i386: move kernel/cpu

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

15 files changed, 4540 insertions, 0 deletions

diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile
new file mode 100644
index 00000000000..6687f6d5ad2
--- /dev/null
+++ b/arch/x86/kernel/cpu/Makefile
@@ -0,0 +1,20 @@
+#
+# Makefile for x86-compatible CPU details and quirks
+#
+
+obj-y	:=	common.o proc.o bugs.o
+
+obj-y	+=	amd.o
+obj-y	+=	cyrix.o
+obj-y	+=	centaur.o
+obj-y	+=	transmeta.o
+obj-y	+=	intel.o intel_cacheinfo.o addon_cpuid_features.o
+obj-y	+=	nexgen.o
+obj-y	+=	umc.o
+
+obj-$(CONFIG_X86_MCE)	+=	../../../x86/kernel/cpu/mcheck/
+
+obj-$(CONFIG_MTRR)	+= 	../../../x86/kernel/cpu/mtrr/
+obj-$(CONFIG_CPU_FREQ)	+=	../../../x86/kernel/cpu/cpufreq/
+
+obj-$(CONFIG_X86_LOCAL_APIC) += perfctr-watchdog.o
diff --git a/arch/x86/kernel/cpu/addon_cpuid_features.c b/arch/x86/kernel/cpu/addon_cpuid_features.c
new file mode 100644
index 00000000000..3e91d3ee26e
--- /dev/null
+++ b/arch/x86/kernel/cpu/addon_cpuid_features.c
@@ -0,0 +1,50 @@
+
+/*
+ *	Routines to indentify additional cpu features that are scattered in
+ *	cpuid space.
+ */
+
+#include <linux/cpu.h>
+
+#include <asm/processor.h>
+
+struct cpuid_bit {
+	u16 feature;
+	u8 reg;
+	u8 bit;
+	u32 level;
+};
+
+enum cpuid_regs {
+	CR_EAX = 0,
+	CR_ECX,
+	CR_EDX,
+	CR_EBX
+};
+
+void __cpuinit init_scattered_cpuid_features(struct cpuinfo_x86 *c)
+{
+	u32 max_level;
+	u32 regs[4];
+	const struct cpuid_bit *cb;
+
+	static const struct cpuid_bit cpuid_bits[] = {
+		{ X86_FEATURE_IDA, CR_EAX, 1, 0x00000006 },
+		{ 0, 0, 0, 0 }
+	};
+
+	for (cb = cpuid_bits; cb->feature; cb++) {
+
+		/* Verify that the level is valid */
+		max_level = cpuid_eax(cb->level & 0xffff0000);
+		if (max_level < cb->level ||
+		    max_level > (cb->level | 0xffff))
+			continue;
+
+		cpuid(cb->level, &regs[CR_EAX], &regs[CR_EBX],
+			&regs[CR_ECX], &regs[CR_EDX]);
+
+		if (regs[cb->reg] & (1 << cb->bit))
+			set_bit(cb->feature, c->x86_capability);
+	}
+}
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
new file mode 100644
index 00000000000..dcf6bbb1c7c
--- /dev/null
+++ b/arch/x86/kernel/cpu/amd.c
@@ -0,0 +1,337 @@
+#include <linux/init.h>
+#include <linux/bitops.h>
+#include <linux/mm.h>
+#include <asm/io.h>
+#include <asm/processor.h>
+#include <asm/apic.h>
+
+#include "cpu.h"
+
+/*
+ *	B step AMD K6 before B 9730xxxx have hardware bugs that can cause
+ *	misexecution of code under Linux. Owners of such processors should
+ *	contact AMD for precise details and a CPU swap.
+ *
+ *	See	http://www.multimania.com/poulot/k6bug.html
+ *		http://www.amd.com/K6/k6docs/revgd.html
+ *
+ *	The following test is erm.. interesting. AMD neglected to up
+ *	the chip setting when fixing the bug but they also tweaked some
+ *	performance at the same time..
+ */
+ 
+extern void vide(void);
+__asm__(".align 4\nvide: ret");
+
+#ifdef CONFIG_X86_LOCAL_APIC
+#define ENABLE_C1E_MASK         0x18000000
+#define CPUID_PROCESSOR_SIGNATURE       1
+#define CPUID_XFAM              0x0ff00000
+#define CPUID_XFAM_K8           0x00000000
+#define CPUID_XFAM_10H          0x00100000
+#define CPUID_XFAM_11H          0x00200000
+#define CPUID_XMOD              0x000f0000
+#define CPUID_XMOD_REV_F        0x00040000
+
+/* AMD systems with C1E don't have a working lAPIC timer. Check for that. */
+static __cpuinit int amd_apic_timer_broken(void)
+{
+	u32 lo, hi;
+	u32 eax = cpuid_eax(CPUID_PROCESSOR_SIGNATURE);
+	switch (eax & CPUID_XFAM) {
+	case CPUID_XFAM_K8:
+		if ((eax & CPUID_XMOD) < CPUID_XMOD_REV_F)
+			break;
+	case CPUID_XFAM_10H:
+	case CPUID_XFAM_11H:
+		rdmsr(MSR_K8_ENABLE_C1E, lo, hi);
+		if (lo & ENABLE_C1E_MASK)
+			return 1;
+                break;
+        default:
+                /* err on the side of caution */
+		return 1;
+        }
+	return 0;
+}
+#endif
+
+int force_mwait __cpuinitdata;
+
+static void __cpuinit init_amd(struct cpuinfo_x86 *c)
+{
+	u32 l, h;
+	int mbytes = num_physpages >> (20-PAGE_SHIFT);
+	int r;
+
+#ifdef CONFIG_SMP
+	unsigned long long value;
+
+	/* Disable TLB flush filter by setting HWCR.FFDIS on K8
+	 * bit 6 of msr C001_0015
+	 *
+	 * Errata 63 for SH-B3 steppings
+	 * Errata 122 for all steppings (F+ have it disabled by default)
+	 */
+	if (c->x86 == 15) {
+		rdmsrl(MSR_K7_HWCR, value);
+		value |= 1 << 6;
+		wrmsrl(MSR_K7_HWCR, value);
+	}
+#endif
+
+	/*
+	 *	FIXME: We should handle the K5 here. Set up the write
+	 *	range and also turn on MSR 83 bits 4 and 31 (write alloc,
+	 *	no bus pipeline)
+	 */
+
+	/* Bit 31 in normal CPUID used for nonstandard 3DNow ID;
+	   3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway */
+	clear_bit(0*32+31, c->x86_capability);
+	
+	r = get_model_name(c);
+
+	switch(c->x86)
+	{
+		case 4:
+		/*
+		 * General Systems BIOSen alias the cpu frequency registers
+		 * of the Elan at 0x000df000. Unfortuantly, one of the Linux
+		 * drivers subsequently pokes it, and changes the CPU speed.
+		 * Workaround : Remove the unneeded alias.
+		 */
+#define CBAR		(0xfffc) /* Configuration Base Address  (32-bit) */
+#define CBAR_ENB	(0x80000000)
+#define CBAR_KEY	(0X000000CB)
+			if (c->x86_model==9 || c->x86_model == 10) {
+				if (inl (CBAR) & CBAR_ENB)
+					outl (0 | CBAR_KEY, CBAR);
+			}
+			break;
+		case 5:
+			if( c->x86_model < 6 )
+			{
+				/* Based on AMD doc 20734R - June 2000 */
+				if ( c->x86_model == 0 ) {
+					clear_bit(X86_FEATURE_APIC, c->x86_capability);
+					set_bit(X86_FEATURE_PGE, c->x86_capability);
+				}
+				break;
+			}
+			
+			if ( c->x86_model == 6 && c->x86_mask == 1 ) {
+				const int K6_BUG_LOOP = 1000000;
+				int n;
+				void (*f_vide)(void);
+				unsigned long d, d2;
+				
+				printk(KERN_INFO "AMD K6 stepping B detected - ");
+				
+				/*
+				 * It looks like AMD fixed the 2.6.2 bug and improved indirect 
+				 * calls at the same time.
+				 */
+
+				n = K6_BUG_LOOP;
+				f_vide = vide;
+				rdtscl(d);
+				while (n--) 
+					f_vide();
+				rdtscl(d2);
+				d = d2-d;
+
+				if (d > 20*K6_BUG_LOOP) 
+					printk("system stability may be impaired when more than 32 MB are used.\n");
+				else 
+					printk("probably OK (after B9730xxxx).\n");
+				printk(KERN_INFO "Please see http://membres.lycos.fr/poulot/k6bug.html\n");
+			}
+
+			/* K6 with old style WHCR */
+			if (c->x86_model < 8 ||
+			   (c->x86_model== 8 && c->x86_mask < 8)) {
+				/* We can only write allocate on the low 508Mb */
+				if(mbytes>508)
+					mbytes=508;
+
+				rdmsr(MSR_K6_WHCR, l, h);
+				if ((l&0x0000FFFF)==0) {
+					unsigned long flags;
+					l=(1<<0)|((mbytes/4)<<1);
+					local_irq_save(flags);
+					wbinvd();
+					wrmsr(MSR_K6_WHCR, l, h);
+					local_irq_restore(flags);
+					printk(KERN_INFO "Enabling old style K6 write allocation for %d Mb\n",
+						mbytes);
+				}
+				break;
+			}
+
+			if ((c->x86_model == 8 && c->x86_mask >7) ||
+			     c->x86_model == 9 || c->x86_model == 13) {
+				/* The more serious chips .. */
+
+				if(mbytes>4092)
+					mbytes=4092;
+
+				rdmsr(MSR_K6_WHCR, l, h);
+				if ((l&0xFFFF0000)==0) {
+					unsigned long flags;
+					l=((mbytes>>2)<<22)|(1<<16);
+					local_irq_save(flags);
+					wbinvd();
+					wrmsr(MSR_K6_WHCR, l, h);
+					local_irq_restore(flags);
+					printk(KERN_INFO "Enabling new style K6 write allocation for %d Mb\n",
+						mbytes);
+				}
+
+				/*  Set MTRR capability flag if appropriate */
+				if (c->x86_model == 13 || c->x86_model == 9 ||
+				   (c->x86_model == 8 && c->x86_mask >= 8))
+					set_bit(X86_FEATURE_K6_MTRR, c->x86_capability);
+				break;
+			}
+
+			if (c->x86_model == 10) {
+				/* AMD Geode LX is model 10 */
+				/* placeholder for any needed mods */
+				break;
+			}
+			break;
+		case 6: /* An Athlon/Duron */
+ 
+			/* Bit 15 of Athlon specific MSR 15, needs to be 0
+ 			 * to enable SSE on Palomino/Morgan/Barton CPU's.
+			 * If the BIOS didn't enable it already, enable it here.
+			 */
+			if (c->x86_model >= 6 && c->x86_model <= 10) {
+				if (!cpu_has(c, X86_FEATURE_XMM)) {
+					printk(KERN_INFO "Enabling disabled K7/SSE Support.\n");
+					rdmsr(MSR_K7_HWCR, l, h);
+					l &= ~0x00008000;
+					wrmsr(MSR_K7_HWCR, l, h);
+					set_bit(X86_FEATURE_XMM, c->x86_capability);
+				}
+			}
+
+			/* It's been determined by AMD that Athlons since model 8 stepping 1
+			 * are more robust with CLK_CTL set to 200xxxxx instead of 600xxxxx
+			 * As per AMD technical note 27212 0.2
+			 */
+			if ((c->x86_model == 8 && c->x86_mask>=1) || (c->x86_model > 8)) {
+				rdmsr(MSR_K7_CLK_CTL, l, h);
+				if ((l & 0xfff00000) != 0x20000000) {
+					printk ("CPU: CLK_CTL MSR was %x. Reprogramming to %x\n", l,
+						((l & 0x000fffff)|0x20000000));
+					wrmsr(MSR_K7_CLK_CTL, (l & 0x000fffff)|0x20000000, h);
+				}
+			}
+			break;
+	}
+
+	switch (c->x86) {
+	case 15:
+	/* Use K8 tuning for Fam10h and Fam11h */
+	case 0x10:
+	case 0x11:
+		set_bit(X86_FEATURE_K8, c->x86_capability);
+		break;
+	case 6:
+		set_bit(X86_FEATURE_K7, c->x86_capability); 
+		break;
+	}
+	if (c->x86 >= 6)
+		set_bit(X86_FEATURE_FXSAVE_LEAK, c->x86_capability);
+
+	display_cacheinfo(c);
+
+	if (cpuid_eax(0x80000000) >= 0x80000008) {
+		c->x86_max_cores = (cpuid_ecx(0x80000008) & 0xff) + 1;
+	}
+
+	if (cpuid_eax(0x80000000) >= 0x80000007) {
+		c->x86_power = cpuid_edx(0x80000007);
+		if (c->x86_power & (1<<8))
+			set_bit(X86_FEATURE_CONSTANT_TSC, c->x86_capability);
+	}
+
+#ifdef CONFIG_X86_HT
+	/*
+	 * On a AMD multi core setup the lower bits of the APIC id
+	 * distingush the cores.
+	 */
+	if (c->x86_max_cores > 1) {
+		int cpu = smp_processor_id();
+		unsigned bits = (cpuid_ecx(0x80000008) >> 12) & 0xf;
+
+		if (bits == 0) {
+			while ((1 << bits) < c->x86_max_cores)
+				bits++;
+		}
+		c->cpu_core_id = c->phys_proc_id & ((1<<bits)-1);
+		c->phys_proc_id >>= bits;
+		printk(KERN_INFO "CPU %d(%d) -> Core %d\n",
+		       cpu, c->x86_max_cores, c->cpu_core_id);
+	}
+#endif
+
+	if (cpuid_eax(0x80000000) >= 0x80000006) {
+		if ((c->x86 == 0x10) && (cpuid_edx(0x80000006) & 0xf000))
+			num_cache_leaves = 4;
+		else
+			num_cache_leaves = 3;
+	}
+
+#ifdef CONFIG_X86_LOCAL_APIC
+	if (amd_apic_timer_broken())
+		local_apic_timer_disabled = 1;
+#endif
+
+	if (c->x86 == 0x10 && !force_mwait)
+		clear_bit(X86_FEATURE_MWAIT, c->x86_capability);
+
+	/* K6s reports MCEs but don't actually have all the MSRs */
+	if (c->x86 < 6)
+		clear_bit(X86_FEATURE_MCE, c->x86_capability);
+}
+
+static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 * c, unsigned int size)
+{
+	/* AMD errata T13 (order #21922) */
+	if ((c->x86 == 6)) {
+		if (c->x86_model == 3 && c->x86_mask == 0)	/* Duron Rev A0 */
+			size = 64;
+		if (c->x86_model == 4 &&
+		    (c->x86_mask==0 || c->x86_mask==1))	/* Tbird rev A1/A2 */
+			size = 256;
+	}
+	return size;
+}
+
+static struct cpu_dev amd_cpu_dev __cpuinitdata = {
+	.c_vendor	= "AMD",
+	.c_ident 	= { "AuthenticAMD" },
+	.c_models = {
+		{ .vendor = X86_VENDOR_AMD, .family = 4, .model_names =
+		  {
+			  [3] = "486 DX/2",
+			  [7] = "486 DX/2-WB",
+			  [8] = "486 DX/4", 
+			  [9] = "486 DX/4-WB", 
+			  [14] = "Am5x86-WT",
+			  [15] = "Am5x86-WB" 
+		  }
+		},
+	},
+	.c_init		= init_amd,
+	.c_size_cache	= amd_size_cache,
+};
+
+int __init amd_init_cpu(void)
+{
+	cpu_devs[X86_VENDOR_AMD] = &amd_cpu_dev;
+	return 0;
+}
diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c
new file mode 100644
index 00000000000..59266f03d1c
--- /dev/null
+++ b/arch/x86/kernel/cpu/bugs.c
@@ -0,0 +1,192 @@
+/*
+ *  arch/i386/cpu/bugs.c
+ *
+ *  Copyright (C) 1994  Linus Torvalds
+ *
+ *  Cyrix stuff, June 1998 by:
+ *	- Rafael R. Reilova (moved everything from head.S),
+ *        <rreilova@ececs.uc.edu>
+ *	- Channing Corn (tests & fixes),
+ *	- Andrew D. Balsa (code cleanup).
+ */
+#include <linux/init.h>
+#include <linux/utsname.h>
+#include <asm/bugs.h>
+#include <asm/processor.h>
+#include <asm/i387.h>
+#include <asm/msr.h>
+#include <asm/paravirt.h>
+#include <asm/alternative.h>
+
+static int __init no_halt(char *s)
+{
+	boot_cpu_data.hlt_works_ok = 0;
+	return 1;
+}
+
+__setup("no-hlt", no_halt);
+
+static int __init mca_pentium(char *s)
+{
+	mca_pentium_flag = 1;
+	return 1;
+}
+
+__setup("mca-pentium", mca_pentium);
+
+static int __init no_387(char *s)
+{
+	boot_cpu_data.hard_math = 0;
+	write_cr0(0xE | read_cr0());
+	return 1;
+}
+
+__setup("no387", no_387);
+
+static double __initdata x = 4195835.0;
+static double __initdata y = 3145727.0;
+
+/*
+ * This used to check for exceptions..
+ * However, it turns out that to support that,
+ * the XMM trap handlers basically had to
+ * be buggy. So let's have a correct XMM trap
+ * handler, and forget about printing out
+ * some status at boot.
+ *
+ * We should really only care about bugs here
+ * anyway. Not features.
+ */
+static void __init check_fpu(void)
+{
+	if (!boot_cpu_data.hard_math) {
+#ifndef CONFIG_MATH_EMULATION
+		printk(KERN_EMERG "No coprocessor found and no math emulation present.\n");
+		printk(KERN_EMERG "Giving up.\n");
+		for (;;) ;
+#endif
+		return;
+	}
+
+/* trap_init() enabled FXSR and company _before_ testing for FP problems here. */
+	/* Test for the divl bug.. */
+	__asm__("fninit\n\t"
+		"fldl %1\n\t"
+		"fdivl %2\n\t"
+		"fmull %2\n\t"
+		"fldl %1\n\t"
+		"fsubp %%st,%%st(1)\n\t"
+		"fistpl %0\n\t"
+		"fwait\n\t"
+		"fninit"
+		: "=m" (*&boot_cpu_data.fdiv_bug)
+		: "m" (*&x), "m" (*&y));
+	if (boot_cpu_data.fdiv_bug)
+		printk("Hmm, FPU with FDIV bug.\n");
+}
+
+static void __init check_hlt(void)
+{
+	if (paravirt_enabled())
+		return;
+
+	printk(KERN_INFO "Checking 'hlt' instruction... ");
+	if (!boot_cpu_data.hlt_works_ok) {
+		printk("disabled\n");
+		return;
+	}
+	halt();
+	halt();
+	halt();
+	halt();
+	printk("OK.\n");
+}
+
+/*
+ *	Most 386 processors have a bug where a POPAD can lock the
+ *	machine even from user space.
+ */
+
+static void __init check_popad(void)
+{
+#ifndef CONFIG_X86_POPAD_OK
+	int res, inp = (int) &res;
+
+	printk(KERN_INFO "Checking for popad bug... ");
+	__asm__ __volatile__(
+	  "movl $12345678,%%eax; movl $0,%%edi; pusha; popa; movl (%%edx,%%edi),%%ecx "
+	  : "=&a" (res)
+	  : "d" (inp)
+	  : "ecx", "edi" );
+	/* If this fails, it means that any user program may lock the CPU hard. Too bad. */
+	if (res != 12345678) printk( "Buggy.\n" );
+		        else printk( "OK.\n" );
+#endif
+}
+
+/*
+ * Check whether we are able to run this kernel safely on SMP.
+ *
+ * - In order to run on a i386, we need to be compiled for i386
+ *   (for due to lack of "invlpg" and working WP on a i386)
+ * - In order to run on anything without a TSC, we need to be
+ *   compiled for a i486.
+ * - In order to support the local APIC on a buggy Pentium machine,
+ *   we need to be compiled with CONFIG_X86_GOOD_APIC disabled,
+ *   which happens implicitly if compiled for a Pentium or lower
+ *   (unless an advanced selection of CPU features is used) as an
+ *   otherwise config implies a properly working local APIC without
+ *   the need to do extra reads from the APIC.
+*/
+
+static void __init check_config(void)
+{
+/*
+ * We'd better not be a i386 if we're configured to use some
+ * i486+ only features! (WP works in supervisor mode and the
+ * new "invlpg" and "bswap" instructions)
+ */
+#if defined(CONFIG_X86_WP_WORKS_OK) || defined(CONFIG_X86_INVLPG) || defined(CONFIG_X86_BSWAP)
+	if (boot_cpu_data.x86 == 3)
+		panic("Kernel requires i486+ for 'invlpg' and other features");
+#endif
+
+/*
+ * If we configured ourselves for a TSC, we'd better have one!
+ */
+#ifdef CONFIG_X86_TSC
+	if (!cpu_has_tsc && !tsc_disable)
+		panic("Kernel compiled for Pentium+, requires TSC feature!");
+#endif
+
+/*
+ * If we were told we had a good local APIC, check for buggy Pentia,
+ * i.e. all B steppings and the C2 stepping of P54C when using their
+ * integrated APIC (see 11AP erratum in "Pentium Processor
+ * Specification Update").
+ */
+#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86_GOOD_APIC)
+	if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL
+	    && cpu_has_apic
+	    && boot_cpu_data.x86 == 5
+	    && boot_cpu_data.x86_model == 2
+	    && (boot_cpu_data.x86_mask < 6 || boot_cpu_data.x86_mask == 11))
+		panic("Kernel compiled for PMMX+, assumes a local APIC without the read-before-write bug!");
+#endif
+}
+
+
+void __init check_bugs(void)
+{
+	identify_boot_cpu();
+#ifndef CONFIG_SMP
+	printk("CPU: ");
+	print_cpu_info(&boot_cpu_data);
+#endif
+	check_config();
+	check_fpu();
+	check_hlt();
+	check_popad();
+	init_utsname()->machine[1] = '0' + (boot_cpu_data.x86 > 6 ? 6 : boot_cpu_data.x86);
+	alternative_instructions();
+}
diff --git a/arch/x86/kernel/cpu/centaur.c b/arch/x86/kernel/cpu/centaur.c
new file mode 100644
index 00000000000..473eac883c7
--- /dev/null
+++ b/arch/x86/kernel/cpu/centaur.c
@@ -0,0 +1,471 @@
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/bitops.h>
+#include <asm/processor.h>
+#include <asm/msr.h>
+#include <asm/e820.h>
+#include <asm/mtrr.h>
+#include "cpu.h"
+
+#ifdef CONFIG_X86_OOSTORE
+
+static u32 __cpuinit power2(u32 x)
+{
+	u32 s=1;
+	while(s<=x)
+		s<<=1;
+	return s>>=1;
+}
+
+
+/*
+ *	Set up an actual MCR
+ */
+ 
+static void __cpuinit centaur_mcr_insert(int reg, u32 base, u32 size, int key)
+{
+	u32 lo, hi;
+	
+	hi = base & ~0xFFF;
+	lo = ~(size-1);		/* Size is a power of 2 so this makes a mask */
+	lo &= ~0xFFF;		/* Remove the ctrl value bits */
+	lo |= key;		/* Attribute we wish to set */
+	wrmsr(reg+MSR_IDT_MCR0, lo, hi);
+	mtrr_centaur_report_mcr(reg, lo, hi);	/* Tell the mtrr driver */
+}
+
+/*
+ *	Figure what we can cover with MCR's
+ *
+ *	Shortcut: We know you can't put 4Gig of RAM on a winchip
+ */
+
+static u32 __cpuinit ramtop(void)		/* 16388 */
+{
+	int i;
+	u32 top = 0;
+	u32 clip = 0xFFFFFFFFUL;
+	
+	for (i = 0; i < e820.nr_map; i++) {
+		unsigned long start, end;
+
+		if (e820.map[i].addr > 0xFFFFFFFFUL)
+			continue;
+		/*
+		 *	Don't MCR over reserved space. Ignore the ISA hole
+		 *	we frob around that catastrophy already
+		 */
+		 			
+		if (e820.map[i].type == E820_RESERVED)
+		{
+			if(e820.map[i].addr >= 0x100000UL && e820.map[i].addr < clip)
+				clip = e820.map[i].addr;
+			continue;
+		}
+		start = e820.map[i].addr;
+		end = e820.map[i].addr + e820.map[i].size;
+		if (start >= end)
+			continue;
+		if (end > top)
+			top = end;
+	}
+	/* Everything below 'top' should be RAM except for the ISA hole.
+	   Because of the limited MCR's we want to map NV/ACPI into our
+	   MCR range for gunk in RAM 
+	   
+	   Clip might cause us to MCR insufficient RAM but that is an
+	   acceptable failure mode and should only bite obscure boxes with
+	   a VESA hole at 15Mb
+	   
+	   The second case Clip sometimes kicks in is when the EBDA is marked
+	   as reserved. Again we fail safe with reasonable results
+	*/
+	
+	if(top>clip)
+		top=clip;
+		
+	return top;
+}
+
+/*
+ *	Compute a set of MCR's to give maximum coverage
+ */
+
+static int __cpuinit centaur_mcr_compute(int nr, int key)
+{
+	u32 mem = ramtop();
+	u32 root = power2(mem);
+	u32 base = root;
+	u32 top = root;
+	u32 floor = 0;
+	int ct = 0;
+	
+	while(ct<nr)
+	{
+		u32 fspace = 0;
+
+		/*
+		 *	Find the largest block we will fill going upwards
+		 */
+
+		u32 high = power2(mem-top);	
+
+		/*
+		 *	Find the largest block we will fill going downwards
+		 */
+
+		u32 low = base/2;
+
+		/*
+		 *	Don't fill below 1Mb going downwards as there
+		 *	is an ISA hole in the way.
+		 */		
+		 
+		if(base <= 1024*1024)
+			low = 0;
+			
+		/*
+		 *	See how much space we could cover by filling below
+		 *	the ISA hole
+		 */
+		 
+		if(floor == 0)
+			fspace = 512*1024;
+		else if(floor ==512*1024)
+			fspace = 128*1024;
+
+		/* And forget ROM space */
+		
+		/*
+		 *	Now install the largest coverage we get
+		 */
+		 
+		if(fspace > high && fspace > low)
+		{
+			centaur_mcr_insert(ct, floor, fspace, key);
+			floor += fspace;
+		}
+		else if(high > low)
+		{
+			centaur_mcr_insert(ct, top, high, key);
+			top += high;
+		}
+		else if(low > 0)
+		{
+			base -= low;
+			centaur_mcr_insert(ct, base, low, key);
+		}
+		else break;
+		ct++;
+	}
+	/*
+	 *	We loaded ct values. We now need to set the mask. The caller
+	 *	must do this bit.
+	 */
+	 
+	return ct;
+}
+
+static void __cpuinit centaur_create_optimal_mcr(void)
+{
+	int i;
+	/*
+	 *	Allocate up to 6 mcrs to mark as much of ram as possible
+	 *	as write combining and weak write ordered.
+	 *
+	 *	To experiment with: Linux never uses stack operations for 
+	 *	mmio spaces so we could globally enable stack operation wc
+	 *
+	 *	Load the registers with type 31 - full write combining, all
+	 *	writes weakly ordered.
+	 */
+	int used = centaur_mcr_compute(6, 31);
+
+	/*
+	 *	Wipe unused MCRs
+	 */
+	 
+	for(i=used;i<8;i++)
+		wrmsr(MSR_IDT_MCR0+i, 0, 0);
+}
+
+static void __cpuinit winchip2_create_optimal_mcr(void)
+{
+	u32 lo, hi;
+	int i;
+
+	/*
+	 *	Allocate up to 6 mcrs to mark as much of ram as possible
+	 *	as write combining, weak store ordered.
+	 *
+	 *	Load the registers with type 25
+	 *		8	-	weak write ordering
+	 *		16	-	weak read ordering
+	 *		1	-	write combining
+	 */
+
+	int used = centaur_mcr_compute(6, 25);
+	
+	/*
+	 *	Mark the registers we are using.
+	 */
+	 
+	rdmsr(MSR_IDT_MCR_CTRL, lo, hi);
+	for(i=0;i<used;i++)
+		lo|=1<<(9+i);
+	wrmsr(MSR_IDT_MCR_CTRL, lo, hi);
+	
+	/*
+	 *	Wipe unused MCRs
+	 */
+	 
+	for(i=used;i<8;i++)
+		wrmsr(MSR_IDT_MCR0+i, 0, 0);
+}
+
+/*
+ *	Handle the MCR key on the Winchip 2.
+ */
+
+static void __cpuinit winchip2_unprotect_mcr(void)
+{
+	u32 lo, hi;
+	u32 key;
+	
+	rdmsr(MSR_IDT_MCR_CTRL, lo, hi);
+	lo&=~0x1C0;	/* blank bits 8-6 */
+	key = (lo>>17) & 7;
+	lo |= key<<6;	/* replace with unlock key */
+	wrmsr(MSR_IDT_MCR_CTRL, lo, hi);
+}
+
+static void __cpuinit winchip2_protect_mcr(void)
+{
+	u32 lo, hi;
+	
+	rdmsr(MSR_IDT_MCR_CTRL, lo, hi);
+	lo&=~0x1C0;	/* blank bits 8-6 */
+	wrmsr(MSR_IDT_MCR_CTRL, lo, hi);
+}
+#endif /* CONFIG_X86_OOSTORE */
+
+#define ACE_PRESENT	(1 << 6)
+#define ACE_ENABLED	(1 << 7)
+#define ACE_FCR		(1 << 28)	/* MSR_VIA_FCR */
+
+#define RNG_PRESENT	(1 << 2)
+#define RNG_ENABLED	(1 << 3)
+#define RNG_ENABLE	(1 << 6)	/* MSR_VIA_RNG */
+
+static void __cpuinit init_c3(struct cpuinfo_x86 *c)
+{
+	u32  lo, hi;
+
+	/* Test for Centaur Extended Feature Flags presence */
+	if (cpuid_eax(0xC0000000) >= 0xC0000001) {
+		u32 tmp = cpuid_edx(0xC0000001);
+
+		/* enable ACE unit, if present and disabled */
+		if ((tmp & (ACE_PRESENT | ACE_ENABLED)) == ACE_PRESENT) {
+			rdmsr (MSR_VIA_FCR, lo, hi);
+			lo |= ACE_FCR;		/* enable ACE unit */
+			wrmsr (MSR_VIA_FCR, lo, hi);
+			printk(KERN_INFO "CPU: Enabled ACE h/w crypto\n");
+		}
+
+		/* enable RNG unit, if present and disabled */
+		if ((tmp & (RNG_PRESENT | RNG_ENABLED)) == RNG_PRESENT) {
+			rdmsr (MSR_VIA_RNG, lo, hi);
+			lo |= RNG_ENABLE;	/* enable RNG unit */
+			wrmsr (MSR_VIA_RNG, lo, hi);
+			printk(KERN_INFO "CPU: Enabled h/w RNG\n");
+		}
+
+		/* store Centaur Extended Feature Flags as
+		 * word 5 of the CPU capability bit array
+		 */
+		c->x86_capability[5] = cpuid_edx(0xC0000001);
+	}
+
+	/* Cyrix III family needs CX8 & PGE explicity enabled. */
+	if (c->x86_model >=6 && c->x86_model <= 9) {
+		rdmsr (MSR_VIA_FCR, lo, hi);
+		lo |= (1<<1 | 1<<7);
+		wrmsr (MSR_VIA_FCR, lo, hi);
+		set_bit(X86_FEATURE_CX8, c->x86_capability);
+	}
+
+	/* Before Nehemiah, the C3's had 3dNOW! */
+	if (c->x86_model >=6 && c->x86_model <9)
+		set_bit(X86_FEATURE_3DNOW, c->x86_capability);
+
+	get_model_name(c);
+	display_cacheinfo(c);
+}
+
+static void __cpuinit init_centaur(struct cpuinfo_x86 *c)
+{
+	enum {
+		ECX8=1<<1,
+		EIERRINT=1<<2,
+		DPM=1<<3,
+		DMCE=1<<4,
+		DSTPCLK=1<<5,
+		ELINEAR=1<<6,
+		DSMC=1<<7,
+		DTLOCK=1<<8,
+		EDCTLB=1<<8,
+		EMMX=1<<9,
+		DPDC=1<<11,
+		EBRPRED=1<<12,
+		DIC=1<<13,
+		DDC=1<<14,
+		DNA=1<<15,
+		ERETSTK=1<<16,
+		E2MMX=1<<19,
+		EAMD3D=1<<20,
+	};
+
+	char *name;
+	u32  fcr_set=0;
+	u32  fcr_clr=0;
+	u32  lo,hi,newlo;
+	u32  aa,bb,cc,dd;
+
+	/* Bit 31 in normal CPUID used for nonstandard 3DNow ID;
+	   3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway */
+	clear_bit(0*32+31, c->x86_capability);
+
+	switch (c->x86) {
+
+		case 5:
+			switch(c->x86_model) {
+			case 4:
+				name="C6";
+				fcr_set=ECX8|DSMC|EDCTLB|EMMX|ERETSTK;
+				fcr_clr=DPDC;
+				printk(KERN_NOTICE "Disabling bugged TSC.\n");
+				clear_bit(X86_FEATURE_TSC, c->x86_capability);
+#ifdef CONFIG_X86_OOSTORE
+				centaur_create_optimal_mcr();
+				/* Enable
+					write combining on non-stack, non-string
+					write combining on string, all types
+					weak write ordering 
+					
+				   The C6 original lacks weak read order 
+				   
+				   Note 0x120 is write only on Winchip 1 */
+				   
+				wrmsr(MSR_IDT_MCR_CTRL, 0x01F0001F, 0);
+#endif				
+				break;
+			case 8:
+				switch(c->x86_mask) {
+				default:
+					name="2";
+					break;
+				case 7 ... 9:
+					name="2A";
+					break;
+				case 10 ... 15:
+					name="2B";
+					break;
+				}
+				fcr_set=ECX8|DSMC|DTLOCK|EMMX|EBRPRED|ERETSTK|E2MMX|EAMD3D;
+				fcr_clr=DPDC;
+#ifdef CONFIG_X86_OOSTORE
+				winchip2_unprotect_mcr();
+				winchip2_create_optimal_mcr();
+				rdmsr(MSR_IDT_MCR_CTRL, lo, hi);
+				/* Enable
+					write combining on non-stack, non-string
+					write combining on string, all types
+					weak write ordering 
+				*/
+				lo|=31;				
+				wrmsr(MSR_IDT_MCR_CTRL, lo, hi);
+				winchip2_protect_mcr();
+#endif
+				break;
+			case 9:
+				name="3";
+				fcr_set=ECX8|DSMC|DTLOCK|EMMX|EBRPRED|ERETSTK|E2MMX|EAMD3D;
+				fcr_clr=DPDC;
+#ifdef CONFIG_X86_OOSTORE
+				winchip2_unprotect_mcr();
+				winchip2_create_optimal_mcr();
+				rdmsr(MSR_IDT_MCR_CTRL, lo, hi);
+				/* Enable
+					write combining on non-stack, non-string
+					write combining on string, all types
+					weak write ordering 
+				*/
+				lo|=31;				
+				wrmsr(MSR_IDT_MCR_CTRL, lo, hi);
+				winchip2_protect_mcr();
+#endif
+				break;
+			default:
+				name="??";
+			}
+
+			rdmsr(MSR_IDT_FCR1, lo, hi);
+			newlo=(lo|fcr_set) & (~fcr_clr);
+
+			if (newlo!=lo) {
+				printk(KERN_INFO "Centaur FCR was 0x%X now 0x%X\n", lo, newlo );
+				wrmsr(MSR_IDT_FCR1, newlo, hi );
+			} else {
+				printk(KERN_INFO "Centaur FCR is 0x%X\n",lo);
+			}
+			/* Emulate MTRRs using Centaur's MCR. */
+			set_bit(X86_FEATURE_CENTAUR_MCR, c->x86_capability);
+			/* Report CX8 */
+			set_bit(X86_FEATURE_CX8, c->x86_capability);
+			/* Set 3DNow! on Winchip 2 and above. */
+			if (c->x86_model >=8)
+				set_bit(X86_FEATURE_3DNOW, c->x86_capability);
+			/* See if we can find out some more. */
+			if ( cpuid_eax(0x80000000) >= 0x80000005 ) {
+				/* Yes, we can. */
+				cpuid(0x80000005,&aa,&bb,&cc,&dd);
+				/* Add L1 data and code cache sizes. */
+				c->x86_cache_size = (cc>>24)+(dd>>24);
+			}
+			sprintf( c->x86_model_id, "WinChip %s", name );
+			break;
+
+		case 6:
+			init_c3(c);
+			break;
+	}
+}
+
+static unsigned int __cpuinit centaur_size_cache(struct cpuinfo_x86 * c, unsigned int size)
+{
+	/* VIA C3 CPUs (670-68F) need further shifting. */
+	if ((c->x86 == 6) && ((c->x86_model == 7) || (c->x86_model == 8)))
+		size >>= 8;
+
+	/* VIA also screwed up Nehemiah stepping 1, and made
+	   it return '65KB' instead of '64KB'
+	   - Note, it seems this may only be in engineering samples. */
+	if ((c->x86==6) && (c->x86_model==9) && (c->x86_mask==1) && (size==65))
+		size -=1;
+
+	return size;
+}
+
+static struct cpu_dev centaur_cpu_dev __cpuinitdata = {
+	.c_vendor	= "Centaur",
+	.c_ident	= { "CentaurHauls" },
+	.c_init		= init_centaur,
+	.c_size_cache	= centaur_size_cache,
+};
+
+int __init centaur_init_cpu(void)
+{
+	cpu_devs[X86_VENDOR_CENTAUR] = &centaur_cpu_dev;
+	return 0;
+}
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
new file mode 100644
index 00000000000..d506201d397
--- /dev/null
+++ b/arch/x86/kernel/cpu/common.c
@@ -0,0 +1,733 @@
+#include <linux/init.h>
+#include <linux/string.h>
+#include <linux/delay.h>
+#include <linux/smp.h>
+#include <linux/module.h>
+#include <linux/percpu.h>
+#include <linux/bootmem.h>
+#include <asm/semaphore.h>
+#include <asm/processor.h>
+#include <asm/i387.h>
+#include <asm/msr.h>
+#include <asm/io.h>
+#include <asm/mmu_context.h>
+#include <asm/mtrr.h>
+#include <asm/mce.h>
+#ifdef CONFIG_X86_LOCAL_APIC
+#include <asm/mpspec.h>
+#include <asm/apic.h>
+#include <mach_apic.h>
+#endif
+
+#include "cpu.h"
+
+DEFINE_PER_CPU(struct gdt_page, gdt_page) = { .gdt = {
+	[GDT_ENTRY_KERNEL_CS] = { 0x0000ffff, 0x00cf9a00 },
+	[GDT_ENTRY_KERNEL_DS] = { 0x0000ffff, 0x00cf9200 },
+	[GDT_ENTRY_DEFAULT_USER_CS] = { 0x0000ffff, 0x00cffa00 },
+	[GDT_ENTRY_DEFAULT_USER_DS] = { 0x0000ffff, 0x00cff200 },
+	/*
+	 * Segments used for calling PnP BIOS have byte granularity.
+	 * They code segments and data segments have fixed 64k limits,
+	 * the transfer segment sizes are set at run time.
+	 */
+	[GDT_ENTRY_PNPBIOS_CS32] = { 0x0000ffff, 0x00409a00 },/* 32-bit code */
+	[GDT_ENTRY_PNPBIOS_CS16] = { 0x0000ffff, 0x00009a00 },/* 16-bit code */
+	[GDT_ENTRY_PNPBIOS_DS] = { 0x0000ffff, 0x00009200 }, /* 16-bit data */
+	[GDT_ENTRY_PNPBIOS_TS1] = { 0x00000000, 0x00009200 },/* 16-bit data */
+	[GDT_ENTRY_PNPBIOS_TS2] = { 0x00000000, 0x00009200 },/* 16-bit data */
+	/*
+	 * The APM segments have byte granularity and their bases
+	 * are set at run time.  All have 64k limits.
+	 */
+	[GDT_ENTRY_APMBIOS_BASE] = { 0x0000ffff, 0x00409a00 },/* 32-bit code */
+	/* 16-bit code */
+	[GDT_ENTRY_APMBIOS_BASE+1] = { 0x0000ffff, 0x00009a00 },
+	[GDT_ENTRY_APMBIOS_BASE+2] = { 0x0000ffff, 0x00409200 }, /* data */
+
+	[GDT_ENTRY_ESPFIX_SS] = { 0x00000000, 0x00c09200 },
+	[GDT_ENTRY_PERCPU] = { 0x00000000, 0x00000000 },
+} };
+EXPORT_PER_CPU_SYMBOL_GPL(gdt_page);
+
+static int cachesize_override __cpuinitdata = -1;
+static int disable_x86_fxsr __cpuinitdata;
+static int disable_x86_serial_nr __cpuinitdata = 1;
+static int disable_x86_sep __cpuinitdata;
+
+struct cpu_dev * cpu_devs[X86_VENDOR_NUM] = {};
+
+extern int disable_pse;
+
+static void __cpuinit default_init(struct cpuinfo_x86 * c)
+{
+	/* Not much we can do here... */
+	/* Check if at least it has cpuid */
+	if (c->cpuid_level == -1) {
+		/* No cpuid. It must be an ancient CPU */
+		if (c->x86 == 4)
+			strcpy(c->x86_model_id, "486");
+		else if (c->x86 == 3)
+			strcpy(c->x86_model_id, "386");
+	}
+}
+
+static struct cpu_dev __cpuinitdata default_cpu = {
+	.c_init	= default_init,
+	.c_vendor = "Unknown",
+};
+static struct cpu_dev * this_cpu __cpuinitdata = &default_cpu;
+
+static int __init cachesize_setup(char *str)
+{
+	get_option (&str, &cachesize_override);
+	return 1;
+}
+__setup("cachesize=", cachesize_setup);
+
+int __cpuinit get_model_name(struct cpuinfo_x86 *c)
+{
+	unsigned int *v;
+	char *p, *q;
+
+	if (cpuid_eax(0x80000000) < 0x80000004)
+		return 0;
+
+	v = (unsigned int *) c->x86_model_id;
+	cpuid(0x80000002, &v[0], &v[1], &v[2], &v[3]);
+	cpuid(0x80000003, &v[4], &v[5], &v[6], &v[7]);
+	cpuid(0x80000004, &v[8], &v[9], &v[10], &v[11]);
+	c->x86_model_id[48] = 0;
+
+	/* Intel chips right-justify this string for some dumb reason;
+	   undo that brain damage */
+	p = q = &c->x86_model_id[0];
+	while ( *p == ' ' )
+	     p++;
+	if ( p != q ) {
+	     while ( *p )
+		  *q++ = *p++;
+	     while ( q <= &c->x86_model_id[48] )
+		  *q++ = '\0';	/* Zero-pad the rest */
+	}
+
+	return 1;
+}
+
+
+void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c)
+{
+	unsigned int n, dummy, ecx, edx, l2size;
+
+	n = cpuid_eax(0x80000000);
+
+	if (n >= 0x80000005) {
+		cpuid(0x80000005, &dummy, &dummy, &ecx, &edx);
+		printk(KERN_INFO "CPU: L1 I Cache: %dK (%d bytes/line), D cache %dK (%d bytes/line)\n",
+			edx>>24, edx&0xFF, ecx>>24, ecx&0xFF);
+		c->x86_cache_size=(ecx>>24)+(edx>>24);	
+	}
+
+	if (n < 0x80000006)	/* Some chips just has a large L1. */
+		return;
+
+	ecx = cpuid_ecx(0x80000006);
+	l2size = ecx >> 16;
+	
+	/* do processor-specific cache resizing */
+	if (this_cpu->c_size_cache)
+		l2size = this_cpu->c_size_cache(c,l2size);
+
+	/* Allow user to override all this if necessary. */
+	if (cachesize_override != -1)
+		l2size = cachesize_override;
+
+	if ( l2size == 0 )
+		return;		/* Again, no L2 cache is possible */
+
+	c->x86_cache_size = l2size;
+
+	printk(KERN_INFO "CPU: L2 Cache: %dK (%d bytes/line)\n",
+	       l2size, ecx & 0xFF);
+}
+
+/* Naming convention should be: <Name> [(<Codename>)] */
+/* This table only is used unless init_<vendor>() below doesn't set it; */
+/* in particular, if CPUID levels 0x80000002..4 are supported, this isn't used */
+
+/* Look up CPU names by table lookup. */
+static char __cpuinit *table_lookup_model(struct cpuinfo_x86 *c)
+{
+	struct cpu_model_info *info;
+
+	if ( c->x86_model >= 16 )
+		return NULL;	/* Range check */
+
+	if (!this_cpu)
+		return NULL;
+
+	info = this_cpu->c_models;
+
+	while (info && info->family) {
+		if (info->family == c->x86)
+			return info->model_names[c->x86_model];
+		info++;
+	}
+	return NULL;		/* Not found */
+}
+
+
+static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c, int early)
+{
+	char *v = c->x86_vendor_id;
+	int i;
+	static int printed;
+
+	for (i = 0; i < X86_VENDOR_NUM; i++) {
+		if (cpu_devs[i]) {
+			if (!strcmp(v,cpu_devs[i]->c_ident[0]) ||
+			    (cpu_devs[i]->c_ident[1] && 
+			     !strcmp(v,cpu_devs[i]->c_ident[1]))) {
+				c->x86_vendor = i;
+				if (!early)
+					this_cpu = cpu_devs[i];
+				return;
+			}
+		}
+	}
+	if (!printed) {
+		printed++;
+		printk(KERN_ERR "CPU: Vendor unknown, using generic init.\n");
+		printk(KERN_ERR "CPU: Your system may be unstable.\n");
+	}
+	c->x86_vendor = X86_VENDOR_UNKNOWN;
+	this_cpu = &default_cpu;
+}
+
+
+static int __init x86_fxsr_setup(char * s)
+{
+	/* Tell all the other CPU's to not use it... */
+	disable_x86_fxsr = 1;
+
+	/*
+	 * ... and clear the bits early in the boot_cpu_data
+	 * so that the bootup process doesn't try to do this
+	 * either.
+	 */
+	clear_bit(X86_FEATURE_FXSR, boot_cpu_data.x86_capability);
+	clear_bit(X86_FEATURE_XMM, boot_cpu_data.x86_capability);
+	return 1;
+}
+__setup("nofxsr", x86_fxsr_setup);
+
+
+static int __init x86_sep_setup(char * s)
+{
+	disable_x86_sep = 1;
+	return 1;
+}
+__setup("nosep", x86_sep_setup);
+
+
+/* Standard macro to see if a specific flag is changeable */
+static inline int flag_is_changeable_p(u32 flag)
+{
+	u32 f1, f2;
+
+	asm("pushfl\n\t"
+	    "pushfl\n\t"
+	    "popl %0\n\t"
+	    "movl %0,%1\n\t"
+	    "xorl %2,%0\n\t"
+	    "pushl %0\n\t"
+	    "popfl\n\t"
+	    "pushfl\n\t"
+	    "popl %0\n\t"
+	    "popfl\n\t"
+	    : "=&r" (f1), "=&r" (f2)
+	    : "ir" (flag));
+
+	return ((f1^f2) & flag) != 0;
+}
+
+
+/* Probe for the CPUID instruction */
+static int __cpuinit have_cpuid_p(void)
+{
+	return flag_is_changeable_p(X86_EFLAGS_ID);
+}
+
+void __init cpu_detect(struct cpuinfo_x86 *c)
+{
+	/* Get vendor name */
+	cpuid(0x00000000, &c->cpuid_level,
+	      (int *)&c->x86_vendor_id[0],
+	      (int *)&c->x86_vendor_id[8],
+	      (int *)&c->x86_vendor_id[4]);
+
+	c->x86 = 4;
+	if (c->cpuid_level >= 0x00000001) {
+		u32 junk, tfms, cap0, misc;
+		cpuid(0x00000001, &tfms, &misc, &junk, &cap0);
+		c->x86 = (tfms >> 8) & 15;
+		c->x86_model = (tfms >> 4) & 15;
+		if (c->x86 == 0xf)
+			c->x86 += (tfms >> 20) & 0xff;
+		if (c->x86 >= 0x6)
+			c->x86_model += ((tfms >> 16) & 0xF) << 4;
+		c->x86_mask = tfms & 15;
+		if (cap0 & (1<<19))
+			c->x86_cache_alignment = ((misc >> 8) & 0xff) * 8;
+	}
+}
+
+/* Do minimum CPU detection early.
+   Fields really needed: vendor, cpuid_level, family, model, mask, cache alignment.
+   The others are not touched to avoid unwanted side effects.
+
+   WARNING: this function is only called on the BP.  Don't add code here
+   that is supposed to run on all CPUs. */
+static void __init early_cpu_detect(void)
+{
+	struct cpuinfo_x86 *c = &boot_cpu_data;
+
+	c->x86_cache_alignment = 32;
+
+	if (!have_cpuid_p())
+		return;
+
+	cpu_detect(c);
+
+	get_cpu_vendor(c, 1);
+}
+
+static void __cpuinit generic_identify(struct cpuinfo_x86 * c)
+{
+	u32 tfms, xlvl;
+	int ebx;
+
+	if (have_cpuid_p()) {
+		/* Get vendor name */
+		cpuid(0x00000000, &c->cpuid_level,
+		      (int *)&c->x86_vendor_id[0],
+		      (int *)&c->x86_vendor_id[8],
+		      (int *)&c->x86_vendor_id[4]);
+		
+		get_cpu_vendor(c, 0);
+		/* Initialize the standard set of capabilities */
+		/* Note that the vendor-specific code below might override */
+	
+		/* Intel-defined flags: level 0x00000001 */
+		if ( c->cpuid_level >= 0x00000001 ) {
+			u32 capability, excap;
+			cpuid(0x00000001, &tfms, &ebx, &excap, &capability);
+			c->x86_capability[0] = capability;
+			c->x86_capability[4] = excap;
+			c->x86 = (tfms >> 8) & 15;
+			c->x86_model = (tfms >> 4) & 15;
+			if (c->x86 == 0xf)
+				c->x86 += (tfms >> 20) & 0xff;
+			if (c->x86 >= 0x6)
+				c->x86_model += ((tfms >> 16) & 0xF) << 4;
+			c->x86_mask = tfms & 15;
+#ifdef CONFIG_X86_HT
+			c->apicid = phys_pkg_id((ebx >> 24) & 0xFF, 0);
+#else
+			c->apicid = (ebx >> 24) & 0xFF;
+#endif
+			if (c->x86_capability[0] & (1<<19))
+				c->x86_clflush_size = ((ebx >> 8) & 0xff) * 8;
+		} else {
+			/* Have CPUID level 0 only - unheard of */
+			c->x86 = 4;
+		}
+
+		/* AMD-defined flags: level 0x80000001 */
+		xlvl = cpuid_eax(0x80000000);
+		if ( (xlvl & 0xffff0000) == 0x80000000 ) {
+			if ( xlvl >= 0x80000001 ) {
+				c->x86_capability[1] = cpuid_edx(0x80000001);
+				c->x86_capability[6] = cpuid_ecx(0x80000001);
+			}
+			if ( xlvl >= 0x80000004 )
+				get_model_name(c); /* Default name */
+		}
+
+		init_scattered_cpuid_features(c);
+	}
+
+	early_intel_workaround(c);
+
+#ifdef CONFIG_X86_HT
+	c->phys_proc_id = (cpuid_ebx(1) >> 24) & 0xff;
+#endif
+}
+
+static void __cpuinit squash_the_stupid_serial_number(struct cpuinfo_x86 *c)
+{
+	if (cpu_has(c, X86_FEATURE_PN) && disable_x86_serial_nr ) {
+		/* Disable processor serial number */
+		unsigned long lo,hi;
+		rdmsr(MSR_IA32_BBL_CR_CTL,lo,hi);
+		lo |= 0x200000;
+		wrmsr(MSR_IA32_BBL_CR_CTL,lo,hi);
+		printk(KERN_NOTICE "CPU serial number disabled.\n");
+		clear_bit(X86_FEATURE_PN, c->x86_capability);
+
+		/* Disabling the serial number may affect the cpuid level */
+		c->cpuid_level = cpuid_eax(0);
+	}
+}
+
+static int __init x86_serial_nr_setup(char *s)
+{
+	disable_x86_serial_nr = 0;
+	return 1;
+}
+__setup("serialnumber", x86_serial_nr_setup);
+
+
+
+/*
+ * This does the hard work of actually picking apart the CPU stuff...
+ */
+static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
+{
+	int i;
+
+	c->loops_per_jiffy = loops_per_jiffy;
+	c->x86_cache_size = -1;
+	c->x86_vendor = X86_VENDOR_UNKNOWN;
+	c->cpuid_level = -1;	/* CPUID not detected */
+	c->x86_model = c->x86_mask = 0;	/* So far unknown... */
+	c->x86_vendor_id[0] = '\0'; /* Unset */
+	c->x86_model_id[0] = '\0';  /* Unset */
+	c->x86_max_cores = 1;
+	c->x86_clflush_size = 32;
+	memset(&c->x86_capability, 0, sizeof c->x86_capability);
+
+	if (!have_cpuid_p()) {
+		/* First of all, decide if this is a 486 or higher */
+		/* It's a 486 if we can modify the AC flag */
+		if ( flag_is_changeable_p(X86_EFLAGS_AC) )
+			c->x86 = 4;
+		else
+			c->x86 = 3;
+	}
+
+	generic_identify(c);
+
+	printk(KERN_DEBUG "CPU: After generic identify, caps:");
+	for (i = 0; i < NCAPINTS; i++)
+		printk(" %08lx", c->x86_capability[i]);
+	printk("\n");
+
+	if (this_cpu->c_identify) {
+		this_cpu->c_identify(c);
+
+		printk(KERN_DEBUG "CPU: After vendor identify, caps:");
+		for (i = 0; i < NCAPINTS; i++)
+			printk(" %08lx", c->x86_capability[i]);
+		printk("\n");
+	}
+
+	/*
+	 * Vendor-specific initialization.  In this section we
+	 * canonicalize the feature flags, meaning if there are
+	 * features a certain CPU supports which CPUID doesn't
+	 * tell us, CPUID claiming incorrect flags, or other bugs,
+	 * we handle them here.
+	 *
+	 * At the end of this section, c->x86_capability better
+	 * indicate the features this CPU genuinely supports!
+	 */
+	if (this_cpu->c_init)
+		this_cpu->c_init(c);
+
+	/* Disable the PN if appropriate */
+	squash_the_stupid_serial_number(c);
+
+	/*
+	 * The vendor-specific functions might have changed features.  Now
+	 * we do "generic changes."
+	 */
+
+	/* TSC disabled? */
+	if ( tsc_disable )
+		clear_bit(X86_FEATURE_TSC, c->x86_capability);
+
+	/* FXSR disabled? */
+	if (disable_x86_fxsr) {
+		clear_bit(X86_FEATURE_FXSR, c->x86_capability);
+		clear_bit(X86_FEATURE_XMM, c->x86_capability);
+	}
+
+	/* SEP disabled? */
+	if (disable_x86_sep)
+		clear_bit(X86_FEATURE_SEP, c->x86_capability);
+
+	if (disable_pse)
+		clear_bit(X86_FEATURE_PSE, c->x86_capability);
+
+	/* If the model name is still unset, do table lookup. */
+	if ( !c->x86_model_id[0] ) {
+		char *p;
+		p = table_lookup_model(c);
+		if ( p )
+			strcpy(c->x86_model_id, p);
+		else
+			/* Last resort... */
+			sprintf(c->x86_model_id, "%02x/%02x",
+				c->x86, c->x86_model);
+	}
+
+	/* Now the feature flags better reflect actual CPU features! */
+
+	printk(KERN_DEBUG "CPU: After all inits, caps:");
+	for (i = 0; i < NCAPINTS; i++)
+		printk(" %08lx", c->x86_capability[i]);
+	printk("\n");
+
+	/*
+	 * On SMP, boot_cpu_data holds the common feature set between
+	 * all CPUs; so make sure that we indicate which features are
+	 * common between the CPUs.  The first time this routine gets
+	 * executed, c == &boot_cpu_data.
+	 */
+	if ( c != &boot_cpu_data ) {
+		/* AND the already accumulated flags with these */
+		for ( i = 0 ; i < NCAPINTS ; i++ )
+			boot_cpu_data.x86_capability[i] &= c->x86_capability[i];
+	}
+
+	/* Init Machine Check Exception if available. */
+	mcheck_init(c);
+}
+
+void __init identify_boot_cpu(void)
+{
+	identify_cpu(&boot_cpu_data);
+	sysenter_setup();
+	enable_sep_cpu();
+	mtrr_bp_init();
+}
+
+void __cpuinit identify_secondary_cpu(struct cpuinfo_x86 *c)
+{
+	BUG_ON(c == &boot_cpu_data);
+	identify_cpu(c);
+	enable_sep_cpu();
+	mtrr_ap_init();
+}
+
+#ifdef CONFIG_X86_HT
+void __cpuinit detect_ht(struct cpuinfo_x86 *c)
+{
+	u32 	eax, ebx, ecx, edx;
+	int 	index_msb, core_bits;
+
+	cpuid(1, &eax, &ebx, &ecx, &edx);
+
+	if (!cpu_has(c, X86_FEATURE_HT) || cpu_has(c, X86_FEATURE_CMP_LEGACY))
+		return;
+
+	smp_num_siblings = (ebx & 0xff0000) >> 16;
+
+	if (smp_num_siblings == 1) {
+		printk(KERN_INFO  "CPU: Hyper-Threading is disabled\n");
+	} else if (smp_num_siblings > 1 ) {
+
+		if (smp_num_siblings > NR_CPUS) {
+			printk(KERN_WARNING "CPU: Unsupported number of the "
+					"siblings %d", smp_num_siblings);
+			smp_num_siblings = 1;
+			return;
+		}
+
+		index_msb = get_count_order(smp_num_siblings);
+		c->phys_proc_id = phys_pkg_id((ebx >> 24) & 0xFF, index_msb);
+
+		printk(KERN_INFO  "CPU: Physical Processor ID: %d\n",
+		       c->phys_proc_id);
+
+		smp_num_siblings = smp_num_siblings / c->x86_max_cores;
+
+		index_msb = get_count_order(smp_num_siblings) ;
+
+		core_bits = get_count_order(c->x86_max_cores);
+
+		c->cpu_core_id = phys_pkg_id((ebx >> 24) & 0xFF, index_msb) &
+					       ((1 << core_bits) - 1);
+
+		if (c->x86_max_cores > 1)
+			printk(KERN_INFO  "CPU: Processor Core ID: %d\n",
+			       c->cpu_core_id);
+	}
+}
+#endif
+
+void __cpuinit print_cpu_info(struct cpuinfo_x86 *c)
+{
+	char *vendor = NULL;
+
+	if (c->x86_vendor < X86_VENDOR_NUM)
+		vendor = this_cpu->c_vendor;
+	else if (c->cpuid_level >= 0)
+		vendor = c->x86_vendor_id;
+
+	if (vendor && strncmp(c->x86_model_id, vendor, strlen(vendor)))
+		printk("%s ", vendor);
+
+	if (!c->x86_model_id[0])
+		printk("%d86", c->x86);
+	else
+		printk("%s", c->x86_model_id);
+
+	if (c->x86_mask || c->cpuid_level >= 0) 
+		printk(" stepping %02x\n", c->x86_mask);
+	else
+		printk("\n");
+}
+
+cpumask_t cpu_initialized __cpuinitdata = CPU_MASK_NONE;
+
+/* This is hacky. :)
+ * We're emulating future behavior.
+ * In the future, the cpu-specific init functions will be called implicitly
+ * via the magic of initcalls.
+ * They will insert themselves into the cpu_devs structure.
+ * Then, when cpu_init() is called, we can just iterate over that array.
+ */
+
+extern int intel_cpu_init(void);
+extern int cyrix_init_cpu(void);
+extern int nsc_init_cpu(void);
+extern int amd_init_cpu(void);
+extern int centaur_init_cpu(void);
+extern int transmeta_init_cpu(void);
+extern int nexgen_init_cpu(void);
+extern int umc_init_cpu(void);
+
+void __init early_cpu_init(void)
+{
+	intel_cpu_init();
+	cyrix_init_cpu();
+	nsc_init_cpu();
+	amd_init_cpu();
+	centaur_init_cpu();
+	transmeta_init_cpu();
+	nexgen_init_cpu();
+	umc_init_cpu();
+	early_cpu_detect();
+
+#ifdef CONFIG_DEBUG_PAGEALLOC
+	/* pse is not compatible with on-the-fly unmapping,
+	 * disable it even if the cpus claim to support it.
+	 */
+	clear_bit(X86_FEATURE_PSE, boot_cpu_data.x86_capability);
+	disable_pse = 1;
+#endif
+}
+
+/* Make sure %fs is initialized properly in idle threads */
+struct pt_regs * __devinit idle_regs(struct pt_regs *regs)
+{
+	memset(regs, 0, sizeof(struct pt_regs));
+	regs->xfs = __KERNEL_PERCPU;
+	return regs;
+}
+
+/* Current gdt points %fs at the "master" per-cpu area: after this,
+ * it's on the real one. */
+void switch_to_new_gdt(void)
+{
+	struct Xgt_desc_struct gdt_descr;
+
+	gdt_descr.address = (long)get_cpu_gdt_table(smp_processor_id());
+	gdt_descr.size = GDT_SIZE - 1;
+	load_gdt(&gdt_descr);
+	asm("mov %0, %%fs" : : "r" (__KERNEL_PERCPU) : "memory");
+}
+
+/*
+ * cpu_init() initializes state that is per-CPU. Some data is already
+ * initialized (naturally) in the bootstrap process, such as the GDT
+ * and IDT. We reload them nevertheless, this function acts as a
+ * 'CPU state barrier', nothing should get across.
+ */
+void __cpuinit cpu_init(void)
+{
+	int cpu = smp_processor_id();
+	struct task_struct *curr = current;
+	struct tss_struct * t = &per_cpu(init_tss, cpu);
+	struct thread_struct *thread = &curr->thread;
+
+	if (cpu_test_and_set(cpu, cpu_initialized)) {
+		printk(KERN_WARNING "CPU#%d already initialized!\n", cpu);
+		for (;;) local_irq_enable();
+	}
+
+	printk(KERN_INFO "Initializing CPU#%d\n", cpu);
+
+	if (cpu_has_vme || cpu_has_tsc || cpu_has_de)
+		clear_in_cr4(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE);
+	if (tsc_disable && cpu_has_tsc) {
+		printk(KERN_NOTICE "Disabling TSC...\n");
+		/**** FIX-HPA: DOES THIS REALLY BELONG HERE? ****/
+		clear_bit(X86_FEATURE_TSC, boot_cpu_data.x86_capability);
+		set_in_cr4(X86_CR4_TSD);
+	}
+
+	load_idt(&idt_descr);
+	switch_to_new_gdt();
+
+	/*
+	 * Set up and load the per-CPU TSS and LDT
+	 */
+	atomic_inc(&init_mm.mm_count);
+	curr->active_mm = &init_mm;
+	if (curr->mm)
+		BUG();
+	enter_lazy_tlb(&init_mm, curr);
+
+	load_esp0(t, thread);
+	set_tss_desc(cpu,t);
+	load_TR_desc();
+	load_LDT(&init_mm.context);
+
+#ifdef CONFIG_DOUBLEFAULT
+	/* Set up doublefault TSS pointer in the GDT */
+	__set_tss_desc(cpu, GDT_ENTRY_DOUBLEFAULT_TSS, &doublefault_tss);
+#endif
+
+	/* Clear %gs. */
+	asm volatile ("mov %0, %%gs" : : "r" (0));
+
+	/* Clear all 6 debug registers: */
+	set_debugreg(0, 0);
+	set_debugreg(0, 1);
+	set_debugreg(0, 2);
+	set_debugreg(0, 3);
+	set_debugreg(0, 6);
+	set_debugreg(0, 7);
+
+	/*
+	 * Force FPU initialization:
+	 */
+	current_thread_info()->status = 0;
+	clear_used_math();
+	mxcsr_feature_mask_init();
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+void __cpuinit cpu_uninit(void)
+{
+	int cpu = raw_smp_processor_id();
+	cpu_clear(cpu, cpu_initialized);
+
+	/* lazy TLB state */
+	per_cpu(cpu_tlbstate, cpu).state = 0;
+	per_cpu(cpu_tlbstate, cpu).active_mm = &init_mm;
+}
+#endif
diff --git a/arch/x86/kernel/cpu/cpu.h b/arch/x86/kernel/cpu/cpu.h
new file mode 100644
index 00000000000..2f6432cef6f
--- /dev/null
+++ b/arch/x86/kernel/cpu/cpu.h
@@ -0,0 +1,28 @@
+
+struct cpu_model_info {
+	int vendor;
+	int family;
+	char *model_names[16];
+};
+
+/* attempt to consolidate cpu attributes */
+struct cpu_dev {
+	char	* c_vendor;
+
+	/* some have two possibilities for cpuid string */
+	char	* c_ident[2];	
+
+	struct		cpu_model_info c_models[4];
+
+	void		(*c_init)(struct cpuinfo_x86 * c);
+	void		(*c_identify)(struct cpuinfo_x86 * c);
+	unsigned int	(*c_size_cache)(struct cpuinfo_x86 * c, unsigned int size);
+};
+
+extern struct cpu_dev * cpu_devs [X86_VENDOR_NUM];
+
+extern int get_model_name(struct cpuinfo_x86 *c);
+extern void display_cacheinfo(struct cpuinfo_x86 *c);
+
+extern void early_intel_workaround(struct cpuinfo_x86 *c);
+
diff --git a/arch/x86/kernel/cpu/cyrix.c b/arch/x86/kernel/cpu/cyrix.c
new file mode 100644
index 00000000000..122d2d75aa9
--- /dev/null
+++ b/arch/x86/kernel/cpu/cyrix.c
@@ -0,0 +1,463 @@
+#include <linux/init.h>
+#include <linux/bitops.h>
+#include <linux/delay.h>
+#include <linux/pci.h>
+#include <asm/dma.h>
+#include <asm/io.h>
+#include <asm/processor-cyrix.h>
+#include <asm/timer.h>
+#include <asm/pci-direct.h>
+#include <asm/tsc.h>
+
+#include "cpu.h"
+
+/*
+ * Read NSC/Cyrix DEVID registers (DIR) to get more detailed info. about the CPU
+ */
+static void __cpuinit do_cyrix_devid(unsigned char *dir0, unsigned char *dir1)
+{
+	unsigned char ccr2, ccr3;
+	unsigned long flags;
+	
+	/* we test for DEVID by checking whether CCR3 is writable */
+	local_irq_save(flags);
+	ccr3 = getCx86(CX86_CCR3);
+	setCx86(CX86_CCR3, ccr3 ^ 0x80);
+	getCx86(0xc0);   /* dummy to change bus */
+
+	if (getCx86(CX86_CCR3) == ccr3) {       /* no DEVID regs. */
+		ccr2 = getCx86(CX86_CCR2);
+		setCx86(CX86_CCR2, ccr2 ^ 0x04);
+		getCx86(0xc0);  /* dummy */
+
+		if (getCx86(CX86_CCR2) == ccr2) /* old Cx486SLC/DLC */
+			*dir0 = 0xfd;
+		else {                          /* Cx486S A step */
+			setCx86(CX86_CCR2, ccr2);
+			*dir0 = 0xfe;
+		}
+	}
+	else {
+		setCx86(CX86_CCR3, ccr3);  /* restore CCR3 */
+
+		/* read DIR0 and DIR1 CPU registers */
+		*dir0 = getCx86(CX86_DIR0);
+		*dir1 = getCx86(CX86_DIR1);
+	}
+	local_irq_restore(flags);
+}
+
+/*
+ * Cx86_dir0_msb is a HACK needed by check_cx686_cpuid/slop in bugs.h in
+ * order to identify the Cyrix CPU model after we're out of setup.c
+ *
+ * Actually since bugs.h doesn't even reference this perhaps someone should
+ * fix the documentation ???
+ */
+static unsigned char Cx86_dir0_msb __cpuinitdata = 0;
+
+static char Cx86_model[][9] __cpuinitdata = {
+	"Cx486", "Cx486", "5x86 ", "6x86", "MediaGX ", "6x86MX ",
+	"M II ", "Unknown"
+};
+static char Cx486_name[][5] __cpuinitdata = {
+	"SLC", "DLC", "SLC2", "DLC2", "SRx", "DRx",
+	"SRx2", "DRx2"
+};
+static char Cx486S_name[][4] __cpuinitdata = {
+	"S", "S2", "Se", "S2e"
+};
+static char Cx486D_name[][4] __cpuinitdata = {
+	"DX", "DX2", "?", "?", "?", "DX4"
+};
+static char Cx86_cb[] __cpuinitdata = "?.5x Core/Bus Clock";
+static char cyrix_model_mult1[] __cpuinitdata = "12??43";
+static char cyrix_model_mult2[] __cpuinitdata = "12233445";
+
+/*
+ * Reset the slow-loop (SLOP) bit on the 686(L) which is set by some old
+ * BIOSes for compatibility with DOS games.  This makes the udelay loop
+ * work correctly, and improves performance.
+ *
+ * FIXME: our newer udelay uses the tsc. We don't need to frob with SLOP
+ */
+
+extern void calibrate_delay(void) __init;
+
+static void __cpuinit check_cx686_slop(struct cpuinfo_x86 *c)
+{
+	unsigned long flags;
+	
+	if (Cx86_dir0_msb == 3) {
+		unsigned char ccr3, ccr5;
+
+		local_irq_save(flags);
+		ccr3 = getCx86(CX86_CCR3);
+		setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10); /* enable MAPEN  */
+		ccr5 = getCx86(CX86_CCR5);
+		if (ccr5 & 2)
+			setCx86(CX86_CCR5, ccr5 & 0xfd);  /* reset SLOP */
+		setCx86(CX86_CCR3, ccr3);                 /* disable MAPEN */
+		local_irq_restore(flags);
+
+		if (ccr5 & 2) { /* possible wrong calibration done */
+			printk(KERN_INFO "Recalibrating delay loop with SLOP bit reset\n");
+			calibrate_delay();
+			c->loops_per_jiffy = loops_per_jiffy;
+		}
+	}
+}
+
+
+static void __cpuinit set_cx86_reorder(void)
+{
+	u8 ccr3;
+
+	printk(KERN_INFO "Enable Memory access reorder on Cyrix/NSC processor.\n");
+	ccr3 = getCx86(CX86_CCR3);
+	setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10); /* enable MAPEN  */
+
+	/* Load/Store Serialize to mem access disable (=reorder it)  */
+	setCx86(CX86_PCR0, getCx86(CX86_PCR0) & ~0x80);
+	/* set load/store serialize from 1GB to 4GB */
+	ccr3 |= 0xe0;
+	setCx86(CX86_CCR3, ccr3);
+}
+
+static void __cpuinit set_cx86_memwb(void)
+{
+	u32 cr0;
+
+	printk(KERN_INFO "Enable Memory-Write-back mode on Cyrix/NSC processor.\n");
+
+	/* CCR2 bit 2: unlock NW bit */
+	setCx86(CX86_CCR2, getCx86(CX86_CCR2) & ~0x04);
+	/* set 'Not Write-through' */
+	cr0 = 0x20000000;
+	write_cr0(read_cr0() | cr0);
+	/* CCR2 bit 2: lock NW bit and set WT1 */
+	setCx86(CX86_CCR2, getCx86(CX86_CCR2) | 0x14 );
+}
+
+static void __cpuinit set_cx86_inc(void)
+{
+	unsigned char ccr3;
+
+	printk(KERN_INFO "Enable Incrementor on Cyrix/NSC processor.\n");
+
+	ccr3 = getCx86(CX86_CCR3);
+	setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10); /* enable MAPEN  */
+	/* PCR1 -- Performance Control */
+	/* Incrementor on, whatever that is */
+	setCx86(CX86_PCR1, getCx86(CX86_PCR1) | 0x02);
+	/* PCR0 -- Performance Control */
+	/* Incrementor Margin 10 */
+	setCx86(CX86_PCR0, getCx86(CX86_PCR0) | 0x04); 
+	setCx86(CX86_CCR3, ccr3);	/* disable MAPEN */
+}
+
+/*
+ *	Configure later MediaGX and/or Geode processor.
+ */
+
+static void __cpuinit geode_configure(void)
+{
+	unsigned long flags;
+	u8 ccr3;
+	local_irq_save(flags);
+
+	/* Suspend on halt power saving and enable #SUSP pin */
+	setCx86(CX86_CCR2, getCx86(CX86_CCR2) | 0x88);
+
+	ccr3 = getCx86(CX86_CCR3);
+	setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10);	/* enable MAPEN */
+	
+
+	/* FPU fast, DTE cache, Mem bypass */
+	setCx86(CX86_CCR4, getCx86(CX86_CCR4) | 0x38);
+	setCx86(CX86_CCR3, ccr3);			/* disable MAPEN */
+	
+	set_cx86_memwb();
+	set_cx86_reorder();	
+	set_cx86_inc();
+	
+	local_irq_restore(flags);
+}
+
+
+static void __cpuinit init_cyrix(struct cpuinfo_x86 *c)
+{
+	unsigned char dir0, dir0_msn, dir0_lsn, dir1 = 0;
+	char *buf = c->x86_model_id;
+	const char *p = NULL;
+
+	/* Bit 31 in normal CPUID used for nonstandard 3DNow ID;
+	   3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway */
+	clear_bit(0*32+31, c->x86_capability);
+
+	/* Cyrix used bit 24 in extended (AMD) CPUID for Cyrix MMX extensions */
+	if ( test_bit(1*32+24, c->x86_capability) ) {
+		clear_bit(1*32+24, c->x86_capability);
+		set_bit(X86_FEATURE_CXMMX, c->x86_capability);
+	}
+
+	do_cyrix_devid(&dir0, &dir1);
+
+	check_cx686_slop(c);
+
+	Cx86_dir0_msb = dir0_msn = dir0 >> 4; /* identifies CPU "family"   */
+	dir0_lsn = dir0 & 0xf;                /* model or clock multiplier */
+
+	/* common case step number/rev -- exceptions handled below */
+	c->x86_model = (dir1 >> 4) + 1;
+	c->x86_mask = dir1 & 0xf;
+
+	/* Now cook; the original recipe is by Channing Corn, from Cyrix.
+	 * We do the same thing for each generation: we work out
+	 * the model, multiplier and stepping.  Black magic included,
+	 * to make the silicon step/rev numbers match the printed ones.
+	 */
+	 
+	switch (dir0_msn) {
+		unsigned char tmp;
+
+	case 0: /* Cx486SLC/DLC/SRx/DRx */
+		p = Cx486_name[dir0_lsn & 7];
+		break;
+
+	case 1: /* Cx486S/DX/DX2/DX4 */
+		p = (dir0_lsn & 8) ? Cx486D_name[dir0_lsn & 5]
+			: Cx486S_name[dir0_lsn & 3];
+		break;
+
+	case 2: /* 5x86 */
+		Cx86_cb[2] = cyrix_model_mult1[dir0_lsn & 5];
+		p = Cx86_cb+2;
+		break;
+
+	case 3: /* 6x86/6x86L */
+		Cx86_cb[1] = ' ';
+		Cx86_cb[2] = cyrix_model_mult1[dir0_lsn & 5];
+		if (dir1 > 0x21) { /* 686L */
+			Cx86_cb[0] = 'L';
+			p = Cx86_cb;
+			(c->x86_model)++;
+		} else             /* 686 */
+			p = Cx86_cb+1;
+		/* Emulate MTRRs using Cyrix's ARRs. */
+		set_bit(X86_FEATURE_CYRIX_ARR, c->x86_capability);
+		/* 6x86's contain this bug */
+		c->coma_bug = 1;
+		break;
+
+	case 4: /* MediaGX/GXm or Geode GXM/GXLV/GX1 */
+#ifdef CONFIG_PCI
+	{
+		u32 vendor, device;
+		/* It isn't really a PCI quirk directly, but the cure is the
+		   same. The MediaGX has deep magic SMM stuff that handles the
+		   SB emulation. It thows away the fifo on disable_dma() which
+		   is wrong and ruins the audio. 
+
+		   Bug2: VSA1 has a wrap bug so that using maximum sized DMA 
+		   causes bad things. According to NatSemi VSA2 has another
+		   bug to do with 'hlt'. I've not seen any boards using VSA2
+		   and X doesn't seem to support it either so who cares 8).
+		   VSA1 we work around however.
+		*/
+
+		printk(KERN_INFO "Working around Cyrix MediaGX virtual DMA bugs.\n");
+		isa_dma_bridge_buggy = 2;
+
+		/* We do this before the PCI layer is running. However we
+		   are safe here as we know the bridge must be a Cyrix
+		   companion and must be present */
+		vendor = read_pci_config_16(0, 0, 0x12, PCI_VENDOR_ID);
+		device = read_pci_config_16(0, 0, 0x12, PCI_DEVICE_ID);
+
+		/*
+		 *  The 5510/5520 companion chips have a funky PIT.
+		 */  
+		if (vendor == PCI_VENDOR_ID_CYRIX &&
+	 (device == PCI_DEVICE_ID_CYRIX_5510 || device == PCI_DEVICE_ID_CYRIX_5520))
+			mark_tsc_unstable("cyrix 5510/5520 detected");
+	}
+#endif
+		c->x86_cache_size=16;	/* Yep 16K integrated cache thats it */
+
+		/* GXm supports extended cpuid levels 'ala' AMD */
+		if (c->cpuid_level == 2) {
+			/* Enable cxMMX extensions (GX1 Datasheet 54) */
+			setCx86(CX86_CCR7, getCx86(CX86_CCR7) | 1);
+			
+			/*
+			 * GXm : 0x30 ... 0x5f GXm  datasheet 51
+			 * GXlv: 0x6x          GXlv datasheet 54
+			 *  ?  : 0x7x
+			 * GX1 : 0x8x          GX1  datasheet 56
+			 */
+			if((0x30 <= dir1 && dir1 <= 0x6f) || (0x80 <=dir1 && dir1 <= 0x8f))
+				geode_configure();
+			get_model_name(c);  /* get CPU marketing name */
+			return;
+		}
+		else {  /* MediaGX */
+			Cx86_cb[2] = (dir0_lsn & 1) ? '3' : '4';
+			p = Cx86_cb+2;
+			c->x86_model = (dir1 & 0x20) ? 1 : 2;
+		}
+		break;
+
+        case 5: /* 6x86MX/M II */
+		if (dir1 > 7)
+		{
+			dir0_msn++;  /* M II */
+			/* Enable MMX extensions (App note 108) */
+			setCx86(CX86_CCR7, getCx86(CX86_CCR7)|1);
+		}
+		else
+		{
+			c->coma_bug = 1;      /* 6x86MX, it has the bug. */
+		}
+		tmp = (!(dir0_lsn & 7) || dir0_lsn & 1) ? 2 : 0;
+		Cx86_cb[tmp] = cyrix_model_mult2[dir0_lsn & 7];
+		p = Cx86_cb+tmp;
+        	if (((dir1 & 0x0f) > 4) || ((dir1 & 0xf0) == 0x20))
+			(c->x86_model)++;
+		/* Emulate MTRRs using Cyrix's ARRs. */
+		set_bit(X86_FEATURE_CYRIX_ARR, c->x86_capability);
+		break;
+
+	case 0xf:  /* Cyrix 486 without DEVID registers */
+		switch (dir0_lsn) {
+		case 0xd:  /* either a 486SLC or DLC w/o DEVID */
+			dir0_msn = 0;
+			p = Cx486_name[(c->hard_math) ? 1 : 0];
+			break;
+
+		case 0xe:  /* a 486S A step */
+			dir0_msn = 0;
+			p = Cx486S_name[0];
+			break;
+		}
+		break;
+
+	default:  /* unknown (shouldn't happen, we know everyone ;-) */
+		dir0_msn = 7;
+		break;
+	}
+	strcpy(buf, Cx86_model[dir0_msn & 7]);
+	if (p) strcat(buf, p);
+	return;
+}
+
+/*
+ * Handle National Semiconductor branded processors
+ */
+static void __cpuinit init_nsc(struct cpuinfo_x86 *c)
+{
+	/* There may be GX1 processors in the wild that are branded
+	 * NSC and not Cyrix.
+	 *
+	 * This function only handles the GX processor, and kicks every
+	 * thing else to the Cyrix init function above - that should
+	 * cover any processors that might have been branded differently
+	 * after NSC acquired Cyrix.
+	 *
+	 * If this breaks your GX1 horribly, please e-mail
+	 * info-linux@ldcmail.amd.com to tell us.
+	 */
+
+	/* Handle the GX (Formally known as the GX2) */
+
+	if (c->x86 == 5 && c->x86_model == 5)
+		display_cacheinfo(c);
+	else
+		init_cyrix(c);
+}
+
+/*
+ * Cyrix CPUs without cpuid or with cpuid not yet enabled can be detected
+ * by the fact that they preserve the flags across the division of 5/2.
+ * PII and PPro exhibit this behavior too, but they have cpuid available.
+ */
+ 
+/*
+ * Perform the Cyrix 5/2 test. A Cyrix won't change
+ * the flags, while other 486 chips will.
+ */
+static inline int test_cyrix_52div(void)
+{
+	unsigned int test;
+
+	__asm__ __volatile__(
+	     "sahf\n\t"		/* clear flags (%eax = 0x0005) */
+	     "div %b2\n\t"	/* divide 5 by 2 */
+	     "lahf"		/* store flags into %ah */
+	     : "=a" (test)
+	     : "0" (5), "q" (2)
+	     : "cc");
+
+	/* AH is 0x02 on Cyrix after the divide.. */
+	return (unsigned char) (test >> 8) == 0x02;
+}
+
+static void __cpuinit cyrix_identify(struct cpuinfo_x86 * c)
+{
+	/* Detect Cyrix with disabled CPUID */
+	if ( c->x86 == 4 && test_cyrix_52div() ) {
+		unsigned char dir0, dir1;
+		
+		strcpy(c->x86_vendor_id, "CyrixInstead");
+	        c->x86_vendor = X86_VENDOR_CYRIX;
+	        
+	        /* Actually enable cpuid on the older cyrix */
+	    
+	    	/* Retrieve CPU revisions */
+	    	
+		do_cyrix_devid(&dir0, &dir1);
+
+		dir0>>=4;		
+		
+		/* Check it is an affected model */
+		
+   	        if (dir0 == 5 || dir0 == 3)
+   	        {
+			unsigned char ccr3;
+			unsigned long flags;
+			printk(KERN_INFO "Enabling CPUID on Cyrix processor.\n");
+			local_irq_save(flags);
+			ccr3 = getCx86(CX86_CCR3);
+			setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10);       /* enable MAPEN  */
+			setCx86(CX86_CCR4, getCx86(CX86_CCR4) | 0x80);  /* enable cpuid  */
+			setCx86(CX86_CCR3, ccr3);                       /* disable MAPEN */
+			local_irq_restore(flags);
+		}
+	}
+}
+
+static struct cpu_dev cyrix_cpu_dev __cpuinitdata = {
+	.c_vendor	= "Cyrix",
+	.c_ident 	= { "CyrixInstead" },
+	.c_init		= init_cyrix,
+	.c_identify	= cyrix_identify,
+};
+
+int __init cyrix_init_cpu(void)
+{
+	cpu_devs[X86_VENDOR_CYRIX] = &cyrix_cpu_dev;
+	return 0;
+}
+
+static struct cpu_dev nsc_cpu_dev __cpuinitdata = {
+	.c_vendor	= "NSC",
+	.c_ident 	= { "Geode by NSC" },
+	.c_init		= init_nsc,
+};
+
+int __init nsc_init_cpu(void)
+{
+	cpu_devs[X86_VENDOR_NSC] = &nsc_cpu_dev;
+	return 0;
+}
+
diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c
new file mode 100644
index 00000000000..dc4e08147b1
--- /dev/null
+++ b/arch/x86/kernel/cpu/intel.c
@@ -0,0 +1,333 @@
+#include <linux/init.h>
+#include <linux/kernel.h>
+
+#include <linux/string.h>
+#include <linux/bitops.h>
+#include <linux/smp.h>
+#include <linux/thread_info.h>
+#include <linux/module.h>
+
+#include <asm/processor.h>
+#include <asm/msr.h>
+#include <asm/uaccess.h>
+
+#include "cpu.h"
+
+#ifdef CONFIG_X86_LOCAL_APIC
+#include <asm/mpspec.h>
+#include <asm/apic.h>
+#include <mach_apic.h>
+#endif
+
+extern int trap_init_f00f_bug(void);
+
+#ifdef CONFIG_X86_INTEL_USERCOPY
+/*
+ * Alignment at which movsl is preferred for bulk memory copies.
+ */
+struct movsl_mask movsl_mask __read_mostly;
+#endif
+
+void __cpuinit early_intel_workaround(struct cpuinfo_x86 *c)
+{
+	if (c->x86_vendor != X86_VENDOR_INTEL)
+		return;
+	/* Netburst reports 64 bytes clflush size, but does IO in 128 bytes */
+	if (c->x86 == 15 && c->x86_cache_alignment == 64)
+		c->x86_cache_alignment = 128;
+}
+
+/*
+ *	Early probe support logic for ppro memory erratum #50
+ *
+ *	This is called before we do cpu ident work
+ */
+ 
+int __cpuinit ppro_with_ram_bug(void)
+{
+	/* Uses data from early_cpu_detect now */
+	if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL &&
+	    boot_cpu_data.x86 == 6 &&
+	    boot_cpu_data.x86_model == 1 &&
+	    boot_cpu_data.x86_mask < 8) {
+		printk(KERN_INFO "Pentium Pro with Errata#50 detected. Taking evasive action.\n");
+		return 1;
+	}
+	return 0;
+}
+	
+
+/*
+ * P4 Xeon errata 037 workaround.
+ * Hardware prefetcher may cause stale data to be loaded into the cache.
+ */
+static void __cpuinit Intel_errata_workarounds(struct cpuinfo_x86 *c)
+{
+	unsigned long lo, hi;
+
+	if ((c->x86 == 15) && (c->x86_model == 1) && (c->x86_mask == 1)) {
+		rdmsr (MSR_IA32_MISC_ENABLE, lo, hi);
+		if ((lo & (1<<9)) == 0) {
+			printk (KERN_INFO "CPU: C0 stepping P4 Xeon detected.\n");
+			printk (KERN_INFO "CPU: Disabling hardware prefetching (Errata 037)\n");
+			lo |= (1<<9);	/* Disable hw prefetching */
+			wrmsr (MSR_IA32_MISC_ENABLE, lo, hi);
+		}
+	}
+}
+
+
+/*
+ * find out the number of processor cores on the die
+ */
+static int __cpuinit num_cpu_cores(struct cpuinfo_x86 *c)
+{
+	unsigned int eax, ebx, ecx, edx;
+
+	if (c->cpuid_level < 4)
+		return 1;
+
+	/* Intel has a non-standard dependency on %ecx for this CPUID level. */
+	cpuid_count(4, 0, &eax, &ebx, &ecx, &edx);
+	if (eax & 0x1f)
+		return ((eax >> 26) + 1);
+	else
+		return 1;
+}
+
+static void __cpuinit init_intel(struct cpuinfo_x86 *c)
+{
+	unsigned int l2 = 0;
+	char *p = NULL;
+
+#ifdef CONFIG_X86_F00F_BUG
+	/*
+	 * All current models of Pentium and Pentium with MMX technology CPUs
+	 * have the F0 0F bug, which lets nonprivileged users lock up the system.
+	 * Note that the workaround only should be initialized once...
+	 */
+	c->f00f_bug = 0;
+	if (!paravirt_enabled() && c->x86 == 5) {
+		static int f00f_workaround_enabled = 0;
+
+		c->f00f_bug = 1;
+		if ( !f00f_workaround_enabled ) {
+			trap_init_f00f_bug();
+			printk(KERN_NOTICE "Intel Pentium with F0 0F bug - workaround enabled.\n");
+			f00f_workaround_enabled = 1;
+		}
+	}
+#endif
+
+	select_idle_routine(c);
+	l2 = init_intel_cacheinfo(c);
+	if (c->cpuid_level > 9 ) {
+		unsigned eax = cpuid_eax(10);
+		/* Check for version and the number of counters */
+		if ((eax & 0xff) && (((eax>>8) & 0xff) > 1))
+			set_bit(X86_FEATURE_ARCH_PERFMON, c->x86_capability);
+	}
+
+	/* SEP CPUID bug: Pentium Pro reports SEP but doesn't have it until model 3 mask 3 */
+	if ((c->x86<<8 | c->x86_model<<4 | c->x86_mask) < 0x633)
+		clear_bit(X86_FEATURE_SEP, c->x86_capability);
+
+	/* Names for the Pentium II/Celeron processors 
+	   detectable only by also checking the cache size.
+	   Dixon is NOT a Celeron. */
+	if (c->x86 == 6) {
+		switch (c->x86_model) {
+		case 5:
+			if (c->x86_mask == 0) {
+				if (l2 == 0)
+					p = "Celeron (Covington)";
+				else if (l2 == 256)
+					p = "Mobile Pentium II (Dixon)";
+			}
+			break;
+			
+		case 6:
+			if (l2 == 128)
+				p = "Celeron (Mendocino)";
+			else if (c->x86_mask == 0 || c->x86_mask == 5)
+				p = "Celeron-A";
+			break;
+			
+		case 8:
+			if (l2 == 128)
+				p = "Celeron (Coppermine)";
+			break;
+		}
+	}
+
+	if ( p )
+		strcpy(c->x86_model_id, p);
+	
+	c->x86_max_cores = num_cpu_cores(c);
+
+	detect_ht(c);
+
+	/* Work around errata */
+	Intel_errata_workarounds(c);
+
+#ifdef CONFIG_X86_INTEL_USERCOPY
+	/*
+	 * Set up the preferred alignment for movsl bulk memory moves
+	 */
+	switch (c->x86) {
+	case 4:		/* 486: untested */
+		break;
+	case 5:		/* Old Pentia: untested */
+		break;
+	case 6:		/* PII/PIII only like movsl with 8-byte alignment */
+		movsl_mask.mask = 7;
+		break;
+	case 15:	/* P4 is OK down to 8-byte alignment */
+		movsl_mask.mask = 7;
+		break;
+	}
+#endif
+
+	if (c->x86 == 15) {
+		set_bit(X86_FEATURE_P4, c->x86_capability);
+		set_bit(X86_FEATURE_SYNC_RDTSC, c->x86_capability);
+	}
+	if (c->x86 == 6) 
+		set_bit(X86_FEATURE_P3, c->x86_capability);
+	if ((c->x86 == 0xf && c->x86_model >= 0x03) ||
+		(c->x86 == 0x6 && c->x86_model >= 0x0e))
+		set_bit(X86_FEATURE_CONSTANT_TSC, c->x86_capability);
+
+	if (cpu_has_ds) {
+		unsigned int l1;
+		rdmsr(MSR_IA32_MISC_ENABLE, l1, l2);
+		if (!(l1 & (1<<11)))
+			set_bit(X86_FEATURE_BTS, c->x86_capability);
+		if (!(l1 & (1<<12)))
+			set_bit(X86_FEATURE_PEBS, c->x86_capability);
+	}
+}
+
+static unsigned int __cpuinit intel_size_cache(struct cpuinfo_x86 * c, unsigned int size)
+{
+	/* Intel PIII Tualatin. This comes in two flavours.
+	 * One has 256kb of cache, the other 512. We have no way
+	 * to determine which, so we use a boottime override
+	 * for the 512kb model, and assume 256 otherwise.
+	 */
+	if ((c->x86 == 6) && (c->x86_model == 11) && (size == 0))
+		size = 256;
+	return size;
+}
+
+static struct cpu_dev intel_cpu_dev __cpuinitdata = {
+	.c_vendor	= "Intel",
+	.c_ident 	= { "GenuineIntel" },
+	.c_models = {
+		{ .vendor = X86_VENDOR_INTEL, .family = 4, .model_names = 
+		  { 
+			  [0] = "486 DX-25/33", 
+			  [1] = "486 DX-50", 
+			  [2] = "486 SX", 
+			  [3] = "486 DX/2", 
+			  [4] = "486 SL", 
+			  [5] = "486 SX/2", 
+			  [7] = "486 DX/2-WB", 
+			  [8] = "486 DX/4", 
+			  [9] = "486 DX/4-WB"
+		  }
+		},
+		{ .vendor = X86_VENDOR_INTEL, .family = 5, .model_names =
+		  { 
+			  [0] = "Pentium 60/66 A-step", 
+			  [1] = "Pentium 60/66", 
+			  [2] = "Pentium 75 - 200",
+			  [3] = "OverDrive PODP5V83", 
+			  [4] = "Pentium MMX",
+			  [7] = "Mobile Pentium 75 - 200", 
+			  [8] = "Mobile Pentium MMX"
+		  }
+		},
+		{ .vendor = X86_VENDOR_INTEL, .family = 6, .model_names =
+		  { 
+			  [0] = "Pentium Pro A-step",
+			  [1] = "Pentium Pro", 
+			  [3] = "Pentium II (Klamath)", 
+			  [4] = "Pentium II (Deschutes)", 
+			  [5] = "Pentium II (Deschutes)", 
+			  [6] = "Mobile Pentium II",
+			  [7] = "Pentium III (Katmai)", 
+			  [8] = "Pentium III (Coppermine)", 
+			  [10] = "Pentium III (Cascades)",
+			  [11] = "Pentium III (Tualatin)",
+		  }
+		},
+		{ .vendor = X86_VENDOR_INTEL, .family = 15, .model_names =
+		  {
+			  [0] = "Pentium 4 (Unknown)",
+			  [1] = "Pentium 4 (Willamette)",
+			  [2] = "Pentium 4 (Northwood)",
+			  [4] = "Pentium 4 (Foster)",
+			  [5] = "Pentium 4 (Foster)",
+		  }
+		},
+	},
+	.c_init		= init_intel,
+	.c_size_cache	= intel_size_cache,
+};
+
+__init int intel_cpu_init(void)
+{
+	cpu_devs[X86_VENDOR_INTEL] = &intel_cpu_dev;
+	return 0;
+}
+
+#ifndef CONFIG_X86_CMPXCHG
+unsigned long cmpxchg_386_u8(volatile void *ptr, u8 old, u8 new)
+{
+	u8 prev;
+	unsigned long flags;
+
+	/* Poor man's cmpxchg for 386. Unsuitable for SMP */
+	local_irq_save(flags);
+	prev = *(u8 *)ptr;
+	if (prev == old)
+		*(u8 *)ptr = new;
+	local_irq_restore(flags);
+	return prev;
+}
+EXPORT_SYMBOL(cmpxchg_386_u8);
+
+unsigned long cmpxchg_386_u16(volatile void *ptr, u16 old, u16 new)
+{
+	u16 prev;
+	unsigned long flags;
+
+	/* Poor man's cmpxchg for 386. Unsuitable for SMP */
+	local_irq_save(flags);
+	prev = *(u16 *)ptr;
+	if (prev == old)
+		*(u16 *)ptr = new;
+	local_irq_restore(flags);
+	return prev;
+}
+EXPORT_SYMBOL(cmpxchg_386_u16);
+
+unsigned long cmpxchg_386_u32(volatile void *ptr, u32 old, u32 new)
+{
+	u32 prev;
+	unsigned long flags;
+
+	/* Poor man's cmpxchg for 386. Unsuitable for SMP */
+	local_irq_save(flags);
+	prev = *(u32 *)ptr;
+	if (prev == old)
+		*(u32 *)ptr = new;
+	local_irq_restore(flags);
+	return prev;
+}
+EXPORT_SYMBOL(cmpxchg_386_u32);
+#endif
+
+// arch_initcall(intel_cpu_init);
+
diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c
new file mode 100644
index 00000000000..db6c25aa577
--- /dev/null
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -0,0 +1,806 @@
+/*
+ *      Routines to indentify caches on Intel CPU.
+ *
+ *      Changes:
+ *      Venkatesh Pallipadi	: Adding cache identification through cpuid(4)
+ *		Ashok Raj <ashok.raj@intel.com>: Work with CPU hotplug infrastructure.
+ *	Andi Kleen / Andreas Herrmann	: CPUID4 emulation on AMD.
+ */
+
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/device.h>
+#include <linux/compiler.h>
+#include <linux/cpu.h>
+#include <linux/sched.h>
+
+#include <asm/processor.h>
+#include <asm/smp.h>
+
+#define LVL_1_INST	1
+#define LVL_1_DATA	2
+#define LVL_2		3
+#define LVL_3		4
+#define LVL_TRACE	5
+
+struct _cache_table
+{
+	unsigned char descriptor;
+	char cache_type;
+	short size;
+};
+
+/* all the cache descriptor types we care about (no TLB or trace cache entries) */
+static struct _cache_table cache_table[] __cpuinitdata =
+{
+	{ 0x06, LVL_1_INST, 8 },	/* 4-way set assoc, 32 byte line size */
+	{ 0x08, LVL_1_INST, 16 },	/* 4-way set assoc, 32 byte line size */
+	{ 0x0a, LVL_1_DATA, 8 },	/* 2 way set assoc, 32 byte line size */
+	{ 0x0c, LVL_1_DATA, 16 },	/* 4-way set assoc, 32 byte line size */
+	{ 0x22, LVL_3,      512 },	/* 4-way set assoc, sectored cache, 64 byte line size */
+	{ 0x23, LVL_3,      1024 },	/* 8-way set assoc, sectored cache, 64 byte line size */
+	{ 0x25, LVL_3,      2048 },	/* 8-way set assoc, sectored cache, 64 byte line size */
+	{ 0x29, LVL_3,      4096 },	/* 8-way set assoc, sectored cache, 64 byte line size */
+	{ 0x2c, LVL_1_DATA, 32 },	/* 8-way set assoc, 64 byte line size */
+	{ 0x30, LVL_1_INST, 32 },	/* 8-way set assoc, 64 byte line size */
+	{ 0x39, LVL_2,      128 },	/* 4-way set assoc, sectored cache, 64 byte line size */
+	{ 0x3a, LVL_2,      192 },	/* 6-way set assoc, sectored cache, 64 byte line size */
+	{ 0x3b, LVL_2,      128 },	/* 2-way set assoc, sectored cache, 64 byte line size */
+	{ 0x3c, LVL_2,      256 },	/* 4-way set assoc, sectored cache, 64 byte line size */
+	{ 0x3d, LVL_2,      384 },	/* 6-way set assoc, sectored cache, 64 byte line size */
+	{ 0x3e, LVL_2,      512 },	/* 4-way set assoc, sectored cache, 64 byte line size */
+	{ 0x41, LVL_2,      128 },	/* 4-way set assoc, 32 byte line size */
+	{ 0x42, LVL_2,      256 },	/* 4-way set assoc, 32 byte line size */
+	{ 0x43, LVL_2,      512 },	/* 4-way set assoc, 32 byte line size */
+	{ 0x44, LVL_2,      1024 },	/* 4-way set assoc, 32 byte line size */
+	{ 0x45, LVL_2,      2048 },	/* 4-way set assoc, 32 byte line size */
+	{ 0x46, LVL_3,      4096 },	/* 4-way set assoc, 64 byte line size */
+	{ 0x47, LVL_3,      8192 },	/* 8-way set assoc, 64 byte line size */
+	{ 0x49, LVL_3,      4096 },	/* 16-way set assoc, 64 byte line size */
+	{ 0x4a, LVL_3,      6144 },	/* 12-way set assoc, 64 byte line size */
+	{ 0x4b, LVL_3,      8192 },	/* 16-way set assoc, 64 byte line size */
+	{ 0x4c, LVL_3,     12288 },	/* 12-way set assoc, 64 byte line size */
+	{ 0x4d, LVL_3,     16384 },	/* 16-way set assoc, 64 byte line size */
+	{ 0x60, LVL_1_DATA, 16 },	/* 8-way set assoc, sectored cache, 64 byte line size */
+	{ 0x66, LVL_1_DATA, 8 },	/* 4-way set assoc, sectored cache, 64 byte line size */
+	{ 0x67, LVL_1_DATA, 16 },	/* 4-way set assoc, sectored cache, 64 byte line size */
+	{ 0x68, LVL_1_DATA, 32 },	/* 4-way set assoc, sectored cache, 64 byte line size */
+	{ 0x70, LVL_TRACE,  12 },	/* 8-way set assoc */
+	{ 0x71, LVL_TRACE,  16 },	/* 8-way set assoc */
+	{ 0x72, LVL_TRACE,  32 },	/* 8-way set assoc */
+	{ 0x73, LVL_TRACE,  64 },	/* 8-way set assoc */
+	{ 0x78, LVL_2,    1024 },	/* 4-way set assoc, 64 byte line size */
+	{ 0x79, LVL_2,     128 },	/* 8-way set assoc, sectored cache, 64 byte line size */
+	{ 0x7a, LVL_2,     256 },	/* 8-way set assoc, sectored cache, 64 byte line size */
+	{ 0x7b, LVL_2,     512 },	/* 8-way set assoc, sectored cache, 64 byte line size */
+	{ 0x7c, LVL_2,    1024 },	/* 8-way set assoc, sectored cache, 64 byte line size */
+	{ 0x7d, LVL_2,    2048 },	/* 8-way set assoc, 64 byte line size */
+	{ 0x7f, LVL_2,     512 },	/* 2-way set assoc, 64 byte line size */
+	{ 0x82, LVL_2,     256 },	/* 8-way set assoc, 32 byte line size */
+	{ 0x83, LVL_2,     512 },	/* 8-way set assoc, 32 byte line size */
+	{ 0x84, LVL_2,    1024 },	/* 8-way set assoc, 32 byte line size */
+	{ 0x85, LVL_2,    2048 },	/* 8-way set assoc, 32 byte line size */
+	{ 0x86, LVL_2,     512 },	/* 4-way set assoc, 64 byte line size */
+	{ 0x87, LVL_2,    1024 },	/* 8-way set assoc, 64 byte line size */
+	{ 0x00, 0, 0}
+};
+
+
+enum _cache_type
+{
+	CACHE_TYPE_NULL	= 0,
+	CACHE_TYPE_DATA = 1,
+	CACHE_TYPE_INST = 2,
+	CACHE_TYPE_UNIFIED = 3
+};
+
+union _cpuid4_leaf_eax {
+	struct {
+		enum _cache_type	type:5;
+		unsigned int		level:3;
+		unsigned int		is_self_initializing:1;
+		unsigned int		is_fully_associative:1;
+		unsigned int		reserved:4;
+		unsigned int		num_threads_sharing:12;
+		unsigned int		num_cores_on_die:6;
+	} split;
+	u32 full;
+};
+
+union _cpuid4_leaf_ebx {
+	struct {
+		unsigned int		coherency_line_size:12;
+		unsigned int		physical_line_partition:10;
+		unsigned int		ways_of_associativity:10;
+	} split;
+	u32 full;
+};
+
+union _cpuid4_leaf_ecx {
+	struct {
+		unsigned int		number_of_sets:32;
+	} split;
+	u32 full;
+};
+
+struct _cpuid4_info {
+	union _cpuid4_leaf_eax eax;
+	union _cpuid4_leaf_ebx ebx;
+	union _cpuid4_leaf_ecx ecx;
+	unsigned long size;
+	cpumask_t shared_cpu_map;
+};
+
+unsigned short			num_cache_leaves;
+
+/* AMD doesn't have CPUID4. Emulate it here to report the same
+   information to the user.  This makes some assumptions about the machine:
+   L2 not shared, no SMT etc. that is currently true on AMD CPUs.
+
+   In theory the TLBs could be reported as fake type (they are in "dummy").
+   Maybe later */
+union l1_cache {
+	struct {
+		unsigned line_size : 8;
+		unsigned lines_per_tag : 8;
+		unsigned assoc : 8;
+		unsigned size_in_kb : 8;
+	};
+	unsigned val;
+};
+
+union l2_cache {
+	struct {
+		unsigned line_size : 8;
+		unsigned lines_per_tag : 4;
+		unsigned assoc : 4;
+		unsigned size_in_kb : 16;
+	};
+	unsigned val;
+};
+
+union l3_cache {
+	struct {
+		unsigned line_size : 8;
+		unsigned lines_per_tag : 4;
+		unsigned assoc : 4;
+		unsigned res : 2;
+		unsigned size_encoded : 14;
+	};
+	unsigned val;
+};
+
+static const unsigned short assocs[] = {
+	[1] = 1, [2] = 2, [4] = 4, [6] = 8,
+	[8] = 16, [0xa] = 32, [0xb] = 48,
+	[0xc] = 64,
+	[0xf] = 0xffff // ??
+};
+
+static const unsigned char levels[] = { 1, 1, 2, 3 };
+static const unsigned char types[] = { 1, 2, 3, 3 };
+
+static void __cpuinit amd_cpuid4(int leaf, union _cpuid4_leaf_eax *eax,
+		       union _cpuid4_leaf_ebx *ebx,
+		       union _cpuid4_leaf_ecx *ecx)
+{
+	unsigned dummy;
+	unsigned line_size, lines_per_tag, assoc, size_in_kb;
+	union l1_cache l1i, l1d;
+	union l2_cache l2;
+	union l3_cache l3;
+	union l1_cache *l1 = &l1d;
+
+	eax->full = 0;
+	ebx->full = 0;
+	ecx->full = 0;
+
+	cpuid(0x80000005, &dummy, &dummy, &l1d.val, &l1i.val);
+	cpuid(0x80000006, &dummy, &dummy, &l2.val, &l3.val);
+
+	switch (leaf) {
+	case 1:
+		l1 = &l1i;
+	case 0:
+		if (!l1->val)
+			return;
+		assoc = l1->assoc;
+		line_size = l1->line_size;
+		lines_per_tag = l1->lines_per_tag;
+		size_in_kb = l1->size_in_kb;
+		break;
+	case 2:
+		if (!l2.val)
+			return;
+		assoc = l2.assoc;
+		line_size = l2.line_size;
+		lines_per_tag = l2.lines_per_tag;
+		/* cpu_data has errata corrections for K7 applied */
+		size_in_kb = current_cpu_data.x86_cache_size;
+		break;
+	case 3:
+		if (!l3.val)
+			return;
+		assoc = l3.assoc;
+		line_size = l3.line_size;
+		lines_per_tag = l3.lines_per_tag;
+		size_in_kb = l3.size_encoded * 512;
+		break;
+	default:
+		return;
+	}
+
+	eax->split.is_self_initializing = 1;
+	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;
+	else
+		eax->split.num_threads_sharing = 0;
+	eax->split.num_cores_on_die = current_cpu_data.x86_max_cores - 1;
+
+
+	if (assoc == 0xf)
+		eax->split.is_fully_associative = 1;
+	ebx->split.coherency_line_size = line_size - 1;
+	ebx->split.ways_of_associativity = assocs[assoc] - 1;
+	ebx->split.physical_line_partition = lines_per_tag - 1;
+	ecx->split.number_of_sets = (size_in_kb * 1024) / line_size /
+		(ebx->split.ways_of_associativity + 1) - 1;
+}
+
+static int __cpuinit cpuid4_cache_lookup(int index, struct _cpuid4_info *this_leaf)
+{
+	union _cpuid4_leaf_eax 	eax;
+	union _cpuid4_leaf_ebx 	ebx;
+	union _cpuid4_leaf_ecx 	ecx;
+	unsigned		edx;
+
+	if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD)
+		amd_cpuid4(index, &eax, &ebx, &ecx);
+	else
+		cpuid_count(4, index, &eax.full, &ebx.full, &ecx.full,  &edx);
+	if (eax.split.type == CACHE_TYPE_NULL)
+		return -EIO; /* better error ? */
+
+	this_leaf->eax = eax;
+	this_leaf->ebx = ebx;
+	this_leaf->ecx = ecx;
+	this_leaf->size = (ecx.split.number_of_sets + 1) *
+		(ebx.split.coherency_line_size + 1) *
+		(ebx.split.physical_line_partition + 1) *
+		(ebx.split.ways_of_associativity + 1);
+	return 0;
+}
+
+static int __cpuinit find_num_cache_leaves(void)
+{
+	unsigned int		eax, ebx, ecx, edx;
+	union _cpuid4_leaf_eax	cache_eax;
+	int 			i = -1;
+
+	do {
+		++i;
+		/* Do cpuid(4) loop to find out num_cache_leaves */
+		cpuid_count(4, i, &eax, &ebx, &ecx, &edx);
+		cache_eax.full = eax;
+	} while (cache_eax.split.type != CACHE_TYPE_NULL);
+	return i;
+}
+
+unsigned int __cpuinit init_intel_cacheinfo(struct cpuinfo_x86 *c)
+{
+	unsigned int trace = 0, l1i = 0, l1d = 0, l2 = 0, l3 = 0; /* Cache sizes */
+	unsigned int new_l1d = 0, new_l1i = 0; /* Cache sizes from cpuid(4) */
+	unsigned int new_l2 = 0, new_l3 = 0, i; /* Cache sizes from cpuid(4) */
+	unsigned int l2_id = 0, l3_id = 0, num_threads_sharing, index_msb;
+#ifdef CONFIG_X86_HT
+	unsigned int cpu = (c == &boot_cpu_data) ? 0 : (c - cpu_data);
+#endif
+
+	if (c->cpuid_level > 3) {
+		static int is_initialized;
+
+		if (is_initialized == 0) {
+			/* Init num_cache_leaves from boot CPU */
+			num_cache_leaves = find_num_cache_leaves();
+			is_initialized++;
+		}
+
+		/*
+		 * Whenever possible use cpuid(4), deterministic cache
+		 * parameters cpuid leaf to find the cache details
+		 */
+		for (i = 0; i < num_cache_leaves; i++) {
+			struct _cpuid4_info this_leaf;
+
+			int retval;
+
+			retval = cpuid4_cache_lookup(i, &this_leaf);
+			if (retval >= 0) {
+				switch(this_leaf.eax.split.level) {
+				    case 1:
+					if (this_leaf.eax.split.type ==
+							CACHE_TYPE_DATA)
+						new_l1d = this_leaf.size/1024;
+					else if (this_leaf.eax.split.type ==
+							CACHE_TYPE_INST)
+						new_l1i = this_leaf.size/1024;
+					break;
+				    case 2:
+					new_l2 = this_leaf.size/1024;
+					num_threads_sharing = 1 + this_leaf.eax.split.num_threads_sharing;
+					index_msb = get_count_order(num_threads_sharing);
+					l2_id = c->apicid >> index_msb;
+					break;
+				    case 3:
+					new_l3 = this_leaf.size/1024;
+					num_threads_sharing = 1 + this_leaf.eax.split.num_threads_sharing;
+					index_msb = get_count_order(num_threads_sharing);
+					l3_id = c->apicid >> index_msb;
+					break;
+				    default:
+					break;
+				}
+			}
+		}
+	}
+	/*
+	 * Don't use cpuid2 if cpuid4 is supported. For P4, we use cpuid2 for
+	 * trace cache
+	 */
+	if ((num_cache_leaves == 0 || c->x86 == 15) && c->cpuid_level > 1) {
+		/* supports eax=2  call */
+		int i, j, n;
+		int regs[4];
+		unsigned char *dp = (unsigned char *)regs;
+		int only_trace = 0;
+
+		if (num_cache_leaves != 0 && c->x86 == 15)
+			only_trace = 1;
+
+		/* Number of times to iterate */
+		n = cpuid_eax(2) & 0xFF;
+
+		for ( i = 0 ; i < n ; i++ ) {
+			cpuid(2, &regs[0], &regs[1], &regs[2], &regs[3]);
+
+			/* If bit 31 is set, this is an unknown format */
+			for ( j = 0 ; j < 3 ; j++ ) {
+				if ( regs[j] < 0 ) regs[j] = 0;
+			}
+
+			/* Byte 0 is level count, not a descriptor */
+			for ( j = 1 ; j < 16 ; j++ ) {
+				unsigned char des = dp[j];
+				unsigned char k = 0;
+
+				/* look up this descriptor in the table */
+				while (cache_table[k].descriptor != 0)
+				{
+					if (cache_table[k].descriptor == des) {
+						if (only_trace && cache_table[k].cache_type != LVL_TRACE)
+							break;
+						switch (cache_table[k].cache_type) {
+						case LVL_1_INST:
+							l1i += cache_table[k].size;
+							break;
+						case LVL_1_DATA:
+							l1d += cache_table[k].size;
+							break;
+						case LVL_2:
+							l2 += cache_table[k].size;
+							break;
+						case LVL_3:
+							l3 += cache_table[k].size;
+							break;
+						case LVL_TRACE:
+							trace += cache_table[k].size;
+							break;
+						}
+
+						break;
+					}
+
+					k++;
+				}
+			}
+		}
+	}
+
+	if (new_l1d)
+		l1d = new_l1d;
+
+	if (new_l1i)
+		l1i = new_l1i;
+
+	if (new_l2) {
+		l2 = new_l2;
+#ifdef CONFIG_X86_HT
+		cpu_llc_id[cpu] = l2_id;
+#endif
+	}
+
+	if (new_l3) {
+		l3 = new_l3;
+#ifdef CONFIG_X86_HT
+		cpu_llc_id[cpu] = l3_id;
+#endif
+	}
+
+	if (trace)
+		printk (KERN_INFO "CPU: Trace cache: %dK uops", trace);
+	else if ( l1i )
+		printk (KERN_INFO "CPU: L1 I cache: %dK", l1i);
+
+	if (l1d)
+		printk(", L1 D cache: %dK\n", l1d);
+	else
+		printk("\n");
+
+	if (l2)
+		printk(KERN_INFO "CPU: L2 cache: %dK\n", l2);
+
+	if (l3)
+		printk(KERN_INFO "CPU: L3 cache: %dK\n", l3);
+
+	c->x86_cache_size = l3 ? l3 : (l2 ? l2 : (l1i+l1d));
+
+	return l2;
+}
+
+/* pointer to _cpuid4_info array (for each cache leaf) */
+static struct _cpuid4_info *cpuid4_info[NR_CPUS];
+#define CPUID4_INFO_IDX(x,y)    (&((cpuid4_info[x])[y]))
+
+#ifdef CONFIG_SMP
+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;
+	struct cpuinfo_x86 *c = cpu_data;
+
+	this_leaf = CPUID4_INFO_IDX(cpu, index);
+	num_threads_sharing = 1 + this_leaf->eax.split.num_threads_sharing;
+
+	if (num_threads_sharing == 1)
+		cpu_set(cpu, this_leaf->shared_cpu_map);
+	else {
+		index_msb = get_count_order(num_threads_sharing);
+
+		for_each_online_cpu(i) {
+			if (c[i].apicid >> index_msb ==
+			    c[cpu].apicid >> index_msb) {
+				cpu_set(i, this_leaf->shared_cpu_map);
+				if (i != cpu && cpuid4_info[i])  {
+					sibling_leaf = CPUID4_INFO_IDX(i, index);
+					cpu_set(cpu, sibling_leaf->shared_cpu_map);
+				}
+			}
+		}
+	}
+}
+static void __cpuinit cache_remove_shared_cpu_map(unsigned int cpu, int index)
+{
+	struct _cpuid4_info	*this_leaf, *sibling_leaf;
+	int sibling;
+
+	this_leaf = CPUID4_INFO_IDX(cpu, index);
+	for_each_cpu_mask(sibling, this_leaf->shared_cpu_map) {
+		sibling_leaf = CPUID4_INFO_IDX(sibling, index);	
+		cpu_clear(cpu, sibling_leaf->shared_cpu_map);
+	}
+}
+#else
+static void __init cache_shared_cpu_map_setup(unsigned int cpu, int index) {}
+static void __init cache_remove_shared_cpu_map(unsigned int cpu, int index) {}
+#endif
+
+static void free_cache_attributes(unsigned int cpu)
+{
+	kfree(cpuid4_info[cpu]);
+	cpuid4_info[cpu] = NULL;
+}
+
+static int __cpuinit detect_cache_attributes(unsigned int cpu)
+{
+	struct _cpuid4_info	*this_leaf;
+	unsigned long 		j;
+	int 			retval;
+	cpumask_t		oldmask;
+
+	if (num_cache_leaves == 0)
+		return -ENOENT;
+
+	cpuid4_info[cpu] = kzalloc(
+	    sizeof(struct _cpuid4_info) * num_cache_leaves, GFP_KERNEL);
+	if (cpuid4_info[cpu] == NULL)
+		return -ENOMEM;
+
+	oldmask = current->cpus_allowed;
+	retval = set_cpus_allowed(current, cpumask_of_cpu(cpu));
+	if (retval)
+		goto out;
+
+	/* Do cpuid and store the results */
+	retval = 0;
+	for (j = 0; j < num_cache_leaves; j++) {
+		this_leaf = CPUID4_INFO_IDX(cpu, j);
+		retval = cpuid4_cache_lookup(j, this_leaf);
+		if (unlikely(retval < 0))
+			break;
+		cache_shared_cpu_map_setup(cpu, j);
+	}
+	set_cpus_allowed(current, oldmask);
+
+out:
+	if (retval)
+		free_cache_attributes(cpu);
+	return retval;
+}
+
+#ifdef CONFIG_SYSFS
+
+#include <linux/kobject.h>
+#include <linux/sysfs.h>
+
+extern struct sysdev_class cpu_sysdev_class; /* from drivers/base/cpu.c */
+
+/* pointer to kobject for cpuX/cache */
+static struct kobject * cache_kobject[NR_CPUS];
+
+struct _index_kobject {
+	struct kobject kobj;
+	unsigned int cpu;
+	unsigned short index;
+};
+
+/* pointer to array of kobjects for cpuX/cache/indexY */
+static struct _index_kobject *index_kobject[NR_CPUS];
+#define INDEX_KOBJECT_PTR(x,y)    (&((index_kobject[x])[y]))
+
+#define show_one_plus(file_name, object, val)				\
+static ssize_t show_##file_name						\
+			(struct _cpuid4_info *this_leaf, char *buf)	\
+{									\
+	return sprintf (buf, "%lu\n", (unsigned long)this_leaf->object + val); \
+}
+
+show_one_plus(level, eax.split.level, 0);
+show_one_plus(coherency_line_size, ebx.split.coherency_line_size, 1);
+show_one_plus(physical_line_partition, ebx.split.physical_line_partition, 1);
+show_one_plus(ways_of_associativity, ebx.split.ways_of_associativity, 1);
+show_one_plus(number_of_sets, ecx.split.number_of_sets, 1);
+
+static ssize_t show_size(struct _cpuid4_info *this_leaf, char *buf)
+{
+	return sprintf (buf, "%luK\n", this_leaf->size / 1024);
+}
+
+static ssize_t show_shared_cpu_map(struct _cpuid4_info *this_leaf, char *buf)
+{
+	char mask_str[NR_CPUS];
+	cpumask_scnprintf(mask_str, NR_CPUS, this_leaf->shared_cpu_map);
+	return sprintf(buf, "%s\n", mask_str);
+}
+
+static ssize_t show_type(struct _cpuid4_info *this_leaf, char *buf) {
+	switch(this_leaf->eax.split.type) {
+	    case CACHE_TYPE_DATA:
+		return sprintf(buf, "Data\n");
+		break;
+	    case CACHE_TYPE_INST:
+		return sprintf(buf, "Instruction\n");
+		break;
+	    case CACHE_TYPE_UNIFIED:
+		return sprintf(buf, "Unified\n");
+		break;
+	    default:
+		return sprintf(buf, "Unknown\n");
+		break;
+	}
+}
+
+struct _cache_attr {
+	struct attribute attr;
+	ssize_t (*show)(struct _cpuid4_info *, char *);
+	ssize_t (*store)(struct _cpuid4_info *, const char *, size_t count);
+};
+
+#define define_one_ro(_name) \
+static struct _cache_attr _name = \
+	__ATTR(_name, 0444, show_##_name, NULL)
+
+define_one_ro(level);
+define_one_ro(type);
+define_one_ro(coherency_line_size);
+define_one_ro(physical_line_partition);
+define_one_ro(ways_of_associativity);
+define_one_ro(number_of_sets);
+define_one_ro(size);
+define_one_ro(shared_cpu_map);
+
+static struct attribute * default_attrs[] = {
+	&type.attr,
+	&level.attr,
+	&coherency_line_size.attr,
+	&physical_line_partition.attr,
+	&ways_of_associativity.attr,
+	&number_of_sets.attr,
+	&size.attr,
+	&shared_cpu_map.attr,
+	NULL
+};
+
+#define to_object(k) container_of(k, struct _index_kobject, kobj)
+#define to_attr(a) container_of(a, struct _cache_attr, attr)
+
+static ssize_t show(struct kobject * kobj, struct attribute * attr, char * buf)
+{
+	struct _cache_attr *fattr = to_attr(attr);
+	struct _index_kobject *this_leaf = to_object(kobj);
+	ssize_t ret;
+
+	ret = fattr->show ?
+		fattr->show(CPUID4_INFO_IDX(this_leaf->cpu, this_leaf->index),
+			buf) :
+	       	0;
+	return ret;
+}
+
+static ssize_t store(struct kobject * kobj, struct attribute * attr,
+		     const char * buf, size_t count)
+{
+	return 0;
+}
+
+static struct sysfs_ops sysfs_ops = {
+	.show   = show,
+	.store  = store,
+};
+
+static struct kobj_type ktype_cache = {
+	.sysfs_ops	= &sysfs_ops,
+	.default_attrs	= default_attrs,
+};
+
+static struct kobj_type ktype_percpu_entry = {
+	.sysfs_ops	= &sysfs_ops,
+};
+
+static void cpuid4_cache_sysfs_exit(unsigned int cpu)
+{
+	kfree(cache_kobject[cpu]);
+	kfree(index_kobject[cpu]);
+	cache_kobject[cpu] = NULL;
+	index_kobject[cpu] = NULL;
+	free_cache_attributes(cpu);
+}
+
+static int __cpuinit cpuid4_cache_sysfs_init(unsigned int cpu)
+{
+
+	if (num_cache_leaves == 0)
+		return -ENOENT;
+
+	detect_cache_attributes(cpu);
+	if (cpuid4_info[cpu] == NULL)
+		return -ENOENT;
+
+	/* Allocate all required memory */
+	cache_kobject[cpu] = kzalloc(sizeof(struct kobject), GFP_KERNEL);
+	if (unlikely(cache_kobject[cpu] == NULL))
+		goto err_out;
+
+	index_kobject[cpu] = kzalloc(
+	    sizeof(struct _index_kobject ) * num_cache_leaves, GFP_KERNEL);
+	if (unlikely(index_kobject[cpu] == NULL))
+		goto err_out;
+
+	return 0;
+
+err_out:
+	cpuid4_cache_sysfs_exit(cpu);
+	return -ENOMEM;
+}
+
+/* Add/Remove cache interface for CPU device */
+static int __cpuinit cache_add_dev(struct sys_device * sys_dev)
+{
+	unsigned int cpu = sys_dev->id;
+	unsigned long i, j;
+	struct _index_kobject *this_object;
+	int retval = 0;
+
+	retval = cpuid4_cache_sysfs_init(cpu);
+	if (unlikely(retval < 0))
+		return retval;
+
+	cache_kobject[cpu]->parent = &sys_dev->kobj;
+	kobject_set_name(cache_kobject[cpu], "%s", "cache");
+	cache_kobject[cpu]->ktype = &ktype_percpu_entry;
+	retval = kobject_register(cache_kobject[cpu]);
+
+	for (i = 0; i < num_cache_leaves; i++) {
+		this_object = INDEX_KOBJECT_PTR(cpu,i);
+		this_object->cpu = cpu;
+		this_object->index = i;
+		this_object->kobj.parent = cache_kobject[cpu];
+		kobject_set_name(&(this_object->kobj), "index%1lu", i);
+		this_object->kobj.ktype = &ktype_cache;
+		retval = kobject_register(&(this_object->kobj));
+		if (unlikely(retval)) {
+			for (j = 0; j < i; j++) {
+				kobject_unregister(
+					&(INDEX_KOBJECT_PTR(cpu,j)->kobj));
+			}
+			kobject_unregister(cache_kobject[cpu]);
+			cpuid4_cache_sysfs_exit(cpu);
+			break;
+		}
+	}
+	return retval;
+}
+
+static void __cpuinit cache_remove_dev(struct sys_device * sys_dev)
+{
+	unsigned int cpu = sys_dev->id;
+	unsigned long i;
+
+	if (cpuid4_info[cpu] == NULL)
+		return;
+	for (i = 0; i < num_cache_leaves; i++) {
+		cache_remove_shared_cpu_map(cpu, i);
+		kobject_unregister(&(INDEX_KOBJECT_PTR(cpu,i)->kobj));
+	}
+	kobject_unregister(cache_kobject[cpu]);
+	cpuid4_cache_sysfs_exit(cpu);
+	return;
+}
+
+static int __cpuinit cacheinfo_cpu_callback(struct notifier_block *nfb,
+					unsigned long action, void *hcpu)
+{
+	unsigned int cpu = (unsigned long)hcpu;
+	struct sys_device *sys_dev;
+
+	sys_dev = get_cpu_sysdev(cpu);
+	switch (action) {
+	case CPU_ONLINE:
+	case CPU_ONLINE_FROZEN:
+		cache_add_dev(sys_dev);
+		break;
+	case CPU_DEAD:
+	case CPU_DEAD_FROZEN:
+		cache_remove_dev(sys_dev);
+		break;
+	}
+	return NOTIFY_OK;
+}
+
+static struct notifier_block __cpuinitdata cacheinfo_cpu_notifier =
+{
+    .notifier_call = cacheinfo_cpu_callback,
+};
+
+static int __cpuinit cache_sysfs_init(void)
+{
+	int i;
+
+	if (num_cache_leaves == 0)
+		return 0;
+
+	register_hotcpu_notifier(&cacheinfo_cpu_notifier);
+
+	for_each_online_cpu(i) {
+		cacheinfo_cpu_callback(&cacheinfo_cpu_notifier, CPU_ONLINE,
+			(void *)(long)i);
+	}
+
+	return 0;
+}
+
+device_initcall(cache_sysfs_init);
+
+#endif
diff --git a/arch/x86/kernel/cpu/nexgen.c b/arch/x86/kernel/cpu/nexgen.c
new file mode 100644
index 00000000000..961fbe1a748
--- /dev/null
+++ b/arch/x86/kernel/cpu/nexgen.c
@@ -0,0 +1,60 @@
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/string.h>
+#include <asm/processor.h>
+
+#include "cpu.h"
+
+/*
+ *	Detect a NexGen CPU running without BIOS hypercode new enough
+ *	to have CPUID. (Thanks to Herbert Oppmann)
+ */
+ 
+static int __cpuinit deep_magic_nexgen_probe(void)
+{
+	int ret;
+	
+	__asm__ __volatile__ (
+		"	movw	$0x5555, %%ax\n"
+		"	xorw	%%dx,%%dx\n"
+		"	movw	$2, %%cx\n"
+		"	divw	%%cx\n"
+		"	movl	$0, %%eax\n"
+		"	jnz	1f\n"
+		"	movl	$1, %%eax\n"
+		"1:\n" 
+		: "=a" (ret) : : "cx", "dx" );
+	return  ret;
+}
+
+static void __cpuinit init_nexgen(struct cpuinfo_x86 * c)
+{
+	c->x86_cache_size = 256; /* A few had 1 MB... */
+}
+
+static void __cpuinit nexgen_identify(struct cpuinfo_x86 * c)
+{
+	/* Detect NexGen with old hypercode */
+	if ( deep_magic_nexgen_probe() ) {
+		strcpy(c->x86_vendor_id, "NexGenDriven");
+	}
+}
+
+static struct cpu_dev nexgen_cpu_dev __cpuinitdata = {
+	.c_vendor	= "Nexgen",
+	.c_ident	= { "NexGenDriven" },
+	.c_models = {
+			{ .vendor = X86_VENDOR_NEXGEN,
+			  .family = 5,
+			  .model_names = { [1] = "Nx586" }
+			},
+	},
+	.c_init		= init_nexgen,
+	.c_identify	= nexgen_identify,
+};
+
+int __init nexgen_init_cpu(void)
+{
+	cpu_devs[X86_VENDOR_NEXGEN] = &nexgen_cpu_dev;
+	return 0;
+}
diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c
new file mode 100644
index 00000000000..93fecd4b03d
--- /dev/null
+++ b/arch/x86/kernel/cpu/perfctr-watchdog.c
@@ -0,0 +1,713 @@
+/* local apic based NMI watchdog for various CPUs.
+   This file also handles reservation of performance counters for coordination
+   with other users (like oprofile).
+
+   Note that these events normally don't tick when the CPU idles. This means
+   the frequency varies with CPU load.
+
+   Original code for K7/P6 written by Keith Owens */
+
+#include <linux/percpu.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/bitops.h>
+#include <linux/smp.h>
+#include <linux/nmi.h>
+#include <asm/apic.h>
+#include <asm/intel_arch_perfmon.h>
+
+struct nmi_watchdog_ctlblk {
+	unsigned int cccr_msr;
+	unsigned int perfctr_msr;  /* the MSR to reset in NMI handler */
+	unsigned int evntsel_msr;  /* the MSR to select the events to handle */
+};
+
+/* Interface defining a CPU specific perfctr watchdog */
+struct wd_ops {
+	int (*reserve)(void);
+	void (*unreserve)(void);
+	int (*setup)(unsigned nmi_hz);
+	void (*rearm)(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz);
+	void (*stop)(void);
+	unsigned perfctr;
+	unsigned evntsel;
+	u64 checkbit;
+};
+
+static struct wd_ops *wd_ops;
+
+/* this number is calculated from Intel's MSR_P4_CRU_ESCR5 register and it's
+ * offset from MSR_P4_BSU_ESCR0.  It will be the max for all platforms (for now)
+ */
+#define NMI_MAX_COUNTER_BITS 66
+
+/* perfctr_nmi_owner tracks the ownership of the perfctr registers:
+ * evtsel_nmi_owner tracks the ownership of the event selection
+ * - different performance counters/ event selection may be reserved for
+ *   different subsystems this reservation system just tries to coordinate
+ *   things a little
+ */
+static DECLARE_BITMAP(perfctr_nmi_owner, NMI_MAX_COUNTER_BITS);
+static DECLARE_BITMAP(evntsel_nmi_owner, NMI_MAX_COUNTER_BITS);
+
+static DEFINE_PER_CPU(struct nmi_watchdog_ctlblk, nmi_watchdog_ctlblk);
+
+/* converts an msr to an appropriate reservation bit */
+static inline unsigned int nmi_perfctr_msr_to_bit(unsigned int msr)
+{
+	/* returns the bit offset of the performance counter register */
+	switch (boot_cpu_data.x86_vendor) {
+	case X86_VENDOR_AMD:
+		return (msr - MSR_K7_PERFCTR0);
+	case X86_VENDOR_INTEL:
+		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
+			return (msr - MSR_ARCH_PERFMON_PERFCTR0);
+
+		switch (boot_cpu_data.x86) {
+		case 6:
+			return (msr - MSR_P6_PERFCTR0);
+		case 15:
+			return (msr - MSR_P4_BPU_PERFCTR0);
+		}
+	}
+	return 0;
+}
+
+/* converts an msr to an appropriate reservation bit */
+/* returns the bit offset of the event selection register */
+static inline unsigned int nmi_evntsel_msr_to_bit(unsigned int msr)
+{
+	/* returns the bit offset of the event selection register */
+	switch (boot_cpu_data.x86_vendor) {
+	case X86_VENDOR_AMD:
+		return (msr - MSR_K7_EVNTSEL0);
+	case X86_VENDOR_INTEL:
+		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
+			return (msr - MSR_ARCH_PERFMON_EVENTSEL0);
+
+		switch (boot_cpu_data.x86) {
+		case 6:
+			return (msr - MSR_P6_EVNTSEL0);
+		case 15:
+			return (msr - MSR_P4_BSU_ESCR0);
+		}
+	}
+	return 0;
+
+}
+
+/* checks for a bit availability (hack for oprofile) */
+int avail_to_resrv_perfctr_nmi_bit(unsigned int counter)
+{
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	return (!test_bit(counter, perfctr_nmi_owner));
+}
+
+/* checks the an msr for availability */
+int avail_to_resrv_perfctr_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_perfctr_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	return (!test_bit(counter, perfctr_nmi_owner));
+}
+
+int reserve_perfctr_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_perfctr_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	if (!test_and_set_bit(counter, perfctr_nmi_owner))
+		return 1;
+	return 0;
+}
+
+void release_perfctr_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_perfctr_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	clear_bit(counter, perfctr_nmi_owner);
+}
+
+int reserve_evntsel_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_evntsel_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	if (!test_and_set_bit(counter, evntsel_nmi_owner))
+		return 1;
+	return 0;
+}
+
+void release_evntsel_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_evntsel_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	clear_bit(counter, evntsel_nmi_owner);
+}
+
+EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi);
+EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi_bit);
+EXPORT_SYMBOL(reserve_perfctr_nmi);
+EXPORT_SYMBOL(release_perfctr_nmi);
+EXPORT_SYMBOL(reserve_evntsel_nmi);
+EXPORT_SYMBOL(release_evntsel_nmi);
+
+void disable_lapic_nmi_watchdog(void)
+{
+	BUG_ON(nmi_watchdog != NMI_LOCAL_APIC);
+
+	if (atomic_read(&nmi_active) <= 0)
+		return;
+
+	on_each_cpu(stop_apic_nmi_watchdog, NULL, 0, 1);
+	wd_ops->unreserve();
+
+	BUG_ON(atomic_read(&nmi_active) != 0);
+}
+
+void enable_lapic_nmi_watchdog(void)
+{
+	BUG_ON(nmi_watchdog != NMI_LOCAL_APIC);
+
+	/* are we already enabled */
+	if (atomic_read(&nmi_active) != 0)
+		return;
+
+	/* are we lapic aware */
+	if (!wd_ops)
+		return;
+	if (!wd_ops->reserve()) {
+		printk(KERN_ERR "NMI watchdog: cannot reserve perfctrs\n");
+		return;
+	}
+
+	on_each_cpu(setup_apic_nmi_watchdog, NULL, 0, 1);
+	touch_nmi_watchdog();
+}
+
+/*
+ * Activate the NMI watchdog via the local APIC.
+ */
+
+static unsigned int adjust_for_32bit_ctr(unsigned int hz)
+{
+	u64 counter_val;
+	unsigned int retval = hz;
+
+	/*
+	 * On Intel CPUs with P6/ARCH_PERFMON only 32 bits in the counter
+	 * are writable, with higher bits sign extending from bit 31.
+	 * So, we can only program the counter with 31 bit values and
+	 * 32nd bit should be 1, for 33.. to be 1.
+	 * Find the appropriate nmi_hz
+	 */
+	counter_val = (u64)cpu_khz * 1000;
+	do_div(counter_val, retval);
+ 	if (counter_val > 0x7fffffffULL) {
+		u64 count = (u64)cpu_khz * 1000;
+		do_div(count, 0x7fffffffUL);
+		retval = count + 1;
+	}
+	return retval;
+}
+
+static void
+write_watchdog_counter(unsigned int perfctr_msr, const char *descr, unsigned nmi_hz)
+{
+	u64 count = (u64)cpu_khz * 1000;
+
+	do_div(count, nmi_hz);
+	if(descr)
+		Dprintk("setting %s to -0x%08Lx\n", descr, count);
+	wrmsrl(perfctr_msr, 0 - count);
+}
+
+static void write_watchdog_counter32(unsigned int perfctr_msr,
+		const char *descr, unsigned nmi_hz)
+{
+	u64 count = (u64)cpu_khz * 1000;
+
+	do_div(count, nmi_hz);
+	if(descr)
+		Dprintk("setting %s to -0x%08Lx\n", descr, count);
+	wrmsr(perfctr_msr, (u32)(-count), 0);
+}
+
+/* AMD K7/K8/Family10h/Family11h support. AMD keeps this interface
+   nicely stable so there is not much variety */
+
+#define K7_EVNTSEL_ENABLE	(1 << 22)
+#define K7_EVNTSEL_INT		(1 << 20)
+#define K7_EVNTSEL_OS		(1 << 17)
+#define K7_EVNTSEL_USR		(1 << 16)
+#define K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING	0x76
+#define K7_NMI_EVENT		K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING
+
+static int setup_k7_watchdog(unsigned nmi_hz)
+{
+	unsigned int perfctr_msr, evntsel_msr;
+	unsigned int evntsel;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	perfctr_msr = wd_ops->perfctr;
+	evntsel_msr = wd_ops->evntsel;
+
+	wrmsrl(perfctr_msr, 0UL);
+
+	evntsel = K7_EVNTSEL_INT
+		| K7_EVNTSEL_OS
+		| K7_EVNTSEL_USR
+		| K7_NMI_EVENT;
+
+	/* setup the timer */
+	wrmsr(evntsel_msr, evntsel, 0);
+	write_watchdog_counter(perfctr_msr, "K7_PERFCTR0",nmi_hz);
+	apic_write(APIC_LVTPC, APIC_DM_NMI);
+	evntsel |= K7_EVNTSEL_ENABLE;
+	wrmsr(evntsel_msr, evntsel, 0);
+
+	wd->perfctr_msr = perfctr_msr;
+	wd->evntsel_msr = evntsel_msr;
+	wd->cccr_msr = 0;  //unused
+	return 1;
+}
+
+static void single_msr_stop_watchdog(void)
+{
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	wrmsr(wd->evntsel_msr, 0, 0);
+}
+
+static int single_msr_reserve(void)
+{
+	if (!reserve_perfctr_nmi(wd_ops->perfctr))
+		return 0;
+
+	if (!reserve_evntsel_nmi(wd_ops->evntsel)) {
+		release_perfctr_nmi(wd_ops->perfctr);
+		return 0;
+	}
+	return 1;
+}
+
+static void single_msr_unreserve(void)
+{
+	release_evntsel_nmi(wd_ops->evntsel);
+	release_perfctr_nmi(wd_ops->perfctr);
+}
+
+static void single_msr_rearm(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz)
+{
+	/* start the cycle over again */
+	write_watchdog_counter(wd->perfctr_msr, NULL, nmi_hz);
+}
+
+static struct wd_ops k7_wd_ops = {
+	.reserve = single_msr_reserve,
+	.unreserve = single_msr_unreserve,
+	.setup = setup_k7_watchdog,
+	.rearm = single_msr_rearm,
+	.stop = single_msr_stop_watchdog,
+	.perfctr = MSR_K7_PERFCTR0,
+	.evntsel = MSR_K7_EVNTSEL0,
+	.checkbit = 1ULL<<47,
+};
+
+/* Intel Model 6 (PPro+,P2,P3,P-M,Core1) */
+
+#define P6_EVNTSEL0_ENABLE	(1 << 22)
+#define P6_EVNTSEL_INT		(1 << 20)
+#define P6_EVNTSEL_OS		(1 << 17)
+#define P6_EVNTSEL_USR		(1 << 16)
+#define P6_EVENT_CPU_CLOCKS_NOT_HALTED	0x79
+#define P6_NMI_EVENT		P6_EVENT_CPU_CLOCKS_NOT_HALTED
+
+static int setup_p6_watchdog(unsigned nmi_hz)
+{
+	unsigned int perfctr_msr, evntsel_msr;
+	unsigned int evntsel;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	perfctr_msr = wd_ops->perfctr;
+	evntsel_msr = wd_ops->evntsel;
+
+	/* KVM doesn't implement this MSR */
+	if (wrmsr_safe(perfctr_msr, 0, 0) < 0)
+		return 0;
+
+	evntsel = P6_EVNTSEL_INT
+		| P6_EVNTSEL_OS
+		| P6_EVNTSEL_USR
+		| P6_NMI_EVENT;
+
+	/* setup the timer */
+	wrmsr(evntsel_msr, evntsel, 0);
+	nmi_hz = adjust_for_32bit_ctr(nmi_hz);
+	write_watchdog_counter32(perfctr_msr, "P6_PERFCTR0",nmi_hz);
+	apic_write(APIC_LVTPC, APIC_DM_NMI);
+	evntsel |= P6_EVNTSEL0_ENABLE;
+	wrmsr(evntsel_msr, evntsel, 0);
+
+	wd->perfctr_msr = perfctr_msr;
+	wd->evntsel_msr = evntsel_msr;
+	wd->cccr_msr = 0;  //unused
+	return 1;
+}
+
+static void p6_rearm(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz)
+{
+	/* P6 based Pentium M need to re-unmask
+	 * the apic vector but it doesn't hurt
+	 * other P6 variant.
+	 * ArchPerfom/Core Duo also needs this */
+	apic_write(APIC_LVTPC, APIC_DM_NMI);
+	/* P6/ARCH_PERFMON has 32 bit counter write */
+	write_watchdog_counter32(wd->perfctr_msr, NULL,nmi_hz);
+}
+
+static struct wd_ops p6_wd_ops = {
+	.reserve = single_msr_reserve,
+	.unreserve = single_msr_unreserve,
+	.setup = setup_p6_watchdog,
+	.rearm = p6_rearm,
+	.stop = single_msr_stop_watchdog,
+	.perfctr = MSR_P6_PERFCTR0,
+	.evntsel = MSR_P6_EVNTSEL0,
+	.checkbit = 1ULL<<39,
+};
+
+/* Intel P4 performance counters. By far the most complicated of all. */
+
+#define MSR_P4_MISC_ENABLE_PERF_AVAIL	(1<<7)
+#define P4_ESCR_EVENT_SELECT(N)	((N)<<25)
+#define P4_ESCR_OS		(1<<3)
+#define P4_ESCR_USR		(1<<2)
+#define P4_CCCR_OVF_PMI0	(1<<26)
+#define P4_CCCR_OVF_PMI1	(1<<27)
+#define P4_CCCR_THRESHOLD(N)	((N)<<20)
+#define P4_CCCR_COMPLEMENT	(1<<19)
+#define P4_CCCR_COMPARE		(1<<18)
+#define P4_CCCR_REQUIRED	(3<<16)
+#define P4_CCCR_ESCR_SELECT(N)	((N)<<13)
+#define P4_CCCR_ENABLE		(1<<12)
+#define P4_CCCR_OVF 		(1<<31)
+
+/* Set up IQ_COUNTER0 to behave like a clock, by having IQ_CCCR0 filter
+   CRU_ESCR0 (with any non-null event selector) through a complemented
+   max threshold. [IA32-Vol3, Section 14.9.9] */
+
+static int setup_p4_watchdog(unsigned nmi_hz)
+{
+	unsigned int perfctr_msr, evntsel_msr, cccr_msr;
+	unsigned int evntsel, cccr_val;
+	unsigned int misc_enable, dummy;
+	unsigned int ht_num;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	rdmsr(MSR_IA32_MISC_ENABLE, misc_enable, dummy);
+	if (!(misc_enable & MSR_P4_MISC_ENABLE_PERF_AVAIL))
+		return 0;
+
+#ifdef CONFIG_SMP
+	/* detect which hyperthread we are on */
+	if (smp_num_siblings == 2) {
+		unsigned int ebx, apicid;
+
+        	ebx = cpuid_ebx(1);
+	        apicid = (ebx >> 24) & 0xff;
+        	ht_num = apicid & 1;
+	} else
+#endif
+		ht_num = 0;
+
+	/* performance counters are shared resources
+	 * assign each hyperthread its own set
+	 * (re-use the ESCR0 register, seems safe
+	 * and keeps the cccr_val the same)
+	 */
+	if (!ht_num) {
+		/* logical cpu 0 */
+		perfctr_msr = MSR_P4_IQ_PERFCTR0;
+		evntsel_msr = MSR_P4_CRU_ESCR0;
+		cccr_msr = MSR_P4_IQ_CCCR0;
+		cccr_val = P4_CCCR_OVF_PMI0 | P4_CCCR_ESCR_SELECT(4);
+	} else {
+		/* logical cpu 1 */
+		perfctr_msr = MSR_P4_IQ_PERFCTR1;
+		evntsel_msr = MSR_P4_CRU_ESCR0;
+		cccr_msr = MSR_P4_IQ_CCCR1;
+		cccr_val = P4_CCCR_OVF_PMI1 | P4_CCCR_ESCR_SELECT(4);
+	}
+
+	evntsel = P4_ESCR_EVENT_SELECT(0x3F)
+	 	| P4_ESCR_OS
+		| P4_ESCR_USR;
+
+	cccr_val |= P4_CCCR_THRESHOLD(15)
+		 | P4_CCCR_COMPLEMENT
+		 | P4_CCCR_COMPARE
+		 | P4_CCCR_REQUIRED;
+
+	wrmsr(evntsel_msr, evntsel, 0);
+	wrmsr(cccr_msr, cccr_val, 0);
+	write_watchdog_counter(perfctr_msr, "P4_IQ_COUNTER0", nmi_hz);
+	apic_write(APIC_LVTPC, APIC_DM_NMI);
+	cccr_val |= P4_CCCR_ENABLE;
+	wrmsr(cccr_msr, cccr_val, 0);
+	wd->perfctr_msr = perfctr_msr;
+	wd->evntsel_msr = evntsel_msr;
+	wd->cccr_msr = cccr_msr;
+	return 1;
+}
+
+static void stop_p4_watchdog(void)
+{
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+	wrmsr(wd->cccr_msr, 0, 0);
+	wrmsr(wd->evntsel_msr, 0, 0);
+}
+
+static int p4_reserve(void)
+{
+	if (!reserve_perfctr_nmi(MSR_P4_IQ_PERFCTR0))
+		return 0;
+#ifdef CONFIG_SMP
+	if (smp_num_siblings > 1 && !reserve_perfctr_nmi(MSR_P4_IQ_PERFCTR1))
+		goto fail1;
+#endif
+	if (!reserve_evntsel_nmi(MSR_P4_CRU_ESCR0))
+		goto fail2;
+	/* RED-PEN why is ESCR1 not reserved here? */
+	return 1;
+ fail2:
+#ifdef CONFIG_SMP
+	if (smp_num_siblings > 1)
+		release_perfctr_nmi(MSR_P4_IQ_PERFCTR1);
+ fail1:
+#endif
+	release_perfctr_nmi(MSR_P4_IQ_PERFCTR0);
+	return 0;
+}
+
+static void p4_unreserve(void)
+{
+#ifdef CONFIG_SMP
+	if (smp_num_siblings > 1)
+		release_perfctr_nmi(MSR_P4_IQ_PERFCTR1);
+#endif
+	release_evntsel_nmi(MSR_P4_CRU_ESCR0);
+	release_perfctr_nmi(MSR_P4_IQ_PERFCTR0);
+}
+
+static void p4_rearm(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz)
+{
+	unsigned dummy;
+	/*
+ 	 * P4 quirks:
+	 * - An overflown perfctr will assert its interrupt
+	 *   until the OVF flag in its CCCR is cleared.
+	 * - LVTPC is masked on interrupt and must be
+	 *   unmasked by the LVTPC handler.
+	 */
+	rdmsrl(wd->cccr_msr, dummy);
+	dummy &= ~P4_CCCR_OVF;
+	wrmsrl(wd->cccr_msr, dummy);
+	apic_write(APIC_LVTPC, APIC_DM_NMI);
+	/* start the cycle over again */
+	write_watchdog_counter(wd->perfctr_msr, NULL, nmi_hz);
+}
+
+static struct wd_ops p4_wd_ops = {
+	.reserve = p4_reserve,
+	.unreserve = p4_unreserve,
+	.setup = setup_p4_watchdog,
+	.rearm = p4_rearm,
+	.stop = stop_p4_watchdog,
+	/* RED-PEN this is wrong for the other sibling */
+	.perfctr = MSR_P4_BPU_PERFCTR0,
+	.evntsel = MSR_P4_BSU_ESCR0,
+	.checkbit = 1ULL<<39,
+};
+
+/* Watchdog using the Intel architected PerfMon. Used for Core2 and hopefully
+   all future Intel CPUs. */
+
+#define ARCH_PERFMON_NMI_EVENT_SEL	ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL
+#define ARCH_PERFMON_NMI_EVENT_UMASK	ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK
+
+static int setup_intel_arch_watchdog(unsigned nmi_hz)
+{
+	unsigned int ebx;
+	union cpuid10_eax eax;
+	unsigned int unused;
+	unsigned int perfctr_msr, evntsel_msr;
+	unsigned int evntsel;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	/*
+	 * Check whether the Architectural PerfMon supports
+	 * Unhalted Core Cycles Event or not.
+	 * NOTE: Corresponding bit = 0 in ebx indicates event present.
+	 */
+	cpuid(10, &(eax.full), &ebx, &unused, &unused);
+	if ((eax.split.mask_length < (ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX+1)) ||
+	    (ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT))
+		return 0;
+
+	perfctr_msr = wd_ops->perfctr;
+	evntsel_msr = wd_ops->evntsel;
+
+	wrmsrl(perfctr_msr, 0UL);
+
+	evntsel = ARCH_PERFMON_EVENTSEL_INT
+		| ARCH_PERFMON_EVENTSEL_OS
+		| ARCH_PERFMON_EVENTSEL_USR
+		| ARCH_PERFMON_NMI_EVENT_SEL
+		| ARCH_PERFMON_NMI_EVENT_UMASK;
+
+	/* setup the timer */
+	wrmsr(evntsel_msr, evntsel, 0);
+	nmi_hz = adjust_for_32bit_ctr(nmi_hz);
+	write_watchdog_counter32(perfctr_msr, "INTEL_ARCH_PERFCTR0", nmi_hz);
+	apic_write(APIC_LVTPC, APIC_DM_NMI);
+	evntsel |= ARCH_PERFMON_EVENTSEL0_ENABLE;
+	wrmsr(evntsel_msr, evntsel, 0);
+
+	wd->perfctr_msr = perfctr_msr;
+	wd->evntsel_msr = evntsel_msr;
+	wd->cccr_msr = 0;  //unused
+	wd_ops->checkbit = 1ULL << (eax.split.bit_width - 1);
+	return 1;
+}
+
+static struct wd_ops intel_arch_wd_ops = {
+	.reserve = single_msr_reserve,
+	.unreserve = single_msr_unreserve,
+	.setup = setup_intel_arch_watchdog,
+	.rearm = p6_rearm,
+	.stop = single_msr_stop_watchdog,
+	.perfctr = MSR_ARCH_PERFMON_PERFCTR1,
+	.evntsel = MSR_ARCH_PERFMON_EVENTSEL1,
+};
+
+static struct wd_ops coreduo_wd_ops = {
+	.reserve = single_msr_reserve,
+	.unreserve = single_msr_unreserve,
+	.setup = setup_intel_arch_watchdog,
+	.rearm = p6_rearm,
+	.stop = single_msr_stop_watchdog,
+	.perfctr = MSR_ARCH_PERFMON_PERFCTR0,
+	.evntsel = MSR_ARCH_PERFMON_EVENTSEL0,
+};
+
+static void probe_nmi_watchdog(void)
+{
+	switch (boot_cpu_data.x86_vendor) {
+	case X86_VENDOR_AMD:
+		if (boot_cpu_data.x86 != 6 && boot_cpu_data.x86 != 15 &&
+		    boot_cpu_data.x86 != 16)
+			return;
+		wd_ops = &k7_wd_ops;
+		break;
+	case X86_VENDOR_INTEL:
+		/* Work around Core Duo (Yonah) errata AE49 where perfctr1
+		   doesn't have a working enable bit. */
+		if (boot_cpu_data.x86 == 6 && boot_cpu_data.x86_model == 14) {
+			wd_ops = &coreduo_wd_ops;
+			break;
+		}
+		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) {
+			wd_ops = &intel_arch_wd_ops;
+			break;
+		}
+		switch (boot_cpu_data.x86) {
+		case 6:
+			if (boot_cpu_data.x86_model > 0xd)
+				return;
+
+			wd_ops = &p6_wd_ops;
+			break;
+		case 15:
+			if (boot_cpu_data.x86_model > 0x4)
+				return;
+
+			wd_ops = &p4_wd_ops;
+			break;
+		default:
+			return;
+		}
+		break;
+	}
+}
+
+/* Interface to nmi.c */
+
+int lapic_watchdog_init(unsigned nmi_hz)
+{
+	if (!wd_ops) {
+		probe_nmi_watchdog();
+		if (!wd_ops)
+			return -1;
+
+		if (!wd_ops->reserve()) {
+			printk(KERN_ERR
+				"NMI watchdog: cannot reserve perfctrs\n");
+			return -1;
+		}
+	}
+
+	if (!(wd_ops->setup(nmi_hz))) {
+		printk(KERN_ERR "Cannot setup NMI watchdog on CPU %d\n",
+		       raw_smp_processor_id());
+		return -1;
+	}
+
+	return 0;
+}
+
+void lapic_watchdog_stop(void)
+{
+	if (wd_ops)
+		wd_ops->stop();
+}
+
+unsigned lapic_adjust_nmi_hz(unsigned hz)
+{
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+	if (wd->perfctr_msr == MSR_P6_PERFCTR0 ||
+	    wd->perfctr_msr == MSR_ARCH_PERFMON_PERFCTR1)
+		hz = adjust_for_32bit_ctr(hz);
+	return hz;
+}
+
+int lapic_wd_event(unsigned nmi_hz)
+{
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+	u64 ctr;
+	rdmsrl(wd->perfctr_msr, ctr);
+	if (ctr & wd_ops->checkbit) { /* perfctr still running? */
+		return 0;
+	}
+	wd_ops->rearm(wd, nmi_hz);
+	return 1;
+}
+
+int lapic_watchdog_ok(void)
+{
+	return wd_ops != NULL;
+}
diff --git a/arch/x86/kernel/cpu/proc.c b/arch/x86/kernel/cpu/proc.c
new file mode 100644
index 00000000000..1e31b6caffb
--- /dev/null
+++ b/arch/x86/kernel/cpu/proc.c
@@ -0,0 +1,192 @@
+#include <linux/smp.h>
+#include <linux/timex.h>
+#include <linux/string.h>
+#include <asm/semaphore.h>
+#include <linux/seq_file.h>
+#include <linux/cpufreq.h>
+
+/*
+ *	Get CPU information for use by the procfs.
+ */
+static int show_cpuinfo(struct seq_file *m, void *v)
+{
+	/* 
+	 * These flag bits must match the definitions in <asm/cpufeature.h>.
+	 * NULL means this bit is undefined or reserved; either way it doesn't
+	 * have meaning as far as Linux is concerned.  Note that it's important
+	 * to realize there is a difference between this table and CPUID -- if
+	 * applications want to get the raw CPUID data, they should access
+	 * /dev/cpu/<cpu_nr>/cpuid instead.
+	 */
+	static const char * const x86_cap_flags[] = {
+		/* Intel-defined */
+	        "fpu", "vme", "de", "pse", "tsc", "msr", "pae", "mce",
+	        "cx8", "apic", NULL, "sep", "mtrr", "pge", "mca", "cmov",
+	        "pat", "pse36", "pn", "clflush", NULL, "dts", "acpi", "mmx",
+	        "fxsr", "sse", "sse2", "ss", "ht", "tm", "ia64", "pbe",
+
+		/* AMD-defined */
+		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+		NULL, NULL, NULL, "syscall", NULL, NULL, NULL, NULL,
+		NULL, NULL, NULL, "mp", "nx", NULL, "mmxext", NULL,
+		NULL, "fxsr_opt", "pdpe1gb", "rdtscp", NULL, "lm",
+		"3dnowext", "3dnow",
+
+		/* Transmeta-defined */
+		"recovery", "longrun", NULL, "lrti", NULL, NULL, NULL, NULL,
+		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+
+		/* Other (Linux-defined) */
+		"cxmmx", "k6_mtrr", "cyrix_arr", "centaur_mcr",
+		NULL, NULL, NULL, NULL,
+		"constant_tsc", "up", NULL, "arch_perfmon",
+		"pebs", "bts", NULL, "sync_rdtsc",
+		"rep_good", NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+
+		/* Intel-defined (#2) */
+		"pni", NULL, NULL, "monitor", "ds_cpl", "vmx", "smx", "est",
+		"tm2", "ssse3", "cid", NULL, NULL, "cx16", "xtpr", NULL,
+		NULL, NULL, "dca", NULL, NULL, NULL, NULL, "popcnt",
+		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+
+		/* VIA/Cyrix/Centaur-defined */
+		NULL, NULL, "rng", "rng_en", NULL, NULL, "ace", "ace_en",
+		"ace2", "ace2_en", "phe", "phe_en", "pmm", "pmm_en", NULL, NULL,
+		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+
+		/* AMD-defined (#2) */
+		"lahf_lm", "cmp_legacy", "svm", "extapic", "cr8_legacy",
+		"altmovcr8", "abm", "sse4a",
+		"misalignsse", "3dnowprefetch",
+		"osvw", "ibs", NULL, NULL, NULL, NULL,
+		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+
+		/* Auxiliary (Linux-defined) */
+		"ida", NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+	};
+	static const char * const x86_power_flags[] = {
+		"ts",	/* temperature sensor */
+		"fid",  /* frequency id control */
+		"vid",  /* voltage id control */
+		"ttp",  /* thermal trip */
+		"tm",
+		"stc",
+		"100mhzsteps",
+		"hwpstate",
+		"",	/* constant_tsc - moved to flags */
+		/* nothing */
+	};
+	struct cpuinfo_x86 *c = v;
+	int i, n = c - cpu_data;
+	int fpu_exception;
+
+#ifdef CONFIG_SMP
+	if (!cpu_online(n))
+		return 0;
+#endif
+	seq_printf(m, "processor\t: %d\n"
+		"vendor_id\t: %s\n"
+		"cpu family\t: %d\n"
+		"model\t\t: %d\n"
+		"model name\t: %s\n",
+		n,
+		c->x86_vendor_id[0] ? c->x86_vendor_id : "unknown",
+		c->x86,
+		c->x86_model,
+		c->x86_model_id[0] ? c->x86_model_id : "unknown");
+
+	if (c->x86_mask || c->cpuid_level >= 0)
+		seq_printf(m, "stepping\t: %d\n", c->x86_mask);
+	else
+		seq_printf(m, "stepping\t: unknown\n");
+
+	if ( cpu_has(c, X86_FEATURE_TSC) ) {
+		unsigned int freq = cpufreq_quick_get(n);
+		if (!freq)
+			freq = cpu_khz;
+		seq_printf(m, "cpu MHz\t\t: %u.%03u\n",
+			freq / 1000, (freq % 1000));
+	}
+
+	/* Cache size */
+	if (c->x86_cache_size >= 0)
+		seq_printf(m, "cache size\t: %d KB\n", c->x86_cache_size);
+#ifdef CONFIG_X86_HT
+	if (c->x86_max_cores * smp_num_siblings > 1) {
+		seq_printf(m, "physical id\t: %d\n", c->phys_proc_id);
+		seq_printf(m, "siblings\t: %d\n", cpus_weight(cpu_core_map[n]));
+		seq_printf(m, "core id\t\t: %d\n", c->cpu_core_id);
+		seq_printf(m, "cpu cores\t: %d\n", c->booted_cores);
+	}
+#endif
+	
+	/* We use exception 16 if we have hardware math and we've either seen it or the CPU claims it is internal */
+	fpu_exception = c->hard_math && (ignore_fpu_irq || cpu_has_fpu);
+	seq_printf(m, "fdiv_bug\t: %s\n"
+			"hlt_bug\t\t: %s\n"
+			"f00f_bug\t: %s\n"
+			"coma_bug\t: %s\n"
+			"fpu\t\t: %s\n"
+			"fpu_exception\t: %s\n"
+			"cpuid level\t: %d\n"
+			"wp\t\t: %s\n"
+			"flags\t\t:",
+		     c->fdiv_bug ? "yes" : "no",
+		     c->hlt_works_ok ? "no" : "yes",
+		     c->f00f_bug ? "yes" : "no",
+		     c->coma_bug ? "yes" : "no",
+		     c->hard_math ? "yes" : "no",
+		     fpu_exception ? "yes" : "no",
+		     c->cpuid_level,
+		     c->wp_works_ok ? "yes" : "no");
+
+	for ( i = 0 ; i < 32*NCAPINTS ; i++ )
+		if ( test_bit(i, c->x86_capability) &&
+		     x86_cap_flags[i] != NULL )
+			seq_printf(m, " %s", x86_cap_flags[i]);
+
+	for (i = 0; i < 32; i++)
+		if (c->x86_power & (1 << i)) {
+			if (i < ARRAY_SIZE(x86_power_flags) &&
+			    x86_power_flags[i])
+				seq_printf(m, "%s%s",
+					   x86_power_flags[i][0]?" ":"",
+					   x86_power_flags[i]);
+			else
+				seq_printf(m, " [%d]", i);
+		}
+
+	seq_printf(m, "\nbogomips\t: %lu.%02lu\n",
+		     c->loops_per_jiffy/(500000/HZ),
+		     (c->loops_per_jiffy/(5000/HZ)) % 100);
+	seq_printf(m, "clflush size\t: %u\n\n", c->x86_clflush_size);
+
+	return 0;
+}
+
+static void *c_start(struct seq_file *m, loff_t *pos)
+{
+	return *pos < NR_CPUS ? cpu_data + *pos : NULL;
+}
+static void *c_next(struct seq_file *m, void *v, loff_t *pos)
+{
+	++*pos;
+	return c_start(m, pos);
+}
+static void c_stop(struct seq_file *m, void *v)
+{
+}
+struct seq_operations cpuinfo_op = {
+	.start	= c_start,
+	.next	= c_next,
+	.stop	= c_stop,
+	.show	= show_cpuinfo,
+};
diff --git a/arch/x86/kernel/cpu/transmeta.c b/arch/x86/kernel/cpu/transmeta.c
new file mode 100644
index 00000000000..200fb3f9ebf
--- /dev/null
+++ b/arch/x86/kernel/cpu/transmeta.c
@@ -0,0 +1,116 @@
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <asm/processor.h>
+#include <asm/msr.h>
+#include "cpu.h"
+
+static void __cpuinit init_transmeta(struct cpuinfo_x86 *c)
+{
+	unsigned int cap_mask, uk, max, dummy;
+	unsigned int cms_rev1, cms_rev2;
+	unsigned int cpu_rev, cpu_freq = 0, cpu_flags, new_cpu_rev;
+	char cpu_info[65];
+
+	get_model_name(c);	/* Same as AMD/Cyrix */
+	display_cacheinfo(c);
+
+	/* Print CMS and CPU revision */
+	max = cpuid_eax(0x80860000);
+	cpu_rev = 0;
+	if ( max >= 0x80860001 ) {
+		cpuid(0x80860001, &dummy, &cpu_rev, &cpu_freq, &cpu_flags); 
+		if (cpu_rev != 0x02000000) {
+			printk(KERN_INFO "CPU: Processor revision %u.%u.%u.%u, %u MHz\n",
+				(cpu_rev >> 24) & 0xff,
+				(cpu_rev >> 16) & 0xff,
+				(cpu_rev >> 8) & 0xff,
+				cpu_rev & 0xff,
+				cpu_freq);
+		}
+	}
+	if ( max >= 0x80860002 ) {
+		cpuid(0x80860002, &new_cpu_rev, &cms_rev1, &cms_rev2, &dummy);
+		if (cpu_rev == 0x02000000) {
+			printk(KERN_INFO "CPU: Processor revision %08X, %u MHz\n",
+				new_cpu_rev, cpu_freq);
+		}
+		printk(KERN_INFO "CPU: Code Morphing Software revision %u.%u.%u-%u-%u\n",
+		       (cms_rev1 >> 24) & 0xff,
+		       (cms_rev1 >> 16) & 0xff,
+		       (cms_rev1 >> 8) & 0xff,
+		       cms_rev1 & 0xff,
+		       cms_rev2);
+	}
+	if ( max >= 0x80860006 ) {
+		cpuid(0x80860003,
+		      (void *)&cpu_info[0],
+		      (void *)&cpu_info[4],
+		      (void *)&cpu_info[8],
+		      (void *)&cpu_info[12]);
+		cpuid(0x80860004,
+		      (void *)&cpu_info[16],
+		      (void *)&cpu_info[20],
+		      (void *)&cpu_info[24],
+		      (void *)&cpu_info[28]);
+		cpuid(0x80860005,
+		      (void *)&cpu_info[32],
+		      (void *)&cpu_info[36],
+		      (void *)&cpu_info[40],
+		      (void *)&cpu_info[44]);
+		cpuid(0x80860006,
+		      (void *)&cpu_info[48],
+		      (void *)&cpu_info[52],
+		      (void *)&cpu_info[56],
+		      (void *)&cpu_info[60]);
+		cpu_info[64] = '\0';
+		printk(KERN_INFO "CPU: %s\n", cpu_info);
+	}
+
+	/* Unhide possibly hidden capability flags */
+	rdmsr(0x80860004, cap_mask, uk);
+	wrmsr(0x80860004, ~0, uk);
+	c->x86_capability[0] = cpuid_edx(0x00000001);
+	wrmsr(0x80860004, cap_mask, uk);
+
+	/* All Transmeta CPUs have a constant TSC */
+	set_bit(X86_FEATURE_CONSTANT_TSC, c->x86_capability);
+	
+	/* If we can run i686 user-space code, call us an i686 */
+#define USER686 ((1 << X86_FEATURE_TSC)|\
+		 (1 << X86_FEATURE_CX8)|\
+		 (1 << X86_FEATURE_CMOV))
+        if (c->x86 == 5 && (c->x86_capability[0] & USER686) == USER686)
+		c->x86 = 6;
+
+#ifdef CONFIG_SYSCTL
+	/* randomize_va_space slows us down enormously;
+	   it probably triggers retranslation of x86->native bytecode */
+	randomize_va_space = 0;
+#endif
+}
+
+static void __cpuinit transmeta_identify(struct cpuinfo_x86 * c)
+{
+	u32 xlvl;
+
+	/* Transmeta-defined flags: level 0x80860001 */
+	xlvl = cpuid_eax(0x80860000);
+	if ( (xlvl & 0xffff0000) == 0x80860000 ) {
+		if (  xlvl >= 0x80860001 )
+			c->x86_capability[2] = cpuid_edx(0x80860001);
+	}
+}
+
+static struct cpu_dev transmeta_cpu_dev __cpuinitdata = {
+	.c_vendor	= "Transmeta",
+	.c_ident	= { "GenuineTMx86", "TransmetaCPU" },
+	.c_init		= init_transmeta,
+	.c_identify	= transmeta_identify,
+};
+
+int __init transmeta_init_cpu(void)
+{
+	cpu_devs[X86_VENDOR_TRANSMETA] = &transmeta_cpu_dev;
+	return 0;
+}
diff --git a/arch/x86/kernel/cpu/umc.c b/arch/x86/kernel/cpu/umc.c
new file mode 100644
index 00000000000..a7a4e75bdcd
--- /dev/null
+++ b/arch/x86/kernel/cpu/umc.c
@@ -0,0 +1,26 @@
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <asm/processor.h>
+#include "cpu.h"
+
+/* UMC chips appear to be only either 386 or 486, so no special init takes place.
+ */
+
+static struct cpu_dev umc_cpu_dev __cpuinitdata = {
+	.c_vendor	= "UMC",
+	.c_ident 	= { "UMC UMC UMC" },
+	.c_models = {
+		{ .vendor = X86_VENDOR_UMC, .family = 4, .model_names =
+		  { 
+			  [1] = "U5D", 
+			  [2] = "U5S", 
+		  }
+		},
+	},
+};
+
+int __init umc_init_cpu(void)
+{
+	cpu_devs[X86_VENDOR_UMC] = &umc_cpu_dev;
+	return 0;
+}