115 files changed, 3540 insertions, 2793 deletions
diff --git a/arch/ppc64/Kconfig b/arch/ppc64/Kconfig
index 2ce87836c671..deca68ad644a 100644
--- a/arch/ppc64/Kconfig
+++ b/arch/ppc64/Kconfig
@@ -44,6 +44,10 @@ config SCHED_NO_NO_OMIT_FRAME_POINTER
        bool
        default y
+config ARCH_MAY_HAVE_PC_FDC
+        bool
+        default y
 # We optimistically allocate largepages from the VM, so make the limit
 # large enough (16MB). This badly named config option is actually
 # max order + 1
@@ -302,12 +306,6 @@ config GENERIC_HARDIRQS
        bool
        default y
-config MSCHUNKS
-        bool
-        depends on PPC_ISERIES
-        default y
 config PPC_RTAS
        bool
        depends on PPC_PSERIES || PPC_BPA
@@ -350,13 +348,46 @@ config SECCOMP
          If unsure, say Y. Only embedded should say N here.
+source "fs/Kconfig.binfmt"
+config HOTPLUG_CPU
+        bool "Support for hot-pluggable CPUs"
+        depends on SMP && EXPERIMENTAL && (PPC_PSERIES || PPC_PMAC)
+        select HOTPLUG
+        ---help---
+          Say Y here to be able to turn CPUs off and on.
+          Say N if you are unsure.
+config PROC_DEVICETREE
+        bool "Support for Open Firmware device tree in /proc"
+        depends on !PPC_ISERIES
+        help
+          This option adds a device-tree directory under /proc which contains
+          an image of the device tree that the kernel copies from Open
+          Firmware. If unsure, say Y here.
+config CMDLINE_BOOL
+        bool "Default bootloader kernel arguments"
+        depends on !PPC_ISERIES
+config CMDLINE
+        string "Initial kernel command string"
+        depends on CMDLINE_BOOL
+        default "console=ttyS0,9600 console=tty0 root=/dev/sda2"
+        help
+          On some platforms, there is currently no way for the boot loader to
+          pass arguments to the kernel. For these platforms, you can supply
+          some command-line options at build time by entering them here.  In
+          most cases you will need to specify the root device here.
 endmenu
 config ISA_DMA_API
        bool
        default y
-menu "General setup"
+menu "Bus Options"
 config ISA
        bool
@@ -389,45 +420,12 @@ config PCI_DOMAINS
        bool
        default PCI
-source "fs/Kconfig.binfmt"
 source "drivers/pci/Kconfig"
-config HOTPLUG_CPU
-        bool "Support for hot-pluggable CPUs"
-        depends on SMP && EXPERIMENTAL && (PPC_PSERIES || PPC_PMAC)
-        select HOTPLUG
-        ---help---
-          Say Y here to be able to turn CPUs off and on.
-          Say N if you are unsure.
 source "drivers/pcmcia/Kconfig"
 source "drivers/pci/hotplug/Kconfig"
-config PROC_DEVICETREE
-        bool "Support for Open Firmware device tree in /proc"
-        depends on !PPC_ISERIES
-        help
-          This option adds a device-tree directory under /proc which contains
-          an image of the device tree that the kernel copies from Open
-          Firmware. If unsure, say Y here.
-config CMDLINE_BOOL
-        bool "Default bootloader kernel arguments"
-        depends on !PPC_ISERIES
-config CMDLINE
-        string "Initial kernel command string"
-        depends on CMDLINE_BOOL
-        default "console=ttyS0,9600 console=tty0 root=/dev/sda2"
-        help
-          On some platforms, there is currently no way for the boot loader to
-          pass arguments to the kernel. For these platforms, you can supply
-          some command-line options at build time by entering them here.  In
-          most cases you will need to specify the root device here.
 endmenu
 source "net/Kconfig"
diff --git a/arch/ppc64/Kconfig.debug b/arch/ppc64/Kconfig.debug
index 46b1ce58da3b..f16a5030527b 100644
--- a/arch/ppc64/Kconfig.debug
+++ b/arch/ppc64/Kconfig.debug
@@ -41,10 +41,19 @@ config XMON
        help
          Include in-kernel hooks for the xmon kernel monitor/debugger.
          Unless you are intending to debug the kernel, say N here.
+          Make sure to enable also CONFIG_BOOTX_TEXT on Macs. Otherwise
+          nothing will appear on the screen (xmon writes directly to the
+          framebuffer memory).
+          The cmdline option 'xmon' or 'xmon=early' will drop into xmon very
+          early during boot. 'xmon=on' will just enable the xmon debugger hooks.
+          'xmon=off' will disable the debugger hooks if CONFIG_XMON_DEFAULT is set.
 config XMON_DEFAULT
        bool "Enable xmon by default"
        depends on XMON
+        help
+          xmon is normally disabled unless booted with 'xmon=on'.
+          Use 'xmon=off' to disable xmon init during runtime.
 config PPCDBG
        bool "Include PPCDBG realtime debugging"
diff --git a/arch/ppc64/Makefile b/arch/ppc64/Makefile
index 731b84758331..17d2c1eac3b8 100644
--- a/arch/ppc64/Makefile
+++ b/arch/ppc64/Makefile
@@ -49,12 +49,14 @@ NM		:= $(NM) --synthetic
 endif
-CHECKFLAGS      += -m64 -D__powerpc__
+CHECKFLAGS      += -m64 -D__powerpc__ -D__powerpc64__
 LDFLAGS         := -m elf64ppc
 LDFLAGS_vmlinux := -Bstatic -e $(KERNELLOAD) -Ttext $(KERNELLOAD)
 CFLAGS          += -msoft-float -pipe -mminimal-toc -mtraceback=none \
                   -mcall-aixdesc
+# Temporary hack until we have migrated to asm-powerpc
+CPPFLAGS        += -Iarch/$(ARCH)/include
 GCC_VERSION     := $(call cc-version)
 GCC_BROKEN_VEC  := $(shell if [ $(GCC_VERSION) -lt 0400 ] ; then echo "y"; fi ;)
@@ -87,11 +89,12 @@ drivers-$(CONFIG_OPROFILE)	+= arch/ppc64/oprofile/
 boot := arch/ppc64/boot
-boottarget-$(CONFIG_PPC_PSERIES) := zImage zImage.initrd
+boottargets-$(CONFIG_PPC_PSERIES) += zImage zImage.initrd
-boottarget-$(CONFIG_PPC_MAPLE) := zImage zImage.initrd
+boottargets-$(CONFIG_PPC_PMAC) += zImage.vmode zImage.initrd.vmode
-boottarget-$(CONFIG_PPC_ISERIES) := vmlinux.sminitrd vmlinux.initrd vmlinux.sm
+boottargets-$(CONFIG_PPC_MAPLE) += zImage zImage.initrd
-boottarget-$(CONFIG_PPC_BPA) := zImage zImage.initrd
+boottargets-$(CONFIG_PPC_ISERIES) += vmlinux.sminitrd vmlinux.initrd vmlinux.sm
-$(boottarget-y): vmlinux
+boottargets-$(CONFIG_PPC_BPA) += zImage zImage.initrd
+$(boottargets-y): vmlinux
        $(Q)$(MAKE) $(build)=$(boot) $(boot)/$@
 bootimage-$(CONFIG_PPC_PSERIES) := $(boot)/zImage
@@ -112,20 +115,21 @@ all: $(KBUILD_IMAGE)
 archclean:
        $(Q)$(MAKE) $(clean)=$(boot)
+        # Temporary hack until we have migrated to asm-powerpc
+        $(Q)rm -rf arch/$(ARCH)/include
-prepare: include/asm-ppc64/offsets.h
-arch/ppc64/kernel/asm-offsets.s: include/asm include/linux/version.h \
+# Temporary hack until we have migrated to asm-powerpc
-                                   include/config/MARKER
+include/asm: arch/$(ARCH)/include/asm
+arch/$(ARCH)/include/asm:
-include/asm-ppc64/offsets.h: arch/ppc64/kernel/asm-offsets.s
+        $(Q)if [ ! -d arch/$(ARCH)/include ]; then mkdir -p arch/$(ARCH)/include; fi
-        $(call filechk,gen-asm-offsets)
+        $(Q)ln -fsn $(srctree)/include/asm-powerpc arch/$(ARCH)/include/asm
 define archhelp
-  echo  '* zImage       - Compressed kernel image (arch/$(ARCH)/boot/zImage)'
+  echo  '  zImage.vmode        - Compressed kernel image (arch/$(ARCH)/boot/zImage.vmode)'
-  echo  '  zImage.initrd- Compressed kernel image with initrd attached,'
+  echo  '  zImage.initrd.vmode - Compressed kernel image with initrd attached,'
-  echo  '                  sourced from arch/$(ARCH)/boot/ramdisk.image.gz'
+  echo  '                        sourced from arch/$(ARCH)/boot/ramdisk.image.gz'
-  echo  '                  (arch/$(ARCH)/boot/zImage.initrd)'
+  echo  '                        (arch/$(ARCH)/boot/zImage.initrd.vmode)'
+  echo  '  zImage              - zImage for pSeries machines'
+  echo  '  zImage.initrd       - zImage with initrd for pSeries machines'
 endef
-CLEAN_FILES += include/asm-ppc64/offsets.h
diff --git a/arch/ppc64/boot/Makefile b/arch/ppc64/boot/Makefile
index 683b2d43c15f..33fdc8710891 100644
--- a/arch/ppc64/boot/Makefile
+++ b/arch/ppc64/boot/Makefile
@@ -22,8 +22,8 @@
 HOSTCC          := gcc
-BOOTCFLAGS      := $(HOSTCFLAGS) $(LINUXINCLUDE) -fno-builtin 
+BOOTCFLAGS      := $(HOSTCFLAGS) -fno-builtin -nostdinc -isystem $(shell $(CROSS32CC) -print-file-name=include)
-BOOTAFLAGS      := -D__ASSEMBLY__ $(BOOTCFLAGS) -traditional
+BOOTAFLAGS      := -D__ASSEMBLY__ $(BOOTCFLAGS) -traditional -nostdinc
 BOOTLFLAGS      := -Ttext 0x00400000 -e _start -T $(srctree)/$(src)/zImage.lds
 OBJCOPYFLAGS    := contents,alloc,load,readonly,data
@@ -37,6 +37,9 @@ quiet_cmd_bootcc = BOOTCC  $@
 quiet_cmd_bootas = BOOTAS  $@
      cmd_bootas = $(CROSS32CC) -Wp,-MD,$(depfile) $(BOOTAFLAGS) -c -o $@ $<
+quiet_cmd_bootld = BOOTLD  $@
+      cmd_bootld = $(CROSS32LD) $(BOOTLFLAGS) -o $@ $(2)
 $(patsubst %.c,%.o, $(filter %.c, $(src-boot))): %.o: %.c
        $(call if_changed_dep,bootcc)
 $(patsubst %.S,%.o, $(filter %.S, $(src-boot))): %.o: %.S
@@ -53,7 +56,7 @@ src-sec = $(foreach section, $(1), $(patsubst %,$(obj)/kernel-%.c, $(section)))
 gz-sec  = $(foreach section, $(1), $(patsubst %,$(obj)/kernel-%.gz, $(section)))
 hostprogs-y             := addnote addRamDisk
-targets                 += zImage zImage.initrd imagesize.c \
+targets                 += zImage.vmode zImage.initrd.vmode zImage zImage.initrd imagesize.c \
                           $(patsubst $(obj)/%,%, $(call obj-sec, $(required) $(initrd))) \
                           $(patsubst $(obj)/%,%, $(call src-sec, $(required) $(initrd))) \
                           $(patsubst $(obj)/%,%, $(call gz-sec, $(required) $(initrd))) \
@@ -63,7 +66,7 @@ extra-y			:= initrd.o
 quiet_cmd_ramdisk = RAMDISK $@
      cmd_ramdisk = $(obj)/addRamDisk $(obj)/ramdisk.image.gz $< $@
-quiet_cmd_stripvm = STRIP $@
+quiet_cmd_stripvm = STRIP   $@
      cmd_stripvm = $(STRIP) -s $< -o $@
 vmlinux.strip: vmlinux FORCE
@@ -71,12 +74,20 @@ vmlinux.strip: vmlinux FORCE
 $(obj)/vmlinux.initrd: vmlinux.strip $(obj)/addRamDisk $(obj)/ramdisk.image.gz FORCE
        $(call if_changed,ramdisk)
-addsection = $(CROSS32OBJCOPY) $(1) \
+quiet_cmd_addsection = ADDSEC  $@
-                --add-section=.kernel:$(strip $(patsubst $(obj)/kernel-%.o,%, $(1)))=$(patsubst %.o,%.gz, $(1)) \
+      cmd_addsection = $(CROSS32OBJCOPY) $@ \
-                --set-section-flags=.kernel:$(strip $(patsubst $(obj)/kernel-%.o,%, $(1)))=$(OBJCOPYFLAGS)
+                --add-section=.kernel:$(strip $(patsubst $(obj)/kernel-%.o,%, $@))=$(patsubst %.o,%.gz, $@) \
+                --set-section-flags=.kernel:$(strip $(patsubst $(obj)/kernel-%.o,%, $@))=$(OBJCOPYFLAGS)
+quiet_cmd_imagesize = GENSIZE $@
+      cmd_imagesize = ls -l vmlinux.strip | \
+                awk '{printf "/* generated -- do not edit! */\n" "unsigned long vmlinux_filesize = %d;\n", $$5}' \
+                > $(obj)/imagesize.c && \
+                $(CROSS_COMPILE)nm -n vmlinux | tail -n 1 | \
+                awk '{printf "unsigned long vmlinux_memsize = 0x%s;\n", substr($$1,8)}' >> $(obj)/imagesize.c
-quiet_cmd_addnote = ADDNOTE $@ 
+quiet_cmd_addnote = ADDNOTE $@
-      cmd_addnote = $(CROSS32LD) $(BOOTLFLAGS) -o $@ $(obj-boot) && $(obj)/addnote $@
+      cmd_addnote = $(obj)/addnote $@
 $(call gz-sec, $(required)): $(obj)/kernel-%.gz: % FORCE
        $(call if_changed,gzip)
@@ -85,28 +96,30 @@ $(obj)/kernel-initrd.gz: $(obj)/ramdisk.image.gz
        cp -f $(obj)/ramdisk.image.gz $@
 $(call src-sec, $(required) $(initrd)): $(obj)/kernel-%.c: $(obj)/kernel-%.gz FORCE
-        touch $@
+        @touch $@
 $(call obj-sec, $(required) $(initrd)): $(obj)/kernel-%.o: $(obj)/kernel-%.c FORCE
        $(call if_changed_dep,bootcc)
-        $(call addsection, $@)
+        $(call cmd,addsection)
+$(obj)/zImage.vmode: obj-boot += $(call obj-sec, $(required))
+$(obj)/zImage.vmode: $(call obj-sec, $(required)) $(obj-boot) FORCE
+        $(call cmd,bootld,$(obj-boot))
+$(obj)/zImage.initrd.vmode: obj-boot += $(call obj-sec, $(required) $(initrd))
+$(obj)/zImage.initrd.vmode: $(call obj-sec, $(required) $(initrd)) $(obj-boot) FORCE
+        $(call cmd,bootld,$(obj-boot))
-$(obj)/zImage: obj-boot += $(call obj-sec, $(required))
+$(obj)/zImage: $(obj)/zImage.vmode $(obj)/addnote FORCE
-$(obj)/zImage: $(call obj-sec, $(required)) $(obj-boot) $(obj)/addnote FORCE
+        @cp -f $< $@
        $(call if_changed,addnote)
-$(obj)/zImage.initrd: obj-boot += $(call obj-sec, $(required) $(initrd))
+$(obj)/zImage.initrd: $(obj)/zImage.initrd.vmode $(obj)/addnote FORCE
-$(obj)/zImage.initrd: $(call obj-sec, $(required) $(initrd)) $(obj-boot) $(obj)/addnote FORCE
+        @cp -f $< $@
        $(call if_changed,addnote)
 $(obj)/imagesize.c: vmlinux.strip
-        @echo Generating $@
+        $(call cmd,imagesize)
-        ls -l vmlinux.strip | \
-        awk '{printf "/* generated -- do not edit! */\n" \
-                "unsigned long vmlinux_filesize = %d;\n", $$5}' > $(obj)/imagesize.c
-        $(CROSS_COMPILE)nm -n vmlinux | tail -n 1 | \
-        awk '{printf "unsigned long vmlinux_memsize = 0x%s;\n", substr($$1,8)}' \
-                >> $(obj)/imagesize.c
 install: $(CONFIGURE) $(BOOTIMAGE)
        sh -x $(srctree)/$(src)/install.sh "$(KERNELRELEASE)" vmlinux System.map "$(INSTALL_PATH)" "$(BOOTIMAGE)"
diff --git a/arch/ppc64/boot/addnote.c b/arch/ppc64/boot/addnote.c
index 719663a694bb..8041a9845ab7 100644
--- a/arch/ppc64/boot/addnote.c
+++ b/arch/ppc64/boot/addnote.c
@@ -157,7 +157,7 @@ main(int ac, char **av)
        PUT_32BE(ns, strlen(arch) + 1);
        PUT_32BE(ns + 4, N_DESCR * 4);
        PUT_32BE(ns + 8, 0x1275);
-        strcpy(&buf[ns + 12], arch);
+        strcpy((char *) &buf[ns + 12], arch);
        ns += 12 + strlen(arch) + 1;
        for (i = 0; i < N_DESCR; ++i, ns += 4)
                PUT_32BE(ns, descr[i]);
@@ -172,7 +172,7 @@ main(int ac, char **av)
        PUT_32BE(ns, strlen(rpaname) + 1);
        PUT_32BE(ns + 4, sizeof(rpanote));
        PUT_32BE(ns + 8, 0x12759999);
-        strcpy(&buf[ns + 12], rpaname);
+        strcpy((char *) &buf[ns + 12], rpaname);
        ns += 12 + ROUNDUP(strlen(rpaname) + 1);
        for (i = 0; i < N_RPA_DESCR; ++i, ns += 4)
                PUT_32BE(ns, rpanote[i]);
diff --git a/arch/ppc64/boot/crt0.S b/arch/ppc64/boot/crt0.S
index 04d3e74cd72f..3861e7f9cf19 100644
--- a/arch/ppc64/boot/crt0.S
+++ b/arch/ppc64/boot/crt0.S
@@ -9,7 +9,7 @@
 * NOTE: this code runs in 32 bit mode and is packaged as ELF32.
 */
-#include <asm/ppc_asm.h>
+#include "ppc_asm.h"
        .text
        .globl  _start
diff --git a/arch/ppc64/boot/div64.S b/arch/ppc64/boot/div64.S
index 38f7e466d7d6..722f360a32a9 100644
--- a/arch/ppc64/boot/div64.S
+++ b/arch/ppc64/boot/div64.S
@@ -13,7 +13,7 @@
 * as published by the Free Software Foundation; either version
 * 2 of the License, or (at your option) any later version.
 */
-#include <asm/ppc_asm.h>
+#include "ppc_asm.h"
        .globl __div64_32
 __div64_32:
diff --git a/arch/ppc64/boot/elf.h b/arch/ppc64/boot/elf.h
new file mode 100644
index 000000000000..d4828fcf1cb9
--- /dev/null
+++ b/arch/ppc64/boot/elf.h
@@ -0,0 +1,149 @@
+#ifndef _PPC_BOOT_ELF_H_
+#define _PPC_BOOT_ELF_H_
+/* 32-bit ELF base types. */
+typedef unsigned int Elf32_Addr;
+typedef unsigned short Elf32_Half;
+typedef unsigned int Elf32_Off;
+typedef signed int Elf32_Sword;
+typedef unsigned int Elf32_Word;
+/* 64-bit ELF base types. */
+typedef unsigned long long Elf64_Addr;
+typedef unsigned short Elf64_Half;
+typedef signed short Elf64_SHalf;
+typedef unsigned long long Elf64_Off;
+typedef signed int Elf64_Sword;
+typedef unsigned int Elf64_Word;
+typedef unsigned long long Elf64_Xword;
+typedef signed long long Elf64_Sxword;
+/* These constants are for the segment types stored in the image headers */
+#define PT_NULL    0
+#define PT_LOAD    1
+#define PT_DYNAMIC 2
+#define PT_INTERP  3
+#define PT_NOTE    4
+#define PT_SHLIB   5
+#define PT_PHDR    6
+#define PT_TLS     7            /* Thread local storage segment */
+#define PT_LOOS    0x60000000   /* OS-specific */
+#define PT_HIOS    0x6fffffff   /* OS-specific */
+#define PT_LOPROC  0x70000000
+#define PT_HIPROC  0x7fffffff
+#define PT_GNU_EH_FRAME         0x6474e550
+#define PT_GNU_STACK    (PT_LOOS + 0x474e551)
+/* These constants define the different elf file types */
+#define ET_NONE   0
+#define ET_REL    1
+#define ET_EXEC   2
+#define ET_DYN    3
+#define ET_CORE   4
+#define ET_LOPROC 0xff00
+#define ET_HIPROC 0xffff
+/* These constants define the various ELF target machines */
+#define EM_NONE  0
+#define EM_PPC         20       /* PowerPC */
+#define EM_PPC64       21       /* PowerPC64 */
+#define EI_NIDENT       16
+typedef struct elf32_hdr {
+        unsigned char e_ident[EI_NIDENT];
+        Elf32_Half e_type;
+        Elf32_Half e_machine;
+        Elf32_Word e_version;
+        Elf32_Addr e_entry;     /* Entry point */
+        Elf32_Off e_phoff;
+        Elf32_Off e_shoff;
+        Elf32_Word e_flags;
+        Elf32_Half e_ehsize;
+        Elf32_Half e_phentsize;
+        Elf32_Half e_phnum;
+        Elf32_Half e_shentsize;
+        Elf32_Half e_shnum;
+        Elf32_Half e_shstrndx;
+} Elf32_Ehdr;
+typedef struct elf64_hdr {
+        unsigned char e_ident[16];      /* ELF "magic number" */
+        Elf64_Half e_type;
+        Elf64_Half e_machine;
+        Elf64_Word e_version;
+        Elf64_Addr e_entry;     /* Entry point virtual address */
+        Elf64_Off e_phoff;      /* Program header table file offset */
+        Elf64_Off e_shoff;      /* Section header table file offset */
+        Elf64_Word e_flags;
+        Elf64_Half e_ehsize;
+        Elf64_Half e_phentsize;
+        Elf64_Half e_phnum;
+        Elf64_Half e_shentsize;
+        Elf64_Half e_shnum;
+        Elf64_Half e_shstrndx;
+} Elf64_Ehdr;
+/* These constants define the permissions on sections in the program
+   header, p_flags. */
+#define PF_R            0x4
+#define PF_W            0x2
+#define PF_X            0x1
+typedef struct elf32_phdr {
+        Elf32_Word p_type;
+        Elf32_Off p_offset;
+        Elf32_Addr p_vaddr;
+        Elf32_Addr p_paddr;
+        Elf32_Word p_filesz;
+        Elf32_Word p_memsz;
+        Elf32_Word p_flags;
+        Elf32_Word p_align;
+} Elf32_Phdr;
+typedef struct elf64_phdr {
+        Elf64_Word p_type;
+        Elf64_Word p_flags;
+        Elf64_Off p_offset;     /* Segment file offset */
+        Elf64_Addr p_vaddr;     /* Segment virtual address */
+        Elf64_Addr p_paddr;     /* Segment physical address */
+        Elf64_Xword p_filesz;   /* Segment size in file */
+        Elf64_Xword p_memsz;    /* Segment size in memory */
+        Elf64_Xword p_align;    /* Segment alignment, file & memory */
+} Elf64_Phdr;
+#define EI_MAG0         0       /* e_ident[] indexes */
+#define EI_MAG1         1
+#define EI_MAG2         2
+#define EI_MAG3         3
+#define EI_CLASS        4
+#define EI_DATA         5
+#define EI_VERSION      6
+#define EI_OSABI        7
+#define EI_PAD          8
+#define ELFMAG0         0x7f    /* EI_MAG */
+#define ELFMAG1         'E'
+#define ELFMAG2         'L'
+#define ELFMAG3         'F'
+#define ELFMAG          "\177ELF"
+#define SELFMAG         4
+#define ELFCLASSNONE    0       /* EI_CLASS */
+#define ELFCLASS32      1
+#define ELFCLASS64      2
+#define ELFCLASSNUM     3
+#define ELFDATANONE     0       /* e_ident[EI_DATA] */
+#define ELFDATA2LSB     1
+#define ELFDATA2MSB     2
+#define EV_NONE         0       /* e_version, EI_VERSION */
+#define EV_CURRENT      1
+#define EV_NUM          2
+#define ELFOSABI_NONE   0
+#define ELFOSABI_LINUX  3
+#endif                          /* _PPC_BOOT_ELF_H_ */
diff --git a/arch/ppc64/boot/main.c b/arch/ppc64/boot/main.c
index 199d9804f61c..f7ec19a2d0b0 100644
--- a/arch/ppc64/boot/main.c
+++ b/arch/ppc64/boot/main.c
@@ -8,38 +8,32 @@
 * as published by the Free Software Foundation; either version
 * 2 of the License, or (at your option) any later version.
 */
-#include "ppc32-types.h"
+#include <stdarg.h>
+#include <stddef.h>
+#include "elf.h"
+#include "page.h"
+#include "string.h"
+#include "stdio.h"
+#include "prom.h"
 #include "zlib.h"
-#include <linux/elf.h>
-#include <linux/string.h>
+static void gunzip(void *, int, unsigned char *, int *);
-#include <asm/processor.h>
+extern void flush_cache(void *, unsigned long);
-#include <asm/page.h>
-extern void *finddevice(const char *);
-extern int getprop(void *, const char *, void *, int);
-extern void printf(const char *fmt, ...);
-extern int sprintf(char *buf, const char *fmt, ...);
-void gunzip(void *, int, unsigned char *, int *);
-void *claim(unsigned int, unsigned int, unsigned int);
-void flush_cache(void *, unsigned long);
-void pause(void);
-extern void exit(void);
-unsigned long strlen(const char *s);
-void *memmove(void *dest, const void *src, unsigned long n);
-void *memcpy(void *dest, const void *src, unsigned long n);
 /* Value picked to match that used by yaboot */
 #define PROG_START      0x01400000
-#define RAM_END         (256<<20) // Fixme: use OF */
+#define RAM_END         (512<<20) // Fixme: use OF */
+#define ONE_MB          0x100000
-char *avail_ram;
+static char *avail_ram;
-char *begin_avail, *end_avail;
+static char *begin_avail, *end_avail;
-char *avail_high;
+static char *avail_high;
-unsigned int heap_use;
+static unsigned int heap_use;
-unsigned int heap_max;
+static unsigned int heap_max;
 extern char _start[];
+extern char _end[];
 extern char _vmlinux_start[];
 extern char _vmlinux_end[];
 extern char _initrd_start[];
@@ -52,9 +46,9 @@ struct addr_range {
        unsigned long size;
        unsigned long memsize;
 };
-struct addr_range vmlinux = {0, 0, 0};
+static struct addr_range vmlinux = {0, 0, 0};
-struct addr_range vmlinuz = {0, 0, 0};
+static struct addr_range vmlinuz = {0, 0, 0};
-struct addr_range initrd  = {0, 0, 0};
+static struct addr_range initrd  = {0, 0, 0};
 static char scratch[128<<10];   /* 128kB of scratch space for gunzip */
@@ -64,22 +58,15 @@ typedef void (*kernel_entry_t)( unsigned long,
                                void *);
-int (*prom)(void *);
-void *chosen_handle;
-void *stdin;
-void *stdout;
-void *stderr;
 #undef DEBUG
-static unsigned long claim_base = PROG_START;
+static unsigned long claim_base;
 static unsigned long try_claim(unsigned long size)
 {
        unsigned long addr = 0;
-        for(; claim_base < RAM_END; claim_base += 0x100000) {
+        for(; claim_base < RAM_END; claim_base += ONE_MB) {
 #ifdef DEBUG
                printf("    trying: 0x%08lx\n\r", claim_base);
 #endif
@@ -110,7 +97,26 @@ void start(unsigned long a1, unsigned long a2, void *promptr)
        if (getprop(chosen_handle, "stdin", &stdin, sizeof(stdin)) != 4)
                exit();
-        printf("\n\rzImage starting: loaded at 0x%x\n\r", (unsigned)_start);
+        printf("\n\rzImage starting: loaded at 0x%lx\n\r", (unsigned long) _start);
+        /*
+         * The first available claim_base must be above the end of the
+         * the loaded kernel wrapper file (_start to _end includes the
+         * initrd image if it is present) and rounded up to a nice
+         * 1 MB boundary for good measure.
+         */
+        claim_base = _ALIGN_UP((unsigned long)_end, ONE_MB);
+#if defined(PROG_START)
+        /*
+         * Maintain a "magic" minimum address. This keeps some older
+         * firmware platforms running.
+         */
+        if (claim_base < PROG_START)
+                claim_base = PROG_START;
+#endif
        /*
         * Now we try to claim some memory for the kernel itself
@@ -120,7 +126,7 @@ void start(unsigned long a1, unsigned long a2, void *promptr)
         * size... In practice we add 1Mb, that is enough, but we should really
         * consider fixing the Makefile to put a _raw_ kernel in there !
         */
-        vmlinux_memsize += 0x100000;
+        vmlinux_memsize += ONE_MB;
        printf("Allocating 0x%lx bytes for kernel ...\n\r", vmlinux_memsize);
        vmlinux.addr = try_claim(vmlinux_memsize);
        if (vmlinux.addr == 0) {
@@ -277,7 +283,7 @@ void zfree(void *x, void *addr, unsigned nb)
 #define DEFLATED        8
-void gunzip(void *dst, int dstlen, unsigned char *src, int *lenp)
+static void gunzip(void *dst, int dstlen, unsigned char *src, int *lenp)
 {
        z_stream s;
        int r, i, flags;
diff --git a/arch/ppc64/boot/page.h b/arch/ppc64/boot/page.h
new file mode 100644
index 000000000000..14eca30fef64
--- /dev/null
+++ b/arch/ppc64/boot/page.h
@@ -0,0 +1,34 @@
+#ifndef _PPC_BOOT_PAGE_H
+#define _PPC_BOOT_PAGE_H
+/*
+ * Copyright (C) 2001 PPC64 Team, IBM Corp
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+#ifdef __ASSEMBLY__
+#define ASM_CONST(x) x
+#else
+#define __ASM_CONST(x) x##UL
+#define ASM_CONST(x) __ASM_CONST(x)
+#endif
+/* PAGE_SHIFT determines the page size */
+#define PAGE_SHIFT      12
+#define PAGE_SIZE       (ASM_CONST(1) << PAGE_SHIFT)
+#define PAGE_MASK       (~(PAGE_SIZE-1))
+/* align addr on a size boundary - adjust address up/down if needed */
+#define _ALIGN_UP(addr,size)    (((addr)+((size)-1))&(~((size)-1)))
+#define _ALIGN_DOWN(addr,size)  ((addr)&(~((size)-1)))
+/* align addr on a size boundary - adjust address up if needed */
+#define _ALIGN(addr,size)     _ALIGN_UP(addr,size)
+/* to align the pointer to the (next) page boundary */
+#define PAGE_ALIGN(addr)        _ALIGN(addr, PAGE_SIZE)
+#endif                          /* _PPC_BOOT_PAGE_H */
diff --git a/arch/ppc64/boot/ppc32-types.h b/arch/ppc64/boot/ppc32-types.h
deleted file mode 100644
index f7b8884f8f70..000000000000
--- a/arch/ppc64/boot/ppc32-types.h
+++ /dev/null
@@ -1,36 +0,0 @@
-#ifndef _PPC64_TYPES_H
-#define _PPC64_TYPES_H
-typedef __signed__ char __s8;
-typedef unsigned char __u8;
-typedef __signed__ short __s16;
-typedef unsigned short __u16;
-typedef __signed__ int __s32;
-typedef unsigned int __u32;
-typedef __signed__ long long __s64;
-typedef unsigned long long __u64;
-typedef signed char s8;
-typedef unsigned char u8;
-typedef signed short s16;
-typedef unsigned short u16;
-typedef signed int s32;
-typedef unsigned int u32;
-typedef signed long long s64;
-typedef unsigned long long u64;
-typedef struct {
-        __u32 u[4];
-} __attribute((aligned(16))) __vector128;
-#define BITS_PER_LONG 32
-typedef __vector128 vector128;
-#endif /* _PPC64_TYPES_H */
diff --git a/arch/ppc64/boot/ppc_asm.h b/arch/ppc64/boot/ppc_asm.h
new file mode 100644
index 000000000000..1c2c2817f9b7
--- /dev/null
+++ b/arch/ppc64/boot/ppc_asm.h
@@ -0,0 +1,62 @@
+#ifndef _PPC64_PPC_ASM_H
+#define _PPC64_PPC_ASM_H
+/*
+ *
+ * Definitions used by various bits of low-level assembly code on PowerPC.
+ *
+ * Copyright (C) 1995-1999 Gary Thomas, Paul Mackerras, Cort Dougan.
+ *
+ *  This program is free software; you can redistribute it and/or
+ *  modify it under the terms of the GNU General Public License
+ *  as published by the Free Software Foundation; either version
+ *  2 of the License, or (at your option) any later version.
+ */
+/* Condition Register Bit Fields */
+#define cr0     0
+#define cr1     1
+#define cr2     2
+#define cr3     3
+#define cr4     4
+#define cr5     5
+#define cr6     6
+#define cr7     7
+/* General Purpose Registers (GPRs) */
+#define r0      0
+#define r1      1
+#define r2      2
+#define r3      3
+#define r4      4
+#define r5      5
+#define r6      6
+#define r7      7
+#define r8      8
+#define r9      9
+#define r10     10
+#define r11     11
+#define r12     12
+#define r13     13
+#define r14     14
+#define r15     15
+#define r16     16
+#define r17     17
+#define r18     18
+#define r19     19
+#define r20     20
+#define r21     21
+#define r22     22
+#define r23     23
+#define r24     24
+#define r25     25
+#define r26     26
+#define r27     27
+#define r28     28
+#define r29     29
+#define r30     30
+#define r31     31
+#endif /* _PPC64_PPC_ASM_H */
diff --git a/arch/ppc64/boot/prom.c b/arch/ppc64/boot/prom.c
index 5e48b80ff5a0..4bea2f4dcb06 100644
--- a/arch/ppc64/boot/prom.c
+++ b/arch/ppc64/boot/prom.c
@@ -7,43 +7,19 @@
 * 2 of the License, or (at your option) any later version.
 */
 #include <stdarg.h>
-#include <linux/types.h>
+#include <stddef.h>
-#include <linux/string.h>
+#include "string.h"
-#include <linux/ctype.h>
+#include "stdio.h"
+#include "prom.h"
-extern __u32 __div64_32(unsigned long long *dividend, __u32 divisor);
-/* The unnecessary pointer compare is there
- * to check for type safety (n must be 64bit)
- */
-# define do_div(n,base) ({                              \
-        __u32 __base = (base);                  \
-        __u32 __rem;                                    \
-        (void)(((typeof((n)) *)0) == ((unsigned long long *)0));        \
-        if (((n) >> 32) == 0) {                 \
-                __rem = (__u32)(n) % __base;            \
-                (n) = (__u32)(n) / __base;              \
-        } else                                          \
-                __rem = __div64_32(&(n), __base);       \
-        __rem;                                          \
- })
 int (*prom)(void *);
 void *chosen_handle;
 void *stdin;
 void *stdout;
 void *stderr;
-void exit(void);
-void *finddevice(const char *name);
-int getprop(void *phandle, const char *name, void *buf, int buflen);
-void chrpboot(int a1, int a2, void *prom);      /* in main.c */
-int printf(char *fmt, ...);
-/* there is no convenient header to get this from...  -- paulus */
-extern unsigned long strlen(const char *);
 int
 write(void *handle, void *ptr, int nb)
@@ -210,107 +186,6 @@ fputs(char *str, void *f)
        return write(f, str, n) == n? 0: -1;
 }
-int
-readchar(void)
-{
-        char ch;
-        for (;;) {
-                switch (read(stdin, &ch, 1)) {
-                case 1:
-                        return ch;
-                case -1:
-                        printf("read(stdin) returned -1\r\n");
-                        return -1;
-                }
-        }
-}
-static char line[256];
-static char *lineptr;
-static int lineleft;
-int
-getchar(void)
-{
-        int c;
-        if (lineleft == 0) {
-                lineptr = line;
-                for (;;) {
-                        c = readchar();
-                        if (c == -1 || c == 4)
-                                break;
-                        if (c == '\r' || c == '\n') {
-                                *lineptr++ = '\n';
-                                putchar('\n');
-                                break;
-                        }
-                        switch (c) {
-                        case 0177:
-                        case '\b':
-                                if (lineptr > line) {
-                                        putchar('\b');
-                                        putchar(' ');
-                                        putchar('\b');
-                                        --lineptr;
-                                }
-                                break;
-                        case 'U' & 0x1F:
-                                while (lineptr > line) {
-                                        putchar('\b');
-                                        putchar(' ');
-                                        putchar('\b');
-                                        --lineptr;
-                                }
-                                break;
-                        default:
-                                if (lineptr >= &line[sizeof(line) - 1])
-                                        putchar('\a');
-                                else {
-                                        putchar(c);
-                                        *lineptr++ = c;
-                                }
-                        }
-                }
-                lineleft = lineptr - line;
-                lineptr = line;
-        }
-        if (lineleft == 0)
-                return -1;
-        --lineleft;
-        return *lineptr++;
-}
-/* String functions lifted from lib/vsprintf.c and lib/ctype.c */
-unsigned char _ctype[] = {
-_C,_C,_C,_C,_C,_C,_C,_C,                        /* 0-7 */
-_C,_C|_S,_C|_S,_C|_S,_C|_S,_C|_S,_C,_C,         /* 8-15 */
-_C,_C,_C,_C,_C,_C,_C,_C,                        /* 16-23 */
-_C,_C,_C,_C,_C,_C,_C,_C,                        /* 24-31 */
-_S|_SP,_P,_P,_P,_P,_P,_P,_P,                    /* 32-39 */
-_P,_P,_P,_P,_P,_P,_P,_P,                        /* 40-47 */
-_D,_D,_D,_D,_D,_D,_D,_D,                        /* 48-55 */
-_D,_D,_P,_P,_P,_P,_P,_P,                        /* 56-63 */
-_P,_U|_X,_U|_X,_U|_X,_U|_X,_U|_X,_U|_X,_U,      /* 64-71 */
-_U,_U,_U,_U,_U,_U,_U,_U,                        /* 72-79 */
-_U,_U,_U,_U,_U,_U,_U,_U,                        /* 80-87 */
-_U,_U,_U,_P,_P,_P,_P,_P,                        /* 88-95 */
-_P,_L|_X,_L|_X,_L|_X,_L|_X,_L|_X,_L|_X,_L,      /* 96-103 */
-_L,_L,_L,_L,_L,_L,_L,_L,                        /* 104-111 */
-_L,_L,_L,_L,_L,_L,_L,_L,                        /* 112-119 */
-_L,_L,_L,_P,_P,_P,_P,_C,                        /* 120-127 */
-0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,                /* 128-143 */
-0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,                /* 144-159 */
-_S|_SP,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,   /* 160-175 */
-_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,_P,       /* 176-191 */
-_U,_U,_U,_U,_U,_U,_U,_U,_U,_U,_U,_U,_U,_U,_U,_U,       /* 192-207 */
-_U,_U,_U,_U,_U,_U,_U,_P,_U,_U,_U,_U,_U,_U,_U,_L,       /* 208-223 */
-_L,_L,_L,_L,_L,_L,_L,_L,_L,_L,_L,_L,_L,_L,_L,_L,       /* 224-239 */
-_L,_L,_L,_L,_L,_L,_L,_P,_L,_L,_L,_L,_L,_L,_L,_L};      /* 240-255 */
 size_t strnlen(const char * s, size_t count)
 {
        const char *sc;
@@ -320,44 +195,30 @@ size_t strnlen(const char * s, size_t count)
        return sc - s;
 }
-unsigned long simple_strtoul(const char *cp,char **endp,unsigned int base)
+extern unsigned int __div64_32(unsigned long long *dividend,
-{
+                               unsigned int divisor);
-        unsigned long result = 0,value;
-        if (!base) {
+/* The unnecessary pointer compare is there
-                base = 10;
+ * to check for type safety (n must be 64bit)
-                if (*cp == '0') {
+ */
-                        base = 8;
+# define do_div(n,base) ({                                              \
-                        cp++;
+        unsigned int __base = (base);                                   \
-                        if ((*cp == 'x') && isxdigit(cp[1])) {
+        unsigned int __rem;                                             \
-                                cp++;
+        (void)(((typeof((n)) *)0) == ((unsigned long long *)0));        \
-                                base = 16;
+        if (((n) >> 32) == 0) {                                         \
-                        }
+                __rem = (unsigned int)(n) % __base;                     \
-                }
+                (n) = (unsigned int)(n) / __base;                       \
-        }
+        } else                                                          \
-        while (isxdigit(*cp) &&
+                __rem = __div64_32(&(n), __base);                       \
-               (value = isdigit(*cp) ? *cp-'0' : toupper(*cp)-'A'+10) < base) {
+        __rem;                                                          \
-                result = result*base + value;
+ })
-                cp++;
-        }
-        if (endp)
-                *endp = (char *)cp;
-        return result;
-}
-long simple_strtol(const char *cp,char **endp,unsigned int base)
-{
-        if(*cp=='-')
-                return -simple_strtoul(cp+1,endp,base);
-        return simple_strtoul(cp,endp,base);
-}
 static int skip_atoi(const char **s)
 {
-        int i=0;
+        int i, c;
-        while (isdigit(**s))
+        for (i = 0; '0' <= (c = **s) && c <= '9'; ++*s)
-                i = i*10 + *((*s)++) - '0';
+                i = i*10 + c - '0';
        return i;
 }
@@ -436,9 +297,6 @@ static char * number(char * str, unsigned long long num, int base, int size, int
        return str;
 }
-/* Forward decl. needed for IP address printing stuff... */
-int sprintf(char * buf, const char *fmt, ...);
 int vsprintf(char *buf, const char *fmt, va_list args)
 {
        int len;
@@ -477,7 +335,7 @@ int vsprintf(char *buf, const char *fmt, va_list args)
                
                /* get field width */
                field_width = -1;
-                if (isdigit(*fmt))
+                if ('0' <= *fmt && *fmt <= '9')
                        field_width = skip_atoi(&fmt);
                else if (*fmt == '*') {
                        ++fmt;
@@ -493,7 +351,7 @@ int vsprintf(char *buf, const char *fmt, va_list args)
                precision = -1;
                if (*fmt == '.') {
                        ++fmt;  
-                        if (isdigit(*fmt))
+                        if ('0' <= *fmt && *fmt <= '9')
                                precision = skip_atoi(&fmt);
                        else if (*fmt == '*') {
                                ++fmt;
@@ -628,7 +486,7 @@ int sprintf(char * buf, const char *fmt, ...)
 static char sprint_buf[1024];
 int
-printf(char *fmt, ...)
+printf(const char *fmt, ...)
 {
        va_list args;
        int n;
diff --git a/arch/ppc64/boot/prom.h b/arch/ppc64/boot/prom.h
new file mode 100644
index 000000000000..96ab5aec740c
--- /dev/null
+++ b/arch/ppc64/boot/prom.h
@@ -0,0 +1,18 @@
+#ifndef _PPC_BOOT_PROM_H_
+#define _PPC_BOOT_PROM_H_
+extern int (*prom) (void *);
+extern void *chosen_handle;
+extern void *stdin;
+extern void *stdout;
+extern void *stderr;
+extern int write(void *handle, void *ptr, int nb);
+extern int read(void *handle, void *ptr, int nb);
+extern void exit(void);
+extern void pause(void);
+extern void *finddevice(const char *);
+extern void *claim(unsigned long virt, unsigned long size, unsigned long align);
+extern int getprop(void *phandle, const char *name, void *buf, int buflen);
+#endif                          /* _PPC_BOOT_PROM_H_ */
diff --git a/arch/ppc64/boot/stdio.h b/arch/ppc64/boot/stdio.h
new file mode 100644
index 000000000000..24bd3a8dee94
--- /dev/null
+++ b/arch/ppc64/boot/stdio.h
@@ -0,0 +1,16 @@
+#ifndef _PPC_BOOT_STDIO_H_
+#define _PPC_BOOT_STDIO_H_
+extern int printf(const char *fmt, ...);
+extern int sprintf(char *buf, const char *fmt, ...);
+extern int vsprintf(char *buf, const char *fmt, va_list args);
+extern int putc(int c, void *f);
+extern int putchar(int c);
+extern int getchar(void);
+extern int fputs(char *str, void *f);
+#endif                          /* _PPC_BOOT_STDIO_H_ */
diff --git a/arch/ppc64/boot/string.S b/arch/ppc64/boot/string.S
index ba5f2d21c9ea..7ade87ae7718 100644
--- a/arch/ppc64/boot/string.S
+++ b/arch/ppc64/boot/string.S
@@ -9,7 +9,7 @@
 * NOTE: this code runs in 32 bit mode and is packaged as ELF32.
 */
-#include <asm/ppc_asm.h>
+#include "ppc_asm.h"
        .text
        .globl  strcpy
diff --git a/arch/ppc64/boot/string.h b/arch/ppc64/boot/string.h
new file mode 100644
index 000000000000..9289258bcbd6
--- /dev/null
+++ b/arch/ppc64/boot/string.h
@@ -0,0 +1,16 @@
+#ifndef _PPC_BOOT_STRING_H_
+#define _PPC_BOOT_STRING_H_
+extern char *strcpy(char *dest, const char *src);
+extern char *strncpy(char *dest, const char *src, size_t n);
+extern char *strcat(char *dest, const char *src);
+extern int strcmp(const char *s1, const char *s2);
+extern size_t strlen(const char *s);
+extern size_t strnlen(const char *s, size_t count);
+extern void *memset(void *s, int c, size_t n);
+extern void *memmove(void *dest, const void *src, unsigned long n);
+extern void *memcpy(void *dest, const void *src, unsigned long n);
+extern int memcmp(const void *s1, const void *s2, size_t n);
+#endif  /* _PPC_BOOT_STRING_H_ */
diff --git a/arch/ppc64/boot/zlib.c b/arch/ppc64/boot/zlib.c
index 78837e884b8b..0d910cd2079d 100644
--- a/arch/ppc64/boot/zlib.c
+++ b/arch/ppc64/boot/zlib.c
@@ -107,7 +107,7 @@ extern void *memcpy(void *, const void *, unsigned long);
 /* Diagnostic functions */
 #ifdef DEBUG_ZLIB
-#  include <stdio.h>
+#  include "stdio.h"
 #  ifndef verbose
 #    define verbose 0
 #  endif
diff --git a/arch/ppc64/configs/g5_defconfig b/arch/ppc64/configs/g5_defconfig
index ab567741e80e..fc83d9330282 100644
--- a/arch/ppc64/configs/g5_defconfig
+++ b/arch/ppc64/configs/g5_defconfig
@@ -103,10 +103,10 @@ CONFIG_PREEMPT_NONE=y
 # CONFIG_PREEMPT_VOLUNTARY is not set
 # CONFIG_PREEMPT is not set
 # CONFIG_PREEMPT_BKL is not set
-CONFIG_HZ_100=y
+# CONFIG_HZ_100 is not set
-# CONFIG_HZ_250 is not set
+CONFIG_HZ_250=y
 # CONFIG_HZ_1000 is not set
-CONFIG_HZ=100
+CONFIG_HZ=250
 CONFIG_GENERIC_HARDIRQS=y
 CONFIG_SECCOMP=y
 CONFIG_ISA_DMA_API=y
diff --git a/arch/ppc64/configs/iSeries_defconfig b/arch/ppc64/configs/iSeries_defconfig
index 394ba18b58c7..013d4e0e4003 100644
--- a/arch/ppc64/configs/iSeries_defconfig
+++ b/arch/ppc64/configs/iSeries_defconfig
@@ -94,12 +94,11 @@ CONFIG_PREEMPT_NONE=y
 # CONFIG_PREEMPT_VOLUNTARY is not set
 # CONFIG_PREEMPT is not set
 # CONFIG_PREEMPT_BKL is not set
-CONFIG_HZ_100=y
+# CONFIG_HZ_100 is not set
-# CONFIG_HZ_250 is not set
+CONFIG_HZ_250=y
 # CONFIG_HZ_1000 is not set
-CONFIG_HZ=100
+CONFIG_HZ=250
 CONFIG_GENERIC_HARDIRQS=y
-CONFIG_MSCHUNKS=y
 CONFIG_LPARCFG=y
 CONFIG_SECCOMP=y
 CONFIG_ISA_DMA_API=y
diff --git a/arch/ppc64/configs/maple_defconfig b/arch/ppc64/configs/maple_defconfig
index 2033fe663dbe..dd42892cd873 100644
--- a/arch/ppc64/configs/maple_defconfig
+++ b/arch/ppc64/configs/maple_defconfig
@@ -103,10 +103,10 @@ CONFIG_PREEMPT_NONE=y
 # CONFIG_PREEMPT_VOLUNTARY is not set
 # CONFIG_PREEMPT is not set
 # CONFIG_PREEMPT_BKL is not set
-CONFIG_HZ_100=y
+# CONFIG_HZ_100 is not set
-# CONFIG_HZ_250 is not set
+CONFIG_HZ_250=y
 # CONFIG_HZ_1000 is not set
-CONFIG_HZ=100
+CONFIG_HZ=250
 CONFIG_GENERIC_HARDIRQS=y
 CONFIG_SECCOMP=y
 CONFIG_ISA_DMA_API=y
diff --git a/arch/ppc64/configs/pSeries_defconfig b/arch/ppc64/configs/pSeries_defconfig
index 297fd5229487..29f7b80b0efc 100644
--- a/arch/ppc64/configs/pSeries_defconfig
+++ b/arch/ppc64/configs/pSeries_defconfig
@@ -112,10 +112,10 @@ CONFIG_PREEMPT_NONE=y
 # CONFIG_PREEMPT_VOLUNTARY is not set
 # CONFIG_PREEMPT is not set
 # CONFIG_PREEMPT_BKL is not set
-CONFIG_HZ_100=y
+# CONFIG_HZ_100 is not set
-# CONFIG_HZ_250 is not set
+CONFIG_HZ_250=y
 # CONFIG_HZ_1000 is not set
-CONFIG_HZ=100
+CONFIG_HZ=250
 CONFIG_EEH=y
 CONFIG_GENERIC_HARDIRQS=y
 CONFIG_PPC_RTAS=y
diff --git a/arch/ppc64/defconfig b/arch/ppc64/defconfig
index c361e7727b7a..7cb4750bb7a9 100644
--- a/arch/ppc64/defconfig
+++ b/arch/ppc64/defconfig
@@ -114,10 +114,10 @@ CONFIG_PREEMPT_NONE=y
 # CONFIG_PREEMPT_VOLUNTARY is not set
 # CONFIG_PREEMPT is not set
 # CONFIG_PREEMPT_BKL is not set
-CONFIG_HZ_100=y
+# CONFIG_HZ_100 is not set
-# CONFIG_HZ_250 is not set
+CONFIG_HZ_250=y
 # CONFIG_HZ_1000 is not set
-CONFIG_HZ=100
+CONFIG_HZ=250
 CONFIG_EEH=y
 CONFIG_GENERIC_HARDIRQS=y
 CONFIG_PPC_RTAS=y
diff --git a/arch/ppc64/kernel/LparData.c b/arch/ppc64/kernel/LparData.c
index 1c11031c838e..0a9c23ca2f0c 100644
--- a/arch/ppc64/kernel/LparData.c
+++ b/arch/ppc64/kernel/LparData.c
@@ -51,6 +51,17 @@ struct HvReleaseData hvReleaseData = {
                0xf4, 0x4b, 0xf6, 0xf4 },
 };
+/*
+ * The NACA.  The first dword of the naca is required by the iSeries
+ * hypervisor to point to itVpdAreas.  The hypervisor finds the NACA
+ * through the pointer in hvReleaseData.
+ */
+struct naca_struct naca = {
+        .xItVpdAreas = &itVpdAreas,
+        .xRamDisk = 0,
+        .xRamDiskSize = 0,
+};
 extern void system_reset_iSeries(void);
 extern void machine_check_iSeries(void);
 extern void data_access_iSeries(void);
@@ -214,29 +225,3 @@ struct ItVpdAreas itVpdAreas = {
                0,0
        }
 };
-struct msChunks msChunks;
-EXPORT_SYMBOL(msChunks);
-/* Depending on whether this is called from iSeries or pSeries setup
- * code, the location of the msChunks struct may or may not have
- * to be reloc'd, so we force the caller to do that for us by passing
- * in a pointer to the structure.
- */
-unsigned long
-msChunks_alloc(unsigned long mem, unsigned long num_chunks, unsigned long chunk_size)
-{
-        unsigned long offset = reloc_offset();
-        struct msChunks *_msChunks = PTRRELOC(&msChunks);
-        _msChunks->num_chunks  = num_chunks;
-        _msChunks->chunk_size  = chunk_size;
-        _msChunks->chunk_shift = __ilog2(chunk_size);
-        _msChunks->chunk_mask  = (1UL<<_msChunks->chunk_shift)-1;
-        mem = _ALIGN(mem, sizeof(msChunks_entry));
-        _msChunks->abs = (msChunks_entry *)(mem + offset);
-        mem += num_chunks * sizeof(msChunks_entry);
-        return mem;
-}
diff --git a/arch/ppc64/kernel/Makefile b/arch/ppc64/kernel/Makefile
index 2ecccb6b4f8c..ae60eb1193c6 100644
--- a/arch/ppc64/kernel/Makefile
+++ b/arch/ppc64/kernel/Makefile
@@ -11,7 +11,7 @@ obj-y               :=	setup.o entry.o traps.o irq.o idle.o dma.o \
                        udbg.o binfmt_elf32.o sys_ppc32.o ioctl32.o \
                        ptrace32.o signal32.o rtc.o init_task.o \
                        lmb.o cputable.o cpu_setup_power4.o idle_power4.o \
-                        iommu.o sysfs.o vdso.o pmc.o
+                        iommu.o sysfs.o vdso.o pmc.o firmware.o
 obj-y += vdso32/ vdso64/
 obj-$(CONFIG_PPC_OF) += of_device.o
@@ -31,7 +31,7 @@ obj-$(CONFIG_PPC_MULTIPLATFORM) += nvram.o i8259.o prom_init.o prom.o
 obj-$(CONFIG_PPC_PSERIES) += pSeries_pci.o pSeries_lpar.o pSeries_hvCall.o \
                             pSeries_nvram.o rtasd.o ras.o pSeries_reconfig.o \
-                             pSeries_setup.o pSeries_iommu.o
+                             pSeries_setup.o pSeries_iommu.o udbg_16550.o
 obj-$(CONFIG_PPC_BPA) += bpa_setup.o bpa_iommu.o bpa_nvram.o \
                         bpa_iic.o spider-pic.o
@@ -50,14 +50,19 @@ obj-$(CONFIG_LPARCFG)		+= lparcfg.o
 obj-$(CONFIG_HVC_CONSOLE)       += hvconsole.o
 obj-$(CONFIG_BOOTX_TEXT)        += btext.o
 obj-$(CONFIG_HVCS)              += hvcserver.o
-obj-$(CONFIG_IBMVIO)            += vio.o
+vio-obj-$(CONFIG_PPC_PSERIES)   += pSeries_vio.o
+vio-obj-$(CONFIG_PPC_ISERIES)   += iSeries_vio.o
+obj-$(CONFIG_IBMVIO)            += vio.o $(vio-obj-y)
 obj-$(CONFIG_XICS)              += xics.o
 obj-$(CONFIG_MPIC)              += mpic.o
 obj-$(CONFIG_PPC_PMAC)          += pmac_setup.o pmac_feature.o pmac_pci.o \
-                                   pmac_time.o pmac_nvram.o pmac_low_i2c.o
+                                   pmac_time.o pmac_nvram.o pmac_low_i2c.o \
+                                   udbg_scc.o
-obj-$(CONFIG_PPC_MAPLE)         += maple_setup.o maple_pci.o maple_time.o
+obj-$(CONFIG_PPC_MAPLE)         += maple_setup.o maple_pci.o maple_time.o \
+                                   udbg_16550.o
 obj-$(CONFIG_U3_DART)           += u3_iommu.o
diff --git a/arch/ppc64/kernel/asm-offsets.c b/arch/ppc64/kernel/asm-offsets.c
index 6f910fa2746f..1ff4fa05a973 100644
--- a/arch/ppc64/kernel/asm-offsets.c
+++ b/arch/ppc64/kernel/asm-offsets.c
@@ -95,7 +95,8 @@ int main(void)
        DEFINE(PACASLBCACHEPTR, offsetof(struct paca_struct, slb_cache_ptr));
        DEFINE(PACACONTEXTID, offsetof(struct paca_struct, context.id));
 #ifdef CONFIG_HUGETLB_PAGE
-        DEFINE(PACAHTLBSEGS, offsetof(struct paca_struct, context.htlb_segs));
+        DEFINE(PACALOWHTLBAREAS, offsetof(struct paca_struct, context.low_htlb_areas));
+        DEFINE(PACAHIGHHTLBAREAS, offsetof(struct paca_struct, context.high_htlb_areas));
 #endif /* CONFIG_HUGETLB_PAGE */
        DEFINE(PACADEFAULTDECR, offsetof(struct paca_struct, default_decr));
        DEFINE(PACA_EXGEN, offsetof(struct paca_struct, exgen));
diff --git a/arch/ppc64/kernel/bpa_iic.c b/arch/ppc64/kernel/bpa_iic.c
index c8f3dc3fad70..0aaa878e19d3 100644
--- a/arch/ppc64/kernel/bpa_iic.c
+++ b/arch/ppc64/kernel/bpa_iic.c
@@ -205,6 +205,18 @@ static struct iic_regs __iomem *find_iic(int cpu)
 }
 #ifdef CONFIG_SMP
+/* Use the highest interrupt priorities for IPI */
+static inline int iic_ipi_to_irq(int ipi)
+{
+        return IIC_IPI_OFFSET + IIC_NUM_IPIS - 1 - ipi;
+}
+static inline int iic_irq_to_ipi(int irq)
+{
+        return IIC_NUM_IPIS - 1 - (irq - IIC_IPI_OFFSET);
+}
 void iic_setup_cpu(void)
 {
        out_be64(&__get_cpu_var(iic).regs->prio, 0xff);
@@ -212,18 +224,20 @@ void iic_setup_cpu(void)
 void iic_cause_IPI(int cpu, int mesg)
 {
-        out_be64(&per_cpu(iic, cpu).regs->generate, mesg);
+        out_be64(&per_cpu(iic, cpu).regs->generate, (IIC_NUM_IPIS - 1 - mesg) << 4);
 }
 static irqreturn_t iic_ipi_action(int irq, void *dev_id, struct pt_regs *regs)
 {
+        smp_message_recv(iic_irq_to_ipi(irq), regs);
-        smp_message_recv(irq - IIC_IPI_OFFSET, regs);
        return IRQ_HANDLED;
 }
-static void iic_request_ipi(int irq, const char *name)
+static void iic_request_ipi(int ipi, const char *name)
 {
+        int irq;
+        irq = iic_ipi_to_irq(ipi);
        /* IPIs are marked SA_INTERRUPT as they must run with irqs
         * disabled */
        get_irq_desc(irq)->handler = &iic_pic;
@@ -233,10 +247,10 @@ static void iic_request_ipi(int irq, const char *name)
 void iic_request_IPIs(void)
 {
-        iic_request_ipi(IIC_IPI_OFFSET + PPC_MSG_CALL_FUNCTION, "IPI-call");
+        iic_request_ipi(PPC_MSG_CALL_FUNCTION, "IPI-call");
-        iic_request_ipi(IIC_IPI_OFFSET + PPC_MSG_RESCHEDULE, "IPI-resched");
+        iic_request_ipi(PPC_MSG_RESCHEDULE, "IPI-resched");
 #ifdef CONFIG_DEBUGGER
-        iic_request_ipi(IIC_IPI_OFFSET + PPC_MSG_DEBUGGER_BREAK, "IPI-debug");
+        iic_request_ipi(PPC_MSG_DEBUGGER_BREAK, "IPI-debug");
 #endif /* CONFIG_DEBUGGER */
 }
 #endif /* CONFIG_SMP */
diff --git a/arch/ppc64/kernel/btext.c b/arch/ppc64/kernel/btext.c
index c53f079e9b77..b6fbfbe9032d 100644
--- a/arch/ppc64/kernel/btext.c
+++ b/arch/ppc64/kernel/btext.c
@@ -7,7 +7,6 @@
 #include <linux/kernel.h>
 #include <linux/string.h>
 #include <linux/init.h>
-#include <linux/version.h>
 #include <asm/sections.h>
 #include <asm/prom.h>
diff --git a/arch/ppc64/kernel/cpu_setup_power4.S b/arch/ppc64/kernel/cpu_setup_power4.S
index 0482c063c26e..1fb673c511ff 100644
--- a/arch/ppc64/kernel/cpu_setup_power4.S
+++ b/arch/ppc64/kernel/cpu_setup_power4.S
@@ -12,10 +12,9 @@
 #include <linux/config.h>
 #include <asm/processor.h>
 #include <asm/page.h>
-#include <asm/ppc_asm.h>
 #include <asm/cputable.h>
 #include <asm/ppc_asm.h>
-#include <asm/offsets.h>
+#include <asm/asm-offsets.h>
 #include <asm/cache.h>
 _GLOBAL(__970_cpu_preinit)
diff --git a/arch/ppc64/kernel/cputable.c b/arch/ppc64/kernel/cputable.c
index 77cec42f9525..8831a28c3c4e 100644
--- a/arch/ppc64/kernel/cputable.c
+++ b/arch/ppc64/kernel/cputable.c
@@ -5,7 +5,7 @@
 *
 *  Modifications for ppc64:
 *      Copyright (C) 2003 Dave Engebretsen <engebret@us.ibm.com>
- * 
+ *
 *  This program is free software; you can redistribute it and/or
 *  modify it under the terms of the GNU General Public License
 *  as published by the Free Software Foundation; either version
@@ -19,6 +19,7 @@
 #include <linux/init.h>
 #include <linux/module.h>
+#include <asm/oprofile_impl.h>
 #include <asm/cputable.h>
 struct cpu_spec* cur_cpu_spec = NULL;
@@ -54,26 +55,32 @@ struct cpu_spec	cpu_specs[] = {
                .pvr_value              = 0x00400000,
                .cpu_name               = "POWER3 (630)",
                .cpu_features           = CPU_FTR_SPLIT_ID_CACHE |
-                        CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | CPU_FTR_IABR |
+                        CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | CPU_FTR_IABR,
-                        CPU_FTR_PMC8,
                .cpu_user_features = COMMON_USER_PPC64,
                .icache_bsize           = 128,
                .dcache_bsize           = 128,
+                .num_pmcs               = 8,
                .cpu_setup              = __setup_cpu_power3,
-                .firmware_features      = COMMON_PPC64_FW,
+#ifdef CONFIG_OPROFILE
+                .oprofile_cpu_type      = "ppc64/power3",
+                .oprofile_model         = &op_model_rs64,
+#endif
        },
        {       /* Power3+ */
                .pvr_mask               = 0xffff0000,
                .pvr_value              = 0x00410000,
                .cpu_name               = "POWER3 (630+)",
                .cpu_features           = CPU_FTR_SPLIT_ID_CACHE |
-                        CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | CPU_FTR_IABR |
+                        CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | CPU_FTR_IABR,
-                        CPU_FTR_PMC8,
                .cpu_user_features      = COMMON_USER_PPC64,
                .icache_bsize           = 128,
                .dcache_bsize           = 128,
+                .num_pmcs               = 8,
                .cpu_setup              = __setup_cpu_power3,
-                .firmware_features      = COMMON_PPC64_FW,
+#ifdef CONFIG_OPROFILE
+                .oprofile_cpu_type      = "ppc64/power3",
+                .oprofile_model         = &op_model_rs64,
+#endif
        },
        {       /* Northstar */
                .pvr_mask               = 0xffff0000,
@@ -81,12 +88,16 @@ struct cpu_spec	cpu_specs[] = {
                .cpu_name               = "RS64-II (northstar)",
                .cpu_features           = CPU_FTR_SPLIT_ID_CACHE |
                        CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | CPU_FTR_IABR |
-                        CPU_FTR_PMC8 | CPU_FTR_MMCRA | CPU_FTR_CTRL,
+                        CPU_FTR_MMCRA | CPU_FTR_CTRL,
                .cpu_user_features      = COMMON_USER_PPC64,
                .icache_bsize           = 128,
                .dcache_bsize           = 128,
+                .num_pmcs               = 8,
                .cpu_setup              = __setup_cpu_power3,
-                .firmware_features      = COMMON_PPC64_FW,
+#ifdef CONFIG_OPROFILE
+                .oprofile_cpu_type      = "ppc64/rs64",
+                .oprofile_model         = &op_model_rs64,
+#endif
        },
        {       /* Pulsar */
                .pvr_mask               = 0xffff0000,
@@ -94,12 +105,16 @@ struct cpu_spec	cpu_specs[] = {
                .cpu_name               = "RS64-III (pulsar)",
                .cpu_features           = CPU_FTR_SPLIT_ID_CACHE |
                        CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | CPU_FTR_IABR |
-                        CPU_FTR_PMC8 | CPU_FTR_MMCRA | CPU_FTR_CTRL,
+                        CPU_FTR_MMCRA | CPU_FTR_CTRL,
                .cpu_user_features      = COMMON_USER_PPC64,
                .icache_bsize           = 128,
                .dcache_bsize           = 128,
+                .num_pmcs               = 8,
                .cpu_setup              = __setup_cpu_power3,
-                .firmware_features      = COMMON_PPC64_FW,
+#ifdef CONFIG_OPROFILE
+                .oprofile_cpu_type      = "ppc64/rs64",
+                .oprofile_model         = &op_model_rs64,
+#endif
        },
        {       /* I-star */
                .pvr_mask               = 0xffff0000,
@@ -107,12 +122,16 @@ struct cpu_spec	cpu_specs[] = {
                .cpu_name               = "RS64-III (icestar)",
                .cpu_features           = CPU_FTR_SPLIT_ID_CACHE |
                        CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | CPU_FTR_IABR |
-                        CPU_FTR_PMC8 | CPU_FTR_MMCRA | CPU_FTR_CTRL,
+                        CPU_FTR_MMCRA | CPU_FTR_CTRL,
                .cpu_user_features      = COMMON_USER_PPC64,
                .icache_bsize           = 128,
                .dcache_bsize           = 128,
+                .num_pmcs               = 8,
                .cpu_setup              = __setup_cpu_power3,
-                .firmware_features      = COMMON_PPC64_FW,
+#ifdef CONFIG_OPROFILE
+                .oprofile_cpu_type      = "ppc64/rs64",
+                .oprofile_model         = &op_model_rs64,
+#endif
        },
        {       /* S-star */
                .pvr_mask               = 0xffff0000,
@@ -120,12 +139,16 @@ struct cpu_spec	cpu_specs[] = {
                .cpu_name               = "RS64-IV (sstar)",
                .cpu_features           = CPU_FTR_SPLIT_ID_CACHE |
                        CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE | CPU_FTR_IABR |
-                        CPU_FTR_PMC8 | CPU_FTR_MMCRA | CPU_FTR_CTRL,
+                        CPU_FTR_MMCRA | CPU_FTR_CTRL,
                .cpu_user_features      = COMMON_USER_PPC64,
                .icache_bsize           = 128,
                .dcache_bsize           = 128,
+                .num_pmcs               = 8,
                .cpu_setup              = __setup_cpu_power3,
-                .firmware_features      = COMMON_PPC64_FW,
+#ifdef CONFIG_OPROFILE
+                .oprofile_cpu_type      = "ppc64/rs64",
+                .oprofile_model         = &op_model_rs64,
+#endif
        },
        {       /* Power4 */
                .pvr_mask               = 0xffff0000,
@@ -133,12 +156,16 @@ struct cpu_spec	cpu_specs[] = {
                .cpu_name               = "POWER4 (gp)",
                .cpu_features           = CPU_FTR_SPLIT_ID_CACHE |
                        CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE |
-                        CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_PMC8 | CPU_FTR_MMCRA,
+                        CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_MMCRA,
                .cpu_user_features      = COMMON_USER_PPC64,
                .icache_bsize           = 128,
                .dcache_bsize           = 128,
+                .num_pmcs               = 8,
                .cpu_setup              = __setup_cpu_power4,
-                .firmware_features      = COMMON_PPC64_FW,
+#ifdef CONFIG_OPROFILE
+                .oprofile_cpu_type      = "ppc64/power4",
+                .oprofile_model         = &op_model_rs64,
+#endif
        },
        {       /* Power4+ */
                .pvr_mask               = 0xffff0000,
@@ -146,12 +173,16 @@ struct cpu_spec	cpu_specs[] = {
                .cpu_name               = "POWER4+ (gq)",
                .cpu_features           = CPU_FTR_SPLIT_ID_CACHE |
                        CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE |
-                        CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_PMC8 | CPU_FTR_MMCRA,
+                        CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_MMCRA,
                .cpu_user_features      = COMMON_USER_PPC64,
                .icache_bsize           = 128,
                .dcache_bsize           = 128,
+                .num_pmcs               = 8,
                .cpu_setup              = __setup_cpu_power4,
-                .firmware_features      = COMMON_PPC64_FW,
+#ifdef CONFIG_OPROFILE
+                .oprofile_cpu_type      = "ppc64/power4",
+                .oprofile_model         = &op_model_power4,
+#endif
        },
        {       /* PPC970 */
                .pvr_mask               = 0xffff0000,
@@ -160,13 +191,17 @@ struct cpu_spec	cpu_specs[] = {
                .cpu_features           = CPU_FTR_SPLIT_ID_CACHE |
                        CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE |
                        CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_ALTIVEC_COMP |
-                        CPU_FTR_CAN_NAP | CPU_FTR_PMC8 | CPU_FTR_MMCRA,
+                        CPU_FTR_CAN_NAP | CPU_FTR_MMCRA,
                .cpu_user_features      = COMMON_USER_PPC64 |
                        PPC_FEATURE_HAS_ALTIVEC_COMP,
                .icache_bsize           = 128,
                .dcache_bsize           = 128,
+                .num_pmcs               = 8,
                .cpu_setup              = __setup_cpu_ppc970,
-                .firmware_features      = COMMON_PPC64_FW,
+#ifdef CONFIG_OPROFILE
+                .oprofile_cpu_type      = "ppc64/970",
+                .oprofile_model         = &op_model_power4,
+#endif
        },
        {       /* PPC970FX */
                .pvr_mask               = 0xffff0000,
@@ -175,13 +210,17 @@ struct cpu_spec	cpu_specs[] = {
                .cpu_features           = CPU_FTR_SPLIT_ID_CACHE |
                        CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE |
                        CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_ALTIVEC_COMP |
-                        CPU_FTR_CAN_NAP | CPU_FTR_PMC8 | CPU_FTR_MMCRA,
+                        CPU_FTR_CAN_NAP | CPU_FTR_MMCRA,
                .cpu_user_features      = COMMON_USER_PPC64 |
                        PPC_FEATURE_HAS_ALTIVEC_COMP,
                .icache_bsize           = 128,
                .dcache_bsize           = 128,
+                .num_pmcs               = 8,
                .cpu_setup              = __setup_cpu_ppc970,
-                .firmware_features      = COMMON_PPC64_FW,
+#ifdef CONFIG_OPROFILE
+                .oprofile_cpu_type      = "ppc64/970",
+                .oprofile_model         = &op_model_power4,
+#endif
        },
        {       /* PPC970MP */
                .pvr_mask               = 0xffff0000,
@@ -190,13 +229,16 @@ struct cpu_spec	cpu_specs[] = {
                .cpu_features           = CPU_FTR_SPLIT_ID_CACHE |
                        CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE |
                        CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_ALTIVEC_COMP |
-                        CPU_FTR_CAN_NAP | CPU_FTR_PMC8 | CPU_FTR_MMCRA,
+                        CPU_FTR_CAN_NAP | CPU_FTR_MMCRA,
                .cpu_user_features      = COMMON_USER_PPC64 |
                        PPC_FEATURE_HAS_ALTIVEC_COMP,
                .icache_bsize           = 128,
                .dcache_bsize           = 128,
                .cpu_setup              = __setup_cpu_ppc970,
-                .firmware_features      = COMMON_PPC64_FW,
+#ifdef CONFIG_OPROFILE
+                .oprofile_cpu_type      = "ppc64/970",
+                .oprofile_model         = &op_model_power4,
+#endif
        },
        {       /* Power5 */
                .pvr_mask               = 0xffff0000,
@@ -210,8 +252,12 @@ struct cpu_spec	cpu_specs[] = {
                .cpu_user_features      = COMMON_USER_PPC64,
                .icache_bsize           = 128,
                .dcache_bsize           = 128,
+                .num_pmcs               = 6,
                .cpu_setup              = __setup_cpu_power4,
-                .firmware_features      = COMMON_PPC64_FW,
+#ifdef CONFIG_OPROFILE
+                .oprofile_cpu_type      = "ppc64/power5",
+                .oprofile_model         = &op_model_power4,
+#endif
        },
        {       /* Power5 */
                .pvr_mask               = 0xffff0000,
@@ -225,8 +271,12 @@ struct cpu_spec	cpu_specs[] = {
                .cpu_user_features      = COMMON_USER_PPC64,
                .icache_bsize           = 128,
                .dcache_bsize           = 128,
+                .num_pmcs               = 6,
                .cpu_setup              = __setup_cpu_power4,
-                .firmware_features      = COMMON_PPC64_FW,
+#ifdef CONFIG_OPROFILE
+                .oprofile_cpu_type      = "ppc64/power5",
+                .oprofile_model         = &op_model_power4,
+#endif
        },
        {       /* BE DD1.x */
                .pvr_mask               = 0xffff0000,
@@ -241,7 +291,6 @@ struct cpu_spec	cpu_specs[] = {
                .icache_bsize           = 128,
                .dcache_bsize           = 128,
                .cpu_setup              = __setup_cpu_be,
-                .firmware_features      = COMMON_PPC64_FW,
        },
        {       /* default match */
                .pvr_mask               = 0x00000000,
@@ -253,30 +302,7 @@ struct cpu_spec	cpu_specs[] = {
                .cpu_user_features      = COMMON_USER_PPC64,
                .icache_bsize           = 128,
                .dcache_bsize           = 128,
+                .num_pmcs               = 6,
                .cpu_setup              = __setup_cpu_power4,
-                .firmware_features      = COMMON_PPC64_FW,
        }
 };
-firmware_feature_t firmware_features_table[FIRMWARE_MAX_FEATURES] = {
-        {FW_FEATURE_PFT,                "hcall-pft"},
-        {FW_FEATURE_TCE,                "hcall-tce"},
-        {FW_FEATURE_SPRG0,              "hcall-sprg0"},
-        {FW_FEATURE_DABR,               "hcall-dabr"},
-        {FW_FEATURE_COPY,               "hcall-copy"},
-        {FW_FEATURE_ASR,                "hcall-asr"},
-        {FW_FEATURE_DEBUG,              "hcall-debug"},
-        {FW_FEATURE_PERF,               "hcall-perf"},
-        {FW_FEATURE_DUMP,               "hcall-dump"},
-        {FW_FEATURE_INTERRUPT,          "hcall-interrupt"},
-        {FW_FEATURE_MIGRATE,            "hcall-migrate"},
-        {FW_FEATURE_PERFMON,            "hcall-perfmon"},
-        {FW_FEATURE_CRQ,                "hcall-crq"},
-        {FW_FEATURE_VIO,                "hcall-vio"},
-        {FW_FEATURE_RDMA,               "hcall-rdma"},
-        {FW_FEATURE_LLAN,               "hcall-lLAN"},
-        {FW_FEATURE_BULK,               "hcall-bulk"},
-        {FW_FEATURE_XDABR,              "hcall-xdabr"},
-        {FW_FEATURE_MULTITCE,           "hcall-multi-tce"},
-        {FW_FEATURE_SPLPAR,             "hcall-splpar"},
-};
diff --git a/arch/ppc64/kernel/eeh.c b/arch/ppc64/kernel/eeh.c
index af5272fedadf..ba93fd731222 100644
--- a/arch/ppc64/kernel/eeh.c
+++ b/arch/ppc64/kernel/eeh.c
@@ -202,10 +202,9 @@ static void pci_addr_cache_print(struct pci_io_addr_cache *cache)
        while (n) {
                struct pci_io_addr_range *piar;
                piar = rb_entry(n, struct pci_io_addr_range, rb_node);
-                printk(KERN_DEBUG "PCI: %s addr range %d [%lx-%lx]: %s %s\n",
+                printk(KERN_DEBUG "PCI: %s addr range %d [%lx-%lx]: %s\n",
                       (piar->flags & IORESOURCE_IO) ? "i/o" : "mem", cnt,
-                       piar->addr_lo, piar->addr_hi, pci_name(piar->pcidev),
+                       piar->addr_lo, piar->addr_hi, pci_name(piar->pcidev));
-                       pci_pretty_name(piar->pcidev));
                cnt++;
                n = rb_next(n);
        }
@@ -255,22 +254,24 @@ pci_addr_cache_insert(struct pci_dev *dev, unsigned long alo,
 static void __pci_addr_cache_insert_device(struct pci_dev *dev)
 {
        struct device_node *dn;
+        struct pci_dn *pdn;
        int i;
        int inserted = 0;
        dn = pci_device_to_OF_node(dev);
        if (!dn) {
-                printk(KERN_WARNING "PCI: no pci dn found for dev=%s %s\n",
+                printk(KERN_WARNING "PCI: no pci dn found for dev=%s\n",
-                        pci_name(dev), pci_pretty_name(dev));
+                        pci_name(dev));
                return;
        }
        /* Skip any devices for which EEH is not enabled. */
-        if (!(dn->eeh_mode & EEH_MODE_SUPPORTED) ||
+        pdn = dn->data;
-            dn->eeh_mode & EEH_MODE_NOCHECK) {
+        if (!(pdn->eeh_mode & EEH_MODE_SUPPORTED) ||
+            pdn->eeh_mode & EEH_MODE_NOCHECK) {
 #ifdef DEBUG
-                printk(KERN_INFO "PCI: skip building address cache for=%s %s\n",
+                printk(KERN_INFO "PCI: skip building address cache for=%s\n",
-                       pci_name(dev), pci_pretty_name(dev));
+                       pci_name(dev));
 #endif
                return;
        }
@@ -416,6 +417,7 @@ int eeh_unregister_notifier(struct notifier_block *nb)
 static int read_slot_reset_state(struct device_node *dn, int rets[])
 {
        int token, outputs;
+        struct pci_dn *pdn = dn->data;
        if (ibm_read_slot_reset_state2 != RTAS_UNKNOWN_SERVICE) {
                token = ibm_read_slot_reset_state2;
@@ -425,8 +427,8 @@ static int read_slot_reset_state(struct device_node *dn, int rets[])
                outputs = 3;
        }
-        return rtas_call(token, 3, outputs, rets, dn->eeh_config_addr,
+        return rtas_call(token, 3, outputs, rets, pdn->eeh_config_addr,
-                         BUID_HI(dn->phb->buid), BUID_LO(dn->phb->buid));
+                         BUID_HI(pdn->phb->buid), BUID_LO(pdn->phb->buid));
 }
 /**
@@ -447,12 +449,12 @@ static void eeh_panic(struct pci_dev *dev, int reset_state)
         * in light of potential corruption, we can use it here.
         */
        if (panic_on_oops)
-                panic("EEH: MMIO failure (%d) on device:%s %s\n", reset_state,
+                panic("EEH: MMIO failure (%d) on device:%s\n", reset_state,
-                      pci_name(dev), pci_pretty_name(dev));
+                      pci_name(dev));
        else {
                __get_cpu_var(ignored_failures)++;
-                printk(KERN_INFO "EEH: Ignored MMIO failure (%d) on device:%s %s\n",
+                printk(KERN_INFO "EEH: Ignored MMIO failure (%d) on device:%s\n",
-                       reset_state, pci_name(dev), pci_pretty_name(dev));
+                       reset_state, pci_name(dev));
        }
 }
@@ -482,8 +484,8 @@ static void eeh_event_handler(void *dummy)
                        break;
                printk(KERN_INFO "EEH: MMIO failure (%d), notifiying device "
-                       "%s %s\n", event->reset_state,
+                       "%s\n", event->reset_state,
-                       pci_name(event->dev), pci_pretty_name(event->dev));
+                       pci_name(event->dev));
                atomic_set(&eeh_fail_count, 0);
                notifier_call_chain (&eeh_notifier_chain,
@@ -535,6 +537,7 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev)
        unsigned long flags;
        int rc, reset_state;
        struct eeh_event  *event;
+        struct pci_dn *pdn;
        __get_cpu_var(total_mmio_ffs)++;
@@ -543,14 +546,15 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev)
        if (!dn)
                return 0;
+        pdn = dn->data;
        /* Access to IO BARs might get this far and still not want checking. */
-        if (!(dn->eeh_mode & EEH_MODE_SUPPORTED) ||
+        if (!pdn->eeh_capable || !(pdn->eeh_mode & EEH_MODE_SUPPORTED) ||
-            dn->eeh_mode & EEH_MODE_NOCHECK) {
+            pdn->eeh_mode & EEH_MODE_NOCHECK) {
                return 0;
        }
-        if (!dn->eeh_config_addr) {
+        if (!pdn->eeh_config_addr) {
                return 0;
        }
@@ -558,7 +562,7 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev)
         * If we already have a pending isolation event for this
         * slot, we know it's bad already, we don't need to check...
         */
-        if (dn->eeh_mode & EEH_MODE_ISOLATED) {
+        if (pdn->eeh_mode & EEH_MODE_ISOLATED) {
                atomic_inc(&eeh_fail_count);
                if (atomic_read(&eeh_fail_count) >= EEH_MAX_FAILS) {
                        /* re-read the slot reset state */
@@ -583,7 +587,7 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev)
        }
        /* prevent repeated reports of this failure */
-        dn->eeh_mode |= EEH_MODE_ISOLATED;
+        pdn->eeh_mode |= EEH_MODE_ISOLATED;
        reset_state = rets[0];
@@ -591,9 +595,9 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev)
        memset(slot_errbuf, 0, eeh_error_buf_size);
        rc = rtas_call(ibm_slot_error_detail,
-                       8, 1, NULL, dn->eeh_config_addr,
+                       8, 1, NULL, pdn->eeh_config_addr,
-                       BUID_HI(dn->phb->buid),
+                       BUID_HI(pdn->phb->buid),
-                       BUID_LO(dn->phb->buid), NULL, 0,
+                       BUID_LO(pdn->phb->buid), NULL, 0,
                       virt_to_phys(slot_errbuf),
                       eeh_error_buf_size,
                       1 /* Temporary Error */);
@@ -680,8 +684,9 @@ static void *early_enable_eeh(struct device_node *dn, void *data)
        u32 *device_id = (u32 *)get_property(dn, "device-id", NULL);
        u32 *regs;
        int enable;
+        struct pci_dn *pdn = dn->data;
-        dn->eeh_mode = 0;
+        pdn->eeh_mode = 0;
        if (status && strcmp(status, "ok") != 0)
                return NULL;    /* ignore devices with bad status */
@@ -692,7 +697,7 @@ static void *early_enable_eeh(struct device_node *dn, void *data)
        /* There is nothing to check on PCI to ISA bridges */
        if (dn->type && !strcmp(dn->type, "isa")) {
-                dn->eeh_mode |= EEH_MODE_NOCHECK;
+                pdn->eeh_mode |= EEH_MODE_NOCHECK;
                return NULL;
        }
@@ -709,7 +714,7 @@ static void *early_enable_eeh(struct device_node *dn, void *data)
                enable = 0;
        if (!enable)
-                dn->eeh_mode |= EEH_MODE_NOCHECK;
+                pdn->eeh_mode |= EEH_MODE_NOCHECK;
        /* Ok... see if this device supports EEH.  Some do, some don't,
         * and the only way to find out is to check each and every one. */
@@ -722,8 +727,8 @@ static void *early_enable_eeh(struct device_node *dn, void *data)
                                EEH_ENABLE);
                if (ret == 0) {
                        eeh_subsystem_enabled = 1;
-                        dn->eeh_mode |= EEH_MODE_SUPPORTED;
+                        pdn->eeh_mode |= EEH_MODE_SUPPORTED;
-                        dn->eeh_config_addr = regs[0];
+                        pdn->eeh_config_addr = regs[0];
 #ifdef DEBUG
                        printk(KERN_DEBUG "EEH: %s: eeh enabled\n", dn->full_name);
 #endif
@@ -731,10 +736,11 @@ static void *early_enable_eeh(struct device_node *dn, void *data)
                        /* This device doesn't support EEH, but it may have an
                         * EEH parent, in which case we mark it as supported. */
-                        if (dn->parent && (dn->parent->eeh_mode & EEH_MODE_SUPPORTED)) {
+                        if (dn->parent && dn->parent->data
+                            && (PCI_DN(dn->parent)->eeh_mode & EEH_MODE_SUPPORTED)) {
                                /* Parent supports EEH. */
-                                dn->eeh_mode |= EEH_MODE_SUPPORTED;
+                                pdn->eeh_mode |= EEH_MODE_SUPPORTED;
-                                dn->eeh_config_addr = dn->parent->eeh_config_addr;
+                                pdn->eeh_config_addr = PCI_DN(dn->parent)->eeh_config_addr;
                                return NULL;
                        }
                }
@@ -791,11 +797,13 @@ void __init eeh_init(void)
        for (phb = of_find_node_by_name(NULL, "pci"); phb;
             phb = of_find_node_by_name(phb, "pci")) {
                unsigned long buid;
+                struct pci_dn *pci;
                buid = get_phb_buid(phb);
-                if (buid == 0)
+                if (buid == 0 || phb->data == NULL)
                        continue;
+                pci = phb->data;
                info.buid_lo = BUID_LO(buid);
                info.buid_hi = BUID_HI(buid);
                traverse_pci_devices(phb, early_enable_eeh, &info);
@@ -824,9 +832,9 @@ void eeh_add_device_early(struct device_node *dn)
        struct pci_controller *phb;
        struct eeh_early_enable_info info;
-        if (!dn)
+        if (!dn || !dn->data)
                return;
-        phb = dn->phb;
+        phb = PCI_DN(dn)->phb;
        if (NULL == phb || 0 == phb->buid) {
                printk(KERN_WARNING "EEH: Expected buid but found none\n");
                return;
@@ -851,8 +859,7 @@ void eeh_add_device_late(struct pci_dev *dev)
                return;
 #ifdef DEBUG
-        printk(KERN_DEBUG "EEH: adding device %s %s\n", pci_name(dev),
+        printk(KERN_DEBUG "EEH: adding device %s\n", pci_name(dev));
-               pci_pretty_name(dev));
 #endif
        pci_addr_cache_insert_device (dev);
@@ -873,8 +880,7 @@ void eeh_remove_device(struct pci_dev *dev)
        /* Unregister the device with the EEH/PCI address search system */
 #ifdef DEBUG
-        printk(KERN_DEBUG "EEH: remove device %s %s\n", pci_name(dev),
+        printk(KERN_DEBUG "EEH: remove device %s\n", pci_name(dev));
-               pci_pretty_name(dev));
 #endif
        pci_addr_cache_remove_device(dev);
 }
diff --git a/arch/ppc64/kernel/entry.S b/arch/ppc64/kernel/entry.S
index 14cd56ac40dd..e8c0bbf4d000 100644
--- a/arch/ppc64/kernel/entry.S
+++ b/arch/ppc64/kernel/entry.S
@@ -28,7 +28,7 @@
 #include <asm/mmu.h>
 #include <asm/thread_info.h>
 #include <asm/ppc_asm.h>
-#include <asm/offsets.h>
+#include <asm/asm-offsets.h>
 #include <asm/cputable.h>
 #ifdef CONFIG_PPC_ISERIES
@@ -410,15 +410,14 @@ BEGIN_FTR_SECTION
        cmpd    cr1,r6,r9       /* or is new ESID the same as current ESID? */
        cror    eq,4*cr1+eq,eq
        beq     2f              /* if yes, don't slbie it */
-        oris    r0,r6,0x0800    /* set C (class) bit */
        /* Bolt in the new stack SLB entry */
        ld      r7,KSP_VSID(r4) /* Get new stack's VSID */
-        oris    r6,r6,(SLB_ESID_V)@h
+        oris    r0,r6,(SLB_ESID_V)@h
-        ori     r6,r6,(SLB_NUM_BOLTED-1)@l
+        ori     r0,r0,(SLB_NUM_BOLTED-1)@l
-        slbie   r0
+        slbie   r6
-        slbie   r0              /* Workaround POWER5 < DD2.1 issue */
+        slbie   r6              /* Workaround POWER5 < DD2.1 issue */
-        slbmte  r7,r6
+        slbmte  r7,r0
        isync
 2:
diff --git a/arch/ppc64/kernel/firmware.c b/arch/ppc64/kernel/firmware.c
new file mode 100644
index 000000000000..d8432c0fb27d
--- /dev/null
+++ b/arch/ppc64/kernel/firmware.c
@@ -0,0 +1,47 @@
+/*
+ *  arch/ppc64/kernel/firmware.c
+ *
+ *  Extracted from cputable.c
+ *
+ *  Copyright (C) 2001 Ben. Herrenschmidt (benh@kernel.crashing.org)
+ *
+ *  Modifications for ppc64:
+ *      Copyright (C) 2003 Dave Engebretsen <engebret@us.ibm.com>
+ *  Copyright (C) 2005 Stephen Rothwell, IBM Corporation
+ *
+ *  This program is free software; you can redistribute it and/or
+ *  modify it under the terms of the GNU General Public License
+ *  as published by the Free Software Foundation; either version
+ *  2 of the License, or (at your option) any later version.
+ */
+#include <linux/config.h>
+#include <asm/firmware.h>
+unsigned long ppc64_firmware_features;
+#ifdef CONFIG_PPC_PSERIES
+firmware_feature_t firmware_features_table[FIRMWARE_MAX_FEATURES] = {
+        {FW_FEATURE_PFT,                "hcall-pft"},
+        {FW_FEATURE_TCE,                "hcall-tce"},
+        {FW_FEATURE_SPRG0,              "hcall-sprg0"},
+        {FW_FEATURE_DABR,               "hcall-dabr"},
+        {FW_FEATURE_COPY,               "hcall-copy"},
+        {FW_FEATURE_ASR,                "hcall-asr"},
+        {FW_FEATURE_DEBUG,              "hcall-debug"},
+        {FW_FEATURE_PERF,               "hcall-perf"},
+        {FW_FEATURE_DUMP,               "hcall-dump"},
+        {FW_FEATURE_INTERRUPT,          "hcall-interrupt"},
+        {FW_FEATURE_MIGRATE,            "hcall-migrate"},
+        {FW_FEATURE_PERFMON,            "hcall-perfmon"},
+        {FW_FEATURE_CRQ,                "hcall-crq"},
+        {FW_FEATURE_VIO,                "hcall-vio"},
+        {FW_FEATURE_RDMA,               "hcall-rdma"},
+        {FW_FEATURE_LLAN,               "hcall-lLAN"},
+        {FW_FEATURE_BULK,               "hcall-bulk"},
+        {FW_FEATURE_XDABR,              "hcall-xdabr"},
+        {FW_FEATURE_MULTITCE,           "hcall-multi-tce"},
+        {FW_FEATURE_SPLPAR,             "hcall-splpar"},
+};
+#endif
diff --git a/arch/ppc64/kernel/head.S b/arch/ppc64/kernel/head.S
index accaa052d31f..58c314738c99 100644
--- a/arch/ppc64/kernel/head.S
+++ b/arch/ppc64/kernel/head.S
@@ -23,17 +23,14 @@
 *  2 of the License, or (at your option) any later version.
 */
-#define SECONDARY_PROCESSORS
 #include <linux/config.h>
 #include <linux/threads.h>
 #include <asm/processor.h>
 #include <asm/page.h>
 #include <asm/mmu.h>
-#include <asm/naca.h>
 #include <asm/systemcfg.h>
 #include <asm/ppc_asm.h>
-#include <asm/offsets.h>
+#include <asm/asm-offsets.h>
 #include <asm/bug.h>
 #include <asm/cputable.h>
 #include <asm/setup.h>
@@ -45,18 +42,13 @@
 #endif
 /*
- * hcall interface to pSeries LPAR
- */
-#define H_SET_ASR       0x30
-/*
 * We layout physical memory as follows:
 * 0x0000 - 0x00ff : Secondary processor spin code
 * 0x0100 - 0x2fff : pSeries Interrupt prologs
- * 0x3000 - 0x3fff : Interrupt support
+ * 0x3000 - 0x5fff : interrupt support, iSeries and common interrupt prologs
- * 0x4000 - 0x4fff : NACA
+ * 0x6000 - 0x6fff : Initial (CPU0) segment table
- * 0x6000          : iSeries and common interrupt prologs
+ * 0x7000 - 0x7fff : FWNMI data area
- * 0x9000 - 0x9fff : Initial segment table
+ * 0x8000 -        : Early init and support code
 */
 /*
@@ -94,6 +86,7 @@ END_FTR_SECTION(0, 1)
        /* Catch branch to 0 in real mode */
        trap
 #ifdef CONFIG_PPC_ISERIES
        /*
         * At offset 0x20, there is a pointer to iSeries LPAR data.
@@ -103,12 +96,12 @@ END_FTR_SECTION(0, 1)
        .llong hvReleaseData-KERNELBASE
        /*
-         * At offset 0x28 and 0x30 are offsets to the msChunks
+         * At offset 0x28 and 0x30 are offsets to the mschunks_map
         * array (used by the iSeries LPAR debugger to do translation
         * between physical addresses and absolute addresses) and
         * to the pidhash table (also used by the debugger)
         */
-        .llong msChunks-KERNELBASE
+        .llong mschunks_map-KERNELBASE
        .llong 0        /* pidhash-KERNELBASE SFRXXX */
        /* Offset 0x38 - Pointer to start of embedded System.map */
@@ -120,7 +113,7 @@ embedded_sysmap_start:
 embedded_sysmap_end:
        .llong  0
-#else /* CONFIG_PPC_ISERIES */
+#endif /* CONFIG_PPC_ISERIES */
        /* Secondary processors spin on this value until it goes to 1. */
        .globl  __secondary_hold_spinloop
@@ -155,7 +148,7 @@ _GLOBAL(__secondary_hold)
        std     r24,__secondary_hold_acknowledge@l(0)
        sync
-        /* All secondary cpu's wait here until told to start. */
+        /* All secondary cpus wait here until told to start. */
 100:    ld      r4,__secondary_hold_spinloop@l(0)
        cmpdi   0,r4,1
        bne     100b
@@ -170,7 +163,6 @@ _GLOBAL(__secondary_hold)
        BUG_OPCODE
 #endif
 #endif
-#endif
 /* This value is used to mark exception frames on the stack. */
        .section ".toc","aw"
@@ -502,33 +494,37 @@ system_call_pSeries:
        STD_EXCEPTION_PSERIES(0x1300, instruction_breakpoint)
        STD_EXCEPTION_PSERIES(0x1700, altivec_assist)
+        . = 0x3000
+/*** pSeries interrupt support ***/
        /* moved from 0xf00 */
-        STD_EXCEPTION_PSERIES(0x3000, performance_monitor)
+        STD_EXCEPTION_PSERIES(., performance_monitor)
-        . = 0x3100
+        .align  7
 _GLOBAL(do_stab_bolted_pSeries)
        mtcrf   0x80,r12
        mfspr   r12,SPRG2
        EXCEPTION_PROLOG_PSERIES(PACA_EXSLB, .do_stab_bolted)
-        
+/*
-        /* Space for the naca.  Architected to be located at real address
+ * Vectors for the FWNMI option.  Share common code.
-         * NACA_PHYS_ADDR.  Various tools rely on this location being fixed.
+ */
-         * The first dword of the naca is required by iSeries LPAR to
+      .globl system_reset_fwnmi
-         * point to itVpdAreas.  On pSeries native, this value is not used.
+system_reset_fwnmi:
-         */
+      HMT_MEDIUM
-        . = NACA_PHYS_ADDR
+      mtspr   SPRG1,r13               /* save r13 */
-        .globl __end_interrupts
+      RUNLATCH_ON(r13)
-__end_interrupts:
+      EXCEPTION_PROLOG_PSERIES(PACA_EXGEN, system_reset_common)
-#ifdef CONFIG_PPC_ISERIES
-        .globl naca
-naca:
-        .llong  itVpdAreas
-        .llong  0               /* xRamDisk */
-        .llong  0               /* xRamDiskSize */
-        . = 0x6100
+      .globl machine_check_fwnmi
+machine_check_fwnmi:
+      HMT_MEDIUM
+      mtspr   SPRG1,r13               /* save r13 */
+      RUNLATCH_ON(r13)
+      EXCEPTION_PROLOG_PSERIES(PACA_EXMC, machine_check_common)
+#ifdef CONFIG_PPC_ISERIES
 /***  ISeries-LPAR interrupt handlers ***/
        STD_EXCEPTION_ISERIES(0x200, machine_check, PACA_EXMC)
@@ -626,9 +622,7 @@ system_reset_iSeries:
        cmpwi   0,r23,0
        beq     iSeries_secondary_smp_loop      /* Loop until told to go */
-#ifdef SECONDARY_PROCESSORS
        bne     .__secondary_start              /* Loop until told to go */
-#endif
 iSeries_secondary_smp_loop:
        /* Let the Hypervisor know we are alive */
        /* 8002 is a call to HvCallCfg::getLps, a harmless Hypervisor function */
@@ -671,51 +665,8 @@ hardware_interrupt_iSeries_masked:
        ld      r13,PACA_EXGEN+EX_R13(r13)
        rfid
        b       .       /* prevent speculative execution */
-#endif
-/*
- * Data area reserved for FWNMI option.
- */
-        .= 0x7000
-        .globl fwnmi_data_area
-fwnmi_data_area:
-#ifdef CONFIG_PPC_ISERIES
-        . = LPARMAP_PHYS
-#include "lparmap.s"
 #endif /* CONFIG_PPC_ISERIES */
-/*
- * Vectors for the FWNMI option.  Share common code.
- */
-        . = 0x8000
-        .globl system_reset_fwnmi
-system_reset_fwnmi:
-        HMT_MEDIUM
-        mtspr   SPRG1,r13               /* save r13 */
-        RUNLATCH_ON(r13)
-        EXCEPTION_PROLOG_PSERIES(PACA_EXGEN, system_reset_common)
-        .globl machine_check_fwnmi
-machine_check_fwnmi:
-        HMT_MEDIUM
-        mtspr   SPRG1,r13               /* save r13 */
-        RUNLATCH_ON(r13)
-        EXCEPTION_PROLOG_PSERIES(PACA_EXMC, machine_check_common)
-        /*
-         * Space for the initial segment table
-         * For LPAR, the hypervisor must fill in at least one entry
-         * before we get control (with relocate on)
-         */
-        . = STAB0_PHYS_ADDR
-        .globl __start_stab
-__start_stab:
-        . = (STAB0_PHYS_ADDR + PAGE_SIZE)
-        .globl __end_stab
-__end_stab:
 /*** Common interrupt handlers ***/
        STD_EXCEPTION_COMMON(0x100, system_reset, .system_reset_exception)
@@ -752,8 +703,8 @@ machine_check_common:
 * R9 contains the saved CR, r13 points to the paca,
 * r10 contains the (bad) kernel stack pointer,
 * r11 and r12 contain the saved SRR0 and SRR1.
- * We switch to using the paca guard page as an emergency stack,
+ * We switch to using an emergency stack, save the registers there,
- * save the registers there, and call kernel_bad_stack(), which panics.
+ * and call kernel_bad_stack(), which panics.
 */
 bad_stack:
        ld      r1,PACAEMERGSP(r13)
@@ -906,6 +857,62 @@ fp_unavailable_common:
        bl      .kernel_fp_unavailable_exception
        BUG_OPCODE
+/*
+ * load_up_fpu(unused, unused, tsk)
+ * Disable FP for the task which had the FPU previously,
+ * and save its floating-point registers in its thread_struct.
+ * Enables the FPU for use in the kernel on return.
+ * On SMP we know the fpu is free, since we give it up every
+ * switch (ie, no lazy save of the FP registers).
+ * On entry: r13 == 'current' && last_task_used_math != 'current'
+ */
+_STATIC(load_up_fpu)
+        mfmsr   r5                      /* grab the current MSR */
+        ori     r5,r5,MSR_FP
+        mtmsrd  r5                      /* enable use of fpu now */
+        isync
+/*
+ * For SMP, we don't do lazy FPU switching because it just gets too
+ * horrendously complex, especially when a task switches from one CPU
+ * to another.  Instead we call giveup_fpu in switch_to.
+ *
+ */
+#ifndef CONFIG_SMP
+        ld      r3,last_task_used_math@got(r2)
+        ld      r4,0(r3)
+        cmpdi   0,r4,0
+        beq     1f
+        /* Save FP state to last_task_used_math's THREAD struct */
+        addi    r4,r4,THREAD
+        SAVE_32FPRS(0, r4)
+        mffs    fr0
+        stfd    fr0,THREAD_FPSCR(r4)
+        /* Disable FP for last_task_used_math */
+        ld      r5,PT_REGS(r4)
+        ld      r4,_MSR-STACK_FRAME_OVERHEAD(r5)
+        li      r6,MSR_FP|MSR_FE0|MSR_FE1
+        andc    r4,r4,r6
+        std     r4,_MSR-STACK_FRAME_OVERHEAD(r5)
+1:
+#endif /* CONFIG_SMP */
+        /* enable use of FP after return */
+        ld      r4,PACACURRENT(r13)
+        addi    r5,r4,THREAD            /* Get THREAD */
+        ld      r4,THREAD_FPEXC_MODE(r5)
+        ori     r12,r12,MSR_FP
+        or      r12,r12,r4
+        std     r12,_MSR(r1)
+        lfd     fr0,THREAD_FPSCR(r5)
+        mtfsf   0xff,fr0
+        REST_32FPRS(0, r5)
+#ifndef CONFIG_SMP
+        /* Update last_task_used_math to 'current' */
+        subi    r4,r5,THREAD            /* Back to 'current' */
+        std     r4,0(r3)
+#endif /* CONFIG_SMP */
+        /* restore registers and return */
+        b       fast_exception_return
        .align  7
        .globl altivec_unavailable_common
 altivec_unavailable_common:
@@ -921,6 +928,80 @@ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
        bl      .altivec_unavailable_exception
        b       .ret_from_except
+#ifdef CONFIG_ALTIVEC
+/*
+ * load_up_altivec(unused, unused, tsk)
+ * Disable VMX for the task which had it previously,
+ * and save its vector registers in its thread_struct.
+ * Enables the VMX for use in the kernel on return.
+ * On SMP we know the VMX is free, since we give it up every
+ * switch (ie, no lazy save of the vector registers).
+ * On entry: r13 == 'current' && last_task_used_altivec != 'current'
+ */
+_STATIC(load_up_altivec)
+        mfmsr   r5                      /* grab the current MSR */
+        oris    r5,r5,MSR_VEC@h
+        mtmsrd  r5                      /* enable use of VMX now */
+        isync
+/*
+ * For SMP, we don't do lazy VMX switching because it just gets too
+ * horrendously complex, especially when a task switches from one CPU
+ * to another.  Instead we call giveup_altvec in switch_to.
+ * VRSAVE isn't dealt with here, that is done in the normal context
+ * switch code. Note that we could rely on vrsave value to eventually
+ * avoid saving all of the VREGs here...
+ */
+#ifndef CONFIG_SMP
+        ld      r3,last_task_used_altivec@got(r2)
+        ld      r4,0(r3)
+        cmpdi   0,r4,0
+        beq     1f
+        /* Save VMX state to last_task_used_altivec's THREAD struct */
+        addi    r4,r4,THREAD
+        SAVE_32VRS(0,r5,r4)
+        mfvscr  vr0
+        li      r10,THREAD_VSCR
+        stvx    vr0,r10,r4
+        /* Disable VMX for last_task_used_altivec */
+        ld      r5,PT_REGS(r4)
+        ld      r4,_MSR-STACK_FRAME_OVERHEAD(r5)
+        lis     r6,MSR_VEC@h
+        andc    r4,r4,r6
+        std     r4,_MSR-STACK_FRAME_OVERHEAD(r5)
+1:
+#endif /* CONFIG_SMP */
+        /* Hack: if we get an altivec unavailable trap with VRSAVE
+         * set to all zeros, we assume this is a broken application
+         * that fails to set it properly, and thus we switch it to
+         * all 1's
+         */
+        mfspr   r4,SPRN_VRSAVE
+        cmpdi   0,r4,0
+        bne+    1f
+        li      r4,-1
+        mtspr   SPRN_VRSAVE,r4
+1:
+        /* enable use of VMX after return */
+        ld      r4,PACACURRENT(r13)
+        addi    r5,r4,THREAD            /* Get THREAD */
+        oris    r12,r12,MSR_VEC@h
+        std     r12,_MSR(r1)
+        li      r4,1
+        li      r10,THREAD_VSCR
+        stw     r4,THREAD_USED_VR(r5)
+        lvx     vr0,r10,r5
+        mtvscr  vr0
+        REST_32VRS(0,r4,r5)
+#ifndef CONFIG_SMP
+        /* Update last_task_used_math to 'current' */
+        subi    r4,r5,THREAD            /* Back to 'current' */
+        std     r4,0(r3)
+#endif /* CONFIG_SMP */
+        /* restore registers and return */
+        b       fast_exception_return
+#endif /* CONFIG_ALTIVEC */
 /*
 * Hash table stuff
 */
@@ -1167,6 +1248,42 @@ unrecov_slb:
        bl      .unrecoverable_exception
        b       1b
+/*
+ * Space for CPU0's segment table.
+ *
+ * On iSeries, the hypervisor must fill in at least one entry before
+ * we get control (with relocate on).  The address is give to the hv
+ * as a page number (see xLparMap in LparData.c), so this must be at a
+ * fixed address (the linker can't compute (u64)&initial_stab >>
+ * PAGE_SHIFT).
+ */
+        . = STAB0_PHYS_ADDR     /* 0x6000 */
+        .globl initial_stab
+initial_stab:
+        .space  4096
+/*
+ * Data area reserved for FWNMI option.
+ * This address (0x7000) is fixed by the RPA.
+ */
+        .= 0x7000
+        .globl fwnmi_data_area
+fwnmi_data_area:
+        /* iSeries does not use the FWNMI stuff, so it is safe to put
+         * this here, even if we later allow kernels that will boot on
+         * both pSeries and iSeries */
+#ifdef CONFIG_PPC_ISERIES
+        . = LPARMAP_PHYS
+#include "lparmap.s"
+/*
+ * This ".text" is here for old compilers that generate a trailing
+ * .note section when compiling .c files to .s
+ */
+        .text
+#endif /* CONFIG_PPC_ISERIES */
+        . = 0x8000
 /*
 * On pSeries, secondary processors spin in the following code.
@@ -1200,7 +1317,7 @@ _GLOBAL(pSeries_secondary_smp_init)
        b       .kexec_wait             /* next kernel might do better   */
 2:      mtspr   SPRG3,r13               /* Save vaddr of paca in SPRG3   */
-        /* From now on, r24 is expected to be logica cpuid */
+        /* From now on, r24 is expected to be logical cpuid */
        mr      r24,r5
 3:      HMT_LOW
        lbz     r23,PACAPROCSTART(r13)  /* Test if this processor should */
@@ -1213,10 +1330,8 @@ _GLOBAL(pSeries_secondary_smp_init)
        cmpwi   0,r23,0
 #ifdef CONFIG_SMP
-#ifdef SECONDARY_PROCESSORS
        bne     .__secondary_start
 #endif
-#endif
        b       3b                      /* Loop until told to go         */
 #ifdef CONFIG_PPC_ISERIES
@@ -1430,228 +1545,6 @@ _GLOBAL(copy_and_flush)
 .align 8
 copy_to_here:
-/*
- * load_up_fpu(unused, unused, tsk)
- * Disable FP for the task which had the FPU previously,
- * and save its floating-point registers in its thread_struct.
- * Enables the FPU for use in the kernel on return.
- * On SMP we know the fpu is free, since we give it up every
- * switch (ie, no lazy save of the FP registers).
- * On entry: r13 == 'current' && last_task_used_math != 'current'
- */
-_STATIC(load_up_fpu)
-        mfmsr   r5                      /* grab the current MSR */
-        ori     r5,r5,MSR_FP
-        mtmsrd  r5                      /* enable use of fpu now */
-        isync
-/*
- * For SMP, we don't do lazy FPU switching because it just gets too
- * horrendously complex, especially when a task switches from one CPU
- * to another.  Instead we call giveup_fpu in switch_to.
- *
- */
-#ifndef CONFIG_SMP
-        ld      r3,last_task_used_math@got(r2)
-        ld      r4,0(r3)
-        cmpdi   0,r4,0
-        beq     1f
-        /* Save FP state to last_task_used_math's THREAD struct */
-        addi    r4,r4,THREAD
-        SAVE_32FPRS(0, r4)
-        mffs    fr0
-        stfd    fr0,THREAD_FPSCR(r4)
-        /* Disable FP for last_task_used_math */
-        ld      r5,PT_REGS(r4)
-        ld      r4,_MSR-STACK_FRAME_OVERHEAD(r5)
-        li      r6,MSR_FP|MSR_FE0|MSR_FE1
-        andc    r4,r4,r6
-        std     r4,_MSR-STACK_FRAME_OVERHEAD(r5)
-1:
-#endif /* CONFIG_SMP */
-        /* enable use of FP after return */
-        ld      r4,PACACURRENT(r13)
-        addi    r5,r4,THREAD            /* Get THREAD */
-        ld      r4,THREAD_FPEXC_MODE(r5)
-        ori     r12,r12,MSR_FP
-        or      r12,r12,r4
-        std     r12,_MSR(r1)
-        lfd     fr0,THREAD_FPSCR(r5)
-        mtfsf   0xff,fr0
-        REST_32FPRS(0, r5)
-#ifndef CONFIG_SMP
-        /* Update last_task_used_math to 'current' */
-        subi    r4,r5,THREAD            /* Back to 'current' */
-        std     r4,0(r3)
-#endif /* CONFIG_SMP */
-        /* restore registers and return */
-        b       fast_exception_return
-/*
- * disable_kernel_fp()
- * Disable the FPU.
- */
-_GLOBAL(disable_kernel_fp)
-        mfmsr   r3
-        rldicl  r0,r3,(63-MSR_FP_LG),1
-        rldicl  r3,r0,(MSR_FP_LG+1),0
-        mtmsrd  r3                      /* disable use of fpu now */
-        isync
-        blr
-/*
- * giveup_fpu(tsk)
- * Disable FP for the task given as the argument,
- * and save the floating-point registers in its thread_struct.
- * Enables the FPU for use in the kernel on return.
- */
-_GLOBAL(giveup_fpu)
-        mfmsr   r5
-        ori     r5,r5,MSR_FP
-        mtmsrd  r5                      /* enable use of fpu now */
-        isync
-        cmpdi   0,r3,0
-        beqlr-                          /* if no previous owner, done */
-        addi    r3,r3,THREAD            /* want THREAD of task */
-        ld      r5,PT_REGS(r3)
-        cmpdi   0,r5,0
-        SAVE_32FPRS(0, r3)
-        mffs    fr0
-        stfd    fr0,THREAD_FPSCR(r3)
-        beq     1f
-        ld      r4,_MSR-STACK_FRAME_OVERHEAD(r5)
-        li      r3,MSR_FP|MSR_FE0|MSR_FE1
-        andc    r4,r4,r3                /* disable FP for previous task */
-        std     r4,_MSR-STACK_FRAME_OVERHEAD(r5)
-1:
-#ifndef CONFIG_SMP
-        li      r5,0
-        ld      r4,last_task_used_math@got(r2)
-        std     r5,0(r4)
-#endif /* CONFIG_SMP */
-        blr
-#ifdef CONFIG_ALTIVEC
-                
-/*
- * load_up_altivec(unused, unused, tsk)
- * Disable VMX for the task which had it previously,
- * and save its vector registers in its thread_struct.
- * Enables the VMX for use in the kernel on return.
- * On SMP we know the VMX is free, since we give it up every
- * switch (ie, no lazy save of the vector registers).
- * On entry: r13 == 'current' && last_task_used_altivec != 'current'
- */
-_STATIC(load_up_altivec)
-        mfmsr   r5                      /* grab the current MSR */
-        oris    r5,r5,MSR_VEC@h
-        mtmsrd  r5                      /* enable use of VMX now */
-        isync
-        
-/*
- * For SMP, we don't do lazy VMX switching because it just gets too
- * horrendously complex, especially when a task switches from one CPU
- * to another.  Instead we call giveup_altvec in switch_to.
- * VRSAVE isn't dealt with here, that is done in the normal context
- * switch code. Note that we could rely on vrsave value to eventually
- * avoid saving all of the VREGs here...
- */
-#ifndef CONFIG_SMP
-        ld      r3,last_task_used_altivec@got(r2)
-        ld      r4,0(r3)
-        cmpdi   0,r4,0
-        beq     1f
-        /* Save VMX state to last_task_used_altivec's THREAD struct */
-        addi    r4,r4,THREAD
-        SAVE_32VRS(0,r5,r4)
-        mfvscr  vr0
-        li      r10,THREAD_VSCR
-        stvx    vr0,r10,r4
-        /* Disable VMX for last_task_used_altivec */
-        ld      r5,PT_REGS(r4)
-        ld      r4,_MSR-STACK_FRAME_OVERHEAD(r5)
-        lis     r6,MSR_VEC@h
-        andc    r4,r4,r6
-        std     r4,_MSR-STACK_FRAME_OVERHEAD(r5)
-1:
-#endif /* CONFIG_SMP */
-        /* Hack: if we get an altivec unavailable trap with VRSAVE
-         * set to all zeros, we assume this is a broken application
-         * that fails to set it properly, and thus we switch it to
-         * all 1's
-         */
-        mfspr   r4,SPRN_VRSAVE
-        cmpdi   0,r4,0
-        bne+    1f
-        li      r4,-1
-        mtspr   SPRN_VRSAVE,r4
-1:
-        /* enable use of VMX after return */
-        ld      r4,PACACURRENT(r13)
-        addi    r5,r4,THREAD            /* Get THREAD */
-        oris    r12,r12,MSR_VEC@h
-        std     r12,_MSR(r1)
-        li      r4,1
-        li      r10,THREAD_VSCR
-        stw     r4,THREAD_USED_VR(r5)
-        lvx     vr0,r10,r5
-        mtvscr  vr0
-        REST_32VRS(0,r4,r5)
-#ifndef CONFIG_SMP
-        /* Update last_task_used_math to 'current' */
-        subi    r4,r5,THREAD            /* Back to 'current' */
-        std     r4,0(r3)
-#endif /* CONFIG_SMP */
-        /* restore registers and return */
-        b       fast_exception_return
-/*
- * disable_kernel_altivec()
- * Disable the VMX.
- */
-_GLOBAL(disable_kernel_altivec)
-        mfmsr   r3
-        rldicl  r0,r3,(63-MSR_VEC_LG),1
-        rldicl  r3,r0,(MSR_VEC_LG+1),0
-        mtmsrd  r3                      /* disable use of VMX now */
-        isync
-        blr
-/*
- * giveup_altivec(tsk)
- * Disable VMX for the task given as the argument,
- * and save the vector registers in its thread_struct.
- * Enables the VMX for use in the kernel on return.
- */
-_GLOBAL(giveup_altivec)
-        mfmsr   r5
-        oris    r5,r5,MSR_VEC@h
-        mtmsrd  r5                      /* enable use of VMX now */
-        isync
-        cmpdi   0,r3,0
-        beqlr-                          /* if no previous owner, done */
-        addi    r3,r3,THREAD            /* want THREAD of task */
-        ld      r5,PT_REGS(r3)
-        cmpdi   0,r5,0
-        SAVE_32VRS(0,r4,r3)
-        mfvscr  vr0
-        li      r4,THREAD_VSCR
-        stvx    vr0,r4,r3
-        beq     1f
-        ld      r4,_MSR-STACK_FRAME_OVERHEAD(r5)
-        lis     r3,MSR_VEC@h
-        andc    r4,r4,r3                /* disable FP for previous task */
-        std     r4,_MSR-STACK_FRAME_OVERHEAD(r5)
-1:
-#ifndef CONFIG_SMP
-        li      r5,0
-        ld      r4,last_task_used_altivec@got(r2)
-        std     r5,0(r4)
-#endif /* CONFIG_SMP */
-        blr
-#endif /* CONFIG_ALTIVEC */
 #ifdef CONFIG_SMP
 #ifdef CONFIG_PPC_PMAC
 /*
@@ -1753,8 +1646,9 @@ _GLOBAL(__secondary_start)
 #else
        /* set the ASR */
        ld      r3,systemcfg@got(r2)    /* r3 = ptr to systemcfg         */
+        ld      r3,0(r3)
        lwz     r3,PLATFORM(r3)         /* r3 = platform flags           */
-        cmpldi  r3,PLATFORM_PSERIES_LPAR
+        andi.   r3,r3,PLATFORM_LPAR     /* Test if bit 0 is set (LPAR bit) */
        bne     98f
        mfspr   r3,PVR
        srwi    r3,r3,16
@@ -1916,8 +1810,9 @@ _STATIC(start_here_multiplatform)
        ld      r3,PACASTABREAL(r13)
        ori     r4,r3,1                 /* turn on valid bit             */
        ld      r3,systemcfg@got(r2)    /* r3 = ptr to systemcfg */
+        ld      r3,0(r3)
        lwz     r3,PLATFORM(r3)         /* r3 = platform flags */
-        cmpldi  r3,PLATFORM_PSERIES_LPAR
+        andi.   r3,r3,PLATFORM_LPAR     /* Test if bit 0 is set (LPAR bit) */
        bne     98f
        mfspr   r3,PVR
        srwi    r3,r3,16
@@ -1935,9 +1830,10 @@ _STATIC(start_here_multiplatform)
 99:
        /* Set SDR1 (hash table pointer) */
        ld      r3,systemcfg@got(r2)    /* r3 = ptr to systemcfg */
+        ld      r3,0(r3)
        lwz     r3,PLATFORM(r3)         /* r3 = platform flags */
        /* Test if bit 0 is set (LPAR bit) */
-        andi.   r3,r3,0x1
+        andi.   r3,r3,PLATFORM_LPAR
        bne     98f
        LOADADDR(r6,_SDR1)              /* Only if NOT LPAR */
        sub     r6,r6,r26
@@ -2002,9 +1898,6 @@ _STATIC(start_here_common)
        bl .start_kernel
-_GLOBAL(__setup_cpu_power3)
-        blr
 _GLOBAL(hmt_init)
 #ifdef CONFIG_HMT
        LOADADDR(r5, hmt_thread_data)
@@ -2095,20 +1988,19 @@ _GLOBAL(smp_release_cpus)
 /*
 * We put a few things here that have to be page-aligned.
- * This stuff goes at the beginning of the data segment,
+ * This stuff goes at the beginning of the bss, which is page-aligned.
- * which is page-aligned.
 */
-        .data
+        .section ".bss"
        .align  12
-        .globl  sdata
-sdata:
        .globl  empty_zero_page
 empty_zero_page:
-        .space  4096
+        .space  PAGE_SIZE
        .globl  swapper_pg_dir
 swapper_pg_dir:
-        .space  4096
+        .space  PAGE_SIZE
 /*
 * This space gets a copy of optional info passed to us by the bootstrap
diff --git a/arch/ppc64/kernel/iSeries_VpdInfo.c b/arch/ppc64/kernel/iSeries_VpdInfo.c
index d11c732daf81..5d921792571f 100644
--- a/arch/ppc64/kernel/iSeries_VpdInfo.c
+++ b/arch/ppc64/kernel/iSeries_VpdInfo.c
@@ -264,8 +264,5 @@ void __init iSeries_Device_Information(struct pci_dev *PciDev, int count)
        printk("%d. PCI: Bus%3d, Device%3d, Vendor %04X Frame%3d, Card %4s  ",
                        count, bus, PCI_SLOT(PciDev->devfn), PciDev->vendor,
                        frame, card);
-        if (pci_class_name(PciDev->class >> 8) == 0)
+        printk("0x%04X\n", (int)(PciDev->class >> 8));
-                printk("0x%04X\n", (int)(PciDev->class >> 8));
-        else
-                printk("%s\n", pci_class_name(PciDev->class >> 8));
 }
diff --git a/arch/ppc64/kernel/iSeries_htab.c b/arch/ppc64/kernel/iSeries_htab.c
index b0250ae4a72a..2192055a90a0 100644
--- a/arch/ppc64/kernel/iSeries_htab.c
+++ b/arch/ppc64/kernel/iSeries_htab.c
@@ -41,6 +41,7 @@ static long iSeries_hpte_insert(unsigned long hpte_group, unsigned long va,
                                unsigned long prpn, unsigned long vflags,
                                unsigned long rflags)
 {
+        unsigned long arpn;
        long slot;
        hpte_t lhpte;
        int secondary = 0;
@@ -70,8 +71,10 @@ static long iSeries_hpte_insert(unsigned long hpte_group, unsigned long va,
                slot &= 0x7fffffffffffffff;
        }
+        arpn = phys_to_abs(prpn << PAGE_SHIFT) >> PAGE_SHIFT;
        lhpte.v = (va >> 23) << HPTE_V_AVPN_SHIFT | vflags | HPTE_V_VALID;
-        lhpte.r = (physRpn_to_absRpn(prpn) << HPTE_R_RPN_SHIFT) | rflags;
+        lhpte.r = (arpn << HPTE_R_RPN_SHIFT) | rflags;
        /* Now fill in the actual HPTE */
        HvCallHpt_addValidate(slot, secondary, &lhpte);
diff --git a/arch/ppc64/kernel/iSeries_pci.c b/arch/ppc64/kernel/iSeries_pci.c
index 356e4fd9a94f..fbc273c32bcc 100644
--- a/arch/ppc64/kernel/iSeries_pci.c
+++ b/arch/ppc64/kernel/iSeries_pci.c
@@ -252,7 +252,7 @@ unsigned long __init find_and_init_phbs(void)
                        phb = (struct pci_controller *)kmalloc(sizeof(struct pci_controller), GFP_KERNEL);
                        if (phb == NULL)
                                return -ENOMEM;
-                        pci_setup_pci_controller(phb);
+                        pci_setup_pci_controller(phb);
                        phb->pci_mem_offset = phb->local_number = bus;
                        phb->first_busno = bus;
diff --git a/arch/ppc64/kernel/iSeries_setup.c b/arch/ppc64/kernel/iSeries_setup.c
index a649edbb23b6..3ffefbbc6623 100644
--- a/arch/ppc64/kernel/iSeries_setup.c
+++ b/arch/ppc64/kernel/iSeries_setup.c
@@ -39,6 +39,7 @@
 #include <asm/cputable.h>
 #include <asm/sections.h>
 #include <asm/iommu.h>
+#include <asm/firmware.h>
 #include <asm/time.h>
 #include "iSeries_setup.h"
@@ -314,6 +315,8 @@ static void __init iSeries_init_early(void)
        DBG(" -> iSeries_init_early()\n");
+        ppc64_firmware_features = FW_FEATURE_ISERIES;
        ppcdbg_initialize();
 #if defined(CONFIG_BLK_DEV_INITRD)
@@ -412,6 +415,22 @@ static void __init iSeries_init_early(void)
        DBG(" <- iSeries_init_early()\n");
 }
+struct mschunks_map mschunks_map = {
+        /* XXX We don't use these, but Piranha might need them. */
+        .chunk_size  = MSCHUNKS_CHUNK_SIZE,
+        .chunk_shift = MSCHUNKS_CHUNK_SHIFT,
+        .chunk_mask  = MSCHUNKS_OFFSET_MASK,
+};
+EXPORT_SYMBOL(mschunks_map);
+void mschunks_alloc(unsigned long num_chunks)
+{
+        klimit = _ALIGN(klimit, sizeof(u32));
+        mschunks_map.mapping = (u32 *)klimit;
+        klimit += num_chunks * sizeof(u32);
+        mschunks_map.num_chunks = num_chunks;
+}
 /*
 * The iSeries may have very large memories ( > 128 GB ) and a partition
 * may get memory in "chunks" that may be anywhere in the 2**52 real
@@ -449,7 +468,7 @@ static void __init build_iSeries_Memory_Map(void)
        /* Chunk size on iSeries is 256K bytes */
        totalChunks = (u32)HvLpConfig_getMsChunks();
-        klimit = msChunks_alloc(klimit, totalChunks, 1UL << 18);
+        mschunks_alloc(totalChunks);
        /*
         * Get absolute address of our load area
@@ -486,7 +505,7 @@ static void __init build_iSeries_Memory_Map(void)
        printk("Load area size %dK\n", loadAreaSize * 256);
        for (nextPhysChunk = 0; nextPhysChunk < loadAreaSize; ++nextPhysChunk)
-                msChunks.abs[nextPhysChunk] =
+                mschunks_map.mapping[nextPhysChunk] =
                        loadAreaFirstChunk + nextPhysChunk;
        /*
@@ -495,7 +514,7 @@ static void __init build_iSeries_Memory_Map(void)
         */
        hptFirstChunk = (u32)addr_to_chunk(HvCallHpt_getHptAddress());
        hptSizePages = (u32)HvCallHpt_getHptPages();
-        hptSizeChunks = hptSizePages >> (msChunks.chunk_shift - PAGE_SHIFT);
+        hptSizeChunks = hptSizePages >> (MSCHUNKS_CHUNK_SHIFT - PAGE_SHIFT);
        hptLastChunk = hptFirstChunk + hptSizeChunks - 1;
        printk("HPT absolute addr = %016lx, size = %dK\n",
@@ -552,7 +571,8 @@ static void __init build_iSeries_Memory_Map(void)
                                     (absChunk > hptLastChunk)) &&
                                    ((absChunk < loadAreaFirstChunk) ||
                                     (absChunk > loadAreaLastChunk))) {
-                                        msChunks.abs[nextPhysChunk] = absChunk;
+                                        mschunks_map.mapping[nextPhysChunk] =
+                                                absChunk;
                                        ++nextPhysChunk;
                                }
                        }
@@ -944,6 +964,8 @@ void __init iSeries_early_setup(void)
        ppc_md.calibrate_decr = iSeries_calibrate_decr;
        ppc_md.progress = iSeries_progress;
+        /* XXX Implement enable_pmcs for iSeries */
        if (get_paca()->lppaca.shared_proc) {
                ppc_md.idle_loop = iseries_shared_idle;
                printk(KERN_INFO "Using shared processor idle loop\n");
diff --git a/arch/ppc64/kernel/iSeries_vio.c b/arch/ppc64/kernel/iSeries_vio.c
new file mode 100644
index 000000000000..6b754b0c8344
--- /dev/null
+++ b/arch/ppc64/kernel/iSeries_vio.c
@@ -0,0 +1,155 @@
+/*
+ * IBM PowerPC iSeries Virtual I/O Infrastructure Support.
+ *
+ *    Copyright (c) 2005 Stephen Rothwell, IBM Corp.
+ *
+ *      This program is free software; you can redistribute it and/or
+ *      modify it under the terms of the GNU General Public License
+ *      as published by the Free Software Foundation; either version
+ *      2 of the License, or (at your option) any later version.
+ */
+#include <linux/types.h>
+#include <linux/device.h>
+#include <linux/init.h>
+#include <asm/vio.h>
+#include <asm/iommu.h>
+#include <asm/abs_addr.h>
+#include <asm/page.h>
+#include <asm/iSeries/vio.h>
+#include <asm/iSeries/HvTypes.h>
+#include <asm/iSeries/HvLpConfig.h>
+#include <asm/iSeries/HvCallXm.h>
+struct device *iSeries_vio_dev = &vio_bus_device.dev;
+EXPORT_SYMBOL(iSeries_vio_dev);
+static struct iommu_table veth_iommu_table;
+static struct iommu_table vio_iommu_table;
+static void __init iommu_vio_init(void)
+{
+        struct iommu_table *t;
+        struct iommu_table_cb cb;
+        unsigned long cbp;
+        unsigned long itc_entries;
+        cb.itc_busno = 255;    /* Bus 255 is the virtual bus */
+        cb.itc_virtbus = 0xff; /* Ask for virtual bus */
+        cbp = virt_to_abs(&cb);
+        HvCallXm_getTceTableParms(cbp);
+        itc_entries = cb.itc_size * PAGE_SIZE / sizeof(union tce_entry);
+        veth_iommu_table.it_size        = itc_entries / 2;
+        veth_iommu_table.it_busno       = cb.itc_busno;
+        veth_iommu_table.it_offset      = cb.itc_offset;
+        veth_iommu_table.it_index       = cb.itc_index;
+        veth_iommu_table.it_type        = TCE_VB;
+        veth_iommu_table.it_blocksize   = 1;
+        t = iommu_init_table(&veth_iommu_table);
+        if (!t)
+                printk("Virtual Bus VETH TCE table failed.\n");
+        vio_iommu_table.it_size         = itc_entries - veth_iommu_table.it_size;
+        vio_iommu_table.it_busno        = cb.itc_busno;
+        vio_iommu_table.it_offset       = cb.itc_offset +
+                                          veth_iommu_table.it_size;
+        vio_iommu_table.it_index        = cb.itc_index;
+        vio_iommu_table.it_type         = TCE_VB;
+        vio_iommu_table.it_blocksize    = 1;
+        t = iommu_init_table(&vio_iommu_table);
+        if (!t)
+                printk("Virtual Bus VIO TCE table failed.\n");
+}
+/**
+ * vio_register_device_iseries: - Register a new iSeries vio device.
+ * @voidev:     The device to register.
+ */
+static struct vio_dev *__init vio_register_device_iseries(char *type,
+                uint32_t unit_num)
+{
+        struct vio_dev *viodev;
+        /* allocate a vio_dev for this device */
+        viodev = kmalloc(sizeof(struct vio_dev), GFP_KERNEL);
+        if (!viodev)
+                return NULL;
+        memset(viodev, 0, sizeof(struct vio_dev));
+        snprintf(viodev->dev.bus_id, BUS_ID_SIZE, "%s%d", type, unit_num);
+        viodev->name = viodev->dev.bus_id;
+        viodev->type = type;
+        viodev->unit_address = unit_num;
+        viodev->iommu_table = &vio_iommu_table;
+        if (vio_register_device(viodev) == NULL) {
+                kfree(viodev);
+                return NULL;
+        }
+        return viodev;
+}
+void __init probe_bus_iseries(void)
+{
+        HvLpIndexMap vlan_map;
+        struct vio_dev *viodev;
+        int i;
+        /* there is only one of each of these */
+        vio_register_device_iseries("viocons", 0);
+        vio_register_device_iseries("vscsi", 0);
+        vlan_map = HvLpConfig_getVirtualLanIndexMap();
+        for (i = 0; i < HVMAXARCHITECTEDVIRTUALLANS; i++) {
+                if ((vlan_map & (0x8000 >> i)) == 0)
+                        continue;
+                viodev = vio_register_device_iseries("vlan", i);
+                /* veth is special and has it own iommu_table */
+                viodev->iommu_table = &veth_iommu_table;
+        }
+        for (i = 0; i < HVMAXARCHITECTEDVIRTUALDISKS; i++)
+                vio_register_device_iseries("viodasd", i);
+        for (i = 0; i < HVMAXARCHITECTEDVIRTUALCDROMS; i++)
+                vio_register_device_iseries("viocd", i);
+        for (i = 0; i < HVMAXARCHITECTEDVIRTUALTAPES; i++)
+                vio_register_device_iseries("viotape", i);
+}
+/**
+ * vio_match_device_iseries: - Tell if a iSeries VIO device matches a
+ *      vio_device_id
+ */
+static int vio_match_device_iseries(const struct vio_device_id *id,
+                const struct vio_dev *dev)
+{
+        return strncmp(dev->type, id->type, strlen(id->type)) == 0;
+}
+static struct vio_bus_ops vio_bus_ops_iseries = {
+        .match = vio_match_device_iseries,
+};
+/**
+ * vio_bus_init_iseries: - Initialize the iSeries virtual IO bus
+ */
+static int __init vio_bus_init_iseries(void)
+{
+        int err;
+        err = vio_bus_init(&vio_bus_ops_iseries);
+        if (err == 0) {
+                iommu_vio_init();
+                vio_bus_device.iommu_table = &vio_iommu_table;
+                iSeries_vio_dev = &vio_bus_device.dev;
+                probe_bus_iseries();
+        }
+        return err;
+}
+__initcall(vio_bus_init_iseries);
diff --git a/arch/ppc64/kernel/idle_power4.S b/arch/ppc64/kernel/idle_power4.S
index 97e4a2655040..ca02afe2a795 100644
--- a/arch/ppc64/kernel/idle_power4.S
+++ b/arch/ppc64/kernel/idle_power4.S
@@ -20,7 +20,7 @@
 #include <asm/cputable.h>
 #include <asm/thread_info.h>
 #include <asm/ppc_asm.h>
-#include <asm/offsets.h>
+#include <asm/asm-offsets.h>
 #undef DEBUG
diff --git a/arch/ppc64/kernel/iomap.c b/arch/ppc64/kernel/iomap.c
index 153cc8b0f136..6160c8dbb7c5 100644
--- a/arch/ppc64/kernel/iomap.c
+++ b/arch/ppc64/kernel/iomap.c
@@ -22,13 +22,23 @@ unsigned int fastcall ioread16(void __iomem *addr)
 {
        return readw(addr);
 }
+unsigned int fastcall ioread16be(void __iomem *addr)
+{
+        return in_be16(addr);
+}
 unsigned int fastcall ioread32(void __iomem *addr)
 {
        return readl(addr);
 }
+unsigned int fastcall ioread32be(void __iomem *addr)
+{
+        return in_be32(addr);
+}
 EXPORT_SYMBOL(ioread8);
 EXPORT_SYMBOL(ioread16);
+EXPORT_SYMBOL(ioread16be);
 EXPORT_SYMBOL(ioread32);
+EXPORT_SYMBOL(ioread32be);
 void fastcall iowrite8(u8 val, void __iomem *addr)
 {
@@ -38,13 +48,23 @@ void fastcall iowrite16(u16 val, void __iomem *addr)
 {
        writew(val, addr);
 }
+void fastcall iowrite16be(u16 val, void __iomem *addr)
+{
+        out_be16(addr, val);
+}
 void fastcall iowrite32(u32 val, void __iomem *addr)
 {
        writel(val, addr);
 }
+void fastcall iowrite32be(u32 val, void __iomem *addr)
+{
+        out_be32(addr, val);
+}
 EXPORT_SYMBOL(iowrite8);
 EXPORT_SYMBOL(iowrite16);
+EXPORT_SYMBOL(iowrite16be);
 EXPORT_SYMBOL(iowrite32);
+EXPORT_SYMBOL(iowrite32be);
 /*
 * These are the "repeat read/write" functions. Note the
@@ -56,15 +76,15 @@ EXPORT_SYMBOL(iowrite32);
 */
 void ioread8_rep(void __iomem *addr, void *dst, unsigned long count)
 {
-        _insb((u8 __force *) addr, dst, count);
+        _insb((u8 __iomem *) addr, dst, count);
 }
 void ioread16_rep(void __iomem *addr, void *dst, unsigned long count)
 {
-        _insw_ns((u16 __force *) addr, dst, count);
+        _insw_ns((u16 __iomem *) addr, dst, count);
 }
 void ioread32_rep(void __iomem *addr, void *dst, unsigned long count)
 {
-        _insl_ns((u32 __force *) addr, dst, count);
+        _insl_ns((u32 __iomem *) addr, dst, count);
 }
 EXPORT_SYMBOL(ioread8_rep);
 EXPORT_SYMBOL(ioread16_rep);
@@ -72,15 +92,15 @@ EXPORT_SYMBOL(ioread32_rep);
 void iowrite8_rep(void __iomem *addr, const void *src, unsigned long count)
 {
-        _outsb((u8 __force *) addr, src, count);
+        _outsb((u8 __iomem *) addr, src, count);
 }
 void iowrite16_rep(void __iomem *addr, const void *src, unsigned long count)
 {
-        _outsw_ns((u16 __force *) addr, src, count);
+        _outsw_ns((u16 __iomem *) addr, src, count);
 }
 void iowrite32_rep(void __iomem *addr, const void *src, unsigned long count)
 {
-        _outsl_ns((u32 __force *) addr, src, count);
+        _outsl_ns((u32 __iomem *) addr, src, count);
 }
 EXPORT_SYMBOL(iowrite8_rep);
 EXPORT_SYMBOL(iowrite16_rep);
diff --git a/arch/ppc64/kernel/iommu.c b/arch/ppc64/kernel/iommu.c
index 845eebd1e28d..9032b6bfe036 100644
--- a/arch/ppc64/kernel/iommu.c
+++ b/arch/ppc64/kernel/iommu.c
@@ -438,7 +438,8 @@ struct iommu_table *iommu_init_table(struct iommu_table *tbl)
 void iommu_free_table(struct device_node *dn)
 {
-        struct iommu_table *tbl = dn->iommu_table;
+        struct pci_dn *pdn = dn->data;
+        struct iommu_table *tbl = pdn->iommu_table;
        unsigned long bitmap_sz, i;
        unsigned int order;
diff --git a/arch/ppc64/kernel/kprobes.c b/arch/ppc64/kernel/kprobes.c
index a3d519518fb8..7e80d49c589a 100644
--- a/arch/ppc64/kernel/kprobes.c
+++ b/arch/ppc64/kernel/kprobes.c
@@ -44,7 +44,7 @@ static struct kprobe *kprobe_prev;
 static unsigned long kprobe_status_prev, kprobe_saved_msr_prev;
 static struct pt_regs jprobe_saved_regs;
-int arch_prepare_kprobe(struct kprobe *p)
+int __kprobes arch_prepare_kprobe(struct kprobe *p)
 {
        int ret = 0;
        kprobe_opcode_t insn = *p->addr;
@@ -68,27 +68,27 @@ int arch_prepare_kprobe(struct kprobe *p)
        return ret;
 }
-void arch_copy_kprobe(struct kprobe *p)
+void __kprobes arch_copy_kprobe(struct kprobe *p)
 {
        memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
        p->opcode = *p->addr;
 }
-void arch_arm_kprobe(struct kprobe *p)
+void __kprobes arch_arm_kprobe(struct kprobe *p)
 {
        *p->addr = BREAKPOINT_INSTRUCTION;
        flush_icache_range((unsigned long) p->addr,
                           (unsigned long) p->addr + sizeof(kprobe_opcode_t));
 }
-void arch_disarm_kprobe(struct kprobe *p)
+void __kprobes arch_disarm_kprobe(struct kprobe *p)
 {
        *p->addr = p->opcode;
        flush_icache_range((unsigned long) p->addr,
                           (unsigned long) p->addr + sizeof(kprobe_opcode_t));
 }
-void arch_remove_kprobe(struct kprobe *p)
+void __kprobes arch_remove_kprobe(struct kprobe *p)
 {
        up(&kprobe_mutex);
        free_insn_slot(p->ainsn.insn);
@@ -102,7 +102,7 @@ static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
        regs->msr |= MSR_SE;
        /* single step inline if it is a trap variant */
-        if (IS_TW(insn) || IS_TD(insn) || IS_TWI(insn) || IS_TDI(insn))
+        if (is_trap(insn))
                regs->nip = (unsigned long)p->addr;
        else
                regs->nip = (unsigned long)p->ainsn.insn;
@@ -122,7 +122,8 @@ static inline void restore_previous_kprobe(void)
        kprobe_saved_msr = kprobe_saved_msr_prev;
 }
-void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs)
+void __kprobes arch_prepare_kretprobe(struct kretprobe *rp,
+                                      struct pt_regs *regs)
 {
        struct kretprobe_instance *ri;
@@ -151,7 +152,9 @@ static inline int kprobe_handler(struct pt_regs *regs)
                   Disarm the probe we just hit, and ignore it. */
                p = get_kprobe(addr);
                if (p) {
-                        if (kprobe_status == KPROBE_HIT_SS) {
+                        kprobe_opcode_t insn = *p->ainsn.insn;
+                        if (kprobe_status == KPROBE_HIT_SS &&
+                                        is_trap(insn)) {
                                regs->msr &= ~MSR_SE;
                                regs->msr |= kprobe_saved_msr;
                                unlock_kprobes();
@@ -191,8 +194,7 @@ static inline int kprobe_handler(struct pt_regs *regs)
                         * trap variant, it could belong to someone else
                         */
                        kprobe_opcode_t cur_insn = *addr;
-                        if (IS_TW(cur_insn) || IS_TD(cur_insn) ||
+                        if (is_trap(cur_insn))
-                                        IS_TWI(cur_insn) || IS_TDI(cur_insn))
                                goto no_kprobe;
                        /*
                         * The breakpoint instruction was removed right
@@ -244,7 +246,7 @@ void kretprobe_trampoline_holder(void)
 /*
 * Called when the probe at kretprobe trampoline is hit
 */
-int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
+int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
 {
        struct kretprobe_instance *ri = NULL;
        struct hlist_head *head;
@@ -308,7 +310,7 @@ int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
 * single-stepped a copy of the instruction.  The address of this
 * copy is p->ainsn.insn.
 */
-static void resume_execution(struct kprobe *p, struct pt_regs *regs)
+static void __kprobes resume_execution(struct kprobe *p, struct pt_regs *regs)
 {
        int ret;
        unsigned int insn = *p->ainsn.insn;
@@ -373,8 +375,8 @@ static inline int kprobe_fault_handler(struct pt_regs *regs, int trapnr)
 /*
 * Wrapper routine to for handling exceptions.
 */
-int kprobe_exceptions_notify(struct notifier_block *self, unsigned long val,
+int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
-                             void *data)
+                                       unsigned long val, void *data)
 {
        struct die_args *args = (struct die_args *)data;
        int ret = NOTIFY_DONE;
@@ -402,11 +404,11 @@ int kprobe_exceptions_notify(struct notifier_block *self, unsigned long val,
        default:
                break;
        }
-        preempt_enable();
+        preempt_enable_no_resched();
        return ret;
 }
-int setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
+int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
 {
        struct jprobe *jp = container_of(p, struct jprobe, kp);
@@ -419,16 +421,16 @@ int setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
        return 1;
 }
-void jprobe_return(void)
+void __kprobes jprobe_return(void)
 {
        asm volatile("trap" ::: "memory");
 }
-void jprobe_return_end(void)
+void __kprobes jprobe_return_end(void)
 {
 };
-int longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
+int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 {
        /*
         * FIXME - we should ideally be validating that we got here 'cos
diff --git a/arch/ppc64/kernel/lmb.c b/arch/ppc64/kernel/lmb.c
index d6c6bd03d2a4..5adaca2ddc9d 100644
--- a/arch/ppc64/kernel/lmb.c
+++ b/arch/ppc64/kernel/lmb.c
@@ -28,33 +28,28 @@ void lmb_dump_all(void)
 {
 #ifdef DEBUG
        unsigned long i;
-        struct lmb *_lmb  = &lmb;
        udbg_printf("lmb_dump_all:\n");
        udbg_printf("    memory.cnt               = 0x%lx\n",
-                    _lmb->memory.cnt);
+                    lmb.memory.cnt);
        udbg_printf("    memory.size              = 0x%lx\n",
-                    _lmb->memory.size);
+                    lmb.memory.size);
-        for (i=0; i < _lmb->memory.cnt ;i++) {
+        for (i=0; i < lmb.memory.cnt ;i++) {
                udbg_printf("    memory.region[0x%x].base       = 0x%lx\n",
-                            i, _lmb->memory.region[i].base);
+                            i, lmb.memory.region[i].base);
-                udbg_printf("                 .physbase = 0x%lx\n",
-                            _lmb->memory.region[i].physbase);
                udbg_printf("                 .size     = 0x%lx\n",
-                            _lmb->memory.region[i].size);
+                            lmb.memory.region[i].size);
        }
        udbg_printf("\n    reserved.cnt   = 0x%lx\n",
-                    _lmb->reserved.cnt);
+                    lmb.reserved.cnt);
        udbg_printf("    reserved.size    = 0x%lx\n",
-                    _lmb->reserved.size);
+                    lmb.reserved.size);
-        for (i=0; i < _lmb->reserved.cnt ;i++) {
+        for (i=0; i < lmb.reserved.cnt ;i++) {
                udbg_printf("    reserved.region[0x%x].base       = 0x%lx\n",
-                            i, _lmb->reserved.region[i].base);
+                            i, lmb.reserved.region[i].base);
-                udbg_printf("                 .physbase = 0x%lx\n",
-                            _lmb->reserved.region[i].physbase);
                udbg_printf("                 .size     = 0x%lx\n",
-                            _lmb->reserved.region[i].size);
+                            lmb.reserved.region[i].size);
        }
 #endif /* DEBUG */
 }
@@ -98,7 +93,6 @@ lmb_coalesce_regions(struct lmb_region *rgn, unsigned long r1, unsigned long r2)
        rgn->region[r1].size += rgn->region[r2].size;
        for (i=r2; i < rgn->cnt-1; i++) {
                rgn->region[i].base = rgn->region[i+1].base;
-                rgn->region[i].physbase = rgn->region[i+1].physbase;
                rgn->region[i].size = rgn->region[i+1].size;
        }
        rgn->cnt--;
@@ -108,49 +102,29 @@ lmb_coalesce_regions(struct lmb_region *rgn, unsigned long r1, unsigned long r2)
 void __init
 lmb_init(void)
 {
-        struct lmb *_lmb = &lmb;
        /* Create a dummy zero size LMB which will get coalesced away later.
         * This simplifies the lmb_add() code below...
         */
-        _lmb->memory.region[0].base = 0;
+        lmb.memory.region[0].base = 0;
-        _lmb->memory.region[0].size = 0;
+        lmb.memory.region[0].size = 0;
-        _lmb->memory.cnt = 1;
+        lmb.memory.cnt = 1;
        /* Ditto. */
-        _lmb->reserved.region[0].base = 0;
+        lmb.reserved.region[0].base = 0;
-        _lmb->reserved.region[0].size = 0;
+        lmb.reserved.region[0].size = 0;
-        _lmb->reserved.cnt = 1;
+        lmb.reserved.cnt = 1;
 }
 /* This routine called with relocation disabled. */
 void __init
 lmb_analyze(void)
 {
-        unsigned long i;
+        int i;
-        unsigned long mem_size = 0;
-        unsigned long size_mask = 0;
+        lmb.memory.size = 0;
-        struct lmb *_lmb = &lmb;
-#ifdef CONFIG_MSCHUNKS
-        unsigned long physbase = 0;
-#endif
-        for (i=0; i < _lmb->memory.cnt; i++) {
-                unsigned long lmb_size;
-                lmb_size = _lmb->memory.region[i].size;
-#ifdef CONFIG_MSCHUNKS
-                _lmb->memory.region[i].physbase = physbase;
-                physbase += lmb_size;
-#else
-                _lmb->memory.region[i].physbase = _lmb->memory.region[i].base;
-#endif
-                mem_size += lmb_size;
-                size_mask |= lmb_size;
-        }
-        _lmb->memory.size = mem_size;
+        for (i = 0; i < lmb.memory.cnt; i++)
+                lmb.memory.size += lmb.memory.region[i].size;
 }
 /* This routine called with relocation disabled. */
@@ -168,7 +142,6 @@ lmb_add_region(struct lmb_region *rgn, unsigned long base, unsigned long size)
                adjacent = lmb_addrs_adjacent(base,size,rgnbase,rgnsize);
                if ( adjacent > 0 ) {
                        rgn->region[i].base -= size;
-                        rgn->region[i].physbase -= size;
                        rgn->region[i].size += size;
                        coalesced++;
                        break;
@@ -195,11 +168,9 @@ lmb_add_region(struct lmb_region *rgn, unsigned long base, unsigned long size)
        for (i=rgn->cnt-1; i >= 0; i--) {
                if (base < rgn->region[i].base) {
                        rgn->region[i+1].base = rgn->region[i].base;
-                        rgn->region[i+1].physbase = rgn->region[i].physbase;
                        rgn->region[i+1].size = rgn->region[i].size;
                }  else {
                        rgn->region[i+1].base = base;
-                        rgn->region[i+1].physbase = lmb_abs_to_phys(base);
                        rgn->region[i+1].size = size;
                        break;
                }
@@ -213,12 +184,11 @@ lmb_add_region(struct lmb_region *rgn, unsigned long base, unsigned long size)
 long __init
 lmb_add(unsigned long base, unsigned long size)
 {
-        struct lmb *_lmb = &lmb;
+        struct lmb_region *_rgn = &(lmb.memory);
-        struct lmb_region *_rgn = &(_lmb->memory);
        /* On pSeries LPAR systems, the first LMB is our RMO region. */
        if ( base == 0 )
-                _lmb->rmo_size = size;
+                lmb.rmo_size = size;
        return lmb_add_region(_rgn, base, size);
@@ -227,8 +197,7 @@ lmb_add(unsigned long base, unsigned long size)
 long __init
 lmb_reserve(unsigned long base, unsigned long size)
 {
-        struct lmb *_lmb = &lmb;
+        struct lmb_region *_rgn = &(lmb.reserved);
-        struct lmb_region *_rgn = &(_lmb->reserved);
        return lmb_add_region(_rgn, base, size);
 }
@@ -260,13 +229,10 @@ lmb_alloc_base(unsigned long size, unsigned long align, unsigned long max_addr)
 {
        long i, j;
        unsigned long base = 0;
-        struct lmb *_lmb = &lmb;
-        struct lmb_region *_mem = &(_lmb->memory);
-        struct lmb_region *_rsv = &(_lmb->reserved);
-        for (i=_mem->cnt-1; i >= 0; i--) {
+        for (i=lmb.memory.cnt-1; i >= 0; i--) {
-                unsigned long lmbbase = _mem->region[i].base;
+                unsigned long lmbbase = lmb.memory.region[i].base;
-                unsigned long lmbsize = _mem->region[i].size;
+                unsigned long lmbsize = lmb.memory.region[i].size;
                if ( max_addr == LMB_ALLOC_ANYWHERE )
                        base = _ALIGN_DOWN(lmbbase+lmbsize-size, align);
@@ -276,8 +242,8 @@ lmb_alloc_base(unsigned long size, unsigned long align, unsigned long max_addr)
                        continue;
                while ( (lmbbase <= base) &&
-                        ((j = lmb_overlaps_region(_rsv,base,size)) >= 0) ) {
+                        ((j = lmb_overlaps_region(&lmb.reserved,base,size)) >= 0) ) {
-                        base = _ALIGN_DOWN(_rsv->region[j].base-size, align);
+                        base = _ALIGN_DOWN(lmb.reserved.region[j].base-size, align);
                }
                if ( (base != 0) && (lmbbase <= base) )
@@ -287,62 +253,24 @@ lmb_alloc_base(unsigned long size, unsigned long align, unsigned long max_addr)
        if ( i < 0 )
                return 0;
-        lmb_add_region(_rsv, base, size);
+        lmb_add_region(&lmb.reserved, base, size);
        return base;
 }
+/* You must call lmb_analyze() before this. */
 unsigned long __init
 lmb_phys_mem_size(void)
 {
-        struct lmb *_lmb = &lmb;
+        return lmb.memory.size;
-#ifdef CONFIG_MSCHUNKS
-        return _lmb->memory.size;
-#else
-        struct lmb_region *_mem = &(_lmb->memory);
-        unsigned long total = 0;
-        int i;
-        /* add all physical memory to the bootmem map */
-        for (i=0; i < _mem->cnt; i++)
-                total += _mem->region[i].size;
-        return total;
-#endif /* CONFIG_MSCHUNKS */
 }
 unsigned long __init
 lmb_end_of_DRAM(void)
 {
-        struct lmb *_lmb = &lmb;
+        int idx = lmb.memory.cnt - 1;
-        struct lmb_region *_mem = &(_lmb->memory);
-        int idx = _mem->cnt - 1;
-#ifdef CONFIG_MSCHUNKS
-        return (_mem->region[idx].physbase + _mem->region[idx].size);
-#else
-        return (_mem->region[idx].base + _mem->region[idx].size);
-#endif /* CONFIG_MSCHUNKS */
-        return 0;
-}
-unsigned long __init
-lmb_abs_to_phys(unsigned long aa)
-{
-        unsigned long i, pa = aa;
-        struct lmb *_lmb = &lmb;
-        struct lmb_region *_mem = &(_lmb->memory);
-        for (i=0; i < _mem->cnt; i++) {
-                unsigned long lmbbase = _mem->region[i].base;
-                unsigned long lmbsize = _mem->region[i].size;
-                if ( lmb_addrs_overlap(aa,1,lmbbase,lmbsize) ) {
-                        pa = _mem->region[i].physbase + (aa - lmbbase);
-                        break;
-                }
-        }
-        return pa;
+        return (lmb.memory.region[idx].base + lmb.memory.region[idx].size);
 }
 /*
@@ -353,20 +281,19 @@ void __init lmb_enforce_memory_limit(void)
 {
        extern unsigned long memory_limit;
        unsigned long i, limit;
-        struct lmb_region *mem = &(lmb.memory);
        if (! memory_limit)
                return;
        limit = memory_limit;
-        for (i = 0; i < mem->cnt; i++) {
+        for (i = 0; i < lmb.memory.cnt; i++) {
-                if (limit > mem->region[i].size) {
+                if (limit > lmb.memory.region[i].size) {
-                        limit -= mem->region[i].size;
+                        limit -= lmb.memory.region[i].size;
                        continue;
                }
-                mem->region[i].size = limit;
+                lmb.memory.region[i].size = limit;
-                mem->cnt = i + 1;
+                lmb.memory.cnt = i + 1;
                break;
        }
 }
diff --git a/arch/ppc64/kernel/lparcfg.c b/arch/ppc64/kernel/lparcfg.c
index 02e96627fa66..cae19bbd5acd 100644
--- a/arch/ppc64/kernel/lparcfg.c
+++ b/arch/ppc64/kernel/lparcfg.c
@@ -29,7 +29,7 @@
 #include <asm/iSeries/HvLpConfig.h>
 #include <asm/lppaca.h>
 #include <asm/hvcall.h>
-#include <asm/cputable.h>
+#include <asm/firmware.h>
 #include <asm/rtas.h>
 #include <asm/system.h>
 #include <asm/time.h>
@@ -273,6 +273,7 @@ static void parse_system_parameter_string(struct seq_file *m)
                if (!workbuffer) {
                        printk(KERN_ERR "%s %s kmalloc failure at line %d \n",
                               __FILE__, __FUNCTION__, __LINE__);
+                        kfree(local_buffer);                    
                        return;
                }
 #ifdef LPARCFG_DEBUG
@@ -377,7 +378,7 @@ static int lparcfg_data(struct seq_file *m, void *v)
        partition_active_processors = lparcfg_count_active_processors();
-        if (cur_cpu_spec->firmware_features & FW_FEATURE_SPLPAR) {
+        if (firmware_has_feature(FW_FEATURE_SPLPAR)) {
                unsigned long h_entitled, h_unallocated;
                unsigned long h_aggregation, h_resource;
                unsigned long pool_idle_time, pool_procs;
@@ -568,10 +569,10 @@ struct file_operations lparcfg_fops = {
 int __init lparcfg_init(void)
 {
        struct proc_dir_entry *ent;
-        mode_t mode = S_IRUSR;
+        mode_t mode = S_IRUSR | S_IRGRP | S_IROTH;
        /* Allow writing if we have FW_FEATURE_SPLPAR */
-        if (cur_cpu_spec->firmware_features & FW_FEATURE_SPLPAR) {
+        if (firmware_has_feature(FW_FEATURE_SPLPAR)) {
                lparcfg_fops.write = lparcfg_write;
                mode |= S_IWUSR;
        }
diff --git a/arch/ppc64/kernel/maple_pci.c b/arch/ppc64/kernel/maple_pci.c
index 53993999b265..1d297e0edfc0 100644
--- a/arch/ppc64/kernel/maple_pci.c
+++ b/arch/ppc64/kernel/maple_pci.c
@@ -283,7 +283,7 @@ static void __init setup_u3_agp(struct pci_controller* hose)
         * the reg address cell, we shall fix that by killing struct
         * reg_property and using some accessor functions instead
         */
-        hose->first_busno = 0xf0;
+        hose->first_busno = 0xf0;
        hose->last_busno = 0xff;
        hose->ops = &u3_agp_pci_ops;
        hose->cfg_addr = ioremap(0xf0000000 + 0x800000, 0x1000);
@@ -315,24 +315,24 @@ static int __init add_bridge(struct device_node *dev)
        char* disp_name;
        int *bus_range;
        int primary = 1;
-        struct property *of_prop;
+        struct property *of_prop;
        DBG("Adding PCI host bridge %s\n", dev->full_name);
-        bus_range = (int *) get_property(dev, "bus-range", &len);
+        bus_range = (int *) get_property(dev, "bus-range", &len);
-        if (bus_range == NULL || len < 2 * sizeof(int)) {
+        if (bus_range == NULL || len < 2 * sizeof(int)) {
-                printk(KERN_WARNING "Can't get bus-range for %s, assume bus 0\n",
+                printk(KERN_WARNING "Can't get bus-range for %s, assume bus 0\n",
-                               dev->full_name);
+                dev->full_name);
-        }
+        }
        hose = alloc_bootmem(sizeof(struct pci_controller));
        if (hose == NULL)
                return -ENOMEM;
-        pci_setup_pci_controller(hose);
+        pci_setup_pci_controller(hose);
-        hose->arch_data = dev;
+        hose->arch_data = dev;
-        hose->first_busno = bus_range ? bus_range[0] : 0;
+        hose->first_busno = bus_range ? bus_range[0] : 0;
-        hose->last_busno = bus_range ? bus_range[1] : 0xff;
+        hose->last_busno = bus_range ? bus_range[1] : 0xff;
        of_prop = alloc_bootmem(sizeof(struct property) +
                                sizeof(hose->global_number));
@@ -346,25 +346,25 @@ static int __init add_bridge(struct device_node *dev)
        }
        disp_name = NULL;
-        if (device_is_compatible(dev, "u3-agp")) {
+        if (device_is_compatible(dev, "u3-agp")) {
-                setup_u3_agp(hose);
+                setup_u3_agp(hose);
-                disp_name = "U3-AGP";
+                disp_name = "U3-AGP";
-                primary = 0;
+                primary = 0;
-        } else if (device_is_compatible(dev, "u3-ht")) {
+        } else if (device_is_compatible(dev, "u3-ht")) {
-                setup_u3_ht(hose);
+                setup_u3_ht(hose);
-                disp_name = "U3-HT";
+                disp_name = "U3-HT";
-                primary = 1;
+                primary = 1;
-        }
+        }
-        printk(KERN_INFO "Found %s PCI host bridge. Firmware bus number: %d->%d\n",
+        printk(KERN_INFO "Found %s PCI host bridge. Firmware bus number: %d->%d\n",
-                disp_name, hose->first_busno, hose->last_busno);
+                disp_name, hose->first_busno, hose->last_busno);
-        /* Interpret the "ranges" property */
+        /* Interpret the "ranges" property */
-        /* This also maps the I/O region and sets isa_io/mem_base */
+        /* This also maps the I/O region and sets isa_io/mem_base */
-        pci_process_bridge_OF_ranges(hose, dev);
+        pci_process_bridge_OF_ranges(hose, dev);
        pci_setup_phb_io(hose, primary);
-        /* Fixup "bus-range" OF property */
+        /* Fixup "bus-range" OF property */
-        fixup_bus_range(dev);
+        fixup_bus_range(dev);
        return 0;
 }
@@ -447,9 +447,9 @@ void __init maple_pci_init(void)
         */
        if (u3_agp) {
                struct device_node *np = u3_agp->arch_data;
-                np->busno = 0xf0;
+                PCI_DN(np)->busno = 0xf0;
                for (np = np->child; np; np = np->sibling)
-                        np->busno = 0xf0;
+                        PCI_DN(np)->busno = 0xf0;
        }
        /* Tell pci.c to use the common resource allocation mecanism */
diff --git a/arch/ppc64/kernel/maple_setup.c b/arch/ppc64/kernel/maple_setup.c
index bb55b5a56910..fc0567498a3a 100644
--- a/arch/ppc64/kernel/maple_setup.c
+++ b/arch/ppc64/kernel/maple_setup.c
@@ -207,9 +207,6 @@ static void __init maple_init_early(void)
                comport = (void *)ioremap(physport, 16);
                udbg_init_uart(comport, default_speed);
-                ppc_md.udbg_putc = udbg_putc;
-                ppc_md.udbg_getc = udbg_getc;
-                ppc_md.udbg_getc_poll = udbg_getc_poll;
                DBG("Hello World !\n");
        }
diff --git a/arch/ppc64/kernel/misc.S b/arch/ppc64/kernel/misc.S
index a05b50b738e9..e7241ad80a08 100644
--- a/arch/ppc64/kernel/misc.S
+++ b/arch/ppc64/kernel/misc.S
@@ -26,7 +26,7 @@
 #include <asm/page.h>
 #include <asm/cache.h>
 #include <asm/ppc_asm.h>
-#include <asm/offsets.h>
+#include <asm/asm-offsets.h>
 #include <asm/cputable.h>
        .text
@@ -183,7 +183,7 @@ PPC64_CACHES:
 *   flush all bytes from start through stop-1 inclusive
 */
-_GLOBAL(__flush_icache_range)
+_KPROBE(__flush_icache_range)
 /*
 * Flush the data cache to memory 
@@ -223,7 +223,7 @@ _GLOBAL(__flush_icache_range)
        bdnz    2b
        isync
        blr
-        
+        .previous .text
 /*
 * Like above, but only do the D-cache.
 *
@@ -680,6 +680,104 @@ _GLOBAL(kernel_thread)
        ld      r30,-16(r1)
        blr
+/*
+ * disable_kernel_fp()
+ * Disable the FPU.
+ */
+_GLOBAL(disable_kernel_fp)
+        mfmsr   r3
+        rldicl  r0,r3,(63-MSR_FP_LG),1
+        rldicl  r3,r0,(MSR_FP_LG+1),0
+        mtmsrd  r3                      /* disable use of fpu now */
+        isync
+        blr
+/*
+ * giveup_fpu(tsk)
+ * Disable FP for the task given as the argument,
+ * and save the floating-point registers in its thread_struct.
+ * Enables the FPU for use in the kernel on return.
+ */
+_GLOBAL(giveup_fpu)
+        mfmsr   r5
+        ori     r5,r5,MSR_FP
+        mtmsrd  r5                      /* enable use of fpu now */
+        isync
+        cmpdi   0,r3,0
+        beqlr-                          /* if no previous owner, done */
+        addi    r3,r3,THREAD            /* want THREAD of task */
+        ld      r5,PT_REGS(r3)
+        cmpdi   0,r5,0
+        SAVE_32FPRS(0, r3)
+        mffs    fr0
+        stfd    fr0,THREAD_FPSCR(r3)
+        beq     1f
+        ld      r4,_MSR-STACK_FRAME_OVERHEAD(r5)
+        li      r3,MSR_FP|MSR_FE0|MSR_FE1
+        andc    r4,r4,r3                /* disable FP for previous task */
+        std     r4,_MSR-STACK_FRAME_OVERHEAD(r5)
+1:
+#ifndef CONFIG_SMP
+        li      r5,0
+        ld      r4,last_task_used_math@got(r2)
+        std     r5,0(r4)
+#endif /* CONFIG_SMP */
+        blr
+#ifdef CONFIG_ALTIVEC
+#if 0 /* this has no callers for now */
+/*
+ * disable_kernel_altivec()
+ * Disable the VMX.
+ */
+_GLOBAL(disable_kernel_altivec)
+        mfmsr   r3
+        rldicl  r0,r3,(63-MSR_VEC_LG),1
+        rldicl  r3,r0,(MSR_VEC_LG+1),0
+        mtmsrd  r3                      /* disable use of VMX now */
+        isync
+        blr
+#endif /* 0 */
+/*
+ * giveup_altivec(tsk)
+ * Disable VMX for the task given as the argument,
+ * and save the vector registers in its thread_struct.
+ * Enables the VMX for use in the kernel on return.
+ */
+_GLOBAL(giveup_altivec)
+        mfmsr   r5
+        oris    r5,r5,MSR_VEC@h
+        mtmsrd  r5                      /* enable use of VMX now */
+        isync
+        cmpdi   0,r3,0
+        beqlr-                          /* if no previous owner, done */
+        addi    r3,r3,THREAD            /* want THREAD of task */
+        ld      r5,PT_REGS(r3)
+        cmpdi   0,r5,0
+        SAVE_32VRS(0,r4,r3)
+        mfvscr  vr0
+        li      r4,THREAD_VSCR
+        stvx    vr0,r4,r3
+        beq     1f
+        ld      r4,_MSR-STACK_FRAME_OVERHEAD(r5)
+        lis     r3,MSR_VEC@h
+        andc    r4,r4,r3                /* disable FP for previous task */
+        std     r4,_MSR-STACK_FRAME_OVERHEAD(r5)
+1:
+#ifndef CONFIG_SMP
+        li      r5,0
+        ld      r4,last_task_used_altivec@got(r2)
+        std     r5,0(r4)
+#endif /* CONFIG_SMP */
+        blr
+#endif /* CONFIG_ALTIVEC */
+_GLOBAL(__setup_cpu_power3)
+        blr
 /* kexec_wait(phys_cpu)
 *
 * wait for the flag to change, indicating this kernel is going away but
@@ -859,7 +957,7 @@ _GLOBAL(sys_call_table32)
        .llong .ppc_fork
        .llong .sys_read
        .llong .sys_write
-        .llong .sys32_open              /* 5 */
+        .llong .compat_sys_open         /* 5 */
        .llong .sys_close
        .llong .sys32_waitpid
        .llong .sys32_creat
@@ -1333,9 +1431,9 @@ _GLOBAL(sys_call_table)
        .llong .sys_ni_syscall          /* 195 - 32bit only stat64 */
        .llong .sys_ni_syscall          /* 32bit only lstat64 */
        .llong .sys_ni_syscall          /* 32bit only fstat64 */
-        .llong .sys_ni_syscall          /* 32bit only pciconfig_read */
+        .llong .sys_pciconfig_read
-        .llong .sys_ni_syscall          /* 32bit only pciconfig_write */
+        .llong .sys_pciconfig_write
-        .llong .sys_ni_syscall          /* 32bit only pciconfig_iobase */
+        .llong .sys_pciconfig_iobase    /* 200 - pciconfig_iobase */
        .llong .sys_ni_syscall          /* reserved for MacOnLinux */
        .llong .sys_getdents64
        .llong .sys_pivot_root
diff --git a/arch/ppc64/kernel/of_device.c b/arch/ppc64/kernel/of_device.c
index b80e81984ba8..da580812ddfe 100644
--- a/arch/ppc64/kernel/of_device.c
+++ b/arch/ppc64/kernel/of_device.c
@@ -236,7 +236,6 @@ void of_device_unregister(struct of_device *ofdev)
 struct of_device* of_platform_device_create(struct device_node *np, const char *bus_id)
 {
        struct of_device *dev;
-        u32 *reg;
        dev = kmalloc(sizeof(*dev), GFP_KERNEL);
        if (!dev)
@@ -250,7 +249,6 @@ struct of_device* of_platform_device_create(struct device_node *np, const char *
        dev->dev.bus = &of_platform_bus_type;
        dev->dev.release = of_release_dev;
-        reg = (u32 *)get_property(np, "reg", NULL);
        strlcpy(dev->dev.bus_id, bus_id, BUS_ID_SIZE);
        if (of_device_register(dev) != 0) {
diff --git a/arch/ppc64/kernel/pSeries_iommu.c b/arch/ppc64/kernel/pSeries_iommu.c
index 69130522a87e..f0fd7fbd6531 100644
--- a/arch/ppc64/kernel/pSeries_iommu.c
+++ b/arch/ppc64/kernel/pSeries_iommu.c
@@ -45,6 +45,7 @@
 #include <asm/plpar_wrappers.h>
 #include <asm/pSeries_reconfig.h>
 #include <asm/systemcfg.h>
+#include <asm/firmware.h>
 #include "pci.h"
 #define DBG(fmt...)
@@ -294,7 +295,7 @@ static void iommu_table_setparms_lpar(struct pci_controller *phb,
                                      struct iommu_table *tbl,
                                      unsigned int *dma_window)
 {
-        tbl->it_busno  = dn->bussubno;
+        tbl->it_busno  = PCI_DN(dn)->bussubno;
        /* TODO: Parse field size properties properly. */
        tbl->it_size   = (((unsigned long)dma_window[4] << 32) |
@@ -310,6 +311,7 @@ static void iommu_table_setparms_lpar(struct pci_controller *phb,
 static void iommu_bus_setup_pSeries(struct pci_bus *bus)
 {
        struct device_node *dn, *pdn;
+        struct pci_dn *pci;
        struct iommu_table *tbl;
        DBG("iommu_bus_setup_pSeries, bus %p, bus->self %p\n", bus, bus->self);
@@ -324,6 +326,7 @@ static void iommu_bus_setup_pSeries(struct pci_bus *bus)
         */
        dn = pci_bus_to_OF_node(bus);
+        pci = dn->data;
        if (!bus->self) {
                /* Root bus */
@@ -340,18 +343,18 @@ static void iommu_bus_setup_pSeries(struct pci_bus *bus)
                                 * alltogether. This leaves 768MB for the window.
                                 */
                                DBG("PHB has io-hole, reserving 256MB\n");
-                                dn->phb->dma_window_size = 3 << 28;
+                                pci->phb->dma_window_size = 3 << 28;
-                                dn->phb->dma_window_base_cur = 1 << 28;
+                                pci->phb->dma_window_base_cur = 1 << 28;
                        } else {
                                /* 1GB window by default */
-                                dn->phb->dma_window_size = 1 << 30;
+                                pci->phb->dma_window_size = 1 << 30;
-                                dn->phb->dma_window_base_cur = 0;
+                                pci->phb->dma_window_base_cur = 0;
                        }
                        tbl = kmalloc(sizeof(struct iommu_table), GFP_KERNEL);
-                        iommu_table_setparms(dn->phb, dn, tbl);
+                        iommu_table_setparms(pci->phb, dn, tbl);
-                        dn->iommu_table = iommu_init_table(tbl);
+                        pci->iommu_table = iommu_init_table(tbl);
                } else {
                        /* Do a 128MB table at root. This is used for the IDE
                         * controller on some SMP-mode POWER4 machines. It
@@ -362,16 +365,16 @@ static void iommu_bus_setup_pSeries(struct pci_bus *bus)
                         * Allocate at offset 128MB to avoid having to deal
                         * with ISA holes; 128MB table for IDE is plenty.
                         */
-                        dn->phb->dma_window_size = 1 << 27;
+                        pci->phb->dma_window_size = 1 << 27;
-                        dn->phb->dma_window_base_cur = 1 << 27;
+                        pci->phb->dma_window_base_cur = 1 << 27;
                        tbl = kmalloc(sizeof(struct iommu_table), GFP_KERNEL);
-                        iommu_table_setparms(dn->phb, dn, tbl);
+                        iommu_table_setparms(pci->phb, dn, tbl);
-                        dn->iommu_table = iommu_init_table(tbl);
+                        pci->iommu_table = iommu_init_table(tbl);
                        /* All child buses have 256MB tables */
-                        dn->phb->dma_window_size = 1 << 28;
+                        pci->phb->dma_window_size = 1 << 28;
                }
        } else {
                pdn = pci_bus_to_OF_node(bus->parent);
@@ -385,12 +388,12 @@ static void iommu_bus_setup_pSeries(struct pci_bus *bus)
                        tbl = kmalloc(sizeof(struct iommu_table), GFP_KERNEL);
-                        iommu_table_setparms(dn->phb, dn, tbl);
+                        iommu_table_setparms(pci->phb, dn, tbl);
-                        dn->iommu_table = iommu_init_table(tbl);
+                        pci->iommu_table = iommu_init_table(tbl);
                } else {
                        /* Lower than first child or under python, use parent table */
-                        dn->iommu_table = pdn->iommu_table;
+                        pci->iommu_table = PCI_DN(pdn)->iommu_table;
                }
        }
 }
@@ -400,6 +403,7 @@ static void iommu_bus_setup_pSeriesLP(struct pci_bus *bus)
 {
        struct iommu_table *tbl;
        struct device_node *dn, *pdn;
+        struct pci_dn *ppci;
        unsigned int *dma_window = NULL;
        DBG("iommu_bus_setup_pSeriesLP, bus %p, bus->self %p\n", bus, bus->self);
@@ -418,22 +422,24 @@ static void iommu_bus_setup_pSeriesLP(struct pci_bus *bus)
                return;
        }
-        if (!pdn->iommu_table) {
+        ppci = pdn->data;
+        if (!ppci->iommu_table) {
                /* Bussubno hasn't been copied yet.
                 * Do it now because iommu_table_setparms_lpar needs it.
                 */
-                pdn->bussubno = bus->number;
+                ppci->bussubno = bus->number;
                tbl = (struct iommu_table *)kmalloc(sizeof(struct iommu_table),
                                                    GFP_KERNEL);
        
-                iommu_table_setparms_lpar(pdn->phb, pdn, tbl, dma_window);
+                iommu_table_setparms_lpar(ppci->phb, pdn, tbl, dma_window);
-                pdn->iommu_table = iommu_init_table(tbl);
+                ppci->iommu_table = iommu_init_table(tbl);
        }
        if (pdn != dn)
-                dn->iommu_table = pdn->iommu_table;
+                PCI_DN(dn)->iommu_table = ppci->iommu_table;
 }
@@ -448,11 +454,11 @@ static void iommu_dev_setup_pSeries(struct pci_dev *dev)
         */
        mydn = dn = pci_device_to_OF_node(dev);
-        while (dn && dn->iommu_table == NULL)
+        while (dn && dn->data && PCI_DN(dn)->iommu_table == NULL)
                dn = dn->parent;
-        if (dn) {
+        if (dn && dn->data) {
-                mydn->iommu_table = dn->iommu_table;
+                PCI_DN(mydn)->iommu_table = PCI_DN(dn)->iommu_table;
        } else {
                DBG("iommu_dev_setup_pSeries, dev %p (%s) has no iommu table\n", dev, dev->pretty_name);
        }
@@ -462,10 +468,11 @@ static int iommu_reconfig_notifier(struct notifier_block *nb, unsigned long acti
 {
        int err = NOTIFY_OK;
        struct device_node *np = node;
+        struct pci_dn *pci = np->data;
        switch (action) {
        case PSERIES_RECONFIG_REMOVE:
-                if (np->iommu_table &&
+                if (pci->iommu_table &&
                    get_property(np, "ibm,dma-window", NULL))
                        iommu_free_table(np);
                break;
@@ -485,6 +492,7 @@ static void iommu_dev_setup_pSeriesLP(struct pci_dev *dev)
        struct device_node *pdn, *dn;
        struct iommu_table *tbl;
        int *dma_window = NULL;
+        struct pci_dn *pci;
        DBG("iommu_dev_setup_pSeriesLP, dev %p (%s)\n", dev, dev->pretty_name);
@@ -496,8 +504,10 @@ static void iommu_dev_setup_pSeriesLP(struct pci_dev *dev)
         */
        dn = pci_device_to_OF_node(dev);
-        for (pdn = dn; pdn && !pdn->iommu_table; pdn = pdn->parent) {
+        for (pdn = dn; pdn && pdn->data && !PCI_DN(pdn)->iommu_table;
-                dma_window = (unsigned int *)get_property(pdn, "ibm,dma-window", NULL);
+             pdn = pdn->parent) {
+                dma_window = (unsigned int *)
+                        get_property(pdn, "ibm,dma-window", NULL);
                if (dma_window)
                        break;
        }
@@ -514,20 +524,21 @@ static void iommu_dev_setup_pSeriesLP(struct pci_dev *dev)
                DBG("Found DMA window, allocating table\n");
        }
-        if (!pdn->iommu_table) {
+        pci = pdn->data;
+        if (!pci->iommu_table) {
                /* iommu_table_setparms_lpar needs bussubno. */
-                pdn->bussubno = pdn->phb->bus->number;
+                pci->bussubno = pci->phb->bus->number;
                tbl = (struct iommu_table *)kmalloc(sizeof(struct iommu_table),
                                                    GFP_KERNEL);
-                iommu_table_setparms_lpar(pdn->phb, pdn, tbl, dma_window);
+                iommu_table_setparms_lpar(pci->phb, pdn, tbl, dma_window);
-                pdn->iommu_table = iommu_init_table(tbl);
+                pci->iommu_table = iommu_init_table(tbl);
        }
        if (pdn != dn)
-                dn->iommu_table = pdn->iommu_table;
+                PCI_DN(dn)->iommu_table = pci->iommu_table;
 }
 static void iommu_bus_setup_null(struct pci_bus *b) { }
@@ -546,7 +557,7 @@ void iommu_init_early_pSeries(void)
        }
        if (systemcfg->platform & PLATFORM_LPAR) {
-                if (cur_cpu_spec->firmware_features & FW_FEATURE_MULTITCE) {
+                if (firmware_has_feature(FW_FEATURE_MULTITCE)) {
                        ppc_md.tce_build = tce_buildmulti_pSeriesLP;
                        ppc_md.tce_free  = tce_freemulti_pSeriesLP;
                } else {
diff --git a/arch/ppc64/kernel/pSeries_lpar.c b/arch/ppc64/kernel/pSeries_lpar.c
index 74dd144dcce8..a6de83f2078f 100644
--- a/arch/ppc64/kernel/pSeries_lpar.c
+++ b/arch/ppc64/kernel/pSeries_lpar.c
@@ -52,7 +52,6 @@ EXPORT_SYMBOL(plpar_hcall_4out);
 EXPORT_SYMBOL(plpar_hcall_norets);
 EXPORT_SYMBOL(plpar_hcall_8arg_2ret);
-extern void fw_feature_init(void);
 extern void pSeries_find_serial_port(void);
@@ -193,9 +192,9 @@ static unsigned char udbg_getcLP(void)
 void udbg_init_debug_lpar(void)
 {
        vtermno = 0;
-        ppc_md.udbg_putc = udbg_putcLP;
+        udbg_putc = udbg_putcLP;
-        ppc_md.udbg_getc = udbg_getcLP;
+        udbg_getc = udbg_getcLP;
-        ppc_md.udbg_getc_poll = udbg_getc_pollLP;
+        udbg_getc_poll = udbg_getc_pollLP;
 }
 /* returns 0 if couldn't find or use /chosen/stdout as console */
@@ -228,18 +227,18 @@ int find_udbg_vterm(void)
                        termno = (u32 *)get_property(stdout_node, "reg", NULL);
                        if (termno) {
                                vtermno = termno[0];
-                                ppc_md.udbg_putc = udbg_putcLP;
+                                udbg_putc = udbg_putcLP;
-                                ppc_md.udbg_getc = udbg_getcLP;
+                                udbg_getc = udbg_getcLP;
-                                ppc_md.udbg_getc_poll = udbg_getc_pollLP;
+                                udbg_getc_poll = udbg_getc_pollLP;
                                found = 1;
                        }
                } else if (device_is_compatible(stdout_node, "hvterm-protocol")) {
                        termno = (u32 *)get_property(stdout_node, "reg", NULL);
                        if (termno) {
                                vtermno = termno[0];
-                                ppc_md.udbg_putc = udbg_hvsi_putc;
+                                udbg_putc = udbg_hvsi_putc;
-                                ppc_md.udbg_getc = udbg_hvsi_getc;
+                                udbg_getc = udbg_hvsi_getc;
-                                ppc_md.udbg_getc_poll = udbg_hvsi_getc_poll;
+                                udbg_getc_poll = udbg_hvsi_getc_poll;
                                found = 1;
                        }
                }
@@ -267,6 +266,10 @@ void vpa_init(int cpu)
        /* Register the Virtual Processor Area (VPA) */
        flags = 1UL << (63 - 18);
+        if (cpu_has_feature(CPU_FTR_ALTIVEC))
+                paca[cpu].lppaca.vmxregs_in_use = 1;
        ret = register_vpa(flags, hwcpu, __pa(vpa));
        if (ret)
@@ -279,7 +282,6 @@ long pSeries_lpar_hpte_insert(unsigned long hpte_group,
                              unsigned long va, unsigned long prpn,
                              unsigned long vflags, unsigned long rflags)
 {
-        unsigned long arpn = physRpn_to_absRpn(prpn);
        unsigned long lpar_rc;
        unsigned long flags;
        unsigned long slot;
@@ -290,7 +292,7 @@ long pSeries_lpar_hpte_insert(unsigned long hpte_group,
        if (vflags & HPTE_V_LARGE)
                hpte_v &= ~(1UL << HPTE_V_AVPN_SHIFT);
-        hpte_r = (arpn << HPTE_R_RPN_SHIFT) | rflags;
+        hpte_r = (prpn << HPTE_R_RPN_SHIFT) | rflags;
        /* Now fill in the actual HPTE */
        /* Set CEC cookie to 0         */
diff --git a/arch/ppc64/kernel/pSeries_reconfig.c b/arch/ppc64/kernel/pSeries_reconfig.c
index dc2a69d412a2..58c61219d08e 100644
--- a/arch/ppc64/kernel/pSeries_reconfig.c
+++ b/arch/ppc64/kernel/pSeries_reconfig.c
@@ -111,7 +111,7 @@ static int pSeries_reconfig_add_node(const char *path, struct property *proplist
        struct device_node *np;
        int err = -ENOMEM;
-        np = kcalloc(1, sizeof(*np), GFP_KERNEL);
+        np = kzalloc(sizeof(*np), GFP_KERNEL);
        if (!np)
                goto out_err;
diff --git a/arch/ppc64/kernel/pSeries_setup.c b/arch/ppc64/kernel/pSeries_setup.c
index 5bec956e44a0..3009701eb90d 100644
--- a/arch/ppc64/kernel/pSeries_setup.c
+++ b/arch/ppc64/kernel/pSeries_setup.c
@@ -37,7 +37,7 @@
 #include <linux/ioport.h>
 #include <linux/console.h>
 #include <linux/pci.h>
-#include <linux/version.h>
+#include <linux/utsname.h>
 #include <linux/adb.h>
 #include <linux/module.h>
 #include <linux/delay.h>
@@ -60,7 +60,8 @@
 #include <asm/nvram.h>
 #include <asm/plpar_wrappers.h>
 #include <asm/xics.h>
-#include <asm/cputable.h>
+#include <asm/firmware.h>
+#include <asm/pmc.h>
 #include "i8259.h"
 #include "mpic.h"
@@ -187,6 +188,21 @@ static void __init pSeries_setup_mpic(void)
                                  " MPIC     ");
 }
+static void pseries_lpar_enable_pmcs(void)
+{
+        unsigned long set, reset;
+        power4_enable_pmcs();
+        set = 1UL << 63;
+        reset = 0;
+        plpar_hcall_norets(H_PERFMON, set, reset);
+        /* instruct hypervisor to maintain PMCs */
+        if (firmware_has_feature(FW_FEATURE_SPLPAR))
+                get_paca()->lppaca.pmcregs_in_use = 1;
+}
 static void __init pSeries_setup_arch(void)
 {
        /* Fixup ppc_md depending on the type of interrupt controller */
@@ -222,8 +238,8 @@ static void __init pSeries_setup_arch(void)
        /* Find and initialize PCI host bridges */
        init_pci_config_tokens();
-        eeh_init();
        find_and_init_phbs();
+        eeh_init();
 #ifdef CONFIG_DUMMY_CONSOLE
        conswitchp = &dummy_con;
@@ -231,11 +247,9 @@ static void __init pSeries_setup_arch(void)
        pSeries_nvram_init();
-        if (cur_cpu_spec->firmware_features & FW_FEATURE_SPLPAR)
-                vpa_init(boot_cpuid);
        /* Choose an idle loop */
-        if (cur_cpu_spec->firmware_features & FW_FEATURE_SPLPAR) {
+        if (firmware_has_feature(FW_FEATURE_SPLPAR)) {
+                vpa_init(boot_cpuid);
                if (get_paca()->lppaca.shared_proc) {
                        printk(KERN_INFO "Using shared processor idle loop\n");
                        ppc_md.idle_loop = pseries_shared_idle;
@@ -247,24 +261,29 @@ static void __init pSeries_setup_arch(void)
                printk(KERN_INFO "Using default idle loop\n");
                ppc_md.idle_loop = default_idle;
        }
+        if (systemcfg->platform & PLATFORM_LPAR)
+                ppc_md.enable_pmcs = pseries_lpar_enable_pmcs;
+        else
+                ppc_md.enable_pmcs = power4_enable_pmcs;
 }
 static int __init pSeries_init_panel(void)
 {
        /* Manually leave the kernel version on the panel. */
        ppc_md.progress("Linux ppc64\n", 0);
-        ppc_md.progress(UTS_RELEASE, 0);
+        ppc_md.progress(system_utsname.version, 0);
        return 0;
 }
 arch_initcall(pSeries_init_panel);
-/* Build up the firmware_features bitmask field
+/* Build up the ppc64_firmware_features bitmask field
 * using contents of device-tree/ibm,hypertas-functions.
 * Ultimately this functionality may be moved into prom.c prom_init().
 */
-void __init fw_feature_init(void)
+static void __init fw_feature_init(void)
 {
        struct device_node * dn;
        char * hypertas;
@@ -272,7 +291,7 @@ void __init fw_feature_init(void)
        DBG(" -> fw_feature_init()\n");
-        cur_cpu_spec->firmware_features = 0;
+        ppc64_firmware_features = 0;
        dn = of_find_node_by_path("/rtas");
        if (dn == NULL) {
                printk(KERN_ERR "WARNING ! Cannot find RTAS in device-tree !\n");
@@ -288,7 +307,7 @@ void __init fw_feature_init(void)
                                if ((firmware_features_table[i].name) &&
                                    (strcmp(firmware_features_table[i].name,hypertas))==0) {
                                        /* we have a match */
-                                        cur_cpu_spec->firmware_features |= 
+                                        ppc64_firmware_features |= 
                                                (firmware_features_table[i].val);
                                        break;
                                } 
@@ -302,7 +321,7 @@ void __init fw_feature_init(void)
        of_node_put(dn);
 no_rtas:
        printk(KERN_INFO "firmware_features = 0x%lx\n", 
-               cur_cpu_spec->firmware_features);
+               ppc64_firmware_features);
        DBG(" <- fw_feature_init()\n");
 }
@@ -378,9 +397,6 @@ static void __init pSeries_init_early(void)
                comport = (void *)ioremap(physport, 16);
                udbg_init_uart(comport, default_speed);
-                ppc_md.udbg_putc = udbg_putc;
-                ppc_md.udbg_getc = udbg_getc;
-                ppc_md.udbg_getc_poll = udbg_getc_poll;
                DBG("Hello World !\n");
        }
@@ -574,6 +590,13 @@ static int pseries_shared_idle(void)
        return 0;
 }
+static int pSeries_pci_probe_mode(struct pci_bus *bus)
+{
+        if (systemcfg->platform & PLATFORM_LPAR)
+                return PCI_PROBE_DEVTREE;
+        return PCI_PROBE_NORMAL;
+}
 struct machdep_calls __initdata pSeries_md = {
        .probe                  = pSeries_probe,
        .setup_arch             = pSeries_setup_arch,
@@ -581,6 +604,7 @@ struct machdep_calls __initdata pSeries_md = {
        .get_cpuinfo            = pSeries_get_cpuinfo,
        .log_error              = pSeries_log_error,
        .pcibios_fixup          = pSeries_final_fixup,
+        .pci_probe_mode         = pSeries_pci_probe_mode,
        .irq_bus_setup          = pSeries_irq_bus_setup,
        .restart                = rtas_restart,
        .power_off              = rtas_power_off,
diff --git a/arch/ppc64/kernel/pSeries_smp.c b/arch/ppc64/kernel/pSeries_smp.c
index 62c55a123560..d2c7e2c4733b 100644
--- a/arch/ppc64/kernel/pSeries_smp.c
+++ b/arch/ppc64/kernel/pSeries_smp.c
@@ -41,6 +41,7 @@
 #include <asm/machdep.h>
 #include <asm/xics.h>
 #include <asm/cputable.h>
+#include <asm/firmware.h>
 #include <asm/system.h>
 #include <asm/rtas.h>
 #include <asm/plpar_wrappers.h>
@@ -271,6 +272,7 @@ static inline int __devinit smp_startup_cpu(unsigned int lcpu)
        unsigned long start_here = __pa((u32)*((unsigned long *)
                                               pSeries_secondary_smp_init));
        unsigned int pcpu;
+        int start_cpu;
        if (cpu_isset(lcpu, of_spin_map))
                /* Already started by OF and sitting in spin loop */
@@ -281,12 +283,20 @@ static inline int __devinit smp_startup_cpu(unsigned int lcpu)
        /* Fixup atomic count: it exited inside IRQ handler. */
        paca[lcpu].__current->thread_info->preempt_count        = 0;
-        status = rtas_call(rtas_token("start-cpu"), 3, 1, NULL,
+        /* 
-                           pcpu, start_here, lcpu);
+         * If the RTAS start-cpu token does not exist then presume the
+         * cpu is already spinning.
+         */
+        start_cpu = rtas_token("start-cpu");
+        if (start_cpu == RTAS_UNKNOWN_SERVICE)
+                return 1;
+        status = rtas_call(start_cpu, 3, 1, NULL, pcpu, start_here, lcpu);
        if (status != 0) {
                printk(KERN_ERR "start-cpu failed: %i\n", status);
                return 0;
        }
        return 1;
 }
@@ -326,7 +336,7 @@ static void __devinit smp_xics_setup_cpu(int cpu)
        if (cpu != boot_cpuid)
                xics_setup_cpu();
-        if (cur_cpu_spec->firmware_features & FW_FEATURE_SPLPAR)
+        if (firmware_has_feature(FW_FEATURE_SPLPAR))
                vpa_init(cpu);
        cpu_clear(cpu, of_spin_map);
diff --git a/arch/ppc64/kernel/pSeries_vio.c b/arch/ppc64/kernel/pSeries_vio.c
new file mode 100644
index 000000000000..e0ae06f58f86
--- /dev/null
+++ b/arch/ppc64/kernel/pSeries_vio.c
@@ -0,0 +1,273 @@
+/*
+ * IBM PowerPC pSeries Virtual I/O Infrastructure Support.
+ *
+ *    Copyright (c) 2003-2005 IBM Corp.
+ *     Dave Engebretsen engebret@us.ibm.com
+ *     Santiago Leon santil@us.ibm.com
+ *     Hollis Blanchard <hollisb@us.ibm.com>
+ *     Stephen Rothwell
+ *
+ *      This program is free software; you can redistribute it and/or
+ *      modify it under the terms of the GNU General Public License
+ *      as published by the Free Software Foundation; either version
+ *      2 of the License, or (at your option) any later version.
+ */
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/mm.h>
+#include <linux/kobject.h>
+#include <asm/iommu.h>
+#include <asm/dma.h>
+#include <asm/prom.h>
+#include <asm/vio.h>
+#include <asm/hvcall.h>
+extern struct subsystem devices_subsys; /* needed for vio_find_name() */
+static void probe_bus_pseries(void)
+{
+        struct device_node *node_vroot, *of_node;
+        node_vroot = find_devices("vdevice");
+        if ((node_vroot == NULL) || (node_vroot->child == NULL))
+                /* this machine doesn't do virtual IO, and that's ok */
+                return;
+        /*
+         * Create struct vio_devices for each virtual device in the device tree.
+         * Drivers will associate with them later.
+         */
+        for (of_node = node_vroot->child; of_node != NULL;
+                        of_node = of_node->sibling) {
+                printk(KERN_DEBUG "%s: processing %p\n", __FUNCTION__, of_node);
+                vio_register_device_node(of_node);
+        }
+}
+/**
+ * vio_match_device_pseries: - Tell if a pSeries VIO device matches a
+ *      vio_device_id
+ */
+static int vio_match_device_pseries(const struct vio_device_id *id,
+                const struct vio_dev *dev)
+{
+        return (strncmp(dev->type, id->type, strlen(id->type)) == 0) &&
+                        device_is_compatible(dev->dev.platform_data, id->compat);
+}
+static void vio_release_device_pseries(struct device *dev)
+{
+        /* XXX free TCE table */
+        of_node_put(dev->platform_data);
+}
+static ssize_t viodev_show_devspec(struct device *dev,
+                struct device_attribute *attr, char *buf)
+{
+        struct device_node *of_node = dev->platform_data;
+        return sprintf(buf, "%s\n", of_node->full_name);
+}
+DEVICE_ATTR(devspec, S_IRUSR | S_IRGRP | S_IROTH, viodev_show_devspec, NULL);
+static void vio_unregister_device_pseries(struct vio_dev *viodev)
+{
+        device_remove_file(&viodev->dev, &dev_attr_devspec);
+}
+static struct vio_bus_ops vio_bus_ops_pseries = {
+        .match = vio_match_device_pseries,
+        .unregister_device = vio_unregister_device_pseries,
+        .release_device = vio_release_device_pseries,
+};
+/**
+ * vio_bus_init_pseries: - Initialize the pSeries virtual IO bus
+ */
+static int __init vio_bus_init_pseries(void)
+{
+        int err;
+        err = vio_bus_init(&vio_bus_ops_pseries);
+        if (err == 0)
+                probe_bus_pseries();
+        return err;
+}
+__initcall(vio_bus_init_pseries);
+/**
+ * vio_build_iommu_table: - gets the dma information from OF and
+ *      builds the TCE tree.
+ * @dev: the virtual device.
+ *
+ * Returns a pointer to the built tce tree, or NULL if it can't
+ * find property.
+*/
+static struct iommu_table *vio_build_iommu_table(struct vio_dev *dev)
+{
+        unsigned int *dma_window;
+        struct iommu_table *newTceTable;
+        unsigned long offset;
+        int dma_window_property_size;
+        dma_window = (unsigned int *) get_property(dev->dev.platform_data, "ibm,my-dma-window", &dma_window_property_size);
+        if(!dma_window) {
+                return NULL;
+        }
+        newTceTable = (struct iommu_table *) kmalloc(sizeof(struct iommu_table), GFP_KERNEL);
+        /*  There should be some code to extract the phys-encoded offset
+                using prom_n_addr_cells(). However, according to a comment
+                on earlier versions, it's always zero, so we don't bother */
+        offset = dma_window[1] >>  PAGE_SHIFT;
+        /* TCE table size - measured in tce entries */
+        newTceTable->it_size            = dma_window[4] >> PAGE_SHIFT;
+        /* offset for VIO should always be 0 */
+        newTceTable->it_offset          = offset;
+        newTceTable->it_busno           = 0;
+        newTceTable->it_index           = (unsigned long)dma_window[0];
+        newTceTable->it_type            = TCE_VB;
+        return iommu_init_table(newTceTable);
+}
+/**
+ * vio_register_device_node: - Register a new vio device.
+ * @of_node:    The OF node for this device.
+ *
+ * Creates and initializes a vio_dev structure from the data in
+ * of_node (dev.platform_data) and adds it to the list of virtual devices.
+ * Returns a pointer to the created vio_dev or NULL if node has
+ * NULL device_type or compatible fields.
+ */
+struct vio_dev * __devinit vio_register_device_node(struct device_node *of_node)
+{
+        struct vio_dev *viodev;
+        unsigned int *unit_address;
+        unsigned int *irq_p;
+        /* we need the 'device_type' property, in order to match with drivers */
+        if ((NULL == of_node->type)) {
+                printk(KERN_WARNING
+                        "%s: node %s missing 'device_type'\n", __FUNCTION__,
+                        of_node->name ? of_node->name : "<unknown>");
+                return NULL;
+        }
+        unit_address = (unsigned int *)get_property(of_node, "reg", NULL);
+        if (!unit_address) {
+                printk(KERN_WARNING "%s: node %s missing 'reg'\n", __FUNCTION__,
+                        of_node->name ? of_node->name : "<unknown>");
+                return NULL;
+        }
+        /* allocate a vio_dev for this node */
+        viodev = kmalloc(sizeof(struct vio_dev), GFP_KERNEL);
+        if (!viodev) {
+                return NULL;
+        }
+        memset(viodev, 0, sizeof(struct vio_dev));
+        viodev->dev.platform_data = of_node_get(of_node);
+        viodev->irq = NO_IRQ;
+        irq_p = (unsigned int *)get_property(of_node, "interrupts", NULL);
+        if (irq_p) {
+                int virq = virt_irq_create_mapping(*irq_p);
+                if (virq == NO_IRQ) {
+                        printk(KERN_ERR "Unable to allocate interrupt "
+                               "number for %s\n", of_node->full_name);
+                } else
+                        viodev->irq = irq_offset_up(virq);
+        }
+        snprintf(viodev->dev.bus_id, BUS_ID_SIZE, "%x", *unit_address);
+        viodev->name = of_node->name;
+        viodev->type = of_node->type;
+        viodev->unit_address = *unit_address;
+        viodev->iommu_table = vio_build_iommu_table(viodev);
+        /* register with generic device framework */
+        if (vio_register_device(viodev) == NULL) {
+                /* XXX free TCE table */
+                kfree(viodev);
+                return NULL;
+        }
+        device_create_file(&viodev->dev, &dev_attr_devspec);
+        return viodev;
+}
+EXPORT_SYMBOL(vio_register_device_node);
+/**
+ * vio_get_attribute: - get attribute for virtual device
+ * @vdev:       The vio device to get property.
+ * @which:      The property/attribute to be extracted.
+ * @length:     Pointer to length of returned data size (unused if NULL).
+ *
+ * Calls prom.c's get_property() to return the value of the
+ * attribute specified by the preprocessor constant @which
+*/
+const void * vio_get_attribute(struct vio_dev *vdev, void* which, int* length)
+{
+        return get_property(vdev->dev.platform_data, (char*)which, length);
+}
+EXPORT_SYMBOL(vio_get_attribute);
+/* vio_find_name() - internal because only vio.c knows how we formatted the
+ * kobject name
+ * XXX once vio_bus_type.devices is actually used as a kset in
+ * drivers/base/bus.c, this function should be removed in favor of
+ * "device_find(kobj_name, &vio_bus_type)"
+ */
+static struct vio_dev *vio_find_name(const char *kobj_name)
+{
+        struct kobject *found;
+        found = kset_find_obj(&devices_subsys.kset, kobj_name);
+        if (!found)
+                return NULL;
+        return to_vio_dev(container_of(found, struct device, kobj));
+}
+/**
+ * vio_find_node - find an already-registered vio_dev
+ * @vnode: device_node of the virtual device we're looking for
+ */
+struct vio_dev *vio_find_node(struct device_node *vnode)
+{
+        uint32_t *unit_address;
+        char kobj_name[BUS_ID_SIZE];
+        /* construct the kobject name from the device node */
+        unit_address = (uint32_t *)get_property(vnode, "reg", NULL);
+        if (!unit_address)
+                return NULL;
+        snprintf(kobj_name, BUS_ID_SIZE, "%x", *unit_address);
+        return vio_find_name(kobj_name);
+}
+EXPORT_SYMBOL(vio_find_node);
+int vio_enable_interrupts(struct vio_dev *dev)
+{
+        int rc = h_vio_signal(dev->unit_address, VIO_IRQ_ENABLE);
+        if (rc != H_Success)
+                printk(KERN_ERR "vio: Error 0x%x enabling interrupts\n", rc);
+        return rc;
+}
+EXPORT_SYMBOL(vio_enable_interrupts);
+int vio_disable_interrupts(struct vio_dev *dev)
+{
+        int rc = h_vio_signal(dev->unit_address, VIO_IRQ_DISABLE);
+        if (rc != H_Success)
+                printk(KERN_ERR "vio: Error 0x%x disabling interrupts\n", rc);
+        return rc;
+}
+EXPORT_SYMBOL(vio_disable_interrupts);
diff --git a/arch/ppc64/kernel/pacaData.c b/arch/ppc64/kernel/pacaData.c
index 6316188737b6..33a2d8db3f21 100644
--- a/arch/ppc64/kernel/pacaData.c
+++ b/arch/ppc64/kernel/pacaData.c
@@ -59,6 +59,7 @@ extern unsigned long __toc_start;
                .fpregs_in_use = 1,                                         \
                .end_of_quantum = 0xfffffffffffffffful,                     \
                .slb_count = 64,                                            \
+                .vmxregs_in_use = 0,                                        \
        },                                                                  \
 #ifdef CONFIG_PPC_ISERIES
@@ -78,7 +79,7 @@ extern unsigned long __toc_start;
 #define BOOTCPU_PACA_INIT(number)                                           \
 {                                                                           \
-        PACA_INIT_COMMON(number, 1, 0, STAB0_VIRT_ADDR)                     \
+        PACA_INIT_COMMON(number, 1, 0, (u64)&initial_stab)                  \
        PACA_INIT_ISERIES(number)                                           \
 }
@@ -90,7 +91,7 @@ extern unsigned long __toc_start;
 #define BOOTCPU_PACA_INIT(number)                                           \
 {                                                                           \
-        PACA_INIT_COMMON(number, 1, STAB0_PHYS_ADDR, STAB0_VIRT_ADDR)       \
+        PACA_INIT_COMMON(number, 1, STAB0_PHYS_ADDR, (u64)&initial_stab)    \
 }
 #endif
diff --git a/arch/ppc64/kernel/pci.c b/arch/ppc64/kernel/pci.c
index d0d55c7908ef..861138ad092c 100644
--- a/arch/ppc64/kernel/pci.c
+++ b/arch/ppc64/kernel/pci.c
@@ -21,6 +21,7 @@
 #include <linux/bootmem.h>
 #include <linux/mm.h>
 #include <linux/list.h>
+#include <linux/syscalls.h>
 #include <asm/processor.h>
 #include <asm/io.h>
@@ -50,6 +51,10 @@ unsigned long io_page_mask;
 EXPORT_SYMBOL(io_page_mask);
+#ifdef CONFIG_PPC_MULTIPLATFORM
+static void fixup_resource(struct resource *res, struct pci_dev *dev);
+static void do_bus_setup(struct pci_bus *bus);
+#endif
 unsigned int pcibios_assign_all_busses(void)
 {
@@ -84,7 +89,6 @@ static void fixup_broken_pcnet32(struct pci_dev* dev)
        if ((dev->class>>8 == PCI_CLASS_NETWORK_ETHERNET)) {
                dev->vendor = PCI_VENDOR_ID_AMD;
                pci_write_config_word(dev, PCI_VENDOR_ID, PCI_VENDOR_ID_AMD);
-                pci_name_device(dev);
        }
 }
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_TRIDENT, PCI_ANY_ID, fixup_broken_pcnet32);
@@ -225,10 +229,287 @@ static void __init pcibios_claim_of_setup(void)
 }
 #endif
+#ifdef CONFIG_PPC_MULTIPLATFORM
+static u32 get_int_prop(struct device_node *np, const char *name, u32 def)
+{
+        u32 *prop;
+        int len;
+        prop = (u32 *) get_property(np, name, &len);
+        if (prop && len >= 4)
+                return *prop;
+        return def;
+}
+static unsigned int pci_parse_of_flags(u32 addr0)
+{
+        unsigned int flags = 0;
+        if (addr0 & 0x02000000) {
+                flags |= IORESOURCE_MEM;
+                if (addr0 & 0x40000000)
+                        flags |= IORESOURCE_PREFETCH;
+        } else if (addr0 & 0x01000000)
+                flags |= IORESOURCE_IO;
+        return flags;
+}
+#define GET_64BIT(prop, i)      ((((u64) (prop)[(i)]) << 32) | (prop)[(i)+1])
+static void pci_parse_of_addrs(struct device_node *node, struct pci_dev *dev)
+{
+        u64 base, size;
+        unsigned int flags;
+        struct resource *res;
+        u32 *addrs, i;
+        int proplen;
+        addrs = (u32 *) get_property(node, "assigned-addresses", &proplen);
+        if (!addrs)
+                return;
+        for (; proplen >= 20; proplen -= 20, addrs += 5) {
+                flags = pci_parse_of_flags(addrs[0]);
+                if (!flags)
+                        continue;
+                base = GET_64BIT(addrs, 1);
+                size = GET_64BIT(addrs, 3);
+                if (!size)
+                        continue;
+                i = addrs[0] & 0xff;
+                if (PCI_BASE_ADDRESS_0 <= i && i <= PCI_BASE_ADDRESS_5) {
+                        res = &dev->resource[(i - PCI_BASE_ADDRESS_0) >> 2];
+                } else if (i == dev->rom_base_reg) {
+                        res = &dev->resource[PCI_ROM_RESOURCE];
+                        flags |= IORESOURCE_READONLY | IORESOURCE_CACHEABLE;
+                } else {
+                        printk(KERN_ERR "PCI: bad cfg reg num 0x%x\n", i);
+                        continue;
+                }
+                res->start = base;
+                res->end = base + size - 1;
+                res->flags = flags;
+                res->name = pci_name(dev);
+                fixup_resource(res, dev);
+        }
+}
+static struct pci_dev *of_create_pci_dev(struct device_node *node,
+                                         struct pci_bus *bus, int devfn)
+{
+        struct pci_dev *dev;
+        const char *type;
+        dev = kmalloc(sizeof(struct pci_dev), GFP_KERNEL);
+        if (!dev)
+                return NULL;
+        type = get_property(node, "device_type", NULL);
+        if (type == NULL)
+                type = "";
+        memset(dev, 0, sizeof(struct pci_dev));
+        dev->bus = bus;
+        dev->sysdata = node;
+        dev->dev.parent = bus->bridge;
+        dev->dev.bus = &pci_bus_type;
+        dev->devfn = devfn;
+        dev->multifunction = 0;         /* maybe a lie? */
+        dev->vendor = get_int_prop(node, "vendor-id", 0xffff);
+        dev->device = get_int_prop(node, "device-id", 0xffff);
+        dev->subsystem_vendor = get_int_prop(node, "subsystem-vendor-id", 0);
+        dev->subsystem_device = get_int_prop(node, "subsystem-id", 0);
+        dev->cfg_size = 256; /*pci_cfg_space_size(dev);*/
+        sprintf(pci_name(dev), "%04x:%02x:%02x.%d", pci_domain_nr(bus),
+                dev->bus->number, PCI_SLOT(devfn), PCI_FUNC(devfn));
+        dev->class = get_int_prop(node, "class-code", 0);
+        dev->current_state = 4;         /* unknown power state */
+        if (!strcmp(type, "pci")) {
+                /* a PCI-PCI bridge */
+                dev->hdr_type = PCI_HEADER_TYPE_BRIDGE;
+                dev->rom_base_reg = PCI_ROM_ADDRESS1;
+        } else if (!strcmp(type, "cardbus")) {
+                dev->hdr_type = PCI_HEADER_TYPE_CARDBUS;
+        } else {
+                dev->hdr_type = PCI_HEADER_TYPE_NORMAL;
+                dev->rom_base_reg = PCI_ROM_ADDRESS;
+                dev->irq = NO_IRQ;
+                if (node->n_intrs > 0) {
+                        dev->irq = node->intrs[0].line;
+                        pci_write_config_byte(dev, PCI_INTERRUPT_LINE,
+                                              dev->irq);
+                }
+        }
+        pci_parse_of_addrs(node, dev);
+        pci_device_add(dev, bus);
+        /* XXX pci_scan_msi_device(dev); */
+        return dev;
+}
+static void of_scan_pci_bridge(struct device_node *node, struct pci_dev *dev);
+static void __devinit of_scan_bus(struct device_node *node,
+                                  struct pci_bus *bus)
+{
+        struct device_node *child = NULL;
+        u32 *reg;
+        int reglen, devfn;
+        struct pci_dev *dev;
+        while ((child = of_get_next_child(node, child)) != NULL) {
+                reg = (u32 *) get_property(child, "reg", &reglen);
+                if (reg == NULL || reglen < 20)
+                        continue;
+                devfn = (reg[0] >> 8) & 0xff;
+                /* create a new pci_dev for this device */
+                dev = of_create_pci_dev(child, bus, devfn);
+                if (!dev)
+                        continue;
+                if (dev->hdr_type == PCI_HEADER_TYPE_BRIDGE ||
+                    dev->hdr_type == PCI_HEADER_TYPE_CARDBUS)
+                        of_scan_pci_bridge(child, dev);
+        }
+        do_bus_setup(bus);
+}
+static void __devinit of_scan_pci_bridge(struct device_node *node,
+                                         struct pci_dev *dev)
+{
+        struct pci_bus *bus;
+        u32 *busrange, *ranges;
+        int len, i, mode;
+        struct resource *res;
+        unsigned int flags;
+        u64 size;
+        /* parse bus-range property */
+        busrange = (u32 *) get_property(node, "bus-range", &len);
+        if (busrange == NULL || len != 8) {
+                printk(KERN_ERR "Can't get bus-range for PCI-PCI bridge %s\n",
+                       node->full_name);
+                return;
+        }
+        ranges = (u32 *) get_property(node, "ranges", &len);
+        if (ranges == NULL) {
+                printk(KERN_ERR "Can't get ranges for PCI-PCI bridge %s\n",
+                       node->full_name);
+                return;
+        }
+        bus = pci_add_new_bus(dev->bus, dev, busrange[0]);
+        if (!bus) {
+                printk(KERN_ERR "Failed to create pci bus for %s\n",
+                       node->full_name);
+                return;
+        }
+        bus->primary = dev->bus->number;
+        bus->subordinate = busrange[1];
+        bus->bridge_ctl = 0;
+        bus->sysdata = node;
+        /* parse ranges property */
+        /* PCI #address-cells == 3 and #size-cells == 2 always */
+        res = &dev->resource[PCI_BRIDGE_RESOURCES];
+        for (i = 0; i < PCI_NUM_RESOURCES - PCI_BRIDGE_RESOURCES; ++i) {
+                res->flags = 0;
+                bus->resource[i] = res;
+                ++res;
+        }
+        i = 1;
+        for (; len >= 32; len -= 32, ranges += 8) {
+                flags = pci_parse_of_flags(ranges[0]);
+                size = GET_64BIT(ranges, 6);
+                if (flags == 0 || size == 0)
+                        continue;
+                if (flags & IORESOURCE_IO) {
+                        res = bus->resource[0];
+                        if (res->flags) {
+                                printk(KERN_ERR "PCI: ignoring extra I/O range"
+                                       " for bridge %s\n", node->full_name);
+                                continue;
+                        }
+                } else {
+                        if (i >= PCI_NUM_RESOURCES - PCI_BRIDGE_RESOURCES) {
+                                printk(KERN_ERR "PCI: too many memory ranges"
+                                       " for bridge %s\n", node->full_name);
+                                continue;
+                        }
+                        res = bus->resource[i];
+                        ++i;
+                }
+                res->start = GET_64BIT(ranges, 1);
+                res->end = res->start + size - 1;
+                res->flags = flags;
+                fixup_resource(res, dev);
+        }
+        sprintf(bus->name, "PCI Bus %04x:%02x", pci_domain_nr(bus),
+                bus->number);
+        mode = PCI_PROBE_NORMAL;
+        if (ppc_md.pci_probe_mode)
+                mode = ppc_md.pci_probe_mode(bus);
+        if (mode == PCI_PROBE_DEVTREE)
+                of_scan_bus(node, bus);
+        else if (mode == PCI_PROBE_NORMAL)
+                pci_scan_child_bus(bus);
+}
+#endif /* CONFIG_PPC_MULTIPLATFORM */
+static void __devinit scan_phb(struct pci_controller *hose)
+{
+        struct pci_bus *bus;
+        struct device_node *node = hose->arch_data;
+        int i, mode;
+        struct resource *res;
+        bus = pci_create_bus(NULL, hose->first_busno, hose->ops, node);
+        if (bus == NULL) {
+                printk(KERN_ERR "Failed to create bus for PCI domain %04x\n",
+                       hose->global_number);
+                return;
+        }
+        bus->secondary = hose->first_busno;
+        hose->bus = bus;
+        bus->resource[0] = res = &hose->io_resource;
+        if (res->flags && request_resource(&ioport_resource, res))
+                printk(KERN_ERR "Failed to request PCI IO region "
+                       "on PCI domain %04x\n", hose->global_number);
+        for (i = 0; i < 3; ++i) {
+                res = &hose->mem_resources[i];
+                bus->resource[i+1] = res;
+                if (res->flags && request_resource(&iomem_resource, res))
+                        printk(KERN_ERR "Failed to request PCI memory region "
+                               "on PCI domain %04x\n", hose->global_number);
+        }
+        mode = PCI_PROBE_NORMAL;
+#ifdef CONFIG_PPC_MULTIPLATFORM
+        if (ppc_md.pci_probe_mode)
+                mode = ppc_md.pci_probe_mode(bus);
+        if (mode == PCI_PROBE_DEVTREE) {
+                bus->subordinate = hose->last_busno;
+                of_scan_bus(node, bus);
+        }
+#endif /* CONFIG_PPC_MULTIPLATFORM */
+        if (mode == PCI_PROBE_NORMAL)
+                hose->last_busno = bus->subordinate = pci_scan_child_bus(bus);
+        pci_bus_add_devices(bus);
+}
 static int __init pcibios_init(void)
 {
        struct pci_controller *hose, *tmp;
-        struct pci_bus *bus;
        /* For now, override phys_mem_access_prot. If we need it,
         * later, we may move that initialization to each ppc_md
@@ -242,13 +523,8 @@ static int __init pcibios_init(void)
        printk("PCI: Probing PCI hardware\n");
        /* Scan all of the recorded PCI controllers.  */
-        list_for_each_entry_safe(hose, tmp, &hose_list, list_node) {
+        list_for_each_entry_safe(hose, tmp, &hose_list, list_node)
-                hose->last_busno = 0xff;
+                scan_phb(hose);
-                bus = pci_scan_bus(hose->first_busno, hose->ops,
-                                   hose->arch_data);
-                hose->bus = bus;
-                hose->last_busno = bus->subordinate;
-        }
 #ifndef CONFIG_PPC_ISERIES
        if (pci_probe_only)
@@ -820,118 +1096,89 @@ void phbs_remap_io(void)
 /*
 * ppc64 can have multifunction devices that do not respond to function 0.
 * In this case we must scan all functions.
+ * XXX this can go now, we use the OF device tree in all the
+ * cases that caused problems. -- paulus
 */
 int pcibios_scan_all_fns(struct pci_bus *bus, int devfn)
 {
-       struct device_node *busdn, *dn;
+       return 0;
+}
+static void __devinit fixup_resource(struct resource *res, struct pci_dev *dev)
+{
+        struct pci_controller *hose = pci_bus_to_host(dev->bus);
+        unsigned long start, end, mask, offset;
-       if (bus->self)
+        if (res->flags & IORESOURCE_IO) {
-               busdn = pci_device_to_OF_node(bus->self);
+                offset = (unsigned long)hose->io_base_virt - pci_io_base;
-       else
-               busdn = bus->sysdata;   /* must be a phb */
-       if (busdn == NULL)
+                start = res->start += offset;
-               return 0;
+                end = res->end += offset;
-       /*
+                /* Need to allow IO access to pages that are in the
-        * Check to see if there is any of the 8 functions are in the
+                   ISA range */
-        * device tree.  If they are then we need to scan all the
+                if (start < MAX_ISA_PORT) {
-        * functions of this slot.
+                        if (end > MAX_ISA_PORT)
-        */
+                                end = MAX_ISA_PORT;
-       for (dn = busdn->child; dn; dn = dn->sibling)
-               if ((dn->devfn >> 3) == (devfn >> 3))
-                       return 1;
-       return 0;
+                        start >>= PAGE_SHIFT;
-}
+                        end >>= PAGE_SHIFT;
+                        /* get the range of pages for the map */
+                        mask = ((1 << (end+1)) - 1) ^ ((1 << start) - 1);
+                        io_page_mask |= mask;
+                }
+        } else if (res->flags & IORESOURCE_MEM) {
+                res->start += hose->pci_mem_offset;
+                res->end += hose->pci_mem_offset;
+        }
+}
 void __devinit pcibios_fixup_device_resources(struct pci_dev *dev,
-                                           struct pci_bus *bus)
+                                              struct pci_bus *bus)
 {
        /* Update device resources.  */
-        struct pci_controller *hose = pci_bus_to_host(bus);
        int i;
-        for (i = 0; i < PCI_NUM_RESOURCES; i++) {
+        for (i = 0; i < PCI_NUM_RESOURCES; i++)
-                if (dev->resource[i].flags & IORESOURCE_IO) {
+                if (dev->resource[i].flags)
-                        unsigned long offset = (unsigned long)hose->io_base_virt
+                        fixup_resource(&dev->resource[i], dev);
-                                - pci_io_base;
-                        unsigned long start, end, mask;
-                        start = dev->resource[i].start += offset;
-                        end = dev->resource[i].end += offset;
-                        /* Need to allow IO access to pages that are in the
-                           ISA range */
-                        if (start < MAX_ISA_PORT) {
-                                if (end > MAX_ISA_PORT)
-                                        end = MAX_ISA_PORT;
-                                start >>= PAGE_SHIFT;
-                                end >>= PAGE_SHIFT;
-                                /* get the range of pages for the map */
-                                mask = ((1 << (end+1))-1) ^ ((1 << start)-1);
-                                io_page_mask |= mask;
-                        }
-                }
-                else if (dev->resource[i].flags & IORESOURCE_MEM) {
-                        dev->resource[i].start += hose->pci_mem_offset;
-                        dev->resource[i].end += hose->pci_mem_offset;
-                }
-        }
 }
 EXPORT_SYMBOL(pcibios_fixup_device_resources);
-void __devinit pcibios_fixup_bus(struct pci_bus *bus)
+static void __devinit do_bus_setup(struct pci_bus *bus)
 {
-        struct pci_controller *hose = pci_bus_to_host(bus);
+        struct pci_dev *dev;
-        struct pci_dev *dev = bus->self;
-        struct resource *res;
-        int i;
-        if (!dev) {
+        ppc_md.iommu_bus_setup(bus);
-                /* Root bus. */
-                hose->bus = bus;
+        list_for_each_entry(dev, &bus->devices, bus_list)
-                bus->resource[0] = res = &hose->io_resource;
+                ppc_md.iommu_dev_setup(dev);
-                if (res->flags && request_resource(&ioport_resource, res))
+        if (ppc_md.irq_bus_setup)
-                        printk(KERN_ERR "Failed to request IO on "
+                ppc_md.irq_bus_setup(bus);
-                                        "PCI domain %d\n", pci_domain_nr(bus));
+}
-                for (i = 0; i < 3; ++i) {
+void __devinit pcibios_fixup_bus(struct pci_bus *bus)
-                        res = &hose->mem_resources[i];
+{
-                        bus->resource[i+1] = res;
+        struct pci_dev *dev = bus->self;
-                        if (res->flags && request_resource(&iomem_resource, res))
-                                printk(KERN_ERR "Failed to request MEM on "
+        if (dev && pci_probe_only &&
-                                                "PCI domain %d\n",
+            (dev->class >> 8) == PCI_CLASS_BRIDGE_PCI) {
-                                                pci_domain_nr(bus));
-                }
-        } else if (pci_probe_only &&
-                   (dev->class >> 8) == PCI_CLASS_BRIDGE_PCI) {
                /* This is a subordinate bridge */
                pci_read_bridge_bases(bus);
                pcibios_fixup_device_resources(dev, bus);
        }
-        ppc_md.iommu_bus_setup(bus);
+        do_bus_setup(bus);
-        list_for_each_entry(dev, &bus->devices, bus_list)
-                ppc_md.iommu_dev_setup(dev);
-        if (ppc_md.irq_bus_setup)
-                ppc_md.irq_bus_setup(bus);
        if (!pci_probe_only)
                return;
-        list_for_each_entry(dev, &bus->devices, bus_list) {
+        list_for_each_entry(dev, &bus->devices, bus_list)
                if ((dev->class >> 8) != PCI_CLASS_BRIDGE_PCI)
                        pcibios_fixup_device_resources(dev, bus);
-        }
 }
 EXPORT_SYMBOL(pcibios_fixup_bus);
@@ -983,3 +1230,62 @@ void pci_resource_to_user(const struct pci_dev *dev, int bar,
 }
 #endif /* CONFIG_PPC_MULTIPLATFORM */
+#define IOBASE_BRIDGE_NUMBER    0
+#define IOBASE_MEMORY           1
+#define IOBASE_IO               2
+#define IOBASE_ISA_IO           3
+#define IOBASE_ISA_MEM          4
+long sys_pciconfig_iobase(long which, unsigned long in_bus,
+                          unsigned long in_devfn)
+{
+        struct pci_controller* hose;
+        struct list_head *ln;
+        struct pci_bus *bus = NULL;
+        struct device_node *hose_node;
+        /* Argh ! Please forgive me for that hack, but that's the
+         * simplest way to get existing XFree to not lockup on some
+         * G5 machines... So when something asks for bus 0 io base
+         * (bus 0 is HT root), we return the AGP one instead.
+         */
+#ifdef CONFIG_PPC_PMAC
+        if (systemcfg->platform == PLATFORM_POWERMAC &&
+            machine_is_compatible("MacRISC4"))
+                if (in_bus == 0)
+                        in_bus = 0xf0;
+#endif /* CONFIG_PPC_PMAC */
+        /* That syscall isn't quite compatible with PCI domains, but it's
+         * used on pre-domains setup. We return the first match
+         */
+        for (ln = pci_root_buses.next; ln != &pci_root_buses; ln = ln->next) {
+                bus = pci_bus_b(ln);
+                if (in_bus >= bus->number && in_bus < (bus->number + bus->subordinate))
+                        break;
+                bus = NULL;
+        }
+        if (bus == NULL || bus->sysdata == NULL)
+                return -ENODEV;
+        hose_node = (struct device_node *)bus->sysdata;
+        hose = PCI_DN(hose_node)->phb;
+        switch (which) {
+        case IOBASE_BRIDGE_NUMBER:
+                return (long)hose->first_busno;
+        case IOBASE_MEMORY:
+                return (long)hose->pci_mem_offset;
+        case IOBASE_IO:
+                return (long)hose->io_base_phys;
+        case IOBASE_ISA_IO:
+                return (long)isa_io_base;
+        case IOBASE_ISA_MEM:
+                return -EINVAL;
+        }
+        return -EOPNOTSUPP;
+}
diff --git a/arch/ppc64/kernel/pci.h b/arch/ppc64/kernel/pci.h
index 26be78b13af1..5eb2cc320566 100644
--- a/arch/ppc64/kernel/pci.h
+++ b/arch/ppc64/kernel/pci.h
@@ -34,7 +34,6 @@ void *traverse_pci_devices(struct device_node *start, traverse_func pre,
 void pci_devs_phb_init(void);
 void pci_devs_phb_init_dynamic(struct pci_controller *phb);
-struct device_node *fetch_dev_dn(struct pci_dev *dev);
 /* PCI address cache management routines */
 void pci_addr_cache_insert_device(struct pci_dev *dev);
diff --git a/arch/ppc64/kernel/pci_dn.c b/arch/ppc64/kernel/pci_dn.c
index ec345462afc3..a86389d07d57 100644
--- a/arch/ppc64/kernel/pci_dn.c
+++ b/arch/ppc64/kernel/pci_dn.c
@@ -23,6 +23,8 @@
 #include <linux/pci.h>
 #include <linux/string.h>
 #include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/bootmem.h>
 #include <asm/io.h>
 #include <asm/prom.h>
@@ -40,16 +42,26 @@ static void * __devinit update_dn_pci_info(struct device_node *dn, void *data)
        struct pci_controller *phb = data;
        int *type = (int *)get_property(dn, "ibm,pci-config-space-type", NULL);
        u32 *regs;
+        struct pci_dn *pdn;
-        dn->phb = phb;
+        if (phb->is_dynamic)
+                pdn = kmalloc(sizeof(*pdn), GFP_KERNEL);
+        else
+                pdn = alloc_bootmem(sizeof(*pdn));
+        if (pdn == NULL)
+                return NULL;
+        memset(pdn, 0, sizeof(*pdn));
+        dn->data = pdn;
+        pdn->node = dn;
+        pdn->phb = phb;
        regs = (u32 *)get_property(dn, "reg", NULL);
        if (regs) {
                /* First register entry is addr (00BBSS00)  */
-                dn->busno = (regs[0] >> 16) & 0xff;
+                pdn->busno = (regs[0] >> 16) & 0xff;
-                dn->devfn = (regs[0] >> 8) & 0xff;
+                pdn->devfn = (regs[0] >> 8) & 0xff;
        }
-        dn->pci_ext_config_space = (type && *type == 1);
+        pdn->pci_ext_config_space = (type && *type == 1);
        return NULL;
 }
@@ -112,10 +124,15 @@ void *traverse_pci_devices(struct device_node *start, traverse_func pre,
 void __devinit pci_devs_phb_init_dynamic(struct pci_controller *phb)
 {
        struct device_node * dn = (struct device_node *) phb->arch_data;
+        struct pci_dn *pdn;
        /* PHB nodes themselves must not match */
-        dn->devfn = dn->busno = -1;
+        update_dn_pci_info(dn, phb);
-        dn->phb = phb;
+        pdn = dn->data;
+        if (pdn) {
+                pdn->devfn = pdn->busno = -1;
+                pdn->phb = phb;
+        }
        /* Update dn->phb ptrs for new phb and children devices */
        traverse_pci_devices(dn, update_dn_pci_info, phb);
@@ -123,14 +140,17 @@ void __devinit pci_devs_phb_init_dynamic(struct pci_controller *phb)
 /*
 * Traversal func that looks for a <busno,devfcn> value.
- * If found, the device_node is returned (thus terminating the traversal).
+ * If found, the pci_dn is returned (thus terminating the traversal).
 */
 static void *is_devfn_node(struct device_node *dn, void *data)
 {
        int busno = ((unsigned long)data >> 8) & 0xff;
        int devfn = ((unsigned long)data) & 0xff;
+        struct pci_dn *pci = dn->data;
-        return ((devfn == dn->devfn) && (busno == dn->busno)) ? dn : NULL;
+        if (pci && (devfn == pci->devfn) && (busno == pci->busno))
+                return dn;
+        return NULL;
 }
 /*
@@ -149,13 +169,10 @@ static void *is_devfn_node(struct device_node *dn, void *data)
 struct device_node *fetch_dev_dn(struct pci_dev *dev)
 {
        struct device_node *orig_dn = dev->sysdata;
-        struct pci_controller *phb = orig_dn->phb; /* assume same phb as orig_dn */
-        struct device_node *phb_dn;
        struct device_node *dn;
        unsigned long searchval = (dev->bus->number << 8) | dev->devfn;
-        phb_dn = phb->arch_data;
+        dn = traverse_pci_devices(orig_dn, is_devfn_node, (void *)searchval);
-        dn = traverse_pci_devices(phb_dn, is_devfn_node, (void *)searchval);
        if (dn)
                dev->sysdata = dn;
        return dn;
@@ -165,11 +182,13 @@ EXPORT_SYMBOL(fetch_dev_dn);
 static int pci_dn_reconfig_notifier(struct notifier_block *nb, unsigned long action, void *node)
 {
        struct device_node *np = node;
+        struct pci_dn *pci;
        int err = NOTIFY_OK;
        switch (action) {
        case PSERIES_RECONFIG_ADD:
-                update_dn_pci_info(np, np->parent->phb);
+                pci = np->parent->data;
+                update_dn_pci_info(np, pci->phb);
                break;
        default:
                err = NOTIFY_DONE;
diff --git a/arch/ppc64/kernel/pci_iommu.c b/arch/ppc64/kernel/pci_iommu.c
index ef0a62b916be..14647e09c9cd 100644
--- a/arch/ppc64/kernel/pci_iommu.c
+++ b/arch/ppc64/kernel/pci_iommu.c
@@ -66,7 +66,7 @@ static inline struct iommu_table *devnode_table(struct device *dev)
 #endif /* CONFIG_PPC_ISERIES */
 #ifdef CONFIG_PPC_MULTIPLATFORM
-        return PCI_GET_DN(pdev)->iommu_table;
+        return PCI_DN(PCI_GET_DN(pdev))->iommu_table;
 #endif /* CONFIG_PPC_MULTIPLATFORM */
 }
diff --git a/arch/ppc64/kernel/pmac_feature.c b/arch/ppc64/kernel/pmac_feature.c
index 98ed2bccab1a..eb4e6c3f694d 100644
--- a/arch/ppc64/kernel/pmac_feature.c
+++ b/arch/ppc64/kernel/pmac_feature.c
@@ -674,6 +674,7 @@ void __init pmac_check_ht_link(void)
 #if 0 /* Disabled for now */
        u32     ufreq, freq, ucfg, cfg;
        struct device_node *pcix_node;
+        struct pci_dn *pdn;
        u8      px_bus, px_devfn;
        struct pci_controller *px_hose;
@@ -687,9 +688,10 @@ void __init pmac_check_ht_link(void)
                printk("No PCI-X bridge found\n");
                return;
        }
-        px_hose = pcix_node->phb;
+        pdn = pcix_node->data;
-        px_bus = pcix_node->busno;
+        px_hose = pdn->phb;
-        px_devfn = pcix_node->devfn;
+        px_bus = pdn->busno;
+        px_devfn = pdn->devfn;
        
        early_read_config_dword(px_hose, px_bus, px_devfn, 0xc4, &cfg);
        early_read_config_dword(px_hose, px_bus, px_devfn, 0xcc, &freq);
diff --git a/arch/ppc64/kernel/pmac_pci.c b/arch/ppc64/kernel/pmac_pci.c
index 71fe911ad183..dc40a0cad0b4 100644
--- a/arch/ppc64/kernel/pmac_pci.c
+++ b/arch/ppc64/kernel/pmac_pci.c
@@ -242,7 +242,7 @@ static int u3_ht_skip_device(struct pci_controller *hose,
        else
                busdn = hose->arch_data;
        for (dn = busdn->child; dn; dn = dn->sibling)
-                if (dn->devfn == devfn)
+                if (dn->data && PCI_DN(dn)->devfn == devfn)
                        break;
        if (dn == NULL)
                return -1;
@@ -388,7 +388,7 @@ static void __init setup_u3_agp(struct pci_controller* hose)
         * the reg address cell, we shall fix that by killing struct
         * reg_property and using some accessor functions instead
         */
-        hose->first_busno = 0xf0;
+        hose->first_busno = 0xf0;
        hose->last_busno = 0xff;
        has_uninorth = 1;
        hose->ops = &macrisc_pci_ops;
@@ -473,7 +473,7 @@ static void __init setup_u3_ht(struct pci_controller* hose)
                        continue;
                }               
                cur++;
-                DBG("U3/HT: hole, %d end at %08lx, %d start at %08lx\n",
+                DBG("U3/HT: hole, %d end at %08lx, %d start at %08lx\n",
                    cur-1, res->start - 1, cur, res->end + 1);
                hose->mem_resources[cur].name = np->full_name;
                hose->mem_resources[cur].flags = IORESOURCE_MEM;
@@ -603,24 +603,24 @@ static int __init add_bridge(struct device_node *dev)
        char* disp_name;
        int *bus_range;
        int primary = 1;
-        struct property *of_prop;
+        struct property *of_prop;
        DBG("Adding PCI host bridge %s\n", dev->full_name);
-        bus_range = (int *) get_property(dev, "bus-range", &len);
+        bus_range = (int *) get_property(dev, "bus-range", &len);
-        if (bus_range == NULL || len < 2 * sizeof(int)) {
+        if (bus_range == NULL || len < 2 * sizeof(int)) {
-                printk(KERN_WARNING "Can't get bus-range for %s, assume bus 0\n",
+                printk(KERN_WARNING "Can't get bus-range for %s, assume bus 0\n",
-                               dev->full_name);
+                        dev->full_name);
-        }
+        }
        hose = alloc_bootmem(sizeof(struct pci_controller));
        if (hose == NULL)
                return -ENOMEM;
-        pci_setup_pci_controller(hose);
+        pci_setup_pci_controller(hose);
-        hose->arch_data = dev;
+        hose->arch_data = dev;
-        hose->first_busno = bus_range ? bus_range[0] : 0;
+        hose->first_busno = bus_range ? bus_range[0] : 0;
-        hose->last_busno = bus_range ? bus_range[1] : 0xff;
+        hose->last_busno = bus_range ? bus_range[1] : 0xff;
        of_prop = alloc_bootmem(sizeof(struct property) +
                                sizeof(hose->global_number));
@@ -634,24 +634,24 @@ static int __init add_bridge(struct device_node *dev)
        }
        disp_name = NULL;
-        if (device_is_compatible(dev, "u3-agp")) {
+        if (device_is_compatible(dev, "u3-agp")) {
-                setup_u3_agp(hose);
+                setup_u3_agp(hose);
-                disp_name = "U3-AGP";
+                disp_name = "U3-AGP";
-                primary = 0;
+                primary = 0;
-        } else if (device_is_compatible(dev, "u3-ht")) {
+        } else if (device_is_compatible(dev, "u3-ht")) {
-                setup_u3_ht(hose);
+                setup_u3_ht(hose);
-                disp_name = "U3-HT";
+                disp_name = "U3-HT";
-                primary = 1;
+                primary = 1;
-        }
+        }
-        printk(KERN_INFO "Found %s PCI host bridge. Firmware bus number: %d->%d\n",
+        printk(KERN_INFO "Found %s PCI host bridge. Firmware bus number: %d->%d\n",
-                disp_name, hose->first_busno, hose->last_busno);
+                disp_name, hose->first_busno, hose->last_busno);
-        /* Interpret the "ranges" property */
+        /* Interpret the "ranges" property */
-        /* This also maps the I/O region and sets isa_io/mem_base */
+        /* This also maps the I/O region and sets isa_io/mem_base */
-        pmac_process_bridge_OF_ranges(hose, dev, primary);
+        pmac_process_bridge_OF_ranges(hose, dev, primary);
-        /* Fixup "bus-range" OF property */
+        /* Fixup "bus-range" OF property */
-        fixup_bus_range(dev);
+        fixup_bus_range(dev);
        return 0;
 }
@@ -746,9 +746,9 @@ void __init pmac_pci_init(void)
         */
        if (u3_agp) {
                struct device_node *np = u3_agp->arch_data;
-                np->busno = 0xf0;
+                PCI_DN(np)->busno = 0xf0;
                for (np = np->child; np; np = np->sibling)
-                        np->busno = 0xf0;
+                        PCI_DN(np)->busno = 0xf0;
        }
        pmac_check_ht_link();
diff --git a/arch/ppc64/kernel/pmac_setup.c b/arch/ppc64/kernel/pmac_setup.c
index e40877fa67cd..325426c7bed0 100644
--- a/arch/ppc64/kernel/pmac_setup.c
+++ b/arch/ppc64/kernel/pmac_setup.c
@@ -71,6 +71,7 @@
 #include <asm/of_device.h>
 #include <asm/lmb.h>
 #include <asm/smu.h>
+#include <asm/pmc.h>
 #include "pmac.h"
 #include "mpic.h"
@@ -273,16 +274,6 @@ static void __pmac pmac_halt(void)
 }
 #ifdef CONFIG_BOOTX_TEXT
-static int dummy_getc_poll(void)
-{
-        return -1;
-}
-static unsigned char dummy_getc(void)
-{
-        return 0;
-}
 static void btext_putc(unsigned char c)
 {
        btext_drawchar(c);
@@ -341,16 +332,13 @@ static void __init pmac_init_early(void)
                sccdbg = 1;
                udbg_init_scc(NULL);
        }
-        else {
 #ifdef CONFIG_BOOTX_TEXT
+        else {
                init_boot_display();
-                ppc_md.udbg_putc = btext_putc;
+                udbg_putc = btext_putc;
-                ppc_md.udbg_getc = dummy_getc;
-                ppc_md.udbg_getc_poll = dummy_getc_poll;
-#endif /* CONFIG_BOOTX_TEXT */
        }
+#endif /* CONFIG_BOOTX_TEXT */
        /* Setup interrupt mapping options */
        ppc64_interrupt_controller = IC_OPEN_PIC;
@@ -489,6 +477,18 @@ static int __init pmac_probe(int platform)
        return 1;
 }
+static int pmac_probe_mode(struct pci_bus *bus)
+{
+        struct device_node *node = bus->sysdata;
+        /* We need to use normal PCI probing for the AGP bus,
+           since the device for the AGP bridge isn't in the tree. */
+        if (bus->self == NULL && device_is_compatible(node, "u3-agp"))
+                return PCI_PROBE_NORMAL;
+        return PCI_PROBE_DEVTREE;
+}
 struct machdep_calls __initdata pmac_md = {
 #ifdef CONFIG_HOTPLUG_CPU
        .cpu_die                = generic_mach_cpu_die,
@@ -500,6 +500,7 @@ struct machdep_calls __initdata pmac_md = {
        .init_IRQ               = pmac_init_IRQ,
        .get_irq                = mpic_get_irq,
        .pcibios_fixup          = pmac_pcibios_fixup,
+        .pci_probe_mode         = pmac_probe_mode,
        .restart                = pmac_restart,
        .power_off              = pmac_power_off,
        .halt                   = pmac_halt,
@@ -511,4 +512,5 @@ struct machdep_calls __initdata pmac_md = {
        .progress               = pmac_progress,
        .check_legacy_ioport    = pmac_check_legacy_ioport,
        .idle_loop              = native_idle,
+        .enable_pmcs            = power4_enable_pmcs,
 };
diff --git a/arch/ppc64/kernel/pmc.c b/arch/ppc64/kernel/pmc.c
index 67be773f9c00..63d9481c3ec2 100644
--- a/arch/ppc64/kernel/pmc.c
+++ b/arch/ppc64/kernel/pmc.c
@@ -26,7 +26,7 @@ static void dummy_perf(struct pt_regs *regs)
        mtspr(SPRN_MMCR0, mmcr0);
 }
-static spinlock_t pmc_owner_lock = SPIN_LOCK_UNLOCKED;
+static DEFINE_SPINLOCK(pmc_owner_lock);
 static void *pmc_owner_caller; /* mostly for debugging */
 perf_irq_t perf_irq = dummy_perf;
@@ -65,3 +65,24 @@ void release_pmc_hardware(void)
        spin_unlock(&pmc_owner_lock);
 }
 EXPORT_SYMBOL_GPL(release_pmc_hardware);
+void power4_enable_pmcs(void)
+{
+        unsigned long hid0;
+        hid0 = mfspr(HID0);
+        hid0 |= 1UL << (63 - 20);
+        /* POWER4 requires the following sequence */
+        asm volatile(
+                "sync\n"
+                "mtspr     %1, %0\n"
+                "mfspr     %0, %1\n"
+                "mfspr     %0, %1\n"
+                "mfspr     %0, %1\n"
+                "mfspr     %0, %1\n"
+                "mfspr     %0, %1\n"
+                "mfspr     %0, %1\n"
+                "isync" : "=&r" (hid0) : "i" (HID0), "0" (hid0):
+                "memory");
+}
diff --git a/arch/ppc64/kernel/process.c b/arch/ppc64/kernel/process.c
index f7cae05e40fb..887005358eb1 100644
--- a/arch/ppc64/kernel/process.c
+++ b/arch/ppc64/kernel/process.c
@@ -50,9 +50,11 @@
 #include <asm/machdep.h>
 #include <asm/iSeries/HvCallHpt.h>
 #include <asm/cputable.h>
+#include <asm/firmware.h>
 #include <asm/sections.h>
 #include <asm/tlbflush.h>
 #include <asm/time.h>
+#include <asm/plpar_wrappers.h>
 #ifndef CONFIG_SMP
 struct task_struct *last_task_used_math = NULL;
@@ -162,7 +164,30 @@ int dump_task_altivec(struct pt_regs *regs, elf_vrregset_t *vrregs)
 #endif /* CONFIG_ALTIVEC */
+static void set_dabr_spr(unsigned long val)
+{
+        mtspr(SPRN_DABR, val);
+}
+int set_dabr(unsigned long dabr)
+{
+        int ret = 0;
+        if (firmware_has_feature(FW_FEATURE_XDABR)) {
+                /* We want to catch accesses from kernel and userspace */
+                unsigned long flags = H_DABRX_KERNEL|H_DABRX_USER;
+                ret = plpar_set_xdabr(dabr, flags);
+        } else if (firmware_has_feature(FW_FEATURE_DABR)) {
+                ret = plpar_set_dabr(dabr);
+        } else {
+                set_dabr_spr(dabr);
+        }
+        return ret;
+}
 DEFINE_PER_CPU(struct cpu_usage, cpu_usage_array);
+static DEFINE_PER_CPU(unsigned long, current_dabr);
 struct task_struct *__switch_to(struct task_struct *prev,
                                struct task_struct *new)
@@ -197,16 +222,20 @@ struct task_struct *__switch_to(struct task_struct *prev,
                new->thread.regs->msr |= MSR_VEC;
 #endif /* CONFIG_ALTIVEC */
+        if (unlikely(__get_cpu_var(current_dabr) != new->thread.dabr)) {
+                set_dabr(new->thread.dabr);
+                __get_cpu_var(current_dabr) = new->thread.dabr;
+        }
        flush_tlb_pending();
        new_thread = &new->thread;
        old_thread = &current->thread;
-/* Collect purr utilization data per process and per processor wise */
+        /* Collect purr utilization data per process and per processor
-/* purr is nothing but processor time base                          */
+         * wise purr is nothing but processor time base
+         */
-#if defined(CONFIG_PPC_PSERIES)
+        if (firmware_has_feature(FW_FEATURE_SPLPAR)) {
-        if (cur_cpu_spec->firmware_features & FW_FEATURE_SPLPAR) {
                struct cpu_usage *cu = &__get_cpu_var(cpu_usage_array);
                long unsigned start_tb, current_tb;
                start_tb = old_thread->start_tb;
@@ -214,8 +243,6 @@ struct task_struct *__switch_to(struct task_struct *prev,
                old_thread->accum_tb += (current_tb - start_tb);
                new_thread->start_tb = current_tb;
        }
-#endif
        local_irq_save(flags);
        last = _switch(old_thread, new_thread);
@@ -336,6 +363,11 @@ void flush_thread(void)
                last_task_used_altivec = NULL;
 #endif /* CONFIG_ALTIVEC */
 #endif /* CONFIG_SMP */
+        if (current->thread.dabr) {
+                current->thread.dabr = 0;
+                set_dabr(0);
+        }
 }
 void
diff --git a/arch/ppc64/kernel/prom.c b/arch/ppc64/kernel/prom.c
index 5aca01ddd81f..7035deb6de92 100644
--- a/arch/ppc64/kernel/prom.c
+++ b/arch/ppc64/kernel/prom.c
@@ -22,7 +22,6 @@
 #include <linux/kernel.h>
 #include <linux/string.h>
 #include <linux/init.h>
-#include <linux/version.h>
 #include <linux/threads.h>
 #include <linux/spinlock.h>
 #include <linux/types.h>
@@ -625,8 +624,8 @@ void __init finish_device_tree(void)
 static inline char *find_flat_dt_string(u32 offset)
 {
-        return ((char *)initial_boot_params) + initial_boot_params->off_dt_strings
+        return ((char *)initial_boot_params) +
-                + offset;
+                initial_boot_params->off_dt_strings + offset;
 }
 /**
@@ -635,26 +634,33 @@ static inline char *find_flat_dt_string(u32 offset)
 * unflatten the tree
 */
 static int __init scan_flat_dt(int (*it)(unsigned long node,
-                                         const char *full_path, void *data),
+                                         const char *uname, int depth,
+                                         void *data),
                               void *data)
 {
        unsigned long p = ((unsigned long)initial_boot_params) +
                initial_boot_params->off_dt_struct;
        int rc = 0;
+        int depth = -1;
        do {
                u32 tag = *((u32 *)p);
                char *pathp;
                
                p += 4;
-                if (tag == OF_DT_END_NODE)
+                if (tag == OF_DT_END_NODE) {
+                        depth --;
+                        continue;
+                }
+                if (tag == OF_DT_NOP)
                        continue;
                if (tag == OF_DT_END)
                        break;
                if (tag == OF_DT_PROP) {
                        u32 sz = *((u32 *)p);
                        p += 8;
-                        p = _ALIGN(p, sz >= 8 ? 8 : 4);
+                        if (initial_boot_params->version < 0x10)
+                                p = _ALIGN(p, sz >= 8 ? 8 : 4);
                        p += sz;
                        p = _ALIGN(p, 4);
                        continue;
@@ -664,9 +670,18 @@ static int __init scan_flat_dt(int (*it)(unsigned long node,
                               " device tree !\n", tag);
                        return -EINVAL;
                }
+                depth++;
                pathp = (char *)p;
                p = _ALIGN(p + strlen(pathp) + 1, 4);
-                rc = it(p, pathp, data);
+                if ((*pathp) == '/') {
+                        char *lp, *np;
+                        for (lp = NULL, np = pathp; *np; np++)
+                                if ((*np) == '/')
+                                        lp = np+1;
+                        if (lp != NULL)
+                                pathp = lp;
+                }
+                rc = it(p, pathp, depth, data);
                if (rc != 0)
                        break;          
        } while(1);
@@ -689,17 +704,21 @@ static void* __init get_flat_dt_prop(unsigned long node, const char *name,
                const char *nstr;
                p += 4;
+                if (tag == OF_DT_NOP)
+                        continue;
                if (tag != OF_DT_PROP)
                        return NULL;
                sz = *((u32 *)p);
                noff = *((u32 *)(p + 4));
                p += 8;
-                p = _ALIGN(p, sz >= 8 ? 8 : 4);
+                if (initial_boot_params->version < 0x10)
+                        p = _ALIGN(p, sz >= 8 ? 8 : 4);
                nstr = find_flat_dt_string(noff);
                if (nstr == NULL) {
-                        printk(KERN_WARNING "Can't find property index name !\n");
+                        printk(KERN_WARNING "Can't find property index"
+                               " name !\n");
                        return NULL;
                }
                if (strcmp(name, nstr) == 0) {
@@ -713,7 +732,7 @@ static void* __init get_flat_dt_prop(unsigned long node, const char *name,
 }
 static void *__init unflatten_dt_alloc(unsigned long *mem, unsigned long size,
-                                               unsigned long align)
+                                       unsigned long align)
 {
        void *res;
@@ -727,13 +746,16 @@ static void *__init unflatten_dt_alloc(unsigned long *mem, unsigned long size,
 static unsigned long __init unflatten_dt_node(unsigned long mem,
                                              unsigned long *p,
                                              struct device_node *dad,
-                                              struct device_node ***allnextpp)
+                                              struct device_node ***allnextpp,
+                                              unsigned long fpsize)
 {
        struct device_node *np;
        struct property *pp, **prev_pp = NULL;
        char *pathp;
        u32 tag;
-        unsigned int l;
+        unsigned int l, allocl;
+        int has_name = 0;
+        int new_format = 0;
        tag = *((u32 *)(*p));
        if (tag != OF_DT_BEGIN_NODE) {
@@ -742,21 +764,62 @@ static unsigned long __init unflatten_dt_node(unsigned long mem,
        }
        *p += 4;
        pathp = (char *)*p;
-        l = strlen(pathp) + 1;
+        l = allocl = strlen(pathp) + 1;
        *p = _ALIGN(*p + l, 4);
-        np = unflatten_dt_alloc(&mem, sizeof(struct device_node) + l,
+        /* version 0x10 has a more compact unit name here instead of the full
+         * path. we accumulate the full path size using "fpsize", we'll rebuild
+         * it later. We detect this because the first character of the name is
+         * not '/'.
+         */
+        if ((*pathp) != '/') {
+                new_format = 1;
+                if (fpsize == 0) {
+                        /* root node: special case. fpsize accounts for path
+                         * plus terminating zero. root node only has '/', so
+                         * fpsize should be 2, but we want to avoid the first
+                         * level nodes to have two '/' so we use fpsize 1 here
+                         */
+                        fpsize = 1;
+                        allocl = 2;
+                } else {
+                        /* account for '/' and path size minus terminal 0
+                         * already in 'l'
+                         */
+                        fpsize += l;
+                        allocl = fpsize;
+                }
+        }
+        np = unflatten_dt_alloc(&mem, sizeof(struct device_node) + allocl,
                                __alignof__(struct device_node));
        if (allnextpp) {
                memset(np, 0, sizeof(*np));
                np->full_name = ((char*)np) + sizeof(struct device_node);
-                memcpy(np->full_name, pathp, l);
+                if (new_format) {
+                        char *p = np->full_name;
+                        /* rebuild full path for new format */
+                        if (dad && dad->parent) {
+                                strcpy(p, dad->full_name);
+#ifdef DEBUG
+                                if ((strlen(p) + l + 1) != allocl) {
+                                        DBG("%s: p: %d, l: %d, a: %d\n",
+                                            pathp, strlen(p), l, allocl);
+                                }
+#endif
+                                p += strlen(p);
+                        }
+                        *(p++) = '/';
+                        memcpy(p, pathp, l);
+                } else
+                        memcpy(np->full_name, pathp, l);
                prev_pp = &np->properties;
                **allnextpp = np;
                *allnextpp = &np->allnext;
                if (dad != NULL) {
                        np->parent = dad;
-                        /* we temporarily use the `next' field as `last_child'. */
+                        /* we temporarily use the next field as `last_child'*/
                        if (dad->next == 0)
                                dad->child = np;
                        else
@@ -770,18 +833,26 @@ static unsigned long __init unflatten_dt_node(unsigned long mem,
                char *pname;
                tag = *((u32 *)(*p));
+                if (tag == OF_DT_NOP) {
+                        *p += 4;
+                        continue;
+                }
                if (tag != OF_DT_PROP)
                        break;
                *p += 4;
                sz = *((u32 *)(*p));
                noff = *((u32 *)((*p) + 4));
-                *p = _ALIGN((*p) + 8, sz >= 8 ? 8 : 4);
+                *p += 8;
+                if (initial_boot_params->version < 0x10)
+                        *p = _ALIGN(*p, sz >= 8 ? 8 : 4);
                pname = find_flat_dt_string(noff);
                if (pname == NULL) {
                        printk("Can't find property name in list !\n");
                        break;
                }
+                if (strcmp(pname, "name") == 0)
+                        has_name = 1;
                l = strlen(pname) + 1;
                pp = unflatten_dt_alloc(&mem, sizeof(struct property),
                                        __alignof__(struct property));
@@ -801,6 +872,36 @@ static unsigned long __init unflatten_dt_node(unsigned long mem,
                }
                *p = _ALIGN((*p) + sz, 4);
        }
+        /* with version 0x10 we may not have the name property, recreate
+         * it here from the unit name if absent
+         */
+        if (!has_name) {
+                char *p = pathp, *ps = pathp, *pa = NULL;
+                int sz;
+                while (*p) {
+                        if ((*p) == '@')
+                                pa = p;
+                        if ((*p) == '/')
+                                ps = p + 1;
+                        p++;
+                }
+                if (pa < ps)
+                        pa = p;
+                sz = (pa - ps) + 1;
+                pp = unflatten_dt_alloc(&mem, sizeof(struct property) + sz,
+                                        __alignof__(struct property));
+                if (allnextpp) {
+                        pp->name = "name";
+                        pp->length = sz;
+                        pp->value = (unsigned char *)(pp + 1);
+                        *prev_pp = pp;
+                        prev_pp = &pp->next;
+                        memcpy(pp->value, ps, sz - 1);
+                        ((char *)pp->value)[sz - 1] = 0;
+                        DBG("fixed up name for %s -> %s\n", pathp, pp->value);
+                }
+        }
        if (allnextpp) {
                *prev_pp = NULL;
                np->name = get_property(np, "name", NULL);
@@ -812,11 +913,11 @@ static unsigned long __init unflatten_dt_node(unsigned long mem,
                        np->type = "<NULL>";
        }
        while (tag == OF_DT_BEGIN_NODE) {
-                mem = unflatten_dt_node(mem, p, np, allnextpp);
+                mem = unflatten_dt_node(mem, p, np, allnextpp, fpsize);
                tag = *((u32 *)(*p));
        }
        if (tag != OF_DT_END_NODE) {
-                printk("Weird tag at start of node: %x\n", tag);
+                printk("Weird tag at end of node: %x\n", tag);
                return mem;
        }
        *p += 4;
@@ -842,21 +943,32 @@ void __init unflatten_device_tree(void)
        /* First pass, scan for size */
        start = ((unsigned long)initial_boot_params) +
                initial_boot_params->off_dt_struct;
-        size = unflatten_dt_node(0, &start, NULL, NULL);
+        size = unflatten_dt_node(0, &start, NULL, NULL, 0);
+        size = (size | 3) + 1;
        DBG("  size is %lx, allocating...\n", size);
        /* Allocate memory for the expanded device tree */
-        mem = (unsigned long)abs_to_virt(lmb_alloc(size,
+        mem = lmb_alloc(size + 4, __alignof__(struct device_node));
-                                                   __alignof__(struct device_node)));
+        if (!mem) {
+                DBG("Couldn't allocate memory with lmb_alloc()!\n");
+                panic("Couldn't allocate memory with lmb_alloc()!\n");
+        }
+        mem = (unsigned long)abs_to_virt(mem);
+        ((u32 *)mem)[size / 4] = 0xdeadbeef;
        DBG("  unflattening...\n", mem);
        /* Second pass, do actual unflattening */
        start = ((unsigned long)initial_boot_params) +
                initial_boot_params->off_dt_struct;
-        unflatten_dt_node(mem, &start, NULL, &allnextp);
+        unflatten_dt_node(mem, &start, NULL, &allnextp, 0);
        if (*((u32 *)start) != OF_DT_END)
-                printk(KERN_WARNING "Weird tag at end of tree: %x\n", *((u32 *)start));
+                printk(KERN_WARNING "Weird tag at end of tree: %08x\n", *((u32 *)start));
+        if (((u32 *)mem)[size / 4] != 0xdeadbeef)
+                printk(KERN_WARNING "End of tree marker overwritten: %08x\n",
+                       ((u32 *)mem)[size / 4] );
        *allnextp = NULL;
        /* Get pointer to OF "/chosen" node for use everywhere */
@@ -880,7 +992,7 @@ void __init unflatten_device_tree(void)
 static int __init early_init_dt_scan_cpus(unsigned long node,
-                                          const char *full_path, void *data)
+                                          const char *uname, int depth, void *data)
 {
        char *type = get_flat_dt_prop(node, "device_type", NULL);
        u32 *prop;
@@ -947,13 +1059,15 @@ static int __init early_init_dt_scan_cpus(unsigned long node,
 }
 static int __init early_init_dt_scan_chosen(unsigned long node,
-                                            const char *full_path, void *data)
+                                            const char *uname, int depth, void *data)
 {
        u32 *prop;
        u64 *prop64;
        extern unsigned long memory_limit, tce_alloc_start, tce_alloc_end;
-        if (strcmp(full_path, "/chosen") != 0)
+        DBG("search \"chosen\", depth: %d, uname: %s\n", depth, uname);
+        if (depth != 1 || strcmp(uname, "chosen") != 0)
                return 0;
        /* get platform type */
@@ -1003,18 +1117,20 @@ static int __init early_init_dt_scan_chosen(unsigned long node,
 }
 static int __init early_init_dt_scan_root(unsigned long node,
-                                          const char *full_path, void *data)
+                                          const char *uname, int depth, void *data)
 {
        u32 *prop;
-        if (strcmp(full_path, "/") != 0)
+        if (depth != 0)
                return 0;
        prop = (u32 *)get_flat_dt_prop(node, "#size-cells", NULL);
        dt_root_size_cells = (prop == NULL) ? 1 : *prop;
-                
+        DBG("dt_root_size_cells = %x\n", dt_root_size_cells);
        prop = (u32 *)get_flat_dt_prop(node, "#address-cells", NULL);
        dt_root_addr_cells = (prop == NULL) ? 2 : *prop;
+        DBG("dt_root_addr_cells = %x\n", dt_root_addr_cells);
        
        /* break now */
        return 1;
@@ -1042,7 +1158,7 @@ static unsigned long __init dt_mem_next_cell(int s, cell_t **cellp)
 static int __init early_init_dt_scan_memory(unsigned long node,
-                                            const char *full_path, void *data)
+                                            const char *uname, int depth, void *data)
 {
        char *type = get_flat_dt_prop(node, "device_type", NULL);
        cell_t *reg, *endp;
@@ -1058,7 +1174,9 @@ static int __init early_init_dt_scan_memory(unsigned long node,
        endp = reg + (l / sizeof(cell_t));
-        DBG("memory scan node %s ...\n", full_path);
+        DBG("memory scan node %s ..., reg size %ld, data: %x %x %x %x, ...\n",
+            uname, l, reg[0], reg[1], reg[2], reg[3]);
        while ((endp - reg) >= (dt_root_addr_cells + dt_root_size_cells)) {
                unsigned long base, size;
@@ -1469,10 +1587,11 @@ struct device_node *of_find_node_by_path(const char *path)
        struct device_node *np = allnodes;
        read_lock(&devtree_lock);
-        for (; np != 0; np = np->allnext)
+        for (; np != 0; np = np->allnext) {
                if (np->full_name != 0 && strcasecmp(np->full_name, path) == 0
                    && of_node_get(np))
                        break;
+        }
        read_unlock(&devtree_lock);
        return np;
 }
@@ -1614,6 +1733,7 @@ static void of_node_release(struct kref *kref)
        kfree(node->intrs);
        kfree(node->addrs);
        kfree(node->full_name);
+        kfree(node->data);
        kfree(node);
 }
diff --git a/arch/ppc64/kernel/prom_init.c b/arch/ppc64/kernel/prom_init.c
index dbbe6c79d8da..9979919cdf92 100644
--- a/arch/ppc64/kernel/prom_init.c
+++ b/arch/ppc64/kernel/prom_init.c
@@ -22,7 +22,6 @@
 #include <linux/kernel.h>
 #include <linux/string.h>
 #include <linux/init.h>
-#include <linux/version.h>
 #include <linux/threads.h>
 #include <linux/spinlock.h>
 #include <linux/types.h>
@@ -892,7 +891,10 @@ static void __init prom_init_mem(void)
        if ( RELOC(of_platform) == PLATFORM_PSERIES_LPAR )
                RELOC(alloc_top) = RELOC(rmo_top);
        else
-                RELOC(alloc_top) = RELOC(rmo_top) = min(0x40000000ul, RELOC(ram_top));
+                /* Some RS64 machines have buggy firmware where claims up at 1GB
+                 * fails. Cap at 768MB as a workaround. Still plenty of room.
+                 */
+                RELOC(alloc_top) = RELOC(rmo_top) = min(0x30000000ul, RELOC(ram_top));
        prom_printf("memory layout at init:\n");
        prom_printf("  memory_limit : %x (16 MB aligned)\n", RELOC(prom_memory_limit));
@@ -1534,7 +1536,8 @@ static unsigned long __init dt_find_string(char *str)
 */
 #define MAX_PROPERTY_NAME 64
-static void __init scan_dt_build_strings(phandle node, unsigned long *mem_start,
+static void __init scan_dt_build_strings(phandle node,
+                                         unsigned long *mem_start,
                                         unsigned long *mem_end)
 {
        unsigned long offset = reloc_offset();
@@ -1547,16 +1550,21 @@ static void __init scan_dt_build_strings(phandle node, unsigned long *mem_start,
        /* get and store all property names */
        prev_name = RELOC("");
        for (;;) {
-                int rc;
                /* 64 is max len of name including nul. */
                namep = make_room(mem_start, mem_end, MAX_PROPERTY_NAME, 1);
-                rc = call_prom("nextprop", 3, 1, node, prev_name, namep);
+                if (call_prom("nextprop", 3, 1, node, prev_name, namep) != 1) {
-                if (rc != 1) {
                        /* No more nodes: unwind alloc */
                        *mem_start = (unsigned long)namep;
                        break;
                }
+                /* skip "name" */
+                if (strcmp(namep, RELOC("name")) == 0) {
+                        *mem_start = (unsigned long)namep;
+                        prev_name = RELOC("name");
+                        continue;
+                }
+                /* get/create string entry */
                soff = dt_find_string(namep);
                if (soff != 0) {
                        *mem_start = (unsigned long)namep;
@@ -1571,7 +1579,7 @@ static void __init scan_dt_build_strings(phandle node, unsigned long *mem_start,
        /* do all our children */
        child = call_prom("child", 1, 1, node);
-        while (child != (phandle)0) {
+        while (child != 0) {
                scan_dt_build_strings(child, mem_start, mem_end);
                child = call_prom("peer", 1, 1, child);
        }
@@ -1580,16 +1588,13 @@ static void __init scan_dt_build_strings(phandle node, unsigned long *mem_start,
 static void __init scan_dt_build_struct(phandle node, unsigned long *mem_start,
                                        unsigned long *mem_end)
 {
-        int l, align;
        phandle child;
-        char *namep, *prev_name, *sstart, *p, *ep;
+        char *namep, *prev_name, *sstart, *p, *ep, *lp, *path;
        unsigned long soff;
        unsigned char *valp;
        unsigned long offset = reloc_offset();
-        char pname[MAX_PROPERTY_NAME];
+        static char pname[MAX_PROPERTY_NAME];
-        char *path;
+        int l;
-        path = RELOC(prom_scratch);
        dt_push_token(OF_DT_BEGIN_NODE, mem_start, mem_end);
@@ -1599,23 +1604,33 @@ static void __init scan_dt_build_struct(phandle node, unsigned long *mem_start,
                      namep, *mem_end - *mem_start);
        if (l >= 0) {
                /* Didn't fit?  Get more room. */
-                if (l+1 > *mem_end - *mem_start) {
+                if ((l+1) > (*mem_end - *mem_start)) {
                        namep = make_room(mem_start, mem_end, l+1, 1);
                        call_prom("package-to-path", 3, 1, node, namep, l);
                }
                namep[l] = '\0';
                /* Fixup an Apple bug where they have bogus \0 chars in the
                 * middle of the path in some properties
                 */
                for (p = namep, ep = namep + l; p < ep; p++)
                        if (*p == '\0') {
                                memmove(p, p+1, ep - p);
-                                ep--; l--;
+                                ep--; l--; p--;
                        }
-                *mem_start = _ALIGN(((unsigned long) namep) + strlen(namep) + 1, 4);
+                /* now try to extract the unit name in that mess */
+                for (p = namep, lp = NULL; *p; p++)
+                        if (*p == '/')
+                                lp = p + 1;
+                if (lp != NULL)
+                        memmove(namep, lp, strlen(lp) + 1);
+                *mem_start = _ALIGN(((unsigned long) namep) +
+                                    strlen(namep) + 1, 4);
        }
        /* get it again for debugging */
+        path = RELOC(prom_scratch);
        memset(path, 0, PROM_SCRATCH_SIZE);
        call_prom("package-to-path", 3, 1, node, path, PROM_SCRATCH_SIZE-1);
@@ -1623,23 +1638,27 @@ static void __init scan_dt_build_struct(phandle node, unsigned long *mem_start,
        prev_name = RELOC("");
        sstart = (char *)RELOC(dt_string_start);
        for (;;) {
-                int rc;
+                if (call_prom("nextprop", 3, 1, node, prev_name,
+                              RELOC(pname)) != 1)
-                rc = call_prom("nextprop", 3, 1, node, prev_name, pname);
-                if (rc != 1)
                        break;
+                /* skip "name" */
+                if (strcmp(RELOC(pname), RELOC("name")) == 0) {
+                        prev_name = RELOC("name");
+                        continue;
+                }
                /* find string offset */
-                soff = dt_find_string(pname);
+                soff = dt_find_string(RELOC(pname));
                if (soff == 0) {
-                        prom_printf("WARNING: Can't find string index for <%s>, node %s\n",
+                        prom_printf("WARNING: Can't find string index for"
-                                    pname, path);
+                                    " <%s>, node %s\n", RELOC(pname), path);
                        break;
                }
                prev_name = sstart + soff;
                /* get length */
-                l = call_prom("getproplen", 2, 1, node, pname);
+                l = call_prom("getproplen", 2, 1, node, RELOC(pname));
                /* sanity checks */
                if (l == PROM_ERROR)
@@ -1648,7 +1667,7 @@ static void __init scan_dt_build_struct(phandle node, unsigned long *mem_start,
                        prom_printf("WARNING: ignoring large property ");
                        /* It seems OF doesn't null-terminate the path :-( */
                        prom_printf("[%s] ", path);
-                        prom_printf("%s length 0x%x\n", pname, l);
+                        prom_printf("%s length 0x%x\n", RELOC(pname), l);
                        continue;
                }
@@ -1658,17 +1677,16 @@ static void __init scan_dt_build_struct(phandle node, unsigned long *mem_start,
                dt_push_token(soff, mem_start, mem_end);
                /* push property content */
-                align = (l >= 8) ? 8 : 4;
+                valp = make_room(mem_start, mem_end, l, 4);
-                valp = make_room(mem_start, mem_end, l, align);
+                call_prom("getprop", 4, 1, node, RELOC(pname), valp, l);
-                call_prom("getprop", 4, 1, node, pname, valp, l);
                *mem_start = _ALIGN(*mem_start, 4);
        }
        /* Add a "linux,phandle" property. */
        soff = dt_find_string(RELOC("linux,phandle"));
        if (soff == 0)
-                prom_printf("WARNING: Can't find string index for <linux-phandle>"
+                prom_printf("WARNING: Can't find string index for"
-                            " node %s\n", path);
+                            " <linux-phandle> node %s\n", path);
        else {
                dt_push_token(OF_DT_PROP, mem_start, mem_end);
                dt_push_token(4, mem_start, mem_end);
@@ -1679,7 +1697,7 @@ static void __init scan_dt_build_struct(phandle node, unsigned long *mem_start,
        /* do all our children */
        child = call_prom("child", 1, 1, node);
-        while (child != (phandle)0) {
+        while (child != 0) {
                scan_dt_build_struct(child, mem_start, mem_end);
                child = call_prom("peer", 1, 1, child);
        }
@@ -1718,7 +1736,8 @@ static void __init flatten_device_tree(void)
        /* Build header and make room for mem rsv map */ 
        mem_start = _ALIGN(mem_start, 4);
-        hdr = make_room(&mem_start, &mem_end, sizeof(struct boot_param_header), 4);
+        hdr = make_room(&mem_start, &mem_end,
+                        sizeof(struct boot_param_header), 4);
        RELOC(dt_header_start) = (unsigned long)hdr;
        rsvmap = make_room(&mem_start, &mem_end, sizeof(mem_reserve_map), 8);
@@ -1731,11 +1750,11 @@ static void __init flatten_device_tree(void)
        namep = make_room(&mem_start, &mem_end, 16, 1);
        strcpy(namep, RELOC("linux,phandle"));
        mem_start = (unsigned long)namep + strlen(namep) + 1;
-        RELOC(dt_string_end) = mem_start;
        /* Build string array */
        prom_printf("Building dt strings...\n"); 
        scan_dt_build_strings(root, &mem_start, &mem_end);
+        RELOC(dt_string_end) = mem_start;
        /* Build structure */
        mem_start = PAGE_ALIGN(mem_start);
@@ -1750,9 +1769,11 @@ static void __init flatten_device_tree(void)
        hdr->totalsize = RELOC(dt_struct_end) - RELOC(dt_header_start);
        hdr->off_dt_struct = RELOC(dt_struct_start) - RELOC(dt_header_start);
        hdr->off_dt_strings = RELOC(dt_string_start) - RELOC(dt_header_start);
+        hdr->dt_strings_size = RELOC(dt_string_end) - RELOC(dt_string_start);
        hdr->off_mem_rsvmap = ((unsigned long)rsvmap) - RELOC(dt_header_start);
        hdr->version = OF_DT_VERSION;
-        hdr->last_comp_version = 1;
+        /* Version 16 is not backward compatible */
+        hdr->last_comp_version = 0x10;
        /* Reserve the whole thing and copy the reserve map in, we
         * also bump mem_reserve_cnt to cause further reservations to
@@ -1808,6 +1829,9 @@ static void __init fixup_device_tree(void)
        /* does it need fixup ? */
        if (prom_getproplen(i2c, "interrupts") > 0)
                return;
+        prom_printf("fixing up bogus interrupts for u3 i2c...\n");
        /* interrupt on this revision of u3 is number 0 and level */
        interrupts[0] = 0;
        interrupts[1] = 1;
diff --git a/arch/ppc64/kernel/ptrace.c b/arch/ppc64/kernel/ptrace.c
index 2993f108d96d..85ed3188a91d 100644
--- a/arch/ppc64/kernel/ptrace.c
+++ b/arch/ppc64/kernel/ptrace.c
@@ -17,6 +17,7 @@
 * this archive for more details.
 */
+#include <linux/config.h>
 #include <linux/kernel.h>
 #include <linux/sched.h>
 #include <linux/mm.h>
@@ -206,6 +207,19 @@ int sys_ptrace(long request, long pid, long addr, long data)
                break;
        }
+        case PTRACE_GET_DEBUGREG: {
+                ret = -EINVAL;
+                /* We only support one DABR and no IABRS at the moment */
+                if (addr > 0)
+                        break;
+                ret = put_user(child->thread.dabr,
+                               (unsigned long __user *)data);
+                break;
+        }
+        case PTRACE_SET_DEBUGREG:
+                ret = ptrace_set_debugreg(child, addr, data);
        case PTRACE_DETACH:
                ret = ptrace_detach(child, data);
                break;
@@ -274,6 +288,20 @@ int sys_ptrace(long request, long pid, long addr, long data)
                break;
        }
+#ifdef CONFIG_ALTIVEC
+        case PTRACE_GETVRREGS:
+                /* Get the child altivec register state. */
+                flush_altivec_to_thread(child);
+                ret = get_vrregs((unsigned long __user *)data, child);
+                break;
+        case PTRACE_SETVRREGS:
+                /* Set the child altivec register state. */
+                flush_altivec_to_thread(child);
+                ret = set_vrregs(child, (unsigned long __user *)data);
+                break;
+#endif
        default:
                ret = ptrace_request(child, request, addr, data);
                break;
diff --git a/arch/ppc64/kernel/ptrace32.c b/arch/ppc64/kernel/ptrace32.c
index 16436426c7e2..fb8c22d6084a 100644
--- a/arch/ppc64/kernel/ptrace32.c
+++ b/arch/ppc64/kernel/ptrace32.c
@@ -17,6 +17,7 @@
 * this archive for more details.
 */
+#include <linux/config.h>
 #include <linux/kernel.h>
 #include <linux/sched.h>
 #include <linux/mm.h>
@@ -337,6 +338,19 @@ int sys32_ptrace(long request, long pid, unsigned long addr, unsigned long data)
                break;
        }
+        case PTRACE_GET_DEBUGREG: {
+                ret = -EINVAL;
+                /* We only support one DABR and no IABRS at the moment */
+                if (addr > 0)
+                        break;
+                ret = put_user(child->thread.dabr, (u32 __user *)data);
+                break;
+        }
+        case PTRACE_SET_DEBUGREG:
+                ret = ptrace_set_debugreg(child, addr, data);
+                break;
        case PTRACE_DETACH:
                ret = ptrace_detach(child, data);
                break;
@@ -405,9 +419,23 @@ int sys32_ptrace(long request, long pid, unsigned long addr, unsigned long data)
                break;
        }
-       case PTRACE_GETEVENTMSG:
+        case PTRACE_GETEVENTMSG:
-                ret = put_user(child->ptrace_message, (unsigned int __user *) data);
+                ret = put_user(child->ptrace_message, (unsigned int __user *) data);
-                break;
+                break;
+#ifdef CONFIG_ALTIVEC
+        case PTRACE_GETVRREGS:
+                /* Get the child altivec register state. */
+                flush_altivec_to_thread(child);
+                ret = get_vrregs((unsigned long __user *)data, child);
+                break;
+        case PTRACE_SETVRREGS:
+                /* Set the child altivec register state. */
+                flush_altivec_to_thread(child);
+                ret = set_vrregs(child, (unsigned long __user *)data);
+                break;
+#endif
        default:
                ret = ptrace_request(child, request, addr, data);
diff --git a/arch/ppc64/kernel/ras.c b/arch/ppc64/kernel/ras.c
index 3c00f7bfc1b5..41b97dc9cc0a 100644
--- a/arch/ppc64/kernel/ras.c
+++ b/arch/ppc64/kernel/ras.c
@@ -59,8 +59,6 @@ char mce_data_buf[RTAS_ERROR_LOG_MAX]
 /* This is true if we are using the firmware NMI handler (typically LPAR) */
 extern int fwnmi_active;
-extern void _exception(int signr, struct pt_regs *regs, int code, unsigned long addr);
 static int ras_get_sensor_state_token;
 static int ras_check_exception_token;
diff --git a/arch/ppc64/kernel/rtas_pci.c b/arch/ppc64/kernel/rtas_pci.c
index 1048817befb8..4a9719b48abe 100644
--- a/arch/ppc64/kernel/rtas_pci.c
+++ b/arch/ppc64/kernel/rtas_pci.c
@@ -48,7 +48,7 @@ static int write_pci_config;
 static int ibm_read_pci_config;
 static int ibm_write_pci_config;
-static int config_access_valid(struct device_node *dn, int where)
+static int config_access_valid(struct pci_dn *dn, int where)
 {
        if (where < 256)
                return 1;
@@ -58,20 +58,37 @@ static int config_access_valid(struct device_node *dn, int where)
        return 0;
 }
+static int of_device_available(struct device_node * dn)
+{
+        char * status;
+        status = get_property(dn, "status", NULL);
+        if (!status)
+                return 1;
+        if (!strcmp(status, "okay"))
+                return 1;
+        return 0;
+}
 static int rtas_read_config(struct device_node *dn, int where, int size, u32 *val)
 {
        int returnval = -1;
        unsigned long buid, addr;
        int ret;
+        struct pci_dn *pdn;
-        if (!dn)
+        if (!dn || !dn->data)
                return PCIBIOS_DEVICE_NOT_FOUND;
-        if (!config_access_valid(dn, where))
+        pdn = dn->data;
+        if (!config_access_valid(pdn, where))
                return PCIBIOS_BAD_REGISTER_NUMBER;
-        addr = ((where & 0xf00) << 20) | (dn->busno << 16) |
+        addr = ((where & 0xf00) << 20) | (pdn->busno << 16) |
-                (dn->devfn << 8) | (where & 0xff);
+                (pdn->devfn << 8) | (where & 0xff);
-        buid = dn->phb->buid;
+        buid = pdn->phb->buid;
        if (buid) {
                ret = rtas_call(ibm_read_pci_config, 4, 2, &returnval,
                                addr, buid >> 32, buid & 0xffffffff, size);
@@ -83,8 +100,8 @@ static int rtas_read_config(struct device_node *dn, int where, int size, u32 *va
        if (ret)
                return PCIBIOS_DEVICE_NOT_FOUND;
-        if (returnval == EEH_IO_ERROR_VALUE(size)
+        if (returnval == EEH_IO_ERROR_VALUE(size) &&
-            && eeh_dn_check_failure (dn, NULL))
+            eeh_dn_check_failure (dn, NULL))
                return PCIBIOS_DEVICE_NOT_FOUND;
        return PCIBIOS_SUCCESSFUL;
@@ -103,24 +120,28 @@ static int rtas_pci_read_config(struct pci_bus *bus,
        /* Search only direct children of the bus */
        for (dn = busdn->child; dn; dn = dn->sibling)
-                if (dn->devfn == devfn)
+                if (dn->data && PCI_DN(dn)->devfn == devfn
+                    && of_device_available(dn))
                        return rtas_read_config(dn, where, size, val);
        return PCIBIOS_DEVICE_NOT_FOUND;
 }
-static int rtas_write_config(struct device_node *dn, int where, int size, u32 val)
+int rtas_write_config(struct device_node *dn, int where, int size, u32 val)
 {
        unsigned long buid, addr;
        int ret;
+        struct pci_dn *pdn;
-        if (!dn)
+        if (!dn || !dn->data)
                return PCIBIOS_DEVICE_NOT_FOUND;
-        if (!config_access_valid(dn, where))
+        pdn = dn->data;
+        if (!config_access_valid(pdn, where))
                return PCIBIOS_BAD_REGISTER_NUMBER;
-        addr = ((where & 0xf00) << 20) | (dn->busno << 16) |
+        addr = ((where & 0xf00) << 20) | (pdn->busno << 16) |
-                (dn->devfn << 8) | (where & 0xff);
+                (pdn->devfn << 8) | (where & 0xff);
-        buid = dn->phb->buid;
+        buid = pdn->phb->buid;
        if (buid) {
                ret = rtas_call(ibm_write_pci_config, 5, 1, NULL, addr, buid >> 32, buid & 0xffffffff, size, (ulong) val);
        } else {
@@ -146,7 +167,8 @@ static int rtas_pci_write_config(struct pci_bus *bus,
        /* Search only direct children of the bus */
        for (dn = busdn->child; dn; dn = dn->sibling)
-                if (dn->devfn == devfn)
+                if (dn->data && PCI_DN(dn)->devfn == devfn
+                    && of_device_available(dn))
                        return rtas_write_config(dn, where, size, val);
        return PCIBIOS_DEVICE_NOT_FOUND;
 }
diff --git a/arch/ppc64/kernel/rtasd.c b/arch/ppc64/kernel/rtasd.c
index b0c3b829fe47..e26b0420b6dd 100644
--- a/arch/ppc64/kernel/rtasd.c
+++ b/arch/ppc64/kernel/rtasd.c
@@ -19,6 +19,7 @@
 #include <linux/vmalloc.h>
 #include <linux/spinlock.h>
 #include <linux/cpu.h>
+#include <linux/delay.h>
 #include <asm/uaccess.h>
 #include <asm/io.h>
@@ -412,8 +413,7 @@ static void do_event_scan_all_cpus(long delay)
                /* Drop hotplug lock, and sleep for the specified delay */
                unlock_cpu_hotplug();
-                set_current_state(TASK_INTERRUPTIBLE);
+                msleep_interruptible(delay);
-                schedule_timeout(delay);
                lock_cpu_hotplug();
                cpu = next_cpu(cpu, cpu_online_map);
@@ -442,7 +442,7 @@ static int rtasd(void *unused)
        printk(KERN_INFO "RTAS daemon started\n");
-        DEBUG("will sleep for %d jiffies\n", (HZ*60/rtas_event_scan_rate) / 2);
+        DEBUG("will sleep for %d milliseconds\n", (30000/rtas_event_scan_rate));
        /* See if we have any error stored in NVRAM */
        memset(logdata, 0, rtas_error_log_max);
@@ -459,7 +459,7 @@ static int rtasd(void *unused)
        }
        /* First pass. */
-        do_event_scan_all_cpus(HZ);
+        do_event_scan_all_cpus(1000);
        if (surveillance_timeout != -1) {
                DEBUG("enabling surveillance\n");
@@ -471,7 +471,7 @@ static int rtasd(void *unused)
         * machines have problems if we call event-scan too
         * quickly. */
        for (;;)
-                do_event_scan_all_cpus((HZ*60/rtas_event_scan_rate) / 2);
+                do_event_scan_all_cpus(30000/rtas_event_scan_rate);
 error:
        /* Should delete proc entries */
diff --git a/arch/ppc64/kernel/rtc.c b/arch/ppc64/kernel/rtc.c
index d729fefa0df5..6ff52bc61325 100644
--- a/arch/ppc64/kernel/rtc.c
+++ b/arch/ppc64/kernel/rtc.c
@@ -35,6 +35,7 @@
 #include <linux/spinlock.h>
 #include <linux/bcd.h>
 #include <linux/interrupt.h>
+#include <linux/delay.h>
 #include <asm/io.h>
 #include <asm/uaccess.h>
@@ -351,8 +352,7 @@ void rtas_get_rtc_time(struct rtc_time *rtc_tm)
                                return; /* delay not allowed */
                        }
                        wait_time = rtas_extended_busy_delay_time(error);
-                        set_current_state(TASK_INTERRUPTIBLE);
+                        msleep_interruptible(wait_time);
-                        schedule_timeout(wait_time);
                        error = RTAS_CLOCK_BUSY;
                }
        } while (error == RTAS_CLOCK_BUSY && (__get_tb() < max_wait_tb));
@@ -386,8 +386,7 @@ int rtas_set_rtc_time(struct rtc_time *tm)
                        if (in_interrupt())
                                return 1;       /* probably decrementer */
                        wait_time = rtas_extended_busy_delay_time(error);
-                        set_current_state(TASK_INTERRUPTIBLE);
+                        msleep_interruptible(wait_time);
-                        schedule_timeout(wait_time);
                        error = RTAS_CLOCK_BUSY;
                }
        } while (error == RTAS_CLOCK_BUSY && (__get_tb() < max_wait_tb));
diff --git a/arch/ppc64/kernel/scanlog.c b/arch/ppc64/kernel/scanlog.c
index 4d70736619c7..215bf8900304 100644
--- a/arch/ppc64/kernel/scanlog.c
+++ b/arch/ppc64/kernel/scanlog.c
@@ -25,6 +25,7 @@
 #include <linux/errno.h>
 #include <linux/proc_fs.h>
 #include <linux/init.h>
+#include <linux/delay.h>
 #include <asm/uaccess.h>
 #include <asm/rtas.h>
 #include <asm/prom.h>
@@ -77,7 +78,7 @@ static ssize_t scanlog_read(struct file *file, char __user *buf,
                return -EFAULT;
        for (;;) {
-                wait_time = HZ/2;       /* default wait if no data */
+                wait_time = 500;        /* default wait if no data */
                spin_lock(&rtas_data_buf_lock);
                memcpy(rtas_data_buf, data, RTAS_DATA_BUF_SIZE);
                status = rtas_call(ibm_scan_log_dump, 2, 1, NULL,
@@ -107,24 +108,14 @@ static ssize_t scanlog_read(struct file *file, char __user *buf,
                        break;
                    default:
                        if (status > 9900 && status <= 9905) {
-                                /* No data.  RTAS is hinting at a delay required
+                                wait_time = rtas_extended_busy_delay_time(status);
-                                 * between 1-100000 milliseconds
-                                 */
-                                int ms = 1;
-                                for (; status > 9900; status--)
-                                        ms = ms * 10;
-                                /* Use microseconds for reasonable accuracy */
-                                ms *= 1000;
-                                wait_time = ms / (1000000/HZ); /* round down is fine */
-                                /* Fall through to sleep */
                        } else {
                                printk(KERN_ERR "scanlog: unknown error from rtas: %d\n", status);
                                return -EIO;
                        }
                }
                /* Apparently no data yet.  Wait and try again. */
-                set_current_state(TASK_INTERRUPTIBLE);
+                msleep_interruptible(wait_time);
-                schedule_timeout(wait_time);
        }
        /*NOTREACHED*/
 }
diff --git a/arch/ppc64/kernel/setup.c b/arch/ppc64/kernel/setup.c
index e9c24d2dbd91..5ac48bd64891 100644
--- a/arch/ppc64/kernel/setup.c
+++ b/arch/ppc64/kernel/setup.c
@@ -25,7 +25,7 @@
 #include <linux/seq_file.h>
 #include <linux/ioport.h>
 #include <linux/console.h>
-#include <linux/version.h>
+#include <linux/utsname.h>
 #include <linux/tty.h>
 #include <linux/root_dev.h>
 #include <linux/notifier.h>
@@ -89,7 +89,7 @@ extern void udbg_init_maple_realmode(void);
 #define EARLY_DEBUG_INIT() udbg_init_maple_realmode()
 #define EARLY_DEBUG_INIT() udbg_init_pmac_realmode()
 #define EARLY_DEBUG_INIT()                                              \
-        do { ppc_md.udbg_putc = call_rtas_display_status_delay; } while(0)
+        do { udbg_putc = call_rtas_display_status_delay; } while(0)
 #endif
 /* extern void *stab; */
@@ -108,7 +108,6 @@ int boot_cpuid = 0;
 int boot_cpuid_phys = 0;
 dev_t boot_dev;
 u64 ppc64_pft_size;
-u64 ppc64_debug_switch;
 struct ppc64_caches ppc64_caches;
 EXPORT_SYMBOL_GPL(ppc64_caches);
@@ -154,34 +153,6 @@ struct screen_info screen_info = {
        .orig_video_points = 16
 };
-/*
- * Initialize the PPCDBG state.  Called before relocation has been enabled.
- */
-void __init ppcdbg_initialize(void)
-{
-        ppc64_debug_switch = PPC_DEBUG_DEFAULT; /* | PPCDBG_BUSWALK | */
-        /* PPCDBG_PHBINIT | PPCDBG_MM | PPCDBG_MMINIT | PPCDBG_TCEINIT | PPCDBG_TCE */;
-}
-/*
- * Early boot console based on udbg
- */
-static struct console udbg_console = {
-        .name   = "udbg",
-        .write  = udbg_console_write,
-        .flags  = CON_PRINTBUFFER,
-        .index  = -1,
-};
-static int early_console_initialized;
-void __init disable_early_printk(void)
-{
-        if (!early_console_initialized)
-                return;
-        unregister_console(&udbg_console);
-        early_console_initialized = 0;
-}
 #if defined(CONFIG_PPC_MULTIPLATFORM) && defined(CONFIG_SMP)
 static int smt_enabled_cmdline;
@@ -425,12 +396,6 @@ void __init early_setup(unsigned long dt_ptr)
        }
        ppc_md = **mach;
-        /* our udbg callbacks got overriden by the above, let's put them
-         * back in. Ultimately, I want those things to be split from the
-         * main ppc_md
-         */
-        EARLY_DEBUG_INIT();
        DBG("Found, Initializing memory management...\n");
        /*
@@ -536,15 +501,19 @@ static void __init check_for_initrd(void)
        DBG(" -> check_for_initrd()\n");
-        prop = (u64 *)get_property(of_chosen, "linux,initrd-start", NULL);
+        if (of_chosen) {
-        if (prop != NULL) {
+                prop = (u64 *)get_property(of_chosen,
-                initrd_start = (unsigned long)__va(*prop);
+                                "linux,initrd-start", NULL);
-                prop = (u64 *)get_property(of_chosen, "linux,initrd-end", NULL);
                if (prop != NULL) {
-                        initrd_end = (unsigned long)__va(*prop);
+                        initrd_start = (unsigned long)__va(*prop);
-                        initrd_below_start_ok = 1;
+                        prop = (u64 *)get_property(of_chosen,
-                } else
+                                        "linux,initrd-end", NULL);
-                        initrd_start = 0;
+                        if (prop != NULL) {
+                                initrd_end = (unsigned long)__va(*prop);
+                                initrd_below_start_ok = 1;
+                        } else
+                                initrd_start = 0;
+                }
        }
        /* If we were passed an initrd, set the ROOT_DEV properly if the values
@@ -627,13 +596,12 @@ void __init setup_system(void)
         * Initialize xmon
         */
 #ifdef CONFIG_XMON_DEFAULT
-        xmon_init();
+        xmon_init(1);
 #endif
        /*
         * Register early console
         */
-        early_console_initialized = 1;
+        register_early_udbg_console();
-        register_console(&udbg_console);
        /* Save unparsed command line copy for /proc/cmdline */
        strlcpy(saved_command_line, cmd_line, COMMAND_LINE_SIZE);
@@ -653,7 +621,7 @@ void __init setup_system(void)
        smp_release_cpus();
 #endif /* defined(CONFIG_SMP) && !defined(CONFIG_PPC_ISERIES) */
-        printk("Starting Linux PPC64 %s\n", UTS_RELEASE);
+        printk("Starting Linux PPC64 %s\n", system_utsname.version);
        printk("-----------------------------------------------------\n");
        printk("ppc64_pft_size                = 0x%lx\n", ppc64_pft_size);
@@ -1096,8 +1064,6 @@ void __init setup_arch(char **cmdline_p)
 #define PPC64_LINUX_FUNCTION 0x0f000000
 #define PPC64_IPL_MESSAGE 0xc0000000
 #define PPC64_TERM_MESSAGE 0xb0000000
-#define PPC64_ATTN_MESSAGE 0xa0000000
-#define PPC64_DUMP_MESSAGE 0xd0000000
 static void ppc64_do_msg(unsigned int src, const char *msg)
 {
@@ -1125,20 +1091,6 @@ void ppc64_terminate_msg(unsigned int src, const char *msg)
        printk("[terminate]%04x %s\n", src, msg);
 }
-/* Print something that needs attention (device error, etc) */
-void ppc64_attention_msg(unsigned int src, const char *msg)
-{
-        ppc64_do_msg(PPC64_LINUX_FUNCTION|PPC64_ATTN_MESSAGE|src, msg);
-        printk("[attention]%04x %s\n", src, msg);
-}
-/* Print a dump progress message. */
-void ppc64_dump_msg(unsigned int src, const char *msg)
-{
-        ppc64_do_msg(PPC64_LINUX_FUNCTION|PPC64_DUMP_MESSAGE|src, msg);
-        printk("[dump]%04x %s\n", src, msg);
-}
 /* This should only be called on processor 0 during calibrate decr */
 void __init setup_default_decr(void)
 {
@@ -1315,7 +1267,7 @@ void __init generic_find_legacy_serial_ports(u64 *physport,
 static struct platform_device serial_device = {
        .name   = "serial8250",
-        .id     = 0,
+        .id     = PLAT8250_DEV_PLATFORM,
        .dev    = {
                .platform_data = serial_ports,
        },
@@ -1343,11 +1295,13 @@ static int __init early_xmon(char *p)
        /* ensure xmon is enabled */
        if (p) {
                if (strncmp(p, "on", 2) == 0)
-                        xmon_init();
+                        xmon_init(1);
+                if (strncmp(p, "off", 3) == 0)
+                        xmon_init(0);
                if (strncmp(p, "early", 5) != 0)
                        return 0;
        }
-        xmon_init();
+        xmon_init(1);
        debugger(NULL);
        return 0;
diff --git a/arch/ppc64/kernel/signal.c b/arch/ppc64/kernel/signal.c
index bf782276984c..347112cca3c0 100644
--- a/arch/ppc64/kernel/signal.c
+++ b/arch/ppc64/kernel/signal.c
@@ -481,10 +481,11 @@ static int handle_signal(unsigned long sig, struct k_sigaction *ka,
        /* Set up Signal Frame */
        ret = setup_rt_frame(sig, ka, info, oldset, regs);
-        if (ret && !(ka->sa.sa_flags & SA_NODEFER)) {
+        if (ret) {
                spin_lock_irq(&current->sighand->siglock);
                sigorsets(&current->blocked, &current->blocked, &ka->sa.sa_mask);
-                sigaddset(&current->blocked,sig);
+                if (!(ka->sa.sa_flags & SA_NODEFER))
+                        sigaddset(&current->blocked,sig);
                recalc_sigpending();
                spin_unlock_irq(&current->sighand->siglock);
        }
@@ -549,6 +550,15 @@ int do_signal(sigset_t *oldset, struct pt_regs *regs)
                /* Whee!  Actually deliver the signal.  */
                if (TRAP(regs) == 0x0C00)
                        syscall_restart(regs, &ka);
+                /*
+                 * Reenable the DABR before delivering the signal to
+                 * user space. The DABR will have been cleared if it
+                 * triggered inside the kernel.
+                 */
+                if (current->thread.dabr)
+                        set_dabr(current->thread.dabr);
                return handle_signal(signr, &ka, &info, oldset, regs);
        }
diff --git a/arch/ppc64/kernel/signal32.c b/arch/ppc64/kernel/signal32.c
index 3c2fa5c284c0..a8b7a5a56bb4 100644
--- a/arch/ppc64/kernel/signal32.c
+++ b/arch/ppc64/kernel/signal32.c
@@ -970,17 +970,26 @@ int do_signal32(sigset_t *oldset, struct pt_regs *regs)
                newsp = regs->gpr[1];
        newsp &= ~0xfUL;
+        /*
+         * Reenable the DABR before delivering the signal to
+         * user space. The DABR will have been cleared if it
+         * triggered inside the kernel.
+         */
+        if (current->thread.dabr)
+                set_dabr(current->thread.dabr);
        /* Whee!  Actually deliver the signal.  */
        if (ka.sa.sa_flags & SA_SIGINFO)
                ret = handle_rt_signal32(signr, &ka, &info, oldset, regs, newsp);
        else
                ret = handle_signal32(signr, &ka, &info, oldset, regs, newsp);
-        if (ret && !(ka.sa.sa_flags & SA_NODEFER)) {
+        if (ret) {
                spin_lock_irq(&current->sighand->siglock);
                sigorsets(&current->blocked, &current->blocked,
                          &ka.sa.sa_mask);
-                sigaddset(&current->blocked, signr);
+                if (!(ka.sa.sa_flags & SA_NODEFER))
+                        sigaddset(&current->blocked, signr);
                recalc_sigpending();
                spin_unlock_irq(&current->sighand->siglock);
        }
diff --git a/arch/ppc64/kernel/sys_ppc32.c b/arch/ppc64/kernel/sys_ppc32.c
index 206619080e66..e93c13458910 100644
--- a/arch/ppc64/kernel/sys_ppc32.c
+++ b/arch/ppc64/kernel/sys_ppc32.c
@@ -708,62 +708,9 @@ asmlinkage int sys32_pciconfig_write(u32 bus, u32 dfn, u32 off, u32 len, u32 ubu
                                   compat_ptr(ubuf));
 }
-#define IOBASE_BRIDGE_NUMBER    0
-#define IOBASE_MEMORY           1
-#define IOBASE_IO               2
-#define IOBASE_ISA_IO           3
-#define IOBASE_ISA_MEM          4
 asmlinkage int sys32_pciconfig_iobase(u32 which, u32 in_bus, u32 in_devfn)
 {
-#ifdef CONFIG_PCI
+        return sys_pciconfig_iobase(which, in_bus, in_devfn);
-        struct pci_controller* hose;
-        struct list_head *ln;
-        struct pci_bus *bus = NULL;
-        struct device_node *hose_node;
-        /* Argh ! Please forgive me for that hack, but that's the
-         * simplest way to get existing XFree to not lockup on some
-         * G5 machines... So when something asks for bus 0 io base
-         * (bus 0 is HT root), we return the AGP one instead.
-         */
-#ifdef CONFIG_PPC_PMAC
-        if (systemcfg->platform == PLATFORM_POWERMAC &&
-            machine_is_compatible("MacRISC4"))
-                if (in_bus == 0)
-                        in_bus = 0xf0;
-#endif /* CONFIG_PPC_PMAC */
-        /* That syscall isn't quite compatible with PCI domains, but it's
-         * used on pre-domains setup. We return the first match
-         */
-        for (ln = pci_root_buses.next; ln != &pci_root_buses; ln = ln->next) {
-                bus = pci_bus_b(ln);
-                if (in_bus >= bus->number && in_bus < (bus->number + bus->subordinate))
-                        break;
-                bus = NULL;
-        }
-        if (bus == NULL || bus->sysdata == NULL)
-                return -ENODEV;
-        hose_node = (struct device_node *)bus->sysdata;
-        hose = hose_node->phb;
-        switch (which) {
-        case IOBASE_BRIDGE_NUMBER:
-                return (long)hose->first_busno;
-        case IOBASE_MEMORY:
-                return (long)hose->pci_mem_offset;
-        case IOBASE_IO:
-                return (long)hose->io_base_phys;
-        case IOBASE_ISA_IO:
-                return (long)isa_io_base;
-        case IOBASE_ISA_MEM:
-                return -EINVAL;
-        }
-#endif  /* CONFIG_PCI */
-        return -EOPNOTSUPP;
 }
@@ -867,37 +814,6 @@ off_t ppc32_lseek(unsigned int fd, u32 offset, unsigned int origin)
        return sys_lseek(fd, (int)offset, origin);
 }
-/*
- * This is just a version for 32-bit applications which does
- * not force O_LARGEFILE on.
- */
-asmlinkage long sys32_open(const char __user * filename, int flags, int mode)
-{
-        char * tmp;
-        int fd, error;
-        tmp = getname(filename);
-        fd = PTR_ERR(tmp);
-        if (!IS_ERR(tmp)) {
-                fd = get_unused_fd();
-                if (fd >= 0) {
-                        struct file * f = filp_open(tmp, flags, mode);
-                        error = PTR_ERR(f);
-                        if (IS_ERR(f))
-                                goto out_error;
-                        fd_install(fd, f);
-                }
-out:
-                putname(tmp);
-        }
-        return fd;
-out_error:
-        put_unused_fd(fd);
-        fd = error;
-        goto out;
-}
 /* Note: it is necessary to treat bufsiz as an unsigned int,
 * with the corresponding cast to a signed int to insure that the 
 * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode)
diff --git a/arch/ppc64/kernel/syscalls.c b/arch/ppc64/kernel/syscalls.c
index a8cbb202b8cd..05f16633bd2c 100644
--- a/arch/ppc64/kernel/syscalls.c
+++ b/arch/ppc64/kernel/syscalls.c
@@ -46,10 +46,6 @@
 extern unsigned long wall_jiffies;
-void
-check_bugs(void)
-{
-}
 /*
 * sys_ipc() is the de-multiplexer for the SysV IPC calls..
diff --git a/arch/ppc64/kernel/sysfs.c b/arch/ppc64/kernel/sysfs.c
index 02b8ac4e0168..6654b350979c 100644
--- a/arch/ppc64/kernel/sysfs.c
+++ b/arch/ppc64/kernel/sysfs.c
@@ -13,6 +13,7 @@
 #include <asm/current.h>
 #include <asm/processor.h>
 #include <asm/cputable.h>
+#include <asm/firmware.h>
 #include <asm/hvcall.h>
 #include <asm/prom.h>
 #include <asm/systemcfg.h>
@@ -100,6 +101,8 @@ static int __init setup_smt_snooze_delay(char *str)
 }
 __setup("smt-snooze-delay=", setup_smt_snooze_delay);
+#endif /* CONFIG_PPC_MULTIPLATFORM */
 /*
 * Enabling PMCs will slow partition context switch times so we only do
 * it the first time we write to the PMCs.
@@ -109,65 +112,15 @@ static DEFINE_PER_CPU(char, pmcs_enabled);
 void ppc64_enable_pmcs(void)
 {
-        unsigned long hid0;
-#ifdef CONFIG_PPC_PSERIES
-        unsigned long set, reset;
-#endif /* CONFIG_PPC_PSERIES */
        /* Only need to enable them once */
        if (__get_cpu_var(pmcs_enabled))
                return;
        __get_cpu_var(pmcs_enabled) = 1;
-        switch (systemcfg->platform) {
+        if (ppc_md.enable_pmcs)
-        case PLATFORM_PSERIES:
+                ppc_md.enable_pmcs();
-        case PLATFORM_POWERMAC:
-                hid0 = mfspr(HID0);
-                hid0 |= 1UL << (63 - 20);
-                /* POWER4 requires the following sequence */
-                asm volatile(
-                             "sync\n"
-                             "mtspr     %1, %0\n"
-                             "mfspr     %0, %1\n"
-                             "mfspr     %0, %1\n"
-                             "mfspr     %0, %1\n"
-                             "mfspr     %0, %1\n"
-                             "mfspr     %0, %1\n"
-                             "mfspr     %0, %1\n"
-                             "isync" : "=&r" (hid0) : "i" (HID0), "0" (hid0):
-                             "memory");
-                break;
-#ifdef CONFIG_PPC_PSERIES
-        case PLATFORM_PSERIES_LPAR:
-                set = 1UL << 63;
-                reset = 0;
-                plpar_hcall_norets(H_PERFMON, set, reset);
-                break;
-#endif /* CONFIG_PPC_PSERIES */
-        default:
-                break;
-        }
-#ifdef CONFIG_PPC_PSERIES
-        /* instruct hypervisor to maintain PMCs */
-        if (cur_cpu_spec->firmware_features & FW_FEATURE_SPLPAR)
-                get_paca()->lppaca.pmcregs_in_use = 1;
-#endif /* CONFIG_PPC_PSERIES */
 }
-#else
-/* PMC stuff */
-void ppc64_enable_pmcs(void)
-{
-        /* XXX Implement for iseries */
-}
-#endif /* CONFIG_PPC_MULTIPLATFORM */
 EXPORT_SYMBOL(ppc64_enable_pmcs);
 /* XXX convert to rusty's on_one_cpu */
@@ -262,18 +215,23 @@ static void register_cpu_online(unsigned int cpu)
        if (cpu_has_feature(CPU_FTR_MMCRA))
                sysdev_create_file(s, &attr_mmcra);
-        sysdev_create_file(s, &attr_pmc1);
+        if (cur_cpu_spec->num_pmcs >= 1)
-        sysdev_create_file(s, &attr_pmc2);
+                sysdev_create_file(s, &attr_pmc1);
-        sysdev_create_file(s, &attr_pmc3);
+        if (cur_cpu_spec->num_pmcs >= 2)
-        sysdev_create_file(s, &attr_pmc4);
+                sysdev_create_file(s, &attr_pmc2);
-        sysdev_create_file(s, &attr_pmc5);
+        if (cur_cpu_spec->num_pmcs >= 3)
-        sysdev_create_file(s, &attr_pmc6);
+                sysdev_create_file(s, &attr_pmc3);
+        if (cur_cpu_spec->num_pmcs >= 4)
-        if (cpu_has_feature(CPU_FTR_PMC8)) {
+                sysdev_create_file(s, &attr_pmc4);
+        if (cur_cpu_spec->num_pmcs >= 5)
+                sysdev_create_file(s, &attr_pmc5);
+        if (cur_cpu_spec->num_pmcs >= 6)
+                sysdev_create_file(s, &attr_pmc6);
+        if (cur_cpu_spec->num_pmcs >= 7)
                sysdev_create_file(s, &attr_pmc7);
+        if (cur_cpu_spec->num_pmcs >= 8)
                sysdev_create_file(s, &attr_pmc8);
-        }
+  
        if (cpu_has_feature(CPU_FTR_SMT))
                sysdev_create_file(s, &attr_purr);
 }
@@ -299,17 +257,22 @@ static void unregister_cpu_online(unsigned int cpu)
        if (cpu_has_feature(CPU_FTR_MMCRA))
                sysdev_remove_file(s, &attr_mmcra);
-        sysdev_remove_file(s, &attr_pmc1);
+        if (cur_cpu_spec->num_pmcs >= 1)
-        sysdev_remove_file(s, &attr_pmc2);
+                sysdev_remove_file(s, &attr_pmc1);
-        sysdev_remove_file(s, &attr_pmc3);
+        if (cur_cpu_spec->num_pmcs >= 2)
-        sysdev_remove_file(s, &attr_pmc4);
+                sysdev_remove_file(s, &attr_pmc2);
-        sysdev_remove_file(s, &attr_pmc5);
+        if (cur_cpu_spec->num_pmcs >= 3)
-        sysdev_remove_file(s, &attr_pmc6);
+                sysdev_remove_file(s, &attr_pmc3);
+        if (cur_cpu_spec->num_pmcs >= 4)
-        if (cpu_has_feature(CPU_FTR_PMC8)) {
+                sysdev_remove_file(s, &attr_pmc4);
+        if (cur_cpu_spec->num_pmcs >= 5)
+                sysdev_remove_file(s, &attr_pmc5);
+        if (cur_cpu_spec->num_pmcs >= 6)
+                sysdev_remove_file(s, &attr_pmc6);
+        if (cur_cpu_spec->num_pmcs >= 7)
                sysdev_remove_file(s, &attr_pmc7);
+        if (cur_cpu_spec->num_pmcs >= 8)
                sysdev_remove_file(s, &attr_pmc8);
-        }
        if (cpu_has_feature(CPU_FTR_SMT))
                sysdev_remove_file(s, &attr_purr);
diff --git a/arch/ppc64/kernel/time.c b/arch/ppc64/kernel/time.c
index 909462e1adea..9939c206afa4 100644
--- a/arch/ppc64/kernel/time.c
+++ b/arch/ppc64/kernel/time.c
@@ -51,7 +51,6 @@
 #include <linux/cpu.h>
 #include <linux/security.h>
-#include <asm/segment.h>
 #include <asm/io.h>
 #include <asm/processor.h>
 #include <asm/nvram.h>
@@ -67,6 +66,7 @@
 #include <asm/prom.h>
 #include <asm/sections.h>
 #include <asm/systemcfg.h>
+#include <asm/firmware.h>
 u64 jiffies_64 __cacheline_aligned_in_smp = INITIAL_JIFFIES;
@@ -128,7 +128,7 @@ static __inline__ void timer_check_rtc(void)
         * We should have an rtc call that only sets the minutes and
         * seconds like on Intel to avoid problems with non UTC clocks.
         */
-        if ( (time_status & STA_UNSYNC) == 0 &&
+        if (ntp_synced() &&
             xtime.tv_sec - last_rtc_update >= 659 &&
             abs((xtime.tv_nsec/1000) - (1000000-1000000/HZ)) < 500000/HZ &&
             jiffies - wall_jiffies == 1) {
@@ -370,13 +370,11 @@ int timer_interrupt(struct pt_regs * regs)
                process_hvlpevents(regs);
 #endif
-/* collect purr register values often, for accurate calculations */
+        /* collect purr register values often, for accurate calculations */
-#if defined(CONFIG_PPC_PSERIES)
+        if (firmware_has_feature(FW_FEATURE_SPLPAR)) {
-        if (cur_cpu_spec->firmware_features & FW_FEATURE_SPLPAR) {
                struct cpu_usage *cu = &__get_cpu_var(cpu_usage_array);
                cu->current_tb = mfspr(SPRN_PURR);
        }
-#endif
        irq_exit();
@@ -437,10 +435,7 @@ int do_settimeofday(struct timespec *tv)
         */
        last_rtc_update = new_sec - 658;
-        time_adjust = 0;                /* stop active adjtime() */
+        ntp_clear();
-        time_status |= STA_UNSYNC;
-        time_maxerror = NTP_PHASE_LIMIT;
-        time_esterror = NTP_PHASE_LIMIT;
        delta_xsec = mulhdu( (tb_last_stamp-do_gtod.varp->tb_orig_stamp),
                             do_gtod.varp->tb_to_xs );
diff --git a/arch/ppc64/kernel/traps.c b/arch/ppc64/kernel/traps.c
index a8d5e83ee89f..7467ae508e6e 100644
--- a/arch/ppc64/kernel/traps.c
+++ b/arch/ppc64/kernel/traps.c
@@ -30,6 +30,7 @@
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/delay.h>
+#include <linux/kprobes.h>
 #include <asm/kdebug.h>
 #include <asm/pgtable.h>
@@ -220,7 +221,7 @@ void instruction_breakpoint_exception(struct pt_regs *regs)
        _exception(SIGTRAP, regs, TRAP_BRKPT, regs->nip);
 }
-void single_step_exception(struct pt_regs *regs)
+void __kprobes single_step_exception(struct pt_regs *regs)
 {
        regs->msr &= ~MSR_SE;  /* Turn off 'trace' bit */
@@ -398,7 +399,7 @@ check_bug_trap(struct pt_regs *regs)
        return 0;
 }
-void program_check_exception(struct pt_regs *regs)
+void __kprobes program_check_exception(struct pt_regs *regs)
 {
        if (debugger_fault_handler(regs))
                return;
diff --git a/arch/ppc64/kernel/u3_iommu.c b/arch/ppc64/kernel/u3_iommu.c
index b6e3bca4102d..41ea09cb9ac7 100644
--- a/arch/ppc64/kernel/u3_iommu.c
+++ b/arch/ppc64/kernel/u3_iommu.c
@@ -276,7 +276,7 @@ static void iommu_dev_setup_u3(struct pci_dev *dev)
        dn = pci_device_to_OF_node(dev);
        if (dn)
-                dn->iommu_table = &iommu_table_u3;
+                PCI_DN(dn)->iommu_table = &iommu_table_u3;
 }
 static void iommu_bus_setup_u3(struct pci_bus *bus)
@@ -291,7 +291,7 @@ static void iommu_bus_setup_u3(struct pci_bus *bus)
        dn = pci_bus_to_OF_node(bus);
        if (dn)
-                dn->iommu_table = &iommu_table_u3;
+                PCI_DN(dn)->iommu_table = &iommu_table_u3;
 }
 static void iommu_dev_setup_null(struct pci_dev *dev) { }
diff --git a/arch/ppc64/kernel/udbg.c b/arch/ppc64/kernel/udbg.c
index c0da45540f0f..d49c3613c8ec 100644
--- a/arch/ppc64/kernel/udbg.c
+++ b/arch/ppc64/kernel/udbg.c
@@ -1,5 +1,5 @@
 /*
- * NS16550 Serial Port (uart) debugging stuff.
+ * polling mode stateless debugging stuff, originally for NS16550 Serial Ports
 *
 * c 2001 PPC 64 Team, IBM Corp
 *
@@ -13,249 +13,24 @@
 #define WANT_PPCDBG_TAB /* Only defined here */
 #include <linux/config.h>
 #include <linux/types.h>
+#include <linux/sched.h>
+#include <linux/console.h>
 #include <asm/ppcdebug.h>
 #include <asm/processor.h>
-#include <asm/uaccess.h>
-#include <asm/machdep.h>
-#include <asm/io.h>
-#include <asm/prom.h>
-#include <asm/pmac_feature.h>
-extern u8 real_readb(volatile u8 __iomem  *addr);
+void (*udbg_putc)(unsigned char c);
-extern void real_writeb(u8 data, volatile u8 __iomem *addr);
+unsigned char (*udbg_getc)(void);
+int (*udbg_getc_poll)(void);
-struct NS16550 {
-        /* this struct must be packed */
-        unsigned char rbr;  /* 0 */
-        unsigned char ier;  /* 1 */
-        unsigned char fcr;  /* 2 */
-        unsigned char lcr;  /* 3 */
-        unsigned char mcr;  /* 4 */
-        unsigned char lsr;  /* 5 */
-        unsigned char msr;  /* 6 */
-        unsigned char scr;  /* 7 */
-};
-#define thr rbr
-#define iir fcr
-#define dll rbr
-#define dlm ier
-#define dlab lcr
-#define LSR_DR   0x01  /* Data ready */
-#define LSR_OE   0x02  /* Overrun */
-#define LSR_PE   0x04  /* Parity error */
-#define LSR_FE   0x08  /* Framing error */
-#define LSR_BI   0x10  /* Break */
-#define LSR_THRE 0x20  /* Xmit holding register empty */
-#define LSR_TEMT 0x40  /* Xmitter empty */
-#define LSR_ERR  0x80  /* Error */
-static volatile struct NS16550 __iomem *udbg_comport;
-void udbg_init_uart(void __iomem *comport, unsigned int speed)
-{
-        u16 dll = speed ? (115200 / speed) : 12;
-        if (comport) {
-                udbg_comport = (struct NS16550 __iomem *)comport;
-                out_8(&udbg_comport->lcr, 0x00);
-                out_8(&udbg_comport->ier, 0xff);
-                out_8(&udbg_comport->ier, 0x00);
-                out_8(&udbg_comport->lcr, 0x80);        /* Access baud rate */
-                out_8(&udbg_comport->dll, dll & 0xff);  /* 1 = 115200,  2 = 57600,
-                                                           3 = 38400, 12 = 9600 baud */
-                out_8(&udbg_comport->dlm, dll >> 8);    /* dll >> 8 which should be zero
-                                                           for fast rates; */
-                out_8(&udbg_comport->lcr, 0x03);        /* 8 data, 1 stop, no parity */
-                out_8(&udbg_comport->mcr, 0x03);        /* RTS/DTR */
-                out_8(&udbg_comport->fcr ,0x07);        /* Clear & enable FIFOs */
-        }
-}
-#ifdef CONFIG_PPC_PMAC
-#define SCC_TXRDY       4
-#define SCC_RXRDY       1
-static volatile u8 __iomem *sccc;
-static volatile u8 __iomem *sccd;
-static unsigned char scc_inittab[] = {
-    13, 0,              /* set baud rate divisor */
-    12, 0,
-    14, 1,              /* baud rate gen enable, src=rtxc */
-    11, 0x50,           /* clocks = br gen */
-    5,  0xea,           /* tx 8 bits, assert DTR & RTS */
-    4,  0x46,           /* x16 clock, 1 stop */
-    3,  0xc1,           /* rx enable, 8 bits */
-};
-void udbg_init_scc(struct device_node *np)
-{
-        u32 *reg;
-        unsigned long addr;
-        int i, x;
-        if (np == NULL)
-                np = of_find_node_by_name(NULL, "escc");
-        if (np == NULL || np->parent == NULL)
-                return;
-        udbg_printf("found SCC...\n");
-        /* Get address within mac-io ASIC */ 
-        reg = (u32 *)get_property(np, "reg", NULL);
-        if (reg == NULL)
-                return;
-        addr = reg[0];
-        udbg_printf("local addr: %lx\n", addr);
-        /* Get address of mac-io PCI itself */
-        reg = (u32 *)get_property(np->parent, "assigned-addresses", NULL);
-        if (reg == NULL)
-                return;
-        addr += reg[2];
-        udbg_printf("final addr: %lx\n", addr);
-        /* Setup for 57600 8N1 */
-        addr += 0x20;
-        sccc = (volatile u8 * __iomem) ioremap(addr & PAGE_MASK, PAGE_SIZE) ;
-        sccc += addr & ~PAGE_MASK;
-        sccd = sccc + 0x10;
-        udbg_printf("ioremap result sccc: %p\n", sccc);
-        mb();
-        for (i = 20000; i != 0; --i)
-                x = in_8(sccc);
-        out_8(sccc, 0x09);              /* reset A or B side */
-        out_8(sccc, 0xc0);
-        for (i = 0; i < sizeof(scc_inittab); ++i)
-                out_8(sccc, scc_inittab[i]);
-        ppc_md.udbg_putc = udbg_putc;
-        ppc_md.udbg_getc = udbg_getc;
-        ppc_md.udbg_getc_poll = udbg_getc_poll;
-        udbg_puts("Hello World !\n");
-}
-#endif /* CONFIG_PPC_PMAC */
-#ifdef CONFIG_PPC_PMAC
-static void udbg_real_putc(unsigned char c)
-{
-        while ((real_readb(sccc) & SCC_TXRDY) == 0)
-                ;
-        real_writeb(c, sccd);
-        if (c == '\n')
-                udbg_real_putc('\r');
-}
-void udbg_init_pmac_realmode(void)
-{
-        sccc = (volatile u8 __iomem *)0x80013020ul;
-        sccd = (volatile u8 __iomem *)0x80013030ul;
-        ppc_md.udbg_putc = udbg_real_putc;
-        ppc_md.udbg_getc = NULL;
-        ppc_md.udbg_getc_poll = NULL;
-}
-#endif /* CONFIG_PPC_PMAC */
-#ifdef CONFIG_PPC_MAPLE
-void udbg_maple_real_putc(unsigned char c)
-{
-        if (udbg_comport) {
-                while ((real_readb(&udbg_comport->lsr) & LSR_THRE) == 0)
-                        /* wait for idle */;
-                real_writeb(c, &udbg_comport->thr); eieio();
-                if (c == '\n') {
-                        /* Also put a CR.  This is for convenience. */
-                        while ((real_readb(&udbg_comport->lsr) & LSR_THRE) == 0)
-                                /* wait for idle */;
-                        real_writeb('\r', &udbg_comport->thr); eieio();
-                }
-        }
-}
-void udbg_init_maple_realmode(void)
-{
-        udbg_comport = (volatile struct NS16550 __iomem *)0xf40003f8;
-        ppc_md.udbg_putc = udbg_maple_real_putc;
-        ppc_md.udbg_getc = NULL;
-        ppc_md.udbg_getc_poll = NULL;
-}
-#endif /* CONFIG_PPC_MAPLE */
-void udbg_putc(unsigned char c)
-{
-        if (udbg_comport) {
-                while ((in_8(&udbg_comport->lsr) & LSR_THRE) == 0)
-                        /* wait for idle */;
-                out_8(&udbg_comport->thr, c);
-                if (c == '\n') {
-                        /* Also put a CR.  This is for convenience. */
-                        while ((in_8(&udbg_comport->lsr) & LSR_THRE) == 0)
-                                /* wait for idle */; 
-                        out_8(&udbg_comport->thr, '\r');
-                }
-        }
-#ifdef CONFIG_PPC_PMAC
-        else if (sccc) {
-                while ((in_8(sccc) & SCC_TXRDY) == 0)
-                        ;
-                out_8(sccd,  c);                
-                if (c == '\n')
-                        udbg_putc('\r');
-        }
-#endif /* CONFIG_PPC_PMAC */
-}
-int udbg_getc_poll(void)
-{
-        if (udbg_comport) {
-                if ((in_8(&udbg_comport->lsr) & LSR_DR) != 0)
-                        return in_8(&udbg_comport->rbr);
-                else
-                        return -1;
-        }
-#ifdef CONFIG_PPC_PMAC
-        else if (sccc) {
-                if ((in_8(sccc) & SCC_RXRDY) != 0)
-                        return in_8(sccd);
-                else
-                        return -1;
-        }
-#endif /* CONFIG_PPC_PMAC */
-        return -1;
-}
-unsigned char udbg_getc(void)
-{
-        if (udbg_comport) {
-                while ((in_8(&udbg_comport->lsr) & LSR_DR) == 0)
-                        /* wait for char */;
-                return in_8(&udbg_comport->rbr);
-        }
-#ifdef CONFIG_PPC_PMAC
-        else if (sccc) {
-                while ((in_8(sccc) & SCC_RXRDY) == 0)
-                        ;
-                return in_8(sccd);
-        }
-#endif /* CONFIG_PPC_PMAC */
-        return 0;
-}
+/* udbg library, used by xmon et al */
 void udbg_puts(const char *s)
 {
-        if (ppc_md.udbg_putc) {
+        if (udbg_putc) {
                char c;
                if (s && *s != '\0') {
                        while ((c = *s++) != '\0')
-                                ppc_md.udbg_putc(c);
+                                udbg_putc(c);
                }
        }
 #if 0
@@ -270,12 +45,12 @@ int udbg_write(const char *s, int n)
        int remain = n;
        char c;
-        if (!ppc_md.udbg_putc)
+        if (!udbg_putc)
                return 0;
        if (s && *s != '\0') {
                while (((c = *s++) != '\0') && (remain-- > 0)) {
-                        ppc_md.udbg_putc(c);
+                        udbg_putc(c);
                }
        }
@@ -287,12 +62,12 @@ int udbg_read(char *buf, int buflen)
        char c, *p = buf;
        int i;
-        if (!ppc_md.udbg_getc)
+        if (!udbg_getc)
                return 0;
        for (i = 0; i < buflen; ++i) {
                do {
-                        c = ppc_md.udbg_getc();
+                        c = udbg_getc();
                } while (c == 0x11 || c == 0x13);
                if (c == 0)
                        break;
@@ -302,11 +77,6 @@ int udbg_read(char *buf, int buflen)
        return i;
 }
-void udbg_console_write(struct console *con, const char *s, unsigned int n)
-{
-        udbg_write(s, n);
-}
 #define UDBG_BUFSIZE 256
 void udbg_printf(const char *fmt, ...)
 {
@@ -319,6 +89,10 @@ void udbg_printf(const char *fmt, ...)
        va_end(args);
 }
+/* PPCDBG stuff */
+u64 ppc64_debug_switch;
 /* Special print used by PPCDBG() macro */
 void udbg_ppcdbg(unsigned long debug_flags, const char *fmt, ...)
 {
@@ -358,3 +132,49 @@ unsigned long udbg_ifdebug(unsigned long flags)
 {
        return (flags & ppc64_debug_switch);
 }
+/*
+ * Initialize the PPCDBG state.  Called before relocation has been enabled.
+ */
+void __init ppcdbg_initialize(void)
+{
+        ppc64_debug_switch = PPC_DEBUG_DEFAULT; /* | PPCDBG_BUSWALK | */
+        /* PPCDBG_PHBINIT | PPCDBG_MM | PPCDBG_MMINIT | PPCDBG_TCEINIT | PPCDBG_TCE */;
+}
+/*
+ * Early boot console based on udbg
+ */
+static void udbg_console_write(struct console *con, const char *s,
+                unsigned int n)
+{
+        udbg_write(s, n);
+}
+static struct console udbg_console = {
+        .name   = "udbg",
+        .write  = udbg_console_write,
+        .flags  = CON_PRINTBUFFER,
+        .index  = -1,
+};
+static int early_console_initialized;
+void __init disable_early_printk(void)
+{
+        if (!early_console_initialized)
+                return;
+        unregister_console(&udbg_console);
+        early_console_initialized = 0;
+}
+/* called by setup_system */
+void register_early_udbg_console(void)
+{
+        early_console_initialized = 1;
+        register_console(&udbg_console);
+}
+#if 0   /* if you want to use this as a regular output console */
+console_initcall(register_udbg_console);
+#endif
diff --git a/arch/ppc64/kernel/udbg_16550.c b/arch/ppc64/kernel/udbg_16550.c
new file mode 100644
index 000000000000..9313574ab935
--- /dev/null
+++ b/arch/ppc64/kernel/udbg_16550.c
@@ -0,0 +1,123 @@
+/*
+ * udbg for for NS16550 compatable serial ports
+ *
+ * Copyright (C) 2001-2005 PPC 64 Team, IBM Corp
+ *
+ *      This program is free software; you can redistribute it and/or
+ *      modify it under the terms of the GNU General Public License
+ *      as published by the Free Software Foundation; either version
+ *      2 of the License, or (at your option) any later version.
+ */
+#include <linux/config.h>
+#include <linux/types.h>
+#include <asm/udbg.h>
+#include <asm/io.h>
+extern u8 real_readb(volatile u8 __iomem  *addr);
+extern void real_writeb(u8 data, volatile u8 __iomem *addr);
+struct NS16550 {
+        /* this struct must be packed */
+        unsigned char rbr;  /* 0 */
+        unsigned char ier;  /* 1 */
+        unsigned char fcr;  /* 2 */
+        unsigned char lcr;  /* 3 */
+        unsigned char mcr;  /* 4 */
+        unsigned char lsr;  /* 5 */
+        unsigned char msr;  /* 6 */
+        unsigned char scr;  /* 7 */
+};
+#define thr rbr
+#define iir fcr
+#define dll rbr
+#define dlm ier
+#define dlab lcr
+#define LSR_DR   0x01  /* Data ready */
+#define LSR_OE   0x02  /* Overrun */
+#define LSR_PE   0x04  /* Parity error */
+#define LSR_FE   0x08  /* Framing error */
+#define LSR_BI   0x10  /* Break */
+#define LSR_THRE 0x20  /* Xmit holding register empty */
+#define LSR_TEMT 0x40  /* Xmitter empty */
+#define LSR_ERR  0x80  /* Error */
+static volatile struct NS16550 __iomem *udbg_comport;
+static void udbg_550_putc(unsigned char c)
+{
+        if (udbg_comport) {
+                while ((in_8(&udbg_comport->lsr) & LSR_THRE) == 0)
+                        /* wait for idle */;
+                out_8(&udbg_comport->thr, c);
+                if (c == '\n')
+                        udbg_550_putc('\r');
+        }
+}
+static int udbg_550_getc_poll(void)
+{
+        if (udbg_comport) {
+                if ((in_8(&udbg_comport->lsr) & LSR_DR) != 0)
+                        return in_8(&udbg_comport->rbr);
+                else
+                        return -1;
+        }
+        return -1;
+}
+static unsigned char udbg_550_getc(void)
+{
+        if (udbg_comport) {
+                while ((in_8(&udbg_comport->lsr) & LSR_DR) == 0)
+                        /* wait for char */;
+                return in_8(&udbg_comport->rbr);
+        }
+        return 0;
+}
+void udbg_init_uart(void __iomem *comport, unsigned int speed)
+{
+        u16 dll = speed ? (115200 / speed) : 12;
+        if (comport) {
+                udbg_comport = (struct NS16550 __iomem *)comport;
+                out_8(&udbg_comport->lcr, 0x00);
+                out_8(&udbg_comport->ier, 0xff);
+                out_8(&udbg_comport->ier, 0x00);
+                out_8(&udbg_comport->lcr, 0x80);        /* Access baud rate */
+                out_8(&udbg_comport->dll, dll & 0xff);  /* 1 = 115200,  2 = 57600,
+                                                           3 = 38400, 12 = 9600 baud */
+                out_8(&udbg_comport->dlm, dll >> 8);    /* dll >> 8 which should be zero
+                                                           for fast rates; */
+                out_8(&udbg_comport->lcr, 0x03);        /* 8 data, 1 stop, no parity */
+                out_8(&udbg_comport->mcr, 0x03);        /* RTS/DTR */
+                out_8(&udbg_comport->fcr ,0x07);        /* Clear & enable FIFOs */
+                udbg_putc = udbg_550_putc;
+                udbg_getc = udbg_550_getc;
+                udbg_getc_poll = udbg_550_getc_poll;
+        }
+}
+#ifdef CONFIG_PPC_MAPLE
+void udbg_maple_real_putc(unsigned char c)
+{
+        if (udbg_comport) {
+                while ((real_readb(&udbg_comport->lsr) & LSR_THRE) == 0)
+                        /* wait for idle */;
+                real_writeb(c, &udbg_comport->thr); eieio();
+                if (c == '\n')
+                        udbg_maple_real_putc('\r');
+        }
+}
+void udbg_init_maple_realmode(void)
+{
+        udbg_comport = (volatile struct NS16550 __iomem *)0xf40003f8;
+        udbg_putc = udbg_maple_real_putc;
+        udbg_getc = NULL;
+        udbg_getc_poll = NULL;
+}
+#endif /* CONFIG_PPC_MAPLE */
diff --git a/arch/ppc64/kernel/udbg_scc.c b/arch/ppc64/kernel/udbg_scc.c
new file mode 100644
index 000000000000..c47fd6c63531
--- /dev/null
+++ b/arch/ppc64/kernel/udbg_scc.c
@@ -0,0 +1,136 @@
+/*
+ * udbg for for zilog scc ports as found on Apple PowerMacs
+ *
+ * Copyright (C) 2001-2005 PPC 64 Team, IBM Corp
+ *
+ *      This program is free software; you can redistribute it and/or
+ *      modify it under the terms of the GNU General Public License
+ *      as published by the Free Software Foundation; either version
+ *      2 of the License, or (at your option) any later version.
+ */
+#include <linux/config.h>
+#include <linux/types.h>
+#include <asm/udbg.h>
+#include <asm/processor.h>
+#include <asm/naca.h>
+#include <asm/io.h>
+#include <asm/prom.h>
+#include <asm/pmac_feature.h>
+extern u8 real_readb(volatile u8 __iomem  *addr);
+extern void real_writeb(u8 data, volatile u8 __iomem *addr);
+#define SCC_TXRDY       4
+#define SCC_RXRDY       1
+static volatile u8 __iomem *sccc;
+static volatile u8 __iomem *sccd;
+static void udbg_scc_putc(unsigned char c)
+{
+        if (sccc) {
+                while ((in_8(sccc) & SCC_TXRDY) == 0)
+                        ;
+                out_8(sccd,  c);
+                if (c == '\n')
+                        udbg_scc_putc('\r');
+        }
+}
+static int udbg_scc_getc_poll(void)
+{
+        if (sccc) {
+                if ((in_8(sccc) & SCC_RXRDY) != 0)
+                        return in_8(sccd);
+                else
+                        return -1;
+        }
+        return -1;
+}
+static unsigned char udbg_scc_getc(void)
+{
+        if (sccc) {
+                while ((in_8(sccc) & SCC_RXRDY) == 0)
+                        ;
+                return in_8(sccd);
+        }
+        return 0;
+}
+static unsigned char scc_inittab[] = {
+    13, 0,              /* set baud rate divisor */
+    12, 0,
+    14, 1,              /* baud rate gen enable, src=rtxc */
+    11, 0x50,           /* clocks = br gen */
+    5,  0xea,           /* tx 8 bits, assert DTR & RTS */
+    4,  0x46,           /* x16 clock, 1 stop */
+    3,  0xc1,           /* rx enable, 8 bits */
+};
+void udbg_init_scc(struct device_node *np)
+{
+        u32 *reg;
+        unsigned long addr;
+        int i, x;
+        if (np == NULL)
+                np = of_find_node_by_name(NULL, "escc");
+        if (np == NULL || np->parent == NULL)
+                return;
+        udbg_printf("found SCC...\n");
+        /* Get address within mac-io ASIC */
+        reg = (u32 *)get_property(np, "reg", NULL);
+        if (reg == NULL)
+                return;
+        addr = reg[0];
+        udbg_printf("local addr: %lx\n", addr);
+        /* Get address of mac-io PCI itself */
+        reg = (u32 *)get_property(np->parent, "assigned-addresses", NULL);
+        if (reg == NULL)
+                return;
+        addr += reg[2];
+        udbg_printf("final addr: %lx\n", addr);
+        /* Setup for 57600 8N1 */
+        addr += 0x20;
+        sccc = (volatile u8 * __iomem) ioremap(addr & PAGE_MASK, PAGE_SIZE) ;
+        sccc += addr & ~PAGE_MASK;
+        sccd = sccc + 0x10;
+        udbg_printf("ioremap result sccc: %p\n", sccc);
+        mb();
+        for (i = 20000; i != 0; --i)
+                x = in_8(sccc);
+        out_8(sccc, 0x09);              /* reset A or B side */
+        out_8(sccc, 0xc0);
+        for (i = 0; i < sizeof(scc_inittab); ++i)
+                out_8(sccc, scc_inittab[i]);
+        udbg_putc = udbg_scc_putc;
+        udbg_getc = udbg_scc_getc;
+        udbg_getc_poll = udbg_scc_getc_poll;
+        udbg_puts("Hello World !\n");
+}
+static void udbg_real_scc_putc(unsigned char c)
+{
+        while ((real_readb(sccc) & SCC_TXRDY) == 0)
+                ;
+        real_writeb(c, sccd);
+        if (c == '\n')
+                udbg_real_scc_putc('\r');
+}
+void udbg_init_pmac_realmode(void)
+{
+        sccc = (volatile u8 __iomem *)0x80013020ul;
+        sccd = (volatile u8 __iomem *)0x80013030ul;
+        udbg_putc = udbg_real_scc_putc;
+        udbg_getc = NULL;
+        udbg_getc_poll = NULL;
+}
diff --git a/arch/ppc64/kernel/vdso32/cacheflush.S b/arch/ppc64/kernel/vdso32/cacheflush.S
index 0ed7ea721715..c8db993574ee 100644
--- a/arch/ppc64/kernel/vdso32/cacheflush.S
+++ b/arch/ppc64/kernel/vdso32/cacheflush.S
@@ -13,7 +13,7 @@
 #include <asm/processor.h>
 #include <asm/ppc_asm.h>
 #include <asm/vdso.h>
-#include <asm/offsets.h>
+#include <asm/asm-offsets.h>
        .text
diff --git a/arch/ppc64/kernel/vdso32/datapage.S b/arch/ppc64/kernel/vdso32/datapage.S
index 29b6bd32e1f1..4f4eb0be3992 100644
--- a/arch/ppc64/kernel/vdso32/datapage.S
+++ b/arch/ppc64/kernel/vdso32/datapage.S
@@ -12,7 +12,7 @@
 #include <linux/config.h>
 #include <asm/processor.h>
 #include <asm/ppc_asm.h>
-#include <asm/offsets.h>
+#include <asm/asm-offsets.h>
 #include <asm/unistd.h>
 #include <asm/vdso.h>
diff --git a/arch/ppc64/kernel/vdso32/gettimeofday.S b/arch/ppc64/kernel/vdso32/gettimeofday.S
index 2b48bf1fb109..07f1c1c650c8 100644
--- a/arch/ppc64/kernel/vdso32/gettimeofday.S
+++ b/arch/ppc64/kernel/vdso32/gettimeofday.S
@@ -13,7 +13,7 @@
 #include <asm/processor.h>
 #include <asm/ppc_asm.h>
 #include <asm/vdso.h>
-#include <asm/offsets.h>
+#include <asm/asm-offsets.h>
 #include <asm/unistd.h>
        .text
diff --git a/arch/ppc64/kernel/vdso64/cacheflush.S b/arch/ppc64/kernel/vdso64/cacheflush.S
index e0725b7b7003..d4a0ad28d534 100644
--- a/arch/ppc64/kernel/vdso64/cacheflush.S
+++ b/arch/ppc64/kernel/vdso64/cacheflush.S
@@ -13,7 +13,7 @@
 #include <asm/processor.h>
 #include <asm/ppc_asm.h>
 #include <asm/vdso.h>
-#include <asm/offsets.h>
+#include <asm/asm-offsets.h>
        .text
diff --git a/arch/ppc64/kernel/vdso64/datapage.S b/arch/ppc64/kernel/vdso64/datapage.S
index 18afd971c9d9..ed6e599ae824 100644
--- a/arch/ppc64/kernel/vdso64/datapage.S
+++ b/arch/ppc64/kernel/vdso64/datapage.S
@@ -12,7 +12,7 @@
 #include <linux/config.h>
 #include <asm/processor.h>
 #include <asm/ppc_asm.h>
-#include <asm/offsets.h>
+#include <asm/asm-offsets.h>
 #include <asm/unistd.h>
 #include <asm/vdso.h>
diff --git a/arch/ppc64/kernel/vdso64/gettimeofday.S b/arch/ppc64/kernel/vdso64/gettimeofday.S
index ed3f970ff05e..f6df8028570a 100644
--- a/arch/ppc64/kernel/vdso64/gettimeofday.S
+++ b/arch/ppc64/kernel/vdso64/gettimeofday.S
@@ -14,7 +14,7 @@
 #include <asm/processor.h>
 #include <asm/ppc_asm.h>
 #include <asm/vdso.h>
-#include <asm/offsets.h>
+#include <asm/asm-offsets.h>
        .text
 /*
diff --git a/arch/ppc64/kernel/vio.c b/arch/ppc64/kernel/vio.c
index 0c0ba71ac0e8..c90e1dd875ce 100644
--- a/arch/ppc64/kernel/vio.c
+++ b/arch/ppc64/kernel/vio.c
@@ -1,10 +1,11 @@
 /*
 * IBM PowerPC Virtual I/O Infrastructure Support.
 *
- *    Copyright (c) 2003 IBM Corp.
+ *    Copyright (c) 2003-2005 IBM Corp.
 *     Dave Engebretsen engebret@us.ibm.com
 *     Santiago Leon santil@us.ibm.com
 *     Hollis Blanchard <hollisb@us.ibm.com>
+ *     Stephen Rothwell
 *
 *      This program is free software; you can redistribute it and/or
 *      modify it under the terms of the GNU General Public License
@@ -14,61 +15,30 @@
 #include <linux/init.h>
 #include <linux/console.h>
-#include <linux/version.h>
 #include <linux/module.h>
-#include <linux/kobject.h>
 #include <linux/mm.h>
 #include <linux/dma-mapping.h>
-#include <asm/rtas.h>
 #include <asm/iommu.h>
 #include <asm/dma.h>
-#include <asm/ppcdebug.h>
 #include <asm/vio.h>
-#include <asm/hvcall.h>
-#include <asm/iSeries/vio.h>
-#include <asm/iSeries/HvTypes.h>
-#include <asm/iSeries/HvCallXm.h>
-#include <asm/iSeries/HvLpConfig.h>
-#define DBGENTER() pr_debug("%s entered\n", __FUNCTION__)
-extern struct subsystem devices_subsys; /* needed for vio_find_name() */
 static const struct vio_device_id *vio_match_device(
                const struct vio_device_id *, const struct vio_dev *);
-#ifdef CONFIG_PPC_PSERIES
+struct vio_dev vio_bus_device  = { /* fake "parent" device */
-static struct iommu_table *vio_build_iommu_table(struct vio_dev *);
-static int vio_num_address_cells;
-#endif
-#ifdef CONFIG_PPC_ISERIES
-static struct iommu_table veth_iommu_table;
-static struct iommu_table vio_iommu_table;
-#endif
-static struct vio_dev vio_bus_device  = { /* fake "parent" device */
        .name = vio_bus_device.dev.bus_id,
        .type = "",
-#ifdef CONFIG_PPC_ISERIES
-        .iommu_table = &vio_iommu_table,
-#endif
        .dev.bus_id = "vio",
        .dev.bus = &vio_bus_type,
 };
-#ifdef CONFIG_PPC_ISERIES
+static struct vio_bus_ops vio_bus_ops;
-static struct vio_dev *__init vio_register_device_iseries(char *type,
-                uint32_t unit_num);
-struct device *iSeries_vio_dev = &vio_bus_device.dev;
-EXPORT_SYMBOL(iSeries_vio_dev);
-#define device_is_compatible(a, b)      1
-#endif
+/*
+ * Convert from struct device to struct vio_dev and pass to driver.
-/* convert from struct device to struct vio_dev and pass to driver.
 * dev->driver has already been set by generic code because vio_bus_match
- * succeeded. */
+ * succeeded.
+ */
 static int vio_bus_probe(struct device *dev)
 {
        struct vio_dev *viodev = to_vio_dev(dev);
@@ -76,15 +46,12 @@ static int vio_bus_probe(struct device *dev)
        const struct vio_device_id *id;
        int error = -ENODEV;
-        DBGENTER();
        if (!viodrv->probe)
                return error;
        id = vio_match_device(viodrv->id_table, viodev);
-        if (id) {
+        if (id)
                error = viodrv->probe(viodev, id);
-        }
        return error;
 }
@@ -95,11 +62,8 @@ static int vio_bus_remove(struct device *dev)
        struct vio_dev *viodev = to_vio_dev(dev);
        struct vio_driver *viodrv = to_vio_driver(dev->driver);
-        DBGENTER();
+        if (viodrv->remove)
-        if (viodrv->remove) {
                return viodrv->remove(viodev);
-        }
        /* driver can't remove */
        return 1;
@@ -135,193 +99,72 @@ void vio_unregister_driver(struct vio_driver *viodrv)
 EXPORT_SYMBOL(vio_unregister_driver);
 /**
- * vio_match_device: - Tell if a VIO device has a matching VIO device id structure.
+ * vio_match_device: - Tell if a VIO device has a matching
- * @ids:        array of VIO device id structures to search in
+ *                      VIO device id structure.
- * @dev:        the VIO device structure to match against
+ * @ids:        array of VIO device id structures to search in
+ * @dev:        the VIO device structure to match against
 *
 * Used by a driver to check whether a VIO device present in the
 * system is in its list of supported devices. Returns the matching
 * vio_device_id structure or NULL if there is no match.
 */
-static const struct vio_device_id * vio_match_device(const struct vio_device_id *ids,
+static const struct vio_device_id *vio_match_device(
-        const struct vio_dev *dev)
+                const struct vio_device_id *ids, const struct vio_dev *dev)
 {
-        DBGENTER();
+        while (ids->type[0] != '\0') {
+                if (vio_bus_ops.match(ids, dev))
-        while (ids->type) {
-                if ((strncmp(dev->type, ids->type, strlen(ids->type)) == 0) &&
-                        device_is_compatible(dev->dev.platform_data, ids->compat))
                        return ids;
                ids++;
        }
        return NULL;
 }
-#ifdef CONFIG_PPC_ISERIES
-void __init iommu_vio_init(void)
-{
-        struct iommu_table *t;
-        struct iommu_table_cb cb;
-        unsigned long cbp;
-        unsigned long itc_entries;
-        cb.itc_busno = 255;    /* Bus 255 is the virtual bus */
-        cb.itc_virtbus = 0xff; /* Ask for virtual bus */
-        cbp = virt_to_abs(&cb);
-        HvCallXm_getTceTableParms(cbp);
-        itc_entries = cb.itc_size * PAGE_SIZE / sizeof(union tce_entry);
-        veth_iommu_table.it_size        = itc_entries / 2;
-        veth_iommu_table.it_busno       = cb.itc_busno;
-        veth_iommu_table.it_offset      = cb.itc_offset;
-        veth_iommu_table.it_index       = cb.itc_index;
-        veth_iommu_table.it_type        = TCE_VB;
-        veth_iommu_table.it_blocksize   = 1;
-        t = iommu_init_table(&veth_iommu_table);
-        if (!t)
-                printk("Virtual Bus VETH TCE table failed.\n");
-        vio_iommu_table.it_size         = itc_entries - veth_iommu_table.it_size;
-        vio_iommu_table.it_busno        = cb.itc_busno;
-        vio_iommu_table.it_offset       = cb.itc_offset +
-                                          veth_iommu_table.it_size;
-        vio_iommu_table.it_index        = cb.itc_index;
-        vio_iommu_table.it_type         = TCE_VB;
-        vio_iommu_table.it_blocksize    = 1;
-        t = iommu_init_table(&vio_iommu_table);
-        if (!t)
-                printk("Virtual Bus VIO TCE table failed.\n");
-}
-#endif
-#ifdef CONFIG_PPC_PSERIES
-static void probe_bus_pseries(void)
-{
-        struct device_node *node_vroot, *of_node;
-        node_vroot = find_devices("vdevice");
-        if ((node_vroot == NULL) || (node_vroot->child == NULL))
-                /* this machine doesn't do virtual IO, and that's ok */
-                return;
-        vio_num_address_cells = prom_n_addr_cells(node_vroot->child);
-        /*
-         * Create struct vio_devices for each virtual device in the device tree.
-         * Drivers will associate with them later.
-         */
-        for (of_node = node_vroot->child; of_node != NULL;
-                        of_node = of_node->sibling) {
-                printk(KERN_DEBUG "%s: processing %p\n", __FUNCTION__, of_node);
-                vio_register_device_node(of_node);
-        }
-}
-#endif
-#ifdef CONFIG_PPC_ISERIES
-static void probe_bus_iseries(void)
-{
-        HvLpIndexMap vlan_map = HvLpConfig_getVirtualLanIndexMap();
-        struct vio_dev *viodev;
-        int i;
-        /* there is only one of each of these */
-        vio_register_device_iseries("viocons", 0);
-        vio_register_device_iseries("vscsi", 0);
-        vlan_map = HvLpConfig_getVirtualLanIndexMap();
-        for (i = 0; i < HVMAXARCHITECTEDVIRTUALLANS; i++) {
-                if ((vlan_map & (0x8000 >> i)) == 0)
-                        continue;
-                viodev = vio_register_device_iseries("vlan", i);
-                /* veth is special and has it own iommu_table */
-                viodev->iommu_table = &veth_iommu_table;
-        }
-        for (i = 0; i < HVMAXARCHITECTEDVIRTUALDISKS; i++)
-                vio_register_device_iseries("viodasd", i);
-        for (i = 0; i < HVMAXARCHITECTEDVIRTUALCDROMS; i++)
-                vio_register_device_iseries("viocd", i);
-        for (i = 0; i < HVMAXARCHITECTEDVIRTUALTAPES; i++)
-                vio_register_device_iseries("viotape", i);
-}
-#endif
 /**
 * vio_bus_init: - Initialize the virtual IO bus
 */
-static int __init vio_bus_init(void)
+int __init vio_bus_init(struct vio_bus_ops *ops)
 {
        int err;
+        vio_bus_ops = *ops;
        err = bus_register(&vio_bus_type);
        if (err) {
                printk(KERN_ERR "failed to register VIO bus\n");
                return err;
        }
-        /* the fake parent of all vio devices, just to give us a nice directory */
+        /*
+         * The fake parent of all vio devices, just to give us
+         * a nice directory
+         */
        err = device_register(&vio_bus_device.dev);
        if (err) {
-                printk(KERN_WARNING "%s: device_register returned %i\n", __FUNCTION__,
+                printk(KERN_WARNING "%s: device_register returned %i\n",
-                        err);
+                                __FUNCTION__, err);
                return err;
        }
-#ifdef CONFIG_PPC_PSERIES
-        probe_bus_pseries();
-#endif
-#ifdef CONFIG_PPC_ISERIES
-        probe_bus_iseries();
-#endif
        return 0;
 }
-__initcall(vio_bus_init);
 /* vio_dev refcount hit 0 */
 static void __devinit vio_dev_release(struct device *dev)
 {
-        DBGENTER();
+        if (vio_bus_ops.release_device)
+                vio_bus_ops.release_device(dev);
-#ifdef CONFIG_PPC_PSERIES
-        /* XXX free TCE table */
-        of_node_put(dev->platform_data);
-#endif
        kfree(to_vio_dev(dev));
 }
-#ifdef CONFIG_PPC_PSERIES
+static ssize_t viodev_show_name(struct device *dev,
-static ssize_t viodev_show_devspec(struct device *dev, struct device_attribute *attr, char *buf)
+                struct device_attribute *attr, char *buf)
-{
-        struct device_node *of_node = dev->platform_data;
-        return sprintf(buf, "%s\n", of_node->full_name);
-}
-DEVICE_ATTR(devspec, S_IRUSR | S_IRGRP | S_IROTH, viodev_show_devspec, NULL);
-#endif
-static ssize_t viodev_show_name(struct device *dev, struct device_attribute *attr, char *buf)
 {
        return sprintf(buf, "%s\n", to_vio_dev(dev)->name);
 }
 DEVICE_ATTR(name, S_IRUSR | S_IRGRP | S_IROTH, viodev_show_name, NULL);
-static struct vio_dev * __devinit vio_register_device_common(
+struct vio_dev * __devinit vio_register_device(struct vio_dev *viodev)
-                struct vio_dev *viodev, char *name, char *type,
-                uint32_t unit_address, struct iommu_table *iommu_table)
 {
-        DBGENTER();
-        viodev->name = name;
-        viodev->type = type;
-        viodev->unit_address = unit_address;
-        viodev->iommu_table = iommu_table;
        /* init generic 'struct device' fields: */
        viodev->dev.parent = &vio_bus_device.dev;
        viodev->dev.bus = &vio_bus_type;
@@ -338,222 +181,15 @@ static struct vio_dev * __devinit vio_register_device_common(
        return viodev;
 }
-#ifdef CONFIG_PPC_PSERIES
-/**
- * vio_register_device_node: - Register a new vio device.
- * @of_node:    The OF node for this device.
- *
- * Creates and initializes a vio_dev structure from the data in
- * of_node (dev.platform_data) and adds it to the list of virtual devices.
- * Returns a pointer to the created vio_dev or NULL if node has
- * NULL device_type or compatible fields.
- */
-struct vio_dev * __devinit vio_register_device_node(struct device_node *of_node)
-{
-        struct vio_dev *viodev;
-        unsigned int *unit_address;
-        unsigned int *irq_p;
-        DBGENTER();
-        /* we need the 'device_type' property, in order to match with drivers */
-        if ((NULL == of_node->type)) {
-                printk(KERN_WARNING
-                        "%s: node %s missing 'device_type'\n", __FUNCTION__,
-                        of_node->name ? of_node->name : "<unknown>");
-                return NULL;
-        }
-        unit_address = (unsigned int *)get_property(of_node, "reg", NULL);
-        if (!unit_address) {
-                printk(KERN_WARNING "%s: node %s missing 'reg'\n", __FUNCTION__,
-                        of_node->name ? of_node->name : "<unknown>");
-                return NULL;
-        }
-        /* allocate a vio_dev for this node */
-        viodev = kmalloc(sizeof(struct vio_dev), GFP_KERNEL);
-        if (!viodev) {
-                return NULL;
-        }
-        memset(viodev, 0, sizeof(struct vio_dev));
-        viodev->dev.platform_data = of_node_get(of_node);
-        viodev->irq = NO_IRQ;
-        irq_p = (unsigned int *)get_property(of_node, "interrupts", NULL);
-        if (irq_p) {
-                int virq = virt_irq_create_mapping(*irq_p);
-                if (virq == NO_IRQ) {
-                        printk(KERN_ERR "Unable to allocate interrupt "
-                               "number for %s\n", of_node->full_name);
-                } else
-                        viodev->irq = irq_offset_up(virq);
-        }
-        snprintf(viodev->dev.bus_id, BUS_ID_SIZE, "%x", *unit_address);
-        /* register with generic device framework */
-        if (vio_register_device_common(viodev, of_node->name, of_node->type,
-                                *unit_address, vio_build_iommu_table(viodev))
-                        == NULL) {
-                /* XXX free TCE table */
-                kfree(viodev);
-                return NULL;
-        }
-        device_create_file(&viodev->dev, &dev_attr_devspec);
-        return viodev;
-}
-EXPORT_SYMBOL(vio_register_device_node);
-#endif
-#ifdef CONFIG_PPC_ISERIES
-/**
- * vio_register_device: - Register a new vio device.
- * @voidev:     The device to register.
- */
-static struct vio_dev *__init vio_register_device_iseries(char *type,
-                uint32_t unit_num)
-{
-        struct vio_dev *viodev;
-        DBGENTER();
-        /* allocate a vio_dev for this node */
-        viodev = kmalloc(sizeof(struct vio_dev), GFP_KERNEL);
-        if (!viodev)
-                return NULL;
-        memset(viodev, 0, sizeof(struct vio_dev));
-        snprintf(viodev->dev.bus_id, BUS_ID_SIZE, "%s%d", type, unit_num);
-        return vio_register_device_common(viodev, viodev->dev.bus_id, type,
-                        unit_num, &vio_iommu_table);
-}
-#endif
 void __devinit vio_unregister_device(struct vio_dev *viodev)
 {
-        DBGENTER();
+        if (vio_bus_ops.unregister_device)
-#ifdef CONFIG_PPC_PSERIES
+                vio_bus_ops.unregister_device(viodev);
-        device_remove_file(&viodev->dev, &dev_attr_devspec);
-#endif
        device_remove_file(&viodev->dev, &dev_attr_name);
        device_unregister(&viodev->dev);
 }
 EXPORT_SYMBOL(vio_unregister_device);
-#ifdef CONFIG_PPC_PSERIES
-/**
- * vio_get_attribute: - get attribute for virtual device
- * @vdev:       The vio device to get property.
- * @which:      The property/attribute to be extracted.
- * @length:     Pointer to length of returned data size (unused if NULL).
- *
- * Calls prom.c's get_property() to return the value of the
- * attribute specified by the preprocessor constant @which
-*/
-const void * vio_get_attribute(struct vio_dev *vdev, void* which, int* length)
-{
-        return get_property(vdev->dev.platform_data, (char*)which, length);
-}
-EXPORT_SYMBOL(vio_get_attribute);
-/* vio_find_name() - internal because only vio.c knows how we formatted the
- * kobject name
- * XXX once vio_bus_type.devices is actually used as a kset in
- * drivers/base/bus.c, this function should be removed in favor of
- * "device_find(kobj_name, &vio_bus_type)"
- */
-static struct vio_dev *vio_find_name(const char *kobj_name)
-{
-        struct kobject *found;
-        found = kset_find_obj(&devices_subsys.kset, kobj_name);
-        if (!found)
-                return NULL;
-        return to_vio_dev(container_of(found, struct device, kobj));
-}
-/**
- * vio_find_node - find an already-registered vio_dev
- * @vnode: device_node of the virtual device we're looking for
- */
-struct vio_dev *vio_find_node(struct device_node *vnode)
-{
-        uint32_t *unit_address;
-        char kobj_name[BUS_ID_SIZE];
-        /* construct the kobject name from the device node */
-        unit_address = (uint32_t *)get_property(vnode, "reg", NULL);
-        if (!unit_address)
-                return NULL;
-        snprintf(kobj_name, BUS_ID_SIZE, "%x", *unit_address);
-        return vio_find_name(kobj_name);
-}
-EXPORT_SYMBOL(vio_find_node);
-/**
- * vio_build_iommu_table: - gets the dma information from OF and builds the TCE tree.
- * @dev: the virtual device.
- *
- * Returns a pointer to the built tce tree, or NULL if it can't
- * find property.
-*/
-static struct iommu_table * vio_build_iommu_table(struct vio_dev *dev)
-{
-        unsigned int *dma_window;
-        struct iommu_table *newTceTable;
-        unsigned long offset;
-        int dma_window_property_size;
-        dma_window = (unsigned int *) get_property(dev->dev.platform_data, "ibm,my-dma-window", &dma_window_property_size);
-        if(!dma_window) {
-                return NULL;
-        }
-        newTceTable = (struct iommu_table *) kmalloc(sizeof(struct iommu_table), GFP_KERNEL);
-        /*  There should be some code to extract the phys-encoded offset
-                using prom_n_addr_cells(). However, according to a comment
-                on earlier versions, it's always zero, so we don't bother */
-        offset = dma_window[1] >>  PAGE_SHIFT;
-        /* TCE table size - measured in tce entries */
-        newTceTable->it_size            = dma_window[4] >> PAGE_SHIFT;
-        /* offset for VIO should always be 0 */
-        newTceTable->it_offset          = offset;
-        newTceTable->it_busno           = 0;
-        newTceTable->it_index           = (unsigned long)dma_window[0];
-        newTceTable->it_type            = TCE_VB;
-        return iommu_init_table(newTceTable);
-}
-int vio_enable_interrupts(struct vio_dev *dev)
-{
-        int rc = h_vio_signal(dev->unit_address, VIO_IRQ_ENABLE);
-        if (rc != H_Success) {
-                printk(KERN_ERR "vio: Error 0x%x enabling interrupts\n", rc);
-        }
-        return rc;
-}
-EXPORT_SYMBOL(vio_enable_interrupts);
-int vio_disable_interrupts(struct vio_dev *dev)
-{
-        int rc = h_vio_signal(dev->unit_address, VIO_IRQ_DISABLE);
-        if (rc != H_Success) {
-                printk(KERN_ERR "vio: Error 0x%x disabling interrupts\n", rc);
-        }
-        return rc;
-}
-EXPORT_SYMBOL(vio_disable_interrupts);
-#endif
 static dma_addr_t vio_map_single(struct device *dev, void *vaddr,
                          size_t size, enum dma_data_direction direction)
 {
@@ -615,18 +251,8 @@ static int vio_bus_match(struct device *dev, struct device_driver *drv)
        const struct vio_dev *vio_dev = to_vio_dev(dev);
        struct vio_driver *vio_drv = to_vio_driver(drv);
        const struct vio_device_id *ids = vio_drv->id_table;
-        const struct vio_device_id *found_id;
-        DBGENTER();
-        if (!ids)
+        return (ids != NULL) && (vio_match_device(ids, vio_dev) != NULL);
-                return 0;
-        found_id = vio_match_device(ids, vio_dev);
-        if (found_id)
-                return 1;
-        return 0;
 }
 struct bus_type vio_bus_type = {
diff --git a/arch/ppc64/kernel/vmlinux.lds.S b/arch/ppc64/kernel/vmlinux.lds.S
index 4103cc13f8d6..0306510bc4ff 100644
--- a/arch/ppc64/kernel/vmlinux.lds.S
+++ b/arch/ppc64/kernel/vmlinux.lds.S
@@ -15,6 +15,7 @@ SECTIONS
        *(.text .text.*)
        SCHED_TEXT
        LOCK_TEXT
+        KPROBES_TEXT
        *(.fixup)
        . = ALIGN(4096);
        _etext = .;
diff --git a/arch/ppc64/kernel/xics.c b/arch/ppc64/kernel/xics.c
index d9dc6f28d050..daf93885dcfa 100644
--- a/arch/ppc64/kernel/xics.c
+++ b/arch/ppc64/kernel/xics.c
@@ -38,7 +38,7 @@ static void xics_mask_and_ack_irq(unsigned int irq);
 static void xics_end_irq(unsigned int irq);
 static void xics_set_affinity(unsigned int irq_nr, cpumask_t cpumask);
-struct hw_interrupt_type xics_pic = {
+static struct hw_interrupt_type xics_pic = {
        .typename = " XICS     ",
        .startup = xics_startup,
        .enable = xics_enable_irq,
@@ -48,7 +48,7 @@ struct hw_interrupt_type xics_pic = {
        .set_affinity = xics_set_affinity
 };
-struct hw_interrupt_type xics_8259_pic = {
+static struct hw_interrupt_type xics_8259_pic = {
        .typename = " XICS/8259",
        .ack = xics_mask_and_ack_irq,
 };
@@ -89,9 +89,8 @@ static struct xics_ipl __iomem *xics_per_cpu[NR_CPUS];
 static int xics_irq_8259_cascade = 0;
 static int xics_irq_8259_cascade_real = 0;
 static unsigned int default_server = 0xFF;
-/* also referenced in smp.c... */
+static unsigned int default_distrib_server = 0;
-unsigned int default_distrib_server = 0;
+static unsigned int interrupt_server_size = 8;
-unsigned int interrupt_server_size = 8;
 /*
 * XICS only has a single IPI, so encode the messages per CPU
@@ -99,10 +98,10 @@ unsigned int interrupt_server_size = 8;
 struct xics_ipi_struct xics_ipi_message[NR_CPUS] __cacheline_aligned;
 /* RTAS service tokens */
-int ibm_get_xive;
+static int ibm_get_xive;
-int ibm_set_xive;
+static int ibm_set_xive;
-int ibm_int_on;
+static int ibm_int_on;
-int ibm_int_off;
+static int ibm_int_off;
 typedef struct {
        int (*xirr_info_get)(int cpu);
@@ -284,16 +283,17 @@ static void xics_enable_irq(unsigned int virq)
        call_status = rtas_call(ibm_set_xive, 3, 1, NULL, irq, server,
                                DEFAULT_PRIORITY);
        if (call_status != 0) {
-                printk(KERN_ERR "xics_enable_irq: irq=%d: ibm_set_xive "
+                printk(KERN_ERR "xics_enable_irq: irq=%u: ibm_set_xive "
-                       "returned %x\n", irq, call_status);
+                       "returned %d\n", irq, call_status);
+                printk("set_xive %x, server %x\n", ibm_set_xive, server);
                return;
        }
        /* Now unmask the interrupt (often a no-op) */
        call_status = rtas_call(ibm_int_on, 1, 1, NULL, irq);
        if (call_status != 0) {
-                printk(KERN_ERR "xics_enable_irq: irq=%d: ibm_int_on "
+                printk(KERN_ERR "xics_enable_irq: irq=%u: ibm_int_on "
-                       "returned %x\n", irq, call_status);
+                       "returned %d\n", irq, call_status);
                return;
        }
 }
@@ -308,8 +308,8 @@ static void xics_disable_real_irq(unsigned int irq)
        call_status = rtas_call(ibm_int_off, 1, 1, NULL, irq);
        if (call_status != 0) {
-                printk(KERN_ERR "xics_disable_real_irq: irq=%d: "
+                printk(KERN_ERR "xics_disable_real_irq: irq=%u: "
-                       "ibm_int_off returned %x\n", irq, call_status);
+                       "ibm_int_off returned %d\n", irq, call_status);
                return;
        }
@@ -317,8 +317,8 @@ static void xics_disable_real_irq(unsigned int irq)
        /* Have to set XIVE to 0xff to be able to remove a slot */
        call_status = rtas_call(ibm_set_xive, 3, 1, NULL, irq, server, 0xff);
        if (call_status != 0) {
-                printk(KERN_ERR "xics_disable_irq: irq=%d: ibm_set_xive(0xff)"
+                printk(KERN_ERR "xics_disable_irq: irq=%u: ibm_set_xive(0xff)"
-                       " returned %x\n", irq, call_status);
+                       " returned %d\n", irq, call_status);
                return;
        }
 }
@@ -380,7 +380,7 @@ int xics_get_irq(struct pt_regs *regs)
                if (irq == NO_IRQ)
                        irq = real_irq_to_virt_slowpath(vec);
                if (irq == NO_IRQ) {
-                        printk(KERN_ERR "Interrupt %d (real) is invalid,"
+                        printk(KERN_ERR "Interrupt %u (real) is invalid,"
                               " disabling it.\n", vec);
                        xics_disable_real_irq(vec);
                } else
@@ -622,7 +622,7 @@ static void xics_set_affinity(unsigned int virq, cpumask_t cpumask)
        status = rtas_call(ibm_get_xive, 1, 3, xics_status, irq);
        if (status) {
-                printk(KERN_ERR "xics_set_affinity: irq=%d ibm,get-xive "
+                printk(KERN_ERR "xics_set_affinity: irq=%u ibm,get-xive "
                       "returns %d\n", irq, status);
                return;
        }
@@ -641,7 +641,7 @@ static void xics_set_affinity(unsigned int virq, cpumask_t cpumask)
                                irq, newmask, xics_status[1]);
        if (status) {
-                printk(KERN_ERR "xics_set_affinity: irq=%d ibm,set-xive "
+                printk(KERN_ERR "xics_set_affinity: irq=%u ibm,set-xive "
                       "returns %d\n", irq, status);
                return;
        }
@@ -720,7 +720,7 @@ void xics_migrate_irqs_away(void)
                status = rtas_call(ibm_get_xive, 1, 3, xics_status, irq);
                if (status) {
-                        printk(KERN_ERR "migrate_irqs_away: irq=%d "
+                        printk(KERN_ERR "migrate_irqs_away: irq=%u "
                                        "ibm,get-xive returns %d\n",
                                        virq, status);
                        goto unlock;
@@ -734,7 +734,7 @@ void xics_migrate_irqs_away(void)
                if (xics_status[0] != get_hard_smp_processor_id(cpu))
                        goto unlock;
-                printk(KERN_WARNING "IRQ %d affinity broken off cpu %u\n",
+                printk(KERN_WARNING "IRQ %u affinity broken off cpu %u\n",
                       virq, cpu);
                /* Reset affinity to all cpus */
diff --git a/arch/ppc64/lib/dec_and_lock.c b/arch/ppc64/lib/dec_and_lock.c
index 6e8d8591708c..7b9d4da5cf92 100644
--- a/arch/ppc64/lib/dec_and_lock.c
+++ b/arch/ppc64/lib/dec_and_lock.c
@@ -20,14 +20,7 @@
 * has a cmpxchg, and where atomic->value is an int holding
 * the value of the atomic (i.e. the high bits aren't used
 * for a lock or anything like that).
- *
- * N.B. ATOMIC_DEC_AND_LOCK gets defined in include/linux/spinlock.h
- * if spinlocks are empty and thus atomic_dec_and_lock is defined
- * to be atomic_dec_and_test - in that case we don't need it
- * defined here as well.
 */
-#ifndef ATOMIC_DEC_AND_LOCK
 int _atomic_dec_and_lock(atomic_t *atomic, spinlock_t *lock)
 {
        int counter;
@@ -52,4 +45,3 @@ int _atomic_dec_and_lock(atomic_t *atomic, spinlock_t *lock)
 }
 EXPORT_SYMBOL(_atomic_dec_and_lock);
-#endif /* ATOMIC_DEC_AND_LOCK */
diff --git a/arch/ppc64/lib/locks.c b/arch/ppc64/lib/locks.c
index ef70ef91abe2..033643ab69e0 100644
--- a/arch/ppc64/lib/locks.c
+++ b/arch/ppc64/lib/locks.c
@@ -23,12 +23,12 @@
 /* waiting for a spinlock... */
 #if defined(CONFIG_PPC_SPLPAR) || defined(CONFIG_PPC_ISERIES)
-void __spin_yield(spinlock_t *lock)
+void __spin_yield(raw_spinlock_t *lock)
 {
        unsigned int lock_value, holder_cpu, yield_count;
        struct paca_struct *holder_paca;
-        lock_value = lock->lock;
+        lock_value = lock->slock;
        if (lock_value == 0)
                return;
        holder_cpu = lock_value & 0xffff;
@@ -38,7 +38,7 @@ void __spin_yield(spinlock_t *lock)
        if ((yield_count & 1) == 0)
                return;         /* virtual cpu is currently running */
        rmb();
-        if (lock->lock != lock_value)
+        if (lock->slock != lock_value)
                return;         /* something has changed */
 #ifdef CONFIG_PPC_ISERIES
        HvCall2(HvCallBaseYieldProcessor, HvCall_YieldToProc,
@@ -54,7 +54,7 @@ void __spin_yield(spinlock_t *lock)
 * This turns out to be the same for read and write locks, since
 * we only know the holder if it is write-locked.
 */
-void __rw_yield(rwlock_t *rw)
+void __rw_yield(raw_rwlock_t *rw)
 {
        int lock_value;
        unsigned int holder_cpu, yield_count;
@@ -82,9 +82,9 @@ void __rw_yield(rwlock_t *rw)
 }
 #endif
-void spin_unlock_wait(spinlock_t *lock)
+void __raw_spin_unlock_wait(raw_spinlock_t *lock)
 {
-        while (lock->lock) {
+        while (lock->slock) {
                HMT_low();
                if (SHARED_PROCESSOR)
                        __spin_yield(lock);
@@ -92,4 +92,4 @@ void spin_unlock_wait(spinlock_t *lock)
        HMT_medium();
 }
-EXPORT_SYMBOL(spin_unlock_wait);
+EXPORT_SYMBOL(__raw_spin_unlock_wait);
diff --git a/arch/ppc64/mm/fault.c b/arch/ppc64/mm/fault.c
index 20b0f37e8bf8..7fbc68bbb739 100644
--- a/arch/ppc64/mm/fault.c
+++ b/arch/ppc64/mm/fault.c
@@ -29,6 +29,7 @@
 #include <linux/interrupt.h>
 #include <linux/smp_lock.h>
 #include <linux/module.h>
+#include <linux/kprobes.h>
 #include <asm/page.h>
 #include <asm/pgtable.h>
@@ -76,6 +77,28 @@ static int store_updates_sp(struct pt_regs *regs)
        return 0;
 }
+static void do_dabr(struct pt_regs *regs, unsigned long error_code)
+{
+        siginfo_t info;
+        if (notify_die(DIE_DABR_MATCH, "dabr_match", regs, error_code,
+                        11, SIGSEGV) == NOTIFY_STOP)
+                return;
+        if (debugger_dabr_match(regs))
+                return;
+        /* Clear the DABR */
+        set_dabr(0);
+        /* Deliver the signal to userspace */
+        info.si_signo = SIGTRAP;
+        info.si_errno = 0;
+        info.si_code = TRAP_HWBKPT;
+        info.si_addr = (void __user *)regs->nip;
+        force_sig_info(SIGTRAP, &info, current);
+}
 /*
 * The error_code parameter is
 *  - DSISR for a non-SLB data access fault,
@@ -84,8 +107,8 @@ static int store_updates_sp(struct pt_regs *regs)
 * The return value is 0 if the fault was handled, or the signal
 * number if this is a kernel fault that can't be handled here.
 */
-int do_page_fault(struct pt_regs *regs, unsigned long address,
+int __kprobes do_page_fault(struct pt_regs *regs, unsigned long address,
-                  unsigned long error_code)
+                            unsigned long error_code)
 {
        struct vm_area_struct * vma;
        struct mm_struct *mm = current->mm;
@@ -110,12 +133,9 @@ int do_page_fault(struct pt_regs *regs, unsigned long address,
        if (!user_mode(regs) && (address >= TASK_SIZE))
                return SIGSEGV;
-        if (error_code & DSISR_DABRMATCH) {
+        if (error_code & DSISR_DABRMATCH) {
-                if (notify_die(DIE_DABR_MATCH, "dabr_match", regs, error_code,
+                do_dabr(regs, error_code);
-                                        11, SIGSEGV) == NOTIFY_STOP)
+                return 0;
-                        return 0;
-                if (debugger_dabr_match(regs))
-                        return 0;
        }
        if (in_atomic() || mm == NULL) {
diff --git a/arch/ppc64/mm/hash_low.S b/arch/ppc64/mm/hash_low.S
index fbff24827ae7..ee5a5d36bfa8 100644
--- a/arch/ppc64/mm/hash_low.S
+++ b/arch/ppc64/mm/hash_low.S
@@ -16,7 +16,7 @@
 #include <asm/page.h>
 #include <asm/types.h>
 #include <asm/ppc_asm.h>
-#include <asm/offsets.h>
+#include <asm/asm-offsets.h>
 #include <asm/cputable.h>
        .text
@@ -129,12 +129,10 @@ _GLOBAL(__hash_page)
         * code rather than call a C function...) 
         */
 BEGIN_FTR_SECTION
-BEGIN_FTR_SECTION
        mr      r4,r30
        mr      r5,r7
        bl      .hash_page_do_lazy_icache
-END_FTR_SECTION_IFSET(CPU_FTR_NOEXECUTE)
+END_FTR_SECTION(CPU_FTR_NOEXECUTE|CPU_FTR_COHERENT_ICACHE, CPU_FTR_NOEXECUTE)
-END_FTR_SECTION_IFCLR(CPU_FTR_COHERENT_ICACHE)
        /* At this point, r3 contains new PP bits, save them in
         * place of "access" in the param area (sic)
diff --git a/arch/ppc64/mm/hash_native.c b/arch/ppc64/mm/hash_native.c
index a6abd3a979bf..7626bb59954d 100644
--- a/arch/ppc64/mm/hash_native.c
+++ b/arch/ppc64/mm/hash_native.c
@@ -51,7 +51,6 @@ long native_hpte_insert(unsigned long hpte_group, unsigned long va,
                        unsigned long prpn, unsigned long vflags,
                        unsigned long rflags)
 {
-        unsigned long arpn = physRpn_to_absRpn(prpn);
        hpte_t *hptep = htab_address + hpte_group;
        unsigned long hpte_v, hpte_r;
        int i;
@@ -74,7 +73,7 @@ long native_hpte_insert(unsigned long hpte_group, unsigned long va,
        hpte_v = (va >> 23) << HPTE_V_AVPN_SHIFT | vflags | HPTE_V_VALID;
        if (vflags & HPTE_V_LARGE)
                va &= ~(1UL << HPTE_V_AVPN_SHIFT);
-        hpte_r = (arpn << HPTE_R_RPN_SHIFT) | rflags;
+        hpte_r = (prpn << HPTE_R_RPN_SHIFT) | rflags;
        hptep->r = hpte_r;
        /* Guarantee the second dword is visible before the valid bit */
diff --git a/arch/ppc64/mm/hash_utils.c b/arch/ppc64/mm/hash_utils.c
index 623b5d130c31..09475c8edf7c 100644
--- a/arch/ppc64/mm/hash_utils.c
+++ b/arch/ppc64/mm/hash_utils.c
@@ -210,7 +210,7 @@ void __init htab_initialize(void)
        /* create bolted the linear mapping in the hash table */
        for (i=0; i < lmb.memory.cnt; i++) {
-                base = lmb.memory.region[i].physbase + KERNELBASE;
+                base = lmb.memory.region[i].base + KERNELBASE;
                size = lmb.memory.region[i].size;
                DBG("creating mapping for region: %lx : %lx\n", base, size);
@@ -302,7 +302,7 @@ int hash_page(unsigned long ea, unsigned long access, unsigned long trap)
        int local = 0;
        cpumask_t tmp;
-        if ((ea & ~REGION_MASK) > EADDR_MASK)
+        if ((ea & ~REGION_MASK) >= PGTABLE_RANGE)
                return 1;
        switch (REGION_ID(ea)) {
diff --git a/arch/ppc64/mm/hugetlbpage.c b/arch/ppc64/mm/hugetlbpage.c
index f9524602818d..338771ec70d7 100644
--- a/arch/ppc64/mm/hugetlbpage.c
+++ b/arch/ppc64/mm/hugetlbpage.c
@@ -27,124 +27,94 @@
 #include <linux/sysctl.h>
-#define HUGEPGDIR_SHIFT         (HPAGE_SHIFT + PAGE_SHIFT - 3)
+#define NUM_LOW_AREAS   (0x100000000UL >> SID_SHIFT)
-#define HUGEPGDIR_SIZE          (1UL << HUGEPGDIR_SHIFT)
+#define NUM_HIGH_AREAS  (PGTABLE_RANGE >> HTLB_AREA_SHIFT)
-#define HUGEPGDIR_MASK          (~(HUGEPGDIR_SIZE-1))
-#define HUGEPTE_INDEX_SIZE      9
+/* Modelled after find_linux_pte() */
-#define HUGEPGD_INDEX_SIZE      10
+pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
-#define PTRS_PER_HUGEPTE        (1 << HUGEPTE_INDEX_SIZE)
-#define PTRS_PER_HUGEPGD        (1 << HUGEPGD_INDEX_SIZE)
-static inline int hugepgd_index(unsigned long addr)
-{
-        return (addr & ~REGION_MASK) >> HUGEPGDIR_SHIFT;
-}
-static pud_t *hugepgd_offset(struct mm_struct *mm, unsigned long addr)
 {
-        int index;
+        pgd_t *pg;
+        pud_t *pu;
+        pmd_t *pm;
+        pte_t *pt;
-        if (! mm->context.huge_pgdir)
+        BUG_ON(! in_hugepage_area(mm->context, addr));
-                return NULL;
+        addr &= HPAGE_MASK;
+        pg = pgd_offset(mm, addr);
+        if (!pgd_none(*pg)) {
+                pu = pud_offset(pg, addr);
+                if (!pud_none(*pu)) {
+                        pm = pmd_offset(pu, addr);
+                        pt = (pte_t *)pm;
+                        BUG_ON(!pmd_none(*pm)
+                               && !(pte_present(*pt) && pte_huge(*pt)));
+                        return pt;
+                }
+        }
-        index = hugepgd_index(addr);
+        return NULL;
-        BUG_ON(index >= PTRS_PER_HUGEPGD);
-        return (pud_t *)(mm->context.huge_pgdir + index);
 }
-static inline pte_t *hugepte_offset(pud_t *dir, unsigned long addr)
+pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
 {
-        int index;
+        pgd_t *pg;
+        pud_t *pu;
-        if (pud_none(*dir))
+        pmd_t *pm;
-                return NULL;
+        pte_t *pt;
-        index = (addr >> HPAGE_SHIFT) % PTRS_PER_HUGEPTE;
-        return (pte_t *)pud_page(*dir) + index;
-}
-static pud_t *hugepgd_alloc(struct mm_struct *mm, unsigned long addr)
-{
        BUG_ON(! in_hugepage_area(mm->context, addr));
-        if (! mm->context.huge_pgdir) {
+        addr &= HPAGE_MASK;
-                pgd_t *new;
-                spin_unlock(&mm->page_table_lock);
-                /* Don't use pgd_alloc(), because we want __GFP_REPEAT */
-                new = kmem_cache_alloc(zero_cache, GFP_KERNEL | __GFP_REPEAT);
-                BUG_ON(memcmp(new, empty_zero_page, PAGE_SIZE));
-                spin_lock(&mm->page_table_lock);
-                /*
+        pg = pgd_offset(mm, addr);
-                 * Because we dropped the lock, we should re-check the
+        pu = pud_alloc(mm, pg, addr);
-                 * entry, as somebody else could have populated it..
-                 */
-                if (mm->context.huge_pgdir)
-                        pgd_free(new);
-                else
-                        mm->context.huge_pgdir = new;
-        }
-        return hugepgd_offset(mm, addr);
-}
-static pte_t *hugepte_alloc(struct mm_struct *mm, pud_t *dir, unsigned long addr)
+        if (pu) {
-{
+                pm = pmd_alloc(mm, pu, addr);
-        if (! pud_present(*dir)) {
+                if (pm) {
-                pte_t *new;
+                        pt = (pte_t *)pm;
+                        BUG_ON(!pmd_none(*pm)
-                spin_unlock(&mm->page_table_lock);
+                               && !(pte_present(*pt) && pte_huge(*pt)));
-                new = kmem_cache_alloc(zero_cache, GFP_KERNEL | __GFP_REPEAT);
+                        return pt;
-                BUG_ON(memcmp(new, empty_zero_page, PAGE_SIZE));
-                spin_lock(&mm->page_table_lock);
-                /*
-                 * Because we dropped the lock, we should re-check the
-                 * entry, as somebody else could have populated it..
-                 */
-                if (pud_present(*dir)) {
-                        if (new)
-                                kmem_cache_free(zero_cache, new);
-                } else {
-                        struct page *ptepage;
-                        if (! new)
-                                return NULL;
-                        ptepage = virt_to_page(new);
-                        ptepage->mapping = (void *) mm;
-                        ptepage->index = addr & HUGEPGDIR_MASK;
-                        pud_populate(mm, dir, new);
                }
        }
-        return hugepte_offset(dir, addr);
+        return NULL;
 }
-pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
+#define HUGEPTE_BATCH_SIZE      (HPAGE_SIZE / PMD_SIZE)
-{
-        pud_t *pud;
-        BUG_ON(! in_hugepage_area(mm->context, addr));
+void set_huge_pte_at(struct mm_struct *mm, unsigned long addr,
+                     pte_t *ptep, pte_t pte)
+{
+        int i;
-        pud = hugepgd_offset(mm, addr);
+        if (pte_present(*ptep)) {
-        if (! pud)
+                pte_clear(mm, addr, ptep);
-                return NULL;
+                flush_tlb_pending();
+        }
-        return hugepte_offset(pud, addr);
+        for (i = 0; i < HUGEPTE_BATCH_SIZE; i++) {
+                *ptep = __pte(pte_val(pte) & ~_PAGE_HPTEFLAGS);
+                ptep++;
+        }
 }
-pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
+pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr,
+                              pte_t *ptep)
 {
-        pud_t *pud;
+        unsigned long old = pte_update(ptep, ~0UL);
+        int i;
-        BUG_ON(! in_hugepage_area(mm->context, addr));
+        if (old & _PAGE_HASHPTE)
+                hpte_update(mm, addr, old, 0);
-        pud = hugepgd_alloc(mm, addr);
+        for (i = 1; i < HUGEPTE_BATCH_SIZE; i++)
-        if (! pud)
+                ptep[i] = __pte(0);
-                return NULL;
-        return hugepte_alloc(mm, pud, addr);
+        return __pte(old);
 }
 /*
@@ -162,29 +132,53 @@ int is_aligned_hugepage_range(unsigned long addr, unsigned long len)
        return 0;
 }
-static void flush_segments(void *parm)
+static void flush_low_segments(void *parm)
 {
-        u16 segs = (unsigned long) parm;
+        u16 areas = (unsigned long) parm;
        unsigned long i;
        asm volatile("isync" : : : "memory");
-        for (i = 0; i < 16; i++) {
+        BUILD_BUG_ON((sizeof(areas)*8) != NUM_LOW_AREAS);
-                if (! (segs & (1U << i)))
+        for (i = 0; i < NUM_LOW_AREAS; i++) {
+                if (! (areas & (1U << i)))
                        continue;
-                asm volatile("slbie %0" : : "r" (i << SID_SHIFT));
+                asm volatile("slbie %0"
+                             : : "r" ((i << SID_SHIFT) | SLBIE_C));
        }
        asm volatile("isync" : : : "memory");
 }
-static int prepare_low_seg_for_htlb(struct mm_struct *mm, unsigned long seg)
+static void flush_high_segments(void *parm)
 {
-        unsigned long start = seg << SID_SHIFT;
+        u16 areas = (unsigned long) parm;
-        unsigned long end = (seg+1) << SID_SHIFT;
+        unsigned long i, j;
+        asm volatile("isync" : : : "memory");
+        BUILD_BUG_ON((sizeof(areas)*8) != NUM_HIGH_AREAS);
+        for (i = 0; i < NUM_HIGH_AREAS; i++) {
+                if (! (areas & (1U << i)))
+                        continue;
+                for (j = 0; j < (1UL << (HTLB_AREA_SHIFT-SID_SHIFT)); j++)
+                        asm volatile("slbie %0"
+                                     :: "r" (((i << HTLB_AREA_SHIFT)
+                                             + (j << SID_SHIFT)) | SLBIE_C));
+        }
+        asm volatile("isync" : : : "memory");
+}
+static int prepare_low_area_for_htlb(struct mm_struct *mm, unsigned long area)
+{
+        unsigned long start = area << SID_SHIFT;
+        unsigned long end = (area+1) << SID_SHIFT;
        struct vm_area_struct *vma;
-        BUG_ON(seg >= 16);
+        BUG_ON(area >= NUM_LOW_AREAS);
        /* Check no VMAs are in the region */
        vma = find_vma(mm, start);
@@ -194,20 +188,39 @@ static int prepare_low_seg_for_htlb(struct mm_struct *mm, unsigned long seg)
        return 0;
 }
-static int open_low_hpage_segs(struct mm_struct *mm, u16 newsegs)
+static int prepare_high_area_for_htlb(struct mm_struct *mm, unsigned long area)
+{
+        unsigned long start = area << HTLB_AREA_SHIFT;
+        unsigned long end = (area+1) << HTLB_AREA_SHIFT;
+        struct vm_area_struct *vma;
+        BUG_ON(area >= NUM_HIGH_AREAS);
+        /* Check no VMAs are in the region */
+        vma = find_vma(mm, start);
+        if (vma && (vma->vm_start < end))
+                return -EBUSY;
+        return 0;
+}
+static int open_low_hpage_areas(struct mm_struct *mm, u16 newareas)
 {
        unsigned long i;
-        newsegs &= ~(mm->context.htlb_segs);
+        BUILD_BUG_ON((sizeof(newareas)*8) != NUM_LOW_AREAS);
-        if (! newsegs)
+        BUILD_BUG_ON((sizeof(mm->context.low_htlb_areas)*8) != NUM_LOW_AREAS);
+        newareas &= ~(mm->context.low_htlb_areas);
+        if (! newareas)
                return 0; /* The segments we want are already open */
-        for (i = 0; i < 16; i++)
+        for (i = 0; i < NUM_LOW_AREAS; i++)
-                if ((1 << i) & newsegs)
+                if ((1 << i) & newareas)
-                        if (prepare_low_seg_for_htlb(mm, i) != 0)
+                        if (prepare_low_area_for_htlb(mm, i) != 0)
                                return -EBUSY;
-        mm->context.htlb_segs |= newsegs;
+        mm->context.low_htlb_areas |= newareas;
        /* update the paca copy of the context struct */
        get_paca()->context = mm->context;
@@ -215,29 +228,63 @@ static int open_low_hpage_segs(struct mm_struct *mm, u16 newsegs)
        /* the context change must make it to memory before the flush,
         * so that further SLB misses do the right thing. */
        mb();
-        on_each_cpu(flush_segments, (void *)(unsigned long)newsegs, 0, 1);
+        on_each_cpu(flush_low_segments, (void *)(unsigned long)newareas, 0, 1);
+        return 0;
+}
+static int open_high_hpage_areas(struct mm_struct *mm, u16 newareas)
+{
+        unsigned long i;
+        BUILD_BUG_ON((sizeof(newareas)*8) != NUM_HIGH_AREAS);
+        BUILD_BUG_ON((sizeof(mm->context.high_htlb_areas)*8)
+                     != NUM_HIGH_AREAS);
+        newareas &= ~(mm->context.high_htlb_areas);
+        if (! newareas)
+                return 0; /* The areas we want are already open */
+        for (i = 0; i < NUM_HIGH_AREAS; i++)
+                if ((1 << i) & newareas)
+                        if (prepare_high_area_for_htlb(mm, i) != 0)
+                                return -EBUSY;
+        mm->context.high_htlb_areas |= newareas;
+        /* update the paca copy of the context struct */
+        get_paca()->context = mm->context;
+        /* the context change must make it to memory before the flush,
+         * so that further SLB misses do the right thing. */
+        mb();
+        on_each_cpu(flush_high_segments, (void *)(unsigned long)newareas, 0, 1);
        return 0;
 }
 int prepare_hugepage_range(unsigned long addr, unsigned long len)
 {
-        if (within_hugepage_high_range(addr, len))
+        int err;
-                return 0;
-        else if ((addr < 0x100000000UL) && ((addr+len) < 0x100000000UL)) {
+        if ( (addr+len) < addr )
-                int err;
+                return -EINVAL;
-                /* Yes, we need both tests, in case addr+len overflows
-                 * 64-bit arithmetic */
+        if ((addr + len) < 0x100000000UL)
-                err = open_low_hpage_segs(current->mm,
+                err = open_low_hpage_areas(current->mm,
                                          LOW_ESID_MASK(addr, len));
-                if (err)
+        else
-                        printk(KERN_DEBUG "prepare_hugepage_range(%lx, %lx)"
+                err = open_high_hpage_areas(current->mm,
-                               " failed (segs: 0x%04hx)\n", addr, len,
+                                            HTLB_AREA_MASK(addr, len));
-                               LOW_ESID_MASK(addr, len));
+        if (err) {
+                printk(KERN_DEBUG "prepare_hugepage_range(%lx, %lx)"
+                       " failed (lowmask: 0x%04hx, highmask: 0x%04hx)\n",
+                       addr, len,
+                       LOW_ESID_MASK(addr, len), HTLB_AREA_MASK(addr, len));
                return err;
        }
-        return -EINVAL;
+        return 0;
 }
 struct page *
@@ -309,8 +356,8 @@ full_search:
                        vma = find_vma(mm, addr);
                        continue;
                }
-                if (touches_hugepage_high_range(addr, len)) {
+                if (touches_hugepage_high_range(mm, addr, len)) {
-                        addr = TASK_HPAGE_END;
+                        addr = ALIGN(addr+1, 1UL<<HTLB_AREA_SHIFT);
                        vma = find_vma(mm, addr);
                        continue;
                }
@@ -389,8 +436,9 @@ hugepage_recheck:
                if (touches_hugepage_low_range(mm, addr, len)) {
                        addr = (addr & ((~0) << SID_SHIFT)) - len;
                        goto hugepage_recheck;
-                } else if (touches_hugepage_high_range(addr, len)) {
+                } else if (touches_hugepage_high_range(mm, addr, len)) {
-                        addr = TASK_HPAGE_BASE - len;
+                        addr = (addr & ((~0UL) << HTLB_AREA_SHIFT)) - len;
+                        goto hugepage_recheck;
                }
                /*
@@ -481,23 +529,28 @@ static unsigned long htlb_get_low_area(unsigned long len, u16 segmask)
        return -ENOMEM;
 }
-static unsigned long htlb_get_high_area(unsigned long len)
+static unsigned long htlb_get_high_area(unsigned long len, u16 areamask)
 {
-        unsigned long addr = TASK_HPAGE_BASE;
+        unsigned long addr = 0x100000000UL;
        struct vm_area_struct *vma;
        vma = find_vma(current->mm, addr);
-        for (vma = find_vma(current->mm, addr);
+        while (addr + len <= TASK_SIZE_USER64) {
-             addr + len <= TASK_HPAGE_END;
-             vma = vma->vm_next) {
                BUG_ON(vma && (addr >= vma->vm_end)); /* invariant */
-                BUG_ON(! within_hugepage_high_range(addr, len));
+                if (! __within_hugepage_high_range(addr, len, areamask)) {
+                        addr = ALIGN(addr+1, 1UL<<HTLB_AREA_SHIFT);
+                        vma = find_vma(current->mm, addr);
+                        continue;
+                }
                if (!vma || (addr + len) <= vma->vm_start)
                        return addr;
                addr = ALIGN(vma->vm_end, HPAGE_SIZE);
-                /* Because we're in a hugepage region, this alignment
+                /* Depending on segmask this might not be a confirmed
-                 * should not skip us over any VMAs */
+                 * hugepage region, so the ALIGN could have skipped
+                 * some VMAs */
+                vma = find_vma(current->mm, addr);
        }
        return -ENOMEM;
@@ -507,6 +560,9 @@ unsigned long hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
                                        unsigned long len, unsigned long pgoff,
                                        unsigned long flags)
 {
+        int lastshift;
+        u16 areamask, curareas;
        if (len & ~HPAGE_MASK)
                return -EINVAL;
@@ -514,67 +570,49 @@ unsigned long hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
                return -EINVAL;
        if (test_thread_flag(TIF_32BIT)) {
-                int lastshift = 0;
+                curareas = current->mm->context.low_htlb_areas;
-                u16 segmask, cursegs = current->mm->context.htlb_segs;
                /* First see if we can do the mapping in the existing
-                 * low hpage segments */
+                 * low areas */
-                addr = htlb_get_low_area(len, cursegs);
+                addr = htlb_get_low_area(len, curareas);
                if (addr != -ENOMEM)
                        return addr;
-                for (segmask = LOW_ESID_MASK(0x100000000UL-len, len);
+                lastshift = 0;
-                     ! lastshift; segmask >>=1) {
+                for (areamask = LOW_ESID_MASK(0x100000000UL-len, len);
-                        if (segmask & 1)
+                     ! lastshift; areamask >>=1) {
+                        if (areamask & 1)
                                lastshift = 1;
-                        addr = htlb_get_low_area(len, cursegs | segmask);
+                        addr = htlb_get_low_area(len, curareas | areamask);
                        if ((addr != -ENOMEM)
-                            && open_low_hpage_segs(current->mm, segmask) == 0)
+                            && open_low_hpage_areas(current->mm, areamask) == 0)
                                return addr;
                }
-                printk(KERN_DEBUG "hugetlb_get_unmapped_area() unable to open"
-                       " enough segments\n");
-                return -ENOMEM;
        } else {
-                return htlb_get_high_area(len);
+                curareas = current->mm->context.high_htlb_areas;
-        }
-}
-void hugetlb_mm_free_pgd(struct mm_struct *mm)
-{
-        int i;
-        pgd_t *pgdir;
-        spin_lock(&mm->page_table_lock);
-        pgdir = mm->context.huge_pgdir;
-        if (! pgdir)
-                goto out;
-        mm->context.huge_pgdir = NULL;
-        /* cleanup any hugepte pages leftover */
+                /* First see if we can do the mapping in the existing
-        for (i = 0; i < PTRS_PER_HUGEPGD; i++) {
+                 * high areas */
-                pud_t *pud = (pud_t *)(pgdir + i);
+                addr = htlb_get_high_area(len, curareas);
+                if (addr != -ENOMEM)
-                if (! pud_none(*pud)) {
+                        return addr;
-                        pte_t *pte = (pte_t *)pud_page(*pud);
-                        struct page *ptepage = virt_to_page(pte);
-                        ptepage->mapping = NULL;
+                lastshift = 0;
+                for (areamask = HTLB_AREA_MASK(TASK_SIZE_USER64-len, len);
+                     ! lastshift; areamask >>=1) {
+                        if (areamask & 1)
+                                lastshift = 1;
-                        BUG_ON(memcmp(pte, empty_zero_page, PAGE_SIZE));
+                        addr = htlb_get_high_area(len, curareas | areamask);
-                        kmem_cache_free(zero_cache, pte);
+                        if ((addr != -ENOMEM)
+                            && open_high_hpage_areas(current->mm, areamask) == 0)
+                                return addr;
                }
-                pud_clear(pud);
        }
+        printk(KERN_DEBUG "hugetlb_get_unmapped_area() unable to open"
-        BUG_ON(memcmp(pgdir, empty_zero_page, PAGE_SIZE));
+               " enough areas\n");
-        kmem_cache_free(zero_cache, pgdir);
+        return -ENOMEM;
- out:
-        spin_unlock(&mm->page_table_lock);
 }
 int hash_huge_page(struct mm_struct *mm, unsigned long access,
diff --git a/arch/ppc64/mm/imalloc.c b/arch/ppc64/mm/imalloc.c
index b6e75b891ac0..c65b87b92756 100644
--- a/arch/ppc64/mm/imalloc.c
+++ b/arch/ppc64/mm/imalloc.c
@@ -31,7 +31,7 @@ static int get_free_im_addr(unsigned long size, unsigned long *im_addr)
                        break;
                if ((unsigned long)tmp->addr >= ioremap_bot)
                        addr = tmp->size + (unsigned long) tmp->addr;
-                if (addr > IMALLOC_END-size) 
+                if (addr >= IMALLOC_END-size)
                        return 1;
        }
        *im_addr = addr;
diff --git a/arch/ppc64/mm/init.c b/arch/ppc64/mm/init.c
index e58a24d42879..c2157c9c3acb 100644
--- a/arch/ppc64/mm/init.c
+++ b/arch/ppc64/mm/init.c
@@ -42,7 +42,6 @@
 #include <asm/pgalloc.h>
 #include <asm/page.h>
-#include <asm/abs_addr.h>
 #include <asm/prom.h>
 #include <asm/lmb.h>
 #include <asm/rtas.h>
@@ -66,6 +65,14 @@
 #include <asm/vdso.h>
 #include <asm/imalloc.h>
+#if PGTABLE_RANGE > USER_VSID_RANGE
+#warning Limited user VSID range means pagetable space is wasted
+#endif
+#if (TASK_SIZE_USER64 < PGTABLE_RANGE) && (TASK_SIZE_USER64 < USER_VSID_RANGE)
+#warning TASK_SIZE is smaller than it needs to be.
+#endif
 int mem_init_done;
 unsigned long ioremap_bot = IMALLOC_BASE;
 static unsigned long phbs_io_bot = PHBS_IO_BASE;
@@ -159,7 +166,6 @@ static int map_io_page(unsigned long ea, unsigned long pa, int flags)
                ptep = pte_alloc_kernel(&init_mm, pmdp, ea);
                if (!ptep)
                        return -ENOMEM;
-                pa = abs_to_phys(pa);
                set_pte_at(&init_mm, ea, ptep, pfn_pte(pa >> PAGE_SHIFT,
                                                          __pgprot(flags)));
                spin_unlock(&init_mm.page_table_lock);
@@ -226,7 +232,7 @@ void __iomem * __ioremap(unsigned long addr, unsigned long size,
         * Before that, we map using addresses going
         * up from ioremap_bot.  imalloc will use
         * the addresses from ioremap_bot through
-         * IMALLOC_END (0xE000001fffffffff)
+         * IMALLOC_END
         * 
         */
        pa = addr & PAGE_MASK;
@@ -386,6 +392,7 @@ void free_initmem(void)
        addr = (unsigned long)__init_begin;
        for (; addr < (unsigned long)__init_end; addr += PAGE_SIZE) {
+                memset((void *)addr, 0xcc, PAGE_SIZE);
                ClearPageReserved(virt_to_page(addr));
                set_page_count(virt_to_page(addr), 1);
                free_page(addr);
@@ -417,12 +424,6 @@ int init_new_context(struct task_struct *tsk, struct mm_struct *mm)
        int index;
        int err;
-#ifdef CONFIG_HUGETLB_PAGE
-        /* We leave htlb_segs as it was, but for a fork, we need to
-         * clear the huge_pgdir. */
-        mm->context.huge_pgdir = NULL;
-#endif
 again:
        if (!idr_pre_get(&mmu_context_idr, GFP_KERNEL))
                return -ENOMEM;
@@ -453,8 +454,6 @@ void destroy_context(struct mm_struct *mm)
        spin_unlock(&mmu_context_lock);
        mm->context.id = NO_CONTEXT;
-        hugetlb_mm_free_pgd(mm);
 }
 /*
@@ -484,9 +483,9 @@ void __init mm_init_ppc64(void)
        for (i = 1; i < lmb.memory.cnt; i++) {
                unsigned long base, prevbase, prevsize;
-                prevbase = lmb.memory.region[i-1].physbase;
+                prevbase = lmb.memory.region[i-1].base;
                prevsize = lmb.memory.region[i-1].size;
-                base = lmb.memory.region[i].physbase;
+                base = lmb.memory.region[i].base;
                if (base > (prevbase + prevsize)) {
                        io_hole_start = prevbase + prevsize;
                        io_hole_size = base  - (prevbase + prevsize);
@@ -513,11 +512,8 @@ int page_is_ram(unsigned long pfn)
        for (i=0; i < lmb.memory.cnt; i++) {
                unsigned long base;
-#ifdef CONFIG_MSCHUNKS
-                base = lmb.memory.region[i].physbase;
-#else
                base = lmb.memory.region[i].base;
-#endif
                if ((paddr >= base) &&
                        (paddr < (base + lmb.memory.region[i].size))) {
                        return 1;
@@ -547,7 +543,7 @@ void __init do_init_bootmem(void)
         */
        bootmap_pages = bootmem_bootmap_pages(total_pages);
-        start = abs_to_phys(lmb_alloc(bootmap_pages<<PAGE_SHIFT, PAGE_SIZE));
+        start = lmb_alloc(bootmap_pages<<PAGE_SHIFT, PAGE_SIZE);
        BUG_ON(!start);
        boot_mapsize = init_bootmem(start >> PAGE_SHIFT, total_pages);
@@ -557,27 +553,18 @@ void __init do_init_bootmem(void)
        /* Add all physical memory to the bootmem map, mark each area
         * present.
         */
-        for (i=0; i < lmb.memory.cnt; i++) {
+        for (i=0; i < lmb.memory.cnt; i++)
-                unsigned long physbase, size;
+                free_bootmem(lmb.memory.region[i].base,
-                unsigned long start_pfn, end_pfn;
+                             lmb_size_bytes(&lmb.memory, i));
-                physbase = lmb.memory.region[i].physbase;
-                size = lmb.memory.region[i].size;
-                start_pfn = physbase >> PAGE_SHIFT;
-                end_pfn = start_pfn + (size >> PAGE_SHIFT);
-                memory_present(0, start_pfn, end_pfn);
-                free_bootmem(physbase, size);
-        }
        /* reserve the sections we're already using */
-        for (i=0; i < lmb.reserved.cnt; i++) {
+        for (i=0; i < lmb.reserved.cnt; i++)
-                unsigned long physbase = lmb.reserved.region[i].physbase;
+                reserve_bootmem(lmb.reserved.region[i].base,
-                unsigned long size = lmb.reserved.region[i].size;
+                                lmb_size_bytes(&lmb.reserved, i));
-                reserve_bootmem(physbase, size);
+        for (i=0; i < lmb.memory.cnt; i++)
-        }
+                memory_present(0, lmb_start_pfn(&lmb.memory, i),
+                               lmb_end_pfn(&lmb.memory, i));
 }
 /*
@@ -615,10 +602,10 @@ static int __init setup_kcore(void)
        int i;
        for (i=0; i < lmb.memory.cnt; i++) {
-                unsigned long physbase, size;
+                unsigned long base, size;
                struct kcore_list *kcore_mem;
-                physbase = lmb.memory.region[i].physbase;
+                base = lmb.memory.region[i].base;
                size = lmb.memory.region[i].size;
                /* GFP_ATOMIC to avoid might_sleep warnings during boot */
@@ -626,7 +613,7 @@ static int __init setup_kcore(void)
                if (!kcore_mem)
                        panic("mem_init: kmalloc failed\n");
-                kclist_add(kcore_mem, __va(physbase), size);
+                kclist_add(kcore_mem, __va(base), size);
        }
        kclist_add(&kcore_vmem, (void *)VMALLOC_START, VMALLOC_END-VMALLOC_START);
@@ -686,9 +673,6 @@ void __init mem_init(void)
        mem_init_done = 1;
-#ifdef CONFIG_PPC_ISERIES
-        iommu_vio_init();
-#endif
        /* Initialize the vDSO */
        vdso_init();
 }
@@ -833,23 +817,43 @@ void __iomem * reserve_phb_iospace(unsigned long size)
        return virt_addr;
 }
-kmem_cache_t *zero_cache;
+static void zero_ctor(void *addr, kmem_cache_t *cache, unsigned long flags)
-static void zero_ctor(void *pte, kmem_cache_t *cache, unsigned long flags)
 {
-        memset(pte, 0, PAGE_SIZE);
+        memset(addr, 0, kmem_cache_size(cache));
 }
+static const int pgtable_cache_size[2] = {
+        PTE_TABLE_SIZE, PMD_TABLE_SIZE
+};
+static const char *pgtable_cache_name[ARRAY_SIZE(pgtable_cache_size)] = {
+        "pgd_pte_cache", "pud_pmd_cache",
+};
+kmem_cache_t *pgtable_cache[ARRAY_SIZE(pgtable_cache_size)];
 void pgtable_cache_init(void)
 {
-        zero_cache = kmem_cache_create("zero",
+        int i;
-                                PAGE_SIZE,
-                                0,
+        BUILD_BUG_ON(PTE_TABLE_SIZE != pgtable_cache_size[PTE_CACHE_NUM]);
-                                SLAB_HWCACHE_ALIGN | SLAB_MUST_HWCACHE_ALIGN,
+        BUILD_BUG_ON(PMD_TABLE_SIZE != pgtable_cache_size[PMD_CACHE_NUM]);
-                                zero_ctor,
+        BUILD_BUG_ON(PUD_TABLE_SIZE != pgtable_cache_size[PUD_CACHE_NUM]);
-                                NULL);
+        BUILD_BUG_ON(PGD_TABLE_SIZE != pgtable_cache_size[PGD_CACHE_NUM]);
-        if (!zero_cache)
-                panic("pgtable_cache_init(): could not create zero_cache!\n");
+        for (i = 0; i < ARRAY_SIZE(pgtable_cache_size); i++) {
+                int size = pgtable_cache_size[i];
+                const char *name = pgtable_cache_name[i];
+                pgtable_cache[i] = kmem_cache_create(name,
+                                                     size, size,
+                                                     SLAB_HWCACHE_ALIGN
+                                                     | SLAB_MUST_HWCACHE_ALIGN,
+                                                     zero_ctor,
+                                                     NULL);
+                if (! pgtable_cache[i])
+                        panic("pgtable_cache_init(): could not create %s!\n",
+                              name);
+        }
 }
 pgprot_t phys_mem_access_prot(struct file *file, unsigned long addr,
diff --git a/arch/ppc64/mm/numa.c b/arch/ppc64/mm/numa.c
index 0b191f2de016..cb864b8f2750 100644
--- a/arch/ppc64/mm/numa.c
+++ b/arch/ppc64/mm/numa.c
@@ -440,8 +440,6 @@ new_range:
                for (i = start ; i < (start+size); i += MEMORY_INCREMENT)
                        numa_memory_lookup_table[i >> MEMORY_INCREMENT_SHIFT] =
                                numa_domain;
-                memory_present(numa_domain, start >> PAGE_SHIFT,
-                                                (start + size) >> PAGE_SHIFT);
                if (--ranges)
                        goto new_range;
@@ -483,7 +481,6 @@ static void __init setup_nonnuma(void)
        for (i = 0 ; i < top_of_ram; i += MEMORY_INCREMENT)
                numa_memory_lookup_table[i >> MEMORY_INCREMENT_SHIFT] = 0;
-        memory_present(0, 0, init_node_data[0].node_end_pfn);
 }
 static void __init dump_numa_topology(void)
@@ -671,7 +668,7 @@ new_range:
                 * Mark reserved regions on this node
                 */
                for (i = 0; i < lmb.reserved.cnt; i++) {
-                        unsigned long physbase = lmb.reserved.region[i].physbase;
+                        unsigned long physbase = lmb.reserved.region[i].base;
                        unsigned long size = lmb.reserved.region[i].size;
                        if (pa_to_nid(physbase) != nid &&
@@ -695,6 +692,46 @@ new_range:
                                                     size);
                        }
                }
+                /*
+                 * This loop may look famaliar, but we have to do it again
+                 * after marking our reserved memory to mark memory present
+                 * for sparsemem.
+                 */
+                addr_cells = get_mem_addr_cells();
+                size_cells = get_mem_size_cells();
+                memory = NULL;
+                while ((memory = of_find_node_by_type(memory, "memory")) != NULL) {
+                        unsigned long mem_start, mem_size;
+                        int numa_domain, ranges;
+                        unsigned int *memcell_buf;
+                        unsigned int len;
+                        memcell_buf = (unsigned int *)get_property(memory, "reg", &len);
+                        if (!memcell_buf || len <= 0)
+                                continue;
+                        ranges = memory->n_addrs;       /* ranges in cell */
+new_range2:
+                        mem_start = read_n_cells(addr_cells, &memcell_buf);
+                        mem_size = read_n_cells(size_cells, &memcell_buf);
+                        if (numa_enabled) {
+                                numa_domain = of_node_numa_domain(memory);
+                                if (numa_domain  >= MAX_NUMNODES)
+                                        numa_domain = 0;
+                        } else
+                                numa_domain =  0;
+                        if (numa_domain != nid)
+                                continue;
+                        mem_size = numa_enforce_memory_limit(mem_start, mem_size);
+                        memory_present(numa_domain, mem_start >> PAGE_SHIFT,
+                                       (mem_start + mem_size) >> PAGE_SHIFT);
+                        if (--ranges)           /* process all ranges in cell */
+                                goto new_range2;
+                }
        }
 }
diff --git a/arch/ppc64/mm/slb.c b/arch/ppc64/mm/slb.c
index 244150a0bc18..0473953f6a37 100644
--- a/arch/ppc64/mm/slb.c
+++ b/arch/ppc64/mm/slb.c
@@ -87,8 +87,8 @@ void switch_slb(struct task_struct *tsk, struct mm_struct *mm)
                int i;
                asm volatile("isync" : : : "memory");
                for (i = 0; i < offset; i++) {
-                        esid_data = (unsigned long)get_paca()->slb_cache[i]
+                        esid_data = ((unsigned long)get_paca()->slb_cache[i]
-                                << SID_SHIFT;
+                                << SID_SHIFT) | SLBIE_C;
                        asm volatile("slbie %0" : : "r" (esid_data));
                }
                asm volatile("isync" : : : "memory");
diff --git a/arch/ppc64/mm/slb_low.S b/arch/ppc64/mm/slb_low.S
index 8379d678f70f..a3a03da503bc 100644
--- a/arch/ppc64/mm/slb_low.S
+++ b/arch/ppc64/mm/slb_low.S
@@ -21,7 +21,7 @@
 #include <asm/page.h>
 #include <asm/mmu.h>
 #include <asm/ppc_asm.h>
-#include <asm/offsets.h>
+#include <asm/asm-offsets.h>
 #include <asm/cputable.h>
 /* void slb_allocate(unsigned long ea);
@@ -89,32 +89,29 @@ END_FTR_SECTION_IFSET(CPU_FTR_16M_PAGE)
        b       9f
 0:      /* user address: proto-VSID = context<<15 | ESID */
-        li      r11,SLB_VSID_USER
+        srdi.   r9,r3,USER_ESID_BITS
-        srdi.   r9,r3,13
        bne-    8f                      /* invalid ea bits set */
 #ifdef CONFIG_HUGETLB_PAGE
 BEGIN_FTR_SECTION
-        /* check against the hugepage ranges */
+        lhz     r9,PACAHIGHHTLBAREAS(r13)
-        cmpldi  r3,(TASK_HPAGE_END>>SID_SHIFT)
+        srdi    r11,r3,(HTLB_AREA_SHIFT-SID_SHIFT)
-        bge     6f                      /* >= TASK_HPAGE_END */
+        srd     r9,r9,r11
-        cmpldi  r3,(TASK_HPAGE_BASE>>SID_SHIFT)
+        lhz     r11,PACALOWHTLBAREAS(r13)
-        bge     5f                      /* TASK_HPAGE_BASE..TASK_HPAGE_END */
+        srd     r11,r11,r3
-        cmpldi  r3,16
+        or      r9,r9,r11
-        bge     6f                      /* 4GB..TASK_HPAGE_BASE */
+END_FTR_SECTION_IFSET(CPU_FTR_16M_PAGE)
+#endif /* CONFIG_HUGETLB_PAGE */
-        lhz     r9,PACAHTLBSEGS(r13)
-        srd     r9,r9,r3
+        li      r11,SLB_VSID_USER
-        andi.   r9,r9,1
-        beq     6f
+#ifdef CONFIG_HUGETLB_PAGE
+BEGIN_FTR_SECTION
-5:      /* this is a hugepage user address */
+        rldimi  r11,r9,8,55             /* shift masked bit into SLB_VSID_L */
-        li      r11,(SLB_VSID_USER|SLB_VSID_L)
 END_FTR_SECTION_IFSET(CPU_FTR_16M_PAGE)
 #endif /* CONFIG_HUGETLB_PAGE */
-6:      ld      r9,PACACONTEXTID(r13)
+        ld      r9,PACACONTEXTID(r13)
        rldimi  r3,r9,USER_ESID_BITS,0
 9:      /* r3 = protovsid, r11 = flags, r10 = esid_data, cr7 = <>KERNELBASE */
diff --git a/arch/ppc64/mm/tlb.c b/arch/ppc64/mm/tlb.c
index 26f0172c4527..d8a6593a13f0 100644
--- a/arch/ppc64/mm/tlb.c
+++ b/arch/ppc64/mm/tlb.c
@@ -41,7 +41,58 @@ DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
 DEFINE_PER_CPU(struct pte_freelist_batch *, pte_freelist_cur);
 unsigned long pte_freelist_forced_free;
-void __pte_free_tlb(struct mmu_gather *tlb, struct page *ptepage)
+struct pte_freelist_batch
+{
+        struct rcu_head rcu;
+        unsigned int    index;
+        pgtable_free_t  tables[0];
+};
+DEFINE_PER_CPU(struct pte_freelist_batch *, pte_freelist_cur);
+unsigned long pte_freelist_forced_free;
+#define PTE_FREELIST_SIZE \
+        ((PAGE_SIZE - sizeof(struct pte_freelist_batch)) \
+          / sizeof(pgtable_free_t))
+#ifdef CONFIG_SMP
+static void pte_free_smp_sync(void *arg)
+{
+        /* Do nothing, just ensure we sync with all CPUs */
+}
+#endif
+/* This is only called when we are critically out of memory
+ * (and fail to get a page in pte_free_tlb).
+ */
+static void pgtable_free_now(pgtable_free_t pgf)
+{
+        pte_freelist_forced_free++;
+        smp_call_function(pte_free_smp_sync, NULL, 0, 1);
+        pgtable_free(pgf);
+}
+static void pte_free_rcu_callback(struct rcu_head *head)
+{
+        struct pte_freelist_batch *batch =
+                container_of(head, struct pte_freelist_batch, rcu);
+        unsigned int i;
+        for (i = 0; i < batch->index; i++)
+                pgtable_free(batch->tables[i]);
+        free_page((unsigned long)batch);
+}
+static void pte_free_submit(struct pte_freelist_batch *batch)
+{
+        INIT_RCU_HEAD(&batch->rcu);
+        call_rcu(&batch->rcu, pte_free_rcu_callback);
+}
+void pgtable_free_tlb(struct mmu_gather *tlb, pgtable_free_t pgf)
 {
        /* This is safe as we are holding page_table_lock */
        cpumask_t local_cpumask = cpumask_of_cpu(smp_processor_id());
@@ -49,19 +100,19 @@ void __pte_free_tlb(struct mmu_gather *tlb, struct page *ptepage)
        if (atomic_read(&tlb->mm->mm_users) < 2 ||
            cpus_equal(tlb->mm->cpu_vm_mask, local_cpumask)) {
-                pte_free(ptepage);
+                pgtable_free(pgf);
                return;
        }
        if (*batchp == NULL) {
                *batchp = (struct pte_freelist_batch *)__get_free_page(GFP_ATOMIC);
                if (*batchp == NULL) {
-                        pte_free_now(ptepage);
+                        pgtable_free_now(pgf);
                        return;
                }
                (*batchp)->index = 0;
        }
-        (*batchp)->pages[(*batchp)->index++] = ptepage;
+        (*batchp)->tables[(*batchp)->index++] = pgf;
        if ((*batchp)->index == PTE_FREELIST_SIZE) {
                pte_free_submit(*batchp);
                *batchp = NULL;
@@ -132,42 +183,6 @@ void __flush_tlb_pending(struct ppc64_tlb_batch *batch)
        put_cpu();
 }
-#ifdef CONFIG_SMP
-static void pte_free_smp_sync(void *arg)
-{
-        /* Do nothing, just ensure we sync with all CPUs */
-}
-#endif
-/* This is only called when we are critically out of memory
- * (and fail to get a page in pte_free_tlb).
- */
-void pte_free_now(struct page *ptepage)
-{
-        pte_freelist_forced_free++;
-        smp_call_function(pte_free_smp_sync, NULL, 0, 1);
-        pte_free(ptepage);
-}
-static void pte_free_rcu_callback(struct rcu_head *head)
-{
-        struct pte_freelist_batch *batch =
-                container_of(head, struct pte_freelist_batch, rcu);
-        unsigned int i;
-        for (i = 0; i < batch->index; i++)
-                pte_free(batch->pages[i]);
-        free_page((unsigned long)batch);
-}
-void pte_free_submit(struct pte_freelist_batch *batch)
-{
-        INIT_RCU_HEAD(&batch->rcu);
-        call_rcu(&batch->rcu, pte_free_rcu_callback);
-}
 void pte_free_finish(void)
 {
        /* This is safe as we are holding page_table_lock */
diff --git a/arch/ppc64/oprofile/common.c b/arch/ppc64/oprofile/common.c
index b28bfda23d94..e5f572710aa0 100644
--- a/arch/ppc64/oprofile/common.c
+++ b/arch/ppc64/oprofile/common.c
@@ -16,11 +16,9 @@
 #include <asm/ptrace.h>
 #include <asm/system.h>
 #include <asm/pmc.h>
+#include <asm/cputable.h>
+#include <asm/oprofile_impl.h>
-#include "op_impl.h"
-extern struct op_ppc64_model op_model_rs64;
-extern struct op_ppc64_model op_model_power4;
 static struct op_ppc64_model *model;
 static struct op_counter_config ctr[OP_MAX_COUNTER];
@@ -123,52 +121,13 @@ static int op_ppc64_create_files(struct super_block *sb, struct dentry *root)
 int __init oprofile_arch_init(struct oprofile_operations *ops)
 {
-        unsigned int pvr;
+        if (!cur_cpu_spec->oprofile_model || !cur_cpu_spec->oprofile_cpu_type)
+                return -ENODEV;
-        pvr = mfspr(SPRN_PVR);
+        model = cur_cpu_spec->oprofile_model;
-        switch (PVR_VER(pvr)) {
+        model->num_counters = cur_cpu_spec->num_pmcs;
-                case PV_630:
-                case PV_630p:
-                        model = &op_model_rs64;
-                        model->num_counters = 8;
-                        ops->cpu_type = "ppc64/power3";
-                        break;
-                case PV_NORTHSTAR:
-                case PV_PULSAR:
-                case PV_ICESTAR:
-                case PV_SSTAR:
-                        model = &op_model_rs64;
-                        model->num_counters = 8;
-                        ops->cpu_type = "ppc64/rs64";
-                        break;
-                case PV_POWER4:
-                case PV_POWER4p:
-                        model = &op_model_power4;
-                        model->num_counters = 8;
-                        ops->cpu_type = "ppc64/power4";
-                        break;
-                case PV_970:
-                case PV_970FX:
-                        model = &op_model_power4;
-                        model->num_counters = 8;
-                        ops->cpu_type = "ppc64/970";
-                        break;
-                case PV_POWER5:
-                case PV_POWER5p:
-                        model = &op_model_power4;
-                        model->num_counters = 6;
-                        ops->cpu_type = "ppc64/power5";
-                        break;
-                default:
-                        return -ENODEV;
-        }
+        ops->cpu_type = cur_cpu_spec->oprofile_cpu_type;
        ops->create_files = op_ppc64_create_files;
        ops->setup = op_ppc64_setup;
        ops->shutdown = op_ppc64_shutdown;
diff --git a/arch/ppc64/oprofile/op_impl.h b/arch/ppc64/oprofile/op_impl.h
deleted file mode 100644
index 7fa7eaabc035..000000000000
--- a/arch/ppc64/oprofile/op_impl.h
+++ /dev/null
@@ -1,108 +0,0 @@
-/*
- * Copyright (C) 2004 Anton Blanchard <anton@au.ibm.com>, IBM
- *
- * Based on alpha version.
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation; either version
- * 2 of the License, or (at your option) any later version.
- */
-#ifndef OP_IMPL_H
-#define OP_IMPL_H 1
-#define OP_MAX_COUNTER 8
-/* Per-counter configuration as set via oprofilefs.  */
-struct op_counter_config {
-        unsigned long valid;
-        unsigned long enabled;
-        unsigned long event;
-        unsigned long count;
-        unsigned long kernel;
-        /* We dont support per counter user/kernel selection */
-        unsigned long user;
-        unsigned long unit_mask;
-};
-/* System-wide configuration as set via oprofilefs.  */
-struct op_system_config {
-        unsigned long mmcr0;
-        unsigned long mmcr1;
-        unsigned long mmcra;
-        unsigned long enable_kernel;
-        unsigned long enable_user;
-        unsigned long backtrace_spinlocks;
-};
-/* Per-arch configuration */
-struct op_ppc64_model {
-        void (*reg_setup) (struct op_counter_config *,
-                           struct op_system_config *,
-                           int num_counters);
-        void (*cpu_setup) (void *);
-        void (*start) (struct op_counter_config *);
-        void (*stop) (void);
-        void (*handle_interrupt) (struct pt_regs *,
-                                  struct op_counter_config *);
-        int num_counters;
-};
-static inline unsigned int ctr_read(unsigned int i)
-{
-        switch(i) {
-        case 0:
-                return mfspr(SPRN_PMC1);
-        case 1:
-                return mfspr(SPRN_PMC2);
-        case 2:
-                return mfspr(SPRN_PMC3);
-        case 3:
-                return mfspr(SPRN_PMC4);
-        case 4:
-                return mfspr(SPRN_PMC5);
-        case 5:
-                return mfspr(SPRN_PMC6);
-        case 6:
-                return mfspr(SPRN_PMC7);
-        case 7:
-                return mfspr(SPRN_PMC8);
-        default:
-                return 0;
-        }
-}
-static inline void ctr_write(unsigned int i, unsigned int val)
-{
-        switch(i) {
-        case 0:
-                mtspr(SPRN_PMC1, val);
-                break;
-        case 1:
-                mtspr(SPRN_PMC2, val);
-                break;
-        case 2:
-                mtspr(SPRN_PMC3, val);
-                break;
-        case 3:
-                mtspr(SPRN_PMC4, val);
-                break;
-        case 4:
-                mtspr(SPRN_PMC5, val);
-                break;
-        case 5:
-                mtspr(SPRN_PMC6, val);
-                break;
-        case 6:
-                mtspr(SPRN_PMC7, val);
-                break;
-        case 7:
-                mtspr(SPRN_PMC8, val);
-                break;
-        default:
-                break;
-        }
-}
-#endif
diff --git a/arch/ppc64/oprofile/op_model_power4.c b/arch/ppc64/oprofile/op_model_power4.c
index 3d103d66870d..32b2bb5625fe 100644
--- a/arch/ppc64/oprofile/op_model_power4.c
+++ b/arch/ppc64/oprofile/op_model_power4.c
@@ -16,14 +16,12 @@
 #include <asm/cputable.h>
 #include <asm/systemcfg.h>
 #include <asm/rtas.h>
+#include <asm/oprofile_impl.h>
 #define dbg(args...)
-#include "op_impl.h"
 static unsigned long reset_value[OP_MAX_COUNTER];
-static int num_counters;
 static int oprofile_running;
 static int mmcra_has_sihv;
@@ -45,8 +43,6 @@ static void power4_reg_setup(struct op_counter_config *ctr,
 {
        int i;
-        num_counters = num_ctrs;
        /*
         * SIHV / SIPR bits are only implemented on POWER4+ (GQ) and above.
         * However we disable it on all POWER4 until we verify it works
@@ -68,7 +64,7 @@ static void power4_reg_setup(struct op_counter_config *ctr,
        backtrace_spinlocks = sys->backtrace_spinlocks;
-        for (i = 0; i < num_counters; ++i)
+        for (i = 0; i < cur_cpu_spec->num_pmcs; ++i)
                reset_value[i] = 0x80000000UL - ctr[i].count;
        /* setup user and kernel profiling */
@@ -121,7 +117,7 @@ static void power4_start(struct op_counter_config *ctr)
        /* set the PMM bit (see comment below) */
        mtmsrd(mfmsr() | MSR_PMM);
-        for (i = 0; i < num_counters; ++i) {
+        for (i = 0; i < cur_cpu_spec->num_pmcs; ++i) {
                if (ctr[i].enabled) {
                        ctr_write(i, reset_value[i]);
                } else {
@@ -272,7 +268,7 @@ static void power4_handle_interrupt(struct pt_regs *regs,
        /* set the PMM bit (see comment below) */
        mtmsrd(mfmsr() | MSR_PMM);
-        for (i = 0; i < num_counters; ++i) {
+        for (i = 0; i < cur_cpu_spec->num_pmcs; ++i) {
                val = ctr_read(i);
                if (val < 0) {
                        if (oprofile_running && ctr[i].enabled) {
diff --git a/arch/ppc64/oprofile/op_model_rs64.c b/arch/ppc64/oprofile/op_model_rs64.c
index bcec506c266a..08c5b333f5c4 100644
--- a/arch/ppc64/oprofile/op_model_rs64.c
+++ b/arch/ppc64/oprofile/op_model_rs64.c
@@ -14,11 +14,10 @@
 #include <asm/system.h>
 #include <asm/processor.h>
 #include <asm/cputable.h>
+#include <asm/oprofile_impl.h>
 #define dbg(args...)
-#include "op_impl.h"
 static void ctrl_write(unsigned int i, unsigned int val)
 {
        unsigned int tmp = 0;
diff --git a/arch/ppc64/xmon/privinst.h b/arch/ppc64/xmon/privinst.h
index 183c3e400258..02eb40dac0b3 100644
--- a/arch/ppc64/xmon/privinst.h
+++ b/arch/ppc64/xmon/privinst.h
@@ -46,7 +46,6 @@ GSETSPR(287, pvr)
 GSETSPR(1008, hid0)
 GSETSPR(1009, hid1)
 GSETSPR(1010, iabr)
-GSETSPR(1013, dabr)
 GSETSPR(1023, pir)
 static inline void store_inst(void *p)
diff --git a/arch/ppc64/xmon/start.c b/arch/ppc64/xmon/start.c
index a9265bcc79b2..e50c158191e1 100644
--- a/arch/ppc64/xmon/start.c
+++ b/arch/ppc64/xmon/start.c
@@ -27,7 +27,7 @@ static void sysrq_handle_xmon(int key, struct pt_regs *pt_regs,
                              struct tty_struct *tty) 
 {
        /* ensure xmon is enabled */
-        xmon_init();
+        xmon_init(1);
        debugger(pt_regs);
 }
@@ -61,7 +61,9 @@ xmon_read(void *handle, void *ptr, int nb)
 int
 xmon_read_poll(void)
 {
-        return udbg_getc_poll();
+        if (udbg_getc_poll)
+                return udbg_getc_poll();
+        return -1;
 }
 
 FILE *xmon_stdin;
diff --git a/arch/ppc64/xmon/xmon.c b/arch/ppc64/xmon/xmon.c
index 05539439e6bc..74e63a886a69 100644
--- a/arch/ppc64/xmon/xmon.c
+++ b/arch/ppc64/xmon/xmon.c
@@ -586,6 +586,8 @@ int xmon_dabr_match(struct pt_regs *regs)
 {
        if ((regs->msr & (MSR_IR|MSR_PR|MSR_SF)) != (MSR_IR|MSR_SF))
                return 0;
+        if (dabr.enabled == 0)
+                return 0;
        xmon_core(regs, 0);
        return 1;
 }
@@ -628,20 +630,6 @@ int xmon_fault_handler(struct pt_regs *regs)
        return 0;
 }
-/* On systems with a hypervisor, we can't set the DABR
-   (data address breakpoint register) directly. */
-static void set_controlled_dabr(unsigned long val)
-{
-#ifdef CONFIG_PPC_PSERIES
-        if (systemcfg->platform == PLATFORM_PSERIES_LPAR) {
-                int rc = plpar_hcall_norets(H_SET_DABR, val);
-                if (rc != H_Success)
-                        xmon_printf("Warning: setting DABR failed (%d)\n", rc);
-        } else
-#endif
-                set_dabr(val);
-}
 static struct bpt *at_breakpoint(unsigned long pc)
 {
        int i;
@@ -728,7 +716,7 @@ static void insert_bpts(void)
 static void insert_cpu_bpts(void)
 {
        if (dabr.enabled)
-                set_controlled_dabr(dabr.address | (dabr.enabled & 7));
+                set_dabr(dabr.address | (dabr.enabled & 7));
        if (iabr && cpu_has_feature(CPU_FTR_IABR))
                set_iabr(iabr->address
                         | (iabr->enabled & (BP_IABR|BP_IABR_TE)));
@@ -756,7 +744,7 @@ static void remove_bpts(void)
 static void remove_cpu_bpts(void)
 {
-        set_controlled_dabr(0);
+        set_dabr(0);
        if (cpu_has_feature(CPU_FTR_IABR))
                set_iabr(0);
 }
@@ -2496,15 +2484,25 @@ static void dump_stab(void)
        }
 }
-void xmon_init(void)
+void xmon_init(int enable)
 {
-        __debugger = xmon;
+        if (enable) {
-        __debugger_ipi = xmon_ipi;
+                __debugger = xmon;
-        __debugger_bpt = xmon_bpt;
+                __debugger_ipi = xmon_ipi;
-        __debugger_sstep = xmon_sstep;
+                __debugger_bpt = xmon_bpt;
-        __debugger_iabr_match = xmon_iabr_match;
+                __debugger_sstep = xmon_sstep;
-        __debugger_dabr_match = xmon_dabr_match;
+                __debugger_iabr_match = xmon_iabr_match;
-        __debugger_fault_handler = xmon_fault_handler;
+                __debugger_dabr_match = xmon_dabr_match;
+                __debugger_fault_handler = xmon_fault_handler;
+        } else {
+                __debugger = NULL;
+                __debugger_ipi = NULL;
+                __debugger_bpt = NULL;
+                __debugger_sstep = NULL;
+                __debugger_iabr_match = NULL;
+                __debugger_dabr_match = NULL;
+                __debugger_fault_handler = NULL;
+        }
 }
 void dump_segments(void)