Merge branch 'x86-64'

* x86-64: (83 commits)
  [PATCH] x86_64: x86_64 stack usage debugging
  [PATCH] x86_64: (resend) x86_64 stack overflow debugging
  [PATCH] x86_64: msi_apic.c build fix
  [PATCH] x86_64: i386/x86-64 Add nmi watchdog support for new Intel CPUs
  [PATCH] x86_64: Avoid broadcasting NMI IPIs
  [PATCH] x86_64: fix apic error on bootup
  [PATCH] x86_64: enlarge window for stack growth
  [PATCH] x86_64: Minor string functions optimizations
  [PATCH] x86_64: Move export symbols to their C functions
  [PATCH] x86_64: Standardize i386/x86_64 handling of NMI_VECTOR
  [PATCH] x86_64: Fix modular pc speaker
  [PATCH] x86_64: remove sys32_ni_syscall()
  [PATCH] x86_64: Do not use -ffunction-sections for modules
  [PATCH] x86_64: Add cpu_relax to apic_wait_icr_idle
  [PATCH] x86_64: adjust kstack_depth_to_print default
  [PATCH] i386/x86-64: adjust /proc/interrupts column headings
  [PATCH] x86_64: Fix race in cpu_local_* on preemptible kernels
  [PATCH] x86_64: Fix fast check in safe_smp_processor_id
  [PATCH] x86_64: x86_64 setup.c - printing cmp related boottime information
  [PATCH] i386/x86-64/ia64: Move polling flag into thread_info_status
  ...

Manual resolve of trivial conflict in arch/i386/kernel/Makefile

23 files changed, 689 insertions, 140 deletions

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index 374fb50608a..f3eaf22f273 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -328,6 +328,15 @@ config X86_MCE_P4THERMAL
 	  Enabling this feature will cause a message to be printed when the P4
 	  enters thermal throttling.
 
+config VM86
+	default y
+	bool "Enable VM86 support" if EMBEDDED
+	help
+          This option is required by programs like DOSEMU to run 16-bit legacy
+	  code on X86 processors. It also may be needed by software like
+          XFree86 to initialize some video cards via BIOS. Disabling this
+          option saves about 6k.
+
 config TOSHIBA
 	tristate "Toshiba Laptop support"
 	---help---
@@ -1068,6 +1077,10 @@ config SCx200HR_TIMER
 	  processor goes idle (as is done by the scheduler).  The
 	  other workaround is idle=poll boot option.
 
+config K8_NB
+	def_bool y
+	depends on AGP_AMD64
+
 source "drivers/pcmcia/Kconfig"
 
 source "drivers/pci/hotplug/Kconfig"
diff --git a/arch/i386/boot/Makefile b/arch/i386/boot/Makefile
index 33e55476381..e9794662606 100644
--- a/arch/i386/boot/Makefile
+++ b/arch/i386/boot/Makefile
@@ -109,8 +109,13 @@ fdimage288: $(BOOTIMAGE) $(obj)/mtools.conf
 isoimage: $(BOOTIMAGE)
 	-rm -rf $(obj)/isoimage
 	mkdir $(obj)/isoimage
-	cp `echo /usr/lib*/syslinux/isolinux.bin | awk '{ print $1; }'` \
-		$(obj)/isoimage
+	for i in lib lib64 share end ; do \
+		if [ -f /usr/$$i/syslinux/isolinux.bin ] ; then \
+			cp /usr/$$i/syslinux/isolinux.bin $(obj)/isoimage ; \
+			break ; \
+		fi ; \
+		if [ $$i = end ] ; then exit 1 ; fi ; \
+	done
 	cp $(BOOTIMAGE) $(obj)/isoimage/linux
 	echo '$(image_cmdline)' > $(obj)/isoimage/isolinux.cfg
 	if [ -f '$(FDINITRD)' ] ; then \
diff --git a/arch/i386/boot/compressed/misc.c b/arch/i386/boot/compressed/misc.c
index f19f3a7492a..b2ccd543410 100644
--- a/arch/i386/boot/compressed/misc.c
+++ b/arch/i386/boot/compressed/misc.c
@@ -24,14 +24,6 @@
 
 #undef memset
 #undef memcpy
-
-/*
- * Why do we do this? Don't ask me..
- *
- * Incomprehensible are the ways of bootloaders.
- */
-static void* memset(void *, int, size_t);
-static void* memcpy(void *, __const void *, size_t);
 #define memzero(s, n)     memset ((s), 0, (n))
 
 typedef unsigned char  uch;
@@ -93,7 +85,7 @@ static unsigned char *real_mode; /* Pointer to real-mode data */
 #endif
 #define RM_SCREEN_INFO (*(struct screen_info *)(real_mode+0))
 
-extern char input_data[];
+extern unsigned char input_data[];
 extern int input_len;
 
 static long bytes_out = 0;
@@ -103,6 +95,9 @@ static unsigned long output_ptr = 0;
 static void *malloc(int size);
 static void free(void *where);
 
+static void *memset(void *s, int c, unsigned n);
+static void *memcpy(void *dest, const void *src, unsigned n);
+
 static void putstr(const char *);
 
 extern int end;
@@ -205,7 +200,7 @@ static void putstr(const char *s)
 	outb_p(0xff & (pos >> 1), vidport+1);
 }
 
-static void* memset(void* s, int c, size_t n)
+static void* memset(void* s, int c, unsigned n)
 {
 	int i;
 	char *ss = (char*)s;
@@ -214,14 +209,13 @@ static void* memset(void* s, int c, size_t n)
 	return s;
 }
 
-static void* memcpy(void* __dest, __const void* __src,
-			    size_t __n)
+static void* memcpy(void* dest, const void* src, unsigned n)
 {
 	int i;
-	char *d = (char *)__dest, *s = (char *)__src;
+	char *d = (char *)dest, *s = (char *)src;
 
-	for (i=0;i<__n;i++) d[i] = s[i];
-	return __dest;
+	for (i=0;i<n;i++) d[i] = s[i];
+	return dest;
 }
 
 /* ===========================================================================
@@ -309,7 +303,7 @@ static void setup_normal_output_buffer(void)
 #else
 	if ((RM_ALT_MEM_K > RM_EXT_MEM_K ? RM_ALT_MEM_K : RM_EXT_MEM_K) < 1024) error("Less than 2MB of memory");
 #endif
-	output_data = (char *)__PHYSICAL_START; /* Normally Points to 1M */
+	output_data = (unsigned char *)__PHYSICAL_START; /* Normally Points to 1M */
 	free_mem_end_ptr = (long)real_mode;
 }
 
@@ -324,11 +318,9 @@ static void setup_output_buffer_if_we_run_high(struct moveparams *mv)
 #ifdef STANDARD_MEMORY_BIOS_CALL
 	if (RM_EXT_MEM_K < (3*1024)) error("Less than 4MB of memory");
 #else
-	if ((RM_ALT_MEM_K > RM_EXT_MEM_K ? RM_ALT_MEM_K : RM_EXT_MEM_K) <
-			(3*1024))
-		error("Less than 4MB of memory");
+	if ((RM_ALT_MEM_K > RM_EXT_MEM_K ? RM_ALT_MEM_K : RM_EXT_MEM_K) < (3*1024)) error("Less than 4MB of memory");
 #endif	
-	mv->low_buffer_start = output_data = (char *)LOW_BUFFER_START;
+	mv->low_buffer_start = output_data = (unsigned char *)LOW_BUFFER_START;
 	low_buffer_end = ((unsigned int)real_mode > LOW_BUFFER_MAX
 	  ? LOW_BUFFER_MAX : (unsigned int)real_mode) & ~0xfff;
 	low_buffer_size = low_buffer_end - LOW_BUFFER_START;
diff --git a/arch/i386/kernel/Makefile b/arch/i386/kernel/Makefile
index 0fac85df64f..5e70c2fb273 100644
--- a/arch/i386/kernel/Makefile
+++ b/arch/i386/kernel/Makefile
@@ -37,6 +37,7 @@ obj-$(CONFIG_DOUBLEFAULT) 	+= doublefault.o
 obj-$(CONFIG_VM86)		+= vm86.o
 obj-$(CONFIG_EARLY_PRINTK)	+= early_printk.o
 obj-$(CONFIG_HPET_TIMER) 	+= hpet.o
+obj-$(CONFIG_K8_NB)		+= k8.o
 
 EXTRA_AFLAGS   := -traditional
 
@@ -76,3 +77,6 @@ SYSCFLAGS_vsyscall-syms.o = -r
 $(obj)/vsyscall-syms.o: $(src)/vsyscall.lds \
 			$(obj)/vsyscall-sysenter.o $(obj)/vsyscall-note.o FORCE
 	$(call if_changed,syscall)
+
+k8-y                      += ../../x86_64/kernel/k8.o
+
diff --git a/arch/i386/kernel/alternative.c b/arch/i386/kernel/alternative.c
index 5cbd6f99fb2..50eb0e03777 100644
--- a/arch/i386/kernel/alternative.c
+++ b/arch/i386/kernel/alternative.c
@@ -4,27 +4,41 @@
 #include <asm/alternative.h>
 #include <asm/sections.h>
 
-#define DEBUG 0
-#if DEBUG
-# define DPRINTK(fmt, args...) printk(fmt, args)
-#else
-# define DPRINTK(fmt, args...)
-#endif
+static int no_replacement    = 0;
+static int smp_alt_once      = 0;
+static int debug_alternative = 0;
+
+static int __init noreplacement_setup(char *s)
+{
+	no_replacement = 1;
+	return 1;
+}
+static int __init bootonly(char *str)
+{
+	smp_alt_once = 1;
+	return 1;
+}
+static int __init debug_alt(char *str)
+{
+	debug_alternative = 1;
+	return 1;
+}
 
+__setup("noreplacement", noreplacement_setup);
+__setup("smp-alt-boot", bootonly);
+__setup("debug-alternative", debug_alt);
+
+#define DPRINTK(fmt, args...) if (debug_alternative) \
+	printk(KERN_DEBUG fmt, args)
+
+#ifdef GENERIC_NOP1
 /* Use inline assembly to define this because the nops are defined
    as inline assembly strings in the include files and we cannot
    get them easily into strings. */
 asm("\t.data\nintelnops: "
 	GENERIC_NOP1 GENERIC_NOP2 GENERIC_NOP3 GENERIC_NOP4 GENERIC_NOP5 GENERIC_NOP6
 	GENERIC_NOP7 GENERIC_NOP8);
-asm("\t.data\nk8nops: "
-	K8_NOP1 K8_NOP2 K8_NOP3 K8_NOP4 K8_NOP5 K8_NOP6
-	K8_NOP7 K8_NOP8);
-asm("\t.data\nk7nops: "
-	K7_NOP1 K7_NOP2 K7_NOP3 K7_NOP4 K7_NOP5 K7_NOP6
-	K7_NOP7 K7_NOP8);
-
-extern unsigned char intelnops[], k8nops[], k7nops[];
+extern unsigned char intelnops[];
 static unsigned char *intel_nops[ASM_NOP_MAX+1] = {
 	NULL,
 	intelnops,
@@ -36,6 +50,13 @@ static unsigned char *intel_nops[ASM_NOP_MAX+1] = {
 	intelnops + 1 + 2 + 3 + 4 + 5 + 6,
 	intelnops + 1 + 2 + 3 + 4 + 5 + 6 + 7,
 };
+#endif
+
+#ifdef K8_NOP1
+asm("\t.data\nk8nops: "
+	K8_NOP1 K8_NOP2 K8_NOP3 K8_NOP4 K8_NOP5 K8_NOP6
+	K8_NOP7 K8_NOP8);
+extern unsigned char k8nops[];
 static unsigned char *k8_nops[ASM_NOP_MAX+1] = {
 	NULL,
 	k8nops,
@@ -47,6 +68,13 @@ static unsigned char *k8_nops[ASM_NOP_MAX+1] = {
 	k8nops + 1 + 2 + 3 + 4 + 5 + 6,
 	k8nops + 1 + 2 + 3 + 4 + 5 + 6 + 7,
 };
+#endif
+
+#ifdef K7_NOP1
+asm("\t.data\nk7nops: "
+	K7_NOP1 K7_NOP2 K7_NOP3 K7_NOP4 K7_NOP5 K7_NOP6
+	K7_NOP7 K7_NOP8);
+extern unsigned char k7nops[];
 static unsigned char *k7_nops[ASM_NOP_MAX+1] = {
 	NULL,
 	k7nops,
@@ -58,6 +86,18 @@ static unsigned char *k7_nops[ASM_NOP_MAX+1] = {
 	k7nops + 1 + 2 + 3 + 4 + 5 + 6,
 	k7nops + 1 + 2 + 3 + 4 + 5 + 6 + 7,
 };
+#endif
+
+#ifdef CONFIG_X86_64
+
+extern char __vsyscall_0;
+static inline unsigned char** find_nop_table(void)
+{
+	return k8_nops;
+}
+
+#else /* CONFIG_X86_64 */
+
 static struct nop {
 	int cpuid;
 	unsigned char **noptable;
@@ -67,14 +107,6 @@ static struct nop {
 	{ -1, NULL }
 };
 
-
-extern struct alt_instr __alt_instructions[], __alt_instructions_end[];
-extern struct alt_instr __smp_alt_instructions[], __smp_alt_instructions_end[];
-extern u8 *__smp_locks[], *__smp_locks_end[];
-
-extern u8 __smp_alt_begin[], __smp_alt_end[];
-
-
 static unsigned char** find_nop_table(void)
 {
 	unsigned char **noptable = intel_nops;
@@ -89,6 +121,14 @@ static unsigned char** find_nop_table(void)
 	return noptable;
 }
 
+#endif /* CONFIG_X86_64 */
+
+extern struct alt_instr __alt_instructions[], __alt_instructions_end[];
+extern struct alt_instr __smp_alt_instructions[], __smp_alt_instructions_end[];
+extern u8 *__smp_locks[], *__smp_locks_end[];
+
+extern u8 __smp_alt_begin[], __smp_alt_end[];
+
 /* Replace instructions with better alternatives for this CPU type.
    This runs before SMP is initialized to avoid SMP problems with
    self modifying code. This implies that assymetric systems where
@@ -99,6 +139,7 @@ void apply_alternatives(struct alt_instr *start, struct alt_instr *end)
 {
 	unsigned char **noptable = find_nop_table();
 	struct alt_instr *a;
+	u8 *instr;
 	int diff, i, k;
 
 	DPRINTK("%s: alt table %p -> %p\n", __FUNCTION__, start, end);
@@ -106,7 +147,16 @@ void apply_alternatives(struct alt_instr *start, struct alt_instr *end)
 		BUG_ON(a->replacementlen > a->instrlen);
 		if (!boot_cpu_has(a->cpuid))
 			continue;
-		memcpy(a->instr, a->replacement, a->replacementlen);
+		instr = a->instr;
+#ifdef CONFIG_X86_64
+		/* vsyscall code is not mapped yet. resolve it manually. */
+		if (instr >= (u8 *)VSYSCALL_START && instr < (u8*)VSYSCALL_END) {
+			instr = __va(instr - (u8*)VSYSCALL_START + (u8*)__pa_symbol(&__vsyscall_0));
+			DPRINTK("%s: vsyscall fixup: %p => %p\n",
+				__FUNCTION__, a->instr, instr);
+		}
+#endif
+		memcpy(instr, a->replacement, a->replacementlen);
 		diff = a->instrlen - a->replacementlen;
 		/* Pad the rest with nops */
 		for (i = a->replacementlen; diff > 0; diff -= k, i += k) {
@@ -186,14 +236,6 @@ struct smp_alt_module {
 static LIST_HEAD(smp_alt_modules);
 static DEFINE_SPINLOCK(smp_alt);
 
-static int smp_alt_once = 0;
-static int __init bootonly(char *str)
-{
-	smp_alt_once = 1;
-	return 1;
-}
-__setup("smp-alt-boot", bootonly);
-
 void alternatives_smp_module_add(struct module *mod, char *name,
 				 void *locks, void *locks_end,
 				 void *text,  void *text_end)
@@ -201,6 +243,9 @@ void alternatives_smp_module_add(struct module *mod, char *name,
 	struct smp_alt_module *smp;
 	unsigned long flags;
 
+	if (no_replacement)
+		return;
+
 	if (smp_alt_once) {
 		if (boot_cpu_has(X86_FEATURE_UP))
 			alternatives_smp_unlock(locks, locks_end,
@@ -235,7 +280,7 @@ void alternatives_smp_module_del(struct module *mod)
 	struct smp_alt_module *item;
 	unsigned long flags;
 
-	if (smp_alt_once)
+	if (no_replacement || smp_alt_once)
 		return;
 
 	spin_lock_irqsave(&smp_alt, flags);
@@ -256,7 +301,7 @@ void alternatives_smp_switch(int smp)
 	struct smp_alt_module *mod;
 	unsigned long flags;
 
-	if (smp_alt_once)
+	if (no_replacement || smp_alt_once)
 		return;
 	BUG_ON(!smp && (num_online_cpus() > 1));
 
@@ -285,6 +330,13 @@ void alternatives_smp_switch(int smp)
 
 void __init alternative_instructions(void)
 {
+	if (no_replacement) {
+		printk(KERN_INFO "(SMP-)alternatives turned off\n");
+		free_init_pages("SMP alternatives",
+				(unsigned long)__smp_alt_begin,
+				(unsigned long)__smp_alt_end);
+		return;
+	}
 	apply_alternatives(__alt_instructions, __alt_instructions_end);
 
 	/* switch to patch-once-at-boottime-only mode and free the
diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c
index 5ab59c12335..7ce09492fc0 100644
--- a/arch/i386/kernel/apic.c
+++ b/arch/i386/kernel/apic.c
@@ -36,6 +36,7 @@
 #include <asm/arch_hooks.h>
 #include <asm/hpet.h>
 #include <asm/i8253.h>
+#include <asm/nmi.h>
 
 #include <mach_apic.h>
 #include <mach_apicdef.h>
@@ -156,7 +157,7 @@ void clear_local_APIC(void)
 	maxlvt = get_maxlvt();
 
 	/*
-	 * Masking an LVT entry on a P6 can trigger a local APIC error
+	 * Masking an LVT entry can trigger a local APIC error
 	 * if the vector is zero. Mask LVTERR first to prevent this.
 	 */
 	if (maxlvt >= 3) {
@@ -1117,7 +1118,18 @@ void disable_APIC_timer(void)
 		unsigned long v;
 
 		v = apic_read(APIC_LVTT);
-		apic_write_around(APIC_LVTT, v | APIC_LVT_MASKED);
+		/*
+		 * When an illegal vector value (0-15) is written to an LVT
+		 * entry and delivery mode is Fixed, the APIC may signal an
+		 * illegal vector error, with out regard to whether the mask
+		 * bit is set or whether an interrupt is actually seen on input.
+		 *
+		 * Boot sequence might call this function when the LVTT has
+		 * '0' vector value. So make sure vector field is set to
+		 * valid value.
+		 */
+		v |= (APIC_LVT_MASKED | LOCAL_TIMER_VECTOR);
+		apic_write_around(APIC_LVTT, v);
 	}
 }
 
diff --git a/arch/i386/kernel/apm.c b/arch/i386/kernel/apm.c
index 9e819eb6822..7c5729d1fd0 100644
--- a/arch/i386/kernel/apm.c
+++ b/arch/i386/kernel/apm.c
@@ -764,9 +764,9 @@ static int apm_do_idle(void)
 	int	idled = 0;
 	int	polling;
 
-	polling = test_thread_flag(TIF_POLLING_NRFLAG);
+	polling = !!(current_thread_info()->status & TS_POLLING);
 	if (polling) {
-		clear_thread_flag(TIF_POLLING_NRFLAG);
+		current_thread_info()->status &= ~TS_POLLING;
 		smp_mb__after_clear_bit();
 	}
 	if (!need_resched()) {
@@ -774,7 +774,7 @@ static int apm_do_idle(void)
 		ret = apm_bios_call_simple(APM_FUNC_IDLE, 0, 0, &eax);
 	}
 	if (polling)
-		set_thread_flag(TIF_POLLING_NRFLAG);
+		current_thread_info()->status |= TS_POLLING;
 
 	if (!idled)
 		return 0;
diff --git a/arch/i386/kernel/cpu/amd.c b/arch/i386/kernel/cpu/amd.c
index 786d1a57048..fd0457c9c82 100644
--- a/arch/i386/kernel/cpu/amd.c
+++ b/arch/i386/kernel/cpu/amd.c
@@ -224,15 +224,17 @@ static void __init init_amd(struct cpuinfo_x86 *c)
 
 #ifdef CONFIG_X86_HT
 	/*
-	 * On a AMD dual core setup the lower bits of the APIC id
-	 * distingush the cores.  Assumes number of cores is a power
-	 * of two.
+	 * 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 = 0;
-		while ((1 << bits) < c->x86_max_cores)
-			bits++;
+		unsigned bits = (cpuid_ecx(0x80000008) >> 12) & 0xf;
+
+		if (bits == 0) {
+			while ((1 << bits) < c->x86_max_cores)
+				bits++;
+		}
 		cpu_core_id[cpu] = phys_proc_id[cpu] & ((1<<bits)-1);
 		phys_proc_id[cpu] >>= bits;
 		printk(KERN_INFO "CPU %d(%d) -> Core %d\n",
@@ -240,6 +242,8 @@ static void __init init_amd(struct cpuinfo_x86 *c)
 	}
 #endif
 
+	if (cpuid_eax(0x80000000) >= 0x80000006)
+		num_cache_leaves = 3;
 }
 
 static unsigned int amd_size_cache(struct cpuinfo_x86 * c, unsigned int size)
diff --git a/arch/i386/kernel/cpu/intel.c b/arch/i386/kernel/cpu/intel.c
index 5386b29bb5a..10afc645c54 100644
--- a/arch/i386/kernel/cpu/intel.c
+++ b/arch/i386/kernel/cpu/intel.c
@@ -122,6 +122,12 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 
 	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)
diff --git a/arch/i386/kernel/cpu/intel_cacheinfo.c b/arch/i386/kernel/cpu/intel_cacheinfo.c
index c8547a6fa7e..6c37b4fd8ce 100644
--- a/arch/i386/kernel/cpu/intel_cacheinfo.c
+++ b/arch/i386/kernel/cpu/intel_cacheinfo.c
@@ -4,6 +4,7 @@
  *      Changes:
  *      Venkatesh Pallipadi	: Adding cache identification through cpuid(4)
  *		Ashok Raj <ashok.raj@intel.com>: Work with CPU hotplug infrastructure.
+ *	Andi Kleen		: CPUID4 emulation on AMD.
  */
 
 #include <linux/init.h>
@@ -130,25 +131,111 @@ struct _cpuid4_info {
 	cpumask_t shared_cpu_map;
 };
 
-static unsigned short			num_cache_leaves;
+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:
+   No L3, 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;
+};
+
+static unsigned short assocs[] = {
+	[1] = 1, [2] = 2, [4] = 4, [6] = 8,
+	[8] = 16,
+	[0xf] = 0xffff // ??
+	};
+static unsigned char levels[] = { 1, 1, 2 };
+static unsigned char types[] = { 1, 2, 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;
+
+	eax->full = 0;
+	ebx->full = 0;
+	ecx->full = 0;
+
+	cpuid(0x80000005, &dummy, &dummy, &l1d.val, &l1i.val);
+	cpuid(0x80000006, &dummy, &dummy, &l2.val, &dummy);
+
+	if (leaf > 2 || !l1d.val || !l1i.val || !l2.val)
+		return;
+
+	eax->split.is_self_initializing = 1;
+	eax->split.type = types[leaf];
+	eax->split.level = levels[leaf];
+	eax->split.num_threads_sharing = 0;
+	eax->split.num_cores_on_die = current_cpu_data.x86_max_cores - 1;
+
+	if (leaf <= 1) {
+		union l1_cache *l1 = leaf == 0 ? &l1d : &l1i;
+		assoc = l1->assoc;
+		line_size = l1->line_size;
+		lines_per_tag = l1->lines_per_tag;
+		size_in_kb = l1->size_in_kb;
+	} else {
+		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;
+	}
+
+	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)
 {
-	unsigned int		eax, ebx, ecx, edx;
-	union _cpuid4_leaf_eax	cache_eax;
+	union _cpuid4_leaf_eax 	eax;
+	union _cpuid4_leaf_ebx 	ebx;
+	union _cpuid4_leaf_ecx 	ecx;
+	unsigned		edx;
 
-	cpuid_count(4, index, &eax, &ebx, &ecx, &edx);
-	cache_eax.full = eax;
-	if (cache_eax.split.type == CACHE_TYPE_NULL)
+	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.full = eax;
-	this_leaf->ebx.full = ebx;
-	this_leaf->ecx.full = ecx;
-	this_leaf->size = (this_leaf->ecx.split.number_of_sets + 1) *
-		(this_leaf->ebx.split.coherency_line_size + 1) *
-		(this_leaf->ebx.split.physical_line_partition + 1) *
-		(this_leaf->ebx.split.ways_of_associativity + 1);
+	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;
 }
 
diff --git a/arch/i386/kernel/crash.c b/arch/i386/kernel/crash.c
index 21dc1bbb806..0c88d3ec8c1 100644
--- a/arch/i386/kernel/crash.c
+++ b/arch/i386/kernel/crash.c
@@ -120,14 +120,9 @@ static int crash_nmi_callback(struct pt_regs *regs, int cpu)
 	return 1;
 }
 
-/*
- * By using the NMI code instead of a vector we just sneak thru the
- * word generator coming out with just what we want.  AND it does
- * not matter if clustered_apic_mode is set or not.
- */
 static void smp_send_nmi_allbutself(void)
 {
-	send_IPI_allbutself(APIC_DM_NMI);
+	send_IPI_allbutself(NMI_VECTOR);
 }
 
 static void nmi_shootdown_cpus(void)
diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S
index cfc683f153b..e6e4506e749 100644
--- a/arch/i386/kernel/entry.S
+++ b/arch/i386/kernel/entry.S
@@ -48,6 +48,7 @@
 #include <asm/smp.h>
 #include <asm/page.h>
 #include <asm/desc.h>
+#include <asm/dwarf2.h>
 #include "irq_vectors.h"
 
 #define nr_syscalls ((syscall_table_size)/4)
@@ -85,31 +86,67 @@ VM_MASK		= 0x00020000
 #define SAVE_ALL \
 	cld; \
 	pushl %es; \
+	CFI_ADJUST_CFA_OFFSET 4;\
+	/*CFI_REL_OFFSET es, 0;*/\
 	pushl %ds; \
+	CFI_ADJUST_CFA_OFFSET 4;\
+	/*CFI_REL_OFFSET ds, 0;*/\
 	pushl %eax; \
+	CFI_ADJUST_CFA_OFFSET 4;\
+	CFI_REL_OFFSET eax, 0;\
 	pushl %ebp; \
+	CFI_ADJUST_CFA_OFFSET 4;\
+	CFI_REL_OFFSET ebp, 0;\
 	pushl %edi; \
+	CFI_ADJUST_CFA_OFFSET 4;\
+	CFI_REL_OFFSET edi, 0;\
 	pushl %esi; \
+	CFI_ADJUST_CFA_OFFSET 4;\
+	CFI_REL_OFFSET esi, 0;\
 	pushl %edx; \
+	CFI_ADJUST_CFA_OFFSET 4;\
+	CFI_REL_OFFSET edx, 0;\
 	pushl %ecx; \
+	CFI_ADJUST_CFA_OFFSET 4;\
+	CFI_REL_OFFSET ecx, 0;\
 	pushl %ebx; \
+	CFI_ADJUST_CFA_OFFSET 4;\
+	CFI_REL_OFFSET ebx, 0;\
 	movl $(__USER_DS), %edx; \
 	movl %edx, %ds; \
 	movl %edx, %es;
 
 #define RESTORE_INT_REGS \
 	popl %ebx;	\
+	CFI_ADJUST_CFA_OFFSET -4;\
+	CFI_RESTORE ebx;\
 	popl %ecx;	\
+	CFI_ADJUST_CFA_OFFSET -4;\
+	CFI_RESTORE ecx;\
 	popl %edx;	\
+	CFI_ADJUST_CFA_OFFSET -4;\
+	CFI_RESTORE edx;\
 	popl %esi;	\
+	CFI_ADJUST_CFA_OFFSET -4;\
+	CFI_RESTORE esi;\
 	popl %edi;	\
+	CFI_ADJUST_CFA_OFFSET -4;\
+	CFI_RESTORE edi;\
 	popl %ebp;	\
-	popl %eax
+	CFI_ADJUST_CFA_OFFSET -4;\
+	CFI_RESTORE ebp;\
+	popl %eax;	\
+	CFI_ADJUST_CFA_OFFSET -4;\
+	CFI_RESTORE eax
 
 #define RESTORE_REGS	\
 	RESTORE_INT_REGS; \
 1:	popl %ds;	\
+	CFI_ADJUST_CFA_OFFSET -4;\
+	/*CFI_RESTORE ds;*/\
 2:	popl %es;	\
+	CFI_ADJUST_CFA_OFFSET -4;\
+	/*CFI_RESTORE es;*/\
 .section .fixup,"ax";	\
 3:	movl $0,(%esp);	\
 	jmp 1b;		\
@@ -122,13 +159,43 @@ VM_MASK		= 0x00020000
 	.long 2b,4b;	\
 .previous
 
+#define RING0_INT_FRAME \
+	CFI_STARTPROC simple;\
+	CFI_DEF_CFA esp, 3*4;\
+	/*CFI_OFFSET cs, -2*4;*/\
+	CFI_OFFSET eip, -3*4
+
+#define RING0_EC_FRAME \
+	CFI_STARTPROC simple;\
+	CFI_DEF_CFA esp, 4*4;\
+	/*CFI_OFFSET cs, -2*4;*/\
+	CFI_OFFSET eip, -3*4
+
+#define RING0_PTREGS_FRAME \
+	CFI_STARTPROC simple;\
+	CFI_DEF_CFA esp, OLDESP-EBX;\
+	/*CFI_OFFSET cs, CS-OLDESP;*/\
+	CFI_OFFSET eip, EIP-OLDESP;\
+	/*CFI_OFFSET es, ES-OLDESP;*/\
+	/*CFI_OFFSET ds, DS-OLDESP;*/\
+	CFI_OFFSET eax, EAX-OLDESP;\
+	CFI_OFFSET ebp, EBP-OLDESP;\
+	CFI_OFFSET edi, EDI-OLDESP;\
+	CFI_OFFSET esi, ESI-OLDESP;\
+	CFI_OFFSET edx, EDX-OLDESP;\
+	CFI_OFFSET ecx, ECX-OLDESP;\
+	CFI_OFFSET ebx, EBX-OLDESP
 
 ENTRY(ret_from_fork)
+	CFI_STARTPROC
 	pushl %eax
+	CFI_ADJUST_CFA_OFFSET -4
 	call schedule_tail
 	GET_THREAD_INFO(%ebp)
 	popl %eax
+	CFI_ADJUST_CFA_OFFSET -4
 	jmp syscall_exit
+	CFI_ENDPROC
 
 /*
  * Return to user mode is not as complex as all this looks,
@@ -139,6 +206,7 @@ ENTRY(ret_from_fork)
 
 	# userspace resumption stub bypassing syscall exit tracing
 	ALIGN
+	RING0_PTREGS_FRAME
 ret_from_exception:
 	preempt_stop
 ret_from_intr:
@@ -171,20 +239,33 @@ need_resched:
 	call preempt_schedule_irq
 	jmp need_resched
 #endif
+	CFI_ENDPROC
 
 /* SYSENTER_RETURN points to after the "sysenter" instruction in
    the vsyscall page.  See vsyscall-sysentry.S, which defines the symbol.  */
 
 	# sysenter call handler stub
 ENTRY(sysenter_entry)
+	CFI_STARTPROC simple
+	CFI_DEF_CFA esp, 0
+	CFI_REGISTER esp, ebp
 	movl TSS_sysenter_esp0(%esp),%esp
 sysenter_past_esp:
 	sti
 	pushl $(__USER_DS)
+	CFI_ADJUST_CFA_OFFSET 4
+	/*CFI_REL_OFFSET ss, 0*/
 	pushl %ebp
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET esp, 0
 	pushfl
+	CFI_ADJUST_CFA_OFFSET 4
 	pushl $(__USER_CS)
+	CFI_ADJUST_CFA_OFFSET 4
+	/*CFI_REL_OFFSET cs, 0*/
 	pushl $SYSENTER_RETURN
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET eip, 0
 
 /*
  * Load the potential sixth argument from user stack.
@@ -199,6 +280,7 @@ sysenter_past_esp:
 .previous
 
 	pushl %eax
+	CFI_ADJUST_CFA_OFFSET 4
 	SAVE_ALL
 	GET_THREAD_INFO(%ebp)
 
@@ -219,11 +301,14 @@ sysenter_past_esp:
 	xorl %ebp,%ebp
 	sti
 	sysexit
+	CFI_ENDPROC
 
 
 	# system call handler stub
 ENTRY(system_call)
+	RING0_INT_FRAME			# can't unwind into user space anyway
 	pushl %eax			# save orig_eax
+	CFI_ADJUST_CFA_OFFSET 4
 	SAVE_ALL
 	GET_THREAD_INFO(%ebp)
 	testl $TF_MASK,EFLAGS(%esp)
@@ -256,10 +341,12 @@ restore_all:
 	movb CS(%esp), %al
 	andl $(VM_MASK | (4 << 8) | 3), %eax
 	cmpl $((4 << 8) | 3), %eax
+	CFI_REMEMBER_STATE
 	je ldt_ss			# returning to user-space with LDT SS
 restore_nocheck:
 	RESTORE_REGS
 	addl $4, %esp
+	CFI_ADJUST_CFA_OFFSET -4
 1:	iret
 .section .fixup,"ax"
 iret_exc:
@@ -273,6 +360,7 @@ iret_exc:
 	.long 1b,iret_exc
 .previous
 
+	CFI_RESTORE_STATE
 ldt_ss:
 	larl OLDSS(%esp), %eax
 	jnz restore_nocheck
@@ -285,11 +373,13 @@ ldt_ss:
 	 * CPUs, which we can try to work around to make
 	 * dosemu and wine happy. */
 	subl $8, %esp		# reserve space for switch16 pointer
+	CFI_ADJUST_CFA_OFFSET 8
 	cli
 	movl %esp, %eax
 	/* Set up the 16bit stack frame with switch32 pointer on top,
 	 * and a switch16 pointer on top of the current frame. */
 	call setup_x86_bogus_stack
+	CFI_ADJUST_CFA_OFFSET -8	# frame has moved
 	RESTORE_REGS
 	lss 20+4(%esp), %esp	# switch to 16bit stack
 1:	iret
@@ -297,9 +387,11 @@ ldt_ss:
 	.align 4
 	.long 1b,iret_exc
 .previous
+	CFI_ENDPROC
 
 	# perform work that needs to be done immediately before resumption
 	ALIGN
+	RING0_PTREGS_FRAME		# can't unwind into user space anyway
 work_pending:
 	testb $_TIF_NEED_RESCHED, %cl