned int cccr_msr;