Merge branch 'x86-stage-3-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip

* 'x86-stage-3-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (190 commits)
  Revert "cpuacct: reduce one NULL check in fast-path"
  Revert "x86: don't compile vsmp_64 for 32bit"
  x86: Correct behaviour of irq affinity
  x86: early_ioremap_init(), use __fix_to_virt(), because we are sure it's safe
  x86: use default_cpu_mask_to_apicid for 64bit
  x86: fix set_extra_move_desc calling
  x86, PAT, PCI: Change vma prot in pci_mmap to reflect inherited prot
  x86/dmi: fix dmi_alloc() section mismatches
  x86: e820 fix various signedness issues in setup.c and e820.c
  x86: apic/io_apic.c define msi_ir_chip and ir_ioapic_chip all the time
  x86: irq.c keep CONFIG_X86_LOCAL_APIC interrupts together
  x86: irq.c use same path for show_interrupts
  x86: cpu/cpu.h cleanup
  x86: Fix a couple of sparse warnings in arch/x86/kernel/apic/io_apic.c
  Revert "x86: create a non-zero sized bm_pte only when needed"
  x86: pci-nommu.c cleanup
  x86: io_delay.c cleanup
  x86: rtc.c cleanup
  x86: i8253 cleanup
  x86: kdebugfs.c cleanup
  ...

Documentation/x86/earlyprintk.txt

@@ -0,0 +1,101 @@
+
+Mini-HOWTO for using the earlyprintk=dbgp boot option with a
+USB2 Debug port key and a debug cable, on x86 systems.
+
+You need two computers, the 'USB debug key' special gadget and
+and two USB cables, connected like this:
+
+  [host/target] <-------> [USB debug key] <-------> [client/console]
+
+1. There are three specific hardware requirements:
+
+ a.) Host/target system needs to have USB debug port capability.
+
+ You can check this capability by looking at a 'Debug port' bit in
+ the lspci -vvv output:
+
+ # lspci -vvv
+ ...
+ 00:1d.7 USB Controller: Intel Corporation 82801H (ICH8 Family) USB2 EHCI Controller #1 (rev 03) (prog-if 20 [EHCI])
+         Subsystem: Lenovo ThinkPad T61
+         Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR+ FastB2B- DisINTx-
+         Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
+         Latency: 0
+         Interrupt: pin D routed to IRQ 19
+         Region 0: Memory at fe227000 (32-bit, non-prefetchable) [size=1K]
+         Capabilities: [50] Power Management version 2
+                 Flags: PMEClk- DSI- D1- D2- AuxCurrent=375mA PME(D0+,D1-,D2-,D3hot+,D3cold+)
+                 Status: D0 PME-Enable- DSel=0 DScale=0 PME+
+         Capabilities: [58] Debug port: BAR=1 offset=00a0
+                            ^^^^^^^^^^^ <==================== [ HERE ]
+	 Kernel driver in use: ehci_hcd
+         Kernel modules: ehci-hcd
+ ...
+
+( If your system does not list a debug port capability then you probably
+  wont be able to use the USB debug key. )
+
+ b.) You also need a Netchip USB debug cable/key:
+
+        http://www.plxtech.com/products/NET2000/NET20DC/default.asp
+
+     This is a small blue plastic connector with two USB connections,
+     it draws power from its USB connections.
+
+ c.) Thirdly, you need a second client/console system with a regular USB port.
+
+2. Software requirements:
+
+ a.) On the host/target system:
+
+    You need to enable the following kernel config option:
+
+      CONFIG_EARLY_PRINTK_DBGP=y
+
+    And you need to add the boot command line: "earlyprintk=dbgp".
+    (If you are using Grub, append it to the 'kernel' line in
+     /etc/grub.conf)
+
+    NOTE: normally earlyprintk console gets turned off once the
+    regular console is alive - use "earlyprintk=dbgp,keep" to keep
+    this channel open beyond early bootup. This can be useful for
+    debugging crashes under Xorg, etc.
+
+ b.) On the client/console system:
+
+    You should enable the following kernel config option:
+
+      CONFIG_USB_SERIAL_DEBUG=y
+
+    On the next bootup with the modified kernel you should
+    get a /dev/ttyUSBx device(s).
+
+    Now this channel of kernel messages is ready to be used: start
+    your favorite terminal emulator (minicom, etc.) and set
+    it up to use /dev/ttyUSB0 - or use a raw 'cat /dev/ttyUSBx' to
+    see the raw output.
+
+ c.) On Nvidia Southbridge based systems: the kernel will try to probe
+     and find out which port has debug device connected.
+
+3. Testing that it works fine:
+
+   You can test the output by using earlyprintk=dbgp,keep and provoking
+   kernel messages on the host/target system. You can provoke a harmless
+   kernel message by for example doing:
+
+     echo h > /proc/sysrq-trigger
+
+   On the host/target system you should see this help line in "dmesg" output:
+
+     SysRq : HELP : loglevel(0-9) reBoot Crashdump terminate-all-tasks(E) memory-full-oom-kill(F) kill-all-tasks(I) saK show-backtrace-all-active-cpus(L) show-memory-usage(M) nice-all-RT-tasks(N) powerOff show-registers(P) show-all-timers(Q) unRaw Sync show-task-states(T) Unmount show-blocked-tasks(W) dump-ftrace-buffer(Z)
+
+   On the client/console system do:
+
+       cat /dev/ttyUSB0
+
+   And you should see the help line above displayed shortly after you've
+   provoked it on the host system.
+
+If it does not work then please ask about it on the linux-kernel@vger.kernel.org
+mailing list or contact the x86 maintainers.

arch/x86/Kconfig

@@ -786,6 +786,11 @@ config X86_MCE_AMD
 	   Additional support for AMD specific MCE features such as
 	   the DRAM Error Threshold.
 
+config X86_MCE_THRESHOLD
+	depends on X86_MCE_AMD || X86_MCE_INTEL
+	bool
+	default y
+
 config X86_MCE_NONFATAL
 	tristate "Check for non-fatal errors on AMD Athlon/Duron / Intel Pentium 4"
 	depends on X86_32 && X86_MCE
@@ -929,6 +934,12 @@ config X86_CPUID
 	  with major 203 and minors 0 to 31 for /dev/cpu/0/cpuid to
 	  /dev/cpu/31/cpuid.
 
+config X86_CPU_DEBUG
+	tristate "/sys/kernel/debug/x86/cpu/* - CPU Debug support"
+	---help---
+	  If you select this option, this will provide various x86 CPUs
+	  information through debugfs.
+
 choice
 	prompt "High Memory Support"
 	default HIGHMEM4G if !X86_NUMAQ
@@ -1121,7 +1132,7 @@ config NUMA_EMU
 
 config NODES_SHIFT
 	int "Maximum NUMA Nodes (as a power of 2)" if !MAXSMP
-	range 1 9   if X86_64
+	range 1 9
 	default "9" if MAXSMP
 	default "6" if X86_64
 	default "4" if X86_NUMAQ
@@ -1429,7 +1440,7 @@ config CRASH_DUMP
 config KEXEC_JUMP
 	bool "kexec jump (EXPERIMENTAL)"
 	depends on EXPERIMENTAL
-	depends on KEXEC && HIBERNATION && X86_32
+	depends on KEXEC && HIBERNATION
 	---help---
 	  Jump between original kernel and kexeced kernel and invoke
 	  code in physical address mode via KEXEC

arch/x86/Kconfig.cpu

@@ -456,24 +456,9 @@ config CPU_SUP_AMD
 
 	  If unsure, say N.
 
-config CPU_SUP_CENTAUR_32
+config CPU_SUP_CENTAUR
 	default y
 	bool "Support Centaur processors" if PROCESSOR_SELECT
-	depends on !64BIT
-	---help---
-	  This enables detection, tunings and quirks for Centaur processors
-
-	  You need this enabled if you want your kernel to run on a
-	  Centaur CPU. Disabling this option on other types of CPUs
-	  makes the kernel a tiny bit smaller. Disabling it on a Centaur
-	  CPU might render the kernel unbootable.
-
-	  If unsure, say N.
-
-config CPU_SUP_CENTAUR_64
-	default y
-	bool "Support Centaur processors" if PROCESSOR_SELECT
-	depends on 64BIT
 	---help---
 	  This enables detection, tunings and quirks for Centaur processors
 

arch/x86/Makefile

@@ -153,34 +153,23 @@ endif
 
 boot := arch/x86/boot
 
-PHONY += zImage bzImage compressed zlilo bzlilo \
+BOOT_TARGETS = bzlilo bzdisk fdimage fdimage144 fdimage288 isoimage install
-         zdisk bzdisk fdimage fdimage144 fdimage288 isoimage install
+
+PHONY += bzImage $(BOOT_TARGETS)
 
 # Default kernel to build
 all: bzImage
 
 # KBUILD_IMAGE specify target image being built
-                    KBUILD_IMAGE := $(boot)/bzImage
+KBUILD_IMAGE := $(boot)/bzImage
-zImage zlilo zdisk: KBUILD_IMAGE := $(boot)/zImage
 
-zImage bzImage: vmlinux
+bzImage: vmlinux
 	$(Q)$(MAKE) $(build)=$(boot) $(KBUILD_IMAGE)
 	$(Q)mkdir -p $(objtree)/arch/$(UTS_MACHINE)/boot
 	$(Q)ln -fsn ../../x86/boot/bzImage $(objtree)/arch/$(UTS_MACHINE)/boot/$@
 
-compressed: zImage
+$(BOOT_TARGETS): vmlinux
-
+	$(Q)$(MAKE) $(build)=$(boot) $@
-zlilo bzlilo: vmlinux
-	$(Q)$(MAKE) $(build)=$(boot) BOOTIMAGE=$(KBUILD_IMAGE) zlilo
-
-zdisk bzdisk: vmlinux
-	$(Q)$(MAKE) $(build)=$(boot) BOOTIMAGE=$(KBUILD_IMAGE) zdisk
-
-fdimage fdimage144 fdimage288 isoimage: vmlinux
-	$(Q)$(MAKE) $(build)=$(boot) BOOTIMAGE=$(KBUILD_IMAGE) $@
-
-install:
-	$(Q)$(MAKE) $(build)=$(boot) BOOTIMAGE=$(KBUILD_IMAGE) install
 
 PHONY += vdso_install
 vdso_install:
@@ -205,7 +194,3 @@ define archhelp
   echo  '                  FDARGS="..."  arguments for the booted kernel'
   echo  '                  FDINITRD=file initrd for the booted kernel'
 endef
-
-CLEAN_FILES += arch/x86/boot/fdimage \
-	       arch/x86/boot/image.iso \
-	       arch/x86/boot/mtools.conf

arch/x86/boot/Makefile

@@ -6,26 +6,24 @@
 # for more details.
 #
 # Copyright (C) 1994 by Linus Torvalds
+# Changed by many, many contributors over the years.
 #
 
 # ROOT_DEV specifies the default root-device when making the image.
 # This can be either FLOPPY, CURRENT, /dev/xxxx or empty, in which case
 # the default of FLOPPY is used by 'build'.
 
-ROOT_DEV := CURRENT
+ROOT_DEV	:= CURRENT
 
 # If you want to preset the SVGA mode, uncomment the next line and
 # set SVGA_MODE to whatever number you want.
 # Set it to -DSVGA_MODE=NORMAL_VGA if you just want the EGA/VGA mode.
 # The number is the same as you would ordinarily press at bootup.
 
-SVGA_MODE := -DSVGA_MODE=NORMAL_VGA
+SVGA_MODE	:= -DSVGA_MODE=NORMAL_VGA
 
-# If you want the RAM disk device, define this to be the size in blocks.
+targets		:= vmlinux.bin setup.bin setup.elf bzImage
-
+targets		+= fdimage fdimage144 fdimage288 image.iso mtools.conf
-#RAMDISK := -DRAMDISK=512
-
-targets		:= vmlinux.bin setup.bin setup.elf zImage bzImage
 subdir-		:= compressed
 
 setup-y		+= a20.o cmdline.o copy.o cpu.o cpucheck.o edd.o
@@ -71,17 +69,13 @@ KBUILD_CFLAGS	:= $(LINUXINCLUDE) -g -Os -D_SETUP -D__KERNEL__ \
 KBUILD_CFLAGS +=   $(call cc-option,-m32)
 KBUILD_AFLAGS	:= $(KBUILD_CFLAGS) -D__ASSEMBLY__
 
-$(obj)/zImage:  asflags-y := $(SVGA_MODE) $(RAMDISK)
+$(obj)/bzImage: asflags-y  := $(SVGA_MODE)
-$(obj)/bzImage: ccflags-y := -D__BIG_KERNEL__
-$(obj)/bzImage: asflags-y := $(SVGA_MODE) $(RAMDISK) -D__BIG_KERNEL__
-$(obj)/bzImage: BUILDFLAGS   := -b
 
 quiet_cmd_image = BUILD   $@
-cmd_image = $(obj)/tools/build $(BUILDFLAGS) $(obj)/setup.bin \
+cmd_image = $(obj)/tools/build $(obj)/setup.bin $(obj)/vmlinux.bin \
-	    $(obj)/vmlinux.bin $(ROOT_DEV) > $@
+	$(ROOT_DEV) > $@
 
-$(obj)/zImage $(obj)/bzImage: $(obj)/setup.bin \
+$(obj)/bzImage: $(obj)/setup.bin $(obj)/vmlinux.bin $(obj)/tools/build FORCE
-			      $(obj)/vmlinux.bin $(obj)/tools/build FORCE
 	$(call if_changed,image)
 	@echo 'Kernel: $@ is ready' ' (#'`cat .version`')'
 
@@ -116,9 +110,11 @@ $(obj)/setup.bin: $(obj)/setup.elf FORCE
 $(obj)/compressed/vmlinux: FORCE
 	$(Q)$(MAKE) $(build)=$(obj)/compressed $@
 
-# Set this if you want to pass append arguments to the zdisk/fdimage/isoimage kernel
+# Set this if you want to pass append arguments to the
+# bzdisk/fdimage/isoimage kernel
 FDARGS =
-# Set this if you want an initrd included with the zdisk/fdimage/isoimage kernel
+# Set this if you want an initrd included with the
+# bzdisk/fdimage/isoimage kernel
 FDINITRD =
 
 image_cmdline = default linux $(FDARGS) $(if $(FDINITRD),initrd=initrd.img,)
@@ -127,7 +123,7 @@ $(obj)/mtools.conf: $(src)/mtools.conf.in
 	sed -e 's|@OBJ@|$(obj)|g' < $< > $@
 
 # This requires write access to /dev/fd0
-zdisk: $(BOOTIMAGE) $(obj)/mtools.conf
+bzdisk: $(obj)/bzImage $(obj)/mtools.conf
 	MTOOLSRC=$(obj)/mtools.conf mformat a:			; sync
 	syslinux /dev/fd0					; sync
 	echo '$(image_cmdline)' | \
@@ -135,10 +131,10 @@ zdisk: $(BOOTIMAGE) $(obj)/mtools.conf
 	if [ -f '$(FDINITRD)' ] ; then \
 		MTOOLSRC=$(obj)/mtools.conf mcopy '$(FDINITRD)' a:initrd.img ; \
 	fi
-	MTOOLSRC=$(obj)/mtools.conf mcopy $(BOOTIMAGE) a:linux	; sync
+	MTOOLSRC=$(obj)/mtools.conf mcopy $(obj)/bzImage a:linux	; sync
 
 # These require being root or having syslinux 2.02 or higher installed
-fdimage fdimage144: $(BOOTIMAGE) $(obj)/mtools.conf
+fdimage fdimage144: $(obj)/bzImage $(obj)/mtools.conf
 	dd if=/dev/zero of=$(obj)/fdimage bs=1024 count=1440
 	MTOOLSRC=$(obj)/mtools.conf mformat v:			; sync
 	syslinux $(obj)/fdimage					; sync
@@ -147,9 +143,9 @@ fdimage fdimage144: $(BOOTIMAGE) $(obj)/mtools.conf
 	if [ -f '$(FDINITRD)' ] ; then \
 		MTOOLSRC=$(obj)/mtools.conf mcopy '$(FDINITRD)' v:initrd.img ; \
 	fi
-	MTOOLSRC=$(obj)/mtools.conf mcopy $(BOOTIMAGE) v:linux	; sync
+	MTOOLSRC=$(obj)/mtools.conf mcopy $(obj)/bzImage v:linux	; sync
 
-fdimage288: $(BOOTIMAGE) $(obj)/mtools.conf
+fdimage288: $(obj)/bzImage $(obj)/mtools.conf
 	dd if=/dev/zero of=$(obj)/fdimage bs=1024 count=2880
 	MTOOLSRC=$(obj)/mtools.conf mformat w:			; sync
 	syslinux $(obj)/fdimage					; sync
@@ -158,9 +154,9 @@ fdimage288: $(BOOTIMAGE) $(obj)/mtools.conf
 	if [ -f '$(FDINITRD)' ] ; then \
 		MTOOLSRC=$(obj)/mtools.conf mcopy '$(FDINITRD)' w:initrd.img ; \
 	fi
-	MTOOLSRC=$(obj)/mtools.conf mcopy $(BOOTIMAGE) w:linux	; sync
+	MTOOLSRC=$(obj)/mtools.conf mcopy $(obj)/bzImage w:linux	; sync
 
-isoimage: $(BOOTIMAGE)
+isoimage: $(obj)/bzImage
 	-rm -rf $(obj)/isoimage
 	mkdir $(obj)/isoimage
 	for i in lib lib64 share end ; do \
@@ -170,7 +166,7 @@ isoimage: $(BOOTIMAGE)
 		fi ; \
 		if [ $$i = end ] ; then exit 1 ; fi ; \
 	done
-	cp $(BOOTIMAGE) $(obj)/isoimage/linux
+	cp $(obj)/bzImage $(obj)/isoimage/linux
 	echo '$(image_cmdline)' > $(obj)/isoimage/isolinux.cfg
 	if [ -f '$(FDINITRD)' ] ; then \
 		cp '$(FDINITRD)' $(obj)/isoimage/initrd.img ; \
@@ -181,12 +177,13 @@ isoimage: $(BOOTIMAGE)
 	isohybrid $(obj)/image.iso 2>/dev/null || true
 	rm -rf $(obj)/isoimage
 
-zlilo: $(BOOTIMAGE)
+bzlilo: $(obj)/bzImage
 	if [ -f $(INSTALL_PATH)/vmlinuz ]; then mv $(INSTALL_PATH)/vmlinuz $(INSTALL_PATH)/vmlinuz.old; fi
 	if [ -f $(INSTALL_PATH)/System.map ]; then mv $(INSTALL_PATH)/System.map $(INSTALL_PATH)/System.old; fi
-	cat $(BOOTIMAGE) > $(INSTALL_PATH)/vmlinuz
+	cat $(obj)/bzImage > $(INSTALL_PATH)/vmlinuz
 	cp System.map $(INSTALL_PATH)/
 	if [ -x /sbin/lilo ]; then /sbin/lilo; else /etc/lilo/install; fi
 
 install:
-	sh $(srctree)/$(src)/install.sh $(KERNELRELEASE) $(BOOTIMAGE) System.map "$(INSTALL_PATH)"
+	sh $(srctree)/$(src)/install.sh $(KERNELRELEASE) $(obj)/bzImage \
+		System.map "$(INSTALL_PATH)"

arch/x86/boot/header.S

@@ -24,12 +24,8 @@
 #include "boot.h"
 #include "offsets.h"
 
-SETUPSECTS	= 4			/* default nr of setup-sectors */
 BOOTSEG		= 0x07C0		/* original address of boot-sector */
-SYSSEG		= DEF_SYSSEG		/* system loaded at 0x10000 (65536) */
+SYSSEG		= 0x1000		/* historical load address >> 4 */
-SYSSIZE		= DEF_SYSSIZE		/* system size: # of 16-byte clicks */
-					/* to be loaded */
-ROOT_DEV	= 0			/* ROOT_DEV is now written by "build" */
 
 #ifndef SVGA_MODE
 #define SVGA_MODE ASK_VGA
@@ -97,12 +93,12 @@ bugger_off_msg:
 	.section ".header", "a"
 	.globl	hdr
 hdr:
-setup_sects:	.byte SETUPSECTS
+setup_sects:	.byte 0			/* Filled in by build.c */
 root_flags:	.word ROOT_RDONLY
-syssize:	.long SYSSIZE
+syssize:	.long 0			/* Filled in by build.c */
-ram_size:	.word RAMDISK
+ram_size:	.word 0			/* Obsolete */
 vid_mode:	.word SVGA_MODE
-root_dev:	.word ROOT_DEV
+root_dev:	.word 0			/* Filled in by build.c */
 boot_flag:	.word 0xAA55
 
 	# offset 512, entry point
@@ -123,14 +119,15 @@ _start:
 					# or else old loadlin-1.5 will fail)
 		.globl realmode_swtch
 realmode_swtch:	.word	0, 0		# default_switch, SETUPSEG
-start_sys_seg:	.word	SYSSEG
+start_sys_seg:	.word	SYSSEG		# obsolete and meaningless, but just
+					# in case something decided to "use" it
 		.word	kernel_version-512 # pointing to kernel version string
 					# above section of header is compatible
 					# with loadlin-1.5 (header v1.5). Don't
 					# change it.
 
-type_of_loader:	.byte	0		# = 0, old one (LILO, Loadlin,
+type_of_loader:	.byte	0		# 0 means ancient bootloader, newer
-					#      Bootlin, SYSLX, bootsect...)
+					# bootloaders know to change this.
 					# See Documentation/i386/boot.txt for
 					# assigned ids
 
@@ -142,11 +139,7 @@ CAN_USE_HEAP	= 0x80			# If set, the loader also has set
 					# space behind setup.S can be used for
 					# heap purposes.
 					# Only the loader knows what is free
-#ifndef __BIG_KERNEL__
-		.byte	0
-#else
 		.byte	LOADED_HIGH
-#endif
 
 setup_move_size: .word  0x8000		# size to move, when setup is not
 					# loaded at 0x90000. We will move setup
@@ -157,11 +150,7 @@ setup_move_size: .word  0x8000		# size to move, when setup is not
 
 code32_start:				# here loaders can put a different
 					# start address for 32-bit code.
-#ifndef __BIG_KERNEL__
-		.long	0x1000		#   0x1000 = default for zImage
-#else
 		.long	0x100000	# 0x100000 = default for big kernel
-#endif
 
 ramdisk_image:	.long	0		# address of loaded ramdisk image
 					# Here the loader puts the 32-bit

arch/x86/boot/pm.c

@@ -32,47 +32,6 @@ static void realmode_switch_hook(void)
 	}
 }
 
-/*
- * A zImage kernel is loaded at 0x10000 but wants to run at 0x1000.
- * A bzImage kernel is loaded and runs at 0x100000.
- */
-static void move_kernel_around(void)
-{
-	/* Note: rely on the compile-time option here rather than
-	   the LOADED_HIGH flag.  The Qemu kernel loader unconditionally
-	   sets the loadflags to zero. */
-#ifndef __BIG_KERNEL__
-	u16 dst_seg, src_seg;
-	u32 syssize;
-
-	dst_seg =  0x1000 >> 4;
-	src_seg = 0x10000 >> 4;
-	syssize = boot_params.hdr.syssize; /* Size in 16-byte paragraphs */
-
-	while (syssize) {
-		int paras  = (syssize >= 0x1000) ? 0x1000 : syssize;
-		int dwords = paras << 2;
-
-		asm volatile("pushw %%es ; "
-			     "pushw %%ds ; "
-			     "movw %1,%%es ; "
-			     "movw %2,%%ds ; "
-			     "xorw %%di,%%di ; "
-			     "xorw %%si,%%si ; "
-			     "rep;movsl ; "
-			     "popw %%ds ; "
-			     "popw %%es"
-			     : "+c" (dwords)
-			     : "r" (dst_seg), "r" (src_seg)
-			     : "esi", "edi");
-
-		syssize -= paras;
-		dst_seg += paras;
-		src_seg += paras;
-	}
-#endif
-}
-
 /*
  * Disable all interrupts at the legacy PIC.
  */
@@ -147,9 +106,6 @@ void go_to_protected_mode(void)
 	/* Hook before leaving real mode, also disables interrupts */
 	realmode_switch_hook();
 
-	/* Move the kernel/setup to their final resting places */
-	move_kernel_around();
-
 	/* Enable the A20 gate */
 	if (enable_a20()) {
 		puts("A20 gate not responding, unable to boot...\n");

arch/x86/boot/pmjump.S

@@ -47,6 +47,7 @@ GLOBAL(protected_mode_jump)
 ENDPROC(protected_mode_jump)
 
 	.code32
+	.section ".text32","ax"
 GLOBAL(in_pm32)
 	# Set up data segments for flat 32-bit mode
 	movl	%ecx, %ds

arch/x86/boot/setup.ld

@@ -17,7 +17,8 @@ SECTIONS
 	.header		: { *(.header) }
 	.inittext	: { *(.inittext) }
 	.initdata	: { *(.initdata) }
-	.text		: { *(.text*) }
+	.text		: { *(.text) }
+	.text32		: { *(.text32) }
 
 	. = ALIGN(16);
 	.rodata		: { *(.rodata*) }

arch/x86/boot/tools/build.c

@@ -130,7 +130,7 @@ static void die(const char * str, ...)
 
 static void usage(void)
 {
-	die("Usage: build [-b] setup system [rootdev] [> image]");
+	die("Usage: build setup system [rootdev] [> image]");
 }
 
 int main(int argc, char ** argv)
@@ -145,11 +145,6 @@ int main(int argc, char ** argv)
 	void *kernel;
 	u32 crc = 0xffffffffUL;
 
-	if (argc > 2 && !strcmp(argv[1], "-b"))
-	  {
-	    is_big_kernel = 1;
-	    argc--, argv++;
-	  }
 	if ((argc < 3) || (argc > 4))
 		usage();
 	if (argc > 3) {
@@ -216,8 +211,6 @@ int main(int argc, char ** argv)
 		die("Unable to mmap '%s': %m", argv[2]);
 	/* Number of 16-byte paragraphs, including space for a 4-byte CRC */
 	sys_size = (sz + 15 + 4) / 16;
-	if (!is_big_kernel && sys_size > DEF_SYSSIZE)
-		die("System is too big. Try using bzImage or modules.");
 
 	/* Patch the setup code with the appropriate size parameters */
 	buf[0x1f1] = setup_sectors-1;

arch/x86/boot/video-vga.c

@@ -129,41 +129,45 @@ u16 vga_crtc(void)
 	return (inb(0x3cc) & 1) ? 0x3d4 : 0x3b4;
 }
 
-static void vga_set_480_scanlines(int end)
+static void vga_set_480_scanlines(int lines)
 {
-	u16 crtc;
+	u16 crtc;		/* CRTC base address */
-	u8  csel;
+	u8  csel;		/* CRTC miscellaneous output register */
+	u8  ovfw;		/* CRTC overflow register */
+	int end = lines-1;
 
 	crtc = vga_crtc();
 
+	ovfw = 0x3c | ((end >> (8-1)) & 0x02) | ((end >> (9-6)) & 0x40);
+
 	out_idx(0x0c, crtc, 0x11); /* Vertical sync end, unlock CR0-7 */
 	out_idx(0x0b, crtc, 0x06); /* Vertical total */
-	out_idx(0x3e, crtc, 0x07); /* Vertical overflow */
+	out_idx(ovfw, crtc, 0x07); /* Vertical overflow */
 	out_idx(0xea, crtc, 0x10); /* Vertical sync start */
-	out_idx(end, crtc, 0x12); /* Vertical display end */
+	out_idx(end,  crtc, 0x12); /* Vertical display end */
 	out_idx(0xe7, crtc, 0x15); /* Vertical blank start */
 	out_idx(0x04, crtc, 0x16); /* Vertical blank end */
 	csel = inb(0x3cc);
 	csel &= 0x0d;
 	csel |= 0xe2;
-	outb(csel, 0x3cc);
+	outb(csel, 0x3c2);
 }
 
 static void vga_set_80x30(void)
 {
-	vga_set_480_scanlines(0xdf);
+	vga_set_480_scanlines(30*16);
 }
 
 static void vga_set_80x34(void)
 {
 	vga_set_14font();
-	vga_set_480_scanlines(0xdb);
+	vga_set_480_scanlines(34*14);
 }
 
 static void vga_set_80x60(void)
 {
 	vga_set_8font();
-	vga_set_480_scanlines(0xdf);
+	vga_set_480_scanlines(60*8);
 }
 
 static int vga_set_mode(struct mode_info *mode)

arch/x86/include/asm/apic.h

@@ -75,7 +75,7 @@ static inline void default_inquire_remote_apic(int apicid)
 #define setup_secondary_clock setup_secondary_APIC_clock
 #endif
 
-#ifdef CONFIG_X86_VSMP
+#ifdef CONFIG_X86_64
 extern int is_vsmp_box(void);
 #else
 static inline int is_vsmp_box(void)
@@ -108,6 +108,16 @@ extern void native_apic_icr_write(u32 low, u32 id);
 extern u64 native_apic_icr_read(void);
 
 #ifdef CONFIG_X86_X2APIC
+/*
+ * Make previous memory operations globally visible before
+ * sending the IPI through x2apic wrmsr. We need a serializing instruction or
+ * mfence for this.
+ */
+static inline void x2apic_wrmsr_fence(void)
+{
+	asm volatile("mfence" : : : "memory");
+}
+
 static inline void native_apic_msr_write(u32 reg, u32 v)
 {
 	if (reg == APIC_DFR || reg == APIC_ID || reg == APIC_LDR ||
@@ -184,6 +194,9 @@ static inline int x2apic_enabled(void)
 {
 	return 0;
 }
+
+#define	x2apic	0
+
 #endif
 
 extern int get_physical_broadcast(void);
@@ -379,6 +392,7 @@ static inline u32 safe_apic_wait_icr_idle(void)
 
 static inline void ack_APIC_irq(void)
 {
+#ifdef CONFIG_X86_LOCAL_APIC
 	/*
 	 * ack_APIC_irq() actually gets compiled as a single instruction
 	 * ... yummie.
@@ -386,6 +400,7 @@ static inline void ack_APIC_irq(void)
 
 	/* Docs say use 0 for future compatibility */
 	apic_write(APIC_EOI, 0);
+#endif
 }
 
 static inline unsigned default_get_apic_id(unsigned long x)
@@ -474,10 +489,19 @@ static inline int default_apic_id_registered(void)
 	return physid_isset(read_apic_id(), phys_cpu_present_map);
 }
 
+static inline int default_phys_pkg_id(int cpuid_apic, int index_msb)
+{
+	return cpuid_apic >> index_msb;
+}
+
+extern int default_apicid_to_node(int logical_apicid);
+
+#endif
+
 static inline unsigned int
 default_cpu_mask_to_apicid(const struct cpumask *cpumask)
 {
-	return cpumask_bits(cpumask)[0];
+	return cpumask_bits(cpumask)[0] & APIC_ALL_CPUS;
 }
 
 static inline unsigned int
@@ -491,15 +515,6 @@ default_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
 	return (unsigned int)(mask1 & mask2 & mask3);
 }
 
-static inline int default_phys_pkg_id(int cpuid_apic, int index_msb)
-{
-	return cpuid_apic >> index_msb;
-}
-
-extern int default_apicid_to_node(int logical_apicid);
-
-#endif
-
 static inline unsigned long default_check_apicid_used(physid_mask_t bitmap, int apicid)
 {
 	return physid_isset(apicid, bitmap);

arch/x86/include/asm/apicdef.h

@@ -53,6 +53,7 @@
 #define		APIC_ESR_SENDILL	0x00020
 #define		APIC_ESR_RECVILL	0x00040
 #define		APIC_ESR_ILLREGA	0x00080
+#define 	APIC_LVTCMCI	0x2f0
 #define	APIC_ICR	0x300
 #define		APIC_DEST_SELF		0x40000
 #define		APIC_DEST_ALLINC	0x80000

arch/x86/include/asm/boot.h

@@ -1,10 +1,6 @@
 #ifndef _ASM_X86_BOOT_H
 #define _ASM_X86_BOOT_H
 
-/* Don't touch these, unless you really know what you're doing. */
-#define DEF_SYSSEG	0x1000
-#define DEF_SYSSIZE	0x7F00
-
 /* Internal svga startup constants */
 #define NORMAL_VGA	0xffff		/* 80x25 mode */
 #define EXTENDED_VGA	0xfffe		/* 80x50 mode */

arch/x86/include/asm/cacheflush.h

@@ -90,6 +90,9 @@ int set_memory_4k(unsigned long addr, int numpages);
 int set_memory_array_uc(unsigned long *addr, int addrinarray);
 int set_memory_array_wb(unsigned long *addr, int addrinarray);
 
+int set_pages_array_uc(struct page **pages, int addrinarray);
+int set_pages_array_wb(struct page **pages, int addrinarray);
+
 /*
  * For legacy compatibility with the old APIs, a few functions
  * are provided that work on a "struct page".

arch/x86/include/asm/cpu_debug.h

@@ -0,0 +1,226 @@
+#ifndef _ASM_X86_CPU_DEBUG_H
+#define _ASM_X86_CPU_DEBUG_H
+
+/*
+ * CPU x86 architecture debug
+ *
+ * Copyright(C) 2009 Jaswinder Singh Rajput
+ */
+
+/* Register flags */
+enum cpu_debug_bit {
+/* Model Specific Registers (MSRs)					*/
+	CPU_MC_BIT,				/* Machine Check	*/
+	CPU_MONITOR_BIT,			/* Monitor		*/
+	CPU_TIME_BIT,				/* Time			*/
+	CPU_PMC_BIT,				/* Performance Monitor	*/
+	CPU_PLATFORM_BIT,			/* Platform		*/
+	CPU_APIC_BIT,				/* APIC			*/
+	CPU_POWERON_BIT,			/* Power-on		*/
+	CPU_CONTROL_BIT,			/* Control		*/
+	CPU_FEATURES_BIT,			/* Features control	*/
+	CPU_LBRANCH_BIT,			/* Last Branch		*/
+	CPU_BIOS_BIT,				/* BIOS			*/
+	CPU_FREQ_BIT,				/* Frequency		*/
+	CPU_MTTR_BIT,				/* MTRR			*/
+	CPU_PERF_BIT,				/* Performance		*/
+	CPU_CACHE_BIT,				/* Cache		*/
+	CPU_SYSENTER_BIT,			/* Sysenter		*/
+	CPU_THERM_BIT,				/* Thermal		*/
+	CPU_MISC_BIT,				/* Miscellaneous	*/
+	CPU_DEBUG_BIT,				/* Debug		*/
+	CPU_PAT_BIT,				/* PAT			*/
+	CPU_VMX_BIT,				/* VMX			*/
+	CPU_CALL_BIT,				/* System Call		*/
+	CPU_BASE_BIT,				/* BASE Address		*/
+	CPU_VER_BIT,				/* Version ID		*/
+	CPU_CONF_BIT,				/* Configuration	*/
+	CPU_SMM_BIT,				/* System mgmt mode	*/
+	CPU_SVM_BIT,				/*Secure Virtual Machine*/
+	CPU_OSVM_BIT,				/* OS-Visible Workaround*/
+/* Standard Registers							*/
+	CPU_TSS_BIT,				/* Task Stack Segment	*/
+	CPU_CR_BIT,				/* Control Registers	*/
+	CPU_DT_BIT,				/* Descriptor Table	*/
+/* End of Registers flags						*/
+	CPU_REG_ALL_BIT,			/* Select all Registers	*/
+};
+
+#define	CPU_REG_ALL		(~0)		/* Select all Registers	*/
+
+#define	CPU_MC			(1 << CPU_MC_BIT)
+#define	CPU_MONITOR		(1 << CPU_MONITOR_BIT)
+#define	CPU_TIME		(1 << CPU_TIME_BIT)
+#define	CPU_PMC			(1 << CPU_PMC_BIT)
+#define	CPU_PLATFORM		(1 << CPU_PLATFORM_BIT)
+#define	CPU_APIC		(1 << CPU_APIC_BIT)
+#define	CPU_POWERON		(1 << CPU_POWERON_BIT)
+#define	CPU_CONTROL		(1 << CPU_CONTROL_BIT)
+#define	CPU_FEATURES		(1 << CPU_FEATURES_BIT)
+#define	CPU_LBRANCH		(1 << CPU_LBRANCH_BIT)
+#define	CPU_BIOS		(1 << CPU_BIOS_BIT)
+#define	CPU_FREQ		(1 << CPU_FREQ_BIT)
+#define	CPU_MTRR		(1 << CPU_MTTR_BIT)
+#define	CPU_PERF		(1 << CPU_PERF_BIT)
+#define	CPU_CACHE		(1 << CPU_CACHE_BIT)
+#define	CPU_SYSENTER		(1 << CPU_SYSENTER_BIT)
+#define	CPU_THERM		(1 << CPU_THERM_BIT)
+#define	CPU_MISC		(1 << CPU_MISC_BIT)
+#define	CPU_DEBUG		(1 << CPU_DEBUG_BIT)
+#define	CPU_PAT			(1 << CPU_PAT_BIT)
+#define	CPU_VMX			(1 << CPU_VMX_BIT)
+#define	CPU_CALL		(1 << CPU_CALL_BIT)
+#define	CPU_BASE		(1 << CPU_BASE_BIT)
+#define	CPU_VER			(1 << CPU_VER_BIT)
+#define	CPU_CONF		(1 << CPU_CONF_BIT)
+#define	CPU_SMM			(1 << CPU_SMM_BIT)
+#define	CPU_SVM			(1 << CPU_SVM_BIT)
+#define	CPU_OSVM		(1 << CPU_OSVM_BIT)
+#define	CPU_TSS			(1 << CPU_TSS_BIT)
+#define	CPU_CR			(1 << CPU_CR_BIT)
+#define	CPU_DT			(1 << CPU_DT_BIT)
+
+/* Register file flags */
+enum cpu_file_bit {
+	CPU_INDEX_BIT,				/* index		*/
+	CPU_VALUE_BIT,				/* value		*/
+};
+
+#define	CPU_FILE_VALUE			(1 << CPU_VALUE_BIT)
+
+/*
+ * DisplayFamily_DisplayModel	Processor Families/Processor Number Series
+ * --------------------------	------------------------------------------
+ * 05_01, 05_02, 05_04		Pentium, Pentium with MMX
+ *
+ * 06_01			Pentium Pro
+ * 06_03, 06_05			Pentium II Xeon, Pentium II
+ * 06_07, 06_08, 06_0A, 06_0B	Pentium III Xeon, Pentum III
+ *
+ * 06_09, 060D			Pentium M
+ *
+ * 06_0E			Core Duo, Core Solo
+ *
+ * 06_0F			Xeon 3000, 3200, 5100, 5300, 7300 series,
+ *				Core 2 Quad, Core 2 Extreme, Core 2 Duo,
+ *				Pentium dual-core
+ * 06_17			Xeon 5200, 5400 series, Core 2 Quad Q9650
+ *
+ * 06_1C			Atom
+ *
+ * 0F_00, 0F_01, 0F_02		Xeon, Xeon MP, Pentium 4
+ * 0F_03, 0F_04			Xeon, Xeon MP, Pentium 4, Pentium D
+ *
+ * 0F_06			Xeon 7100, 5000 Series, Xeon MP,
+ *				Pentium 4, Pentium D
+ */
+
+/* Register processors bits */
+enum cpu_processor_bit {
+	CPU_NONE,
+/* Intel */
+	CPU_INTEL_PENTIUM_BIT,
+	CPU_INTEL_P6_BIT,
+	CPU_INTEL_PENTIUM_M_BIT,
+	CPU_INTEL_CORE_BIT,
+	CPU_INTEL_CORE2_BIT,
+	CPU_INTEL_ATOM_BIT,
+	CPU_INTEL_XEON_P4_BIT,
+	CPU_INTEL_XEON_MP_BIT,
+/* AMD */
+	CPU_AMD_K6_BIT,
+	CPU_AMD_K7_BIT,
+	CPU_AMD_K8_BIT,
+	CPU_AMD_0F_BIT,
+	CPU_AMD_10_BIT,
+	CPU_AMD_11_BIT,
+};
+
+#define	CPU_INTEL_PENTIUM	(1 << CPU_INTEL_PENTIUM_BIT)
+#define	CPU_INTEL_P6		(1 << CPU_INTEL_P6_BIT)
+#define	CPU_INTEL_PENTIUM_M	(1 << CPU_INTEL_PENTIUM_M_BIT)
+#define	CPU_INTEL_CORE		(1 << CPU_INTEL_CORE_BIT)
+#define	CPU_INTEL_CORE2		(1 << CPU_INTEL_CORE2_BIT)
+#define	CPU_INTEL_ATOM		(1 << CPU_INTEL_ATOM_BIT)
+#define	CPU_INTEL_XEON_P4	(1 << CPU_INTEL_XEON_P4_BIT)
+#define	CPU_INTEL_XEON_MP	(1 << CPU_INTEL_XEON_MP_BIT)
+
+#define	CPU_INTEL_PX		(CPU_INTEL_P6 | CPU_INTEL_PENTIUM_M)
+#define	CPU_INTEL_COREX		(CPU_INTEL_CORE | CPU_INTEL_CORE2)
+#define	CPU_INTEL_XEON		(CPU_INTEL_XEON_P4 | CPU_INTEL_XEON_MP)
+#define	CPU_CO_AT		(CPU_INTEL_CORE | CPU_INTEL_ATOM)
+#define	CPU_C2_AT		(CPU_INTEL_CORE2 | CPU_INTEL_ATOM)
+#define	CPU_CX_AT		(CPU_INTEL_COREX | CPU_INTEL_ATOM)
+#define	CPU_CX_XE		(CPU_INTEL_COREX | CPU_INTEL_XEON)
+#define	CPU_P6_XE		(CPU_INTEL_P6 | CPU_INTEL_XEON)
+#define	CPU_PM_CO_AT		(CPU_INTEL_PENTIUM_M | CPU_CO_AT)
+#define	CPU_C2_AT_XE		(CPU_C2_AT | CPU_INTEL_XEON)
+#define	CPU_CX_AT_XE		(CPU_CX_AT | CPU_INTEL_XEON)
+#define	CPU_P6_CX_AT		(CPU_INTEL_P6 | CPU_CX_AT)
+#define	CPU_P6_CX_XE		(CPU_P6_XE | CPU_INTEL_COREX)
+#define	CPU_P6_CX_AT_XE		(CPU_INTEL_P6 | CPU_CX_AT_XE)
+#define	CPU_PM_CX_AT_XE		(CPU_INTEL_PENTIUM_M | CPU_CX_AT_XE)
+#define	CPU_PM_CX_AT		(CPU_INTEL_PENTIUM_M | CPU_CX_AT)
+#define	CPU_PM_CX_XE		(CPU_INTEL_PENTIUM_M | CPU_CX_XE)
+#define	CPU_PX_CX_AT		(CPU_INTEL_PX | CPU_CX_AT)
+#define	CPU_PX_CX_AT_XE		(CPU_INTEL_PX | CPU_CX_AT_XE)
+
+/* Select all supported Intel CPUs */
+#define	CPU_INTEL_ALL		(CPU_INTEL_PENTIUM | CPU_PX_CX_AT_XE)
+
+#define	CPU_AMD_K6		(1 << CPU_AMD_K6_BIT)
+#define	CPU_AMD_K7		(1 << CPU_AMD_K7_BIT)
+#define	CPU_AMD_K8		(1 << CPU_AMD_K8_BIT)
+#define	CPU_AMD_0F		(1 << CPU_AMD_0F_BIT)
+#define	CPU_AMD_10		(1 << CPU_AMD_10_BIT)
+#define	CPU_AMD_11		(1 << CPU_AMD_11_BIT)
+
+#define	CPU_K10_PLUS		(CPU_AMD_10 | CPU_AMD_11)
+#define	CPU_K0F_PLUS		(CPU_AMD_0F | CPU_K10_PLUS)
+#define	CPU_K8_PLUS		(CPU_AMD_K8 | CPU_K0F_PLUS)
+#define	CPU_K7_PLUS		(CPU_AMD_K7 | CPU_K8_PLUS)
+
+/* Select all supported AMD CPUs */
+#define	CPU_AMD_ALL		(CPU_AMD_K6 | CPU_K7_PLUS)
+
+/* Select all supported CPUs */
+#define	CPU_ALL			(CPU_INTEL_ALL | CPU_AMD_ALL)
+
+#define MAX_CPU_FILES		512
+
+struct cpu_private {
+	unsigned		cpu;
+	unsigned		type;
+	unsigned		reg;
+	unsigned		file;
+};
+
+struct cpu_debug_base {
+	char			*name;		/* Register name	*/
+	unsigned		flag;		/* Register flag	*/
+	unsigned		write;		/* Register write flag	*/
+};
+
+/*
+ * Currently it looks similar to cpu_debug_base but once we add more files
+ * cpu_file_base will go in different direction
+ */
+struct cpu_file_base {
+	char			*name;		/* Register file name	*/
+	unsigned		flag;		/* Register file flag	*/
+	unsigned		write;		/* Register write flag	*/
+};
+
+struct cpu_cpuX_base {
+	struct dentry		*dentry;	/* Register dentry	*/
+	int			init;		/* Register index file	*/
+};
+
+struct cpu_debug_range {
+	unsigned		min;		/* Register range min	*/
+	unsigned		max;		/* Register range max	*/
+	unsigned		flag;		/* Supported flags	*/
+	unsigned		model;		/* Supported models	*/
+};
+
+#endif /* _ASM_X86_CPU_DEBUG_H */

arch/x86/include/asm/desc.h

@@ -91,7 +91,6 @@ static inline int desc_empty(const void *ptr)
 #define store_gdt(dtr) native_store_gdt(dtr)
 #define store_idt(dtr) native_store_idt(dtr)
 #define store_tr(tr) (tr = native_store_tr())
-#define store_ldt(ldt) asm("sldt %0":"=m" (ldt))
 
 #define load_TLS(t, cpu) native_load_tls(t, cpu)
 #define set_ldt native_set_ldt
@@ -112,6 +111,8 @@ static inline void paravirt_free_ldt(struct desc_struct *ldt, unsigned entries)
 }
 #endif	/* CONFIG_PARAVIRT */
 
+#define store_ldt(ldt) asm("sldt %0" : "=m"(ldt))
+
 static inline void native_write_idt_entry(gate_desc *idt, int entry,
 					  const gate_desc *gate)
 {

arch/x86/include/asm/dmi.h

@@ -1,22 +1,15 @@
 #ifndef _ASM_X86_DMI_H
 #define _ASM_X86_DMI_H
 
+#include <linux/compiler.h>
+#include <linux/init.h>
+
 #include <asm/io.h>
+#include <asm/setup.h>
 
-#define DMI_MAX_DATA 2048
+static __always_inline __init void *dmi_alloc(unsigned len)
-
-extern int dmi_alloc_index;
-extern char dmi_alloc_data[DMI_MAX_DATA];
-
-/* This is so early that there is no good way to allocate dynamic memory.
-   Allocate data in an BSS array. */
-static inline void *dmi_alloc(unsigned len)
 {
-	int idx = dmi_alloc_index;
+	return extend_brk(len, sizeof(int));
-	if ((dmi_alloc_index + len) > DMI_MAX_DATA)
-		return NULL;
-	dmi_alloc_index += len;
-	return dmi_alloc_data + idx;
 }
 
 /* Use early IO mappings for DMI because it's initialized early */

arch/x86/include/asm/e820.h

@@ -72,7 +72,7 @@ extern int e820_all_mapped(u64 start, u64 end, unsigned type);
 extern void e820_add_region(u64 start, u64 size, int type);
 extern void e820_print_map(char *who);
 extern int
-sanitize_e820_map(struct e820entry *biosmap, int max_nr_map, int *pnr_map);
+sanitize_e820_map(struct e820entry *biosmap, int max_nr_map, u32 *pnr_map);
 extern u64 e820_update_range(u64 start, u64 size, unsigned old_type,
 			       unsigned new_type);
 extern u64 e820_remove_range(u64 start, u64 size, unsigned old_type,

arch/x86/include/asm/entry_arch.h

@@ -33,6 +33,8 @@ BUILD_INTERRUPT3(invalidate_interrupt7,INVALIDATE_TLB_VECTOR_START+7,
 		 smp_invalidate_interrupt)
 #endif
 
+BUILD_INTERRUPT(generic_interrupt, GENERIC_INTERRUPT_VECTOR)
+
 /*
  * every pentium local APIC has two 'local interrupts', with a
  * soft-definable vector attached to both interrupts, one of

arch/x86/include/asm/hardirq.h

@@ -12,6 +12,7 @@ typedef struct {
 	unsigned int apic_timer_irqs;	/* arch dependent */
 	unsigned int irq_spurious_count;
 #endif
+	unsigned int generic_irqs;	/* arch dependent */
 #ifdef CONFIG_SMP
 	unsigned int irq_resched_count;
 	unsigned int irq_call_count;

arch/x86/include/asm/highmem.h

@@ -63,6 +63,7 @@ void *kmap_atomic_prot(struct page *page, enum km_type type, pgprot_t prot);
 void *kmap_atomic(struct page *page, enum km_type type);
 void kunmap_atomic(void *kvaddr, enum km_type type);
 void *kmap_atomic_pfn(unsigned long pfn, enum km_type type);
+void *kmap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot);
 struct page *kmap_atomic_to_page(void *ptr);
 
 #ifndef CONFIG_PARAVIRT

arch/x86/include/asm/hw_irq.h

@@ -27,6 +27,7 @@
 
 /* Interrupt handlers registered during init_IRQ */
 extern void apic_timer_interrupt(void);
+extern void generic_interrupt(void);
 extern void error_interrupt(void);
 extern void spurious_interrupt(void);
 extern void thermal_interrupt(void);

arch/x86/include/asm/init.h

@@ -0,0 +1,18 @@
+#ifndef _ASM_X86_INIT_32_H
+#define _ASM_X86_INIT_32_H
+
+#ifdef CONFIG_X86_32
+extern void __init early_ioremap_page_table_range_init(void);
+#endif
+
+extern unsigned long __init
+kernel_physical_mapping_init(unsigned long start,
+			     unsigned long end,
+			     unsigned long page_size_mask);
+
+
+extern unsigned long __initdata e820_table_start;
+extern unsigned long __meminitdata e820_table_end;
+extern unsigned long __meminitdata e820_table_top;
+
+#endif /* _ASM_X86_INIT_32_H */

arch/x86/include/asm/io_apic.h

@@ -162,7 +162,8 @@ extern int (*ioapic_renumber_irq)(int ioapic, int irq);
 extern void ioapic_init_mappings(void);
 
 #ifdef CONFIG_X86_64
-extern int save_mask_IO_APIC_setup(void);
+extern int save_IO_APIC_setup(void);
+extern void mask_IO_APIC_setup(void);
 extern void restore_IO_APIC_setup(void);
 extern void reinit_intr_remapped_IO_APIC(int);
 #endif
@@ -172,7 +173,7 @@ extern void probe_nr_irqs_gsi(void);
 extern int setup_ioapic_entry(int apic, int irq,
 			      struct IO_APIC_route_entry *entry,
 			      unsigned int destination, int trigger,
-			      int polarity, int vector);
+			      int polarity, int vector, int pin);
 extern void ioapic_write_entry(int apic, int pin,
 			       struct IO_APIC_route_entry e);
 #else  /* !CONFIG_X86_IO_APIC */

arch/x86/include/asm/irq.h

@@ -36,6 +36,7 @@ static inline int irq_canonicalize(int irq)
 extern void fixup_irqs(void);
 #endif
 
+extern void (*generic_interrupt_extension)(void);
 extern void init_IRQ(void);
 extern void native_init_IRQ(void);
 extern bool handle_irq(unsigned irq, struct pt_regs *regs);

arch/x86/include/asm/irq_remapping.h

@@ -1,8 +1,6 @@
 #ifndef _ASM_X86_IRQ_REMAPPING_H
 #define _ASM_X86_IRQ_REMAPPING_H
 
-extern int x2apic;
-
 #define IRTE_DEST(dest) ((x2apic) ? dest : dest << 8)
 
 #endif	/* _ASM_X86_IRQ_REMAPPING_H */

arch/x86/include/asm/irq_vectors.h

@@ -111,6 +111,11 @@
  */
 #define LOCAL_PERF_VECTOR		0xee
 
+/*
+ * Generic system vector for platform specific use
+ */
+#define GENERIC_INTERRUPT_VECTOR	0xed
+
 /*
  * First APIC vector available to drivers: (vectors 0x30-0xee) we
  * start at 0x31(0x41) to spread out vectors evenly between priority

arch/x86/include/asm/kexec.h

@@ -9,13 +9,13 @@
 # define PAGES_NR		4
 #else
 # define PA_CONTROL_PAGE	0
-# define PA_TABLE_PAGE		1
+# define VA_CONTROL_PAGE	1
-# define PAGES_NR		2
+# define PA_TABLE_PAGE		2
+# define PA_SWAP_PAGE		3
+# define PAGES_NR		4
 #endif
 
-#ifdef CONFIG_X86_32
 # define KEXEC_CONTROL_CODE_MAX_SIZE	2048
-#endif
 
 #ifndef __ASSEMBLY__
 
@@ -136,10 +136,11 @@ relocate_kernel(unsigned long indirection_page,
 		unsigned int has_pae,
 		unsigned int preserve_context);
 #else
-NORET_TYPE void
+unsigned long
 relocate_kernel(unsigned long indirection_page,
 		unsigned long page_list,
-		unsigned long start_address) ATTRIB_NORET;
+		unsigned long start_address,
+		unsigned int preserve_context);
 #endif
 
 #define ARCH_HAS_KIMAGE_ARCH

arch/x86/include/asm/linkage.h

@@ -1,14 +1,11 @@
 #ifndef _ASM_X86_LINKAGE_H
 #define _ASM_X86_LINKAGE_H
 
+#include <linux/stringify.h>
+
 #undef notrace
 #define notrace __attribute__((no_instrument_function))
 
-#ifdef CONFIG_X86_64
-#define __ALIGN .p2align 4,,15
-#define __ALIGN_STR ".p2align 4,,15"
-#endif
-
 #ifdef CONFIG_X86_32
 #define asmlinkage CPP_ASMLINKAGE __attribute__((regparm(0)))
 /*
@@ -50,16 +47,20 @@
 	__asmlinkage_protect_n(ret, "g" (arg1), "g" (arg2), "g" (arg3), \
 			      "g" (arg4), "g" (arg5), "g" (arg6))
 
-#endif
+#endif /* CONFIG_X86_32 */
+
+#ifdef __ASSEMBLY__
 
 #define GLOBAL(name)	\
 	.globl name;	\
 	name:
 
-#ifdef CONFIG_X86_ALIGNMENT_16
+#if defined(CONFIG_X86_64) || defined(CONFIG_X86_ALIGNMENT_16)
-#define __ALIGN .align 16,0x90
+#define __ALIGN		.p2align 4, 0x90
-#define __ALIGN_STR ".align 16,0x90"
+#define __ALIGN_STR	__stringify(__ALIGN)
 #endif
 
+#endif /* __ASSEMBLY__ */
+
 #endif /* _ASM_X86_LINKAGE_H */
 

arch/x86/include/asm/mce.h

@@ -11,6 +11,8 @@
  */
 
 #define MCG_CTL_P	 (1UL<<8)   /* MCG_CAP register available */
+#define MCG_EXT_P	 (1ULL<<9)   /* Extended registers available */
+#define MCG_CMCI_P	 (1ULL<<10)  /* CMCI supported */
 
 #define MCG_STATUS_RIPV  (1UL<<0)   /* restart ip valid */
 #define MCG_STATUS_EIPV  (1UL<<1)   /* ip points to correct instruction */
@@ -90,14 +92,29 @@ extern int mce_disabled;
 
 #include <asm/atomic.h>
 
+void mce_setup(struct mce *m);
 void mce_log(struct mce *m);
 DECLARE_PER_CPU(struct sys_device, device_mce);
 extern void (*threshold_cpu_callback)(unsigned long action, unsigned int cpu);
 
+/*
+ * To support more than 128 would need to escape the predefined
+ * Linux defined extended banks first.
+ */
+#define MAX_NR_BANKS (MCE_EXTENDED_BANK - 1)
+
 #ifdef CONFIG_X86_MCE_INTEL
 void mce_intel_feature_init(struct cpuinfo_x86 *c);
+void cmci_clear(void);
+void cmci_reenable(void);
+void cmci_rediscover(int dying);
+void cmci_recheck(void);
 #else
 static inline void mce_intel_feature_init(struct cpuinfo_x86 *c) { }
+static inline void cmci_clear(void) {}
+static inline void cmci_reenable(void) {}
+static inline void cmci_rediscover(int dying) {}
+static inline void cmci_recheck(void) {}
 #endif
 
 #ifdef CONFIG_X86_MCE_AMD
@@ -106,11 +123,23 @@ void mce_amd_feature_init(struct cpuinfo_x86 *c);
 static inline void mce_amd_feature_init(struct cpuinfo_x86 *c) { }
 #endif
 
-void mce_log_therm_throt_event(unsigned int cpu, __u64 status);
+extern int mce_available(struct cpuinfo_x86 *c);
+
+void mce_log_therm_throt_event(__u64 status);
 
 extern atomic_t mce_entry;
 
 extern void do_machine_check(struct pt_regs *, long);
+
+typedef DECLARE_BITMAP(mce_banks_t, MAX_NR_BANKS);
+DECLARE_PER_CPU(mce_banks_t, mce_poll_banks);
+
+enum mcp_flags {
+	MCP_TIMESTAMP = (1 << 0),	/* log time stamp */
+	MCP_UC = (1 << 1),		/* log uncorrected errors */
+};
+extern void machine_check_poll(enum mcp_flags flags, mce_banks_t *b);
+
 extern int mce_notify_user(void);
 
 #endif /* !CONFIG_X86_32 */
@@ -120,8 +149,8 @@ extern void mcheck_init(struct cpuinfo_x86 *c);
 #else
 #define mcheck_init(c) do { } while (0)
 #endif
-extern void stop_mce(void);
+
-extern void restart_mce(void);
+extern void (*mce_threshold_vector)(void);
 
 #endif /* __KERNEL__ */
 #endif /* _ASM_X86_MCE_H */

arch/x86/include/asm/msidef.h

@@ -47,6 +47,7 @@
 #define	 MSI_ADDR_DEST_ID_MASK		0x00ffff0
 #define  MSI_ADDR_DEST_ID(dest)		(((dest) << MSI_ADDR_DEST_ID_SHIFT) & \
 					 MSI_ADDR_DEST_ID_MASK)
+#define MSI_ADDR_EXT_DEST_ID(dest)	((dest) & 0xffffff00)
 
 #define MSI_ADDR_IR_EXT_INT		(1 << 4)
 #define MSI_ADDR_IR_SHV			(1 << 3)

arch/x86/include/asm/msr-index.h

@@ -81,6 +81,11 @@
 #define MSR_IA32_MC0_ADDR		0x00000402
 #define MSR_IA32_MC0_MISC		0x00000403
 
+/* These are consecutive and not in the normal 4er MCE bank block */
+#define MSR_IA32_MC0_CTL2		0x00000280
+#define CMCI_EN			(1ULL << 30)
+#define CMCI_THRESHOLD_MASK		0xffffULL
+
 #define MSR_P6_PERFCTR0			0x000000c1
 #define MSR_P6_PERFCTR1			0x000000c2
 #define MSR_P6_EVNTSEL0			0x00000186

arch/x86/include/asm/page_32_types.h

@@ -39,6 +39,11 @@
 #define __VIRTUAL_MASK_SHIFT	32
 #endif	/* CONFIG_X86_PAE */
 
+/*
+ * Kernel image size is limited to 512 MB (see in arch/x86/kernel/head_32.S)
+ */
+#define KERNEL_IMAGE_SIZE	(512 * 1024 * 1024)
+
 #ifndef __ASSEMBLY__
 
 /*

arch/x86/include/asm/page_types.h

@@ -40,14 +40,8 @@
 
 #ifndef __ASSEMBLY__
 
-struct pgprot;
-
 extern int page_is_ram(unsigned long pagenr);
 extern int devmem_is_allowed(unsigned long pagenr);
-extern void map_devmem(unsigned long pfn, unsigned long size,
-		       struct pgprot vma_prot);
-extern void unmap_devmem(unsigned long pfn, unsigned long size,
-			 struct pgprot vma_prot);
 
 extern unsigned long max_low_pfn_mapped;
 extern unsigned long max_pfn_mapped;

arch/x86/include/asm/paravirt.h

@@ -317,8 +317,6 @@ struct pv_mmu_ops {
 #if PAGETABLE_LEVELS >= 3
 #ifdef CONFIG_X86_PAE
 	void (*set_pte_atomic)(pte_t *ptep, pte_t pteval);
-	void (*set_pte_present)(struct mm_struct *mm, unsigned long addr,
-				pte_t *ptep, pte_t pte);
 	void (*pte_clear)(struct mm_struct *mm, unsigned long addr,
 			  pte_t *ptep);
 	void (*pmd_clear)(pmd_t *pmdp);
@@ -389,7 +387,7 @@ extern struct pv_lock_ops pv_lock_ops;
 
 #define paravirt_type(op)				\
 	[paravirt_typenum] "i" (PARAVIRT_PATCH(op)),	\
-	[paravirt_opptr] "m" (op)
+	[paravirt_opptr] "i" (&(op))
 #define paravirt_clobber(clobber)		\
 	[paravirt_clobber] "i" (clobber)
 
@@ -443,7 +441,7 @@ int paravirt_disable_iospace(void);
  * offset into the paravirt_patch_template structure, and can therefore be
  * freely converted back into a structure offset.
  */
-#define PARAVIRT_CALL	"call *%[paravirt_opptr];"
+#define PARAVIRT_CALL	"call *%c[paravirt_opptr];"
 
 /*
  * These macros are intended to wrap calls through one of the paravirt
@@ -1365,13 +1363,6 @@ static inline void set_pte_atomic(pte_t *ptep, pte_t pte)
 		    pte.pte, pte.pte >> 32);
 }
 
-static inline void set_pte_present(struct mm_struct *mm, unsigned long addr,
-				   pte_t *ptep, pte_t pte)
-{
-	/* 5 arg words */
-	pv_mmu_ops.set_pte_present(mm, addr, ptep, pte);
-}
-
 static inline void pte_clear(struct mm_struct *mm, unsigned long addr,
 			     pte_t *ptep)
 {
@@ -1388,12 +1379,6 @@ static inline void set_pte_atomic(pte_t *ptep, pte_t pte)
 	set_pte(ptep, pte);
 }
 
-static inline void set_pte_present(struct mm_struct *mm, unsigned long addr,
-				   pte_t *ptep, pte_t pte)
-{
-	set_pte(ptep, pte);
-}
-
 static inline void pte_clear(struct mm_struct *mm, unsigned long addr,
 			     pte_t *ptep)
 {

arch/x86/include/asm/pat.h

@@ -2,6 +2,7 @@
 #define _ASM_X86_PAT_H
 
 #include <linux/types.h>
+#include <asm/pgtable_types.h>
 
 #ifdef CONFIG_X86_PAT
 extern int pat_enabled;
@@ -17,5 +18,9 @@ extern int free_memtype(u64 start, u64 end);
 
 extern int kernel_map_sync_memtype(u64 base, unsigned long size,
 		unsigned long flag);
+extern void map_devmem(unsigned long pfn, unsigned long size,
+		       struct pgprot vma_prot);
+extern void unmap_devmem(unsigned long pfn, unsigned long size,
+			 struct pgprot vma_prot);
 
 #endif /* _ASM_X86_PAT_H */

arch/x86/include/asm/pgtable-2level.h

@@ -26,13 +26,6 @@ static inline void native_set_pte_atomic(pte_t *ptep, pte_t pte)
 	native_set_pte(ptep, pte);
 }
 
-static inline void native_set_pte_present(struct mm_struct *mm,
-					  unsigned long addr,
-					  pte_t *ptep, pte_t pte)
-{
-	native_set_pte(ptep, pte);
-}
-
 static inline void native_pmd_clear(pmd_t *pmdp)
 {
 	native_set_pmd(pmdp, __pmd(0));

arch/x86/include/asm/pgtable-3level.h

@@ -31,23 +31,6 @@ static inline void native_set_pte(pte_t *ptep, pte_t pte)
 	ptep->pte_low = pte.pte_low;
 }
 
-/*
- * Since this is only called on user PTEs, and the page fault handler
- * must handle the already racy situation of simultaneous page faults,
- * we are justified in merely clearing the PTE present bit, followed
- * by a set.  The ordering here is important.
- */
-static inline void native_set_pte_present(struct mm_struct *mm,
-					  unsigned long addr,
-					  pte_t *ptep, pte_t pte)
-{
-	ptep->pte_low = 0;
-	smp_wmb();
-	ptep->pte_high = pte.pte_high;
-	smp_wmb();
-	ptep->pte_low = pte.pte_low;
-}
-
 static inline void native_set_pte_atomic(pte_t *ptep, pte_t pte)
 {
 	set_64bit((unsigned long long *)(ptep), native_pte_val(pte));

arch/x86/include/asm/pgtable.h

@@ -31,8 +31,6 @@ extern struct list_head pgd_list;
 #define set_pte(ptep, pte)		native_set_pte(ptep, pte)
 #define set_pte_at(mm, addr, ptep, pte)	native_set_pte_at(mm, addr, ptep, pte)
 
-#define set_pte_present(mm, addr, ptep, pte)				\
-	native_set_pte_present(mm, addr, ptep, pte)
 #define set_pte_atomic(ptep, pte)					\
 	native_set_pte_atomic(ptep, pte)
 

arch/x86/include/asm/pgtable_32.h

@@ -42,9 +42,6 @@ extern void set_pmd_pfn(unsigned long, unsigned long, pgprot_t);
  */
 #undef TEST_ACCESS_OK
 
-/* The boot page tables (all created as a single array) */
-extern unsigned long pg0[];
-
 #ifdef CONFIG_X86_PAE
 # include <asm/pgtable-3level.h>
 #else

arch/x86/include/asm/pgtable_32_types.h

@@ -25,6 +25,11 @@
  * area for the same reason. ;)
  */
 #define VMALLOC_OFFSET	(8 * 1024 * 1024)
+
+#ifndef __ASSEMBLER__
+extern bool __vmalloc_start_set; /* set once high_memory is set */
+#endif
+
 #define VMALLOC_START	((unsigned long)high_memory + VMALLOC_OFFSET)
 #ifdef CONFIG_X86_PAE
 #define LAST_PKMAP 512

arch/x86/include/asm/pgtable_types.h

@@ -273,6 +273,7 @@ typedef struct page *pgtable_t;
 
 extern pteval_t __supported_pte_mask;
 extern int nx_enabled;
+extern void set_nx(void);
 
 #define pgprot_writecombine	pgprot_writecombine
 extern pgprot_t pgprot_writecombine(pgprot_t prot);

arch/x86/include/asm/processor.h

@@ -75,9 +75,9 @@ struct cpuinfo_x86 {
 #else
 	/* Number of 4K pages in DTLB/ITLB combined(in pages): */
 	int			x86_tlbsize;
+#endif
 	__u8			x86_virt_bits;
 	__u8			x86_phys_bits;
-#endif
 	/* CPUID returned core id bits: */
 	__u8			x86_coreid_bits;
 	/* Max extended CPUID function supported: */
@@ -391,6 +391,9 @@ DECLARE_PER_CPU(union irq_stack_union, irq_stack_union);
 DECLARE_INIT_PER_CPU(irq_stack_union);
 
 DECLARE_PER_CPU(char *, irq_stack_ptr);
+DECLARE_PER_CPU(unsigned int, irq_count);
+extern unsigned long kernel_eflags;
+extern asmlinkage void ignore_sysret(void);
 #else	/* X86_64 */
 #ifdef CONFIG_CC_STACKPROTECTOR
 DECLARE_PER_CPU(unsigned long, stack_canary);

arch/x86/include/asm/sections.h

@@ -1 +1,8 @@
+#ifndef _ASM_X86_SECTIONS_H
+#define _ASM_X86_SECTIONS_H
+
 #include <asm-generic/sections.h>
+
+extern char __brk_base[], __brk_limit[];
+
+#endif	/* _ASM_X86_SECTIONS_H */

arch/x86/include/asm/setup.h

@@ -64,7 +64,7 @@ extern void x86_quirk_time_init(void);
 #include <asm/bootparam.h>
 
 /* Interrupt control for vSMPowered x86_64 systems */
-#ifdef CONFIG_X86_VSMP
+#ifdef CONFIG_X86_64
 void vsmp_init(void);
 #else
 static inline void vsmp_init(void) { }
@@ -100,20 +100,51 @@ extern struct boot_params boot_params;
  */
 #define LOWMEMSIZE()	(0x9f000)
 
+/* exceedingly early brk-like allocator */
+extern unsigned long _brk_end;
+void *extend_brk(size_t size, size_t align);
+
+/*
+ * Reserve space in the brk section.  The name must be unique within
+ * the file, and somewhat descriptive.  The size is in bytes.  Must be
+ * used at file scope.
+ *
+ * (This uses a temp function to wrap the asm so we can pass it the
+ * size parameter; otherwise we wouldn't be able to.  We can't use a
+ * "section" attribute on a normal variable because it always ends up
+ * being @progbits, which ends up allocating space in the vmlinux
+ * executable.)
+ */
+#define RESERVE_BRK(name,sz)						\
+	static void __section(.discard) __used				\
+	__brk_reservation_fn_##name##__(void) {				\
+		asm volatile (						\
+			".pushsection .brk_reservation,\"aw\",@nobits;" \
+			".brk." #name ":"				\
+			" 1:.skip %c0;"					\
+			" .size .brk." #name ", . - 1b;"		\
+			" .popsection"					\
+			: : "i" (sz));					\
+	}
+
 #ifdef __i386__
 
 void __init i386_start_kernel(void);
 extern void probe_roms(void);
 
-extern unsigned long init_pg_tables_start;
-extern unsigned long init_pg_tables_end;
-
 #else
 void __init x86_64_start_kernel(char *real_mode);
 void __init x86_64_start_reservations(char *real_mode_data);
 
 #endif /* __i386__ */
 #endif /* _SETUP */
+#else
+#define RESERVE_BRK(name,sz)				\
+	.pushsection .brk_reservation,"aw",@nobits;	\
+.brk.name:						\
+1:	.skip sz;					\
+	.size .brk.name,.-1b;				\
+	.popsection
 #endif /* __ASSEMBLY__ */
 #endif  /*  __KERNEL__  */
 

arch/x86/include/asm/uv/uv_hub.h

@@ -199,6 +199,10 @@ DECLARE_PER_CPU(struct uv_hub_info_s, __uv_hub_info);
 #define SCIR_CPU_ACTIVITY	0x02	/* not idle */
 #define SCIR_CPU_HB_INTERVAL	(HZ)	/* once per second */
 
+/* Loop through all installed blades */
+#define for_each_possible_blade(bid)		\
+	for ((bid) = 0; (bid) < uv_num_possible_blades(); (bid)++)
+
 /*
  * Macros for converting between kernel virtual addresses, socket local physical
  * addresses, and UV global physical addresses.

arch/x86/include/asm/xen/hypercall.h

@@ -296,6 +296,8 @@ HYPERVISOR_get_debugreg(int reg)
 static inline int
 HYPERVISOR_update_descriptor(u64 ma, u64 desc)
 {
+	if (sizeof(u64) == sizeof(long))
+		return _hypercall2(int, update_descriptor, ma, desc);
 	return _hypercall4(int, update_descriptor, ma, ma>>32, desc, desc>>32);
 }
 

arch/x86/kernel/Makefile

@@ -70,7 +70,6 @@ obj-$(CONFIG_FUNCTION_GRAPH_TRACER)	+= ftrace.o
 obj-$(CONFIG_KEXEC)		+= machine_kexec_$(BITS).o
 obj-$(CONFIG_KEXEC)		+= relocate_kernel_$(BITS).o crash.o
 obj-$(CONFIG_CRASH_DUMP)	+= crash_dump_$(BITS).o
-obj-$(CONFIG_X86_VSMP)		+= vsmp_64.o
 obj-$(CONFIG_KPROBES)		+= kprobes.o
 obj-$(CONFIG_MODULES)		+= module_$(BITS).o
 obj-$(CONFIG_EFI) 		+= efi.o efi_$(BITS).o efi_stub_$(BITS).o
@@ -111,7 +110,7 @@ obj-$(CONFIG_SWIOTLB)			+= pci-swiotlb_64.o # NB rename without _64
 ###
 # 64 bit specific files
 ifeq ($(CONFIG_X86_64),y)
-	obj-$(CONFIG_X86_UV)		+= tlb_uv.o bios_uv.o uv_irq.o uv_sysfs.o
+	obj-$(CONFIG_X86_UV)		+= tlb_uv.o bios_uv.o uv_irq.o uv_sysfs.o uv_time.o
 	obj-$(CONFIG_X86_PM_TIMER)	+= pmtimer_64.o
 	obj-$(CONFIG_AUDIT)		+= audit_64.o
 
@@ -120,4 +119,5 @@ ifeq ($(CONFIG_X86_64),y)
 	obj-$(CONFIG_AMD_IOMMU)		+= amd_iommu_init.o amd_iommu.o
 
 	obj-$(CONFIG_PCI_MMCONFIG)	+= mmconf-fam10h_64.o
+	obj-y				+= vsmp_64.o
 endif

arch/x86/kernel/alternative.c

@@ -414,9 +414,17 @@ void __init alternative_instructions(void)
 	   that might execute the to be patched code.
 	   Other CPUs are not running. */
 	stop_nmi();
-#ifdef CONFIG_X86_MCE
+
-	stop_mce();
+	/*
-#endif
+	 * Don't stop machine check exceptions while patching.
+	 * MCEs only happen when something got corrupted and in this
+	 * case we must do something about the corruption.
+	 * Ignoring it is worse than a unlikely patching race.
+	 * Also machine checks tend to be broadcast and if one CPU
+	 * goes into machine check the others follow quickly, so we don't
+	 * expect a machine check to cause undue problems during to code
+	 * patching.
+	 */
 
 	apply_alternatives(__alt_instructions, __alt_instructions_end);
 
@@ -456,9 +464,6 @@ void __init alternative_instructions(void)
 				(unsigned long)__smp_locks_end);
 
 	restart_nmi();
-#ifdef CONFIG_X86_MCE
-	restart_mce();
-#endif
 }
 
 /**

arch/x86/kernel/apic/apic.c

@@ -46,6 +46,7 @@
 #include <asm/idle.h>
 #include <asm/mtrr.h>
 #include <asm/smp.h>
+#include <asm/mce.h>
 
 unsigned int num_processors;
 
@@ -808,7 +809,7 @@ void clear_local_APIC(void)
 	u32 v;
 
 	/* APIC hasn't been mapped yet */
-	if (!apic_phys)
+	if (!x2apic && !apic_phys)
 		return;
 
 	maxlvt = lapic_get_maxlvt();
@@ -842,6 +843,14 @@ void clear_local_APIC(void)
 		apic_write(APIC_LVTTHMR, v | APIC_LVT_MASKED);
 	}
 #endif
+#ifdef CONFIG_X86_MCE_INTEL
+	if (maxlvt >= 6) {
+		v = apic_read(APIC_LVTCMCI);
+		if (!(v & APIC_LVT_MASKED))
+			apic_write(APIC_LVTCMCI, v | APIC_LVT_MASKED);
+	}
+#endif
+
 	/*
 	 * Clean APIC state for other OSs:
 	 */
@@ -1241,6 +1250,12 @@ void __cpuinit setup_local_APIC(void)
 	apic_write(APIC_LVT1, value);
 
 	preempt_enable();
+
+#ifdef CONFIG_X86_MCE_INTEL
+	/* Recheck CMCI information after local APIC is up on CPU #0 */
+	if (smp_processor_id() == 0)
+		cmci_recheck();
+#endif
 }
 
 void __cpuinit end_local_APIC_setup(void)
@@ -1319,15 +1334,16 @@ void __init enable_IR_x2apic(void)
 		return;
 	}
 
-	local_irq_save(flags);
+	ret = save_IO_APIC_setup();
-	mask_8259A();
-
-	ret = save_mask_IO_APIC_setup();
 	if (ret) {
 		pr_info("Saving IO-APIC state failed: %d\n", ret);
 		goto end;
 	}
 
+	local_irq_save(flags);
+	mask_IO_APIC_setup();
+	mask_8259A();
+
 	ret = enable_intr_remapping(1);
 
 	if (ret && x2apic_preenabled) {
@@ -1352,10 +1368,10 @@ void __init enable_IR_x2apic(void)
 	else
 		reinit_intr_remapped_IO_APIC(x2apic_preenabled);
 
-end:
 	unmask_8259A();
 	local_irq_restore(flags);
 
+end:
 	if (!ret) {
 		if (!x2apic_preenabled)
 			pr_info("Enabled x2apic and interrupt-remapping\n");
@@ -1508,12 +1524,10 @@ void __init early_init_lapic_mapping(void)
  */
 void __init init_apic_mappings(void)
 {
-#ifdef CONFIG_X86_X2APIC
 	if (x2apic) {
 		boot_cpu_physical_apicid = read_apic_id();
 		return;
 	}
-#endif
 
 	/*
 	 * If no local APIC can be found then set up a fake all
@@ -1957,12 +1971,9 @@ static int lapic_resume(struct sys_device *dev)
 
 	local_irq_save(flags);
 
-#ifdef CONFIG_X86_X2APIC
 	if (x2apic)
 		enable_x2apic();
-	else
+	else {
-#endif
-	{
 		/*
 		 * Make sure the APICBASE points to the right address
 		 *

arch/x86/kernel/apic/apic_flat_64.c

@@ -159,20 +159,6 @@ static int flat_apic_id_registered(void)
 	return physid_isset(read_xapic_id(), phys_cpu_present_map);
 }
 
-static unsigned int flat_cpu_mask_to_apicid(const struct cpumask *cpumask)
-{
-	return cpumask_bits(cpumask)[0] & APIC_ALL_CPUS;
-}
-
-static unsigned int flat_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
-						const struct cpumask *andmask)
-{
-	unsigned long mask1 = cpumask_bits(cpumask)[0] & APIC_ALL_CPUS;
-	unsigned long mask2 = cpumask_bits(andmask)[0] & APIC_ALL_CPUS;
-
-	return mask1 & mask2;
-}
-
 static int flat_phys_pkg_id(int initial_apic_id, int index_msb)
 {
 	return hard_smp_processor_id() >> index_msb;
@@ -213,8 +199,8 @@ struct apic apic_flat =  {
 	.set_apic_id			= set_apic_id,
 	.apic_id_mask			= 0xFFu << 24,
 
-	.cpu_mask_to_apicid		= flat_cpu_mask_to_apicid,
+	.cpu_mask_to_apicid		= default_cpu_mask_to_apicid,
-	.cpu_mask_to_apicid_and		= flat_cpu_mask_to_apicid_and,
+	.cpu_mask_to_apicid_and		= default_cpu_mask_to_apicid_and,
 
 	.send_IPI_mask			= flat_send_IPI_mask,
 	.send_IPI_mask_allbutself	= flat_send_IPI_mask_allbutself,

arch/x86/kernel/apic/io_apic.c

@@ -389,6 +389,8 @@ struct io_apic {
 	unsigned int index;
 	unsigned int unused[3];
 	unsigned int data;
+	unsigned int unused2[11];
+	unsigned int eoi;
 };
 
 static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx)
@@ -397,6 +399,12 @@ static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx)
 		+ (mp_ioapics[idx].apicaddr & ~PAGE_MASK);
 }
 
+static inline void io_apic_eoi(unsigned int apic, unsigned int vector)
+{
+	struct io_apic __iomem *io_apic = io_apic_base(apic);
+	writel(vector, &io_apic->eoi);
+}
+
 static inline unsigned int io_apic_read(unsigned int apic, unsigned int reg)
 {
 	struct io_apic __iomem *io_apic = io_apic_base(apic);
@@ -546,16 +554,12 @@ static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq
 
 		apic = entry->apic;
 		pin = entry->pin;
-#ifdef CONFIG_INTR_REMAP
 		/*
 		 * With interrupt-remapping, destination information comes
 		 * from interrupt-remapping table entry.
 		 */
 		if (!irq_remapped(irq))
 			io_apic_write(apic, 0x11 + pin*2, dest);
-#else
-		io_apic_write(apic, 0x11 + pin*2, dest);
-#endif
 		reg = io_apic_read(apic, 0x10 + pin*2);
 		reg &= ~IO_APIC_REDIR_VECTOR_MASK;
 		reg |= vector;
@@ -588,10 +592,12 @@ set_desc_affinity(struct irq_desc *desc, const struct cpumask *mask)
 	if (assign_irq_vector(irq, cfg, mask))
 		return BAD_APICID;
 
-	cpumask_and(desc->affinity, cfg->domain, mask);
+	/* check that before desc->addinity get updated */
 	set_extra_move_desc(desc, mask);
 
-	return apic->cpu_mask_to_apicid_and(desc->affinity, cpu_online_mask);
+	cpumask_copy(desc->affinity, mask);
+
+	return apic->cpu_mask_to_apicid_and(desc->affinity, cfg->domain);
 }
 
 static void
@@ -849,9 +855,9 @@ __setup("pirq=", ioapic_pirq_setup);
 static struct IO_APIC_route_entry *early_ioapic_entries[MAX_IO_APICS];
 
 /*
- * Saves and masks all the unmasked IO-APIC RTE's
+ * Saves all the IO-APIC RTE's
  */
-int save_mask_IO_APIC_setup(void)
+int save_IO_APIC_setup(void)
 {
 	union IO_APIC_reg_01 reg_01;
 	unsigned long flags;
@@ -876,16 +882,9 @@ int save_mask_IO_APIC_setup(void)
 	}
 
 	for (apic = 0; apic < nr_ioapics; apic++)
-		for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) {
+		for (pin = 0; pin < nr_ioapic_registers[apic]; pin++)
-			struct IO_APIC_route_entry entry;
+			early_ioapic_entries[apic][pin] =
-
-			entry = early_ioapic_entries[apic][pin] =
 				ioapic_read_entry(apic, pin);
-			if (!entry.mask) {
-				entry.mask = 1;
-				ioapic_write_entry(apic, pin, entry);
-			}
-		}
 
 	return 0;
 
@@ -898,6 +897,25 @@ int save_mask_IO_APIC_setup(void)
 	return -ENOMEM;
 }
 
+void mask_IO_APIC_setup(void)
+{
+	int apic, pin;
+
+	for (apic = 0; apic < nr_ioapics; apic++) {
+		if (!early_ioapic_entries[apic])
+			break;
+		for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) {
+			struct IO_APIC_route_entry entry;
+
+			entry = early_ioapic_entries[apic][pin];
+			if (!entry.mask) {
+				entry.mask = 1;
+				ioapic_write_entry(apic, pin, entry);
+			}
+		}
+	}
+}
+
 void restore_IO_APIC_setup(void)
 {
 	int apic, pin;
@@ -1411,9 +1429,7 @@ void __setup_vector_irq(int cpu)
 }
 
 static struct irq_chip ioapic_chip;
-#ifdef CONFIG_INTR_REMAP
 static struct irq_chip ir_ioapic_chip;
-#endif
 
 #define IOAPIC_AUTO     -1
 #define IOAPIC_EDGE     0
@@ -1452,7 +1468,6 @@ static void ioapic_register_intr(int irq, struct irq_desc *desc, unsigned long t
 	else
 		desc->status &= ~IRQ_LEVEL;
 
-#ifdef CONFIG_INTR_REMAP
 	if (irq_remapped(irq)) {
 		desc->status |= IRQ_MOVE_PCNTXT;
 		if (trigger)
@@ -1464,7 +1479,7 @@ static void ioapic_register_intr(int irq, struct irq_desc *desc, unsigned long t
 						      handle_edge_irq, "edge");
 		return;
 	}
-#endif
+
 	if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) ||
 	    trigger == IOAPIC_LEVEL)
 		set_irq_chip_and_handler_name(irq, &ioapic_chip,
@@ -1478,14 +1493,13 @@ static void ioapic_register_intr(int irq, struct irq_desc *desc, unsigned long t
 int setup_ioapic_entry(int apic_id, int irq,
 		       struct IO_APIC_route_entry *entry,
 		       unsigned int destination, int trigger,
-		       int polarity, int vector)
+		       int polarity, int vector, int pin)
 {
 	/*
 	 * add it to the IO-APIC irq-routing table:
 	 */
 	memset(entry,0,sizeof(*entry));
 
-#ifdef CONFIG_INTR_REMAP
 	if (intr_remapping_enabled) {
 		struct intel_iommu *iommu = map_ioapic_to_ir(apic_id);
 		struct irte irte;
@@ -1504,7 +1518,14 @@ int setup_ioapic_entry(int apic_id, int irq,
 
 		irte.present = 1;
 		irte.dst_mode = apic->irq_dest_mode;
-		irte.trigger_mode = trigger;
+		/*
+		 * Trigger mode in the IRTE will always be edge, and the
+		 * actual level or edge trigger will be setup in the IO-APIC
+		 * RTE. This will help simplify level triggered irq migration.
+		 * For more details, see the comments above explainig IO-APIC
+		 * irq migration in the presence of interrupt-remapping.
+		 */
+		irte.trigger_mode = 0;
 		irte.dlvry_mode = apic->irq_delivery_mode;
 		irte.vector = vector;
 		irte.dest_id = IRTE_DEST(destination);
@@ -1515,18 +1536,21 @@ int setup_ioapic_entry(int apic_id, int irq,
 		ir_entry->zero = 0;
 		ir_entry->format = 1;
 		ir_entry->index = (index & 0x7fff);
-	} else
+		/*
-#endif
+		 * IO-APIC RTE will be configured with virtual vector.
-	{
+		 * irq handler will do the explicit EOI to the io-apic.
+		 */
+		ir_entry->vector = pin;
+	} else {
 		entry->delivery_mode = apic->irq_delivery_mode;
 		entry->dest_mode = apic->irq_dest_mode;
 		entry->dest = destination;
+		entry->vector = vector;
 	}
 
 	entry->mask = 0;				/* enable IRQ */
 	entry->trigger = trigger;
 	entry->polarity = polarity;
-	entry->vector = vector;
 
 	/* Mask level triggered irqs.
 	 * Use IRQ_DELAYED_DISABLE for edge triggered irqs.
@@ -1561,7 +1585,7 @@ static void setup_IO_APIC_irq(int apic_id, int pin, unsigned int irq, struct irq
 
 
 	if (setup_ioapic_entry(mp_ioapics[apic_id].apicid, irq, &entry,
-			       dest, trigger, polarity, cfg->vector)) {
+			       dest, trigger, polarity, cfg->vector, pin)) {
 		printk("Failed to setup ioapic entry for ioapic  %d, pin %d\n",
 		       mp_ioapics[apic_id].apicid, pin);
 		__clear_irq_vector(irq, cfg);
@@ -1642,10 +1666,8 @@ static void __init setup_timer_IRQ0_pin(unsigned int apic_id, unsigned int pin,
 {
 	struct IO_APIC_route_entry entry;
 
-#ifdef CONFIG_INTR_REMAP
 	if (intr_remapping_enabled)
 		return;
-#endif
 
 	memset(&entry, 0, sizeof(entry));
 
@@ -2040,8 +2062,13 @@ void disable_IO_APIC(void)
 	 * If the i8259 is routed through an IOAPIC
 	 * Put that IOAPIC in virtual wire mode
 	 * so legacy interrupts can be delivered.
+	 *
+	 * With interrupt-remapping, for now we will use virtual wire A mode,
+	 * as virtual wire B is little complex (need to configure both
+	 * IOAPIC RTE aswell as interrupt-remapping table entry).
+	 * As this gets called during crash dump, keep this simple for now.
 	 */
-	if (ioapic_i8259.pin != -1) {
+	if (ioapic_i8259.pin != -1 && !intr_remapping_enabled) {
 		struct IO_APIC_route_entry entry;
 
 		memset(&entry, 0, sizeof(entry));
@@ -2061,7 +2088,10 @@ void disable_IO_APIC(void)
 		ioapic_write_entry(ioapic_i8259.apic, ioapic_i8259.pin, entry);
 	}
 
-	disconnect_bsp_APIC(ioapic_i8259.pin != -1);
+	/*
+	 * Use virtual wire A mode when interrupt remapping is enabled.
+	 */
+	disconnect_bsp_APIC(!intr_remapping_enabled && ioapic_i8259.pin != -1);
 }
 
 #ifdef CONFIG_X86_32
@@ -2303,37 +2333,24 @@ static int ioapic_retrigger_irq(unsigned int irq)
 #ifdef CONFIG_SMP
 
 #ifdef CONFIG_INTR_REMAP
-static void ir_irq_migration(struct work_struct *work);
-
-static DECLARE_DELAYED_WORK(ir_migration_work, ir_irq_migration);
 
 /*
  * Migrate the IO-APIC irq in the presence of intr-remapping.
  *
- * For edge triggered, irq migration is a simple atomic update(of vector
+ * For both level and edge triggered, irq migration is a simple atomic
- * and cpu destination) of IRTE and flush the hardware cache.
+ * update(of vector and cpu destination) of IRTE and flush the hardware cache.
  *
- * For level triggered, we need to modify the io-apic RTE aswell with the update
+ * For level triggered, we eliminate the io-apic RTE modification (with the
- * vector information, along with modifying IRTE with vector and destination.
+ * updated vector information), by using a virtual vector (io-apic pin number).
- * So irq migration for level triggered is little  bit more complex compared to
+ * Real vector that is used for interrupting cpu will be coming from
- * edge triggered migration. But the good news is, we use the same algorithm
+ * the interrupt-remapping table entry.
- * for level triggered migration as we have today, only difference being,
- * we now initiate the irq migration from process context instead of the
- * interrupt context.
- *
- * In future, when we do a directed EOI (combined with cpu EOI broadcast
- * suppression) to the IO-APIC, level triggered irq migration will also be
- * as simple as edge triggered migration and we can do the irq migration
- * with a simple atomic update to IO-APIC RTE.
  */
 static void
 migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask)
 {
 	struct irq_cfg *cfg;
 	struct irte irte;
-	int modify_ioapic_rte;
 	unsigned int dest;
-	unsigned long flags;
 	unsigned int irq;
 
 	if (!cpumask_intersects(mask, cpu_online_mask))
@@ -2351,13 +2368,6 @@ migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask)
 
 	dest = apic->cpu_mask_to_apicid_and(cfg->domain, mask);
 
-	modify_ioapic_rte = desc->status & IRQ_LEVEL;
-	if (modify_ioapic_rte) {
-		spin_lock_irqsave(&ioapic_lock, flags);
-		__target_IO_APIC_irq(irq, dest, cfg);
-		spin_unlock_irqrestore(&ioapic_lock, flags);
-	}
-
 	irte.vector = cfg->vector;
 	irte.dest_id = IRTE_DEST(dest);
 
@@ -2372,73 +2382,12 @@ migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask)
 	cpumask_copy(desc->affinity, mask);
 }
 
-static int migrate_irq_remapped_level_desc(struct irq_desc *desc)
-{
-	int ret = -1;
-	struct irq_cfg *cfg = desc->chip_data;
-
-	mask_IO_APIC_irq_desc(desc);
-
-	if (io_apic_level_ack_pending(cfg)) {
-		/*
-		 * Interrupt in progress. Migrating irq now will change the
-		 * vector information in the IO-APIC RTE and that will confuse
-		 * the EOI broadcast performed by cpu.
-		 * So, delay the irq migration to the next instance.
-		 */
-		schedule_delayed_work(&ir_migration_work, 1);
-		goto unmask;
-	}
-
-	/* everthing is clear. we have right of way */
-	migrate_ioapic_irq_desc(desc, desc->pending_mask);
-
-	ret = 0;
-	desc->status &= ~IRQ_MOVE_PENDING;
-	cpumask_clear(desc->pending_mask);
-
-unmask:
-	unmask_IO_APIC_irq_desc(desc);
-
-	return ret;
-}
-
-static void ir_irq_migration(struct work_struct *work)
-{
-	unsigned int irq;
-	struct irq_desc *desc;
-
-	for_each_irq_desc(irq, desc) {
-		if (desc->status & IRQ_MOVE_PENDING) {
-			unsigned long flags;
-
-			spin_lock_irqsave(&desc->lock, flags);
-			if (!desc->chip->set_affinity ||
-			    !(desc->status & IRQ_MOVE_PENDING)) {
-				desc->status &= ~IRQ_MOVE_PENDING;
-				spin_unlock_irqrestore(&desc->lock, flags);
-				continue;
-			}
-
-			desc->chip->set_affinity(irq, desc->pending_mask);
-			spin_unlock_irqrestore(&desc->lock, flags);
-		}
-	}
-}
-
 /*
  * Migrates the IRQ destination in the process context.
  */
 static void set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc,
 					    const struct cpumask *mask)
 {
-	if (desc->status & IRQ_LEVEL) {
-		desc->status |= IRQ_MOVE_PENDING;
-		cpumask_copy(desc->pending_mask, mask);
-		migrate_irq_remapped_level_desc(desc);
-		return;
-	}
-
 	migrate_ioapic_irq_desc(desc, mask);
 }
 static void set_ir_ioapic_affinity_irq(unsigned int irq,
@@ -2448,6 +2397,11 @@ static void set_ir_ioapic_affinity_irq(unsigned int irq,
 
 	set_ir_ioapic_affinity_irq_desc(desc, mask);
 }
+#else
+static inline void set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc,
+						   const struct cpumask *mask)
+{
+}
 #endif
 
 asmlinkage void smp_irq_move_cleanup_interrupt(void)
@@ -2461,6 +2415,7 @@ asmlinkage void smp_irq_move_cleanup_interrupt(void)
 	me = smp_processor_id();
 	for (vector = FIRST_EXTERNAL_VECTOR; vector < NR_VECTORS; vector++) {
 		unsigned int irq;
+		unsigned int irr;
 		struct irq_desc *desc;
 		struct irq_cfg *cfg;
 		irq = __get_cpu_var(vector_irq)[vector];
@@ -2480,6 +2435,18 @@ asmlinkage void smp_irq_move_cleanup_interrupt(void)
 		if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain))
 			goto unlock;
 
+		irr = apic_read(APIC_IRR + (vector / 32 * 0x10));
+		/*
+		 * Check if the vector that needs to be cleanedup is
+		 * registered at the cpu's IRR. If so, then this is not
+		 * the best time to clean it up. Lets clean it up in the
+		 * next attempt by sending another IRQ_MOVE_CLEANUP_VECTOR
+		 * to myself.
+		 */
+		if (irr  & (1 << (vector % 32))) {
+			apic->send_IPI_self(IRQ_MOVE_CLEANUP_VECTOR);
+			goto unlock;
+		}
 		__get_cpu_var(vector_irq)[vector] = -1;
 		cfg->move_cleanup_count--;
 unlock:
@@ -2529,9 +2496,44 @@ static inline void irq_complete_move(struct irq_desc **descp) {}
 #endif
 
 #ifdef CONFIG_INTR_REMAP
+static void __eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg)
+{
+	int apic, pin;
+	struct irq_pin_list *entry;
+
+	entry = cfg->irq_2_pin;
+	for (;;) {
+
+		if (!entry)
+			break;
+
+		apic = entry->apic;
+		pin = entry->pin;
+		io_apic_eoi(apic, pin);
+		entry = entry->next;
+	}
+}
+
+static void
+eoi_ioapic_irq(struct irq_desc *desc)
+{
+	struct irq_cfg *cfg;
+	unsigned long flags;
+	unsigned int irq;
+
+	irq = desc->irq;
+	cfg = desc->chip_data;
+
+	spin_lock_irqsave(&ioapic_lock, flags);
+	__eoi_ioapic_irq(irq, cfg);
+	spin_unlock_irqrestore(&ioapic_lock, flags);
+}
+
 static void ack_x2apic_level(unsigned int irq)
 {
+	struct irq_desc *desc = irq_to_desc(irq);
 	ack_x2APIC_irq();
+	eoi_ioapic_irq(desc);
 }
 
 static void ack_x2apic_edge(unsigned int irq)
@@ -2662,20 +2664,20 @@ static struct irq_chip ioapic_chip __read_mostly = {
 	.retrigger	= ioapic_retrigger_irq,
 };
 
-#ifdef CONFIG_INTR_REMAP
 static struct irq_chip ir_ioapic_chip __read_mostly = {
 	.name		= "IR-IO-APIC",
 	.startup	= startup_ioapic_irq,
 	.mask		= mask_IO_APIC_irq,
 	.unmask		= unmask_IO_APIC_irq,
+#ifdef CONFIG_INTR_REMAP
 	.ack		= ack_x2apic_edge,
 	.eoi		= ack_x2apic_level,
 #ifdef CONFIG_SMP
 	.set_affinity	= set_ir_ioapic_affinity_irq,
+#endif
 #endif
 	.retrigger	= ioapic_retrigger_irq,
 };
-#endif
 
 static inline void init_IO_APIC_traps(void)
 {
@@ -2901,10 +2903,8 @@ static inline void __init check_timer(void)
 	 * 8259A.
 	 */
 	if (pin1 == -1) {
-#ifdef CONFIG_INTR_REMAP
 		if (intr_remapping_enabled)
 			panic("BIOS bug: timer not connected to IO-APIC");
-#endif
 		pin1 = pin2;
 		apic1 = apic2;
 		no_pin1 = 1;
@@ -2940,10 +2940,8 @@ static inline void __init check_timer(void)
 				clear_IO_APIC_pin(0, pin1);
 			goto out;
 		}
-#ifdef CONFIG_INTR_REMAP
 		if (intr_remapping_enabled)
 			panic("timer doesn't work through Interrupt-remapped IO-APIC");
-#endif
 		local_irq_disable();
 		clear_IO_APIC_pin(apic1, pin1);
 		if (!no_pin1)
@@ -3237,9 +3235,7 @@ void destroy_irq(unsigned int irq)
 	if (desc)
 		desc->chip_data = cfg;
 
-#ifdef CONFIG_INTR_REMAP
 	free_irte(irq);
-#endif
 	spin_lock_irqsave(&vector_lock, flags);
 	__clear_irq_vector(irq, cfg);
 	spin_unlock_irqrestore(&vector_lock, flags);
@@ -3265,7 +3261,6 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms
 
 	dest = apic->cpu_mask_to_apicid_and(cfg->domain, apic->target_cpus());
 
-#ifdef CONFIG_INTR_REMAP
 	if (irq_remapped(irq)) {
 		struct irte irte;
 		int ir_index;
@@ -3291,10 +3286,13 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms
 				  MSI_ADDR_IR_SHV |
 				  MSI_ADDR_IR_INDEX1(ir_index) |
 				  MSI_ADDR_IR_INDEX2(ir_index);
-	} else
+	} else {
-#endif
+		if (x2apic_enabled())
-	{
+			msg->address_hi = MSI_ADDR_BASE_HI |
-		msg->address_hi = MSI_ADDR_BASE_HI;
+					  MSI_ADDR_EXT_DEST_ID(dest);
+		else
+			msg->address_hi = MSI_ADDR_BASE_HI;
+
 		msg->address_lo =
 			MSI_ADDR_BASE_LO |
 			((apic->irq_dest_mode == 0) ?
@@ -3394,14 +3392,15 @@ static struct irq_chip msi_chip = {
 	.retrigger	= ioapic_retrigger_irq,
 };
 
-#ifdef CONFIG_INTR_REMAP
 static struct irq_chip msi_ir_chip = {
 	.name		= "IR-PCI-MSI",
 	.unmask		= unmask_msi_irq,
 	.mask		= mask_msi_irq,
+#ifdef CONFIG_INTR_REMAP
 	.ack		= ack_x2apic_edge,
 #ifdef CONFIG_SMP
 	.set_affinity	= ir_set_msi_irq_affinity,
+#endif
 #endif
 	.retrigger	= ioapic_retrigger_irq,
 };
@@ -3432,7 +3431,6 @@ static int msi_alloc_irte(struct pci_dev *dev, int irq, int nvec)
 	}
 	return index;
 }
-#endif
 
 static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq)
 {
@@ -3446,7 +3444,6 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq)
 	set_irq_msi(irq, msidesc);
 	write_msi_msg(irq, &msg);
 
-#ifdef CONFIG_INTR_REMAP
 	if (irq_remapped(irq)) {
 		struct irq_desc *desc = irq_to_desc(irq);
 		/*
@@ -3455,7 +3452,6 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq)
 		desc->status |= IRQ_MOVE_PCNTXT;
 		set_irq_chip_and_handler_name(irq, &msi_ir_chip, handle_edge_irq, "edge");
 	} else
-#endif
 		set_irq_chip_and_handler_name(irq, &msi_chip, handle_edge_irq, "edge");
 
 	dev_printk(KERN_DEBUG, &dev->dev, "irq %d for MSI/MSI-X\n", irq);
@@ -3469,11 +3465,8 @@ int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
 	int ret, sub_handle;
 	struct msi_desc *msidesc;
 	unsigned int irq_want;
-
+	struct intel_iommu *iommu = NULL;
-#ifdef CONFIG_INTR_REMAP
-	struct intel_iommu *iommu = 0;
 	int index = 0;
-#endif
 
 	irq_want = nr_irqs_gsi;
 	sub_handle = 0;
@@ -3482,7 +3475,6 @@ int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
 		if (irq == 0)
 			return -1;
 		irq_want = irq + 1;
-#ifdef CONFIG_INTR_REMAP
 		if (!intr_remapping_enabled)
 			goto no_ir;
 
@@ -3510,7 +3502,6 @@ int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
 			set_irte_irq(irq, iommu, index, sub_handle);
 		}
 no_ir:
-#endif
 		ret = setup_msi_irq(dev, msidesc, irq);
 		if (ret < 0)
 			goto error;
@@ -3528,7 +3519,7 @@ void arch_teardown_msi_irq(unsigned int irq)
 	destroy_irq(irq);
 }
 
-#ifdef CONFIG_DMAR
+#if defined (CONFIG_DMAR) || defined (CONFIG_INTR_REMAP)
 #ifdef CONFIG_SMP
 static void dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
@@ -3609,7 +3600,7 @@ static void hpet_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
 
 #endif /* CONFIG_SMP */
 
-struct irq_chip hpet_msi_type = {
+static struct irq_chip hpet_msi_type = {
 	.name = "HPET_MSI",
 	.unmask = hpet_msi_unmask,
 	.mask = hpet_msi_mask,
@@ -4045,11 +4036,9 @@ void __init setup_ioapic_dest(void)
 			else
 				mask = apic->target_cpus();
 
-#ifdef CONFIG_INTR_REMAP
 			if (intr_remapping_enabled)
 				set_ir_ioapic_affinity_irq_desc(desc, mask);
 			else
-#endif
 				set_ioapic_affinity_irq_desc(desc, mask);
 		}
 
@@ -4142,9 +4131,12 @@ static int __init ioapic_insert_resources(void)
 	struct resource *r = ioapic_resources;
 
 	if (!r) {
-		printk(KERN_ERR
+		if (nr_ioapics > 0) {
-		       "IO APIC resources could be not be allocated.\n");
+			printk(KERN_ERR
-		return -1;
+				"IO APIC resources couldn't be allocated.\n");
+			return -1;
+		}
+		return 0;
 	}
 
 	for (i = 0; i < nr_ioapics; i++) {

arch/x86/kernel/apic/probe_64.c

@@ -68,6 +68,13 @@ void __init default_setup_apic_routing(void)
 			apic = &apic_physflat;
 		printk(KERN_INFO "Setting APIC routing to %s\n", apic->name);
 	}
+
+	/*
+	 * Now that apic routing model is selected, configure the
+	 * fault handling for intr remapping.
+	 */
+	if (intr_remapping_enabled)
+		enable_drhd_fault_handling();
 }
 
 /* Same for both flat and physical. */

arch/x86/kernel/apic/x2apic_cluster.c

@@ -57,6 +57,8 @@ static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector)
 	unsigned long query_cpu;
 	unsigned long flags;
 
+	x2apic_wrmsr_fence();
+
 	local_irq_save(flags);
 	for_each_cpu(query_cpu, mask) {
 		__x2apic_send_IPI_dest(
@@ -73,6 +75,8 @@ static void
 	unsigned long query_cpu;
 	unsigned long flags;
 
+	x2apic_wrmsr_fence();
+
 	local_irq_save(flags);
 	for_each_cpu(query_cpu, mask) {
 		if (query_cpu == this_cpu)
@@ -90,6 +94,8 @@ static void x2apic_send_IPI_allbutself(int vector)
 	unsigned long query_cpu;
 	unsigned long flags;
 
+	x2apic_wrmsr_fence();
+
 	local_irq_save(flags);
 	for_each_online_cpu(query_cpu) {
 		if (query_cpu == this_cpu)

arch/x86/kernel/apic/x2apic_phys.c

@@ -58,6 +58,8 @@ static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector)
 	unsigned long query_cpu;
 	unsigned long flags;
 
+	x2apic_wrmsr_fence();
+
 	local_irq_save(flags);
 	for_each_cpu(query_cpu, mask) {
 		__x2apic_send_IPI_dest(per_cpu(x86_cpu_to_apicid, query_cpu),
@@ -73,6 +75,8 @@ static void
 	unsigned long query_cpu;
 	unsigned long flags;
 
+	x2apic_wrmsr_fence();
+
 	local_irq_save(flags);
 	for_each_cpu(query_cpu, mask) {
 		if (query_cpu != this_cpu)
@@ -89,6 +93,8 @@ static void x2apic_send_IPI_allbutself(int vector)
 	unsigned long query_cpu;
 	unsigned long flags;
 
+	x2apic_wrmsr_fence();
+
 	local_irq_save(flags);
 	for_each_online_cpu(query_cpu) {
 		if (query_cpu == this_cpu)

arch/x86/kernel/check.c

@@ -83,15 +83,15 @@ void __init setup_bios_corruption_check(void)
 		u64 size;
 		addr = find_e820_area_size(addr, &size, PAGE_SIZE);
 
-		if (addr == 0)
+		if (!(addr + 1))
+			break;
+
+		if (addr >= corruption_check_size)
 			break;
 
 		if ((addr + size) > corruption_check_size)
 			size = corruption_check_size - addr;
 
-		if (size == 0)
-			break;
-
 		e820_update_range(addr, size, E820_RAM, E820_RESERVED);
 		scan_areas[num_scan_areas].addr = addr;
 		scan_areas[num_scan_areas].size = size;

arch/x86/kernel/cpu/Makefile

@@ -14,11 +14,12 @@ obj-y			+= vmware.o hypervisor.o
 obj-$(CONFIG_X86_32)	+= bugs.o cmpxchg.o
 obj-$(CONFIG_X86_64)	+= bugs_64.o
 
+obj-$(CONFIG_X86_CPU_DEBUG)		+= cpu_debug.o
+
 obj-$(CONFIG_CPU_SUP_INTEL)		+= intel.o
 obj-$(CONFIG_CPU_SUP_AMD)		+= amd.o
 obj-$(CONFIG_CPU_SUP_CYRIX_32)		+= cyrix.o
-obj-$(CONFIG_CPU_SUP_CENTAUR_32)	+= centaur.o
+obj-$(CONFIG_CPU_SUP_CENTAUR)		+= centaur.o
-obj-$(CONFIG_CPU_SUP_CENTAUR_64)	+= centaur_64.o
 obj-$(CONFIG_CPU_SUP_TRANSMETA_32)	+= transmeta.o
 obj-$(CONFIG_CPU_SUP_UMC_32)		+= umc.o
 

arch/x86/kernel/cpu/addon_cpuid_features.c

@@ -29,7 +29,7 @@ void __cpuinit init_scattered_cpuid_features(struct cpuinfo_x86 *c)
 	u32 regs[4];
 	const struct cpuid_bit *cb;
 
-	static const struct cpuid_bit cpuid_bits[] = {
+	static const struct cpuid_bit __cpuinitconst cpuid_bits[] = {
 		{ X86_FEATURE_IDA, CR_EAX, 1, 0x00000006 },
 		{ 0, 0, 0, 0 }
 	};

arch/x86/kernel/cpu/amd.c

@@ -5,6 +5,7 @@
 #include <asm/io.h>
 #include <asm/processor.h>
 #include <asm/apic.h>
+#include <asm/cpu.h>
 
 #ifdef CONFIG_X86_64
 # include <asm/numa_64.h>
@@ -141,6 +142,55 @@ static void __cpuinit init_amd_k6(struct cpuinfo_x86 *c)
 	}
 }
 
+static void __cpuinit amd_k7_smp_check(struct cpuinfo_x86 *c)
+{
+#ifdef CONFIG_SMP
+	/* calling is from identify_secondary_cpu() ? */
+	if (c->cpu_index == boot_cpu_id)
+		return;
+
+	/*
+	 * Certain Athlons might work (for various values of 'work') in SMP
+	 * but they are not certified as MP capable.
+	 */
+	/* Athlon 660/661 is valid. */
+	if ((c->x86_model == 6) && ((c->x86_mask == 0) ||
+	    (c->x86_mask == 1)))
+		goto valid_k7;
+
+	/* Duron 670 is valid */
+	if ((c->x86_model == 7) && (c->x86_mask == 0))
+		goto valid_k7;
+
+	/*
+	 * Athlon 662, Duron 671, and Athlon >model 7 have capability
+	 * bit. It's worth noting that the A5 stepping (662) of some
+	 * Athlon XP's have the MP bit set.
+	 * See http://www.heise.de/newsticker/data/jow-18.10.01-000 for
+	 * more.
+	 */
+	if (((c->x86_model == 6) && (c->x86_mask >= 2)) ||
+	    ((c->x86_model == 7) && (c->x86_mask >= 1)) ||
+	     (c->x86_model > 7))
+		if (cpu_has_mp)
+			goto valid_k7;
+
+	/* If we get here, not a certified SMP capable AMD system. */
+
+	/*
+	 * Don't taint if we are running SMP kernel on a single non-MP
+	 * approved Athlon
+	 */
+	WARN_ONCE(1, "WARNING: This combination of AMD"
+		"processors is not suitable for SMP.\n");
+	if (!test_taint(TAINT_UNSAFE_SMP))
+		add_taint(TAINT_UNSAFE_SMP);
+
+valid_k7:
+	;
+#endif
+}
+
 static void __cpuinit init_amd_k7(struct cpuinfo_x86 *c)
 {
 	u32 l, h;
@@ -175,6 +225,8 @@ static void __cpuinit init_amd_k7(struct cpuinfo_x86 *c)
 	}
 
 	set_cpu_cap(c, X86_FEATURE_K7);
+
+	amd_k7_smp_check(c);
 }
 #endif
 
@@ -450,7 +502,7 @@ static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 *c, unsigned int
 }
 #endif
 
-static struct cpu_dev amd_cpu_dev __cpuinitdata = {
+static const struct cpu_dev __cpuinitconst amd_cpu_dev = {
 	.c_vendor	= "AMD",
 	.c_ident	= { "AuthenticAMD" },
 #ifdef CONFIG_X86_32

arch/x86/kernel/cpu/centaur.c

@@ -1,11 +1,11 @@
+#include <linux/bitops.h>
 #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 <asm/msr.h>
 
 #include "cpu.h"
 
@@ -276,7 +276,7 @@ static void __cpuinit init_c3(struct cpuinfo_x86 *c)
 		 */
 		c->x86_capability[5] = cpuid_edx(0xC0000001);
 	}
-
+#ifdef CONFIG_X86_32
 	/* Cyrix III family needs CX8 & PGE explicitly enabled. */
 	if (c->x86_model >= 6 && c->x86_model <= 9) {
 		rdmsr(MSR_VIA_FCR, lo, hi);
@@ -288,6 +288,11 @@ static void __cpuinit init_c3(struct cpuinfo_x86 *c)
 	/* Before Nehemiah, the C3's had 3dNOW! */
 	if (c->x86_model >= 6 && c->x86_model < 9)
 		set_cpu_cap(c, X86_FEATURE_3DNOW);
+#endif
+	if (c->x86 == 0x6 && c->x86_model >= 0xf) {
+		c->x86_cache_alignment = c->x86_clflush_size * 2;
+		set_cpu_cap(c, X86_FEATURE_REP_GOOD);
+	}
 
 	display_cacheinfo(c);
 }
@@ -316,16 +321,25 @@ enum {
 static void __cpuinit early_init_centaur(struct cpuinfo_x86 *c)
 {
 	switch (c->x86) {
+#ifdef CONFIG_X86_32
 	case 5:
 		/* Emulate MTRRs using Centaur's MCR. */
 		set_cpu_cap(c, X86_FEATURE_CENTAUR_MCR);
 		break;
+#endif
+	case 6:
+		if (c->x86_model >= 0xf)
+			set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
+		break;
 	}
+#ifdef CONFIG_X86_64
+	set_cpu_cap(c, X86_FEATURE_SYSENTER32);
+#endif
 }
 
 static void __cpuinit init_centaur(struct cpuinfo_x86 *c)
 {
-
+#ifdef CONFIG_X86_32
 	char *name;
 	u32  fcr_set = 0;
 	u32  fcr_clr = 0;
@@ -337,8 +351,10 @@ static void __cpuinit init_centaur(struct cpuinfo_x86 *c)
 	 * 3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway
 	 */
 	clear_cpu_cap(c, 0*32+31);
-
+#endif
+	early_init_centaur(c);
 	switch (c->x86) {
+#ifdef CONFIG_X86_32
 	case 5:
 		switch (c->x86_model) {
 		case 4:
@@ -442,16 +458,20 @@ static void __cpuinit init_centaur(struct cpuinfo_x86 *c)
 		}
 		sprintf(c->x86_model_id, "WinChip %s", name);
 		break;
-
+#endif
 	case 6:
 		init_c3(c);
 		break;
 	}
+#ifdef CONFIG_X86_64
+	set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC);
+#endif
 }
 
 static unsigned int __cpuinit
 centaur_size_cache(struct cpuinfo_x86 *c, unsigned int size)
 {
+#ifdef CONFIG_X86_32
 	/* VIA C3 CPUs (670-68F) need further shifting. */
 	if ((c->x86 == 6) && ((c->x86_model == 7) || (c->x86_model == 8)))
 		size >>= 8;
@@ -464,11 +484,11 @@ centaur_size_cache(struct cpuinfo_x86 *c, unsigned int size)
 	if ((c->x86 == 6) && (c->x86_model == 9) &&
 				(c->x86_mask == 1) && (size == 65))
 		size -= 1;
-
+#endif
 	return size;
 }
 
-static struct cpu_dev centaur_cpu_dev __cpuinitdata = {
+static const struct cpu_dev __cpuinitconst centaur_cpu_dev = {
 	.c_vendor	= "Centaur",
 	.c_ident	= { "CentaurHauls" },
 	.c_early_init	= early_init_centaur,

arch/x86/kernel/cpu/centaur_64.c

@@ -1,37 +0,0 @@
-#include <linux/init.h>
-#include <linux/smp.h>
-
-#include <asm/cpufeature.h>
-#include <asm/processor.h>
-
-#include "cpu.h"
-
-static void __cpuinit early_init_centaur(struct cpuinfo_x86 *c)
-{
-	if (c->x86 == 0x6 && c->x86_model >= 0xf)
-		set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
-
-	set_cpu_cap(c, X86_FEATURE_SYSENTER32);
-}
-
-static void __cpuinit init_centaur(struct cpuinfo_x86 *c)
-{
-	early_init_centaur(c);
-
-	if (c->x86 == 0x6 && c->x86_model >= 0xf) {
-		c->x86_cache_alignment = c->x86_clflush_size * 2;
-		set_cpu_cap(c, X86_FEATURE_REP_GOOD);
-	}
-	set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC);
-}
-
-static struct cpu_dev centaur_cpu_dev __cpuinitdata = {
-	.c_vendor	= "Centaur",
-	.c_ident	= { "CentaurHauls" },
-	.c_early_init	= early_init_centaur,
-	.c_init		= init_centaur,
-	.c_x86_vendor	= X86_VENDOR_CENTAUR,
-};
-
-cpu_dev_register(centaur_cpu_dev);
-

arch/x86/kernel/cpu/common.c

@@ -1,52 +1,52 @@
-#include <linux/init.h>
-#include <linux/kernel.h>
-#include <linux/sched.h>
-#include <linux/string.h>
 #include <linux/bootmem.h>
+#include <linux/linkage.h>
 #include <linux/bitops.h>
+#include <linux/kernel.h>
 #include <linux/module.h>
-#include <linux/kgdb.h>
-#include <linux/topology.h>
-#include <linux/delay.h>
-#include <linux/smp.h>
 #include <linux/percpu.h>
-#include <asm/i387.h>
+#include <linux/string.h>
-#include <asm/msr.h>
+#include <linux/delay.h>
-#include <asm/io.h>
+#include <linux/sched.h>
-#include <asm/linkage.h>
+#include <linux/init.h>
+#include <linux/kgdb.h>
+#include <linux/smp.h>
+#include <linux/io.h>
+
+#include <asm/stackprotector.h>
 #include <asm/mmu_context.h>
-#include <asm/mtrr.h>
+#include <asm/hypervisor.h>
-#include <asm/mce.h>
+#include <asm/processor.h>
-#include <asm/pat.h>
+#include <asm/sections.h>
-#include <asm/asm.h>
+#include <asm/topology.h>
-#include <asm/numa.h>
-#include <asm/smp.h>
-#include <asm/cpu.h>
 #include <asm/cpumask.h>
+#include <asm/pgtable.h>
+#include <asm/atomic.h>
+#include <asm/proto.h>
+#include <asm/setup.h>
 #include <asm/apic.h>
+#include <asm/desc.h>
+#include <asm/i387.h>
+#include <asm/mtrr.h>
+#include <asm/numa.h>
+#include <asm/asm.h>
+#include <asm/cpu.h>
+#include <asm/mce.h>
+#include <asm/msr.h>
+#include <asm/pat.h>
+#include <asm/smp.h>
 
 #ifdef CONFIG_X86_LOCAL_APIC
 #include <asm/uv/uv.h>
 #endif
 
-#include <asm/pgtable.h>
-#include <asm/processor.h>
-#include <asm/desc.h>
-#include <asm/atomic.h>
-#include <asm/proto.h>
-#include <asm/sections.h>
-#include <asm/setup.h>
-#include <asm/hypervisor.h>
-#include <asm/stackprotector.h>
-
 #include "cpu.h"
 
 #ifdef CONFIG_X86_64
 
 /* all of these masks are initialized in setup_cpu_local_masks() */
-cpumask_var_t cpu_callin_mask;
-cpumask_var_t cpu_callout_mask;
 cpumask_var_t cpu_initialized_mask;
+cpumask_var_t cpu_callout_mask;
+cpumask_var_t cpu_callin_mask;
 
 /* representing cpus for which sibling maps can be computed */
 cpumask_var_t cpu_sibling_setup_mask;
@@ -62,15 +62,15 @@ void __init setup_cpu_local_masks(void)
 
 #else /* CONFIG_X86_32 */
 
-cpumask_t cpu_callin_map;
+cpumask_t cpu_sibling_setup_map;
 cpumask_t cpu_callout_map;
 cpumask_t cpu_initialized;
-cpumask_t cpu_sibling_setup_map;
+cpumask_t cpu_callin_map;
 
 #endif /* CONFIG_X86_32 */
 
 
-static struct cpu_dev *this_cpu __cpuinitdata;
+static const struct cpu_dev *this_cpu __cpuinitdata;
 
 DEFINE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page) = { .gdt = {
 #ifdef CONFIG_X86_64
@@ -79,48 +79,48 @@ DEFINE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page) = { .gdt = {
 	 * IRET will check the segment types  kkeil 2000/10/28
 	 * Also sysret mandates a special GDT layout
 	 *
-	 * The TLS descriptors are currently at a different place compared to i386.
+	 * TLS descriptors are currently at a different place compared to i386.
 	 * Hopefully nobody expects them at a fixed place (Wine?)
 	 */
-	[GDT_ENTRY_KERNEL32_CS] = { { { 0x0000ffff, 0x00cf9b00 } } },
+	[GDT_ENTRY_KERNEL32_CS]		= { { { 0x0000ffff, 0x00cf9b00 } } },
-	[GDT_ENTRY_KERNEL_CS] = { { { 0x0000ffff, 0x00af9b00 } } },
+	[GDT_ENTRY_KERNEL_CS]		= { { { 0x0000ffff, 0x00af9b00 } } },
-	[GDT_ENTRY_KERNEL_DS] = { { { 0x0000ffff, 0x00cf9300 } } },
+	[GDT_ENTRY_KERNEL_DS]		= { { { 0x0000ffff, 0x00cf9300 } } },
-	[GDT_ENTRY_DEFAULT_USER32_CS] = { { { 0x0000ffff, 0x00cffb00 } } },
+	[GDT_ENTRY_DEFAULT_USER32_CS]	= { { { 0x0000ffff, 0x00cffb00 } } },
-	[GDT_ENTRY_DEFAULT_USER_DS] = { { { 0x0000ffff, 0x00cff300 } } },
+	[GDT_ENTRY_DEFAULT_USER_DS]	= { { { 0x0000ffff, 0x00cff300 } } },
-	[GDT_ENTRY_DEFAULT_USER_CS] = { { { 0x0000ffff, 0x00affb00 } } },
+	[GDT_ENTRY_DEFAULT_USER_CS]	= { { { 0x0000ffff, 0x00affb00 } } },
 #else
-	[GDT_ENTRY_KERNEL_CS] = { { { 0x0000ffff, 0x00cf9a00 } } },
+	[GDT_ENTRY_KERNEL_CS]		= { { { 0x0000ffff, 0x00cf9a00 } } },
-	[GDT_ENTRY_KERNEL_DS] = { { { 0x0000ffff, 0x00cf9200 } } },
+	[GDT_ENTRY_KERNEL_DS]		= { { { 0x0000ffff, 0x00cf9200 } } },
-	[GDT_ENTRY_DEFAULT_USER_CS] = { { { 0x0000ffff, 0x00cffa00 } } },
+	[GDT_ENTRY_DEFAULT_USER_CS]	= { { { 0x0000ffff, 0x00cffa00 } } },
-	[GDT_ENTRY_DEFAULT_USER_DS] = { { { 0x0000ffff, 0x00cff200 } } },
+	[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.
 	 */
 	/* 32-bit code */
-	[GDT_ENTRY_PNPBIOS_CS32] = { { { 0x0000ffff, 0x00409a00 } } },
+	[GDT_ENTRY_PNPBIOS_CS32]	= { { { 0x0000ffff, 0x00409a00 } } },
 	/* 16-bit code */
-	[GDT_ENTRY_PNPBIOS_CS16] = { { { 0x0000ffff, 0x00009a00 } } },
+	[GDT_ENTRY_PNPBIOS_CS16]	= { { { 0x0000ffff, 0x00009a00 } } },
 	/* 16-bit data */
-	[GDT_ENTRY_PNPBIOS_DS] = { { { 0x0000ffff, 0x00009200 } } },
+	[GDT_ENTRY_PNPBIOS_DS]		= { { { 0x0000ffff, 0x00009200 } } },
 	/* 16-bit data */
-	[GDT_ENTRY_PNPBIOS_TS1] = { { { 0x00000000, 0x00009200 } } },
+	[GDT_ENTRY_PNPBIOS_TS1]		= { { { 0x00000000, 0x00009200 } } },
 	/* 16-bit data */
-	[GDT_ENTRY_PNPBIOS_TS2] = { { { 0x00000000, 0x00009200 } } },
+	[GDT_ENTRY_PNPBIOS_TS2]		= { { { 0x00000000, 0x00009200 } } },
 	/*
 	 * The APM segments have byte granularity and their bases
 	 * are set at run time.  All have 64k limits.
 	 */
 	/* 32-bit code */
-	[GDT_ENTRY_APMBIOS_BASE] = { { { 0x0000ffff, 0x00409a00 } } },
+	[GDT_ENTRY_APMBIOS_BASE]	= { { { 0x0000ffff, 0x00409a00 } } },
 	/* 16-bit code */
-	[GDT_ENTRY_APMBIOS_BASE+1] = { { { 0x0000ffff, 0x00009a00 } } },
+	[GDT_ENTRY_APMBIOS_BASE+1]	= { { { 0x0000ffff, 0x00009a00 } } },
 	/* data */
-	[GDT_ENTRY_APMBIOS_BASE+2] = { { { 0x0000ffff, 0x00409200 } } },
+	[GDT_ENTRY_APMBIOS_BASE+2]	= { { { 0x0000ffff, 0x00409200 } } },
 
-	[GDT_ENTRY_ESPFIX_SS] = { { { 0x00000000, 0x00c09200 } } },
+	[GDT_ENTRY_ESPFIX_SS]		= { { { 0x00000000, 0x00c09200 } } },
-	[GDT_ENTRY_PERCPU] = { { { 0x0000ffff, 0x00cf9200 } } },
+	[GDT_ENTRY_PERCPU]		= { { { 0x0000ffff, 0x00cf9200 } } },
 	GDT_STACK_CANARY_INIT
 #endif
 } };
@@ -164,16 +164,17 @@ static inline int flag_is_changeable_p(u32 flag)
 	 * the CPUID. Add "volatile" to not allow gcc to
 	 * optimize the subsequent calls to this function.
 	 */
-	asm volatile ("pushfl\n\t"
+	asm volatile ("pushfl		\n\t"
-		      "pushfl\n\t"
+		      "pushfl		\n\t"
-		      "popl %0\n\t"
+		      "popl %0		\n\t"
-		      "movl %0,%1\n\t"
+		      "movl %0, %1	\n\t"
-		      "xorl %2,%0\n\t"
+		      "xorl %2, %0	\n\t"
-		      "pushl %0\n\t"
+		      "pushl %0		\n\t"
-		      "popfl\n\t"
+		      "popfl		\n\t"
-		      "pushfl\n\t"
+		      "pushfl		\n\t"
-		      "popl %0\n\t"
+		      "popl %0		\n\t"
-		      "popfl\n\t"
+		      "popfl		\n\t"
+
 		      : "=&r" (f1), "=&r" (f2)
 		      : "ir" (flag));
 
@@ -188,18 +189,22 @@ static int __cpuinit have_cpuid_p(void)
 
 static void __cpuinit squash_the_stupid_serial_number(struct cpuinfo_x86 *c)
 {
-	if (cpu_has(c, X86_FEATURE_PN) && disable_x86_serial_nr) {
+	unsigned long lo, hi;
-		/* 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_cpu_cap(c, X86_FEATURE_PN);
 
-		/* Disabling the serial number may affect the cpuid level */
+	if (!cpu_has(c, X86_FEATURE_PN) || !disable_x86_serial_nr)
-		c->cpuid_level = cpuid_eax(0);
+		return;
-	}
+
+	/* Disable processor serial number: */
+
+	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_cpu_cap(c, X86_FEATURE_PN);
+
+	/* Disabling the serial number may affect the cpuid level */
+	c->cpuid_level = cpuid_eax(0);
 }
 
 static int __init x86_serial_nr_setup(char *s)
@@ -232,6 +237,7 @@ struct cpuid_dependent_feature {
 	u32 feature;
 	u32 level;
 };
+
 static const struct cpuid_dependent_feature __cpuinitconst
 cpuid_dependent_features[] = {
 	{ X86_FEATURE_MWAIT,		0x00000005 },
@@ -243,7 +249,11 @@ cpuid_dependent_features[] = {
 static void __cpuinit filter_cpuid_features(struct cpuinfo_x86 *c, bool warn)
 {
 	const struct cpuid_dependent_feature *df;
+
 	for (df = cpuid_dependent_features; df->feature; df++) {
+
+		if (!cpu_has(c, df->feature))
+			continue;
 		/*
 		 * Note: cpuid_level is set to -1 if unavailable, but
 		 * extended_extended_level is set to 0 if unavailable
@@ -251,32 +261,32 @@ static void __cpuinit filter_cpuid_features(struct cpuinfo_x86 *c, bool warn)
 		 * when signed; hence the weird messing around with
 		 * signs here...
 		 */
-		if (cpu_has(c, df->feature) &&
+		if (!((s32)df->level < 0 ?
-		    ((s32)df->level < 0 ?
 		     (u32)df->level > (u32)c->extended_cpuid_level :
-		     (s32)df->level > (s32)c->cpuid_level)) {
+		     (s32)df->level > (s32)c->cpuid_level))
-			clear_cpu_cap(c, df->feature);
+			continue;
-			if (warn)
+
-				printk(KERN_WARNING
+		clear_cpu_cap(c, df->feature);
-				       "CPU: CPU feature %s disabled "
+		if (!warn)
-				       "due to lack of CPUID level 0x%x\n",
+			continue;
-				       x86_cap_flags[df->feature],
+
-				       df->level);
+		printk(KERN_WARNING
-		}
+		       "CPU: CPU feature %s disabled, no CPUID level 0x%x\n",
+				x86_cap_flags[df->feature], df->level);
 	}
 }
 
 /*
  * 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
+ * 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)
+static const char *__cpuinit table_lookup_model(struct cpuinfo_x86 *c)
 {
-	struct cpu_model_info *info;
+	const struct cpu_model_info *info;
 
 	if (c->x86_model >= 16)
 		return NULL;	/* Range check */
@@ -307,8 +317,10 @@ void load_percpu_segment(int cpu)
 	load_stack_canary_segment();
 }
 
-/* Current gdt points %fs at the "master" per-cpu area: after this,
+/*
- * it's on the real one. */
+ * Current gdt points %fs at the "master" per-cpu area: after this,
+ * it's on the real one.
+ */
 void switch_to_new_gdt(int cpu)
 {
 	struct desc_ptr gdt_descr;
@@ -321,7 +333,7 @@ void switch_to_new_gdt(int cpu)
 	load_percpu_segment(cpu);
 }
 
-static struct cpu_dev *cpu_devs[X86_VENDOR_NUM] = {};
+static const struct cpu_dev *__cpuinitdata cpu_devs[X86_VENDOR_NUM] = {};
 
 static void __cpuinit default_init(struct cpuinfo_x86 *c)
 {
@@ -340,7 +352,7 @@ static void __cpuinit default_init(struct cpuinfo_x86 *c)
 #endif
 }
 
-static struct cpu_dev __cpuinitdata default_cpu = {
+static const struct cpu_dev __cpuinitconst default_cpu = {
 	.c_init	= default_init,
 	.c_vendor = "Unknown",
 	.c_x86_vendor = X86_VENDOR_UNKNOWN,
@@ -354,22 +366,24 @@ static void __cpuinit get_model_name(struct cpuinfo_x86 *c)
 	if (c->extended_cpuid_level < 0x80000004)
 		return;
 
-	v = (unsigned int *) c->x86_model_id;
+	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 */
+	 * Intel chips right-justify this string for some dumb reason;
+	 * undo that brain damage:
+	 */
 	p = q = &c->x86_model_id[0];
 	while (*p == ' ')
-	     p++;
+		p++;
 	if (p != q) {
-	     while (*p)
+		while (*p)
-		  *q++ = *p++;
+			*q++ = *p++;
-	     while (q <= &c->x86_model_id[48])
+		while (q <= &c->x86_model_id[48])
-		  *q++ = '\0';	/* Zero-pad the rest */
+			*q++ = '\0';	/* Zero-pad the rest */
 	}
 }
 
@@ -438,28 +452,31 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c)
 
 	if (smp_num_siblings == 1) {
 		printk(KERN_INFO  "CPU: Hyper-Threading is disabled\n");
-	} else if (smp_num_siblings > 1) {
+		goto out;
-
-		if (smp_num_siblings > nr_cpu_ids) {
-			printk(KERN_WARNING "CPU: Unsupported number of siblings %d",
-					smp_num_siblings);
-			smp_num_siblings = 1;
-			return;
-		}
-
-		index_msb = get_count_order(smp_num_siblings);
-		c->phys_proc_id = apic->phys_pkg_id(c->initial_apicid, index_msb);
-
-		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 = apic->phys_pkg_id(c->initial_apicid, index_msb) &
-					       ((1 << core_bits) - 1);
 	}
 
+	if (smp_num_siblings <= 1)
+		goto out;
+
+	if (smp_num_siblings > nr_cpu_ids) {
+		pr_warning("CPU: Unsupported number of siblings %d",
+			   smp_num_siblings);
+		smp_num_siblings = 1;
+		return;
+	}
+
+	index_msb = get_count_order(smp_num_siblings);
+	c->phys_proc_id = apic->phys_pkg_id(c->initial_apicid, index_msb);
+
+	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 = apic->phys_pkg_id(c->initial_apicid, index_msb) &
+				       ((1 << core_bits) - 1);
+
 out:
 	if ((c->x86_max_cores * smp_num_siblings) > 1) {
 		printk(KERN_INFO  "CPU: Physical Processor ID: %d\n",
@@ -473,8 +490,8 @@ void __cpuinit detect_ht(struct cpuinfo_x86 *c)
 static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c)
 {
 	char *v = c->x86_vendor_id;
-	int i;
 	static int printed;
+	int i;
 
 	for (i = 0; i < X86_VENDOR_NUM; i++) {
 		if (!cpu_devs[i])
@@ -483,6 +500,7 @@ static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c)
 		if (!strcmp(v, cpu_devs[i]->c_ident[0]) ||
 		    (cpu_devs[i]->c_ident[1] &&
 		     !strcmp(v, cpu_devs[i]->c_ident[1]))) {
+
 			this_cpu = cpu_devs[i];
 			c->x86_vendor = this_cpu->c_x86_vendor;
 			return;
@@ -491,7 +509,9 @@ static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c)
 
 	if (!printed) {
 		printed++;
-		printk(KERN_ERR "CPU: vendor_id '%s' unknown, using generic init.\n", v);
+		printk(KERN_ERR
+		    "CPU: vendor_id '%s' unknown, using generic init.\n", v);
+
 		printk(KERN_ERR "CPU: Your system may be unstable.\n");
 	}
 
@@ -511,14 +531,17 @@ void __cpuinit cpu_detect(struct cpuinfo_x86 *c)
 	/* Intel-defined flags: level 0x00000001 */
 	if (c->cpuid_level >= 0x00000001) {
 		u32 junk, tfms, cap0, misc;
+
 		cpuid(0x00000001, &tfms, &misc, &junk, &cap0);
 		c->x86 = (tfms >> 8) & 0xf;
 		c->x86_model = (tfms >> 4) & 0xf;
 		c->x86_mask = tfms & 0xf;
+
 		if (c->x86 == 0xf)
 			c->x86 += (tfms >> 20) & 0xff;
 		if (c->x86 >= 0x6)
 			c->x86_model += ((tfms >> 16) & 0xf) << 4;
+
 		if (cap0 & (1<<19)) {
 			c->x86_clflush_size = ((misc >> 8) & 0xff) * 8;
 			c->x86_cache_alignment = c->x86_clflush_size;
@@ -534,6 +557,7 @@ static void __cpuinit get_cpu_cap(struct cpuinfo_x86 *c)
 	/* 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;
@@ -542,6 +566,7 @@ static void __cpuinit get_cpu_cap(struct cpuinfo_x86 *c)
 	/* AMD-defined flags: level 0x80000001 */
 	xlvl = cpuid_eax(0x80000000);
 	c->extended_cpuid_level = xlvl;
+
 	if ((xlvl & 0xffff0000) == 0x80000000) {
 		if (xlvl >= 0x80000001) {
 			c->x86_capability[1] = cpuid_edx(0x80000001);
@@ -549,13 +574,15 @@ static void __cpuinit get_cpu_cap(struct cpuinfo_x86 *c)
 		}
 	}
 
-#ifdef CONFIG_X86_64
 	if (c->extended_cpuid_level >= 0x80000008) {
 		u32 eax = cpuid_eax(0x80000008);
 
 		c->x86_virt_bits = (eax >> 8) & 0xff;
 		c->x86_phys_bits = eax & 0xff;
 	}
+#ifdef CONFIG_X86_32
+	else if (cpu_has(c, X86_FEATURE_PAE) || cpu_has(c, X86_FEATURE_PSE36))
+		c->x86_phys_bits = 36;
 #endif
 
 	if (c->extended_cpuid_level >= 0x80000007)
@@ -602,8 +629,12 @@ static void __init early_identify_cpu(struct cpuinfo_x86 *c)
 {
 #ifdef CONFIG_X86_64
 	c->x86_clflush_size = 64;
+	c->x86_phys_bits = 36;
+	c->x86_virt_bits = 48;
 #else
 	c->x86_clflush_size = 32;
+	c->x86_phys_bits = 32;
+	c->x86_virt_bits = 32;
 #endif
 	c->x86_cache_alignment = c->x86_clflush_size;
 
@@ -634,12 +665,12 @@ static void __init early_identify_cpu(struct cpuinfo_x86 *c)
 
 void __init early_cpu_init(void)
 {
-	struct cpu_dev **cdev;
+	const struct cpu_dev *const *cdev;
 	int count = 0;
 
-	printk("KERNEL supported cpus:\n");
+	printk(KERN_INFO "KERNEL supported cpus:\n");
 	for (cdev = __x86_cpu_dev_start; cdev < __x86_cpu_dev_end; cdev++) {
-		struct cpu_dev *cpudev = *cdev;
+		const struct cpu_dev *cpudev = *cdev;
 		unsigned int j;
 
 		if (count >= X86_VENDOR_NUM)
@@ -650,7 +681,7 @@ void __init early_cpu_init(void)
 		for (j = 0; j < 2; j++) {
 			if (!cpudev->c_ident[j])
 				continue;
-			printk("  %s %s\n", cpudev->c_vendor,
+			printk(KERN_INFO "  %s %s\n", cpudev->c_vendor,
 				cpudev->c_ident[j]);
 		}
 	}
@@ -726,9 +757,13 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 	c->x86_coreid_bits = 0;
 #ifdef CONFIG_X86_64
 	c->x86_clflush_size = 64;
+	c->x86_phys_bits = 36;
+	c->x86_virt_bits = 48;
 #else
 	c->cpuid_level = -1;	/* CPUID not detected */
 	c->x86_clflush_size = 32;
+	c->x86_phys_bits = 32;
+	c->x86_virt_bits = 32;
 #endif
 	c->x86_cache_alignment = c->x86_clflush_size;
 	memset(&c->x86_capability, 0, sizeof c->x86_capability);
@@ -759,8 +794,8 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 	squash_the_stupid_serial_number(c);
 
 	/*
-	 * The vendor-specific functions might have changed features.  Now
+	 * The vendor-specific functions might have changed features.
-	 * we do "generic changes."
+	 * Now we do "generic changes."
 	 */
 
 	/* Filter out anything that depends on CPUID levels we don't have */
@@ -768,7 +803,7 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 
 	/* If the model name is still unset, do table lookup. */
 	if (!c->x86_model_id[0]) {
-		char *p;
+		const char *p;
 		p = table_lookup_model(c);
 		if (p)
 			strcpy(c->x86_model_id, p);
@@ -843,11 +878,11 @@ void __cpuinit identify_secondary_cpu(struct cpuinfo_x86 *c)
 }
 
 struct msr_range {
-	unsigned min;
+	unsigned	min;
-	unsigned max;
+	unsigned	max;
 };
 
-static struct msr_range msr_range_array[] __cpuinitdata = {
+static const struct msr_range msr_range_array[] __cpuinitconst = {
 	{ 0x00000000, 0x00000418},
 	{ 0xc0000000, 0xc000040b},
 	{ 0xc0010000, 0xc0010142},
@@ -856,14 +891,15 @@ static struct msr_range msr_range_array[] __cpuinitdata = {
 
 static void __cpuinit print_cpu_msr(void)
 {
+	unsigned index_min, index_max;
 	unsigned index;
 	u64 val;
 	int i;
-	unsigned index_min, index_max;
 
 	for (i = 0; i < ARRAY_SIZE(msr_range_array); i++) {
 		index_min = msr_range_array[i].min;
 		index_max = msr_range_array[i].max;
+
 		for (index = index_min; index < index_max; index++) {
 			if (rdmsrl_amd_safe(index, &val))
 				continue;
@@ -873,6 +909,7 @@ static void __cpuinit print_cpu_msr(void)
 }
 
 static int show_msr __cpuinitdata;
+
 static __init int setup_show_msr(char *arg)
 {
 	int num;
@@ -894,12 +931,14 @@ __setup("noclflush", setup_noclflush);
 
 void __cpuinit print_cpu_info(struct cpuinfo_x86 *c)
 {
-	char *vendor = NULL;
+	const char *vendor = NULL;
 
-	if (c->x86_vendor < X86_VENDOR_NUM)
+	if (c->x86_vendor < X86_VENDOR_NUM) {
 		vendor = this_cpu->c_vendor;
-	else if (c->cpuid_level >= 0)
+	} else {
-		vendor = c->x86_vendor_id;
+		if (c->cpuid_level >= 0)
+			vendor = c->x86_vendor_id;
+	}
 
 	if (vendor && !strstr(c->x86_model_id, vendor))
 		printk(KERN_CONT "%s ", vendor);
@@ -926,10 +965,12 @@ void __cpuinit print_cpu_info(struct cpuinfo_x86 *c)
 static __init int setup_disablecpuid(char *arg)
 {
 	int bit;
+
 	if (get_option(&arg, &bit) && bit < NCAPINTS*32)
 		setup_clear_cpu_cap(bit);
 	else
 		return 0;
+
 	return 1;
 }
 __setup("clearcpuid=", setup_disablecpuid);
@@ -939,6 +980,7 @@ struct desc_ptr idt_descr = { 256 * 16 - 1, (unsigned long) idt_table };
 
 DEFINE_PER_CPU_FIRST(union irq_stack_union,
 		     irq_stack_union) __aligned(PAGE_SIZE);
+
 DEFINE_PER_CPU(char *, irq_stack_ptr) =
 	init_per_cpu_var(irq_stack_union.irq_stack) + IRQ_STACK_SIZE - 64;
 
@@ -948,12 +990,21 @@ EXPORT_PER_CPU_SYMBOL(kernel_stack);
 
 DEFINE_PER_CPU(unsigned int, irq_count) = -1;
 
+/*
+ * Special IST stacks which the CPU switches to when it calls
+ * an IST-marked descriptor entry. Up to 7 stacks (hardware
+ * limit), all of them are 4K, except the debug stack which
+ * is 8K.
+ */
+static const unsigned int exception_stack_sizes[N_EXCEPTION_STACKS] = {
+	  [0 ... N_EXCEPTION_STACKS - 1]	= EXCEPTION_STKSZ,
+	  [DEBUG_STACK - 1]			= DEBUG_STKSZ
+};
+
 static DEFINE_PER_CPU_PAGE_ALIGNED(char, exception_stacks
 	[(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ + DEBUG_STKSZ])
 	__aligned(PAGE_SIZE);
 
-extern asmlinkage void ignore_sysret(void);
-
 /* May not be marked __init: used by software suspend */
 void syscall_init(void)
 {
@@ -983,7 +1034,7 @@ unsigned long kernel_eflags;
  */
 DEFINE_PER_CPU(struct orig_ist, orig_ist);
 
-#else	/* x86_64 */
+#else	/* CONFIG_X86_64 */
 
 #ifdef CONFIG_CC_STACKPROTECTOR
 DEFINE_PER_CPU(unsigned long, stack_canary);
@@ -995,9 +1046,26 @@ struct pt_regs * __cpuinit idle_regs(struct pt_regs *regs)
 	memset(regs, 0, sizeof(struct pt_regs));
 	regs->fs = __KERNEL_PERCPU;
 	regs->gs = __KERNEL_STACK_CANARY;
+
 	return regs;
 }
-#endif	/* x86_64 */
+#endif	/* CONFIG_X86_64 */
+
+/*
+ * Clear all 6 debug registers:
+ */
+static void clear_all_debug_regs(void)
+{
+	int i;
+
+	for (i = 0; i < 8; i++) {
+		/* Ignore db4, db5 */
+		if ((i == 4) || (i == 5))
+			continue;
+
+		set_debugreg(0, i);
+	}
+}
 
 /*
  * cpu_init() initializes state that is per-CPU. Some data is already
@@ -1007,15 +1075,20 @@ struct pt_regs * __cpuinit idle_regs(struct pt_regs *regs)
  * A lot of state is already set up in PDA init for 64 bit
  */
 #ifdef CONFIG_X86_64
+
 void __cpuinit cpu_init(void)
 {
-	int cpu = stack_smp_processor_id();
+	struct orig_ist *orig_ist;
-	struct tss_struct *t = &per_cpu(init_tss, cpu);
-	struct orig_ist *orig_ist = &per_cpu(orig_ist, cpu);
-	unsigned long v;
 	struct task_struct *me;
+	struct tss_struct *t;
+	unsigned long v;
+	int cpu;
 	int i;
 
+	cpu = stack_smp_processor_id();
+	t = &per_cpu(init_tss, cpu);
+	orig_ist = &per_cpu(orig_ist, cpu);
+
 #ifdef CONFIG_NUMA
 	if (cpu != 0 && percpu_read(node_number) == 0 &&
 	    cpu_to_node(cpu) != NUMA_NO_NODE)
@@ -1056,19 +1129,17 @@ void __cpuinit cpu_init(void)
 	 * set up and load the per-CPU TSS
 	 */
 	if (!orig_ist->ist[0]) {
-		static const unsigned int sizes[N_EXCEPTION_STACKS] = {
-		  [0 ... N_EXCEPTION_STACKS - 1] = EXCEPTION_STKSZ,
-		  [DEBUG_STACK - 1] = DEBUG_STKSZ
-		};
 		char *estacks = per_cpu(exception_stacks, cpu);
+
 		for (v = 0; v < N_EXCEPTION_STACKS; v++) {
-			estacks += sizes[v];
+			estacks += exception_stack_sizes[v];
 			orig_ist->ist[v] = t->x86_tss.ist[v] =
 					(unsigned long)estacks;
 		}
 	}
 
 	t->x86_tss.io_bitmap_base = offsetof(struct tss_struct, io_bitmap);
+
 	/*
 	 * <= is required because the CPU will access up to
 	 * 8 bits beyond the end of the IO permission bitmap.
@@ -1078,8 +1149,7 @@ void __cpuinit cpu_init(void)
 
 	atomic_inc(&init_mm.mm_count);
 	me->active_mm = &init_mm;
-	if (me->mm)
+	BUG_ON(me->mm);
-		BUG();
 	enter_lazy_tlb(&init_mm, me);
 
 	load_sp0(t, &current->thread);
@@ -1098,17 +1168,7 @@ void __cpuinit cpu_init(void)
 		arch_kgdb_ops.correct_hw_break();
 	else
 #endif
-	{
+		clear_all_debug_regs();
-		/*
-		 * Clear all 6 debug registers:
-		 */
-		set_debugreg(0UL, 0);
-		set_debugreg(0UL, 1);
-		set_debugreg(0UL, 2);
-		set_debugreg(0UL, 3);
-		set_debugreg(0UL, 6);
-		set_debugreg(0UL, 7);
-	}
 
 	fpu_init();
 
@@ -1129,7 +1189,8 @@ void __cpuinit cpu_init(void)
 
 	if (cpumask_test_and_set_cpu(cpu, cpu_initialized_mask)) {
 		printk(KERN_WARNING "CPU#%d already initialized!\n", cpu);
-		for (;;) local_irq_enable();
+		for (;;)
+			local_irq_enable();
 	}
 
 	printk(KERN_INFO "Initializing CPU#%d\n", cpu);
@@ -1145,8 +1206,7 @@ void __cpuinit cpu_init(void)
 	 */
 	atomic_inc(&init_mm.mm_count);
 	curr->active_mm = &init_mm;
-	if (curr->mm)
+	BUG_ON(curr->mm);
-		BUG();
 	enter_lazy_tlb(&init_mm, curr);
 
 	load_sp0(t, thread);
@@ -1159,13 +1219,7 @@ void __cpuinit cpu_init(void)
 	__set_tss_desc(cpu, GDT_ENTRY_DOUBLEFAULT_TSS, &doublefault_tss);
 #endif
 
-	/* Clear all 6 debug registers: */
+	clear_all_debug_regs();
-	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:
@@ -1185,6 +1239,4 @@ void __cpuinit cpu_init(void)
 
 	xsave_init();
 }
-
-
 #endif

arch/x86/kernel/cpu/cpu.h

@@ -3,33 +3,34 @@
 #define ARCH_X86_CPU_H
 
 struct cpu_model_info {
-	int vendor;
+	int		vendor;
-	int family;
+	int		family;
-	char *model_names[16];
+	const char	*model_names[16];
 };
 
 /* attempt to consolidate cpu attributes */
 struct cpu_dev {
-	char	* c_vendor;
+	const char	*c_vendor;
 
 	/* some have two possibilities for cpuid string */
-	char	* c_ident[2];	
+	const char	*c_ident[2];
 
 	struct		cpu_model_info c_models[4];
 
-	void            (*c_early_init)(struct cpuinfo_x86 *c);
+	void            (*c_early_init)(struct cpuinfo_x86 *);
-	void		(*c_init)(struct cpuinfo_x86 * c);
+	void		(*c_init)(struct cpuinfo_x86 *);
-	void		(*c_identify)(struct cpuinfo_x86 * c);
+	void		(*c_identify)(struct cpuinfo_x86 *);
-	unsigned int	(*c_size_cache)(struct cpuinfo_x86 * c, unsigned int size);
+	unsigned int	(*c_size_cache)(struct cpuinfo_x86 *, unsigned int);
-	int	c_x86_vendor;
+	int		c_x86_vendor;
 };
 
 #define cpu_dev_register(cpu_devX) \
-	static struct cpu_dev *__cpu_dev_##cpu_devX __used \
+	static const struct cpu_dev *const __cpu_dev_##cpu_devX __used \
 	__attribute__((__section__(".x86_cpu_dev.init"))) = \
 	&cpu_devX;
 
-extern struct cpu_dev *__x86_cpu_dev_start[], *__x86_cpu_dev_end[];
+extern const struct cpu_dev *const __x86_cpu_dev_start[],
+			    *const __x86_cpu_dev_end[];
 
 extern void display_cacheinfo(struct cpuinfo_x86 *c);
 

arch/x86/kernel/cpu/cpu_debug.c

@@ -0,0 +1,901 @@
+/*
+ * CPU x86 architecture debug code
+ *
+ * Copyright(C) 2009 Jaswinder Singh Rajput
+ *
+ * For licencing details see kernel-base/COPYING
+ */
+
+#include <linux/interrupt.h>
+#include <linux/compiler.h>
+#include <linux/seq_file.h>
+#include <linux/debugfs.h>
+#include <linux/kprobes.h>
+#include <linux/uaccess.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/percpu.h>
+#include <linux/signal.h>
+#include <linux/errno.h>
+#include <linux/sched.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/smp.h>
+
+#include <asm/cpu_debug.h>
+#include <asm/paravirt.h>
+#include <asm/system.h>
+#include <asm/traps.h>
+#include <asm/apic.h>
+#include <asm/desc.h>
+
+static DEFINE_PER_CPU(struct cpu_cpuX_base, cpu_arr[CPU_REG_ALL_BIT]);
+static DEFINE_PER_CPU(struct cpu_private *, priv_arr[MAX_CPU_FILES]);
+static DEFINE_PER_CPU(unsigned, cpu_modelflag);
+static DEFINE_PER_CPU(int, cpu_priv_count);
+static DEFINE_PER_CPU(unsigned, cpu_model);
+
+static DEFINE_MUTEX(cpu_debug_lock);
+
+static struct dentry *cpu_debugfs_dir;
+
+static struct cpu_debug_base cpu_base[] = {
+	{ "mc",		CPU_MC,		0	},
+	{ "monitor",	CPU_MONITOR,	0	},
+	{ "time",	CPU_TIME,	0	},
+	{ "pmc",	CPU_PMC,	1	},
+	{ "platform",	CPU_PLATFORM,	0	},
+	{ "apic",	CPU_APIC,	0	},
+	{ "poweron",	CPU_POWERON,	0	},
+	{ "control",	CPU_CONTROL,	0	},
+	{ "features",	CPU_FEATURES,	0	},
+	{ "lastbranch",	CPU_LBRANCH,	0	},
+	{ "bios",	CPU_BIOS,	0	},
+	{ "freq",	CPU_FREQ,	0	},
+	{ "mtrr",	CPU_MTRR,	0	},
+	{ "perf",	CPU_PERF,	0	},
+	{ "cache",	CPU_CACHE,	0	},
+	{ "sysenter",	CPU_SYSENTER,	0	},
+	{ "therm",	CPU_THERM,	0	},
+	{ "misc",	CPU_MISC,	0	},
+	{ "debug",	CPU_DEBUG,	0	},
+	{ "pat",	CPU_PAT,	0	},
+	{ "vmx",	CPU_VMX,	0	},
+	{ "call",	CPU_CALL,	0	},
+	{ "base",	CPU_BASE,	0	},
+	{ "ver",	CPU_VER,	0	},
+	{ "conf",	CPU_CONF,	0	},
+	{ "smm",	CPU_SMM,	0	},
+	{ "svm",	CPU_SVM,	0	},
+	{ "osvm",	CPU_OSVM,	0	},
+	{ "tss",	CPU_TSS,	0	},
+	{ "cr",		CPU_CR,		0	},
+	{ "dt",		CPU_DT,		0	},
+	{ "registers",	CPU_REG_ALL,	0	},
+};
+
+static struct cpu_file_base cpu_file[] = {
+	{ "index",	CPU_REG_ALL,	0	},
+	{ "value",	CPU_REG_ALL,	1	},
+};
+
+/* Intel Registers Range */
+static struct cpu_debug_range cpu_intel_range[] = {
+	{ 0x00000000, 0x00000001, CPU_MC,	CPU_INTEL_ALL		},
+	{ 0x00000006, 0x00000007, CPU_MONITOR,	CPU_CX_AT_XE		},
+	{ 0x00000010, 0x00000010, CPU_TIME,	CPU_INTEL_ALL		},
+	{ 0x00000011, 0x00000013, CPU_PMC,	CPU_INTEL_PENTIUM	},
+	{ 0x00000017, 0x00000017, CPU_PLATFORM,	CPU_PX_CX_AT_XE		},
+	{ 0x0000001B, 0x0000001B, CPU_APIC,	CPU_P6_CX_AT_XE		},
+
+	{ 0x0000002A, 0x0000002A, CPU_POWERON,	CPU_PX_CX_AT_XE		},
+	{ 0x0000002B, 0x0000002B, CPU_POWERON,	CPU_INTEL_XEON		},
+	{ 0x0000002C, 0x0000002C, CPU_FREQ,	CPU_INTEL_XEON		},
+	{ 0x0000003A, 0x0000003A, CPU_CONTROL,	CPU_CX_AT_XE		},
+
+	{ 0x00000040, 0x00000043, CPU_LBRANCH,	CPU_PM_CX_AT_XE		},
+	{ 0x00000044, 0x00000047, CPU_LBRANCH,	CPU_PM_CO_AT		},
+	{ 0x00000060, 0x00000063, CPU_LBRANCH,	CPU_C2_AT		},
+	{ 0x00000064, 0x00000067, CPU_LBRANCH,	CPU_INTEL_ATOM		},
+
+	{ 0x00000079, 0x00000079, CPU_BIOS,	CPU_P6_CX_AT_XE		},
+	{ 0x00000088, 0x0000008A, CPU_CACHE,	CPU_INTEL_P6		},
+	{ 0x0000008B, 0x0000008B, CPU_BIOS,	CPU_P6_CX_AT_XE		},
+	{ 0x0000009B, 0x0000009B, CPU_MONITOR,	CPU_INTEL_XEON		},
+
+	{ 0x000000C1, 0x000000C2, CPU_PMC,	CPU_P6_CX_AT		},
+	{ 0x000000CD, 0x000000CD, CPU_FREQ,	CPU_CX_AT		},
+	{ 0x000000E7, 0x000000E8, CPU_PERF,	CPU_CX_AT		},
+	{ 0x000000FE, 0x000000FE, CPU_MTRR,	CPU_P6_CX_XE		},
+
+	{ 0x00000116, 0x00000116, CPU_CACHE,	CPU_INTEL_P6		},
+	{ 0x00000118, 0x00000118, CPU_CACHE,	CPU_INTEL_P6		},
+	{ 0x00000119, 0x00000119, CPU_CACHE,	CPU_INTEL_PX		},
+	{ 0x0000011A, 0x0000011B, CPU_CACHE,	CPU_INTEL_P6		},
+	{ 0x0000011E, 0x0000011E, CPU_CACHE,	CPU_PX_CX_AT		},
+
+	{ 0x00000174, 0x00000176, CPU_SYSENTER,	CPU_P6_CX_AT_XE		},
+	{ 0x00000179, 0x0000017A, CPU_MC,	CPU_PX_CX_AT_XE		},
+	{ 0x0000017B, 0x0000017B, CPU_MC,	CPU_P6_XE		},
+	{ 0x00000186, 0x00000187, CPU_PMC,	CPU_P6_CX_AT		},
+	{ 0x00000198, 0x00000199, CPU_PERF,	CPU_PM_CX_AT_XE		},
+	{ 0x0000019A, 0x0000019A, CPU_TIME,	CPU_PM_CX_AT_XE		},
+	{ 0x0000019B, 0x0000019D, CPU_THERM,	CPU_PM_CX_AT_XE		},
+	{ 0x000001A0, 0x000001A0, CPU_MISC,	CPU_PM_CX_AT_XE		},
+
+	{ 0x000001C9, 0x000001C9, CPU_LBRANCH,	CPU_PM_CX_AT		},
+	{ 0x000001D7, 0x000001D8, CPU_LBRANCH,	CPU_INTEL_XEON		},
+	{ 0x000001D9, 0x000001D9, CPU_DEBUG,	CPU_CX_AT_XE		},
+	{ 0x000001DA, 0x000001DA, CPU_LBRANCH,	CPU_INTEL_XEON		},
+	{ 0x000001DB, 0x000001DB, CPU_LBRANCH,	CPU_P6_XE		},
+	{ 0x000001DC, 0x000001DC, CPU_LBRANCH,	CPU_INTEL_P6		},
+	{ 0x000001DD, 0x000001DE, CPU_LBRANCH,	CPU_PX_CX_AT_XE		},
+	{ 0x000001E0, 0x000001E0, CPU_LBRANCH,	CPU_INTEL_P6		},
+
+	{ 0x00000200, 0x0000020F, CPU_MTRR,	CPU_P6_CX_XE		},
+	{ 0x00000250, 0x00000250, CPU_MTRR,	CPU_P6_CX_XE		},
+	{ 0x00000258, 0x00000259, CPU_MTRR,	CPU_P6_CX_XE		},
+	{ 0x00000268, 0x0000026F, CPU_MTRR,	CPU_P6_CX_XE		},
+	{ 0x00000277, 0x00000277, CPU_PAT,	CPU_C2_AT_XE		},
+	{ 0x000002FF, 0x000002FF, CPU_MTRR,	CPU_P6_CX_XE		},
+
+	{ 0x00000300, 0x00000308, CPU_PMC,	CPU_INTEL_XEON		},
+	{ 0x00000309, 0x0000030B, CPU_PMC,	CPU_C2_AT_XE		},
+	{ 0x0000030C, 0x00000311, CPU_PMC,	CPU_INTEL_XEON		},
+	{ 0x00000345, 0x00000345, CPU_PMC,	CPU_C2_AT		},
+	{ 0x00000360, 0x00000371, CPU_PMC,	CPU_INTEL_XEON		},
+	{ 0x0000038D, 0x00000390, CPU_PMC,	CPU_C2_AT		},
+	{ 0x000003A0, 0x000003BE, CPU_PMC,	CPU_INTEL_XEON		},
+	{ 0x000003C0, 0x000003CD, CPU_PMC,	CPU_INTEL_XEON		},
+	{ 0x000003E0, 0x000003E1, CPU_PMC,	CPU_INTEL_XEON		},
+	{ 0x000003F0, 0x000003F0, CPU_PMC,	CPU_INTEL_XEON		},
+	{ 0x000003F1, 0x000003F1, CPU_PMC,	CPU_C2_AT_XE		},
+	{ 0x000003F2, 0x000003F2, CPU_PMC,	CPU_INTEL_XEON		},
+
+	{ 0x00000400, 0x00000402, CPU_MC,	CPU_PM_CX_AT_XE		},
+	{ 0x00000403, 0x00000403, CPU_MC,	CPU_INTEL_XEON		},
+	{ 0x00000404, 0x00000406, CPU_MC,	CPU_PM_CX_AT_XE		},
+	{ 0x00000407, 0x00000407, CPU_MC,	CPU_INTEL_XEON		},
+	{ 0x00000408, 0x0000040A, CPU_MC,	CPU_PM_CX_AT_XE		},
+	{ 0x0000040B, 0x0000040B, CPU_MC,	CPU_INTEL_XEON		},
+	{ 0x0000040C, 0x0000040E, CPU_MC,	CPU_PM_CX_XE		},
+	{ 0x0000040F, 0x0000040F, CPU_MC,	CPU_INTEL_XEON		},
+	{ 0x00000410, 0x00000412, CPU_MC,	CPU_PM_CX_AT_XE		},
+	{ 0x00000413, 0x00000417, CPU_MC,	CPU_CX_AT_XE		},
+	{ 0x00000480, 0x0000048B, CPU_VMX,	CPU_CX_AT_XE		},
+
+	{ 0x00000600, 0x00000600, CPU_DEBUG,	CPU_PM_CX_AT_XE		},
+	{ 0x00000680, 0x0000068F, CPU_LBRANCH,	CPU_INTEL_XEON		},
+	{ 0x000006C0, 0x000006CF, CPU_LBRANCH,	CPU_INTEL_XEON		},
+
+	{ 0x000107CC, 0x000107D3, CPU_PMC,	CPU_INTEL_XEON_MP	},
+
+	{ 0xC0000080, 0xC0000080, CPU_FEATURES,	CPU_INTEL_XEON		},
+	{ 0xC0000081, 0xC0000082, CPU_CALL,	CPU_INTEL_XEON		},
+	{ 0xC0000084, 0xC0000084, CPU_CALL,	CPU_INTEL_XEON		},
+	{ 0xC0000100, 0xC0000102, CPU_BASE,	CPU_INTEL_XEON		},
+};
+
+/* AMD Registers Range */
+static struct cpu_debug_range cpu_amd_range[] = {
+	{ 0x00000000, 0x00000001, CPU_MC,	CPU_K10_PLUS,		},
+	{ 0x00000010, 0x00000010, CPU_TIME,	CPU_K8_PLUS,		},
+	{ 0x0000001B, 0x0000001B, CPU_APIC,	CPU_K8_PLUS,		},
+	{ 0x0000002A, 0x0000002A, CPU_POWERON,	CPU_K7_PLUS		},
+	{ 0x0000008B, 0x0000008B, CPU_VER,	CPU_K8_PLUS		},
+	{ 0x000000FE, 0x000000FE, CPU_MTRR,	CPU_K8_PLUS,		},
+
+	{ 0x00000174, 0x00000176, CPU_SYSENTER,	CPU_K8_PLUS,		},
+	{ 0x00000179, 0x0000017B, CPU_MC,	CPU_K8_PLUS,		},
+	{ 0x000001D9, 0x000001D9, CPU_DEBUG,	CPU_K8_PLUS,		},
+	{ 0x000001DB, 0x000001DE, CPU_LBRANCH,	CPU_K8_PLUS,		},
+
+	{ 0x00000200, 0x0000020F, CPU_MTRR,	CPU_K8_PLUS,		},
+	{ 0x00000250, 0x00000250, CPU_MTRR,	CPU_K8_PLUS,		},
+	{ 0x00000258, 0x00000259, CPU_MTRR,	CPU_K8_PLUS,		},
+	{ 0x00000268, 0x0000026F, CPU_MTRR,	CPU_K8_PLUS,		},
+	{ 0x00000277, 0x00000277, CPU_PAT,	CPU_K8_PLUS,		},
+	{ 0x000002FF, 0x000002FF, CPU_MTRR,	CPU_K8_PLUS,		},
+
+	{ 0x00000400, 0x00000413, CPU_MC,	CPU_K8_PLUS,		},
+
+	{ 0xC0000080, 0xC0000080, CPU_FEATURES,	CPU_AMD_ALL,		},
+	{ 0xC0000081, 0xC0000084, CPU_CALL,	CPU_K8_PLUS,		},
+	{ 0xC0000100, 0xC0000102, CPU_BASE,	CPU_K8_PLUS,		},
+	{ 0xC0000103, 0xC0000103, CPU_TIME,	CPU_K10_PLUS,		},
+
+	{ 0xC0010000, 0xC0010007, CPU_PMC,	CPU_K8_PLUS,		},
+	{ 0xC0010010, 0xC0010010, CPU_CONF,	CPU_K7_PLUS,		},
+	{ 0xC0010015, 0xC0010015, CPU_CONF,	CPU_K7_PLUS,		},
+	{ 0xC0010016, 0xC001001A, CPU_MTRR,	CPU_K8_PLUS,		},
+	{ 0xC001001D, 0xC001001D, CPU_MTRR,	CPU_K8_PLUS,		},
+	{ 0xC001001F, 0xC001001F, CPU_CONF,	CPU_K8_PLUS,		},
+	{ 0xC0010030, 0xC0010035, CPU_BIOS,	CPU_K8_PLUS,		},
+	{ 0xC0010044, 0xC0010048, CPU_MC,	CPU_K8_PLUS,		},
+	{ 0xC0010050, 0xC0010056, CPU_SMM,	CPU_K0F_PLUS,		},
+	{ 0xC0010058, 0xC0010058, CPU_CONF,	CPU_K10_PLUS,		},
+	{ 0xC0010060, 0xC0010060, CPU_CACHE,	CPU_AMD_11,		},
+	{ 0xC0010061, 0xC0010068, CPU_SMM,	CPU_K10_PLUS,		},
+	{ 0xC0010069, 0xC001006B, CPU_SMM,	CPU_AMD_11,		},
+	{ 0xC0010070, 0xC0010071, CPU_SMM,	CPU_K10_PLUS,		},
+	{ 0xC0010111, 0xC0010113, CPU_SMM,	CPU_K8_PLUS,		},
+	{ 0xC0010114, 0xC0010118, CPU_SVM,	CPU_K10_PLUS,		},
+	{ 0xC0010140, 0xC0010141, CPU_OSVM,	CPU_K10_PLUS,		},
+	{ 0xC0011022, 0xC0011023, CPU_CONF,	CPU_K10_PLUS,		},
+};
+
+
+/* Intel */
+static int get_intel_modelflag(unsigned model)
+{
+	int flag;
+
+	switch (model) {
+	case 0x0501:
+	case 0x0502:
+	case 0x0504:
+		flag = CPU_INTEL_PENTIUM;
+		break;
+	case 0x0601:
+	case 0x0603:
+	case 0x0605:
+	case 0x0607:
+	case 0x0608:
+	case 0x060A:
+	case 0x060B:
+		flag = CPU_INTEL_P6;
+		break;
+	case 0x0609:
+	case 0x060D:
+		flag = CPU_INTEL_PENTIUM_M;
+		break;
+	case 0x060E:
+		flag = CPU_INTEL_CORE;
+		break;
+	case 0x060F:
+	case 0x0617:
+		flag = CPU_INTEL_CORE2;
+		break;
+	case 0x061C:
+		flag = CPU_INTEL_ATOM;
+		break;
+	case 0x0F00:
+	case 0x0F01:
+	case 0x0F02:
+	case 0x0F03:
+	case 0x0F04:
+		flag = CPU_INTEL_XEON_P4;
+		break;
+	case 0x0F06:
+		flag = CPU_INTEL_XEON_MP;
+		break;
+	default:
+		flag = CPU_NONE;
+		break;
+	}
+
+	return flag;
+}
+
+/* AMD */
+static int get_amd_modelflag(unsigned model)
+{
+	int flag;
+
+	switch (model >> 8) {
+	case 0x6:
+		flag = CPU_AMD_K6;
+		break;
+	case 0x7:
+		flag = CPU_AMD_K7;
+		break;
+	case 0x8:
+		flag = CPU_AMD_K8;
+		break;
+	case 0xf:
+		flag = CPU_AMD_0F;
+		break;
+	case 0x10:
+		flag = CPU_AMD_10;
+		break;
+	case 0x11:
+		flag = CPU_AMD_11;
+		break;
+	default:
+		flag = CPU_NONE;
+		break;
+	}
+
+	return flag;
+}
+
+static int get_cpu_modelflag(unsigned cpu)
+{
+	int flag;
+
+	flag = per_cpu(cpu_model, cpu);
+
+	switch (flag >> 16) {
+	case X86_VENDOR_INTEL:
+		flag = get_intel_modelflag(flag);
+		break;
+	case X86_VENDOR_AMD:
+		flag = get_amd_modelflag(flag & 0xffff);
+		break;
+	default:
+		flag = CPU_NONE;
+		break;
+	}
+
+	return flag;
+}
+
+static int get_cpu_range_count(unsigned cpu)
+{
+	int index;
+
+	switch (per_cpu(cpu_model, cpu) >> 16) {
+	case X86_VENDOR_INTEL:
+		index = ARRAY_SIZE(cpu_intel_range);
+		break;
+	case X86_VENDOR_AMD:
+		index = ARRAY_SIZE(cpu_amd_range);
+		break;
+	default:
+		index = 0;
+		break;
+	}
+
+	return index;
+}
+
+static int is_typeflag_valid(unsigned cpu, unsigned flag)
+{
+	unsigned vendor, modelflag;
+	int i, index;
+
+	/* Standard Registers should be always valid */
+	if (flag >= CPU_TSS)
+		return 1;
+
+	modelflag = per_cpu(cpu_modelflag, cpu);
+	vendor = per_cpu(cpu_model, cpu) >> 16;
+	index = get_cpu_range_count(cpu);
+
+	for (i = 0; i < index; i++) {
+		switch (vendor) {
+		case X86_VENDOR_INTEL:
+			if ((cpu_intel_range[i].model & modelflag) &&
+			    (cpu_intel_range[i].flag & flag))
+				return 1;
+			break;
+		case X86_VENDOR_AMD:
+			if ((cpu_amd_range[i].model & modelflag) &&
+			    (cpu_amd_range[i].flag & flag))
+				return 1;
+			break;
+		}
+	}
+
+	/* Invalid */
+	return 0;
+}
+
+static unsigned get_cpu_range(unsigned cpu, unsigned *min, unsigned *max,
+			      int index, unsigned flag)
+{
+	unsigned modelflag;
+
+	modelflag = per_cpu(cpu_modelflag, cpu);
+	*max = 0;
+	switch (per_cpu(cpu_model, cpu) >> 16) {
+	case X86_VENDOR_INTEL:
+		if ((cpu_intel_range[index].model & modelflag) &&
+		    (cpu_intel_range[index].flag & flag)) {
+			*min = cpu_intel_range[index].min;
+			*max = cpu_intel_range[index].max;
+		}
+		break;
+	case X86_VENDOR_AMD:
+		if ((cpu_amd_range[index].model & modelflag) &&
+		    (cpu_amd_range[index].flag & flag)) {
+			*min = cpu_amd_range[index].min;
+			*max = cpu_amd_range[index].max;
+		}
+		break;
+	}
+
+	return *max;
+}
+
+/* This function can also be called with seq = NULL for printk */
+static void print_cpu_data(struct seq_file *seq, unsigned type,
+			   u32 low, u32 high)
+{
+	struct cpu_private *priv;
+	u64 val = high;
+
+	if (seq) {
+		priv = seq->private;
+		if (priv->file) {
+			val = (val << 32) | low;
+			seq_printf(seq, "0x%llx\n", val);
+		} else
+			seq_printf(seq, " %08x: %08x_%08x\n",
+				   type, high, low);
+	} else
+		printk(KERN_INFO " %08x: %08x_%08x\n", type, high, low);
+}
+
+/* This function can also be called with seq = NULL for printk */
+static void print_msr(struct seq_file *seq, unsigned cpu, unsigned flag)
+{
+	unsigned msr, msr_min, msr_max;
+	struct cpu_private *priv;
+	u32 low, high;
+	int i, range;
+
+	if (seq) {
+		priv = seq->private;
+		if (priv->file) {
+			if (!rdmsr_safe_on_cpu(priv->cpu, priv->reg,
+					       &low, &high))
+				print_cpu_data(seq, priv->reg, low, high);
+			return;
+		}
+	}
+
+	range = get_cpu_range_count(cpu);
+
+	for (i = 0; i < range; i++) {
+		if (!get_cpu_range(cpu, &msr_min, &msr_max, i, flag))
+			continue;
+
+		for (msr = msr_min; msr <= msr_max; msr++) {
+			if (rdmsr_safe_on_cpu(cpu, msr, &low, &high))
+				continue;
+			print_cpu_data(seq, msr, low, high);
+		}
+	}
+}
+
+static void print_tss(void *arg)
+{
+	struct pt_regs *regs = task_pt_regs(current);
+	struct seq_file *seq = arg;
+	unsigned int seg;
+
+	seq_printf(seq, " RAX\t: %016lx\n", regs->ax);
+	seq_printf(seq, " RBX\t: %016lx\n", regs->bx);
+	seq_printf(seq, " RCX\t: %016lx\n", regs->cx);
+	seq_printf(seq, " RDX\t: %016lx\n", regs->dx);
+
+	seq_printf(seq, " RSI\t: %016lx\n", regs->si);
+	seq_printf(seq, " RDI\t: %016lx\n", regs->di);
+	seq_printf(seq, " RBP\t: %016lx\n", regs->bp);
+	seq_printf(seq, " ESP\t: %016lx\n", regs->sp);
+
+#ifdef CONFIG_X86_64
+	seq_printf(seq, " R08\t: %016lx\n", regs->r8);
+	seq_printf(seq, " R09\t: %016lx\n", regs->r9);
+	seq_printf(seq, " R10\t: %016lx\n", regs->r10);
+	seq_printf(seq, " R11\t: %016lx\n", regs->r11);
+	seq_printf(seq, " R12\t: %016lx\n", regs->r12);
+	seq_printf(seq, " R13\t: %016lx\n", regs->r13);
+	seq_printf(seq, " R14\t: %016lx\n", regs->r14);
+	seq_printf(seq, " R15\t: %016lx\n", regs->r15);
+#endif
+
+	asm("movl %%cs,%0" : "=r" (seg));
+	seq_printf(seq, " CS\t:             %04x\n", seg);
+	asm("movl %%ds,%0" : "=r" (seg));
+	seq_printf(seq, " DS\t:             %04x\n", seg);
+	seq_printf(seq, " SS\t:             %04lx\n", regs->ss & 0xffff);
+	asm("movl %%es,%0" : "=r" (seg));
+	seq_printf(seq, " ES\t:             %04x\n", seg);
+	asm("movl %%fs,%0" : "=r" (seg));
+	seq_printf(seq, " FS\t:             %04x\n", seg);
+	asm("movl %%gs,%0" : "=r" (seg));
+	seq_printf(seq, " GS\t:             %04x\n", seg);
+
+	seq_printf(seq, " EFLAGS\t: %016lx\n", regs->flags);
+
+	seq_printf(seq, " EIP\t: %016lx\n", regs->ip);
+}
+
+static void print_cr(void *arg)
+{
+	struct seq_file *seq = arg;
+
+	seq_printf(seq, " cr0\t: %016lx\n", read_cr0());
+	seq_printf(seq, " cr2\t: %016lx\n", read_cr2());
+	seq_printf(seq, " cr3\t: %016lx\n", read_cr3());
+	seq_printf(seq, " cr4\t: %016lx\n", read_cr4_safe());
+#ifdef CONFIG_X86_64
+	seq_printf(seq, " cr8\t: %016lx\n", read_cr8());
+#endif
+}
+
+static void print_desc_ptr(char *str, struct seq_file *seq, struct desc_ptr dt)
+{
+	seq_printf(seq, " %s\t: %016llx\n", str, (u64)(dt.address | dt.size));
+}
+
+static void print_dt(void *seq)
+{
+	struct desc_ptr dt;
+	unsigned long ldt;
+
+	/* IDT */
+	store_idt((struct desc_ptr *)&dt);
+	print_desc_ptr("IDT", seq, dt);
+
+	/* GDT */
+	store_gdt((struct desc_ptr *)&dt);
+	print_desc_ptr("GDT", seq, dt);
+
+	/* LDT */
+	store_ldt(ldt);
+	seq_printf(seq, " LDT\t: %016lx\n", ldt);
+
+	/* TR */
+	store_tr(ldt);
+	seq_printf(seq, " TR\t: %016lx\n", ldt);
+}
+
+static void print_dr(void *arg)
+{
+	struct seq_file *seq = arg;
+	unsigned long dr;
+	int i;
+
+	for (i = 0; i < 8; i++) {
+		/* Ignore db4, db5 */
+		if ((i == 4) || (i == 5))
+			continue;
+		get_debugreg(dr, i);
+		seq_printf(seq, " dr%d\t: %016lx\n", i, dr);
+	}
+
+	seq_printf(seq, "\n MSR\t:\n");
+}
+
+static void print_apic(void *arg)
+{
+	struct seq_file *seq = arg;
+
+#ifdef CONFIG_X86_LOCAL_APIC
+	seq_printf(seq, " LAPIC\t:\n");
+	seq_printf(seq, " ID\t\t: %08x\n",  apic_read(APIC_ID) >> 24);
+	seq_printf(seq, " LVR\t\t: %08x\n",  apic_read(APIC_LVR));
+	seq_printf(seq, " TASKPRI\t: %08x\n",  apic_read(APIC_TASKPRI));
+	seq_printf(seq, " ARBPRI\t\t: %08x\n",  apic_read(APIC_ARBPRI));
+	seq_printf(seq, " PROCPRI\t: %08x\n",  apic_read(APIC_PROCPRI));
+	seq_printf(seq, " LDR\t\t: %08x\n",  apic_read(APIC_LDR));
+	seq_printf(seq, " DFR\t\t: %08x\n",  apic_read(APIC_DFR));
+	seq_printf(seq, " SPIV\t\t: %08x\n",  apic_read(APIC_SPIV));
+	seq_printf(seq, " ISR\t\t: %08x\n",  apic_read(APIC_ISR));
+	seq_printf(seq, " ESR\t\t: %08x\n",  apic_read(APIC_ESR));
+	seq_printf(seq, " ICR\t\t: %08x\n",  apic_read(APIC_ICR));
+	seq_printf(seq, " ICR2\t\t: %08x\n",  apic_read(APIC_ICR2));
+	seq_printf(seq, " LVTT\t\t: %08x\n",  apic_read(APIC_LVTT));
+	seq_printf(seq, " LVTTHMR\t: %08x\n",  apic_read(APIC_LVTTHMR));
+	seq_printf(seq, " LVTPC\t\t: %08x\n",  apic_read(APIC_LVTPC));
+	seq_printf(seq, " LVT0\t\t: %08x\n",  apic_read(APIC_LVT0));
+	seq_printf(seq, " LVT1\t\t: %08x\n",  apic_read(APIC_LVT1));
+	seq_printf(seq, " LVTERR\t\t: %08x\n",  apic_read(APIC_LVTERR));
+	seq_printf(seq, " TMICT\t\t: %08x\n",  apic_read(APIC_TMICT));
+	seq_printf(seq, " TMCCT\t\t: %08x\n",  apic_read(APIC_TMCCT));
+	seq_printf(seq, " TDCR\t\t: %08x\n",  apic_read(APIC_TDCR));
+#endif /* CONFIG_X86_LOCAL_APIC */
+
+	seq_printf(seq, "\n MSR\t:\n");
+}
+
+static int cpu_seq_show(struct seq_file *seq, void *v)
+{
+	struct cpu_private *priv = seq->private;
+
+	if (priv == NULL)
+		return -EINVAL;
+
+	switch (cpu_base[priv->type].flag) {
+	case CPU_TSS:
+		smp_call_function_single(priv->cpu, print_tss, seq, 1);
+		break;
+	case CPU_CR:
+		smp_call_function_single(priv->cpu, print_cr, seq, 1);
+		break;
+	case CPU_DT:
+		smp_call_function_single(priv->cpu, print_dt, seq, 1);
+		break;
+	case CPU_DEBUG:
+		if (priv->file == CPU_INDEX_BIT)
+			smp_call_function_single(priv->cpu, print_dr, seq, 1);
+		print_msr(seq, priv->cpu, cpu_base[priv->type].flag);
+		break;
+	case CPU_APIC:
+		if (priv->file == CPU_INDEX_BIT)
+			smp_call_function_single(priv->cpu, print_apic, seq, 1);
+		print_msr(seq, priv->cpu, cpu_base[priv->type].flag);
+		break;
+
+	default:
+		print_msr(seq, priv->cpu, cpu_base[priv->type].flag);
+		break;
+	}
+	seq_printf(seq, "\n");
+
+	return 0;
+}
+
+static void *cpu_seq_start(struct seq_file *seq, loff_t *pos)
+{
+	if (*pos == 0) /* One time is enough ;-) */
+		return seq;
+
+	return NULL;
+}
+
+static void *cpu_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+	(*pos)++;
+
+	return cpu_seq_start(seq, pos);
+}
+
+static void cpu_seq_stop(struct seq_file *seq, void *v)
+{
+}
+
+static const struct seq_operations cpu_seq_ops = {
+	.start		= cpu_seq_start,
+	.next		= cpu_seq_next,
+	.stop		= cpu_seq_stop,
+	.show		= cpu_seq_show,
+};
+
+static int cpu_seq_open(struct inode *inode, struct file *file)
+{
+	struct cpu_private *priv = inode->i_private;
+	struct seq_file *seq;
+	int err;
+
+	err = seq_open(file, &cpu_seq_ops);
+	if (!err) {
+		seq = file->private_data;
+		seq->private = priv;
+	}
+
+	return err;
+}
+
+static int write_msr(struct cpu_private *priv, u64 val)
+{
+	u32 low, high;
+
+	high = (val >> 32) & 0xffffffff;
+	low = val & 0xffffffff;
+
+	if (!wrmsr_safe_on_cpu(priv->cpu, priv->reg, low, high))
+		return 0;
+
+	return -EPERM;
+}
+
+static int write_cpu_register(struct cpu_private *priv, const char *buf)
+{
+	int ret = -EPERM;
+	u64 val;
+
+	ret = strict_strtoull(buf, 0, &val);
+	if (ret < 0)
+		return ret;
+
+	/* Supporting only MSRs */
+	if (priv->type < CPU_TSS_BIT)
+		return write_msr(priv, val);
+
+	return ret;
+}
+
+static ssize_t cpu_write(struct file *file, const char __user *ubuf,
+			     size_t count, loff_t *off)
+{
+	struct seq_file *seq = file->private_data;
+	struct cpu_private *priv = seq->private;
+	char buf[19];
+
+	if ((priv == NULL) || (count >= sizeof(buf)))
+		return -EINVAL;
+
+	if (copy_from_user(&buf, ubuf, count))
+		return -EFAULT;
+
+	buf[count] = 0;
+
+	if ((cpu_base[priv->type].write) && (cpu_file[priv->file].write))
+		if (!write_cpu_register(priv, buf))
+			return count;
+
+	return -EACCES;
+}
+
+static const struct file_operations cpu_fops = {
+	.owner		= THIS_MODULE,
+	.open		= cpu_seq_open,
+	.read		= seq_read,
+	.write		= cpu_write,
+	.llseek		= seq_lseek,
+	.release	= seq_release,
+};
+
+static int cpu_create_file(unsigned cpu, unsigned type, unsigned reg,
+			   unsigned file, struct dentry *dentry)
+{
+	struct cpu_private *priv = NULL;
+
+	/* Already intialized */
+	if (file == CPU_INDEX_BIT)
+		if (per_cpu(cpu_arr[type].init, cpu))
+			return 0;
+
+	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+	if (priv == NULL)
+		return -ENOMEM;
+
+	priv->cpu = cpu;
+	priv->type = type;
+	priv->reg = reg;
+	priv->file = file;
+	mutex_lock(&cpu_debug_lock);
+	per_cpu(priv_arr[type], cpu) = priv;
+	per_cpu(cpu_priv_count, cpu)++;
+	mutex_unlock(&cpu_debug_lock);
+
+	if (file)
+		debugfs_create_file(cpu_file[file].name, S_IRUGO,
+				    dentry, (void *)priv, &cpu_fops);
+	else {
+		debugfs_create_file(cpu_base[type].name, S_IRUGO,
+				    per_cpu(cpu_arr[type].dentry, cpu),
+				    (void *)priv, &cpu_fops);
+		mutex_lock(&cpu_debug_lock);
+		per_cpu(cpu_arr[type].init, cpu) = 1;
+		mutex_unlock(&cpu_debug_lock);
+	}
+
+	return 0;
+}
+
+static int cpu_init_regfiles(unsigned cpu, unsigned int type, unsigned reg,
+			     struct dentry *dentry)
+{
+	unsigned file;
+	int err = 0;
+
+	for (file = 0; file <  ARRAY_SIZE(cpu_file); file++) {
+		err = cpu_create_file(cpu, type, reg, file, dentry);
+		if (err)
+			return err;
+	}
+
+	return err;
+}
+
+static int cpu_init_msr(unsigned cpu, unsigned type, struct dentry *dentry)
+{
+	struct dentry *cpu_dentry = NULL;
+	unsigned reg, reg_min, reg_max;
+	int i, range, err = 0;
+	char reg_dir[12];
+	u32 low, high;
+
+	range = get_cpu_range_count(cpu);
+
+	for (i = 0; i < range; i++) {
+		if (!get_cpu_range(cpu, &reg_min, &reg_max, i,
+				   cpu_base[type].flag))
+			continue;
+
+		for (reg = reg_min; reg <= reg_max; reg++) {
+			if (rdmsr_safe_on_cpu(cpu, reg, &low, &high))
+				continue;
+
+			sprintf(reg_dir, "0x%x", reg);
+			cpu_dentry = debugfs_create_dir(reg_dir, dentry);
+			err = cpu_init_regfiles(cpu, type, reg, cpu_dentry);
+			if (err)
+				return err;
+		}
+	}
+
+	return err;
+}
+
+static int cpu_init_allreg(unsigned cpu, struct dentry *dentry)
+{
+	struct dentry *cpu_dentry = NULL;
+	unsigned type;
+	int err = 0;
+
+	for (type = 0; type <  ARRAY_SIZE(cpu_base) - 1; type++) {
+		if (!is_typeflag_valid(cpu, cpu_base[type].flag))
+			continue;
+		cpu_dentry = debugfs_create_dir(cpu_base[type].name, dentry);
+		per_cpu(cpu_arr[type].dentry, cpu) = cpu_dentry;
+
+		if (type < CPU_TSS_BIT)
+			err = cpu_init_msr(cpu, type, cpu_dentry);
+		else
+			err = cpu_create_file(cpu, type, 0, CPU_INDEX_BIT,
+					      cpu_dentry);
+		if (err)
+			return err;
+	}
+
+	return err;
+}
+
+static int cpu_init_cpu(void)
+{
+	struct dentry *cpu_dentry = NULL;
+	struct cpuinfo_x86 *cpui;
+	char cpu_dir[12];
+	unsigned cpu;
+	int err = 0;
+
+	for (cpu = 0; cpu < nr_cpu_ids; cpu++) {
+		cpui = &cpu_data(cpu);
+		if (!cpu_has(cpui, X86_FEATURE_MSR))
+			continue;
+		per_cpu(cpu_model, cpu) = ((cpui->x86_vendor << 16) |
+					   (cpui->x86 << 8) |
+					   (cpui->x86_model));
+		per_cpu(cpu_modelflag, cpu) = get_cpu_modelflag(cpu);
+
+		sprintf(cpu_dir, "cpu%d", cpu);
+		cpu_dentry = debugfs_create_dir(cpu_dir, cpu_debugfs_dir);
+		err = cpu_init_allreg(cpu, cpu_dentry);
+
+		pr_info("cpu%d(%d) debug files %d\n",
+			cpu, nr_cpu_ids, per_cpu(cpu_priv_count, cpu));
+		if (per_cpu(cpu_priv_count, cpu) > MAX_CPU_FILES) {
+			pr_err("Register files count %d exceeds limit %d\n",
+				per_cpu(cpu_priv_count, cpu), MAX_CPU_FILES);
+			per_cpu(cpu_priv_count, cpu) = MAX_CPU_FILES;
+			err = -ENFILE;
+		}
+		if (err)
+			return err;
+	}
+
+	return err;
+}
+
+static int __init cpu_debug_init(void)
+{
+	cpu_debugfs_dir = debugfs_create_dir("cpu", arch_debugfs_dir);
+
+	return cpu_init_cpu();
+}
+
+static void __exit cpu_debug_exit(void)
+{
+	int i, cpu;
+
+	if (cpu_debugfs_dir)
+		debugfs_remove_recursive(cpu_debugfs_dir);
+
+	for (cpu = 0; cpu <  nr_cpu_ids; cpu++)
+		for (i = 0; i < per_cpu(cpu_priv_count, cpu); i++)
+			kfree(per_cpu(priv_arr[i], cpu));
+}
+
+module_init(cpu_debug_init);
+module_exit(cpu_debug_exit);
+
+MODULE_AUTHOR("Jaswinder Singh Rajput");
+MODULE_DESCRIPTION("CPU Debug module");
+MODULE_LICENSE("GPL");

arch/x86/kernel/cpu/cyrix.c

@@ -61,23 +61,23 @@ static void __cpuinit do_cyrix_devid(unsigned char *dir0, unsigned char *dir1)
  */
 static unsigned char Cx86_dir0_msb __cpuinitdata = 0;
 
-static char Cx86_model[][9] __cpuinitdata = {
+static const char __cpuinitconst Cx86_model[][9] = {
 	"Cx486", "Cx486", "5x86 ", "6x86", "MediaGX ", "6x86MX ",
 	"M II ", "Unknown"
 };
-static char Cx486_name[][5] __cpuinitdata = {
+static const char __cpuinitconst Cx486_name[][5] = {
 	"SLC", "DLC", "SLC2", "DLC2", "SRx", "DRx",
 	"SRx2", "DRx2"
 };
-static char Cx486S_name[][4] __cpuinitdata = {
+static const char __cpuinitconst Cx486S_name[][4] = {
 	"S", "S2", "Se", "S2e"
 };
-static char Cx486D_name[][4] __cpuinitdata = {
+static const char __cpuinitconst Cx486D_name[][4] = {
 	"DX", "DX2", "?", "?", "?", "DX4"
 };
 static char Cx86_cb[] __cpuinitdata = "?.5x Core/Bus Clock";
-static char cyrix_model_mult1[] __cpuinitdata = "12??43";
+static const char __cpuinitconst cyrix_model_mult1[] = "12??43";
-static char cyrix_model_mult2[] __cpuinitdata = "12233445";
+static const char __cpuinitconst cyrix_model_mult2[] = "12233445";
 
 /*
  * Reset the slow-loop (SLOP) bit on the 686(L) which is set by some old
@@ -435,7 +435,7 @@ static void __cpuinit cyrix_identify(struct cpuinfo_x86 *c)
 	}
 }
 
-static struct cpu_dev cyrix_cpu_dev __cpuinitdata = {
+static const struct cpu_dev __cpuinitconst cyrix_cpu_dev = {
 	.c_vendor	= "Cyrix",
 	.c_ident	= { "CyrixInstead" },
 	.c_early_init	= early_init_cyrix,
@@ -446,7 +446,7 @@ static struct cpu_dev cyrix_cpu_dev __cpuinitdata = {
 
 cpu_dev_register(cyrix_cpu_dev);
 
-static struct cpu_dev nsc_cpu_dev __cpuinitdata = {
+static const struct cpu_dev __cpuinitconst nsc_cpu_dev = {
 	.c_vendor	= "NSC",
 	.c_ident	= { "Geode by NSC" },
 	.c_init		= init_nsc,

arch/x86/kernel/cpu/intel.c

@@ -14,6 +14,7 @@
 #include <asm/uaccess.h>
 #include <asm/ds.h>
 #include <asm/bugs.h>
+#include <asm/cpu.h>
 
 #ifdef CONFIG_X86_64
 #include <asm/topology.h>
@@ -54,6 +55,11 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c)
 		c->x86_cache_alignment = 128;
 #endif
 
+	/* CPUID workaround for 0F33/0F34 CPU */
+	if (c->x86 == 0xF && c->x86_model == 0x3
+	    && (c->x86_mask == 0x3 || c->x86_mask == 0x4))
+		c->x86_phys_bits = 36;
+
 	/*
 	 * c->x86_power is 8000_0007 edx. Bit 8 is TSC runs at constant rate
 	 * with P/T states and does not stop in deep C-states.
@@ -116,6 +122,28 @@ static void __cpuinit trap_init_f00f_bug(void)
 }
 #endif
 
+static void __cpuinit intel_smp_check(struct cpuinfo_x86 *c)
+{
+#ifdef CONFIG_SMP
+	/* calling is from identify_secondary_cpu() ? */
+	if (c->cpu_index == boot_cpu_id)
+		return;
+
+	/*
+	 * Mask B, Pentium, but not Pentium MMX
+	 */
+	if (c->x86 == 5 &&
+	    c->x86_mask >= 1 && c->x86_mask <= 4 &&
+	    c->x86_model <= 3) {
+		/*
+		 * Remember we have B step Pentia with bugs
+		 */
+		WARN_ONCE(1, "WARNING: SMP operation may be unreliable"
+				    "with B stepping processors.\n");
+	}
+#endif
+}
+
 static void __cpuinit intel_workarounds(struct cpuinfo_x86 *c)
 {
 	unsigned long lo, hi;
@@ -192,6 +220,8 @@ static void __cpuinit intel_workarounds(struct cpuinfo_x86 *c)
 #ifdef CONFIG_X86_NUMAQ
 	numaq_tsc_disable();
 #endif
+
+	intel_smp_check(c);
 }
 #else
 static void __cpuinit intel_workarounds(struct cpuinfo_x86 *c)
@@ -391,7 +421,7 @@ static unsigned int __cpuinit intel_size_cache(struct cpuinfo_x86 *c, unsigned i
 }
 #endif
 
-static struct cpu_dev intel_cpu_dev __cpuinitdata = {
+static const struct cpu_dev __cpuinitconst intel_cpu_dev = {
 	.c_vendor	= "Intel",
 	.c_ident	= { "GenuineIntel" },
 #ifdef CONFIG_X86_32

arch/x86/kernel/cpu/intel_cacheinfo.c

@@ -32,7 +32,7 @@ struct _cache_table
 };
 
 /* all the cache descriptor types we care about (no TLB or trace cache entries) */
-static struct _cache_table cache_table[] __cpuinitdata =
+static const struct _cache_table __cpuinitconst cache_table[] =
 {
 	{ 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 */
@@ -206,15 +206,15 @@ union l3_cache {
 	unsigned val;
 };
 
-static unsigned short assocs[] __cpuinitdata = {
+static const unsigned short __cpuinitconst assocs[] = {
 	[1] = 1, [2] = 2, [4] = 4, [6] = 8,
 	[8] = 16, [0xa] = 32, [0xb] = 48,
 	[0xc] = 64,
 	[0xf] = 0xffff // ??
 };
 
-static unsigned char levels[] __cpuinitdata = { 1, 1, 2, 3 };
+static const unsigned char __cpuinitconst levels[] = { 1, 1, 2, 3 };
-static unsigned char types[] __cpuinitdata = { 1, 2, 3, 3 };
+static const unsigned char __cpuinitconst types[] = { 1, 2, 3, 3 };
 
 static void __cpuinit
 amd_cpuid4(int leaf, union _cpuid4_leaf_eax *eax,

arch/x86/kernel/cpu/mcheck/Makefile

@@ -4,3 +4,4 @@ obj-$(CONFIG_X86_32)		+= k7.o p4.o p5.o p6.o winchip.o
 obj-$(CONFIG_X86_MCE_INTEL)	+= mce_intel_64.o
 obj-$(CONFIG_X86_MCE_AMD)	+= mce_amd_64.o
 obj-$(CONFIG_X86_MCE_NONFATAL)	+= non-fatal.o
+obj-$(CONFIG_X86_MCE_THRESHOLD) += threshold.o

arch/x86/kernel/cpu/mcheck/mce_32.c

@@ -60,20 +60,6 @@ void mcheck_init(struct cpuinfo_x86 *c)
 	}
 }
 
-static unsigned long old_cr4 __initdata;
-
-void __init stop_mce(void)
-{
-	old_cr4 = read_cr4();
-	clear_in_cr4(X86_CR4_MCE);
-}
-
-void __init restart_mce(void)
-{
-	if (old_cr4 & X86_CR4_MCE)
-		set_in_cr4(X86_CR4_MCE);
-}
-
 static int __init mcheck_disable(char *str)
 {
 	mce_disabled = 1;

arch/x86/kernel/cpu/mcheck/mce_64.c

@@ -3,6 +3,8 @@
  * K8 parts Copyright 2002,2003 Andi Kleen, SuSE Labs.
  * Rest from unknown author(s).
  * 2004 Andi Kleen. Rewrote most of it.
+ * Copyright 2008 Intel Corporation
+ * Author: Andi Kleen
  */
 
 #include <linux/init.h>
@@ -24,6 +26,9 @@
 #include <linux/ctype.h>
 #include <linux/kmod.h>
 #include <linux/kdebug.h>
+#include <linux/kobject.h>
+#include <linux/sysfs.h>
+#include <linux/ratelimit.h>
 #include <asm/processor.h>
 #include <asm/msr.h>
 #include <asm/mce.h>
@@ -32,7 +37,6 @@
 #include <asm/idle.h>
 
 #define MISC_MCELOG_MINOR 227
-#define NR_SYSFS_BANKS 6
 
 atomic_t mce_entry;
 
@@ -47,7 +51,7 @@ static int mce_dont_init;
  */
 static int tolerant = 1;
 static int banks;
-static unsigned long bank[NR_SYSFS_BANKS] = { [0 ... NR_SYSFS_BANKS-1] = ~0UL };
+static u64 *bank;
 static unsigned long notify_user;
 static int rip_msr;
 static int mce_bootlog = -1;
@@ -58,6 +62,19 @@ static char *trigger_argv[2] = { trigger, NULL };
 
 static DECLARE_WAIT_QUEUE_HEAD(mce_wait);
 
+/* MCA banks polled by the period polling timer for corrected events */
+DEFINE_PER_CPU(mce_banks_t, mce_poll_banks) = {
+	[0 ... BITS_TO_LONGS(MAX_NR_BANKS)-1] = ~0UL
+};
+
+/* Do initial initialization of a struct mce */
+void mce_setup(struct mce *m)
+{
+	memset(m, 0, sizeof(struct mce));
+	m->cpu = smp_processor_id();
+	rdtscll(m->tsc);
+}
+
 /*
  * Lockless MCE logging infrastructure.
  * This avoids deadlocks on printk locks without having to break locks. Also
@@ -119,11 +136,11 @@ static void print_mce(struct mce *m)
 			print_symbol("{%s}", m->ip);
 		printk("\n");
 	}
-	printk(KERN_EMERG "TSC %Lx ", m->tsc);
+	printk(KERN_EMERG "TSC %llx ", m->tsc);
 	if (m->addr)
-		printk("ADDR %Lx ", m->addr);
+		printk("ADDR %llx ", m->addr);
 	if (m->misc)
-		printk("MISC %Lx ", m->misc);
+		printk("MISC %llx ", m->misc);
 	printk("\n");
 	printk(KERN_EMERG "This is not a software problem!\n");
 	printk(KERN_EMERG "Run through mcelog --ascii to decode "
@@ -149,8 +166,10 @@ static void mce_panic(char *msg, struct mce *backup, unsigned long start)
 	panic(msg);
 }
 
-static int mce_available(struct cpuinfo_x86 *c)
+int mce_available(struct cpuinfo_x86 *c)
 {
+	if (mce_dont_init)
+		return 0;
 	return cpu_has(c, X86_FEATURE_MCE) && cpu_has(c, X86_FEATURE_MCA);
 }
 
@@ -172,7 +191,77 @@ static inline void mce_get_rip(struct mce *m, struct pt_regs *regs)
 }
 
 /*
- * The actual machine check handler
+ * Poll for corrected events or events that happened before reset.
+ * Those are just logged through /dev/mcelog.
+ *
+ * This is executed in standard interrupt context.
+ */
+void machine_check_poll(enum mcp_flags flags, mce_banks_t *b)
+{
+	struct mce m;
+	int i;
+
+	mce_setup(&m);
+
+	rdmsrl(MSR_IA32_MCG_STATUS, m.mcgstatus);
+	for (i = 0; i < banks; i++) {
+		if (!bank[i] || !test_bit(i, *b))
+			continue;
+
+		m.misc = 0;
+		m.addr = 0;
+		m.bank = i;
+		m.tsc = 0;
+
+		barrier();
+		rdmsrl(MSR_IA32_MC0_STATUS + i*4, m.status);
+		if (!(m.status & MCI_STATUS_VAL))
+			continue;
+
+		/*
+		 * Uncorrected events are handled by the exception handler
+		 * when it is enabled. But when the exception is disabled log
+		 * everything.
+		 *
+		 * TBD do the same check for MCI_STATUS_EN here?
+		 */
+		if ((m.status & MCI_STATUS_UC) && !(flags & MCP_UC))
+			continue;
+
+		if (m.status & MCI_STATUS_MISCV)
+			rdmsrl(MSR_IA32_MC0_MISC + i*4, m.misc);
+		if (m.status & MCI_STATUS_ADDRV)
+			rdmsrl(MSR_IA32_MC0_ADDR + i*4, m.addr);
+
+		if (!(flags & MCP_TIMESTAMP))
+			m.tsc = 0;
+		/*
+		 * Don't get the IP here because it's unlikely to
+		 * have anything to do with the actual error location.
+		 */
+
+		mce_log(&m);
+		add_taint(TAINT_MACHINE_CHECK);
+
+		/*
+		 * Clear state for this bank.
+		 */
+		wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0);
+	}
+
+	/*
+	 * Don't clear MCG_STATUS here because it's only defined for
+	 * exceptions.
+	 */
+}
+
+/*
+ * The actual machine check handler. This only handles real
+ * exceptions when something got corrupted coming in through int 18.
+ *
+ * This is executed in NMI context not subject to normal locking rules. This
+ * implies that most kernel services cannot be safely used. Don't even
+ * think about putting a printk in there!
  */
 void do_machine_check(struct pt_regs * regs, long error_code)
 {
@@ -190,17 +279,18 @@ void do_machine_check(struct pt_regs * regs, long error_code)
 	 * error.
 	 */
 	int kill_it = 0;
+	DECLARE_BITMAP(toclear, MAX_NR_BANKS);
 
 	atomic_inc(&mce_entry);
 
-	if ((regs
+	if (notify_die(DIE_NMI, "machine check", regs, error_code,
-	     && notify_die(DIE_NMI, "machine check", regs, error_code,
 			   18, SIGKILL) == NOTIFY_STOP)
-	    || !banks)
+		goto out2;
+	if (!banks)
 		goto out2;
 
-	memset(&m, 0, sizeof(struct mce));
+	mce_setup(&m);
-	m.cpu = smp_processor_id();
+
 	rdmsrl(MSR_IA32_MCG_STATUS, m.mcgstatus);
 	/* if the restart IP is not valid, we're done for */
 	if (!(m.mcgstatus & MCG_STATUS_RIPV))
@@ -210,18 +300,32 @@ void do_machine_check(struct pt_regs * regs, long error_code)
 	barrier();
 
 	for (i = 0; i < banks; i++) {
-		if (i < NR_SYSFS_BANKS && !bank[i])
+		__clear_bit(i, toclear);
+		if (!bank[i])
 			continue;
 
 		m.misc = 0;
 		m.addr = 0;
 		m.bank = i;
-		m.tsc = 0;
 
 		rdmsrl(MSR_IA32_MC0_STATUS + i*4, m.status);
 		if ((m.status & MCI_STATUS_VAL) == 0)
 			continue;
 
+		/*
+		 * Non uncorrected errors are handled by machine_check_poll
+		 * Leave them alone.
+		 */
+		if ((m.status & MCI_STATUS_UC) == 0)
+			continue;
+
+		/*
+		 * Set taint even when machine check was not enabled.
+		 */
+		add_taint(TAINT_MACHINE_CHECK);
+
+		__set_bit(i, toclear);
+
 		if (m.status & MCI_STATUS_EN) {
 			/* if PCC was set, there's no way out */
 			no_way_out |= !!(m.status & MCI_STATUS_PCC);
@@ -235,6 +339,12 @@ void do_machine_check(struct pt_regs * regs, long error_code)
 					no_way_out = 1;
 				kill_it = 1;
 			}
+		} else {
+			/*
+			 * Machine check event was not enabled. Clear, but
+			 * ignore.
+			 */
+			continue;
 		}
 
 		if (m.status & MCI_STATUS_MISCV)
@@ -243,10 +353,7 @@ void do_machine_check(struct pt_regs * regs, long error_code)
 			rdmsrl(MSR_IA32_MC0_ADDR + i*4, m.addr);
 
 		mce_get_rip(&m, regs);
-		if (error_code >= 0)
+		mce_log(&m);
-			rdtscll(m.tsc);
-		if (error_code != -2)
-			mce_log(&m);
 
 		/* Did this bank cause the exception? */
 		/* Assume that the bank with uncorrectable errors did it,
@@ -255,14 +362,8 @@ void do_machine_check(struct pt_regs * regs, long error_code)
 			panicm = m;
 			panicm_found = 1;
 		}
-
-		add_taint(TAINT_MACHINE_CHECK);
 	}
 
-	/* Never do anything final in the polling timer */
-	if (!regs)
-		goto out;
-
 	/* If we didn't find an uncorrectable error, pick
 	   the last one (shouldn't happen, just being safe). */
 	if (!panicm_found)
@@ -309,10 +410,11 @@ void do_machine_check(struct pt_regs * regs, long error_code)
 	/* notify userspace ASAP */
 	set_thread_flag(TIF_MCE_NOTIFY);
 
- out:
 	/* the last thing we do is clear state */
-	for (i = 0; i < banks; i++)
+	for (i = 0; i < banks; i++) {
-		wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0);
+		if (test_bit(i, toclear))
+			wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0);
+	}
 	wrmsrl(MSR_IA32_MCG_STATUS, 0);
  out2:
 	atomic_dec(&mce_entry);
@@ -332,15 +434,13 @@ void do_machine_check(struct pt_regs * regs, long error_code)
  * and historically has been the register value of the
  * MSR_IA32_THERMAL_STATUS (Intel) msr.
  */
-void mce_log_therm_throt_event(unsigned int cpu, __u64 status)
+void mce_log_therm_throt_event(__u64 status)
 {
 	struct mce m;
 
-	memset(&m, 0, sizeof(m));
+	mce_setup(&m);
-	m.cpu = cpu;
 	m.bank = MCE_THERMAL_BANK;
 	m.status = status;
-	rdtscll(m.tsc);
 	mce_log(&m);
 }
 #endif /* CONFIG_X86_MCE_INTEL */
@@ -353,18 +453,18 @@ void mce_log_therm_throt_event(unsigned int cpu, __u64 status)
 
 static int check_interval = 5 * 60; /* 5 minutes */
 static int next_interval; /* in jiffies */
-static void mcheck_timer(struct work_struct *work);
+static void mcheck_timer(unsigned long);
-static DECLARE_DELAYED_WORK(mcheck_work, mcheck_timer);
+static DEFINE_PER_CPU(struct timer_list, mce_timer);
 
-static void mcheck_check_cpu(void *info)
+static void mcheck_timer(unsigned long data)
 {
+	struct timer_list *t = &per_cpu(mce_timer, data);
+
+	WARN_ON(smp_processor_id() != data);
+
 	if (mce_available(&current_cpu_data))
-		do_machine_check(NULL, 0);
+		machine_check_poll(MCP_TIMESTAMP,
-}
+				&__get_cpu_var(mce_poll_banks));
-
-static void mcheck_timer(struct work_struct *work)
-{
-	on_each_cpu(mcheck_check_cpu, NULL, 1);
 
 	/*
 	 * Alert userspace if needed.  If we logged an MCE, reduce the
@@ -377,31 +477,41 @@ static void mcheck_timer(struct work_struct *work)
 				(int)round_jiffies_relative(check_interval*HZ));
 	}
 
-	schedule_delayed_work(&mcheck_work, next_interval);
+	t->expires = jiffies + next_interval;
+	add_timer(t);
 }
 
+static void mce_do_trigger(struct work_struct *work)
+{
+	call_usermodehelper(trigger, trigger_argv, NULL, UMH_NO_WAIT);
+}
+
+static DECLARE_WORK(mce_trigger_work, mce_do_trigger);
+
 /*
- * This is only called from process context.  This is where we do
+ * Notify the user(s) about new machine check events.
- * anything we need to alert userspace about new MCEs.  This is called
+ * Can be called from interrupt context, but not from machine check/NMI
- * directly from the poller and also from entry.S and idle, thanks to
+ * context.
- * TIF_MCE_NOTIFY.
  */
 int mce_notify_user(void)
 {
+	/* Not more than two messages every minute */
+	static DEFINE_RATELIMIT_STATE(ratelimit, 60*HZ, 2);
+
 	clear_thread_flag(TIF_MCE_NOTIFY);
 	if (test_and_clear_bit(0, &notify_user)) {
-		static unsigned long last_print;
-		unsigned long now = jiffies;
-
 		wake_up_interruptible(&mce_wait);
-		if (trigger[0])
-			call_usermodehelper(trigger, trigger_argv, NULL,
-						UMH_NO_WAIT);
 
-		if (time_after_eq(now, last_print + (check_interval*HZ))) {
+		/*
-			last_print = now;
+		 * There is no risk of missing notifications because
+		 * work_pending is always cleared before the function is
+		 * executed.
+		 */
+		if (trigger[0] && !work_pending(&mce_trigger_work))
+			schedule_work(&mce_trigger_work);
+
+		if (__ratelimit(&ratelimit))
 			printk(KERN_INFO "Machine check events logged\n");
-		}
 
 		return 1;
 	}
@@ -425,63 +535,78 @@ static struct notifier_block mce_idle_notifier = {
 
 static __init int periodic_mcheck_init(void)
 {
-	next_interval = check_interval * HZ;
+       idle_notifier_register(&mce_idle_notifier);
-	if (next_interval)
+       return 0;
-		schedule_delayed_work(&mcheck_work,
-				      round_jiffies_relative(next_interval));
-	idle_notifier_register(&mce_idle_notifier);
-	return 0;
 }
 __initcall(periodic_mcheck_init);
 
-
 /*
  * Initialize Machine Checks for a CPU.
  */
-static void mce_init(void *dummy)
+static int mce_cap_init(void)
 {
 	u64 cap;
-	int i;
+	unsigned b;
 
 	rdmsrl(MSR_IA32_MCG_CAP, cap);
-	banks = cap & 0xff;
+	b = cap & 0xff;
-	if (banks > MCE_EXTENDED_BANK) {
+	if (b > MAX_NR_BANKS) {
-		banks = MCE_EXTENDED_BANK;
+		printk(KERN_WARNING
-		printk(KERN_INFO "MCE: warning: using only %d banks\n",
+		       "MCE: Using only %u machine check banks out of %u\n",
-		       MCE_EXTENDED_BANK);
+			MAX_NR_BANKS, b);
+		b = MAX_NR_BANKS;
 	}
+
+	/* Don't support asymmetric configurations today */
+	WARN_ON(banks != 0 && b != banks);
+	banks = b;
+	if (!bank) {
+		bank = kmalloc(banks * sizeof(u64), GFP_KERNEL);
+		if (!bank)
+			return -ENOMEM;
+		memset(bank, 0xff, banks * sizeof(u64));
+	}
+
 	/* Use accurate RIP reporting if available. */
 	if ((cap & (1<<9)) && ((cap >> 16) & 0xff) >= 9)
 		rip_msr = MSR_IA32_MCG_EIP;
 
-	/* Log the machine checks left over from the previous reset.
+	return 0;
-	   This also clears all registers */
+}
-	do_machine_check(NULL, mce_bootlog ? -1 : -2);
+
+static void mce_init(void *dummy)
+{
+	u64 cap;
+	int i;
+	mce_banks_t all_banks;
+
+	/*
+	 * Log the machine checks left over from the previous reset.
+	 */
+	bitmap_fill(all_banks, MAX_NR_BANKS);
+	machine_check_poll(MCP_UC, &all_banks);
 
 	set_in_cr4(X86_CR4_MCE);
 
+	rdmsrl(MSR_IA32_MCG_CAP, cap);
 	if (cap & MCG_CTL_P)
 		wrmsr(MSR_IA32_MCG_CTL, 0xffffffff, 0xffffffff);
 
 	for (i = 0; i < banks; i++) {
-		if (i < NR_SYSFS_BANKS)
+		wrmsrl(MSR_IA32_MC0_CTL+4*i, bank[i]);
-			wrmsrl(MSR_IA32_MC0_CTL+4*i, bank[i]);
-		else
-			wrmsrl(MSR_IA32_MC0_CTL+4*i, ~0UL);
-
 		wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0);
 	}
 }
 
 /* Add per CPU specific workarounds here */
-static void __cpuinit mce_cpu_quirks(struct cpuinfo_x86 *c)
+static void mce_cpu_quirks(struct cpuinfo_x86 *c)
 {
 	/* This should be disabled by the BIOS, but isn't always */
 	if (c->x86_vendor == X86_VENDOR_AMD) {
-		if(c->x86 == 15)
+		if (c->x86 == 15 && banks > 4)
 			/* disable GART TBL walk error reporting, which trips off
 			   incorrectly with the IOMMU & 3ware & Cerberus. */
-			clear_bit(10, &bank[4]);
+			clear_bit(10, (unsigned long *)&bank[4]);
 		if(c->x86 <= 17 && mce_bootlog < 0)
 			/* Lots of broken BIOS around that don't clear them
 			   by default and leave crap in there. Don't log. */
@@ -504,20 +629,38 @@ static void mce_cpu_features(struct cpuinfo_x86 *c)
 	}
 }
 
+static void mce_init_timer(void)
+{
+	struct timer_list *t = &__get_cpu_var(mce_timer);
+
+	/* data race harmless because everyone sets to the same value */
+	if (!next_interval)
+		next_interval = check_interval * HZ;
+	if (!next_interval)
+		return;
+	setup_timer(t, mcheck_timer, smp_processor_id());
+	t->expires = round_jiffies(jiffies + next_interval);
+	add_timer(t);
+}
+
 /*
  * Called for each booted CPU to set up machine checks.
  * Must be called with preempt off.
  */
 void __cpuinit mcheck_init(struct cpuinfo_x86 *c)
 {
-	mce_cpu_quirks(c);
+	if (!mce_available(c))
-
-	if (mce_dont_init ||
-	    !mce_available(c))
 		return;
 
+	if (mce_cap_init() < 0) {
+		mce_dont_init = 1;
+		return;
+	}
+	mce_cpu_quirks(c);
+
 	mce_init(NULL);
 	mce_cpu_features(c);
+	mce_init_timer();
 }
 
 /*
@@ -573,7 +716,7 @@ static ssize_t mce_read(struct file *filp, char __user *ubuf, size_t usize,
 {
 	unsigned long *cpu_tsc;
 	static DEFINE_MUTEX(mce_read_mutex);
-	unsigned next;
+	unsigned prev, next;
 	char __user *buf = ubuf;
 	int i, err;
 
@@ -592,25 +735,32 @@ static ssize_t mce_read(struct file *filp, char __user *ubuf, size_t usize,
 	}
 
 	err = 0;
-	for (i = 0; i < next; i++) {
+	prev = 0;
-		unsigned long start = jiffies;
+	do {
+		for (i = prev; i < next; i++) {
+			unsigned long start = jiffies;
 
-		while (!mcelog.entry[i].finished) {
+			while (!mcelog.entry[i].finished) {
-			if (time_after_eq(jiffies, start + 2)) {
+				if (time_after_eq(jiffies, start + 2)) {
-				memset(mcelog.entry + i,0, sizeof(struct mce));
+					memset(mcelog.entry + i, 0,
-				goto timeout;
+					       sizeof(struct mce));
+					goto timeout;
+				}
+				cpu_relax();
 			}
-			cpu_relax();
+			smp_rmb();
+			err |= copy_to_user(buf, mcelog.entry + i,
+					    sizeof(struct mce));
+			buf += sizeof(struct mce);
+timeout:
+			;
 		}
-		smp_rmb();
-		err |= copy_to_user(buf, mcelog.entry + i, sizeof(struct mce));
-		buf += sizeof(struct mce);
- timeout:
-		;
-	}
 
-	memset(mcelog.entry, 0, next * sizeof(struct mce));
+		memset(mcelog.entry + prev, 0,
-	mcelog.next = 0;
+		       (next - prev) * sizeof(struct mce));
+		prev = next;
+		next = cmpxchg(&mcelog.next, prev, 0);
+	} while (next != prev);
 
 	synchronize_sched();
 
@@ -680,20 +830,6 @@ static struct miscdevice mce_log_device = {
 	&mce_chrdev_ops,
 };
 
-static unsigned long old_cr4 __initdata;
-
-void __init stop_mce(void)
-{
-	old_cr4 = read_cr4();
-	clear_in_cr4(X86_CR4_MCE);
-}
-
-void __init restart_mce(void)
-{
-	if (old_cr4 & X86_CR4_MCE)
-		set_in_cr4(X86_CR4_MCE);
-}
-
 /*
  * Old style boot options parsing. Only for compatibility.
  */
@@ -703,8 +839,7 @@ static int __init mcheck_disable(char *str)
 	return 1;
 }
 
-/* mce=off disables machine check. Note you can re-enable it later
+/* mce=off disables machine check.
-   using sysfs.
    mce=TOLERANCELEVEL (number, see above)
    mce=bootlog Log MCEs from before booting. Disabled by default on AMD.
    mce=nobootlog Don't log MCEs from before booting. */
@@ -728,6 +863,29 @@ __setup("mce=", mcheck_enable);
  * Sysfs support
  */
 
+/*
+ * Disable machine checks on suspend and shutdown. We can't really handle
+ * them later.
+ */
+static int mce_disable(void)
+{
+	int i;
+
+	for (i = 0; i < banks; i++)
+		wrmsrl(MSR_IA32_MC0_CTL + i*4, 0);
+	return 0;
+}
+
+static int mce_suspend(struct sys_device *dev, pm_message_t state)
+{
+	return mce_disable();
+}
+
+static int mce_shutdown(struct sys_device *dev)
+{
+	return mce_disable();
+}
+
 /* On resume clear all MCE state. Don't want to see leftovers from the BIOS.
    Only one CPU is active at this time, the others get readded later using
    CPU hotplug. */
@@ -738,20 +896,24 @@ static int mce_resume(struct sys_device *dev)
 	return 0;
 }
 
+static void mce_cpu_restart(void *data)
+{
+	del_timer_sync(&__get_cpu_var(mce_timer));
+	if (mce_available(&current_cpu_data))
+		mce_init(NULL);
+	mce_init_timer();
+}
+
 /* Reinit MCEs after user configuration changes */
 static void mce_restart(void)
 {
-	if (next_interval)
-		cancel_delayed_work(&mcheck_work);
-	/* Timer race is harmless here */
-	on_each_cpu(mce_init, NULL, 1);
 	next_interval = check_interval * HZ;
-	if (next_interval)
+	on_each_cpu(mce_cpu_restart, NULL, 1);
-		schedule_delayed_work(&mcheck_work,
-				      round_jiffies_relative(next_interval));
 }
 
 static struct sysdev_class mce_sysclass = {
+	.suspend = mce_suspend,
+	.shutdown = mce_shutdown,
 	.resume = mce_resume,
 	.name = "machinecheck",
 };
@@ -778,16 +940,26 @@ void (*threshold_cpu_callback)(unsigned long action, unsigned int cpu) __cpuinit
 	}								\
 	static SYSDEV_ATTR(name, 0644, show_ ## name, set_ ## name);
 
-/*
+static struct sysdev_attribute *bank_attrs;
- * TBD should generate these dynamically based on number of available banks.
+
- * Have only 6 contol banks in /sysfs until then.
+static ssize_t show_bank(struct sys_device *s, struct sysdev_attribute *attr,
- */
+			 char *buf)
-ACCESSOR(bank0ctl,bank[0],mce_restart())
+{
-ACCESSOR(bank1ctl,bank[1],mce_restart())
+	u64 b = bank[attr - bank_attrs];
-ACCESSOR(bank2ctl,bank[2],mce_restart())
+	return sprintf(buf, "%llx\n", b);
-ACCESSOR(bank3ctl,bank[3],mce_restart())
+}
-ACCESSOR(bank4ctl,bank[4],mce_restart())
+
-ACCESSOR(bank5ctl,bank[5],mce_restart())
+static ssize_t set_bank(struct sys_device *s, struct sysdev_attribute *attr,
+			const char *buf, size_t siz)
+{
+	char *end;
+	u64 new = simple_strtoull(buf, &end, 0);
+	if (end == buf)
+		return -EINVAL;
+	bank[attr - bank_attrs] = new;
+	mce_restart();
+	return end-buf;
+}
 
 static ssize_t show_trigger(struct sys_device *s, struct sysdev_attribute *attr,
 				char *buf)
@@ -814,8 +986,6 @@ static SYSDEV_ATTR(trigger, 0644, show_trigger, set_trigger);
 static SYSDEV_INT_ATTR(tolerant, 0644, tolerant);
 ACCESSOR(check_interval,check_interval,mce_restart())
 static struct sysdev_attribute *mce_attributes[] = {
-	&attr_bank0ctl, &attr_bank1ctl, &attr_bank2ctl,
-	&attr_bank3ctl, &attr_bank4ctl, &attr_bank5ctl,
 	&attr_tolerant.attr, &attr_check_interval, &attr_trigger,
 	NULL
 };
@@ -845,11 +1015,22 @@ static __cpuinit int mce_create_device(unsigned int cpu)
 		if (err)
 			goto error;
 	}
+	for (i = 0; i < banks; i++) {
+		err = sysdev_create_file(&per_cpu(device_mce, cpu),
+					&bank_attrs[i]);
+		if (err)
+			goto error2;
+	}
 	cpu_set(cpu, mce_device_initialized);
 
 	return 0;
+error2:
+	while (--i >= 0) {
+		sysdev_remove_file(&per_cpu(device_mce, cpu),
+					&bank_attrs[i]);
+	}
 error:
-	while (i--) {
+	while (--i >= 0) {
 		sysdev_remove_file(&per_cpu(device_mce,cpu),
 				   mce_attributes[i]);
 	}
@@ -868,15 +1049,46 @@ static __cpuinit void mce_remove_device(unsigned int cpu)
 	for (i = 0; mce_attributes[i]; i++)
 		sysdev_remove_file(&per_cpu(device_mce,cpu),
 			mce_attributes[i]);
+	for (i = 0; i < banks; i++)
+		sysdev_remove_file(&per_cpu(device_mce, cpu),
+			&bank_attrs[i]);
 	sysdev_unregister(&per_cpu(device_mce,cpu));
 	cpu_clear(cpu, mce_device_initialized);
 }
 
+/* Make sure there are no machine checks on offlined CPUs. */
+static void mce_disable_cpu(void *h)
+{
+	int i;
+	unsigned long action = *(unsigned long *)h;
+
+	if (!mce_available(&current_cpu_data))
+		return;
+	if (!(action & CPU_TASKS_FROZEN))
+		cmci_clear();
+	for (i = 0; i < banks; i++)
+		wrmsrl(MSR_IA32_MC0_CTL + i*4, 0);
+}
+
+static void mce_reenable_cpu(void *h)
+{
+	int i;
+	unsigned long action = *(unsigned long *)h;
+
+	if (!mce_available(&current_cpu_data))
+		return;
+	if (!(action & CPU_TASKS_FROZEN))
+		cmci_reenable();
+	for (i = 0; i < banks; i++)
+		wrmsrl(MSR_IA32_MC0_CTL + i*4, bank[i]);
+}
+
 /* Get notified when a cpu comes on/off. Be hotplug friendly. */
 static int __cpuinit mce_cpu_callback(struct notifier_block *nfb,
 				      unsigned long action, void *hcpu)
 {
 	unsigned int cpu = (unsigned long)hcpu;
+	struct timer_list *t = &per_cpu(mce_timer, cpu);
 
 	switch (action) {
 	case CPU_ONLINE:
@@ -891,6 +1103,21 @@ static int __cpuinit mce_cpu_callback(struct notifier_block *nfb,
 			threshold_cpu_callback(action, cpu);
 		mce_remove_device(cpu);
 		break;
+	case CPU_DOWN_PREPARE:
+	case CPU_DOWN_PREPARE_FROZEN:
+		del_timer_sync(t);
+		smp_call_function_single(cpu, mce_disable_cpu, &action, 1);
+		break;
+	case CPU_DOWN_FAILED:
+	case CPU_DOWN_FAILED_FROZEN:
+		t->expires = round_jiffies(jiffies + next_interval);
+		add_timer_on(t, cpu);
+		smp_call_function_single(cpu, mce_reenable_cpu, &action, 1);
+		break;
+	case CPU_POST_DEAD:
+		/* intentionally ignoring frozen here */
+		cmci_rediscover(cpu);
+		break;
 	}
 	return NOTIFY_OK;
 }
@@ -899,6 +1126,34 @@ static struct notifier_block mce_cpu_notifier __cpuinitdata = {
 	.notifier_call = mce_cpu_callback,
 };
 
+static __init int mce_init_banks(void)
+{
+	int i;
+
+	bank_attrs = kzalloc(sizeof(struct sysdev_attribute) * banks,
+				GFP_KERNEL);
+	if (!bank_attrs)
+		return -ENOMEM;
+
+	for (i = 0; i < banks; i++) {
+		struct sysdev_attribute *a = &bank_attrs[i];
+		a->attr.name = kasprintf(GFP_KERNEL, "bank%d", i);
+		if (!a->attr.name)
+			goto nomem;
+		a->attr.mode = 0644;
+		a->show = show_bank;
+		a->store = set_bank;
+	}
+	return 0;
+
+nomem:
+	while (--i >= 0)
+		kfree(bank_attrs[i].attr.name);
+	kfree(bank_attrs);
+	bank_attrs = NULL;
+	return -ENOMEM;
+}
+
 static __init int mce_init_device(void)
 {
 	int err;
@@ -906,6 +1161,11 @@ static __init int mce_init_device(void)
 
 	if (!mce_available(&boot_cpu_data))
 		return -EIO;
+
+	err = mce_init_banks();
+	if (err)
+		return err;
+
 	err = sysdev_class_register(&mce_sysclass);
 	if (err)
 		return err;

arch/x86/kernel/cpu/mcheck/mce_amd_64.c

@@ -79,6 +79,8 @@ static unsigned char shared_bank[NR_BANKS] = {
 
 static DEFINE_PER_CPU(unsigned char, bank_map);	/* see which banks are on */
 
+static void amd_threshold_interrupt(void);
+
 /*
  * CPU Initialization
  */
@@ -90,7 +92,8 @@ struct thresh_restart {
 };
 
 /* must be called with correct cpu affinity */
-static long threshold_restart_bank(void *_tr)
+/* Called via smp_call_function_single() */
+static void threshold_restart_bank(void *_tr)
 {
 	struct thresh_restart *tr = _tr;
 	u32 mci_misc_hi, mci_misc_lo;
@@ -117,7 +120,6 @@ static long threshold_restart_bank(void *_tr)
 
 	mci_misc_hi |= MASK_COUNT_EN_HI;
 	wrmsr(tr->b->address, mci_misc_lo, mci_misc_hi);
-	return 0;
 }
 
 /* cpu init entry point, called from mce.c with preempt off */
@@ -174,6 +176,8 @@ void mce_amd_feature_init(struct cpuinfo_x86 *c)
 			tr.reset = 0;
 			tr.old_limit = 0;
 			threshold_restart_bank(&tr);
+
+			mce_threshold_vector = amd_threshold_interrupt;
 		}
 	}
 }
@@ -187,19 +191,13 @@ void mce_amd_feature_init(struct cpuinfo_x86 *c)
  * the interrupt goes off when error_count reaches threshold_limit.
  * the handler will simply log mcelog w/ software defined bank number.
  */
-asmlinkage void mce_threshold_interrupt(void)
+static void amd_threshold_interrupt(void)
 {
 	unsigned int bank, block;
 	struct mce m;
 	u32 low = 0, high = 0, address = 0;
 
-	ack_APIC_irq();
+	mce_setup(&m);
-	exit_idle();
-	irq_enter();
-
-	memset(&m, 0, sizeof(m));
-	rdtscll(m.tsc);
-	m.cpu = smp_processor_id();
 
 	/* assume first bank caused it */
 	for (bank = 0; bank < NR_BANKS; ++bank) {
@@ -233,7 +231,8 @@ asmlinkage void mce_threshold_interrupt(void)
 
 			/* Log the machine check that caused the threshold
 			   event. */
-			do_machine_check(NULL, 0);
+			machine_check_poll(MCP_TIMESTAMP,
+					&__get_cpu_var(mce_poll_banks));
 
 			if (high & MASK_OVERFLOW_HI) {
 				rdmsrl(address, m.misc);
@@ -243,13 +242,10 @@ asmlinkage void mce_threshold_interrupt(void)
 				       + bank * NR_BLOCKS
 				       + block;
 				mce_log(&m);
-				goto out;
+				return;
 			}
 		}
 	}
-out:
-	inc_irq_stat(irq_threshold_count);
-	irq_exit();
 }
 
 /*
@@ -283,7 +279,7 @@ static ssize_t store_interrupt_enable(struct threshold_block *b,
 	tr.b = b;
 	tr.reset = 0;
 	tr.old_limit = 0;
-	work_on_cpu(b->cpu, threshold_restart_bank, &tr);
+	smp_call_function_single(b->cpu, threshold_restart_bank, &tr, 1);
 
 	return end - buf;
 }
@@ -305,23 +301,32 @@ static ssize_t store_threshold_limit(struct threshold_block *b,
 	tr.b = b;
 	tr.reset = 0;
 
-	work_on_cpu(b->cpu, threshold_restart_bank, &tr);
+	smp_call_function_single(b->cpu, threshold_restart_bank, &tr, 1);
 
 	return end - buf;
 }
 
-static long local_error_count(void *_b)
+struct threshold_block_cross_cpu {
+	struct threshold_block *tb;
+	long retval;
+};
+
+static void local_error_count_handler(void *_tbcc)
 {
-	struct threshold_block *b = _b;
+	struct threshold_block_cross_cpu *tbcc = _tbcc;
+	struct threshold_block *b = tbcc->tb;
 	u32 low, high;
 
 	rdmsr(b->address, low, high);
-	return (high & 0xFFF) - (THRESHOLD_MAX - b->threshold_limit);
+	tbcc->retval = (high & 0xFFF) - (THRESHOLD_MAX - b->threshold_limit);
 }
 
 static ssize_t show_error_count(struct threshold_block *b, char *buf)
 {
-	return sprintf(buf, "%lx\n", work_on_cpu(b->cpu, local_error_count, b));
+	struct threshold_block_cross_cpu tbcc = { .tb = b, };
+
+	smp_call_function_single(b->cpu, local_error_count_handler, &tbcc, 1);
+	return sprintf(buf, "%lx\n", tbcc.retval);
 }
 
 static ssize_t store_error_count(struct threshold_block *b,
@@ -329,7 +334,7 @@ static ssize_t store_error_count(struct threshold_block *b,
 {
 	struct thresh_restart tr = { .b = b, .reset = 1, .old_limit = 0 };
 
-	work_on_cpu(b->cpu, threshold_restart_bank, &tr);
+	smp_call_function_single(b->cpu, threshold_restart_bank, &tr, 1);
 	return 1;
 }
 
@@ -398,7 +403,7 @@ static __cpuinit int allocate_threshold_blocks(unsigned int cpu,
 	if ((bank >= NR_BANKS) || (block >= NR_BLOCKS))
 		return 0;
 
-	if (rdmsr_safe(address, &low, &high))
+	if (rdmsr_safe_on_cpu(cpu, address, &low, &high))
 		return 0;
 
 	if (!(high & MASK_VALID_HI)) {
@@ -462,12 +467,11 @@ static __cpuinit int allocate_threshold_blocks(unsigned int cpu,
 	return err;
 }
 
-static __cpuinit long local_allocate_threshold_blocks(void *_bank)
+static __cpuinit long
+local_allocate_threshold_blocks(int cpu, unsigned int bank)
 {
-	unsigned int *bank = _bank;
+	return allocate_threshold_blocks(cpu, bank, 0,
-
+					 MSR_IA32_MC0_MISC + bank * 4);
-	return allocate_threshold_blocks(smp_processor_id(), *bank, 0,
-					 MSR_IA32_MC0_MISC + *bank * 4);
 }
 
 /* symlinks sibling shared banks to first core.  first core owns dir/files. */
@@ -530,7 +534,7 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank)
 
 	per_cpu(threshold_banks, cpu)[bank] = b;
 
-	err = work_on_cpu(cpu, local_allocate_threshold_blocks, &bank);
+	err = local_allocate_threshold_blocks(cpu, bank);
 	if (err)
 		goto out_free;
 

arch/x86/kernel/cpu/mcheck/mce_intel_64.c

@@ -1,6 +1,8 @@
 /*
  * Intel specific MCE features.
  * Copyright 2004 Zwane Mwaikambo <zwane@linuxpower.ca>
+ * Copyright (C) 2008, 2009 Intel Corporation
+ * Author: Andi Kleen
  */
 
 #include <linux/init.h>
@@ -13,6 +15,7 @@
 #include <asm/hw_irq.h>
 #include <asm/idle.h>
 #include <asm/therm_throt.h>
+#include <asm/apic.h>
 
 asmlinkage void smp_thermal_interrupt(void)
 {
@@ -25,7 +28,7 @@ asmlinkage void smp_thermal_interrupt(void)
 
 	rdmsrl(MSR_IA32_THERM_STATUS, msr_val);
 	if (therm_throt_process(msr_val & 1))
-		mce_log_therm_throt_event(smp_processor_id(), msr_val);
+		mce_log_therm_throt_event(msr_val);
 
 	inc_irq_stat(irq_thermal_count);
 	irq_exit();
@@ -85,7 +88,209 @@ static void intel_init_thermal(struct cpuinfo_x86 *c)
 	return;
 }
 
+/*
+ * Support for Intel Correct Machine Check Interrupts. This allows
+ * the CPU to raise an interrupt when a corrected machine check happened.
+ * Normally we pick those up using a regular polling timer.
+ * Also supports reliable discovery of shared banks.
+ */
+
+static DEFINE_PER_CPU(mce_banks_t, mce_banks_owned);
+
+/*
+ * cmci_discover_lock protects against parallel discovery attempts
+ * which could race against each other.
+ */
+static DEFINE_SPINLOCK(cmci_discover_lock);
+
+#define CMCI_THRESHOLD 1
+
+static int cmci_supported(int *banks)
+{
+	u64 cap;
+
+	/*
+	 * Vendor check is not strictly needed, but the initial
+	 * initialization is vendor keyed and this
+	 * makes sure none of the backdoors are entered otherwise.
+	 */
+	if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL)
+		return 0;
+	if (!cpu_has_apic || lapic_get_maxlvt() < 6)
+		return 0;
+	rdmsrl(MSR_IA32_MCG_CAP, cap);
+	*banks = min_t(unsigned, MAX_NR_BANKS, cap & 0xff);
+	return !!(cap & MCG_CMCI_P);
+}
+
+/*
+ * The interrupt handler. This is called on every event.
+ * Just call the poller directly to log any events.
+ * This could in theory increase the threshold under high load,
+ * but doesn't for now.
+ */
+static void intel_threshold_interrupt(void)
+{
+	machine_check_poll(MCP_TIMESTAMP, &__get_cpu_var(mce_banks_owned));
+	mce_notify_user();
+}
+
+static void print_update(char *type, int *hdr, int num)
+{
+	if (*hdr == 0)
+		printk(KERN_INFO "CPU %d MCA banks", smp_processor_id());
+	*hdr = 1;
+	printk(KERN_CONT " %s:%d", type, num);
+}
+
+/*
+ * Enable CMCI (Corrected Machine Check Interrupt) for available MCE banks
+ * on this CPU. Use the algorithm recommended in the SDM to discover shared
+ * banks.
+ */
+static void cmci_discover(int banks, int boot)
+{
+	unsigned long *owned = (void *)&__get_cpu_var(mce_banks_owned);
+	int hdr = 0;
+	int i;
+
+	spin_lock(&cmci_discover_lock);
+	for (i = 0; i < banks; i++) {
+		u64 val;
+
+		if (test_bit(i, owned))
+			continue;
+
+		rdmsrl(MSR_IA32_MC0_CTL2 + i, val);
+
+		/* Already owned by someone else? */
+		if (val & CMCI_EN) {
+			if (test_and_clear_bit(i, owned) || boot)
+				print_update("SHD", &hdr, i);
+			__clear_bit(i, __get_cpu_var(mce_poll_banks));
+			continue;
+		}
+
+		val |= CMCI_EN | CMCI_THRESHOLD;
+		wrmsrl(MSR_IA32_MC0_CTL2 + i, val);
+		rdmsrl(MSR_IA32_MC0_CTL2 + i, val);
+
+		/* Did the enable bit stick? -- the bank supports CMCI */
+		if (val & CMCI_EN) {
+			if (!test_and_set_bit(i, owned) || boot)
+				print_update("CMCI", &hdr, i);
+			__clear_bit(i, __get_cpu_var(mce_poll_banks));
+		} else {
+			WARN_ON(!test_bit(i, __get_cpu_var(mce_poll_banks)));
+		}
+	}
+	spin_unlock(&cmci_discover_lock);
+	if (hdr)
+		printk(KERN_CONT "\n");
+}
+
+/*
+ * Just in case we missed an event during initialization check
+ * all the CMCI owned banks.
+ */
+void cmci_recheck(void)
+{
+	unsigned long flags;
+	int banks;
+
+	if (!mce_available(&current_cpu_data) || !cmci_supported(&banks))
+		return;
+	local_irq_save(flags);
+	machine_check_poll(MCP_TIMESTAMP, &__get_cpu_var(mce_banks_owned));
+	local_irq_restore(flags);
+}
+
+/*
+ * Disable CMCI on this CPU for all banks it owns when it goes down.
+ * This allows other CPUs to claim the banks on rediscovery.
+ */
+void cmci_clear(void)
+{
+	int i;
+	int banks;
+	u64 val;
+
+	if (!cmci_supported(&banks))
+		return;
+	spin_lock(&cmci_discover_lock);
+	for (i = 0; i < banks; i++) {
+		if (!test_bit(i, __get_cpu_var(mce_banks_owned)))
+			continue;
+		/* Disable CMCI */
+		rdmsrl(MSR_IA32_MC0_CTL2 + i, val);
+		val &= ~(CMCI_EN|CMCI_THRESHOLD_MASK);
+		wrmsrl(MSR_IA32_MC0_CTL2 + i, val);
+		__clear_bit(i, __get_cpu_var(mce_banks_owned));
+	}
+	spin_unlock(&cmci_discover_lock);
+}
+
+/*
+ * After a CPU went down cycle through all the others and rediscover
+ * Must run in process context.
+ */
+void cmci_rediscover(int dying)
+{
+	int banks;
+	int cpu;
+	cpumask_var_t old;
+
+	if (!cmci_supported(&banks))
+		return;
+	if (!alloc_cpumask_var(&old, GFP_KERNEL))
+		return;
+	cpumask_copy(old, &current->cpus_allowed);
+
+	for_each_online_cpu (cpu) {
+		if (cpu == dying)
+			continue;
+		if (set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu)))
+			continue;
+		/* Recheck banks in case CPUs don't all have the same */
+		if (cmci_supported(&banks))
+			cmci_discover(banks, 0);
+	}
+
+	set_cpus_allowed_ptr(current, old);
+	free_cpumask_var(old);
+}
+
+/*
+ * Reenable CMCI on this CPU in case a CPU down failed.
+ */
+void cmci_reenable(void)
+{
+	int banks;
+	if (cmci_supported(&banks))
+		cmci_discover(banks, 0);
+}
+
+static void intel_init_cmci(void)
+{
+	int banks;
+
+	if (!cmci_supported(&banks))
+		return;
+
+	mce_threshold_vector = intel_threshold_interrupt;
+	cmci_discover(banks, 1);
+	/*
+	 * For CPU #0 this runs with still disabled APIC, but that's
+	 * ok because only the vector is set up. We still do another
+	 * check for the banks later for CPU #0 just to make sure
+	 * to not miss any events.
+	 */
+	apic_write(APIC_LVTCMCI, THRESHOLD_APIC_VECTOR|APIC_DM_FIXED);
+	cmci_recheck();
+}
+
 void mce_intel_feature_init(struct cpuinfo_x86 *c)
 {
 	intel_init_thermal(c);
+	intel_init_cmci();
 }

arch/x86/kernel/cpu/mcheck/threshold.c

@@ -0,0 +1,29 @@
+/*
+ * Common corrected MCE threshold handler code:
+ */
+#include <linux/interrupt.h>
+#include <linux/kernel.h>
+
+#include <asm/irq_vectors.h>
+#include <asm/apic.h>
+#include <asm/idle.h>
+#include <asm/mce.h>
+
+static void default_threshold_interrupt(void)
+{
+	printk(KERN_ERR "Unexpected threshold interrupt at vector %x\n",
+			 THRESHOLD_APIC_VECTOR);
+}
+
+void (*mce_threshold_vector)(void) = default_threshold_interrupt;
+
+asmlinkage void mce_threshold_interrupt(void)
+{
+	exit_idle();
+	irq_enter();
+	inc_irq_stat(irq_threshold_count);
+	mce_threshold_vector();
+	irq_exit();
+	/* Ack only at the end to avoid potential reentry */
+	ack_APIC_irq();
+}

arch/x86/kernel/cpu/mtrr/Makefile

@@ -1,3 +1,3 @@
-obj-y		:= main.o if.o generic.o state.o
+obj-y		:= main.o if.o generic.o state.o cleanup.o
 obj-$(CONFIG_X86_32) += amd.o cyrix.o centaur.o
 

arch/x86/kernel/cpu/mtrr/generic.c

@@ -33,13 +33,31 @@ u64 mtrr_tom2;
 struct mtrr_state_type mtrr_state = {};
 EXPORT_SYMBOL_GPL(mtrr_state);
 
-static int __initdata mtrr_show;
+/**
-static int __init mtrr_debug(char *opt)
+ * BIOS is expected to clear MtrrFixDramModEn bit, see for example
+ * "BIOS and Kernel Developer's Guide for the AMD Athlon 64 and AMD
+ * Opteron Processors" (26094 Rev. 3.30 February 2006), section
+ * "13.2.1.2 SYSCFG Register": "The MtrrFixDramModEn bit should be set
+ * to 1 during BIOS initalization of the fixed MTRRs, then cleared to
+ * 0 for operation."
+ */
+static inline void k8_check_syscfg_dram_mod_en(void)
 {
-	mtrr_show = 1;
+	u32 lo, hi;
-	return 0;
+
+	if (!((boot_cpu_data.x86_vendor == X86_VENDOR_AMD) &&
+	      (boot_cpu_data.x86 >= 0x0f)))
+		return;
+
+	rdmsr(MSR_K8_SYSCFG, lo, hi);
+	if (lo & K8_MTRRFIXRANGE_DRAM_MODIFY) {
+		printk(KERN_ERR FW_WARN "MTRR: CPU %u: SYSCFG[MtrrFixDramModEn]"
+		       " not cleared by BIOS, clearing this bit\n",
+		       smp_processor_id());
+		lo &= ~K8_MTRRFIXRANGE_DRAM_MODIFY;
+		mtrr_wrmsr(MSR_K8_SYSCFG, lo, hi);
+	}
 }
-early_param("mtrr.show", mtrr_debug);
 
 /*
  * Returns the effective MTRR type for the region
@@ -174,6 +192,8 @@ get_fixed_ranges(mtrr_type * frs)
 	unsigned int *p = (unsigned int *) frs;
 	int i;
 
+	k8_check_syscfg_dram_mod_en();
+
 	rdmsr(MTRRfix64K_00000_MSR, p[0], p[1]);
 
 	for (i = 0; i < 2; i++)
@@ -188,18 +208,94 @@ void mtrr_save_fixed_ranges(void *info)
 		get_fixed_ranges(mtrr_state.fixed_ranges);
 }
 
-static void print_fixed(unsigned base, unsigned step, const mtrr_type*types)
+static unsigned __initdata last_fixed_start;
+static unsigned __initdata last_fixed_end;
+static mtrr_type __initdata last_fixed_type;
+
+static void __init print_fixed_last(void)
+{
+	if (!last_fixed_end)
+		return;
+
+	printk(KERN_DEBUG "  %05X-%05X %s\n", last_fixed_start,
+		last_fixed_end - 1, mtrr_attrib_to_str(last_fixed_type));
+
+	last_fixed_end = 0;
+}
+
+static void __init update_fixed_last(unsigned base, unsigned end,
+				       mtrr_type type)
+{
+	last_fixed_start = base;
+	last_fixed_end = end;
+	last_fixed_type = type;
+}
+
+static void __init print_fixed(unsigned base, unsigned step,
+			       const mtrr_type *types)
 {
 	unsigned i;
 
-	for (i = 0; i < 8; ++i, ++types, base += step)
+	for (i = 0; i < 8; ++i, ++types, base += step) {
-		printk(KERN_INFO "MTRR %05X-%05X %s\n",
+		if (last_fixed_end == 0) {
-			base, base + step - 1, mtrr_attrib_to_str(*types));
+			update_fixed_last(base, base + step, *types);
+			continue;
+		}
+		if (last_fixed_end == base && last_fixed_type == *types) {
+			last_fixed_end = base + step;
+			continue;
+		}
+		/* new segments: gap or different type */
+		print_fixed_last();
+		update_fixed_last(base, base + step, *types);
+	}
 }
 
 static void prepare_set(void);
 static void post_set(void);
 
+static void __init print_mtrr_state(void)
+{
+	unsigned int i;
+	int high_width;
+
+	printk(KERN_DEBUG "MTRR default type: %s\n",
+			 mtrr_attrib_to_str(mtrr_state.def_type));
+	if (mtrr_state.have_fixed) {
+		printk(KERN_DEBUG "MTRR fixed ranges %sabled:\n",
+		       mtrr_state.enabled & 1 ? "en" : "dis");
+		print_fixed(0x00000, 0x10000, mtrr_state.fixed_ranges + 0);
+		for (i = 0; i < 2; ++i)
+			print_fixed(0x80000 + i * 0x20000, 0x04000, mtrr_state.fixed_ranges + (i + 1) * 8);
+		for (i = 0; i < 8; ++i)
+			print_fixed(0xC0000 + i * 0x08000, 0x01000, mtrr_state.fixed_ranges + (i + 3) * 8);
+
+		/* tail */
+		print_fixed_last();
+	}
+	printk(KERN_DEBUG "MTRR variable ranges %sabled:\n",
+	       mtrr_state.enabled & 2 ? "en" : "dis");
+	high_width = ((size_or_mask ? ffs(size_or_mask) - 1 : 32) - (32 - PAGE_SHIFT) + 3) / 4;
+	for (i = 0; i < num_var_ranges; ++i) {
+		if (mtrr_state.var_ranges[i].mask_lo & (1 << 11))
+			printk(KERN_DEBUG "  %u base %0*X%05X000 mask %0*X%05X000 %s\n",
+			       i,
+			       high_width,
+			       mtrr_state.var_ranges[i].base_hi,
+			       mtrr_state.var_ranges[i].base_lo >> 12,
+			       high_width,
+			       mtrr_state.var_ranges[i].mask_hi,
+			       mtrr_state.var_ranges[i].mask_lo >> 12,
+			       mtrr_attrib_to_str(mtrr_state.var_ranges[i].base_lo & 0xff));
+		else
+			printk(KERN_DEBUG "  %u disabled\n", i);
+	}
+	if (mtrr_tom2) {
+		printk(KERN_DEBUG "TOM2: %016llx aka %lldM\n",
+				  mtrr_tom2, mtrr_tom2>>20);
+	}
+}
+
 /*  Grab all of the MTRR state for this CPU into *state  */
 void __init get_mtrr_state(void)
 {
@@ -231,41 +327,9 @@ void __init get_mtrr_state(void)
 		mtrr_tom2 |= low;
 		mtrr_tom2 &= 0xffffff800000ULL;
 	}
-	if (mtrr_show) {
-		int high_width;
 
-		printk(KERN_INFO "MTRR default type: %s\n", mtrr_attrib_to_str(mtrr_state.def_type));
+	print_mtrr_state();
-		if (mtrr_state.have_fixed) {
+
-			printk(KERN_INFO "MTRR fixed ranges %sabled:\n",
-			       mtrr_state.enabled & 1 ? "en" : "dis");
-			print_fixed(0x00000, 0x10000, mtrr_state.fixed_ranges + 0);
-			for (i = 0; i < 2; ++i)
-				print_fixed(0x80000 + i * 0x20000, 0x04000, mtrr_state.fixed_ranges + (i + 1) * 8);
-			for (i = 0; i < 8; ++i)
-				print_fixed(0xC0000 + i * 0x08000, 0x01000, mtrr_state.fixed_ranges + (i + 3) * 8);
-		}
-		printk(KERN_INFO "MTRR variable ranges %sabled:\n",
-		       mtrr_state.enabled & 2 ? "en" : "dis");
-		high_width = ((size_or_mask ? ffs(size_or_mask) - 1 : 32) - (32 - PAGE_SHIFT) + 3) / 4;
-		for (i = 0; i < num_var_ranges; ++i) {
-			if (mtrr_state.var_ranges[i].mask_lo & (1 << 11))
-				printk(KERN_INFO "MTRR %u base %0*X%05X000 mask %0*X%05X000 %s\n",
-				       i,
-				       high_width,
-				       mtrr_state.var_ranges[i].base_hi,
-				       mtrr_state.var_ranges[i].base_lo >> 12,
-				       high_width,
-				       mtrr_state.var_ranges[i].mask_hi,
-				       mtrr_state.var_ranges[i].mask_lo >> 12,
-				       mtrr_attrib_to_str(mtrr_state.var_ranges[i].base_lo & 0xff));
-			else
-				printk(KERN_INFO "MTRR %u disabled\n", i);
-		}
-		if (mtrr_tom2) {
-			printk(KERN_INFO "TOM2: %016llx aka %lldM\n",
-					  mtrr_tom2, mtrr_tom2>>20);
-		}
-	}
 	mtrr_state_set = 1;
 
 	/* PAT setup for BP. We need to go through sync steps here */
@@ -307,28 +371,11 @@ void mtrr_wrmsr(unsigned msr, unsigned a, unsigned b)
 			smp_processor_id(), msr, a, b);
 }
 
-/**
- * Enable and allow read/write of extended fixed-range MTRR bits on K8 CPUs
- * see AMD publication no. 24593, chapter 3.2.1 for more information
- */
-static inline void k8_enable_fixed_iorrs(void)
-{
-	unsigned lo, hi;
-
-	rdmsr(MSR_K8_SYSCFG, lo, hi);
-	mtrr_wrmsr(MSR_K8_SYSCFG, lo
-				| K8_MTRRFIXRANGE_DRAM_ENABLE
-				| K8_MTRRFIXRANGE_DRAM_MODIFY, hi);
-}
-
 /**
  * set_fixed_range - checks & updates a fixed-range MTRR if it differs from the value it should have
  * @msr: MSR address of the MTTR which should be checked and updated
  * @changed: pointer which indicates whether the MTRR needed to be changed
  * @msrwords: pointer to the MSR values which the MSR should have
- *
- * If K8 extentions are wanted, update the K8 SYSCFG MSR also.
- * See AMD publication no. 24593, chapter 7.8.1, page 233 for more information.
  */
 static void set_fixed_range(int msr, bool *changed, unsigned int *msrwords)
 {
@@ -337,10 +384,6 @@ static void set_fixed_range(int msr, bool *changed, unsigned int *msrwords)
 	rdmsr(msr, lo, hi);
 
 	if (lo != msrwords[0] || hi != msrwords[1]) {
-		if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD &&
-		    (boot_cpu_data.x86 >= 0x0f && boot_cpu_data.x86 <= 0x11) &&
-		    ((msrwords[0] | msrwords[1]) & K8_MTRR_RDMEM_WRMEM_MASK))
-			k8_enable_fixed_iorrs();
 		mtrr_wrmsr(msr, msrwords[0], msrwords[1]);
 		*changed = true;
 	}
@@ -376,22 +419,31 @@ static void generic_get_mtrr(unsigned int reg, unsigned long *base,
 {
 	unsigned int mask_lo, mask_hi, base_lo, base_hi;
 	unsigned int tmp, hi;
+	int cpu;
+
+	/*
+	 * get_mtrr doesn't need to update mtrr_state, also it could be called
+	 * from any cpu, so try to print it out directly.
+	 */
+	cpu = get_cpu();
 
 	rdmsr(MTRRphysMask_MSR(reg), mask_lo, mask_hi);
+
 	if ((mask_lo & 0x800) == 0) {
 		/*  Invalid (i.e. free) range  */
 		*base = 0;
 		*size = 0;
 		*type = 0;
-		return;
+		goto out_put_cpu;
 	}
 
 	rdmsr(MTRRphysBase_MSR(reg), base_lo, base_hi);
 
-	/* Work out the shifted address mask. */
+	/* Work out the shifted address mask: */
 	tmp = mask_hi << (32 - PAGE_SHIFT) | mask_lo >> PAGE_SHIFT;
 	mask_lo = size_or_mask | tmp;
-	/* Expand tmp with high bits to all 1s*/
+
+	/* Expand tmp with high bits to all 1s: */
 	hi = fls(tmp);
 	if (hi > 0) {
 		tmp |= ~((1<<(hi - 1)) - 1);
@@ -402,11 +454,19 @@ static void generic_get_mtrr(unsigned int reg, unsigned long *base,
 		}
 	}
 
-	/* This works correctly if size is a power of two, i.e. a
+	/*
-	   contiguous range. */
+	 * This works correctly if size is a power of two, i.e. a
+	 * contiguous range:
+	 */
 	*size = -mask_lo;
 	*base = base_hi << (32 - PAGE_SHIFT) | base_lo >> PAGE_SHIFT;
 	*type = base_lo & 0xff;
+
+	printk(KERN_DEBUG "  get_mtrr: cpu%d reg%02d base=%010lx size=%010lx %s\n",
+			cpu, reg, *base, *size,
+			mtrr_attrib_to_str(*type & 0xff));
+out_put_cpu:
+	put_cpu();
 }
 
 /**
@@ -419,6 +479,8 @@ static int set_fixed_ranges(mtrr_type * frs)
 	bool changed = false;
 	int block=-1, range;
 
+	k8_check_syscfg_dram_mod_en();
+
 	while (fixed_range_blocks[++block].ranges)
 	    for (range=0; range < fixed_range_blocks[block].ranges; range++)
 		set_fixed_range(fixed_range_blocks[block].base_msr + range,

arch/x86/kernel/cpu/mtrr/mtrr.h

@@ -79,6 +79,7 @@ extern struct mtrr_ops * mtrr_if;
 
 extern unsigned int num_var_ranges;
 extern u64 mtrr_tom2;
+extern struct mtrr_state_type mtrr_state;
 
 void mtrr_state_warn(void);
 const char *mtrr_attrib_to_str(int x);
@@ -88,3 +89,6 @@ void mtrr_wrmsr(unsigned, unsigned, unsigned);
 int amd_init_mtrr(void);
 int cyrix_init_mtrr(void);
 int centaur_init_mtrr(void);
+
+extern int changed_by_mtrr_cleanup;
+extern int mtrr_cleanup(unsigned address_bits);

arch/x86/kernel/cpu/transmeta.c

@@ -98,7 +98,7 @@ static void __cpuinit init_transmeta(struct cpuinfo_x86 *c)
 #endif
 }
 
-static struct cpu_dev transmeta_cpu_dev __cpuinitdata = {
+static const struct cpu_dev __cpuinitconst transmeta_cpu_dev = {
 	.c_vendor	= "Transmeta",
 	.c_ident	= { "GenuineTMx86", "TransmetaCPU" },
 	.c_early_init	= early_init_transmeta,

arch/x86/kernel/cpu/umc.c

@@ -8,7 +8,7 @@
  * so no special init takes place.
  */
 
-static struct cpu_dev umc_cpu_dev __cpuinitdata = {
+static const struct cpu_dev __cpuinitconst umc_cpu_dev = {
 	.c_vendor	= "UMC",
 	.c_ident	= { "UMC UMC UMC" },
 	.c_models = {

arch/x86/kernel/e820.c

@@ -110,19 +110,50 @@ int __init e820_all_mapped(u64 start, u64 end, unsigned type)
 /*
  * Add a memory region to the kernel e820 map.
  */
-void __init e820_add_region(u64 start, u64 size, int type)
+static void __init __e820_add_region(struct e820map *e820x, u64 start, u64 size,
+					 int type)
 {
-	int x = e820.nr_map;
+	int x = e820x->nr_map;
 
-	if (x == ARRAY_SIZE(e820.map)) {
+	if (x == ARRAY_SIZE(e820x->map)) {
 		printk(KERN_ERR "Ooops! Too many entries in the memory map!\n");
 		return;
 	}
 
-	e820.map[x].addr = start;
+	e820x->map[x].addr = start;
-	e820.map[x].size = size;
+	e820x->map[x].size = size;
-	e820.map[x].type = type;
+	e820x->map[x].type = type;
-	e820.nr_map++;
+	e820x->nr_map++;
+}
+
+void __init e820_add_region(u64 start, u64 size, int type)
+{
+	__e820_add_region(&e820, start, size, type);
+}
+
+static void __init e820_print_type(u32 type)
+{
+	switch (type) {
+	case E820_RAM:
+	case E820_RESERVED_KERN:
+		printk(KERN_CONT "(usable)");
+		break;
+	case E820_RESERVED:
+		printk(KERN_CONT "(reserved)");
+		break;
+	case E820_ACPI:
+		printk(KERN_CONT "(ACPI data)");
+		break;
+	case E820_NVS:
+		printk(KERN_CONT "(ACPI NVS)");
+		break;
+	case E820_UNUSABLE:
+		printk(KERN_CONT "(unusable)");
+		break;
+	default:
+		printk(KERN_CONT "type %u", type);
+		break;
+	}
 }
 
 void __init e820_print_map(char *who)
@@ -134,27 +165,8 @@ void __init e820_print_map(char *who)
 		       (unsigned long long) e820.map[i].addr,
 		       (unsigned long long)
 		       (e820.map[i].addr + e820.map[i].size));
-		switch (e820.map[i].type) {
+		e820_print_type(e820.map[i].type);
-		case E820_RAM:
+		printk(KERN_CONT "\n");
-		case E820_RESERVED_KERN:
-			printk(KERN_CONT "(usable)\n");
-			break;
-		case E820_RESERVED:
-			printk(KERN_CONT "(reserved)\n");
-			break;
-		case E820_ACPI:
-			printk(KERN_CONT "(ACPI data)\n");
-			break;
-		case E820_NVS:
-			printk(KERN_CONT "(ACPI NVS)\n");
-			break;
-		case E820_UNUSABLE:
-			printk("(unusable)\n");
-			break;
-		default:
-			printk(KERN_CONT "type %u\n", e820.map[i].type);
-			break;
-		}
 	}
 }
 
@@ -221,7 +233,7 @@ void __init e820_print_map(char *who)
  */
 
 int __init sanitize_e820_map(struct e820entry *biosmap, int max_nr_map,
-				int *pnr_map)
+			     u32 *pnr_map)
 {
 	struct change_member {
 		struct e820entry *pbios; /* pointer to original bios entry */
@@ -417,11 +429,12 @@ static int __init append_e820_map(struct e820entry *biosmap, int nr_map)
 	return __append_e820_map(biosmap, nr_map);
 }
 
-static u64 __init e820_update_range_map(struct e820map *e820x, u64 start,
+static u64 __init __e820_update_range(struct e820map *e820x, u64 start,
 					u64 size, unsigned old_type,
 					unsigned new_type)
 {
-	int i;
+	u64 end;
+	unsigned int i;
 	u64 real_updated_size = 0;
 
 	BUG_ON(old_type == new_type);
@@ -429,27 +442,55 @@ static u64 __init e820_update_range_map(struct e820map *e820x, u64 start,
 	if (size > (ULLONG_MAX - start))
 		size = ULLONG_MAX - start;
 
-	for (i = 0; i < e820.nr_map; i++) {
+	end = start + size;
+	printk(KERN_DEBUG "e820 update range: %016Lx - %016Lx ",
+		       (unsigned long long) start,
+		       (unsigned long long) end);
+	e820_print_type(old_type);
+	printk(KERN_CONT " ==> ");
+	e820_print_type(new_type);
+	printk(KERN_CONT "\n");
+
+	for (i = 0; i < e820x->nr_map; i++) {
 		struct e820entry *ei = &e820x->map[i];
 		u64 final_start, final_end;
+		u64 ei_end;
+
 		if (ei->type != old_type)
 			continue;
-		/* totally covered? */
+
-		if (ei->addr >= start &&
+		ei_end = ei->addr + ei->size;
-		    (ei->addr + ei->size) <= (start + size)) {
+		/* totally covered by new range? */
+		if (ei->addr >= start && ei_end <= end) {
 			ei->type = new_type;
 			real_updated_size += ei->size;
 			continue;
 		}
+
+		/* new range is totally covered? */
+		if (ei->addr < start && ei_end > end) {
+			__e820_add_region(e820x, start, size, new_type);
+			__e820_add_region(e820x, end, ei_end - end, ei->type);
+			ei->size = start - ei->addr;
+			real_updated_size += size;
+			continue;
+		}
+
 		/* partially covered */
 		final_start = max(start, ei->addr);
-		final_end = min(start + size, ei->addr + ei->size);
+		final_end = min(end, ei_end);
 		if (final_start >= final_end)
 			continue;
-		e820_add_region(final_start, final_end - final_start,
+
-					 new_type);
+		__e820_add_region(e820x, final_start, final_end - final_start,
+				  new_type);
+
 		real_updated_size += final_end - final_start;
 
+		/*
+		 * left range could be head or tail, so need to update
+		 * size at first.
+		 */
 		ei->size -= final_end - final_start;
 		if (ei->addr < final_start)
 			continue;
@@ -461,13 +502,13 @@ static u64 __init e820_update_range_map(struct e820map *e820x, u64 start,
 u64 __init e820_update_range(u64 start, u64 size, unsigned old_type,
 			     unsigned new_type)
 {
-	return e820_update_range_map(&e820, start, size, old_type, new_type);
+	return __e820_update_range(&e820, start, size, old_type, new_type);
 }
 
 static u64 __init e820_update_range_saved(u64 start, u64 size,
 					  unsigned old_type, unsigned new_type)
 {
-	return e820_update_range_map(&e820_saved, start, size, old_type,
+	return __e820_update_range(&e820_saved, start, size, old_type,
 				     new_type);
 }
 
@@ -511,7 +552,7 @@ u64 __init e820_remove_range(u64 start, u64 size, unsigned old_type,
 
 void __init update_e820(void)
 {
-	int nr_map;
+	u32 nr_map;
 
 	nr_map = e820.nr_map;
 	if (sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &nr_map))
@@ -522,7 +563,7 @@ void __init update_e820(void)
 }
 static void __init update_e820_saved(void)
 {
-	int nr_map;
+	u32 nr_map;
 
 	nr_map = e820_saved.nr_map;
 	if (sanitize_e820_map(e820_saved.map, ARRAY_SIZE(e820_saved.map), &nr_map))
@@ -1020,8 +1061,8 @@ u64 __init find_e820_area_size(u64 start, u64 *sizep, u64 align)
 			continue;
 		return addr;
 	}
-	return -1UL;
 
+	return -1ULL;
 }
 
 /*
@@ -1034,13 +1075,22 @@ u64 __init early_reserve_e820(u64 startt, u64 sizet, u64 align)
 	u64 start;
 
 	start = startt;
-	while (size < sizet)
+	while (size < sizet && (start + 1))
 		start = find_e820_area_size(start, &size, align);
 
 	if (size < sizet)
 		return 0;
 
+#ifdef CONFIG_X86_32
+	if (start >= MAXMEM)
+		return 0;
+	if (start + size > MAXMEM)
+		size = MAXMEM - start;
+#endif
+
 	addr = round_down(start + size - sizet, align);
+	if (addr < start)
+		return 0;
 	e820_update_range(addr, sizet, E820_RAM, E820_RESERVED);
 	e820_update_range_saved(addr, sizet, E820_RAM, E820_RESERVED);
 	printk(KERN_INFO "update e820 for early_reserve_e820\n");
@@ -1253,7 +1303,7 @@ early_param("memmap", parse_memmap_opt);
 void __init finish_e820_parsing(void)
 {
 	if (userdef) {
-		int nr = e820.nr_map;
+		u32 nr = e820.nr_map;
 
 		if (sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &nr) < 0)
 			early_panic("Invalid user supplied memory map");
@@ -1336,7 +1386,7 @@ void __init e820_reserve_resources_late(void)
 char *__init default_machine_specific_memory_setup(void)
 {
 	char *who = "BIOS-e820";
-	int new_nr;
+	u32 new_nr;
 	/*
 	 * Try to copy the BIOS-supplied E820-map.
 	 *

arch/x86/kernel/early_printk.c

@@ -250,7 +250,7 @@ static int dbgp_wait_until_complete(void)
 	return (ctrl & DBGP_ERROR) ? -DBGP_ERRCODE(ctrl) : DBGP_LEN(ctrl);
 }
 
-static void dbgp_mdelay(int ms)
+static void __init dbgp_mdelay(int ms)
 {
 	int i;
 
@@ -311,7 +311,7 @@ static void dbgp_set_data(const void *buf, int size)
 	writel(hi, &ehci_debug->data47);
 }
 
-static void dbgp_get_data(void *buf, int size)
+static void __init dbgp_get_data(void *buf, int size)
 {
 	unsigned char *bytes = buf;
 	u32 lo, hi;
@@ -355,7 +355,7 @@ static int dbgp_bulk_write(unsigned devnum, unsigned endpoint,
 	return ret;
 }
 
-static int dbgp_bulk_read(unsigned devnum, unsigned endpoint, void *data,
+static int __init dbgp_bulk_read(unsigned devnum, unsigned endpoint, void *data,
 				 int size)
 {
 	u32 pids, addr, ctrl;
@@ -386,8 +386,8 @@ static int dbgp_bulk_read(unsigned devnum, unsigned endpoint, void *data,
 	return ret;
 }
 
-static int dbgp_control_msg(unsigned devnum, int requesttype, int request,
+static int __init dbgp_control_msg(unsigned devnum, int requesttype,
-	int value, int index, void *data, int size)
+	int request, int value, int index, void *data, int size)
 {
 	u32 pids, addr, ctrl;
 	struct usb_ctrlrequest req;
@@ -489,7 +489,7 @@ static u32 __init find_dbgp(int ehci_num, u32 *rbus, u32 *rslot, u32 *rfunc)
 	return 0;
 }
 
-static int ehci_reset_port(int port)
+static int __init ehci_reset_port(int port)
 {
 	u32 portsc;
 	u32 delay_time, delay;
@@ -532,7 +532,7 @@ static int ehci_reset_port(int port)
 	return -EBUSY;
 }
 
-static int ehci_wait_for_port(int port)
+static int __init ehci_wait_for_port(int port)
 {
 	u32 status;
 	int ret, reps;
@@ -557,13 +557,13 @@ static inline void dbgp_printk(const char *fmt, ...) { }
 
 typedef void (*set_debug_port_t)(int port);
 
-static void default_set_debug_port(int port)
+static void __init default_set_debug_port(int port)
 {
 }
 
-static set_debug_port_t set_debug_port = default_set_debug_port;
+static set_debug_port_t __initdata set_debug_port = default_set_debug_port;
 
-static void nvidia_set_debug_port(int port)
+static void __init nvidia_set_debug_port(int port)
 {
 	u32 dword;
 	dword = read_pci_config(ehci_dev.bus, ehci_dev.slot, ehci_dev.func,

arch/x86/kernel/entry_32.S

@@ -442,8 +442,7 @@ sysenter_past_esp:
 
 	GET_THREAD_INFO(%ebp)
 
-	/* Note, _TIF_SECCOMP is bit number 8, and so it needs testw and not testb */
+	testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%ebp)
-	testw $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%ebp)
 	jnz sysenter_audit
 sysenter_do_call:
 	cmpl $(nr_syscalls), %eax
@@ -454,7 +453,7 @@ sysenter_do_call:
 	DISABLE_INTERRUPTS(CLBR_ANY)
 	TRACE_IRQS_OFF
 	movl TI_flags(%ebp), %ecx
-	testw $_TIF_ALLWORK_MASK, %cx
+	testl $_TIF_ALLWORK_MASK, %ecx
 	jne sysexit_audit
 sysenter_exit:
 /* if something modifies registers it must also disable sysexit */
@@ -468,7 +467,7 @@ sysenter_exit:
 
 #ifdef CONFIG_AUDITSYSCALL
 sysenter_audit:
-	testw $(_TIF_WORK_SYSCALL_ENTRY & ~_TIF_SYSCALL_AUDIT),TI_flags(%ebp)
+	testl $(_TIF_WORK_SYSCALL_ENTRY & ~_TIF_SYSCALL_AUDIT),TI_flags(%ebp)
 	jnz syscall_trace_entry
 	addl $4,%esp
 	CFI_ADJUST_CFA_OFFSET -4
@@ -485,7 +484,7 @@ sysenter_audit:
 	jmp sysenter_do_call
 
 sysexit_audit:
-	testw $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT), %cx
+	testl $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT), %ecx
 	jne syscall_exit_work
 	TRACE_IRQS_ON
 	ENABLE_INTERRUPTS(CLBR_ANY)
@@ -498,7 +497,7 @@ sysexit_audit:
 	DISABLE_INTERRUPTS(CLBR_ANY)
 	TRACE_IRQS_OFF
 	movl TI_flags(%ebp), %ecx
-	testw $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT), %cx
+	testl $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT), %ecx
 	jne syscall_exit_work
 	movl PT_EAX(%esp),%eax	/* reload syscall return value */
 	jmp sysenter_exit
@@ -523,8 +522,7 @@ ENTRY(system_call)
 	SAVE_ALL
 	GET_THREAD_INFO(%ebp)
 					# system call tracing in operation / emulation
-	/* Note, _TIF_SECCOMP is bit number 8, and so it needs testw and not testb */
+	testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%ebp)
-	testw $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%ebp)
 	jnz syscall_trace_entry
 	cmpl $(nr_syscalls), %eax
 	jae syscall_badsys
@@ -538,7 +536,7 @@ syscall_exit:
 					# between sampling and the iret
 	TRACE_IRQS_OFF
 	movl TI_flags(%ebp), %ecx
-	testw $_TIF_ALLWORK_MASK, %cx	# current->work
+	testl $_TIF_ALLWORK_MASK, %ecx	# current->work
 	jne syscall_exit_work
 
 restore_all:
@@ -673,7 +671,7 @@ END(syscall_trace_entry)
 	# perform syscall exit tracing
 	ALIGN
 syscall_exit_work:
-	testb $_TIF_WORK_SYSCALL_EXIT, %cl
+	testl $_TIF_WORK_SYSCALL_EXIT, %ecx
 	jz work_pending
 	TRACE_IRQS_ON
 	ENABLE_INTERRUPTS(CLBR_ANY)	# could let syscall_trace_leave() call

arch/x86/kernel/entry_64.S

@@ -368,6 +368,7 @@ ENTRY(save_rest)
 END(save_rest)
 
 /* save complete stack frame */
+	.pushsection .kprobes.text, "ax"
 ENTRY(save_paranoid)
 	XCPT_FRAME 1 RDI+8
 	cld
@@ -396,6 +397,7 @@ ENTRY(save_paranoid)
 1:	ret
 	CFI_ENDPROC
 END(save_paranoid)
+	.popsection
 
 /*
  * A newly forked process directly context switches into this address.
@@ -416,7 +418,6 @@ ENTRY(ret_from_fork)
 
 	GET_THREAD_INFO(%rcx)
 
-	CFI_REMEMBER_STATE
 	RESTORE_REST
 
 	testl $3, CS-ARGOFFSET(%rsp)		# from kernel_thread?
@@ -428,7 +429,6 @@ ENTRY(ret_from_fork)
 	RESTORE_TOP_OF_STACK %rdi, -ARGOFFSET
 	jmp ret_from_sys_call			# go to the SYSRET fastpath
 
-	CFI_RESTORE_STATE
 	CFI_ENDPROC
 END(ret_from_fork)
 
@@ -984,6 +984,8 @@ apicinterrupt UV_BAU_MESSAGE \
 #endif
 apicinterrupt LOCAL_TIMER_VECTOR \
 	apic_timer_interrupt smp_apic_timer_interrupt
+apicinterrupt GENERIC_INTERRUPT_VECTOR \
+	generic_interrupt smp_generic_interrupt
 
 #ifdef CONFIG_SMP
 apicinterrupt INVALIDATE_TLB_VECTOR_START+0 \

arch/x86/kernel/head32.c

@@ -18,7 +18,7 @@ void __init i386_start_kernel(void)
 {
 	reserve_trampoline_memory();
 
-	reserve_early(__pa_symbol(&_text), __pa_symbol(&_end), "TEXT DATA BSS");
+	reserve_early(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS");
 
 #ifdef CONFIG_BLK_DEV_INITRD
 	/* Reserve INITRD */
@@ -29,9 +29,6 @@ void __init i386_start_kernel(void)
 		reserve_early(ramdisk_image, ramdisk_end, "RAMDISK");
 	}
 #endif
-	reserve_early(init_pg_tables_start, init_pg_tables_end,
-			"INIT_PG_TABLE");
-
 	reserve_ebda_region();
 
 	/*

arch/x86/kernel/head64.c

@@ -100,7 +100,7 @@ void __init x86_64_start_reservations(char *real_mode_data)
 
 	reserve_trampoline_memory();
 
-	reserve_early(__pa_symbol(&_text), __pa_symbol(&_end), "TEXT DATA BSS");
+	reserve_early(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS");
 
 #ifdef CONFIG_BLK_DEV_INITRD
 	/* Reserve INITRD */

arch/x86/kernel/head_32.S

@@ -38,42 +38,40 @@
 #define X86_VENDOR_ID	new_cpu_data+CPUINFO_x86_vendor_id
 
 /*
- * This is how much memory *in addition to the memory covered up to
+ * This is how much memory in addition to the memory covered up to
- * and including _end* we need mapped initially.
+ * and including _end we need mapped initially.
  * We need:
- *  - one bit for each possible page, but only in low memory, which means
+ *     (KERNEL_IMAGE_SIZE/4096) / 1024 pages (worst case, non PAE)
- *     2^32/4096/8 = 128K worst case (4G/4G split.)
+ *     (KERNEL_IMAGE_SIZE/4096) / 512 + 4 pages (worst case for PAE)
- *  - enough space to map all low memory, which means
- *     (2^32/4096) / 1024 pages (worst case, non PAE)
- *     (2^32/4096) / 512 + 4 pages (worst case for PAE)
- *  - a few pages for allocator use before the kernel pagetable has
- *     been set up
  *
  * Modulo rounding, each megabyte assigned here requires a kilobyte of
  * memory, which is currently unreclaimed.
  *
  * This should be a multiple of a page.
+ *
+ * KERNEL_IMAGE_SIZE should be greater than pa(_end)
+ * and small than max_low_pfn, otherwise will waste some page table entries
  */
-LOW_PAGES = 1<<(32-PAGE_SHIFT_asm)
-
-/*
- * To preserve the DMA pool in PAGEALLOC kernels, we'll allocate
- * pagetables from above the 16MB DMA limit, so we'll have to set
- * up pagetables 16MB more (worst-case):
- */
-#ifdef CONFIG_DEBUG_PAGEALLOC
-LOW_PAGES = LOW_PAGES + 0x1000000
-#endif
 
 #if PTRS_PER_PMD > 1
-PAGE_TABLE_SIZE = (LOW_PAGES / PTRS_PER_PMD) + PTRS_PER_PGD
+#define PAGE_TABLE_SIZE(pages) (((pages) / PTRS_PER_PMD) + PTRS_PER_PGD)
 #else
-PAGE_TABLE_SIZE = (LOW_PAGES / PTRS_PER_PGD)
+#define PAGE_TABLE_SIZE(pages) ((pages) / PTRS_PER_PGD)
 #endif
-BOOTBITMAP_SIZE = LOW_PAGES / 8
-ALLOCATOR_SLOP = 4
 
-INIT_MAP_BEYOND_END = BOOTBITMAP_SIZE + (PAGE_TABLE_SIZE + ALLOCATOR_SLOP)*PAGE_SIZE_asm
+/* Enough space to fit pagetables for the low memory linear map */
+MAPPING_BEYOND_END = \
+	PAGE_TABLE_SIZE(((1<<32) - __PAGE_OFFSET) >> PAGE_SHIFT) << PAGE_SHIFT
+
+/*
+ * Worst-case size of the kernel mapping we need to make:
+ * the worst-case size of the kernel itself, plus the extra we need
+ * to map for the linear map.
+ */
+KERNEL_PAGES = (KERNEL_IMAGE_SIZE + MAPPING_BEYOND_END)>>PAGE_SHIFT
+
+INIT_MAP_SIZE = PAGE_TABLE_SIZE(KERNEL_PAGES) * PAGE_SIZE_asm
+RESERVE_BRK(pagetables, INIT_MAP_SIZE)
 
 /*
  * 32-bit kernel entrypoint; only used by the boot CPU.  On entry,
@@ -166,10 +164,10 @@ num_subarch_entries = (. - subarch_entries) / 4
 
 /*
  * Initialize page tables.  This creates a PDE and a set of page
- * tables, which are located immediately beyond _end.  The variable
+ * tables, which are located immediately beyond __brk_base.  The variable
- * init_pg_tables_end is set up to point to the first "safe" location.
+ * _brk_end is set up to point to the first "safe" location.
  * Mappings are created both at virtual address 0 (identity mapping)
- * and PAGE_OFFSET for up to _end+sizeof(page tables)+INIT_MAP_BEYOND_END.
+ * and PAGE_OFFSET for up to _end.
  *
  * Note that the stack is not yet set up!
  */
@@ -190,8 +188,7 @@ default_entry:
 
 	xorl %ebx,%ebx				/* %ebx is kept at zero */
 
-	movl $pa(pg0), %edi
+	movl $pa(__brk_base), %edi
-	movl %edi, pa(init_pg_tables_start)
 	movl $pa(swapper_pg_pmd), %edx
 	movl $PTE_IDENT_ATTR, %eax
 10:
@@ -209,14 +206,14 @@ default_entry:
 	loop 11b
 
 	/*
-	 * End condition: we must map up to and including INIT_MAP_BEYOND_END
+	 * End condition: we must map up to the end + MAPPING_BEYOND_END.
-	 * bytes beyond the end of our own page tables.
 	 */
-	leal (INIT_MAP_BEYOND_END+PTE_IDENT_ATTR)(%edi),%ebp
+	movl $pa(_end) + MAPPING_BEYOND_END + PTE_IDENT_ATTR, %ebp
 	cmpl %ebp,%eax
 	jb 10b
 1:
-	movl %edi,pa(init_pg_tables_end)
+	addl $__PAGE_OFFSET, %edi
+	movl %edi, pa(_brk_end)
 	shrl $12, %eax
 	movl %eax, pa(max_pfn_mapped)
 
@@ -227,8 +224,7 @@ default_entry:
 
 page_pde_offset = (__PAGE_OFFSET >> 20);
 
-	movl $pa(pg0), %edi
+	movl $pa(__brk_base), %edi
-	movl %edi, pa(init_pg_tables_start)
 	movl $pa(swapper_pg_dir), %edx
 	movl $PTE_IDENT_ATTR, %eax
 10:
@@ -242,14 +238,13 @@ page_pde_offset = (__PAGE_OFFSET >> 20);
 	addl $0x1000,%eax
 	loop 11b
 	/*
-	 * End condition: we must map up to and including INIT_MAP_BEYOND_END
+	 * End condition: we must map up to the end + MAPPING_BEYOND_END.
-	 * bytes beyond the end of our own page tables; the +0x007 is
-	 * the attribute bits
 	 */
-	leal (INIT_MAP_BEYOND_END+PTE_IDENT_ATTR)(%edi),%ebp
+	movl $pa(_end) + MAPPING_BEYOND_END + PTE_IDENT_ATTR, %ebp
 	cmpl %ebp,%eax
 	jb 10b
-	movl %edi,pa(init_pg_tables_end)
+	addl $__PAGE_OFFSET, %edi
+	movl %edi, pa(_brk_end)
 	shrl $12, %eax
 	movl %eax, pa(max_pfn_mapped)
 
@@ -636,6 +631,7 @@ swapper_pg_fixmap:
 	.fill 1024,4,0
 ENTRY(empty_zero_page)
 	.fill 4096,1,0
+
 /*
  * This starts the data section.
  */

arch/x86/kernel/i8253.c

@@ -3,17 +3,17 @@
  *
  */
 #include <linux/clockchips.h>
-#include <linux/init.h>
 #include <linux/interrupt.h>
+#include <linux/spinlock.h>
 #include <linux/jiffies.h>
 #include <linux/module.h>
-#include <linux/spinlock.h>
+#include <linux/delay.h>
+#include <linux/init.h>
+#include <linux/io.h>
 
-#include <asm/smp.h>
-#include <asm/delay.h>
 #include <asm/i8253.h>
-#include <asm/io.h>
 #include <asm/hpet.h>
+#include <asm/smp.h>
 
 DEFINE_SPINLOCK(i8253_lock);
 EXPORT_SYMBOL(i8253_lock);
@@ -40,7 +40,7 @@ static void init_pit_timer(enum clock_event_mode mode,
 {
 	spin_lock(&i8253_lock);
 
-	switch(mode) {
+	switch (mode) {
 	case CLOCK_EVT_MODE_PERIODIC:
 		/* binary, mode 2, LSB/MSB, ch 0 */
 		outb_pit(0x34, PIT_MODE);
@@ -95,7 +95,7 @@ static int pit_next_event(unsigned long delta, struct clock_event_device *evt)
  * registered. This mechanism replaces the previous #ifdef LOCAL_APIC -
  * !using_apic_timer decisions in do_timer_interrupt_hook()
  */
-static struct clock_event_device pit_clockevent = {
+static struct clock_event_device pit_ce = {
 	.name		= "pit",
 	.features	= CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
 	.set_mode	= init_pit_timer,
@@ -114,15 +114,13 @@ void __init setup_pit_timer(void)
 	 * Start pit with the boot cpu mask and make it global after the
 	 * IO_APIC has been initialized.
 	 */
-	pit_clockevent.cpumask = cpumask_of(smp_processor_id());
+	pit_ce.cpumask = cpumask_of(smp_processor_id());
-	pit_clockevent.mult = div_sc(CLOCK_TICK_RATE, NSEC_PER_SEC,
+	pit_ce.mult = div_sc(CLOCK_TICK_RATE, NSEC_PER_SEC, pit_ce.shift);
-				     pit_clockevent.shift);
+	pit_ce.max_delta_ns = clockevent_delta2ns(0x7FFF, &pit_ce);
-	pit_clockevent.max_delta_ns =
+	pit_ce.min_delta_ns = clockevent_delta2ns(0xF, &pit_ce);
-		clockevent_delta2ns(0x7FFF, &pit_clockevent);
+
-	pit_clockevent.min_delta_ns =
+	clockevents_register_device(&pit_ce);
-		clockevent_delta2ns(0xF, &pit_clockevent);
+	global_clock_event = &pit_ce;
-	clockevents_register_device(&pit_clockevent);
-	global_clock_event = &pit_clockevent;
 }
 
 #ifndef CONFIG_X86_64
@@ -133,11 +131,11 @@ void __init setup_pit_timer(void)
  */
 static cycle_t pit_read(void)
 {
+	static int old_count;
+	static u32 old_jifs;
 	unsigned long flags;
 	int count;
 	u32 jifs;
-	static int old_count;
-	static u32 old_jifs;
 
 	spin_lock_irqsave(&i8253_lock, flags);
 	/*
@@ -179,9 +177,9 @@ static cycle_t pit_read(void)
 	 * Previous attempts to handle these cases intelligently were
 	 * buggy, so we just do the simple thing now.
 	 */
-	if (count > old_count && jifs == old_jifs) {
+	if (count > old_count && jifs == old_jifs)
 		count = old_count;
-	}
+
 	old_count = count;
 	old_jifs = jifs;
 
@@ -192,13 +190,13 @@ static cycle_t pit_read(void)
 	return (cycle_t)(jifs * LATCH) + count;
 }
 
-static struct clocksource clocksource_pit = {
+static struct clocksource pit_cs = {
-	.name	= "pit",
+	.name		= "pit",
-	.rating = 110,
+	.rating		= 110,
-	.read	= pit_read,
+	.read		= pit_read,
-	.mask	= CLOCKSOURCE_MASK(32),
+	.mask		= CLOCKSOURCE_MASK(32),
-	.mult	= 0,
+	.mult		= 0,
-	.shift	= 20,
+	.shift		= 20,
 };
 
 static void pit_disable_clocksource(void)
@@ -206,9 +204,9 @@ static void pit_disable_clocksource(void)
 	/*
 	 * Use mult to check whether it is registered or not
 	 */
-	if (clocksource_pit.mult) {
+	if (pit_cs.mult) {
-		clocksource_unregister(&clocksource_pit);
+		clocksource_unregister(&pit_cs);
-		clocksource_pit.mult = 0;
+		pit_cs.mult = 0;
 	}
 }
 
@@ -222,13 +220,13 @@ static int __init init_pit_clocksource(void)
 	  * - when local APIC timer is active (PIT is switched off)
 	  */
 	if (num_possible_cpus() > 1 || is_hpet_enabled() ||
-	    pit_clockevent.mode != CLOCK_EVT_MODE_PERIODIC)
+	    pit_ce.mode != CLOCK_EVT_MODE_PERIODIC)
 		return 0;
 
-	clocksource_pit.mult = clocksource_hz2mult(CLOCK_TICK_RATE,
+	pit_cs.mult = clocksource_hz2mult(CLOCK_TICK_RATE, pit_cs.shift);
-						   clocksource_pit.shift);
+
-	return clocksource_register(&clocksource_pit);
+	return clocksource_register(&pit_cs);
 }
 arch_initcall(init_pit_clocksource);
 
-#endif
+#endif /* !CONFIG_X86_64 */

arch/x86/kernel/io_delay.c

@@ -7,10 +7,10 @@
  */
 #include <linux/kernel.h>
 #include <linux/module.h>
-#include <linux/init.h>
 #include <linux/delay.h>
+#include <linux/init.h>
 #include <linux/dmi.h>
-#include <asm/io.h>
+#include <linux/io.h>
 
 int io_delay_type __read_mostly = CONFIG_DEFAULT_IO_DELAY_TYPE;
 
@@ -47,8 +47,7 @@ EXPORT_SYMBOL(native_io_delay);
 static int __init dmi_io_delay_0xed_port(const struct dmi_system_id *id)
 {
 	if (io_delay_type == CONFIG_IO_DELAY_TYPE_0X80) {
-		printk(KERN_NOTICE "%s: using 0xed I/O delay port\n",
+		pr_notice("%s: using 0xed I/O delay port\n", id->ident);
-			id->ident);
 		io_delay_type = CONFIG_IO_DELAY_TYPE_0XED;
 	}
 
@@ -64,40 +63,40 @@ static struct dmi_system_id __initdata io_delay_0xed_port_dmi_table[] = {
 		.callback	= dmi_io_delay_0xed_port,
 		.ident		= "Compaq Presario V6000",
 		.matches	= {
-			DMI_MATCH(DMI_BOARD_VENDOR, "Quanta"),
+			DMI_MATCH(DMI_BOARD_VENDOR,	"Quanta"),
-			DMI_MATCH(DMI_BOARD_NAME, "30B7")
+			DMI_MATCH(DMI_BOARD_NAME,	"30B7")
 		}
 	},
 	{
 		.callback	= dmi_io_delay_0xed_port,
 		.ident		= "HP Pavilion dv9000z",
 		.matches	= {
-			DMI_MATCH(DMI_BOARD_VENDOR, "Quanta"),
+			DMI_MATCH(DMI_BOARD_VENDOR,	"Quanta"),
-			DMI_MATCH(DMI_BOARD_NAME, "30B9")
+			DMI_MATCH(DMI_BOARD_NAME,	"30B9")
 		}
 	},
 	{
 		.callback	= dmi_io_delay_0xed_port,
 		.ident		= "HP Pavilion dv6000",
 		.matches	= {
-			DMI_MATCH(DMI_BOARD_VENDOR, "Quanta"),
+			DMI_MATCH(DMI_BOARD_VENDOR,	"Quanta"),
-			DMI_MATCH(DMI_BOARD_NAME, "30B8")
+			DMI_MATCH(DMI_BOARD_NAME,	"30B8")
 		}
 	},
 	{
 		.callback	= dmi_io_delay_0xed_port,
 		.ident		= "HP Pavilion tx1000",
 		.matches	= {
-			DMI_MATCH(DMI_BOARD_VENDOR, "Quanta"),
+			DMI_MATCH(DMI_BOARD_VENDOR,	"Quanta"),
-			DMI_MATCH(DMI_BOARD_NAME, "30BF")
+			DMI_MATCH(DMI_BOARD_NAME,	"30BF")
 		}
 	},
 	{
 		.callback	= dmi_io_delay_0xed_port,
 		.ident		= "Presario F700",
 		.matches	= {
-			DMI_MATCH(DMI_BOARD_VENDOR, "Quanta"),
+			DMI_MATCH(DMI_BOARD_VENDOR,	"Quanta"),
-			DMI_MATCH(DMI_BOARD_NAME, "30D3")
+			DMI_MATCH(DMI_BOARD_NAME,	"30D3")
 		}
 	},
 	{ }

arch/x86/kernel/irq.c

@@ -15,6 +15,9 @@
 
 atomic_t irq_err_count;
 
+/* Function pointer for generic interrupt vector handling */
+void (*generic_interrupt_extension)(void) = NULL;
+
 /*
  * 'what should we do if we get a hw irq event on an illegal vector'.
  * each architecture has to answer this themselves.
@@ -42,55 +45,60 @@ void ack_bad_irq(unsigned int irq)
 /*
  * /proc/interrupts printing:
  */
-static int show_other_interrupts(struct seq_file *p)
+static int show_other_interrupts(struct seq_file *p, int prec)
 {
 	int j;
 
-	seq_printf(p, "NMI: ");
+	seq_printf(p, "%*s: ", prec, "NMI");
 	for_each_online_cpu(j)
 		seq_printf(p, "%10u ", irq_stats(j)->__nmi_count);
 	seq_printf(p, "  Non-maskable interrupts\n");
 #ifdef CONFIG_X86_LOCAL_APIC
-	seq_printf(p, "LOC: ");
+	seq_printf(p, "%*s: ", prec, "LOC");
 	for_each_online_cpu(j)
 		seq_printf(p, "%10u ", irq_stats(j)->apic_timer_irqs);
 	seq_printf(p, "  Local timer interrupts\n");
+
+	seq_printf(p, "%*s: ", prec, "SPU");
+	for_each_online_cpu(j)
+		seq_printf(p, "%10u ", irq_stats(j)->irq_spurious_count);
+	seq_printf(p, "  Spurious interrupts\n");
 #endif
+	if (generic_interrupt_extension) {
+		seq_printf(p, "PLT: ");
+		for_each_online_cpu(j)
+			seq_printf(p, "%10u ", irq_stats(j)->generic_irqs);
+		seq_printf(p, "  Platform interrupts\n");
+	}
 #ifdef CONFIG_SMP
-	seq_printf(p, "RES: ");
+	seq_printf(p, "%*s: ", prec, "RES");
 	for_each_online_cpu(j)
 		seq_printf(p, "%10u ", irq_stats(j)->irq_resched_count);
 	seq_printf(p, "  Rescheduling interrupts\n");
-	seq_printf(p, "CAL: ");
+	seq_printf(p, "%*s: ", prec, "CAL");
 	for_each_online_cpu(j)
 		seq_printf(p, "%10u ", irq_stats(j)->irq_call_count);
 	seq_printf(p, "  Function call interrupts\n");
-	seq_printf(p, "TLB: ");
+	seq_printf(p, "%*s: ", prec, "TLB");
 	for_each_online_cpu(j)
 		seq_printf(p, "%10u ", irq_stats(j)->irq_tlb_count);
 	seq_printf(p, "  TLB shootdowns\n");
 #endif
 #ifdef CONFIG_X86_MCE
-	seq_printf(p, "TRM: ");
+	seq_printf(p, "%*s: ", prec, "TRM");
 	for_each_online_cpu(j)
 		seq_printf(p, "%10u ", irq_stats(j)->irq_thermal_count);
 	seq_printf(p, "  Thermal event interrupts\n");
 # ifdef CONFIG_X86_64
-	seq_printf(p, "THR: ");
+	seq_printf(p, "%*s: ", prec, "THR");
 	for_each_online_cpu(j)
 		seq_printf(p, "%10u ", irq_stats(j)->irq_threshold_count);
 	seq_printf(p, "  Threshold APIC interrupts\n");
 # endif
 #endif
-#ifdef CONFIG_X86_LOCAL_APIC
+	seq_printf(p, "%*s: %10u\n", prec, "ERR", atomic_read(&irq_err_count));
-	seq_printf(p, "SPU: ");
-	for_each_online_cpu(j)
-		seq_printf(p, "%10u ", irq_stats(j)->irq_spurious_count);
-	seq_printf(p, "  Spurious interrupts\n");
-#endif
-	seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count));
 #if defined(CONFIG_X86_IO_APIC)
-	seq_printf(p, "MIS: %10u\n", atomic_read(&irq_mis_count));
+	seq_printf(p, "%*s: %10u\n", prec, "MIS", atomic_read(&irq_mis_count));
 #endif
 	return 0;
 }
@@ -98,19 +106,22 @@ static int show_other_interrupts(struct seq_file *p)
 int show_interrupts(struct seq_file *p, void *v)
 {
 	unsigned long flags, any_count = 0;
-	int i = *(loff_t *) v, j;
+	int i = *(loff_t *) v, j, prec;
 	struct irqaction *action;
 	struct irq_desc *desc;
 
 	if (i > nr_irqs)
 		return 0;
 
+	for (prec = 3, j = 1000; prec < 10 && j <= nr_irqs; ++prec)
+		j *= 10;
+
 	if (i == nr_irqs)
-		return show_other_interrupts(p);
+		return show_other_interrupts(p, prec);
 
 	/* print header */
 	if (i == 0) {
-		seq_printf(p, "           ");
+		seq_printf(p, "%*s", prec + 8, "");
 		for_each_online_cpu(j)
 			seq_printf(p, "CPU%-8d", j);
 		seq_putc(p, '\n');
@@ -121,23 +132,15 @@ int show_interrupts(struct seq_file *p, void *v)
 		return 0;
 
 	spin_lock_irqsave(&desc->lock, flags);
-#ifndef CONFIG_SMP
-	any_count = kstat_irqs(i);
-#else
 	for_each_online_cpu(j)
 		any_count |= kstat_irqs_cpu(i, j);
-#endif
 	action = desc->action;
 	if (!action && !any_count)
 		goto out;
 
-	seq_printf(p, "%3d: ", i);
+	seq_printf(p, "%*d: ", prec, i);
-#ifndef CONFIG_SMP
-	seq_printf(p, "%10u ", kstat_irqs(i));
-#else
 	for_each_online_cpu(j)
 		seq_printf(p, "%10u ", kstat_irqs_cpu(i, j));
-#endif
 	seq_printf(p, " %8s", desc->chip->name);
 	seq_printf(p, "-%-8s", desc->name);
 
@@ -162,7 +165,10 @@ u64 arch_irq_stat_cpu(unsigned int cpu)
 
 #ifdef CONFIG_X86_LOCAL_APIC
 	sum += irq_stats(cpu)->apic_timer_irqs;
+	sum += irq_stats(cpu)->irq_spurious_count;
 #endif
+	if (generic_interrupt_extension)
+		sum += irq_stats(cpu)->generic_irqs;
 #ifdef CONFIG_SMP
 	sum += irq_stats(cpu)->irq_resched_count;
 	sum += irq_stats(cpu)->irq_call_count;
@@ -173,9 +179,6 @@ u64 arch_irq_stat_cpu(unsigned int cpu)
 # ifdef CONFIG_X86_64
 	sum += irq_stats(cpu)->irq_threshold_count;
 #endif
-#endif
-#ifdef CONFIG_X86_LOCAL_APIC
-	sum += irq_stats(cpu)->irq_spurious_count;
 #endif
 	return sum;
 }
@@ -226,4 +229,27 @@ unsigned int __irq_entry do_IRQ(struct pt_regs *regs)
 	return 1;
 }
 
+/*
+ * Handler for GENERIC_INTERRUPT_VECTOR.
+ */
+void smp_generic_interrupt(struct pt_regs *regs)
+{
+	struct pt_regs *old_regs = set_irq_regs(regs);
+
+	ack_APIC_irq();
+
+	exit_idle();
+
+	irq_enter();
+
+	inc_irq_stat(generic_irqs);
+
+	if (generic_interrupt_extension)
+		generic_interrupt_extension();
+
+	irq_exit();
+
+	set_irq_regs(old_regs);
+}
+
 EXPORT_SYMBOL_GPL(vector_used_by_percpu_irq);

arch/x86/kernel/irqinit_32.c

@@ -175,6 +175,9 @@ void __init native_init_IRQ(void)
 	/* self generated IPI for local APIC timer */
 	alloc_intr_gate(LOCAL_TIMER_VECTOR, apic_timer_interrupt);
 
+	/* generic IPI for platform specific use */
+	alloc_intr_gate(GENERIC_INTERRUPT_VECTOR, generic_interrupt);
+
 	/* IPI vectors for APIC spurious and error interrupts */
 	alloc_intr_gate(SPURIOUS_APIC_VECTOR, spurious_interrupt);
 	alloc_intr_gate(ERROR_APIC_VECTOR, error_interrupt);

arch/x86/kernel/irqinit_64.c

@@ -147,6 +147,9 @@ static void __init apic_intr_init(void)
 	/* self generated IPI for local APIC timer */
 	alloc_intr_gate(LOCAL_TIMER_VECTOR, apic_timer_interrupt);
 
+	/* generic IPI for platform specific use */
+	alloc_intr_gate(GENERIC_INTERRUPT_VECTOR, generic_interrupt);
+
 	/* IPI vectors for APIC spurious and error interrupts */
 	alloc_intr_gate(SPURIOUS_APIC_VECTOR, spurious_interrupt);
 	alloc_intr_gate(ERROR_APIC_VECTOR, error_interrupt);

arch/x86/kernel/kdebugfs.c

@@ -8,11 +8,11 @@
  */
 #include <linux/debugfs.h>
 #include <linux/uaccess.h>
-#include <linux/stat.h>
+#include <linux/module.h>
 #include <linux/init.h>
+#include <linux/stat.h>
 #include <linux/io.h>
 #include <linux/mm.h>
-#include <linux/module.h>
 
 #include <asm/setup.h>
 
@@ -26,9 +26,8 @@ struct setup_data_node {
 	u32 len;
 };
 
-static ssize_t
+static ssize_t setup_data_read(struct file *file, char __user *user_buf,
-setup_data_read(struct file *file, char __user *user_buf, size_t count,
+			       size_t count, loff_t *ppos)
-		loff_t *ppos)
 {
 	struct setup_data_node *node = file->private_data;
 	unsigned long remain;
@@ -39,20 +38,21 @@ setup_data_read(struct file *file, char __user *user_buf, size_t count,
 
 	if (pos < 0)
 		return -EINVAL;
+
 	if (pos >= node->len)
 		return 0;
 
 	if (count > node->len - pos)
 		count = node->len - pos;
+
 	pa = node->paddr + sizeof(struct setup_data) + pos;
 	pg = pfn_to_page((pa + count - 1) >> PAGE_SHIFT);
 	if (PageHighMem(pg)) {
 		p = ioremap_cache(pa, count);
 		if (!p)
 			return -ENXIO;
-	} else {
+	} else
 		p = __va(pa);
-	}
 
 	remain = copy_to_user(user_buf, p, count);
 
@@ -70,12 +70,13 @@ setup_data_read(struct file *file, char __user *user_buf, size_t count,
 static int setup_data_open(struct inode *inode, struct file *file)
 {
 	file->private_data = inode->i_private;
+
 	return 0;
 }
 
 static const struct file_operations fops_setup_data = {
-	.read =		setup_data_read,
+	.read		= setup_data_read,
-	.open =		setup_data_open,
+	.open		= setup_data_open,
 };
 
 static int __init
@@ -84,57 +85,50 @@ create_setup_data_node(struct dentry *parent, int no,
 {
 	struct dentry *d, *type, *data;
 	char buf[16];
-	int error;
 
 	sprintf(buf, "%d", no);
 	d = debugfs_create_dir(buf, parent);
-	if (!d) {
+	if (!d)
-		error = -ENOMEM;
+		return -ENOMEM;
-		goto err_return;
+
-	}
 	type = debugfs_create_x32("type", S_IRUGO, d, &node->type);
-	if (!type) {
+	if (!type)
-		error = -ENOMEM;
 		goto err_dir;
-	}
+
 	data = debugfs_create_file("data", S_IRUGO, d, node, &fops_setup_data);
-	if (!data) {
+	if (!data)
-		error = -ENOMEM;
 		goto err_type;
-	}
+
 	return 0;
 
 err_type:
 	debugfs_remove(type);
 err_dir:
 	debugfs_remove(d);
-err_return:
+	return -ENOMEM;
-	return error;
 }
 
 static int __init create_setup_data_nodes(struct dentry *parent)
 {
 	struct setup_data_node *node;
 	struct setup_data *data;
-	int error, no = 0;
+	int error = -ENOMEM;
 	struct dentry *d;
 	struct page *pg;
 	u64 pa_data;
+	int no = 0;
 
 	d = debugfs_create_dir("setup_data", parent);
-	if (!d) {
+	if (!d)
-		error = -ENOMEM;
+		return -ENOMEM;
-		goto err_return;
-	}
 
 	pa_data = boot_params.hdr.setup_data;
 
 	while (pa_data) {
 		node = kmalloc(sizeof(*node), GFP_KERNEL);
-		if (!node) {
+		if (!node)
-			error = -ENOMEM;
 			goto err_dir;
-		}
+
 		pg = pfn_to_page((pa_data+sizeof(*data)-1) >> PAGE_SHIFT);
 		if (PageHighMem(pg)) {
 			data = ioremap_cache(pa_data, sizeof(*data));
@@ -143,9 +137,8 @@ static int __init create_setup_data_nodes(struct dentry *parent)
 				error = -ENXIO;
 				goto err_dir;
 			}
-		} else {
+		} else
 			data = __va(pa_data);
-		}
 
 		node->paddr = pa_data;
 		node->type = data->type;
@@ -159,11 +152,11 @@ static int __init create_setup_data_nodes(struct dentry *parent)
 			goto err_dir;
 		no++;
 	}
+
 	return 0;
 
 err_dir:
 	debugfs_remove(d);
-err_return:
 	return error;
 }
 
@@ -175,28 +168,26 @@ static struct debugfs_blob_wrapper boot_params_blob = {
 static int __init boot_params_kdebugfs_init(void)
 {
 	struct dentry *dbp, *version, *data;
-	int error;
+	int error = -ENOMEM;
 
 	dbp = debugfs_create_dir("boot_params", NULL);
-	if (!dbp) {
+	if (!dbp)
-		error = -ENOMEM;
+		return -ENOMEM;
-		goto err_return;
+
-	}
 	version = debugfs_create_x16("version", S_IRUGO, dbp,
 				     &boot_params.hdr.version);
-	if (!version) {
+	if (!version)
-		error = -ENOMEM;
 		goto err_dir;
-	}
+
 	data = debugfs_create_blob("data", S_IRUGO, dbp,
 				   &boot_params_blob);
-	if (!data) {
+	if (!data)
-		error = -ENOMEM;
 		goto err_version;
-	}
+
 	error = create_setup_data_nodes(dbp);
 	if (error)
 		goto err_data;
+
 	return 0;
 
 err_data:
@@ -205,10 +196,9 @@ static int __init boot_params_kdebugfs_init(void)
 	debugfs_remove(version);
 err_dir:
 	debugfs_remove(dbp);
-err_return:
 	return error;
 }
-#endif
+#endif /* CONFIG_DEBUG_BOOT_PARAMS */
 
 static int __init arch_kdebugfs_init(void)
 {

arch/x86/kernel/kprobes.c

@@ -193,7 +193,7 @@ static int __kprobes can_boost(kprobe_opcode_t *opcodes)
 	kprobe_opcode_t opcode;
 	kprobe_opcode_t *orig_opcodes = opcodes;
 
-	if (search_exception_tables(opcodes))
+	if (search_exception_tables((unsigned long)opcodes))
 		return 0;	/* Page fault may occur on this address. */
 
 retry:

arch/x86/kernel/kvm.c

@@ -138,12 +138,6 @@ static void kvm_set_pte_atomic(pte_t *ptep, pte_t pte)
 	kvm_mmu_write(ptep, pte_val(pte));
 }
 
-static void kvm_set_pte_present(struct mm_struct *mm, unsigned long addr,
-				pte_t *ptep, pte_t pte)
-{
-	kvm_mmu_write(ptep, pte_val(pte));
-}
-
 static void kvm_pte_clear(struct mm_struct *mm,
 			  unsigned long addr, pte_t *ptep)
 {
@@ -220,7 +214,6 @@ static void paravirt_ops_setup(void)
 #if PAGETABLE_LEVELS >= 3
 #ifdef CONFIG_X86_PAE
 		pv_mmu_ops.set_pte_atomic = kvm_set_pte_atomic;
-		pv_mmu_ops.set_pte_present = kvm_set_pte_present;
 		pv_mmu_ops.pte_clear = kvm_pte_clear;
 		pv_mmu_ops.pmd_clear = kvm_pmd_clear;
 #endif

arch/x86/kernel/machine_kexec_32.c

@@ -14,12 +14,12 @@
 #include <linux/ftrace.h>
 #include <linux/suspend.h>
 #include <linux/gfp.h>
+#include <linux/io.h>
 
 #include <asm/pgtable.h>
 #include <asm/pgalloc.h>
 #include <asm/tlbflush.h>
 #include <asm/mmu_context.h>
-#include <asm/io.h>
 #include <asm/apic.h>
 #include <asm/cpufeature.h>
 #include <asm/desc.h>
@@ -63,7 +63,7 @@ static void load_segments(void)
 		"\tmovl %%eax,%%fs\n"
 		"\tmovl %%eax,%%gs\n"
 		"\tmovl %%eax,%%ss\n"
-		::: "eax", "memory");
+		: : : "eax", "memory");
 #undef STR
 #undef __STR
 }
@@ -205,7 +205,8 @@ void machine_kexec(struct kimage *image)
 
 	if (image->preserve_context) {
 #ifdef CONFIG_X86_IO_APIC
-		/* We need to put APICs in legacy mode so that we can
+		/*
+		 * We need to put APICs in legacy mode so that we can
 		 * get timer interrupts in second kernel. kexec/kdump
 		 * paths already have calls to disable_IO_APIC() in
 		 * one form or other. kexec jump path also need
@@ -227,7 +228,8 @@ void machine_kexec(struct kimage *image)
 		page_list[PA_SWAP_PAGE] = (page_to_pfn(image->swap_page)
 						<< PAGE_SHIFT);
 
-	/* The segment registers are funny things, they have both a
+	/*
+	 * The segment registers are funny things, they have both a
 	 * visible and an invisible part.  Whenever the visible part is
 	 * set to a specific selector, the invisible part is loaded
 	 * with from a table in memory.  At no other time is the
@@ -237,11 +239,12 @@ void machine_kexec(struct kimage *image)
 	 * segments, before I zap the gdt with an invalid value.
 	 */
 	load_segments();
-	/* The gdt & idt are now invalid.
+	/*
+	 * The gdt & idt are now invalid.
 	 * If you want to load them you must set up your own idt & gdt.
 	 */
-	set_gdt(phys_to_virt(0),0);
+	set_gdt(phys_to_virt(0), 0);
-	set_idt(phys_to_virt(0),0);
+	set_idt(phys_to_virt(0), 0);
 
 	/* now call it */
 	image->start = relocate_kernel_ptr((unsigned long)image->head,

arch/x86/kernel/machine_kexec_64.c

@@ -12,11 +12,47 @@
 #include <linux/reboot.h>
 #include <linux/numa.h>
 #include <linux/ftrace.h>
+#include <linux/io.h>
+#include <linux/suspend.h>
 
 #include <asm/pgtable.h>
 #include <asm/tlbflush.h>
 #include <asm/mmu_context.h>
-#include <asm/io.h>
+
+static int init_one_level2_page(struct kimage *image, pgd_t *pgd,
+				unsigned long addr)
+{
+	pud_t *pud;
+	pmd_t *pmd;
+	struct page *page;
+	int result = -ENOMEM;
+
+	addr &= PMD_MASK;
+	pgd += pgd_index(addr);
+	if (!pgd_present(*pgd)) {
+		page = kimage_alloc_control_pages(image, 0);
+		if (!page)
+			goto out;
+		pud = (pud_t *)page_address(page);
+		memset(pud, 0, PAGE_SIZE);
+		set_pgd(pgd, __pgd(__pa(pud) | _KERNPG_TABLE));
+	}
+	pud = pud_offset(pgd, addr);
+	if (!pud_present(*pud)) {
+		page = kimage_alloc_control_pages(image, 0);
+		if (!page)
+			goto out;
+		pmd = (pmd_t *)page_address(page);
+		memset(pmd, 0, PAGE_SIZE);
+		set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE));
+	}
+	pmd = pmd_offset(pud, addr);
+	if (!pmd_present(*pmd))
+		set_pmd(pmd, __pmd(addr | __PAGE_KERNEL_LARGE_EXEC));
+	result = 0;
+out:
+	return result;
+}
 
 static void init_level2_page(pmd_t *level2p, unsigned long addr)
 {
@@ -83,9 +119,8 @@ static int init_level4_page(struct kimage *image, pgd_t *level4p,
 		}
 		level3p = (pud_t *)page_address(page);
 		result = init_level3_page(image, level3p, addr, last_addr);
-		if (result) {
+		if (result)
 			goto out;
-		}
 		set_pgd(level4p++, __pgd(__pa(level3p) | _KERNPG_TABLE));
 		addr += PGDIR_SIZE;
 	}
@@ -154,6 +189,13 @@ static int init_pgtable(struct kimage *image, unsigned long start_pgtable)
 	int result;
 	level4p = (pgd_t *)__va(start_pgtable);
 	result = init_level4_page(image, level4p, 0, max_pfn << PAGE_SHIFT);
+	if (result)
+		return result;
+	/*
+	 * image->start may be outside 0 ~ max_pfn, for example when
+	 * jump back to original kernel from kexeced kernel
+	 */
+	result = init_one_level2_page(image, level4p, image->start);
 	if (result)
 		return result;
 	return init_transition_pgtable(image, level4p);
@@ -229,20 +271,45 @@ void machine_kexec(struct kimage *image)
 {
 	unsigned long page_list[PAGES_NR];
 	void *control_page;
+	int save_ftrace_enabled;
 
-	tracer_disable();
+#ifdef CONFIG_KEXEC_JUMP
+	if (kexec_image->preserve_context)
+		save_processor_state();
+#endif
+
+	save_ftrace_enabled = __ftrace_enabled_save();
 
 	/* Interrupts aren't acceptable while we reboot */
 	local_irq_disable();
 
+	if (image->preserve_context) {
+#ifdef CONFIG_X86_IO_APIC
+		/*
+		 * We need to put APICs in legacy mode so that we can
+		 * get timer interrupts in second kernel. kexec/kdump
+		 * paths already have calls to disable_IO_APIC() in
+		 * one form or other. kexec jump path also need
+		 * one.
+		 */
+		disable_IO_APIC();
+#endif
+	}
+
 	control_page = page_address(image->control_code_page) + PAGE_SIZE;
-	memcpy(control_page, relocate_kernel, PAGE_SIZE);
+	memcpy(control_page, relocate_kernel, KEXEC_CONTROL_CODE_MAX_SIZE);
 
 	page_list[PA_CONTROL_PAGE] = virt_to_phys(control_page);
+	page_list[VA_CONTROL_PAGE] = (unsigned long)control_page;
 	page_list[PA_TABLE_PAGE] =
 	  (unsigned long)__pa(page_address(image->control_code_page));
 
-	/* The segment registers are funny things, they have both a
+	if (image->type == KEXEC_TYPE_DEFAULT)
+		page_list[PA_SWAP_PAGE] = (page_to_pfn(image->swap_page)
+						<< PAGE_SHIFT);
+
+	/*
+	 * The segment registers are funny things, they have both a
 	 * visible and an invisible part.  Whenever the visible part is
 	 * set to a specific selector, the invisible part is loaded
 	 * with from a table in memory.  At no other time is the
@@ -252,15 +319,25 @@ void machine_kexec(struct kimage *image)
 	 * segments, before I zap the gdt with an invalid value.
 	 */
 	load_segments();
-	/* The gdt & idt are now invalid.
+	/*
+	 * The gdt & idt are now invalid.
 	 * If you want to load them you must set up your own idt & gdt.
 	 */
-	set_gdt(phys_to_virt(0),0);
+	set_gdt(phys_to_virt(0), 0);
-	set_idt(phys_to_virt(0),0);
+	set_idt(phys_to_virt(0), 0);
 
 	/* now call it */
-	relocate_kernel((unsigned long)image->head, (unsigned long)page_list,
+	image->start = relocate_kernel((unsigned long)image->head,
-			image->start);
+				       (unsigned long)page_list,
+				       image->start,
+				       image->preserve_context);
+
+#ifdef CONFIG_KEXEC_JUMP
+	if (kexec_image->preserve_context)
+		restore_processor_state();
+#endif
+
+	__ftrace_enabled_restore(save_ftrace_enabled);
 }
 
 void arch_crash_save_vmcoreinfo(void)

arch/x86/kernel/mmconf-fam10h_64.c

@@ -226,7 +226,7 @@ static int __devinit set_check_enable_amd_mmconf(const struct dmi_system_id *d)
         return 0;
 }
 
-static struct dmi_system_id __devinitdata mmconf_dmi_table[] = {
+static const struct dmi_system_id __cpuinitconst mmconf_dmi_table[] = {
         {
                 .callback = set_check_enable_amd_mmconf,
                 .ident = "Sun Microsystems Machine",

arch/x86/kernel/mpparse.c

@@ -109,9 +109,6 @@ static void __init MP_bus_info(struct mpc_bus *m)
 	} else
 		printk(KERN_WARNING "Unknown bustype %s - ignoring\n", str);
 }
-#endif
-
-#ifdef CONFIG_X86_IO_APIC
 
 static int bad_ioapic(unsigned long address)
 {
@@ -224,8 +221,12 @@ static void __init MP_intsrc_info(struct mpc_intsrc *m)
 	if (++mp_irq_entries == MAX_IRQ_SOURCES)
 		panic("Max # of irq sources exceeded!!\n");
 }
+#else /* CONFIG_X86_IO_APIC */
+static inline void __init MP_bus_info(struct mpc_bus *m) {}
+static inline void __init MP_ioapic_info(struct mpc_ioapic *m) {}
+static inline void __init MP_intsrc_info(struct mpc_intsrc *m) {}
+#endif /* CONFIG_X86_IO_APIC */
 
-#endif
 
 static void __init MP_lintsrc_info(struct mpc_lintsrc *m)
 {
@@ -275,6 +276,20 @@ static int __init smp_check_mpc(struct mpc_table *mpc, char *oem, char *str)
 	return 1;
 }
 
+static void skip_entry(unsigned char **ptr, int *count, int size)
+{
+	*ptr += size;
+	*count += size;
+}
+
+static void __init smp_dump_mptable(struct mpc_table *mpc, unsigned char *mpt)
+{
+	printk(KERN_ERR "Your mptable is wrong, contact your HW vendor!\n"
+		"type %x\n", *mpt);
+	print_hex_dump(KERN_ERR, "  ", DUMP_PREFIX_ADDRESS, 16,
+			1, mpc, mpc->length, 1);
+}
+
 static int __init smp_read_mpc(struct mpc_table *mpc, unsigned early)
 {
 	char str[16];
@@ -310,61 +325,30 @@ static int __init smp_read_mpc(struct mpc_table *mpc, unsigned early)
 	while (count < mpc->length) {
 		switch (*mpt) {
 		case MP_PROCESSOR:
-			{
+			/* ACPI may have already provided this data */
-				struct mpc_cpu *m = (struct mpc_cpu *)mpt;
+			if (!acpi_lapic)
-				/* ACPI may have already provided this data */
+				MP_processor_info((struct mpc_cpu *)mpt);
-				if (!acpi_lapic)
+			skip_entry(&mpt, &count, sizeof(struct mpc_cpu));
-					MP_processor_info(m);
+			break;
-				mpt += sizeof(*m);
-				count += sizeof(*m);
-				break;
-			}
 		case MP_BUS:
-			{
+			MP_bus_info((struct mpc_bus *)mpt);
-				struct mpc_bus *m = (struct mpc_bus *)mpt;
+			skip_entry(&mpt, &count, sizeof(struct mpc_bus));
-#ifdef CONFIG_X86_IO_APIC
+			break;
-				MP_bus_info(m);
-#endif
-				mpt += sizeof(*m);
-				count += sizeof(*m);
-				break;
-			}
 		case MP_IOAPIC:
-			{
+			MP_ioapic_info((struct mpc_ioapic *)mpt);
-#ifdef CONFIG_X86_IO_APIC
+			skip_entry(&mpt, &count, sizeof(struct mpc_ioapic));
-				struct mpc_ioapic *m = (struct mpc_ioapic *)mpt;
+			break;
-				MP_ioapic_info(m);
-#endif
-				mpt += sizeof(struct mpc_ioapic);
-				count += sizeof(struct mpc_ioapic);
-				break;
-			}
 		case MP_INTSRC:
-			{
+			MP_intsrc_info((struct mpc_intsrc *)mpt);
-#ifdef CONFIG_X86_IO_APIC
+			skip_entry(&mpt, &count, sizeof(struct mpc_intsrc));
-				struct mpc_intsrc *m = (struct mpc_intsrc *)mpt;
+			break;
-
-				MP_intsrc_info(m);
-#endif
-				mpt += sizeof(struct mpc_intsrc);
-				count += sizeof(struct mpc_intsrc);
-				break;
-			}
 		case MP_LINTSRC:
-			{
+			MP_lintsrc_info((struct mpc_lintsrc *)mpt);
-				struct mpc_lintsrc *m =
+			skip_entry(&mpt, &count, sizeof(struct mpc_lintsrc));
-				    (struct mpc_lintsrc *)mpt;
+			break;
-				MP_lintsrc_info(m);
-				mpt += sizeof(*m);
-				count += sizeof(*m);
-				break;
-			}
 		default:
 			/* wrong mptable */
-			printk(KERN_ERR "Your mptable is wrong, contact your HW vendor!\n");
+			smp_dump_mptable(mpc, mpt);
-			printk(KERN_ERR "type %x\n", *mpt);
-			print_hex_dump(KERN_ERR, "  ", DUMP_PREFIX_ADDRESS, 16,
-					1, mpc, mpc->length, 1);
 			count = mpc->length;
 			break;
 		}
@@ -558,6 +542,68 @@ static inline void __init construct_default_ISA_mptable(int mpc_default_type)
 
 static struct mpf_intel *mpf_found;
 
+static unsigned long __init get_mpc_size(unsigned long physptr)
+{
+	struct mpc_table *mpc;
+	unsigned long size;
+
+	mpc = early_ioremap(physptr, PAGE_SIZE);
+	size = mpc->length;
+	early_iounmap(mpc, PAGE_SIZE);
+	apic_printk(APIC_VERBOSE, "  mpc: %lx-%lx\n", physptr, physptr + size);
+
+	return size;
+}
+
+static int __init check_physptr(struct mpf_intel *mpf, unsigned int early)
+{
+	struct mpc_table *mpc;
+	unsigned long size;
+
+	size = get_mpc_size(mpf->physptr);
+	mpc = early_ioremap(mpf->physptr, size);
+	/*
+	 * Read the physical hardware table.  Anything here will
+	 * override the defaults.
+	 */
+	if (!smp_read_mpc(mpc, early)) {
+#ifdef CONFIG_X86_LOCAL_APIC
+		smp_found_config = 0;
+#endif
+		printk(KERN_ERR "BIOS bug, MP table errors detected!...\n"
+			"... disabling SMP support. (tell your hw vendor)\n");
+		early_iounmap(mpc, size);
+		return -1;
+	}
+	early_iounmap(mpc, size);
+
+	if (early)
+		return -1;
+
+#ifdef CONFIG_X86_IO_APIC
+	/*
+	 * If there are no explicit MP IRQ entries, then we are
+	 * broken.  We set up most of the low 16 IO-APIC pins to
+	 * ISA defaults and hope it will work.
+	 */
+	if (!mp_irq_entries) {
+		struct mpc_bus bus;
+
+		printk(KERN_ERR "BIOS bug, no explicit IRQ entries, "
+		       "using default mptable. (tell your hw vendor)\n");
+
+		bus.type = MP_BUS;
+		bus.busid = 0;
+		memcpy(bus.bustype, "ISA   ", 6);
+		MP_bus_info(&bus);
+
+		construct_default_ioirq_mptable(0);
+	}
+#endif
+
+	return 0;
+}
+
 /*
  * Scan the memory blocks for an SMP configuration block.
  */
@@ -611,45 +657,8 @@ static void __init __get_smp_config(unsigned int early)
 		construct_default_ISA_mptable(mpf->feature1);
 
 	} else if (mpf->physptr) {
-
+		if (check_physptr(mpf, early))
-		/*
-		 * Read the physical hardware table.  Anything here will
-		 * override the defaults.
-		 */
-		if (!smp_read_mpc(phys_to_virt(mpf->physptr), early)) {
-#ifdef CONFIG_X86_LOCAL_APIC
-			smp_found_config = 0;
-#endif
-			printk(KERN_ERR
-			       "BIOS bug, MP table errors detected!...\n");
-			printk(KERN_ERR "... disabling SMP support. "
-			       "(tell your hw vendor)\n");
 			return;
-		}
-
-		if (early)
-			return;
-#ifdef CONFIG_X86_IO_APIC
-		/*
-		 * If there are no explicit MP IRQ entries, then we are
-		 * broken.  We set up most of the low 16 IO-APIC pins to
-		 * ISA defaults and hope it will work.
-		 */
-		if (!mp_irq_entries) {
-			struct mpc_bus bus;
-
-			printk(KERN_ERR "BIOS bug, no explicit IRQ entries, "
-			       "using default mptable. "
-			       "(tell your hw vendor)\n");
-
-			bus.type = MP_BUS;
-			bus.busid = 0;
-			memcpy(bus.bustype, "ISA   ", 6);
-			MP_bus_info(&bus);
-
-			construct_default_ioirq_mptable(0);
-		}
-#endif
 	} else
 		BUG();
 
@@ -670,6 +679,31 @@ void __init get_smp_config(void)
 	__get_smp_config(0);
 }
 
+static void smp_reserve_bootmem(struct mpf_intel *mpf)
+{
+	unsigned long size = get_mpc_size(mpf->physptr);
+#ifdef CONFIG_X86_32
+	/*
+	 * We cannot access to MPC table to compute table size yet,
+	 * as only few megabytes from the bottom is mapped now.
+	 * PC-9800's MPC table places on the very last of physical
+	 * memory; so that simply reserving PAGE_SIZE from mpf->physptr
+	 * yields BUG() in reserve_bootmem.
+	 * also need to make sure physptr is below than max_low_pfn
+	 * we don't need reserve the area above max_low_pfn
+	 */
+	unsigned long end = max_low_pfn * PAGE_SIZE;
+
+	if (mpf->physptr < end) {
+		if (mpf->physptr + size > end)
+			size = end - mpf->physptr;
+		reserve_bootmem_generic(mpf->physptr, size, BOOTMEM_DEFAULT);
+	}
+#else
+	reserve_bootmem_generic(mpf->physptr, size, BOOTMEM_DEFAULT);
+#endif
+}
+
 static int __init smp_scan_config(unsigned long base, unsigned long length,
 				  unsigned reserve)
 {
@@ -697,36 +731,10 @@ static int __init smp_scan_config(unsigned long base, unsigned long length,
 
 			if (!reserve)
 				return 1;
-			reserve_bootmem_generic(virt_to_phys(mpf), PAGE_SIZE,
+			reserve_bootmem_generic(virt_to_phys(mpf), sizeof(*mpf),
-					BOOTMEM_DEFAULT);
-			if (mpf->physptr) {
-				unsigned long size = PAGE_SIZE;
-#ifdef CONFIG_X86_32
-				/*
-				 * We cannot access to MPC table to compute
-				 * table size yet, as only few megabytes from
-				 * the bottom is mapped now.
-				 * PC-9800's MPC table places on the very last
-				 * of physical memory; so that simply reserving
-				 * PAGE_SIZE from mpf->physptr yields BUG()
-				 * in reserve_bootmem.
-				 * also need to make sure physptr is below than
-				 * max_low_pfn
-				 * we don't need reserve the area above max_low_pfn
-				 */
-				unsigned long end = max_low_pfn * PAGE_SIZE;
-
-				if (mpf->physptr < end) {
-					if (mpf->physptr + size > end)
-						size = end - mpf->physptr;
-					reserve_bootmem_generic(mpf->physptr, size,
-							BOOTMEM_DEFAULT);
-				}
-#else
-				reserve_bootmem_generic(mpf->physptr, size,
 						BOOTMEM_DEFAULT);
-#endif
+			if (mpf->physptr)
-			}
+				smp_reserve_bootmem(mpf);
 
 			return 1;
 		}
@@ -829,7 +837,57 @@ static int  __init get_MP_intsrc_index(struct mpc_intsrc *m)
 #define SPARE_SLOT_NUM 20
 
 static struct mpc_intsrc __initdata *m_spare[SPARE_SLOT_NUM];
-#endif
+
+static void check_irq_src(struct mpc_intsrc *m, int *nr_m_spare)
+{
+	int i;
+
+	apic_printk(APIC_VERBOSE, "OLD ");
+	print_MP_intsrc_info(m);
+
+	i = get_MP_intsrc_index(m);
+	if (i > 0) {
+		assign_to_mpc_intsrc(&mp_irqs[i], m);
+		apic_printk(APIC_VERBOSE, "NEW ");
+		print_mp_irq_info(&mp_irqs[i]);
+		return;
+	}
+	if (!i) {
+		/* legacy, do nothing */
+		return;
+	}
+	if (*nr_m_spare < SPARE_SLOT_NUM) {
+		/*
+		 * not found (-1), or duplicated (-2) are invalid entries,
+		 * we need to use the slot later
+		 */
+		m_spare[*nr_m_spare] = m;
+		*nr_m_spare += 1;
+	}
+}
+#else /* CONFIG_X86_IO_APIC */
+static inline void check_irq_src(struct mpc_intsrc *m, int *nr_m_spare) {}
+#endif /* CONFIG_X86_IO_APIC */
+
+static int check_slot(unsigned long mpc_new_phys, unsigned long mpc_new_length,
+		      int count)
+{
+	if (!mpc_new_phys) {
+		pr_info("No spare slots, try to append...take your risk, "
+			"new mpc_length %x\n", count);
+	} else {
+		if (count <= mpc_new_length)
+			pr_info("No spare slots, try to append..., "
+				"new mpc_length %x\n", count);
+		else {
+			pr_err("mpc_new_length %lx is too small\n",
+				mpc_new_length);
+			return -1;
+		}
+	}
+
+	return 0;
+}
 
 static int  __init replace_intsrc_all(struct mpc_table *mpc,
 					unsigned long mpc_new_phys,
@@ -837,77 +895,33 @@ static int  __init replace_intsrc_all(struct mpc_table *mpc,
 {
 #ifdef CONFIG_X86_IO_APIC
 	int i;
-	int nr_m_spare = 0;
 #endif
-
 	int count = sizeof(*mpc);
+	int nr_m_spare = 0;
 	unsigned char *mpt = ((unsigned char *)mpc) + count;
 
 	printk(KERN_INFO "mpc_length %x\n", mpc->length);
 	while (count < mpc->length) {
 		switch (*mpt) {
 		case MP_PROCESSOR:
-			{
+			skip_entry(&mpt, &count, sizeof(struct mpc_cpu));
-				struct mpc_cpu *m = (struct mpc_cpu *)mpt;
+			break;
-				mpt += sizeof(*m);
-				count += sizeof(*m);
-				break;
-			}
 		case MP_BUS:
-			{
+			skip_entry(&mpt, &count, sizeof(struct mpc_bus));
-				struct mpc_bus *m = (struct mpc_bus *)mpt;
+			break;
-				mpt += sizeof(*m);
-				count += sizeof(*m);
-				break;
-			}
 		case MP_IOAPIC:
-			{
+			skip_entry(&mpt, &count, sizeof(struct mpc_ioapic));
-				mpt += sizeof(struct mpc_ioapic);
+			break;
-				count += sizeof(struct mpc_ioapic);
-				break;
-			}
 		case MP_INTSRC:
-			{
+			check_irq_src((struct mpc_intsrc *)mpt, &nr_m_spare);
-#ifdef CONFIG_X86_IO_APIC
+			skip_entry(&mpt, &count, sizeof(struct mpc_intsrc));
-				struct mpc_intsrc *m = (struct mpc_intsrc *)mpt;
+			break;
-
-				printk(KERN_INFO "OLD ");
-				print_MP_intsrc_info(m);
-				i = get_MP_intsrc_index(m);
-				if (i > 0) {
-					assign_to_mpc_intsrc(&mp_irqs[i], m);
-					printk(KERN_INFO "NEW ");
-					print_mp_irq_info(&mp_irqs[i]);
-				} else if (!i) {
-					/* legacy, do nothing */
-				} else if (nr_m_spare < SPARE_SLOT_NUM) {
-					/*
-					 * not found (-1), or duplicated (-2)
-					 * are invalid entries,
-					 * we need to use the slot  later
-					 */
-					m_spare[nr_m_spare] = m;
-					nr_m_spare++;
-				}
-#endif
-				mpt += sizeof(struct mpc_intsrc);
-				count += sizeof(struct mpc_intsrc);
-				break;
-			}
 		case MP_LINTSRC:
-			{
+			skip_entry(&mpt, &count, sizeof(struct mpc_lintsrc));
-				struct mpc_lintsrc *m =
+			break;
-				    (struct mpc_lintsrc *)mpt;
-				mpt += sizeof(*m);
-				count += sizeof(*m);
-				break;
-			}
 		default:
 			/* wrong mptable */
-			printk(KERN_ERR "Your mptable is wrong, contact your HW vendor!\n");
+			smp_dump_mptable(mpc, mpt);
-			printk(KERN_ERR "type %x\n", *mpt);
-			print_hex_dump(KERN_ERR, "  ", DUMP_PREFIX_ADDRESS, 16,
-					1, mpc, mpc->length, 1);
 			goto out;
 		}
 	}
@@ -924,23 +938,15 @@ static int  __init replace_intsrc_all(struct mpc_table *mpc,
 			continue;
 
 		if (nr_m_spare > 0) {
-			printk(KERN_INFO "*NEW* found ");
+			apic_printk(APIC_VERBOSE, "*NEW* found\n");
 			nr_m_spare--;
 			assign_to_mpc_intsrc(&mp_irqs[i], m_spare[nr_m_spare]);
 			m_spare[nr_m_spare] = NULL;
 		} else {
 			struct mpc_intsrc *m = (struct mpc_intsrc *)mpt;
 			count += sizeof(struct mpc_intsrc);
-			if (!mpc_new_phys) {
+			if (!check_slot(mpc_new_phys, mpc_new_length, count))
-				printk(KERN_INFO "No spare slots, try to append...take your risk, new mpc_length %x\n", count);
+				goto out;
-			} else {
-				if (count <= mpc_new_length)
-					printk(KERN_INFO "No spare slots, try to append..., new mpc_length %x\n", count);
-				else {
-					printk(KERN_ERR "mpc_new_length %lx is too small\n", mpc_new_length);
-					goto out;
-				}
-			}
 			assign_to_mpc_intsrc(&mp_irqs[i], m);
 			mpc->length = count;
 			mpt += sizeof(struct mpc_intsrc);