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

Documentation/x86_64/boot-options.txt

@@ -205,6 +205,27 @@ IOMMU
   pages  Prereserve that many 128K pages for the software IO bounce buffering.
   force  Force all IO through the software TLB.
 
+  calgary=[64k,128k,256k,512k,1M,2M,4M,8M]
+  calgary=[translate_empty_slots]
+  calgary=[disable=<PCI bus number>]
+
+    64k,...,8M - Set the size of each PCI slot's translation table
+    when using the Calgary IOMMU. This is the size of the translation
+    table itself in main memory. The smallest table, 64k, covers an IO
+    space of 32MB; the largest, 8MB table, can cover an IO space of
+    4GB. Normally the kernel will make the right choice by itself.
+
+    translate_empty_slots - Enable translation even on slots that have
+    no devices attached to them, in case a device will be hotplugged
+    in the future.
+
+    disable=<PCI bus number> - Disable translation on a given PHB. For
+    example, the built-in graphics adapter resides on the first bridge
+    (PCI bus number 0); if translation (isolation) is enabled on this
+    bridge, X servers that access the hardware directly from user
+    space might stop working. Use this option if you have devices that
+    are accessed from userspace directly on some PCI host bridge.
+
 Debugging
 
   oops=panic Always panic on oopses. Default is to just kill the process,

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"

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 \

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;

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
+

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
@@ -99,14 +139,24 @@ 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);
 	for (a = start; a < end; a++) {
 		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) {
@@ -186,21 +236,16 @@ 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)
 {
 	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