Stop dhcpcd before starting if it was running

[people/pmueller/ipfire-2.x.git] / src / patches / 60036_xen3-patch-2.6.27.patch1

From: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: [PATCH] Linux: Update to 2.6.27
Patch-mainline: 2.6.27

 This patch contains the differences between Linux 2.6.26 and 2.6.27.

Acked-by: Jeff Mahoney <jeffm@suse.com>
Automatically created from "patches.kernel.org/patch-2.6.27" by xen-port-patches.py

Index: head-2008-12-01/arch/x86/Kconfig
===================================================================
--- head-2008-12-01.orig/arch/x86/Kconfig       2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/Kconfig    2008-12-01 11:49:07.000000000 +0100
@@ -590,7 +590,7 @@ config CALGARY_IOMMU_ENABLED_BY_DEFAULT
 config AMD_IOMMU
        bool "AMD IOMMU support"
        select SWIOTLB
-       depends on X86_64 && PCI && ACPI
+       depends on X86_64 && PCI && ACPI && !X86_64_XEN
        help
          With this option you can enable support for AMD IOMMU hardware in
          your system. An IOMMU is a hardware component which provides
@@ -625,8 +625,10 @@ config MAXSMP
 
 config NR_CPUS
        int "Maximum number of CPUs (2-512)" if !MAXSMP
+       range 2 32 if XEN
        range 2 512
        depends on SMP
+       default "32" if MAXSMP && XEN
        default "4096" if MAXSMP
        default "32" if X86_NUMAQ || X86_SUMMIT || X86_BIGSMP || X86_ES7000
        default "16" if X86_64_XEN
@@ -1223,7 +1225,7 @@ config MTRR
 config MTRR_SANITIZER
        bool
        prompt "MTRR cleanup support"
-       depends on MTRR
+       depends on MTRR && !XEN
        help
          Convert MTRR layout from continuous to discrete, so X drivers can
          add writeback entries.
Index: head-2008-12-01/arch/x86/Kconfig.debug
===================================================================
--- head-2008-12-01.orig/arch/x86/Kconfig.debug 2008-12-01 11:37:10.000000000 +0100
+++ head-2008-12-01/arch/x86/Kconfig.debug      2008-12-01 11:49:07.000000000 +0100
@@ -25,6 +25,7 @@ config STRICT_DEVMEM
 config X86_VERBOSE_BOOTUP
        bool "Enable verbose x86 bootup info messages"
        default y
+       depends on !XEN
        help
          Enables the informational output from the decompression stage
          (e.g. bzImage) of the boot. If you disable this you will still
@@ -166,7 +167,7 @@ config MMIOTRACE_HOOKS
 
 config MMIOTRACE
        bool "Memory mapped IO tracing"
-       depends on DEBUG_KERNEL && PCI
+       depends on DEBUG_KERNEL && PCI && !XEN
        select TRACING
        select MMIOTRACE_HOOKS
        help
Index: head-2008-12-01/arch/x86/Makefile
===================================================================
--- head-2008-12-01.orig/arch/x86/Makefile      2008-12-01 11:36:55.000000000 +0100
+++ head-2008-12-01/arch/x86/Makefile   2008-12-01 11:49:07.000000000 +0100
@@ -116,8 +116,8 @@ mflags-$(CONFIG_X86_VOYAGER)        := -Iinclud
 mcore-$(CONFIG_X86_VOYAGER)    := arch/x86/mach-voyager/
 
 # Xen subarch support
-mflags-$(CONFIG_X86_XEN)       := -Iinclude/asm-x86/mach-xen
-mcore-$(CONFIG_X86_XEN)                := arch/x86/mach-xen/
+mflags-$(CONFIG_XEN)           := -Iinclude/asm-x86/mach-xen
+mcore-$(CONFIG_XEN)            := arch/x86/mach-xen/
 
 # generic subarchitecture
 mflags-$(CONFIG_X86_GENERICARCH):= -Iinclude/asm-x86/mach-generic
@@ -128,8 +128,10 @@ mcore-$(CONFIG_X86_GENERICARCH)    := arch/
 mflags-y += -Iinclude/asm-x86/mach-default
 
 # 64 bit does not support subarch support - clear sub arch variables
+ifneq ($(CONFIG_XEN),y)
 fcore-$(CONFIG_X86_64)  :=
 mcore-$(CONFIG_X86_64)  :=
+endif
 
 KBUILD_CFLAGS += $(mflags-y)
 KBUILD_AFLAGS += $(mflags-y)
Index: head-2008-12-01/arch/x86/ia32/ia32entry-xen.S
===================================================================
--- head-2008-12-01.orig/arch/x86/ia32/ia32entry-xen.S  2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/ia32/ia32entry-xen.S       2008-12-01 11:49:07.000000000 +0100
@@ -15,6 +15,16 @@
 #include <asm/irqflags.h>
 #include <linux/linkage.h>
 
+/* Avoid __ASSEMBLER__'ifying <linux/audit.h> just for this.  */
+#include <linux/elf-em.h>
+#define AUDIT_ARCH_I386                (EM_386|__AUDIT_ARCH_LE)
+#define __AUDIT_ARCH_LE           0x40000000
+
+#ifndef CONFIG_AUDITSYSCALL
+#define sysexit_audit int_ret_from_sys_call
+#define sysretl_audit int_ret_from_sys_call
+#endif
+
 #define IA32_NR_syscalls ((ia32_syscall_end - ia32_sys_call_table)/8)
 
        .macro IA32_ARG_FIXUP noebp=0
@@ -37,6 +47,11 @@
        movq    %rax,R8(%rsp)
        .endm
 
+       /*
+        * Reload arg registers from stack in case ptrace changed them.
+        * We don't reload %eax because syscall_trace_enter() returned
+        * the value it wants us to use in the table lookup.
+        */
        .macro LOAD_ARGS32 offset
        movl \offset(%rsp),%r11d
        movl \offset+8(%rsp),%r10d
@@ -46,7 +61,6 @@
        movl \offset+48(%rsp),%edx
        movl \offset+56(%rsp),%esi
        movl \offset+64(%rsp),%edi
-       movl \offset+72(%rsp),%eax
        .endm
 
        .macro CFI_STARTPROC32 simple
@@ -61,6 +75,19 @@
        CFI_UNDEFINED   r15
        .endm
 
+#ifdef CONFIG_PARAVIRT
+ENTRY(native_usergs_sysret32)
+       swapgs
+       sysretl
+ENDPROC(native_usergs_sysret32)
+
+ENTRY(native_irq_enable_sysexit)
+       swapgs
+       sti
+       sysexit
+ENDPROC(native_irq_enable_sysexit)
+#endif
+
 /*
  * 32bit SYSENTER instruction entry.
  *
@@ -98,7 +125,7 @@ ENTRY(ia32_sysenter_target)
        CFI_RESTORE     rcx
        movl    %ebp,%ebp               /* zero extension */
        movl    %eax,%eax
-       movl    48-THREAD_SIZE+threadinfo_sysenter_return(%rsp),%r10d
+       movl    48-THREAD_SIZE+TI_sysenter_return(%rsp),%r10d
        movl    $__USER32_DS,40(%rsp)
        movq    %rbp,32(%rsp)
        movl    $__USER32_CS,16(%rsp)
@@ -113,19 +140,79 @@ ENTRY(ia32_sysenter_target)
        .quad 1b,ia32_badarg
        .previous       
        GET_THREAD_INFO(%r10)
-       orl    $TS_COMPAT,threadinfo_status(%r10)
-       testl  $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SECCOMP),threadinfo_flags(%r10)
+       orl    $TS_COMPAT,TI_status(%r10)
+       testl  $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%r10)
        jnz  sysenter_tracesys
-sysenter_do_call:      
        cmpl    $(IA32_NR_syscalls-1),%eax
        ja      ia32_badsys
+sysenter_do_call:
        IA32_ARG_FIXUP 1
+sysenter_dispatch:
        call    *ia32_sys_call_table(,%rax,8)
        movq    %rax,RAX-ARGOFFSET(%rsp)
+       GET_THREAD_INFO(%r10)
+       DISABLE_INTERRUPTS(CLBR_NONE)
+       TRACE_IRQS_OFF
+       testl   $_TIF_ALLWORK_MASK,TI_flags(%r10)
+       jnz     sysexit_audit
+       jmp int_ret_from_sys_call
+
+#ifdef CONFIG_AUDITSYSCALL
+       .macro auditsys_entry_common
+       movl %esi,%r9d                  /* 6th arg: 4th syscall arg */
+       movl %edx,%r8d                  /* 5th arg: 3rd syscall arg */
+       /* (already in %ecx)               4th arg: 2nd syscall arg */
+       movl %ebx,%edx                  /* 3rd arg: 1st syscall arg */
+       movl %eax,%esi                  /* 2nd arg: syscall number */
+       movl $AUDIT_ARCH_I386,%edi      /* 1st arg: audit arch */
+       call audit_syscall_entry
+       movl RAX-ARGOFFSET(%rsp),%eax   /* reload syscall number */
+       cmpl $(IA32_NR_syscalls-1),%eax
+       ja ia32_badsys
+       movl %ebx,%edi                  /* reload 1st syscall arg */
+       movl RCX-ARGOFFSET(%rsp),%esi   /* reload 2nd syscall arg */
+       movl RDX-ARGOFFSET(%rsp),%edx   /* reload 3rd syscall arg */
+       movl RSI-ARGOFFSET(%rsp),%ecx   /* reload 4th syscall arg */
+       movl RDI-ARGOFFSET(%rsp),%r8d   /* reload 5th syscall arg */
+       .endm
+
+       .macro auditsys_exit exit,ebpsave=RBP
+       testl $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT),TI_flags(%r10)
+       jnz int_ret_from_sys_call
+       TRACE_IRQS_ON
+       ENABLE_INTERRUPTS(CLBR_NONE)
+       movl %eax,%esi          /* second arg, syscall return value */
+       cmpl $0,%eax            /* is it < 0? */
+       setl %al                /* 1 if so, 0 if not */
+       movzbl %al,%edi         /* zero-extend that into %edi */
+       inc %edi /* first arg, 0->1(AUDITSC_SUCCESS), 1->2(AUDITSC_FAILURE) */
+       call audit_syscall_exit
+       GET_THREAD_INFO(%r10)
+       movl RAX-ARGOFFSET(%rsp),%eax   /* reload syscall return value */
+       movl \ebpsave-ARGOFFSET(%rsp),%ebp /* reload user register value */
+       movl $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT),%edi
+       DISABLE_INTERRUPTS(CLBR_NONE)
+       TRACE_IRQS_OFF
+       testl %edi,TI_flags(%r10)
+       jnz int_with_check
        jmp int_ret_from_sys_call
+       .endm
+
+sysenter_auditsys:
+       auditsys_entry_common
+       movl %ebp,%r9d                  /* reload 6th syscall arg */
+       jmp sysenter_dispatch
+
+sysexit_audit:
+       auditsys_exit sysexit_from_sys_call
+#endif
 
 sysenter_tracesys:
        xchgl   %r9d,%ebp
+#ifdef CONFIG_AUDITSYSCALL
+       testl   $(_TIF_WORK_SYSCALL_ENTRY & ~_TIF_SYSCALL_AUDIT),TI_flags(%r10)
+       jz      sysenter_auditsys
+#endif
        SAVE_REST
        CLEAR_RREGS
        movq    %r9,R9(%rsp)
@@ -186,18 +273,38 @@ ENTRY(ia32_cstar_target)
        .quad 1b,ia32_badarg
        .previous       
        GET_THREAD_INFO(%r10)
-       orl   $TS_COMPAT,threadinfo_status(%r10)
-       testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SECCOMP),threadinfo_flags(%r10)
+       orl   $TS_COMPAT,TI_status(%r10)
+       testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%r10)
        jnz   cstar_tracesys
 cstar_do_call: 
        cmpl $IA32_NR_syscalls-1,%eax
        ja  ia32_badsys
        IA32_ARG_FIXUP 1
+cstar_dispatch:
        call *ia32_sys_call_table(,%rax,8)
        movq %rax,RAX-ARGOFFSET(%rsp)
+       GET_THREAD_INFO(%r10)
+       DISABLE_INTERRUPTS(CLBR_NONE)
+       testl $_TIF_ALLWORK_MASK,TI_flags(%r10)
+       jnz sysretl_audit
        jmp int_ret_from_sys_call
        
-cstar_tracesys:        
+#ifdef CONFIG_AUDITSYSCALL
+cstar_auditsys:
+       movl %r9d,R9-ARGOFFSET(%rsp)    /* register to be clobbered by call */
+       auditsys_entry_common
+       movl R9-ARGOFFSET(%rsp),%r9d    /* reload 6th syscall arg */
+       jmp cstar_dispatch
+
+sysretl_audit:
+       auditsys_exit sysretl_from_sys_call, RCX /* user %ebp in RCX slot */
+#endif
+
+cstar_tracesys:
+#ifdef CONFIG_AUDITSYSCALL
+       testl $(_TIF_WORK_SYSCALL_ENTRY & ~_TIF_SYSCALL_AUDIT),TI_flags(%r10)
+       jz cstar_auditsys
+#endif
        xchgl %r9d,%ebp
        SAVE_REST
        CLEAR_RREGS
@@ -263,8 +370,8 @@ ENTRY(ia32_syscall)
           this could be a problem. */
        SAVE_ARGS 0,0,1
        GET_THREAD_INFO(%r10)
-       orl   $TS_COMPAT,threadinfo_status(%r10)
-       testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SECCOMP),threadinfo_flags(%r10)
+       orl   $TS_COMPAT,TI_status(%r10)
+       testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%r10)
        jnz ia32_tracesys
 ia32_do_syscall:       
        cmpl $(IA32_NR_syscalls-1),%eax
@@ -309,13 +416,11 @@ quiet_ni_syscall:
        PTREGSCALL stub32_rt_sigreturn, sys32_rt_sigreturn, %rdi
        PTREGSCALL stub32_sigreturn, sys32_sigreturn, %rdi
        PTREGSCALL stub32_sigaltstack, sys32_sigaltstack, %rdx
-       PTREGSCALL stub32_sigsuspend, sys32_sigsuspend, %rcx
        PTREGSCALL stub32_execve, sys32_execve, %rcx
        PTREGSCALL stub32_fork, sys_fork, %rdi
        PTREGSCALL stub32_clone, sys32_clone, %rdx
        PTREGSCALL stub32_vfork, sys_vfork, %rdi
        PTREGSCALL stub32_iopl, sys_iopl, %rsi
-       PTREGSCALL stub32_rt_sigsuspend, sys_rt_sigsuspend, %rdx
 
 ENTRY(ia32_ptregs_common)
        popq %r11
@@ -415,7 +520,7 @@ ia32_sys_call_table:
        .quad sys_ssetmask
        .quad sys_setreuid16    /* 70 */
        .quad sys_setregid16
-       .quad stub32_sigsuspend
+       .quad sys32_sigsuspend
        .quad compat_sys_sigpending
        .quad sys_sethostname
        .quad compat_sys_setrlimit      /* 75 */
@@ -522,7 +627,7 @@ ia32_sys_call_table:
        .quad sys32_rt_sigpending
        .quad compat_sys_rt_sigtimedwait
        .quad sys32_rt_sigqueueinfo
-       .quad stub32_rt_sigsuspend
+       .quad sys_rt_sigsuspend
        .quad sys32_pread               /* 180 */
        .quad sys32_pwrite
        .quad sys_chown16
@@ -670,4 +775,10 @@ ia32_sys_call_table:
        .quad sys32_fallocate
        .quad compat_sys_timerfd_settime        /* 325 */
        .quad compat_sys_timerfd_gettime
+       .quad compat_sys_signalfd4
+       .quad sys_eventfd2
+       .quad sys_epoll_create1
+       .quad sys_dup3                  /* 330 */
+       .quad sys_pipe2
+       .quad sys_inotify_init1
 ia32_syscall_end:
Index: head-2008-12-01/arch/x86/kernel/Makefile
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/Makefile       2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/Makefile    2008-12-01 11:49:07.000000000 +0100
@@ -120,9 +120,10 @@ ifeq ($(CONFIG_X86_64),y)
 
         obj-$(CONFIG_PCI_MMCONFIG)     += mmconf-fam10h_64.o
 
-       obj-$(CONFIG_XEN)               += nmi_64.o
+       obj-$(CONFIG_XEN)               += nmi.o
        time_64-$(CONFIG_XEN)           += time_32.o
 endif
 
-disabled-obj-$(CONFIG_XEN) := early-quirks.o hpet.o i8253.o i8259_$(BITS).o \
-       pci-swiotlb_64.o reboot.o smpboot.o tlb_$(BITS).o tsc_$(BITS).o tsc_sync.o vsmp_64.o
+disabled-obj-$(CONFIG_XEN) := bios_uv.o early-quirks.o hpet.o i8253.o \
+       i8259.o irqinit_$(BITS).o pci-swiotlb_64.o reboot.o smpboot.o \
+       tlb_$(BITS).o tsc.o tsc_sync.o vsmp_64.o
Index: head-2008-12-01/arch/x86/kernel/acpi/boot.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/acpi/boot.c    2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/acpi/boot.c 2008-12-01 11:49:07.000000000 +0100
@@ -951,7 +951,9 @@ void __init mp_register_ioapic(int id, u
        mp_ioapics[idx].mp_flags = MPC_APIC_USABLE;
        mp_ioapics[idx].mp_apicaddr = address;
 
+#ifndef CONFIG_XEN
        set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address);
+#endif
        mp_ioapics[idx].mp_apicid = uniq_ioapic_id(id);
 #ifdef CONFIG_X86_32
        mp_ioapics[idx].mp_apicver = io_apic_get_version(idx);
@@ -1108,7 +1110,7 @@ int mp_register_gsi(u32 gsi, int trigger
 {
        int ioapic;
        int ioapic_pin;
-#ifdef CONFIG_X86_32
+#if defined(CONFIG_X86_32) && !defined(CONFIG_XEN)
 #define MAX_GSI_NUM    4096
 #define IRQ_COMPRESSION_START  64
 
@@ -1156,7 +1158,7 @@ int mp_register_gsi(u32 gsi, int trigger
        if (test_bit(ioapic_pin, mp_ioapic_routing[ioapic].pin_programmed)) {
                pr_debug(KERN_DEBUG "Pin %d-%d already programmed\n",
                         mp_ioapic_routing[ioapic].apic_id, ioapic_pin);
-#ifdef CONFIG_X86_32
+#if defined(CONFIG_X86_32) && !defined(CONFIG_XEN)
                return (gsi < IRQ_COMPRESSION_START ? gsi : gsi_to_irq[gsi]);
 #else
                return gsi;
@@ -1164,7 +1166,7 @@ int mp_register_gsi(u32 gsi, int trigger
        }
 
        set_bit(ioapic_pin, mp_ioapic_routing[ioapic].pin_programmed);
-#ifdef CONFIG_X86_32
+#if defined(CONFIG_X86_32) && !defined(CONFIG_XEN)
        /*
         * For GSI >= 64, use IRQ compression
         */
Index: head-2008-12-01/arch/x86/kernel/acpi/sleep-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/acpi/sleep-xen.c       2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/acpi/sleep-xen.c    2008-12-01 11:49:07.000000000 +0100
@@ -9,6 +9,7 @@
 #include <linux/bootmem.h>
 #include <linux/dmi.h>
 #include <linux/cpumask.h>
+#include <asm/segment.h>
 
 #include "realmode/wakeup.h"
 #include "sleep.h"
@@ -20,7 +21,7 @@ unsigned long acpi_realmode_flags;
 /* address in low memory of the wakeup routine. */
 static unsigned long acpi_realmode;
 
-#ifdef CONFIG_64BIT
+#if defined(CONFIG_SMP) && defined(CONFIG_64BIT)
 static char temp_stack[10240];
 #endif
 #endif
@@ -54,18 +55,27 @@ int acpi_save_state_mem(void)
        header->video_mode = saved_video_mode;
 
        header->wakeup_jmp_seg = acpi_wakeup_address >> 4;
+
+       /*
+        * Set up the wakeup GDT.  We set these up as Big Real Mode,
+        * that is, with limits set to 4 GB.  At least the Lenovo
+        * Thinkpad X61 is known to need this for the video BIOS
+        * initialization quirk to work; this is likely to also
+        * be the case for other laptops or integrated video devices.
+        */
+
        /* GDT[0]: GDT self-pointer */
        header->wakeup_gdt[0] =
                (u64)(sizeof(header->wakeup_gdt) - 1) +
                ((u64)(acpi_wakeup_address +
                        ((char *)&header->wakeup_gdt - (char *)acpi_realmode))
                                << 16);
-       /* GDT[1]: real-mode-like code segment */
-       header->wakeup_gdt[1] = (0x009bULL << 40) +
-               ((u64)acpi_wakeup_address << 16) + 0xffff;
-       /* GDT[2]: real-mode-like data segment */
-       header->wakeup_gdt[2] = (0x0093ULL << 40) +
-               ((u64)acpi_wakeup_address << 16) + 0xffff;
+       /* GDT[1]: big real mode-like code segment */
+       header->wakeup_gdt[1] =
+               GDT_ENTRY(0x809b, acpi_wakeup_address, 0xfffff);
+       /* GDT[2]: big real mode-like data segment */
+       header->wakeup_gdt[2] =
+               GDT_ENTRY(0x8093, acpi_wakeup_address, 0xfffff);
 
 #ifndef CONFIG_64BIT
        store_gdt((struct desc_ptr *)&header->pmode_gdt);
@@ -79,7 +89,7 @@ int acpi_save_state_mem(void)
 #endif /* !CONFIG_64BIT */
 
        header->pmode_cr0 = read_cr0();
-       header->pmode_cr4 = read_cr4();
+       header->pmode_cr4 = read_cr4_safe();
        header->realmode_flags = acpi_realmode_flags;
        header->real_magic = 0x12345678;
 
@@ -89,7 +99,9 @@ int acpi_save_state_mem(void)
        saved_magic = 0x12345678;
 #else /* CONFIG_64BIT */
        header->trampoline_segment = setup_trampoline() >> 4;
-       init_rsp = (unsigned long)temp_stack + 4096;
+#ifdef CONFIG_SMP
+       stack_start.sp = temp_stack + 4096;
+#endif
        initial_code = (unsigned long)wakeup_long64;
        saved_magic = 0x123456789abcdef0;
 #endif /* CONFIG_64BIT */
@@ -145,6 +157,12 @@ static int __init acpi_sleep_setup(char 
                        acpi_realmode_flags |= 2;
                if (strncmp(str, "s3_beep", 7) == 0)
                        acpi_realmode_flags |= 4;
+#ifdef CONFIG_HIBERNATION
+               if (strncmp(str, "s4_nohwsig", 10) == 0)
+                       acpi_no_s4_hw_signature();
+#endif
+               if (strncmp(str, "old_ordering", 12) == 0)
+                       acpi_old_suspend_ordering();
                str = strchr(str, ',');
                if (str != NULL)
                        str += strspn(str, ", \t");
Index: head-2008-12-01/arch/x86/kernel/apic_32-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/apic_32-xen.c  2008-12-01 11:37:10.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/apic_32-xen.c       2008-12-01 11:49:07.000000000 +0100
@@ -59,7 +59,10 @@ static cpumask_t timer_bcast_ipi;
 /*
  * Debug level, exported for io_apic.c
  */
-int apic_verbosity;
+unsigned int apic_verbosity;
+
+/* Have we found an MP table */
+int smp_found_config;
 
 #ifndef CONFIG_XEN
 static int modern_apic(void)
Index: head-2008-12-01/arch/x86/kernel/apic_64-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/apic_64-xen.c  2008-12-01 11:37:10.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/apic_64-xen.c       2008-12-01 11:49:07.000000000 +0100
@@ -39,7 +39,10 @@ int disable_apic;
 /*
  * Debug level, exported for io_apic.c
  */
-int apic_verbosity;
+unsigned int apic_verbosity;
+
+/* Have we found an MP table */
+int smp_found_config;
 
 /*
  * The guts of the apic timer interrupt
Index: head-2008-12-01/arch/x86/kernel/asm-offsets_64.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/asm-offsets_64.c       2008-12-03 15:48:43.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/asm-offsets_64.c    2008-12-01 11:49:07.000000000 +0100
@@ -138,7 +138,7 @@ int main(void)
 
        BLANK();
        DEFINE(PAGE_SIZE_asm, PAGE_SIZE);
-#ifdef CONFIG_XEN
+#ifdef CONFIG_PARAVIRT_XEN
        BLANK();
        OFFSET(XEN_vcpu_info_mask, vcpu_info, evtchn_upcall_mask);
        OFFSET(XEN_vcpu_info_pending, vcpu_info, evtchn_upcall_pending);
Index: head-2008-12-01/arch/x86/kernel/cpu/amd_64.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/cpu/amd_64.c   2008-12-03 15:48:43.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/cpu/amd_64.c        2008-12-01 11:49:07.000000000 +0100
@@ -193,6 +193,7 @@ static void __cpuinit init_amd(struct cp
                fam10h_check_enable_mmcfg();
        }
 
+#ifndef CONFIG_XEN
        if (c == &boot_cpu_data && c->x86 >= 0xf && c->x86 <= 0x11) {
                unsigned long long tseg;
 
@@ -211,6 +212,7 @@ static void __cpuinit init_amd(struct cp
                        set_memory_4k((unsigned long)__va(tseg), 1);
                }
        }
+#endif
 }
 
 static struct cpu_dev amd_cpu_dev __cpuinitdata = {
Index: head-2008-12-01/arch/x86/kernel/cpu/bugs_64.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/cpu/bugs_64.c  2008-12-03 15:48:43.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/cpu/bugs_64.c       2008-12-01 11:49:07.000000000 +0100
@@ -20,6 +20,7 @@ void __init check_bugs(void)
 #endif
        alternative_instructions();
 
+#ifndef CONFIG_XEN
        /*
         * Make sure the first 2MB area is not mapped by huge pages
         * There are typically fixed size MTRRs in there and overlapping
@@ -30,4 +31,5 @@ void __init check_bugs(void)
         */
        if (!direct_gbpages)
                set_memory_4k((unsigned long)__va(0), 1);
+#endif
 }
Index: head-2008-12-01/arch/x86/kernel/cpu/common-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/cpu/common-xen.c       2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/cpu/common-xen.c    2008-12-01 11:49:07.000000000 +0100
@@ -13,6 +13,7 @@
 #include <asm/mtrr.h>
 #include <asm/mce.h>
 #include <asm/pat.h>
+#include <asm/asm.h>
 #ifdef CONFIG_X86_LOCAL_APIC
 #include <asm/mpspec.h>
 #include <asm/apic.h>
@@ -341,11 +342,24 @@ static void __init early_cpu_detect(void
 
        get_cpu_vendor(c, 1);
 
+       early_get_cap(c);
+
        if (c->x86_vendor != X86_VENDOR_UNKNOWN &&
            cpu_devs[c->x86_vendor]->c_early_init)
                cpu_devs[c->x86_vendor]->c_early_init(c);
+}
 
-       early_get_cap(c);
+/*
+ * The NOPL instruction is supposed to exist on all CPUs with
+ * family >= 6; unfortunately, that's not true in practice because
+ * of early VIA chips and (more importantly) broken virtualizers that
+ * are not easy to detect.  In the latter case it doesn't even *fail*
+ * reliably, so probing for it doesn't even work.  Disable it completely
+ * unless we can find a reliable way to detect all the broken cases.
+ */
+static void __cpuinit detect_nopl(struct cpuinfo_x86 *c)
+{
+       clear_cpu_cap(c, X86_FEATURE_NOPL);
 }
 
 static void __cpuinit generic_identify(struct cpuinfo_x86 *c)
@@ -402,8 +416,8 @@ static void __cpuinit generic_identify(s
                }
 
                init_scattered_cpuid_features(c);
+               detect_nopl(c);
        }
-
 }
 
 static void __cpuinit squash_the_stupid_serial_number(struct cpuinfo_x86 *c)
@@ -434,7 +448,7 @@ __setup("serialnumber", x86_serial_nr_se
 /*
  * This does the hard work of actually picking apart the CPU stuff...
  */
-void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
+static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 {
        int i;
 
@@ -448,6 +462,8 @@ void __cpuinit identify_cpu(struct cpuin
        c->x86_max_cores = 1;
        c->x86_clflush_size = 32;
        memset(&c->x86_capability, 0, sizeof c->x86_capability);
+       if (boot_cpu_has(X86_FEATURE_SYSCALL32))
+               set_cpu_cap(c, X86_FEATURE_SYSCALL32);
 
        if (!have_cpuid_p()) {
                /*
Index: head-2008-12-01/arch/x86/kernel/cpu/common_64-xen.c
===================================================================
--- /dev/null   1970-01-01 00:00:00.000000000 +0000
+++ head-2008-12-01/arch/x86/kernel/cpu/common_64-xen.c 2008-12-01 11:49:07.000000000 +0100
@@ -0,0 +1,771 @@
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/string.h>
+#include <linux/bootmem.h>
+#include <linux/bitops.h>
+#include <linux/module.h>
+#include <linux/kgdb.h>
+#include <linux/topology.h>
+#include <linux/delay.h>
+#include <linux/smp.h>
+#include <linux/percpu.h>
+#include <asm/i387.h>
+#include <asm/msr.h>
+#include <asm/io.h>
+#include <asm/linkage.h>
+#include <asm/mmu_context.h>
+#include <asm/mtrr.h>
+#include <asm/mce.h>
+#include <asm/pat.h>
+#include <asm/asm.h>
+#include <asm/numa.h>
+#ifdef CONFIG_X86_LOCAL_APIC
+#include <asm/mpspec.h>
+#include <asm/apic.h>
+#include <mach_apic.h>
+#elif defined(CONFIG_XEN)
+#include <mach_apic.h>
+#endif
+#include <asm/pda.h>
+#include <asm/pgtable.h>
+#include <asm/processor.h>
+#include <asm/desc.h>
+#include <asm/atomic.h>
+#include <asm/proto.h>
+#include <asm/sections.h>
+#include <asm/setup.h>
+#include <asm/genapic.h>
+
+#include "cpu.h"
+
+/* We need valid kernel segments for data and code in long mode too
+ * IRET will check the segment types  kkeil 2000/10/28
+ * Also sysret mandates a special GDT layout
+ */
+/* The TLS descriptors are currently at a different place compared to i386.
+   Hopefully nobody expects them at a fixed place (Wine?) */
+DEFINE_PER_CPU(struct gdt_page, gdt_page) = { .gdt = {
+       [GDT_ENTRY_KERNEL32_CS] = { { { 0x0000ffff, 0x00cf9b00 } } },
+       [GDT_ENTRY_KERNEL_CS] = { { { 0x0000ffff, 0x00af9b00 } } },
+       [GDT_ENTRY_KERNEL_DS] = { { { 0x0000ffff, 0x00cf9300 } } },
+       [GDT_ENTRY_DEFAULT_USER32_CS] = { { { 0x0000ffff, 0x00cffb00 } } },
+       [GDT_ENTRY_DEFAULT_USER_DS] = { { { 0x0000ffff, 0x00cff300 } } },
+       [GDT_ENTRY_DEFAULT_USER_CS] = { { { 0x0000ffff, 0x00affb00 } } },
+} };
+EXPORT_PER_CPU_SYMBOL_GPL(gdt_page);
+
+__u32 cleared_cpu_caps[NCAPINTS] __cpuinitdata;
+
+/* Current gdt points %fs at the "master" per-cpu area: after this,
+ * it's on the real one. */
+void switch_to_new_gdt(void)
+{
+#ifndef CONFIG_XEN
+       struct desc_ptr gdt_descr;
+
+       gdt_descr.address = (long)get_cpu_gdt_table(smp_processor_id());
+       gdt_descr.size = GDT_SIZE - 1;
+       load_gdt(&gdt_descr);
+#else
+       void *va, *gdt_addr = get_cpu_gdt_table(smp_processor_id());
+       unsigned long frames[16];
+       unsigned int f = 0;
+
+       for (va = gdt_addr; va < gdt_addr + GDT_SIZE; va += PAGE_SIZE) {
+               frames[f++] = virt_to_mfn(va);
+               make_page_readonly(va, XENFEAT_writable_descriptor_tables);
+       }
+       if (HYPERVISOR_set_gdt(frames, GDT_SIZE / sizeof(struct desc_struct)))
+               BUG();
+#endif
+}
+
+struct cpu_dev *cpu_devs[X86_VENDOR_NUM] = {};
+
+static void __cpuinit default_init(struct cpuinfo_x86 *c)
+{
+       display_cacheinfo(c);
+}
+
+static struct cpu_dev __cpuinitdata default_cpu = {
+       .c_init = default_init,
+       .c_vendor = "Unknown",
+};
+static struct cpu_dev *this_cpu __cpuinitdata = &default_cpu;
+
+int __cpuinit get_model_name(struct cpuinfo_x86 *c)
+{
+       unsigned int *v;
+
+       if (c->extended_cpuid_level < 0x80000004)
+               return 0;
+
+       v = (unsigned int *) c->x86_model_id;
+       cpuid(0x80000002, &v[0], &v[1], &v[2], &v[3]);
+       cpuid(0x80000003, &v[4], &v[5], &v[6], &v[7]);
+       cpuid(0x80000004, &v[8], &v[9], &v[10], &v[11]);
+       c->x86_model_id[48] = 0;
+       return 1;
+}
+
+
+void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c)
+{
+       unsigned int n, dummy, ebx, ecx, edx;
+
+       n = c->extended_cpuid_level;
+
+       if (n >= 0x80000005) {
+               cpuid(0x80000005, &dummy, &ebx, &ecx, &edx);
+               printk(KERN_INFO "CPU: L1 I Cache: %dK (%d bytes/line), "
+                      "D cache %dK (%d bytes/line)\n",
+                      edx>>24, edx&0xFF, ecx>>24, ecx&0xFF);
+               c->x86_cache_size = (ecx>>24) + (edx>>24);
+               /* On K8 L1 TLB is inclusive, so don't count it */
+               c->x86_tlbsize = 0;
+       }
+
+       if (n >= 0x80000006) {
+               cpuid(0x80000006, &dummy, &ebx, &ecx, &edx);
+               ecx = cpuid_ecx(0x80000006);
+               c->x86_cache_size = ecx >> 16;
+               c->x86_tlbsize += ((ebx >> 16) & 0xfff) + (ebx & 0xfff);
+
+               printk(KERN_INFO "CPU: L2 Cache: %dK (%d bytes/line)\n",
+               c->x86_cache_size, ecx & 0xFF);
+       }
+}
+
+void __cpuinit detect_ht(struct cpuinfo_x86 *c)
+{
+#ifdef CONFIG_SMP
+       u32 eax, ebx, ecx, edx;
+       int index_msb, core_bits;
+
+       cpuid(1, &eax, &ebx, &ecx, &edx);
+
+
+       if (!cpu_has(c, X86_FEATURE_HT))
+               return;
+       if (cpu_has(c, X86_FEATURE_CMP_LEGACY))
+               goto out;
+
+       smp_num_siblings = (ebx & 0xff0000) >> 16;
+
+       if (smp_num_siblings == 1) {
+               printk(KERN_INFO  "CPU: Hyper-Threading is disabled\n");
+       } else if (smp_num_siblings > 1) {
+
+               if (smp_num_siblings > NR_CPUS) {
+                       printk(KERN_WARNING "CPU: Unsupported number of "
+                              "siblings %d", smp_num_siblings);
+                       smp_num_siblings = 1;
+                       return;
+               }
+
+               index_msb = get_count_order(smp_num_siblings);
+               c->phys_proc_id = phys_pkg_id(index_msb);
+
+               smp_num_siblings = smp_num_siblings / c->x86_max_cores;
+
+               index_msb = get_count_order(smp_num_siblings);
+
+               core_bits = get_count_order(c->x86_max_cores);
+
+               c->cpu_core_id = phys_pkg_id(index_msb) &
+                                              ((1 << core_bits) - 1);
+       }
+out:
+       if ((c->x86_max_cores * smp_num_siblings) > 1) {
+               printk(KERN_INFO  "CPU: Physical Processor ID: %d\n",
+                      c->phys_proc_id);
+               printk(KERN_INFO  "CPU: Processor Core ID: %d\n",
+                      c->cpu_core_id);
+       }
+
+#endif
+}
+
+static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c)
+{
+       char *v = c->x86_vendor_id;
+       int i;
+       static int printed;
+
+       for (i = 0; i < X86_VENDOR_NUM; i++) {
+               if (cpu_devs[i]) {
+                       if (!strcmp(v, cpu_devs[i]->c_ident[0]) ||
+                           (cpu_devs[i]->c_ident[1] &&
+                           !strcmp(v, cpu_devs[i]->c_ident[1]))) {
+                               c->x86_vendor = i;
+                               this_cpu = cpu_devs[i];
+                               return;
+                       }
+               }
+       }
+       if (!printed) {
+               printed++;
+               printk(KERN_ERR "CPU: Vendor unknown, using generic init.\n");
+               printk(KERN_ERR "CPU: Your system may be unstable.\n");
+       }
+       c->x86_vendor = X86_VENDOR_UNKNOWN;
+}
+
+static void __init early_cpu_support_print(void)
+{
+       int i,j;
+       struct cpu_dev *cpu_devx;
+
+       printk("KERNEL supported cpus:\n");
+       for (i = 0; i < X86_VENDOR_NUM; i++) {
+               cpu_devx = cpu_devs[i];
+               if (!cpu_devx)
+                       continue;
+               for (j = 0; j < 2; j++) {
+                       if (!cpu_devx->c_ident[j])
+                               continue;
+                       printk("  %s %s\n", cpu_devx->c_vendor,
+                               cpu_devx->c_ident[j]);
+               }
+       }
+}
+
+/*
+ * The NOPL instruction is supposed to exist on all CPUs with
+ * family >= 6, unfortunately, that's not true in practice because
+ * of early VIA chips and (more importantly) broken virtualizers that
+ * are not easy to detect.  Hence, probe for it based on first
+ * principles.
+ *
+ * Note: no 64-bit chip is known to lack these, but put the code here
+ * for consistency with 32 bits, and to make it utterly trivial to
+ * diagnose the problem should it ever surface.
+ */
+static void __cpuinit detect_nopl(struct cpuinfo_x86 *c)
+{
+       const u32 nopl_signature = 0x888c53b1; /* Random number */
+       u32 has_nopl = nopl_signature;
+
+       clear_cpu_cap(c, X86_FEATURE_NOPL);
+       if (c->x86 >= 6) {
+               asm volatile("\n"
+                            "1:      .byte 0x0f,0x1f,0xc0\n" /* nopl %eax */
+                            "2:\n"
+                            "        .section .fixup,\"ax\"\n"
+                            "3:      xor %0,%0\n"
+                            "        jmp 2b\n"
+                            "        .previous\n"
+                            _ASM_EXTABLE(1b,3b)
+                            : "+a" (has_nopl));
+
+               if (has_nopl == nopl_signature)
+                       set_cpu_cap(c, X86_FEATURE_NOPL);
+       }
+}
+
+static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c);
+
+void __init early_cpu_init(void)
+{
+        struct cpu_vendor_dev *cvdev;
+
+        for (cvdev = __x86cpuvendor_start ;
+             cvdev < __x86cpuvendor_end   ;
+             cvdev++)
+                cpu_devs[cvdev->vendor] = cvdev->cpu_dev;
+       early_cpu_support_print();
+       early_identify_cpu(&boot_cpu_data);
+}
+
+/* Do some early cpuid on the boot CPU to get some parameter that are
+   needed before check_bugs. Everything advanced is in identify_cpu
+   below. */
+static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c)
+{
+       u32 tfms, xlvl;
+
+       c->loops_per_jiffy = loops_per_jiffy;
+       c->x86_cache_size = -1;
+       c->x86_vendor = X86_VENDOR_UNKNOWN;
+       c->x86_model = c->x86_mask = 0; /* So far unknown... */
+       c->x86_vendor_id[0] = '\0'; /* Unset */
+       c->x86_model_id[0] = '\0';  /* Unset */
+       c->x86_clflush_size = 64;
+       c->x86_cache_alignment = c->x86_clflush_size;
+       c->x86_max_cores = 1;
+       c->x86_coreid_bits = 0;
+       c->extended_cpuid_level = 0;
+       memset(&c->x86_capability, 0, sizeof c->x86_capability);
+
+       /* Get vendor name */
+       cpuid(0x00000000, (unsigned int *)&c->cpuid_level,
+             (unsigned int *)&c->x86_vendor_id[0],
+             (unsigned int *)&c->x86_vendor_id[8],
+             (unsigned int *)&c->x86_vendor_id[4]);
+
+       get_cpu_vendor(c);
+
+       /* Initialize the standard set of capabilities */
+       /* Note that the vendor-specific code below might override */
+
+       /* Intel-defined flags: level 0x00000001 */
+       if (c->cpuid_level >= 0x00000001) {
+               __u32 misc;
+               cpuid(0x00000001, &tfms, &misc, &c->x86_capability[4],
+                     &c->x86_capability[0]);
+               c->x86 = (tfms >> 8) & 0xf;
+               c->x86_model = (tfms >> 4) & 0xf;
+               c->x86_mask = tfms & 0xf;
+               if (c->x86 == 0xf)
+                       c->x86 += (tfms >> 20) & 0xff;
+               if (c->x86 >= 0x6)
+                       c->x86_model += ((tfms >> 16) & 0xF) << 4;
+               if (test_cpu_cap(c, X86_FEATURE_CLFLSH))
+                       c->x86_clflush_size = ((misc >> 8) & 0xff) * 8;
+       } else {
+               /* Have CPUID level 0 only - unheard of */
+               c->x86 = 4;
+       }
+
+       c->initial_apicid = (cpuid_ebx(1) >> 24) & 0xff;
+#ifdef CONFIG_SMP
+       c->phys_proc_id = c->initial_apicid;
+#endif
+       /* AMD-defined flags: level 0x80000001 */
+       xlvl = cpuid_eax(0x80000000);
+       c->extended_cpuid_level = xlvl;
+       if ((xlvl & 0xffff0000) == 0x80000000) {
+               if (xlvl >= 0x80000001) {
+                       c->x86_capability[1] = cpuid_edx(0x80000001);
+                       c->x86_capability[6] = cpuid_ecx(0x80000001);
+               }
+               if (xlvl >= 0x80000004)
+                       get_model_name(c); /* Default name */
+       }
+
+       /* Transmeta-defined flags: level 0x80860001 */
+       xlvl = cpuid_eax(0x80860000);
+       if ((xlvl & 0xffff0000) == 0x80860000) {
+               /* Don't set x86_cpuid_level here for now to not confuse. */
+               if (xlvl >= 0x80860001)
+                       c->x86_capability[2] = cpuid_edx(0x80860001);
+       }
+
+       if (c->extended_cpuid_level >= 0x80000007)
+               c->x86_power = cpuid_edx(0x80000007);
+
+       if (c->extended_cpuid_level >= 0x80000008) {
+               u32 eax = cpuid_eax(0x80000008);
+
+               c->x86_virt_bits = (eax >> 8) & 0xff;
+               c->x86_phys_bits = eax & 0xff;
+       }
+
+       detect_nopl(c);
+
+       if (c->x86_vendor != X86_VENDOR_UNKNOWN &&
+           cpu_devs[c->x86_vendor]->c_early_init)
+               cpu_devs[c->x86_vendor]->c_early_init(c);
+
+       validate_pat_support(c);
+}
+
+/*
+ * This does the hard work of actually picking apart the CPU stuff...
+ */
+static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
+{
+       int i;
+
+       early_identify_cpu(c);
+
+       init_scattered_cpuid_features(c);
+
+       c->apicid = phys_pkg_id(0);
+
+       /*
+        * Vendor-specific initialization.  In this section we
+        * canonicalize the feature flags, meaning if there are
+        * features a certain CPU supports which CPUID doesn't
+        * tell us, CPUID claiming incorrect flags, or other bugs,
+        * we handle them here.
+        *
+        * At the end of this section, c->x86_capability better
+        * indicate the features this CPU genuinely supports!
+        */
+       if (this_cpu->c_init)
+               this_cpu->c_init(c);
+
+       detect_ht(c);
+
+       /*
+        * On SMP, boot_cpu_data holds the common feature set between
+        * all CPUs; so make sure that we indicate which features are
+        * common between the CPUs.  The first time this routine gets
+        * executed, c == &boot_cpu_data.
+        */
+       if (c != &boot_cpu_data) {
+               /* AND the already accumulated flags with these */
+               for (i = 0; i < NCAPINTS; i++)
+                       boot_cpu_data.x86_capability[i] &= c->x86_capability[i];
+       }
+
+       /* Clear all flags overriden by options */
+       for (i = 0; i < NCAPINTS; i++)
+               c->x86_capability[i] &= ~cleared_cpu_caps[i];
+
+#ifdef CONFIG_X86_MCE
+       mcheck_init(c);
+#endif
+       select_idle_routine(c);
+
+#ifdef CONFIG_NUMA
+       numa_add_cpu(smp_processor_id());
+#endif
+
+}
+
+void __cpuinit identify_boot_cpu(void)
+{
+       identify_cpu(&boot_cpu_data);
+}
+
+void __cpuinit identify_secondary_cpu(struct cpuinfo_x86 *c)
+{
+       BUG_ON(c == &boot_cpu_data);
+       identify_cpu(c);
+       mtrr_ap_init();
+}
+
+static __init int setup_noclflush(char *arg)
+{
+       setup_clear_cpu_cap(X86_FEATURE_CLFLSH);
+       return 1;
+}
+__setup("noclflush", setup_noclflush);
+
+void __cpuinit print_cpu_info(struct cpuinfo_x86 *c)
+{
+       if (c->x86_model_id[0])
+               printk(KERN_CONT "%s", c->x86_model_id);
+
+       if (c->x86_mask || c->cpuid_level >= 0)
+               printk(KERN_CONT " stepping %02x\n", c->x86_mask);
+       else
+               printk(KERN_CONT "\n");
+}
+
+static __init int setup_disablecpuid(char *arg)
+{
+       int bit;
+       if (get_option(&arg, &bit) && bit < NCAPINTS*32)
+               setup_clear_cpu_cap(bit);
+       else
+               return 0;
+       return 1;
+}
+__setup("clearcpuid=", setup_disablecpuid);
+
+cpumask_t cpu_initialized __cpuinitdata = CPU_MASK_NONE;
+
+struct x8664_pda **_cpu_pda __read_mostly;
+EXPORT_SYMBOL(_cpu_pda);
+
+#ifndef CONFIG_X86_NO_IDT
+struct desc_ptr idt_descr = { 256 * 16 - 1, (unsigned long) idt_table };
+#endif
+
+char boot_cpu_stack[IRQSTACKSIZE] __page_aligned_bss;
+
+unsigned long __supported_pte_mask __read_mostly = ~0UL;
+EXPORT_SYMBOL_GPL(__supported_pte_mask);
+
+static int do_not_nx __cpuinitdata;
+
+/* noexec=on|off
+Control non executable mappings for 64bit processes.
+
+on     Enable(default)
+off    Disable
+*/
+static int __init nonx_setup(char *str)
+{
+       if (!str)
+               return -EINVAL;
+       if (!strncmp(str, "on", 2)) {
+               __supported_pte_mask |= _PAGE_NX;
+               do_not_nx = 0;
+       } else if (!strncmp(str, "off", 3)) {
+               do_not_nx = 1;
+               __supported_pte_mask &= ~_PAGE_NX;
+       }
+       return 0;
+}
+early_param("noexec", nonx_setup);
+
+int force_personality32;
+
+/* noexec32=on|off
+Control non executable heap for 32bit processes.
+To control the stack too use noexec=off
+
+on     PROT_READ does not imply PROT_EXEC for 32bit processes (default)
+off    PROT_READ implies PROT_EXEC
+*/
+static int __init nonx32_setup(char *str)
+{
+       if (!strcmp(str, "on"))
+               force_personality32 &= ~READ_IMPLIES_EXEC;
+       else if (!strcmp(str, "off"))
+               force_personality32 |= READ_IMPLIES_EXEC;
+       return 1;
+}
+__setup("noexec32=", nonx32_setup);
+
+static void __init_refok switch_pt(int cpu)
+{
+#ifdef CONFIG_XEN
+       if (cpu == 0)
+               xen_init_pt();
+       xen_pt_switch(__pa_symbol(init_level4_pgt));
+       xen_new_user_pt(__pa_symbol(__user_pgd(init_level4_pgt)));
+#endif
+}
+
+void pda_init(int cpu)
+{
+       struct x8664_pda *pda = cpu_pda(cpu);
+
+       /* Setup up data that may be needed in __get_free_pages early */
+       loadsegment(fs, 0);
+       loadsegment(gs, 0);
+#ifndef CONFIG_XEN
+       /* Memory clobbers used to order PDA accessed */
+       mb();
+       wrmsrl(MSR_GS_BASE, pda);
+       mb();
+#else
+       if (HYPERVISOR_set_segment_base(SEGBASE_GS_KERNEL,
+                                       (unsigned long)pda))
+               BUG();
+#endif
+
+       pda->cpunumber = cpu;
+       pda->irqcount = -1;
+       pda->kernelstack = (unsigned long)stack_thread_info() -
+                                PDA_STACKOFFSET + THREAD_SIZE;
+       pda->active_mm = &init_mm;
+       pda->mmu_state = 0;
+
+       if (cpu == 0) {
+               /* others are initialized in smpboot.c */
+               pda->pcurrent = &init_task;
+               pda->irqstackptr = boot_cpu_stack;
+               pda->irqstackptr += IRQSTACKSIZE - 64;
+       } else {
+               if (!pda->irqstackptr) {
+                       pda->irqstackptr = (char *)
+                               __get_free_pages(GFP_ATOMIC, IRQSTACK_ORDER);
+                       if (!pda->irqstackptr)
+                               panic("cannot allocate irqstack for cpu %d",
+                                     cpu);
+                       pda->irqstackptr += IRQSTACKSIZE - 64;
+               }
+
+               if (pda->nodenumber == 0 && cpu_to_node(cpu) != NUMA_NO_NODE)
+                       pda->nodenumber = cpu_to_node(cpu);
+       }
+
+       switch_pt(cpu);
+}
+
+#ifndef CONFIG_X86_NO_TSS
+char boot_exception_stacks[(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ +
+                          DEBUG_STKSZ] __page_aligned_bss;
+#endif
+
+extern asmlinkage void ignore_sysret(void);
+
+void __cpuinit syscall_init(void)
+{
+#ifndef CONFIG_XEN
+       /*
+        * LSTAR and STAR live in a bit strange symbiosis.
+        * They both write to the same internal register. STAR allows to
+        * set CS/DS but only a 32bit target. LSTAR sets the 64bit rip.
+        */
+       wrmsrl(MSR_STAR,  ((u64)__USER32_CS)<<48  | ((u64)__KERNEL_CS)<<32);
+       wrmsrl(MSR_LSTAR, system_call);
+       wrmsrl(MSR_CSTAR, ignore_sysret);
+
+       /* Flags to clear on syscall */
+       wrmsrl(MSR_SYSCALL_MASK,
+              X86_EFLAGS_TF|X86_EFLAGS_DF|X86_EFLAGS_IF|X86_EFLAGS_IOPL);
+#endif
+#ifdef CONFIG_IA32_EMULATION
+       syscall32_cpu_init();
+#else
+       static /*const*/ struct callback_register __cpuinitdata cstar = {
+               .type = CALLBACKTYPE_syscall32,
+               .address = (unsigned long)ignore_sysret
+       };
+
+       if (HYPERVISOR_callback_op(CALLBACKOP_register, &cstar))
+               printk(KERN_WARN "Unable to register CSTAR callback\n");
+#endif
+}
+
+void __cpuinit check_efer(void)
+{
+       unsigned long efer;
+
+       rdmsrl(MSR_EFER, efer);
+       if (!(efer & EFER_NX) || do_not_nx)
+               __supported_pte_mask &= ~_PAGE_NX;
+}
+
+unsigned long kernel_eflags;
+
+#ifndef CONFIG_X86_NO_TSS
+/*
+ * Copies of the original ist values from the tss are only accessed during
+ * debugging, no special alignment required.
+ */
+DEFINE_PER_CPU(struct orig_ist, orig_ist);
+#endif
+
+/*
+ * cpu_init() initializes state that is per-CPU. Some data is already
+ * initialized (naturally) in the bootstrap process, such as the GDT
+ * and IDT. We reload them nevertheless, this function acts as a
+ * 'CPU state barrier', nothing should get across.
+ * A lot of state is already set up in PDA init.
+ */
+void __cpuinit cpu_init(void)
+{
+       int cpu = stack_smp_processor_id();
+#ifndef CONFIG_X86_NO_TSS
+       struct tss_struct *t = &per_cpu(init_tss, cpu);
+       struct orig_ist *orig_ist = &per_cpu(orig_ist, cpu);
+       unsigned long v;
+       char *estacks = NULL;
+       int i;
+#endif
+       struct task_struct *me;
+
+       /* CPU 0 is initialised in head64.c */
+       if (cpu != 0)
+               pda_init(cpu);
+#ifndef CONFIG_X86_NO_TSS
+       else
+               estacks = boot_exception_stacks;
+#endif
+
+       me = current;
+
+       if (cpu_test_and_set(cpu, cpu_initialized))
+               panic("CPU#%d already initialized!\n", cpu);
+
+       printk(KERN_INFO "Initializing CPU#%d\n", cpu);
+
+       clear_in_cr4(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE);
+
+       /*
+        * Initialize the per-CPU GDT with the boot GDT,
+        * and set up the GDT descriptor:
+        */
+
+       switch_to_new_gdt();
+#ifndef CONFIG_X86_NO_IDT
+       load_idt((const struct desc_ptr *)&idt_descr);
+#endif
+
+       memset(me->thread.tls_array, 0, GDT_ENTRY_TLS_ENTRIES * 8);
+       syscall_init();
+
+       wrmsrl(MSR_FS_BASE, 0);
+       wrmsrl(MSR_KERNEL_GS_BASE, 0);
+       barrier();
+
+       check_efer();
+
+#ifndef CONFIG_X86_NO_TSS
+       /*
+        * set up and load the per-CPU TSS
+        */
+       if (!orig_ist->ist[0]) {
+               static const unsigned int order[N_EXCEPTION_STACKS] = {
+                 [0 ... N_EXCEPTION_STACKS - 1] = EXCEPTION_STACK_ORDER,
+                 [DEBUG_STACK - 1] = DEBUG_STACK_ORDER
+               };
+               for (v = 0; v < N_EXCEPTION_STACKS; v++) {
+                       if (cpu) {
+                               estacks = (char *)__get_free_pages(GFP_ATOMIC, order[v]);
+                               if (!estacks)
+                                       panic("Cannot allocate exception "
+                                             "stack %ld %d\n", v, cpu);
+                       }
+                       estacks += PAGE_SIZE << order[v];
+                       orig_ist->ist[v] = t->x86_tss.ist[v] =
+                                       (unsigned long)estacks;
+               }
+       }
+
+       t->x86_tss.io_bitmap_base = offsetof(struct tss_struct, io_bitmap);
+       /*
+        * <= is required because the CPU will access up to
+        * 8 bits beyond the end of the IO permission bitmap.
+        */
+       for (i = 0; i <= IO_BITMAP_LONGS; i++)
+               t->io_bitmap[i] = ~0UL;
+#endif
+
+       atomic_inc(&init_mm.mm_count);
+       me->active_mm = &init_mm;
+       if (me->mm)
+               BUG();
+       enter_lazy_tlb(&init_mm, me);
+
+       load_sp0(t, &current->thread);
+#ifndef CONFIG_X86_NO_TSS
+       set_tss_desc(cpu, t);
+       load_TR_desc();
+#endif
+       load_LDT(&init_mm.context);
+
+#ifdef CONFIG_KGDB
+       /*
+        * If the kgdb is connected no debug regs should be altered.  This
+        * is only applicable when KGDB and a KGDB I/O module are built
+        * into the kernel and you are using early debugging with
+        * kgdbwait. KGDB will control the kernel HW breakpoint registers.
+        */
+       if (kgdb_connected && arch_kgdb_ops.correct_hw_break)
+               arch_kgdb_ops.correct_hw_break();
+       else {
+#endif
+       /*
+        * Clear all 6 debug registers:
+        */
+
+       set_debugreg(0UL, 0);
+       set_debugreg(0UL, 1);
+       set_debugreg(0UL, 2);
+       set_debugreg(0UL, 3);
+       set_debugreg(0UL, 6);
+       set_debugreg(0UL, 7);
+#ifdef CONFIG_KGDB
+       /* If the kgdb is connected no debug regs should be altered. */
+       }
+#endif
+
+       fpu_init();
+
+       asm ("pushfq; popq %0" : "=rm" (kernel_eflags));
+       if (raw_irqs_disabled())
+               kernel_eflags &= ~X86_EFLAGS_IF;
+
+       if (is_uv_system())
+               uv_cpu_init();
+}
Index: head-2008-12-01/arch/x86/kernel/e820-xen.c
===================================================================
--- /dev/null   1970-01-01 00:00:00.000000000 +0000
+++ head-2008-12-01/arch/x86/kernel/e820-xen.c  2008-12-01 11:49:07.000000000 +0100
@@ -0,0 +1,1470 @@
+/*
+ * Handle the memory map.
+ * The functions here do the job until bootmem takes over.
+ *
+ *  Getting sanitize_e820_map() in sync with i386 version by applying change:
+ *  -  Provisions for empty E820 memory regions (reported by certain BIOSes).
+ *     Alex Achenbach <xela@slit.de>, December 2002.
+ *  Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
+ *
+ */
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/bootmem.h>
+#include <linux/ioport.h>
+#include <linux/string.h>
+#include <linux/kexec.h>
+#include <linux/module.h>
+#include <linux/mm.h>
+#include <linux/pfn.h>
+#include <linux/suspend.h>
+#include <linux/firmware-map.h>
+
+#include <asm/pgtable.h>
+#include <asm/page.h>
+#include <asm/e820.h>
+#include <asm/proto.h>
+#include <asm/setup.h>
+#include <xen/interface/memory.h>
+
+/*
+ * The e820 map is the map that gets modified e.g. with command line parameters
+ * and that is also registered with modifications in the kernel resource tree
+ * with the iomem_resource as parent.
+ *
+ * The e820_saved is directly saved after the BIOS-provided memory map is
+ * copied. It doesn't get modified afterwards. It's registered for the
+ * /sys/firmware/memmap interface.
+ *
+ * That memory map is not modified and is used as base for kexec. The kexec'd
+ * kernel should get the same memory map as the firmware provides. Then the
+ * user can e.g. boot the original kernel with mem=1G while still booting the
+ * next kernel with full memory.
+ */
+struct e820map e820;
+struct e820map e820_saved;
+#ifdef CONFIG_XEN
+static struct e820map machine_e820;
+#endif
+
+/* For PCI or other memory-mapped resources */
+unsigned long pci_mem_start = 0xaeedbabe;
+#ifdef CONFIG_PCI
+EXPORT_SYMBOL(pci_mem_start);
+#endif
+
+/*
+ * This function checks if any part of the range <start,end> is mapped
+ * with type.
+ */
+int
+e820_any_mapped(u64 start, u64 end, unsigned type)
+{
+       int i;
+
+#ifndef CONFIG_XEN
+       for (i = 0; i < e820.nr_map; i++) {
+               struct e820entry *ei = &e820.map[i];
+#else
+       if (!is_initial_xendomain())
+               return 0;
+       for (i = 0; i < machine_e820.nr_map; ++i) {
+               const struct e820entry *ei = &machine_e820.map[i];
+#endif
+
+               if (type && ei->type != type)
+                       continue;
+               if (ei->addr >= end || ei->addr + ei->size <= start)
+                       continue;
+               return 1;
+       }
+       return 0;
+}
+EXPORT_SYMBOL_GPL(e820_any_mapped);
+
+/*
+ * This function checks if the entire range <start,end> is mapped with type.
+ *
+ * Note: this function only works correct if the e820 table is sorted and
+ * not-overlapping, which is the case
+ */
+int __init e820_all_mapped(u64 start, u64 end, unsigned type)
+{
+       int i;
+
+#ifndef CONFIG_XEN
+       for (i = 0; i < e820.nr_map; i++) {
+               struct e820entry *ei = &e820.map[i];
+#else
+       if (!is_initial_xendomain())
+               return 0;
+       for (i = 0; i < machine_e820.nr_map; ++i) {
+               const struct e820entry *ei = &machine_e820.map[i];
+#endif
+
+               if (type && ei->type != type)
+                       continue;
+               /* is the region (part) in overlap with the current region ?*/
+               if (ei->addr >= end || ei->addr + ei->size <= start)
+                       continue;
+
+               /* if the region is at the beginning of <start,end> we move
+                * start to the end of the region since it's ok until there
+                */
+               if (ei->addr <= start)
+                       start = ei->addr + ei->size;
+               /*
+                * if start is now at or beyond end, we're done, full
+                * coverage
+                */
+               if (start >= end)
+                       return 1;
+       }
+       return 0;
+}
+
+/*
+ * Add a memory region to the kernel e820 map.
+ */
+void __init e820_add_region(u64 start, u64 size, int type)
+{
+       int x = e820.nr_map;
+
+       if (x == ARRAY_SIZE(e820.map)) {
+               printk(KERN_ERR "Ooops! Too many entries in the memory map!\n");
+               return;
+       }
+
+       e820.map[x].addr = start;
+       e820.map[x].size = size;
+       e820.map[x].type = type;
+       e820.nr_map++;
+}
+
+void __init e820_print_map(char *who)
+{
+       int i;
+
+       for (i = 0; i < e820.nr_map; i++) {
+               printk(KERN_INFO " %s: %016Lx - %016Lx ", who,
+                      (unsigned long long) e820.map[i].addr,
+                      (unsigned long long)
+                      (e820.map[i].addr + e820.map[i].size));
+               switch (e820.map[i].type) {
+               case E820_RAM:
+               case E820_RESERVED_KERN:
+                       printk(KERN_CONT "(usable)\n");
+                       break;
+               case E820_RESERVED:
+                       printk(KERN_CONT "(reserved)\n");
+                       break;
+               case E820_ACPI:
+                       printk(KERN_CONT "(ACPI data)\n");
+                       break;
+               case E820_NVS:
+                       printk(KERN_CONT "(ACPI NVS)\n");
+                       break;
+               default:
+                       printk(KERN_CONT "type %u\n", e820.map[i].type);
+                       break;
+               }
+       }
+}
+
+/*
+ * Sanitize the BIOS e820 map.
+ *
+ * Some e820 responses include overlapping entries. The following
+ * replaces the original e820 map with a new one, removing overlaps,
+ * and resolving conflicting memory types in favor of highest
+ * numbered type.
+ *
+ * The input parameter biosmap points to an array of 'struct
+ * e820entry' which on entry has elements in the range [0, *pnr_map)
+ * valid, and which has space for up to max_nr_map entries.
+ * On return, the resulting sanitized e820 map entries will be in
+ * overwritten in the same location, starting at biosmap.
+ *
+ * The integer pointed to by pnr_map must be valid on entry (the
+ * current number of valid entries located at biosmap) and will
+ * be updated on return, with the new number of valid entries
+ * (something no more than max_nr_map.)
+ *
+ * The return value from sanitize_e820_map() is zero if it
+ * successfully 'sanitized' the map entries passed in, and is -1
+ * if it did nothing, which can happen if either of (1) it was
+ * only passed one map entry, or (2) any of the input map entries
+ * were invalid (start + size < start, meaning that the size was
+ * so big the described memory range wrapped around through zero.)
+ *
+ *     Visually we're performing the following
+ *     (1,2,3,4 = memory types)...
+ *
+ *     Sample memory map (w/overlaps):
+ *        ____22__________________
+ *        ______________________4_
+ *        ____1111________________
+ *        _44_____________________
+ *        11111111________________
+ *        ____________________33__
+ *        ___________44___________
+ *        __________33333_________
+ *        ______________22________
+ *        ___________________2222_
+ *        _________111111111______
+ *        _____________________11_
+ *        _________________4______
+ *
+ *     Sanitized equivalent (no overlap):
+ *        1_______________________
+ *        _44_____________________
+ *        ___1____________________
+ *        ____22__________________
+ *        ______11________________
+ *        _________1______________
+ *        __________3_____________
+ *        ___________44___________
+ *        _____________33_________
+ *        _______________2________
+ *        ________________1_______
+ *        _________________4______
+ *        ___________________2____
+ *        ____________________33__
+ *        ______________________4_
+ */
+
+int __init sanitize_e820_map(struct e820entry *biosmap, int max_nr_map,
+                               int *pnr_map)
+{
+       struct change_member {
+               struct e820entry *pbios; /* pointer to original bios entry */
+               unsigned long long addr; /* address for this change point */
+       };
+       static struct change_member change_point_list[2*E820_X_MAX] __initdata;
+       static struct change_member *change_point[2*E820_X_MAX] __initdata;
+       static struct e820entry *overlap_list[E820_X_MAX] __initdata;
+       static struct e820entry new_bios[E820_X_MAX] __initdata;
+       struct change_member *change_tmp;
+       unsigned long current_type, last_type;
+       unsigned long long last_addr;
+       int chgidx, still_changing;
+       int overlap_entries;
+       int new_bios_entry;
+       int old_nr, new_nr, chg_nr;
+       int i;
+
+       /* if there's only one memory region, don't bother */
+#ifdef CONFIG_XEN
+       if (*pnr_map == 1)
+               return 0;
+#endif
+       if (*pnr_map < 2)
+               return -1;
+
+       old_nr = *pnr_map;
+       BUG_ON(old_nr > max_nr_map);
+
+       /* bail out if we find any unreasonable addresses in bios map */
+       for (i = 0; i < old_nr; i++)
+               if (biosmap[i].addr + biosmap[i].size < biosmap[i].addr)
+                       return -1;
+
+       /* create pointers for initial change-point information (for sorting) */
+       for (i = 0; i < 2 * old_nr; i++)
+               change_point[i] = &change_point_list[i];
+
+       /* record all known change-points (starting and ending addresses),
+          omitting those that are for empty memory regions */
+       chgidx = 0;
+       for (i = 0; i < old_nr; i++)    {
+               if (biosmap[i].size != 0) {
+                       change_point[chgidx]->addr = biosmap[i].addr;
+                       change_point[chgidx++]->pbios = &biosmap[i];
+                       change_point[chgidx]->addr = biosmap[i].addr +
+                               biosmap[i].size;
+                       change_point[chgidx++]->pbios = &biosmap[i];
+               }
+       }
+       chg_nr = chgidx;
+
+       /* sort change-point list by memory addresses (low -> high) */
+       still_changing = 1;
+       while (still_changing)  {
+               still_changing = 0;
+               for (i = 1; i < chg_nr; i++)  {
+                       unsigned long long curaddr, lastaddr;
+                       unsigned long long curpbaddr, lastpbaddr;
+
+                       curaddr = change_point[i]->addr;
+                       lastaddr = change_point[i - 1]->addr;
+                       curpbaddr = change_point[i]->pbios->addr;
+                       lastpbaddr = change_point[i - 1]->pbios->addr;
+
+                       /*
+                        * swap entries, when:
+                        *
+                        * curaddr > lastaddr or
+                        * curaddr == lastaddr and curaddr == curpbaddr and
+                        * lastaddr != lastpbaddr
+                        */
+                       if (curaddr < lastaddr ||
+                           (curaddr == lastaddr && curaddr == curpbaddr &&
+                            lastaddr != lastpbaddr)) {
+                               change_tmp = change_point[i];
+                               change_point[i] = change_point[i-1];
+                               change_point[i-1] = change_tmp;
+                               still_changing = 1;
+                       }
+               }
+       }
+
+       /* create a new bios memory map, removing overlaps */
+       overlap_entries = 0;     /* number of entries in the overlap table */
+       new_bios_entry = 0;      /* index for creating new bios map entries */
+       last_type = 0;           /* start with undefined memory type */
+       last_addr = 0;           /* start with 0 as last starting address */
+
+       /* loop through change-points, determining affect on the new bios map */
+       for (chgidx = 0; chgidx < chg_nr; chgidx++) {
+               /* keep track of all overlapping bios entries */
+               if (change_point[chgidx]->addr ==
+                   change_point[chgidx]->pbios->addr) {
+                       /*
+                        * add map entry to overlap list (> 1 entry
+                        * implies an overlap)
+                        */
+                       overlap_list[overlap_entries++] =
+                               change_point[chgidx]->pbios;
+               } else {
+                       /*
+                        * remove entry from list (order independent,
+                        * so swap with last)
+                        */
+                       for (i = 0; i < overlap_entries; i++) {
+                               if (overlap_list[i] ==
+                                   change_point[chgidx]->pbios)
+                                       overlap_list[i] =
+                                               overlap_list[overlap_entries-1];
+                       }
+                       overlap_entries--;
+               }
+               /*
+                * if there are overlapping entries, decide which
+                * "type" to use (larger value takes precedence --
+                * 1=usable, 2,3,4,4+=unusable)
+                */
+               current_type = 0;
+               for (i = 0; i < overlap_entries; i++)
+                       if (overlap_list[i]->type > current_type)
+                               current_type = overlap_list[i]->type;
+               /*
+                * continue building up new bios map based on this
+                * information
+                */
+               if (current_type != last_type)  {
+                       if (last_type != 0)      {
+                               new_bios[new_bios_entry].size =
+                                       change_point[chgidx]->addr - last_addr;
+                               /*
+                                * move forward only if the new size
+                                * was non-zero
+                                */
+                               if (new_bios[new_bios_entry].size != 0)
+                                       /*
+                                        * no more space left for new
+                                        * bios entries ?
+                                        */
+                                       if (++new_bios_entry >= max_nr_map)
+                                               break;
+                       }
+                       if (current_type != 0)  {
+                               new_bios[new_bios_entry].addr =
+                                       change_point[chgidx]->addr;
+                               new_bios[new_bios_entry].type = current_type;
+                               last_addr = change_point[chgidx]->addr;
+                       }
+                       last_type = current_type;
+               }
+       }
+       /* retain count for new bios entries */
+       new_nr = new_bios_entry;
+
+       /* copy new bios mapping into original location */
+       memcpy(biosmap, new_bios, new_nr * sizeof(struct e820entry));
+       *pnr_map = new_nr;
+
+       return 0;
+}
+
+static int __init __append_e820_map(struct e820entry *biosmap, int nr_map)
+{
+       while (nr_map) {
+               u64 start = biosmap->addr;
+               u64 size = biosmap->size;
+               u64 end = start + size;
+               u32 type = biosmap->type;
+
+               /* Overflow in 64 bits? Ignore the memory map. */
+               if (start > end)
+                       return -1;
+
+               e820_add_region(start, size, type);
+
+               biosmap++;
+               nr_map--;
+       }
+       return 0;
+}
+
+/*
+ * Copy the BIOS e820 map into a safe place.
+ *
+ * Sanity-check it while we're at it..
+ *
+ * If we're lucky and live on a modern system, the setup code
+ * will have given us a memory map that we can use to properly
+ * set up memory.  If we aren't, we'll fake a memory map.
+ */
+static int __init append_e820_map(struct e820entry *biosmap, int nr_map)
+{
+#ifndef CONFIG_XEN
+       /* Only one memory region (or negative)? Ignore it */
+       if (nr_map < 2)
+               return -1;
+#else
+       BUG_ON(nr_map < 1);
+#endif
+
+       return __append_e820_map(biosmap, nr_map);
+}
+
+static u64 __init e820_update_range_map(struct e820map *e820x, u64 start,
+                                       u64 size, unsigned old_type,
+                                       unsigned new_type)
+{
+       int i;
+       u64 real_updated_size = 0;
+
+       BUG_ON(old_type == new_type);
+
+       if (size > (ULLONG_MAX - start))
+               size = ULLONG_MAX - start;
+
+       for (i = 0; i < e820.nr_map; i++) {
+               struct e820entry *ei = &e820x->map[i];
+               u64 final_start, final_end;
+               if (ei->type != old_type)
+                       continue;
+               /* totally covered? */
+               if (ei->addr >= start &&
+                   (ei->addr + ei->size) <= (start + size)) {
+                       ei->type = new_type;
+                       real_updated_size += ei->size;
+                       continue;
+               }
+               /* partially covered */
+               final_start = max(start, ei->addr);
+               final_end = min(start + size, ei->addr + ei->size);
+               if (final_start >= final_end)
+                       continue;
+               e820_add_region(final_start, final_end - final_start,
+                                        new_type);
+               real_updated_size += final_end - final_start;
+
+               ei->size -= final_end - final_start;
+               if (ei->addr < final_start)
+                       continue;
+               ei->addr = final_end;
+       }
+       return real_updated_size;
+}
+
+u64 __init e820_update_range(u64 start, u64 size, unsigned old_type,
+                            unsigned new_type)
+{
+       return e820_update_range_map(&e820, start, size, old_type, new_type);
+}
+
+static u64 __init e820_update_range_saved(u64 start, u64 size,
+                                         unsigned old_type, unsigned new_type)
+{
+       return e820_update_range_map(&e820_saved, start, size, old_type,
+                                    new_type);
+}
+
+/* make e820 not cover the range */
+u64 __init e820_remove_range(u64 start, u64 size, unsigned old_type,
+                            int checktype)
+{
+       int i;
+       u64 real_removed_size = 0;
+
+       if (size > (ULLONG_MAX - start))
+               size = ULLONG_MAX - start;
+
+       for (i = 0; i < e820.nr_map; i++) {
+               struct e820entry *ei = &e820.map[i];
+               u64 final_start, final_end;
+
+               if (checktype && ei->type != old_type)
+                       continue;
+               /* totally covered? */
+               if (ei->addr >= start &&
+                   (ei->addr + ei->size) <= (start + size)) {
+                       real_removed_size += ei->size;
+                       memset(ei, 0, sizeof(struct e820entry));
+                       continue;
+               }
+               /* partially covered */
+               final_start = max(start, ei->addr);
+               final_end = min(start + size, ei->addr + ei->size);
+               if (final_start >= final_end)
+                       continue;
+               real_removed_size += final_end - final_start;
+
+               ei->size -= final_end - final_start;
+               if (ei->addr < final_start)
+                       continue;
+               ei->addr = final_end;
+       }
+       return real_removed_size;
+}
+
+void __init update_e820(void)
+{
+       int nr_map;
+
+       nr_map = e820.nr_map;
+       if (sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &nr_map))
+               return;
+       e820.nr_map = nr_map;
+       printk(KERN_INFO "modified physical RAM map:\n");
+       e820_print_map("modified");
+}
+static void __init update_e820_saved(void)
+{
+       int nr_map;
+
+       nr_map = e820_saved.nr_map;
+       if (sanitize_e820_map(e820_saved.map, ARRAY_SIZE(e820_saved.map), &nr_map))
+               return;
+       e820_saved.nr_map = nr_map;
+}
+
+#ifdef CONFIG_XEN
+#define e820 machine_e820
+#endif
+
+#define MAX_GAP_END 0x100000000ull
+/*
+ * Search for a gap in the e820 memory space from start_addr to end_addr.
+ */
+__init int e820_search_gap(unsigned long *gapstart, unsigned long *gapsize,
+               unsigned long start_addr, unsigned long long end_addr)
+{
+       unsigned long long last;
+       int i = e820.nr_map;
+       int found = 0;
+
+       last = (end_addr && end_addr < MAX_GAP_END) ? end_addr : MAX_GAP_END;
+#ifdef CONFIG_X86_64
+       if (start_addr >= MAX_GAP_END)
+               last = end_addr ?: (1UL << boot_cpu_data.x86_phys_bits);
+#endif
+
+       while (--i >= 0) {
+               unsigned long long start = e820.map[i].addr;
+               unsigned long long end = start + e820.map[i].size;
+
+               if (end < start_addr)
+                       continue;
+
+               /*
+                * Since "last" is at most 4GB, we know we'll
+                * fit in 32 bits if this condition is true
+                */
+               if (last > end) {
+                       unsigned long gap = last - end;
+
+                       if (gap >= *gapsize) {
+                               *gapsize = gap;
+                               *gapstart = end;
+                               found = 1;
+                       }
+               }
+               if (start < last)
+                       last = start;
+       }
+       return found;
+}
+
+/*
+ * Search for the biggest gap in the low 32 bits of the e820
+ * memory space.  We pass this space to PCI to assign MMIO resources
+ * for hotplug or unconfigured devices in.
+ * Hopefully the BIOS let enough space left.
+ */
+__init void e820_setup_gap(void)
+{
+       unsigned long gapstart, gapsize, round;
+       int found;
+
+       gapstart = 0x10000000;
+       gapsize = 0x400000;
+       found  = e820_search_gap(&gapstart, &gapsize, 0, MAX_GAP_END);
+
+#ifdef CONFIG_X86_64
+       if (!found) {
+               printk(KERN_ERR "PCI: Warning: Cannot find a gap in the 32bit "
+                      "address range\n"
+                      KERN_ERR "PCI: Unassigned devices with 32bit resource "
+                      "registers may break!\n");
+               found  = e820_search_gap(&gapstart, &gapsize, MAX_GAP_END, 0);
+               BUG_ON(!found);
+       }
+#endif
+
+       /*
+        * See how much we want to round up: start off with
+        * rounding to the next 1MB area.
+        */
+       round = 0x100000;
+       while ((gapsize >> 4) > round)
+               round += round;
+       /* Fun with two's complement */
+       pci_mem_start = (gapstart + round) & -round;
+
+       printk(KERN_INFO
+              "Allocating PCI resources starting at %lx (gap: %lx:%lx)\n",
+              pci_mem_start, gapstart, gapsize);
+}
+
+#undef e820
+
+#ifndef CONFIG_XEN
+/**
+ * Because of the size limitation of struct boot_params, only first
+ * 128 E820 memory entries are passed to kernel via
+ * boot_params.e820_map, others are passed via SETUP_E820_EXT node of
+ * linked list of struct setup_data, which is parsed here.
+ */
+void __init parse_e820_ext(struct setup_data *sdata, unsigned long pa_data)
+{
+       u32 map_len;
+       int entries;
+       struct e820entry *extmap;
+
+       entries = sdata->len / sizeof(struct e820entry);
+       map_len = sdata->len + sizeof(struct setup_data);
+       if (map_len > PAGE_SIZE)
+               sdata = early_ioremap(pa_data, map_len);
+       extmap = (struct e820entry *)(sdata->data);
+       __append_e820_map(extmap, entries);
+       sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);
+       if (map_len > PAGE_SIZE)
+               early_iounmap(sdata, map_len);
+       printk(KERN_INFO "extended physical RAM map:\n");
+       e820_print_map("extended");
+}
+
+#if defined(CONFIG_X86_64) || \
+       (defined(CONFIG_X86_32) && defined(CONFIG_HIBERNATION))
+/**
+ * Find the ranges of physical addresses that do not correspond to
+ * e820 RAM areas and mark the corresponding pages as nosave for
+ * hibernation (32 bit) or software suspend and suspend to RAM (64 bit).
+ *
+ * This function requires the e820 map to be sorted and without any
+ * overlapping entries and assumes the first e820 area to be RAM.
+ */
+void __init e820_mark_nosave_regions(unsigned long limit_pfn)
+{
+       int i;
+       unsigned long pfn;
+
+       pfn = PFN_DOWN(e820.map[0].addr + e820.map[0].size);
+       for (i = 1; i < e820.nr_map; i++) {
+               struct e820entry *ei = &e820.map[i];
+
+               if (pfn < PFN_UP(ei->addr))
+                       register_nosave_region(pfn, PFN_UP(ei->addr));
+
+               pfn = PFN_DOWN(ei->addr + ei->size);
+               if (ei->type != E820_RAM && ei->type != E820_RESERVED_KERN)
+                       register_nosave_region(PFN_UP(ei->addr), pfn);
+
+               if (pfn >= limit_pfn)
+                       break;
+       }
+}
+#endif
+#endif
+
+/*
+ * Early reserved memory areas.
+ */
+#define MAX_EARLY_RES 20
+
+struct early_res {
+       u64 start, end;
+       char name[16];
+       char overlap_ok;
+};
+static struct early_res early_res[MAX_EARLY_RES] __initdata = {
+#ifndef CONFIG_XEN
+       { 0, PAGE_SIZE, "BIOS data page" },     /* BIOS data page */
+#if defined(CONFIG_X86_64) && defined(CONFIG_X86_TRAMPOLINE)
+       { TRAMPOLINE_BASE, TRAMPOLINE_BASE + 2 * PAGE_SIZE, "TRAMPOLINE" },
+#endif
+#if defined(CONFIG_X86_32) && defined(CONFIG_SMP)
+       /*
+        * But first pinch a few for the stack/trampoline stuff
+        * FIXME: Don't need the extra page at 4K, but need to fix
+        * trampoline before removing it. (see the GDT stuff)
+        */
+       { PAGE_SIZE, PAGE_SIZE + PAGE_SIZE, "EX TRAMPOLINE" },
+       /*
+        * Has to be in very low memory so we can execute
+        * real-mode AP code.
+        */
+       { TRAMPOLINE_BASE, TRAMPOLINE_BASE + PAGE_SIZE, "TRAMPOLINE" },
+#endif
+#endif
+       {}
+};
+
+static int __init find_overlapped_early(u64 start, u64 end)
+{
+       int i;
+       struct early_res *r;
+
+       for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) {
+               r = &early_res[i];
+               if (end > r->start && start < r->end)
+                       break;
+       }
+
+       return i;
+}
+
+/*
+ * Drop the i-th range from the early reservation map,
+ * by copying any higher ranges down one over it, and
+ * clearing what had been the last slot.
+ */
+static void __init drop_range(int i)
+{
+       int j;
+
+       for (j = i + 1; j < MAX_EARLY_RES && early_res[j].end; j++)
+               ;
+
+       memmove(&early_res[i], &early_res[i + 1],
+              (j - 1 - i) * sizeof(struct early_res));
+
+       early_res[j - 1].end = 0;
+}
+
+/*
+ * Split any existing ranges that:
+ *  1) are marked 'overlap_ok', and
+ *  2) overlap with the stated range [start, end)
+ * into whatever portion (if any) of the existing range is entirely
+ * below or entirely above the stated range.  Drop the portion
+ * of the existing range that overlaps with the stated range,
+ * which will allow the caller of this routine to then add that
+ * stated range without conflicting with any existing range.
+ */
+static void __init drop_overlaps_that_are_ok(u64 start, u64 end)
+{
+       int i;
+       struct early_res *r;
+       u64 lower_start, lower_end;
+       u64 upper_start, upper_end;
+       char name[16];
+
+       for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) {
+               r = &early_res[i];
+
+               /* Continue past non-overlapping ranges */
+               if (end <= r->start || start >= r->end)
+                       continue;
+
+               /*
+                * Leave non-ok overlaps as is; let caller
+                * panic "Overlapping early reservations"
+                * when it hits this overlap.
+                */
+               if (!r->overlap_ok)
+                       return;
+
+               /*
+                * We have an ok overlap.  We will drop it from the early
+                * reservation map, and add back in any non-overlapping
+                * portions (lower or upper) as separate, overlap_ok,
+                * non-overlapping ranges.
+                */
+
+               /* 1. Note any non-overlapping (lower or upper) ranges. */
+               strncpy(name, r->name, sizeof(name) - 1);
+
+               lower_start = lower_end = 0;
+               upper_start = upper_end = 0;
+               if (r->start < start) {
+                       lower_start = r->start;
+                       lower_end = start;
+               }
+               if (r->end > end) {
+                       upper_start = end;
+                       upper_end = r->end;
+               }
+
+               /* 2. Drop the original ok overlapping range */
+               drop_range(i);
+
+               i--;            /* resume for-loop on copied down entry */
+
+               /* 3. Add back in any non-overlapping ranges. */
+               if (lower_end)
+                       reserve_early_overlap_ok(lower_start, lower_end, name);
+               if (upper_end)
+                       reserve_early_overlap_ok(upper_start, upper_end, name);
+       }
+}
+
+static void __init __reserve_early(u64 start, u64 end, char *name,
+                                               int overlap_ok)
+{
+       int i;
+       struct early_res *r;
+
+       i = find_overlapped_early(start, end);
+       if (i >= MAX_EARLY_RES)
+               panic("Too many early reservations");
+       r = &early_res[i];
+       if (r->end)
+               panic("Overlapping early reservations "
+                     "%llx-%llx %s to %llx-%llx %s\n",
+                     start, end - 1, name?name:"", r->start,
+                     r->end - 1, r->name);
+       r->start = start;
+       r->end = end;
+       r->overlap_ok = overlap_ok;
+       if (name)
+               strncpy(r->name, name, sizeof(r->name) - 1);
+}
+
+/*
+ * A few early reservtations come here.
+ *
+ * The 'overlap_ok' in the name of this routine does -not- mean it
+ * is ok for these reservations to overlap an earlier reservation.
+ * Rather it means that it is ok for subsequent reservations to
+ * overlap this one.
+ *
+ * Use this entry point to reserve early ranges when you are doing
+ * so out of "Paranoia", reserving perhaps more memory than you need,
+ * just in case, and don't mind a subsequent overlapping reservation
+ * that is known to be needed.
+ *
+ * The drop_overlaps_that_are_ok() call here isn't really needed.
+ * It would be needed if we had two colliding 'overlap_ok'
+ * reservations, so that the second such would not panic on the
+ * overlap with the first.  We don't have any such as of this
+ * writing, but might as well tolerate such if it happens in
+ * the future.
+ */
+void __init reserve_early_overlap_ok(u64 start, u64 end, char *name)
+{
+       drop_overlaps_that_are_ok(start, end);
+       __reserve_early(start, end, name, 1);
+}
+
+/*
+ * Most early reservations come here.
+ *
+ * We first have drop_overlaps_that_are_ok() drop any pre-existing
+ * 'overlap_ok' ranges, so that we can then reserve this memory
+ * range without risk of panic'ing on an overlapping overlap_ok
+ * early reservation.
+ */
+void __init reserve_early(u64 start, u64 end, char *name)
+{
+       drop_overlaps_that_are_ok(start, end);
+       __reserve_early(start, end, name, 0);
+}
+
+void __init free_early(u64 start, u64 end)
+{
+       struct early_res *r;
+       int i;
+
+       i = find_overlapped_early(start, end);
+       r = &early_res[i];
+       if (i >= MAX_EARLY_RES || r->end != end || r->start != start)
+               panic("free_early on not reserved area: %llx-%llx!",
+                        start, end - 1);
+
+       drop_range(i);
+}
+
+void __init early_res_to_bootmem(u64 start, u64 end)
+{
+       int i, count;
+       u64 final_start, final_end;
+
+       count  = 0;
+       for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++)
+               count++;
+
+       printk(KERN_INFO "(%d early reservations) ==> bootmem [%010llx - %010llx]\n",
+                        count, start, end);
+       for (i = 0; i < count; i++) {
+               struct early_res *r = &early_res[i];
+               printk(KERN_INFO "  #%d [%010llx - %010llx] %16s", i,
+                       r->start, r->end, r->name);
+               final_start = max(start, r->start);
+               final_end = min(end, r->end);
+               if (final_start >= final_end) {
+                       printk(KERN_CONT "\n");
+                       continue;
+               }
+               printk(KERN_CONT " ==> [%010llx - %010llx]\n",
+                       final_start, final_end);
+               reserve_bootmem_generic(final_start, final_end - final_start,
+                               BOOTMEM_DEFAULT);
+       }
+}
+
+/* Check for already reserved areas */
+static inline int __init bad_addr(u64 *addrp, u64 size, u64 align)
+{
+       int i;
+       u64 addr = *addrp;
+       int changed = 0;
+       struct early_res *r;
+again:
+       i = find_overlapped_early(addr, addr + size);
+       r = &early_res[i];
+       if (i < MAX_EARLY_RES && r->end) {
+               *addrp = addr = round_up(r->end, align);
+               changed = 1;
+               goto again;
+       }
+       return changed;
+}
+
+/* Check for already reserved areas */
+static inline int __init bad_addr_size(u64 *addrp, u64 *sizep, u64 align)
+{
+       int i;
+       u64 addr = *addrp, last;
+       u64 size = *sizep;
+       int changed = 0;
+again:
+       last = addr + size;
+       for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) {
+               struct early_res *r = &early_res[i];
+               if (last > r->start && addr < r->start) {
+                       size = r->start - addr;
+                       changed = 1;
+                       goto again;
+               }
+               if (last > r->end && addr < r->end) {
+                       addr = round_up(r->end, align);
+                       size = last - addr;
+                       changed = 1;
+                       goto again;
+               }
+               if (last <= r->end && addr >= r->start) {
+                       (*sizep)++;
+                       return 0;
+               }
+       }
+       if (changed) {
+               *addrp = addr;
+               *sizep = size;
+       }
+       return changed;
+}
+
+/*
+ * Find a free area with specified alignment in a specific range.
+ */
+u64 __init find_e820_area(u64 start, u64 end, u64 size, u64 align)
+{
+       int i;
+
+       for (i = 0; i < e820.nr_map; i++) {
+               struct e820entry *ei = &e820.map[i];
+               u64 addr, last;
+               u64 ei_last;
+
+               if (ei->type != E820_RAM)
+                       continue;
+               addr = round_up(ei->addr, align);
+               ei_last = ei->addr + ei->size;
+               if (addr < start)
+                       addr = round_up(start, align);
+               if (addr >= ei_last)
+                       continue;
+               while (bad_addr(&addr, size, align) && addr+size <= ei_last)
+                       ;
+               last = addr + size;
+               if (last > ei_last)
+                       continue;
+               if (last > end)
+                       continue;
+               return addr;
+       }
+       return -1ULL;
+}
+
+/*
+ * Find next free range after *start
+ */
+u64 __init find_e820_area_size(u64 start, u64 *sizep, u64 align)
+{
+       int i;
+
+       for (i = 0; i < e820.nr_map; i++) {
+               struct e820entry *ei = &e820.map[i];
+               u64 addr, last;
+               u64 ei_last;
+
+               if (ei->type != E820_RAM)
+                       continue;
+               addr = round_up(ei->addr, align);
+               ei_last = ei->addr + ei->size;
+               if (addr < start)
+                       addr = round_up(start, align);
+               if (addr >= ei_last)
+                       continue;
+               *sizep = ei_last - addr;
+               while (bad_addr_size(&addr, sizep, align) &&
+                       addr + *sizep <= ei_last)
+                       ;
+               last = addr + *sizep;
+               if (last > ei_last)
+                       continue;
+               return addr;
+       }
+       return -1UL;
+
+}
+
+/*
+ * pre allocated 4k and reserved it in e820
+ */
+u64 __init early_reserve_e820(u64 startt, u64 sizet, u64 align)
+{
+       u64 size = 0;
+       u64 addr;
+       u64 start;
+
+       start = startt;
+       while (size < sizet)
+               start = find_e820_area_size(start, &size, align);
+
+       if (size < sizet)
+               return 0;
+
+       addr = round_down(start + size - sizet, align);
+       e820_update_range(addr, sizet, E820_RAM, E820_RESERVED);
+       e820_update_range_saved(addr, sizet, E820_RAM, E820_RESERVED);
+       printk(KERN_INFO "update e820 for early_reserve_e820\n");
+       update_e820();
+       update_e820_saved();
+
+       return addr;
+}
+
+#ifdef CONFIG_X86_32
+# ifdef CONFIG_X86_PAE
+#  define MAX_ARCH_PFN         (1ULL<<(40-PAGE_SHIFT))
+# else
+#  define MAX_ARCH_PFN         (1ULL<<(32-PAGE_SHIFT))
+# endif
+#else /* CONFIG_X86_32 */
+# define MAX_ARCH_PFN MAXMEM>>PAGE_SHIFT
+#endif
+
+/*
+ * Find the highest page frame number we have available
+ */
+static unsigned long __init e820_end_pfn(unsigned long limit_pfn, unsigned type)
+{
+       int i;
+       unsigned long last_pfn = 0;
+       unsigned long max_arch_pfn = MAX_ARCH_PFN;
+
+       for (i = 0; i < e820.nr_map; i++) {
+               struct e820entry *ei = &e820.map[i];
+               unsigned long start_pfn;
+               unsigned long end_pfn;
+
+               if (ei->type != type)
+                       continue;
+
+               start_pfn = ei->addr >> PAGE_SHIFT;
+               end_pfn = (ei->addr + ei->size) >> PAGE_SHIFT;
+
+               if (start_pfn >= limit_pfn)
+                       continue;
+               if (end_pfn > limit_pfn) {
+                       last_pfn = limit_pfn;
+                       break;
+               }
+               if (end_pfn > last_pfn)
+                       last_pfn = end_pfn;
+       }
+
+       if (last_pfn > max_arch_pfn)
+               last_pfn = max_arch_pfn;
+
+       printk(KERN_INFO "last_pfn = %#lx max_arch_pfn = %#lx\n",
+                        last_pfn, max_arch_pfn);
+       return last_pfn;
+}
+unsigned long __init e820_end_of_ram_pfn(void)
+{
+       return e820_end_pfn(MAX_ARCH_PFN, E820_RAM);
+}
+
+unsigned long __init e820_end_of_low_ram_pfn(void)
+{
+       return e820_end_pfn(1UL<<(32 - PAGE_SHIFT), E820_RAM);
+}
+/*
+ * Finds an active region in the address range from start_pfn to last_pfn and
+ * returns its range in ei_startpfn and ei_endpfn for the e820 entry.
+ */
+int __init e820_find_active_region(const struct e820entry *ei,
+                                 unsigned long start_pfn,
+                                 unsigned long last_pfn,
+                                 unsigned long *ei_startpfn,
+                                 unsigned long *ei_endpfn)
+{
+       u64 align = PAGE_SIZE;
+
+       *ei_startpfn = round_up(ei->addr, align) >> PAGE_SHIFT;
+       *ei_endpfn = round_down(ei->addr + ei->size, align) >> PAGE_SHIFT;
+
+       /* Skip map entries smaller than a page */
+       if (*ei_startpfn >= *ei_endpfn)
+               return 0;
+
+       /* Skip if map is outside the node */
+       if (ei->type != E820_RAM || *ei_endpfn <= start_pfn ||
+                                   *ei_startpfn >= last_pfn)
+               return 0;
+
+       /* Check for overlaps */
+       if (*ei_startpfn < start_pfn)
+               *ei_startpfn = start_pfn;
+       if (*ei_endpfn > last_pfn)
+               *ei_endpfn = last_pfn;
+
+       return 1;
+}
+
+/* Walk the e820 map and register active regions within a node */
+void __init e820_register_active_regions(int nid, unsigned long start_pfn,
+                                        unsigned long last_pfn)
+{
+       unsigned long ei_startpfn;
+       unsigned long ei_endpfn;
+       int i;
+
+       for (i = 0; i < e820.nr_map; i++)
+               if (e820_find_active_region(&e820.map[i],
+                                           start_pfn, last_pfn,
+                                           &ei_startpfn, &ei_endpfn))
+                       add_active_range(nid, ei_startpfn, ei_endpfn);
+}
+
+/*
+ * Find the hole size (in bytes) in the memory range.
+ * @start: starting address of the memory range to scan
+ * @end: ending address of the memory range to scan
+ */
+u64 __init e820_hole_size(u64 start, u64 end)
+{
+       unsigned long start_pfn = start >> PAGE_SHIFT;
+       unsigned long last_pfn = end >> PAGE_SHIFT;
+       unsigned long ei_startpfn, ei_endpfn, ram = 0;
+       int i;
+
+       for (i = 0; i < e820.nr_map; i++) {
+               if (e820_find_active_region(&e820.map[i],
+                                           start_pfn, last_pfn,
+                                           &ei_startpfn, &ei_endpfn))
+                       ram += ei_endpfn - ei_startpfn;
+       }
+       return end - start - ((u64)ram << PAGE_SHIFT);
+}
+
+static void early_panic(char *msg)
+{
+       early_printk(msg);
+       panic(msg);
+}
+
+static int userdef __initdata;
+
+/* "mem=nopentium" disables the 4MB page tables. */
+static int __init parse_memopt(char *p)
+{
+       u64 mem_size, current_end;
+       unsigned int i;
+
+       if (!p)
+               return -EINVAL;
+
+#ifdef CONFIG_X86_32
+       if (!strcmp(p, "nopentium")) {
+               setup_clear_cpu_cap(X86_FEATURE_PSE);
+               return 0;
+       }
+#endif
+
+       userdef = 1;
+       mem_size = memparse(p, &p);
+       e820_remove_range(mem_size, ULLONG_MAX - mem_size, E820_RAM, 1);
+
+       i = e820.nr_map - 1;
+       current_end = e820.map[i].addr + e820.map[i].size;
+       if (current_end < mem_size) {
+               /*
+                * The e820 map ends before our requested size so
+                * extend the final entry to the requested address.
+                */
+               if (e820.map[i].type == E820_RAM)
+                       e820.map[i].size = mem_size - e820.map[i].addr;
+               else
+                       e820_add_region(current_end, mem_size - current_end, E820_RAM);
+       }
+
+       return 0;
+}
+early_param("mem", parse_memopt);
+
+#ifndef CONFIG_XEN
+static int __init parse_memmap_opt(char *p)
+{
+       char *oldp;
+       u64 start_at, mem_size;
+
+       if (!p)
+               return -EINVAL;
+
+       if (!strncmp(p, "exactmap", 8)) {
+#ifdef CONFIG_CRASH_DUMP
+               /*
+                * If we are doing a crash dump, we still need to know
+                * the real mem size before original memory map is
+                * reset.
+                */
+               saved_max_pfn = e820_end_of_ram_pfn();
+#endif
+               e820.nr_map = 0;
+               userdef = 1;
+               return 0;
+       }
+
+       oldp = p;
+       mem_size = memparse(p, &p);
+       if (p == oldp)
+               return -EINVAL;
+
+       userdef = 1;
+       if (*p == '@') {
+               start_at = memparse(p+1, &p);
+               e820_add_region(start_at, mem_size, E820_RAM);
+       } else if (*p == '#') {
+               start_at = memparse(p+1, &p);
+               e820_add_region(start_at, mem_size, E820_ACPI);
+       } else if (*p == '$') {
+               start_at = memparse(p+1, &p);
+               e820_add_region(start_at, mem_size, E820_RESERVED);
+       } else
+               e820_remove_range(mem_size, ULLONG_MAX - mem_size, E820_RAM, 1);
+
+       return *p == '\0' ? 0 : -EINVAL;
+}
+early_param("memmap", parse_memmap_opt);
+
+void __init finish_e820_parsing(void)
+{
+       if (userdef) {
+               int nr = e820.nr_map;
+
+               if (sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &nr) < 0)
+                       early_panic("Invalid user supplied memory map");
+               e820.nr_map = nr;
+
+               printk(KERN_INFO "user-defined physical RAM map:\n");
+               e820_print_map("user");
+       }
+}
+#endif
+
+static inline const char *e820_type_to_string(int e820_type)
+{
+       switch (e820_type) {
+       case E820_RESERVED_KERN:
+       case E820_RAM:  return "System RAM";
+       case E820_ACPI: return "ACPI Tables";
+       case E820_NVS:  return "ACPI Non-volatile Storage";
+       default:        return "reserved";
+       }
+}
+
+#ifdef CONFIG_XEN
+#define e820 machine_e820
+#endif
+
+/*
+ * Mark e820 reserved areas as busy for the resource manager.
+ */
+void __init e820_reserve_resources(void)
+{
+       int i;
+       struct resource *res;
+       u64 end;
+
+       res = alloc_bootmem_low(sizeof(struct resource) * e820.nr_map);
+       for (i = 0; i < e820.nr_map; i++) {
+               end = e820.map[i].addr + e820.map[i].size - 1;
+#ifndef CONFIG_RESOURCES_64BIT
+               if (end > 0x100000000ULL) {
+                       res++;
+                       continue;
+               }
+#endif
+               res->name = e820_type_to_string(e820.map[i].type);
+               res->start = e820.map[i].addr;
+               res->end = end;
+
+               res->flags = IORESOURCE_MEM | IORESOURCE_BUSY;
+               insert_resource(&iomem_resource, res);
+               res++;
+       }
+
+       for (i = 0; i < e820_saved.nr_map; i++) {
+               struct e820entry *entry = &e820_saved.map[i];
+               firmware_map_add_early(entry->addr,
+                       entry->addr + entry->size - 1,
+                       e820_type_to_string(entry->type));
+       }
+}
+
+#undef e820
+
+#ifndef CONFIG_XEN
+char *__init default_machine_specific_memory_setup(void)
+{
+       char *who = "BIOS-e820";
+       int new_nr;
+       /*
+        * Try to copy the BIOS-supplied E820-map.
+        *
+        * Otherwise fake a memory map; one section from 0k->640k,
+        * the next section from 1mb->appropriate_mem_k
+        */
+       new_nr = boot_params.e820_entries;
+       sanitize_e820_map(boot_params.e820_map,
+                       ARRAY_SIZE(boot_params.e820_map),
+                       &new_nr);
+       boot_params.e820_entries = new_nr;
+       if (append_e820_map(boot_params.e820_map, boot_params.e820_entries)
+         < 0) {
+               u64 mem_size;
+
+               /* compare results from other methods and take the greater */
+               if (boot_params.alt_mem_k
+                   < boot_params.screen_info.ext_mem_k) {
+                       mem_size = boot_params.screen_info.ext_mem_k;
+                       who = "BIOS-88";
+               } else {
+                       mem_size = boot_params.alt_mem_k;
+                       who = "BIOS-e801";
+               }
+
+               e820.nr_map = 0;
+               e820_add_region(0, LOWMEMSIZE(), E820_RAM);
+               e820_add_region(HIGH_MEMORY, mem_size << 10, E820_RAM);
+       }
+
+       /* In case someone cares... */
+       return who;
+}
+
+char *__init __attribute__((weak)) machine_specific_memory_setup(void)
+{
+       if (x86_quirks->arch_memory_setup) {
+               char *who = x86_quirks->arch_memory_setup();
+
+               if (who)
+                       return who;
+       }
+       return default_machine_specific_memory_setup();
+}
+#endif
+
+char * __init memory_setup(void)
+{
+       int rc, nr_map;
+       struct xen_memory_map memmap;
+       /*
+        * This is rather large for a stack variable but this early in
+        * the boot process we know we have plenty slack space.
+        */
+       struct e820entry map[E820MAX];
+
+       memmap.nr_entries = E820MAX;
+       set_xen_guest_handle(memmap.buffer, map);
+
+       rc = HYPERVISOR_memory_op(XENMEM_memory_map, &memmap);
+       if (rc == -ENOSYS) {
+               memmap.nr_entries = 1;
+               map[0].addr = 0ULL;
+               map[0].size = PFN_PHYS((unsigned long long)xen_start_info->nr_pages);
+               /* 8MB slack (to balance backend allocations). */
+               map[0].size += 8ULL << 20;
+               map[0].type = E820_RAM;
+               rc = 0;
+       }
+       BUG_ON(rc);
+
+       nr_map = memmap.nr_entries;
+       sanitize_e820_map(map, ARRAY_SIZE(map), &nr_map);
+
+       if (append_e820_map(map, nr_map) < 0)
+               BUG();
+
+#ifdef CONFIG_XEN
+       if (is_initial_xendomain()) {
+               memmap.nr_entries = E820MAX;
+               set_xen_guest_handle(memmap.buffer, machine_e820.map);
+
+               if (HYPERVISOR_memory_op(XENMEM_machine_memory_map, &memmap))
+                       BUG();
+               machine_e820.nr_map = memmap.nr_entries;
+       } else
+               machine_e820 = e820;
+#endif
+
+       return "Xen";
+}
+
+void __init setup_memory_map(void)
+{
+       char *who;
+
+       who = memory_setup();
+       memcpy(&e820_saved, &e820, sizeof(struct e820map));
+       printk(KERN_INFO "Xen-provided physical RAM map:\n");
+       e820_print_map(who);
+}
Index: head-2008-12-01/arch/x86/kernel/e820_32-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/e820_32-xen.c  2008-12-01 11:44:55.000000000 +0100
+++ /dev/null   1970-01-01 00:00:00.000000000 +0000
@@ -1,873 +0,0 @@
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <linux/init.h>
-#include <linux/bootmem.h>
-#include <linux/ioport.h>
-#include <linux/string.h>
-#include <linux/kexec.h>
-#include <linux/module.h>
-#include <linux/mm.h>
-#include <linux/pfn.h>
-#include <linux/uaccess.h>
-#include <linux/suspend.h>
-
-#include <asm/pgtable.h>
-#include <asm/page.h>
-#include <asm/e820.h>
-#include <asm/setup.h>
-#include <xen/interface/memory.h>
-
-struct e820map e820;
-struct change_member {
-       struct e820entry *pbios; /* pointer to original bios entry */
-       unsigned long long addr; /* address for this change point */
-};
-static struct change_member change_point_list[2*E820MAX] __initdata;
-static struct change_member *change_point[2*E820MAX] __initdata;
-static struct e820entry *overlap_list[E820MAX] __initdata;
-static struct e820entry new_bios[E820MAX] __initdata;
-/* For PCI or other memory-mapped resources */
-unsigned long pci_mem_start = 0x10000000;
-#ifdef CONFIG_PCI
-EXPORT_SYMBOL(pci_mem_start);
-#endif
-extern int user_defined_memmap;
-
-static struct resource system_rom_resource = {
-       .name   = "System ROM",
-       .start  = 0xf0000,
-       .end    = 0xfffff,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
-};
-
-static struct resource extension_rom_resource = {
-       .name   = "Extension ROM",
-       .start  = 0xe0000,
-       .end    = 0xeffff,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
-};
-
-static struct resource adapter_rom_resources[] = { {
-       .name   = "Adapter ROM",
-       .start  = 0xc8000,
-       .end    = 0,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
-}, {
-       .name   = "Adapter ROM",
-       .start  = 0,
-       .end    = 0,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
-}, {
-       .name   = "Adapter ROM",
-       .start  = 0,
-       .end    = 0,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
-}, {
-       .name   = "Adapter ROM",
-       .start  = 0,
-       .end    = 0,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
-}, {
-       .name   = "Adapter ROM",
-       .start  = 0,
-       .end    = 0,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
-}, {
-       .name   = "Adapter ROM",
-       .start  = 0,
-       .end    = 0,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
-} };
-
-static struct resource video_rom_resource = {
-       .name   = "Video ROM",
-       .start  = 0xc0000,
-       .end    = 0xc7fff,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
-};
-
-#define ROMSIGNATURE 0xaa55
-
-static int __init romsignature(const unsigned char *rom)
-{
-       const unsigned short * const ptr = (const unsigned short *)rom;
-       unsigned short sig;
-
-       return probe_kernel_address(ptr, sig) == 0 && sig == ROMSIGNATURE;
-}
-
-static int __init romchecksum(const unsigned char *rom, unsigned long length)
-{
-       unsigned char sum, c;
-
-       for (sum = 0; length && probe_kernel_address(rom++, c) == 0; length--)
-               sum += c;
-       return !length && !sum;
-}
-
-static void __init probe_roms(void)
-{
-       const unsigned char *rom;
-       unsigned long start, length, upper;
-       unsigned char c;
-       int i;
-
-#ifdef CONFIG_XEN
-       /* Nothing to do if not running in dom0. */
-       if (!is_initial_xendomain())
-               return;
-#endif
-
-       /* video rom */
-       upper = adapter_rom_resources[0].start;
-       for (start = video_rom_resource.start; start < upper; start += 2048) {
-               rom = isa_bus_to_virt(start);
-               if (!romsignature(rom))
-                       continue;
-
-               video_rom_resource.start = start;
-
-               if (probe_kernel_address(rom + 2, c) != 0)
-                       continue;
-
-               /* 0 < length <= 0x7f * 512, historically */
-               length = c * 512;
-
-               /* if checksum okay, trust length byte */
-               if (length && romchecksum(rom, length))
-                       video_rom_resource.end = start + length - 1;
-
-               request_resource(&iomem_resource, &video_rom_resource);
-               break;
-       }
-
-       start = (video_rom_resource.end + 1 + 2047) & ~2047UL;
-       if (start < upper)
-               start = upper;
-
-       /* system rom */
-       request_resource(&iomem_resource, &system_rom_resource);
-       upper = system_rom_resource.start;
-
-       /* check for extension rom (ignore length byte!) */
-       rom = isa_bus_to_virt((unsigned long)extension_rom_resource.start);
-       if (romsignature(rom)) {
-               length = extension_rom_resource.end - extension_rom_resource.start + 1;
-               if (romchecksum(rom, length)) {
-                       request_resource(&iomem_resource, &extension_rom_resource);
-                       upper = extension_rom_resource.start;
-               }
-       }
-
-       /* check for adapter roms on 2k boundaries */
-       for (i = 0; i < ARRAY_SIZE(adapter_rom_resources) && start < upper; start += 2048) {
-               rom = isa_bus_to_virt(start);
-               if (!romsignature(rom))
-                       continue;
-
-               if (probe_kernel_address(rom + 2, c) != 0)
-                       continue;
-
-               /* 0 < length <= 0x7f * 512, historically */
-               length = c * 512;
-
-               /* but accept any length that fits if checksum okay */
-               if (!length || start + length > upper || !romchecksum(rom, length))
-                       continue;
-
-               adapter_rom_resources[i].start = start;
-               adapter_rom_resources[i].end = start + length - 1;
-               request_resource(&iomem_resource, &adapter_rom_resources[i]);
-
-               start = adapter_rom_resources[i++].end & ~2047UL;
-       }
-}
-
-#ifdef CONFIG_XEN
-static struct e820map machine_e820;
-#define e820 machine_e820
-#endif
-
-/*
- * Request address space for all standard RAM and ROM resources
- * and also for regions reported as reserved by the e820.
- */
-void __init init_iomem_resources(struct resource *code_resource,
-               struct resource *data_resource,
-               struct resource *bss_resource)
-{
-       int i;
-
-       probe_roms();
-       for (i = 0; i < e820.nr_map; i++) {
-               struct resource *res;
-#ifndef CONFIG_RESOURCES_64BIT
-               if (e820.map[i].addr + e820.map[i].size > 0x100000000ULL)
-                       continue;
-#endif
-               res = kzalloc(sizeof(struct resource), GFP_ATOMIC);
-               switch (e820.map[i].type) {
-               case E820_RAM:  res->name = "System RAM"; break;
-               case E820_ACPI: res->name = "ACPI Tables"; break;
-               case E820_NVS:  res->name = "ACPI Non-volatile Storage"; break;
-               default:        res->name = "reserved";
-               }
-               res->start = e820.map[i].addr;
-               res->end = res->start + e820.map[i].size - 1;
-               res->flags = IORESOURCE_MEM | IORESOURCE_BUSY;
-               if (request_resource(&iomem_resource, res)) {
-                       kfree(res);
-                       continue;
-               }
-               if (e820.map[i].type == E820_RAM) {
-                       /*
-                        *  We don't know which RAM region contains kernel data,
-                        *  so we try it repeatedly and let the resource manager
-                        *  test it.
-                        */
-#ifndef CONFIG_XEN
-                       request_resource(res, code_resource);
-                       request_resource(res, data_resource);
-                       request_resource(res, bss_resource);
-#endif
-#ifdef CONFIG_KEXEC
-                       if (crashk_res.start != crashk_res.end)
-                               request_resource(res, &crashk_res);
-#ifdef CONFIG_XEN
-                       xen_machine_kexec_register_resources(res);
-#endif
-#endif
-               }
-       }
-}
-
-#undef e820
-
-#if defined(CONFIG_PM) && defined(CONFIG_HIBERNATION)
-/**
- * e820_mark_nosave_regions - Find the ranges of physical addresses that do not
- * correspond to e820 RAM areas and mark the corresponding pages as nosave for
- * hibernation.
- *
- * This function requires the e820 map to be sorted and without any
- * overlapping entries and assumes the first e820 area to be RAM.
- */
-void __init e820_mark_nosave_regions(void)
-{
-       int i;
-       unsigned long pfn;
-
-       pfn = PFN_DOWN(e820.map[0].addr + e820.map[0].size);
-       for (i = 1; i < e820.nr_map; i++) {
-               struct e820entry *ei = &e820.map[i];
-
-               if (pfn < PFN_UP(ei->addr))
-                       register_nosave_region(pfn, PFN_UP(ei->addr));
-
-               pfn = PFN_DOWN(ei->addr + ei->size);
-               if (ei->type != E820_RAM)
-                       register_nosave_region(PFN_UP(ei->addr), pfn);
-
-               if (pfn >= max_low_pfn)
-                       break;
-       }
-}
-#endif
-
-void __init add_memory_region(unsigned long long start,
-                             unsigned long long size, int type)
-{
-       int x;
-
-       x = e820.nr_map;
-
-       if (x == E820MAX) {
-               printk(KERN_ERR "Ooops! Too many entries in the memory map!\n");
-               return;
-       }
-
-       e820.map[x].addr = start;
-       e820.map[x].size = size;
-       e820.map[x].type = type;
-       e820.nr_map++;
-} /* add_memory_region */
-
-/*
- * Sanitize the BIOS e820 map.
- *
- * Some e820 responses include overlapping entries.  The following
- * replaces the original e820 map with a new one, removing overlaps.
- *
- */
-int __init sanitize_e820_map(struct e820entry * biosmap, char * pnr_map)
-{
-       struct change_member *change_tmp;
-       unsigned long current_type, last_type;
-       unsigned long long last_addr;
-       int chgidx, still_changing;
-       int overlap_entries;
-       int new_bios_entry;
-       int old_nr, new_nr, chg_nr;
-       int i;
-
-       /*
-               Visually we're performing the following (1,2,3,4 = memory types)...
-
-               Sample memory map (w/overlaps):
-                  ____22__________________
-                  ______________________4_
-                  ____1111________________
-                  _44_____________________
-                  11111111________________
-                  ____________________33__
-                  ___________44___________
-                  __________33333_________
-                  ______________22________
-                  ___________________2222_
-                  _________111111111______
-                  _____________________11_
-                  _________________4______
-
-               Sanitized equivalent (no overlap):
-                  1_______________________
-                  _44_____________________
-                  ___1____________________
-                  ____22__________________
-                  ______11________________
-                  _________1______________
-                  __________3_____________
-                  ___________44___________
-                  _____________33_________
-                  _______________2________
-                  ________________1_______
-                  _________________4______
-                  ___________________2____
-                  ____________________33__
-                  ______________________4_
-       */
-       /* if there's only one memory region, don't bother */
-       if (*pnr_map < 2) {
-               return -1;
-       }
-
-       old_nr = *pnr_map;
-
-       /* bail out if we find any unreasonable addresses in bios map */
-       for (i=0; i<old_nr; i++)
-               if (biosmap[i].addr + biosmap[i].size < biosmap[i].addr) {
-                       return -1;
-               }
-
-       /* create pointers for initial change-point information (for sorting) */
-       for (i=0; i < 2*old_nr; i++)
-               change_point[i] = &change_point_list[i];
-
-       /* record all known change-points (starting and ending addresses),
-          omitting those that are for empty memory regions */
-       chgidx = 0;
-       for (i=0; i < old_nr; i++)      {
-               if (biosmap[i].size != 0) {
-                       change_point[chgidx]->addr = biosmap[i].addr;
-                       change_point[chgidx++]->pbios = &biosmap[i];
-                       change_point[chgidx]->addr = biosmap[i].addr + biosmap[i].size;
-                       change_point[chgidx++]->pbios = &biosmap[i];
-               }
-       }
-       chg_nr = chgidx;        /* true number of change-points */
-
-       /* sort change-point list by memory addresses (low -> high) */
-       still_changing = 1;
-       while (still_changing)  {
-               still_changing = 0;
-               for (i=1; i < chg_nr; i++)  {
-                       /* if <current_addr> > <last_addr>, swap */
-                       /* or, if current=<start_addr> & last=<end_addr>, swap */
-                       if ((change_point[i]->addr < change_point[i-1]->addr) ||
-                               ((change_point[i]->addr == change_point[i-1]->addr) &&
-                                (change_point[i]->addr == change_point[i]->pbios->addr) &&
-                                (change_point[i-1]->addr != change_point[i-1]->pbios->addr))
-                          )
-                       {
-                               change_tmp = change_point[i];
-                               change_point[i] = change_point[i-1];
-                               change_point[i-1] = change_tmp;
-                               still_changing=1;
-                       }
-               }
-       }
-
-       /* create a new bios memory map, removing overlaps */
-       overlap_entries=0;       /* number of entries in the overlap table */
-       new_bios_entry=0;        /* index for creating new bios map entries */
-       last_type = 0;           /* start with undefined memory type */
-       last_addr = 0;           /* start with 0 as last starting address */
-       /* loop through change-points, determining affect on the new bios map */
-       for (chgidx=0; chgidx < chg_nr; chgidx++)
-       {
-               /* keep track of all overlapping bios entries */
-               if (change_point[chgidx]->addr == change_point[chgidx]->pbios->addr)
-               {
-                       /* add map entry to overlap list (> 1 entry implies an overlap) */
-                       overlap_list[overlap_entries++]=change_point[chgidx]->pbios;
-               }
-               else
-               {
-                       /* remove entry from list (order independent, so swap with last) */
-                       for (i=0; i<overlap_entries; i++)
-                       {
-                               if (overlap_list[i] == change_point[chgidx]->pbios)
-                                       overlap_list[i] = overlap_list[overlap_entries-1];
-                       }
-                       overlap_entries--;
-               }
-               /* if there are overlapping entries, decide which "type" to use */
-               /* (larger value takes precedence -- 1=usable, 2,3,4,4+=unusable) */
-               current_type = 0;
-               for (i=0; i<overlap_entries; i++)
-                       if (overlap_list[i]->type > current_type)
-                               current_type = overlap_list[i]->type;
-               /* continue building up new bios map based on this information */
-               if (current_type != last_type)  {
-                       if (last_type != 0)      {
-                               new_bios[new_bios_entry].size =
-                                       change_point[chgidx]->addr - last_addr;
-                               /* move forward only if the new size was non-zero */
-                               if (new_bios[new_bios_entry].size != 0)
-                                       if (++new_bios_entry >= E820MAX)
-                                               break;  /* no more space left for new bios entries */
-                       }
-                       if (current_type != 0)  {
-                               new_bios[new_bios_entry].addr = change_point[chgidx]->addr;
-                               new_bios[new_bios_entry].type = current_type;
-                               last_addr=change_point[chgidx]->addr;
-                       }
-                       last_type = current_type;
-               }
-       }
-       new_nr = new_bios_entry;   /* retain count for new bios entries */
-
-       /* copy new bios mapping into original location */
-       memcpy(biosmap, new_bios, new_nr*sizeof(struct e820entry));
-       *pnr_map = new_nr;
-
-       return 0;
-}
-
-/*
- * Copy the BIOS e820 map into a safe place.
- *
- * Sanity-check it while we're at it..
- *
- * If we're lucky and live on a modern system, the setup code
- * will have given us a memory map that we can use to properly
- * set up memory.  If we aren't, we'll fake a memory map.
- *
- * We check to see that the memory map contains at least 2 elements
- * before we'll use it, because the detection code in setup.S may
- * not be perfect and most every PC known to man has two memory
- * regions: one from 0 to 640k, and one from 1mb up.  (The IBM
- * thinkpad 560x, for example, does not cooperate with the memory
- * detection code.)
- */
-int __init copy_e820_map(struct e820entry *biosmap, int nr_map)
-{
-#ifndef CONFIG_XEN
-       /* Only one memory region (or negative)? Ignore it */
-       if (nr_map < 2)
-               return -1;
-#else
-       BUG_ON(nr_map < 1);
-#endif
-
-       do {
-               u64 start = biosmap->addr;
-               u64 size = biosmap->size;
-               u64 end = start + size;
-               u32 type = biosmap->type;
-
-               /* Overflow in 64 bits? Ignore the memory map. */
-               if (start > end)
-                       return -1;
-
-               add_memory_region(start, size, type);
-       } while (biosmap++, --nr_map);
-
-#ifdef CONFIG_XEN
-       if (is_initial_xendomain()) {
-               struct xen_memory_map memmap;
-
-               memmap.nr_entries = E820MAX;
-               set_xen_guest_handle(memmap.buffer, machine_e820.map);
-
-               if (HYPERVISOR_memory_op(XENMEM_machine_memory_map, &memmap))
-                       BUG();
-               machine_e820.nr_map = memmap.nr_entries;
-       } else
-               machine_e820 = e820;
-#endif
-
-       return 0;
-}
-
-/*
- * Find the highest page frame number we have available
- */
-void __init propagate_e820_map(void)
-{
-       int i;
-
-       max_pfn = 0;
-
-       for (i = 0; i < e820.nr_map; i++) {
-               unsigned long start, end;
-               /* RAM? */
-               if (e820.map[i].type != E820_RAM)
-                       continue;
-               start = PFN_UP(e820.map[i].addr);
-               end = PFN_DOWN(e820.map[i].addr + e820.map[i].size);
-               if (start >= end)
-                       continue;
-               if (end > max_pfn)
-                       max_pfn = end;
-               memory_present(0, start, end);
-       }
-}
-
-/*
- * Register fully available low RAM pages with the bootmem allocator.
- */
-void __init register_bootmem_low_pages(unsigned long max_low_pfn)
-{
-       int i;
-
-       for (i = 0; i < e820.nr_map; i++) {
-               unsigned long curr_pfn, last_pfn, size;
-               /*
-                * Reserve usable low memory
-                */
-               if (e820.map[i].type != E820_RAM)
-                       continue;
-               /*
-                * We are rounding up the start address of usable memory:
-                */
-               curr_pfn = PFN_UP(e820.map[i].addr);
-               if (curr_pfn >= max_low_pfn)
-                       continue;
-               /*
-                * ... and at the end of the usable range downwards:
-                */
-               last_pfn = PFN_DOWN(e820.map[i].addr + e820.map[i].size);
-
-#ifdef CONFIG_XEN
-               /*
-                * Truncate to the number of actual pages currently
-                * present.
-                */
-               if (last_pfn > xen_start_info->nr_pages)
-                       last_pfn = xen_start_info->nr_pages;
-#endif
-
-               if (last_pfn > max_low_pfn)
-                       last_pfn = max_low_pfn;
-
-               /*
-                * .. finally, did all the rounding and playing
-                * around just make the area go away?
-                */
-               if (last_pfn <= curr_pfn)
-                       continue;
-
-               size = last_pfn - curr_pfn;
-               free_bootmem(PFN_PHYS(curr_pfn), PFN_PHYS(size));
-       }
-}
-
-void __init e820_register_memory(void)
-{
-       unsigned long gapstart, gapsize, round;
-       unsigned long long last;
-       int i;
-
-#ifdef CONFIG_XEN
-       if (is_initial_xendomain()) {
-               struct xen_memory_map memmap;
-
-               memmap.nr_entries = E820MAX;
-               set_xen_guest_handle(memmap.buffer, machine_e820.map);
-
-               if (HYPERVISOR_memory_op(XENMEM_machine_memory_map, &memmap))
-                       BUG();
-               machine_e820.nr_map = memmap.nr_entries;
-       }
-       else
-               machine_e820 = e820;
-#define e820 machine_e820
-#endif
-
-       /*
-        * Search for the biggest gap in the low 32 bits of the e820
-        * memory space.
-        */
-       last = 0x100000000ull;
-       gapstart = 0x10000000;
-       gapsize = 0x400000;
-       i = e820.nr_map;
-       while (--i >= 0) {
-               unsigned long long start = e820.map[i].addr;
-               unsigned long long end = start + e820.map[i].size;
-
-               /*
-                * Since "last" is at most 4GB, we know we'll
-                * fit in 32 bits if this condition is true
-                */
-               if (last > end) {
-                       unsigned long gap = last - end;
-
-                       if (gap > gapsize) {
-                               gapsize = gap;
-                               gapstart = end;
-                       }
-               }
-               if (start < last)
-                       last = start;
-       }
-#undef e820
-
-       /*
-        * See how much we want to round up: start off with
-        * rounding to the next 1MB area.
-        */
-       round = 0x100000;
-       while ((gapsize >> 4) > round)
-               round += round;
-       /* Fun with two's complement */
-       pci_mem_start = (gapstart + round) & -round;
-
-       printk("Allocating PCI resources starting at %08lx (gap: %08lx:%08lx)\n",
-               pci_mem_start, gapstart, gapsize);
-}
-
-void __init print_memory_map(char *who)
-{
-       int i;
-
-       for (i = 0; i < e820.nr_map; i++) {
-               printk(" %s: %016Lx - %016Lx ", who,
-                       e820.map[i].addr,
-                       e820.map[i].addr + e820.map[i].size);
-               switch (e820.map[i].type) {
-               case E820_RAM:  printk("(usable)\n");
-                               break;
-               case E820_RESERVED:
-                               printk("(reserved)\n");
-                               break;
-               case E820_ACPI:
-                               printk("(ACPI data)\n");
-                               break;
-               case E820_NVS:
-                               printk("(ACPI NVS)\n");
-                               break;
-               default:        printk("type %u\n", e820.map[i].type);
-                               break;
-               }
-       }
-}
-
-void __init limit_regions(unsigned long long size)
-{
-       unsigned long long current_addr = 0;
-       int i;
-
-       print_memory_map("limit_regions start");
-       for (i = 0; i < e820.nr_map; i++) {
-               current_addr = e820.map[i].addr + e820.map[i].size;
-               if (current_addr < size)
-                       continue;
-
-               if (e820.map[i].type != E820_RAM)
-                       continue;
-
-               if (e820.map[i].addr >= size) {
-                       /*
-                        * This region starts past the end of the
-                        * requested size, skip it completely.
-                        */
-                       e820.nr_map = i;
-               } else {
-                       e820.nr_map = i + 1;
-                       e820.map[i].size -= current_addr - size;
-               }
-               print_memory_map("limit_regions endfor");
-               return;
-       }
-#ifdef CONFIG_XEN
-       if (current_addr < size) {
-               /*
-                * The e820 map finished before our requested size so
-                * extend the final entry to the requested address.
-                */
-               --i;
-               if (e820.map[i].type == E820_RAM)
-                       e820.map[i].size -= current_addr - size;
-               else
-                       add_memory_region(current_addr, size - current_addr, E820_RAM);
-       }
-#endif
-       print_memory_map("limit_regions endfunc");
-}
-
-/*
- * This function checks if any part of the range <start,end> is mapped
- * with type.
- */
-int
-e820_any_mapped(u64 start, u64 end, unsigned type)
-{
-       int i;
-
-#ifndef CONFIG_XEN
-       for (i = 0; i < e820.nr_map; i++) {
-               const struct e820entry *ei = &e820.map[i];
-#else
-       if (!is_initial_xendomain())
-               return 0;
-       for (i = 0; i < machine_e820.nr_map; ++i) {
-               const struct e820entry *ei = &machine_e820.map[i];
-#endif
-
-               if (type && ei->type != type)
-                       continue;
-               if (ei->addr >= end || ei->addr + ei->size <= start)
-                       continue;
-               return 1;
-       }
-       return 0;
-}
-EXPORT_SYMBOL_GPL(e820_any_mapped);
-
- /*
-  * This function checks if the entire range <start,end> is mapped with type.
-  *
-  * Note: this function only works correct if the e820 table is sorted and
-  * not-overlapping, which is the case
-  */
-int __init
-e820_all_mapped(unsigned long s, unsigned long e, unsigned type)
-{
-       u64 start = s;
-       u64 end = e;
-       int i;
-
-#ifndef CONFIG_XEN
-       for (i = 0; i < e820.nr_map; i++) {
-               struct e820entry *ei = &e820.map[i];
-#else
-       if (!is_initial_xendomain())
-               return 0;
-       for (i = 0; i < machine_e820.nr_map; ++i) {
-               const struct e820entry *ei = &machine_e820.map[i];
-#endif
-
-               if (type && ei->type != type)
-                       continue;
-               /* is the region (part) in overlap with the current region ?*/
-               if (ei->addr >= end || ei->addr + ei->size <= start)
-                       continue;
-               /* if the region is at the beginning of <start,end> we move
-                * start to the end of the region since it's ok until there
-                */
-               if (ei->addr <= start)
-                       start = ei->addr + ei->size;
-               /* if start is now at or beyond end, we're done, full
-                * coverage */
-               if (start >= end)
-                       return 1; /* we're done */
-       }
-       return 0;
-}
-
-static int __init parse_memmap(char *arg)
-{
-       if (!arg)
-               return -EINVAL;
-
-       if (strcmp(arg, "exactmap") == 0) {
-#ifdef CONFIG_CRASH_DUMP
-               /* If we are doing a crash dump, we
-                * still need to know the real mem
-                * size before original memory map is
-                * reset.
-                */
-               propagate_e820_map();
-               saved_max_pfn = max_pfn;
-#endif
-               e820.nr_map = 0;
-               user_defined_memmap = 1;
-       } else {
-               /* If the user specifies memory size, we
-                * limit the BIOS-provided memory map to
-                * that size. exactmap can be used to specify
-                * the exact map. mem=number can be used to
-                * trim the existing memory map.
-                */
-               unsigned long long start_at, mem_size;
-
-               mem_size = memparse(arg, &arg);
-               if (*arg == '@') {
-                       start_at = memparse(arg+1, &arg);
-                       add_memory_region(start_at, mem_size, E820_RAM);
-               } else if (*arg == '#') {
-                       start_at = memparse(arg+1, &arg);
-                       add_memory_region(start_at, mem_size, E820_ACPI);
-               } else if (*arg == '$') {
-                       start_at = memparse(arg+1, &arg);
-                       add_memory_region(start_at, mem_size, E820_RESERVED);
-               } else {
-                       limit_regions(mem_size);
-                       user_defined_memmap = 1;
-               }
-       }
-       return 0;
-}
-early_param("memmap", parse_memmap);
-
-#ifndef CONFIG_XEN
-void __init update_memory_range(u64 start, u64 size, unsigned old_type,
-                               unsigned new_type)
-{
-       int i;
-
-       BUG_ON(old_type == new_type);
-
-       for (i = 0; i < e820.nr_map; i++) {
-               struct e820entry *ei = &e820.map[i];
-               u64 final_start, final_end;
-               if (ei->type != old_type)
-                       continue;
-               /* totally covered? */
-               if (ei->addr >= start && ei->size <= size) {
-                       ei->type = new_type;
-                       continue;
-               }
-               /* partially covered */
-               final_start = max(start, ei->addr);
-               final_end = min(start + size, ei->addr + ei->size);
-               if (final_start >= final_end)
-                       continue;
-               add_memory_region(final_start, final_end - final_start,
-                                        new_type);
-       }
-}
-
-void __init update_e820(void)
-{
-       u8 nr_map;
-
-       nr_map = e820.nr_map;
-       if (sanitize_e820_map(e820.map, &nr_map))
-               return;
-       e820.nr_map = nr_map;
-       printk(KERN_INFO "modified physical RAM map:\n");
-       print_memory_map("modified");
-}
-#endif
Index: head-2008-12-01/arch/x86/kernel/e820_64-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/e820_64-xen.c  2008-12-01 11:44:55.000000000 +0100
+++ /dev/null   1970-01-01 00:00:00.000000000 +0000
@@ -1,1045 +0,0 @@
-/*
- * Handle the memory map.
- * The functions here do the job until bootmem takes over.
- *
- *  Getting sanitize_e820_map() in sync with i386 version by applying change:
- *  -  Provisions for empty E820 memory regions (reported by certain BIOSes).
- *     Alex Achenbach <xela@slit.de>, December 2002.
- *  Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
- *
- */
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <linux/init.h>
-#include <linux/bootmem.h>
-#include <linux/ioport.h>
-#include <linux/string.h>
-#include <linux/kexec.h>
-#include <linux/module.h>
-#include <linux/mm.h>
-#include <linux/suspend.h>
-#include <linux/pfn.h>
-
-#include <asm/pgtable.h>
-#include <asm/page.h>
-#include <asm/e820.h>
-#include <asm/proto.h>
-#include <asm/setup.h>
-#include <asm/sections.h>
-#include <asm/kdebug.h>
-#include <xen/interface/memory.h>
-
-struct e820map e820 __initdata;
-#ifdef CONFIG_XEN
-struct e820map machine_e820;
-#endif
-
-/*
- * PFN of last memory page.
- */
-unsigned long end_pfn;
-
-/*
- * end_pfn only includes RAM, while max_pfn_mapped includes all e820 entries.
- * The direct mapping extends to max_pfn_mapped, so that we can directly access
- * apertures, ACPI and other tables without having to play with fixmaps.
- */
-unsigned long max_pfn_mapped;
-
-/*
- * Last pfn which the user wants to use.
- */
-static unsigned long __initdata end_user_pfn = MAXMEM>>PAGE_SHIFT;
-
-/*
- * Early reserved memory areas.
- */
-#define MAX_EARLY_RES 20
-
-struct early_res {
-       unsigned long start, end;
-       char name[16];
-};
-static struct early_res early_res[MAX_EARLY_RES] __initdata = {
-#ifndef CONFIG_XEN
-       { 0, PAGE_SIZE, "BIOS data page" },                     /* BIOS data page */
-#ifdef CONFIG_X86_TRAMPOLINE
-       { TRAMPOLINE_BASE, TRAMPOLINE_BASE + 2 * PAGE_SIZE, "TRAMPOLINE" },
-#endif
-#endif
-       {}
-};
-
-void __init reserve_early(unsigned long start, unsigned long end, char *name)
-{
-       int i;
-       struct early_res *r;
-       for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) {
-               r = &early_res[i];
-               if (end > r->start && start < r->end)
-                       panic("Overlapping early reservations %lx-%lx %s to %lx-%lx %s\n",
-                             start, end - 1, name?name:"", r->start, r->end - 1, r->name);
-       }
-       if (i >= MAX_EARLY_RES)
-               panic("Too many early reservations");
-       r = &early_res[i];
-       r->start = start;
-       r->end = end;
-       if (name)
-               strncpy(r->name, name, sizeof(r->name) - 1);
-}
-
-void __init free_early(unsigned long start, unsigned long end)
-{
-       struct early_res *r;
-       int i, j;
-
-       for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) {
-               r = &early_res[i];
-               if (start == r->start && end == r->end)
-                       break;
-       }
-       if (i >= MAX_EARLY_RES || !early_res[i].end)
-               panic("free_early on not reserved area: %lx-%lx!", start, end);
-
-       for (j = i + 1; j < MAX_EARLY_RES && early_res[j].end; j++)
-               ;
-
-       memmove(&early_res[i], &early_res[i + 1],
-              (j - 1 - i) * sizeof(struct early_res));
-
-       early_res[j - 1].end = 0;
-}
-
-void __init early_res_to_bootmem(unsigned long start, unsigned long end)
-{
-       int i;
-       unsigned long final_start, final_end;
-       for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) {
-               struct early_res *r = &early_res[i];
-               final_start = max(start, r->start);
-               final_end = min(end, r->end);
-               if (final_start >= final_end)
-                       continue;
-               printk(KERN_INFO "  early res: %d [%lx-%lx] %s\n", i,
-                       final_start, final_end - 1, r->name);
-               reserve_bootmem_generic(final_start, final_end - final_start);
-       }
-}
-
-/* Check for already reserved areas */
-static inline int __init
-bad_addr(unsigned long *addrp, unsigned long size, unsigned long align)
-{
-       int i;
-       unsigned long addr = *addrp, last;
-       int changed = 0;
-again:
-       last = addr + size;
-       for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) {
-               struct early_res *r = &early_res[i];
-               if (last >= r->start && addr < r->end) {
-                       *addrp = addr = round_up(r->end, align);
-                       changed = 1;
-                       goto again;
-               }
-       }
-       return changed;
-}
-
-/* Check for already reserved areas */
-static inline int __init
-bad_addr_size(unsigned long *addrp, unsigned long *sizep, unsigned long align)
-{
-       int i;
-       unsigned long addr = *addrp, last;
-       unsigned long size = *sizep;
-       int changed = 0;
-again:
-       last = addr + size;
-       for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) {
-               struct early_res *r = &early_res[i];
-               if (last > r->start && addr < r->start) {
-                       size = r->start - addr;
-                       changed = 1;
-                       goto again;
-               }
-               if (last > r->end && addr < r->end) {
-                       addr = round_up(r->end, align);
-                       size = last - addr;
-                       changed = 1;
-                       goto again;
-               }
-               if (last <= r->end && addr >= r->start) {
-                       (*sizep)++;
-                       return 0;
-               }
-       }
-       if (changed) {
-               *addrp = addr;
-               *sizep = size;
-       }
-       return changed;
-}
-/*
- * This function checks if any part of the range <start,end> is mapped
- * with type.
- */
-int
-e820_any_mapped(unsigned long start, unsigned long end, unsigned type)
-{
-       int i;
-
-#ifndef CONFIG_XEN
-       for (i = 0; i < e820.nr_map; i++) {
-               struct e820entry *ei = &e820.map[i];
-#else
-       if (!is_initial_xendomain())
-               return 0;
-       for (i = 0; i < machine_e820.nr_map; i++) {
-               const struct e820entry *ei = &machine_e820.map[i];
-#endif
-
-               if (type && ei->type != type)
-                       continue;
-               if (ei->addr >= end || ei->addr + ei->size <= start)
-                       continue;
-               return 1;
-       }
-       return 0;
-}
-EXPORT_SYMBOL_GPL(e820_any_mapped);
-
-/*
- * This function checks if the entire range <start,end> is mapped with type.
- *
- * Note: this function only works correct if the e820 table is sorted and
- * not-overlapping, which is the case
- */
-int __init e820_all_mapped(unsigned long start, unsigned long end,
-                          unsigned type)
-{
-       int i;
-
-#ifndef CONFIG_XEN
-       for (i = 0; i < e820.nr_map; i++) {
-               struct e820entry *ei = &e820.map[i];
-#else
-       if (!is_initial_xendomain())
-               return 0;
-       for (i = 0; i < machine_e820.nr_map; i++) {
-               const struct e820entry *ei = &machine_e820.map[i];
-#endif
-
-               if (type && ei->type != type)
-                       continue;
-               /* is the region (part) in overlap with the current region ?*/
-               if (ei->addr >= end || ei->addr + ei->size <= start)
-                       continue;
-
-               /* if the region is at the beginning of <start,end> we move
-                * start to the end of the region since it's ok until there
-                */
-               if (ei->addr <= start)
-                       start = ei->addr + ei->size;
-               /*
-                * if start is now at or beyond end, we're done, full
-                * coverage
-                */
-               if (start >= end)
-                       return 1;
-       }
-       return 0;
-}
-
-/*
- * Find a free area with specified alignment in a specific range.
- */
-unsigned long __init find_e820_area(unsigned long start, unsigned long end,
-                                   unsigned long size, unsigned long align)
-{
-       int i;
-
-       for (i = 0; i < e820.nr_map; i++) {
-               struct e820entry *ei = &e820.map[i];
-               unsigned long addr, last;
-               unsigned long ei_last;
-
-               if (ei->type != E820_RAM)
-                       continue;
-               addr = round_up(ei->addr, align);
-               ei_last = ei->addr + ei->size;
-               if (addr < start)
-                       addr = round_up(start, align);
-               if (addr >= ei_last)
-                       continue;
-               while (bad_addr(&addr, size, align) && addr+size <= ei_last)
-                       ;
-               last = addr + size;
-               if (last > ei_last)
-                       continue;
-               if (last > end)
-                       continue;
-               return addr;
-       }
-       return -1UL;
-}
-
-/*
- * Find next free range after *start
- */
-unsigned long __init find_e820_area_size(unsigned long start,
-                                        unsigned long *sizep,
-                                        unsigned long align)
-{
-       int i;
-
-       for (i = 0; i < e820.nr_map; i++) {
-               struct e820entry *ei = &e820.map[i];
-               unsigned long addr, last;
-               unsigned long ei_last;
-
-               if (ei->type != E820_RAM)
-                       continue;
-               addr = round_up(ei->addr, align);
-               ei_last = ei->addr + ei->size;
-               if (addr < start)
-                       addr = round_up(start, align);
-               if (addr >= ei_last)
-                       continue;
-               *sizep = ei_last - addr;
-               while (bad_addr_size(&addr, sizep, align) &&
-                       addr + *sizep <= ei_last)
-                       ;
-               last = addr + *sizep;
-               if (last > ei_last)
-                       continue;
-               return addr;
-       }
-       return -1UL;
-
-}
-/*
- * Find the highest page frame number we have available
- */
-unsigned long __init e820_end_of_ram(void)
-{
-       unsigned long end_pfn;
-
-       end_pfn = find_max_pfn_with_active_regions();
-
-       if (end_pfn > max_pfn_mapped)
-               max_pfn_mapped = end_pfn;
-       if (max_pfn_mapped > MAXMEM>>PAGE_SHIFT)
-               max_pfn_mapped = MAXMEM>>PAGE_SHIFT;
-       if (end_pfn > end_user_pfn)
-               end_pfn = end_user_pfn;
-       if (end_pfn > max_pfn_mapped)
-               end_pfn = max_pfn_mapped;
-
-       printk(KERN_INFO "max_pfn_mapped = %lu\n", max_pfn_mapped);
-       return end_pfn;
-}
-
-/*
- * Mark e820 reserved areas as busy for the resource manager.
- */
-void __init e820_reserve_resources(struct e820entry *e820, int nr_map)
-{
-       int i;
-       struct resource *res;
-
-       res = alloc_bootmem_low(sizeof(struct resource) * nr_map);
-       for (i = 0; i < nr_map; i++) {
-               switch (e820[i].type) {
-               case E820_RAM:  res->name = "System RAM"; break;
-               case E820_ACPI: res->name = "ACPI Tables"; break;
-               case E820_NVS:  res->name = "ACPI Non-volatile Storage"; break;
-               default:        res->name = "reserved";
-               }
-               res->start = e820[i].addr;
-               res->end = res->start + e820[i].size - 1;
-               res->flags = IORESOURCE_MEM | IORESOURCE_BUSY;
-               insert_resource(&iomem_resource, res);
-               res++;
-       }
-}
-
-#ifndef CONFIG_XEN
-/*
- * Find the ranges of physical addresses that do not correspond to
- * e820 RAM areas and mark the corresponding pages as nosave for software
- * suspend and suspend to RAM.
- *
- * This function requires the e820 map to be sorted and without any
- * overlapping entries and assumes the first e820 area to be RAM.
- */
-void __init e820_mark_nosave_regions(void)
-{
-       int i;
-       unsigned long paddr;
-
-       paddr = round_down(e820.map[0].addr + e820.map[0].size, PAGE_SIZE);
-       for (i = 1; i < e820.nr_map; i++) {
-               struct e820entry *ei = &e820.map[i];
-
-               if (paddr < ei->addr)
-                       register_nosave_region(PFN_DOWN(paddr),
-                                               PFN_UP(ei->addr));
-
-               paddr = round_down(ei->addr + ei->size, PAGE_SIZE);
-               if (ei->type != E820_RAM)
-                       register_nosave_region(PFN_UP(ei->addr),
-                                               PFN_DOWN(paddr));
-
-               if (paddr >= (end_pfn << PAGE_SHIFT))
-                       break;
-       }
-}
-#endif
-
-/*
- * Finds an active region in the address range from start_pfn to end_pfn and
- * returns its range in ei_startpfn and ei_endpfn for the e820 entry.
- */
-static int __init e820_find_active_region(const struct e820entry *ei,
-                                         unsigned long start_pfn,
-                                         unsigned long end_pfn,
-                                         unsigned long *ei_startpfn,
-                                         unsigned long *ei_endpfn)
-{
-       *ei_startpfn = round_up(ei->addr, PAGE_SIZE) >> PAGE_SHIFT;
-       *ei_endpfn = round_down(ei->addr + ei->size, PAGE_SIZE) >> PAGE_SHIFT;
-
-       /* Skip map entries smaller than a page */
-       if (*ei_startpfn >= *ei_endpfn)
-               return 0;
-
-       /* Check if max_pfn_mapped should be updated */
-       if (ei->type != E820_RAM && *ei_endpfn > max_pfn_mapped)
-               max_pfn_mapped = *ei_endpfn;
-
-       /* Skip if map is outside the node */
-       if (ei->type != E820_RAM || *ei_endpfn <= start_pfn ||
-                                   *ei_startpfn >= end_pfn)
-               return 0;
-
-       /* Check for overlaps */
-       if (*ei_startpfn < start_pfn)
-               *ei_startpfn = start_pfn;
-       if (*ei_endpfn > end_pfn)
-               *ei_endpfn = end_pfn;
-
-       /* Obey end_user_pfn to save on memmap */
-       if (*ei_startpfn >= end_user_pfn)
-               return 0;
-       if (*ei_endpfn > end_user_pfn)
-               *ei_endpfn = end_user_pfn;
-
-       return 1;
-}
-
-/* Walk the e820 map and register active regions within a node */
-void __init
-e820_register_active_regions(int nid, unsigned long start_pfn,
-                                                       unsigned long end_pfn)
-{
-       unsigned long ei_startpfn;
-       unsigned long ei_endpfn;
-       int i;
-
-       for (i = 0; i < e820.nr_map; i++)
-               if (e820_find_active_region(&e820.map[i],
-                                           start_pfn, end_pfn,
-                                           &ei_startpfn, &ei_endpfn))
-                       add_active_range(nid, ei_startpfn, ei_endpfn);
-}
-
-/*
- * Add a memory region to the kernel e820 map.
- */
-void __init add_memory_region(unsigned long start, unsigned long size, int type)
-{
-       int x = e820.nr_map;
-
-       if (x == E820MAX) {
-               printk(KERN_ERR "Ooops! Too many entries in the memory map!\n");
-               return;
-       }
-
-       e820.map[x].addr = start;
-       e820.map[x].size = size;
-       e820.map[x].type = type;
-       e820.nr_map++;
-}
-
-/*
- * Find the hole size (in bytes) in the memory range.
- * @start: starting address of the memory range to scan
- * @end: ending address of the memory range to scan
- */
-unsigned long __init e820_hole_size(unsigned long start, unsigned long end)
-{
-       unsigned long start_pfn = start >> PAGE_SHIFT;
-       unsigned long end_pfn = end >> PAGE_SHIFT;
-       unsigned long ei_startpfn, ei_endpfn, ram = 0;
-       int i;
-
-       for (i = 0; i < e820.nr_map; i++) {
-               if (e820_find_active_region(&e820.map[i],
-                                           start_pfn, end_pfn,
-                                           &ei_startpfn, &ei_endpfn))
-                       ram += ei_endpfn - ei_startpfn;
-       }
-       return end - start - (ram << PAGE_SHIFT);
-}
-
-static void __init e820_print_map(char *who)
-{
-       int i;
-
-       for (i = 0; i < e820.nr_map; i++) {
-               printk(KERN_INFO " %s: %016Lx - %016Lx ", who,
-                      (unsigned long long) e820.map[i].addr,
-                      (unsigned long long)
-                      (e820.map[i].addr + e820.map[i].size));
-               switch (e820.map[i].type) {
-               case E820_RAM:
-                       printk(KERN_CONT "(usable)\n");
-                       break;
-               case E820_RESERVED:
-                       printk(KERN_CONT "(reserved)\n");
-                       break;
-               case E820_ACPI:
-                       printk(KERN_CONT "(ACPI data)\n");
-                       break;
-               case E820_NVS:
-                       printk(KERN_CONT "(ACPI NVS)\n");
-                       break;
-               default:
-                       printk(KERN_CONT "type %u\n", e820.map[i].type);
-                       break;
-               }
-       }
-}
-
-/*
- * Sanitize the BIOS e820 map.
- *
- * Some e820 responses include overlapping entries. The following
- * replaces the original e820 map with a new one, removing overlaps.
- *
- */
-static int __init sanitize_e820_map(struct e820entry *biosmap, char *pnr_map)
-{
-       struct change_member {
-               struct e820entry *pbios; /* pointer to original bios entry */
-               unsigned long long addr; /* address for this change point */
-       };
-       static struct change_member change_point_list[2*E820MAX] __initdata;
-       static struct change_member *change_point[2*E820MAX] __initdata;
-       static struct e820entry *overlap_list[E820MAX] __initdata;
-       static struct e820entry new_bios[E820MAX] __initdata;
-       struct change_member *change_tmp;
-       unsigned long current_type, last_type;
-       unsigned long long last_addr;
-       int chgidx, still_changing;
-       int overlap_entries;
-       int new_bios_entry;
-       int old_nr, new_nr, chg_nr;
-       int i;
-
-       /*
-               Visually we're performing the following
-               (1,2,3,4 = memory types)...
-
-               Sample memory map (w/overlaps):
-                  ____22__________________
-                  ______________________4_
-                  ____1111________________
-                  _44_____________________
-                  11111111________________
-                  ____________________33__
-                  ___________44___________
-                  __________33333_________
-                  ______________22________
-                  ___________________2222_
-                  _________111111111______
-                  _____________________11_
-                  _________________4______
-
-               Sanitized equivalent (no overlap):
-                  1_______________________
-                  _44_____________________
-                  ___1____________________
-                  ____22__________________
-                  ______11________________
-                  _________1______________
-                  __________3_____________
-                  ___________44___________
-                  _____________33_________
-                  _______________2________
-                  ________________1_______
-                  _________________4______
-                  ___________________2____
-                  ____________________33__
-                  ______________________4_
-       */
-
-       /* if there's only one memory region, don't bother */
-       if (*pnr_map < 2)
-               return -1;
-
-       old_nr = *pnr_map;
-
-       /* bail out if we find any unreasonable addresses in bios map */
-       for (i = 0; i < old_nr; i++)
-               if (biosmap[i].addr + biosmap[i].size < biosmap[i].addr)
-                       return -1;
-
-       /* create pointers for initial change-point information (for sorting) */
-       for (i = 0; i < 2 * old_nr; i++)
-               change_point[i] = &change_point_list[i];
-
-       /* record all known change-points (starting and ending addresses),
-          omitting those that are for empty memory regions */
-       chgidx = 0;
-       for (i = 0; i < old_nr; i++)    {
-               if (biosmap[i].size != 0) {
-                       change_point[chgidx]->addr = biosmap[i].addr;
-                       change_point[chgidx++]->pbios = &biosmap[i];
-                       change_point[chgidx]->addr = biosmap[i].addr +
-                               biosmap[i].size;
-                       change_point[chgidx++]->pbios = &biosmap[i];
-               }
-       }
-       chg_nr = chgidx;
-
-       /* sort change-point list by memory addresses (low -> high) */
-       still_changing = 1;
-       while (still_changing)  {
-               still_changing = 0;
-               for (i = 1; i < chg_nr; i++)  {
-                       unsigned long long curaddr, lastaddr;
-                       unsigned long long curpbaddr, lastpbaddr;
-
-                       curaddr = change_point[i]->addr;
-                       lastaddr = change_point[i - 1]->addr;
-                       curpbaddr = change_point[i]->pbios->addr;
-                       lastpbaddr = change_point[i - 1]->pbios->addr;
-
-                       /*
-                        * swap entries, when:
-                        *
-                        * curaddr > lastaddr or
-                        * curaddr == lastaddr and curaddr == curpbaddr and
-                        * lastaddr != lastpbaddr
-                        */
-                       if (curaddr < lastaddr ||
-                           (curaddr == lastaddr && curaddr == curpbaddr &&
-                            lastaddr != lastpbaddr)) {
-                               change_tmp = change_point[i];
-                               change_point[i] = change_point[i-1];
-                               change_point[i-1] = change_tmp;
-                               still_changing = 1;
-                       }
-               }
-       }
-
-       /* create a new bios memory map, removing overlaps */
-       overlap_entries = 0;     /* number of entries in the overlap table */
-       new_bios_entry = 0;      /* index for creating new bios map entries */
-       last_type = 0;           /* start with undefined memory type */
-       last_addr = 0;           /* start with 0 as last starting address */
-
-       /* loop through change-points, determining affect on the new bios map */
-       for (chgidx = 0; chgidx < chg_nr; chgidx++) {
-               /* keep track of all overlapping bios entries */
-               if (change_point[chgidx]->addr ==
-                   change_point[chgidx]->pbios->addr) {
-                       /*
-                        * add map entry to overlap list (> 1 entry
-                        * implies an overlap)
-                        */
-                       overlap_list[overlap_entries++] =
-                               change_point[chgidx]->pbios;
-               } else {
-                       /*
-                        * remove entry from list (order independent,
-                        * so swap with last)
-                        */
-                       for (i = 0; i < overlap_entries; i++) {
-                               if (overlap_list[i] ==
-                                   change_point[chgidx]->pbios)
-                                       overlap_list[i] =
-                                               overlap_list[overlap_entries-1];
-                       }
-                       overlap_entries--;
-               }
-               /*
-                * if there are overlapping entries, decide which
-                * "type" to use (larger value takes precedence --
-                * 1=usable, 2,3,4,4+=unusable)
-                */
-               current_type = 0;
-               for (i = 0; i < overlap_entries; i++)
-                       if (overlap_list[i]->type > current_type)
-                               current_type = overlap_list[i]->type;
-               /*
-                * continue building up new bios map based on this
-                * information
-                */
-               if (current_type != last_type)  {
-                       if (last_type != 0)      {
-                               new_bios[new_bios_entry].size =
-                                       change_point[chgidx]->addr - last_addr;
-                               /*
-                                * move forward only if the new size
-                                * was non-zero
-                                */
-                               if (new_bios[new_bios_entry].size != 0)
-                                       /*
-                                        * no more space left for new
-                                        * bios entries ?
-                                        */
-                                       if (++new_bios_entry >= E820MAX)
-                                               break;
-                       }
-                       if (current_type != 0)  {
-                               new_bios[new_bios_entry].addr =
-                                       change_point[chgidx]->addr;
-                               new_bios[new_bios_entry].type = current_type;
-                               last_addr = change_point[chgidx]->addr;
-                       }
-                       last_type = current_type;
-               }
-       }
-       /* retain count for new bios entries */
-       new_nr = new_bios_entry;
-
-       /* copy new bios mapping into original location */
-       memcpy(biosmap, new_bios, new_nr * sizeof(struct e820entry));
-       *pnr_map = new_nr;
-
-       return 0;
-}
-
-/*
- * Copy the BIOS e820 map into a safe place.
- *
- * Sanity-check it while we're at it..
- *
- * If we're lucky and live on a modern system, the setup code
- * will have given us a memory map that we can use to properly
- * set up memory.  If we aren't, we'll fake a memory map.
- */
-static int __init copy_e820_map(struct e820entry *biosmap, int nr_map)
-{
-#ifndef CONFIG_XEN
-       /* Only one memory region (or negative)? Ignore it */
-       if (nr_map < 2)
-               return -1;
-#else
-       BUG_ON(nr_map < 1);
-#endif
-
-       do {
-               u64 start = biosmap->addr;
-               u64 size = biosmap->size;
-               u64 end = start + size;
-               u32 type = biosmap->type;
-
-               /* Overflow in 64 bits? Ignore the memory map. */
-               if (start > end)
-                       return -1;
-
-               add_memory_region(start, size, type);
-       } while (biosmap++, --nr_map);
-
-#ifdef CONFIG_XEN
-       if (is_initial_xendomain()) {
-               struct xen_memory_map memmap;
-
-               memmap.nr_entries = E820MAX;
-               set_xen_guest_handle(memmap.buffer, machine_e820.map);
-
-               if (HYPERVISOR_memory_op(XENMEM_machine_memory_map, &memmap))
-                       BUG();
-               machine_e820.nr_map = memmap.nr_entries;
-       } else
-               machine_e820 = e820;
-#endif
-
-       return 0;
-}
-
-static void early_panic(char *msg)
-{
-       early_printk(msg);
-       panic(msg);
-}
-
-/* We're not void only for x86 32-bit compat */
-char * __init machine_specific_memory_setup(void)
-{
-#ifndef CONFIG_XEN
-       char *who = "BIOS-e820";
-       /*
-        * Try to copy the BIOS-supplied E820-map.
-        *
-        * Otherwise fake a memory map; one section from 0k->640k,
-        * the next section from 1mb->appropriate_mem_k
-        */
-       sanitize_e820_map(boot_params.e820_map, &boot_params.e820_entries);
-       if (copy_e820_map(boot_params.e820_map, boot_params.e820_entries) < 0)
-               early_panic("Cannot find a valid memory map");
-#else  /* CONFIG_XEN */
-       char *who = "Xen";
-       int rc;
-       struct xen_memory_map memmap;
-       /*
-        * This is rather large for a stack variable but this early in
-        * the boot process we know we have plenty slack space.
-        */
-       struct e820entry map[E820MAX];
-
-       memmap.nr_entries = E820MAX;
-       set_xen_guest_handle(memmap.buffer, map);
-
-       rc = HYPERVISOR_memory_op(XENMEM_memory_map, &memmap);
-       if ( rc == -ENOSYS ) {
-               memmap.nr_entries = 1;
-               map[0].addr = 0ULL;
-               map[0].size = xen_start_info->nr_pages << PAGE_SHIFT;
-               /* 8MB slack (to balance backend allocations). */
-               map[0].size += 8 << 20;
-               map[0].type = E820_RAM;
-               rc = 0;
-       }
-       BUG_ON(rc);
-
-       sanitize_e820_map(map, (char *)&memmap.nr_entries);
-
-       if (copy_e820_map(map, (char)memmap.nr_entries) < 0)
-               early_panic("Cannot find a valid memory map");
-#endif
-       printk(KERN_INFO "BIOS-provided physical RAM map:\n");
-       e820_print_map(who);
-
-       /* In case someone cares... */
-       return who;
-}
-
-static int __init parse_memopt(char *p)
-{
-       int i;
-       unsigned long current_end;
-       unsigned long end;
-
-       if (!p)
-               return -EINVAL;
-       end_user_pfn = memparse(p, &p);
-       end_user_pfn >>= PAGE_SHIFT;
-
-       end = end_user_pfn<<PAGE_SHIFT;
-       i = e820.nr_map-1;
-       current_end = e820.map[i].addr + e820.map[i].size;
-
-       if (current_end < end) {
-               /*
-                 * The e820 map ends before our requested size so
-                 * extend the final entry to the requested address.
-                 */
-               if (e820.map[i].type == E820_RAM)
-                       e820.map[i].size = end - e820.map[i].addr;
-               else
-                       add_memory_region(current_end, end - current_end, E820_RAM);
-       }
-
-       return 0;
-}
-early_param("mem", parse_memopt);
-
-static int userdef __initdata;
-
-static int __init parse_memmap_opt(char *p)
-{
-       char *oldp;
-       unsigned long long start_at, mem_size;
-
-       if (!strcmp(p, "exactmap")) {
-#ifdef CONFIG_CRASH_DUMP
-               /*
-                * If we are doing a crash dump, we still need to know
-                * the real mem size before original memory map is
-                * reset.
-                */
-               e820_register_active_regions(0, 0, -1UL);
-               saved_max_pfn = e820_end_of_ram();
-               remove_all_active_ranges();
-#endif
-               max_pfn_mapped = 0;
-               e820.nr_map = 0;
-               userdef = 1;
-               return 0;
-       }
-
-       oldp = p;
-       mem_size = memparse(p, &p);
-       if (p == oldp)
-               return -EINVAL;
-
-       userdef = 1;
-       if (*p == '@') {
-               start_at = memparse(p+1, &p);
-               add_memory_region(start_at, mem_size, E820_RAM);
-       } else if (*p == '#') {
-               start_at = memparse(p+1, &p);
-               add_memory_region(start_at, mem_size, E820_ACPI);
-       } else if (*p == '$') {
-               start_at = memparse(p+1, &p);
-               add_memory_region(start_at, mem_size, E820_RESERVED);
-       } else {
-               end_user_pfn = (mem_size >> PAGE_SHIFT);
-       }
-       return *p == '\0' ? 0 : -EINVAL;
-}
-early_param("memmap", parse_memmap_opt);
-
-void __init finish_e820_parsing(void)
-{
-       if (userdef) {
-               char nr = e820.nr_map;
-
-               if (sanitize_e820_map(e820.map, &nr) < 0)
-                       early_panic("Invalid user supplied memory map");
-               e820.nr_map = nr;
-
-               printk(KERN_INFO "user-defined physical RAM map:\n");
-               e820_print_map("user");
-       }
-}
-
-#ifndef CONFIG_XEN
-void __init update_memory_range(u64 start, u64 size, unsigned old_type,
-                               unsigned new_type)
-{
-       int i;
-
-       BUG_ON(old_type == new_type);
-
-       for (i = 0; i < e820.nr_map; i++) {
-               struct e820entry *ei = &e820.map[i];
-               u64 final_start, final_end;
-               if (ei->type != old_type)
-                       continue;
-               /* totally covered? */
-               if (ei->addr >= start && ei->size <= size) {
-                       ei->type = new_type;
-                       continue;
-               }
-               /* partially covered */
-               final_start = max(start, ei->addr);
-               final_end = min(start + size, ei->addr + ei->size);
-               if (final_start >= final_end)
-                       continue;
-               add_memory_region(final_start, final_end - final_start,
-                                        new_type);
-       }
-}
-
-void __init update_e820(void)
-{
-       u8 nr_map;
-
-       nr_map = e820.nr_map;
-       if (sanitize_e820_map(e820.map, &nr_map))
-               return;
-       e820.nr_map = nr_map;
-       printk(KERN_INFO "modified physical RAM map:\n");
-       e820_print_map("modified");
-}
-#endif
-
-unsigned long pci_mem_start = 0xaeedbabe;
-EXPORT_SYMBOL(pci_mem_start);
-
-/*
- * Search for the biggest gap in the low 32 bits of the e820
- * memory space.  We pass this space to PCI to assign MMIO resources
- * for hotplug or unconfigured devices in.
- * Hopefully the BIOS let enough space left.
- */
-__init void e820_setup_gap(struct e820entry *e820, int nr_map)
-{
-       unsigned long gapstart, gapsize, round;
-       unsigned long last;
-       int i;
-       int found = 0;
-
-       last = 0x100000000ull;
-       gapstart = 0x10000000;
-       gapsize = 0x400000;
-       i = nr_map;
-       while (--i >= 0) {
-               unsigned long long start = e820[i].addr;
-               unsigned long long end = start + e820[i].size;
-
-               /*
-                * Since "last" is at most 4GB, we know we'll
-                * fit in 32 bits if this condition is true
-                */
-               if (last > end) {
-                       unsigned long gap = last - end;
-
-                       if (gap > gapsize) {
-                               gapsize = gap;
-                               gapstart = end;
-                               found = 1;
-                       }
-               }
-               if (start < last)
-                       last = start;
-       }
-
-       if (!found) {
-               gapstart = (end_pfn << PAGE_SHIFT) + 1024*1024;
-               printk(KERN_ERR "PCI: Warning: Cannot find a gap in the 32bit "
-                      "address range\n"
-                      KERN_ERR "PCI: Unassigned devices with 32bit resource "
-                      "registers may break!\n");
-       }
-
-       /*
-        * See how much we want to round up: start off with
-        * rounding to the next 1MB area.
-        */
-       round = 0x100000;
-       while ((gapsize >> 4) > round)
-               round += round;
-       /* Fun with two's complement */
-       pci_mem_start = (gapstart + round) & -round;
-
-       printk(KERN_INFO
-              "Allocating PCI resources starting at %lx (gap: %lx:%lx)\n",
-              pci_mem_start, gapstart, gapsize);
-}
-
-int __init arch_get_ram_range(int slot, u64 *addr, u64 *size)
-{
-       int i;
-
-       if (slot < 0 || slot >= e820.nr_map)
-               return -1;
-       for (i = slot; i < e820.nr_map; i++) {
-               if (e820.map[i].type != E820_RAM)
-                       continue;
-               break;
-       }
-       if (i == e820.nr_map || e820.map[i].addr > (max_pfn << PAGE_SHIFT))
-               return -1;
-       *addr = e820.map[i].addr;
-       *size = min_t(u64, e820.map[i].size + e820.map[i].addr,
-               max_pfn << PAGE_SHIFT) - *addr;
-       return i + 1;
-}
Index: head-2008-12-01/arch/x86/kernel/early_printk-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/early_printk-xen.c     2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/early_printk-xen.c  2008-12-01 11:49:07.000000000 +0100
@@ -225,7 +225,7 @@ static struct console simnow_console = {
 static struct console *early_console = &early_vga_console;
 static int early_console_initialized;
 
-void early_printk(const char *fmt, ...)
+asmlinkage void early_printk(const char *fmt, ...)
 {
        char buf[512];
        int n;
Index: head-2008-12-01/arch/x86/kernel/entry_32-xen.S
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/entry_32-xen.S 2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/entry_32-xen.S      2008-12-01 11:49:07.000000000 +0100
@@ -51,15 +51,26 @@
 #include <asm/percpu.h>
 #include <asm/dwarf2.h>
 #include <asm/processor-flags.h>
-#include "irq_vectors.h"
+#include <asm/ftrace.h>
+#include <asm/irq_vectors.h>
 #include <xen/interface/xen.h>
 
+/* Avoid __ASSEMBLER__'ifying <linux/audit.h> just for this.  */
+#include <linux/elf-em.h>
+#define AUDIT_ARCH_I386                (EM_386|__AUDIT_ARCH_LE)
+#define __AUDIT_ARCH_LE           0x40000000
+
+#ifndef CONFIG_AUDITSYSCALL
+#define sysenter_audit syscall_trace_entry
+#define sysexit_audit  syscall_exit_work
+#endif
+
 /*
  * We use macros for low-level operations which need to be overridden
  * for paravirtualization.  The following will never clobber any registers:
  *   INTERRUPT_RETURN (aka. "iret")
  *   GET_CR0_INTO_EAX (aka. "movl %cr0, %eax")
- *   ENABLE_INTERRUPTS_SYSCALL_RET (aka "sti; sysexit").
+ *   ENABLE_INTERRUPTS_SYSEXIT (aka "sti; sysexit").
  *
  * For DISABLE_INTERRUPTS/ENABLE_INTERRUPTS (aka "cli"/"sti"), you must
  * specify what registers can be overwritten (CLBR_NONE, CLBR_EAX/EDX/ECX/ANY).
@@ -277,11 +288,6 @@ END(resume_kernel)
 #endif
        CFI_ENDPROC
 
-       .macro test_tif ti_reg          # system call tracing in operation / emulation
-       /* Note, _TIF_SECCOMP is bit number 8, and so it needs testw and not testb */
-       testw $(_TIF_SYSCALL_EMU|_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT),TI_flags(\ti_reg)
-       .endm
-
 /* SYSENTER_RETURN points to after the "sysenter" instruction in
    the vsyscall page.  See vsyscall-sysentry.S, which defines the symbol.  */
 
@@ -338,8 +344,9 @@ sysenter_past_esp:
 .previous
 
        GET_THREAD_INFO(%ebp)
-       test_tif %ebp
-       jnz syscall_trace_entry
+       testw $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%ebp)
+       jnz sysenter_audit
+sysenter_do_call:
        cmpl $(nr_syscalls), %eax
        jae syscall_badsys
        call *sys_call_table(,%eax,4)
@@ -349,14 +356,54 @@ sysenter_past_esp:
        TRACE_IRQS_OFF
        movl TI_flags(%ebp), %ecx
        testw $_TIF_ALLWORK_MASK, %cx
-       jne syscall_exit_work
+       jne sysexit_audit
+sysenter_exit:
 /* if something modifies registers it must also disable sysexit */
        movl PT_EIP(%esp), %edx
        movl PT_OLDESP(%esp), %ecx
        xorl %ebp,%ebp
        TRACE_IRQS_ON
 1:     mov  PT_FS(%esp), %fs
-       ENABLE_INTERRUPTS_SYSCALL_RET
+       ENABLE_INTERRUPTS_SYSEXIT
+
+#ifdef CONFIG_AUDITSYSCALL
+sysenter_audit:
+       testw $(_TIF_WORK_SYSCALL_ENTRY & ~_TIF_SYSCALL_AUDIT),TI_flags(%ebp)
+       jnz syscall_trace_entry
+       addl $4,%esp
+       CFI_ADJUST_CFA_OFFSET -4
+       /* %esi already in 8(%esp)         6th arg: 4th syscall arg */
+       /* %edx already in 4(%esp)         5th arg: 3rd syscall arg */
+       /* %ecx already in 0(%esp)         4th arg: 2nd syscall arg */
+       movl %ebx,%ecx                  /* 3rd arg: 1st syscall arg */
+       movl %eax,%edx                  /* 2nd arg: syscall number */
+       movl $AUDIT_ARCH_I386,%eax      /* 1st arg: audit arch */
+       call audit_syscall_entry
+       pushl %ebx
+       CFI_ADJUST_CFA_OFFSET 4
+       movl PT_EAX(%esp),%eax          /* reload syscall number */
+       jmp sysenter_do_call
+
+sysexit_audit:
+       testw $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT), %cx
+       jne syscall_exit_work
+       TRACE_IRQS_ON
+       ENABLE_INTERRUPTS(CLBR_ANY)
+       movl %eax,%edx          /* second arg, syscall return value */
+       cmpl $0,%eax            /* is it < 0? */
+       setl %al                /* 1 if so, 0 if not */
+       movzbl %al,%eax         /* zero-extend that */
+       inc %eax /* first arg, 0->1(AUDITSC_SUCCESS), 1->2(AUDITSC_FAILURE) */
+       call audit_syscall_exit
+       DISABLE_INTERRUPTS(CLBR_ANY)
+       TRACE_IRQS_OFF
+       movl TI_flags(%ebp), %ecx
+       testw $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT), %cx
+       jne syscall_exit_work
+       movl PT_EAX(%esp),%eax  /* reload syscall return value */
+       jmp sysenter_exit
+#endif
+
        CFI_ENDPROC
 .pushsection .fixup,"ax"
 2:     movl $0,PT_FS(%esp)
@@ -400,7 +447,7 @@ ENTRY(system_call)
        CFI_ADJUST_CFA_OFFSET 4
        SAVE_ALL
        GET_THREAD_INFO(%ebp)
-       test_tif %ebp
+       testw $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%ebp)
        jnz syscall_trace_entry
        cmpl $(nr_syscalls), %eax
        jae syscall_badsys
@@ -413,10 +460,6 @@ syscall_exit:
                                        # setting need_resched or sigpending
                                        # between sampling and the iret
        TRACE_IRQS_OFF
-       testl $X86_EFLAGS_TF,PT_EFLAGS(%esp)    # If tracing set singlestep flag on exit
-       jz no_singlestep
-       orl $_TIF_SINGLESTEP,TI_flags(%ebp)
-no_singlestep:
        movl TI_flags(%ebp), %ecx
        testw $_TIF_ALLWORK_MASK, %cx   # current->work
        jne syscall_exit_work
@@ -588,12 +631,8 @@ END(work_pending)
 syscall_trace_entry:
        movl $-ENOSYS,PT_EAX(%esp)
        movl %esp, %eax
-       xorl %edx,%edx
-       call do_syscall_trace
-       cmpl $0, %eax
-       jne resume_userspace            # ret != 0 -> running under PTRACE_SYSEMU,
-                                       # so must skip actual syscall
-       movl PT_ORIG_EAX(%esp), %eax
+       call syscall_trace_enter
+       /* What it returned is what we'll actually use.  */
        cmpl $(nr_syscalls), %eax
        jnae syscall_call
        jmp syscall_exit
@@ -602,14 +641,13 @@ END(syscall_trace_entry)
        # perform syscall exit tracing
        ALIGN
 syscall_exit_work:
-       testb $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP), %cl
+       testb $_TIF_WORK_SYSCALL_EXIT, %cl
        jz work_pending
        TRACE_IRQS_ON
-       ENABLE_INTERRUPTS(CLBR_ANY)     # could let do_syscall_trace() call
+       ENABLE_INTERRUPTS(CLBR_ANY)     # could let syscall_trace_leave() call
                                        # schedule() instead
        movl %esp, %eax
-       movl $1, %edx
-       call do_syscall_trace
+       call syscall_trace_leave
        jmp resume_userspace
 END(syscall_exit_work)
        CFI_ENDPROC
@@ -1109,10 +1147,10 @@ ENTRY(native_iret)
 .previous
 END(native_iret)
 
-ENTRY(native_irq_enable_syscall_ret)
+ENTRY(native_irq_enable_sysexit)
        sti
        sysexit
-END(native_irq_enable_syscall_ret)
+END(native_irq_enable_sysexit)
 #endif
 
 KPROBE_ENTRY(int3)
@@ -1261,6 +1299,77 @@ ENTRY(kernel_thread_helper)
        CFI_ENDPROC
 ENDPROC(kernel_thread_helper)
 
+#ifdef CONFIG_FTRACE
+#ifdef CONFIG_DYNAMIC_FTRACE
+
+ENTRY(mcount)
+       pushl %eax
+       pushl %ecx
+       pushl %edx
+       movl 0xc(%esp), %eax
+       subl $MCOUNT_INSN_SIZE, %eax
+
+.globl mcount_call
+mcount_call:
+       call ftrace_stub
+
+       popl %edx
+       popl %ecx
+       popl %eax
+
+       ret
+END(mcount)
+
+ENTRY(ftrace_caller)
+       pushl %eax
+       pushl %ecx
+       pushl %edx
+       movl 0xc(%esp), %eax
+       movl 0x4(%ebp), %edx
+       subl $MCOUNT_INSN_SIZE, %eax
+
+.globl ftrace_call
+ftrace_call:
+       call ftrace_stub
+
+       popl %edx
+       popl %ecx
+       popl %eax
+
+.globl ftrace_stub
+ftrace_stub:
+       ret
+END(ftrace_caller)
+
+#else /* ! CONFIG_DYNAMIC_FTRACE */
+
+ENTRY(mcount)
+       cmpl $ftrace_stub, ftrace_trace_function
+       jnz trace
+.globl ftrace_stub
+ftrace_stub:
+       ret
+
+       /* taken from glibc */
+trace:
+       pushl %eax
+       pushl %ecx
+       pushl %edx
+       movl 0xc(%esp), %eax
+       movl 0x4(%ebp), %edx
+       subl $MCOUNT_INSN_SIZE, %eax
+
+       call *ftrace_trace_function
+
+       popl %edx
+       popl %ecx
+       popl %eax
+
+       jmp ftrace_stub
+END(mcount)
+#endif /* CONFIG_DYNAMIC_FTRACE */
+#endif /* CONFIG_FTRACE */
+
 #include <asm/alternative-asm.h>
 
        # pv syscall call handler stub
@@ -1286,7 +1395,7 @@ ENTRY(ia32pv_cstar_target)
 .previous
        SAVE_ALL
        GET_THREAD_INFO(%ebp)
-       test_tif %ebp
+       testw $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%ebp)
        jnz cstar_trace_entry
        cmpl $nr_syscalls,%eax
        jae cstar_badsys
@@ -1320,29 +1429,21 @@ cstar_trace_entry:
        btl %eax,cstar_special
        jc .Lcstar_trace_special
 1:     movl %esp,%eax
-       xorl %edx,%edx
        LOCK_PREFIX
        orl $_TIF_CSTAR,TI_flags(%ebp)
-       call do_syscall_trace
+       call syscall_trace_enter
        LOCK_PREFIX
        andl $~_TIF_CSTAR,TI_flags(%ebp)
-       testl %eax,%eax
-       jne .Lcstar_resume              # ret != 0 -> running under PTRACE_SYSEMU,
-                                       # so must skip actual syscall
-       movl PT_ORIG_EAX(%esp),%eax
+       /* What it returned is what we'll actually use.  */
        cmpl $nr_syscalls,%eax
        jb .Lcstar_call
        jmp .Lcstar_exit
 .Lcstar_trace_special:
        movl PT_ECX(%esp),%ecx
        movl %esp,%eax
-       xorl %edx,%edx
        movl %ecx,PT_EBP(%esp)          # put user EBP back in place
-       call do_syscall_trace
-       testl %eax,%eax
-       jne resume_userspace            # ret != 0 -> running under PTRACE_SYSEMU,
-                                       # so must skip actual syscall
-       movl PT_ORIG_EAX(%esp),%eax
+       call syscall_trace_enter
+       /* What it returned is what we'll actually use.  */
        cmpl $nr_syscalls,%eax
        jb syscall_call
        jmp syscall_exit
Index: head-2008-12-01/arch/x86/kernel/entry_64.S
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/entry_64.S     2008-12-03 15:48:43.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/entry_64.S  2008-12-01 11:49:07.000000000 +0100
@@ -1409,7 +1409,7 @@ ENTRY(arch_unwind_init_running)
 ENDPROC(arch_unwind_init_running)
 #endif
 
-#ifdef CONFIG_XEN
+#ifdef CONFIG_PARAVIRT_XEN
 ENTRY(xen_hypervisor_callback)
        zeroentry xen_do_hypervisor_callback
 END(xen_hypervisor_callback)
@@ -1507,4 +1507,4 @@ ENTRY(xen_failsafe_callback)
        CFI_ENDPROC
 END(xen_failsafe_callback)
 
-#endif /* CONFIG_XEN */
+#endif /* CONFIG_PARAVIRT_XEN */
Index: head-2008-12-01/arch/x86/kernel/entry_64-xen.S
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/entry_64-xen.S 2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/entry_64-xen.S      2008-12-01 11:49:07.000000000 +0100
@@ -53,19 +53,130 @@
 #include <asm/hw_irq.h>
 #include <asm/page.h>
 #include <asm/irqflags.h>
+#include <asm/ftrace.h>
 #include <asm/errno.h>
 #include <xen/interface/xen.h>
 #include <xen/interface/features.h>
 
+/* Avoid __ASSEMBLER__'ifying <linux/audit.h> just for this.  */
+#include <linux/elf-em.h>
+#define AUDIT_ARCH_X86_64      (EM_X86_64|__AUDIT_ARCH_64BIT|__AUDIT_ARCH_LE)
+#define __AUDIT_ARCH_64BIT 0x80000000
+#define __AUDIT_ARCH_LE           0x40000000
+
        .code64
 
+#ifdef CONFIG_FTRACE
+#ifdef CONFIG_DYNAMIC_FTRACE
+ENTRY(mcount)
+
+       subq $0x38, %rsp
+       movq %rax, (%rsp)
+       movq %rcx, 8(%rsp)
+       movq %rdx, 16(%rsp)
+       movq %rsi, 24(%rsp)
+       movq %rdi, 32(%rsp)
+       movq %r8, 40(%rsp)
+       movq %r9, 48(%rsp)
+
+       movq 0x38(%rsp), %rdi
+       subq $MCOUNT_INSN_SIZE, %rdi
+
+.globl mcount_call
+mcount_call:
+       call ftrace_stub
+
+       movq 48(%rsp), %r9
+       movq 40(%rsp), %r8
+       movq 32(%rsp), %rdi
+       movq 24(%rsp), %rsi
+       movq 16(%rsp), %rdx
+       movq 8(%rsp), %rcx
+       movq (%rsp), %rax
+       addq $0x38, %rsp
+
+       retq
+END(mcount)
+
+ENTRY(ftrace_caller)
+
+       /* taken from glibc */
+       subq $0x38, %rsp
+       movq %rax, (%rsp)
+       movq %rcx, 8(%rsp)
+       movq %rdx, 16(%rsp)
+       movq %rsi, 24(%rsp)
+       movq %rdi, 32(%rsp)
+       movq %r8, 40(%rsp)
+       movq %r9, 48(%rsp)
+
+       movq 0x38(%rsp), %rdi
+       movq 8(%rbp), %rsi
+       subq $MCOUNT_INSN_SIZE, %rdi
+
+.globl ftrace_call
+ftrace_call:
+       call ftrace_stub
+
+       movq 48(%rsp), %r9
+       movq 40(%rsp), %r8
+       movq 32(%rsp), %rdi
+       movq 24(%rsp), %rsi
+       movq 16(%rsp), %rdx
+       movq 8(%rsp), %rcx
+       movq (%rsp), %rax
+       addq $0x38, %rsp
+
+.globl ftrace_stub
+ftrace_stub:
+       retq
+END(ftrace_caller)
+
+#else /* ! CONFIG_DYNAMIC_FTRACE */
+ENTRY(mcount)
+       cmpq $ftrace_stub, ftrace_trace_function
+       jnz trace
+.globl ftrace_stub
+ftrace_stub:
+       retq
+
+trace:
+       /* taken from glibc */
+       subq $0x38, %rsp
+       movq %rax, (%rsp)
+       movq %rcx, 8(%rsp)
+       movq %rdx, 16(%rsp)
+       movq %rsi, 24(%rsp)
+       movq %rdi, 32(%rsp)
+       movq %r8, 40(%rsp)
+       movq %r9, 48(%rsp)
+
+       movq 0x38(%rsp), %rdi
+       movq 8(%rbp), %rsi
+       subq $MCOUNT_INSN_SIZE, %rdi
+
+       call   *ftrace_trace_function
+
+       movq 48(%rsp), %r9
+       movq 40(%rsp), %r8
+       movq 32(%rsp), %rdi
+       movq 24(%rsp), %rsi
+       movq 16(%rsp), %rdx
+       movq 8(%rsp), %rcx
+       movq (%rsp), %rax
+       addq $0x38, %rsp
+
+       jmp ftrace_stub
+END(mcount)
+#endif /* CONFIG_DYNAMIC_FTRACE */
+#endif /* CONFIG_FTRACE */
+
 #ifndef CONFIG_PREEMPT
 #define retint_kernel retint_restore_args
 #endif 
 
 #ifdef CONFIG_PARAVIRT
-ENTRY(native_irq_enable_syscall_ret)
-       movq    %gs:pda_oldrsp,%rsp
+ENTRY(native_usergs_sysret64)
        swapgs
        sysretq
 #endif /* CONFIG_PARAVIRT */
@@ -102,7 +213,7 @@ NMI_MASK = 0x80000000
        .macro FAKE_STACK_FRAME child_rip
        /* push in order ss, rsp, eflags, cs, rip */
        xorl %eax, %eax
-       pushq %rax /* ss */
+       pushq $__KERNEL_DS /* ss */
        CFI_ADJUST_CFA_OFFSET   8
        /*CFI_REL_OFFSET        ss,0*/
        pushq %rax /* rsp */
@@ -197,13 +308,13 @@ ENTRY(ret_from_fork)
        CFI_ADJUST_CFA_OFFSET -4
        call schedule_tail
        GET_THREAD_INFO(%rcx)
-       testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT),threadinfo_flags(%rcx)
+       testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT),TI_flags(%rcx)
        jnz rff_trace
 rff_action:    
        RESTORE_REST
        testl $3,CS-ARGOFFSET(%rsp)     # from kernel_thread?
        je   int_ret_from_sys_call
-       testl $_TIF_IA32,threadinfo_flags(%rcx)
+       testl $_TIF_IA32,TI_flags(%rcx)
        jnz  int_ret_from_sys_call
        RESTORE_TOP_OF_STACK %rdi,ARGOFFSET
        jmp ret_from_sys_call
@@ -265,8 +376,9 @@ ENTRY(system_call)
        SAVE_ARGS -8,0
        movq  %rax,ORIG_RAX-ARGOFFSET(%rsp) 
        GET_THREAD_INFO(%rcx)
-       testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SECCOMP),threadinfo_flags(%rcx)
+       testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%rcx)
        jnz tracesys
+system_call_fastpath:
        cmpq $__NR_syscall_max,%rax
        ja badsys
        movq %r10,%rcx
@@ -284,7 +396,7 @@ sysret_check:               
        GET_THREAD_INFO(%rcx)
        DISABLE_INTERRUPTS(CLBR_NONE)
        TRACE_IRQS_OFF
-       movl threadinfo_flags(%rcx),%edx
+       movl TI_flags(%rcx),%edx
        andl %edi,%edx
        jnz  sysret_careful 
        CFI_REMEMBER_STATE
@@ -315,16 +427,16 @@ sysret_careful:
 sysret_signal:
        TRACE_IRQS_ON
        ENABLE_INTERRUPTS(CLBR_NONE)
-       testl $_TIF_DO_NOTIFY_MASK,%edx
-       jz    1f
-
-       /* Really a signal */
+#ifdef CONFIG_AUDITSYSCALL
+       bt $TIF_SYSCALL_AUDIT,%edx
+       jc sysret_audit
+#endif
        /* edx: work flags (arg3) */
        leaq do_notify_resume(%rip),%rax
        leaq -ARGOFFSET(%rsp),%rdi # &pt_regs -> arg1
        xorl %esi,%esi # oldset -> arg2
        call ptregscall_common
-1:     movl $_TIF_NEED_RESCHED,%edi
+       movl $_TIF_WORK_MASK,%edi
        /* Use IRET because user could have changed frame. This
           works because ptregscall_common has called FIXUP_TOP_OF_STACK. */
        DISABLE_INTERRUPTS(CLBR_NONE)
@@ -335,14 +447,56 @@ badsys:
        movq $-ENOSYS,RAX-ARGOFFSET(%rsp)
        jmp ret_from_sys_call
 
+#ifdef CONFIG_AUDITSYSCALL
+       /*
+        * Fast path for syscall audit without full syscall trace.
+        * We just call audit_syscall_entry() directly, and then
+        * jump back to the normal fast path.
+        */
+auditsys:
+       movq %r10,%r9                   /* 6th arg: 4th syscall arg */
+       movq %rdx,%r8                   /* 5th arg: 3rd syscall arg */
+       movq %rsi,%rcx                  /* 4th arg: 2nd syscall arg */
+       movq %rdi,%rdx                  /* 3rd arg: 1st syscall arg */
+       movq %rax,%rsi                  /* 2nd arg: syscall number */
+       movl $AUDIT_ARCH_X86_64,%edi    /* 1st arg: audit arch */
+       call audit_syscall_entry
+       LOAD_ARGS 0             /* reload call-clobbered registers */
+       jmp system_call_fastpath
+
+       /*
+        * Return fast path for syscall audit.  Call audit_syscall_exit()
+        * directly and then jump back to the fast path with TIF_SYSCALL_AUDIT
+        * masked off.
+        */
+sysret_audit:
+       movq %rax,%rsi          /* second arg, syscall return value */
+       cmpq $0,%rax            /* is it < 0? */
+       setl %al                /* 1 if so, 0 if not */
+       movzbl %al,%edi         /* zero-extend that into %edi */
+       inc %edi /* first arg, 0->1(AUDITSC_SUCCESS), 1->2(AUDITSC_FAILURE) */
+       call audit_syscall_exit
+       movl $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT),%edi
+       jmp sysret_check
+#endif /* CONFIG_AUDITSYSCALL */
+
        /* Do syscall tracing */
 tracesys:                       
+#ifdef CONFIG_AUDITSYSCALL
+       testl $(_TIF_WORK_SYSCALL_ENTRY & ~_TIF_SYSCALL_AUDIT),TI_flags(%rcx)
+       jz auditsys
+#endif
        SAVE_REST
        movq $-ENOSYS,RAX(%rsp) /* ptrace can change this for a bad syscall */
        FIXUP_TOP_OF_STACK %rdi
        movq %rsp,%rdi
        call syscall_trace_enter
-       LOAD_ARGS ARGOFFSET  /* reload args from stack in case ptrace changed it */
+       /*
+        * Reload arg registers from stack in case ptrace changed them.
+        * We don't reload %rax because syscall_trace_enter() returned
+        * the value it wants us to use in the table lookup.
+        */
+       LOAD_ARGS ARGOFFSET, 1
        RESTORE_REST
        cmpq $__NR_syscall_max,%rax
        ja   int_ret_from_sys_call      /* RAX(%rsp) set to -ENOSYS above */
@@ -356,6 +510,7 @@ tracesys:                    
  * Has correct top of stack, but partial stack frame.
  */
        .globl int_ret_from_sys_call
+       .globl int_with_check
 int_ret_from_sys_call:
        DISABLE_INTERRUPTS(CLBR_NONE)
        TRACE_IRQS_OFF
@@ -370,10 +525,10 @@ int_ret_from_sys_call:
 int_with_check:
        LOCKDEP_SYS_EXIT_IRQ
        GET_THREAD_INFO(%rcx)
-       movl threadinfo_flags(%rcx),%edx
+       movl TI_flags(%rcx),%edx
        andl %edi,%edx
        jnz   int_careful
-       andl    $~TS_COMPAT,threadinfo_status(%rcx)
+       andl    $~TS_COMPAT,TI_status(%rcx)
        jmp   retint_restore_args
 
        /* Either reschedule or signal or syscall exit tracking needed. */
@@ -399,7 +554,7 @@ int_very_careful:
        ENABLE_INTERRUPTS(CLBR_NONE)
        SAVE_REST
        /* Check for syscall exit trace */      
-       testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP),%edx
+       testl $_TIF_WORK_SYSCALL_EXIT,%edx
        jz int_signal
        pushq %rdi
        CFI_ADJUST_CFA_OFFSET 8
@@ -407,7 +562,7 @@ int_very_careful:
        call syscall_trace_leave
        popq %rdi
        CFI_ADJUST_CFA_OFFSET -8
-       andl $~(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP),%edi
+       andl $~(_TIF_WORK_SYSCALL_EXIT|_TIF_SYSCALL_EMU),%edi
        jmp int_restore_rest
        
 int_signal:
@@ -416,7 +571,7 @@ int_signal:
        movq %rsp,%rdi          # &ptregs -> arg1
        xorl %esi,%esi          # oldset -> arg2
        call do_notify_resume
-1:     movl $_TIF_NEED_RESCHED,%edi    
+1:     movl $_TIF_WORK_MASK,%edi
 int_restore_rest:
        RESTORE_REST
        DISABLE_INTERRUPTS(CLBR_NONE)
@@ -443,7 +598,6 @@ END(\label)
        PTREGSCALL stub_clone, sys_clone, %r8
        PTREGSCALL stub_fork, sys_fork, %rdi
        PTREGSCALL stub_vfork, sys_vfork, %rdi
-       PTREGSCALL stub_rt_sigsuspend, sys_rt_sigsuspend, %rdx
        PTREGSCALL stub_sigaltstack, sys_sigaltstack, %rdx
        PTREGSCALL stub_iopl, sys_iopl, %rsi
 
@@ -517,10 +671,12 @@ END(stub_rt_sigreturn)
  *
  */ 
 
-retint_check:
+retint_with_reschedule:
        CFI_DEFAULT_STACK adj=1
+       movl $_TIF_WORK_MASK,%edi
+retint_check:
        LOCKDEP_SYS_EXIT_IRQ
-       movl threadinfo_flags(%rcx),%edx
+       movl TI_flags(%rcx),%edx
        andl %edi,%edx
        CFI_REMEMBER_STATE
        jnz  retint_careful
@@ -565,17 +721,16 @@ retint_signal:
        RESTORE_REST
        DISABLE_INTERRUPTS(CLBR_NONE)
        TRACE_IRQS_OFF
-       movl $_TIF_NEED_RESCHED,%edi
        GET_THREAD_INFO(%rcx)
-       jmp retint_check
+       jmp retint_with_reschedule
 
 #ifdef CONFIG_PREEMPT
        /* Returning to kernel space. Check if we need preemption */
        /* rcx:  threadinfo. interrupts off. */
 ENTRY(retint_kernel)
-       cmpl $0,threadinfo_preempt_count(%rcx)
+       cmpl $0,TI_preempt_count(%rcx)
        jnz  retint_restore_args
-       bt  $TIF_NEED_RESCHED,threadinfo_flags(%rcx)
+       bt  $TIF_NEED_RESCHED,TI_flags(%rcx)
        jnc  retint_restore_args
        bt   $9,EFLAGS-ARGOFFSET(%rsp)  /* interrupts off? */
        jnc  retint_restore_args
@@ -630,6 +785,9 @@ END(invalidate_interrupt\num)
 ENTRY(call_function_interrupt)
        apicinterrupt CALL_FUNCTION_VECTOR,smp_call_function_interrupt
 END(call_function_interrupt)
+ENTRY(call_function_single_interrupt)
+       apicinterrupt CALL_FUNCTION_SINGLE_VECTOR,smp_call_function_single_interrupt
+END(call_function_single_interrupt)
 ENTRY(irq_move_cleanup_interrupt)
        apicinterrupt IRQ_MOVE_CLEANUP_VECTOR,smp_irq_move_cleanup_interrupt
 END(irq_move_cleanup_interrupt)
@@ -639,6 +797,10 @@ ENTRY(apic_timer_interrupt)
        apicinterrupt LOCAL_TIMER_VECTOR,smp_apic_timer_interrupt
 END(apic_timer_interrupt)
 
+ENTRY(uv_bau_message_intr1)
+       apicinterrupt 220,uv_bau_message_interrupt
+END(uv_bau_message_intr1)
+
 ENTRY(error_interrupt)
        apicinterrupt ERROR_APIC_VECTOR,smp_error_interrupt
 END(error_interrupt)
@@ -752,7 +914,7 @@ paranoid_restore\trace:
        jmp irq_return
 paranoid_userspace\trace:
        GET_THREAD_INFO(%rcx)
-       movl threadinfo_flags(%rcx),%ebx
+       movl TI_flags(%rcx),%ebx
        andl $_TIF_WORK_MASK,%ebx
        jz paranoid_swapgs\trace
        movq %rsp,%rdi                  /* &pt_regs */
@@ -849,7 +1011,7 @@ error_exit:
        testb $3,CS-ARGOFFSET(%rsp)
        jz retint_kernel
        LOCKDEP_SYS_EXIT_IRQ
-       movl  threadinfo_flags(%rcx),%edx
+       movl  TI_flags(%rcx),%edx
        movl  $_TIF_WORK_MASK,%edi      
        andl  %edi,%edx
        jnz   retint_careful
@@ -871,11 +1033,11 @@ error_kernelspace:
           iret run with kernel gs again, so don't set the user space flag.
           B stepping K8s sometimes report an truncated RIP for IRET 
           exceptions returning to compat mode. Check for these here too. */
-       leaq irq_return(%rip),%rbp
-       cmpq %rbp,RIP(%rsp) 
+       leaq irq_return(%rip),%rcx
+       cmpq %rcx,RIP(%rsp)
        je   error_swapgs
-       movl %ebp,%ebp  /* zero extend */
-       cmpq %rbp,RIP(%rsp) 
+       movl %ecx,%ecx  /* zero extend */
+       cmpq %rcx,RIP(%rsp)
        je   error_swapgs
        cmpq $gs_change,RIP(%rsp)
         je   error_swapgs
@@ -1121,6 +1283,7 @@ END(device_not_available)
        /* runs on exception stack */
 KPROBE_ENTRY(debug)
 /*     INTR_FRAME
+       PARAVIRT_ADJUST_EXCEPTION_FRAME
        pushq $0
        CFI_ADJUST_CFA_OFFSET 8 */
        zeroentry do_debug
@@ -1148,6 +1311,7 @@ END(do_nmi_callback)
 
 KPROBE_ENTRY(int3)
 /*     INTR_FRAME
+       PARAVIRT_ADJUST_EXCEPTION_FRAME
        pushq $0
        CFI_ADJUST_CFA_OFFSET 8 */
        zeroentry do_int3
@@ -1171,14 +1335,11 @@ ENTRY(coprocessor_segment_overrun)
        zeroentry do_coprocessor_segment_overrun
 END(coprocessor_segment_overrun)
 
-ENTRY(reserved)
-       zeroentry do_reserved
-END(reserved)
-
 #if 0
        /* runs on exception stack */
 ENTRY(double_fault)
        XCPT_FRAME
+       PARAVIRT_ADJUST_EXCEPTION_FRAME
        paranoidentry do_double_fault
        jmp paranoid_exit1
        CFI_ENDPROC
@@ -1196,6 +1357,7 @@ END(segment_not_present)
        /* runs on exception stack */
 ENTRY(stack_segment)
 /*     XCPT_FRAME
+       PARAVIRT_ADJUST_EXCEPTION_FRAME
        paranoidentry do_stack_segment */
        errorentry do_stack_segment
 /*     jmp paranoid_exit1
@@ -1222,6 +1384,7 @@ END(spurious_interrupt_bug)
        /* runs on exception stack */
 ENTRY(machine_check)
        INTR_FRAME
+       PARAVIRT_ADJUST_EXCEPTION_FRAME
        pushq $0
        CFI_ADJUST_CFA_OFFSET 8 
        paranoidentry do_machine_check
Index: head-2008-12-01/arch/x86/kernel/genapic_64-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/genapic_64-xen.c       2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/genapic_64-xen.c    2008-12-01 11:49:07.000000000 +0100
@@ -58,7 +58,7 @@ void __init setup_apic_routing(void)
        else
 #endif
 
-       if (num_possible_cpus() <= 8)
+       if (max_physical_apicid < 8)
                genapic = &apic_flat;
        else
                genapic = &apic_physflat;
@@ -121,4 +121,5 @@ int is_uv_system(void)
 {
        return uv_system_type != UV_NONE;
 }
+EXPORT_SYMBOL_GPL(is_uv_system);
 #endif
Index: head-2008-12-01/arch/x86/kernel/genapic_xen_64.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/genapic_xen_64.c       2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/genapic_xen_64.c    2008-12-01 11:49:07.000000000 +0100
@@ -43,7 +43,7 @@ void xen_send_IPI_shortcut(unsigned int 
                __send_IPI_one(smp_processor_id(), vector);
                break;
        case APIC_DEST_ALLBUT:
-               for (cpu = 0; cpu < NR_CPUS; ++cpu) {
+               for_each_possible_cpu(cpu) {
                        if (cpu == smp_processor_id())
                                continue;
                        if (cpu_isset(cpu, cpu_online_map)) {
@@ -52,7 +52,7 @@ void xen_send_IPI_shortcut(unsigned int 
                }
                break;
        case APIC_DEST_ALLINC:
-               for (cpu = 0; cpu < NR_CPUS; ++cpu) {
+               for_each_possible_cpu(cpu) {
                        if (cpu_isset(cpu, cpu_online_map)) {
                                __send_IPI_one(cpu, vector);
                        }
@@ -81,8 +81,6 @@ static cpumask_t xen_vector_allocation_d
  */
 static void xen_init_apic_ldr(void)
 {
-       Dprintk("%s\n", __FUNCTION__);
-       return;
 }
 
 static void xen_send_IPI_allbutself(int vector)
@@ -92,14 +90,12 @@ static void xen_send_IPI_allbutself(int 
         * we get an APIC send error if we try to broadcast.
         * thus we have to avoid sending IPIs in this case.
         */
-       Dprintk("%s\n", __FUNCTION__);
        if (num_online_cpus() > 1)
                xen_send_IPI_shortcut(APIC_DEST_ALLBUT, vector, APIC_DEST_LOGICAL);
 }
 
 static void xen_send_IPI_all(int vector)
 {
-       Dprintk("%s\n", __FUNCTION__);
        xen_send_IPI_shortcut(APIC_DEST_ALLINC, vector, APIC_DEST_LOGICAL);
 }
 
@@ -109,11 +105,10 @@ static void xen_send_IPI_mask(cpumask_t 
        unsigned int cpu;
        unsigned long flags;
 
-       Dprintk("%s\n", __FUNCTION__);
        local_irq_save(flags);
        WARN_ON(mask & ~cpus_addr(cpu_online_map)[0]);
 
-       for (cpu = 0; cpu < NR_CPUS; ++cpu) {
+       for_each_possible_cpu(cpu) {
                if (cpu_isset(cpu, cpumask)) {
                        __send_IPI_one(cpu, vector);
                }
@@ -125,14 +120,12 @@ static void xen_send_IPI_mask(cpumask_t 
 static int xen_apic_id_registered(void)
 {
        /* better be set */
-       Dprintk("%s\n", __FUNCTION__);
        return physid_isset(smp_processor_id(), phys_cpu_present_map);
 }
 #endif
 
 static unsigned int xen_cpu_mask_to_apicid(cpumask_t cpumask)
 {
-       Dprintk("%s\n", __FUNCTION__);
        return cpus_addr(cpumask)[0] & APIC_ALL_CPUS;
 }
 
@@ -140,7 +133,6 @@ static unsigned int phys_pkg_id(int inde
 {
        u32 ebx;
 
-       Dprintk("%s\n", __FUNCTION__);
        ebx = cpuid_ebx(1);
        return ((ebx >> 24) & 0xFF) >> index_msb;
 }
Index: head-2008-12-01/arch/x86/kernel/head-xen.c
===================================================================
--- /dev/null   1970-01-01 00:00:00.000000000 +0000
+++ head-2008-12-01/arch/x86/kernel/head-xen.c  2008-12-01 11:49:07.000000000 +0100
@@ -0,0 +1,57 @@
+#include <linux/kernel.h>
+#include <linux/init.h>
+
+#include <asm/setup.h>
+#include <asm/bios_ebda.h>
+
+#define BIOS_LOWMEM_KILOBYTES 0x413
+
+/*
+ * The BIOS places the EBDA/XBDA at the top of conventional
+ * memory, and usually decreases the reported amount of
+ * conventional memory (int 0x12) too. This also contains a
+ * workaround for Dell systems that neglect to reserve EBDA.
+ * The same workaround also avoids a problem with the AMD768MPX
+ * chipset: reserve a page before VGA to prevent PCI prefetch
+ * into it (errata #56). Usually the page is reserved anyways,
+ * unless you have no PS/2 mouse plugged in.
+ */
+void __init reserve_ebda_region(void)
+{
+#ifndef CONFIG_XEN
+       unsigned int lowmem, ebda_addr;
+
+       /* To determine the position of the EBDA and the */
+       /* end of conventional memory, we need to look at */
+       /* the BIOS data area. In a paravirtual environment */
+       /* that area is absent. We'll just have to assume */
+       /* that the paravirt case can handle memory setup */
+       /* correctly, without our help. */
+       if (paravirt_enabled())
+               return;
+
+       /* end of low (conventional) memory */
+       lowmem = *(unsigned short *)__va(BIOS_LOWMEM_KILOBYTES);
+       lowmem <<= 10;
+
+       /* start of EBDA area */
+       ebda_addr = get_bios_ebda();
+
+       /* Fixup: bios puts an EBDA in the top 64K segment */
+       /* of conventional memory, but does not adjust lowmem. */
+       if ((lowmem - ebda_addr) <= 0x10000)
+               lowmem = ebda_addr;
+
+       /* Fixup: bios does not report an EBDA at all. */
+       /* Some old Dells seem to need 4k anyhow (bugzilla 2990) */
+       if ((ebda_addr == 0) && (lowmem >= 0x9f000))
+               lowmem = 0x9f000;
+
+       /* Paranoia: should never happen, but... */
+       if ((lowmem == 0) || (lowmem >= 0x100000))
+               lowmem = 0x9f000;
+
+       /* reserve all memory between lowmem and the 1MB mark */
+       reserve_early_overlap_ok(lowmem, 0x100000, "BIOS reserved");
+#endif
+}
Index: head-2008-12-01/arch/x86/kernel/head32-xen.c
===================================================================
--- /dev/null   1970-01-01 00:00:00.000000000 +0000
+++ head-2008-12-01/arch/x86/kernel/head32-xen.c        2008-12-01 11:49:07.000000000 +0100
@@ -0,0 +1,57 @@
+/*
+ *  linux/arch/i386/kernel/head32.c -- prepare to run common code
+ *
+ *  Copyright (C) 2000 Andrea Arcangeli <andrea@suse.de> SuSE
+ *  Copyright (C) 2007 Eric Biederman <ebiederm@xmission.com>
+ */
+
+#include <linux/init.h>
+#include <linux/start_kernel.h>
+
+#include <asm/setup.h>
+#include <asm/sections.h>
+#include <asm/e820.h>
+#include <asm/bios_ebda.h>
+
+void __init i386_start_kernel(void)
+{
+       reserve_early(__pa_symbol(&_text), __pa_symbol(&_end), "TEXT DATA BSS");
+
+#ifndef CONFIG_XEN
+#ifdef CONFIG_BLK_DEV_INITRD
+       /* Reserve INITRD */
+       if (boot_params.hdr.type_of_loader && boot_params.hdr.ramdisk_image) {
+               u64 ramdisk_image = boot_params.hdr.ramdisk_image;
+               u64 ramdisk_size  = boot_params.hdr.ramdisk_size;
+               u64 ramdisk_end   = ramdisk_image + ramdisk_size;
+               reserve_early(ramdisk_image, ramdisk_end, "RAMDISK");
+       }
+#endif
+       reserve_early(init_pg_tables_start, init_pg_tables_end,
+                       "INIT_PG_TABLE");
+#else
+       reserve_early(ALIGN(__pa_symbol(&_end), PAGE_SIZE),
+                     __pa(xen_start_info->pt_base)
+                     + (xen_start_info->nr_pt_frames << PAGE_SHIFT),
+                     "Xen provided");
+
+       {
+               int max_cmdline;
+
+               if ((max_cmdline = MAX_GUEST_CMDLINE) > COMMAND_LINE_SIZE)
+                       max_cmdline = COMMAND_LINE_SIZE;
+               memcpy(boot_command_line, xen_start_info->cmd_line, max_cmdline);
+               boot_command_line[max_cmdline-1] = '\0';
+       }
+#endif
+
+       reserve_ebda_region();
+
+       /*
+        * At this point everything still needed from the boot loader
+        * or BIOS or kernel text should be early reserved or marked not
+        * RAM in e820. All other memory is free game.
+        */
+
+       start_kernel();
+}
Index: head-2008-12-01/arch/x86/kernel/head64-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/head64-xen.c   2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/head64-xen.c        2008-12-01 11:49:07.000000000 +0100
@@ -32,7 +32,26 @@
 #include <asm/e820.h>
 #include <asm/bios_ebda.h>
 
-unsigned long start_pfn;
+/* boot cpu pda */
+static struct x8664_pda _boot_cpu_pda __read_mostly;
+
+#ifdef CONFIG_SMP
+/*
+ * We install an empty cpu_pda pointer table to indicate to early users
+ * (numa_set_node) that the cpu_pda pointer table for cpus other than
+ * the boot cpu is not yet setup.
+ */
+static struct x8664_pda *__cpu_pda[NR_CPUS] __initdata;
+#else
+static struct x8664_pda *__cpu_pda[NR_CPUS] __read_mostly;
+#endif
+
+void __init x86_64_init_pda(void)
+{
+       _cpu_pda = __cpu_pda;
+       cpu_pda(0) = &_boot_cpu_pda;
+       pda_init(0);
+}
 
 #ifndef CONFIG_XEN
 static void __init zap_identity_mappings(void)
@@ -77,83 +96,10 @@ EXPORT_SYMBOL(machine_to_phys_mapping);
 unsigned int machine_to_phys_order;
 EXPORT_SYMBOL(machine_to_phys_order);
 
-#define BIOS_LOWMEM_KILOBYTES 0x413
-
-/*
- * The BIOS places the EBDA/XBDA at the top of conventional
- * memory, and usually decreases the reported amount of
- * conventional memory (int 0x12) too. This also contains a
- * workaround for Dell systems that neglect to reserve EBDA.
- * The same workaround also avoids a problem with the AMD768MPX
- * chipset: reserve a page before VGA to prevent PCI prefetch
- * into it (errata #56). Usually the page is reserved anyways,
- * unless you have no PS/2 mouse plugged in.
- */
-static void __init reserve_ebda_region(void)
-{
-#ifndef CONFIG_XEN
-       unsigned int lowmem, ebda_addr;
-
-       /* To determine the position of the EBDA and the */
-       /* end of conventional memory, we need to look at */
-       /* the BIOS data area. In a paravirtual environment */
-       /* that area is absent. We'll just have to assume */
-       /* that the paravirt case can handle memory setup */
-       /* correctly, without our help. */
-       if (paravirt_enabled())
-               return;
-
-       /* end of low (conventional) memory */
-       lowmem = *(unsigned short *)__va(BIOS_LOWMEM_KILOBYTES);
-       lowmem <<= 10;
-
-       /* start of EBDA area */
-       ebda_addr = get_bios_ebda();
-
-       /* Fixup: bios puts an EBDA in the top 64K segment */
-       /* of conventional memory, but does not adjust lowmem. */
-       if ((lowmem - ebda_addr) <= 0x10000)
-               lowmem = ebda_addr;
-
-       /* Fixup: bios does not report an EBDA at all. */
-       /* Some old Dells seem to need 4k anyhow (bugzilla 2990) */
-       if ((ebda_addr == 0) && (lowmem >= 0x9f000))
-               lowmem = 0x9f000;
-
-       /* Paranoia: should never happen, but... */
-       if ((lowmem == 0) || (lowmem >= 0x100000))
-               lowmem = 0x9f000;
-
-       /* reserve all memory between lowmem and the 1MB mark */
-       reserve_early(lowmem, 0x100000, "BIOS reserved");
-#endif
-}
-
-static void __init reserve_setup_data(void)
-{
-#ifndef CONFIG_XEN
-       struct setup_data *data;
-       unsigned long pa_data;
-       char buf[32];
-
-       if (boot_params.hdr.version < 0x0209)
-               return;
-       pa_data = boot_params.hdr.setup_data;
-       while (pa_data) {
-               data = early_ioremap(pa_data, sizeof(*data));
-               sprintf(buf, "setup data %x", data->type);
-               reserve_early(pa_data, pa_data+sizeof(*data)+data->len, buf);
-               pa_data = data->next;
-               early_iounmap(data, sizeof(*data));
-       }
-#endif
-}
-
 void __init x86_64_start_kernel(char * real_mode_data)
 {
        struct xen_machphys_mapping mapping;
        unsigned long machine_to_phys_nr_ents;
-       int i;
 
        /*
         * Build-time sanity checks on the kernel image and module
@@ -167,6 +113,7 @@ void __init x86_64_start_kernel(char * r
        BUILD_BUG_ON(!(MODULES_VADDR > __START_KERNEL));
        BUILD_BUG_ON(!(((MODULES_END - 1) & PGDIR_MASK) ==
                                (__START_KERNEL & PGDIR_MASK)));
+       BUILD_BUG_ON(__fix_to_virt(__end_of_fixed_addresses) <= MODULES_END);
 
        xen_setup_features();
 
@@ -174,8 +121,6 @@ void __init x86_64_start_kernel(char * r
        if (!xen_feature(XENFEAT_auto_translated_physmap))
                phys_to_machine_mapping =
                        (unsigned long *)xen_start_info->mfn_list;
-       start_pfn = (__pa(xen_start_info->pt_base) >> PAGE_SHIFT) +
-               xen_start_info->nr_pt_frames;
 
        machine_to_phys_mapping = (unsigned long *)MACH2PHYS_VIRT_START;
        machine_to_phys_nr_ents = MACH2PHYS_NR_ENTRIES;
@@ -208,19 +153,23 @@ void __init x86_64_start_kernel(char * r
 
        early_printk("Kernel alive\n");
 
-       for (i = 0; i < NR_CPUS; i++)
-               cpu_pda(i) = &boot_cpu_pda[i];
+       x86_64_init_pda();
 
-       pda_init(0);
+       early_printk("Kernel really alive\n");
+
+       x86_64_start_reservations(real_mode_data);
+}
+
+void __init x86_64_start_reservations(char *real_mode_data)
+{
        copy_bootdata(__va(real_mode_data));
 
        reserve_early(__pa_symbol(&_text), __pa_symbol(&_end), "TEXT DATA BSS");
 
        reserve_early(round_up(__pa_symbol(&_end), PAGE_SIZE),
-                     start_pfn << PAGE_SHIFT, "Xen provided");
-
-       reserve_ebda_region();
-       reserve_setup_data();
+                     __pa(xen_start_info->pt_base)
+                     + (xen_start_info->nr_pt_frames << PAGE_SHIFT),
+                     "Xen provided");
 
        /*
         * At this point everything still needed from the boot loader
Index: head-2008-12-01/arch/x86/kernel/head_64-xen.S
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/head_64-xen.S  2008-12-01 11:36:47.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/head_64-xen.S       2008-12-01 11:49:07.000000000 +0100
@@ -95,53 +95,6 @@ NEXT_PAGE(hypercall_page)
 
 #undef NEXT_PAGE
 
-       .data
-
-       .align 16
-       .globl cpu_gdt_descr
-cpu_gdt_descr:
-       .word   gdt_end-cpu_gdt_table-1
-gdt:
-       .quad   cpu_gdt_table
-#ifdef CONFIG_SMP
-       .rept   NR_CPUS-1
-       .word   0
-       .quad   0
-       .endr
-#endif
-
-/* We need valid kernel segments for data and code in long mode too
- * IRET will check the segment types  kkeil 2000/10/28
- * Also sysret mandates a special GDT layout 
- */
-                               
-       .section .data.page_aligned, "aw"
-       .align PAGE_SIZE
-
-/* The TLS descriptors are currently at a different place compared to i386.
-   Hopefully nobody expects them at a fixed place (Wine?) */
-
-ENTRY(cpu_gdt_table)
-       .quad   0x0000000000000000      /* NULL descriptor */
-       .quad   0x00cf9b000000ffff      /* __KERNEL32_CS */
-       .quad   0x00af9b000000ffff      /* __KERNEL_CS */
-       .quad   0x00cf93000000ffff      /* __KERNEL_DS */
-       .quad   0x00cffb000000ffff      /* __USER32_CS */
-       .quad   0x00cff3000000ffff      /* __USER_DS, __USER32_DS  */
-       .quad   0x00affb000000ffff      /* __USER_CS */
-       .quad   0x0                     /* unused */
-       .quad   0,0                     /* TSS */
-       .quad   0,0                     /* LDT */
-       .quad   0,0,0                   /* three TLS descriptors */
-       .quad   0x0000f40000000000      /* node/CPU stored in limit */
-gdt_end:
-       /* asm/segment.h:GDT_ENTRIES must match this */
-       /* This should be a multiple of the cache line size */
-       /* GDTs of other CPUs are now dynamically allocated */
-
-       /* zero the remaining page */
-       .fill PAGE_SIZE / 8 - GDT_ENTRIES,8,0
-
        .section .bss.page_aligned, "aw", @nobits
        .align PAGE_SIZE
 ENTRY(empty_zero_page)
Index: head-2008-12-01/arch/x86/kernel/io_apic_32-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/io_apic_32-xen.c       2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/io_apic_32-xen.c    2008-12-01 11:49:07.000000000 +0100
@@ -25,6 +25,7 @@
 #include <linux/init.h>
 #include <linux/delay.h>
 #include <linux/sched.h>
+#include <linux/bootmem.h>
 #include <linux/mc146818rtc.h>
 #include <linux/compiler.h>
 #include <linux/acpi.h>
@@ -75,7 +76,7 @@ static struct { int pin, apic; } ioapic_
 static DEFINE_SPINLOCK(ioapic_lock);
 static DEFINE_SPINLOCK(vector_lock);
 
-int timer_over_8254 __initdata = 1;
+int timer_through_8259 __initdata;
 
 /*
  *     Is the SiS APIC rmw bug present ?
@@ -89,15 +90,21 @@ int sis_apic_bug = -1;
 int nr_ioapic_registers[MAX_IO_APICS];
 
 /* I/O APIC entries */
-struct mpc_config_ioapic mp_ioapics[MAX_IO_APICS];
+struct mp_config_ioapic mp_ioapics[MAX_IO_APICS];
 int nr_ioapics;
 
 /* MP IRQ source entries */
-struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES];
+struct mp_config_intsrc mp_irqs[MAX_IRQ_SOURCES];
 
 /* # of MP IRQ source entries */
 int mp_irq_entries;
 
+#if defined (CONFIG_MCA) || defined (CONFIG_EISA)
+int mp_bus_id_to_type[MAX_MP_BUSSES];
+#endif
+
+DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES);
+
 static int disable_timer_pin_1 __initdata;
 
 /*
@@ -128,7 +135,7 @@ struct io_apic {
 static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx)
 {
        return (void __iomem *) __fix_to_virt(FIX_IO_APIC_BASE_0 + idx)
-               + (mp_ioapics[idx].mpc_apicaddr & ~PAGE_MASK);
+               + (mp_ioapics[idx].mp_apicaddr & ~PAGE_MASK);
 }
 #endif
 
@@ -142,7 +149,7 @@ static inline unsigned int io_apic_read(
        struct physdev_apic apic_op;
        int ret;
 
-       apic_op.apic_physbase = mp_ioapics[apic].mpc_apicaddr;
+       apic_op.apic_physbase = mp_ioapics[apic].mp_apicaddr;
        apic_op.reg = reg;
        ret = HYPERVISOR_physdev_op(PHYSDEVOP_apic_read, &apic_op);
        if (ret)
@@ -160,7 +167,7 @@ static inline void io_apic_write(unsigne
 #else
        struct physdev_apic apic_op;
 
-       apic_op.apic_physbase = mp_ioapics[apic].mpc_apicaddr;
+       apic_op.apic_physbase = mp_ioapics[apic].mp_apicaddr;
        apic_op.reg = reg;
        apic_op.value = value;
        WARN_ON(HYPERVISOR_physdev_op(PHYSDEVOP_apic_write, &apic_op));
@@ -288,7 +295,7 @@ static void __init replace_pin_at_irq(un
        }
 }
 
-static void __modify_IO_APIC_irq (unsigned int irq, unsigned long enable, unsigned long disable)
+static void __modify_IO_APIC_irq(unsigned int irq, unsigned long enable, unsigned long disable)
 {
        struct irq_pin_list *entry = irq_2_pin + irq;
        unsigned int pin, reg;
@@ -308,30 +315,32 @@ static void __modify_IO_APIC_irq (unsign
 }
 
 /* mask = 1 */
-static void __mask_IO_APIC_irq (unsigned int irq)
+static void __mask_IO_APIC_irq(unsigned int irq)
 {
-       __modify_IO_APIC_irq(irq, 0x00010000, 0);
+       __modify_IO_APIC_irq(irq, IO_APIC_REDIR_MASKED, 0);
 }
 
 /* mask = 0 */
-static void __unmask_IO_APIC_irq (unsigned int irq)
+static void __unmask_IO_APIC_irq(unsigned int irq)
 {
-       __modify_IO_APIC_irq(irq, 0, 0x00010000);
+       __modify_IO_APIC_irq(irq, 0, IO_APIC_REDIR_MASKED);
 }
 
 /* mask = 1, trigger = 0 */
-static void __mask_and_edge_IO_APIC_irq (unsigned int irq)
+static void __mask_and_edge_IO_APIC_irq(unsigned int irq)
 {
-       __modify_IO_APIC_irq(irq, 0x00010000, 0x00008000);
+       __modify_IO_APIC_irq(irq, IO_APIC_REDIR_MASKED,
+                               IO_APIC_REDIR_LEVEL_TRIGGER);
 }
 
 /* mask = 0, trigger = 1 */
-static void __unmask_and_level_IO_APIC_irq (unsigned int irq)
+static void __unmask_and_level_IO_APIC_irq(unsigned int irq)
 {
-       __modify_IO_APIC_irq(irq, 0x00008000, 0x00010000);
+       __modify_IO_APIC_irq(irq, IO_APIC_REDIR_LEVEL_TRIGGER,
+                               IO_APIC_REDIR_MASKED);
 }
 
-static void mask_IO_APIC_irq (unsigned int irq)
+static void mask_IO_APIC_irq(unsigned int irq)
 {
        unsigned long flags;
 
@@ -340,7 +349,7 @@ static void mask_IO_APIC_irq (unsigned i
        spin_unlock_irqrestore(&ioapic_lock, flags);
 }
 
-static void unmask_IO_APIC_irq (unsigned int irq)
+static void unmask_IO_APIC_irq(unsigned int irq)
 {
        unsigned long flags;
 
@@ -352,7 +361,7 @@ static void unmask_IO_APIC_irq (unsigned
 static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin)
 {
        struct IO_APIC_route_entry entry;
-       
+
        /* Check delivery_mode to be sure we're not clearing an SMI pin */
        entry = ioapic_read_entry(apic, pin);
        if (entry.delivery_mode == dest_SMI)
@@ -364,7 +373,7 @@ static void clear_IO_APIC_pin(unsigned i
        ioapic_mask_entry(apic, pin);
 }
 
-static void clear_IO_APIC (void)
+static void clear_IO_APIC(void)
 {
        int apic, pin;
 
@@ -381,7 +390,7 @@ static void set_ioapic_affinity_irq(unsi
        struct irq_pin_list *entry = irq_2_pin + irq;
        unsigned int apicid_value;
        cpumask_t tmp;
-       
+
        cpus_and(tmp, cpumask, cpu_online_map);
        if (cpus_empty(tmp))
                tmp = TARGET_CPUS;
@@ -410,7 +419,7 @@ static void set_ioapic_affinity_irq(unsi
 # include <linux/kernel_stat.h>        /* kstat */
 # include <linux/slab.h>               /* kmalloc() */
 # include <linux/timer.h>
- 
+
 #define IRQBALANCE_CHECK_ARCH -999
 #define MAX_BALANCED_IRQ_INTERVAL      (5*HZ)
 #define MIN_BALANCED_IRQ_INTERVAL      (HZ/2)
@@ -422,14 +431,14 @@ static int physical_balance __read_mostl
 static long balanced_irq_interval __read_mostly = MAX_BALANCED_IRQ_INTERVAL;
 
 static struct irq_cpu_info {
-       unsigned long * last_irq;
-       unsigned long * irq_delta;
+       unsigned long *last_irq;
+       unsigned long *irq_delta;
        unsigned long irq;
 } irq_cpu_data[NR_CPUS];
 
 #define CPU_IRQ(cpu)           (irq_cpu_data[cpu].irq)
-#define LAST_CPU_IRQ(cpu,irq)   (irq_cpu_data[cpu].last_irq[irq])
-#define IRQ_DELTA(cpu,irq)     (irq_cpu_data[cpu].irq_delta[irq])
+#define LAST_CPU_IRQ(cpu, irq)   (irq_cpu_data[cpu].last_irq[irq])
+#define IRQ_DELTA(cpu, irq)    (irq_cpu_data[cpu].irq_delta[irq])
 
 #define IDLE_ENOUGH(cpu,now) \
        (idle_cpu(cpu) && ((now) - per_cpu(irq_stat, (cpu)).idle_timestamp > 1))
@@ -468,8 +477,8 @@ inside:
                        if (cpu == -1)
                                cpu = NR_CPUS-1;
                }
-       } while (!cpu_online(cpu) || !IRQ_ALLOWED(cpu,allowed_mask) ||
-                       (search_idle && !IDLE_ENOUGH(cpu,now)));
+       } while (!cpu_online(cpu) || !IRQ_ALLOWED(cpu, allowed_mask) ||
+                       (search_idle && !IDLE_ENOUGH(cpu, now)));
 
        return cpu;
 }
@@ -479,15 +488,14 @@ static inline void balance_irq(int cpu, 
        unsigned long now = jiffies;
        cpumask_t allowed_mask;
        unsigned int new_cpu;
-               
+
        if (irqbalance_disabled)
-               return; 
+               return;
 
        cpus_and(allowed_mask, cpu_online_map, balance_irq_affinity[irq]);
        new_cpu = move(cpu, allowed_mask, now, 1);
-       if (cpu != new_cpu) {
+       if (cpu != new_cpu)
                set_pending_irq(irq, cpumask_of_cpu(new_cpu));
-       }
 }
 
 static inline void rotate_irqs_among_cpus(unsigned long useful_load_threshold)
@@ -499,14 +507,14 @@ static inline void rotate_irqs_among_cpu
                        if (!irq_desc[j].action)
                                continue;
                        /* Is it a significant load ?  */
-                       if (IRQ_DELTA(CPU_TO_PACKAGEINDEX(i),j) <
+                       if (IRQ_DELTA(CPU_TO_PACKAGEINDEX(i), j) <
                                                useful_load_threshold)
                                continue;
                        balance_irq(i, j);
                }
        }
        balanced_irq_interval = max((long)MIN_BALANCED_IRQ_INTERVAL,
-               balanced_irq_interval - BALANCED_IRQ_LESS_DELTA);       
+               balanced_irq_interval - BALANCED_IRQ_LESS_DELTA);
        return;
 }
 
@@ -535,22 +543,22 @@ static void do_irq_balance(void)
                        /* Is this an active IRQ or balancing disabled ? */
                        if (!irq_desc[j].action || irq_balancing_disabled(j))
                                continue;
-                       if ( package_index == i )
-                               IRQ_DELTA(package_index,j) = 0;
+                       if (package_index == i)
+                               IRQ_DELTA(package_index, j) = 0;
                        /* Determine the total count per processor per IRQ */
                        value_now = (unsigned long) kstat_cpu(i).irqs[j];
 
                        /* Determine the activity per processor per IRQ */
-                       delta = value_now - LAST_CPU_IRQ(i,j);
+                       delta = value_now - LAST_CPU_IRQ(i, j);
 
                        /* Update last_cpu_irq[][] for the next time */
-                       LAST_CPU_IRQ(i,j) = value_now;
+                       LAST_CPU_IRQ(i, j) = value_now;
 
                        /* Ignore IRQs whose rate is less than the clock */
                        if (delta < useful_load_threshold)
                                continue;
                        /* update the load for the processor or package total */
-                       IRQ_DELTA(package_index,j) += delta;
+                       IRQ_DELTA(package_index, j) += delta;
 
                        /* Keep track of the higher numbered sibling as well */
                        if (i != package_index)
@@ -576,7 +584,8 @@ static void do_irq_balance(void)
        max_cpu_irq = ULONG_MAX;
 
 tryanothercpu:
-       /* Look for heaviest loaded processor.
+       /*
+        * Look for heaviest loaded processor.
         * We may come back to get the next heaviest loaded processor.
         * Skip processors with trivial loads.
         */
@@ -585,7 +594,7 @@ tryanothercpu:
        for_each_online_cpu(i) {
                if (i != CPU_TO_PACKAGEINDEX(i))
                        continue;
-               if (max_cpu_irq <= CPU_IRQ(i)) 
+               if (max_cpu_irq <= CPU_IRQ(i))
                        continue;
                if (tmp_cpu_irq < CPU_IRQ(i)) {
                        tmp_cpu_irq = CPU_IRQ(i);
@@ -594,8 +603,9 @@ tryanothercpu:
        }
 
        if (tmp_loaded == -1) {
-        /* In the case of small number of heavy interrupt sources, 
-         * loading some of the cpus too much. We use Ingo's original 
+        /*
+         * In the case of small number of heavy interrupt sources,
+         * loading some of the cpus too much. We use Ingo's original
          * approach to rotate them around.
          */
                if (!first_attempt && imbalance >= useful_load_threshold) {
@@ -604,13 +614,14 @@ tryanothercpu:
                }
                goto not_worth_the_effort;
        }
-       
+
        first_attempt = 0;              /* heaviest search */
        max_cpu_irq = tmp_cpu_irq;      /* load */
        max_loaded = tmp_loaded;        /* processor */
        imbalance = (max_cpu_irq - min_cpu_irq) / 2;
-       
-       /* if imbalance is less than approx 10% of max load, then
+
+       /*
+        * if imbalance is less than approx 10% of max load, then
         * observe diminishing returns action. - quit
         */
        if (imbalance < (max_cpu_irq >> 3))
@@ -626,26 +637,25 @@ tryanotherirq:
                /* Is this an active IRQ? */
                if (!irq_desc[j].action)
                        continue;
-               if (imbalance <= IRQ_DELTA(max_loaded,j))
+               if (imbalance <= IRQ_DELTA(max_loaded, j))
                        continue;
                /* Try to find the IRQ that is closest to the imbalance
                 * without going over.
                 */
-               if (move_this_load < IRQ_DELTA(max_loaded,j)) {
-                       move_this_load = IRQ_DELTA(max_loaded,j);
+               if (move_this_load < IRQ_DELTA(max_loaded, j)) {
+                       move_this_load = IRQ_DELTA(max_loaded, j);
                        selected_irq = j;
                }
        }
-       if (selected_irq == -1) {
+       if (selected_irq == -1)
                goto tryanothercpu;
-       }
 
        imbalance = move_this_load;
-       
+
        /* For physical_balance case, we accumulated both load
         * values in the one of the siblings cpu_irq[],
         * to use the same code for physical and logical processors
-        * as much as possible. 
+        * as much as possible.
         *
         * NOTE: the cpu_irq[] array holds the sum of the load for
         * sibling A and sibling B in the slot for the lowest numbered
@@ -674,11 +684,11 @@ tryanotherirq:
                /* mark for change destination */
                set_pending_irq(selected_irq, cpumask_of_cpu(min_loaded));
 
-               /* Since we made a change, come back sooner to 
+               /* Since we made a change, come back sooner to
                 * check for more variation.
                 */
                balanced_irq_interval = max((long)MIN_BALANCED_IRQ_INTERVAL,
-                       balanced_irq_interval - BALANCED_IRQ_LESS_DELTA);       
+                       balanced_irq_interval - BALANCED_IRQ_LESS_DELTA);
                return;
        }
        goto tryanotherirq;
@@ -689,7 +699,7 @@ not_worth_the_effort:
         * upward
         */
        balanced_irq_interval = min((long)MAX_BALANCED_IRQ_INTERVAL,
-               balanced_irq_interval + BALANCED_IRQ_MORE_DELTA);       
+               balanced_irq_interval + BALANCED_IRQ_MORE_DELTA);
        return;
 }
 
@@ -728,13 +738,13 @@ static int __init balanced_irq_init(void
        cpumask_t tmp;
 
        cpus_shift_right(tmp, cpu_online_map, 2);
-        c = &boot_cpu_data;
+       c = &boot_cpu_data;
        /* When not overwritten by the command line ask subarchitecture. */
        if (irqbalance_disabled == IRQBALANCE_CHECK_ARCH)
                irqbalance_disabled = NO_BALANCE_IRQ;
        if (irqbalance_disabled)
                return 0;
-       
+
         /* disable irqbalance completely if there is only one processor online */
        if (num_online_cpus() < 2) {
                irqbalance_disabled = 1;
@@ -748,16 +758,14 @@ static int __init balanced_irq_init(void
                physical_balance = 1;
 
        for_each_online_cpu(i) {
-               irq_cpu_data[i].irq_delta = kmalloc(sizeof(unsigned long) * NR_IRQS, GFP_KERNEL);
-               irq_cpu_data[i].last_irq = kmalloc(sizeof(unsigned long) * NR_IRQS, GFP_KERNEL);
+               irq_cpu_data[i].irq_delta = kzalloc(sizeof(unsigned long) * NR_IRQS, GFP_KERNEL);
+               irq_cpu_data[i].last_irq = kzalloc(sizeof(unsigned long) * NR_IRQS, GFP_KERNEL);
                if (irq_cpu_data[i].irq_delta == NULL || irq_cpu_data[i].last_irq == NULL) {
                        printk(KERN_ERR "balanced_irq_init: out of memory");
                        goto failed;
                }
-               memset(irq_cpu_data[i].irq_delta,0,sizeof(unsigned long) * NR_IRQS);
-               memset(irq_cpu_data[i].last_irq,0,sizeof(unsigned long) * NR_IRQS);
        }
-       
+
        printk(KERN_INFO "Starting balanced_irq\n");
        if (!IS_ERR(kthread_run(balanced_irq, NULL, "kirqd")))
                return 0;
@@ -799,7 +807,7 @@ void send_IPI_self(int vector)
        /*
         * Send the IPI. The write to APIC_ICR fires this off.
         */
-       apic_write_around(APIC_ICR, cfg);
+       apic_write(APIC_ICR, cfg);
 #endif
 }
 #endif /* !CONFIG_SMP */
@@ -853,10 +861,10 @@ static int find_irq_entry(int apic, int 
        int i;
 
        for (i = 0; i < mp_irq_entries; i++)
-               if (mp_irqs[i].mpc_irqtype == type &&
-                   (mp_irqs[i].mpc_dstapic == mp_ioapics[apic].mpc_apicid ||
-                    mp_irqs[i].mpc_dstapic == MP_APIC_ALL) &&
-                   mp_irqs[i].mpc_dstirq == pin)
+               if (mp_irqs[i].mp_irqtype == type &&
+                   (mp_irqs[i].mp_dstapic == mp_ioapics[apic].mp_apicid ||
+                    mp_irqs[i].mp_dstapic == MP_APIC_ALL) &&
+                   mp_irqs[i].mp_dstirq == pin)
                        return i;
 
        return -1;
@@ -871,13 +879,13 @@ static int __init find_isa_irq_pin(int i
        int i;
 
        for (i = 0; i < mp_irq_entries; i++) {
-               int lbus = mp_irqs[i].mpc_srcbus;
+               int lbus = mp_irqs[i].mp_srcbus;
 
                if (test_bit(lbus, mp_bus_not_pci) &&
-                   (mp_irqs[i].mpc_irqtype == type) &&
-                   (mp_irqs[i].mpc_srcbusirq == irq))
+                   (mp_irqs[i].mp_irqtype == type) &&
+                   (mp_irqs[i].mp_srcbusirq == irq))
 
-                       return mp_irqs[i].mpc_dstirq;
+                       return mp_irqs[i].mp_dstirq;
        }
        return -1;
 }
@@ -887,17 +895,17 @@ static int __init find_isa_irq_apic(int 
        int i;
 
        for (i = 0; i < mp_irq_entries; i++) {
-               int lbus = mp_irqs[i].mpc_srcbus;
+               int lbus = mp_irqs[i].mp_srcbus;
 
                if (test_bit(lbus, mp_bus_not_pci) &&
-                   (mp_irqs[i].mpc_irqtype == type) &&
-                   (mp_irqs[i].mpc_srcbusirq == irq))
+                   (mp_irqs[i].mp_irqtype == type) &&
+                   (mp_irqs[i].mp_srcbusirq == irq))
                        break;
        }
        if (i < mp_irq_entries) {
                int apic;
-               for(apic = 0; apic < nr_ioapics; apic++) {
-                       if (mp_ioapics[apic].mpc_apicid == mp_irqs[i].mpc_dstapic)
+               for (apic = 0; apic < nr_ioapics; apic++) {
+                       if (mp_ioapics[apic].mp_apicid == mp_irqs[i].mp_dstapic)
                                return apic;
                }
        }
@@ -918,28 +926,28 @@ int IO_APIC_get_PCI_irq_vector(int bus, 
 
        apic_printk(APIC_DEBUG, "querying PCI -> IRQ mapping bus:%d, "
                "slot:%d, pin:%d.\n", bus, slot, pin);
-       if (mp_bus_id_to_pci_bus[bus] == -1) {
+       if (test_bit(bus, mp_bus_not_pci)) {
                printk(KERN_WARNING "PCI BIOS passed nonexistent PCI bus %d!\n", bus);
                return -1;
        }
        for (i = 0; i < mp_irq_entries; i++) {
-               int lbus = mp_irqs[i].mpc_srcbus;
+               int lbus = mp_irqs[i].mp_srcbus;
 
                for (apic = 0; apic < nr_ioapics; apic++)
-                       if (mp_ioapics[apic].mpc_apicid == mp_irqs[i].mpc_dstapic ||
-                           mp_irqs[i].mpc_dstapic == MP_APIC_ALL)
+                       if (mp_ioapics[apic].mp_apicid == mp_irqs[i].mp_dstapic ||
+                           mp_irqs[i].mp_dstapic == MP_APIC_ALL)
                                break;
 
                if (!test_bit(lbus, mp_bus_not_pci) &&
-                   !mp_irqs[i].mpc_irqtype &&
+                   !mp_irqs[i].mp_irqtype &&
                    (bus == lbus) &&
-                   (slot == ((mp_irqs[i].mpc_srcbusirq >> 2) & 0x1f))) {
-                       int irq = pin_2_irq(i,apic,mp_irqs[i].mpc_dstirq);
+                   (slot == ((mp_irqs[i].mp_srcbusirq >> 2) & 0x1f))) {
+                       int irq = pin_2_irq(i, apic, mp_irqs[i].mp_dstirq);
 
                        if (!(apic || IO_APIC_IRQ(irq)))
                                continue;
 
-                       if (pin == (mp_irqs[i].mpc_srcbusirq & 3))
+                       if (pin == (mp_irqs[i].mp_srcbusirq & 3))
                                return irq;
                        /*
                         * Use the first all-but-pin matching entry as a
@@ -954,7 +962,7 @@ int IO_APIC_get_PCI_irq_vector(int bus, 
 EXPORT_SYMBOL(IO_APIC_get_PCI_irq_vector);
 
 /*
- * This function currently is only a helper for the i386 smp boot process where 
+ * This function currently is only a helper for the i386 smp boot process where
  * we need to reprogram the ioredtbls to cater for the cpus which have come online
  * so mask in all cases should simply be TARGET_CPUS
  */
@@ -1008,7 +1016,7 @@ static int EISA_ELCR(unsigned int irq)
  * EISA conforming in the MP table, that means its trigger type must
  * be read in from the ELCR */
 
-#define default_EISA_trigger(idx)      (EISA_ELCR(mp_irqs[idx].mpc_srcbusirq))
+#define default_EISA_trigger(idx)      (EISA_ELCR(mp_irqs[idx].mp_srcbusirq))
 #define default_EISA_polarity(idx)     default_ISA_polarity(idx)
 
 /* PCI interrupts are always polarity one level triggered,
@@ -1025,118 +1033,115 @@ static int EISA_ELCR(unsigned int irq)
 
 static int MPBIOS_polarity(int idx)
 {
-       int bus = mp_irqs[idx].mpc_srcbus;
+       int bus = mp_irqs[idx].mp_srcbus;
        int polarity;
 
        /*
         * Determine IRQ line polarity (high active or low active):
         */
-       switch (mp_irqs[idx].mpc_irqflag & 3)
+       switch (mp_irqs[idx].mp_irqflag & 3) {
+       case 0: /* conforms, ie. bus-type dependent polarity */
        {
-               case 0: /* conforms, ie. bus-type dependent polarity */
-               {
-                       polarity = test_bit(bus, mp_bus_not_pci)?
-                               default_ISA_polarity(idx):
-                               default_PCI_polarity(idx);
-                       break;
-               }
-               case 1: /* high active */
-               {
-                       polarity = 0;
-                       break;
-               }
-               case 2: /* reserved */
-               {
-                       printk(KERN_WARNING "broken BIOS!!\n");
-                       polarity = 1;
-                       break;
-               }
-               case 3: /* low active */
-               {
-                       polarity = 1;
-                       break;
-               }
-               default: /* invalid */
-               {
-                       printk(KERN_WARNING "broken BIOS!!\n");
-                       polarity = 1;
-                       break;
-               }
+               polarity = test_bit(bus, mp_bus_not_pci)?
+                       default_ISA_polarity(idx):
+                       default_PCI_polarity(idx);
+               break;
+       }
+       case 1: /* high active */
+       {
+               polarity = 0;
+               break;
+       }
+       case 2: /* reserved */
+       {
+               printk(KERN_WARNING "broken BIOS!!\n");
+               polarity = 1;
+               break;
+       }
+       case 3: /* low active */
+       {
+               polarity = 1;
+               break;
+       }
+       default: /* invalid */
+       {
+               printk(KERN_WARNING "broken BIOS!!\n");
+               polarity = 1;
+               break;
+       }
        }
        return polarity;
 }
 
 static int MPBIOS_trigger(int idx)
 {
-       int bus = mp_irqs[idx].mpc_srcbus;
+       int bus = mp_irqs[idx].mp_srcbus;
        int trigger;
 
        /*
         * Determine IRQ trigger mode (edge or level sensitive):
         */
-       switch ((mp_irqs[idx].mpc_irqflag>>2) & 3)
+       switch ((mp_irqs[idx].mp_irqflag>>2) & 3) {
+       case 0: /* conforms, ie. bus-type dependent */
        {
-               case 0: /* conforms, ie. bus-type dependent */
-               {
-                       trigger = test_bit(bus, mp_bus_not_pci)?
-                                       default_ISA_trigger(idx):
-                                       default_PCI_trigger(idx);
+               trigger = test_bit(bus, mp_bus_not_pci)?
+                               default_ISA_trigger(idx):
+                               default_PCI_trigger(idx);
 #if defined(CONFIG_EISA) || defined(CONFIG_MCA)
-                       switch (mp_bus_id_to_type[bus])
-                       {
-                               case MP_BUS_ISA: /* ISA pin */
-                               {
-                                       /* set before the switch */
-                                       break;
-                               }
-                               case MP_BUS_EISA: /* EISA pin */
-                               {
-                                       trigger = default_EISA_trigger(idx);
-                                       break;
-                               }
-                               case MP_BUS_PCI: /* PCI pin */
-                               {
-                                       /* set before the switch */
-                                       break;
-                               }
-                               case MP_BUS_MCA: /* MCA pin */
-                               {
-                                       trigger = default_MCA_trigger(idx);
-                                       break;
-                               }
-                               default:
-                               {
-                                       printk(KERN_WARNING "broken BIOS!!\n");
-                                       trigger = 1;
-                                       break;
-                               }
-                       }
-#endif
+               switch (mp_bus_id_to_type[bus]) {
+               case MP_BUS_ISA: /* ISA pin */
+               {
+                       /* set before the switch */
                        break;
                }
-               case 1: /* edge */
+               case MP_BUS_EISA: /* EISA pin */
                {
-                       trigger = 0;
+                       trigger = default_EISA_trigger(idx);
                        break;
                }
-               case 2: /* reserved */
+               case MP_BUS_PCI: /* PCI pin */
                {
-                       printk(KERN_WARNING "broken BIOS!!\n");
-                       trigger = 1;
+                       /* set before the switch */
                        break;
                }
-               case 3: /* level */
+               case MP_BUS_MCA: /* MCA pin */
                {
-                       trigger = 1;
+                       trigger = default_MCA_trigger(idx);
                        break;
                }
-               default: /* invalid */
+               default:
                {
                        printk(KERN_WARNING "broken BIOS!!\n");
-                       trigger = 0;
+                       trigger = 1;
                        break;
                }
        }
+#endif
+               break;
+       }
+       case 1: /* edge */
+       {
+               trigger = 0;
+               break;
+       }
+       case 2: /* reserved */
+       {
+               printk(KERN_WARNING "broken BIOS!!\n");
+               trigger = 1;
+               break;
+       }
+       case 3: /* level */
+       {
+               trigger = 1;
+               break;
+       }
+       default: /* invalid */
+       {
+               printk(KERN_WARNING "broken BIOS!!\n");
+               trigger = 0;
+               break;
+       }
+       }
        return trigger;
 }
 
@@ -1153,16 +1158,16 @@ static inline int irq_trigger(int idx)
 static int pin_2_irq(int idx, int apic, int pin)
 {
        int irq, i;
-       int bus = mp_irqs[idx].mpc_srcbus;
+       int bus = mp_irqs[idx].mp_srcbus;
 
        /*
         * Debugging check, we are in big trouble if this message pops up!
         */
-       if (mp_irqs[idx].mpc_dstirq != pin)
+       if (mp_irqs[idx].mp_dstirq != pin)
                printk(KERN_ERR "broken BIOS or MPTABLE parser, ayiee!!\n");
 
        if (test_bit(bus, mp_bus_not_pci))
-               irq = mp_irqs[idx].mpc_srcbusirq;
+               irq = mp_irqs[idx].mp_srcbusirq;
        else {
                /*
                 * PCI IRQs are mapped in order
@@ -1204,8 +1209,8 @@ static inline int IO_APIC_irq_trigger(in
 
        for (apic = 0; apic < nr_ioapics; apic++) {
                for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) {
-                       idx = find_irq_entry(apic,pin,mp_INT);
-                       if ((idx != -1) && (irq == pin_2_irq(idx,apic,pin)))
+                       idx = find_irq_entry(apic, pin, mp_INT);
+                       if ((idx != -1) && (irq == pin_2_irq(idx, apic, pin)))
                                return irq_trigger(idx);
                }
        }
@@ -1291,25 +1296,25 @@ static void __init setup_IO_APIC_irqs(vo
                /*
                 * add it to the IO-APIC irq-routing table:
                 */
-               memset(&entry,0,sizeof(entry));
+               memset(&entry, 0, sizeof(entry));
 
                entry.delivery_mode = INT_DELIVERY_MODE;
                entry.dest_mode = INT_DEST_MODE;
                entry.mask = 0;                         /* enable IRQ */
-               entry.dest.logical.logical_dest = 
+               entry.dest.logical.logical_dest =
                                        cpu_mask_to_apicid(TARGET_CPUS);
 
-               idx = find_irq_entry(apic,pin,mp_INT);
+               idx = find_irq_entry(apic, pin, mp_INT);
                if (idx == -1) {
                        if (first_notcon) {
                                apic_printk(APIC_VERBOSE, KERN_DEBUG
                                                " IO-APIC (apicid-pin) %d-%d",
-                                               mp_ioapics[apic].mpc_apicid,
+                                               mp_ioapics[apic].mp_apicid,
                                                pin);
                                first_notcon = 0;
                        } else
                                apic_printk(APIC_VERBOSE, ", %d-%d",
-                                       mp_ioapics[apic].mpc_apicid, pin);
+                                       mp_ioapics[apic].mp_apicid, pin);
                        continue;
                }
 
@@ -1343,7 +1348,7 @@ static void __init setup_IO_APIC_irqs(vo
                        vector = assign_irq_vector(irq);
                        entry.vector = vector;
                        ioapic_register_intr(irq, vector, IOAPIC_AUTO);
-               
+
                        if (!apic && (irq < 16))
                                disable_8259A_irq(irq);
                }
@@ -1355,27 +1360,23 @@ static void __init setup_IO_APIC_irqs(vo
                apic_printk(APIC_VERBOSE, " not connected.\n");
 }
 
+#ifndef CONFIG_XEN
 /*
- * Set up the 8259A-master output pin:
+ * Set up the timer pin, possibly with the 8259A-master behind.
  */
-#ifndef CONFIG_XEN
-static void __init setup_ExtINT_IRQ0_pin(unsigned int apic, unsigned int pin, int vector)
+static void __init setup_timer_IRQ0_pin(unsigned int apic, unsigned int pin,
+                                       int vector)
 {
        struct IO_APIC_route_entry entry;
 
-       memset(&entry,0,sizeof(entry));
-
-       disable_8259A_irq(0);
-
-       /* mask LVT0 */
-       apic_write_around(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT);
+       memset(&entry, 0, sizeof(entry));
 
        /*
         * We use logical delivery to get the timer IRQ
         * to the first CPU.
         */
        entry.dest_mode = INT_DEST_MODE;
-       entry.mask = 0;                                 /* unmask IRQ now */
+       entry.mask = 1;                                 /* mask IRQ now */
        entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS);
        entry.delivery_mode = INT_DELIVERY_MODE;
        entry.polarity = 0;
@@ -1384,17 +1385,14 @@ static void __init setup_ExtINT_IRQ0_pin
 
        /*
         * The timer IRQ doesn't have to know that behind the
-        * scene we have a 8259A-master in AEOI mode ...
+        * scene we may have a 8259A-master in AEOI mode ...
         */
-       irq_desc[0].chip = &ioapic_chip;
-       set_irq_handler(0, handle_edge_irq);
+       ioapic_register_intr(0, vector, IOAPIC_EDGE);
 
        /*
         * Add it to the IO-APIC irq-routing table:
         */
        ioapic_write_entry(apic, pin, entry);
-
-       enable_8259A_irq(0);
 }
 
 void __init print_IO_APIC(void)
@@ -1409,10 +1407,10 @@ void __init print_IO_APIC(void)
        if (apic_verbosity == APIC_QUIET)
                return;
 
-       printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries);
+       printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries);
        for (i = 0; i < nr_ioapics; i++)
                printk(KERN_DEBUG "number of IO-APIC #%d registers: %d.\n",
-                      mp_ioapics[i].mpc_apicid, nr_ioapic_registers[i]);
+                      mp_ioapics[i].mp_apicid, nr_ioapic_registers[i]);
 
        /*
         * We are a bit conservative about what we expect.  We have to
@@ -1431,7 +1429,7 @@ void __init print_IO_APIC(void)
                reg_03.raw = io_apic_read(apic, 3);
        spin_unlock_irqrestore(&ioapic_lock, flags);
 
-       printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].mpc_apicid);
+       printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].mp_apicid);
        printk(KERN_DEBUG ".... register #00: %08X\n", reg_00.raw);
        printk(KERN_DEBUG ".......    : physical APIC id: %02X\n", reg_00.bits.ID);
        printk(KERN_DEBUG ".......    : Delivery Type: %X\n", reg_00.bits.delivery_type);
@@ -1512,7 +1510,7 @@ void __init print_IO_APIC(void)
        return;
 }
 
-static void print_APIC_bitfield (int base)
+static void print_APIC_bitfield(int base)
 {
        unsigned int v;
        int i, j;
@@ -1533,7 +1531,7 @@ static void print_APIC_bitfield (int bas
        }
 }
 
-void /*__init*/ print_local_APIC(void * dummy)
+void /*__init*/ print_local_APIC(void *dummy)
 {
        unsigned int v, ver, maxlvt;
 
@@ -1542,6 +1540,7 @@ void /*__init*/ print_local_APIC(void * 
 
        printk("\n" KERN_DEBUG "printing local APIC contents on CPU#%d/%d:\n",
                smp_processor_id(), hard_smp_processor_id());
+       v = apic_read(APIC_ID);
        printk(KERN_INFO "... APIC ID:      %08x (%01x)\n", v,
                        GET_APIC_ID(read_apic_id()));
        v = apic_read(APIC_LVR);
@@ -1616,9 +1615,9 @@ void /*__init*/ print_local_APIC(void * 
        printk("\n");
 }
 
-void print_all_local_APICs (void)
+void print_all_local_APICs(void)
 {
-       on_each_cpu(print_local_APIC, NULL, 1, 1);
+       on_each_cpu(print_local_APIC, NULL, 1);
 }
 
 void /*__init*/ print_PIC(void)
@@ -1639,11 +1638,11 @@ void /*__init*/ print_PIC(void)
        v = inb(0xa0) << 8 | inb(0x20);
        printk(KERN_DEBUG "... PIC  IRR: %04x\n", v);
 
-       outb(0x0b,0xa0);
-       outb(0x0b,0x20);
+       outb(0x0b, 0xa0);
+       outb(0x0b, 0x20);
        v = inb(0xa0) << 8 | inb(0x20);
-       outb(0x0a,0xa0);
-       outb(0x0a,0x20);
+       outb(0x0a, 0xa0);
+       outb(0x0a, 0x20);
 
        spin_unlock_irqrestore(&i8259A_lock, flags);
 
@@ -1652,6 +1651,8 @@ void /*__init*/ print_PIC(void)
        v = inb(0x4d1) << 8 | inb(0x4d0);
        printk(KERN_DEBUG "... PIC ELCR: %04x\n", v);
 }
+#else
+void __init print_IO_APIC(void) {}
 #endif /* !CONFIG_XEN */
 
 static void __init enable_IO_APIC(void)
@@ -1681,7 +1682,7 @@ static void __init enable_IO_APIC(void)
                nr_ioapic_registers[apic] = reg_01.bits.entries+1;
        }
 #ifndef CONFIG_XEN
-       for(apic = 0; apic < nr_ioapics; apic++) {
+       for (apic = 0; apic < nr_ioapics; apic++) {
                int pin;
                /* See if any of the pins is in ExtINT mode */
                for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) {
@@ -1774,7 +1775,7 @@ void disable_IO_APIC(void)
  * by Matt Domsch <Matt_Domsch@dell.com>  Tue Dec 21 12:25:05 CST 1999
  */
 
-#if !defined(CONFIG_XEN) && !defined(CONFIG_X86_NUMAQ)
+#ifndef CONFIG_XEN
 static void __init setup_ioapic_ids_from_mpc(void)
 {
        union IO_APIC_reg_00 reg_00;
@@ -1784,6 +1785,11 @@ static void __init setup_ioapic_ids_from
        unsigned char old_id;
        unsigned long flags;
 
+#ifdef CONFIG_X86_NUMAQ
+       if (found_numaq)
+               return;
+#endif
+
        /*
         * Don't check I/O APIC IDs for xAPIC systems.  They have
         * no meaning without the serial APIC bus.
@@ -1806,15 +1812,15 @@ static void __init setup_ioapic_ids_from
                spin_lock_irqsave(&ioapic_lock, flags);
                reg_00.raw = io_apic_read(apic, 0);
                spin_unlock_irqrestore(&ioapic_lock, flags);
-               
-               old_id = mp_ioapics[apic].mpc_apicid;
 
-               if (mp_ioapics[apic].mpc_apicid >= get_physical_broadcast()) {
+               old_id = mp_ioapics[apic].mp_apicid;
+
+               if (mp_ioapics[apic].mp_apicid >= get_physical_broadcast()) {
                        printk(KERN_ERR "BIOS bug, IO-APIC#%d ID is %d in the MPC table!...\n",
-                               apic, mp_ioapics[apic].mpc_apicid);
+                               apic, mp_ioapics[apic].mp_apicid);
                        printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n",
                                reg_00.bits.ID);
-                       mp_ioapics[apic].mpc_apicid = reg_00.bits.ID;
+                       mp_ioapics[apic].mp_apicid = reg_00.bits.ID;
                }
 
                /*
@@ -1823,9 +1829,9 @@ static void __init setup_ioapic_ids_from
                 * 'stuck on smp_invalidate_needed IPI wait' messages.
                 */
                if (check_apicid_used(phys_id_present_map,
-                                       mp_ioapics[apic].mpc_apicid)) {
+                                       mp_ioapics[apic].mp_apicid)) {
                        printk(KERN_ERR "BIOS bug, IO-APIC#%d ID %d is already used!...\n",
-                               apic, mp_ioapics[apic].mpc_apicid);
+                               apic, mp_ioapics[apic].mp_apicid);
                        for (i = 0; i < get_physical_broadcast(); i++)
                                if (!physid_isset(i, phys_id_present_map))
                                        break;
@@ -1834,13 +1840,13 @@ static void __init setup_ioapic_ids_from
                        printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n",
                                i);
                        physid_set(i, phys_id_present_map);
-                       mp_ioapics[apic].mpc_apicid = i;
+                       mp_ioapics[apic].mp_apicid = i;
                } else {
                        physid_mask_t tmp;
-                       tmp = apicid_to_cpu_present(mp_ioapics[apic].mpc_apicid);
+                       tmp = apicid_to_cpu_present(mp_ioapics[apic].mp_apicid);
                        apic_printk(APIC_VERBOSE, "Setting %d in the "
                                        "phys_id_present_map\n",
-                                       mp_ioapics[apic].mpc_apicid);
+                                       mp_ioapics[apic].mp_apicid);
                        physids_or(phys_id_present_map, phys_id_present_map, tmp);
                }
 
@@ -1849,21 +1855,21 @@ static void __init setup_ioapic_ids_from
                 * We need to adjust the IRQ routing table
                 * if the ID changed.
                 */
-               if (old_id != mp_ioapics[apic].mpc_apicid)
+               if (old_id != mp_ioapics[apic].mp_apicid)
                        for (i = 0; i < mp_irq_entries; i++)
-                               if (mp_irqs[i].mpc_dstapic == old_id)
-                                       mp_irqs[i].mpc_dstapic
-                                               = mp_ioapics[apic].mpc_apicid;
+                               if (mp_irqs[i].mp_dstapic == old_id)
+                                       mp_irqs[i].mp_dstapic
+                                               = mp_ioapics[apic].mp_apicid;
 
                /*
                 * Read the right value from the MPC table and
                 * write it into the ID register.
-                */
+                */
                apic_printk(APIC_VERBOSE, KERN_INFO
                        "...changing IO-APIC physical APIC ID to %d ...",
-                       mp_ioapics[apic].mpc_apicid);
+                       mp_ioapics[apic].mp_apicid);
 
-               reg_00.bits.ID = mp_ioapics[apic].mpc_apicid;
+               reg_00.bits.ID = mp_ioapics[apic].mp_apicid;
                spin_lock_irqsave(&ioapic_lock, flags);
                io_apic_write(apic, 0, reg_00.raw);
                spin_unlock_irqrestore(&ioapic_lock, flags);
@@ -1874,17 +1880,13 @@ static void __init setup_ioapic_ids_from
                spin_lock_irqsave(&ioapic_lock, flags);
                reg_00.raw = io_apic_read(apic, 0);
                spin_unlock_irqrestore(&ioapic_lock, flags);
-               if (reg_00.bits.ID != mp_ioapics[apic].mpc_apicid)
+               if (reg_00.bits.ID != mp_ioapics[apic].mp_apicid)
                        printk("could not set ID!\n");
                else
                        apic_printk(APIC_VERBOSE, " ok.\n");
        }
 }
-#else
-static void __init setup_ioapic_ids_from_mpc(void) { }
-#endif
 
-#ifndef CONFIG_XEN
 int no_timer_check __initdata;
 
 static int __init notimercheck(char *s)
@@ -2077,45 +2079,53 @@ static inline void init_IO_APIC_traps(vo
  * The local APIC irq-chip implementation:
  */
 
-static void ack_apic(unsigned int irq)
+static void ack_lapic_irq(unsigned int irq)
 {
        ack_APIC_irq();
 }
 
-static void mask_lapic_irq (unsigned int irq)
+static void mask_lapic_irq(unsigned int irq)
 {
        unsigned long v;
 
        v = apic_read(APIC_LVT0);
-       apic_write_around(APIC_LVT0, v | APIC_LVT_MASKED);
+       apic_write(APIC_LVT0, v | APIC_LVT_MASKED);
 }
 
-static void unmask_lapic_irq (unsigned int irq)
+static void unmask_lapic_irq(unsigned int irq)
 {
        unsigned long v;
 
        v = apic_read(APIC_LVT0);
-       apic_write_around(APIC_LVT0, v & ~APIC_LVT_MASKED);
+       apic_write(APIC_LVT0, v & ~APIC_LVT_MASKED);
 }
 
 static struct irq_chip lapic_chip __read_mostly = {
-       .name           = "local-APIC-edge",
+       .name           = "local-APIC",
        .mask           = mask_lapic_irq,
        .unmask         = unmask_lapic_irq,
-       .eoi            = ack_apic,
+       .ack            = ack_lapic_irq,
 };
 
+static void lapic_register_intr(int irq, int vector)
+{
+       irq_desc[irq].status &= ~IRQ_LEVEL;
+       set_irq_chip_and_handler_name(irq, &lapic_chip, handle_edge_irq,
+                                     "edge");
+       set_intr_gate(vector, interrupt[irq]);
+}
+
 static void __init setup_nmi(void)
 {
        /*
-        * Dirty trick to enable the NMI watchdog ...
+        * Dirty trick to enable the NMI watchdog ...
         * We put the 8259A master into AEOI mode and
         * unmask on all local APICs LVT0 as NMI.
         *
         * The idea to use the 8259A in AEOI mode ('8259A Virtual Wire')
         * is from Maciej W. Rozycki - so we do not have to EOI from
         * the NMI handler or the timer interrupt.
-        */ 
+        */
        apic_printk(APIC_VERBOSE, KERN_INFO "activating NMI Watchdog ...");
 
        enable_NMI_through_LVT0();
@@ -2191,11 +2201,16 @@ static inline void __init unlock_ExtINT_
 static inline void __init check_timer(void)
 {
        int apic1, pin1, apic2, pin2;
+       int no_pin1 = 0;
        int vector;
+       unsigned int ver;
        unsigned long flags;
 
        local_irq_save(flags);
 
+       ver = apic_read(APIC_LVR);
+       ver = GET_APIC_VERSION(ver);
+
        /*
         * get/set the timer IRQ vector:
         */
@@ -2204,34 +2219,54 @@ static inline void __init check_timer(vo
        set_intr_gate(vector, interrupt[0]);
 
        /*
-        * Subtle, code in do_timer_interrupt() expects an AEOI
-        * mode for the 8259A whenever interrupts are routed
-        * through I/O APICs.  Also IRQ0 has to be enabled in
-        * the 8259A which implies the virtual wire has to be
-        * disabled in the local APIC.
+        * As IRQ0 is to be enabled in the 8259A, the virtual
+        * wire has to be disabled in the local APIC.  Also
+        * timer interrupts need to be acknowledged manually in
+        * the 8259A for the i82489DX when using the NMI
+        * watchdog as that APIC treats NMIs as level-triggered.
+        * The AEOI mode will finish them in the 8259A
+        * automatically.
         */
-       apic_write_around(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT);
+       apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT);
        init_8259A(1);
-       timer_ack = 1;
-       if (timer_over_8254 > 0)
-               enable_8259A_irq(0);
+       timer_ack = (nmi_watchdog == NMI_IO_APIC && !APIC_INTEGRATED(ver));
 
        pin1  = find_isa_irq_pin(0, mp_INT);
        apic1 = find_isa_irq_apic(0, mp_INT);
        pin2  = ioapic_i8259.pin;
        apic2 = ioapic_i8259.apic;
 
-       printk(KERN_INFO "..TIMER: vector=0x%02X apic1=%d pin1=%d apic2=%d pin2=%d\n",
-               vector, apic1, pin1, apic2, pin2);
+       apic_printk(APIC_QUIET, KERN_INFO "..TIMER: vector=0x%02X "
+                   "apic1=%d pin1=%d apic2=%d pin2=%d\n",
+                   vector, apic1, pin1, apic2, pin2);
+
+       /*
+        * Some BIOS writers are clueless and report the ExtINTA
+        * I/O APIC input from the cascaded 8259A as the timer
+        * interrupt input.  So just in case, if only one pin
+        * was found above, try it both directly and through the
+        * 8259A.
+        */
+       if (pin1 == -1) {
+               pin1 = pin2;
+               apic1 = apic2;
+               no_pin1 = 1;
+       } else if (pin2 == -1) {
+               pin2 = pin1;
+               apic2 = apic1;
+       }
 
        if (pin1 != -1) {
                /*
                 * Ok, does IRQ0 through the IOAPIC work?
                 */
+               if (no_pin1) {
+                       add_pin_to_irq(0, apic1, pin1);
+                       setup_timer_IRQ0_pin(apic1, pin1, vector);
+               }
                unmask_IO_APIC_irq(0);
                if (timer_irq_works()) {
                        if (nmi_watchdog == NMI_IO_APIC) {
-                               disable_8259A_irq(0);
                                setup_nmi();
                                enable_8259A_irq(0);
                        }
@@ -2240,71 +2275,77 @@ static inline void __init check_timer(vo
                        goto out;
                }
                clear_IO_APIC_pin(apic1, pin1);
-               printk(KERN_ERR "..MP-BIOS bug: 8254 timer not connected to "
-                               "IO-APIC\n");
-       }
-
-       printk(KERN_INFO "...trying to set up timer (IRQ0) through the 8259A ... ");
-       if (pin2 != -1) {
-               printk("\n..... (found pin %d) ...", pin2);
+               if (!no_pin1)
+                       apic_printk(APIC_QUIET, KERN_ERR "..MP-BIOS bug: "
+                                   "8254 timer not connected to IO-APIC\n");
+
+               apic_printk(APIC_QUIET, KERN_INFO "...trying to set up timer "
+                           "(IRQ0) through the 8259A ...\n");
+               apic_printk(APIC_QUIET, KERN_INFO
+                           "..... (found apic %d pin %d) ...\n", apic2, pin2);
                /*
                 * legacy devices should be connected to IO APIC #0
                 */
-               setup_ExtINT_IRQ0_pin(apic2, pin2, vector);
+               replace_pin_at_irq(0, apic1, pin1, apic2, pin2);
+               setup_timer_IRQ0_pin(apic2, pin2, vector);
+               unmask_IO_APIC_irq(0);
+               enable_8259A_irq(0);
                if (timer_irq_works()) {
-                       printk("works.\n");
-                       if (pin1 != -1)
-                               replace_pin_at_irq(0, apic1, pin1, apic2, pin2);
-                       else
-                               add_pin_to_irq(0, apic2, pin2);
+                       apic_printk(APIC_QUIET, KERN_INFO "....... works.\n");
+                       timer_through_8259 = 1;
                        if (nmi_watchdog == NMI_IO_APIC) {
+                               disable_8259A_irq(0);
                                setup_nmi();
+                               enable_8259A_irq(0);
                        }
                        goto out;
                }
                /*
                 * Cleanup, just in case ...
                 */
+               disable_8259A_irq(0);
                clear_IO_APIC_pin(apic2, pin2);
+               apic_printk(APIC_QUIET, KERN_INFO "....... failed.\n");
        }
-       printk(" failed.\n");
 
        if (nmi_watchdog == NMI_IO_APIC) {
-               printk(KERN_WARNING "timer doesn't work through the IO-APIC - disabling NMI Watchdog!\n");
-               nmi_watchdog = 0;
+               apic_printk(APIC_QUIET, KERN_WARNING "timer doesn't work "
+                           "through the IO-APIC - disabling NMI Watchdog!\n");
+               nmi_watchdog = NMI_NONE;
        }
+       timer_ack = 0;
 
-       printk(KERN_INFO "...trying to set up timer as Virtual Wire IRQ...");
+       apic_printk(APIC_QUIET, KERN_INFO
+                   "...trying to set up timer as Virtual Wire IRQ...\n");
 
-       disable_8259A_irq(0);
-       set_irq_chip_and_handler_name(0, &lapic_chip, handle_fasteoi_irq,
-                                     "fasteoi");
-       apic_write_around(APIC_LVT0, APIC_DM_FIXED | vector);   /* Fixed mode */
+       lapic_register_intr(0, vector);
+       apic_write(APIC_LVT0, APIC_DM_FIXED | vector);  /* Fixed mode */
        enable_8259A_irq(0);
 
        if (timer_irq_works()) {
-               printk(" works.\n");
+               apic_printk(APIC_QUIET, KERN_INFO "..... works.\n");
                goto out;
        }
-       apic_write_around(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_FIXED | vector);
-       printk(" failed.\n");
+       disable_8259A_irq(0);
+       apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_FIXED | vector);
+       apic_printk(APIC_QUIET, KERN_INFO "..... failed.\n");
 
-       printk(KERN_INFO "...trying to set up timer as ExtINT IRQ...");
+       apic_printk(APIC_QUIET, KERN_INFO
+                   "...trying to set up timer as ExtINT IRQ...\n");
 
-       timer_ack = 0;
        init_8259A(0);
        make_8259A_irq(0);
-       apic_write_around(APIC_LVT0, APIC_DM_EXTINT);
+       apic_write(APIC_LVT0, APIC_DM_EXTINT);
 
        unlock_ExtINT_logic();
 
        if (timer_irq_works()) {
-               printk(" works.\n");
+               apic_printk(APIC_QUIET, KERN_INFO "..... works.\n");
                goto out;
        }
-       printk(" failed :(.\n");
+       apic_printk(APIC_QUIET, KERN_INFO "..... failed :(.\n");
        panic("IO-APIC + timer doesn't work!  Boot with apic=debug and send a "
-               "report.  Then try booting with the 'noapic' option");
+               "report.  Then try booting with the 'noapic' option.\n");
 out:
        local_irq_restore(flags);
 }
@@ -2314,11 +2355,21 @@ int timer_uses_ioapic_pin_0 = 0;
 #endif
 
 /*
- *
- * IRQ's that are handled by the PIC in the MPS IOAPIC case.
- * - IRQ2 is the cascade IRQ, and cannot be a io-apic IRQ.
- *   Linux doesn't really care, as it's not actually used
- *   for any interrupt handling anyway.
+ * Traditionally ISA IRQ2 is the cascade IRQ, and is not available
+ * to devices.  However there may be an I/O APIC pin available for
+ * this interrupt regardless.  The pin may be left unconnected, but
+ * typically it will be reused as an ExtINT cascade interrupt for
+ * the master 8259A.  In the MPS case such a pin will normally be
+ * reported as an ExtINT interrupt in the MP table.  With ACPI
+ * there is no provision for ExtINT interrupts, and in the absence
+ * of an override it would be treated as an ordinary ISA I/O APIC
+ * interrupt, that is edge-triggered and unmasked by default.  We
+ * used to do this, but it caused problems on some systems because
+ * of the NMI watchdog and sometimes IRQ0 of the 8254 timer using
+ * the same ExtINT cascade interrupt to drive the local APIC of the
+ * bootstrap processor.  Therefore we refrain from routing IRQ2 to
+ * the I/O APIC in all cases now.  No actual device should request
+ * it anyway.  --macro
  */
 #define PIC_IRQS       (1 << PIC_CASCADE_IR)
 
@@ -2328,25 +2379,22 @@ void __init setup_IO_APIC(void)
        int i;
 
        /* Reserve all the system vectors. */
-       for (i = FIRST_SYSTEM_VECTOR; i < NR_VECTORS; i++)
+       for (i = first_system_vector; i < NR_VECTORS; i++)
                set_bit(i, used_vectors);
 #endif
 
        enable_IO_APIC();
 
-       if (acpi_ioapic)
-               io_apic_irqs = ~0;      /* all IRQs go through IOAPIC */
-       else
-               io_apic_irqs = ~PIC_IRQS;
+       io_apic_irqs = ~PIC_IRQS;
 
        printk("ENABLING IO-APIC IRQs\n");
 
+#ifndef CONFIG_XEN
        /*
         * Set up IO-APIC IRQ routing.
         */
        if (!acpi_ioapic)
                setup_ioapic_ids_from_mpc();
-#ifndef CONFIG_XEN
        sync_Arb_IDs();
 #endif
        setup_IO_APIC_irqs();
@@ -2356,28 +2404,14 @@ void __init setup_IO_APIC(void)
                print_IO_APIC();
 }
 
-static int __init setup_disable_8254_timer(char *s)
-{
-       timer_over_8254 = -1;
-       return 1;
-}
-static int __init setup_enable_8254_timer(char *s)
-{
-       timer_over_8254 = 2;
-       return 1;
-}
-
-__setup("disable_8254_timer", setup_disable_8254_timer);
-__setup("enable_8254_timer", setup_enable_8254_timer);
-
 /*
  *     Called after all the initialization is done. If we didnt find any
  *     APIC bugs then we can allow the modify fast path
  */
- 
+
 static int __init io_apic_bug_finalize(void)
 {
-       if(sis_apic_bug == -1)
+       if (sis_apic_bug == -1)
                sis_apic_bug = 0;
        if (is_initial_xendomain()) {
                struct xen_platform_op op = { .cmd = XENPF_platform_quirk };
@@ -2396,17 +2430,17 @@ struct sysfs_ioapic_data {
        struct sys_device dev;
        struct IO_APIC_route_entry entry[0];
 };
-static struct sysfs_ioapic_data * mp_ioapic_data[MAX_IO_APICS];
+static struct sysfs_ioapic_data *mp_ioapic_data[MAX_IO_APICS];
 
 static int ioapic_suspend(struct sys_device *dev, pm_message_t state)
 {
        struct IO_APIC_route_entry *entry;
        struct sysfs_ioapic_data *data;
        int i;
-       
+
        data = container_of(dev, struct sysfs_ioapic_data, dev);
        entry = data->entry;
-       for (i = 0; i < nr_ioapic_registers[dev->id]; i ++)
+       for (i = 0; i < nr_ioapic_registers[dev->id]; i++)
                entry[i] = ioapic_read_entry(dev->id, i);
 
        return 0;
@@ -2419,18 +2453,18 @@ static int ioapic_resume(struct sys_devi
        unsigned long flags;
        union IO_APIC_reg_00 reg_00;
        int i;
-       
+
        data = container_of(dev, struct sysfs_ioapic_data, dev);
        entry = data->entry;
 
        spin_lock_irqsave(&ioapic_lock, flags);
        reg_00.raw = io_apic_read(dev->id, 0);
-       if (reg_00.bits.ID != mp_ioapics[dev->id].mpc_apicid) {
-               reg_00.bits.ID = mp_ioapics[dev->id].mpc_apicid;
+       if (reg_00.bits.ID != mp_ioapics[dev->id].mp_apicid) {
+               reg_00.bits.ID = mp_ioapics[dev->id].mp_apicid;
                io_apic_write(dev->id, 0, reg_00.raw);
        }
        spin_unlock_irqrestore(&ioapic_lock, flags);
-       for (i = 0; i < nr_ioapic_registers[dev->id]; i ++)
+       for (i = 0; i < nr_ioapic_registers[dev->id]; i++)
                ioapic_write_entry(dev->id, i, entry[i]);
 
        return 0;
@@ -2444,24 +2478,23 @@ static struct sysdev_class ioapic_sysdev
 
 static int __init ioapic_init_sysfs(void)
 {
-       struct sys_device * dev;
+       struct sys_device *dev;
        int i, size, error = 0;
 
        error = sysdev_class_register(&ioapic_sysdev_class);
        if (error)
                return error;
 
-       for (i = 0; i < nr_ioapics; i++ ) {
-               size = sizeof(struct sys_device) + nr_ioapic_registers[i] 
+       for (i = 0; i < nr_ioapics; i++) {
+               size = sizeof(struct sys_device) + nr_ioapic_registers[i]
                        * sizeof(struct IO_APIC_route_entry);
-               mp_ioapic_data[i] = kmalloc(size, GFP_KERNEL);
+               mp_ioapic_data[i] = kzalloc(size, GFP_KERNEL);
                if (!mp_ioapic_data[i]) {
                        printk(KERN_ERR "Can't suspend/resume IOAPIC %d\n", i);
                        continue;
                }
-               memset(mp_ioapic_data[i], 0, size);
                dev = &mp_ioapic_data[i]->dev;
-               dev->id = i; 
+               dev->id = i;
                dev->cls = &ioapic_sysdev_class;
                error = sysdev_register(dev);
                if (error) {
@@ -2538,7 +2571,7 @@ static int msi_compose_msg(struct pci_de
                msg->address_lo =
                        MSI_ADDR_BASE_LO |
                        ((INT_DEST_MODE == 0) ?
-                               MSI_ADDR_DEST_MODE_PHYSICAL:
+MSI_ADDR_DEST_MODE_PHYSICAL:
                                MSI_ADDR_DEST_MODE_LOGICAL) |
                        ((INT_DELIVERY_MODE != dest_LowestPrio) ?
                                MSI_ADDR_REDIRECTION_CPU:
@@ -2549,7 +2582,7 @@ static int msi_compose_msg(struct pci_de
                        MSI_DATA_TRIGGER_EDGE |
                        MSI_DATA_LEVEL_ASSERT |
                        ((INT_DELIVERY_MODE != dest_LowestPrio) ?
-                               MSI_DATA_DELIVERY_FIXED:
+MSI_DATA_DELIVERY_FIXED:
                                MSI_DATA_DELIVERY_LOWPRI) |
                        MSI_DATA_VECTOR(vector);
        }
@@ -2720,12 +2753,12 @@ int arch_setup_ht_irq(unsigned int irq, 
 #endif /* CONFIG_HT_IRQ */
 
 /* --------------------------------------------------------------------------
-                          ACPI-based IOAPIC Configuration
+                       ACPI-based IOAPIC Configuration
    -------------------------------------------------------------------------- */
 
 #ifdef CONFIG_ACPI
 
-int __init io_apic_get_unique_id (int ioapic, int apic_id)
+int __init io_apic_get_unique_id(int ioapic, int apic_id)
 {
 #ifndef CONFIG_XEN
        union IO_APIC_reg_00 reg_00;
@@ -2735,10 +2768,10 @@ int __init io_apic_get_unique_id (int io
        int i = 0;
 
        /*
-        * The P4 platform supports up to 256 APIC IDs on two separate APIC 
-        * buses (one for LAPICs, one for IOAPICs), where predecessors only 
+        * The P4 platform supports up to 256 APIC IDs on two separate APIC
+        * buses (one for LAPICs, one for IOAPICs), where predecessors only
         * supports up to 16 on one shared APIC bus.
-        * 
+        *
         * TBD: Expand LAPIC/IOAPIC support on P4-class systems to take full
         *      advantage of new APIC bus architecture.
         */
@@ -2757,7 +2790,7 @@ int __init io_apic_get_unique_id (int io
        }
 
        /*
-        * Every APIC in a system must have a unique ID or we get lots of nice 
+        * Every APIC in a system must have a unique ID or we get lots of nice
         * 'stuck on smp_invalidate_needed IPI wait' messages.
         */
        if (check_apicid_used(apic_id_map, apic_id)) {
@@ -2774,7 +2807,7 @@ int __init io_apic_get_unique_id (int io
                        "trying %d\n", ioapic, apic_id, i);
 
                apic_id = i;
-       } 
+       }
 
        tmp = apicid_to_cpu_present(apic_id);
        physids_or(apic_id_map, apic_id_map, tmp);
@@ -2802,7 +2835,7 @@ int __init io_apic_get_unique_id (int io
 }
 
 
-int __init io_apic_get_version (int ioapic)
+int __init io_apic_get_version(int ioapic)
 {
        union IO_APIC_reg_01    reg_01;
        unsigned long flags;
@@ -2815,7 +2848,7 @@ int __init io_apic_get_version (int ioap
 }
 
 
-int __init io_apic_get_redir_entries (int ioapic)
+int __init io_apic_get_redir_entries(int ioapic)
 {
        union IO_APIC_reg_01    reg_01;
        unsigned long flags;
@@ -2828,7 +2861,7 @@ int __init io_apic_get_redir_entries (in
 }
 
 
-int io_apic_set_pci_routing (int ioapic, int pin, int irq, int edge_level, int active_high_low)
+int io_apic_set_pci_routing(int ioapic, int pin, int irq, int edge_level, int active_high_low)
 {
        struct IO_APIC_route_entry entry;
 
@@ -2844,7 +2877,7 @@ int io_apic_set_pci_routing (int ioapic,
         * corresponding device driver registers for this IRQ.
         */
 
-       memset(&entry,0,sizeof(entry));
+       memset(&entry, 0, sizeof(entry));
 
        entry.delivery_mode = INT_DELIVERY_MODE;
        entry.dest_mode = INT_DEST_MODE;
@@ -2863,7 +2896,7 @@ int io_apic_set_pci_routing (int ioapic,
 
        apic_printk(APIC_DEBUG, KERN_DEBUG "IOAPIC[%d]: Set PCI routing entry "
                "(%d-%d -> 0x%x -> IRQ %d Mode:%i Active:%i)\n", ioapic,
-               mp_ioapics[ioapic].mpc_apicid, pin, entry.vector, irq,
+               mp_ioapics[ioapic].mp_apicid, pin, entry.vector, irq,
                edge_level, active_high_low);
 
        ioapic_register_intr(irq, entry.vector, edge_level);
@@ -2884,8 +2917,8 @@ int acpi_get_override_irq(int bus_irq, i
                return -1;
 
        for (i = 0; i < mp_irq_entries; i++)
-               if (mp_irqs[i].mpc_irqtype == mp_INT &&
-                   mp_irqs[i].mpc_srcbusirq == bus_irq)
+               if (mp_irqs[i].mp_irqtype == mp_INT &&
+                   mp_irqs[i].mp_srcbusirq == bus_irq)
                        break;
        if (i >= mp_irq_entries)
                return -1;
@@ -2918,3 +2951,35 @@ static int __init parse_noapic(char *arg
        return 0;
 }
 early_param("noapic", parse_noapic);
+
+#ifndef CONFIG_XEN
+void __init ioapic_init_mappings(void)
+{
+       unsigned long ioapic_phys, idx = FIX_IO_APIC_BASE_0;
+       int i;
+
+       for (i = 0; i < nr_ioapics; i++) {
+               if (smp_found_config) {
+                       ioapic_phys = mp_ioapics[i].mp_apicaddr;
+                       if (!ioapic_phys) {
+                               printk(KERN_ERR
+                                      "WARNING: bogus zero IO-APIC "
+                                      "address found in MPTABLE, "
+                                      "disabling IO/APIC support!\n");
+                               smp_found_config = 0;
+                               skip_ioapic_setup = 1;
+                               goto fake_ioapic_page;
+                       }
+               } else {
+fake_ioapic_page:
+                       ioapic_phys = (unsigned long)
+                                     alloc_bootmem_pages(PAGE_SIZE);
+                       ioapic_phys = __pa(ioapic_phys);
+               }
+               set_fixmap_nocache(idx, ioapic_phys);
+               printk(KERN_DEBUG "mapped IOAPIC to %08lx (%08lx)\n",
+                      __fix_to_virt(idx), ioapic_phys);
+               idx++;
+       }
+}
+#endif
Index: head-2008-12-01/arch/x86/kernel/io_apic_64-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/io_apic_64-xen.c       2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/io_apic_64-xen.c    2008-12-01 11:49:07.000000000 +0100
@@ -45,6 +45,7 @@
 #include <asm/proto.h>
 #include <asm/acpi.h>
 #include <asm/dma.h>
+#include <asm/i8259.h>
 #include <asm/nmi.h>
 #include <asm/msidef.h>
 #include <asm/hypertransport.h>
@@ -63,10 +64,16 @@ struct irq_cfg {
 };
 
 /* irq_cfg is indexed by the sum of all RTEs in all I/O APICs. */
-struct irq_cfg irq_cfg[NR_IRQS] __read_mostly;
+static struct irq_cfg irq_cfg[NR_IRQS] __read_mostly;
 
 static int assign_irq_vector(int irq, cpumask_t mask);
 
+#ifndef CONFIG_XEN
+int first_system_vector = 0xfe;
+
+char system_vectors[NR_VECTORS] = { [0 ... NR_VECTORS-1] = SYS_VECTOR_FREE};
+#endif
+
 #define __apicdebuginit  __init
 
 int sis_apic_bug; /* not actually supported, dummy for compile */
@@ -89,14 +96,14 @@ unsigned long io_apic_irqs;
 
 #define clear_IO_APIC() ((void)0)
 #else
-int timer_over_8254 __initdata = 1;
+int timer_through_8259 __initdata;
 
 /* Where if anywhere is the i8259 connect in external int mode */
 static struct { int pin, apic; } ioapic_i8259 = { -1, -1 };
 #endif
 
 static DEFINE_SPINLOCK(ioapic_lock);
-DEFINE_SPINLOCK(vector_lock);
+static DEFINE_SPINLOCK(vector_lock);
 
 /*
  * # of IRQ routing registers
@@ -104,15 +111,17 @@ DEFINE_SPINLOCK(vector_lock);
 int nr_ioapic_registers[MAX_IO_APICS];
 
 /* I/O APIC entries */
-struct mpc_config_ioapic mp_ioapics[MAX_IO_APICS];
+struct mp_config_ioapic mp_ioapics[MAX_IO_APICS];
 int nr_ioapics;
 
 /* MP IRQ source entries */
-struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES];
+struct mp_config_intsrc mp_irqs[MAX_IRQ_SOURCES];
 
 /* # of MP IRQ source entries */
 int mp_irq_entries;
 
+DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES);
+
 /*
  * Rough estimation of how many shared IRQs there are, can
  * be changed anytime.
@@ -141,7 +150,7 @@ struct io_apic {
 static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx)
 {
        return (void __iomem *) __fix_to_virt(FIX_IO_APIC_BASE_0 + idx)
-               + (mp_ioapics[idx].mpc_apicaddr & ~PAGE_MASK);
+               + (mp_ioapics[idx].mp_apicaddr & ~PAGE_MASK);
 }
 #endif
 
@@ -155,7 +164,7 @@ static inline unsigned int io_apic_read(
        struct physdev_apic apic_op;
        int ret;
 
-       apic_op.apic_physbase = mp_ioapics[apic].mpc_apicaddr;
+       apic_op.apic_physbase = mp_ioapics[apic].mp_apicaddr;
        apic_op.reg = reg;
        ret = HYPERVISOR_physdev_op(PHYSDEVOP_apic_read, &apic_op);
        if (ret)
@@ -173,7 +182,7 @@ static inline void io_apic_write(unsigne
 #else
        struct physdev_apic apic_op;
 
-       apic_op.apic_physbase = mp_ioapics[apic].mpc_apicaddr;
+       apic_op.apic_physbase = mp_ioapics[apic].mp_apicaddr;
        apic_op.reg = reg;
        apic_op.value = value;
        WARN_ON(HYPERVISOR_physdev_op(PHYSDEVOP_apic_write, &apic_op));
@@ -209,7 +218,7 @@ static bool io_apic_level_ack_pending(un
                        break;
                reg = io_apic_read(entry->apic, 0x10 + pin*2);
                /* Is the remote IRR bit set? */
-               if ((reg >> 14) & 1) {
+               if (reg & IO_APIC_REDIR_REMOTE_IRR) {
                        spin_unlock_irqrestore(&ioapic_lock, flags);
                        return true;
                }
@@ -308,7 +317,7 @@ static void __target_IO_APIC_irq(unsigne
                        break;
                io_apic_write(apic, 0x11 + pin*2, dest);
                reg = io_apic_read(apic, 0x10 + pin*2);
-               reg &= ~0x000000ff;
+               reg &= ~IO_APIC_REDIR_VECTOR_MASK;
                reg |= vector;
                io_apic_modify(apic, reg);
                if (!entry->next)
@@ -372,6 +381,26 @@ static void add_pin_to_irq(unsigned int 
 }
 
 #ifndef CONFIG_XEN
+/*
+ * Reroute an IRQ to a different pin.
+ */
+static void __init replace_pin_at_irq(unsigned int irq,
+                                     int oldapic, int oldpin,
+                                     int newapic, int newpin)
+{
+       struct irq_pin_list *entry = irq_2_pin + irq;
+
+       while (1) {
+               if (entry->apic == oldapic && entry->pin == oldpin) {
+                       entry->apic = newapic;
+                       entry->pin = newpin;
+               }
+               if (!entry->next)
+                       break;
+               entry = irq_2_pin + entry->next;
+       }
+}
+
 #define __DO_ACTION(R, ACTION, FINAL)                                  \
                                                                        \
 {                                                                      \
@@ -399,10 +428,11 @@ static void add_pin_to_irq(unsigned int 
        static void name##_IO_APIC_irq (unsigned int irq)               \
        __DO_ACTION(R, ACTION, FINAL)
 
-DO_ACTION( __mask,             0, |= 0x00010000, io_apic_sync(entry->apic) )
-                                               /* mask = 1 */
-DO_ACTION( __unmask,           0, &= 0xfffeffff, )
-                                               /* mask = 0 */
+/* mask = 1 */
+DO_ACTION(__mask,      0, |= IO_APIC_REDIR_MASKED, io_apic_sync(entry->apic))
+
+/* mask = 0 */
+DO_ACTION(__unmask,    0, &= ~IO_APIC_REDIR_MASKED, )
 
 static void mask_IO_APIC_irq (unsigned int irq)
 {
@@ -465,22 +495,6 @@ static int __init disable_timer_pin_setu
 }
 __setup("disable_timer_pin_1", disable_timer_pin_setup);
 
-#ifndef CONFIG_XEN
-static int __init setup_disable_8254_timer(char *s)
-{
-       timer_over_8254 = -1;
-       return 1;
-}
-static int __init setup_enable_8254_timer(char *s)
-{
-       timer_over_8254 = 2;
-       return 1;
-}
-
-__setup("disable_8254_timer", setup_disable_8254_timer);
-__setup("enable_8254_timer", setup_enable_8254_timer);
-#endif /* !CONFIG_XEN */
-
 
 /*
  * Find the IRQ entry number of a certain pin.
@@ -490,10 +504,10 @@ static int find_irq_entry(int apic, int 
        int i;
 
        for (i = 0; i < mp_irq_entries; i++)
-               if (mp_irqs[i].mpc_irqtype == type &&
-                   (mp_irqs[i].mpc_dstapic == mp_ioapics[apic].mpc_apicid ||
-                    mp_irqs[i].mpc_dstapic == MP_APIC_ALL) &&
-                   mp_irqs[i].mpc_dstirq == pin)
+               if (mp_irqs[i].mp_irqtype == type &&
+                   (mp_irqs[i].mp_dstapic == mp_ioapics[apic].mp_apicid ||
+                    mp_irqs[i].mp_dstapic == MP_APIC_ALL) &&
+                   mp_irqs[i].mp_dstirq == pin)
                        return i;
 
        return -1;
@@ -508,13 +522,13 @@ static int __init find_isa_irq_pin(int i
        int i;
 
        for (i = 0; i < mp_irq_entries; i++) {
-               int lbus = mp_irqs[i].mpc_srcbus;
+               int lbus = mp_irqs[i].mp_srcbus;
 
                if (test_bit(lbus, mp_bus_not_pci) &&
-                   (mp_irqs[i].mpc_irqtype == type) &&
-                   (mp_irqs[i].mpc_srcbusirq == irq))
+                   (mp_irqs[i].mp_irqtype == type) &&
+                   (mp_irqs[i].mp_srcbusirq == irq))
 
-                       return mp_irqs[i].mpc_dstirq;
+                       return mp_irqs[i].mp_dstirq;
        }
        return -1;
 }
@@ -524,17 +538,17 @@ static int __init find_isa_irq_apic(int 
        int i;
 
        for (i = 0; i < mp_irq_entries; i++) {
-               int lbus = mp_irqs[i].mpc_srcbus;
+               int lbus = mp_irqs[i].mp_srcbus;
 
                if (test_bit(lbus, mp_bus_not_pci) &&
-                   (mp_irqs[i].mpc_irqtype == type) &&
-                   (mp_irqs[i].mpc_srcbusirq == irq))
+                   (mp_irqs[i].mp_irqtype == type) &&
+                   (mp_irqs[i].mp_srcbusirq == irq))
                        break;
        }
        if (i < mp_irq_entries) {
                int apic;
                for(apic = 0; apic < nr_ioapics; apic++) {
-                       if (mp_ioapics[apic].mpc_apicid == mp_irqs[i].mpc_dstapic)
+                       if (mp_ioapics[apic].mp_apicid == mp_irqs[i].mp_dstapic)
                                return apic;
                }
        }
@@ -555,28 +569,28 @@ int IO_APIC_get_PCI_irq_vector(int bus, 
 
        apic_printk(APIC_DEBUG, "querying PCI -> IRQ mapping bus:%d, slot:%d, pin:%d.\n",
                bus, slot, pin);
-       if (mp_bus_id_to_pci_bus[bus] == -1) {
+       if (test_bit(bus, mp_bus_not_pci)) {
                apic_printk(APIC_VERBOSE, "PCI BIOS passed nonexistent PCI bus %d!\n", bus);
                return -1;
        }
        for (i = 0; i < mp_irq_entries; i++) {
-               int lbus = mp_irqs[i].mpc_srcbus;
+               int lbus = mp_irqs[i].mp_srcbus;
 
                for (apic = 0; apic < nr_ioapics; apic++)
-                       if (mp_ioapics[apic].mpc_apicid == mp_irqs[i].mpc_dstapic ||
-                           mp_irqs[i].mpc_dstapic == MP_APIC_ALL)
+                       if (mp_ioapics[apic].mp_apicid == mp_irqs[i].mp_dstapic ||
+                           mp_irqs[i].mp_dstapic == MP_APIC_ALL)
                                break;
 
                if (!test_bit(lbus, mp_bus_not_pci) &&
-                   !mp_irqs[i].mpc_irqtype &&
+                   !mp_irqs[i].mp_irqtype &&
                    (bus == lbus) &&
-                   (slot == ((mp_irqs[i].mpc_srcbusirq >> 2) & 0x1f))) {
-                       int irq = pin_2_irq(i,apic,mp_irqs[i].mpc_dstirq);
+                   (slot == ((mp_irqs[i].mp_srcbusirq >> 2) & 0x1f))) {
+                       int irq = pin_2_irq(i,apic,mp_irqs[i].mp_dstirq);
 
                        if (!(apic || IO_APIC_IRQ(irq)))
                                continue;
 
-                       if (pin == (mp_irqs[i].mpc_srcbusirq & 3))
+                       if (pin == (mp_irqs[i].mp_srcbusirq & 3))
                                return irq;
                        /*
                         * Use the first all-but-pin matching entry as a
@@ -604,13 +618,13 @@ int IO_APIC_get_PCI_irq_vector(int bus, 
 
 static int MPBIOS_polarity(int idx)
 {
-       int bus = mp_irqs[idx].mpc_srcbus;
+       int bus = mp_irqs[idx].mp_srcbus;
        int polarity;
 
        /*
         * Determine IRQ line polarity (high active or low active):
         */
-       switch (mp_irqs[idx].mpc_irqflag & 3)
+       switch (mp_irqs[idx].mp_irqflag & 3)
        {
                case 0: /* conforms, ie. bus-type dependent polarity */
                        if (test_bit(bus, mp_bus_not_pci))
@@ -646,13 +660,13 @@ static int MPBIOS_polarity(int idx)
 
 static int MPBIOS_trigger(int idx)
 {
-       int bus = mp_irqs[idx].mpc_srcbus;
+       int bus = mp_irqs[idx].mp_srcbus;
        int trigger;
 
        /*
         * Determine IRQ trigger mode (edge or level sensitive):
         */
-       switch ((mp_irqs[idx].mpc_irqflag>>2) & 3)
+       switch ((mp_irqs[idx].mp_irqflag>>2) & 3)
        {
                case 0: /* conforms, ie. bus-type dependent */
                        if (test_bit(bus, mp_bus_not_pci))
@@ -699,16 +713,16 @@ static inline int irq_trigger(int idx)
 static int pin_2_irq(int idx, int apic, int pin)
 {
        int irq, i;
-       int bus = mp_irqs[idx].mpc_srcbus;
+       int bus = mp_irqs[idx].mp_srcbus;
 
        /*
         * Debugging check, we are in big trouble if this message pops up!
         */
-       if (mp_irqs[idx].mpc_dstirq != pin)
+       if (mp_irqs[idx].mp_dstirq != pin)
                printk(KERN_ERR "broken BIOS or MPTABLE parser, ayiee!!\n");
 
        if (test_bit(bus, mp_bus_not_pci)) {
-               irq = mp_irqs[idx].mpc_srcbusirq;
+               irq = mp_irqs[idx].mp_srcbusirq;
        } else {
                /*
                 * PCI IRQs are mapped in order
@@ -722,6 +736,19 @@ static int pin_2_irq(int idx, int apic, 
        return irq;
 }
 
+void lock_vector_lock(void)
+{
+       /* Used to the online set of cpus does not change
+        * during assign_irq_vector.
+        */
+       spin_lock(&vector_lock);
+}
+
+void unlock_vector_lock(void)
+{
+       spin_unlock(&vector_lock);
+}
+
 static int __assign_irq_vector(int irq, cpumask_t mask)
 {
        struct physdev_irq irq_op;
@@ -773,7 +800,7 @@ static void __clear_irq_vector(int irq)
 
        vector = cfg->vector;
        cpus_and(mask, cfg->domain, cpu_online_map);
-       for_each_cpu_mask(cpu, mask)
+       for_each_cpu_mask_nr(cpu, mask)
                per_cpu(vector_irq, cpu)[vector] = -1;
 
        cfg->vector = 0;
@@ -842,7 +869,7 @@ static void setup_IO_APIC_irq(int apic, 
        apic_printk(APIC_VERBOSE,KERN_DEBUG
                    "IOAPIC[%d]: Set routing entry (%d-%d -> 0x%x -> "
                    "IRQ %d Mode:%i Active:%i)\n",
-                   apic, mp_ioapics[apic].mpc_apicid, pin, cfg->vector,
+                   apic, mp_ioapics[apic].mp_apicid, pin, cfg->vector,
                    irq, trigger, polarity);
 
        /*
@@ -883,10 +910,10 @@ static void __init setup_IO_APIC_irqs(vo
                idx = find_irq_entry(apic,pin,mp_INT);
                if (idx == -1) {
                        if (first_notcon) {
-                               apic_printk(APIC_VERBOSE, KERN_DEBUG " IO-APIC (apicid-pin) %d-%d", mp_ioapics[apic].mpc_apicid, pin);
+                               apic_printk(APIC_VERBOSE, KERN_DEBUG " IO-APIC (apicid-pin) %d-%d", mp_ioapics[apic].mp_apicid, pin);
                                first_notcon = 0;
                        } else
-                               apic_printk(APIC_VERBOSE, ", %d-%d", mp_ioapics[apic].mpc_apicid, pin);
+                               apic_printk(APIC_VERBOSE, ", %d-%d", mp_ioapics[apic].mp_apicid, pin);
                        continue;
                }
                if (!first_notcon) {
@@ -908,26 +935,21 @@ static void __init setup_IO_APIC_irqs(vo
 
 #ifndef CONFIG_XEN
 /*
- * Set up the 8259A-master output pin as broadcast to all
- * CPUs.
+ * Set up the timer pin, possibly with the 8259A-master behind.
  */
-static void __init setup_ExtINT_IRQ0_pin(unsigned int apic, unsigned int pin, int vector)
+static void __init setup_timer_IRQ0_pin(unsigned int apic, unsigned int pin,
+                                       int vector)
 {
        struct IO_APIC_route_entry entry;
 
        memset(&entry, 0, sizeof(entry));
 
-       disable_8259A_irq(0);
-
-       /* mask LVT0 */
-       apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT);
-
        /*
         * We use logical delivery to get the timer IRQ
         * to the first CPU.
         */
        entry.dest_mode = INT_DEST_MODE;
-       entry.mask = 0;                                 /* unmask IRQ now */
+       entry.mask = 1;                                 /* mask IRQ now */
        entry.dest = cpu_mask_to_apicid(TARGET_CPUS);
        entry.delivery_mode = INT_DELIVERY_MODE;
        entry.polarity = 0;
@@ -936,7 +958,7 @@ static void __init setup_ExtINT_IRQ0_pin
 
        /*
         * The timer IRQ doesn't have to know that behind the
-        * scene we have a 8259A-master in AEOI mode ...
+        * scene we may have a 8259A-master in AEOI mode ...
         */
        set_irq_chip_and_handler_name(0, &ioapic_chip, handle_edge_irq, "edge");
 
@@ -944,8 +966,6 @@ static void __init setup_ExtINT_IRQ0_pin
         * Add it to the IO-APIC irq-routing table:
         */
        ioapic_write_entry(apic, pin, entry);
-
-       enable_8259A_irq(0);
 }
 
 void __apicdebuginit print_IO_APIC(void)
@@ -962,7 +982,7 @@ void __apicdebuginit print_IO_APIC(void)
        printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries);
        for (i = 0; i < nr_ioapics; i++)
                printk(KERN_DEBUG "number of IO-APIC #%d registers: %d.\n",
-                      mp_ioapics[i].mpc_apicid, nr_ioapic_registers[i]);
+                      mp_ioapics[i].mp_apicid, nr_ioapic_registers[i]);
 
        /*
         * We are a bit conservative about what we expect.  We have to
@@ -980,7 +1000,7 @@ void __apicdebuginit print_IO_APIC(void)
        spin_unlock_irqrestore(&ioapic_lock, flags);
 
        printk("\n");
-       printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].mpc_apicid);
+       printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].mp_apicid);
        printk(KERN_DEBUG ".... register #00: %08X\n", reg_00.raw);
        printk(KERN_DEBUG ".......    : physical APIC id: %02X\n", reg_00.bits.ID);
 
@@ -1072,6 +1092,7 @@ void __apicdebuginit print_local_APIC(vo
 
        printk("\n" KERN_DEBUG "printing local APIC contents on CPU#%d/%d:\n",
                smp_processor_id(), hard_smp_processor_id());
+       v = apic_read(APIC_ID);
        printk(KERN_INFO "... APIC ID:      %08x (%01x)\n", v, GET_APIC_ID(read_apic_id()));
        v = apic_read(APIC_LVR);
        printk(KERN_INFO "... APIC VERSION: %08x\n", v);
@@ -1141,7 +1162,7 @@ void __apicdebuginit print_local_APIC(vo
 
 void print_all_local_APICs (void)
 {
-       on_each_cpu(print_local_APIC, NULL, 1, 1);
+       on_each_cpu(print_local_APIC, NULL, 1);
 }
 
 void __apicdebuginit print_PIC(void)
@@ -1175,6 +1196,8 @@ void __apicdebuginit print_PIC(void)
        v = inb(0x4d1) << 8 | inb(0x4d0);
        printk(KERN_DEBUG "... PIC ELCR: %04x\n", v);
 }
+#else
+void __apicdebuginit print_IO_APIC(void) {}
 #endif /* !CONFIG_XEN */
 
 void __init enable_IO_APIC(void)
@@ -1359,12 +1382,10 @@ static unsigned int startup_ioapic_irq(u
 static int ioapic_retrigger_irq(unsigned int irq)
 {
        struct irq_cfg *cfg = &irq_cfg[irq];
-       cpumask_t mask;
        unsigned long flags;
 
        spin_lock_irqsave(&vector_lock, flags);
-       mask = cpumask_of_cpu(first_cpu(cfg->domain));
-       send_IPI_mask(mask, cfg->vector);
+       send_IPI_mask(cpumask_of_cpu(first_cpu(cfg->domain)), cfg->vector);
        spin_unlock_irqrestore(&vector_lock, flags);
 
        return 1;
@@ -1545,7 +1566,7 @@ static inline void init_IO_APIC_traps(vo
 }
 
 #ifndef CONFIG_XEN
-static void enable_lapic_irq (unsigned int irq)
+static void unmask_lapic_irq(unsigned int irq)
 {
        unsigned long v;
 
@@ -1553,7 +1574,7 @@ static void enable_lapic_irq (unsigned i
        apic_write(APIC_LVT0, v & ~APIC_LVT_MASKED);
 }
 
-static void disable_lapic_irq (unsigned int irq)
+static void mask_lapic_irq(unsigned int irq)
 {
        unsigned long v;
 
@@ -1566,19 +1587,20 @@ static void ack_lapic_irq (unsigned int 
        ack_APIC_irq();
 }
 
-static void end_lapic_irq (unsigned int i) { /* nothing */ }
-
-static struct hw_interrupt_type lapic_irq_type __read_mostly = {
-       .name = "local-APIC",
-       .typename = "local-APIC-edge",
-       .startup = NULL, /* startup_irq() not used for IRQ0 */
-       .shutdown = NULL, /* shutdown_irq() not used for IRQ0 */
-       .enable = enable_lapic_irq,
-       .disable = disable_lapic_irq,
-       .ack = ack_lapic_irq,
-       .end = end_lapic_irq,
+static struct irq_chip lapic_chip __read_mostly = {
+       .name           = "local-APIC",
+       .mask           = mask_lapic_irq,
+       .unmask         = unmask_lapic_irq,
+       .ack            = ack_lapic_irq,
 };
 
+static void lapic_register_intr(int irq)
+{
+       irq_desc[irq].status &= ~IRQ_LEVEL;
+       set_irq_chip_and_handler_name(irq, &lapic_chip, handle_edge_irq,
+                                     "edge");
+}
+
 static void __init setup_nmi(void)
 {
        /*
@@ -1664,6 +1686,7 @@ static inline void __init check_timer(vo
        struct irq_cfg *cfg = irq_cfg + 0;
        int apic1, pin1, apic2, pin2;
        unsigned long flags;
+       int no_pin1 = 0;
 
        local_irq_save(flags);
 
@@ -1674,34 +1697,48 @@ static inline void __init check_timer(vo
        assign_irq_vector(0, TARGET_CPUS);
 
        /*
-        * Subtle, code in do_timer_interrupt() expects an AEOI
-        * mode for the 8259A whenever interrupts are routed
-        * through I/O APICs.  Also IRQ0 has to be enabled in
-        * the 8259A which implies the virtual wire has to be
-        * disabled in the local APIC.
+        * As IRQ0 is to be enabled in the 8259A, the virtual
+        * wire has to be disabled in the local APIC.
         */
        apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT);
        init_8259A(1);
-       if (timer_over_8254 > 0)
-               enable_8259A_irq(0);
 
        pin1  = find_isa_irq_pin(0, mp_INT);
        apic1 = find_isa_irq_apic(0, mp_INT);
        pin2  = ioapic_i8259.pin;
        apic2 = ioapic_i8259.apic;
 
-       apic_printk(APIC_VERBOSE,KERN_INFO "..TIMER: vector=0x%02X apic1=%d pin1=%d apic2=%d pin2=%d\n",
-               cfg->vector, apic1, pin1, apic2, pin2);
+       apic_printk(APIC_QUIET, KERN_INFO "..TIMER: vector=0x%02X "
+                   "apic1=%d pin1=%d apic2=%d pin2=%d\n",
+                   cfg->vector, apic1, pin1, apic2, pin2);
+
+       /*
+        * Some BIOS writers are clueless and report the ExtINTA
+        * I/O APIC input from the cascaded 8259A as the timer
+        * interrupt input.  So just in case, if only one pin
+        * was found above, try it both directly and through the
+        * 8259A.
+        */
+       if (pin1 == -1) {
+               pin1 = pin2;
+               apic1 = apic2;
+               no_pin1 = 1;
+       } else if (pin2 == -1) {
+               pin2 = pin1;
+               apic2 = apic1;
+       }
 
        if (pin1 != -1) {
                /*
                 * Ok, does IRQ0 through the IOAPIC work?
                 */
+               if (no_pin1) {
+                       add_pin_to_irq(0, apic1, pin1);
+                       setup_timer_IRQ0_pin(apic1, pin1, cfg->vector);
+               }
                unmask_IO_APIC_irq(0);
                if (!no_timer_check && timer_irq_works()) {
-                       nmi_watchdog_default();
                        if (nmi_watchdog == NMI_IO_APIC) {
-                               disable_8259A_irq(0);
                                setup_nmi();
                                enable_8259A_irq(0);
                        }
@@ -1710,54 +1747,62 @@ static inline void __init check_timer(vo
                        goto out;
                }
                clear_IO_APIC_pin(apic1, pin1);
-               apic_printk(APIC_QUIET,KERN_ERR "..MP-BIOS bug: 8254 timer not "
-                               "connected to IO-APIC\n");
-       }
-
-       apic_printk(APIC_VERBOSE,KERN_INFO "...trying to set up timer (IRQ0) "
-                               "through the 8259A ... ");
-       if (pin2 != -1) {
-               apic_printk(APIC_VERBOSE,"\n..... (found apic %d pin %d) ...",
-                       apic2, pin2);
+               if (!no_pin1)
+                       apic_printk(APIC_QUIET, KERN_ERR "..MP-BIOS bug: "
+                                   "8254 timer not connected to IO-APIC\n");
+
+               apic_printk(APIC_QUIET, KERN_INFO "...trying to set up timer "
+                           "(IRQ0) through the 8259A ...\n");
+               apic_printk(APIC_QUIET, KERN_INFO
+                           "..... (found apic %d pin %d) ...\n", apic2, pin2);
                /*
                 * legacy devices should be connected to IO APIC #0
                 */
-               setup_ExtINT_IRQ0_pin(apic2, pin2, cfg->vector);
+               replace_pin_at_irq(0, apic1, pin1, apic2, pin2);
+               setup_timer_IRQ0_pin(apic2, pin2, cfg->vector);
+               unmask_IO_APIC_irq(0);
+               enable_8259A_irq(0);
                if (timer_irq_works()) {
-                       apic_printk(APIC_VERBOSE," works.\n");
-                       nmi_watchdog_default();
+                       apic_printk(APIC_QUIET, KERN_INFO "....... works.\n");
+                       timer_through_8259 = 1;
                        if (nmi_watchdog == NMI_IO_APIC) {
+                               disable_8259A_irq(0);
                                setup_nmi();
+                               enable_8259A_irq(0);
                        }
                        goto out;
                }
                /*
                 * Cleanup, just in case ...
                 */
+               disable_8259A_irq(0);
                clear_IO_APIC_pin(apic2, pin2);
+               apic_printk(APIC_QUIET, KERN_INFO "....... failed.\n");
        }
-       apic_printk(APIC_VERBOSE," failed.\n");
 
        if (nmi_watchdog == NMI_IO_APIC) {
-               printk(KERN_WARNING "timer doesn't work through the IO-APIC - disabling NMI Watchdog!\n");
-               nmi_watchdog = 0;
+               apic_printk(APIC_QUIET, KERN_WARNING "timer doesn't work "
+                           "through the IO-APIC - disabling NMI Watchdog!\n");
+               nmi_watchdog = NMI_NONE;
        }
 
-       apic_printk(APIC_VERBOSE, KERN_INFO "...trying to set up timer as Virtual Wire IRQ...");
+       apic_printk(APIC_QUIET, KERN_INFO
+                   "...trying to set up timer as Virtual Wire IRQ...\n");
 
-       disable_8259A_irq(0);
-       irq_desc[0].chip = &lapic_irq_type;
+       lapic_register_intr(0);
        apic_write(APIC_LVT0, APIC_DM_FIXED | cfg->vector);     /* Fixed mode */
        enable_8259A_irq(0);
 
        if (timer_irq_works()) {
-               apic_printk(APIC_VERBOSE," works.\n");
+               apic_printk(APIC_QUIET, KERN_INFO "..... works.\n");
                goto out;
        }
+       disable_8259A_irq(0);
        apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_FIXED | cfg->vector);
-       apic_printk(APIC_VERBOSE," failed.\n");
+       apic_printk(APIC_QUIET, KERN_INFO "..... failed.\n");
 
-       apic_printk(APIC_VERBOSE, KERN_INFO "...trying to set up timer as ExtINT IRQ...");
+       apic_printk(APIC_QUIET, KERN_INFO
+                   "...trying to set up timer as ExtINT IRQ...\n");
 
        init_8259A(0);
        make_8259A_irq(0);
@@ -1766,11 +1811,12 @@ static inline void __init check_timer(vo
        unlock_ExtINT_logic();
 
        if (timer_irq_works()) {
-               apic_printk(APIC_VERBOSE," works.\n");
+               apic_printk(APIC_QUIET, KERN_INFO "..... works.\n");
                goto out;
        }
-       apic_printk(APIC_VERBOSE," failed :(.\n");
-       panic("IO-APIC + timer doesn't work! Try using the 'noapic' kernel parameter\n");
+       apic_printk(APIC_QUIET, KERN_INFO "..... failed :(.\n");
+       panic("IO-APIC + timer doesn't work!  Boot with apic=debug and send a "
+               "report.  Then try booting with the 'noapic' option.\n");
 out:
        local_irq_restore(flags);
 }
@@ -1788,10 +1834,21 @@ __setup("no_timer_check", notimercheck);
 
 /*
  *
- * IRQs that are handled by the PIC in the MPS IOAPIC case.
- * - IRQ2 is the cascade IRQ, and cannot be a io-apic IRQ.
- *   Linux doesn't really care, as it's not actually used
- *   for any interrupt handling anyway.
+ * Traditionally ISA IRQ2 is the cascade IRQ, and is not available
+ * to devices.  However there may be an I/O APIC pin available for
+ * this interrupt regardless.  The pin may be left unconnected, but
+ * typically it will be reused as an ExtINT cascade interrupt for
+ * the master 8259A.  In the MPS case such a pin will normally be
+ * reported as an ExtINT interrupt in the MP table.  With ACPI
+ * there is no provision for ExtINT interrupts, and in the absence
+ * of an override it would be treated as an ordinary ISA I/O APIC
+ * interrupt, that is edge-triggered and unmasked by default.  We
+ * used to do this, but it caused problems on some systems because
+ * of the NMI watchdog and sometimes IRQ0 of the 8254 timer using
+ * the same ExtINT cascade interrupt to drive the local APIC of the
+ * bootstrap processor.  Therefore we refrain from routing IRQ2 to
+ * the I/O APIC in all cases now.  No actual device should request
+ * it anyway.  --macro
  */
 #define PIC_IRQS       (1<<2)
 
@@ -1799,10 +1856,7 @@ void __init setup_IO_APIC(void)
 {
        enable_IO_APIC();
 
-       if (acpi_ioapic)
-               io_apic_irqs = ~0;      /* all IRQs go through IOAPIC */
-       else
-               io_apic_irqs = ~PIC_IRQS;
+       io_apic_irqs = ~PIC_IRQS;
 
        apic_printk(APIC_VERBOSE, "ENABLING IO-APIC IRQs\n");
 
@@ -1851,8 +1905,8 @@ static int ioapic_resume(struct sys_devi
 
        spin_lock_irqsave(&ioapic_lock, flags);
        reg_00.raw = io_apic_read(dev->id, 0);
-       if (reg_00.bits.ID != mp_ioapics[dev->id].mpc_apicid) {
-               reg_00.bits.ID = mp_ioapics[dev->id].mpc_apicid;
+       if (reg_00.bits.ID != mp_ioapics[dev->id].mp_apicid) {
+               reg_00.bits.ID = mp_ioapics[dev->id].mp_apicid;
                io_apic_write(dev->id, 0, reg_00.raw);
        }
        spin_unlock_irqrestore(&ioapic_lock, flags);
@@ -2254,8 +2308,8 @@ int acpi_get_override_irq(int bus_irq, i
                return -1;
 
        for (i = 0; i < mp_irq_entries; i++)
-               if (mp_irqs[i].mpc_irqtype == mp_INT &&
-                   mp_irqs[i].mpc_srcbusirq == bus_irq)
+               if (mp_irqs[i].mp_irqtype == mp_INT &&
+                   mp_irqs[i].mp_srcbusirq == bus_irq)
                        break;
        if (i >= mp_irq_entries)
                return -1;
@@ -2349,7 +2403,7 @@ void __init ioapic_init_mappings(void)
        ioapic_res = ioapic_setup_resources();
        for (i = 0; i < nr_ioapics; i++) {
                if (smp_found_config) {
-                       ioapic_phys = mp_ioapics[i].mpc_apicaddr;
+                       ioapic_phys = mp_ioapics[i].mp_apicaddr;
                } else {
                        ioapic_phys = (unsigned long)
                                alloc_bootmem_pages(PAGE_SIZE);
Index: head-2008-12-01/arch/x86/kernel/ipi-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/ipi-xen.c      2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/ipi-xen.c   2008-12-01 11:49:07.000000000 +0100
@@ -8,7 +8,6 @@
 #include <linux/kernel_stat.h>
 #include <linux/mc146818rtc.h>
 #include <linux/cache.h>
-#include <linux/interrupt.h>
 #include <linux/cpu.h>
 #include <linux/module.h>
 
@@ -85,7 +84,7 @@ void __send_IPI_shortcut(unsigned int sh
        /*
         * Send the IPI. The write to APIC_ICR fires this off.
         */
-       apic_write_around(APIC_ICR, cfg);
+       apic_write(APIC_ICR, cfg);
 #else
        int cpu;
 
@@ -132,7 +131,7 @@ static inline void __send_IPI_dest_field
         * prepare target chip field
         */
        cfg = __prepare_ICR2(mask);
-       apic_write_around(APIC_ICR2, cfg);
+       apic_write(APIC_ICR2, cfg);
 
        /*
         * program the ICR
@@ -142,7 +141,7 @@ static inline void __send_IPI_dest_field
        /*
         * Send the IPI. The write to APIC_ICR fires this off.
         */
-       apic_write_around(APIC_ICR, cfg);
+       apic_write(APIC_ICR, cfg);
 }
 #endif
 
Index: head-2008-12-01/arch/x86/kernel/irq_32-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/irq_32-xen.c   2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/irq_32-xen.c        2008-12-01 11:49:07.000000000 +0100
@@ -48,6 +48,29 @@ void ack_bad_irq(unsigned int irq)
 #endif
 }
 
+#ifdef CONFIG_DEBUG_STACKOVERFLOW
+/* Debugging check for stack overflow: is there less than 1KB free? */
+static int check_stack_overflow(void)
+{
+       long sp;
+
+       __asm__ __volatile__("andl %%esp,%0" :
+                            "=r" (sp) : "0" (THREAD_SIZE - 1));
+
+       return sp < (sizeof(struct thread_info) + STACK_WARN);
+}
+
+static void print_stack_overflow(void)
+{
+       printk(KERN_WARNING "low stack detected by irq handler\n");
+       dump_stack();
+}
+
+#else
+static inline int check_stack_overflow(void) { return 0; }
+static inline void print_stack_overflow(void) { }
+#endif
+
 #ifdef CONFIG_4KSTACKS
 /*
  * per-CPU IRQ handling contexts (thread information and stack)
@@ -59,48 +82,26 @@ union irq_ctx {
 
 static union irq_ctx *hardirq_ctx[NR_CPUS] __read_mostly;
 static union irq_ctx *softirq_ctx[NR_CPUS] __read_mostly;
-#endif
-
-/*
- * do_IRQ handles all normal device IRQ's (the special
- * SMP cross-CPU interrupts have their own specific
- * handlers).
- */
-unsigned int do_IRQ(struct pt_regs *regs)
-{      
-       struct pt_regs *old_regs;
-       /* high bit used in ret_from_ code */
-       int irq = ~regs->orig_ax;
-       struct irq_desc *desc = irq_desc + irq;
-#ifdef CONFIG_4KSTACKS
-       union irq_ctx *curctx, *irqctx;
-       u32 *isp;
-#endif
 
-       if (unlikely((unsigned)irq >= NR_IRQS)) {
-               printk(KERN_EMERG "%s: cannot handle IRQ %d\n",
-                                       __func__, irq);
-               BUG();
-       }
+static char softirq_stack[NR_CPUS * THREAD_SIZE] __page_aligned_bss;
+static char hardirq_stack[NR_CPUS * THREAD_SIZE] __page_aligned_bss;
 
-       old_regs = set_irq_regs(regs);
-       /*irq_enter();*/
-#ifdef CONFIG_DEBUG_STACKOVERFLOW
-       /* Debugging check for stack overflow: is there less than 1KB free? */
-       {
-               long sp;
-
-               __asm__ __volatile__("andl %%esp,%0" :
-                                       "=r" (sp) : "0" (THREAD_SIZE - 1));
-               if (unlikely(sp < (sizeof(struct thread_info) + STACK_WARN))) {
-                       printk("do_IRQ: stack overflow: %ld\n",
-                               sp - sizeof(struct thread_info));
-                       dump_stack();
-               }
-       }
-#endif
+static void call_on_stack(void *func, void *stack)
+{
+       asm volatile("xchgl     %%ebx,%%esp     \n"
+                    "call      *%%edi          \n"
+                    "movl      %%ebx,%%esp     \n"
+                    : "=b" (stack)
+                    : "0" (stack),
+                      "D"(func)
+                    : "memory", "cc", "edx", "ecx", "eax");
+}
 
-#ifdef CONFIG_4KSTACKS
+static inline int
+execute_on_irq_stack(int overflow, struct irq_desc *desc, int irq)
+{
+       union irq_ctx *curctx, *irqctx;
+       u32 *isp, arg1, arg2;
 
        curctx = (union irq_ctx *) current_thread_info();
        irqctx = hardirq_ctx[smp_processor_id()];
@@ -111,52 +112,39 @@ unsigned int do_IRQ(struct pt_regs *regs
         * handler) we can't do that and just have to keep using the
         * current stack (which is the irq stack already after all)
         */
-       if (curctx != irqctx) {
-               int arg1, arg2, bx;
-
-               /* build the stack frame on the IRQ stack */
-               isp = (u32*) ((char*)irqctx + sizeof(*irqctx));
-               irqctx->tinfo.task = curctx->tinfo.task;
-               irqctx->tinfo.previous_esp = current_stack_pointer;
+       if (unlikely(curctx == irqctx))
+               return 0;
 
-               /*
-                * Copy the softirq bits in preempt_count so that the
-                * softirq checks work in the hardirq context.
-                */
-               irqctx->tinfo.preempt_count =
-                       (irqctx->tinfo.preempt_count & ~SOFTIRQ_MASK) |
-                       (curctx->tinfo.preempt_count & SOFTIRQ_MASK);
-
-               asm volatile(
-                       "       xchgl  %%ebx,%%esp    \n"
-                       "       call   *%%edi         \n"
-                       "       movl   %%ebx,%%esp    \n"
-                       : "=a" (arg1), "=d" (arg2), "=b" (bx)
-                       :  "0" (irq),   "1" (desc),  "2" (isp),
-                          "D" (desc->handle_irq)
-                       : "memory", "cc", "ecx"
-               );
-       } else
-#endif
-               desc->handle_irq(irq, desc);
+       /* build the stack frame on the IRQ stack */
+       isp = (u32 *) ((char*)irqctx + sizeof(*irqctx));
+       irqctx->tinfo.task = curctx->tinfo.task;
+       irqctx->tinfo.previous_esp = current_stack_pointer;
 
-       /*irq_exit();*/
-       set_irq_regs(old_regs);
+       /*
+        * Copy the softirq bits in preempt_count so that the
+        * softirq checks work in the hardirq context.
+        */
+       irqctx->tinfo.preempt_count =
+               (irqctx->tinfo.preempt_count & ~SOFTIRQ_MASK) |
+               (curctx->tinfo.preempt_count & SOFTIRQ_MASK);
+
+       if (unlikely(overflow))
+               call_on_stack(print_stack_overflow, isp);
+
+       asm volatile("xchgl     %%ebx,%%esp     \n"
+                    "call      *%%edi          \n"
+                    "movl      %%ebx,%%esp     \n"
+                    : "=a" (arg1), "=d" (arg2), "=b" (isp)
+                    :  "0" (irq),   "1" (desc),  "2" (isp),
+                       "D" (desc->handle_irq)
+                    : "memory", "cc", "ecx");
        return 1;
 }
 
-#ifdef CONFIG_4KSTACKS
-
-static char softirq_stack[NR_CPUS * THREAD_SIZE]
-               __attribute__((__section__(".bss.page_aligned")));
-
-static char hardirq_stack[NR_CPUS * THREAD_SIZE]
-               __attribute__((__section__(".bss.page_aligned")));
-
 /*
  * allocate per-cpu stacks for hardirq and for softirq processing
  */
-void irq_ctx_init(int cpu)
+void __cpuinit irq_ctx_init(int cpu)
 {
        union irq_ctx *irqctx;
 
@@ -164,25 +152,25 @@ void irq_ctx_init(int cpu)
                return;
 
        irqctx = (union irq_ctx*) &hardirq_stack[cpu*THREAD_SIZE];
-       irqctx->tinfo.task              = NULL;
-       irqctx->tinfo.exec_domain       = NULL;
-       irqctx->tinfo.cpu               = cpu;
-       irqctx->tinfo.preempt_count     = HARDIRQ_OFFSET;
-       irqctx->tinfo.addr_limit        = MAKE_MM_SEG(0);
+       irqctx->tinfo.task              = NULL;
+       irqctx->tinfo.exec_domain       = NULL;
+       irqctx->tinfo.cpu               = cpu;
+       irqctx->tinfo.preempt_count     = HARDIRQ_OFFSET;
+       irqctx->tinfo.addr_limit        = MAKE_MM_SEG(0);
 
        hardirq_ctx[cpu] = irqctx;
 
        irqctx = (union irq_ctx*) &softirq_stack[cpu*THREAD_SIZE];
-       irqctx->tinfo.task              = NULL;
-       irqctx->tinfo.exec_domain       = NULL;
-       irqctx->tinfo.cpu               = cpu;
-       irqctx->tinfo.preempt_count     = 0;
-       irqctx->tinfo.addr_limit        = MAKE_MM_SEG(0);
+       irqctx->tinfo.task              = NULL;
+       irqctx->tinfo.exec_domain       = NULL;
+       irqctx->tinfo.cpu               = cpu;
+       irqctx->tinfo.preempt_count     = 0;
+       irqctx->tinfo.addr_limit        = MAKE_MM_SEG(0);
 
        softirq_ctx[cpu] = irqctx;
 
-       printk("CPU %u irqstacks, hard=%p soft=%p\n",
-               cpu,hardirq_ctx[cpu],softirq_ctx[cpu]);
+       printk(KERN_DEBUG "CPU %u irqstacks, hard=%p soft=%p\n",
+              cpu,hardirq_ctx[cpu],softirq_ctx[cpu]);
 }
 
 void irq_ctx_exit(int cpu)
@@ -211,25 +199,56 @@ asmlinkage void do_softirq(void)
                /* build the stack frame on the softirq stack */
                isp = (u32*) ((char*)irqctx + sizeof(*irqctx));
 
-               asm volatile(
-                       "       xchgl   %%ebx,%%esp     \n"
-                       "       call    __do_softirq    \n"
-                       "       movl    %%ebx,%%esp     \n"
-                       : "=b"(isp)
-                       : "0"(isp)
-                       : "memory", "cc", "edx", "ecx", "eax"
-               );
+               call_on_stack(__do_softirq, isp);
                /*
                 * Shouldnt happen, we returned above if in_interrupt():
-                */
+                */
                WARN_ON_ONCE(softirq_count());
        }
 
        local_irq_restore(flags);
 }
+
+#else
+static inline int
+execute_on_irq_stack(int overflow, struct irq_desc *desc, int irq) { return 0; }
 #endif
 
 /*
+ * do_IRQ handles all normal device IRQ's (the special
+ * SMP cross-CPU interrupts have their own specific
+ * handlers).
+ */
+unsigned int do_IRQ(struct pt_regs *regs)
+{
+       struct pt_regs *old_regs;
+       /* high bit used in ret_from_ code */
+       int overflow, irq = ~regs->orig_ax;
+       struct irq_desc *desc = irq_desc + irq;
+
+       if (unlikely((unsigned)irq >= NR_IRQS)) {
+               printk(KERN_EMERG "%s: cannot handle IRQ %d\n",
+                                       __func__, irq);
+               BUG();
+       }
+
+       old_regs = set_irq_regs(regs);
+       /*irq_enter();*/
+
+       overflow = check_stack_overflow();
+
+       if (!execute_on_irq_stack(overflow, desc, irq)) {
+               if (unlikely(overflow))
+                       print_stack_overflow();
+               desc->handle_irq(irq, desc);
+       }
+
+       /*irq_exit();*/
+       set_irq_regs(old_regs);
+       return 1;
+}
+
+/*
  * Interrupt statistics:
  */
 
@@ -337,6 +356,42 @@ skip:
        return 0;
 }
 
+/*
+ * /proc/stat helpers
+ */
+u64 arch_irq_stat_cpu(unsigned int cpu)
+{
+       u64 sum = nmi_count(cpu);
+
+#ifdef CONFIG_X86_LOCAL_APIC
+       sum += per_cpu(irq_stat, cpu).apic_timer_irqs;
+#endif
+#ifdef CONFIG_SMP
+       sum += per_cpu(irq_stat, cpu).irq_resched_count;
+       sum += per_cpu(irq_stat, cpu).irq_call_count;
+#ifndef CONFIG_XEN
+       sum += per_cpu(irq_stat, cpu).irq_tlb_count;
+#endif
+#endif
+#ifdef CONFIG_X86_MCE
+       sum += per_cpu(irq_stat, cpu).irq_thermal_count;
+#endif
+#ifdef CONFIG_X86_LOCAL_APIC
+       sum += per_cpu(irq_stat, cpu).irq_spurious_count;
+#endif
+       return sum;
+}
+
+u64 arch_irq_stat(void)
+{
+       u64 sum = atomic_read(&irq_err_count);
+
+#ifdef CONFIG_X86_IO_APIC
+       sum += atomic_read(&irq_mis_count);
+#endif
+       return sum;
+}
+
 #ifdef CONFIG_HOTPLUG_CPU
 
 void fixup_irqs(cpumask_t map)
Index: head-2008-12-01/arch/x86/kernel/irq_64-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/irq_64-xen.c   2008-12-01 11:37:10.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/irq_64-xen.c        2008-12-01 11:49:07.000000000 +0100
@@ -163,6 +163,34 @@ skip:
 }
 
 /*
+ * /proc/stat helpers
+ */
+u64 arch_irq_stat_cpu(unsigned int cpu)
+{
+       u64 sum = cpu_pda(cpu)->__nmi_count;
+
+       sum += cpu_pda(cpu)->apic_timer_irqs;
+#ifdef CONFIG_SMP
+       sum += cpu_pda(cpu)->irq_resched_count;
+       sum += cpu_pda(cpu)->irq_call_count;
+#ifndef CONFIG_XEN
+       sum += cpu_pda(cpu)->irq_tlb_count;
+#endif
+#endif
+#ifdef CONFIG_X86_MCE
+       sum += cpu_pda(cpu)->irq_thermal_count;
+       sum += cpu_pda(cpu)->irq_threshold_count;
+#endif
+       sum += cpu_pda(cpu)->irq_spurious_count;
+       return sum;
+}
+
+u64 arch_irq_stat(void)
+{
+       return atomic_read(&irq_err_count);
+}
+
+/*
  * do_IRQ handles all normal device IRQ's (the special
  * SMP cross-CPU interrupts have their own specific
  * handlers).
Index: head-2008-12-01/arch/x86/kernel/ldt-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/ldt-xen.c      2008-12-01 11:37:10.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/ldt-xen.c   2008-12-01 11:49:07.000000000 +0100
@@ -20,9 +20,9 @@
 #include <asm/mmu_context.h>
 
 #ifdef CONFIG_SMP
-static void flush_ldt(void *null)
+static void flush_ldt(void *current_mm)
 {
-       if (current->active_mm)
+       if (current->active_mm == current_mm)
                load_LDT(&current->active_mm->context);
 }
 #endif
@@ -62,8 +62,6 @@ static int alloc_ldt(mm_context_t *pc, i
 
        if (reload) {
 #ifdef CONFIG_SMP
-               cpumask_t mask;
-
                preempt_disable();
 #endif
                make_pages_readonly(newldt,
@@ -71,9 +69,9 @@ static int alloc_ldt(mm_context_t *pc, i
                                    XENFEAT_writable_descriptor_tables);
                load_LDT(pc);
 #ifdef CONFIG_SMP
-               mask = cpumask_of_cpu(smp_processor_id());
-               if (!cpus_equal(current->mm->cpu_vm_mask, mask))
-                       smp_call_function(flush_ldt, NULL, 1, 1);
+               if (!cpus_equal(current->mm->cpu_vm_mask,
+                               cpumask_of_cpu(smp_processor_id())))
+                       smp_call_function(flush_ldt, current->mm, 1);
                preempt_enable();
 #endif
        }
Index: head-2008-12-01/arch/x86/kernel/microcode-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/microcode-xen.c        2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/microcode-xen.c     2008-12-01 11:49:07.000000000 +0100
@@ -5,13 +5,14 @@
  *                   2006      Shaohua Li <shaohua.li@intel.com>
  *
  *     This driver allows to upgrade microcode on Intel processors
- *     belonging to IA-32 family - PentiumPro, Pentium II, 
+ *     belonging to IA-32 family - PentiumPro, Pentium II,
  *     Pentium III, Xeon, Pentium 4, etc.
  *
- *     Reference: Section 8.10 of Volume III, Intel Pentium 4 Manual, 
- *     Order Number 245472 or free download from:
- *             
- *     http://developer.intel.com/design/pentium4/manuals/245472.htm
+ *     Reference: Section 8.11 of Volume 3a, IA-32 Intel? Architecture
+ *     Software Developer's Manual
+ *     Order Number 253668 or free download from:
+ *
+ *     http://developer.intel.com/design/pentium4/manuals/253668.htm
  *
  *     For more information, go to http://www.urbanmyth.org/microcode
  *
@@ -26,6 +27,7 @@
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/sched.h>
+#include <linux/smp_lock.h>
 #include <linux/cpumask.h>
 #include <linux/module.h>
 #include <linux/slab.h>
@@ -86,6 +88,7 @@ static int do_microcode_update (const vo
 
 static int microcode_open (struct inode *unused1, struct file *unused2)
 {
+       cycle_kernel_lock();
        return capable(CAP_SYS_RAWIO) ? 0 : -EPERM;
 }
 
@@ -162,7 +165,7 @@ static int request_microcode(void)
                c->x86, c->x86_model, c->x86_mask);
        error = request_firmware(&firmware, name, &microcode_pdev->dev);
        if (error) {
-               pr_debug("microcode: ucode data file %s load failed\n", name);
+               pr_debug("microcode: data file %s load failed\n", name);
                return error;
        }
 
@@ -183,6 +186,9 @@ static int __init microcode_init (void)
 {
        int error;
 
+       printk(KERN_INFO
+               "IA-32 Microcode Update Driver: v" MICROCODE_VERSION " <tigran@aivazian.fsnet.co.uk>\n");
+
        error = microcode_dev_init();
        if (error)
                return error;
@@ -195,8 +201,6 @@ static int __init microcode_init (void)
 
        request_microcode();
 
-       printk(KERN_INFO 
-               "IA-32 Microcode Update Driver: v" MICROCODE_VERSION " <tigran@aivazian.fsnet.co.uk>\n");
        return 0;
 }
 
Index: head-2008-12-01/arch/x86/kernel/mpparse-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/mpparse-xen.c  2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/mpparse-xen.c       2008-12-01 11:49:07.000000000 +0100
@@ -25,6 +25,9 @@
 #include <asm/proto.h>
 #include <asm/acpi.h>
 #include <asm/bios_ebda.h>
+#include <asm/e820.h>
+#include <asm/trampoline.h>
+#include <asm/setup.h>
 
 #include <mach_apic.h>
 #ifdef CONFIG_X86_32
@@ -32,28 +35,6 @@
 #include <mach_mpparse.h>
 #endif
 
-/* Have we found an MP table */
-int smp_found_config;
-
-/*
- * Various Linux-internal data structures created from the
- * MP-table.
- */
-#if defined (CONFIG_MCA) || defined (CONFIG_EISA)
-int mp_bus_id_to_type[MAX_MP_BUSSES];
-#endif
-
-DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES);
-int mp_bus_id_to_pci_bus[MAX_MP_BUSSES] = {[0 ... MAX_MP_BUSSES - 1] = -1 };
-
-static int mp_current_pci_id;
-
-int pic_mode;
-
-/*
- * Intel MP BIOS table parsing routines:
- */
-
 /*
  * Checksum an MP configuration block.
  */
@@ -68,20 +49,8 @@ static int __init mpf_checksum(unsigned 
        return sum & 0xFF;
 }
 
-#ifdef CONFIG_X86_NUMAQ
-/*
- * Have to match translation table entries to main table entries by counter
- * hence the mpc_record variable .... can't see a less disgusting way of
- * doing this ....
- */
-
-static int mpc_record;
-static struct mpc_config_translation *translation_table[MAX_MPC_ENTRY]
-    __cpuinitdata;
-#endif
-
 #ifndef CONFIG_XEN
-static void __cpuinit MP_processor_info(struct mpc_config_processor *m)
+static void __init MP_processor_info(struct mpc_config_processor *m)
 {
        int apicid;
        char *bootup_cpu = "";
@@ -90,11 +59,12 @@ static void __cpuinit MP_processor_info(
                disabled_cpus++;
                return;
        }
-#ifdef CONFIG_X86_NUMAQ
-       apicid = mpc_apic_id(m, translation_table[mpc_record]);
-#else
-       apicid = m->mpc_apicid;
-#endif
+
+       if (x86_quirks->mpc_apic_id)
+               apicid = x86_quirks->mpc_apic_id(m);
+       else
+               apicid = m->mpc_apicid;
+
        if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) {
                bootup_cpu = " (Bootup-CPU)";
                boot_cpu_physical_apicid = m->mpc_apicid;
@@ -104,24 +74,23 @@ static void __cpuinit MP_processor_info(
        generic_processor_info(apicid, m->mpc_apicver);
 }
 #else
-static void __cpuinit MP_processor_info(struct mpc_config_processor *m)
+static void __init MP_processor_info(struct mpc_config_processor *m)
 {
        num_processors++;
 }
 #endif /* CONFIG_XEN */
 
+#ifdef CONFIG_X86_IO_APIC
 static void __init MP_bus_info(struct mpc_config_bus *m)
 {
        char str[7];
-
        memcpy(str, m->mpc_bustype, 6);
        str[6] = 0;
 
-#ifdef CONFIG_X86_NUMAQ
-       mpc_oem_bus_info(m, str, translation_table[mpc_record]);
-#else
-       Dprintk("Bus #%d is %s\n", m->mpc_busid, str);
-#endif
+       if (x86_quirks->mpc_oem_bus_info)
+               x86_quirks->mpc_oem_bus_info(m, str);
+       else
+               apic_printk(APIC_VERBOSE, "Bus #%d is %s\n", m->mpc_busid, str);
 
 #if MAX_MP_BUSSES < 256
        if (m->mpc_busid >= MAX_MP_BUSSES) {
@@ -138,12 +107,10 @@ static void __init MP_bus_info(struct mp
                mp_bus_id_to_type[m->mpc_busid] = MP_BUS_ISA;
 #endif
        } else if (strncmp(str, BUSTYPE_PCI, sizeof(BUSTYPE_PCI) - 1) == 0) {
-#ifdef CONFIG_X86_NUMAQ
-               mpc_oem_pci_bus(m, translation_table[mpc_record]);
-#endif
+               if (x86_quirks->mpc_oem_pci_bus)
+                       x86_quirks->mpc_oem_pci_bus(m);
+
                clear_bit(m->mpc_busid, mp_bus_not_pci);
-               mp_bus_id_to_pci_bus[m->mpc_busid] = mp_current_pci_id;
-               mp_current_pci_id++;
 #if defined(CONFIG_EISA) || defined (CONFIG_MCA)
                mp_bus_id_to_type[m->mpc_busid] = MP_BUS_PCI;
        } else if (strncmp(str, BUSTYPE_EISA, sizeof(BUSTYPE_EISA) - 1) == 0) {
@@ -154,6 +121,7 @@ static void __init MP_bus_info(struct mp
        } else
                printk(KERN_WARNING "Unknown bustype %s - ignoring\n", str);
 }
+#endif
 
 #ifdef CONFIG_X86_IO_APIC
 
@@ -183,117 +151,111 @@ static void __init MP_ioapic_info(struct
        if (bad_ioapic(m->mpc_apicaddr))
                return;
 
-       mp_ioapics[nr_ioapics] = *m;
+       mp_ioapics[nr_ioapics].mp_apicaddr = m->mpc_apicaddr;
+       mp_ioapics[nr_ioapics].mp_apicid = m->mpc_apicid;
+       mp_ioapics[nr_ioapics].mp_type = m->mpc_type;
+       mp_ioapics[nr_ioapics].mp_apicver = m->mpc_apicver;
+       mp_ioapics[nr_ioapics].mp_flags = m->mpc_flags;
        nr_ioapics++;
 }
 
-static void __init MP_intsrc_info(struct mpc_config_intsrc *m)
+static void print_MP_intsrc_info(struct mpc_config_intsrc *m)
 {
-       mp_irqs[mp_irq_entries] = *m;
-       Dprintk("Int: type %d, pol %d, trig %d, bus %d,"
+       apic_printk(APIC_VERBOSE, "Int: type %d, pol %d, trig %d, bus %02x,"
                " IRQ %02x, APIC ID %x, APIC INT %02x\n",
                m->mpc_irqtype, m->mpc_irqflag & 3,
                (m->mpc_irqflag >> 2) & 3, m->mpc_srcbus,
                m->mpc_srcbusirq, m->mpc_dstapic, m->mpc_dstirq);
-       if (++mp_irq_entries == MAX_IRQ_SOURCES)
-               panic("Max # of irq sources exceeded!!\n");
 }
 
-#endif
-
-static void __init MP_lintsrc_info(struct mpc_config_lintsrc *m)
+static void __init print_mp_irq_info(struct mp_config_intsrc *mp_irq)
 {
-       Dprintk("Lint: type %d, pol %d, trig %d, bus %d,"
-               " IRQ %02x, APIC ID %x, APIC LINT %02x\n",
-               m->mpc_irqtype, m->mpc_irqflag & 3,
-               (m->mpc_irqflag >> 2) & 3, m->mpc_srcbusid,
-               m->mpc_srcbusirq, m->mpc_destapic, m->mpc_destapiclint);
+       apic_printk(APIC_VERBOSE, "Int: type %d, pol %d, trig %d, bus %02x,"
+               " IRQ %02x, APIC ID %x, APIC INT %02x\n",
+               mp_irq->mp_irqtype, mp_irq->mp_irqflag & 3,
+               (mp_irq->mp_irqflag >> 2) & 3, mp_irq->mp_srcbus,
+               mp_irq->mp_srcbusirq, mp_irq->mp_dstapic, mp_irq->mp_dstirq);
 }
 
-#ifdef CONFIG_X86_NUMAQ
-static void __init MP_translation_info(struct mpc_config_translation *m)
+static void __init assign_to_mp_irq(struct mpc_config_intsrc *m,
+                                   struct mp_config_intsrc *mp_irq)
 {
-       printk(KERN_INFO
-              "Translation: record %d, type %d, quad %d, global %d, local %d\n",
-              mpc_record, m->trans_type, m->trans_quad, m->trans_global,
-              m->trans_local);
+       mp_irq->mp_dstapic = m->mpc_dstapic;
+       mp_irq->mp_type = m->mpc_type;
+       mp_irq->mp_irqtype = m->mpc_irqtype;
+       mp_irq->mp_irqflag = m->mpc_irqflag;
+       mp_irq->mp_srcbus = m->mpc_srcbus;
+       mp_irq->mp_srcbusirq = m->mpc_srcbusirq;
+       mp_irq->mp_dstirq = m->mpc_dstirq;
+}
 
-       if (mpc_record >= MAX_MPC_ENTRY)
-               printk(KERN_ERR "MAX_MPC_ENTRY exceeded!\n");
-       else
-               translation_table[mpc_record] = m;      /* stash this for later */
-       if (m->trans_quad < MAX_NUMNODES && !node_online(m->trans_quad))
-               node_set_online(m->trans_quad);
+static void __init assign_to_mpc_intsrc(struct mp_config_intsrc *mp_irq,
+                                       struct mpc_config_intsrc *m)
+{
+       m->mpc_dstapic = mp_irq->mp_dstapic;
+       m->mpc_type = mp_irq->mp_type;
+       m->mpc_irqtype = mp_irq->mp_irqtype;
+       m->mpc_irqflag = mp_irq->mp_irqflag;
+       m->mpc_srcbus = mp_irq->mp_srcbus;
+       m->mpc_srcbusirq = mp_irq->mp_srcbusirq;
+       m->mpc_dstirq = mp_irq->mp_dstirq;
 }
 
-/*
- * Read/parse the MPC oem tables
- */
+static int __init mp_irq_mpc_intsrc_cmp(struct mp_config_intsrc *mp_irq,
+                                       struct mpc_config_intsrc *m)
+{
+       if (mp_irq->mp_dstapic != m->mpc_dstapic)
+               return 1;
+       if (mp_irq->mp_type != m->mpc_type)
+               return 2;
+       if (mp_irq->mp_irqtype != m->mpc_irqtype)
+               return 3;
+       if (mp_irq->mp_irqflag != m->mpc_irqflag)
+               return 4;
+       if (mp_irq->mp_srcbus != m->mpc_srcbus)
+               return 5;
+       if (mp_irq->mp_srcbusirq != m->mpc_srcbusirq)
+               return 6;
+       if (mp_irq->mp_dstirq != m->mpc_dstirq)
+               return 7;
+
+       return 0;
+}
 
-static void __init smp_read_mpc_oem(struct mp_config_oemtable *oemtable,
-                                   unsigned short oemsize)
+static void __init MP_intsrc_info(struct mpc_config_intsrc *m)
 {
-       int count = sizeof(*oemtable);  /* the header size */
-       unsigned char *oemptr = ((unsigned char *)oemtable) + count;
+       int i;
 
-       mpc_record = 0;
-       printk(KERN_INFO "Found an OEM MPC table at %8p - parsing it ... \n",
-              oemtable);
-       if (memcmp(oemtable->oem_signature, MPC_OEM_SIGNATURE, 4)) {
-               printk(KERN_WARNING
-                      "SMP mpc oemtable: bad signature [%c%c%c%c]!\n",
-                      oemtable->oem_signature[0], oemtable->oem_signature[1],
-                      oemtable->oem_signature[2], oemtable->oem_signature[3]);
-               return;
-       }
-       if (mpf_checksum((unsigned char *)oemtable, oemtable->oem_length)) {
-               printk(KERN_WARNING "SMP oem mptable: checksum error!\n");
-               return;
-       }
-       while (count < oemtable->oem_length) {
-               switch (*oemptr) {
-               case MP_TRANSLATION:
-                       {
-                               struct mpc_config_translation *m =
-                                   (struct mpc_config_translation *)oemptr;
-                               MP_translation_info(m);
-                               oemptr += sizeof(*m);
-                               count += sizeof(*m);
-                               ++mpc_record;
-                               break;
-                       }
-               default:
-                       {
-                               printk(KERN_WARNING
-                                      "Unrecognised OEM table entry type! - %d\n",
-                                      (int)*oemptr);
-                               return;
-                       }
-               }
+       print_MP_intsrc_info(m);
+
+       for (i = 0; i < mp_irq_entries; i++) {
+               if (!mp_irq_mpc_intsrc_cmp(&mp_irqs[i], m))
+                       return;
        }
+
+       assign_to_mp_irq(m, &mp_irqs[mp_irq_entries]);
+       if (++mp_irq_entries == MAX_IRQ_SOURCES)
+               panic("Max # of irq sources exceeded!!\n");
 }
 
-static inline void mps_oem_check(struct mp_config_table *mpc, char *oem,
-                                char *productid)
+#endif
+
+static void __init MP_lintsrc_info(struct mpc_config_lintsrc *m)
 {
-       if (strncmp(oem, "IBM NUMA", 8))
-               printk("Warning!  May not be a NUMA-Q system!\n");
-       if (mpc->mpc_oemptr)
-               smp_read_mpc_oem((struct mp_config_oemtable *)mpc->mpc_oemptr,
-                                mpc->mpc_oemsize);
+       apic_printk(APIC_VERBOSE, "Lint: type %d, pol %d, trig %d, bus %02x,"
+               " IRQ %02x, APIC ID %x, APIC LINT %02x\n",
+               m->mpc_irqtype, m->mpc_irqflag & 3,
+               (m->mpc_irqflag >> 2) & 3, m->mpc_srcbusid,
+               m->mpc_srcbusirq, m->mpc_destapic, m->mpc_destapiclint);
 }
-#endif /* CONFIG_X86_NUMAQ */
 
 /*
  * Read/parse the MPC
  */
 
-static int __init smp_read_mpc(struct mp_config_table *mpc, unsigned early)
+static int __init smp_check_mpc(struct mp_config_table *mpc, char *oem,
+                               char *str)
 {
-       char str[16];
-       char oem[10];
-       int count = sizeof(*mpc);
-       unsigned char *mpt = ((unsigned char *)mpc) + count;
 
        if (memcmp(mpc->mpc_signature, MPC_SIGNATURE, 4)) {
                printk(KERN_ERR "MPTABLE: bad signature [%c%c%c%c]!\n",
@@ -316,19 +278,41 @@ static int __init smp_read_mpc(struct mp
        }
        memcpy(oem, mpc->mpc_oem, 8);
        oem[8] = 0;
-       printk(KERN_INFO "MPTABLE: OEM ID: %s ", oem);
+       printk(KERN_INFO "MPTABLE: OEM ID: %s\n", oem);
 
        memcpy(str, mpc->mpc_productid, 12);
        str[12] = 0;
-       printk("Product ID: %s ", str);
 
-#ifdef CONFIG_X86_32
-       mps_oem_check(mpc, oem, str);
-#endif
-       printk(KERN_INFO "MPTABLE: Product ID: %s ", str);
+       printk(KERN_INFO "MPTABLE: Product ID: %s\n", str);
 
        printk(KERN_INFO "MPTABLE: APIC at: 0x%X\n", mpc->mpc_lapic);
 
+       return 1;
+}
+
+static int __init smp_read_mpc(struct mp_config_table *mpc, unsigned early)
+{
+       char str[16];
+       char oem[10];
+
+       int count = sizeof(*mpc);
+       unsigned char *mpt = ((unsigned char *)mpc) + count;
+
+       if (!smp_check_mpc(mpc, oem, str))
+               return 0;
+
+#ifdef CONFIG_X86_32
+       /*
+        * need to make sure summit and es7000's mps_oem_check is safe to be
+        * called early via genericarch 's mps_oem_check
+        */
+       if (early) {
+#ifdef CONFIG_X86_NUMAQ
+               numaq_mps_oem_check(mpc, oem, str);
+#endif
+       } else
+               mps_oem_check(mpc, oem, str);
+#endif
        /* save the local APIC address, it might be non-default */
        if (!acpi_lapic)
                mp_lapic_addr = mpc->mpc_lapic;
@@ -336,12 +320,17 @@ static int __init smp_read_mpc(struct mp
        if (early)
                return 1;
 
+       if (mpc->mpc_oemptr && x86_quirks->smp_read_mpc_oem) {
+               struct mp_config_oemtable *oem_table = (struct mp_config_oemtable *)(unsigned long)mpc->mpc_oemptr;
+               x86_quirks->smp_read_mpc_oem(oem_table, mpc->mpc_oemsize);
+       }
+
        /*
         *      Now process the configuration blocks.
         */
-#ifdef CONFIG_X86_NUMAQ
-       mpc_record = 0;
-#endif
+       if (x86_quirks->mpc_record)
+               *x86_quirks->mpc_record = 0;
+
        while (count < mpc->mpc_length) {
                switch (*mpt) {
                case MP_PROCESSOR:
@@ -359,7 +348,9 @@ static int __init smp_read_mpc(struct mp
                        {
                                struct mpc_config_bus *m =
                                    (struct mpc_config_bus *)mpt;
+#ifdef CONFIG_X86_IO_APIC
                                MP_bus_info(m);
+#endif
                                mpt += sizeof(*m);
                                count += sizeof(*m);
                                break;
@@ -405,10 +396,14 @@ static int __init smp_read_mpc(struct mp
                        count = mpc->mpc_length;
                        break;
                }
-#ifdef CONFIG_X86_NUMAQ
-               ++mpc_record;
-#endif
+               if (x86_quirks->mpc_record)
+                       (*x86_quirks->mpc_record)++;
        }
+
+#ifdef CONFIG_X86_GENERICARCH
+       generic_bigsmp_probe();
+#endif
+
        setup_apic_routing();
        if (!num_processors)
                printk(KERN_ERR "MPTABLE: no processors registered!\n");
@@ -434,7 +429,7 @@ static void __init construct_default_ioi
        intsrc.mpc_type = MP_INTSRC;
        intsrc.mpc_irqflag = 0; /* conforming */
        intsrc.mpc_srcbus = 0;
-       intsrc.mpc_dstapic = mp_ioapics[0].mpc_apicid;
+       intsrc.mpc_dstapic = mp_ioapics[0].mp_apicid;
 
        intsrc.mpc_irqtype = mp_INT;
 
@@ -495,40 +490,11 @@ static void __init construct_default_ioi
        MP_intsrc_info(&intsrc);
 }
 
-#endif
 
-static inline void __init construct_default_ISA_mptable(int mpc_default_type)
+static void __init construct_ioapic_table(int mpc_default_type)
 {
-       struct mpc_config_processor processor;
-       struct mpc_config_bus bus;
-#ifdef CONFIG_X86_IO_APIC
        struct mpc_config_ioapic ioapic;
-#endif
-       struct mpc_config_lintsrc lintsrc;
-       int linttypes[2] = { mp_ExtINT, mp_NMI };
-       int i;
-
-       /*
-        * local APIC has default address
-        */
-       mp_lapic_addr = APIC_DEFAULT_PHYS_BASE;
-
-       /*
-        * 2 CPUs, numbered 0 & 1.
-        */
-       processor.mpc_type = MP_PROCESSOR;
-       /* Either an integrated APIC or a discrete 82489DX. */
-       processor.mpc_apicver = mpc_default_type > 4 ? 0x10 : 0x01;
-       processor.mpc_cpuflag = CPU_ENABLED;
-       processor.mpc_cpufeature = (boot_cpu_data.x86 << 8) |
-           (boot_cpu_data.x86_model << 4) | boot_cpu_data.x86_mask;
-       processor.mpc_featureflag = boot_cpu_data.x86_capability[0];
-       processor.mpc_reserved[0] = 0;
-       processor.mpc_reserved[1] = 0;
-       for (i = 0; i < 2; i++) {
-               processor.mpc_apicid = i;
-               MP_processor_info(&processor);
-       }
+       struct mpc_config_bus bus;
 
        bus.mpc_type = MP_BUS;
        bus.mpc_busid = 0;
@@ -557,7 +523,6 @@ static inline void __init construct_defa
                MP_bus_info(&bus);
        }
 
-#ifdef CONFIG_X86_IO_APIC
        ioapic.mpc_type = MP_IOAPIC;
        ioapic.mpc_apicid = 2;
        ioapic.mpc_apicver = mpc_default_type > 4 ? 0x10 : 0x01;
@@ -569,7 +534,42 @@ static inline void __init construct_defa
         * We set up most of the low 16 IO-APIC pins according to MPS rules.
         */
        construct_default_ioirq_mptable(mpc_default_type);
+}
+#else
+static inline void __init construct_ioapic_table(int mpc_default_type) { }
 #endif
+
+static inline void __init construct_default_ISA_mptable(int mpc_default_type)
+{
+       struct mpc_config_processor processor;
+       struct mpc_config_lintsrc lintsrc;
+       int linttypes[2] = { mp_ExtINT, mp_NMI };
+       int i;
+
+       /*
+        * local APIC has default address
+        */
+       mp_lapic_addr = APIC_DEFAULT_PHYS_BASE;
+
+       /*
+        * 2 CPUs, numbered 0 & 1.
+        */
+       processor.mpc_type = MP_PROCESSOR;
+       /* Either an integrated APIC or a discrete 82489DX. */
+       processor.mpc_apicver = mpc_default_type > 4 ? 0x10 : 0x01;
+       processor.mpc_cpuflag = CPU_ENABLED;
+       processor.mpc_cpufeature = (boot_cpu_data.x86 << 8) |
+           (boot_cpu_data.x86_model << 4) | boot_cpu_data.x86_mask;
+       processor.mpc_featureflag = boot_cpu_data.x86_capability[0];
+       processor.mpc_reserved[0] = 0;
+       processor.mpc_reserved[1] = 0;
+       for (i = 0; i < 2; i++) {
+               processor.mpc_apicid = i;
+               MP_processor_info(&processor);
+       }
+
+       construct_ioapic_table(mpc_default_type);
+
        lintsrc.mpc_type = MP_LINTSRC;
        lintsrc.mpc_irqflag = 0;        /* conforming */
        lintsrc.mpc_srcbusid = 0;
@@ -587,10 +587,14 @@ static struct intel_mp_floating *mpf_fou
 /*
  * Scan the memory blocks for an SMP configuration block.
  */
-static void __init __get_smp_config(unsigned early)
+static void __init __get_smp_config(unsigned int early)
 {
        struct intel_mp_floating *mpf = mpf_found;
 
+       if (x86_quirks->mach_get_smp_config) {
+               if (x86_quirks->mach_get_smp_config(early))
+                       return;
+       }
        if (acpi_lapic && early)
                return;
        /*
@@ -607,7 +611,7 @@ static void __init __get_smp_config(unsi
 
        printk(KERN_INFO "Intel MultiProcessor Specification v1.%d\n",
               mpf->mpf_specification);
-#ifdef CONFIG_X86_32
+#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86_32) && !defined(CONFIG_XEN)
        if (mpf->mpf_feature2 & (1 << 7)) {
                printk(KERN_INFO "    IMCR and PIC compatibility mode.\n");
                pic_mode = 1;
@@ -639,7 +643,9 @@ static void __init __get_smp_config(unsi
                 * override the defaults.
                 */
                if (!smp_read_mpc(isa_bus_to_virt(mpf->mpf_physptr), early)) {
+#ifdef CONFIG_X86_LOCAL_APIC
                        smp_found_config = 0;
+#endif
                        printk(KERN_ERR
                               "BIOS bug, MP table errors detected!...\n");
                        printk(KERN_ERR "... disabling SMP support. "
@@ -696,7 +702,8 @@ static int __init smp_scan_config(unsign
        unsigned int *bp = isa_bus_to_virt(base);
        struct intel_mp_floating *mpf;
 
-       Dprintk("Scan SMP from %p for %ld bytes.\n", bp, length);
+       apic_printk(APIC_VERBOSE, "Scan SMP from %p for %ld bytes.\n",
+                       bp, length);
        BUILD_BUG_ON(sizeof(*mpf) != 16);
 
        while (length > 0) {
@@ -706,16 +713,22 @@ static int __init smp_scan_config(unsign
                    !mpf_checksum((unsigned char *)bp, 16) &&
                    ((mpf->mpf_specification == 1)
                     || (mpf->mpf_specification == 4))) {
-
+#ifdef CONFIG_X86_LOCAL_APIC
                        smp_found_config = 1;
+#endif
                        mpf_found = mpf;
-#ifdef CONFIG_X86_32
+
 #ifndef CONFIG_XEN
                        printk(KERN_INFO "found SMP MP-table at [%p] %08lx\n",
                               mpf, virt_to_phys(mpf));
-                       reserve_bootmem(virt_to_phys(mpf), PAGE_SIZE,
+
+                       if (!reserve)
+                               return 1;
+                       reserve_bootmem_generic(virt_to_phys(mpf), PAGE_SIZE,
                                        BOOTMEM_DEFAULT);
                        if (mpf->mpf_physptr) {
+                               unsigned long size = PAGE_SIZE;
+#ifdef CONFIG_X86_32
                                /*
                                 * We cannot access to MPC table to compute
                                 * table size yet, as only few megabytes from
@@ -725,27 +738,18 @@ static int __init smp_scan_config(unsign
                                 * PAGE_SIZE from mpg->mpf_physptr yields BUG()
                                 * in reserve_bootmem.
                                 */
-                               unsigned long size = PAGE_SIZE;
                                unsigned long end = max_low_pfn * PAGE_SIZE;
                                if (mpf->mpf_physptr + size > end)
                                        size = end - mpf->mpf_physptr;
-                               reserve_bootmem(mpf->mpf_physptr, size,
+#endif
+                               reserve_bootmem_generic(mpf->mpf_physptr, size,
                                                BOOTMEM_DEFAULT);
                        }
 #else
                        printk(KERN_INFO "found SMP MP-table at [%p] %08lx\n",
                                mpf, ((void *)bp - isa_bus_to_virt(base)) + base);
 #endif
-#elif !defined(CONFIG_XEN)
-                       if (!reserve)
-                               return 1;
-
-                       reserve_bootmem_generic(virt_to_phys(mpf), PAGE_SIZE);
-                       if (mpf->mpf_physptr)
-                               reserve_bootmem_generic(mpf->mpf_physptr,
-                                                       PAGE_SIZE);
-#endif
-               return 1;
+                       return 1;
                }
                bp += 4;
                length -= 16;
@@ -753,10 +757,15 @@ static int __init smp_scan_config(unsign
        return 0;
 }
 
-static void __init __find_smp_config(unsigned reserve)
+static void __init __find_smp_config(unsigned int reserve)
 {
 #ifndef CONFIG_XEN
        unsigned int address;
+
+       if (x86_quirks->mach_find_smp_config) {
+               if (x86_quirks->mach_find_smp_config(reserve))
+                       return;
+       }
 #endif
 
        /*
@@ -805,300 +814,301 @@ void __init find_smp_config(void)
        __find_smp_config(1);
 }
 
-/* --------------------------------------------------------------------------
-                            ACPI-based MP Configuration
-   -------------------------------------------------------------------------- */
-
-/*
- * Keep this outside and initialized to 0, for !CONFIG_ACPI builds:
- */
-int es7000_plat;
-
-#ifdef CONFIG_ACPI
+#ifdef CONFIG_X86_IO_APIC
+static u8 __initdata irq_used[MAX_IRQ_SOURCES];
 
-#ifdef CONFIG_X86_IO_APIC
+static int  __init get_MP_intsrc_index(struct mpc_config_intsrc *m)
+{
+       int i;
 
-#define MP_ISA_BUS             0
+       if (m->mpc_irqtype != mp_INT)
+               return 0;
 
-extern struct mp_ioapic_routing mp_ioapic_routing[MAX_IO_APICS];
+       if (m->mpc_irqflag != 0x0f)
+               return 0;
 
-static int mp_find_ioapic(int gsi)
-{
-       int i = 0;
+       /* not legacy */
 
-       /* Find the IOAPIC that manages this GSI. */
-       for (i = 0; i < nr_ioapics; i++) {
-               if ((gsi >= mp_ioapic_routing[i].gsi_base)
-                   && (gsi <= mp_ioapic_routing[i].gsi_end))
-                       return i;
+       for (i = 0; i < mp_irq_entries; i++) {
+               if (mp_irqs[i].mp_irqtype != mp_INT)
+                       continue;
+
+               if (mp_irqs[i].mp_irqflag != 0x0f)
+                       continue;
+
+               if (mp_irqs[i].mp_srcbus != m->mpc_srcbus)
+                       continue;
+               if (mp_irqs[i].mp_srcbusirq != m->mpc_srcbusirq)
+                       continue;
+               if (irq_used[i]) {
+                       /* already claimed */
+                       return -2;
+               }
+               irq_used[i] = 1;
+               return i;
        }
 
-       printk(KERN_ERR "ERROR: Unable to locate IOAPIC for GSI %d\n", gsi);
+       /* not found */
        return -1;
 }
 
-static u8 __init uniq_ioapic_id(u8 id)
-{
-#ifdef CONFIG_X86_32
-       if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) &&
-           !APIC_XAPIC(apic_version[boot_cpu_physical_apicid]))
-               return io_apic_get_unique_id(nr_ioapics, id);
-       else
-               return id;
-#else
-       int i;
-       DECLARE_BITMAP(used, 256);
-       bitmap_zero(used, 256);
-       for (i = 0; i < nr_ioapics; i++) {
-               struct mpc_config_ioapic *ia = &mp_ioapics[i];
-               __set_bit(ia->mpc_apicid, used);
-       }
-       if (!test_bit(id, used))
-               return id;
-       return find_first_zero_bit(used, 256);
+#define SPARE_SLOT_NUM 20
+
+static struct mpc_config_intsrc __initdata *m_spare[SPARE_SLOT_NUM];
 #endif
-}
 
-void __init mp_register_ioapic(int id, u32 address, u32 gsi_base)
+static int  __init replace_intsrc_all(struct mp_config_table *mpc,
+                                       unsigned long mpc_new_phys,
+                                       unsigned long mpc_new_length)
 {
-       int idx = 0;
-
-       if (bad_ioapic(address))
-               return;
+#ifdef CONFIG_X86_IO_APIC
+       int i;
+       int nr_m_spare = 0;
+#endif
 
-       idx = nr_ioapics;
+       int count = sizeof(*mpc);
+       unsigned char *mpt = ((unsigned char *)mpc) + count;
 
-       mp_ioapics[idx].mpc_type = MP_IOAPIC;
-       mp_ioapics[idx].mpc_flags = MPC_APIC_USABLE;
-       mp_ioapics[idx].mpc_apicaddr = address;
+       printk(KERN_INFO "mpc_length %x\n", mpc->mpc_length);
+       while (count < mpc->mpc_length) {
+               switch (*mpt) {
+               case MP_PROCESSOR:
+                       {
+                               struct mpc_config_processor *m =
+                                   (struct mpc_config_processor *)mpt;
+                               mpt += sizeof(*m);
+                               count += sizeof(*m);
+                               break;
+                       }
+               case MP_BUS:
+                       {
+                               struct mpc_config_bus *m =
+                                   (struct mpc_config_bus *)mpt;
+                               mpt += sizeof(*m);
+                               count += sizeof(*m);
+                               break;
+                       }
+               case MP_IOAPIC:
+                       {
+                               mpt += sizeof(struct mpc_config_ioapic);
+                               count += sizeof(struct mpc_config_ioapic);
+                               break;
+                       }
+               case MP_INTSRC:
+                       {
+#ifdef CONFIG_X86_IO_APIC
+                               struct mpc_config_intsrc *m =
+                                   (struct mpc_config_intsrc *)mpt;
 
-#ifndef CONFIG_XEN
-       set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address);
+                               printk(KERN_INFO "OLD ");
+                               print_MP_intsrc_info(m);
+                               i = get_MP_intsrc_index(m);
+                               if (i > 0) {
+                                       assign_to_mpc_intsrc(&mp_irqs[i], m);
+                                       printk(KERN_INFO "NEW ");
+                                       print_mp_irq_info(&mp_irqs[i]);
+                               } else if (!i) {
+                                       /* legacy, do nothing */
+                               } else if (nr_m_spare < SPARE_SLOT_NUM) {
+                                       /*
+                                        * not found (-1), or duplicated (-2)
+                                        * are invalid entries,
+                                        * we need to use the slot  later
+                                        */
+                                       m_spare[nr_m_spare] = m;
+                                       nr_m_spare++;
+                               }
 #endif
-       mp_ioapics[idx].mpc_apicid = uniq_ioapic_id(id);
-#ifdef CONFIG_X86_32
-       mp_ioapics[idx].mpc_apicver = io_apic_get_version(idx);
-#else
-       mp_ioapics[idx].mpc_apicver = 0;
+                               mpt += sizeof(struct mpc_config_intsrc);
+                               count += sizeof(struct mpc_config_intsrc);
+                               break;
+                       }
+               case MP_LINTSRC:
+                       {
+                               struct mpc_config_lintsrc *m =
+                                   (struct mpc_config_lintsrc *)mpt;
+                               mpt += sizeof(*m);
+                               count += sizeof(*m);
+                               break;
+                       }
+               default:
+                       /* wrong mptable */
+                       printk(KERN_ERR "Your mptable is wrong, contact your HW vendor!\n");
+                       printk(KERN_ERR "type %x\n", *mpt);
+                       print_hex_dump(KERN_ERR, "  ", DUMP_PREFIX_ADDRESS, 16,
+                                       1, mpc, mpc->mpc_length, 1);
+                       goto out;
+               }
+       }
+
+#ifdef CONFIG_X86_IO_APIC
+       for (i = 0; i < mp_irq_entries; i++) {
+               if (irq_used[i])
+                       continue;
+
+               if (mp_irqs[i].mp_irqtype != mp_INT)
+                       continue;
+
+               if (mp_irqs[i].mp_irqflag != 0x0f)
+                       continue;
+
+               if (nr_m_spare > 0) {
+                       printk(KERN_INFO "*NEW* found ");
+                       nr_m_spare--;
+                       assign_to_mpc_intsrc(&mp_irqs[i], m_spare[nr_m_spare]);
+                       m_spare[nr_m_spare] = NULL;
+               } else {
+                       struct mpc_config_intsrc *m =
+                           (struct mpc_config_intsrc *)mpt;
+                       count += sizeof(struct mpc_config_intsrc);
+                       if (!mpc_new_phys) {
+                               printk(KERN_INFO "No spare slots, try to append...take your risk, new mpc_length %x\n", count);
+                       } else {
+                               if (count <= mpc_new_length)
+                                       printk(KERN_INFO "No spare slots, try to append..., new mpc_length %x\n", count);
+                               else {
+                                       printk(KERN_ERR "mpc_new_length %lx is too small\n", mpc_new_length);
+                                       goto out;
+                               }
+                       }
+                       assign_to_mpc_intsrc(&mp_irqs[i], m);
+                       mpc->mpc_length = count;
+                       mpt += sizeof(struct mpc_config_intsrc);
+               }
+               print_mp_irq_info(&mp_irqs[i]);
+       }
 #endif
-       /*
-        * Build basic GSI lookup table to facilitate gsi->io_apic lookups
-        * and to prevent reprogramming of IOAPIC pins (PCI GSIs).
-        */
-       mp_ioapic_routing[idx].apic_id = mp_ioapics[idx].mpc_apicid;
-       mp_ioapic_routing[idx].gsi_base = gsi_base;
-       mp_ioapic_routing[idx].gsi_end = gsi_base +
-           io_apic_get_redir_entries(idx);
-
-       printk(KERN_INFO "IOAPIC[%d]: apic_id %d, version %d, address 0x%x, "
-              "GSI %d-%d\n", idx, mp_ioapics[idx].mpc_apicid,
-              mp_ioapics[idx].mpc_apicver, mp_ioapics[idx].mpc_apicaddr,
-              mp_ioapic_routing[idx].gsi_base, mp_ioapic_routing[idx].gsi_end);
+out:
+       /* update checksum */
+       mpc->mpc_checksum = 0;
+       mpc->mpc_checksum -= mpf_checksum((unsigned char *)mpc,
+                                          mpc->mpc_length);
 
-       nr_ioapics++;
+       return 0;
 }
 
-void __init mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger, u32 gsi)
-{
-       struct mpc_config_intsrc intsrc;
-       int ioapic = -1;
-       int pin = -1;
-
-       /*
-        * Convert 'gsi' to 'ioapic.pin'.
-        */
-       ioapic = mp_find_ioapic(gsi);
-       if (ioapic < 0)
-               return;
-       pin = gsi - mp_ioapic_routing[ioapic].gsi_base;
+static int __initdata enable_update_mptable;
 
-       /*
-        * TBD: This check is for faulty timer entries, where the override
-        *      erroneously sets the trigger to level, resulting in a HUGE
-        *      increase of timer interrupts!
-        */
-       if ((bus_irq == 0) && (trigger == 3))
-               trigger = 1;
+static int __init update_mptable_setup(char *str)
+{
+       enable_update_mptable = 1;
+       return 0;
+}
+early_param("update_mptable", update_mptable_setup);
 
-       intsrc.mpc_type = MP_INTSRC;
-       intsrc.mpc_irqtype = mp_INT;
-       intsrc.mpc_irqflag = (trigger << 2) | polarity;
-       intsrc.mpc_srcbus = MP_ISA_BUS;
-       intsrc.mpc_srcbusirq = bus_irq; /* IRQ */
-       intsrc.mpc_dstapic = mp_ioapics[ioapic].mpc_apicid;     /* APIC ID */
-       intsrc.mpc_dstirq = pin;        /* INTIN# */
+static unsigned long __initdata mpc_new_phys;
+static unsigned long mpc_new_length __initdata = 4096;
 
-       MP_intsrc_info(&intsrc);
+/* alloc_mptable or alloc_mptable=4k */
+static int __initdata alloc_mptable;
+static int __init parse_alloc_mptable_opt(char *p)
+{
+       enable_update_mptable = 1;
+       alloc_mptable = 1;
+       if (!p)
+               return 0;
+       mpc_new_length = PAGE_SIZE << get_order(memparse(p, &p));
+       return 0;
 }
+early_param("alloc_mptable", parse_alloc_mptable_opt);
 
-void __init mp_config_acpi_legacy_irqs(void)
+void __init early_reserve_e820_mpc_new(void)
 {
-       struct mpc_config_intsrc intsrc;
-       int i = 0;
-       int ioapic = -1;
-
-#if defined (CONFIG_MCA) || defined (CONFIG_EISA)
-       /*
-        * Fabricate the legacy ISA bus (bus #31).
-        */
-       mp_bus_id_to_type[MP_ISA_BUS] = MP_BUS_ISA;
+       if (enable_update_mptable && alloc_mptable) {
+               u64 startt = PAGE_SIZE;
+#ifdef CONFIG_X86_TRAMPOLINE
+               startt = TRAMPOLINE_BASE;
 #endif
-       set_bit(MP_ISA_BUS, mp_bus_not_pci);
-       Dprintk("Bus #%d is ISA\n", MP_ISA_BUS);
-
-       /*
-        * Older generations of ES7000 have no legacy identity mappings
-        */
-       if (es7000_plat == 1)
-               return;
-
-       /*
-        * Locate the IOAPIC that manages the ISA IRQs (0-15).
-        */
-       ioapic = mp_find_ioapic(0);
-       if (ioapic < 0)
-               return;
-
-       intsrc.mpc_type = MP_INTSRC;
-       intsrc.mpc_irqflag = 0; /* Conforming */
-       intsrc.mpc_srcbus = MP_ISA_BUS;
-#ifdef CONFIG_X86_IO_APIC
-       intsrc.mpc_dstapic = mp_ioapics[ioapic].mpc_apicid;
-#endif
-       /*
-        * Use the default configuration for the IRQs 0-15.  Unless
-        * overridden by (MADT) interrupt source override entries.
-        */
-       for (i = 0; i < 16; i++) {
-               int idx;
-
-               for (idx = 0; idx < mp_irq_entries; idx++) {
-                       struct mpc_config_intsrc *irq = mp_irqs + idx;
-
-                       /* Do we already have a mapping for this ISA IRQ? */
-                       if (irq->mpc_srcbus == MP_ISA_BUS
-                           && irq->mpc_srcbusirq == i)
-                               break;
-
-                       /* Do we already have a mapping for this IOAPIC pin */
-                       if ((irq->mpc_dstapic == intsrc.mpc_dstapic) &&
-                           (irq->mpc_dstirq == i))
-                               break;
-               }
-
-               if (idx != mp_irq_entries) {
-                       printk(KERN_DEBUG "ACPI: IRQ%d used by override.\n", i);
-                       continue;       /* IRQ already used */
-               }
-
-               intsrc.mpc_irqtype = mp_INT;
-               intsrc.mpc_srcbusirq = i;       /* Identity mapped */
-               intsrc.mpc_dstirq = i;
-
-               MP_intsrc_info(&intsrc);
+               mpc_new_phys = early_reserve_e820(startt, mpc_new_length,
+                                                 mpc_new_length);
        }
 }
 
-int mp_register_gsi(u32 gsi, int triggering, int polarity)
+static int __init update_mp_table(void)
 {
-       int ioapic;
-       int ioapic_pin;
-#if defined(CONFIG_X86_32) && !defined(CONFIG_XEN)
-#define MAX_GSI_NUM    4096
-#define IRQ_COMPRESSION_START  64
+       char str[16];
+       char oem[10];
+       struct intel_mp_floating *mpf;
+       struct mp_config_table *mpc;
+       struct mp_config_table *mpc_new;
+
+       if (!enable_update_mptable)
+               return 0;
+
+       mpf = mpf_found;
+       if (!mpf)
+               return 0;
 
-       static int pci_irq = IRQ_COMPRESSION_START;
        /*
-        * Mapping between Global System Interrupts, which
-        * represent all possible interrupts, and IRQs
-        * assigned to actual devices.
+        * Now see if we need to go further.
         */
-       static int gsi_to_irq[MAX_GSI_NUM];
-#else
+       if (mpf->mpf_feature1 != 0)
+               return 0;
 
-       if (acpi_irq_model != ACPI_IRQ_MODEL_IOAPIC)
-               return gsi;
-#endif
+       if (!mpf->mpf_physptr)
+               return 0;
 
-       /* Don't set up the ACPI SCI because it's already set up */
-       if (acpi_gbl_FADT.sci_interrupt == gsi)
-               return gsi;
+       mpc = isa_bus_to_virt(mpf->mpf_physptr);
 
-       ioapic = mp_find_ioapic(gsi);
-       if (ioapic < 0) {
-               printk(KERN_WARNING "No IOAPIC for GSI %u\n", gsi);
-               return gsi;
-       }
+       if (!smp_check_mpc(mpc, oem, str))
+               return 0;
 
-       ioapic_pin = gsi - mp_ioapic_routing[ioapic].gsi_base;
+       printk(KERN_INFO "mpf: %lx\n", virt_to_phys(mpf));
+       printk(KERN_INFO "mpf_physptr: %x\n", mpf->mpf_physptr);
 
-#ifndef CONFIG_X86_32
-       if (ioapic_renumber_irq)
-               gsi = ioapic_renumber_irq(ioapic, gsi);
-#endif
+       if (mpc_new_phys && mpc->mpc_length > mpc_new_length) {
+               mpc_new_phys = 0;
+               printk(KERN_INFO "mpc_new_length is %ld, please use alloc_mptable=8k\n",
+                        mpc_new_length);
+       }
+
+       if (!mpc_new_phys) {
+               unsigned char old, new;
+               /* check if we can change the postion */
+               mpc->mpc_checksum = 0;
+               old = mpf_checksum((unsigned char *)mpc, mpc->mpc_length);
+               mpc->mpc_checksum = 0xff;
+               new = mpf_checksum((unsigned char *)mpc, mpc->mpc_length);
+               if (old == new) {
+                       printk(KERN_INFO "mpc is readonly, please try alloc_mptable instead\n");
+                       return 0;
+               }
+               printk(KERN_INFO "use in-positon replacing\n");
+       } else {
+               maddr_t mpc_new_bus;
 
-       /*
-        * Avoid pin reprogramming.  PRTs typically include entries
-        * with redundant pin->gsi mappings (but unique PCI devices);
-        * we only program the IOAPIC on the first.
-        */
-       if (ioapic_pin > MP_MAX_IOAPIC_PIN) {
-               printk(KERN_ERR "Invalid reference to IOAPIC pin "
-                      "%d-%d\n", mp_ioapic_routing[ioapic].apic_id,
-                      ioapic_pin);
-               return gsi;
-       }
-       if (test_bit(ioapic_pin, mp_ioapic_routing[ioapic].pin_programmed)) {
-               Dprintk(KERN_DEBUG "Pin %d-%d already programmed\n",
-                       mp_ioapic_routing[ioapic].apic_id, ioapic_pin);
-#if defined(CONFIG_X86_32) && !defined(CONFIG_XEN)
-               return (gsi < IRQ_COMPRESSION_START ? gsi : gsi_to_irq[gsi]);
-#else
-               return gsi;
-#endif
+               if (xen_create_contiguous_region((unsigned long)phys_to_virt(mpc_new_phys),
+                                                get_order(mpc_new_length), 32))
+                       BUG();
+               mpc_new_bus = phys_to_machine(mpc_new_phys);
+               mpf->mpf_physptr = mpc_new_bus;
+               mpc_new = phys_to_virt(mpc_new_phys);
+               memcpy(mpc_new, mpc, mpc->mpc_length);
+               mpc = mpc_new;
+               /* check if we can modify that */
+               if (mpc_new_bus - mpf->mpf_physptr) {
+                       struct intel_mp_floating *mpf_new;
+                       /* steal 16 bytes from [0, 1k) */
+                       printk(KERN_INFO "mpf new: %x\n", 0x400 - 16);
+                       mpf_new = isa_bus_to_virt(0x400 - 16);
+                       memcpy(mpf_new, mpf, 16);
+                       mpf = mpf_new;
+                       mpf->mpf_physptr = mpc_new_bus;
+               }
+               mpf->mpf_checksum = 0;
+               mpf->mpf_checksum -= mpf_checksum((unsigned char *)mpf, 16);
+               printk(KERN_INFO "mpf_physptr new: %x\n", mpf->mpf_physptr);
        }
 
-       set_bit(ioapic_pin, mp_ioapic_routing[ioapic].pin_programmed);
-#if defined(CONFIG_X86_32) && !defined(CONFIG_XEN)
        /*
-        * For GSI >= 64, use IRQ compression
+        * only replace the one with mp_INT and
+        *       MP_IRQ_TRIGGER_LEVEL|MP_IRQ_POLARITY_LOW,
+        * already in mp_irqs , stored by ... and mp_config_acpi_gsi,
+        * may need pci=routeirq for all coverage
         */
-       if ((gsi >= IRQ_COMPRESSION_START)
-           && (triggering == ACPI_LEVEL_SENSITIVE)) {
-               /*
-                * For PCI devices assign IRQs in order, avoiding gaps
-                * due to unused I/O APIC pins.
-                */
-               int irq = gsi;
-               if (gsi < MAX_GSI_NUM) {
-                       /*
-                        * Retain the VIA chipset work-around (gsi > 15), but
-                        * avoid a problem where the 8254 timer (IRQ0) is setup
-                        * via an override (so it's not on pin 0 of the ioapic),
-                        * and at the same time, the pin 0 interrupt is a PCI
-                        * type.  The gsi > 15 test could cause these two pins
-                        * to be shared as IRQ0, and they are not shareable.
-                        * So test for this condition, and if necessary, avoid
-                        * the pin collision.
-                        */
-                       gsi = pci_irq++;
-                       /*
-                        * Don't assign IRQ used by ACPI SCI
-                        */
-                       if (gsi == acpi_gbl_FADT.sci_interrupt)
-                               gsi = pci_irq++;
-                       gsi_to_irq[irq] = gsi;
-               } else {
-                       printk(KERN_ERR "GSI %u is too high\n", gsi);
-                       return gsi;
-               }
-       }
-#endif
-       io_apic_set_pci_routing(ioapic, ioapic_pin, gsi,
-                               triggering == ACPI_EDGE_SENSITIVE ? 0 : 1,
-                               polarity == ACPI_ACTIVE_HIGH ? 0 : 1);
-       return gsi;
+       replace_intsrc_all(mpc, mpc_new_phys, mpc_new_length);
+
+       return 0;
 }
 
-#endif /* CONFIG_X86_IO_APIC */
-#endif /* CONFIG_ACPI */
+late_initcall(update_mp_table);
Index: head-2008-12-01/arch/x86/kernel/nmi.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/nmi.c  2008-12-03 15:48:43.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/nmi.c       2008-12-01 11:49:07.000000000 +0100
@@ -27,7 +27,9 @@
 #include <linux/kdebug.h>
 #include <linux/smp.h>
 
+#ifndef CONFIG_XEN
 #include <asm/i8259.h>
+#endif
 #include <asm/io_apic.h>
 #include <asm/smp.h>
 #include <asm/nmi.h>
@@ -179,8 +181,10 @@ int __init check_nmi_watchdog(void)
        kfree(prev_nmi_count);
        return 0;
 error:
+#ifndef CONFIG_XEN
        if (nmi_watchdog == NMI_IO_APIC && !timer_through_8259)
                disable_8259A_irq(0);
+#endif
 #ifdef CONFIG_X86_32
        timer_ack = 0;
 #endif
Index: head-2008-12-01/arch/x86/kernel/pci-dma-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/pci-dma-xen.c  2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/pci-dma-xen.c       2008-12-01 11:49:07.000000000 +0100
@@ -5,13 +5,13 @@
 
 #include <asm/proto.h>
 #include <asm/dma.h>
-#include <asm/gart.h>
+#include <asm/iommu.h>
 #include <asm/calgary.h>
+#include <asm/amd_iommu.h>
 
-int forbid_dac __read_mostly;
-EXPORT_SYMBOL(forbid_dac);
+static int forbid_dac __read_mostly;
 
-const struct dma_mapping_ops *dma_ops;
+struct dma_mapping_ops *dma_ops;
 EXPORT_SYMBOL(dma_ops);
 
 static int iommu_sac_force __read_mostly;
@@ -74,13 +74,17 @@ early_param("dma32_size", parse_dma32_si
 void __init dma32_reserve_bootmem(void)
 {
        unsigned long size, align;
-       if (end_pfn <= MAX_DMA32_PFN)
+       if (max_pfn <= MAX_DMA32_PFN)
                return;
 
+       /*
+        * check aperture_64.c allocate_aperture() for reason about
+        * using 512M as goal
+        */
        align = 64ULL<<20;
        size = round_up(dma32_bootmem_size, align);
        dma32_bootmem_ptr = __alloc_bootmem_nopanic(size, align,
-                                __pa(MAX_DMA_ADDRESS));
+                                512ULL<<20);
        if (dma32_bootmem_ptr)
                dma32_bootmem_size = size;
        else
@@ -88,17 +92,14 @@ void __init dma32_reserve_bootmem(void)
 }
 static void __init dma32_free_bootmem(void)
 {
-       int node;
 
-       if (end_pfn <= MAX_DMA32_PFN)
+       if (max_pfn <= MAX_DMA32_PFN)
                return;
 
        if (!dma32_bootmem_ptr)
                return;
 
-       for_each_online_node(node)
-               free_bootmem_node(NODE_DATA(node), __pa(dma32_bootmem_ptr),
-                                 dma32_bootmem_size);
+       free_bootmem(__pa(dma32_bootmem_ptr), dma32_bootmem_size);
 
        dma32_bootmem_ptr = NULL;
        dma32_bootmem_size = 0;
@@ -107,7 +108,7 @@ static void __init dma32_free_bootmem(vo
 #define dma32_free_bootmem() ((void)0)
 #endif
 
-static const struct dma_mapping_ops swiotlb_dma_ops = {
+static struct dma_mapping_ops swiotlb_dma_ops = {
        .mapping_error = swiotlb_dma_mapping_error,
        .map_single = swiotlb_map_single_phys,
        .unmap_single = swiotlb_unmap_single,
@@ -130,25 +131,31 @@ void __init pci_iommu_alloc(void)
         * The order of these functions is important for
         * fall-back/fail-over reasons
         */
-#ifdef CONFIG_GART_IOMMU
        gart_iommu_hole_init();
-#endif
 
-#ifdef CONFIG_CALGARY_IOMMU
        detect_calgary();
-#endif
 
        detect_intel_iommu();
 
-#ifdef CONFIG_SWIOTLB
+       amd_iommu_detect();
+
        swiotlb_init();
        if (swiotlb) {
                printk(KERN_INFO "PCI-DMA: Using software bounce buffering for IO (SWIOTLB)\n");
                dma_ops = &swiotlb_dma_ops;
        }
-#endif
 }
 
+#ifndef CONFIG_XEN
+unsigned long iommu_num_pages(unsigned long addr, unsigned long len)
+{
+       unsigned long size = roundup((addr & ~PAGE_MASK) + len, PAGE_SIZE);
+
+       return size >> PAGE_SHIFT;
+}
+EXPORT_SYMBOL(iommu_num_pages);
+#endif
+
 /*
  * See <Documentation/x86_64/boot-options.txt> for the iommu kernel parameter
  * documentation.
@@ -201,9 +208,7 @@ static __init int iommu_setup(char *p)
                        swiotlb = 1;
 #endif
 
-#ifdef CONFIG_GART_IOMMU
                gart_parse_options(p);
-#endif
 
 #ifdef CONFIG_CALGARY_IOMMU
                if (!strncmp(p, "calgary", 7))
@@ -245,136 +250,19 @@ int range_straddles_page_boundary(paddr_
                !check_pages_physically_contiguous(pfn, offset, size));
 }
 
-#ifdef CONFIG_X86_32
-int dma_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr,
-                               dma_addr_t device_addr, size_t size, int flags)
-{
-       void __iomem *mem_base = NULL;
-       int pages = size >> PAGE_SHIFT;
-       int bitmap_size = BITS_TO_LONGS(pages) * sizeof(long);
-
-       if ((flags & (DMA_MEMORY_MAP | DMA_MEMORY_IO)) == 0)
-               goto out;
-       if (!size)
-               goto out;
-       if (dev->dma_mem)
-               goto out;
-
-       /* FIXME: this routine just ignores DMA_MEMORY_INCLUDES_CHILDREN */
-
-       mem_base = ioremap(bus_addr, size);
-       if (!mem_base)
-               goto out;
-
-       dev->dma_mem = kzalloc(sizeof(struct dma_coherent_mem), GFP_KERNEL);
-       if (!dev->dma_mem)
-               goto out;
-       dev->dma_mem->bitmap = kzalloc(bitmap_size, GFP_KERNEL);
-       if (!dev->dma_mem->bitmap)
-               goto free1_out;
-
-       dev->dma_mem->virt_base = mem_base;
-       dev->dma_mem->device_base = device_addr;
-       dev->dma_mem->size = pages;
-       dev->dma_mem->flags = flags;
-
-       if (flags & DMA_MEMORY_MAP)
-               return DMA_MEMORY_MAP;
-
-       return DMA_MEMORY_IO;
-
- free1_out:
-       kfree(dev->dma_mem);
- out:
-       if (mem_base)
-               iounmap(mem_base);
-       return 0;
-}
-EXPORT_SYMBOL(dma_declare_coherent_memory);
-
-void dma_release_declared_memory(struct device *dev)
-{
-       struct dma_coherent_mem *mem = dev->dma_mem;
-
-       if (!mem)
-               return;
-       dev->dma_mem = NULL;
-       iounmap(mem->virt_base);
-       kfree(mem->bitmap);
-       kfree(mem);
-}
-EXPORT_SYMBOL(dma_release_declared_memory);
-
-void *dma_mark_declared_memory_occupied(struct device *dev,
-                                       dma_addr_t device_addr, size_t size)
-{
-       struct dma_coherent_mem *mem = dev->dma_mem;
-       int pos, err;
-       int pages = (size + (device_addr & ~PAGE_MASK) + PAGE_SIZE - 1);
-
-       pages >>= PAGE_SHIFT;
-
-       if (!mem)
-               return ERR_PTR(-EINVAL);
-
-       pos = (device_addr - mem->device_base) >> PAGE_SHIFT;
-       err = bitmap_allocate_region(mem->bitmap, pos, get_order(pages));
-       if (err != 0)
-               return ERR_PTR(err);
-       return mem->virt_base + (pos << PAGE_SHIFT);
-}
-EXPORT_SYMBOL(dma_mark_declared_memory_occupied);
-
-static int dma_alloc_from_coherent_mem(struct device *dev, ssize_t size,
-                                      dma_addr_t *dma_handle, void **ret)
-{
-       struct dma_coherent_mem *mem = dev ? dev->dma_mem : NULL;
-       int order = get_order(size);
-
-       if (mem) {
-               int page = bitmap_find_free_region(mem->bitmap, mem->size,
-                                                    order);
-               if (page >= 0) {
-                       *dma_handle = mem->device_base + (page << PAGE_SHIFT);
-                       *ret = mem->virt_base + (page << PAGE_SHIFT);
-                       memset(*ret, 0, size);
-               }
-               if (mem->flags & DMA_MEMORY_EXCLUSIVE)
-                       *ret = NULL;
-       }
-       return (mem != NULL);
-}
-
-static int dma_release_coherent(struct device *dev, int order, void *vaddr)
-{
-       struct dma_coherent_mem *mem = dev ? dev->dma_mem : NULL;
-
-       if (mem && vaddr >= mem->virt_base && vaddr <
-                  (mem->virt_base + (mem->size << PAGE_SHIFT))) {
-               int page = (vaddr - mem->virt_base) >> PAGE_SHIFT;
-
-               bitmap_release_region(mem->bitmap, page, order);
-               return 1;
-       }
-       return 0;
-}
-#else
-#define dma_alloc_from_coherent_mem(dev, size, handle, ret) (0)
-#define dma_release_coherent(dev, order, vaddr) (0)
-#endif /* CONFIG_X86_32 */
-
 int dma_supported(struct device *dev, u64 mask)
 {
+       struct dma_mapping_ops *ops = get_dma_ops(dev);
+
 #ifdef CONFIG_PCI
        if (mask > 0xffffffff && forbid_dac > 0) {
-               printk(KERN_INFO "PCI: Disallowing DAC for device %s\n",
-                                dev->bus_id);
+               dev_info(dev, "PCI: Disallowing DAC for device\n");
                return 0;
        }
 #endif
 
-       if (dma_ops->dma_supported)
-               return dma_ops->dma_supported(dev, mask);
+       if (ops->dma_supported)
+               return ops->dma_supported(dev, mask);
 
        /* Copied from i386. Doesn't make much sense, because it will
           only work for pci_alloc_coherent.
@@ -395,8 +283,7 @@ int dma_supported(struct device *dev, u6
           type. Normally this doesn't make any difference, but gives
           more gentle handling of IOMMU overflow. */
        if (iommu_sac_force && (mask >= DMA_40BIT_MASK)) {
-               printk(KERN_INFO "%s: Force SAC with mask %Lx\n",
-                                dev->bus_id, mask);
+               dev_info(dev, "Force SAC with mask %Lx\n", mask);
                return 0;
        }
 
@@ -422,6 +309,9 @@ void *
 dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle,
                   gfp_t gfp)
 {
+#ifndef CONFIG_XEN
+       struct dma_mapping_ops *ops = get_dma_ops(dev);
+#endif
        void *memory = NULL;
        struct page *page;
        unsigned long dma_mask = 0;
@@ -431,7 +321,7 @@ dma_alloc_coherent(struct device *dev, s
        /* ignore region specifiers */
        gfp &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32);
 
-       if (dma_alloc_from_coherent_mem(dev, size, dma_handle, &memory))
+       if (dma_alloc_from_coherent(dev, size, dma_handle, &memory))
                return memory;
 
        if (!dev) {
@@ -491,8 +381,8 @@ dma_alloc_coherent(struct device *dev, s
                        /* Let low level make its own zone decisions */
                        gfp &= ~(GFP_DMA32|GFP_DMA);
 
-                       if (dma_ops->alloc_coherent)
-                               return dma_ops->alloc_coherent(dev, size,
+                       if (ops->alloc_coherent)
+                               return ops->alloc_coherent(dev, size,
                                                           dma_handle, gfp);
                        return NULL;
                }
@@ -504,14 +394,14 @@ dma_alloc_coherent(struct device *dev, s
                }
        }
 
-       if (dma_ops->alloc_coherent) {
+       if (ops->alloc_coherent) {
                free_pages((unsigned long)memory, order);
                gfp &= ~(GFP_DMA|GFP_DMA32);
-               return dma_ops->alloc_coherent(dev, size, dma_handle, gfp);
+               return ops->alloc_coherent(dev, size, dma_handle, gfp);
        }
 
-       if (dma_ops->map_simple) {
-               *dma_handle = dma_ops->map_simple(dev, virt_to_bus(memory),
+       if (ops->map_simple) {
+               *dma_handle = ops->map_simple(dev, virt_to_bus(memory),
                                              size,
                                              PCI_DMA_BIDIRECTIONAL);
                if (*dma_handle != bad_dma_address)
@@ -542,13 +432,17 @@ EXPORT_SYMBOL(dma_alloc_coherent);
 void dma_free_coherent(struct device *dev, size_t size,
                         void *vaddr, dma_addr_t bus)
 {
+#ifndef CONFIG_XEN
+       struct dma_mapping_ops *ops = get_dma_ops(dev);
+#endif
+
        int order = get_order(size);
        WARN_ON(irqs_disabled());       /* for portability */
-       if (dma_release_coherent(dev, order, vaddr))
+       if (dma_release_from_coherent(dev, order, vaddr))
                return;
 #ifndef CONFIG_XEN
-       if (dma_ops->unmap_single)
-               dma_ops->unmap_single(dev, bus, size, 0);
+       if (ops->unmap_single)
+               ops->unmap_single(dev, bus, size, 0);
 #endif
        xen_destroy_contiguous_region((unsigned long)vaddr, order);
        free_pages((unsigned long)vaddr, order);
@@ -557,15 +451,13 @@ EXPORT_SYMBOL(dma_free_coherent);
 
 static int __init pci_iommu_init(void)
 {
-#ifdef CONFIG_CALGARY_IOMMU
        calgary_iommu_init();
-#endif
 
        intel_iommu_init();
 
-#ifdef CONFIG_GART_IOMMU
+       amd_iommu_init();
+
        gart_iommu_init();
-#endif
 
        no_iommu_init();
        return 0;
Index: head-2008-12-01/arch/x86/kernel/pci-nommu-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/pci-nommu-xen.c        2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/pci-nommu-xen.c     2008-12-01 11:49:07.000000000 +0100
@@ -79,18 +79,12 @@ gnttab_unmap_single(struct device *dev, 
        gnttab_dma_unmap_page(dma_addr);
 }
 
-static int nommu_mapping_error(dma_addr_t dma_addr)
-{
-       return (dma_addr == bad_dma_address);
-}
-
-static const struct dma_mapping_ops nommu_dma_ops = {
+static struct dma_mapping_ops nommu_dma_ops = {
        .map_single = gnttab_map_single,
        .unmap_single = gnttab_unmap_single,
        .map_sg = gnttab_map_sg,
        .unmap_sg = gnttab_unmap_sg,
        .dma_supported = swiotlb_dma_supported,
-       .mapping_error = nommu_mapping_error
 };
 
 void __init no_iommu_init(void)
Index: head-2008-12-01/arch/x86/kernel/probe_roms_32.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/probe_roms_32.c        2008-12-03 15:48:43.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/probe_roms_32.c     2008-12-01 11:49:07.000000000 +0100
@@ -99,6 +99,11 @@ void __init probe_roms(void)
        unsigned char c;
        int i;
 
+#ifdef CONFIG_XEN
+       if (!is_initial_xendomain())
+               return;
+#endif
+
        /* video rom */
        upper = adapter_rom_resources[0].start;
        for (start = video_rom_resource.start; start < upper; start += 2048) {
@@ -131,7 +136,7 @@ void __init probe_roms(void)
        upper = system_rom_resource.start;
 
        /* check for extension rom (ignore length byte!) */
-       rom = isa_bus_to_virt(extension_rom_resource.start);
+       rom = isa_bus_to_virt((unsigned long)extension_rom_resource.start);
        if (romsignature(rom)) {
                length = extension_rom_resource.end - extension_rom_resource.start + 1;
                if (romchecksum(rom, length)) {
Index: head-2008-12-01/arch/x86/kernel/process-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/process-xen.c  2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/process-xen.c       2008-12-01 11:49:07.000000000 +0100
@@ -6,6 +6,13 @@
 #include <linux/sched.h>
 #include <linux/module.h>
 #include <linux/pm.h>
+#include <linux/clockchips.h>
+#include <asm/system.h>
+
+unsigned long idle_halt;
+EXPORT_SYMBOL(idle_halt);
+unsigned long idle_nomwait;
+EXPORT_SYMBOL(idle_nomwait);
 
 struct kmem_cache *task_xstate_cachep;
 
@@ -45,6 +52,70 @@ void arch_task_cache_init(void)
                                  SLAB_PANIC, NULL);
 }
 
+/*
+ * Idle related variables and functions
+ */
+unsigned long boot_option_idle_override = 0;
+EXPORT_SYMBOL(boot_option_idle_override);
+
+/*
+ * Powermanagement idle function, if any..
+ */
+void (*pm_idle)(void);
+EXPORT_SYMBOL(pm_idle);
+
+#ifdef CONFIG_X86_32
+/*
+ * This halt magic was a workaround for ancient floppy DMA
+ * wreckage. It should be safe to remove.
+ */
+static int hlt_counter;
+void disable_hlt(void)
+{
+       hlt_counter++;
+}
+EXPORT_SYMBOL(disable_hlt);
+
+void enable_hlt(void)
+{
+       hlt_counter--;
+}
+EXPORT_SYMBOL(enable_hlt);
+
+static inline int hlt_use_halt(void)
+{
+       return (!hlt_counter && boot_cpu_data.hlt_works_ok);
+}
+#else
+static inline int hlt_use_halt(void)
+{
+       return 1;
+}
+#endif
+
+/*
+ * We use this if we don't have any better
+ * idle routine..
+ */
+void xen_idle(void)
+{
+       current_thread_info()->status &= ~TS_POLLING;
+       /*
+        * TS_POLLING-cleared state must be visible before we
+        * test NEED_RESCHED:
+        */
+       smp_mb();
+
+       if (!need_resched())
+               safe_halt();    /* enables interrupts racelessly */
+       else
+               local_irq_enable();
+       current_thread_info()->status |= TS_POLLING;
+}
+#ifdef CONFIG_APM_MODULE
+EXPORT_SYMBOL(default_idle);
+#endif
+
 static void do_nothing(void *unused)
 {
 }
@@ -61,7 +132,7 @@ void cpu_idle_wait(void)
 {
        smp_mb();
        /* kick all the CPUs so that they exit out of pm_idle */
-       smp_call_function(do_nothing, NULL, 0, 1);
+       smp_call_function(do_nothing, NULL, 1);
 }
 EXPORT_SYMBOL_GPL(cpu_idle_wait);
 
@@ -125,60 +196,175 @@ static void poll_idle(void)
  *
  * idle=mwait overrides this decision and forces the usage of mwait.
  */
+static int __cpuinitdata force_mwait;
+
+#define MWAIT_INFO                     0x05
+#define MWAIT_ECX_EXTENDED_INFO                0x01
+#define MWAIT_EDX_C1                   0xf0
+
 static int __cpuinit mwait_usable(const struct cpuinfo_x86 *c)
 {
+       u32 eax, ebx, ecx, edx;
+
        if (force_mwait)
                return 1;
 
-       if (c->x86_vendor == X86_VENDOR_AMD) {
-               switch(c->x86) {
-               case 0x10:
-               case 0x11:
-                       return 0;
+       if (c->cpuid_level < MWAIT_INFO)
+               return 0;
+
+       cpuid(MWAIT_INFO, &eax, &ebx, &ecx, &edx);
+       /* Check, whether EDX has extended info about MWAIT */
+       if (!(ecx & MWAIT_ECX_EXTENDED_INFO))
+               return 1;
+
+       /*
+        * edx enumeratios MONITOR/MWAIT extensions. Check, whether
+        * C1  supports MWAIT
+        */
+       return (edx & MWAIT_EDX_C1);
+}
+
+/*
+ * Check for AMD CPUs, which have potentially C1E support
+ */
+static int __cpuinit check_c1e_idle(const struct cpuinfo_x86 *c)
+{
+       if (c->x86_vendor != X86_VENDOR_AMD)
+               return 0;
+
+       if (c->x86 < 0x0F)
+               return 0;
+
+       /* Family 0x0f models < rev F do not have C1E */
+       if (c->x86 == 0x0f && c->x86_model < 0x40)
+               return 0;
+
+       return 1;
+}
+
+static cpumask_t c1e_mask = CPU_MASK_NONE;
+static int c1e_detected;
+
+void c1e_remove_cpu(int cpu)
+{
+       cpu_clear(cpu, c1e_mask);
+}
+
+/*
+ * C1E aware idle routine. We check for C1E active in the interrupt
+ * pending message MSR. If we detect C1E, then we handle it the same
+ * way as C3 power states (local apic timer and TSC stop)
+ */
+static void c1e_idle(void)
+{
+       if (need_resched())
+               return;
+
+       if (!c1e_detected) {
+               u32 lo, hi;
+
+               rdmsr(MSR_K8_INT_PENDING_MSG, lo, hi);
+               if (lo & K8_INTP_C1E_ACTIVE_MASK) {
+                       c1e_detected = 1;
+                       if (!boot_cpu_has(X86_FEATURE_CONSTANT_TSC))
+                               mark_tsc_unstable("TSC halt in AMD C1E");
+                       printk(KERN_INFO "System has AMD C1E enabled\n");
+                       set_cpu_cap(&boot_cpu_data, X86_FEATURE_AMDC1E);
                }
        }
-       return 1;
+
+       if (c1e_detected) {
+               int cpu = smp_processor_id();
+
+               if (!cpu_isset(cpu, c1e_mask)) {
+                       cpu_set(cpu, c1e_mask);
+                       /*
+                        * Force broadcast so ACPI can not interfere. Needs
+                        * to run with interrupts enabled as it uses
+                        * smp_function_call.
+                        */
+                       local_irq_enable();
+                       clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_FORCE,
+                                          &cpu);
+                       printk(KERN_INFO "Switch to broadcast mode on CPU%d\n",
+                              cpu);
+                       local_irq_disable();
+               }
+               clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_ENTER, &cpu);
+
+               default_idle();
+
+               /*
+                * The switch back from broadcast mode needs to be
+                * called with interrupts disabled.
+                */
+                local_irq_disable();
+                clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_EXIT, &cpu);
+                local_irq_enable();
+       } else
+               default_idle();
 }
 #endif
 
 void __cpuinit select_idle_routine(const struct cpuinfo_x86 *c)
 {
 #ifndef CONFIG_XEN
-       static int selected;
-
-       if (selected)
-               return;
 #ifdef CONFIG_X86_SMP
        if (pm_idle == poll_idle && smp_num_siblings > 1) {
                printk(KERN_WARNING "WARNING: polling idle and HT enabled,"
                        " performance may degrade.\n");
        }
 #endif
+       if (pm_idle)
+               return;
+
        if (cpu_has(c, X86_FEATURE_MWAIT) && mwait_usable(c)) {
                /*
-                * Skip, if setup has overridden idle.
                 * One CPU supports mwait => All CPUs supports mwait
                 */
-               if (!pm_idle) {
-                       printk(KERN_INFO "using mwait in idle threads.\n");
-                       pm_idle = mwait_idle;
-               }
-       }
-       selected = 1;
+               printk(KERN_INFO "using mwait in idle threads.\n");
+               pm_idle = mwait_idle;
+       } else if (check_c1e_idle(c)) {
+               printk(KERN_INFO "using C1E aware idle routine\n");
+               pm_idle = c1e_idle;
+       } else
+               pm_idle = default_idle;
 #endif
 }
 
 static int __init idle_setup(char *str)
 {
+       if (!str)
+               return -EINVAL;
+
        if (!strcmp(str, "poll")) {
                printk("using polling idle threads.\n");
                pm_idle = poll_idle;
-       }
 #ifndef CONFIG_XEN
-       else if (!strcmp(str, "mwait"))
+       } else if (!strcmp(str, "mwait"))
                force_mwait = 1;
+       else if (!strcmp(str, "halt")) {
+               /*
+                * When the boot option of idle=halt is added, halt is
+                * forced to be used for CPU idle. In such case CPU C2/C3
+                * won't be used again.
+                * To continue to load the CPU idle driver, don't touch
+                * the boot_option_idle_override.
+                */
+               pm_idle = default_idle;
+               idle_halt = 1;
+               return 0;
+       } else if (!strcmp(str, "nomwait")) {
+               /*
+                * If the boot option of "idle=nomwait" is added,
+                * it means that mwait will be disabled for CPU C2/C3
+                * states. In such case it won't touch the variable
+                * of boot_option_idle_override.
+                */
+               idle_nomwait = 1;
+               return 0;
 #endif
-       else
+       } else
                return -1;
 
        boot_option_idle_override = 1;
Index: head-2008-12-01/arch/x86/kernel/process_32-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/process_32-xen.c       2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/process_32-xen.c    2008-12-01 11:49:07.000000000 +0100
@@ -59,15 +59,11 @@
 #include <asm/tlbflush.h>
 #include <asm/cpu.h>
 #include <asm/kdebug.h>
+#include <asm/idle.h>
 
 asmlinkage void ret_from_fork(void) __asm__("ret_from_fork");
 asmlinkage void cstar_ret_from_fork(void) __asm__("cstar_ret_from_fork");
 
-static int hlt_counter;
-
-unsigned long boot_option_idle_override = 0;
-EXPORT_SYMBOL(boot_option_idle_override);
-
 DEFINE_PER_CPU(struct task_struct *, current_task) = &init_task;
 EXPORT_PER_CPU_SYMBOL(current_task);
 
@@ -82,46 +78,27 @@ unsigned long thread_saved_pc(struct tas
        return ((unsigned long *)tsk->thread.sp)[3];
 }
 
-/*
- * Powermanagement idle function, if any..
- */
-void (*pm_idle)(void);
-EXPORT_SYMBOL(pm_idle);
+#ifdef CONFIG_HOTPLUG_CPU
+#ifndef CONFIG_XEN
+#include <asm/nmi.h>
 
-void disable_hlt(void)
+static void cpu_exit_clear(void)
 {
-       hlt_counter++;
-}
+       int cpu = raw_smp_processor_id();
 
-EXPORT_SYMBOL(disable_hlt);
-
-void enable_hlt(void)
-{
-       hlt_counter--;
-}
+       idle_task_exit();
 
-EXPORT_SYMBOL(enable_hlt);
+       cpu_uninit();
+       irq_ctx_exit(cpu);
 
-static void xen_idle(void)
-{
-       current_thread_info()->status &= ~TS_POLLING;
-       /*
-        * TS_POLLING-cleared state must be visible before we
-        * test NEED_RESCHED:
-        */
-       smp_mb();
+       cpu_clear(cpu, cpu_callout_map);
+       cpu_clear(cpu, cpu_callin_map);
 
-       if (!need_resched())
-               safe_halt();    /* enables interrupts racelessly */
-       else
-               local_irq_enable();
-       current_thread_info()->status |= TS_POLLING;
+       numa_remove_cpu(cpu);
+       c1e_remove_cpu(cpu);
 }
-#ifdef CONFIG_APM_MODULE
-EXPORT_SYMBOL(default_idle);
 #endif
 
-#ifdef CONFIG_HOTPLUG_CPU
 static inline void play_dead(void)
 {
        idle_task_exit();
@@ -152,13 +129,11 @@ void cpu_idle(void)
 
        /* endless idle loop with no priority at all */
        while (1) {
-               tick_nohz_stop_sched_tick();
+               tick_nohz_stop_sched_tick(1);
                while (!need_resched()) {
-                       void (*idle)(void);
 
                        check_pgt_cache();
                        rmb();
-                       idle = xen_idle; /* no alternatives */
 
                        if (rcu_pending(cpu))
                                rcu_check_callbacks(cpu, 0);
@@ -168,7 +143,10 @@ void cpu_idle(void)
 
                        local_irq_disable();
                        __get_cpu_var(irq_stat).idle_timestamp = jiffies;
-                       idle();
+                       /* Don't trace irqs off for idle */
+                       stop_critical_timings();
+                       xen_idle();
+                       start_critical_timings();
                }
                tick_nohz_restart_sched_tick();
                preempt_enable_no_resched();
Index: head-2008-12-01/arch/x86/kernel/process_64-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/process_64-xen.c       2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/process_64-xen.c    2008-12-01 11:49:07.000000000 +0100
@@ -65,15 +65,6 @@ asmlinkage extern void ret_from_fork(voi
 
 unsigned long kernel_thread_flags = CLONE_VM | CLONE_UNTRACED;
 
-unsigned long boot_option_idle_override = 0;
-EXPORT_SYMBOL(boot_option_idle_override);
-
-/*
- * Powermanagement idle function, if any..
- */
-void (*pm_idle)(void);
-EXPORT_SYMBOL(pm_idle);
-
 static ATOMIC_NOTIFIER_HEAD(idle_notifier);
 
 void idle_notifier_register(struct notifier_block *n)
@@ -103,25 +94,13 @@ void exit_idle(void)
        __exit_idle();
 }
 
-static void xen_idle(void)
-{
-       current_thread_info()->status &= ~TS_POLLING;
-       /*
-        * TS_POLLING-cleared state must be visible before we
-        * test NEED_RESCHED:
-        */
-       smp_mb();
-       if (!need_resched())
-               safe_halt();    /* enables interrupts racelessly */
-       else
-               local_irq_enable();
-       current_thread_info()->status |= TS_POLLING;
-}
-
 #ifdef CONFIG_HOTPLUG_CPU
 static inline void play_dead(void)
 {
        idle_task_exit();
+#ifndef CONFIG_XEN
+       c1e_remove_cpu(raw_smp_processor_id());
+#endif
        local_irq_disable();
        cpu_clear(smp_processor_id(), cpu_initialized);
        preempt_enable_no_resched();
@@ -146,12 +125,11 @@ void cpu_idle(void)
        current_thread_info()->status |= TS_POLLING;
        /* endless idle loop with no priority at all */
        while (1) {
-               tick_nohz_stop_sched_tick();
+               tick_nohz_stop_sched_tick(1);
                while (!need_resched()) {
-                       void (*idle)(void);
 
                        rmb();
-                       idle = xen_idle; /* no alternatives */
+
                        if (cpu_is_offline(smp_processor_id()))
                                play_dead();
                        /*
@@ -161,7 +139,10 @@ void cpu_idle(void)
                         */
                        local_irq_disable();
                        enter_idle();
-                       idle();
+                       /* Don't trace irqs off for idle */
+                       stop_critical_timings();
+                       xen_idle();
+                       start_critical_timings();
                        /* In many cases the interrupt that ended idle
                           has already called exit_idle. But some idle
                           loops can be woken up without interrupt. */
@@ -271,7 +252,7 @@ void exit_thread(void)
        }
 }
 
-void load_gs_index(unsigned gs)
+void xen_load_gs_index(unsigned gs)
 {
        WARN_ON(HYPERVISOR_set_segment_base(SEGBASE_GS_USER_SEL, gs));
 }
@@ -373,10 +354,10 @@ int copy_thread(int nr, unsigned long cl
        p->thread.fs = me->thread.fs;
        p->thread.gs = me->thread.gs;
 
-       asm("mov %%gs,%0" : "=m" (p->thread.gsindex));
-       asm("mov %%fs,%0" : "=m" (p->thread.fsindex));
-       asm("mov %%es,%0" : "=m" (p->thread.es));
-       asm("mov %%ds,%0" : "=m" (p->thread.ds));
+       savesegment(gs, p->thread.gsindex);
+       savesegment(fs, p->thread.fsindex);
+       savesegment(es, p->thread.es);
+       savesegment(ds, p->thread.ds);
 
        if (unlikely(test_tsk_thread_flag(me, TIF_IO_BITMAP))) {
                p->thread.io_bitmap_ptr = kmalloc(IO_BITMAP_BYTES, GFP_KERNEL);
@@ -417,7 +398,9 @@ out:
 void
 start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp)
 {
-       asm volatile("movl %0, %%fs; movl %0, %%es; movl %0, %%ds" :: "r"(0));
+       loadsegment(fs, 0);
+       loadsegment(es, 0);
+       loadsegment(ds, 0);
        load_gs_index(0);
        regs->ip                = new_ip;
        regs->sp                = new_sp;
@@ -557,8 +540,8 @@ static inline void __switch_to_xtra(stru
 struct task_struct *
 __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 {
-       struct thread_struct *prev = &prev_p->thread,
-                                *next = &next_p->thread;
+       struct thread_struct *prev = &prev_p->thread;
+       struct thread_struct *next = &next_p->thread;
        int cpu = smp_processor_id();
 #ifndef CONFIG_X86_NO_TSS
        struct tss_struct *tss = &per_cpu(init_tss, cpu);
@@ -661,12 +644,25 @@ __switch_to(struct task_struct *prev_p, 
         */
        if (unlikely(next->es))
                loadsegment(es, next->es); 
-       
+
        if (unlikely(next->ds))
                loadsegment(ds, next->ds);
 
+       /*
+        * Leave lazy mode, flushing any hypercalls made here.
+        * This must be done before restoring TLS segments so
+        * the GDT and LDT are properly updated, and must be
+        * done before math_state_restore, so the TS bit is up
+        * to date.
+        */
+       arch_leave_lazy_cpu_mode();
+
        /* 
         * Switch FS and GS.
+        *
+        * Segment register != 0 always requires a reload.  Also
+        * reload when it has changed.  When prev process used 64bit
+        * base always reload to avoid an information leak.
         */
        if (unlikely(next->fsindex))
                loadsegment(fs, next->fsindex);
@@ -687,7 +683,8 @@ __switch_to(struct task_struct *prev_p, 
        write_pda(oldrsp, next->usersp);
        write_pda(pcurrent, next_p); 
        write_pda(kernelstack,
-       (unsigned long)task_stack_page(next_p) + THREAD_SIZE - PDA_STACKOFFSET);
+                 (unsigned long)task_stack_page(next_p) +
+                 THREAD_SIZE - PDA_STACKOFFSET);
 #ifdef CONFIG_CC_STACKPROTECTOR
        write_pda(stack_canary, next_p->stack_canary);
 
@@ -848,7 +845,7 @@ long do_arch_prctl(struct task_struct *t
                        set_32bit_tls(task, FS_TLS, addr);
                        if (doit) {
                                load_TLS(&task->thread, cpu);
-                               asm volatile("movl %0,%%fs" :: "r"(FS_TLS_SEL));
+                               loadsegment(fs, FS_TLS_SEL);
                        }
                        task->thread.fsindex = FS_TLS_SEL;
                        task->thread.fs = 0;
@@ -858,7 +855,7 @@ long do_arch_prctl(struct task_struct *t
                        if (doit) {
                                /* set the selector to 0 to not confuse
                                   __switch_to */
-                               asm volatile("movl %0,%%fs" :: "r" (0));
+                               loadsegment(fs, 0);
                                 ret = HYPERVISOR_set_segment_base(SEGBASE_FS,
                                                                  addr);
                        }
@@ -882,7 +879,7 @@ long do_arch_prctl(struct task_struct *t
                if (task->thread.gsindex == GS_TLS_SEL)
                        base = read_32bit_tls(task, GS_TLS);
                else if (doit) {
-                       asm("movl %%gs,%0" : "=r" (gsindex));
+                       savesegment(gs, gsindex);
                        if (gsindex)
                                rdmsrl(MSR_KERNEL_GS_BASE, base);
                        else
Index: head-2008-12-01/arch/x86/kernel/quirks-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/quirks-xen.c   2008-12-01 11:37:10.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/quirks-xen.c        2008-12-01 11:49:07.000000000 +0100
@@ -63,6 +63,7 @@ static enum {
        ICH_FORCE_HPET_RESUME,
        VT8237_FORCE_HPET_RESUME,
        NVIDIA_FORCE_HPET_RESUME,
+       ATI_FORCE_HPET_RESUME,
 } force_hpet_resume_type;
 
 static void __iomem *rcba_base;
@@ -156,6 +157,8 @@ static void ich_force_enable_hpet(struct
 
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ESB2_0,
                         ich_force_enable_hpet);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH6_0,
+                        ich_force_enable_hpet);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH6_1,
                         ich_force_enable_hpet);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH7_0,
@@ -172,6 +175,12 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_I
 
 static struct pci_dev *cached_dev;
 
+static void hpet_print_force_info(void)
+{
+       printk(KERN_INFO "HPET not enabled in BIOS. "
+              "You might try hpet=force boot option\n");
+}
+
 static void old_ich_force_hpet_resume(void)
 {
        u32 val;
@@ -251,8 +260,12 @@ static void old_ich_force_enable_hpet_us
 {
        if (hpet_force_user)
                old_ich_force_enable_hpet(dev);
+       else
+               hpet_print_force_info();
 }
 
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ESB_1,
+                        old_ich_force_enable_hpet_user);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801CA_0,
                         old_ich_force_enable_hpet_user);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801CA_12,
@@ -288,9 +301,14 @@ static void vt8237_force_enable_hpet(str
 {
        u32 uninitialized_var(val);
 
-       if (!hpet_force_user || hpet_address || force_hpet_address)
+       if (hpet_address || force_hpet_address)
                return;
 
+       if (!hpet_force_user) {
+               hpet_print_force_info();
+               return;
+       }
+
        pci_read_config_dword(dev, 0x68, &val);
        /*
         * Bit 7 is HPET enable bit.
@@ -328,6 +346,36 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_V
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8237,
                         vt8237_force_enable_hpet);
 
+static void ati_force_hpet_resume(void)
+{
+       pci_write_config_dword(cached_dev, 0x14, 0xfed00000);
+       printk(KERN_DEBUG "Force enabled HPET at resume\n");
+}
+
+static void ati_force_enable_hpet(struct pci_dev *dev)
+{
+       u32 uninitialized_var(val);
+
+       if (hpet_address || force_hpet_address)
+               return;
+
+       if (!hpet_force_user) {
+               hpet_print_force_info();
+               return;
+       }
+
+       pci_write_config_dword(dev, 0x14, 0xfed00000);
+       pci_read_config_dword(dev, 0x14, &val);
+       force_hpet_address = val;
+       force_hpet_resume_type = ATI_FORCE_HPET_RESUME;
+       dev_printk(KERN_DEBUG, &dev->dev, "Force enabled HPET at 0x%lx\n",
+                  force_hpet_address);
+       cached_dev = dev;
+       return;
+}
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_IXP400_SMBUS,
+                        ati_force_enable_hpet);
+
 /*
  * Undocumented chipset feature taken from LinuxBIOS.
  */
@@ -341,8 +389,13 @@ static void nvidia_force_enable_hpet(str
 {
        u32 uninitialized_var(val);
 
-       if (!hpet_force_user || hpet_address || force_hpet_address)
+       if (hpet_address || force_hpet_address)
+               return;
+
+       if (!hpet_force_user) {
+               hpet_print_force_info();
                return;
+       }
 
        pci_write_config_dword(dev, 0x44, 0xfed00001);
        pci_read_config_dword(dev, 0x44, &val);
@@ -395,6 +448,9 @@ void force_hpet_resume(void)
        case NVIDIA_FORCE_HPET_RESUME:
                nvidia_force_hpet_resume();
                return;
+       case ATI_FORCE_HPET_RESUME:
+               ati_force_hpet_resume();
+               return;
        default:
                break;
        }
Index: head-2008-12-01/arch/x86/kernel/setup-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/setup-xen.c    2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/setup-xen.c 2008-12-01 11:49:07.000000000 +0100
@@ -1,141 +1,1147 @@
-#include <linux/kernel.h>
+/*
+ *  Copyright (C) 1995  Linus Torvalds
+ *
+ *  Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999
+ *
+ *  Memory region support
+ *     David Parsons <orc@pell.chi.il.us>, July-August 1999
+ *
+ *  Added E820 sanitization routine (removes overlapping memory regions);
+ *  Brian Moyle <bmoyle@mvista.com>, February 2001
+ *
+ * Moved CPU detection code to cpu/${cpu}.c
+ *    Patrick Mochel <mochel@osdl.org>, March 2002
+ *
+ *  Provisions for empty E820 memory regions (reported by certain BIOSes).
+ *  Alex Achenbach <xela@slit.de>, December 2002.
+ *
+ */
+
+/*
+ * This file handles the architecture-dependent parts of initialization
+ */
+
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/mmzone.h>
+#include <linux/screen_info.h>
+#include <linux/ioport.h>
+#include <linux/acpi.h>
+#include <linux/apm_bios.h>
+#include <linux/initrd.h>
+#include <linux/bootmem.h>
+#include <linux/seq_file.h>
+#include <linux/console.h>
+#include <linux/mca.h>
+#include <linux/root_dev.h>
+#include <linux/highmem.h>
 #include <linux/module.h>
+#include <linux/efi.h>
 #include <linux/init.h>
-#include <linux/bootmem.h>
+#include <linux/edd.h>
+#include <linux/iscsi_ibft.h>
+#include <linux/nodemask.h>
+#include <linux/kexec.h>
+#include <linux/dmi.h>
+#include <linux/pfn.h>
+#include <linux/pci.h>
+#include <asm/pci-direct.h>
+#include <linux/init_ohci1394_dma.h>
+#include <linux/kvm_para.h>
+
+#include <linux/errno.h>
+#include <linux/kernel.h>
+#include <linux/stddef.h>
+#include <linux/unistd.h>
+#include <linux/ptrace.h>
+#include <linux/slab.h>
+#include <linux/user.h>
+#include <linux/delay.h>
+
+#include <linux/kallsyms.h>
+#include <linux/cpufreq.h>
+#include <linux/dma-mapping.h>
+#include <linux/ctype.h>
+#include <linux/uaccess.h>
+
 #include <linux/percpu.h>
-#include <asm/smp.h>
-#include <asm/percpu.h>
+#include <linux/crash_dump.h>
+
+#include <video/edid.h>
+
+#include <asm/mtrr.h>
+#include <asm/apic.h>
+#include <asm/e820.h>
+#include <asm/mpspec.h>
+#include <asm/setup.h>
+#include <asm/arch_hooks.h>
+#include <asm/efi.h>
 #include <asm/sections.h>
+#include <asm/dmi.h>
+#include <asm/io_apic.h>
+#include <asm/ist.h>
+#include <asm/vmi.h>
+#include <setup_arch.h>
+#include <asm/bios_ebda.h>
+#include <asm/cacheflush.h>
 #include <asm/processor.h>
-#include <asm/setup.h>
+#include <asm/bugs.h>
+
+#include <asm/system.h>
+#include <asm/vsyscall.h>
+#include <asm/smp.h>
+#include <asm/desc.h>
+#include <asm/dma.h>
+#include <asm/iommu.h>
+#include <asm/mmu_context.h>
+#include <asm/proto.h>
+
+#include <mach_apic.h>
+#include <asm/paravirt.h>
+
+#include <asm/percpu.h>
 #include <asm/topology.h>
-#include <asm/mpspec.h>
 #include <asm/apicdef.h>
+#ifdef CONFIG_X86_64
+#include <asm/numa_64.h>
+#endif
+
+#ifdef CONFIG_XEN
+#include <asm/hypervisor.h>
+#include <xen/interface/kexec.h>
+#include <xen/interface/memory.h>
+#include <xen/interface/nmi.h>
+#include <xen/interface/physdev.h>
+#include <xen/features.h>
+#include <xen/firmware.h>
+#include <xen/xencons.h>
+
+shared_info_t *HYPERVISOR_shared_info = (shared_info_t *)empty_zero_page;
+EXPORT_SYMBOL(HYPERVISOR_shared_info);
 
-#ifdef CONFIG_X86_LOCAL_APIC
-unsigned int num_processors;
-unsigned disabled_cpus __cpuinitdata;
-/* Processor that is doing the boot up */
-unsigned int boot_cpu_physical_apicid = -1U;
-EXPORT_SYMBOL(boot_cpu_physical_apicid);
+static int xen_panic_event(struct notifier_block *, unsigned long, void *);
+static struct notifier_block xen_panic_block = {
+       xen_panic_event, NULL, 0 /* try to go last */
+};
 
-DEFINE_PER_CPU(u16, x86_cpu_to_apicid) = BAD_APICID;
-EXPORT_PER_CPU_SYMBOL(x86_cpu_to_apicid);
+unsigned long *phys_to_machine_mapping;
+EXPORT_SYMBOL(phys_to_machine_mapping);
 
-/* Bitmask of physically existing CPUs */
-physid_mask_t phys_cpu_present_map;
+unsigned long *pfn_to_mfn_frame_list_list,
+#ifdef CONFIG_X86_64
+       *pfn_to_mfn_frame_list[512];
+#else
+       *pfn_to_mfn_frame_list[128];
+#endif
+
+/* Raw start-of-day parameters from the hypervisor. */
+start_info_t *xen_start_info;
+EXPORT_SYMBOL(xen_start_info);
+#endif
+
+#ifndef ARCH_SETUP
+#define ARCH_SETUP
+#endif
+
+#ifndef CONFIG_XEN
+#ifndef CONFIG_DEBUG_BOOT_PARAMS
+struct boot_params __initdata boot_params;
+#else
+struct boot_params boot_params;
+#endif
 #endif
 
-#if defined(CONFIG_HAVE_SETUP_PER_CPU_AREA) && defined(CONFIG_X86_SMP)
 /*
- * Copy data used in early init routines from the initial arrays to the
- * per cpu data areas.  These arrays then become expendable and the
- * *_early_ptr's are zeroed indicating that the static arrays are gone.
+ * Machine setup..
  */
-static void __init setup_per_cpu_maps(void)
+static struct resource data_resource = {
+       .name   = "Kernel data",
+       .start  = 0,
+       .end    = 0,
+       .flags  = IORESOURCE_BUSY | IORESOURCE_MEM
+};
+
+static struct resource code_resource = {
+       .name   = "Kernel code",
+       .start  = 0,
+       .end    = 0,
+       .flags  = IORESOURCE_BUSY | IORESOURCE_MEM
+};
+
+static struct resource bss_resource = {
+       .name   = "Kernel bss",
+       .start  = 0,
+       .end    = 0,
+       .flags  = IORESOURCE_BUSY | IORESOURCE_MEM
+};
+
+
+#ifdef CONFIG_X86_32
+#ifndef CONFIG_XEN
+/* This value is set up by the early boot code to point to the value
+   immediately after the boot time page tables.  It contains a *physical*
+   address, and must not be in the .bss segment! */
+unsigned long init_pg_tables_start __initdata = ~0UL;
+unsigned long init_pg_tables_end __initdata = ~0UL;
+#endif
+
+static struct resource video_ram_resource = {
+       .name   = "Video RAM area",
+       .start  = 0xa0000,
+       .end    = 0xbffff,
+       .flags  = IORESOURCE_BUSY | IORESOURCE_MEM
+};
+
+/* cpu data as detected by the assembly code in head.S */
+struct cpuinfo_x86 new_cpu_data __cpuinitdata = {0, 0, 0, 0, -1, 1, 0, 0, -1};
+/* common cpu data for all cpus */
+struct cpuinfo_x86 boot_cpu_data __read_mostly = {0, 0, 0, 0, -1, 1, 0, 0, -1};
+EXPORT_SYMBOL(boot_cpu_data);
+#ifndef CONFIG_XEN
+static void set_mca_bus(int x)
 {
+#ifdef CONFIG_MCA
+       MCA_bus = x;
+#endif
+}
+
+unsigned int def_to_bigsmp;
+
+/* for MCA, but anyone else can use it if they want */
+unsigned int machine_id;
+unsigned int machine_submodel_id;
+unsigned int BIOS_revision;
+
+struct apm_info apm_info;
+EXPORT_SYMBOL(apm_info);
+#endif
+
+#if defined(CONFIG_X86_SPEEDSTEP_SMI_MODULE)
+struct ist_info ist_info;
+EXPORT_SYMBOL(ist_info);
+#elif defined(CONFIG_X86_SPEEDSTEP_SMI)
+struct ist_info ist_info;
+#endif
+
+#else
+struct cpuinfo_x86 boot_cpu_data __read_mostly;
+EXPORT_SYMBOL(boot_cpu_data);
+#endif
+
+
+#if !defined(CONFIG_X86_PAE) || defined(CONFIG_X86_64)
+unsigned long mmu_cr4_features;
+#else
+unsigned long mmu_cr4_features = X86_CR4_PAE;
+#endif
+
+/* Boot loader ID as an integer, for the benefit of proc_dointvec */
+int bootloader_type;
+
+/*
+ * Early DMI memory
+ */
+int dmi_alloc_index;
+char dmi_alloc_data[DMI_MAX_DATA];
+
+/*
+ * Setup options
+ */
+struct screen_info screen_info;
+EXPORT_SYMBOL(screen_info);
+struct edid_info edid_info;
+EXPORT_SYMBOL_GPL(edid_info);
+
+extern int root_mountflags;
+
+unsigned long saved_video_mode;
+
+#define RAMDISK_IMAGE_START_MASK       0x07FF
+#define RAMDISK_PROMPT_FLAG            0x8000
+#define RAMDISK_LOAD_FLAG              0x4000
+
+static char __initdata command_line[COMMAND_LINE_SIZE];
+
+#if defined(CONFIG_EDD) || defined(CONFIG_EDD_MODULE)
+struct edd edd;
+#ifdef CONFIG_EDD_MODULE
+EXPORT_SYMBOL(edd);
+#endif
 #ifndef CONFIG_XEN
-       int cpu;
+/**
+ * copy_edd() - Copy the BIOS EDD information
+ *              from boot_params into a safe place.
+ *
+ */
+static inline void copy_edd(void)
+{
+     memcpy(edd.mbr_signature, boot_params.edd_mbr_sig_buffer,
+           sizeof(edd.mbr_signature));
+     memcpy(edd.edd_info, boot_params.eddbuf, sizeof(edd.edd_info));
+     edd.mbr_signature_nr = boot_params.edd_mbr_sig_buf_entries;
+     edd.edd_info_nr = boot_params.eddbuf_entries;
+}
+#endif
+#else
+static inline void copy_edd(void)
+{
+}
+#endif
+
+#ifdef CONFIG_BLK_DEV_INITRD
+
+#if defined(CONFIG_X86_32) && !defined(CONFIG_XEN)
+
+#define MAX_MAP_CHUNK  (NR_FIX_BTMAPS << PAGE_SHIFT)
+static void __init relocate_initrd(void)
+{
+
+       u64 ramdisk_image = boot_params.hdr.ramdisk_image;
+       u64 ramdisk_size  = boot_params.hdr.ramdisk_size;
+       u64 end_of_lowmem = max_low_pfn << PAGE_SHIFT;
+       u64 ramdisk_here;
+       unsigned long slop, clen, mapaddr;
+       char *p, *q;
+
+       /* We need to move the initrd down into lowmem */
+       ramdisk_here = find_e820_area(0, end_of_lowmem, ramdisk_size,
+                                        PAGE_SIZE);
+
+       if (ramdisk_here == -1ULL)
+               panic("Cannot find place for new RAMDISK of size %lld\n",
+                        ramdisk_size);
+
+       /* Note: this includes all the lowmem currently occupied by
+          the initrd, we rely on that fact to keep the data intact. */
+       reserve_early(ramdisk_here, ramdisk_here + ramdisk_size,
+                        "NEW RAMDISK");
+       initrd_start = ramdisk_here + PAGE_OFFSET;
+       initrd_end   = initrd_start + ramdisk_size;
+       printk(KERN_INFO "Allocated new RAMDISK: %08llx - %08llx\n",
+                        ramdisk_here, ramdisk_here + ramdisk_size);
+
+       q = (char *)initrd_start;
+
+       /* Copy any lowmem portion of the initrd */
+       if (ramdisk_image < end_of_lowmem) {
+               clen = end_of_lowmem - ramdisk_image;
+               p = (char *)__va(ramdisk_image);
+               memcpy(q, p, clen);
+               q += clen;
+               ramdisk_image += clen;
+               ramdisk_size  -= clen;
+       }
+
+       /* Copy the highmem portion of the initrd */
+       while (ramdisk_size) {
+               slop = ramdisk_image & ~PAGE_MASK;
+               clen = ramdisk_size;
+               if (clen > MAX_MAP_CHUNK-slop)
+                       clen = MAX_MAP_CHUNK-slop;
+               mapaddr = ramdisk_image & PAGE_MASK;
+               p = early_ioremap(mapaddr, clen+slop);
+               memcpy(q, p+slop, clen);
+               early_iounmap(p, clen+slop);
+               q += clen;
+               ramdisk_image += clen;
+               ramdisk_size  -= clen;
+       }
+       /* high pages is not converted by early_res_to_bootmem */
+       ramdisk_image = boot_params.hdr.ramdisk_image;
+       ramdisk_size  = boot_params.hdr.ramdisk_size;
+       printk(KERN_INFO "Move RAMDISK from %016llx - %016llx to"
+               " %08llx - %08llx\n",
+               ramdisk_image, ramdisk_image + ramdisk_size - 1,
+               ramdisk_here, ramdisk_here + ramdisk_size - 1);
+}
+#endif
+
+static void __init reserve_initrd(void)
+{
+#ifndef CONFIG_XEN
+       u64 ramdisk_image = boot_params.hdr.ramdisk_image;
+       u64 ramdisk_size  = boot_params.hdr.ramdisk_size;
+       u64 ramdisk_end   = ramdisk_image + ramdisk_size;
+       u64 end_of_lowmem = max_low_pfn << PAGE_SHIFT;
 
-       for_each_possible_cpu(cpu) {
-               per_cpu(x86_cpu_to_apicid, cpu) = x86_cpu_to_apicid_init[cpu];
-               per_cpu(x86_bios_cpu_apicid, cpu) =
-                                               x86_bios_cpu_apicid_init[cpu];
-#ifdef CONFIG_NUMA
-               per_cpu(x86_cpu_to_node_map, cpu) =
-                                               x86_cpu_to_node_map_init[cpu];
+       if (!boot_params.hdr.type_of_loader ||
+           !ramdisk_image || !ramdisk_size)
+               return;         /* No initrd provided by bootloader */
+#else
+       unsigned long ramdisk_image = __pa(xen_start_info->mod_start);
+       unsigned long ramdisk_size  = xen_start_info->mod_len;
+       unsigned long ramdisk_end   = ramdisk_image + ramdisk_size;
+       unsigned long end_of_lowmem = max_low_pfn << PAGE_SHIFT;
+
+       if (!xen_start_info->mod_start || !ramdisk_size)
+               return;         /* No initrd provided by bootloader */
 #endif
+
+       initrd_start = 0;
+
+       if (ramdisk_size >= (end_of_lowmem>>1)) {
+               free_early(ramdisk_image, ramdisk_end);
+               printk(KERN_ERR "initrd too large to handle, "
+                      "disabling initrd\n");
+               return;
        }
 
-       /* indicate the early static arrays will soon be gone */
-       x86_cpu_to_apicid_early_ptr = NULL;
-       x86_bios_cpu_apicid_early_ptr = NULL;
-#ifdef CONFIG_NUMA
-       x86_cpu_to_node_map_early_ptr = NULL;
+       printk(KERN_INFO "RAMDISK: %08lx - %08lx\n", ramdisk_image,
+                       ramdisk_end);
+
+
+       if (ramdisk_end <= end_of_lowmem) {
+               /* All in lowmem, easy case */
+               /*
+                * don't need to reserve again, already reserved early
+                * in i386_start_kernel
+                */
+               initrd_start = ramdisk_image + PAGE_OFFSET;
+               initrd_end = initrd_start + ramdisk_size;
+#ifdef CONFIG_X86_64_XEN
+               initrd_below_start_ok = 1;
 #endif
+               return;
+       }
+
+#if defined(CONFIG_X86_32) && !defined(CONFIG_XEN)
+       relocate_initrd();
+#else
+       printk(KERN_ERR "initrd extends beyond end of memory "
+              "(0x%08lx > 0x%08lx)\ndisabling initrd\n",
+              ramdisk_end, end_of_lowmem);
+       initrd_start = 0;
 #endif
+       free_early(ramdisk_image, ramdisk_end);
 }
+#else
+static void __init reserve_initrd(void)
+{
+}
+#endif /* CONFIG_BLK_DEV_INITRD */
 
-#ifdef CONFIG_HAVE_CPUMASK_OF_CPU_MAP
-cpumask_t *cpumask_of_cpu_map __read_mostly;
-EXPORT_SYMBOL(cpumask_of_cpu_map);
+static void __init parse_setup_data(void)
+{
+#ifndef CONFIG_XEN
+       struct setup_data *data;
+       u64 pa_data;
+
+       if (boot_params.hdr.version < 0x0209)
+               return;
+       pa_data = boot_params.hdr.setup_data;
+       while (pa_data) {
+               data = early_ioremap(pa_data, PAGE_SIZE);
+               switch (data->type) {
+               case SETUP_E820_EXT:
+                       parse_e820_ext(data, pa_data);
+                       break;
+               default:
+                       break;
+               }
+               pa_data = data->next;
+               early_iounmap(data, PAGE_SIZE);
+       }
+#endif
+}
 
-/* requires nr_cpu_ids to be initialized */
-static void __init setup_cpumask_of_cpu(void)
+static void __init e820_reserve_setup_data(void)
 {
-       int i;
+#ifndef CONFIG_XEN
+       struct setup_data *data;
+       u64 pa_data;
+       int found = 0;
 
-       /* alloc_bootmem zeroes memory */
-       cpumask_of_cpu_map = alloc_bootmem_low(sizeof(cpumask_t) * nr_cpu_ids);
-       for (i = 0; i < nr_cpu_ids; i++)
-               cpu_set(i, cpumask_of_cpu_map[i]);
+       if (boot_params.hdr.version < 0x0209)
+               return;
+       pa_data = boot_params.hdr.setup_data;
+       while (pa_data) {
+               data = early_ioremap(pa_data, sizeof(*data));
+               e820_update_range(pa_data, sizeof(*data)+data->len,
+                        E820_RAM, E820_RESERVED_KERN);
+               found = 1;
+               pa_data = data->next;
+               early_iounmap(data, sizeof(*data));
+       }
+       if (!found)
+               return;
+
+       sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);
+       memcpy(&e820_saved, &e820, sizeof(struct e820map));
+       printk(KERN_INFO "extended physical RAM map:\n");
+       e820_print_map("reserve setup_data");
+#endif
 }
-#else
-static inline void setup_cpumask_of_cpu(void) { }
+
+static void __init reserve_early_setup_data(void)
+{
+#ifndef CONFIG_XEN
+       struct setup_data *data;
+       u64 pa_data;
+       char buf[32];
+
+       if (boot_params.hdr.version < 0x0209)
+               return;
+       pa_data = boot_params.hdr.setup_data;
+       while (pa_data) {
+               data = early_ioremap(pa_data, sizeof(*data));
+               sprintf(buf, "setup data %x", data->type);
+               reserve_early(pa_data, pa_data+sizeof(*data)+data->len, buf);
+               pa_data = data->next;
+               early_iounmap(data, sizeof(*data));
+       }
 #endif
+}
 
-#ifdef CONFIG_X86_32
 /*
- * Great future not-so-futuristic plan: make i386 and x86_64 do it
- * the same way
+ * --------- Crashkernel reservation ------------------------------
+ */
+
+#ifdef CONFIG_KEXEC
+
+#ifndef CONFIG_XEN
+/**
+ * Reserve @size bytes of crashkernel memory at any suitable offset.
+ *
+ * @size: Size of the crashkernel memory to reserve.
+ * Returns the base address on success, and -1ULL on failure.
+ */
+unsigned long long __init find_and_reserve_crashkernel(unsigned long long size)
+{
+       const unsigned long long alignment = 16<<20;    /* 16M */
+       unsigned long long start = 0LL;
+
+       while (1) {
+               int ret;
+
+               start = find_e820_area(start, ULONG_MAX, size, alignment);
+               if (start == -1ULL)
+                       return start;
+
+               /* try to reserve it */
+               ret = reserve_bootmem_generic(start, size, BOOTMEM_EXCLUSIVE);
+               if (ret >= 0)
+                       return start;
+
+               start += alignment;
+       }
+}
+
+static inline unsigned long long get_total_mem(void)
+{
+       unsigned long long total;
+
+       total = max_low_pfn - min_low_pfn;
+#ifdef CONFIG_HIGHMEM
+       total += highend_pfn - highstart_pfn;
+#endif
+
+       return total << PAGE_SHIFT;
+}
+
+static void __init reserve_crashkernel(void)
+{
+       unsigned long long total_mem;
+       unsigned long long crash_size, crash_base;
+       int ret;
+
+       total_mem = get_total_mem();
+
+       ret = parse_crashkernel(boot_command_line, total_mem,
+                       &crash_size, &crash_base);
+       if (ret != 0 || crash_size <= 0)
+               return;
+
+       /* 0 means: find the address automatically */
+       if (crash_base <= 0) {
+               crash_base = find_and_reserve_crashkernel(crash_size);
+               if (crash_base == -1ULL) {
+                       pr_info("crashkernel reservation failed. "
+                               "No suitable area found.\n");
+                       return;
+               }
+       } else {
+               ret = reserve_bootmem_generic(crash_base, crash_size,
+                                       BOOTMEM_EXCLUSIVE);
+               if (ret < 0) {
+                       pr_info("crashkernel reservation failed - "
+                               "memory is in use\n");
+                       return;
+               }
+       }
+
+       printk(KERN_INFO "Reserving %ldMB of memory at %ldMB "
+                       "for crashkernel (System RAM: %ldMB)\n",
+                       (unsigned long)(crash_size >> 20),
+                       (unsigned long)(crash_base >> 20),
+                       (unsigned long)(total_mem >> 20));
+
+       crashk_res.start = crash_base;
+       crashk_res.end   = crash_base + crash_size - 1;
+       insert_resource(&iomem_resource, &crashk_res);
+}
+#else
+#define reserve_crashkernel xen_machine_kexec_setup_resources
+#endif
+#else
+static void __init reserve_crashkernel(void)
+{
+}
+#endif
+
+static struct resource standard_io_resources[] = {
+       { .name = "dma1", .start = 0x00, .end = 0x1f,
+               .flags = IORESOURCE_BUSY | IORESOURCE_IO },
+       { .name = "pic1", .start = 0x20, .end = 0x21,
+               .flags = IORESOURCE_BUSY | IORESOURCE_IO },
+       { .name = "timer0", .start = 0x40, .end = 0x43,
+               .flags = IORESOURCE_BUSY | IORESOURCE_IO },
+       { .name = "timer1", .start = 0x50, .end = 0x53,
+               .flags = IORESOURCE_BUSY | IORESOURCE_IO },
+       { .name = "keyboard", .start = 0x60, .end = 0x60,
+               .flags = IORESOURCE_BUSY | IORESOURCE_IO },
+       { .name = "keyboard", .start = 0x64, .end = 0x64,
+               .flags = IORESOURCE_BUSY | IORESOURCE_IO },
+       { .name = "dma page reg", .start = 0x80, .end = 0x8f,
+               .flags = IORESOURCE_BUSY | IORESOURCE_IO },
+       { .name = "pic2", .start = 0xa0, .end = 0xa1,
+               .flags = IORESOURCE_BUSY | IORESOURCE_IO },
+       { .name = "dma2", .start = 0xc0, .end = 0xdf,
+               .flags = IORESOURCE_BUSY | IORESOURCE_IO },
+       { .name = "fpu", .start = 0xf0, .end = 0xff,
+               .flags = IORESOURCE_BUSY | IORESOURCE_IO }
+};
+
+static void __init reserve_standard_io_resources(void)
+{
+       int i;
+
+       /* Nothing to do if not running in dom0. */
+       if (!is_initial_xendomain())
+               return;
+
+       /* request I/O space for devices used on all i[345]86 PCs */
+       for (i = 0; i < ARRAY_SIZE(standard_io_resources); i++)
+               request_resource(&ioport_resource, &standard_io_resources[i]);
+
+}
+
+#ifdef CONFIG_PROC_VMCORE
+/* elfcorehdr= specifies the location of elf core header
+ * stored by the crashed kernel. This option will be passed
+ * by kexec loader to the capture kernel.
  */
-unsigned long __per_cpu_offset[NR_CPUS] __read_mostly;
-EXPORT_SYMBOL(__per_cpu_offset);
+static int __init setup_elfcorehdr(char *arg)
+{
+       char *end;
+       if (!arg)
+               return -EINVAL;
+       elfcorehdr_addr = memparse(arg, &end);
+       return end > arg ? 0 : -EINVAL;
+}
+early_param("elfcorehdr", setup_elfcorehdr);
 #endif
 
+static struct x86_quirks default_x86_quirks __initdata;
+
+struct x86_quirks *x86_quirks __initdata = &default_x86_quirks;
+
 /*
- * Great future plan:
- * Declare PDA itself and support (irqstack,tss,pgd) as per cpu data.
- * Always point %gs to its beginning
+ * Determine if we were loaded by an EFI loader.  If so, then we have also been
+ * passed the efi memmap, systab, etc., so we should use these data structures
+ * for initialization.  Note, the efi init code path is determined by the
+ * global efi_enabled. This allows the same kernel image to be used on existing
+ * systems (with a traditional BIOS) as well as on EFI systems.
  */
-void __init setup_per_cpu_areas(void)
+/*
+ * setup_arch - architecture-specific boot-time initializations
+ *
+ * Note: On x86_64, fixmaps are ready for use even before this is called.
+ */
+
+void __init setup_arch(char **cmdline_p)
 {
-       int i, highest_cpu = 0;
-       unsigned long size;
+#ifdef CONFIG_XEN
+       unsigned int i;
+       unsigned long p2m_pages;
+       struct physdev_set_iopl set_iopl;
 
-#ifdef CONFIG_HOTPLUG_CPU
-       prefill_possible_map();
+#ifdef CONFIG_X86_32
+       /* Force a quick death if the kernel panics (not domain 0). */
+       extern int panic_timeout;
+       if (!panic_timeout && !is_initial_xendomain())
+               panic_timeout = 1;
 #endif
 
-       /* Copy section for each CPU (we discard the original) */
-       size = PERCPU_ENOUGH_ROOM;
-       printk(KERN_INFO "PERCPU: Allocating %lu bytes of per cpu data\n",
-                         size);
-
-       for_each_possible_cpu(i) {
-               char *ptr;
-#ifndef CONFIG_NEED_MULTIPLE_NODES
-               ptr = alloc_bootmem_pages(size);
-#else
-               int node = early_cpu_to_node(i);
-               if (!node_online(node) || !NODE_DATA(node)) {
-                       ptr = alloc_bootmem_pages(size);
-                       printk(KERN_INFO
-                              "cpu %d has no node or node-local memory\n", i);
-               }
-               else
-                       ptr = alloc_bootmem_pages_node(NODE_DATA(node), size);
+       /* Register a call for panic conditions. */
+       atomic_notifier_chain_register(&panic_notifier_list, &xen_panic_block);
+
+       WARN_ON(HYPERVISOR_vm_assist(VMASST_CMD_enable,
+                                    VMASST_TYPE_writable_pagetables));
+#ifdef CONFIG_X86_32
+       WARN_ON(HYPERVISOR_vm_assist(VMASST_CMD_enable,
+                                    VMASST_TYPE_4gb_segments));
+#endif
+#endif /* CONFIG_XEN */
+
+#ifdef CONFIG_X86_32
+       memcpy(&boot_cpu_data, &new_cpu_data, sizeof(new_cpu_data));
+       visws_early_detect();
+       pre_setup_arch_hook();
+#else
+       printk(KERN_INFO "Command line: %s\n", boot_command_line);
+#endif
+
+       early_cpu_init();
+       early_ioremap_init();
+
+#ifndef CONFIG_XEN
+       ROOT_DEV = old_decode_dev(boot_params.hdr.root_dev);
+       screen_info = boot_params.screen_info;
+       edid_info = boot_params.edid_info;
+#ifdef CONFIG_X86_32
+       apm_info.bios = boot_params.apm_bios_info;
+       ist_info = boot_params.ist_info;
+       if (boot_params.sys_desc_table.length != 0) {
+               set_mca_bus(boot_params.sys_desc_table.table[3] & 0x2);
+               machine_id = boot_params.sys_desc_table.table[0];
+               machine_submodel_id = boot_params.sys_desc_table.table[1];
+               BIOS_revision = boot_params.sys_desc_table.table[2];
+       }
+#endif
+       saved_video_mode = boot_params.hdr.vid_mode;
+       bootloader_type = boot_params.hdr.type_of_loader;
+
+#ifdef CONFIG_BLK_DEV_RAM
+       rd_image_start = boot_params.hdr.ram_size & RAMDISK_IMAGE_START_MASK;
+       rd_prompt = ((boot_params.hdr.ram_size & RAMDISK_PROMPT_FLAG) != 0);
+       rd_doload = ((boot_params.hdr.ram_size & RAMDISK_LOAD_FLAG) != 0);
+#endif
+#ifdef CONFIG_EFI
+       if (!strncmp((char *)&boot_params.efi_info.efi_loader_signature,
+#ifdef CONFIG_X86_32
+                    "EL32",
+#else
+                    "EL64",
+#endif
+        4)) {
+               efi_enabled = 1;
+               efi_reserve_early();
+       }
+#endif
+#else /* CONFIG_XEN */
+#ifdef CONFIG_X86_32
+       /* This must be initialized to UNNAMED_MAJOR for ipconfig to work
+          properly.  Setting ROOT_DEV to default to /dev/ram0 breaks initrd.
+       */
+       ROOT_DEV = MKDEV(UNNAMED_MAJOR,0);
+#else
+       ROOT_DEV = MKDEV(RAMDISK_MAJOR,0);
+#endif
+       if (is_initial_xendomain()) {
+               const struct dom0_vga_console_info *info =
+                       (void *)((char *)xen_start_info +
+                                xen_start_info->console.dom0.info_off);
+
+               dom0_init_screen_info(info,
+                                     xen_start_info->console.dom0.info_size);
+               xen_start_info->console.domU.mfn = 0;
+               xen_start_info->console.domU.evtchn = 0;
+       } else
+               screen_info.orig_video_isVGA = 0;
+       copy_edid();
+#endif /* CONFIG_XEN */
+
+       ARCH_SETUP
+
+       setup_memory_map();
+       parse_setup_data();
+       /* update the e820_saved too */
+       e820_reserve_setup_data();
+
+       copy_edd();
+
+#ifndef CONFIG_XEN
+       if (!boot_params.hdr.root_flags)
+               root_mountflags &= ~MS_RDONLY;
 #endif
-               if (!ptr)
-                       panic("Cannot allocate cpu data for CPU %d\n", i);
+       init_mm.start_code = (unsigned long) _text;
+       init_mm.end_code = (unsigned long) _etext;
+       init_mm.end_data = (unsigned long) _edata;
+#ifdef CONFIG_X86_32
+#ifndef CONFIG_XEN
+       init_mm.brk = init_pg_tables_end + PAGE_OFFSET;
+#else
+       init_mm.brk = (PFN_UP(__pa(xen_start_info->pt_base)) +
+                      xen_start_info->nr_pt_frames) << PAGE_SHIFT;
+#endif
+#else
+       init_mm.brk = (unsigned long) &_end;
+#endif
+
+       code_resource.start = virt_to_phys(_text);
+       code_resource.end = virt_to_phys(_etext)-1;
+       data_resource.start = virt_to_phys(_etext);
+       data_resource.end = virt_to_phys(_edata)-1;
+       bss_resource.start = virt_to_phys(&__bss_start);
+       bss_resource.end = virt_to_phys(&__bss_stop)-1;
+
+       strlcpy(command_line, boot_command_line, COMMAND_LINE_SIZE);
+       *cmdline_p = command_line;
+
+       parse_early_param();
+
 #ifdef CONFIG_X86_64
-               cpu_pda(i)->data_offset = ptr - __per_cpu_start;
+       check_efer();
+#endif
+
+#if defined(CONFIG_VMI) && defined(CONFIG_X86_32)
+       /*
+        * Must be before kernel pagetables are setup
+        * or fixmap area is touched.
+        */
+       vmi_init();
+#endif
+
+       /* after early param, so could get panic from serial */
+       reserve_early_setup_data();
+
+       if (acpi_mps_check()) {
+#if defined(CONFIG_X86_LOCAL_APIC) && !defined(CONFIG_XEN)
+               disable_apic = 1;
+#endif
+               setup_clear_cpu_cap(X86_FEATURE_APIC);
+       }
+
+#ifdef CONFIG_PCI
+       if (pci_early_dump_regs)
+               early_dump_pci_devices();
+#endif
+
+       finish_e820_parsing();
+
+#ifdef CONFIG_X86_32
+       probe_roms();
+#endif
+
+#ifndef CONFIG_XEN
+       /* after parse_early_param, so could debug it */
+       insert_resource(&iomem_resource, &code_resource);
+       insert_resource(&iomem_resource, &data_resource);
+       insert_resource(&iomem_resource, &bss_resource);
+
+       if (efi_enabled)
+               efi_init();
+
+#ifdef CONFIG_X86_32
+       if (ppro_with_ram_bug()) {
+               e820_update_range(0x70000000ULL, 0x40000ULL, E820_RAM,
+                                 E820_RESERVED);
+               sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);
+               printk(KERN_INFO "fixed physical RAM map:\n");
+               e820_print_map("bad_ppro");
+       }
 #else
-               __per_cpu_offset[i] = ptr - __per_cpu_start;
+       early_gart_iommu_check();
 #endif
-               memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start);
+#endif /* CONFIG_XEN */
 
-               highest_cpu = i;
+       /*
+        * partially used pages are not usable - thus
+        * we are rounding upwards:
+        */
+       max_pfn = e820_end_of_ram_pfn();
+
+       /* preallocate 4k for mptable mpc */
+       early_reserve_e820_mpc_new();
+       /* update e820 for memory not covered by WB MTRRs */
+       mtrr_bp_init();
+#ifndef CONFIG_XEN
+       if (mtrr_trim_uncached_memory(max_pfn))
+               max_pfn = e820_end_of_ram_pfn();
+#endif
+
+#ifdef CONFIG_X86_32
+       /* max_low_pfn get updated here */
+       find_low_pfn_range();
+#else
+       num_physpages = max_pfn;
+       max_mapnr = max_pfn;
+
+
+       /* How many end-of-memory variables you have, grandma! */
+       /* need this before calling reserve_initrd */
+       if (max_pfn > (1UL<<(32 - PAGE_SHIFT)))
+               max_low_pfn = e820_end_of_low_ram_pfn();
+       else
+               max_low_pfn = max_pfn;
+
+       high_memory = (void *)__va(max_pfn * PAGE_SIZE - 1) + 1;
+#endif
+
+       /* max_pfn_mapped is updated here */
+#ifdef CONFIG_X86_64_XEN
+       /*
+        * Due to the way initial table space gets calculated on Xen, we have
+        * to call init_memory_mapping() with the larger end address first.
+        */
+       if (max_pfn > max_low_pfn)
+               max_pfn_mapped = init_memory_mapping(1UL<<32,
+                                                    max_pfn<<PAGE_SHIFT);
+       max_low_pfn_mapped = init_memory_mapping(0, max_low_pfn<<PAGE_SHIFT);
+       if (max_pfn > max_low_pfn)
+               /* can we preserve max_low_pfn ?*/
+               max_low_pfn = max_pfn;
+       else
+               max_pfn_mapped = max_low_pfn_mapped;
+#else
+       max_low_pfn_mapped = init_memory_mapping(0, max_low_pfn<<PAGE_SHIFT);
+       max_pfn_mapped = max_low_pfn_mapped;
+
+#ifdef CONFIG_X86_64
+       if (max_pfn > max_low_pfn) {
+               max_pfn_mapped = init_memory_mapping(1UL<<32,
+                                                    max_pfn<<PAGE_SHIFT);
+               /* can we preseve max_low_pfn ?*/
+               max_low_pfn = max_pfn;
        }
+#endif
+#endif
 
-       nr_cpu_ids = highest_cpu + 1;
-       printk(KERN_DEBUG "NR_CPUS: %d, nr_cpu_ids: %d\n", NR_CPUS, nr_cpu_ids);
+       /*
+        * NOTE: On x86-32, only from this point on, fixmaps are ready for use.
+        */
 
-       /* Setup percpu data maps */
-       setup_per_cpu_maps();
+#ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT
+       if (init_ohci1394_dma_early)
+               init_ohci1394_dma_on_all_controllers();
+#endif
 
-       /* Setup cpumask_of_cpu map */
-       setup_cpumask_of_cpu();
-}
+       reserve_initrd();
+
+#if defined(CONFIG_X86_64) && !defined(CONFIG_XEN)
+       vsmp_init();
+#endif
+
+       if (is_initial_xendomain())
+               dmi_scan_machine();
+
+       io_delay_init();
+
+#ifdef CONFIG_ACPI
+       if (!is_initial_xendomain()) {
+               printk(KERN_INFO "ACPI in unprivileged domain disabled\n");
+               disable_acpi();
+       }
+#endif
+
+       /*
+        * Parse the ACPI tables for possible boot-time SMP configuration.
+        */
+       acpi_boot_table_init();
+
+#ifdef CONFIG_ACPI_NUMA
+       /*
+        * Parse SRAT to discover nodes.
+        */
+       acpi_numa_init();
+#endif
+
+       initmem_init(0, max_pfn);
 
+#ifdef CONFIG_ACPI_SLEEP
+       /*
+        * Reserve low memory region for sleep support.
+        */
+       acpi_reserve_bootmem();
 #endif
+#ifdef CONFIG_X86_FIND_SMP_CONFIG
+       /*
+        * Find and reserve possible boot-time SMP configuration:
+        */
+       find_smp_config();
+#endif
+       reserve_crashkernel();
+
+#if defined(CONFIG_X86_64) && !defined(CONFIG_XEN)
+       /*
+        * dma32_reserve_bootmem() allocates bootmem which may conflict
+        * with the crashkernel command line, so do that after
+        * reserve_crashkernel()
+        */
+       dma32_reserve_bootmem();
+#endif
+
+       reserve_ibft_region();
+
+#ifdef CONFIG_KVM_CLOCK
+       kvmclock_init();
+#endif
+
+       xen_pagetable_setup_start(swapper_pg_dir);
+       paging_init();
+       xen_pagetable_setup_done(swapper_pg_dir);
+       paravirt_post_allocator_init();
+
+#ifdef CONFIG_X86_64
+       map_vsyscall();
+#endif
+
+#ifdef CONFIG_XEN
+       p2m_pages = max_pfn;
+       if (xen_start_info->nr_pages > max_pfn) {
+               /*
+                * the max_pfn was shrunk (probably by mem= or highmem=
+                * kernel parameter); shrink reservation with the HV
+                */
+               struct xen_memory_reservation reservation = {
+                       .address_bits = 0,
+                       .extent_order = 0,
+                       .domid = DOMID_SELF
+               };
+               unsigned int difference;
+               int ret;
+
+               difference = xen_start_info->nr_pages - max_pfn;
+
+               set_xen_guest_handle(reservation.extent_start,
+                                    ((unsigned long *)xen_start_info->mfn_list) + max_pfn);
+               reservation.nr_extents = difference;
+               ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation,
+                                          &reservation);
+               BUG_ON(ret != difference);
+       }
+       else if (max_pfn > xen_start_info->nr_pages)
+               p2m_pages = xen_start_info->nr_pages;
+
+       if (!xen_feature(XENFEAT_auto_translated_physmap)) {
+               unsigned long i, j;
+               unsigned int k, fpp;
+
+               /* Make sure we have a large enough P->M table. */
+               phys_to_machine_mapping = alloc_bootmem_pages(
+                       max_pfn * sizeof(unsigned long));
+               memset(phys_to_machine_mapping, ~0,
+                      max_pfn * sizeof(unsigned long));
+               memcpy(phys_to_machine_mapping,
+                      (unsigned long *)xen_start_info->mfn_list,
+                      p2m_pages * sizeof(unsigned long));
+               free_bootmem(
+                       __pa(xen_start_info->mfn_list),
+                       PFN_PHYS(PFN_UP(xen_start_info->nr_pages *
+                                       sizeof(unsigned long))));
+
+               /*
+                * Initialise the list of the frames that specify the list of
+                * frames that make up the p2m table. Used by save/restore.
+                */
+               pfn_to_mfn_frame_list_list = alloc_bootmem_pages(PAGE_SIZE);
+
+               fpp = PAGE_SIZE/sizeof(unsigned long);
+               for (i = j = 0, k = -1; i < max_pfn; i += fpp, j++) {
+                       if (j == fpp)
+                               j = 0;
+                       if (j == 0) {
+                               k++;
+                               BUG_ON(k>=ARRAY_SIZE(pfn_to_mfn_frame_list));
+                               pfn_to_mfn_frame_list[k] =
+                                       alloc_bootmem_pages(PAGE_SIZE);
+                               pfn_to_mfn_frame_list_list[k] =
+                                       virt_to_mfn(pfn_to_mfn_frame_list[k]);
+                       }
+                       pfn_to_mfn_frame_list[k][j] =
+                               virt_to_mfn(&phys_to_machine_mapping[i]);
+               }
+               HYPERVISOR_shared_info->arch.max_pfn = max_pfn;
+               HYPERVISOR_shared_info->arch.pfn_to_mfn_frame_list_list =
+                       virt_to_mfn(pfn_to_mfn_frame_list_list);
+       }
+
+       /* Mark all ISA DMA channels in-use - using them wouldn't work. */
+       for (i = 0; i < MAX_DMA_CHANNELS; ++i)
+               if (i != 4 && request_dma(i, "xen") != 0)
+                       BUG();
+#endif /* CONFIG_XEN */
+
+#ifdef CONFIG_X86_GENERICARCH
+       generic_apic_probe();
+#endif
+
+#ifndef CONFIG_XEN
+       early_quirks();
+#endif
+
+       /*
+        * Read APIC and some other early information from ACPI tables.
+        */
+       acpi_boot_init();
+
+#if defined(CONFIG_X86_MPPARSE) || defined(CONFIG_X86_VISWS)
+       /*
+        * get boot-time SMP configuration:
+        */
+       if (smp_found_config)
+               get_smp_config();
+#endif
+
+       prefill_possible_map();
+#ifdef CONFIG_X86_64
+       init_cpu_to_node();
+#endif
+
+#ifndef CONFIG_XEN
+       init_apic_mappings();
+       ioapic_init_mappings();
+
+       kvm_guest_init();
+
+       e820_reserve_resources();
+       e820_mark_nosave_regions(max_low_pfn);
+#else
+       if (is_initial_xendomain())
+               e820_reserve_resources();
+#endif
+
+#ifdef CONFIG_X86_32
+       request_resource(&iomem_resource, &video_ram_resource);
+#endif
+       reserve_standard_io_resources();
+
+#ifndef CONFIG_XEN
+       e820_setup_gap();
+
+#ifdef CONFIG_VT
+#if defined(CONFIG_VGA_CONSOLE)
+       if (!efi_enabled || (efi_mem_type(0xa0000) != EFI_CONVENTIONAL_MEMORY))
+               conswitchp = &vga_con;
+#elif defined(CONFIG_DUMMY_CONSOLE)
+       conswitchp = &dummy_con;
+#endif
+#endif
+#else /* CONFIG_XEN */
+       if (is_initial_xendomain())
+               e820_setup_gap();
+
+       set_iopl.iopl = 1;
+       WARN_ON(HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl));
+
+#ifdef CONFIG_VT
+#ifdef CONFIG_DUMMY_CONSOLE
+       conswitchp = &dummy_con;
+#endif
+#ifdef CONFIG_VGA_CONSOLE
+       if (is_initial_xendomain())
+               conswitchp = &vga_con;
+#endif
+#endif
+#endif /* CONFIG_XEN */
+}
+
+#ifdef CONFIG_XEN
+static int
+xen_panic_event(struct notifier_block *this, unsigned long event, void *ptr)
+{
+       HYPERVISOR_shutdown(SHUTDOWN_crash);
+       /* we're never actually going to get here... */
+       return NOTIFY_DONE;
+}
+#endif /* !CONFIG_XEN */
Index: head-2008-12-01/arch/x86/kernel/setup64-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/setup64-xen.c  2008-12-01 11:44:55.000000000 +0100
+++ /dev/null   1970-01-01 00:00:00.000000000 +0000
@@ -1,370 +0,0 @@
-/* 
- * X86-64 specific CPU setup.
- * Copyright (C) 1995  Linus Torvalds
- * Copyright 2001, 2002, 2003 SuSE Labs / Andi Kleen.
- * See setup.c for older changelog.
- *
- * Jun Nakajima <jun.nakajima@intel.com> 
- *   Modified for Xen
- *
- */ 
-#include <linux/init.h>
-#include <linux/kernel.h>
-#include <linux/sched.h>
-#include <linux/string.h>
-#include <linux/bootmem.h>
-#include <linux/bitops.h>
-#include <linux/module.h>
-#include <linux/kgdb.h>
-#include <asm/pda.h>
-#include <asm/pgtable.h>
-#include <asm/processor.h>
-#include <asm/desc.h>
-#include <asm/atomic.h>
-#include <asm/mmu_context.h>
-#include <asm/smp.h>
-#include <asm/i387.h>
-#include <asm/percpu.h>
-#include <asm/proto.h>
-#include <asm/sections.h>
-#include <asm/setup.h>
-#include <asm/genapic.h>
-#ifdef CONFIG_XEN
-#include <asm/hypervisor.h>
-#endif
-
-#ifndef CONFIG_DEBUG_BOOT_PARAMS
-struct boot_params __initdata boot_params;
-#else
-struct boot_params boot_params;
-#endif
-
-cpumask_t cpu_initialized __cpuinitdata = CPU_MASK_NONE;
-
-struct x8664_pda *_cpu_pda[NR_CPUS] __read_mostly;
-EXPORT_SYMBOL(_cpu_pda);
-struct x8664_pda boot_cpu_pda[NR_CPUS] __cacheline_aligned;
-
-#ifndef CONFIG_X86_NO_IDT
-struct desc_ptr idt_descr = { 256 * 16 - 1, (unsigned long) idt_table }; 
-#endif
-
-char boot_cpu_stack[IRQSTACKSIZE] __attribute__((section(".bss.page_aligned")));
-
-unsigned long __supported_pte_mask __read_mostly = ~0UL;
-EXPORT_SYMBOL(__supported_pte_mask);
-
-static int do_not_nx __cpuinitdata = 0;
-
-/* noexec=on|off
-Control non executable mappings for 64bit processes.
-
-on     Enable(default)
-off    Disable
-*/ 
-static int __init nonx_setup(char *str)
-{
-       if (!str)
-               return -EINVAL;
-       if (!strncmp(str, "on", 2)) {
-                __supported_pte_mask |= _PAGE_NX; 
-               do_not_nx = 0; 
-       } else if (!strncmp(str, "off", 3)) {
-               do_not_nx = 1;
-               __supported_pte_mask &= ~_PAGE_NX;
-        }
-       return 0;
-} 
-early_param("noexec", nonx_setup);
-
-int force_personality32 = 0; 
-
-/* noexec32=on|off
-Control non executable heap for 32bit processes.
-To control the stack too use noexec=off
-
-on     PROT_READ does not imply PROT_EXEC for 32bit processes (default)
-off    PROT_READ implies PROT_EXEC
-*/
-static int __init nonx32_setup(char *str)
-{
-       if (!strcmp(str, "on"))
-               force_personality32 &= ~READ_IMPLIES_EXEC;
-       else if (!strcmp(str, "off"))
-               force_personality32 |= READ_IMPLIES_EXEC;
-       return 1;
-}
-__setup("noexec32=", nonx32_setup);
-
-#ifdef CONFIG_XEN
-static void __init_refok switch_pt(int cpu)
-{
-       if (cpu == 0)
-               xen_init_pt();
-       xen_pt_switch(__pa_symbol(init_level4_pgt));
-       xen_new_user_pt(__pa_symbol(__user_pgd(init_level4_pgt)));
-}
-#define switch_pt() switch_pt(cpu)
-
-static void __cpuinit cpu_gdt_init(const struct desc_ptr *gdt_descr)
-{
-       unsigned long frames[16];
-       unsigned long va;
-       int f;
-
-       for (va = gdt_descr->address, f = 0;
-            va < gdt_descr->address + gdt_descr->size;
-            va += PAGE_SIZE, f++) {
-               frames[f] = virt_to_mfn(va);
-               make_page_readonly(
-                       (void *)va, XENFEAT_writable_descriptor_tables);
-       }
-       if (HYPERVISOR_set_gdt(frames, (gdt_descr->size + 1) /
-                               sizeof (struct desc_struct)))
-               BUG();
-}
-#else
-static void switch_pt(void)
-{
-       asm volatile("movq %0,%%cr3" :: "r" (__pa_symbol(&init_level4_pgt)));
-}
-
-static void __cpuinit cpu_gdt_init(const struct desc_ptr *gdt_descr)
-{
-       load_gdt(gdt_descr);
-       load_idt(idt_descr);
-}
-#endif
-
-void pda_init(int cpu)
-{ 
-       struct x8664_pda *pda = cpu_pda(cpu);
-
-       /* Setup up data that may be needed in __get_free_pages early */
-       asm volatile("movl %0,%%fs ; movl %0,%%gs" :: "r" (0)); 
-#ifndef CONFIG_XEN
-       /* Memory clobbers used to order PDA accessed */
-       mb();
-       wrmsrl(MSR_GS_BASE, pda);
-       mb();
-#else
-       if (HYPERVISOR_set_segment_base(SEGBASE_GS_KERNEL,
-                                       (unsigned long)pda))
-               BUG();
-#endif
-       pda->cpunumber = cpu; 
-       pda->irqcount = -1;
-       pda->kernelstack = 
-               (unsigned long)stack_thread_info() - PDA_STACKOFFSET + THREAD_SIZE; 
-       pda->active_mm = &init_mm;
-       pda->mmu_state = 0;
-
-       if (cpu == 0) {
-               /* others are initialized in smpboot.c */
-               pda->pcurrent = &init_task;
-               pda->irqstackptr = boot_cpu_stack; 
-       } else {
-               pda->irqstackptr = (char *)
-                       __get_free_pages(GFP_ATOMIC, IRQSTACK_ORDER);
-               if (!pda->irqstackptr)
-                       panic("cannot allocate irqstack for cpu %d", cpu); 
-       }
-
-       switch_pt();
-
-       pda->irqstackptr += IRQSTACKSIZE-64;
-} 
-
-#ifndef CONFIG_X86_NO_TSS
-char boot_exception_stacks[(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ + DEBUG_STKSZ]
-__attribute__((section(".bss.page_aligned")));
-#endif
-
-extern asmlinkage void ignore_sysret(void);
-
-/* May not be marked __init: used by software suspend */
-void syscall_init(void)
-{
-#ifndef CONFIG_XEN
-       /* 
-        * LSTAR and STAR live in a bit strange symbiosis.
-        * They both write to the same internal register. STAR allows to set CS/DS
-        * but only a 32bit target. LSTAR sets the 64bit rip.    
-        */ 
-       wrmsrl(MSR_STAR,  ((u64)__USER32_CS)<<48  | ((u64)__KERNEL_CS)<<32); 
-       wrmsrl(MSR_LSTAR, system_call); 
-       wrmsrl(MSR_CSTAR, ignore_sysret);
-
-       /* Flags to clear on syscall */
-       wrmsrl(MSR_SYSCALL_MASK,
-              X86_EFLAGS_TF|X86_EFLAGS_DF|X86_EFLAGS_IF|X86_EFLAGS_IOPL);
-#endif
-#ifdef CONFIG_IA32_EMULATION                   
-       syscall32_cpu_init ();
-#else
-       {
-               static const struct callback_register cstar = {
-                       .type = CALLBACKTYPE_syscall32,
-                       .address = (unsigned long)ignore_sysret
-               };
-               if (HYPERVISOR_callback_op(CALLBACKOP_register, &cstar))
-                       printk(KERN_WARN "Unable to register CSTAR callback\n");
-       }
-#endif
-}
-
-void __cpuinit check_efer(void)
-{
-       unsigned long efer;
-
-       rdmsrl(MSR_EFER, efer); 
-        if (!(efer & EFER_NX) || do_not_nx) { 
-                __supported_pte_mask &= ~_PAGE_NX; 
-        }       
-}
-
-unsigned long kernel_eflags;
-
-#ifndef CONFIG_X86_NO_TSS
-/*
- * Copies of the original ist values from the tss are only accessed during
- * debugging, no special alignment required.
- */
-DEFINE_PER_CPU(struct orig_ist, orig_ist);
-#endif
-
-/*
- * cpu_init() initializes state that is per-CPU. Some data is already
- * initialized (naturally) in the bootstrap process, such as the GDT
- * and IDT. We reload them nevertheless, this function acts as a
- * 'CPU state barrier', nothing should get across.
- * A lot of state is already set up in PDA init.
- */
-void __cpuinit cpu_init (void)
-{
-       int cpu = stack_smp_processor_id();
-#ifndef CONFIG_X86_NO_TSS
-       struct tss_struct *t = &per_cpu(init_tss, cpu);
-       struct orig_ist *orig_ist = &per_cpu(orig_ist, cpu);
-       unsigned long v; 
-       char *estacks = NULL; 
-       unsigned i;
-#endif
-       struct task_struct *me;
-
-       /* CPU 0 is initialised in head64.c */
-       if (cpu != 0) {
-               pda_init(cpu);
-       }
-#ifndef CONFIG_X86_NO_TSS
-       else
-               estacks = boot_exception_stacks; 
-#endif
-
-       me = current;
-
-       if (cpu_test_and_set(cpu, cpu_initialized))
-               panic("CPU#%d already initialized!\n", cpu);
-
-       printk("Initializing CPU#%d\n", cpu);
-
-       clear_in_cr4(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE);
-
-       /*
-        * Initialize the per-CPU GDT with the boot GDT,
-        * and set up the GDT descriptor:
-        */
-#ifndef CONFIG_XEN 
-       if (cpu)
-               memcpy(get_cpu_gdt_table(cpu), cpu_gdt_table, GDT_SIZE);
-#endif
-
-       cpu_gdt_descr[cpu].size = GDT_SIZE;
-       cpu_gdt_init(&cpu_gdt_descr[cpu]);
-
-       memset(me->thread.tls_array, 0, GDT_ENTRY_TLS_ENTRIES * 8);
-       syscall_init();
-
-       wrmsrl(MSR_FS_BASE, 0);
-       wrmsrl(MSR_KERNEL_GS_BASE, 0);
-       barrier(); 
-
-       check_efer();
-
-#ifndef CONFIG_X86_NO_TSS
-       /*
-        * set up and load the per-CPU TSS
-        */
-       for (v = 0; v < N_EXCEPTION_STACKS; v++) {
-               static const unsigned int order[N_EXCEPTION_STACKS] = {
-                       [0 ... N_EXCEPTION_STACKS - 1] = EXCEPTION_STACK_ORDER,
-                       [DEBUG_STACK - 1] = DEBUG_STACK_ORDER
-               };
-               if (cpu) {
-                       estacks = (char *)__get_free_pages(GFP_ATOMIC, order[v]);
-                       if (!estacks)
-                               panic("Cannot allocate exception stack %ld %d\n",
-                                     v, cpu); 
-               }
-               estacks += PAGE_SIZE << order[v];
-               orig_ist->ist[v] = t->x86_tss.ist[v] = (unsigned long)estacks;
-       }
-
-       t->x86_tss.io_bitmap_base = offsetof(struct tss_struct, io_bitmap);
-       /*
-        * <= is required because the CPU will access up to
-        * 8 bits beyond the end of the IO permission bitmap.
-        */
-       for (i = 0; i <= IO_BITMAP_LONGS; i++)
-               t->io_bitmap[i] = ~0UL;
-#endif
-
-       atomic_inc(&init_mm.mm_count);
-       me->active_mm = &init_mm;
-       if (me->mm)
-               BUG();
-       enter_lazy_tlb(&init_mm, me);
-
-#ifndef CONFIG_X86_NO_TSS
-       set_tss_desc(cpu, t);
-#endif
-#ifndef CONFIG_XEN
-       load_TR_desc();
-#endif
-       load_LDT(&init_mm.context);
-
-#ifdef CONFIG_KGDB
-       /*
-        * If the kgdb is connected no debug regs should be altered.  This
-        * is only applicable when KGDB and a KGDB I/O module are built
-        * into the kernel and you are using early debugging with
-        * kgdbwait. KGDB will control the kernel HW breakpoint registers.
-        */
-       if (kgdb_connected && arch_kgdb_ops.correct_hw_break)
-               arch_kgdb_ops.correct_hw_break();
-       else {
-#endif
-       /*
-        * Clear all 6 debug registers:
-        */
-
-       set_debugreg(0UL, 0);
-       set_debugreg(0UL, 1);
-       set_debugreg(0UL, 2);
-       set_debugreg(0UL, 3);
-       set_debugreg(0UL, 6);
-       set_debugreg(0UL, 7);
-#ifdef CONFIG_KGDB
-       /* If the kgdb is connected no debug regs should be altered. */
-       }
-#endif
-
-       fpu_init(); 
-
-       asm ("pushfq; popq %0" : "=rm" (kernel_eflags));
-       if (raw_irqs_disabled())
-               kernel_eflags &= ~X86_EFLAGS_IF;
-
-       if (is_uv_system())
-               uv_cpu_init();
-}
Index: head-2008-12-01/arch/x86/kernel/setup_32-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/setup_32-xen.c 2008-12-01 11:44:55.000000000 +0100
+++ /dev/null   1970-01-01 00:00:00.000000000 +0000
@@ -1,1151 +0,0 @@
-/*
- *  Copyright (C) 1995  Linus Torvalds
- *
- *  Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999
- *
- *  Memory region support
- *     David Parsons <orc@pell.chi.il.us>, July-August 1999
- *
- *  Added E820 sanitization routine (removes overlapping memory regions);
- *  Brian Moyle <bmoyle@mvista.com>, February 2001
- *
- * Moved CPU detection code to cpu/${cpu}.c
- *    Patrick Mochel <mochel@osdl.org>, March 2002
- *
- *  Provisions for empty E820 memory regions (reported by certain BIOSes).
- *  Alex Achenbach <xela@slit.de>, December 2002.
- *
- */
-
-/*
- * This file handles the architecture-dependent parts of initialization
- */
-
-#include <linux/sched.h>
-#include <linux/mm.h>
-#include <linux/mmzone.h>
-#include <linux/screen_info.h>
-#include <linux/ioport.h>
-#include <linux/acpi.h>
-#include <linux/apm_bios.h>
-#include <linux/initrd.h>
-#include <linux/bootmem.h>
-#include <linux/seq_file.h>
-#include <linux/console.h>
-#include <linux/mca.h>
-#include <linux/root_dev.h>
-#include <linux/highmem.h>
-#include <linux/module.h>
-#include <linux/efi.h>
-#include <linux/init.h>
-#include <linux/edd.h>
-#include <linux/iscsi_ibft.h>
-#include <linux/nodemask.h>
-#include <linux/kernel.h>
-#include <linux/percpu.h>
-#include <linux/notifier.h>
-#include <linux/kexec.h>
-#include <linux/crash_dump.h>
-#include <linux/dmi.h>
-#include <linux/pfn.h>
-#include <linux/pci.h>
-#include <linux/init_ohci1394_dma.h>
-#include <linux/kvm_para.h>
-
-#include <video/edid.h>
-
-#include <asm/mtrr.h>
-#include <asm/apic.h>
-#include <asm/e820.h>
-#include <asm/mpspec.h>
-#include <asm/mmzone.h>
-#include <asm/setup.h>
-#include <asm/arch_hooks.h>
-#include <asm/sections.h>
-#include <asm/io_apic.h>
-#include <asm/ist.h>
-#include <asm/io.h>
-#include <asm/hypervisor.h>
-#include <xen/interface/physdev.h>
-#include <xen/interface/memory.h>
-#include <xen/features.h>
-#include <xen/firmware.h>
-#include <xen/xencons.h>
-#include <setup_arch.h>
-#include <asm/bios_ebda.h>
-#include <asm/cacheflush.h>
-#include <asm/processor.h>
-
-#ifdef CONFIG_XEN
-#include <xen/interface/kexec.h>
-#endif
-
-static int xen_panic_event(struct notifier_block *, unsigned long, void *);
-static struct notifier_block xen_panic_block = {
-       xen_panic_event, NULL, 0 /* try to go last */
-};
-
-/*
- * Machine setup..
- */
-static struct resource data_resource = {
-       .name   = "Kernel data",
-       .start  = 0,
-       .end    = 0,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_MEM
-};
-
-static struct resource code_resource = {
-       .name   = "Kernel code",
-       .start  = 0,
-       .end    = 0,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_MEM
-};
-
-static struct resource bss_resource = {
-       .name   = "Kernel bss",
-       .start  = 0,
-       .end    = 0,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_MEM
-};
-
-static struct resource video_ram_resource = {
-       .name   = "Video RAM area",
-       .start  = 0xa0000,
-       .end    = 0xbffff,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_MEM
-};
-
-static struct resource standard_io_resources[] = { {
-       .name   = "dma1",
-       .start  = 0x0000,
-       .end    = 0x001f,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_IO
-}, {
-       .name   = "pic1",
-       .start  = 0x0020,
-       .end    = 0x0021,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_IO
-}, {
-       .name   = "timer0",
-       .start  = 0x0040,
-       .end    = 0x0043,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_IO
-}, {
-       .name   = "timer1",
-       .start  = 0x0050,
-       .end    = 0x0053,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_IO
-}, {
-       .name   = "keyboard",
-       .start  = 0x0060,
-       .end    = 0x0060,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_IO
-}, {
-       .name   = "keyboard",
-       .start  = 0x0064,
-       .end    = 0x0064,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_IO
-}, {
-       .name   = "dma page reg",
-       .start  = 0x0080,
-       .end    = 0x008f,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_IO
-}, {
-       .name   = "pic2",
-       .start  = 0x00a0,
-       .end    = 0x00a1,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_IO
-}, {
-       .name   = "dma2",
-       .start  = 0x00c0,
-       .end    = 0x00df,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_IO
-}, {
-       .name   = "fpu",
-       .start  = 0x00f0,
-       .end    = 0x00ff,
-       .flags  = IORESOURCE_BUSY | IORESOURCE_IO
-} };
-
-/* cpu data as detected by the assembly code in head.S */
-struct cpuinfo_x86 new_cpu_data __cpuinitdata = { 0, 0, 0, 0, -1, 1, 0, 0, -1 };
-/* common cpu data for all cpus */
-struct cpuinfo_x86 boot_cpu_data __read_mostly = { 0, 0, 0, 0, -1, 1, 0, 0, -1 };
-EXPORT_SYMBOL(boot_cpu_data);
-
-unsigned int def_to_bigsmp;
-
-#ifndef CONFIG_X86_PAE
-unsigned long mmu_cr4_features;
-#else
-unsigned long mmu_cr4_features = X86_CR4_PAE;
-#endif
-
-/* for MCA, but anyone else can use it if they want */
-unsigned int machine_id;
-unsigned int machine_submodel_id;
-unsigned int BIOS_revision;
-
-/* Boot loader ID as an integer, for the benefit of proc_dointvec */
-int bootloader_type;
-
-/* user-defined highmem size */
-static unsigned int highmem_pages = -1;
-
-/*
- * Setup options
- */
-struct screen_info screen_info;
-EXPORT_SYMBOL(screen_info);
-struct apm_info apm_info;
-EXPORT_SYMBOL(apm_info);
-struct edid_info edid_info;
-EXPORT_SYMBOL_GPL(edid_info);
-#ifndef CONFIG_XEN
-#define copy_edid() (edid_info = boot_params.edid_info)
-#endif
-struct ist_info ist_info;
-#if defined(CONFIG_X86_SPEEDSTEP_SMI) || \
-       defined(CONFIG_X86_SPEEDSTEP_SMI_MODULE)
-EXPORT_SYMBOL(ist_info);
-#endif
-
-extern void early_cpu_init(void);
-extern int root_mountflags;
-
-unsigned long saved_video_mode;
-
-#define RAMDISK_IMAGE_START_MASK       0x07FF
-#define RAMDISK_PROMPT_FLAG            0x8000
-#define RAMDISK_LOAD_FLAG              0x4000
-
-static char __initdata command_line[COMMAND_LINE_SIZE];
-
-#ifndef CONFIG_DEBUG_BOOT_PARAMS
-struct boot_params __initdata boot_params;
-#else
-struct boot_params boot_params;
-#endif
-
-/*
- * Point at the empty zero page to start with. We map the real shared_info
- * page as soon as fixmap is up and running.
- */
-shared_info_t *HYPERVISOR_shared_info = (shared_info_t *)empty_zero_page;
-EXPORT_SYMBOL(HYPERVISOR_shared_info);
-
-unsigned long *phys_to_machine_mapping;
-unsigned long *pfn_to_mfn_frame_list_list, *pfn_to_mfn_frame_list[16];
-EXPORT_SYMBOL(phys_to_machine_mapping);
-
-/* Raw start-of-day parameters from the hypervisor. */
-start_info_t *xen_start_info;
-EXPORT_SYMBOL(xen_start_info);
-
-#if defined(CONFIG_EDD) || defined(CONFIG_EDD_MODULE)
-struct edd edd;
-#ifdef CONFIG_EDD_MODULE
-EXPORT_SYMBOL(edd);
-#endif
-#ifndef CONFIG_XEN
-/**
- * copy_edd() - Copy the BIOS EDD information
- *              from boot_params into a safe place.
- *
- */
-static inline void copy_edd(void)
-{
-     memcpy(edd.mbr_signature, boot_params.edd_mbr_sig_buffer,
-           sizeof(edd.mbr_signature));
-     memcpy(edd.edd_info, boot_params.eddbuf, sizeof(edd.edd_info));
-     edd.mbr_signature_nr = boot_params.edd_mbr_sig_buf_entries;
-     edd.edd_info_nr = boot_params.eddbuf_entries;
-}
-#endif
-#else
-static inline void copy_edd(void)
-{
-}
-#endif
-
-int __initdata user_defined_memmap;
-
-/*
- * "mem=nopentium" disables the 4MB page tables.
- * "mem=XXX[kKmM]" defines a memory region from HIGH_MEM
- * to <mem>, overriding the bios size.
- * "memmap=XXX[KkmM]@XXX[KkmM]" defines a memory region from
- * <start> to <start>+<mem>, overriding the bios size.
- *
- * HPA tells me bootloaders need to parse mem=, so no new
- * option should be mem=  [also see Documentation/i386/boot.txt]
- */
-static int __init parse_mem(char *arg)
-{
-       if (!arg)
-               return -EINVAL;
-
-       if (strcmp(arg, "nopentium") == 0) {
-               setup_clear_cpu_cap(X86_FEATURE_PSE);
-       } else {
-               /* If the user specifies memory size, we
-                * limit the BIOS-provided memory map to
-                * that size. exactmap can be used to specify
-                * the exact map. mem=number can be used to
-                * trim the existing memory map.
-                */
-               unsigned long long mem_size;
-
-               mem_size = memparse(arg, &arg);
-               limit_regions(mem_size);
-               user_defined_memmap = 1;
-       }
-       return 0;
-}
-early_param("mem", parse_mem);
-
-#ifdef CONFIG_PROC_VMCORE
-/* elfcorehdr= specifies the location of elf core header
- * stored by the crashed kernel.
- */
-static int __init parse_elfcorehdr(char *arg)
-{
-       if (!arg)
-               return -EINVAL;
-
-       elfcorehdr_addr = memparse(arg, &arg);
-       return 0;
-}
-early_param("elfcorehdr", parse_elfcorehdr);
-#endif /* CONFIG_PROC_VMCORE */
-
-/*
- * highmem=size forces highmem to be exactly 'size' bytes.
- * This works even on boxes that have no highmem otherwise.
- * This also works to reduce highmem size on bigger boxes.
- */
-static int __init parse_highmem(char *arg)
-{
-       if (!arg)
-               return -EINVAL;
-
-       highmem_pages = memparse(arg, &arg) >> PAGE_SHIFT;
-       return 0;
-}
-early_param("highmem", parse_highmem);
-
-/*
- * vmalloc=size forces the vmalloc area to be exactly 'size'
- * bytes. This can be used to increase (or decrease) the
- * vmalloc area - the default is 128m.
- */
-static int __init parse_vmalloc(char *arg)
-{
-       if (!arg)
-               return -EINVAL;
-
-       __VMALLOC_RESERVE = memparse(arg, &arg);
-       return 0;
-}
-early_param("vmalloc", parse_vmalloc);
-
-#ifndef CONFIG_XEN
-/*
- * reservetop=size reserves a hole at the top of the kernel address space which
- * a hypervisor can load into later.  Needed for dynamically loaded hypervisors,
- * so relocating the fixmap can be done before paging initialization.
- */
-static int __init parse_reservetop(char *arg)
-{
-       unsigned long address;
-
-       if (!arg)
-               return -EINVAL;
-
-       address = memparse(arg, &arg);
-       reserve_top_address(address);
-       return 0;
-}
-early_param("reservetop", parse_reservetop);
-#endif
-
-/*
- * Determine low and high memory ranges:
- */
-unsigned long __init find_max_low_pfn(void)
-{
-       unsigned long max_low_pfn;
-
-       max_low_pfn = max_pfn;
-       if (max_low_pfn > MAXMEM_PFN) {
-               if (highmem_pages == -1)
-                       highmem_pages = max_pfn - MAXMEM_PFN;
-               if (highmem_pages + MAXMEM_PFN < max_pfn)
-                       max_pfn = MAXMEM_PFN + highmem_pages;
-               if (highmem_pages + MAXMEM_PFN > max_pfn) {
-                       printk("only %luMB highmem pages available, ignoring highmem size of %uMB.\n", pages_to_mb(max_pfn - MAXMEM_PFN), pages_to_mb(highmem_pages));
-                       highmem_pages = 0;
-               }
-               max_low_pfn = MAXMEM_PFN;
-#ifndef CONFIG_HIGHMEM
-               /* Maximum memory usable is what is directly addressable */
-               printk(KERN_WARNING "Warning only %ldMB will be used.\n",
-                                       MAXMEM>>20);
-               if (max_pfn > MAX_NONPAE_PFN)
-                       printk(KERN_WARNING "Use a HIGHMEM64G enabled kernel.\n");
-               else
-                       printk(KERN_WARNING "Use a HIGHMEM enabled kernel.\n");
-               max_pfn = MAXMEM_PFN;
-#else /* !CONFIG_HIGHMEM */
-#ifndef CONFIG_HIGHMEM64G
-               if (max_pfn > MAX_NONPAE_PFN) {
-                       max_pfn = MAX_NONPAE_PFN;
-                       printk(KERN_WARNING "Warning only 4GB will be used.\n");
-                       printk(KERN_WARNING "Use a HIGHMEM64G enabled kernel.\n");
-               }
-#endif /* !CONFIG_HIGHMEM64G */
-#endif /* !CONFIG_HIGHMEM */
-       } else {
-               if (highmem_pages == -1)
-                       highmem_pages = 0;
-#ifdef CONFIG_HIGHMEM
-               if (highmem_pages >= max_pfn) {
-                       printk(KERN_ERR "highmem size specified (%uMB) is bigger than pages available (%luMB)!.\n", pages_to_mb(highmem_pages), pages_to_mb(max_pfn));
-                       highmem_pages = 0;
-               }
-               if (highmem_pages) {
-                       if (max_low_pfn-highmem_pages < 64*1024*1024/PAGE_SIZE){
-                               printk(KERN_ERR "highmem size %uMB results in smaller than 64MB lowmem, ignoring it.\n", pages_to_mb(highmem_pages));
-                               highmem_pages = 0;
-                       }
-                       max_low_pfn -= highmem_pages;
-               }
-#else
-               if (highmem_pages)
-                       printk(KERN_ERR "ignoring highmem size on non-highmem kernel!\n");
-#endif
-       }
-       return max_low_pfn;
-}
-
-#ifndef CONFIG_XEN
-#define BIOS_LOWMEM_KILOBYTES 0x413
-
-/*
- * The BIOS places the EBDA/XBDA at the top of conventional
- * memory, and usually decreases the reported amount of
- * conventional memory (int 0x12) too. This also contains a
- * workaround for Dell systems that neglect to reserve EBDA.
- * The same workaround also avoids a problem with the AMD768MPX
- * chipset: reserve a page before VGA to prevent PCI prefetch
- * into it (errata #56). Usually the page is reserved anyways,
- * unless you have no PS/2 mouse plugged in.
- */
-static void __init reserve_ebda_region(void)
-{
-       unsigned int lowmem, ebda_addr;
-
-       /* To determine the position of the EBDA and the */
-       /* end of conventional memory, we need to look at */
-       /* the BIOS data area. In a paravirtual environment */
-       /* that area is absent. We'll just have to assume */
-       /* that the paravirt case can handle memory setup */
-       /* correctly, without our help. */
-       if (paravirt_enabled())
-               return;
-
-       /* end of low (conventional) memory */
-       lowmem = *(unsigned short *)__va(BIOS_LOWMEM_KILOBYTES);
-       lowmem <<= 10;
-
-       /* start of EBDA area */
-       ebda_addr = get_bios_ebda();
-
-       /* Fixup: bios puts an EBDA in the top 64K segment */
-       /* of conventional memory, but does not adjust lowmem. */
-       if ((lowmem - ebda_addr) <= 0x10000)
-               lowmem = ebda_addr;
-
-       /* Fixup: bios does not report an EBDA at all. */
-       /* Some old Dells seem to need 4k anyhow (bugzilla 2990) */
-       if ((ebda_addr == 0) && (lowmem >= 0x9f000))
-               lowmem = 0x9f000;
-
-       /* Paranoia: should never happen, but... */
-       if ((lowmem == 0) || (lowmem >= 0x100000))
-               lowmem = 0x9f000;
-
-       /* reserve all memory between lowmem and the 1MB mark */
-       reserve_bootmem(lowmem, 0x100000 - lowmem, BOOTMEM_DEFAULT);
-}
-#endif
-
-#ifndef CONFIG_NEED_MULTIPLE_NODES
-static void __init setup_bootmem_allocator(void);
-static unsigned long __init setup_memory(void)
-{
-       /*
-        * partially used pages are not usable - thus
-        * we are rounding upwards:
-        */
-       min_low_pfn = PFN_UP(__pa(xen_start_info->pt_base)) +
-               xen_start_info->nr_pt_frames;
-
-       max_low_pfn = find_max_low_pfn();
-
-#ifdef CONFIG_HIGHMEM
-       highstart_pfn = highend_pfn = max_pfn;
-       if (max_pfn > max_low_pfn) {
-               highstart_pfn = max_low_pfn;
-       }
-       printk(KERN_NOTICE "%ldMB HIGHMEM available.\n",
-               pages_to_mb(highend_pfn - highstart_pfn));
-       num_physpages = highend_pfn;
-       high_memory = (void *) __va(highstart_pfn * PAGE_SIZE - 1) + 1;
-#else
-       num_physpages = max_low_pfn;
-       high_memory = (void *) __va(max_low_pfn * PAGE_SIZE - 1) + 1;
-#endif
-#ifdef CONFIG_FLATMEM
-       max_mapnr = num_physpages;
-#endif
-       printk(KERN_NOTICE "%ldMB LOWMEM available.\n",
-                       pages_to_mb(max_low_pfn));
-
-       setup_bootmem_allocator();
-
-       return max_low_pfn;
-}
-
-static void __init zone_sizes_init(void)
-{
-       unsigned long max_zone_pfns[MAX_NR_ZONES];
-       memset(max_zone_pfns, 0, sizeof(max_zone_pfns));
-       max_zone_pfns[ZONE_DMA] =
-               virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT;
-       max_zone_pfns[ZONE_NORMAL] = max_low_pfn;
-#ifdef CONFIG_HIGHMEM
-       max_zone_pfns[ZONE_HIGHMEM] = highend_pfn;
-       add_active_range(0, 0, highend_pfn);
-#else
-       add_active_range(0, 0, max_low_pfn);
-#endif
-
-       free_area_init_nodes(max_zone_pfns);
-}
-#else
-extern unsigned long __init setup_memory(void);
-extern void zone_sizes_init(void);
-#endif /* !CONFIG_NEED_MULTIPLE_NODES */
-
-static inline unsigned long long get_total_mem(void)
-{
-       unsigned long long total;
-
-       total = max_low_pfn - min_low_pfn;
-#ifdef CONFIG_HIGHMEM
-       total += highend_pfn - highstart_pfn;
-#endif
-
-       return total << PAGE_SHIFT;
-}
-
-#ifdef CONFIG_KEXEC
-#ifndef CONFIG_XEN
-static void __init reserve_crashkernel(void)
-{
-       unsigned long long total_mem;
-       unsigned long long crash_size, crash_base;
-       int ret;
-
-       total_mem = get_total_mem();
-
-       ret = parse_crashkernel(boot_command_line, total_mem,
-                       &crash_size, &crash_base);
-       if (ret == 0 && crash_size > 0) {
-               if (crash_base > 0) {
-                       printk(KERN_INFO "Reserving %ldMB of memory at %ldMB "
-                                       "for crashkernel (System RAM: %ldMB)\n",
-                                       (unsigned long)(crash_size >> 20),
-                                       (unsigned long)(crash_base >> 20),
-                                       (unsigned long)(total_mem >> 20));
-
-                       if (reserve_bootmem(crash_base, crash_size,
-                                       BOOTMEM_EXCLUSIVE) < 0) {
-                               printk(KERN_INFO "crashkernel reservation "
-                                       "failed - memory is in use\n");
-                               return;
-                       }
-
-                       crashk_res.start = crash_base;
-                       crashk_res.end   = crash_base + crash_size - 1;
-               } else
-                       printk(KERN_INFO "crashkernel reservation failed - "
-                                       "you have to specify a base address\n");
-       }
-}
-#else
-#define reserve_crashkernel xen_machine_kexec_setup_resources
-#endif
-#else
-static inline void __init reserve_crashkernel(void)
-{}
-#endif
-
-#ifdef CONFIG_BLK_DEV_INITRD
-
-static bool do_relocate_initrd = false;
-
-static void __init reserve_initrd(void)
-{
-       unsigned long ramdisk_image = __pa(xen_start_info->mod_start);
-       unsigned long ramdisk_size  = xen_start_info->mod_len;
-       unsigned long ramdisk_end   = ramdisk_image + ramdisk_size;
-       unsigned long end_of_lowmem = max_low_pfn << PAGE_SHIFT;
-       unsigned long ramdisk_here;
-
-       initrd_start = 0;
-
-       if (!xen_start_info->mod_start || !ramdisk_size)
-               return;         /* No initrd provided by bootloader */
-
-       if (ramdisk_end < ramdisk_image) {
-               printk(KERN_ERR "initrd wraps around end of memory, "
-                      "disabling initrd\n");
-               return;
-       }
-       if (ramdisk_size >= end_of_lowmem/2) {
-               printk(KERN_ERR "initrd too large to handle, "
-                      "disabling initrd\n");
-               return;
-       }
-       if (ramdisk_end <= end_of_lowmem) {
-               /* All in lowmem, easy case */
-               reserve_bootmem(ramdisk_image, ramdisk_size, BOOTMEM_DEFAULT);
-               initrd_start = ramdisk_image + PAGE_OFFSET;
-               initrd_end = initrd_start+ramdisk_size;
-               return;
-       }
-
-       /* We need to move the initrd down into lowmem */
-       ramdisk_here = (end_of_lowmem - ramdisk_size) & PAGE_MASK;
-
-       /* Note: this includes all the lowmem currently occupied by
-          the initrd, we rely on that fact to keep the data intact. */
-       reserve_bootmem(ramdisk_here, ramdisk_size, BOOTMEM_DEFAULT);
-       initrd_start = ramdisk_here + PAGE_OFFSET;
-       initrd_end   = initrd_start + ramdisk_size;
-
-       do_relocate_initrd = true;
-}
-
-#define MAX_MAP_CHUNK  (NR_FIX_BTMAPS << PAGE_SHIFT)
-
-static void __init relocate_initrd(void)
-{
-       unsigned long ramdisk_image = boot_params.hdr.ramdisk_image;
-       unsigned long ramdisk_size  = boot_params.hdr.ramdisk_size;
-       unsigned long end_of_lowmem = max_low_pfn << PAGE_SHIFT;
-       unsigned long ramdisk_here;
-       unsigned long slop, clen, mapaddr;
-       char *p, *q;
-
-       if (!do_relocate_initrd)
-               return;
-
-       ramdisk_here = initrd_start - PAGE_OFFSET;
-
-       q = (char *)initrd_start;
-
-       /* Copy any lowmem portion of the initrd */
-       if (ramdisk_image < end_of_lowmem) {
-               clen = end_of_lowmem - ramdisk_image;
-               p = (char *)__va(ramdisk_image);
-               memcpy(q, p, clen);
-               q += clen;
-               ramdisk_image += clen;
-               ramdisk_size  -= clen;
-       }
-
-       /* Copy the highmem portion of the initrd */
-       while (ramdisk_size) {
-               slop = ramdisk_image & ~PAGE_MASK;
-               clen = ramdisk_size;
-               if (clen > MAX_MAP_CHUNK-slop)
-                       clen = MAX_MAP_CHUNK-slop;
-               mapaddr = ramdisk_image & PAGE_MASK;
-               p = early_ioremap(mapaddr, clen+slop);
-               memcpy(q, p+slop, clen);
-               early_iounmap(p, clen+slop);
-               q += clen;
-               ramdisk_image += clen;
-               ramdisk_size  -= clen;
-       }
-}
-
-#endif /* CONFIG_BLK_DEV_INITRD */
-
-void __init setup_bootmem_allocator(void)
-{
-       unsigned long bootmap_size;
-       /*
-        * Initialize the boot-time allocator (with low memory only):
-        */
-       bootmap_size = init_bootmem(min_low_pfn, max_low_pfn);
-
-       register_bootmem_low_pages(max_low_pfn);
-
-       /*
-        * Reserve the bootmem bitmap itself as well. We do this in two
-        * steps (first step was init_bootmem()) because this catches
-        * the (very unlikely) case of us accidentally initializing the
-        * bootmem allocator with an invalid RAM area.
-        */
-       reserve_bootmem(__pa_symbol(_text), (PFN_PHYS(min_low_pfn) +
-                        bootmap_size + PAGE_SIZE-1) - __pa_symbol(_text),
-                        BOOTMEM_DEFAULT);
-
-#ifndef CONFIG_XEN
-       /*
-        * reserve physical page 0 - it's a special BIOS page on many boxes,
-        * enabling clean reboots, SMP operation, laptop functions.
-        */
-       reserve_bootmem(0, PAGE_SIZE, BOOTMEM_DEFAULT);
-
-       /* reserve EBDA region */
-       reserve_ebda_region();
-
-#ifdef CONFIG_SMP
-       /*
-        * But first pinch a few for the stack/trampoline stuff
-        * FIXME: Don't need the extra page at 4K, but need to fix
-        * trampoline before removing it. (see the GDT stuff)
-        */
-       reserve_bootmem(PAGE_SIZE, PAGE_SIZE, BOOTMEM_DEFAULT);
-#endif
-#ifdef CONFIG_ACPI_SLEEP
-       /*
-        * Reserve low memory region for sleep support.
-        */
-       acpi_reserve_bootmem();
-#endif
-#endif /* !CONFIG_XEN */
-
-#ifdef CONFIG_BLK_DEV_INITRD
-       reserve_initrd();
-#endif
-       numa_kva_reserve();
-       reserve_crashkernel();
-
-       reserve_ibft_region();
-}
-
-/*
- * The node 0 pgdat is initialized before all of these because
- * it's needed for bootmem.  node>0 pgdats have their virtual
- * space allocated before the pagetables are in place to access
- * them, so they can't be cleared then.
- *
- * This should all compile down to nothing when NUMA is off.
- */
-static void __init remapped_pgdat_init(void)
-{
-       int nid;
-
-       for_each_online_node(nid) {
-               if (nid != 0)
-                       memset(NODE_DATA(nid), 0, sizeof(struct pglist_data));
-       }
-}
-
-#ifdef CONFIG_MCA
-static void set_mca_bus(int x)
-{
-       MCA_bus = x;
-}
-#else
-static void set_mca_bus(int x) { }
-#endif
-
-/* Overridden in paravirt.c if CONFIG_PARAVIRT */
-char * __init __attribute__((weak)) memory_setup(void)
-{
-       return machine_specific_memory_setup();
-}
-
-#ifdef CONFIG_NUMA
-/*
- * In the golden day, when everything among i386 and x86_64 will be
- * integrated, this will not live here
- */
-void *x86_cpu_to_node_map_early_ptr;
-int x86_cpu_to_node_map_init[NR_CPUS] = {
-       [0 ... NR_CPUS-1] = NUMA_NO_NODE
-};
-DEFINE_PER_CPU(int, x86_cpu_to_node_map) = NUMA_NO_NODE;
-#endif
-
-/*
- * Determine if we were loaded by an EFI loader.  If so, then we have also been
- * passed the efi memmap, systab, etc., so we should use these data structures
- * for initialization.  Note, the efi init code path is determined by the
- * global efi_enabled. This allows the same kernel image to be used on existing
- * systems (with a traditional BIOS) as well as on EFI systems.
- */
-void __init setup_arch(char **cmdline_p)
-{
-       int i, j, k, fpp;
-       struct physdev_set_iopl set_iopl;
-       unsigned long max_low_pfn;
-       unsigned long p2m_pages;
-
-       /* Force a quick death if the kernel panics (not domain 0). */
-       extern int panic_timeout;
-       if (!panic_timeout && !is_initial_xendomain())
-               panic_timeout = 1;
-
-       /* Register a call for panic conditions. */
-       atomic_notifier_chain_register(&panic_notifier_list, &xen_panic_block);
-
-       WARN_ON(HYPERVISOR_vm_assist(VMASST_CMD_enable,
-                                    VMASST_TYPE_4gb_segments));
-       WARN_ON(HYPERVISOR_vm_assist(VMASST_CMD_enable,
-                                    VMASST_TYPE_writable_pagetables));
-
-       memcpy(&boot_cpu_data, &new_cpu_data, sizeof(new_cpu_data));
-       pre_setup_arch_hook();
-       early_cpu_init();
-       early_ioremap_init();
-#ifdef CONFIG_SMP
-       prefill_possible_map();
-#endif
-
-#ifdef CONFIG_EFI
-       if (!strncmp((char *)&boot_params.efi_info.efi_loader_signature,
-                    "EL32", 4))
-               efi_enabled = 1;
-#endif
-
-       /* This must be initialized to UNNAMED_MAJOR for ipconfig to work
-          properly.  Setting ROOT_DEV to default to /dev/ram0 breaks initrd.
-       */
-       ROOT_DEV = MKDEV(UNNAMED_MAJOR,0);
-       screen_info = boot_params.screen_info;
-       copy_edid();
-       apm_info.bios = boot_params.apm_bios_info;
-       ist_info = boot_params.ist_info;
-       saved_video_mode = boot_params.hdr.vid_mode;
-       if( boot_params.sys_desc_table.length != 0 ) {
-               set_mca_bus(boot_params.sys_desc_table.table[3] & 0x2);
-               machine_id = boot_params.sys_desc_table.table[0];
-               machine_submodel_id = boot_params.sys_desc_table.table[1];
-               BIOS_revision = boot_params.sys_desc_table.table[2];
-       }
-       bootloader_type = boot_params.hdr.type_of_loader;
-
-       if (is_initial_xendomain()) {
-               const struct dom0_vga_console_info *info =
-                       (void *)((char *)xen_start_info +
-                                xen_start_info->console.dom0.info_off);
-
-               dom0_init_screen_info(info,
-                                     xen_start_info->console.dom0.info_size);
-               xen_start_info->console.domU.mfn = 0;
-               xen_start_info->console.domU.evtchn = 0;
-       } else
-               screen_info.orig_video_isVGA = 0;
-
-#ifdef CONFIG_BLK_DEV_RAM
-       rd_image_start = boot_params.hdr.ram_size & RAMDISK_IMAGE_START_MASK;
-       rd_prompt = ((boot_params.hdr.ram_size & RAMDISK_PROMPT_FLAG) != 0);
-       rd_doload = ((boot_params.hdr.ram_size & RAMDISK_LOAD_FLAG) != 0);
-#endif
-
-       ARCH_SETUP
-
-       printk(KERN_INFO "BIOS-provided physical RAM map:\n");
-       print_memory_map(memory_setup());
-
-       copy_edd();
-
-       if (!boot_params.hdr.root_flags)
-               root_mountflags &= ~MS_RDONLY;
-       init_mm.start_code = (unsigned long) _text;
-       init_mm.end_code = (unsigned long) _etext;
-       init_mm.end_data = (unsigned long) _edata;
-       init_mm.brk = (PFN_UP(__pa(xen_start_info->pt_base)) +
-                      xen_start_info->nr_pt_frames) << PAGE_SHIFT;
-
-       code_resource.start = virt_to_phys(_text);
-       code_resource.end = virt_to_phys(_etext)-1;
-       data_resource.start = virt_to_phys(_etext);
-       data_resource.end = virt_to_phys(_edata)-1;
-       bss_resource.start = virt_to_phys(&__bss_start);
-       bss_resource.end = virt_to_phys(&__bss_stop)-1;
-
-       if ((i = MAX_GUEST_CMDLINE) > COMMAND_LINE_SIZE)
-               i = COMMAND_LINE_SIZE;
-       memcpy(boot_command_line, xen_start_info->cmd_line, i);
-       boot_command_line[i - 1] = '\0';
-       parse_early_param();
-
-       if (user_defined_memmap) {
-               printk(KERN_INFO "user-defined physical RAM map:\n");
-               print_memory_map("user");
-       }
-
-       strlcpy(command_line, boot_command_line, COMMAND_LINE_SIZE);
-       *cmdline_p = command_line;
-
-       if (efi_enabled)
-               efi_init();
-
-       /* update e820 for memory not covered by WB MTRRs */
-       propagate_e820_map();
-       mtrr_bp_init();
-#ifndef CONFIG_XEN
-       if (mtrr_trim_uncached_memory(max_pfn))
-               propagate_e820_map();
-#endif
-
-       max_low_pfn = setup_memory();
-
-#ifdef CONFIG_KVM_CLOCK
-       kvmclock_init();
-#endif
-
-#ifdef CONFIG_VMI
-       /*
-        * Must be after max_low_pfn is determined, and before kernel
-        * pagetables are setup.
-        */
-       vmi_init();
-#endif
-       kvm_guest_init();
-
-       /*
-        * NOTE: before this point _nobody_ is allowed to allocate
-        * any memory using the bootmem allocator.  Although the
-        * allocator is now initialised only the first 8Mb of the kernel
-        * virtual address space has been mapped.  All allocations before
-        * paging_init() has completed must use the alloc_bootmem_low_pages()
-        * variant (which allocates DMA'able memory) and care must be taken
-        * not to exceed the 8Mb limit.
-        */
-
-#ifdef CONFIG_SMP
-       smp_alloc_memory(); /* AP processor realmode stacks in low memory*/
-#endif
-       paging_init();
-
-       /*
-        * NOTE: On x86-32, only from this point on, fixmaps are ready for use.
-        */
-
-#ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT
-       if (init_ohci1394_dma_early)
-               init_ohci1394_dma_on_all_controllers();
-#endif
-
-       remapped_pgdat_init();
-       sparse_init();
-       zone_sizes_init();
-
-#ifdef CONFIG_X86_FIND_SMP_CONFIG
-       /*
-        * Find and reserve possible boot-time SMP configuration:
-        */
-       find_smp_config();
-#endif
-
-       p2m_pages = max_pfn;
-       if (xen_start_info->nr_pages > max_pfn) {
-               /*
-                * the max_pfn was shrunk (probably by mem= or highmem=
-                * kernel parameter); shrink reservation with the HV
-                */
-               struct xen_memory_reservation reservation = {
-                       .address_bits = 0,
-                       .extent_order = 0,
-                       .domid = DOMID_SELF
-               };
-               unsigned int difference;
-               int ret;
-
-               difference = xen_start_info->nr_pages - max_pfn;
-
-               set_xen_guest_handle(reservation.extent_start,
-                                    ((unsigned long *)xen_start_info->mfn_list) + max_pfn);
-               reservation.nr_extents = difference;
-               ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation,
-                                          &reservation);
-               BUG_ON (ret != difference);
-       }
-       else if (max_pfn > xen_start_info->nr_pages)
-               p2m_pages = xen_start_info->nr_pages;
-
-       /* Make sure we have a correctly sized P->M table. */
-       if (!xen_feature(XENFEAT_auto_translated_physmap)) {
-               phys_to_machine_mapping = alloc_bootmem_low_pages(
-                    max_pfn * sizeof(unsigned long));
-               memset(phys_to_machine_mapping, ~0,
-                      max_pfn * sizeof(unsigned long));
-               memcpy(phys_to_machine_mapping,
-                      (unsigned long *)xen_start_info->mfn_list,
-                      p2m_pages * sizeof(unsigned long));
-               free_bootmem(
-                    __pa(xen_start_info->mfn_list),
-                    PFN_PHYS(PFN_UP(xen_start_info->nr_pages *
-                                    sizeof(unsigned long))));
-
-               /*
-                * Initialise the list of the frames that specify the list of
-                * frames that make up the p2m table. Used by save/restore
-                */
-               pfn_to_mfn_frame_list_list = alloc_bootmem_low_pages(PAGE_SIZE);
-
-               fpp = PAGE_SIZE/sizeof(unsigned long);
-               for (i=0, j=0, k=-1; i< max_pfn; i+=fpp, j++) {
-                       if ((j % fpp) == 0) {
-                               k++;
-                               BUG_ON(k>=16);
-                               pfn_to_mfn_frame_list[k] =
-                                       alloc_bootmem_low_pages(PAGE_SIZE);
-                               pfn_to_mfn_frame_list_list[k] =
-                                       virt_to_mfn(pfn_to_mfn_frame_list[k]);
-                               j=0;
-                       }
-                       pfn_to_mfn_frame_list[k][j] =
-                               virt_to_mfn(&phys_to_machine_mapping[i]);
-               }
-               HYPERVISOR_shared_info->arch.max_pfn = max_pfn;
-               HYPERVISOR_shared_info->arch.pfn_to_mfn_frame_list_list =
-                    virt_to_mfn(pfn_to_mfn_frame_list_list);
-       }
-
-       /* Mark all ISA DMA channels in-use - using them wouldn't work. */
-       for (i = 0; i < MAX_DMA_CHANNELS; ++i)
-               if (i != 4 && request_dma(i, "xen") != 0)
-                       BUG();
-
-       /*
-        * NOTE: at this point the bootmem allocator is fully available.
-        */
-
-#ifdef CONFIG_BLK_DEV_INITRD
-       relocate_initrd();
-#endif
-
-       paravirt_post_allocator_init();
-
-       if (is_initial_xendomain())
-               dmi_scan_machine();
-
-       io_delay_init();
-
-#if defined(CONFIG_X86_SMP) && !defined(CONFIG_XEN)
-       /*
-        * setup to use the early static init tables during kernel startup
-        * X86_SMP will exclude sub-arches that don't deal well with it.
-        */
-       x86_cpu_to_apicid_early_ptr = (void *)x86_cpu_to_apicid_init;
-       x86_bios_cpu_apicid_early_ptr = (void *)x86_bios_cpu_apicid_init;
-#ifdef CONFIG_NUMA
-       x86_cpu_to_node_map_early_ptr = (void *)x86_cpu_to_node_map_init;
-#endif
-#endif
-
-#ifdef CONFIG_X86_GENERICARCH
-       generic_apic_probe();
-#endif
-
-       set_iopl.iopl = 1;
-       WARN_ON(HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl));
-
-#ifdef CONFIG_ACPI
-       if (!is_initial_xendomain()) {
-               printk(KERN_INFO "ACPI in unprivileged domain disabled\n");
-               acpi_disabled = 1;
-               acpi_ht = 0;
-       }
-
-       /*
-        * Parse the ACPI tables for possible boot-time SMP configuration.
-        */
-       acpi_boot_table_init();
-#endif
-
-#ifndef CONFIG_XEN
-       early_quirks();
-#endif
-
-#ifdef CONFIG_ACPI
-       acpi_boot_init();
-
-#if defined(CONFIG_SMP) && defined(CONFIG_X86_PC)
-       if (def_to_bigsmp)
-               printk(KERN_WARNING "More than 8 CPUs detected and "
-                       "CONFIG_X86_PC cannot handle it.\nUse "
-                       "CONFIG_X86_GENERICARCH or CONFIG_X86_BIGSMP.\n");
-#endif
-#endif
-#ifdef CONFIG_X86_LOCAL_APIC
-       if (smp_found_config)
-               get_smp_config();
-#endif
-
-       e820_register_memory();
-       e820_mark_nosave_regions();
-
-       if (is_initial_xendomain()) {
-#ifdef CONFIG_VT
-#if defined(CONFIG_VGA_CONSOLE)
-               if (!efi_enabled ||
-                   (efi_mem_type(0xa0000) != EFI_CONVENTIONAL_MEMORY))
-                       conswitchp = &vga_con;
-#elif defined(CONFIG_DUMMY_CONSOLE)
-               conswitchp = &dummy_con;
-#endif
-#endif
-       } else {
-#if defined(CONFIG_VT) && defined(CONFIG_DUMMY_CONSOLE)
-               conswitchp = &dummy_con;
-#endif
-       }
-}
-
-static int
-xen_panic_event(struct notifier_block *this, unsigned long event, void *ptr)
-{
-       HYPERVISOR_shutdown(SHUTDOWN_crash);
-       /* we're never actually going to get here... */
-       return NOTIFY_DONE;
-}
-
-/*
- * Request address space for all standard resources
- *
- * This is called just before pcibios_init(), which is also a
- * subsys_initcall, but is linked in later (in arch/i386/pci/common.c).
- */
-static int __init request_standard_resources(void)
-{
-       int i;
-
-       /* Nothing to do if not running in dom0. */
-       if (!is_initial_xendomain())
-               return 0;
-
-       printk(KERN_INFO "Setting up standard PCI resources\n");
-       init_iomem_resources(&code_resource, &data_resource, &bss_resource);
-
-       request_resource(&iomem_resource, &video_ram_resource);
-
-       /* request I/O space for devices used on all i[345]86 PCs */
-       for (i = 0; i < ARRAY_SIZE(standard_io_resources); i++)
-               request_resource(&ioport_resource, &standard_io_resources[i]);
-       return 0;
-}
-
-subsys_initcall(request_standard_resources);
Index: head-2008-12-01/arch/x86/kernel/setup_64-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/setup_64-xen.c 2008-12-01 11:44:55.000000000 +0100
+++ /dev/null   1970-01-01 00:00:00.000000000 +0000
@@ -1,1433 +0,0 @@
-/*
- *  Copyright (C) 1995  Linus Torvalds
- */
-
-/*
- * This file handles the architecture-dependent parts of initialization
- */
-
-#include <linux/errno.h>
-#include <linux/sched.h>
-#include <linux/kernel.h>
-#include <linux/mm.h>
-#include <linux/stddef.h>
-#include <linux/unistd.h>
-#include <linux/ptrace.h>
-#include <linux/slab.h>
-#include <linux/user.h>
-#include <linux/screen_info.h>
-#include <linux/ioport.h>
-#include <linux/delay.h>
-#include <linux/init.h>
-#include <linux/initrd.h>
-#include <linux/highmem.h>
-#include <linux/bootmem.h>
-#include <linux/module.h>
-#include <asm/processor.h>
-#include <linux/console.h>
-#include <linux/seq_file.h>
-#include <linux/crash_dump.h>
-#include <linux/root_dev.h>
-#include <linux/pci.h>
-#include <asm/pci-direct.h>
-#include <linux/efi.h>
-#include <linux/acpi.h>
-#include <linux/kallsyms.h>
-#include <linux/edd.h>
-#include <linux/iscsi_ibft.h>
-#include <linux/mmzone.h>
-#include <linux/kexec.h>
-#include <linux/cpufreq.h>
-#include <linux/dmi.h>
-#include <linux/dma-mapping.h>
-#include <linux/ctype.h>
-#include <linux/sort.h>
-#include <linux/uaccess.h>
-#include <linux/init_ohci1394_dma.h>
-#include <linux/kvm_para.h>
-
-#include <asm/mtrr.h>
-#include <asm/uaccess.h>
-#include <asm/system.h>
-#include <asm/vsyscall.h>
-#include <asm/io.h>
-#include <asm/smp.h>
-#include <asm/msr.h>
-#include <asm/desc.h>
-#include <video/edid.h>
-#include <asm/e820.h>
-#include <asm/dma.h>
-#include <asm/gart.h>
-#include <asm/mpspec.h>
-#include <asm/mmu_context.h>
-#include <asm/proto.h>
-#include <asm/setup.h>
-#include <asm/numa.h>
-#include <asm/sections.h>
-#include <asm/dmi.h>
-#include <asm/cacheflush.h>
-#include <asm/mce.h>
-#include <asm/ds.h>
-#include <asm/topology.h>
-#include <asm/pat.h>
-
-#include <mach_apic.h>
-#ifdef CONFIG_XEN
-#include <linux/percpu.h>
-#include <xen/interface/physdev.h>
-#include "setup_arch_pre.h"
-#include <asm/hypervisor.h>
-#include <xen/interface/nmi.h>
-#include <xen/features.h>
-#include <xen/firmware.h>
-#include <xen/xencons.h>
-#define PFN_UP(x)       (((x) + PAGE_SIZE-1) >> PAGE_SHIFT)
-#define PFN_PHYS(x)     ((x) << PAGE_SHIFT)
-#include <asm/mach-xen/setup_arch_post.h>
-#include <xen/interface/memory.h>
-
-#ifdef CONFIG_XEN
-#include <xen/interface/kexec.h>
-#endif
-
-extern unsigned long start_pfn;
-extern struct edid_info edid_info;
-
-shared_info_t *HYPERVISOR_shared_info = (shared_info_t *)empty_zero_page;
-EXPORT_SYMBOL(HYPERVISOR_shared_info);
-
-static int xen_panic_event(struct notifier_block *, unsigned long, void *);
-static struct notifier_block xen_panic_block = {
-       xen_panic_event, NULL, 0 /* try to go last */
-};
-
-unsigned long *phys_to_machine_mapping;
-unsigned long *pfn_to_mfn_frame_list_list, *pfn_to_mfn_frame_list[512];
-
-EXPORT_SYMBOL(phys_to_machine_mapping);
-
-DEFINE_PER_CPU(multicall_entry_t, multicall_list[8]);
-DEFINE_PER_CPU(int, nr_multicall_ents);
-
-/* Raw start-of-day parameters from the hypervisor. */
-start_info_t *xen_start_info;
-EXPORT_SYMBOL(xen_start_info);
-#endif
-
-/*
- * Machine setup..
- */
-
-struct cpuinfo_x86 boot_cpu_data __read_mostly;
-EXPORT_SYMBOL(boot_cpu_data);
-
-__u32 cleared_cpu_caps[NCAPINTS] __cpuinitdata;
-
-unsigned long mmu_cr4_features;
-
-/* Boot loader ID as an integer, for the benefit of proc_dointvec */
-int bootloader_type;
-
-unsigned long saved_video_mode;
-
-int force_mwait __cpuinitdata;
-
-/*
- * Early DMI memory
- */
-int dmi_alloc_index;
-char dmi_alloc_data[DMI_MAX_DATA];
-
-/*
- * Setup options
- */
-struct screen_info screen_info;
-EXPORT_SYMBOL(screen_info);
-struct sys_desc_table_struct {
-       unsigned short length;
-       unsigned char table[0];
-};
-
-struct edid_info edid_info;
-EXPORT_SYMBOL_GPL(edid_info);
-
-extern int root_mountflags;
-
-char __initdata command_line[COMMAND_LINE_SIZE];
-
-static struct resource standard_io_resources[] = {
-       { .name = "dma1", .start = 0x00, .end = 0x1f,
-               .flags = IORESOURCE_BUSY | IORESOURCE_IO },
-       { .name = "pic1", .start = 0x20, .end = 0x21,
-               .flags = IORESOURCE_BUSY | IORESOURCE_IO },
-       { .name = "timer0", .start = 0x40, .end = 0x43,
-               .flags = IORESOURCE_BUSY | IORESOURCE_IO },
-       { .name = "timer1", .start = 0x50, .end = 0x53,
-               .flags = IORESOURCE_BUSY | IORESOURCE_IO },
-       { .name = "keyboard", .start = 0x60, .end = 0x60,
-               .flags = IORESOURCE_BUSY | IORESOURCE_IO },
-       { .name = "keyboard", .start = 0x64, .end = 0x64,
-               .flags = IORESOURCE_BUSY | IORESOURCE_IO },
-       { .name = "dma page reg", .start = 0x80, .end = 0x8f,
-               .flags = IORESOURCE_BUSY | IORESOURCE_IO },
-       { .name = "pic2", .start = 0xa0, .end = 0xa1,
-               .flags = IORESOURCE_BUSY | IORESOURCE_IO },
-       { .name = "dma2", .start = 0xc0, .end = 0xdf,
-               .flags = IORESOURCE_BUSY | IORESOURCE_IO },
-       { .name = "fpu", .start = 0xf0, .end = 0xff,
-               .flags = IORESOURCE_BUSY | IORESOURCE_IO }
-};
-
-#define IORESOURCE_RAM (IORESOURCE_BUSY | IORESOURCE_MEM)
-
-static struct resource data_resource = {
-       .name = "Kernel data",
-       .start = 0,
-       .end = 0,
-       .flags = IORESOURCE_RAM,
-};
-static struct resource code_resource = {
-       .name = "Kernel code",
-       .start = 0,
-       .end = 0,
-       .flags = IORESOURCE_RAM,
-};
-static struct resource bss_resource = {
-       .name = "Kernel bss",
-       .start = 0,
-       .end = 0,
-       .flags = IORESOURCE_RAM,
-};
-
-static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c);
-
-#ifdef CONFIG_PROC_VMCORE
-/* elfcorehdr= specifies the location of elf core header
- * stored by the crashed kernel. This option will be passed
- * by kexec loader to the capture kernel.
- */
-static int __init setup_elfcorehdr(char *arg)
-{
-       char *end;
-       if (!arg)
-               return -EINVAL;
-       elfcorehdr_addr = memparse(arg, &end);
-       return end > arg ? 0 : -EINVAL;
-}
-early_param("elfcorehdr", setup_elfcorehdr);
-#endif
-
-#ifndef CONFIG_NUMA
-static void __init
-contig_initmem_init(unsigned long start_pfn, unsigned long end_pfn)
-{
-       unsigned long bootmap_size, bootmap;
-
-       bootmap_size = bootmem_bootmap_pages(end_pfn)<<PAGE_SHIFT;
-       bootmap = find_e820_area(0, end_pfn<<PAGE_SHIFT, bootmap_size,
-                                PAGE_SIZE);
-       if (bootmap == -1L)
-               panic("Cannot find bootmem map of size %ld\n", bootmap_size);
-       bootmap_size = init_bootmem(bootmap >> PAGE_SHIFT, end_pfn);
-       e820_register_active_regions(0, start_pfn, end_pfn);
-#ifdef CONFIG_XEN
-       free_bootmem_with_active_regions(0, xen_start_info->nr_pages);
-       early_res_to_bootmem(0, xen_start_info->nr_pages<<PAGE_SHIFT);
-#else
-       free_bootmem_with_active_regions(0, end_pfn);
-       early_res_to_bootmem(0, end_pfn<<PAGE_SHIFT);
-#endif
-       reserve_bootmem(bootmap, bootmap_size, BOOTMEM_DEFAULT);
-}
-#endif
-
-#if defined(CONFIG_EDD) || defined(CONFIG_EDD_MODULE)
-struct edd edd;
-#ifdef CONFIG_EDD_MODULE
-EXPORT_SYMBOL(edd);
-#endif
-#ifndef CONFIG_XEN
-/**
- * copy_edd() - Copy the BIOS EDD information
- *              from boot_params into a safe place.
- *
- */
-static inline void copy_edd(void)
-{
-     memcpy(edd.mbr_signature, boot_params.edd_mbr_sig_buffer,
-           sizeof(edd.mbr_signature));
-     memcpy(edd.edd_info, boot_params.eddbuf, sizeof(edd.edd_info));
-     edd.mbr_signature_nr = boot_params.edd_mbr_sig_buf_entries;
-     edd.edd_info_nr = boot_params.eddbuf_entries;
-}
-#endif
-#else
-static inline void copy_edd(void)
-{
-}
-#endif
-
-#ifdef CONFIG_KEXEC
-#ifndef CONFIG_XEN
-static void __init reserve_crashkernel(void)
-{
-       unsigned long long total_mem;
-       unsigned long long crash_size, crash_base;
-       int ret;
-
-       total_mem = ((unsigned long long)max_low_pfn - min_low_pfn) << PAGE_SHIFT;
-
-       ret = parse_crashkernel(boot_command_line, total_mem,
-                       &crash_size, &crash_base);
-       if (ret == 0 && crash_size) {
-               if (crash_base <= 0) {
-                       printk(KERN_INFO "crashkernel reservation failed - "
-                                       "you have to specify a base address\n");
-                       return;
-               }
-
-               if (reserve_bootmem(crash_base, crash_size,
-                                       BOOTMEM_EXCLUSIVE) < 0) {
-                       printk(KERN_INFO "crashkernel reservation failed - "
-                                       "memory is in use\n");
-                       return;
-               }
-
-               printk(KERN_INFO "Reserving %ldMB of memory at %ldMB "
-                               "for crashkernel (System RAM: %ldMB)\n",
-                               (unsigned long)(crash_size >> 20),
-                               (unsigned long)(crash_base >> 20),
-                               (unsigned long)(total_mem >> 20));
-               crashk_res.start = crash_base;
-               crashk_res.end   = crash_base + crash_size - 1;
-               insert_resource(&iomem_resource, &crashk_res);
-       }
-}
-#else
-#define reserve_crashkernel xen_machine_kexec_setup_resources
-#endif
-#else
-static inline void __init reserve_crashkernel(void)
-{}
-#endif
-
-/* Overridden in paravirt.c if CONFIG_PARAVIRT */
-void __attribute__((weak)) __init memory_setup(void)
-{
-       machine_specific_memory_setup();
-}
-
-static void __init parse_setup_data(void)
-{
-       struct setup_data *data;
-       unsigned long pa_data;
-
-       if (boot_params.hdr.version < 0x0209)
-               return;
-       pa_data = boot_params.hdr.setup_data;
-       while (pa_data) {
-               data = early_ioremap(pa_data, PAGE_SIZE);
-               switch (data->type) {
-               default:
-                       break;
-               }
-#ifndef CONFIG_DEBUG_BOOT_PARAMS
-               free_early(pa_data, pa_data+sizeof(*data)+data->len);
-#endif
-               pa_data = data->next;
-               early_iounmap(data, PAGE_SIZE);
-       }
-}
-
-#ifdef CONFIG_PCI_MMCONFIG
-extern void __cpuinit fam10h_check_enable_mmcfg(void);
-extern void __init check_enable_amd_mmconf_dmi(void);
-#else
-void __cpuinit fam10h_check_enable_mmcfg(void)
-{
-}
-void __init check_enable_amd_mmconf_dmi(void)
-{
-}
-#endif
-
-/*
- * setup_arch - architecture-specific boot-time initializations
- *
- * Note: On x86_64, fixmaps are ready for use even before this is called.
- */
-void __init setup_arch(char **cmdline_p)
-{
-       unsigned i;
-
-#ifdef CONFIG_XEN
-       extern struct e820map machine_e820;
-
-       printk(KERN_INFO "Command line: %s\n", boot_command_line);
-
-       /* Register a call for panic conditions. */
-       atomic_notifier_chain_register(&panic_notifier_list, &xen_panic_block);
-
-       WARN_ON(HYPERVISOR_vm_assist(VMASST_CMD_enable,
-                                    VMASST_TYPE_writable_pagetables));
-
-       early_ioremap_init();
-
-       ROOT_DEV = MKDEV(RAMDISK_MAJOR,0); 
-       screen_info = boot_params.screen_info;
-
-       if (is_initial_xendomain()) {
-               const struct dom0_vga_console_info *info =
-                       (void *)((char *)xen_start_info +
-                                xen_start_info->console.dom0.info_off);
-
-               dom0_init_screen_info(info,
-                                     xen_start_info->console.dom0.info_size);
-               xen_start_info->console.domU.mfn = 0;
-               xen_start_info->console.domU.evtchn = 0;
-       } else
-               screen_info.orig_video_isVGA = 0;
-
-       copy_edid();
-#else
-       printk(KERN_INFO "Command line: %s\n", boot_command_line);
-
-       ROOT_DEV = old_decode_dev(boot_params.hdr.root_dev);
-       screen_info = boot_params.screen_info;
-       edid_info = boot_params.edid_info;
-#endif /* !CONFIG_XEN */
-       saved_video_mode = boot_params.hdr.vid_mode;
-       bootloader_type = boot_params.hdr.type_of_loader;
-
-#ifdef CONFIG_BLK_DEV_RAM
-       rd_image_start = boot_params.hdr.ram_size & RAMDISK_IMAGE_START_MASK;
-       rd_prompt = ((boot_params.hdr.ram_size & RAMDISK_PROMPT_FLAG) != 0);
-       rd_doload = ((boot_params.hdr.ram_size & RAMDISK_LOAD_FLAG) != 0);
-#endif
-#ifdef CONFIG_EFI
-       if (!strncmp((char *)&boot_params.efi_info.efi_loader_signature,
-                    "EL64", 4))
-               efi_enabled = 1;
-#endif
-
-       ARCH_SETUP
-
-       memory_setup();
-       copy_edd();
-
-       if (!boot_params.hdr.root_flags)
-               root_mountflags &= ~MS_RDONLY;
-       init_mm.start_code = (unsigned long) &_text;
-       init_mm.end_code = (unsigned long) &_etext;
-       init_mm.end_data = (unsigned long) &_edata;
-       init_mm.brk = (unsigned long) &_end;
-
-       code_resource.start = virt_to_phys(&_text);
-       code_resource.end = virt_to_phys(&_etext)-1;
-       data_resource.start = virt_to_phys(&_etext);
-       data_resource.end = virt_to_phys(&_edata)-1;
-       bss_resource.start = virt_to_phys(&__bss_start);
-       bss_resource.end = virt_to_phys(&__bss_stop)-1;
-
-       early_identify_cpu(&boot_cpu_data);
-
-       strlcpy(command_line, boot_command_line, COMMAND_LINE_SIZE);
-       *cmdline_p = command_line;
-
-       parse_setup_data();
-
-       parse_early_param();
-
-#ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT
-       if (init_ohci1394_dma_early)
-               init_ohci1394_dma_on_all_controllers();
-#endif
-
-       finish_e820_parsing();
-
-#ifndef CONFIG_XEN
-       /* after parse_early_param, so could debug it */
-       insert_resource(&iomem_resource, &code_resource);
-       insert_resource(&iomem_resource, &data_resource);
-       insert_resource(&iomem_resource, &bss_resource);
-#endif
-
-       early_gart_iommu_check();
-
-       e820_register_active_regions(0, 0, -1UL);
-       /*
-        * partially used pages are not usable - thus
-        * we are rounding upwards:
-        */
-       end_pfn = e820_end_of_ram();
-       /* update e820 for memory not covered by WB MTRRs */
-       mtrr_bp_init();
-#ifndef CONFIG_XEN
-       if (mtrr_trim_uncached_memory(end_pfn)) {
-               e820_register_active_regions(0, 0, -1UL);
-               end_pfn = e820_end_of_ram();
-       }
-#endif
-
-       num_physpages = end_pfn;
-       max_mapnr = end_pfn;
-
-       check_efer();
-
-       max_pfn_mapped = init_memory_mapping(0, (max_pfn_mapped << PAGE_SHIFT));
-       if (efi_enabled)
-               efi_init();
-
-#ifndef CONFIG_XEN
-       vsmp_init();
-#endif
-
-       if (is_initial_xendomain())
-               dmi_scan_machine();
-
-       io_delay_init();
-
-#ifdef CONFIG_KVM_CLOCK
-       kvmclock_init();
-#endif
-
-#if defined(CONFIG_SMP) && !defined(CONFIG_XEN)
-       /* setup to use the early static init tables during kernel startup */
-       x86_cpu_to_apicid_early_ptr = (void *)x86_cpu_to_apicid_init;
-       x86_bios_cpu_apicid_early_ptr = (void *)x86_bios_cpu_apicid_init;
-#ifdef CONFIG_NUMA
-       x86_cpu_to_node_map_early_ptr = (void *)x86_cpu_to_node_map_init;
-#endif
-#endif
-
-       /* How many end-of-memory variables you have, grandma! */
-       max_low_pfn = end_pfn;
-       max_pfn = end_pfn;
-       high_memory = (void *)__va(end_pfn * PAGE_SIZE - 1) + 1;
-
-       /* Remove active ranges so rediscovery with NUMA-awareness happens */
-       remove_all_active_ranges();
-
-#ifdef CONFIG_ACPI_NUMA
-       /*
-        * Parse SRAT to discover nodes.
-        */
-       acpi_numa_init();
-#endif
-
-#ifdef CONFIG_NUMA
-       numa_initmem_init(0, end_pfn);
-#else
-       contig_initmem_init(0, end_pfn);
-#endif
-
-#ifndef CONFIG_XEN
-       dma32_reserve_bootmem();
-
-#ifdef CONFIG_ACPI_SLEEP
-       /*
-        * Reserve low memory region for sleep support.
-        */
-       acpi_reserve_bootmem();
-#endif
-
-       if (efi_enabled)
-               efi_reserve_bootmem();
-#endif
-
-#ifdef CONFIG_BLK_DEV_INITRD
-#ifdef CONFIG_XEN
-       if (xen_start_info->mod_start) {
-               unsigned long ramdisk_image = __pa(xen_start_info->mod_start);
-               unsigned long ramdisk_size  = xen_start_info->mod_len;
-#else
-       if (boot_params.hdr.type_of_loader && boot_params.hdr.ramdisk_image) {
-               unsigned long ramdisk_image = boot_params.hdr.ramdisk_image;
-               unsigned long ramdisk_size  = boot_params.hdr.ramdisk_size;
-#endif
-               unsigned long ramdisk_end   = ramdisk_image + ramdisk_size;
-               unsigned long end_of_mem    = end_pfn << PAGE_SHIFT;
-
-               if (ramdisk_end <= end_of_mem) {
-                       /*
-                        * don't need to reserve again, already reserved early
-                        * in x86_64_start_kernel, and early_res_to_bootmem
-                        * convert that to reserved in bootmem
-                        */
-                       initrd_start = ramdisk_image + PAGE_OFFSET;
-                       initrd_end = initrd_start+ramdisk_size;
-#ifdef CONFIG_XEN
-                       initrd_below_start_ok = 1;
-#endif
-               } else {
-                       free_bootmem(ramdisk_image, ramdisk_size);
-                       printk(KERN_ERR "initrd extends beyond end of memory "
-                              "(0x%08lx > 0x%08lx)\ndisabling initrd\n",
-                              ramdisk_end, end_of_mem);
-                       initrd_start = 0;
-               }
-       }
-#endif
-       reserve_crashkernel();
-
-       reserve_ibft_region();
-
-       paging_init();
-       map_vsyscall();
-#ifdef CONFIG_X86_LOCAL_APIC
-       /*
-       * Find and reserve possible boot-time SMP configuration:
-       */
-       find_smp_config();
-#endif
-#ifdef CONFIG_XEN
-       {
-               int i, j, k, fpp;
-               unsigned long p2m_pages;
-
-               p2m_pages = end_pfn;
-               if (xen_start_info->nr_pages > end_pfn) {
-                       /*
-                        * the end_pfn was shrunk (probably by mem= or highmem=
-                        * kernel parameter); shrink reservation with the HV
-                        */
-                       struct xen_memory_reservation reservation = {
-                               .address_bits = 0,
-                               .extent_order = 0,
-                               .domid = DOMID_SELF
-                       };
-                       unsigned int difference;
-                       int ret;
-                       
-                       difference = xen_start_info->nr_pages - end_pfn;
-                       
-                       set_xen_guest_handle(reservation.extent_start,
-                                            ((unsigned long *)xen_start_info->mfn_list) + end_pfn);
-                       reservation.nr_extents = difference;
-                       ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation,
-                                                  &reservation);
-                       BUG_ON (ret != difference);
-               }
-               else if (end_pfn > xen_start_info->nr_pages)
-                       p2m_pages = xen_start_info->nr_pages;
-
-               if (!xen_feature(XENFEAT_auto_translated_physmap)) {
-                       /* Make sure we have a large enough P->M table. */
-                       phys_to_machine_mapping = alloc_bootmem_pages(
-                               end_pfn * sizeof(unsigned long));
-                       memset(phys_to_machine_mapping, ~0,
-                              end_pfn * sizeof(unsigned long));
-                       memcpy(phys_to_machine_mapping,
-                              (unsigned long *)xen_start_info->mfn_list,
-                              p2m_pages * sizeof(unsigned long));
-                       free_bootmem(
-                               __pa(xen_start_info->mfn_list),
-                               PFN_PHYS(PFN_UP(xen_start_info->nr_pages *
-                                               sizeof(unsigned long))));
-
-                       /*
-                        * Initialise the list of the frames that specify the
-                        * list of frames that make up the p2m table. Used by
-                         * save/restore.
-                        */
-                       pfn_to_mfn_frame_list_list = alloc_bootmem_pages(PAGE_SIZE);
-
-                       fpp = PAGE_SIZE/sizeof(unsigned long);
-                       for (i=0, j=0, k=-1; i< end_pfn; i+=fpp, j++) {
-                               if ((j % fpp) == 0) {
-                                       k++;
-                                       BUG_ON(k>=fpp);
-                                       pfn_to_mfn_frame_list[k] =
-                                               alloc_bootmem_pages(PAGE_SIZE);
-                                       pfn_to_mfn_frame_list_list[k] =
-                                               virt_to_mfn(pfn_to_mfn_frame_list[k]);
-                                       j=0;
-                               }
-                               pfn_to_mfn_frame_list[k][j] =
-                                       virt_to_mfn(&phys_to_machine_mapping[i]);
-                       }
-                       HYPERVISOR_shared_info->arch.max_pfn = end_pfn;
-                       HYPERVISOR_shared_info->arch.pfn_to_mfn_frame_list_list =
-                               virt_to_mfn(pfn_to_mfn_frame_list_list);
-               }
-
-               /* Mark all ISA DMA channels in-use - using them wouldn't work. */
-               for (i = 0; i < MAX_DMA_CHANNELS; ++i)
-                       if (i != 4 && request_dma(i, "xen") != 0)
-                               BUG();
-       }
-
-#ifdef CONFIG_ACPI
-       if (!is_initial_xendomain()) {
-               acpi_disabled = 1;
-               acpi_ht = 0;
-       }
-#endif
-#endif
-
-#ifndef CONFIG_XEN
-       early_quirks();
-#endif
-
-#ifdef CONFIG_ACPI
-       /*
-        * Initialize the ACPI boot-time table parser (gets the RSDP and SDT).
-        * Call this early for SRAT node setup.
-        */
-       acpi_boot_table_init();
-
-       /*
-        * Read APIC and some other early information from ACPI tables.
-        */
-       acpi_boot_init();
-#endif
-
-       init_cpu_to_node();
-
-#ifdef CONFIG_X86_LOCAL_APIC
-       /*
-        * get boot-time SMP configuration:
-        */
-       if (smp_found_config)
-               get_smp_config();
-#ifndef CONFIG_XEN
-       init_apic_mappings();
-       ioapic_init_mappings();
-#endif
-#endif
-#if defined(CONFIG_XEN) && defined(CONFIG_SMP) && !defined(CONFIG_HOTPLUG_CPU)
-       prefill_possible_map();
-#endif
-
-       kvm_guest_init();
-
-       /*
-        * We trust e820 completely. No explicit ROM probing in memory.
-        */
-#ifdef CONFIG_XEN
-       if (is_initial_xendomain())
-               e820_reserve_resources(machine_e820.map, machine_e820.nr_map);
-#else
-       e820_reserve_resources(e820.map, e820.nr_map);
-       e820_mark_nosave_regions();
-#endif
-
-       /* request I/O space for devices used on all i[345]86 PCs */
-       for (i = 0; i < ARRAY_SIZE(standard_io_resources); i++)
-               request_resource(&ioport_resource, &standard_io_resources[i]);
-
-#ifdef CONFIG_XEN
-       if (is_initial_xendomain())
-               e820_setup_gap(machine_e820.map, machine_e820.nr_map);
-#else
-       e820_setup_gap(e820.map, e820.nr_map);
-#endif
-
-#ifdef CONFIG_XEN
-       {
-               struct physdev_set_iopl set_iopl;
-
-               set_iopl.iopl = 1;
-               WARN_ON(HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl));
-
-               if (is_initial_xendomain()) {
-#ifdef CONFIG_VT
-#if defined(CONFIG_VGA_CONSOLE)
-                       conswitchp = &vga_con;
-#elif defined(CONFIG_DUMMY_CONSOLE)
-                       conswitchp = &dummy_con;
-#endif
-#endif
-               } else {
-#if defined(CONFIG_VT) && defined(CONFIG_DUMMY_CONSOLE)
-                       conswitchp = &dummy_con;
-#endif
-                }
-       }
-#else  /* CONFIG_XEN */
-
-#ifdef CONFIG_VT
-#if defined(CONFIG_VGA_CONSOLE)
-       if (!efi_enabled || (efi_mem_type(0xa0000) != EFI_CONVENTIONAL_MEMORY))
-               conswitchp = &vga_con;
-#elif defined(CONFIG_DUMMY_CONSOLE)
-       conswitchp = &dummy_con;
-#endif
-#endif
-
-#endif /* !CONFIG_XEN */
-
-       /* do this before identify_cpu for boot cpu */
-       check_enable_amd_mmconf_dmi();
-}
-
-#ifdef CONFIG_XEN
-static int
-xen_panic_event(struct notifier_block *this, unsigned long event, void *ptr)
-{
-       HYPERVISOR_shutdown(SHUTDOWN_crash);
-       /* we're never actually going to get here... */
-       return NOTIFY_DONE;
-}
-#endif /* !CONFIG_XEN */
-
-
-static int __cpuinit get_model_name(struct cpuinfo_x86 *c)
-{
-       unsigned int *v;
-
-       if (c->extended_cpuid_level < 0x80000004)
-               return 0;
-
-       v = (unsigned int *) c->x86_model_id;
-       cpuid(0x80000002, &v[0], &v[1], &v[2], &v[3]);
-       cpuid(0x80000003, &v[4], &v[5], &v[6], &v[7]);
-       cpuid(0x80000004, &v[8], &v[9], &v[10], &v[11]);
-       c->x86_model_id[48] = 0;
-       return 1;
-}
-
-
-static void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c)
-{
-       unsigned int n, dummy, eax, ebx, ecx, edx;
-
-       n = c->extended_cpuid_level;
-
-       if (n >= 0x80000005) {
-               cpuid(0x80000005, &dummy, &ebx, &ecx, &edx);
-               printk(KERN_INFO "CPU: L1 I Cache: %dK (%d bytes/line), "
-                      "D cache %dK (%d bytes/line)\n",
-                      edx>>24, edx&0xFF, ecx>>24, ecx&0xFF);
-               c->x86_cache_size = (ecx>>24) + (edx>>24);
-               /* On K8 L1 TLB is inclusive, so don't count it */
-               c->x86_tlbsize = 0;
-       }
-
-       if (n >= 0x80000006) {
-               cpuid(0x80000006, &dummy, &ebx, &ecx, &edx);
-               ecx = cpuid_ecx(0x80000006);
-               c->x86_cache_size = ecx >> 16;
-               c->x86_tlbsize += ((ebx >> 16) & 0xfff) + (ebx & 0xfff);
-
-               printk(KERN_INFO "CPU: L2 Cache: %dK (%d bytes/line)\n",
-               c->x86_cache_size, ecx & 0xFF);
-       }
-       if (n >= 0x80000008) {
-               cpuid(0x80000008, &eax, &dummy, &dummy, &dummy);
-               c->x86_virt_bits = (eax >> 8) & 0xff;
-               c->x86_phys_bits = eax & 0xff;
-       }
-}
-
-#ifdef CONFIG_NUMA
-static int __cpuinit nearby_node(int apicid)
-{
-       int i, node;
-
-       for (i = apicid - 1; i >= 0; i--) {
-               node = apicid_to_node[i];
-               if (node != NUMA_NO_NODE && node_online(node))
-                       return node;
-       }
-       for (i = apicid + 1; i < MAX_LOCAL_APIC; i++) {
-               node = apicid_to_node[i];
-               if (node != NUMA_NO_NODE && node_online(node))
-                       return node;
-       }
-       return first_node(node_online_map); /* Shouldn't happen */
-}
-#endif
-
-/*
- * On a AMD dual core setup the lower bits of the APIC id distingush the cores.
- * Assumes number of cores is a power of two.
- */
-static void __cpuinit amd_detect_cmp(struct cpuinfo_x86 *c)
-{
-#ifdef CONFIG_SMP
-       unsigned bits;
-#ifdef CONFIG_NUMA
-       int cpu = smp_processor_id();
-       int node = 0;
-       unsigned apicid = hard_smp_processor_id();
-#endif
-       bits = c->x86_coreid_bits;
-
-       /* Low order bits define the core id (index of core in socket) */
-       c->cpu_core_id = c->initial_apicid & ((1 << bits)-1);
-       /* Convert the initial APIC ID into the socket ID */
-       c->phys_proc_id = c->initial_apicid >> bits;
-
-#ifdef CONFIG_NUMA
-       node = c->phys_proc_id;
-       if (apicid_to_node[apicid] != NUMA_NO_NODE)
-               node = apicid_to_node[apicid];
-       if (!node_online(node)) {
-               /* Two possibilities here:
-                  - The CPU is missing memory and no node was created.
-                  In that case try picking one from a nearby CPU
-                  - The APIC IDs differ from the HyperTransport node IDs
-                  which the K8 northbridge parsing fills in.
-                  Assume they are all increased by a constant offset,
-                  but in the same order as the HT nodeids.
-                  If that doesn't result in a usable node fall back to the
-                  path for the previous case.  */
-
-               int ht_nodeid = c->initial_apicid;
-
-               if (ht_nodeid >= 0 &&
-                   apicid_to_node[ht_nodeid] != NUMA_NO_NODE)
-                       node = apicid_to_node[ht_nodeid];
-               /* Pick a nearby node */
-               if (!node_online(node))
-                       node = nearby_node(apicid);
-       }
-       numa_set_node(cpu, node);
-
-       printk(KERN_INFO "CPU %d/%x -> Node %d\n", cpu, apicid, node);
-#endif
-#endif
-}
-
-static void __cpuinit early_init_amd_mc(struct cpuinfo_x86 *c)
-{
-#ifdef CONFIG_SMP
-       unsigned bits, ecx;
-
-       /* Multi core CPU? */
-       if (c->extended_cpuid_level < 0x80000008)
-               return;
-
-       ecx = cpuid_ecx(0x80000008);
-
-       c->x86_max_cores = (ecx & 0xff) + 1;
-
-       /* CPU telling us the core id bits shift? */
-       bits = (ecx >> 12) & 0xF;
-
-       /* Otherwise recompute */
-       if (bits == 0) {
-               while ((1 << bits) < c->x86_max_cores)
-                       bits++;
-       }
-
-       c->x86_coreid_bits = bits;
-
-#endif
-}
-
-#define ENABLE_C1E_MASK                0x18000000
-#define CPUID_PROCESSOR_SIGNATURE      1
-#define CPUID_XFAM             0x0ff00000
-#define CPUID_XFAM_K8          0x00000000
-#define CPUID_XFAM_10H         0x00100000
-#define CPUID_XFAM_11H         0x00200000
-#define CPUID_XMOD             0x000f0000
-#define CPUID_XMOD_REV_F       0x00040000
-
-#ifndef CONFIG_XEN
-/* AMD systems with C1E don't have a working lAPIC timer. Check for that. */
-static __cpuinit int amd_apic_timer_broken(void)
-{
-       u32 lo, hi, eax = cpuid_eax(CPUID_PROCESSOR_SIGNATURE);
-
-       switch (eax & CPUID_XFAM) {
-       case CPUID_XFAM_K8:
-               if ((eax & CPUID_XMOD) < CPUID_XMOD_REV_F)
-                       break;
-       case CPUID_XFAM_10H:
-       case CPUID_XFAM_11H:
-               rdmsr(MSR_K8_ENABLE_C1E, lo, hi);
-               if (lo & ENABLE_C1E_MASK)
-                       return 1;
-               break;
-       default:
-               /* err on the side of caution */
-               return 1;
-       }
-       return 0;
-}
-#endif
-
-static void __cpuinit early_init_amd(struct cpuinfo_x86 *c)
-{
-       early_init_amd_mc(c);
-
-       /* c->x86_power is 8000_0007 edx. Bit 8 is constant TSC */
-       if (c->x86_power & (1<<8))
-               set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
-}
-
-static void __cpuinit init_amd(struct cpuinfo_x86 *c)
-{
-       unsigned level;
-
-#ifdef CONFIG_SMP
-       unsigned long value;
-
-       /*
-        * Disable TLB flush filter by setting HWCR.FFDIS on K8
-        * bit 6 of msr C001_0015
-        *
-        * Errata 63 for SH-B3 steppings
-        * Errata 122 for all steppings (F+ have it disabled by default)
-        */
-       if (c->x86 == 15) {
-               rdmsrl(MSR_K8_HWCR, value);
-               value |= 1 << 6;
-               wrmsrl(MSR_K8_HWCR, value);
-       }
-#endif
-
-       /* Bit 31 in normal CPUID used for nonstandard 3DNow ID;
-          3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway */
-       clear_cpu_cap(c, 0*32+31);
-
-       /* On C+ stepping K8 rep microcode works well for copy/memset */
-       level = cpuid_eax(1);
-       if (c->x86 == 15 && ((level >= 0x0f48 && level < 0x0f50) ||
-                            level >= 0x0f58))
-               set_cpu_cap(c, X86_FEATURE_REP_GOOD);
-       if (c->x86 == 0x10 || c->x86 == 0x11)
-               set_cpu_cap(c, X86_FEATURE_REP_GOOD);
-
-       /* Enable workaround for FXSAVE leak */
-       if (c->x86 >= 6)
-               set_cpu_cap(c, X86_FEATURE_FXSAVE_LEAK);
-
-       level = get_model_name(c);
-       if (!level) {
-               switch (c->x86) {
-               case 15:
-                       /* Should distinguish Models here, but this is only
-                          a fallback anyways. */
-                       strcpy(c->x86_model_id, "Hammer");
-                       break;
-               }
-       }
-       display_cacheinfo(c);
-
-       /* Multi core CPU? */
-       if (c->extended_cpuid_level >= 0x80000008)
-               amd_detect_cmp(c);
-
-       if (c->extended_cpuid_level >= 0x80000006 &&
-               (cpuid_edx(0x80000006) & 0xf000))
-               num_cache_leaves = 4;
-       else
-               num_cache_leaves = 3;
-
-       if (c->x86 == 0xf || c->x86 == 0x10 || c->x86 == 0x11)
-               set_cpu_cap(c, X86_FEATURE_K8);
-
-       /* MFENCE stops RDTSC speculation */
-       set_cpu_cap(c, X86_FEATURE_MFENCE_RDTSC);
-
-       if (c->x86 == 0x10)
-               fam10h_check_enable_mmcfg();
-
-#ifndef CONFIG_XEN
-       if (amd_apic_timer_broken())
-               disable_apic_timer = 1;
-
-       if (c == &boot_cpu_data && c->x86 >= 0xf && c->x86 <= 0x11) {
-               unsigned long long tseg;
-
-               /*
-                * Split up direct mapping around the TSEG SMM area.
-                * Don't do it for gbpages because there seems very little
-                * benefit in doing so.
-                */
-               if (!rdmsrl_safe(MSR_K8_TSEG_ADDR, &tseg) &&
-               (tseg >> PMD_SHIFT) < (max_pfn_mapped >> (PMD_SHIFT-PAGE_SHIFT)))
-                       set_memory_4k((unsigned long)__va(tseg), 1);
-       }
-#endif
-}
-
-void __cpuinit detect_ht(struct cpuinfo_x86 *c)
-{
-#ifdef CONFIG_SMP
-       u32 eax, ebx, ecx, edx;
-       int index_msb, core_bits;
-
-       cpuid(1, &eax, &ebx, &ecx, &edx);
-
-
-       if (!cpu_has(c, X86_FEATURE_HT))
-               return;
-       if (cpu_has(c, X86_FEATURE_CMP_LEGACY))
-               goto out;
-
-       smp_num_siblings = (ebx & 0xff0000) >> 16;
-
-       if (smp_num_siblings == 1) {
-               printk(KERN_INFO  "CPU: Hyper-Threading is disabled\n");
-       } else if (smp_num_siblings > 1) {
-
-               if (smp_num_siblings > NR_CPUS) {
-                       printk(KERN_WARNING "CPU: Unsupported number of "
-                              "siblings %d", smp_num_siblings);
-                       smp_num_siblings = 1;
-                       return;
-               }
-
-               index_msb = get_count_order(smp_num_siblings);
-               c->phys_proc_id = phys_pkg_id(index_msb);
-
-               smp_num_siblings = smp_num_siblings / c->x86_max_cores;
-
-               index_msb = get_count_order(smp_num_siblings);
-
-               core_bits = get_count_order(c->x86_max_cores);
-
-               c->cpu_core_id = phys_pkg_id(index_msb) &
-                                              ((1 << core_bits) - 1);
-       }
-out:
-       if ((c->x86_max_cores * smp_num_siblings) > 1) {
-               printk(KERN_INFO  "CPU: Physical Processor ID: %d\n",
-                      c->phys_proc_id);
-               printk(KERN_INFO  "CPU: Processor Core ID: %d\n",
-                      c->cpu_core_id);
-       }
-
-#endif
-}
-
-/*
- * find out the number of processor cores on the die
- */
-static int __cpuinit intel_num_cpu_cores(struct cpuinfo_x86 *c)
-{
-       unsigned int eax, t;
-
-       if (c->cpuid_level < 4)
-               return 1;
-
-       cpuid_count(4, 0, &eax, &t, &t, &t);
-
-       if (eax & 0x1f)
-               return ((eax >> 26) + 1);
-       else
-               return 1;
-}
-
-static void __cpuinit srat_detect_node(void)
-{
-#ifdef CONFIG_NUMA
-       unsigned node;
-       int cpu = smp_processor_id();
-       int apicid = hard_smp_processor_id();
-
-       /* Don't do the funky fallback heuristics the AMD version employs
-          for now. */
-       node = apicid_to_node[apicid];
-       if (node == NUMA_NO_NODE || !node_online(node))
-               node = first_node(node_online_map);
-       numa_set_node(cpu, node);
-
-       printk(KERN_INFO "CPU %d/%x -> Node %d\n", cpu, apicid, node);
-#endif
-}
-
-static void __cpuinit early_init_intel(struct cpuinfo_x86 *c)
-{
-       if ((c->x86 == 0xf && c->x86_model >= 0x03) ||
-           (c->x86 == 0x6 && c->x86_model >= 0x0e))
-               set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
-}
-
-static void __cpuinit init_intel(struct cpuinfo_x86 *c)
-{
-       /* Cache sizes */
-       unsigned n;
-
-       init_intel_cacheinfo(c);
-       if (c->cpuid_level > 9) {
-               unsigned eax = cpuid_eax(10);
-               /* Check for version and the number of counters */
-               if ((eax & 0xff) && (((eax>>8) & 0xff) > 1))
-                       set_cpu_cap(c, X86_FEATURE_ARCH_PERFMON);
-       }
-
-       if (cpu_has_ds) {
-               unsigned int l1, l2;
-               rdmsr(MSR_IA32_MISC_ENABLE, l1, l2);
-               if (!(l1 & (1<<11)))
-                       set_cpu_cap(c, X86_FEATURE_BTS);
-               if (!(l1 & (1<<12)))
-                       set_cpu_cap(c, X86_FEATURE_PEBS);
-       }
-
-
-       if (cpu_has_bts)
-               ds_init_intel(c);
-
-       n = c->extended_cpuid_level;
-       if (n >= 0x80000008) {
-               unsigned eax = cpuid_eax(0x80000008);
-               c->x86_virt_bits = (eax >> 8) & 0xff;
-               c->x86_phys_bits = eax & 0xff;
-               /* CPUID workaround for Intel 0F34 CPU */
-               if (c->x86_vendor == X86_VENDOR_INTEL &&
-                   c->x86 == 0xF && c->x86_model == 0x3 &&
-                   c->x86_mask == 0x4)
-                       c->x86_phys_bits = 36;
-       }
-
-       if (c->x86 == 15)
-               c->x86_cache_alignment = c->x86_clflush_size * 2;
-       if (c->x86 == 6)
-               set_cpu_cap(c, X86_FEATURE_REP_GOOD);
-       set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC);
-       c->x86_max_cores = intel_num_cpu_cores(c);
-
-       srat_detect_node();
-}
-
-static void __cpuinit early_init_centaur(struct cpuinfo_x86 *c)
-{
-       if (c->x86 == 0x6 && c->x86_model >= 0xf)
-               set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
-}
-
-static void __cpuinit init_centaur(struct cpuinfo_x86 *c)
-{
-       /* Cache sizes */
-       unsigned n;
-
-       n = c->extended_cpuid_level;
-       if (n >= 0x80000008) {
-               unsigned eax = cpuid_eax(0x80000008);
-               c->x86_virt_bits = (eax >> 8) & 0xff;
-               c->x86_phys_bits = eax & 0xff;
-       }
-
-       if (c->x86 == 0x6 && c->x86_model >= 0xf) {
-               c->x86_cache_alignment = c->x86_clflush_size * 2;
-               set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
-               set_cpu_cap(c, X86_FEATURE_REP_GOOD);
-       }
-       set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC);
-}
-
-static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c)
-{
-       char *v = c->x86_vendor_id;
-
-       if (!strcmp(v, "AuthenticAMD"))
-               c->x86_vendor = X86_VENDOR_AMD;
-       else if (!strcmp(v, "GenuineIntel"))
-               c->x86_vendor = X86_VENDOR_INTEL;
-       else if (!strcmp(v, "CentaurHauls"))
-               c->x86_vendor = X86_VENDOR_CENTAUR;
-       else
-               c->x86_vendor = X86_VENDOR_UNKNOWN;
-}
-
-/* Do some early cpuid on the boot CPU to get some parameter that are
-   needed before check_bugs. Everything advanced is in identify_cpu
-   below. */
-static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c)
-{
-       u32 tfms, xlvl;
-
-       c->loops_per_jiffy = loops_per_jiffy;
-       c->x86_cache_size = -1;
-       c->x86_vendor = X86_VENDOR_UNKNOWN;
-       c->x86_model = c->x86_mask = 0; /* So far unknown... */
-       c->x86_vendor_id[0] = '\0'; /* Unset */
-       c->x86_model_id[0] = '\0';  /* Unset */
-       c->x86_clflush_size = 64;
-       c->x86_cache_alignment = c->x86_clflush_size;
-       c->x86_max_cores = 1;
-       c->x86_coreid_bits = 0;
-       c->extended_cpuid_level = 0;
-       memset(&c->x86_capability, 0, sizeof c->x86_capability);
-
-       /* Get vendor name */
-       cpuid(0x00000000, (unsigned int *)&c->cpuid_level,
-             (unsigned int *)&c->x86_vendor_id[0],
-             (unsigned int *)&c->x86_vendor_id[8],
-             (unsigned int *)&c->x86_vendor_id[4]);
-
-       get_cpu_vendor(c);
-
-       /* Initialize the standard set of capabilities */
-       /* Note that the vendor-specific code below might override */
-
-       /* Intel-defined flags: level 0x00000001 */
-       if (c->cpuid_level >= 0x00000001) {
-               __u32 misc;
-               cpuid(0x00000001, &tfms, &misc, &c->x86_capability[4],
-                     &c->x86_capability[0]);
-               c->x86 = (tfms >> 8) & 0xf;
-               c->x86_model = (tfms >> 4) & 0xf;
-               c->x86_mask = tfms & 0xf;
-               if (c->x86 == 0xf)
-                       c->x86 += (tfms >> 20) & 0xff;
-               if (c->x86 >= 0x6)
-                       c->x86_model += ((tfms >> 16) & 0xF) << 4;
-               if (test_cpu_cap(c, X86_FEATURE_CLFLSH))
-                       c->x86_clflush_size = ((misc >> 8) & 0xff) * 8;
-       } else {
-               /* Have CPUID level 0 only - unheard of */
-               c->x86 = 4;
-       }
-
-       c->initial_apicid = (cpuid_ebx(1) >> 24) & 0xff;
-#ifdef CONFIG_SMP
-       c->phys_proc_id = c->initial_apicid;
-#endif
-       /* AMD-defined flags: level 0x80000001 */
-       xlvl = cpuid_eax(0x80000000);
-       c->extended_cpuid_level = xlvl;
-       if ((xlvl & 0xffff0000) == 0x80000000) {
-               if (xlvl >= 0x80000001) {
-                       c->x86_capability[1] = cpuid_edx(0x80000001);
-                       c->x86_capability[6] = cpuid_ecx(0x80000001);
-               }
-               if (xlvl >= 0x80000004)
-                       get_model_name(c); /* Default name */
-       }
-
-       /* Transmeta-defined flags: level 0x80860001 */
-       xlvl = cpuid_eax(0x80860000);
-       if ((xlvl & 0xffff0000) == 0x80860000) {
-               /* Don't set x86_cpuid_level here for now to not confuse. */
-               if (xlvl >= 0x80860001)
-                       c->x86_capability[2] = cpuid_edx(0x80860001);
-       }
-
-       c->extended_cpuid_level = cpuid_eax(0x80000000);
-       if (c->extended_cpuid_level >= 0x80000007)
-               c->x86_power = cpuid_edx(0x80000007);
-
-       switch (c->x86_vendor) {
-       case X86_VENDOR_AMD:
-               early_init_amd(c);
-               break;
-       case X86_VENDOR_INTEL:
-               early_init_intel(c);
-               break;
-       case X86_VENDOR_CENTAUR:
-               early_init_centaur(c);
-               break;
-       }
-
-       validate_pat_support(c);
-}
-
-/*
- * This does the hard work of actually picking apart the CPU stuff...
- */
-void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
-{
-       int i;
-
-       early_identify_cpu(c);
-
-       init_scattered_cpuid_features(c);
-
-       c->apicid = phys_pkg_id(0);
-
-       /*
-        * Vendor-specific initialization.  In this section we
-        * canonicalize the feature flags, meaning if there are
-        * features a certain CPU supports which CPUID doesn't
-        * tell us, CPUID claiming incorrect flags, or other bugs,
-        * we handle them here.
-        *
-        * At the end of this section, c->x86_capability better
-        * indicate the features this CPU genuinely supports!
-        */
-       switch (c->x86_vendor) {
-       case X86_VENDOR_AMD:
-               init_amd(c);
-               break;
-
-       case X86_VENDOR_INTEL:
-               init_intel(c);
-               break;
-
-       case X86_VENDOR_CENTAUR:
-               init_centaur(c);
-               break;
-
-       case X86_VENDOR_UNKNOWN:
-       default:
-               display_cacheinfo(c);
-               break;
-       }
-
-       detect_ht(c);
-
-       /*
-        * On SMP, boot_cpu_data holds the common feature set between
-        * all CPUs; so make sure that we indicate which features are
-        * common between the CPUs.  The first time this routine gets
-        * executed, c == &boot_cpu_data.
-        */
-       if (c != &boot_cpu_data) {
-               /* AND the already accumulated flags with these */
-               for (i = 0; i < NCAPINTS; i++)
-                       boot_cpu_data.x86_capability[i] &= c->x86_capability[i];
-       }
-
-       /* Clear all flags overriden by options */
-       for (i = 0; i < NCAPINTS; i++)
-               c->x86_capability[i] &= ~cleared_cpu_caps[i];
-
-#ifdef CONFIG_X86_MCE
-       mcheck_init(c);
-#endif
-       select_idle_routine(c);
-
-#ifdef CONFIG_NUMA
-       numa_add_cpu(smp_processor_id());
-#endif
-
-}
-
-void __cpuinit identify_boot_cpu(void)
-{
-       identify_cpu(&boot_cpu_data);
-}
-
-void __cpuinit identify_secondary_cpu(struct cpuinfo_x86 *c)
-{
-       BUG_ON(c == &boot_cpu_data);
-       identify_cpu(c);
-       mtrr_ap_init();
-}
-
-static __init int setup_noclflush(char *arg)
-{
-       setup_clear_cpu_cap(X86_FEATURE_CLFLSH);
-       return 1;
-}
-__setup("noclflush", setup_noclflush);
-
-void __cpuinit print_cpu_info(struct cpuinfo_x86 *c)
-{
-       if (c->x86_model_id[0])
-               printk(KERN_CONT "%s", c->x86_model_id);
-
-       if (c->x86_mask || c->cpuid_level >= 0)
-               printk(KERN_CONT " stepping %02x\n", c->x86_mask);
-       else
-               printk(KERN_CONT "\n");
-}
-
-static __init int setup_disablecpuid(char *arg)
-{
-       int bit;
-       if (get_option(&arg, &bit) && bit < NCAPINTS*32)
-               setup_clear_cpu_cap(bit);
-       else
-               return 0;
-       return 1;
-}
-__setup("clearcpuid=", setup_disablecpuid);
Index: head-2008-12-01/arch/x86/kernel/setup_percpu-xen.c
===================================================================
--- /dev/null   1970-01-01 00:00:00.000000000 +0000
+++ head-2008-12-01/arch/x86/kernel/setup_percpu-xen.c  2008-12-01 11:49:07.000000000 +0100
@@ -0,0 +1,385 @@
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/bootmem.h>
+#include <linux/percpu.h>
+#include <linux/kexec.h>
+#include <linux/crash_dump.h>
+#include <asm/smp.h>
+#include <asm/percpu.h>
+#include <asm/sections.h>
+#include <asm/processor.h>
+#include <asm/setup.h>
+#include <asm/topology.h>
+#include <asm/mpspec.h>
+#include <asm/apicdef.h>
+#include <asm/highmem.h>
+
+#ifdef CONFIG_X86_LOCAL_APIC
+unsigned int num_processors;
+unsigned disabled_cpus __cpuinitdata;
+/* Processor that is doing the boot up */
+unsigned int boot_cpu_physical_apicid = -1U;
+unsigned int max_physical_apicid;
+EXPORT_SYMBOL(boot_cpu_physical_apicid);
+
+/* Bitmask of physically existing CPUs */
+physid_mask_t phys_cpu_present_map;
+#endif
+
+/* map cpu index to physical APIC ID */
+#ifndef CONFIG_XEN
+DEFINE_EARLY_PER_CPU(u16, x86_cpu_to_apicid, BAD_APICID);
+DEFINE_EARLY_PER_CPU(u16, x86_bios_cpu_apicid, BAD_APICID);
+EXPORT_EARLY_PER_CPU_SYMBOL(x86_cpu_to_apicid);
+EXPORT_EARLY_PER_CPU_SYMBOL(x86_bios_cpu_apicid);
+#else
+DEFINE_PER_CPU(u16, x86_cpu_to_apicid) = BAD_APICID;
+EXPORT_PER_CPU_SYMBOL(x86_cpu_to_apicid);
+#endif
+
+#if defined(CONFIG_NUMA) && defined(CONFIG_X86_64)
+#define        X86_64_NUMA     1
+
+/* map cpu index to node index */
+DEFINE_EARLY_PER_CPU(int, x86_cpu_to_node_map, NUMA_NO_NODE);
+EXPORT_EARLY_PER_CPU_SYMBOL(x86_cpu_to_node_map);
+
+/* which logical CPUs are on which nodes */
+cpumask_t *node_to_cpumask_map;
+EXPORT_SYMBOL(node_to_cpumask_map);
+
+/* setup node_to_cpumask_map */
+static void __init setup_node_to_cpumask_map(void);
+
+#else
+static inline void setup_node_to_cpumask_map(void) { }
+#endif
+
+#if defined(CONFIG_HAVE_SETUP_PER_CPU_AREA) && defined(CONFIG_X86_SMP)
+/*
+ * Copy data used in early init routines from the initial arrays to the
+ * per cpu data areas.  These arrays then become expendable and the
+ * *_early_ptr's are zeroed indicating that the static arrays are gone.
+ */
+static void __init setup_per_cpu_maps(void)
+{
+#ifndef CONFIG_XEN
+       int cpu;
+
+       for_each_possible_cpu(cpu) {
+               per_cpu(x86_cpu_to_apicid, cpu) =
+                               early_per_cpu_map(x86_cpu_to_apicid, cpu);
+               per_cpu(x86_bios_cpu_apicid, cpu) =
+                               early_per_cpu_map(x86_bios_cpu_apicid, cpu);
+#ifdef X86_64_NUMA
+               per_cpu(x86_cpu_to_node_map, cpu) =
+                               early_per_cpu_map(x86_cpu_to_node_map, cpu);
+#endif
+       }
+
+       /* indicate the early static arrays will soon be gone */
+       early_per_cpu_ptr(x86_cpu_to_apicid) = NULL;
+       early_per_cpu_ptr(x86_bios_cpu_apicid) = NULL;
+#ifdef X86_64_NUMA
+       early_per_cpu_ptr(x86_cpu_to_node_map) = NULL;
+#endif
+#endif
+}
+
+#ifdef CONFIG_X86_32
+/*
+ * Great future not-so-futuristic plan: make i386 and x86_64 do it
+ * the same way
+ */
+unsigned long __per_cpu_offset[NR_CPUS] __read_mostly;
+EXPORT_SYMBOL(__per_cpu_offset);
+static inline void setup_cpu_pda_map(void) { }
+
+#elif !defined(CONFIG_SMP)
+static inline void setup_cpu_pda_map(void) { }
+
+#else /* CONFIG_SMP && CONFIG_X86_64 */
+
+/*
+ * Allocate cpu_pda pointer table and array via alloc_bootmem.
+ */
+static void __init setup_cpu_pda_map(void)
+{
+       char *pda;
+       struct x8664_pda **new_cpu_pda;
+       unsigned long size;
+       int cpu;
+
+       size = roundup(sizeof(struct x8664_pda), cache_line_size());
+
+       /* allocate cpu_pda array and pointer table */
+       {
+               unsigned long tsize = nr_cpu_ids * sizeof(void *);
+               unsigned long asize = size * (nr_cpu_ids - 1);
+
+               tsize = roundup(tsize, cache_line_size());
+               new_cpu_pda = alloc_bootmem(tsize + asize);
+               pda = (char *)new_cpu_pda + tsize;
+       }
+
+       /* initialize pointer table to static pda's */
+       for_each_possible_cpu(cpu) {
+               if (cpu == 0) {
+                       /* leave boot cpu pda in place */
+                       new_cpu_pda[0] = cpu_pda(0);
+                       continue;
+               }
+               new_cpu_pda[cpu] = (struct x8664_pda *)pda;
+               new_cpu_pda[cpu]->in_bootmem = 1;
+               pda += size;
+       }
+
+       /* point to new pointer table */
+       _cpu_pda = new_cpu_pda;
+}
+#endif
+
+/*
+ * Great future plan:
+ * Declare PDA itself and support (irqstack,tss,pgd) as per cpu data.
+ * Always point %gs to its beginning
+ */
+void __init setup_per_cpu_areas(void)
+{
+       ssize_t size = PERCPU_ENOUGH_ROOM;
+       char *ptr;
+       int cpu;
+
+       /* Setup cpu_pda map */
+       setup_cpu_pda_map();
+
+       /* Copy section for each CPU (we discard the original) */
+       size = PERCPU_ENOUGH_ROOM;
+       printk(KERN_INFO "PERCPU: Allocating %zd bytes of per cpu data\n",
+                         size);
+
+       for_each_possible_cpu(cpu) {
+#ifndef CONFIG_NEED_MULTIPLE_NODES
+               ptr = alloc_bootmem_pages(size);
+#else
+               int node = early_cpu_to_node(cpu);
+               if (!node_online(node) || !NODE_DATA(node)) {
+                       ptr = alloc_bootmem_pages(size);
+                       printk(KERN_INFO
+                              "cpu %d has no node %d or node-local memory\n",
+                               cpu, node);
+               }
+               else
+                       ptr = alloc_bootmem_pages_node(NODE_DATA(node), size);
+#endif
+               per_cpu_offset(cpu) = ptr - __per_cpu_start;
+               memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start);
+
+       }
+
+       printk(KERN_DEBUG "NR_CPUS: %d, nr_cpu_ids: %d, nr_node_ids %d\n",
+               NR_CPUS, nr_cpu_ids, nr_node_ids);
+
+       /* Setup percpu data maps */
+       setup_per_cpu_maps();
+
+       /* Setup node to cpumask map */
+       setup_node_to_cpumask_map();
+}
+
+#endif
+
+#ifdef X86_64_NUMA
+
+/*
+ * Allocate node_to_cpumask_map based on number of available nodes
+ * Requires node_possible_map to be valid.
+ *
+ * Note: node_to_cpumask() is not valid until after this is done.
+ */
+static void __init setup_node_to_cpumask_map(void)
+{
+       unsigned int node, num = 0;
+       cpumask_t *map;
+
+       /* setup nr_node_ids if not done yet */
+       if (nr_node_ids == MAX_NUMNODES) {
+               for_each_node_mask(node, node_possible_map)
+                       num = node;
+               nr_node_ids = num + 1;
+       }
+
+       /* allocate the map */
+       map = alloc_bootmem_low(nr_node_ids * sizeof(cpumask_t));
+
+       pr_debug(KERN_DEBUG "Node to cpumask map at %p for %d nodes\n",
+                map, nr_node_ids);
+
+       /* node_to_cpumask() will now work */
+       node_to_cpumask_map = map;
+}
+
+void __cpuinit numa_set_node(int cpu, int node)
+{
+       int *cpu_to_node_map = early_per_cpu_ptr(x86_cpu_to_node_map);
+
+       if (cpu_pda(cpu) && node != NUMA_NO_NODE)
+               cpu_pda(cpu)->nodenumber = node;
+
+       if (cpu_to_node_map)
+               cpu_to_node_map[cpu] = node;
+
+       else if (per_cpu_offset(cpu))
+               per_cpu(x86_cpu_to_node_map, cpu) = node;
+
+       else
+               pr_debug("Setting node for non-present cpu %d\n", cpu);
+}
+
+void __cpuinit numa_clear_node(int cpu)
+{
+       numa_set_node(cpu, NUMA_NO_NODE);
+}
+
+#ifndef CONFIG_DEBUG_PER_CPU_MAPS
+
+void __cpuinit numa_add_cpu(int cpu)
+{
+       cpu_set(cpu, node_to_cpumask_map[early_cpu_to_node(cpu)]);
+}
+
+void __cpuinit numa_remove_cpu(int cpu)
+{
+       cpu_clear(cpu, node_to_cpumask_map[cpu_to_node(cpu)]);
+}
+
+#else /* CONFIG_DEBUG_PER_CPU_MAPS */
+
+/*
+ * --------- debug versions of the numa functions ---------
+ */
+static void __cpuinit numa_set_cpumask(int cpu, int enable)
+{
+       int node = cpu_to_node(cpu);
+       cpumask_t *mask;
+       char buf[64];
+
+       if (node_to_cpumask_map == NULL) {
+               printk(KERN_ERR "node_to_cpumask_map NULL\n");
+               dump_stack();
+               return;
+       }
+
+       mask = &node_to_cpumask_map[node];
+       if (enable)
+               cpu_set(cpu, *mask);
+       else
+               cpu_clear(cpu, *mask);
+
+       cpulist_scnprintf(buf, sizeof(buf), *mask);
+       printk(KERN_DEBUG "%s cpu %d node %d: mask now %s\n",
+               enable? "numa_add_cpu":"numa_remove_cpu", cpu, node, buf);
+ }
+
+void __cpuinit numa_add_cpu(int cpu)
+{
+       numa_set_cpumask(cpu, 1);
+}
+
+void __cpuinit numa_remove_cpu(int cpu)
+{
+       numa_set_cpumask(cpu, 0);
+}
+
+int cpu_to_node(int cpu)
+{
+       if (early_per_cpu_ptr(x86_cpu_to_node_map)) {
+               printk(KERN_WARNING
+                       "cpu_to_node(%d): usage too early!\n", cpu);
+               dump_stack();
+               return early_per_cpu_ptr(x86_cpu_to_node_map)[cpu];
+       }
+       return per_cpu(x86_cpu_to_node_map, cpu);
+}
+EXPORT_SYMBOL(cpu_to_node);
+
+/*
+ * Same function as cpu_to_node() but used if called before the
+ * per_cpu areas are setup.
+ */
+int early_cpu_to_node(int cpu)
+{
+       if (early_per_cpu_ptr(x86_cpu_to_node_map))
+               return early_per_cpu_ptr(x86_cpu_to_node_map)[cpu];
+
+       if (!per_cpu_offset(cpu)) {
+               printk(KERN_WARNING
+                       "early_cpu_to_node(%d): no per_cpu area!\n", cpu);
+               dump_stack();
+               return NUMA_NO_NODE;
+       }
+       return per_cpu(x86_cpu_to_node_map, cpu);
+}
+
+
+/* empty cpumask */
+static const cpumask_t cpu_mask_none;
+
+/*
+ * Returns a pointer to the bitmask of CPUs on Node 'node'.
+ */
+const cpumask_t *_node_to_cpumask_ptr(int node)
+{
+       if (node_to_cpumask_map == NULL) {
+               printk(KERN_WARNING
+                       "_node_to_cpumask_ptr(%d): no node_to_cpumask_map!\n",
+                       node);
+               dump_stack();
+               return (const cpumask_t *)&cpu_online_map;
+       }
+       if (node >= nr_node_ids) {
+               printk(KERN_WARNING
+                       "_node_to_cpumask_ptr(%d): node > nr_node_ids(%d)\n",
+                       node, nr_node_ids);
+               dump_stack();
+               return &cpu_mask_none;
+       }
+       return &node_to_cpumask_map[node];
+}
+EXPORT_SYMBOL(_node_to_cpumask_ptr);
+
+/*
+ * Returns a bitmask of CPUs on Node 'node'.
+ *
+ * Side note: this function creates the returned cpumask on the stack
+ * so with a high NR_CPUS count, excessive stack space is used.  The
+ * node_to_cpumask_ptr function should be used whenever possible.
+ */
+cpumask_t node_to_cpumask(int node)
+{
+       if (node_to_cpumask_map == NULL) {
+               printk(KERN_WARNING
+                       "node_to_cpumask(%d): no node_to_cpumask_map!\n", node);
+               dump_stack();
+               return cpu_online_map;
+       }
+       if (node >= nr_node_ids) {
+               printk(KERN_WARNING
+                       "node_to_cpumask(%d): node > nr_node_ids(%d)\n",
+                       node, nr_node_ids);
+               dump_stack();
+               return cpu_mask_none;
+       }
+       return node_to_cpumask_map[node];
+}
+EXPORT_SYMBOL(node_to_cpumask);
+
+/*
+ * --------- end of debug versions of the numa functions ---------
+ */
+
+#endif /* CONFIG_DEBUG_PER_CPU_MAPS */
+
+#endif /* X86_64_NUMA */
+
Index: head-2008-12-01/arch/x86/kernel/smp-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/smp-xen.c      2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/smp-xen.c   2008-12-01 11:49:07.000000000 +0100
@@ -121,132 +121,14 @@ void xen_smp_send_reschedule(int cpu)
        send_IPI_mask(cpumask_of_cpu(cpu), RESCHEDULE_VECTOR);
 }
 
-/*
- * Structure and data for smp_call_function(). This is designed to minimise
- * static memory requirements. It also looks cleaner.
- */
-static DEFINE_SPINLOCK(call_lock);
-
-struct call_data_struct {
-       void (*func) (void *info);
-       void *info;
-       atomic_t started;
-       atomic_t finished;
-       int wait;
-};
-
-void lock_ipi_call_lock(void)
+void xen_send_call_func_single_ipi(int cpu)
 {
-       spin_lock_irq(&call_lock);
+       send_IPI_mask(cpumask_of_cpu(cpu), CALL_FUNC_SINGLE_VECTOR);
 }
 
-void unlock_ipi_call_lock(void)
+void xen_send_call_func_ipi(cpumask_t mask)
 {
-       spin_unlock_irq(&call_lock);
-}
-
-static struct call_data_struct *call_data;
-
-static void __smp_call_function(void (*func) (void *info), void *info,
-                               int nonatomic, int wait)
-{
-       struct call_data_struct data;
-       int cpus = num_online_cpus() - 1;
-
-       if (!cpus)
-               return;
-
-       data.func = func;
-       data.info = info;
-       atomic_set(&data.started, 0);
-       data.wait = wait;
-       if (wait)
-               atomic_set(&data.finished, 0);
-
-       call_data = &data;
-       mb();
-
-       /* Send a message to all other CPUs and wait for them to respond */
-       send_IPI_allbutself(CALL_FUNCTION_VECTOR);
-
-       /* Wait for response */
-       while (atomic_read(&data.started) != cpus)
-               cpu_relax();
-
-       if (wait)
-               while (atomic_read(&data.finished) != cpus)
-                       cpu_relax();
-}
-
-
-/**
- * smp_call_function_mask(): Run a function on a set of other CPUs.
- * @mask: The set of cpus to run on.  Must not include the current cpu.
- * @func: The function to run. This must be fast and non-blocking.
- * @info: An arbitrary pointer to pass to the function.
- * @wait: If true, wait (atomically) until function has completed on other CPUs.
- *
-  * Returns 0 on success, else a negative status code.
- *
- * If @wait is true, then returns once @func has returned; otherwise
- * it returns just before the target cpu calls @func.
- *
- * You must not call this function with disabled interrupts or from a
- * hardware interrupt handler or from a bottom half handler.
- */
-int
-xen_smp_call_function_mask(cpumask_t mask,
-                             void (*func)(void *), void *info,
-                             int wait)
-{
-       struct call_data_struct data;
-       cpumask_t allbutself;
-       int cpus;
-
-       /* Can deadlock when called with interrupts disabled */
-       WARN_ON(irqs_disabled());
-
-       /* Holding any lock stops cpus from going down. */
-       spin_lock(&call_lock);
-
-       allbutself = cpu_online_map;
-       cpu_clear(smp_processor_id(), allbutself);
-
-       cpus_and(mask, mask, allbutself);
-       cpus = cpus_weight(mask);
-
-       if (!cpus) {
-               spin_unlock(&call_lock);
-               return 0;
-       }
-
-       data.func = func;
-       data.info = info;
-       atomic_set(&data.started, 0);
-       data.wait = wait;
-       if (wait)
-               atomic_set(&data.finished, 0);
-
-       call_data = &data;
-       wmb();
-
-       /* Send a message to other CPUs */
-       if (cpus_equal(mask, allbutself) &&
-           cpus_equal(cpu_online_map, cpu_callout_map))
-               send_IPI_allbutself(CALL_FUNCTION_VECTOR);
-       else
-               send_IPI_mask(mask, CALL_FUNCTION_VECTOR);
-
-       /* Wait for response */
-       while (atomic_read(&data.started) != cpus)
-               cpu_relax();
-
-       if (wait)
-               while (atomic_read(&data.finished) != cpus)
-                       cpu_relax();
-       spin_unlock(&call_lock);
-
-       return 0;
+       send_IPI_mask(mask, CALL_FUNCTION_VECTOR);
 }
 
 static void stop_this_cpu(void *dummy)
@@ -268,15 +150,10 @@ static void stop_this_cpu(void *dummy)
 
 void xen_smp_send_stop(void)
 {
-       int nolock;
        unsigned long flags;
 
-       /* Don't deadlock on the call lock in panic */
-       nolock = !spin_trylock(&call_lock);
+       smp_call_function(stop_this_cpu, NULL, 0);
        local_irq_save(flags);
-       __smp_call_function(stop_this_cpu, NULL, 0, 0);
-       if (!nolock)
-               spin_unlock(&call_lock);
        disable_all_local_evtchn();
        local_irq_restore(flags);
 }
@@ -298,21 +175,8 @@ irqreturn_t smp_reschedule_interrupt(int
 
 irqreturn_t smp_call_function_interrupt(int irq, void *dev_id)
 {
-       void (*func) (void *info) = call_data->func;
-       void *info = call_data->info;
-       int wait = call_data->wait;
-
-       /*
-        * Notify initiating CPU that I've grabbed the data and am
-        * about to execute the function
-        */
-       mb();
-       atomic_inc(&call_data->started);
-       /*
-        * At this point the info structure may be out of scope unless wait==1
-        */
        irq_enter();
-       (*func)(info);
+       generic_smp_call_function_interrupt();
 #ifdef CONFIG_X86_32
        __get_cpu_var(irq_stat).irq_call_count++;
 #else
@@ -320,10 +184,19 @@ irqreturn_t smp_call_function_interrupt(
 #endif
        irq_exit();
 
-       if (wait) {
-               mb();
-               atomic_inc(&call_data->finished);
-       }
+       return IRQ_HANDLED;
+}
+
+irqreturn_t smp_call_function_single_interrupt(int irq, void *dev_id)
+{
+       irq_enter();
+       generic_smp_call_function_single_interrupt();
+#ifdef CONFIG_X86_32
+       __get_cpu_var(irq_stat).irq_call_count++;
+#else
+       add_pda(irq_call_count, 1);
+#endif
+       irq_exit();
 
        return IRQ_HANDLED;
 }
Index: head-2008-12-01/arch/x86/kernel/time_32-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/time_32-xen.c  2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/time_32-xen.c       2008-12-01 11:58:30.000000000 +0100
@@ -470,7 +470,7 @@ irqreturn_t timer_interrupt(int irq, voi
 
        /* Keep nmi watchdog up to date */
 #ifdef __i386__
-       per_cpu(irq_stat, smp_processor_id()).irq0_irqs++;
+       x86_add_percpu(irq_stat.irq0_irqs, 1);
 #else
        add_pda(irq0_irqs, 1);
 #endif
@@ -748,9 +748,7 @@ void __init time_init(void)
 
        update_wallclock();
 
-#ifndef CONFIG_X86_64
        use_tsc_delay();
-#endif
 
        /* Cannot request_irq() until kmem is initialised. */
        late_time_init = setup_cpu0_timer_irq;
@@ -807,7 +805,8 @@ static void stop_hz_timer(void)
 
        /* Leave ourselves in tick mode if rcu or softirq or timer pending. */
        if (rcu_needs_cpu(cpu) || local_softirq_pending() ||
-           (j = next_timer_interrupt(), time_before_eq(j, jiffies))) {
+           (j = get_next_timer_interrupt(jiffies),
+            time_before_eq(j, jiffies))) {
                cpu_clear(cpu, nohz_cpu_mask);
                j = jiffies + 1;
        }
Index: head-2008-12-01/arch/x86/kernel/traps_32-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/traps_32-xen.c 2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/traps_32-xen.c      2008-12-01 11:49:07.000000000 +0100
@@ -1,5 +1,6 @@
 /*
  *  Copyright (C) 1991, 1992  Linus Torvalds
+ *  Copyright (C) 2000, 2001, 2002 Andi Kleen, SuSE Labs
  *
  *  Pentium III FXSR, SSE support
  *     Gareth Hughes <gareth@valinux.com>, May 2000
@@ -57,11 +58,10 @@
 #include <asm/nmi.h>
 #include <asm/smp.h>
 #include <asm/io.h>
+#include <asm/traps.h>
 
 #include "mach_traps.h"
 
-int panic_on_unrecovered_nmi;
-
 #ifndef CONFIG_XEN
 DECLARE_BITMAP(used_vectors, NR_VECTORS);
 EXPORT_SYMBOL_GPL(used_vectors);
@@ -82,43 +82,22 @@ gate_desc idt_table[256]
        __attribute__((__section__(".data.idt"))) = { { { { 0, 0 } } }, };
 #endif
 
-asmlinkage void divide_error(void);
-asmlinkage void debug(void);
-asmlinkage void nmi(void);
-asmlinkage void int3(void);
-asmlinkage void overflow(void);
-asmlinkage void bounds(void);
-asmlinkage void invalid_op(void);
-asmlinkage void device_not_available(void);
-asmlinkage void coprocessor_segment_overrun(void);
-asmlinkage void invalid_TSS(void);
-asmlinkage void segment_not_present(void);
-asmlinkage void stack_segment(void);
-asmlinkage void general_protection(void);
-asmlinkage void page_fault(void);
-asmlinkage void coprocessor_error(void);
-asmlinkage void simd_coprocessor_error(void);
-asmlinkage void alignment_check(void);
-#ifndef CONFIG_XEN
-asmlinkage void spurious_interrupt_bug(void);
-#else
-asmlinkage void fixup_4gb_segment(void);
-#endif
-asmlinkage void machine_check(void);
-
+int panic_on_unrecovered_nmi;
 int kstack_depth_to_print = 24;
 static unsigned int code_bytes = 64;
+static int ignore_nmis;
+static int die_counter;
 
 void printk_address(unsigned long address, int reliable)
 {
 #ifdef CONFIG_KALLSYMS
-       char namebuf[KSYM_NAME_LEN];
        unsigned long offset = 0;
        unsigned long symsize;
        const char *symname;
-       char reliab[4] = "";
-       char *delim = ":";
        char *modname;
+       char *delim = ":";
+       char namebuf[KSYM_NAME_LEN];
+       char reliab[4] = "";
 
        symname = kallsyms_lookup(address, &symsize, &offset,
                                        &modname, namebuf);
@@ -138,22 +117,23 @@ void printk_address(unsigned long addres
 #endif
 }
 
-static inline int valid_stack_ptr(struct thread_info *tinfo, void *p, unsigned size)
+static inline int valid_stack_ptr(struct thread_info *tinfo,
+                       void *p, unsigned int size)
 {
-       return  p > (void *)tinfo &&
-               p <= (void *)tinfo + THREAD_SIZE - size;
+       void *t = tinfo;
+       return  p > t && p <= t + THREAD_SIZE - size;
 }
 
 /* The form of the top of the frame on the stack */
 struct stack_frame {
-       struct stack_frame      *next_frame;
-       unsigned long           return_address;
+       struct stack_frame *next_frame;
+       unsigned long return_address;
 };
 
 static inline unsigned long
 print_context_stack(struct thread_info *tinfo,
-                   unsigned long *stack, unsigned long bp,
-                   const struct stacktrace_ops *ops, void *data)
+               unsigned long *stack, unsigned long bp,
+               const struct stacktrace_ops *ops, void *data)
 {
        struct stack_frame *frame = (struct stack_frame *)bp;
 
@@ -175,8 +155,6 @@ print_context_stack(struct thread_info *
        return bp;
 }
 
-#define MSG(msg)               ops->warning(data, msg)
-
 void dump_trace(struct task_struct *task, struct pt_regs *regs,
                unsigned long *stack, unsigned long bp,
                const struct stacktrace_ops *ops, void *data)
@@ -186,7 +164,6 @@ void dump_trace(struct task_struct *task
 
        if (!stack) {
                unsigned long dummy;
-
                stack = &dummy;
                if (task != current)
                        stack = (unsigned long *)task->thread.sp;
@@ -204,7 +181,7 @@ void dump_trace(struct task_struct *task
        }
 #endif
 
-       while (1) {
+       for (;;) {
                struct thread_info *context;
 
                context = (struct thread_info *)
@@ -256,15 +233,15 @@ static void print_trace_address(void *da
 }
 
 static const struct stacktrace_ops print_trace_ops = {
-       .warning                = print_trace_warning,
-       .warning_symbol         = print_trace_warning_symbol,
-       .stack                  = print_trace_stack,
-       .address                = print_trace_address,
+       .warning = print_trace_warning,
+       .warning_symbol = print_trace_warning_symbol,
+       .stack = print_trace_stack,
+       .address = print_trace_address,
 };
 
 static void
 show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
-                  unsigned long *stack, unsigned long bp, char *log_lvl)
+               unsigned long *stack, unsigned long bp, char *log_lvl)
 {
        dump_trace(task, regs, stack, bp, &print_trace_ops, log_lvl);
        printk("%s =======================\n", log_lvl);
@@ -359,15 +336,14 @@ void show_registers(struct pt_regs *regs
                printk(KERN_EMERG "Code: ");
 
                ip = (u8 *)regs->ip - code_prologue;
-               if (ip < (u8 *)PAGE_OFFSET ||
-                       probe_kernel_address(ip, c)) {
+               if (ip < (u8 *)PAGE_OFFSET || probe_kernel_address(ip, c)) {
                        /* try starting at EIP */
                        ip = (u8 *)regs->ip;
                        code_len = code_len - code_prologue + 1;
                }
                for (i = 0; i < code_len; i++, ip++) {
                        if (ip < (u8 *)PAGE_OFFSET ||
-                               probe_kernel_address(ip, c)) {
+                                       probe_kernel_address(ip, c)) {
                                printk(" Bad EIP value.");
                                break;
                        }
@@ -392,7 +368,53 @@ int is_valid_bugaddr(unsigned long ip)
        return ud2 == 0x0b0f;
 }
 
-static int die_counter;
+static raw_spinlock_t die_lock = __RAW_SPIN_LOCK_UNLOCKED;
+static int die_owner = -1;
+static unsigned int die_nest_count;
+
+unsigned __kprobes long oops_begin(void)
+{
+       unsigned long flags;
+
+       oops_enter();
+
+       if (die_owner != raw_smp_processor_id()) {
+               console_verbose();
+               raw_local_irq_save(flags);
+               __raw_spin_lock(&die_lock);
+               die_owner = smp_processor_id();
+               die_nest_count = 0;
+               bust_spinlocks(1);
+       } else {
+               raw_local_irq_save(flags);
+       }
+       die_nest_count++;
+       return flags;
+}
+
+void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr)
+{
+       bust_spinlocks(0);
+       die_owner = -1;
+       add_taint(TAINT_DIE);
+       __raw_spin_unlock(&die_lock);
+       raw_local_irq_restore(flags);
+
+       if (!regs)
+               return;
+
+       if (kexec_should_crash(current))
+               crash_kexec(regs);
+
+       if (in_interrupt())
+               panic("Fatal exception in interrupt");
+
+       if (panic_on_oops)
+               panic("Fatal exception");
+
+       oops_exit();
+       do_exit(signr);
+}
 
 int __kprobes __die(const char *str, struct pt_regs *regs, long err)
 {
@@ -410,26 +432,22 @@ int __kprobes __die(const char *str, str
        printk("DEBUG_PAGEALLOC");
 #endif
        printk("\n");
-
        if (notify_die(DIE_OOPS, str, regs, err,
-                       current->thread.trap_no, SIGSEGV) != NOTIFY_STOP) {
+                       current->thread.trap_no, SIGSEGV) == NOTIFY_STOP)
+               return 1;
 
-               show_registers(regs);
-               /* Executive summary in case the oops scrolled away */
-               sp = (unsigned long) (&regs->sp);
-               savesegment(ss, ss);
-               if (user_mode(regs)) {
-                       sp = regs->sp;
-                       ss = regs->ss & 0xffff;
-               }
-               printk(KERN_EMERG "EIP: [<%08lx>] ", regs->ip);
-               print_symbol("%s", regs->ip);
-               printk(" SS:ESP %04x:%08lx\n", ss, sp);
-
-               return 0;
-       }
-
-       return 1;
+       show_registers(regs);
+       /* Executive summary in case the oops scrolled away */
+       sp = (unsigned long) (&regs->sp);
+       savesegment(ss, ss);
+       if (user_mode(regs)) {
+               sp = regs->sp;
+               ss = regs->ss & 0xffff;
+       }
+       printk(KERN_EMERG "EIP: [<%08lx>] ", regs->ip);
+       print_symbol("%s", regs->ip);
+       printk(" SS:ESP %04x:%08lx\n", ss, sp);
+       return 0;
 }
 
 /*
@@ -438,31 +456,9 @@ int __kprobes __die(const char *str, str
  */
 void die(const char *str, struct pt_regs *regs, long err)
 {
-       static struct {
-               raw_spinlock_t lock;
-               u32 lock_owner;
-               int lock_owner_depth;
-       } die = {
-               .lock =                 __RAW_SPIN_LOCK_UNLOCKED,
-               .lock_owner =           -1,
-               .lock_owner_depth =     0
-       };
-       unsigned long flags;
-
-       oops_enter();
+       unsigned long flags = oops_begin();
 
-       if (die.lock_owner != raw_smp_processor_id()) {
-               console_verbose();
-               raw_local_irq_save(flags);
-               __raw_spin_lock(&die.lock);
-               die.lock_owner = smp_processor_id();
-               die.lock_owner_depth = 0;
-               bust_spinlocks(1);
-       } else {
-               raw_local_irq_save(flags);
-       }
-
-       if (++die.lock_owner_depth < 3) {
+       if (die_nest_count < 3) {
                report_bug(regs->ip, regs);
 
                if (__die(str, regs, err))
@@ -471,26 +467,7 @@ void die(const char *str, struct pt_regs
                printk(KERN_EMERG "Recursive die() failure, output suppressed\n");
        }
 
-       bust_spinlocks(0);
-       die.lock_owner = -1;
-       add_taint(TAINT_DIE);
-       __raw_spin_unlock(&die.lock);
-       raw_local_irq_restore(flags);
-
-       if (!regs)
-               return;
-
-       if (kexec_should_crash(current))
-               crash_kexec(regs);
-
-       if (in_interrupt())
-               panic("Fatal exception in interrupt");
-
-       if (panic_on_oops)
-               panic("Fatal exception");
-
-       oops_exit();
-       do_exit(SIGSEGV);
+       oops_end(flags, regs, SIGSEGV);
 }
 
 static inline void
@@ -554,7 +531,7 @@ void do_##name(struct pt_regs *regs, lon
 {                                                                      \
        trace_hardirqs_fixup();                                         \
        if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr)  \
-                                               == NOTIFY_STOP)         \
+                                                       == NOTIFY_STOP) \
                return;                                                 \
        do_trap(trapnr, signr, str, 0, regs, error_code, NULL);         \
 }
@@ -570,7 +547,7 @@ void do_##name(struct pt_regs *regs, lon
        info.si_code = sicode;                                          \
        info.si_addr = (void __user *)siaddr;                           \
        if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr)  \
-                                               == NOTIFY_STOP)         \
+                                                       == NOTIFY_STOP) \
                return;                                                 \
        do_trap(trapnr, signr, str, 0, regs, error_code, &info);        \
 }
@@ -579,7 +556,7 @@ void do_##name(struct pt_regs *regs, lon
 void do_##name(struct pt_regs *regs, long error_code)                  \
 {                                                                      \
        if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr)  \
-                                               == NOTIFY_STOP)         \
+                                                       == NOTIFY_STOP) \
                return;                                                 \
        do_trap(trapnr, signr, str, 1, regs, error_code, NULL);         \
 }
@@ -594,28 +571,29 @@ void do_##name(struct pt_regs *regs, lon
        info.si_addr = (void __user *)siaddr;                           \
        trace_hardirqs_fixup();                                         \
        if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr)  \
-                                               == NOTIFY_STOP)         \
+                                                       == NOTIFY_STOP) \
                return;                                                 \
        do_trap(trapnr, signr, str, 1, regs, error_code, &info);        \
 }
 
-DO_VM86_ERROR_INFO(0, SIGFPE,  "divide error", divide_error, FPE_INTDIV, regs->ip)
+DO_VM86_ERROR_INFO(0, SIGFPE, "divide error", divide_error, FPE_INTDIV, regs->ip)
 #ifndef CONFIG_KPROBES
 DO_VM86_ERROR(3, SIGTRAP, "int3", int3)
 #endif
 DO_VM86_ERROR(4, SIGSEGV, "overflow", overflow)
 DO_VM86_ERROR(5, SIGSEGV, "bounds", bounds)
-DO_ERROR_INFO(6, SIGILL,  "invalid opcode", invalid_op, ILL_ILLOPN, regs->ip, 0)
-DO_ERROR(9, SIGFPE,  "coprocessor segment overrun", coprocessor_segment_overrun)
+DO_ERROR_INFO(6, SIGILL, "invalid opcode", invalid_op, ILL_ILLOPN, regs->ip, 0)
+DO_ERROR(9, SIGFPE, "coprocessor segment overrun", coprocessor_segment_overrun)
 DO_ERROR(10, SIGSEGV, "invalid TSS", invalid_TSS)
-DO_ERROR(11, SIGBUS,  "segment not present", segment_not_present)
-DO_ERROR(12, SIGBUS,  "stack segment", stack_segment)
+DO_ERROR(11, SIGBUS, "segment not present", segment_not_present)
+DO_ERROR(12, SIGBUS, "stack segment", stack_segment)
 DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0, 0)
 DO_ERROR_INFO(32, SIGILL, "iret exception", iret_error, ILL_BADSTK, 0, 1)
 
-void __kprobes do_general_protection(struct pt_regs * regs,
-                                             long error_code)
+void __kprobes
+do_general_protection(struct pt_regs *regs, long error_code)
 {
+       struct task_struct *tsk;
        struct thread_struct *thread;
 
        thread = &current->thread;
@@ -623,23 +601,24 @@ void __kprobes do_general_protection(str
        if (regs->flags & X86_VM_MASK)
                goto gp_in_vm86;
 
+       tsk = current;
        if (!user_mode(regs))
                goto gp_in_kernel;
 
-       current->thread.error_code = error_code;
-       current->thread.trap_no = 13;
+       tsk->thread.error_code = error_code;
+       tsk->thread.trap_no = 13;
 
-       if (show_unhandled_signals && unhandled_signal(current, SIGSEGV) &&
-           printk_ratelimit()) {
+       if (show_unhandled_signals && unhandled_signal(tsk, SIGSEGV) &&
+                       printk_ratelimit()) {
                printk(KERN_INFO
-                   "%s[%d] general protection ip:%lx sp:%lx error:%lx",
-                   current->comm, task_pid_nr(current),
-                   regs->ip, regs->sp, error_code);
+                       "%s[%d] general protection ip:%lx sp:%lx error:%lx",
+                       tsk->comm, task_pid_nr(tsk),
+                       regs->ip, regs->sp, error_code);
                print_vma_addr(" in ", regs->ip);
                printk("\n");
        }
 
-       force_sig(SIGSEGV, current);
+       force_sig(SIGSEGV, tsk);
        return;
 
 gp_in_vm86:
@@ -648,14 +627,15 @@ gp_in_vm86:
        return;
 
 gp_in_kernel:
-       if (!fixup_exception(regs)) {
-               current->thread.error_code = error_code;
-               current->thread.trap_no = 13;
-               if (notify_die(DIE_GPF, "general protection fault", regs,
+       if (fixup_exception(regs))
+               return;
+
+       tsk->thread.error_code = error_code;
+       tsk->thread.trap_no = 13;
+       if (notify_die(DIE_GPF, "general protection fault", regs,
                                error_code, 13, SIGSEGV) == NOTIFY_STOP)
-                       return;
-               die("general protection fault", regs, error_code);
-       }
+               return;
+       die("general protection fault", regs, error_code);
 }
 
 static notrace __kprobes void
@@ -722,9 +702,9 @@ unknown_nmi_error(unsigned char reason, 
 
 static DEFINE_SPINLOCK(nmi_print_lock);
 
-void notrace __kprobes die_nmi(struct pt_regs *regs, const char *msg)
+void notrace __kprobes die_nmi(char *str, struct pt_regs *regs, int do_panic)
 {
-       if (notify_die(DIE_NMIWATCHDOG, msg, regs, 0, 2, SIGINT) == NOTIFY_STOP)
+       if (notify_die(DIE_NMIWATCHDOG, str, regs, 0, 2, SIGINT) == NOTIFY_STOP)
                return;
 
        spin_lock(&nmi_print_lock);
@@ -733,10 +713,12 @@ void notrace __kprobes die_nmi(struct pt
        * to get a message out:
        */
        bust_spinlocks(1);
-       printk(KERN_EMERG "%s", msg);
+       printk(KERN_EMERG "%s", str);
        printk(" on CPU%d, ip %08lx, registers:\n",
                smp_processor_id(), regs->ip);
        show_registers(regs);
+       if (do_panic)
+               panic("Non maskable interrupt");
        console_silent();
        spin_unlock(&nmi_print_lock);
        bust_spinlocks(0);
@@ -756,14 +738,17 @@ void notrace __kprobes die_nmi(struct pt
 static notrace __kprobes void default_do_nmi(struct pt_regs *regs)
 {
        unsigned char reason = 0;
+       int cpu;
 
-       /* Only the BSP gets external NMIs from the system: */
-       if (!smp_processor_id())
+       cpu = smp_processor_id();
+
+       /* Only the BSP gets external NMIs from the system. */
+       if (!cpu)
                reason = get_nmi_reason();
 
        if (!(reason & 0xc0)) {
                if (notify_die(DIE_NMI_IPI, "nmi_ipi", regs, reason, 2, SIGINT)
-                                                       == NOTIFY_STOP)
+                                                               == NOTIFY_STOP)
                        return;
 #ifdef CONFIG_X86_LOCAL_APIC
                /*
@@ -772,7 +757,7 @@ static notrace __kprobes void default_do
                 */
                if (nmi_watchdog_tick(regs, reason))
                        return;
-               if (!do_nmi_callback(regs, smp_processor_id()))
+               if (!do_nmi_callback(regs, cpu))
                        unknown_nmi_error(reason, regs);
 #else
                unknown_nmi_error(reason, regs);
@@ -782,6 +767,8 @@ static notrace __kprobes void default_do
        }
        if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP)
                return;
+
+       /* AK: following checks seem to be broken on modern chipsets. FIXME */
        if (reason & 0x80)
                mem_parity_error(reason, regs);
        if (reason & 0x40)
@@ -793,8 +780,6 @@ static notrace __kprobes void default_do
        reassert_nmi();
 }
 
-static int ignore_nmis;
-
 notrace __kprobes void do_nmi(struct pt_regs *regs, long error_code)
 {
        int cpu;
@@ -879,7 +864,7 @@ void __kprobes do_debug(struct pt_regs *
        tsk->thread.debugctlmsr = 0;
 
        if (notify_die(DIE_DEBUG, "debug", regs, condition, error_code,
-                                       SIGTRAP) == NOTIFY_STOP)
+                                               SIGTRAP) == NOTIFY_STOP)
                return;
        /* It's safe to allow irq's after DR6 has been saved */
        if (regs->flags & X86_EFLAGS_IF)
@@ -940,9 +925,8 @@ clear_TF_reenable:
 void math_error(void __user *ip)
 {
        struct task_struct *task;
-       unsigned short cwd;
-       unsigned short swd;
        siginfo_t info;
+       unsigned short cwd, swd;
 
        /*
         * Save the info for the exception handler and clear the error.
@@ -961,7 +945,7 @@ void math_error(void __user *ip)
         * C1 reg you need in case of a stack fault, 0x040 is the stack
         * fault bit.  We should only be taking one exception at a time,
         * so if this combination doesn't produce any single exception,
-        * then we have a bad program that isn't syncronizing its FPU usage
+        * then we have a bad program that isn't synchronizing its FPU usage
         * and it will suffer the consequences since we won't be able to
         * fully reproduce the context of the exception
         */
@@ -970,7 +954,7 @@ void math_error(void __user *ip)
        switch (swd & ~cwd & 0x3f) {
        case 0x000: /* No unmasked exception */
                return;
-       default:    /* Multiple exceptions */
+       default: /* Multiple exceptions */
                break;
        case 0x001: /* Invalid Op */
                /*
@@ -1006,8 +990,8 @@ void do_coprocessor_error(struct pt_regs
 static void simd_math_error(void __user *ip)
 {
        struct task_struct *task;
-       unsigned short mxcsr;
        siginfo_t info;
+       unsigned short mxcsr;
 
        /*
         * Save the info for the exception handler and clear the error.
@@ -1084,7 +1068,7 @@ void do_spurious_interrupt_bug(struct pt
 
 unsigned long patch_espfix_desc(unsigned long uesp, unsigned long kesp)
 {
-       struct desc_struct *gdt = __get_cpu_var(gdt_page).gdt;
+       struct desc_struct *gdt = get_cpu_gdt_table(smp_processor_id());
        unsigned long base = (kesp - uesp) & -THREAD_SIZE;
        unsigned long new_kesp = kesp - base;
        unsigned long lim_pages = (new_kesp | (THREAD_SIZE - 1)) >> PAGE_SHIFT;
Index: head-2008-12-01/arch/x86/kernel/traps_64-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/traps_64-xen.c 2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/traps_64-xen.c      2008-12-01 11:49:07.000000000 +0100
@@ -10,73 +10,56 @@
  * 'Traps.c' handles hardware traps and faults after we have saved some
  * state in 'entry.S'.
  */
-#include <linux/sched.h>
+#include <linux/moduleparam.h>
+#include <linux/interrupt.h>
+#include <linux/kallsyms.h>
+#include <linux/spinlock.h>
+#include <linux/kprobes.h>
+#include <linux/uaccess.h>
+#include <linux/utsname.h>
+#include <linux/kdebug.h>
 #include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/ptrace.h>
 #include <linux/string.h>
+#include <linux/unwind.h>
+#include <linux/delay.h>
 #include <linux/errno.h>
-#include <linux/ptrace.h>
+#include <linux/kexec.h>
+#include <linux/sched.h>
 #include <linux/timer.h>
-#include <linux/mm.h>
 #include <linux/init.h>
-#include <linux/delay.h>
-#include <linux/spinlock.h>
-#include <linux/interrupt.h>
-#include <linux/kallsyms.h>
-#include <linux/module.h>
-#include <linux/moduleparam.h>
-#include <linux/nmi.h>
-#include <linux/kprobes.h>
-#include <linux/kexec.h>
-#include <linux/unwind.h>
-#include <linux/uaccess.h>
 #include <linux/bug.h>
-#include <linux/kdebug.h>
-#include <linux/utsname.h>
-
-#include <mach_traps.h>
+#include <linux/nmi.h>
+#include <linux/mm.h>
 
 #if defined(CONFIG_EDAC)
 #include <linux/edac.h>
 #endif
 
-#include <asm/system.h>
-#include <asm/io.h>
-#include <asm/atomic.h>
+#include <asm/stacktrace.h>
+#include <asm/processor.h>
 #include <asm/debugreg.h>
+#include <asm/atomic.h>
+#include <asm/system.h>
+#include <asm/unwind.h>
 #include <asm/desc.h>
 #include <asm/i387.h>
-#include <asm/processor.h>
-#include <asm/unwind.h>
+#include <asm/nmi.h>
 #include <asm/smp.h>
+#include <asm/io.h>
 #include <asm/pgalloc.h>
-#include <asm/pda.h>
 #include <asm/proto.h>
-#include <asm/nmi.h>
-#include <asm/stacktrace.h>
+#include <asm/pda.h>
+#include <asm/traps.h>
 
-asmlinkage void divide_error(void);
-asmlinkage void debug(void);
-asmlinkage void nmi(void);
-asmlinkage void int3(void);
-asmlinkage void overflow(void);
-asmlinkage void bounds(void);
-asmlinkage void invalid_op(void);
-asmlinkage void device_not_available(void);
-asmlinkage void double_fault(void);
-asmlinkage void coprocessor_segment_overrun(void);
-asmlinkage void invalid_TSS(void);
-asmlinkage void segment_not_present(void);
-asmlinkage void stack_segment(void);
-asmlinkage void general_protection(void);
-asmlinkage void page_fault(void);
-asmlinkage void coprocessor_error(void);
-asmlinkage void simd_coprocessor_error(void);
-asmlinkage void reserved(void);
-asmlinkage void alignment_check(void);
-asmlinkage void machine_check(void);
-asmlinkage void spurious_interrupt_bug(void);
+#include <mach_traps.h>
 
+int panic_on_unrecovered_nmi;
+int kstack_depth_to_print = 12;
 static unsigned int code_bytes = 64;
+static int ignore_nmis;
+static int die_counter;
 
 static inline void conditional_sti(struct pt_regs *regs)
 {
@@ -100,34 +83,9 @@ static inline void preempt_conditional_c
        dec_preempt_count();
 }
 
-int kstack_depth_to_print = 12;
-
 void printk_address(unsigned long address, int reliable)
 {
-#ifdef CONFIG_KALLSYMS
-       unsigned long offset = 0, symsize;
-       const char *symname;
-       char *modname;
-       char *delim = ":";
-       char namebuf[KSYM_NAME_LEN];
-       char reliab[4] = "";
-
-       symname = kallsyms_lookup(address, &symsize, &offset,
-                                       &modname, namebuf);
-       if (!symname) {
-               printk(" [<%016lx>]\n", address);
-               return;
-       }
-       if (!reliable)
-               strcpy(reliab, "? ");
-
-       if (!modname)
-               modname = delim = "";
-       printk(" [<%016lx>] %s%s%s%s%s+0x%lx/0x%lx\n",
-               address, reliab, delim, modname, delim, symname, offset, symsize);
-#else
-       printk(" [<%016lx>]\n", address);
-#endif
+       printk(" [<%016lx>] %s%pS\n", address, reliable ? "": "? ", (void *) address);
 }
 
 static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack,
@@ -206,8 +164,6 @@ static unsigned long *in_exception_stack
        return NULL;
 }
 
-#define MSG(txt) ops->warning(data, txt)
-
 /*
  * x86-64 can have up to three kernel stacks:
  * process stack
@@ -234,11 +190,11 @@ struct stack_frame {
        unsigned long return_address;
 };
 
-
-static inline unsigned long print_context_stack(struct thread_info *tinfo,
-                               unsigned long *stack, unsigned long bp,
-                               const struct stacktrace_ops *ops, void *data,
-                               unsigned long *end)
+static inline unsigned long
+print_context_stack(struct thread_info *tinfo,
+               unsigned long *stack, unsigned long bp,
+               const struct stacktrace_ops *ops, void *data,
+               unsigned long *end)
 {
        struct stack_frame *frame = (struct stack_frame *)bp;
 
@@ -260,7 +216,7 @@ static inline unsigned long print_contex
        return bp;
 }
 
-void dump_trace(struct task_struct *tsk, struct pt_regs *regs,
+void dump_trace(struct task_struct *task, struct pt_regs *regs,
                unsigned long *stack, unsigned long bp,
                const struct stacktrace_ops *ops, void *data)
 {
@@ -269,36 +225,34 @@ void dump_trace(struct task_struct *tsk,
        unsigned used = 0;
        struct thread_info *tinfo;
 
-       if (!tsk)
-               tsk = current;
-       tinfo = task_thread_info(tsk);
+       if (!task)
+               task = current;
 
        if (!stack) {
                unsigned long dummy;
                stack = &dummy;
-               if (tsk && tsk != current)
-                       stack = (unsigned long *)tsk->thread.sp;
+               if (task && task != current)
+                       stack = (unsigned long *)task->thread.sp;
        }
 
 #ifdef CONFIG_FRAME_POINTER
        if (!bp) {
-               if (tsk == current) {
+               if (task == current) {
                        /* Grab bp right from our regs */
-                       asm("movq %%rbp, %0" : "=r" (bp):);
+                       asm("movq %%rbp, %0" : "=r" (bp) :);
                } else {
                        /* bp is the last reg pushed by switch_to */
-                       bp = *(unsigned long *) tsk->thread.sp;
+                       bp = *(unsigned long *) task->thread.sp;
                }
        }
 #endif
 
-
-
        /*
         * Print function call entries in all stacks, starting at the
         * current stack address. If the stacks consist of nested
         * exceptions
         */
+       tinfo = task_thread_info(task);
        for (;;) {
                char *id;
                unsigned long *estack_end;
@@ -383,18 +337,24 @@ static const struct stacktrace_ops print
        .address = print_trace_address,
 };
 
-void
-show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long *stack,
-               unsigned long bp)
+static void
+show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
+               unsigned long *stack, unsigned long bp, char *log_lvl)
 {
        printk("\nCall Trace:\n");
-       dump_trace(tsk, regs, stack, bp, &print_trace_ops, NULL);
+       dump_trace(task, regs, stack, bp, &print_trace_ops, log_lvl);
        printk("\n");
 }
 
+void show_trace(struct task_struct *task, struct pt_regs *regs,
+               unsigned long *stack, unsigned long bp)
+{
+       show_trace_log_lvl(task, regs, stack, bp, "");
+}
+
 static void
-_show_stack(struct task_struct *tsk, struct pt_regs *regs, unsigned long *sp,
-                                                       unsigned long bp)
+show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs,
+               unsigned long *sp, unsigned long bp, char *log_lvl)
 {
        unsigned long *stack;
        int i;
@@ -406,14 +366,14 @@ _show_stack(struct task_struct *tsk, str
        // back trace for this cpu.
 
        if (sp == NULL) {
-               if (tsk)
-                       sp = (unsigned long *)tsk->thread.sp;
+               if (task)
+                       sp = (unsigned long *)task->thread.sp;
                else
                        sp = (unsigned long *)&sp;
        }
 
        stack = sp;
-       for(i=0; i < kstack_depth_to_print; i++) {
+       for (i = 0; i < kstack_depth_to_print; i++) {
                if (stack >= irqstack && stack <= irqstack_end) {
                        if (stack == irqstack_end) {
                                stack = (unsigned long *) (irqstack_end[-1]);
@@ -428,12 +388,12 @@ _show_stack(struct task_struct *tsk, str
                printk(" %016lx", *stack++);
                touch_nmi_watchdog();
        }
-       show_trace(tsk, regs, sp, bp);
+       show_trace_log_lvl(task, regs, sp, bp, log_lvl);
 }
 
-void show_stack(struct task_struct *tsk, unsigned long * sp)
+void show_stack(struct task_struct *task, unsigned long *sp)
 {
-       _show_stack(tsk, NULL, sp, 0);
+       show_stack_log_lvl(task, NULL, sp, 0, "");
 }
 
 /*
@@ -441,8 +401,8 @@ void show_stack(struct task_struct *tsk,
  */
 void dump_stack(void)
 {
-       unsigned long dummy;
        unsigned long bp = 0;
+       unsigned long stack;
 
 #ifdef CONFIG_FRAME_POINTER
        if (!bp)
@@ -454,7 +414,7 @@ void dump_stack(void)
                init_utsname()->release,
                (int)strcspn(init_utsname()->version, " "),
                init_utsname()->version);
-       show_trace(NULL, NULL, &dummy, bp);
+       show_trace(NULL, NULL, &stack, bp);
 }
 
 EXPORT_SYMBOL(dump_stack);
@@ -465,12 +425,8 @@ void show_registers(struct pt_regs *regs
        unsigned long sp;
        const int cpu = smp_processor_id();
        struct task_struct *cur = cpu_pda(cpu)->pcurrent;
-       u8 *ip;
-       unsigned int code_prologue = code_bytes * 43 / 64;
-       unsigned int code_len = code_bytes;
 
        sp = regs->sp;
-       ip = (u8 *) regs->ip - code_prologue;
        printk("CPU %d ", cpu);
        __show_regs(regs);
        printk("Process %s (pid: %d, threadinfo %p, task %p)\n",
@@ -481,15 +437,22 @@ void show_registers(struct pt_regs *regs
         * time of the fault..
         */
        if (!user_mode(regs)) {
+               unsigned int code_prologue = code_bytes * 43 / 64;
+               unsigned int code_len = code_bytes;
                unsigned char c;
+               u8 *ip;
+
                printk("Stack: ");
-               _show_stack(NULL, regs, (unsigned long *)sp, regs->bp);
+               show_stack_log_lvl(NULL, regs, (unsigned long *)sp,
+                               regs->bp, "");
                printk("\n");
 
                printk(KERN_EMERG "Code: ");
+
+               ip = (u8 *)regs->ip - code_prologue;
                if (ip < (u8 *)PAGE_OFFSET || probe_kernel_address(ip, c)) {
                        /* try starting at RIP */
-                       ip = (u8 *) regs->ip;
+                       ip = (u8 *)regs->ip;
                        code_len = code_len - code_prologue + 1;
                }
                for (i = 0; i < code_len; i++, ip++) {
@@ -505,7 +468,7 @@ void show_registers(struct pt_regs *regs
                }
        }
        printk("\n");
-}      
+}
 
 int is_valid_bugaddr(unsigned long ip)
 {
@@ -545,7 +508,7 @@ unsigned __kprobes long oops_begin(void)
 }
 
 void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr)
-{ 
+{
        die_owner = -1;
        bust_spinlocks(0);
        die_nest_count--;
@@ -563,10 +526,9 @@ void __kprobes oops_end(unsigned long fl
        do_exit(signr);
 }
 
-int __kprobes __die(const char * str, struct pt_regs * regs, long err)
+int __kprobes __die(const char *str, struct pt_regs *regs, long err)
 {
-       static int die_counter;
-       printk(KERN_EMERG "%s: %04lx [%u] ", str, err & 0xffff,++die_counter);
+       printk(KERN_EMERG "%s: %04lx [%u] ", str, err & 0xffff, ++die_counter);
 #ifdef CONFIG_PREEMPT
        printk("PREEMPT ");
 #endif
@@ -577,8 +539,10 @@ int __kprobes __die(const char * str, st
        printk("DEBUG_PAGEALLOC");
 #endif
        printk("\n");
-       if (notify_die(DIE_OOPS, str, regs, err, current->thread.trap_no, SIGSEGV) == NOTIFY_STOP)
+       if (notify_die(DIE_OOPS, str, regs, err,
+                       current->thread.trap_no, SIGSEGV) == NOTIFY_STOP)
                return 1;
+
        show_registers(regs);
        add_taint(TAINT_DIE);
        /* Executive summary in case the oops scrolled away */
@@ -590,7 +554,7 @@ int __kprobes __die(const char * str, st
        return 0;
 }
 
-void die(const char * str, struct pt_regs * regs, long err)
+void die(const char *str, struct pt_regs *regs, long err)
 {
        unsigned long flags = oops_begin();
 
@@ -608,8 +572,7 @@ die_nmi(char *str, struct pt_regs *regs,
 {
        unsigned long flags;
 
-       if (notify_die(DIE_NMIWATCHDOG, str, regs, 0, 2, SIGINT) ==
-           NOTIFY_STOP)
+       if (notify_die(DIE_NMIWATCHDOG, str, regs, 0, 2, SIGINT) == NOTIFY_STOP)
                return;
 
        flags = oops_begin();
@@ -617,7 +580,9 @@ die_nmi(char *str, struct pt_regs *regs,
         * We are in trouble anyway, lets at least try
         * to get a message out.
         */
-       printk(str, smp_processor_id());
+       printk(KERN_EMERG "%s", str);
+       printk(" on CPU%d, ip %08lx, registers:\n",
+               smp_processor_id(), regs->ip);
        show_registers(regs);
        if (kexec_should_crash(current))
                crash_kexec(regs);
@@ -630,44 +595,44 @@ die_nmi(char *str, struct pt_regs *regs,
 }
 #endif
 
-static void __kprobes do_trap(int trapnr, int signr, char *str,
-                             struct pt_regs * regs, long error_code,
-                             siginfo_t *info)
+static void __kprobes
+do_trap(int trapnr, int signr, char *str, struct pt_regs *regs,
+       long error_code, siginfo_t *info)
 {
        struct task_struct *tsk = current;
 
-       if (user_mode(regs)) {
-               /*
-                * We want error_code and trap_no set for userspace
-                * faults and kernelspace faults which result in
-                * die(), but not kernelspace faults which are fixed
-                * up.  die() gives the process no chance to handle
-                * the signal and notice the kernel fault information,
-                * so that won't result in polluting the information
-                * about previously queued, but not yet delivered,
-                * faults.  See also do_general_protection below.
-                */
-               tsk->thread.error_code = error_code;
-               tsk->thread.trap_no = trapnr;
+       if (!user_mode(regs))
+               goto kernel_trap;
 
-               if (show_unhandled_signals && unhandled_signal(tsk, signr) &&
-                   printk_ratelimit()) {
-                       printk(KERN_INFO
-                              "%s[%d] trap %s ip:%lx sp:%lx error:%lx",
-                              tsk->comm, tsk->pid, str,
-                              regs->ip, regs->sp, error_code);
-                       print_vma_addr(" in ", regs->ip);
-                       printk("\n");
-               }
+       /*
+        * We want error_code and trap_no set for userspace faults and
+        * kernelspace faults which result in die(), but not
+        * kernelspace faults which are fixed up.  die() gives the
+        * process no chance to handle the signal and notice the
+        * kernel fault information, so that won't result in polluting
+        * the information about previously queued, but not yet
+        * delivered, faults.  See also do_general_protection below.
+        */
+       tsk->thread.error_code = error_code;
+       tsk->thread.trap_no = trapnr;
 
-               if (info)
-                       force_sig_info(signr, info, tsk);
-               else
-                       force_sig(signr, tsk);
-               return;
+       if (show_unhandled_signals && unhandled_signal(tsk, signr) &&
+           printk_ratelimit()) {
+               printk(KERN_INFO
+                      "%s[%d] trap %s ip:%lx sp:%lx error:%lx",
+                      tsk->comm, tsk->pid, str,
+                      regs->ip, regs->sp, error_code);
+               print_vma_addr(" in ", regs->ip);
+               printk("\n");
        }
 
+       if (info)
+               force_sig_info(signr, info, tsk);
+       else
+               force_sig(signr, tsk);
+       return;
 
+kernel_trap:
        if (!fixup_exception(regs)) {
                tsk->thread.error_code = error_code;
                tsk->thread.trap_no = trapnr;
@@ -677,41 +642,39 @@ static void __kprobes do_trap(int trapnr
 }
 
 #define DO_ERROR(trapnr, signr, str, name) \
-asmlinkage void do_##name(struct pt_regs * regs, long error_code) \
-{ \
-       if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \
-                                                       == NOTIFY_STOP) \
-               return; \
+asmlinkage void do_##name(struct pt_regs * regs, long error_code)      \
+{                                                                      \
+       if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr)  \
+                                                       == NOTIFY_STOP) \
+               return;                                                 \
        conditional_sti(regs);                                          \
-       do_trap(trapnr, signr, str, regs, error_code, NULL); \
+       do_trap(trapnr, signr, str, regs, error_code, NULL);            \
 }
 
-#define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \
-asmlinkage void do_##name(struct pt_regs * regs, long error_code) \
-{ \
-       siginfo_t info; \
-       info.si_signo = signr; \
-       info.si_errno = 0; \
-       info.si_code = sicode; \
-       info.si_addr = (void __user *)siaddr; \
-       trace_hardirqs_fixup(); \
-       if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \
-                                                       == NOTIFY_STOP) \
-               return; \
+#define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr)                \
+asmlinkage void do_##name(struct pt_regs * regs, long error_code)      \
+{                                                                      \
+       siginfo_t info;                                                 \
+       info.si_signo = signr;                                          \
+       info.si_errno = 0;                                              \
+       info.si_code = sicode;                                          \
+       info.si_addr = (void __user *)siaddr;                           \
+       trace_hardirqs_fixup();                                         \
+       if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr)  \
+                                                       == NOTIFY_STOP) \
+               return;                                                 \
        conditional_sti(regs);                                          \
-       do_trap(trapnr, signr, str, regs, error_code, &info); \
+       do_trap(trapnr, signr, str, regs, error_code, &info);           \
 }
 
-DO_ERROR_INFO( 0, SIGFPE,  "divide error", divide_error, FPE_INTDIV, regs->ip)
-DO_ERROR( 4, SIGSEGV, "overflow", overflow)
-DO_ERROR( 5, SIGSEGV, "bounds", bounds)
-DO_ERROR_INFO( 6, SIGILL,  "invalid opcode", invalid_op, ILL_ILLOPN, regs->ip)
-DO_ERROR( 7, SIGSEGV, "device not available", device_not_available)
-DO_ERROR( 9, SIGFPE,  "coprocessor segment overrun", coprocessor_segment_overrun)
+DO_ERROR_INFO(0, SIGFPE, "divide error", divide_error, FPE_INTDIV, regs->ip)
+DO_ERROR(4, SIGSEGV, "overflow", overflow)
+DO_ERROR(5, SIGSEGV, "bounds", bounds)
+DO_ERROR_INFO(6, SIGILL, "invalid opcode", invalid_op, ILL_ILLOPN, regs->ip)
+DO_ERROR(9, SIGFPE, "coprocessor segment overrun", coprocessor_segment_overrun)
 DO_ERROR(10, SIGSEGV, "invalid TSS", invalid_TSS)
-DO_ERROR(11, SIGBUS,  "segment not present", segment_not_present)
+DO_ERROR(11, SIGBUS, "segment not present", segment_not_present)
 DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0)
-DO_ERROR(18, SIGSEGV, "reserved", reserved)
 
 /* Runs on IST stack */
 asmlinkage void do_stack_segment(struct pt_regs *regs, long error_code)
@@ -741,31 +704,34 @@ asmlinkage void do_double_fault(struct p
                die(str, regs, error_code);
 }
 
-asmlinkage void __kprobes do_general_protection(struct pt_regs * regs,
-                                               long error_code)
+asmlinkage void __kprobes
+do_general_protection(struct pt_regs *regs, long error_code)
 {
-       struct task_struct *tsk = current;
+       struct task_struct *tsk;
 
        conditional_sti(regs);
 
-       if (user_mode(regs)) {
-               tsk->thread.error_code = error_code;
-               tsk->thread.trap_no = 13;
+       tsk = current;
+       if (!user_mode(regs))
+               goto gp_in_kernel;
 
-               if (show_unhandled_signals && unhandled_signal(tsk, SIGSEGV) &&
-                   printk_ratelimit()) {
-                       printk(KERN_INFO
-                      "%s[%d] general protection ip:%lx sp:%lx error:%lx",
-                              tsk->comm, tsk->pid,
-                              regs->ip, regs->sp, error_code);
-                       print_vma_addr(" in ", regs->ip);
-                       printk("\n");
-               }
+       tsk->thread.error_code = error_code;
+       tsk->thread.trap_no = 13;
 
-               force_sig(SIGSEGV, tsk);
-               return;
-       } 
+       if (show_unhandled_signals && unhandled_signal(tsk, SIGSEGV) &&
+                       printk_ratelimit()) {
+               printk(KERN_INFO
+                       "%s[%d] general protection ip:%lx sp:%lx error:%lx",
+                       tsk->comm, tsk->pid,
+                       regs->ip, regs->sp, error_code);
+               print_vma_addr(" in ", regs->ip);
+               printk("\n");
+       }
 
+       force_sig(SIGSEGV, tsk);
+       return;
+
+gp_in_kernel:
        if (fixup_exception(regs))
                return;
 
@@ -778,14 +744,14 @@ asmlinkage void __kprobes do_general_pro
 }
 
 static notrace __kprobes void
-mem_parity_error(unsigned char reason, struct pt_regs * regs)
+mem_parity_error(unsigned char reason, struct pt_regs *regs)
 {
        printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x.\n",
                reason);
        printk(KERN_EMERG "You have some hardware problem, likely on the PCI bus.\n");
 
 #if defined(CONFIG_EDAC)
-       if(edac_handler_set()) {
+       if (edac_handler_set()) {
                edac_atomic_assert_error();
                return;
        }
@@ -801,7 +767,7 @@ mem_parity_error(unsigned char reason, s
 }
 
 static notrace __kprobes void
-io_check_error(unsigned char reason, struct pt_regs * regs)
+io_check_error(unsigned char reason, struct pt_regs *regs)
 {
        printk("NMI: IOCK error (debug interrupt?)\n");
        show_registers(regs);
@@ -827,14 +793,14 @@ unknown_nmi_error(unsigned char reason, 
 
 /* Runs on IST stack. This code must keep interrupts off all the time.
    Nested NMIs are prevented by the CPU. */
-asmlinkage notrace  __kprobes void default_do_nmi(struct pt_regs *regs)
+asmlinkage notrace __kprobes void default_do_nmi(struct pt_regs *regs)
 {
        unsigned char reason = 0;
        int cpu;
 
        cpu = smp_processor_id();
 
-       /* Only the BSP gets external NMIs from the system.  */
+       /* Only the BSP gets external NMIs from the system. */
        if (!cpu)
                reason = get_nmi_reason();
 
@@ -847,33 +813,58 @@ asmlinkage notrace  __kprobes void defau
                 * Ok, so this is none of the documented NMI sources,
                 * so it must be the NMI watchdog.
                 */
-               if (nmi_watchdog_tick(regs,reason))
+               if (nmi_watchdog_tick(regs, reason))
                        return;
 #endif
-               if (!do_nmi_callback(regs,cpu))
+               if (!do_nmi_callback(regs, cpu))
                        unknown_nmi_error(reason, regs);
 
                return;
        }
        if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP)
-               return; 
+               return;
 
        /* AK: following checks seem to be broken on modern chipsets. FIXME */
-
        if (reason & 0x80)
                mem_parity_error(reason, regs);
        if (reason & 0x40)
                io_check_error(reason, regs);
 }
 
+asmlinkage notrace __kprobes void
+do_nmi(struct pt_regs *regs, long error_code)
+{
+       nmi_enter();
+
+       add_pda(__nmi_count, 1);
+
+       if (!ignore_nmis)
+               default_do_nmi(regs);
+
+       nmi_exit();
+}
+
+void stop_nmi(void)
+{
+       acpi_nmi_disable();
+       ignore_nmis++;
+}
+
+void restart_nmi(void)
+{
+       ignore_nmis--;
+       acpi_nmi_enable();
+}
+
 /* runs on IST stack. */
 asmlinkage void __kprobes do_int3(struct pt_regs * regs, long error_code)
 {
        trace_hardirqs_fixup();
 
-       if (notify_die(DIE_INT3, "int3", regs, error_code, 3, SIGTRAP) == NOTIFY_STOP) {
+       if (notify_die(DIE_INT3, "int3", regs, error_code, 3, SIGTRAP)
+                       == NOTIFY_STOP)
                return;
-       }
+
        preempt_conditional_sti(regs);
        do_trap(3, SIGTRAP, "int3", regs, error_code, NULL);
        preempt_conditional_cli(regs);
@@ -904,8 +895,8 @@ asmlinkage __kprobes struct pt_regs *syn
 asmlinkage void __kprobes do_debug(struct pt_regs * regs,
                                   unsigned long error_code)
 {
-       unsigned long condition;
        struct task_struct *tsk = current;
+       unsigned long condition;
        siginfo_t info;
 
        trace_hardirqs_fixup();
@@ -926,21 +917,19 @@ asmlinkage void __kprobes do_debug(struc
 
        /* Mask out spurious debug traps due to lazy DR7 setting */
        if (condition & (DR_TRAP0|DR_TRAP1|DR_TRAP2|DR_TRAP3)) {
-               if (!tsk->thread.debugreg7) { 
+               if (!tsk->thread.debugreg7)
                        goto clear_dr7;
-               }
        }
 
        tsk->thread.debugreg6 = condition;
 
-
        /*
         * Single-stepping through TF: make sure we ignore any events in
         * kernel space (but re-enable TF when returning to user mode).
         */
        if (condition & DR_STEP) {
-                if (!user_mode(regs))
-                       goto clear_TF_reenable;
+               if (!user_mode(regs))
+                       goto clear_TF_reenable;
        }
 
        /* Ok, finally something we can handle */
@@ -953,7 +942,7 @@ asmlinkage void __kprobes do_debug(struc
        force_sig_info(SIGTRAP, &info, tsk);
 
 clear_dr7:
-       set_debugreg(0UL, 7);
+       set_debugreg(0, 7);
        preempt_conditional_cli(regs);
        return;
 
@@ -961,6 +950,7 @@ clear_TF_reenable:
        set_tsk_thread_flag(tsk, TIF_SINGLESTEP);
        regs->flags &= ~X86_EFLAGS_TF;
        preempt_conditional_cli(regs);
+       return;
 }
 
 static int kernel_math_error(struct pt_regs *regs, const char *str, int trapnr)
@@ -983,7 +973,7 @@ static int kernel_math_error(struct pt_r
 asmlinkage void do_coprocessor_error(struct pt_regs *regs)
 {
        void __user *ip = (void __user *)(regs->ip);
-       struct task_struct * task;
+       struct task_struct *task;
        siginfo_t info;
        unsigned short cwd, swd;
 
@@ -1016,30 +1006,30 @@ asmlinkage void do_coprocessor_error(str
        cwd = get_fpu_cwd(task);
        swd = get_fpu_swd(task);
        switch (swd & ~cwd & 0x3f) {
-               case 0x000:
-               default:
-                       break;
-               case 0x001: /* Invalid Op */
-                       /*
-                        * swd & 0x240 == 0x040: Stack Underflow
-                        * swd & 0x240 == 0x240: Stack Overflow
-                        * User must clear the SF bit (0x40) if set
-                        */
-                       info.si_code = FPE_FLTINV;
-                       break;
-               case 0x002: /* Denormalize */
-               case 0x010: /* Underflow */
-                       info.si_code = FPE_FLTUND;
-                       break;
-               case 0x004: /* Zero Divide */
-                       info.si_code = FPE_FLTDIV;
-                       break;
-               case 0x008: /* Overflow */
-                       info.si_code = FPE_FLTOVF;
-                       break;
-               case 0x020: /* Precision */
-                       info.si_code = FPE_FLTRES;
-                       break;
+       case 0x000: /* No unmasked exception */
+       default: /* Multiple exceptions */
+               break;
+       case 0x001: /* Invalid Op */
+               /*
+                * swd & 0x240 == 0x040: Stack Underflow
+                * swd & 0x240 == 0x240: Stack Overflow
+                * User must clear the SF bit (0x40) if set
+                */
+               info.si_code = FPE_FLTINV;
+               break;
+       case 0x002: /* Denormalize */
+       case 0x010: /* Underflow */
+               info.si_code = FPE_FLTUND;
+               break;
+       case 0x004: /* Zero Divide */
+               info.si_code = FPE_FLTDIV;
+               break;
+       case 0x008: /* Overflow */
+               info.si_code = FPE_FLTOVF;
+               break;
+       case 0x020: /* Precision */
+               info.si_code = FPE_FLTRES;
+               break;
        }
        force_sig_info(SIGFPE, &info, task);
 }
@@ -1052,7 +1042,7 @@ asmlinkage void bad_intr(void)
 asmlinkage void do_simd_coprocessor_error(struct pt_regs *regs)
 {
        void __user *ip = (void __user *)(regs->ip);
-       struct task_struct * task;
+       struct task_struct *task;
        siginfo_t info;
        unsigned short mxcsr;
 
@@ -1080,25 +1070,25 @@ asmlinkage void do_simd_coprocessor_erro
         */
        mxcsr = get_fpu_mxcsr(task);
        switch (~((mxcsr & 0x1f80) >> 7) & (mxcsr & 0x3f)) {
-               case 0x000:
-               default:
-                       break;
-               case 0x001: /* Invalid Op */
-                       info.si_code = FPE_FLTINV;
-                       break;
-               case 0x002: /* Denormalize */
-               case 0x010: /* Underflow */
-                       info.si_code = FPE_FLTUND;
-                       break;
-               case 0x004: /* Zero Divide */
-                       info.si_code = FPE_FLTDIV;
-                       break;
-               case 0x008: /* Overflow */
-                       info.si_code = FPE_FLTOVF;
-                       break;
-               case 0x020: /* Precision */
-                       info.si_code = FPE_FLTRES;
-                       break;
+       case 0x000:
+       default:
+               break;
+       case 0x001: /* Invalid Op */
+               info.si_code = FPE_FLTINV;
+               break;
+       case 0x002: /* Denormalize */
+       case 0x010: /* Underflow */
+               info.si_code = FPE_FLTUND;
+               break;
+       case 0x004: /* Zero Divide */
+               info.si_code = FPE_FLTDIV;
+               break;
+       case 0x008: /* Overflow */
+               info.si_code = FPE_FLTOVF;
+               break;
+       case 0x020: /* Precision */
+               info.si_code = FPE_FLTRES;
+               break;
        }
        force_sig_info(SIGFPE, &info, task);
 }
@@ -1118,7 +1108,7 @@ asmlinkage void __attribute__((weak)) mc
 }
 
 /*
- *  'math_state_restore()' saves the current math information in the
+ * 'math_state_restore()' saves the current math information in the
  * old math state array, and gets the new ones from the current task
  *
  * Careful.. There are problems with IBM-designed IRQ13 behaviour.
@@ -1145,7 +1135,14 @@ asmlinkage void math_state_restore(void)
 
         /* clts(); */ /* 'clts' is done for us by Xen during virtual trap. */
 
-       restore_fpu_checking(&me->thread.xstate->fxsave);
+       /*
+        * Paranoid restore. send a SIGSEGV if we fail to restore the state.
+        */
+       if (unlikely(restore_fpu_checking(&me->thread.xstate->fxsave))) {
+               stts();
+               force_sig(SIGSEGV, me);
+               return;
+       }
        task_thread_info(me)->status |= TS_USEDFPU;
        me->fpu_counter++;
 }
@@ -1190,13 +1187,12 @@ void __init trap_init(void)
         ret = HYPERVISOR_set_trap_table(trap_table);
         if (ret) 
                printk("HYPERVISOR_set_trap_table failed: error %d\n", ret);
-
        /*
         * initialize the per thread extended state:
         */
-        init_thread_xstate();
+       init_thread_xstate();
        /*
-        * Should be a barrier for any external CPU state.
+        * Should be a barrier for any external CPU state:
         */
        cpu_init();
 }
@@ -1212,27 +1208,25 @@ void __cpuinit smp_trap_init(trap_info_t
        }
 }
 
-
 static int __init oops_setup(char *s)
-{ 
+{
        if (!s)
                return -EINVAL;
        if (!strcmp(s, "panic"))
                panic_on_oops = 1;
        return 0;
-} 
+}
 early_param("oops", oops_setup);
 
 static int __init kstack_setup(char *s)
 {
        if (!s)
                return -EINVAL;
-       kstack_depth_to_print = simple_strtoul(s,NULL,0);
+       kstack_depth_to_print = simple_strtoul(s, NULL, 0);
        return 0;
 }
 early_param("kstack", kstack_setup);
 
-
 static int __init code_bytes_setup(char *s)
 {
        code_bytes = simple_strtoul(s, NULL, 0);
Index: head-2008-12-01/arch/x86/kernel/vsyscall_64-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/kernel/vsyscall_64-xen.c      2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/kernel/vsyscall_64-xen.c   2008-12-01 11:49:07.000000000 +0100
@@ -42,7 +42,8 @@
 #include <asm/topology.h>
 #include <asm/vgtod.h>
 
-#define __vsyscall(nr) __attribute__ ((unused,__section__(".vsyscall_" #nr)))
+#define __vsyscall(nr) \
+               __attribute__ ((unused, __section__(".vsyscall_" #nr))) notrace
 #define __syscall_clobber "r11","cx","memory"
 
 /*
@@ -264,10 +265,7 @@ static void __cpuinit vsyscall_set_cpu(i
        d |= cpu;
        d |= (node & 0xf) << 12;
        d |= (node >> 4) << 48;
-       if (HYPERVISOR_update_descriptor(virt_to_machine(get_cpu_gdt_table(cpu)
-                                                        + GDT_ENTRY_PER_CPU),
-                                        d))
-               BUG();
+       write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_PER_CPU, &d, DESCTYPE_S);
 }
 
 static void __cpuinit cpu_vsyscall_init(void *arg)
@@ -281,7 +279,7 @@ cpu_vsyscall_notifier(struct notifier_bl
 {
        long cpu = (long)arg;
        if (action == CPU_ONLINE || action == CPU_ONLINE_FROZEN)
-               smp_call_function_single(cpu, cpu_vsyscall_init, NULL, 0, 1);
+               smp_call_function_single(cpu, cpu_vsyscall_init, NULL, 1);
        return NOTIFY_DONE;
 }
 
@@ -311,7 +309,7 @@ static int __init vsyscall_init(void)
 #ifdef CONFIG_SYSCTL
        register_sysctl_table(kernel_root_table2);
 #endif
-       on_each_cpu(cpu_vsyscall_init, NULL, 0, 1);
+       on_each_cpu(cpu_vsyscall_init, NULL, 1);
        hotcpu_notifier(cpu_vsyscall_notifier, 0);
        return 0;
 }
Index: head-2008-12-01/arch/x86/mach-xen/setup.c
===================================================================
--- head-2008-12-01.orig/arch/x86/mach-xen/setup.c      2008-12-01 11:37:10.000000000 +0100
+++ head-2008-12-01/arch/x86/mach-xen/setup.c   2008-12-01 11:49:07.000000000 +0100
@@ -17,6 +17,8 @@
 #include <xen/interface/callback.h>
 #include <xen/interface/memory.h>
 
+#ifdef CONFIG_X86_32
+
 #ifdef CONFIG_HOTPLUG_CPU
 #define DEFAULT_SEND_IPI       (1)
 #else
@@ -44,51 +46,6 @@ static int __init print_ipi_mode(void)
 
 late_initcall(print_ipi_mode);
 
-/**
- * machine_specific_memory_setup - Hook for machine specific memory setup.
- *
- * Description:
- *     This is included late in kernel/setup.c so that it can make
- *     use of all of the static functions.
- **/
-
-char * __init machine_specific_memory_setup(void)
-{
-       int rc;
-       struct xen_memory_map memmap;
-       /*
-        * This is rather large for a stack variable but this early in
-        * the boot process we know we have plenty slack space.
-        */
-       struct e820entry map[E820MAX];
-
-       memmap.nr_entries = E820MAX;
-       set_xen_guest_handle(memmap.buffer, map);
-
-       rc = HYPERVISOR_memory_op(XENMEM_memory_map, &memmap);
-       if ( rc == -ENOSYS ) {
-               memmap.nr_entries = 1;
-               map[0].addr = 0ULL;
-               map[0].size = PFN_PHYS((unsigned long long)xen_start_info->nr_pages);
-               /* 8MB slack (to balance backend allocations). */
-               map[0].size += 8ULL << 20;
-               map[0].type = E820_RAM;
-               rc = 0;
-       }
-       BUG_ON(rc);
-
-       sanitize_e820_map(map, (char *)&memmap.nr_entries);
-
-       BUG_ON(copy_e820_map(map, (char)memmap.nr_entries) < 0);
-
-       return "Xen";
-}
-
-
-extern void hypervisor_callback(void);
-extern void failsafe_callback(void);
-extern void nmi(void);
-
 unsigned long *machine_to_phys_mapping = (void *)MACH2PHYS_VIRT_START;
 EXPORT_SYMBOL(machine_to_phys_mapping);
 unsigned int machine_to_phys_order;
@@ -121,33 +78,66 @@ void __init pre_setup_arch_hook(void)
                        (unsigned long *)xen_start_info->mfn_list;
 }
 
+#endif /* CONFIG_X86_32 */
+
+extern void hypervisor_callback(void);
+extern void failsafe_callback(void);
+extern void nmi(void);
+
+#ifdef CONFIG_X86_64
+#include <asm/proto.h>
+#define CALLBACK_ADDR(fn) ((unsigned long)(fn))
+#else
+#define CALLBACK_ADDR(fn) { __KERNEL_CS, (unsigned long)(fn) }
+#endif
+
 void __init machine_specific_arch_setup(void)
 {
        int ret;
        static struct callback_register __initdata event = {
                .type = CALLBACKTYPE_event,
-               .address = { __KERNEL_CS, (unsigned long)hypervisor_callback },
+               .address = CALLBACK_ADDR(hypervisor_callback)
        };
        static struct callback_register __initdata failsafe = {
                .type = CALLBACKTYPE_failsafe,
-               .address = { __KERNEL_CS, (unsigned long)failsafe_callback },
+               .address = CALLBACK_ADDR(failsafe_callback)
+       };
+#ifdef CONFIG_X86_64
+       static struct callback_register __initdata syscall = {
+               .type = CALLBACKTYPE_syscall,
+               .address = CALLBACK_ADDR(system_call)
        };
+#endif
+#if defined(CONFIG_X86_LOCAL_APIC) || defined(CONFIG_X86_32)
        static struct callback_register __initdata nmi_cb = {
                .type = CALLBACKTYPE_nmi,
-               .address = { __KERNEL_CS, (unsigned long)nmi },
+               .address = CALLBACK_ADDR(nmi)
        };
+#endif
 
        ret = HYPERVISOR_callback_op(CALLBACKOP_register, &event);
        if (ret == 0)
                ret = HYPERVISOR_callback_op(CALLBACKOP_register, &failsafe);
+#ifdef CONFIG_X86_64
+       if (ret == 0)
+               ret = HYPERVISOR_callback_op(CALLBACKOP_register, &syscall);
+#endif
 #if CONFIG_XEN_COMPAT <= 0x030002
+#ifdef CONFIG_X86_32
        if (ret == -ENOSYS)
                ret = HYPERVISOR_set_callbacks(
                        event.address.cs, event.address.eip,
                        failsafe.address.cs, failsafe.address.eip);
+#else
+               ret = HYPERVISOR_set_callbacks(
+                       event.address,
+                       failsafe.address,
+                       syscall.address);
+#endif
 #endif
        BUG_ON(ret);
 
+#if defined(CONFIG_X86_LOCAL_APIC) || defined(CONFIG_X86_32)
        ret = HYPERVISOR_callback_op(CALLBACKOP_register, &nmi_cb);
 #if CONFIG_XEN_COMPAT <= 0x030002
        if (ret == -ENOSYS) {
@@ -158,15 +148,43 @@ void __init machine_specific_arch_setup(
                HYPERVISOR_nmi_op(XENNMI_register_callback, &cb);
        }
 #endif
+#endif
 
+#ifdef CONFIG_X86_32
        /* Do an early initialization of the fixmap area */
        {
                extern pte_t swapper_pg_fixmap[PTRS_PER_PTE];
                unsigned long addr = __fix_to_virt(FIX_EARLYCON_MEM_BASE);
                pud_t *pud = pud_offset(swapper_pg_dir + pgd_index(addr), addr);
                pmd_t *pmd = pmd_offset(pud, addr);
+               unsigned int i;
 
                make_lowmem_page_readonly(swapper_pg_fixmap, XENFEAT_writable_page_tables);
                set_pmd(pmd, __pmd(__pa_symbol(swapper_pg_fixmap) | _PAGE_TABLE));
+
+#define __FIXADDR_TOP (-PAGE_SIZE)
+#define FIX_BUG_ON(fix) BUILD_BUG_ON(pmd_index(__fix_to_virt(FIX_##fix)) \
+                       != pmd_index(__fix_to_virt(FIX_EARLYCON_MEM_BASE)))
+               FIX_BUG_ON(SHARED_INFO);
+               FIX_BUG_ON(ISAMAP_BEGIN);
+               FIX_BUG_ON(ISAMAP_END);
+#undef __FIXADDR_TOP
+               BUG_ON(pte_index(hypervisor_virt_start));
+
+               /* Switch to the real shared_info page, and clear the
+                * dummy page. */
+               set_fixmap(FIX_SHARED_INFO, xen_start_info->shared_info);
+               HYPERVISOR_shared_info = (shared_info_t *)fix_to_virt(FIX_SHARED_INFO);
+               memset(empty_zero_page, 0, sizeof(empty_zero_page));
+
+               /* Setup mapping of lower 1st MB */
+               for (i = 0; i < NR_FIX_ISAMAPS; i++)
+                       if (is_initial_xendomain())
+                               set_fixmap(FIX_ISAMAP_BEGIN - i, i * PAGE_SIZE);
+                       else
+                               __set_fixmap(FIX_ISAMAP_BEGIN - i,
+                                            virt_to_machine(empty_zero_page),
+                                            PAGE_KERNEL_RO);
        }
+#endif
 }
Index: head-2008-12-01/arch/x86/mm/fault-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/mm/fault-xen.c        2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/mm/fault-xen.c     2008-12-01 11:49:07.000000000 +0100
@@ -10,6 +10,7 @@
 #include <linux/string.h>
 #include <linux/types.h>
 #include <linux/ptrace.h>
+#include <linux/mmiotrace.h>
 #include <linux/mman.h>
 #include <linux/mm.h>
 #include <linux/smp.h>
@@ -49,17 +50,23 @@
 #define PF_RSVD                (1<<3)
 #define PF_INSTR       (1<<4)
 
+static inline int kmmio_fault(struct pt_regs *regs, unsigned long addr)
+{
+#ifdef CONFIG_MMIOTRACE_HOOKS
+       if (unlikely(is_kmmio_active()))
+               if (kmmio_handler(regs, addr) == 1)
+                       return -1;
+#endif
+       return 0;
+}
+
 static inline int notify_page_fault(struct pt_regs *regs)
 {
 #ifdef CONFIG_KPROBES
        int ret = 0;
 
        /* kprobe_running() needs smp_processor_id() */
-#ifdef CONFIG_X86_32
        if (!user_mode_vm(regs)) {
-#else
-       if (!user_mode(regs)) {
-#endif
                preempt_disable();
                if (kprobe_running() && kprobe_fault_handler(regs, 14))
                        ret = 1;
@@ -409,11 +416,7 @@ static void show_fault_oops(struct pt_re
                printk(KERN_CONT "NULL pointer dereference");
        else
                printk(KERN_CONT "paging request");
-#ifdef CONFIG_X86_32
-       printk(KERN_CONT " at %08lx\n", address);
-#else
-       printk(KERN_CONT " at %016lx\n", address);
-#endif
+       printk(KERN_CONT " at %p\n", (void *) address);
        printk(KERN_ALERT "IP:");
        printk_address(regs->ip, 1);
        dump_pagetable(address);
@@ -628,6 +631,8 @@ void __kprobes do_page_fault(struct pt_r
 
        if (notify_page_fault(regs))
                return;
+       if (unlikely(kmmio_fault(regs, address)))
+               return;
 
        /*
         * We fault-in kernel-space virtual memory on-demand. The
@@ -832,14 +837,10 @@ bad_area_nosemaphore:
                if (show_unhandled_signals && unhandled_signal(tsk, SIGSEGV) &&
                    printk_ratelimit()) {
                        printk(
-#ifdef CONFIG_X86_32
-                       "%s%s[%d]: segfault at %lx ip %08lx sp %08lx error %lx",
-#else
-                       "%s%s[%d]: segfault at %lx ip %lx sp %lx error %lx",
-#endif
+                       "%s%s[%d]: segfault at %lx ip %p sp %p error %lx",
                        task_pid_nr(tsk) > 1 ? KERN_INFO : KERN_EMERG,
-                       tsk->comm, task_pid_nr(tsk), address, regs->ip,
-                       regs->sp, error_code);
+                       tsk->comm, task_pid_nr(tsk), address,
+                       (void *) regs->ip, (void *) regs->sp, error_code);
                        print_vma_addr(" in ", regs->ip);
                        printk("\n");
                }
@@ -947,81 +948,45 @@ LIST_HEAD(pgd_list);
 void vmalloc_sync_all(void)
 {
 #ifdef CONFIG_X86_32
-       /*
-        * Note that races in the updates of insync and start aren't
-        * problematic: insync can only get set bits added, and updates to
-        * start are only improving performance (without affecting correctness
-        * if undone).
-        * XEN: To work on PAE, we need to iterate over PMDs rather than PGDs.
-        *      This change works just fine with 2-level paging too.
-        */
-#define sync_index(a) ((a) >> PMD_SHIFT)
-       static DECLARE_BITMAP(insync, PTRS_PER_PGD*PTRS_PER_PMD);
-       static unsigned long start = TASK_SIZE;
-       unsigned long address;
+       unsigned long address = VMALLOC_START & PGDIR_MASK;
 
        if (SHARED_KERNEL_PMD)
                return;
 
        BUILD_BUG_ON(TASK_SIZE & ~PMD_MASK);
-       for (address = start;
-            address < hypervisor_virt_start;
-            address += PMD_SIZE) {
-               if (!test_bit(sync_index(address), insync)) {
-                       unsigned long flags;
-                       struct page *page;
-
-                       spin_lock_irqsave(&pgd_lock, flags);
-                       /* XEN: failure path assumes non-empty pgd_list. */
-                       if (unlikely(list_empty(&pgd_list))) {
-                               spin_unlock_irqrestore(&pgd_lock, flags);
-                               return;
-                       }
-                       list_for_each_entry(page, &pgd_list, lru) {
-                               if (!vmalloc_sync_one(page_address(page),
-                                                     address))
-                                       break;
-                       }
-                       spin_unlock_irqrestore(&pgd_lock, flags);
-                       if (!page)
-                               set_bit(sync_index(address), insync);
+       for (; address < hypervisor_virt_start; address += PMD_SIZE) {
+               unsigned long flags;
+               struct page *page;
+
+               spin_lock_irqsave(&pgd_lock, flags);
+               list_for_each_entry(page, &pgd_list, lru) {
+                       if (!vmalloc_sync_one(page_address(page),
+                                             address))
+                               break;
                }
-               if (address == start && test_bit(sync_index(address), insync))
-                       start = address + PMD_SIZE;
+               spin_unlock_irqrestore(&pgd_lock, flags);
        }
 #else /* CONFIG_X86_64 */
-       /*
-        * Note that races in the updates of insync and start aren't
-        * problematic: insync can only get set bits added, and updates to
-        * start are only improving performance (without affecting correctness
-        * if undone).
-        */
-       static DECLARE_BITMAP(insync, PTRS_PER_PGD);
-       static unsigned long start = VMALLOC_START & PGDIR_MASK;
+       unsigned long start = VMALLOC_START & PGDIR_MASK;
        unsigned long address;
 
        for (address = start; address <= VMALLOC_END; address += PGDIR_SIZE) {
-               if (!test_bit(pgd_index(address), insync)) {
-                       const pgd_t *pgd_ref = pgd_offset_k(address);
-                       unsigned long flags;
-                       struct page *page;
-
-                       if (pgd_none(*pgd_ref))
-                               continue;
-                       spin_lock_irqsave(&pgd_lock, flags);
-                       list_for_each_entry(page, &pgd_list, lru) {
-                               pgd_t *pgd;
-                               pgd = (pgd_t *)page_address(page) + pgd_index(address);
-                               if (pgd_none(*pgd))
-                                       set_pgd(pgd, *pgd_ref);
-                               else
-                                       BUG_ON(pgd_page_vaddr(*pgd) != pgd_page_vaddr(*pgd_ref));
-                       }
-                       spin_unlock_irqrestore(&pgd_lock, flags);
-                       set_bit(pgd_index(address), insync);
+               const pgd_t *pgd_ref = pgd_offset_k(address);
+               unsigned long flags;
+               struct page *page;
+
+               if (pgd_none(*pgd_ref))
+                       continue;
+               spin_lock_irqsave(&pgd_lock, flags);
+               list_for_each_entry(page, &pgd_list, lru) {
+                       pgd_t *pgd;
+                       pgd = (pgd_t *)page_address(page) + pgd_index(address);
+                       if (pgd_none(*pgd))
+                               set_pgd(pgd, *pgd_ref);
+                       else
+                               BUG_ON(pgd_page_vaddr(*pgd) != pgd_page_vaddr(*pgd_ref));
                }
-               if (address == start)
-                       start = address + PGDIR_SIZE;
+               spin_unlock_irqrestore(&pgd_lock, flags);
        }
 #endif
 }
Index: head-2008-12-01/arch/x86/mm/hypervisor.c
===================================================================
--- head-2008-12-01.orig/arch/x86/mm/hypervisor.c       2008-12-01 11:37:10.000000000 +0100
+++ head-2008-12-01/arch/x86/mm/hypervisor.c    2008-12-01 11:49:07.000000000 +0100
@@ -837,42 +837,9 @@ int write_ldt_entry(struct desc_struct *
        return HYPERVISOR_update_descriptor(mach_lp, *(const u64*)desc);
 }
 
-#define MAX_BATCHED_FULL_PTES 32
-
-int xen_change_pte_range(struct mm_struct *mm, pmd_t *pmd,
-                        unsigned long addr, unsigned long end, pgprot_t newprot,
-                        int dirty_accountable)
+int write_gdt_entry(struct desc_struct *gdt, int entry, const void *desc,
+                   int type)
 {
-       int rc = 0, i = 0;
-       mmu_update_t u[MAX_BATCHED_FULL_PTES];
-       pte_t *pte;
-       spinlock_t *ptl;
-
-       if (!xen_feature(XENFEAT_mmu_pt_update_preserve_ad))
-               return 0;
-
-       pte = pte_offset_map_lock(mm, pmd, addr, &ptl);
-       do {
-               if (pte_present(*pte)) {
-                       pte_t ptent = pte_modify(*pte, newprot);
-
-                       if (dirty_accountable && pte_dirty(ptent))
-                               ptent = pte_mkwrite(ptent);
-                       u[i].ptr = (__pmd_val(*pmd) & PHYSICAL_PAGE_MASK)
-                                  | ((unsigned long)pte & ~PAGE_MASK)
-                                  | MMU_PT_UPDATE_PRESERVE_AD;
-                       u[i].val = __pte_val(ptent);
-                       if (++i == MAX_BATCHED_FULL_PTES) {
-                               if ((rc = HYPERVISOR_mmu_update(
-                                       &u[0], i, NULL, DOMID_SELF)) != 0)
-                                       break;
-                               i = 0;
-                       }
-               }
-       } while (pte++, addr += PAGE_SIZE, addr != end);
-       if (i)
-               rc = HYPERVISOR_mmu_update( &u[0], i, NULL, DOMID_SELF);
-       pte_unmap_unlock(pte - 1, ptl);
-       BUG_ON(rc && rc != -ENOSYS);
-       return !rc;
+       maddr_t mach_gp = virt_to_machine(gdt + entry);
+       return HYPERVISOR_update_descriptor(mach_gp, *(const u64*)desc);
 }
Index: head-2008-12-01/arch/x86/mm/init_32-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/mm/init_32-xen.c      2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/mm/init_32-xen.c   2008-12-01 11:49:07.000000000 +0100
@@ -54,6 +54,7 @@
 
 unsigned int __VMALLOC_RESERVE = 128 << 20;
 
+unsigned long max_low_pfn_mapped;
 unsigned long max_pfn_mapped;
 
 DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
@@ -61,6 +62,27 @@ unsigned long highstart_pfn, highend_pfn
 
 static noinline int do_test_wp_bit(void);
 
+
+static unsigned long __initdata table_start;
+static unsigned long __initdata table_end;
+static unsigned long __initdata table_top;
+
+static int __initdata after_init_bootmem;
+
+static __init void *alloc_low_page(unsigned long *phys)
+{
+       unsigned long pfn = table_end++;
+       void *adr;
+
+       if (pfn >= table_top)
+               panic("alloc_low_page: ran out of memory");
+
+       adr = __va(pfn * PAGE_SIZE);
+       memset(adr, 0, PAGE_SIZE);
+       *phys  = pfn * PAGE_SIZE;
+       return adr;
+}
+
 /*
  * Creates a middle page table and puts a pointer to it in the
  * given global directory entry. This only returns the gd entry
@@ -72,9 +94,12 @@ static pmd_t * __init one_md_table_init(
        pmd_t *pmd_table;
 
 #ifdef CONFIG_X86_PAE
+       unsigned long phys;
        if (!(__pgd_val(*pgd) & _PAGE_PRESENT)) {
-               pmd_table = (pmd_t *) alloc_bootmem_low_pages(PAGE_SIZE);
-
+               if (after_init_bootmem)
+                       pmd_table = (pmd_t *)alloc_bootmem_low_pages(PAGE_SIZE);
+               else
+                       pmd_table = (pmd_t *)alloc_low_page(&phys);
                paravirt_alloc_pmd(&init_mm, __pa(pmd_table) >> PAGE_SHIFT);
                make_lowmem_page_readonly(pmd_table, XENFEAT_writable_page_tables);
                set_pgd(pgd, __pgd(__pa(pmd_table) | _PAGE_PRESENT));
@@ -101,12 +126,16 @@ static pte_t * __init one_page_table_ini
 #endif
                pte_t *page_table = NULL;
 
+               if (after_init_bootmem) {
 #ifdef CONFIG_DEBUG_PAGEALLOC
-               page_table = (pte_t *) alloc_bootmem_pages(PAGE_SIZE);
+                       page_table = (pte_t *) alloc_bootmem_pages(PAGE_SIZE);
 #endif
-               if (!page_table) {
-                       page_table =
+                       if (!page_table)
+                               page_table =
                                (pte_t *)alloc_bootmem_low_pages(PAGE_SIZE);
+               } else {
+                       unsigned long phys;
+                       page_table = (pte_t *)alloc_low_page(&phys);
                }
 
                paravirt_alloc_pte(&init_mm, __pa(page_table) >> PAGE_SHIFT);
@@ -167,24 +196,24 @@ static inline int is_kernel_text(unsigne
  * of max_low_pfn pages, by creating page tables starting from address
  * PAGE_OFFSET:
  */
-static void __init kernel_physical_mapping_init(pgd_t *pgd_base)
+static void __init kernel_physical_mapping_init(pgd_t *pgd_base,
+                                               unsigned long start_pfn,
+                                               unsigned long end_pfn,
+                                               int use_pse)
 {
        int pgd_idx, pmd_idx, pte_ofs;
        unsigned long pfn;
        pgd_t *pgd;
        pmd_t *pmd;
        pte_t *pte;
+       unsigned pages_2m = 0, pages_4k = 0;
 
-       unsigned long max_ram_pfn = xen_start_info->nr_pages;
-       if (max_ram_pfn > max_low_pfn)
-               max_ram_pfn = max_low_pfn;
+       if (!cpu_has_pse)
+               use_pse = 0;
 
-       pgd_idx = pgd_index(PAGE_OFFSET);
+       pfn = start_pfn;
+       pgd_idx = pgd_index((pfn<<PAGE_SHIFT) + PAGE_OFFSET);
        pgd = pgd_base + pgd_idx;
-       pfn = 0;
-       pmd_idx = pmd_index(PAGE_OFFSET);
-       pte_ofs = pte_index(PAGE_OFFSET);
-
        for (; pgd_idx < PTRS_PER_PGD; pgd++, pgd_idx++) {
 #ifdef CONFIG_XEN
                /*
@@ -198,10 +227,16 @@ static void __init kernel_physical_mappi
 #else
                pmd = one_md_table_init(pgd);
 #endif
-               if (pfn >= max_low_pfn)
+
+               if (pfn >= end_pfn)
                        continue;
+#ifdef CONFIG_X86_PAE
+               pmd_idx = pmd_index((pfn<<PAGE_SHIFT) + PAGE_OFFSET);
                pmd += pmd_idx;
-               for (; pmd_idx < PTRS_PER_PMD && pfn < max_low_pfn;
+#else
+               pmd_idx = 0;
+#endif
+               for (; pmd_idx < PTRS_PER_PMD && pfn < end_pfn;
                     pmd++, pmd_idx++) {
                        unsigned int addr = pfn * PAGE_SIZE + PAGE_OFFSET;
 
@@ -211,13 +246,8 @@ static void __init kernel_physical_mappi
                        /*
                         * Map with big pages if possible, otherwise
                         * create normal page tables:
-                        *
-                        * Don't use a large page for the first 2/4MB of memory
-                        * because there are often fixed size MTRRs in there
-                        * and overlapping MTRRs into large pages can cause
-                        * slowdowns.
                         */
-                       if (cpu_has_pse && !(pgd_idx == 0 && pmd_idx == 0)) {
+                       if (use_pse) {
                                unsigned int addr2;
                                pgprot_t prot = PAGE_KERNEL_LARGE;
 
@@ -228,49 +258,35 @@ static void __init kernel_physical_mappi
                                    is_kernel_text(addr2))
                                        prot = PAGE_KERNEL_LARGE_EXEC;
 
+                               pages_2m++;
                                set_pmd(pmd, pfn_pmd(pfn, prot));
 
                                pfn += PTRS_PER_PTE;
-                               max_pfn_mapped = pfn;
                                continue;
                        }
                        pte = one_page_table_init(pmd);
 
-                       for (pte += pte_ofs;
-                            pte_ofs < PTRS_PER_PTE && pfn < max_low_pfn;
+                       pte_ofs = pte_index((pfn<<PAGE_SHIFT) + PAGE_OFFSET);
+                       pte += pte_ofs;
+                       for (; pte_ofs < PTRS_PER_PTE && pfn < end_pfn;
                             pte++, pfn++, pte_ofs++, addr += PAGE_SIZE) {
                                pgprot_t prot = PAGE_KERNEL;
 
                                /* XEN: Only map initial RAM allocation. */
-                               if ((pfn >= max_ram_pfn) || pte_present(*pte))
+                               if (pfn >= xen_start_info->nr_pages || pte_present(*pte))
                                        continue;
                                if (is_kernel_text(addr))
                                        prot = PAGE_KERNEL_EXEC;
 
+                               pages_4k++;
                                set_pte(pte, pfn_pte(pfn, prot));
                        }
-                       max_pfn_mapped = pfn;
-                       pte_ofs = 0;
                }
-               pmd_idx = 0;
        }
+       update_page_count(PG_LEVEL_2M, pages_2m);
+       update_page_count(PG_LEVEL_4K, pages_4k);
 }
 
-#ifndef CONFIG_XEN
-
-static inline int page_kills_ppro(unsigned long pagenr)
-{
-       if (pagenr >= 0x70000 && pagenr <= 0x7003F)
-               return 1;
-       return 0;
-}
-
-#else
-
-#define page_kills_ppro(p)     0
-
-#endif
-
 /*
  * devmem_is_allowed() checks to see if /dev/mem access to a certain address
  * is valid. The argument is a physical page number.
@@ -331,30 +347,63 @@ static void __init permanent_kmaps_init(
        pkmap_page_table = pte;
 }
 
-void __init add_one_highpage_init(struct page *page, int pfn, int bad_ppro)
+static void __init add_one_highpage_init(struct page *page, int pfn)
 {
-       if (page_is_ram(pfn) && !(bad_ppro && page_kills_ppro(pfn))) {
-               ClearPageReserved(page);
-               init_page_count(page);
-               if (pfn < xen_start_info->nr_pages)
-                       __free_page(page);
-               totalhigh_pages++;
-       } else
-               SetPageReserved(page);
+       ClearPageReserved(page);
+       init_page_count(page);
+       if (pfn < xen_start_info->nr_pages)
+               __free_page(page);
+       totalhigh_pages++;
+}
+
+struct add_highpages_data {
+       unsigned long start_pfn;
+       unsigned long end_pfn;
+};
+
+static int __init add_highpages_work_fn(unsigned long start_pfn,
+                                        unsigned long end_pfn, void *datax)
+{
+       int node_pfn;
+       struct page *page;
+       unsigned long final_start_pfn, final_end_pfn;
+       struct add_highpages_data *data;
+
+       data = (struct add_highpages_data *)datax;
+
+       final_start_pfn = max(start_pfn, data->start_pfn);
+       final_end_pfn = min(end_pfn, data->end_pfn);
+       if (final_start_pfn >= final_end_pfn)
+               return 0;
+
+       for (node_pfn = final_start_pfn; node_pfn < final_end_pfn;
+            node_pfn++) {
+               if (!pfn_valid(node_pfn))
+                       continue;
+               page = pfn_to_page(node_pfn);
+               add_one_highpage_init(page, node_pfn);
+       }
+
+       return 0;
+
+}
+
+void __init add_highpages_with_active_regions(int nid, unsigned long start_pfn,
+                                             unsigned long end_pfn)
+{
+       struct add_highpages_data data;
+
+       data.start_pfn = start_pfn;
+       data.end_pfn = end_pfn;
+
+       work_with_active_regions(nid, add_highpages_work_fn, &data);
 }
 
 #ifndef CONFIG_NUMA
-static void __init set_highmem_pages_init(int bad_ppro)
+static void __init set_highmem_pages_init(void)
 {
-       int pfn;
+       add_highpages_with_active_regions(0, highstart_pfn, highend_pfn);
 
-       for (pfn = highstart_pfn; pfn < highend_pfn; pfn++) {
-               /*
-                * Holes under sparsemem might not have no mem_map[]:
-                */
-               if (pfn_valid(pfn))
-                       add_one_highpage_init(pfn_to_page(pfn), pfn, bad_ppro);
-       }
        totalram_pages += totalhigh_pages;
 }
 #endif /* !CONFIG_NUMA */
@@ -362,24 +411,11 @@ static void __init set_highmem_pages_ini
 #else
 # define kmap_init()                           do { } while (0)
 # define permanent_kmaps_init(pgd_base)                do { } while (0)
-# define set_highmem_pages_init(bad_ppro)      do { } while (0)
+# define set_highmem_pages_init()      do { } while (0)
 #endif /* CONFIG_HIGHMEM */
 
-pteval_t __PAGE_KERNEL = _PAGE_KERNEL;
-EXPORT_SYMBOL(__PAGE_KERNEL);
-
-pteval_t __PAGE_KERNEL_EXEC = _PAGE_KERNEL_EXEC;
-
 pgd_t *swapper_pg_dir;
 
-static void __init xen_pagetable_setup_start(pgd_t *base)
-{
-}
-
-static void __init xen_pagetable_setup_done(pgd_t *base)
-{
-}
-
 /*
  * Build a proper pagetable for the kernel mappings.  Up until this
  * point, we've been running on some set of pagetables constructed by
@@ -399,27 +435,10 @@ static void __init xen_pagetable_setup_d
  * be partially populated, and so it avoids stomping on any existing
  * mappings.
  */
-static void __init pagetable_init(void)
+static void __init early_ioremap_page_table_range_init(pgd_t *pgd_base)
 {
-       pgd_t *pgd_base = (pgd_t *)xen_start_info->pt_base;
        unsigned long vaddr, end;
 
-       xen_pagetable_setup_start(pgd_base);
-
-       /* Enable PSE if available */
-       if (cpu_has_pse)
-               set_in_cr4(X86_CR4_PSE);
-
-       /* Enable PGE if available */
-       if (cpu_has_pge) {
-               set_in_cr4(X86_CR4_PGE);
-               __PAGE_KERNEL |= _PAGE_GLOBAL;
-               __PAGE_KERNEL_EXEC |= _PAGE_GLOBAL;
-       }
-
-       kernel_physical_mapping_init(pgd_base);
-       remap_numa_kva();
-
        /*
         * Fixed mappings, only the page table structure has to be
         * created - mappings will be set by set_fixmap():
@@ -429,10 +448,13 @@ static void __init pagetable_init(void)
        end = (FIXADDR_TOP + PMD_SIZE - 1) & PMD_MASK;
        page_table_range_init(vaddr, end, pgd_base);
        early_ioremap_reset();
+}
 
-       permanent_kmaps_init(pgd_base);
+static void __init pagetable_init(void)
+{
+       pgd_t *pgd_base = (pgd_t *)xen_start_info->pt_base;
 
-       xen_pagetable_setup_done(pgd_base);
+       permanent_kmaps_init(pgd_base);
 }
 
 #if defined(CONFIG_ACPI_SLEEP) && !defined(CONFIG_XEN)
@@ -475,7 +497,7 @@ void zap_low_mappings(void)
 
 int nx_enabled;
 
-pteval_t __supported_pte_mask __read_mostly = ~_PAGE_NX;
+pteval_t __supported_pte_mask __read_mostly = ~(_PAGE_NX | _PAGE_GLOBAL);
 EXPORT_SYMBOL_GPL(__supported_pte_mask);
 
 #ifdef CONFIG_X86_PAE
@@ -528,42 +550,369 @@ static void __init set_nx(void)
 }
 #endif
 
+/* user-defined highmem size */
+static unsigned int highmem_pages = -1;
+
 /*
- * paging_init() sets up the page tables - note that the first 8MB are
- * already mapped by head.S.
- *
- * This routines also unmaps the page at virtual kernel address 0, so
- * that we can trap those pesky NULL-reference errors in the kernel.
+ * highmem=size forces highmem to be exactly 'size' bytes.
+ * This works even on boxes that have no highmem otherwise.
+ * This also works to reduce highmem size on bigger boxes.
  */
-void __init paging_init(void)
+static int __init parse_highmem(char *arg)
+{
+       if (!arg)
+               return -EINVAL;
+
+       highmem_pages = memparse(arg, &arg) >> PAGE_SHIFT;
+       return 0;
+}
+early_param("highmem", parse_highmem);
+
+/*
+ * Determine low and high memory ranges:
+ */
+void __init find_low_pfn_range(void)
+{
+       /* it could update max_pfn */
+
+       /* max_low_pfn is 0, we already have early_res support */
+
+       max_low_pfn = max_pfn;
+       if (max_low_pfn > MAXMEM_PFN) {
+               if (highmem_pages == -1)
+                       highmem_pages = max_pfn - MAXMEM_PFN;
+               if (highmem_pages + MAXMEM_PFN < max_pfn)
+                       max_pfn = MAXMEM_PFN + highmem_pages;
+               if (highmem_pages + MAXMEM_PFN > max_pfn) {
+                       printk(KERN_WARNING "only %luMB highmem pages "
+                               "available, ignoring highmem size of %uMB.\n",
+                               pages_to_mb(max_pfn - MAXMEM_PFN),
+                               pages_to_mb(highmem_pages));
+                       highmem_pages = 0;
+               }
+               max_low_pfn = MAXMEM_PFN;
+#ifndef CONFIG_HIGHMEM
+               /* Maximum memory usable is what is directly addressable */
+               printk(KERN_WARNING "Warning only %ldMB will be used.\n",
+                                       MAXMEM>>20);
+               if (max_pfn > MAX_NONPAE_PFN)
+                       printk(KERN_WARNING
+                                "Use a HIGHMEM64G enabled kernel.\n");
+               else
+                       printk(KERN_WARNING "Use a HIGHMEM enabled kernel.\n");
+               max_pfn = MAXMEM_PFN;
+#else /* !CONFIG_HIGHMEM */
+#ifndef CONFIG_HIGHMEM64G
+               if (max_pfn > MAX_NONPAE_PFN) {
+                       max_pfn = MAX_NONPAE_PFN;
+                       printk(KERN_WARNING "Warning only 4GB will be used."
+                               "Use a HIGHMEM64G enabled kernel.\n");
+               }
+#endif /* !CONFIG_HIGHMEM64G */
+#endif /* !CONFIG_HIGHMEM */
+       } else {
+               if (highmem_pages == -1)
+                       highmem_pages = 0;
+#ifdef CONFIG_HIGHMEM
+               if (highmem_pages >= max_pfn) {
+                       printk(KERN_ERR "highmem size specified (%uMB) is "
+                               "bigger than pages available (%luMB)!.\n",
+                               pages_to_mb(highmem_pages),
+                               pages_to_mb(max_pfn));
+                       highmem_pages = 0;
+               }
+               if (highmem_pages) {
+                       if (max_low_pfn - highmem_pages <
+                           64*1024*1024/PAGE_SIZE){
+                               printk(KERN_ERR "highmem size %uMB results in "
+                               "smaller than 64MB lowmem, ignoring it.\n"
+                                       , pages_to_mb(highmem_pages));
+                               highmem_pages = 0;
+                       }
+                       max_low_pfn -= highmem_pages;
+               }
+#else
+               if (highmem_pages)
+                       printk(KERN_ERR "ignoring highmem size on non-highmem"
+                                       " kernel!\n");
+#endif
+       }
+}
+
+#ifndef CONFIG_NEED_MULTIPLE_NODES
+void __init initmem_init(unsigned long start_pfn,
+                                 unsigned long end_pfn)
+{
+#ifdef CONFIG_HIGHMEM
+       highstart_pfn = highend_pfn = max_pfn;
+       if (max_pfn > max_low_pfn)
+               highstart_pfn = max_low_pfn;
+       memory_present(0, 0, highend_pfn);
+       e820_register_active_regions(0, 0, highend_pfn);
+       printk(KERN_NOTICE "%ldMB HIGHMEM available.\n",
+               pages_to_mb(highend_pfn - highstart_pfn));
+       num_physpages = highend_pfn;
+       high_memory = (void *) __va(highstart_pfn * PAGE_SIZE - 1) + 1;
+#else
+       memory_present(0, 0, max_low_pfn);
+       e820_register_active_regions(0, 0, max_low_pfn);
+       num_physpages = max_low_pfn;
+       high_memory = (void *) __va(max_low_pfn * PAGE_SIZE - 1) + 1;
+#endif
+#ifdef CONFIG_FLATMEM
+       max_mapnr = num_physpages;
+#endif
+       printk(KERN_NOTICE "%ldMB LOWMEM available.\n",
+                       pages_to_mb(max_low_pfn));
+
+       setup_bootmem_allocator();
+}
+#endif /* !CONFIG_NEED_MULTIPLE_NODES */
+
+static void __init zone_sizes_init(void)
+{
+       unsigned long max_zone_pfns[MAX_NR_ZONES];
+       memset(max_zone_pfns, 0, sizeof(max_zone_pfns));
+       max_zone_pfns[ZONE_DMA] =
+               virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT;
+       max_zone_pfns[ZONE_NORMAL] = max_low_pfn;
+#ifdef CONFIG_HIGHMEM
+       max_zone_pfns[ZONE_HIGHMEM] = highend_pfn;
+#endif
+
+       free_area_init_nodes(max_zone_pfns);
+}
+
+void __init setup_bootmem_allocator(void)
 {
        int i;
+       unsigned long bootmap_size, bootmap;
+       unsigned long end_pfn = min(max_low_pfn, xen_start_info->nr_pages);
+
+       /*
+        * Initialize the boot-time allocator (with low memory only):
+        */
+       bootmap_size = bootmem_bootmap_pages(end_pfn)<<PAGE_SHIFT;
+       bootmap = find_e820_area(min_low_pfn<<PAGE_SHIFT,
+                                max_pfn_mapped<<PAGE_SHIFT, bootmap_size,
+                                PAGE_SIZE);
+       if (bootmap == -1L)
+               panic("Cannot find bootmem map of size %ld\n", bootmap_size);
+       reserve_early(bootmap, bootmap + bootmap_size, "BOOTMAP");
+
+       /* don't touch min_low_pfn */
+       bootmap_size = init_bootmem_node(NODE_DATA(0), bootmap >> PAGE_SHIFT,
+                                        min_low_pfn, end_pfn);
+       printk(KERN_INFO "  mapped low ram: 0 - %08lx\n",
+                max_pfn_mapped<<PAGE_SHIFT);
+       printk(KERN_INFO "  low ram: %08lx - %08lx\n",
+                min_low_pfn<<PAGE_SHIFT, max_low_pfn<<PAGE_SHIFT);
+       printk(KERN_INFO "  bootmap %08lx - %08lx\n",
+                bootmap, bootmap + bootmap_size);
+       for_each_online_node(i)
+               free_bootmem_with_active_regions(i, end_pfn);
+       early_res_to_bootmem(0, end_pfn<<PAGE_SHIFT);
+
+       after_init_bootmem = 1;
+}
+
+static unsigned long __init extend_init_mapping(unsigned long tables_space)
+{
+       unsigned long start_pfn = (__pa(xen_start_info->pt_base) >> PAGE_SHIFT)
+                                 + xen_start_info->nr_pt_frames;
+       unsigned long start = start_pfn, va;
+       pgd_t *pgd;
+       pud_t *pud;
+       pmd_t *pmd;
+       pte_t *pte;
+
+       /* Kill mapping of low 1MB. */
+       for (va = PAGE_OFFSET; va < (unsigned long)&_text; va += PAGE_SIZE)
+               if (HYPERVISOR_update_va_mapping(va, __pte_ma(0), 0))
+                       BUG();
+
+       /* Ensure init mappings cover kernel text/data and initial tables. */
+       while (va < PAGE_OFFSET + (start_pfn << PAGE_SHIFT) + tables_space) {
+               pgd = pgd_offset_k(va);
+               pud = pud_offset(pgd, va);
+               pmd = pmd_offset(pud, va);
+               if (pmd_none(*pmd)) {
+                       unsigned long pa = start_pfn++ << PAGE_SHIFT;
+
+                       memset(__va(pa), 0, PAGE_SIZE);
+                       make_lowmem_page_readonly(__va(pa),
+                                                 XENFEAT_writable_page_tables);
+                       xen_l2_entry_update(pmd, __pmd(pa | _KERNPG_TABLE));
+               }
+               pte = pte_offset_kernel(pmd, va);
+               if (pte_none(*pte)) {
+                       pte_t new_pte = __pte(__pa(va) | _KERNPG_TABLE);
+
+                       if (HYPERVISOR_update_va_mapping(va, new_pte, 0))
+                               BUG();
+               }
+               va += PAGE_SIZE;
+       }
+
+       /* Finally, blow away any spurious initial mappings. */
+       while (1) {
+               pgd = pgd_offset_k(va);
+               pud = pud_offset(pgd, va);
+               pmd = pmd_offset(pud, va);
+               if (pmd_none(*pmd))
+                       break;
+               if (HYPERVISOR_update_va_mapping(va, __pte_ma(0), 0))
+                       BUG();
+               va += PAGE_SIZE;
+       }
+
+       if (start_pfn > start)
+               reserve_early(start << PAGE_SHIFT,
+                             start_pfn << PAGE_SHIFT, "INITMAP");
+
+       return start_pfn;
+}
+
+static void __init find_early_table_space(unsigned long end)
+{
+       unsigned long puds, pmds, ptes, tables;
+
+       puds = (end + PUD_SIZE - 1) >> PUD_SHIFT;
+       tables = PAGE_ALIGN(puds * sizeof(pud_t));
+
+       pmds = (end + PMD_SIZE - 1) >> PMD_SHIFT;
+       tables += PAGE_ALIGN(pmds * sizeof(pmd_t));
+
+       if (cpu_has_pse) {
+               unsigned long extra;
+
+               extra = end - ((end>>PMD_SHIFT) << PMD_SHIFT);
+               extra += PMD_SIZE;
+               ptes = (extra + PAGE_SIZE - 1) >> PAGE_SHIFT;
+       } else
+               ptes = (end + PAGE_SIZE - 1) >> PAGE_SHIFT;
+
+       tables += PAGE_ALIGN(ptes * sizeof(pte_t));
+
+       /* for fixmap */
+       tables += PAGE_SIZE
+                 * ((((FIXADDR_TOP + PMD_SIZE - 1) & PMD_MASK)
+                     - (__fix_to_virt(__end_of_fixed_addresses - 1) & PMD_MASK))
+                    >> PMD_SHIFT);
+
+       table_start = extend_init_mapping(tables);
+
+       table_end = table_start;
+       table_top = table_start + (tables>>PAGE_SHIFT);
+
+       printk(KERN_DEBUG "kernel direct mapping tables up to %lx @ %lx-%lx\n",
+               end, table_start << PAGE_SHIFT,
+               (table_start << PAGE_SHIFT) + tables);
+}
+
+unsigned long __init_refok init_memory_mapping(unsigned long start,
+                                               unsigned long end)
+{
+       pgd_t *pgd_base = swapper_pg_dir;
+       unsigned long start_pfn, end_pfn;
+       unsigned long big_page_start;
+
+       /*
+        * Find space for the kernel direct mapping tables.
+        */
+       if (!after_init_bootmem)
+               find_early_table_space(end);
 
 #ifdef CONFIG_X86_PAE
        set_nx();
        if (nx_enabled)
                printk(KERN_INFO "NX (Execute Disable) protection: active\n");
 #endif
+
+       /* Enable PSE if available */
+       if (cpu_has_pse)
+               set_in_cr4(X86_CR4_PSE);
+
+       /* Enable PGE if available */
+       if (cpu_has_pge) {
+               set_in_cr4(X86_CR4_PGE);
+               __supported_pte_mask |= _PAGE_GLOBAL;
+       }
+
+       /*
+        * Don't use a large page for the first 2/4MB of memory
+        * because there are often fixed size MTRRs in there
+        * and overlapping MTRRs into large pages can cause
+        * slowdowns.
+        */
+       big_page_start = PMD_SIZE;
+
+       if (start < big_page_start) {
+               start_pfn = start >> PAGE_SHIFT;
+               end_pfn = min(big_page_start>>PAGE_SHIFT, end>>PAGE_SHIFT);
+       } else {
+               /* head is not big page alignment ? */
+               start_pfn = start >> PAGE_SHIFT;
+               end_pfn = ((start + (PMD_SIZE - 1))>>PMD_SHIFT)
+                                << (PMD_SHIFT - PAGE_SHIFT);
+       }
+       if (start_pfn < end_pfn)
+               kernel_physical_mapping_init(pgd_base, start_pfn, end_pfn, 0);
+
+       /* big page range */
+       start_pfn = ((start + (PMD_SIZE - 1))>>PMD_SHIFT)
+                        << (PMD_SHIFT - PAGE_SHIFT);
+       if (start_pfn < (big_page_start >> PAGE_SHIFT))
+               start_pfn =  big_page_start >> PAGE_SHIFT;
+       end_pfn = (end>>PMD_SHIFT) << (PMD_SHIFT - PAGE_SHIFT);
+       if (start_pfn < end_pfn)
+               kernel_physical_mapping_init(pgd_base, start_pfn, end_pfn,
+                                               cpu_has_pse);
+
+       /* tail is not big page alignment ? */
+       start_pfn = end_pfn;
+       if (start_pfn > (big_page_start>>PAGE_SHIFT)) {
+               end_pfn = end >> PAGE_SHIFT;
+               if (start_pfn < end_pfn)
+                       kernel_physical_mapping_init(pgd_base, start_pfn,
+                                                        end_pfn, 0);
+       }
+
+       early_ioremap_page_table_range_init(pgd_base);
+
+       __flush_tlb_all();
+
+       if (!after_init_bootmem)
+               reserve_early(table_start << PAGE_SHIFT,
+                             table_end << PAGE_SHIFT, "PGTABLE");
+
+       if (!after_init_bootmem)
+               early_memtest(start, end);
+
+       return end >> PAGE_SHIFT;
+}
+
+
+/*
+ * paging_init() sets up the page tables - note that the first 8MB are
+ * already mapped by head.S.
+ *
+ * This routines also unmaps the page at virtual kernel address 0, so
+ * that we can trap those pesky NULL-reference errors in the kernel.
+ */
+void __init paging_init(void)
+{
        pagetable_init();
 
        __flush_tlb_all();
 
        kmap_init();
 
-       /* Switch to the real shared_info page, and clear the
-        * dummy page. */
-       set_fixmap(FIX_SHARED_INFO, xen_start_info->shared_info);
-       HYPERVISOR_shared_info = (shared_info_t *)fix_to_virt(FIX_SHARED_INFO);
-       memset(empty_zero_page, 0, sizeof(empty_zero_page));
-
-       /* Setup mapping of lower 1st MB */
-       for (i = 0; i < NR_FIX_ISAMAPS; i++)
-               if (is_initial_xendomain())
-                       set_fixmap(FIX_ISAMAP_BEGIN - i, i * PAGE_SIZE);
-               else
-                       __set_fixmap(FIX_ISAMAP_BEGIN - i,
-                                    virt_to_machine(empty_zero_page),
-                                    PAGE_KERNEL_RO);
+       /*
+        * NOTE: at this point the bootmem allocator is fully available.
+        */
+       sparse_init();
+       zone_sizes_init();
 }
 
 /*
@@ -598,7 +947,7 @@ static struct kcore_list kcore_mem, kcor
 void __init mem_init(void)
 {
        int codesize, reservedpages, datasize, initsize;
-       int tmp, bad_ppro;
+       int tmp;
        unsigned long pfn;
 
        pci_iommu_alloc();
@@ -606,19 +955,6 @@ void __init mem_init(void)
 #ifdef CONFIG_FLATMEM
        BUG_ON(!mem_map);
 #endif
-       bad_ppro = ppro_with_ram_bug();
-
-#ifdef CONFIG_HIGHMEM
-       /* check that fixmap and pkmap do not overlap */
-       if (PKMAP_BASE + LAST_PKMAP*PAGE_SIZE >= FIXADDR_START) {
-               printk(KERN_ERR
-                       "fixmap and kmap areas overlap - this will crash\n");
-               printk(KERN_ERR "pkstart: %lxh pkend: %lxh fixstart %lxh\n",
-                               PKMAP_BASE, PKMAP_BASE + LAST_PKMAP*PAGE_SIZE,
-                               FIXADDR_START);
-               BUG();
-       }
-#endif
        /* this will put all low memory onto the freelists */
        totalram_pages += free_all_bootmem();
        /* XEN: init and count low-mem pages outside initial allocation. */
@@ -636,7 +972,7 @@ void __init mem_init(void)
                if (page_is_ram(tmp) && PageReserved(pfn_to_page(tmp)))
                        reservedpages++;
 
-       set_highmem_pages_init(bad_ppro);
+       set_highmem_pages_init();
 
        codesize =  (unsigned long) &_etext - (unsigned long) &_text;
        datasize =  (unsigned long) &_edata - (unsigned long) &_etext;
@@ -657,7 +993,6 @@ void __init mem_init(void)
                (unsigned long) (totalhigh_pages << (PAGE_SHIFT-10))
               );
 
-#if 1 /* double-sanity-check paranoia */
        printk(KERN_INFO "virtual kernel memory layout:\n"
                "    fixmap  : 0x%08lx - 0x%08lx   (%4ld kB)\n"
 #ifdef CONFIG_HIGHMEM
@@ -698,7 +1033,6 @@ void __init mem_init(void)
 #endif
        BUG_ON(VMALLOC_START                            > VMALLOC_END);
        BUG_ON((unsigned long)high_memory               > VMALLOC_START);
-#endif /* double-sanity-check paranoia */
 
        if (boot_cpu_data.wp_works_ok < 0)
                test_wp_bit();
@@ -755,6 +1089,8 @@ void mark_rodata_ro(void)
        unsigned long start = PFN_ALIGN(_text);
        unsigned long size = PFN_ALIGN(_etext) - start;
 
+#ifndef CONFIG_DYNAMIC_FTRACE
+       /* Dynamic tracing modifies the kernel text section */
        set_pages_ro(virt_to_page(start), size >> PAGE_SHIFT);
        printk(KERN_INFO "Write protecting the kernel text: %luk\n",
                size >> 10);
@@ -767,6 +1103,8 @@ void mark_rodata_ro(void)
        printk(KERN_INFO "Testing CPA: write protecting again\n");
        set_pages_ro(virt_to_page(start), size>>PAGE_SHIFT);
 #endif
+#endif /* CONFIG_DYNAMIC_FTRACE */
+
        start += size;
        size = (unsigned long)__end_rodata - start;
        set_pages_ro(virt_to_page(start), size >> PAGE_SHIFT);
@@ -829,3 +1167,9 @@ void free_initrd_mem(unsigned long start
        free_init_pages("initrd memory", start, end);
 }
 #endif
+
+int __init reserve_bootmem_generic(unsigned long phys, unsigned long len,
+                                  int flags)
+{
+       return reserve_bootmem(phys, len, flags);
+}
Index: head-2008-12-01/arch/x86/mm/init_64-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/mm/init_64-xen.c      2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/mm/init_64-xen.c   2008-12-01 11:49:07.000000000 +0100
@@ -21,6 +21,7 @@
 #include <linux/swap.h>
 #include <linux/smp.h>
 #include <linux/init.h>
+#include <linux/initrd.h>
 #include <linux/pagemap.h>
 #include <linux/bootmem.h>
 #include <linux/proc_fs.h>
@@ -52,6 +53,14 @@
 
 #include <xen/features.h>
 
+/*
+ * end_pfn only includes RAM, while max_pfn_mapped includes all e820 entries.
+ * The direct mapping extends to max_pfn_mapped, so that we can directly access
+ * apertures, ACPI and other tables without having to play with fixmaps.
+ */
+unsigned long max_low_pfn_mapped;
+unsigned long max_pfn_mapped;
+
 #if CONFIG_XEN_COMPAT <= 0x030002
 unsigned int __kernel_page_user;
 EXPORT_SYMBOL(__kernel_page_user);
@@ -60,12 +69,11 @@ EXPORT_SYMBOL(__kernel_page_user);
 int after_bootmem;
 
 DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
-extern unsigned long start_pfn;
 
 extern pmd_t level2_fixmap_pgt[PTRS_PER_PMD];
 extern pte_t level1_fixmap_pgt[PTRS_PER_PTE];
 
-int direct_gbpages __meminitdata
+int direct_gbpages
 #ifdef CONFIG_DIRECT_GBPAGES
                                = 1
 #endif
@@ -145,55 +153,23 @@ void __meminit early_make_page_readonly(
  * around without checking the pgd every time.
  */
 
-void show_mem(void)
-{
-       long i, total = 0, reserved = 0;
-       long shared = 0, cached = 0;
-       struct page *page;
-       pg_data_t *pgdat;
-
-       printk(KERN_INFO "Mem-info:\n");
-       show_free_areas();
-       for_each_online_pgdat(pgdat) {
-               for (i = 0; i < pgdat->node_spanned_pages; ++i) {
-                       /*
-                        * This loop can take a while with 256 GB and
-                        * 4k pages so defer the NMI watchdog:
-                        */
-                       if (unlikely(i % MAX_ORDER_NR_PAGES == 0))
-                               touch_nmi_watchdog();
-
-                       if (!pfn_valid(pgdat->node_start_pfn + i))
-                               continue;
-
-                       page = pfn_to_page(pgdat->node_start_pfn + i);
-                       total++;
-                       if (PageReserved(page))
-                               reserved++;
-                       else if (PageSwapCache(page))
-                               cached++;
-                       else if (page_count(page))
-                               shared += page_count(page) - 1;
-               }
-       }
-       printk(KERN_INFO "%lu pages of RAM\n",          total);
-       printk(KERN_INFO "%lu reserved pages\n",        reserved);
-       printk(KERN_INFO "%lu pages shared\n",          shared);
-       printk(KERN_INFO "%lu pages swap cached\n",     cached);
-}
-
 static unsigned long __meminitdata table_start;
+static unsigned long __meminitdata table_cur;
 static unsigned long __meminitdata table_end;
 
-static __init void *spp_getpage(void)
+/*
+ * NOTE: This function is marked __ref because it calls __init function
+ * (alloc_bootmem_pages). It's safe to do it ONLY when after_bootmem == 0.
+ */
+static __ref void *spp_getpage(void)
 {
        void *ptr;
 
        if (after_bootmem)
                ptr = (void *) get_zeroed_page(GFP_ATOMIC);
-       else if (start_pfn < table_end) {
-               ptr = __va(start_pfn << PAGE_SHIFT);
-               start_pfn++;
+       else if (table_cur < table_end) {
+               ptr = __va(table_cur << PAGE_SHIFT);
+               table_cur++;
                memset(ptr, 0, PAGE_SIZE);
        } else
                ptr = alloc_bootmem_pages(PAGE_SIZE);
@@ -208,30 +184,18 @@ static __init void *spp_getpage(void)
        return ptr;
 }
 
-#define pgd_offset_u(address) (__user_pgd(init_level4_pgt) + pgd_index(address))
-#define pud_offset_u(address) (level3_user_pgt + pud_index(address))
-
-static __init void
-set_pte_phys(unsigned long vaddr, unsigned long phys, pgprot_t prot, int user_mode)
+void
+set_pte_vaddr_pud(pud_t *pud_page, unsigned long vaddr, pte_t new_pte)
 {
-       pgd_t *pgd;
        pud_t *pud;
        pmd_t *pmd;
-       pte_t *pte, new_pte;
-
-       pr_debug("set_pte_phys %lx to %lx\n", vaddr, phys);
+       pte_t *pte;
 
-       pgd = (user_mode ? pgd_offset_u(vaddr) : pgd_offset_k(vaddr));
-       if (pgd_none(*pgd)) {
-               printk(KERN_ERR
-                       "PGD FIXMAP MISSING, it should be setup in head.S!\n");
-               return;
-       }
-       pud = (user_mode ? pud_offset_u(vaddr) : pud_offset(pgd, vaddr));
+       pud = pud_page + pud_index(vaddr);
        if (pud_none(*pud)) {
                pmd = (pmd_t *) spp_getpage();
                make_page_readonly(pmd, XENFEAT_writable_page_tables);
-               set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE | _PAGE_USER));
+               pud_populate(&init_mm, pud, pmd);
                if (pmd != pmd_offset(pud, 0)) {
                        printk(KERN_ERR "PAGETABLE BUG #01! %p <-> %p\n",
                                pmd, pmd_offset(pud, 0));
@@ -242,19 +206,20 @@ set_pte_phys(unsigned long vaddr, unsign
        if (pmd_none(*pmd)) {
                pte = (pte_t *) spp_getpage();
                make_page_readonly(pte, XENFEAT_writable_page_tables);
-               set_pmd(pmd, __pmd(__pa(pte) | _KERNPG_TABLE | _PAGE_USER));
+               pmd_populate_kernel(&init_mm, pmd, pte);
                if (pte != pte_offset_kernel(pmd, 0)) {
                        printk(KERN_ERR "PAGETABLE BUG #02!\n");
                        return;
                }
        }
-       if (pgprot_val(prot))
-               new_pte = pfn_pte(phys >> PAGE_SHIFT, prot);
-       else
-               new_pte = __pte(0);
 
        pte = pte_offset_kernel(pmd, vaddr);
        if (!pte_none(*pte) && __pte_val(new_pte) &&
+#ifdef CONFIG_ACPI
+           /* __acpi_map_table() fails to properly call clear_fixmap() */
+           (vaddr < __fix_to_virt(FIX_ACPI_END) ||
+            vaddr > __fix_to_virt(FIX_ACPI_BEGIN)) &&
+#endif
            __pte_val(*pte) != (__pte_val(new_pte) & __supported_pte_mask))
                pte_ERROR(*pte);
        set_pte(pte, new_pte);
@@ -266,15 +231,13 @@ set_pte_phys(unsigned long vaddr, unsign
        __flush_tlb_one(vaddr);
 }
 
-static __init void
-set_pte_phys_ma(unsigned long vaddr, unsigned long phys, pgprot_t prot)
+void
+set_pte_vaddr(unsigned long vaddr, pte_t pteval)
 {
        pgd_t *pgd;
-       pud_t *pud;
-       pmd_t *pmd;
-       pte_t *pte, new_pte;
+       pud_t *pud_page;
 
-       pr_debug("set_pte_phys_ma %lx to %lx\n", vaddr, phys);
+       pr_debug("set_pte_vaddr %lx to %lx\n", vaddr, __pte_val(pteval));
 
        pgd = pgd_offset_k(vaddr);
        if (pgd_none(*pgd)) {
@@ -282,47 +245,51 @@ set_pte_phys_ma(unsigned long vaddr, uns
                        "PGD FIXMAP MISSING, it should be setup in head.S!\n");
                return;
        }
-       pud = pud_offset(pgd, vaddr);
-       if (pud_none(*pud)) {
-               pmd = (pmd_t *) spp_getpage();
-               make_page_readonly(pmd, XENFEAT_writable_page_tables);
-               set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE | _PAGE_USER));
-               if (pmd != pmd_offset(pud, 0)) {
-                       printk(KERN_ERR "PAGETABLE BUG #01! %p <-> %p\n",
-                               pmd, pmd_offset(pud, 0));
+       pud_page = (pud_t*)pgd_page_vaddr(*pgd);
+       set_pte_vaddr_pud(pud_page, vaddr, pteval);
+}
+
+#ifndef CONFIG_XEN
+/*
+ * Create large page table mappings for a range of physical addresses.
+ */
+static void __init __init_extra_mapping(unsigned long phys, unsigned long size,
+                                               pgprot_t prot)
+{
+       pgd_t *pgd;
+       pud_t *pud;
+       pmd_t *pmd;
+
+       BUG_ON((phys & ~PMD_MASK) || (size & ~PMD_MASK));
+       for (; size; phys += PMD_SIZE, size -= PMD_SIZE) {
+               pgd = pgd_offset_k((unsigned long)__va(phys));
+               if (pgd_none(*pgd)) {
+                       pud = (pud_t *) spp_getpage();
+                       set_pgd(pgd, __pgd(__pa(pud) | _KERNPG_TABLE |
+                                               _PAGE_USER));
                }
-       }
-       pmd = pmd_offset(pud, vaddr);
-       if (pmd_none(*pmd)) {
-               pte = (pte_t *) spp_getpage();
-               make_page_readonly(pte, XENFEAT_writable_page_tables);
-               set_pmd(pmd, __pmd(__pa(pte) | _KERNPG_TABLE | _PAGE_USER));
-               if (pte != pte_offset_kernel(pmd, 0)) {
-                       printk(KERN_ERR "PAGETABLE BUG #02!\n");
-                       return;
+               pud = pud_offset(pgd, (unsigned long)__va(phys));
+               if (pud_none(*pud)) {
+                       pmd = (pmd_t *) spp_getpage();
+                       set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE |
+                                               _PAGE_USER));
                }
+               pmd = pmd_offset(pud, phys);
+               BUG_ON(!pmd_none(*pmd));
+               set_pmd(pmd, __pmd(phys | pgprot_val(prot)));
        }
-       new_pte = pfn_pte_ma(phys >> PAGE_SHIFT, prot);
+}
 
-       pte = pte_offset_kernel(pmd, vaddr);
-       if (!pte_none(*pte) && __pte_val(new_pte) &&
-#ifdef CONFIG_ACPI
-           /* __acpi_map_table() fails to properly call clear_fixmap() */
-           (vaddr < __fix_to_virt(FIX_ACPI_END) ||
-            vaddr > __fix_to_virt(FIX_ACPI_BEGIN)) &&
-#endif
-           __pte_val(*pte) != (__pte_val(new_pte) & __supported_pte_mask))
-               pte_ERROR(*pte);
-       set_pte(pte, new_pte);
+void __init init_extra_mapping_wb(unsigned long phys, unsigned long size)
+{
+       __init_extra_mapping(phys, size, PAGE_KERNEL_LARGE);
+}
 
-       /*
-        * It's enough to flush this one mapping.
-        * (PGE mappings get flushed as well)
-        */
-       __flush_tlb_one(vaddr);
+void __init init_extra_mapping_uc(unsigned long phys, unsigned long size)
+{
+       __init_extra_mapping(phys, size, PAGE_KERNEL_LARGE_NOCACHE);
 }
 
-#ifndef CONFIG_XEN
 /*
  * The head.S code sets up the kernel high mapping:
  *
@@ -352,33 +319,9 @@ void __init cleanup_highmap(void)
 }
 #endif
 
-/* NOTE: this is meant to be run only at boot */
-void __init __set_fixmap(enum fixed_addresses idx, unsigned long phys, pgprot_t prot)
-{
-       unsigned long address = __fix_to_virt(idx);
-
-       if (idx >= __end_of_fixed_addresses) {
-               printk(KERN_ERR "Invalid __set_fixmap\n");
-               return;
-       }
-       switch (idx) {
-       case VSYSCALL_LAST_PAGE ... VSYSCALL_FIRST_PAGE:
-               set_pte_phys(address, phys, prot, 0);
-               set_pte_phys(address, phys, prot, 1);
-               break;
-       case FIX_EARLYCON_MEM_BASE:
-               xen_l1_entry_update(level1_fixmap_pgt + pte_index(address),
-                                   pfn_pte_ma(phys >> PAGE_SHIFT, prot));
-               break;
-       default:
-               set_pte_phys_ma(address, phys, prot);
-               break;
-       }
-}
-
 static __meminit void *alloc_static_page(unsigned long *phys)
 {
-       unsigned long va = (start_pfn << PAGE_SHIFT) + __START_KERNEL_map;
+       unsigned long va = (table_cur << PAGE_SHIFT) + __START_KERNEL_map;
 
        if (after_bootmem) {
                void *adr = (void *)get_zeroed_page(GFP_ATOMIC);
@@ -387,13 +330,12 @@ static __meminit void *alloc_static_page
                return adr;
        }
 
-       *phys = start_pfn << PAGE_SHIFT;
-       start_pfn++;
-       memset((void *)va, 0, PAGE_SIZE);
-       return (void *)va;
+       BUG_ON(!table_cur);
+       *phys = table_cur++ << PAGE_SHIFT;
+       return memset((void *)va, 0, PAGE_SIZE);
 }
 
-#define PTE_SIZE PAGE_SIZE
+#define unmap_low_page(p) ((void)(p))
 
 static inline int __meminit make_readonly(unsigned long paddr)
 {
@@ -408,7 +350,7 @@ static inline int __meminit make_readonl
        /* Make old page tables read-only. */
        if (!xen_feature(XENFEAT_writable_page_tables)
            && (paddr >= (xen_start_info->pt_base - __START_KERNEL_map))
-           && (paddr < (start_pfn << PAGE_SHIFT)))
+           && (paddr < (table_cur << PAGE_SHIFT)))
                readonly = 1;
 
        /*
@@ -425,118 +367,129 @@ static inline int __meminit make_readonl
        return readonly;
 }
 
-#ifndef CONFIG_XEN
-/* Must run before zap_low_mappings */
-__meminit void *early_ioremap(unsigned long addr, unsigned long size)
+static unsigned long __meminit
+phys_pte_init(pte_t *pte_page, unsigned long addr, unsigned long end)
 {
-       pmd_t *pmd, *last_pmd;
-       unsigned long vaddr;
-       int i, pmds;
-
-       pmds = ((addr & ~PMD_MASK) + size + ~PMD_MASK) / PMD_SIZE;
-       vaddr = __START_KERNEL_map;
-       pmd = level2_kernel_pgt;
-       last_pmd = level2_kernel_pgt + PTRS_PER_PMD - 1;
-
-       for (; pmd <= last_pmd; pmd++, vaddr += PMD_SIZE) {
-               for (i = 0; i < pmds; i++) {
-                       if (pmd_present(pmd[i]))
-                               goto continue_outer_loop;
-               }
-               vaddr += addr & ~PMD_MASK;
-               addr &= PMD_MASK;
+       unsigned pages = 0;
+       unsigned long last_map_addr = end;
+       int i;
 
-               for (i = 0; i < pmds; i++, addr += PMD_SIZE)
-                       set_pmd(pmd+i, __pmd(addr | __PAGE_KERNEL_LARGE_EXEC));
-               __flush_tlb_all();
-
-               return (void *)vaddr;
-continue_outer_loop:
-               ;
+       pte_t *pte = pte_page + pte_index(addr);
+
+       for(i = pte_index(addr); i < PTRS_PER_PTE; i++, addr += PAGE_SIZE, pte++) {
+               unsigned long pteval = addr | __PAGE_KERNEL;
+
+               if (addr >= (after_bootmem
+                            ? end
+                            : xen_start_info->nr_pages << PAGE_SHIFT))
+                       break;
+
+               if (__pte_val(*pte))
+                       continue;
+
+               if (make_readonly(addr))
+                       pteval &= ~_PAGE_RW;
+               if (0)
+                       printk("   pte=%p addr=%lx pte=%016lx\n",
+                              pte, addr, pteval);
+               if (!after_bootmem)
+                       *pte = __pte(pteval & __supported_pte_mask);
+               else
+                       set_pte(pte, __pte(pteval & __supported_pte_mask));
+               last_map_addr = (addr & PAGE_MASK) + PAGE_SIZE;
+               pages++;
        }
-       printk("early_ioremap(0x%lx, %lu) failed\n", addr, size);
-       return NULL;
+       update_page_count(PG_LEVEL_4K, pages);
+
+       return last_map_addr;
 }
 
-/*
- * To avoid virtual aliases later:
- */
-__meminit void early_iounmap(void *addr, unsigned long size)
+static unsigned long __meminit
+phys_pte_update(pmd_t *pmd, unsigned long address, unsigned long end)
 {
-       unsigned long vaddr;
-       pmd_t *pmd;
-       int i, pmds;
-
-       vaddr = (unsigned long)addr;
-       pmds = ((vaddr & ~PMD_MASK) + size + ~PMD_MASK) / PMD_SIZE;
-       pmd = level2_kernel_pgt + pmd_index(vaddr);
+       pte_t *pte = (pte_t *)pmd_page_vaddr(*pmd);
 
-       for (i = 0; i < pmds; i++)
-               pmd_clear(pmd + i);
-
-       __flush_tlb_all();
+       BUG_ON(!after_bootmem);
+       return phys_pte_init(pte, address, end);
 }
-#endif
 
 static unsigned long __meminit
-phys_pmd_init(pmd_t *pmd_page, unsigned long address, unsigned long end)
+phys_pmd_init(pmd_t *pmd_page, unsigned long address, unsigned long end,
+                        unsigned long page_size_mask)
 {
+       unsigned long pages = 0;
+       unsigned long last_map_addr = end;
+       unsigned long start = address;
+
        int i = pmd_index(address);
 
-       for (; i < PTRS_PER_PMD; i++) {
+       for (; i < PTRS_PER_PMD; i++, address += PMD_SIZE) {
                unsigned long pte_phys;
-               pmd_t *pmd = pmd_page + i;
-               pte_t *pte, *pte_save;
-               int k;
+               pmd_t *pmd = pmd_page + pmd_index(address);
+               pte_t *pte;
 
                if (address >= end)
                        break;
 
                if (__pmd_val(*pmd)) {
-                       address += PMD_SIZE;
+                       if (!pmd_large(*pmd)) {
+                               spin_lock(&init_mm.page_table_lock);
+                               last_map_addr = phys_pte_update(pmd, address,
+                                                               end);
+                               spin_unlock(&init_mm.page_table_lock);
+                       }
+                       /* Count entries we're using from level2_ident_pgt */
+                       if (start == 0)
+                               pages++;
                        continue;
                }
 
-               pte = alloc_static_page(&pte_phys);
-               pte_save = pte;
-               for (k = 0; k < PTRS_PER_PTE; pte++, k++, address += PTE_SIZE) {
-                       unsigned long pteval = address | _PAGE_NX | _KERNPG_TABLE;
-
-                       if (address >= (after_bootmem
-                                       ? end
-                                       : xen_start_info->nr_pages << PAGE_SHIFT))
-                               pteval = 0;
-                       else if (make_readonly(address))
-                               pteval &= ~_PAGE_RW;
-                       set_pte(pte, __pte(pteval & __supported_pte_mask));
+               if (page_size_mask & (1<<PG_LEVEL_2M)) {
+                       pages++;
+                       spin_lock(&init_mm.page_table_lock);
+                       set_pte((pte_t *)pmd,
+                               pfn_pte(address >> PAGE_SHIFT, PAGE_KERNEL_LARGE));
+                       spin_unlock(&init_mm.page_table_lock);
+                       last_map_addr = (address & PMD_MASK) + PMD_SIZE;
+                       continue;
                }
+
+               pte = alloc_static_page(&pte_phys);
+               last_map_addr = phys_pte_init(pte, address, end);
+               unmap_low_page(pte);
+
                if (!after_bootmem) {
-                       early_make_page_readonly(pte_save, XENFEAT_writable_page_tables);
-                       *pmd = __pmd(pte_phys | _KERNPG_TABLE);
+                       early_make_page_readonly(pte, XENFEAT_writable_page_tables);
+                       *pmd = __pmd(pte_phys | _PAGE_TABLE);
                } else {
-                       make_page_readonly(pte_save, XENFEAT_writable_page_tables);
-                       set_pmd(pmd, __pmd(pte_phys | _KERNPG_TABLE));
+                       make_page_readonly(pte, XENFEAT_writable_page_tables);
+                       spin_lock(&init_mm.page_table_lock);
+                       pmd_populate_kernel(&init_mm, pmd, __va(pte_phys));
+                       spin_unlock(&init_mm.page_table_lock);
                }
        }
-       return address;
+       update_page_count(PG_LEVEL_2M, pages);
+       return last_map_addr;
 }
 
 static unsigned long __meminit
-phys_pmd_update(pud_t *pud, unsigned long address, unsigned long end)
+phys_pmd_update(pud_t *pud, unsigned long address, unsigned long end,
+                        unsigned long page_size_mask)
 {
        pmd_t *pmd = pmd_offset(pud, 0);
        unsigned long last_map_addr;
 
-       spin_lock(&init_mm.page_table_lock);
-       last_map_addr = phys_pmd_init(pmd, address, end);
-       spin_unlock(&init_mm.page_table_lock);
+       BUG_ON(!after_bootmem);
+       last_map_addr = phys_pmd_init(pmd, address, end, page_size_mask);
        __flush_tlb_all();
        return last_map_addr;
 }
 
 static unsigned long __meminit
-phys_pud_init(pud_t *pud_page, unsigned long addr, unsigned long end)
+phys_pud_init(pud_t *pud_page, unsigned long addr, unsigned long end,
+                        unsigned long page_size_mask)
 {
+       unsigned long pages = 0;
        unsigned long last_map_addr = end;
        int i = pud_index(addr);
 
@@ -550,29 +503,59 @@ phys_pud_init(pud_t *pud_page, unsigned 
 
                if (__pud_val(*pud)) {
                        if (!pud_large(*pud))
-                               last_map_addr = phys_pmd_update(pud, addr, end);
+                               last_map_addr = phys_pmd_update(pud, addr, end,
+                                                        page_size_mask);
                        continue;
                }
 
-               if (direct_gbpages) {
+               if (page_size_mask & (1<<PG_LEVEL_1G)) {
+                       pages++;
+                       spin_lock(&init_mm.page_table_lock);
                        set_pte((pte_t *)pud,
                                pfn_pte(addr >> PAGE_SHIFT, PAGE_KERNEL_LARGE));
+                       spin_unlock(&init_mm.page_table_lock);
                        last_map_addr = (addr & PUD_MASK) + PUD_SIZE;
                        continue;
                }
 
                pmd = alloc_static_page(&pmd_phys);
+               last_map_addr = phys_pmd_init(pmd, addr, end, page_size_mask);
+               unmap_low_page(pmd);
 
-               spin_lock(&init_mm.page_table_lock);
-               *pud = __pud(pmd_phys | _KERNPG_TABLE);
-               last_map_addr = phys_pmd_init(pmd, addr, end);
-               spin_unlock(&init_mm.page_table_lock);
-
-               early_make_page_readonly(pmd, XENFEAT_writable_page_tables);
+               if (!after_bootmem) {
+                       early_make_page_readonly(pmd, XENFEAT_writable_page_tables);
+                       if (page_size_mask & (1 << PG_LEVEL_NUM))
+                               xen_l3_entry_update(pud, __pud(pmd_phys | _PAGE_TABLE));
+                       else
+                               *pud = __pud(pmd_phys | _PAGE_TABLE);
+               } else {
+                       make_page_readonly(pmd, XENFEAT_writable_page_tables);
+                       spin_lock(&init_mm.page_table_lock);
+                       pud_populate(&init_mm, pud, __va(pmd_phys));
+                       spin_unlock(&init_mm.page_table_lock);
+               }
        }
        __flush_tlb_all();
+       update_page_count(PG_LEVEL_1G, pages);
 
-       return last_map_addr >> PAGE_SHIFT;
+       return last_map_addr;
+}
+
+static unsigned long __meminit
+phys_pud_update(pgd_t *pgd, unsigned long addr, unsigned long end,
+                unsigned long page_size_mask)
+{
+       pud_t *pud;
+
+       if (!after_bootmem) {
+               unsigned long addr = __pgd_val(*pgd), *page;
+
+               addr_to_page(addr, page);
+               pud = (pud_t *)page;
+       } else
+               pud = (pud_t *)pgd_page_vaddr(*pgd);
+
+       return phys_pud_init(pud, addr, end, page_size_mask | (1 << PG_LEVEL_NUM));
 }
 
 void __init xen_init_pt(void)
@@ -654,7 +637,7 @@ void __init xen_init_pt(void)
 static void __init extend_init_mapping(unsigned long tables_space)
 {
        unsigned long va = __START_KERNEL_map;
-       unsigned long start = start_pfn;
+       unsigned long start = table_cur;
        unsigned long phys, addr, *pte_page;
        pmd_t *pmd;
        pte_t *pte, new_pte;
@@ -674,7 +657,7 @@ static void __init extend_init_mapping(u
 
        /* Ensure init mappings cover kernel text/data and initial tables. */
        while (va < (__START_KERNEL_map
-                    + (start_pfn << PAGE_SHIFT)
+                    + (table_cur << PAGE_SHIFT)
                     + tables_space)) {
                pmd = (pmd_t *)&page[pmd_index(va)];
                if (pmd_none(*pmd)) {
@@ -706,9 +689,9 @@ static void __init extend_init_mapping(u
                va += PAGE_SIZE;
        }
 
-       if (start_pfn > start)
+       if (table_cur > start)
                reserve_early(start << PAGE_SHIFT,
-                             start_pfn << PAGE_SHIFT, "INITMAP");
+                             table_cur << PAGE_SHIFT, "INITMAP");
 }
 
 static void __init find_early_table_space(unsigned long end)
@@ -717,23 +700,25 @@ static void __init find_early_table_spac
 
        puds = (end + PUD_SIZE - 1) >> PUD_SHIFT;
        pmds = (end + PMD_SIZE - 1) >> PMD_SHIFT;
-       ptes = (end + PTE_SIZE - 1) >> PAGE_SHIFT;
+       ptes = (end + PAGE_SIZE - 1) >> PAGE_SHIFT;
 
        tables = round_up(puds * 8, PAGE_SIZE) + 
                round_up(pmds * 8, PAGE_SIZE) + 
                round_up(ptes * 8, PAGE_SIZE); 
 
+       table_cur = (__pa(xen_start_info->pt_base) >> PAGE_SHIFT) +
+               xen_start_info->nr_pt_frames;
+
        extend_init_mapping(tables);
 
-       table_start = start_pfn;
+       table_start = table_cur;
        table_end = table_start + (tables>>PAGE_SHIFT);
 
-       early_printk("kernel direct mapping tables up to %lx @ %lx-%lx\n",
-               end, table_start << PAGE_SHIFT,
-               (table_start << PAGE_SHIFT) + tables);
+       printk(KERN_DEBUG "kernel direct mapping tables up to %lx @ %lx-%lx\n",
+               end, table_start << PAGE_SHIFT, table_end << PAGE_SHIFT);
 }
 
-static void __init xen_finish_init_mapping(void)
+static void __init xen_finish_init_mapping(bool reserve)
 {
        unsigned long i, start, end;
 
@@ -762,7 +747,8 @@ static void __init xen_finish_init_mappi
                if (HYPERVISOR_update_va_mapping(start, __pte_ma(0), 0))
                        BUG();
 
-       /* Allocate pte's for initial fixmaps from 'start_pfn' allocator. */
+       /* Allocate pte's for initial fixmaps from 'table_cur' allocator. */
+       start = table_cur;
        table_end = ~0UL;
 
        /*
@@ -789,8 +775,11 @@ static void __init xen_finish_init_mappi
                                     << PAGE_SHIFT,
                                     PAGE_KERNEL_RO);
 
-       /* Disable the 'start_pfn' allocator. */
-       table_end = start_pfn;
+       /* Disable the 'table_cur' allocator. */
+       table_end = table_cur;
+       if (reserve && table_cur > start)
+               reserve_early(start << PAGE_SHIFT,
+                             table_cur << PAGE_SHIFT, "FIXMAP");
 }
 
 static void __init init_gbpages(void)
@@ -801,126 +790,89 @@ static void __init init_gbpages(void)
                direct_gbpages = 0;
 }
 
-#ifdef CONFIG_MEMTEST_BOOTPARAM
-
-static void __init memtest(unsigned long start_phys, unsigned long size,
-                                unsigned pattern)
+static unsigned long __init kernel_physical_mapping_init(unsigned long start,
+                                               unsigned long end,
+                                               unsigned long page_size_mask)
 {
-       unsigned long i;
-       unsigned long *start;
-       unsigned long start_bad;
-       unsigned long last_bad;
-       unsigned long val;
-       unsigned long start_phys_aligned;
-       unsigned long count;
-       unsigned long incr;
-
-       switch (pattern) {
-       case 0:
-               val = 0UL;
-               break;
-       case 1:
-               val = -1UL;
-               break;
-       case 2:
-               val = 0x5555555555555555UL;
-               break;
-       case 3:
-               val = 0xaaaaaaaaaaaaaaaaUL;
-               break;
-       default:
-               return;
-       }
 
-       incr = sizeof(unsigned long);
-       start_phys_aligned = ALIGN(start_phys, incr);
-       count = (size - (start_phys_aligned - start_phys))/incr;
-       start = __va(start_phys_aligned);
-       start_bad = 0;
-       last_bad = 0;
-
-       for (i = 0; i < count; i++)
-               start[i] = val;
-       for (i = 0; i < count; i++, start++, start_phys_aligned += incr) {
-               if (*start != val) {
-                       if (start_phys_aligned == last_bad + incr) {
-                               last_bad += incr;
-                       } else {
-                               if (start_bad) {
-                                       printk(KERN_CONT "\n  %016lx bad mem addr %016lx - %016lx reserved",
-                                               val, start_bad, last_bad + incr);
-                                       reserve_early(start_bad, last_bad - start_bad, "BAD RAM");
-                               }
-                               start_bad = last_bad = start_phys_aligned;
-                       }
-               }
-       }
-       if (start_bad) {
-               printk(KERN_CONT "\n  %016lx bad mem addr %016lx - %016lx reserved",
-                       val, start_bad, last_bad + incr);
-               reserve_early(start_bad, last_bad - start_bad, "BAD RAM");
-       }
-
-}
-
-static int memtest_pattern __initdata = CONFIG_MEMTEST_BOOTPARAM_VALUE;
-
-static int __init parse_memtest(char *arg)
-{
-       if (arg)
-               memtest_pattern = simple_strtoul(arg, NULL, 0);
-       return 0;
-}
+       unsigned long next, last_map_addr = end;
 
-early_param("memtest", parse_memtest);
+       start = (unsigned long)__va(start);
+       end = (unsigned long)__va(end);
 
-static void __init early_memtest(unsigned long start, unsigned long end)
-{
-       u64 t_start, t_size;
-       unsigned pattern;
+       for (; start < end; start = next) {
+               pgd_t *pgd = pgd_offset_k(start);
+               unsigned long pud_phys;
+               pud_t *pud;
 
-       if (!memtest_pattern)
-               return;
+               next = (start + PGDIR_SIZE) & PGDIR_MASK;
+               if (next > end)
+                       next = end;
 
-       printk(KERN_INFO "early_memtest: pattern num %d", memtest_pattern);
-       for (pattern = 0; pattern < memtest_pattern; pattern++) {
-               t_start = start;
-               t_size = 0;
-               while (t_start < end) {
-                       t_start = find_e820_area_size(t_start, &t_size, 1);
-
-                       /* done ? */
-                       if (t_start >= end)
-                               break;
-                       if (t_start + t_size > end)
-                               t_size = end - t_start;
-
-                       printk(KERN_CONT "\n  %016llx - %016llx pattern %d",
-                               (unsigned long long)t_start,
-                               (unsigned long long)t_start + t_size, pattern);
+               if (__pgd_val(*pgd)) {
+                       last_map_addr = phys_pud_update(pgd, __pa(start),
+                                                __pa(end), page_size_mask);
+                       continue;
+               }
 
-                       memtest(t_start, t_size, pattern);
+               pud = alloc_static_page(&pud_phys);
+               last_map_addr = phys_pud_init(pud, __pa(start), __pa(next),
+                                                page_size_mask);
+               unmap_low_page(pud);
 
-                       t_start += t_size;
+               if(!after_bootmem) {
+                       early_make_page_readonly(pud, XENFEAT_writable_page_tables);
+                       xen_l4_entry_update(pgd, __pgd(pud_phys | _PAGE_TABLE));
+               } else {
+                       make_page_readonly(pud, XENFEAT_writable_page_tables);
+                       spin_lock(&init_mm.page_table_lock);
+                       pgd_populate(&init_mm, pgd, __va(pud_phys));
+                       spin_unlock(&init_mm.page_table_lock);
                }
        }
-       printk(KERN_CONT "\n");
+
+       return last_map_addr;
 }
-#else
-static void __init early_memtest(unsigned long start, unsigned long end)
+
+struct map_range {
+       unsigned long start;
+       unsigned long end;
+       unsigned page_size_mask;
+};
+
+#define NR_RANGE_MR 5
+
+static int save_mr(struct map_range *mr, int nr_range,
+                  unsigned long start_pfn, unsigned long end_pfn,
+                  unsigned long page_size_mask)
 {
+
+       if (start_pfn < end_pfn) {
+               if (nr_range >= NR_RANGE_MR)
+                       panic("run out of range for init_memory_mapping\n");
+               mr[nr_range].start = start_pfn<<PAGE_SHIFT;
+               mr[nr_range].end   = end_pfn<<PAGE_SHIFT;
+               mr[nr_range].page_size_mask = page_size_mask;
+               nr_range++;
+       }
+
+       return nr_range;
 }
-#endif
 
 /*
  * Setup the direct mapping of the physical memory at PAGE_OFFSET.
  * This runs before bootmem is initialized and gets pages directly from
  * the physical memory. To access them they are temporarily mapped.
  */
-unsigned long __init_refok init_memory_mapping(unsigned long start, unsigned long end)
+unsigned long __init_refok init_memory_mapping(unsigned long start,
+                                              unsigned long end)
 {
-       unsigned long next, last_map_addr = end;
-       unsigned long start_phys = start, end_phys = end;
+       unsigned long last_map_addr = 0;
+       unsigned long page_size_mask = 0;
+       unsigned long start_pfn, end_pfn;
+       bool first = !table_start;
+       struct map_range mr[NR_RANGE_MR];
+       int nr_range, i;
 
        printk(KERN_INFO "init_memory_mapping\n");
 
@@ -931,51 +883,123 @@ unsigned long __init_refok init_memory_m
         * memory mapped. Unfortunately this is done currently before the
         * nodes are discovered.
         */
-       if (!after_bootmem) {
+       if (!after_bootmem)
                init_gbpages();
-               find_early_table_space(end);
-       }
 
-       start = (unsigned long)__va(start);
-       end = (unsigned long)__va(end);
+       if (direct_gbpages)
+               page_size_mask |= 1 << PG_LEVEL_1G;
+       if (cpu_has_pse)
+               page_size_mask |= 1 << PG_LEVEL_2M;
+
+       memset(mr, 0, sizeof(mr));
+       nr_range = 0;
+
+       /* head if not big page alignment ?*/
+       start_pfn = start >> PAGE_SHIFT;
+       end_pfn = ((start + (PMD_SIZE - 1)) >> PMD_SHIFT)
+                       << (PMD_SHIFT - PAGE_SHIFT);
+       nr_range = save_mr(mr, nr_range, start_pfn, end_pfn, 0);
+
+       /* big page (2M) range*/
+       start_pfn = ((start + (PMD_SIZE - 1))>>PMD_SHIFT)
+                        << (PMD_SHIFT - PAGE_SHIFT);
+       end_pfn = ((start + (PUD_SIZE - 1))>>PUD_SHIFT)
+                        << (PUD_SHIFT - PAGE_SHIFT);
+       if (end_pfn > ((end>>PUD_SHIFT)<<(PUD_SHIFT - PAGE_SHIFT)))
+               end_pfn = ((end>>PUD_SHIFT)<<(PUD_SHIFT - PAGE_SHIFT));
+       nr_range = save_mr(mr, nr_range, start_pfn, end_pfn,
+                       page_size_mask & (1<<PG_LEVEL_2M));
+
+       /* big page (1G) range */
+       start_pfn = end_pfn;
+       end_pfn = (end>>PUD_SHIFT) << (PUD_SHIFT - PAGE_SHIFT);
+       nr_range = save_mr(mr, nr_range, start_pfn, end_pfn,
+                               page_size_mask &
+                                ((1<<PG_LEVEL_2M)|(1<<PG_LEVEL_1G)));
+
+       /* tail is not big page (1G) alignment */
+       start_pfn = end_pfn;
+       end_pfn = (end>>PMD_SHIFT) << (PMD_SHIFT - PAGE_SHIFT);
+       nr_range = save_mr(mr, nr_range, start_pfn, end_pfn,
+                       page_size_mask & (1<<PG_LEVEL_2M));
+
+       /* tail is not big page (2M) alignment */
+       start_pfn = end_pfn;
+       end_pfn = end>>PAGE_SHIFT;
+       nr_range = save_mr(mr, nr_range, start_pfn, end_pfn, 0);
+
+       /* try to merge same page size and continuous */
+       for (i = 0; nr_range > 1 && i < nr_range - 1; i++) {
+               unsigned long old_start;
+               if (mr[i].end != mr[i+1].start ||
+                   mr[i].page_size_mask != mr[i+1].page_size_mask)
+                       continue;
+               /* move it */
+               old_start = mr[i].start;
+               memmove(&mr[i], &mr[i+1],
+                        (nr_range - 1 - i) * sizeof (struct map_range));
+               mr[i--].start = old_start;
+               nr_range--;
+       }
 
-       for (; start < end; start = next) {
-               pgd_t *pgd = pgd_offset_k(start);
-               unsigned long pud_phys;
-               pud_t *pud;
+       for (i = 0; i < nr_range; i++)
+               printk(KERN_DEBUG " %010lx - %010lx page %s\n",
+                               mr[i].start, mr[i].end,
+                       (mr[i].page_size_mask & (1<<PG_LEVEL_1G))?"1G":(
+                        (mr[i].page_size_mask & (1<<PG_LEVEL_2M))?"2M":"4k"));
 
-               if (after_bootmem)
-                       pud = pud_offset(pgd, start & PGDIR_MASK);
-               else
-                       pud = alloc_static_page(&pud_phys);
-               next = start + PGDIR_SIZE;
-               if (next > end)
-                       next = end;
-               last_map_addr = phys_pud_init(pud, __pa(start), __pa(next));
-               if (!after_bootmem) {
-                       early_make_page_readonly(pud, XENFEAT_writable_page_tables);
-                       set_pgd(pgd_offset_k(start), mk_kernel_pgd(pud_phys));
-               }
-       }
+       if (first)
+               find_early_table_space(end);
 
-       if (!after_bootmem) {
-               BUG_ON(start_pfn != table_end);
-               xen_finish_init_mapping();
+       for (i = 0; i < nr_range; i++)
+               last_map_addr = kernel_physical_mapping_init(
+                                       mr[i].start, mr[i].end,
+                                       mr[i].page_size_mask);
+
+       BUG_ON(table_cur > table_end);
+       if (start < (table_start << PAGE_SHIFT)) {
+               WARN_ON(table_cur != table_end);
+               xen_finish_init_mapping(!first);
        }
 
        __flush_tlb_all();
 
-       if (!after_bootmem)
+       if (first && table_end > table_start)
                reserve_early(table_start << PAGE_SHIFT,
                              table_end << PAGE_SHIFT, "PGTABLE");
 
+       printk(KERN_INFO "last_map_addr: %lx end: %lx\n",
+                        last_map_addr, end);
+
        if (!after_bootmem)
-               early_memtest(start_phys, end_phys);
+               early_memtest(start, end);
 
-       return last_map_addr;
+       return last_map_addr >> PAGE_SHIFT;
 }
 
 #ifndef CONFIG_NUMA
+void __init initmem_init(unsigned long start_pfn, unsigned long end_pfn)
+{
+       unsigned long bootmap_size, bootmap;
+
+       e820_register_active_regions(0, start_pfn, end_pfn);
+#ifdef CONFIG_XEN
+       if (end_pfn > xen_start_info->nr_pages)
+               end_pfn = xen_start_info->nr_pages;
+#endif
+       bootmap_size = bootmem_bootmap_pages(end_pfn)<<PAGE_SHIFT;
+       bootmap = find_e820_area(0, end_pfn<<PAGE_SHIFT, bootmap_size,
+                                PAGE_SIZE);
+       if (bootmap == -1L)
+               panic("Cannot find bootmem map of size %ld\n", bootmap_size);
+       /* don't touch min_low_pfn */
+       bootmap_size = init_bootmem_node(NODE_DATA(0), bootmap >> PAGE_SHIFT,
+                                        0, end_pfn);
+       free_bootmem_with_active_regions(0, end_pfn);
+       early_res_to_bootmem(0, end_pfn<<PAGE_SHIFT);
+       reserve_bootmem(bootmap, bootmap_size, BOOTMEM_DEFAULT);
+}
+
 void __init paging_init(void)
 {
        unsigned long max_zone_pfns[MAX_NR_ZONES];
@@ -983,9 +1007,9 @@ void __init paging_init(void)
        memset(max_zone_pfns, 0, sizeof(max_zone_pfns));
        max_zone_pfns[ZONE_DMA] = MAX_DMA_PFN;
        max_zone_pfns[ZONE_DMA32] = MAX_DMA32_PFN;
-       max_zone_pfns[ZONE_NORMAL] = end_pfn;
+       max_zone_pfns[ZONE_NORMAL] = max_pfn;
 
-       memory_present(0, 0, end_pfn);
+       memory_present(0, 0, max_pfn);
        sparse_init();
        free_area_init_nodes(max_zone_pfns);
 
@@ -1076,8 +1100,8 @@ void __init mem_init(void)
                init_page_count(pfn_to_page(pfn));
                totalram_pages++;
        }
-       reservedpages = end_pfn - totalram_pages -
-                                       absent_pages_in_range(0, end_pfn);
+       reservedpages = max_pfn - totalram_pages -
+                                       absent_pages_in_range(0, max_pfn);
        after_bootmem = 1;
 
        codesize =  (unsigned long) &_etext - (unsigned long) &_text;
@@ -1096,7 +1120,7 @@ void __init mem_init(void)
        printk(KERN_INFO "Memory: %luk/%luk available (%ldk kernel code, "
                                "%ldk reserved, %ldk data, %ldk init)\n",
                (unsigned long) nr_free_pages() << (PAGE_SHIFT-10),
-               end_pfn << (PAGE_SHIFT-10),
+               max_pfn << (PAGE_SHIFT-10),
                codesize >> 10,
                reservedpages << (PAGE_SHIFT-10),
                datasize >> 10,
@@ -1159,6 +1183,13 @@ EXPORT_SYMBOL_GPL(rodata_test_data);
 void mark_rodata_ro(void)
 {
        unsigned long start = PFN_ALIGN(_stext), end = PFN_ALIGN(__end_rodata);
+       unsigned long rodata_start =
+               ((unsigned long)__start_rodata + PAGE_SIZE - 1) & PAGE_MASK;
+
+#ifdef CONFIG_DYNAMIC_FTRACE
+       /* Dynamic tracing modifies the kernel text section */
+       start = rodata_start;
+#endif
 
        printk(KERN_INFO "Write protecting the kernel read-only data: %luk\n",
               (end - start) >> 10);
@@ -1168,8 +1199,7 @@ void mark_rodata_ro(void)
         * The rodata section (but not the kernel text!) should also be
         * not-executable.
         */
-       start = ((unsigned long)__start_rodata + PAGE_SIZE - 1) & PAGE_MASK;
-       set_memory_nx(start, (end - start) >> PAGE_SHIFT);
+       set_memory_nx(rodata_start, (end - rodata_start) >> PAGE_SHIFT);
 
        rodata_test();
 
@@ -1191,24 +1221,26 @@ void free_initrd_mem(unsigned long start
 }
 #endif
 
-void __init reserve_bootmem_generic(unsigned long phys, unsigned len)
+int __init reserve_bootmem_generic(unsigned long phys, unsigned long len,
+                                  int flags)
 {
 #ifdef CONFIG_NUMA
        int nid, next_nid;
+       int ret;
 #endif
        unsigned long pfn = phys >> PAGE_SHIFT;
 
-       if (pfn >= end_pfn) {
+       if (pfn >= max_pfn) {
                /*
                 * This can happen with kdump kernels when accessing
                 * firmware tables:
                 */
                if (pfn < max_pfn_mapped)
-                       return;
+                       return -EFAULT;
 
-               printk(KERN_ERR "reserve_bootmem: illegal reserve %lx %u\n",
+               printk(KERN_ERR "reserve_bootmem: illegal reserve %lx %lu\n",
                                phys, len);
-               return;
+               return -EFAULT;
        }
 
        /* Should check here against the e820 map to avoid double free */
@@ -1216,9 +1248,13 @@ void __init reserve_bootmem_generic(unsi
        nid = phys_to_nid(phys);
        next_nid = phys_to_nid(phys + len - 1);
        if (nid == next_nid)
-               reserve_bootmem_node(NODE_DATA(nid), phys, len, BOOTMEM_DEFAULT);
+               ret = reserve_bootmem_node(NODE_DATA(nid), phys, len, flags);
        else
-               reserve_bootmem(phys, len, BOOTMEM_DEFAULT);
+               ret = reserve_bootmem(phys, len, flags);
+
+       if (ret != 0)
+               return ret;
+
 #else
        reserve_bootmem(phys, len, BOOTMEM_DEFAULT);
 #endif
@@ -1231,6 +1267,8 @@ void __init reserve_bootmem_generic(unsi
                set_dma_reserve(dma_reserve);
        }
 #endif
+
+       return 0;
 }
 
 int kern_addr_valid(unsigned long addr)
@@ -1335,7 +1373,7 @@ vmemmap_populate(struct page *start_page
        pmd_t *pmd;
 
        for (; addr < end; addr = next) {
-               next = pmd_addr_end(addr, end);
+               void *p = NULL;
 
                pgd = vmemmap_pgd_populate(addr, node);
                if (!pgd)
@@ -1345,33 +1383,51 @@ vmemmap_populate(struct page *start_page
                if (!pud)
                        return -ENOMEM;
 
-               pmd = pmd_offset(pud, addr);
-               if (pmd_none(*pmd)) {
-                       pte_t entry;
-                       void *p;
+               if (!cpu_has_pse) {
+                       next = (addr + PAGE_SIZE) & PAGE_MASK;
+                       pmd = vmemmap_pmd_populate(pud, addr, node);
+
+                       if (!pmd)
+                               return -ENOMEM;
+
+                       p = vmemmap_pte_populate(pmd, addr, node);
 
-                       p = vmemmap_alloc_block(PMD_SIZE, node);
                        if (!p)
                                return -ENOMEM;
 
-                       entry = pfn_pte(__pa(p) >> PAGE_SHIFT,
-                                                       PAGE_KERNEL_LARGE);
-                       set_pmd(pmd, __pmd_ma(__pte_val(entry)));
-
-                       /* check to see if we have contiguous blocks */
-                       if (p_end != p || node_start != node) {
-                               if (p_start)
-                                       printk(KERN_DEBUG " [%lx-%lx] PMD -> [%p-%p] on node %d\n",
-                                               addr_start, addr_end-1, p_start, p_end-1, node_start);
-                               addr_start = addr;
-                               node_start = node;
-                               p_start = p;
-                       }
-                       addr_end = addr + PMD_SIZE;
-                       p_end = p + PMD_SIZE;
+                       addr_end = addr + PAGE_SIZE;
+                       p_end = p + PAGE_SIZE;
                } else {
-                       vmemmap_verify((pte_t *)pmd, node, addr, next);
+                       next = pmd_addr_end(addr, end);
+
+                       pmd = pmd_offset(pud, addr);
+                       if (pmd_none(*pmd)) {
+                               pte_t entry;
+
+                               p = vmemmap_alloc_block(PMD_SIZE, node);
+                               if (!p)
+                                       return -ENOMEM;
+
+                               entry = pfn_pte(__pa(p) >> PAGE_SHIFT,
+                                               PAGE_KERNEL_LARGE);
+                               set_pmd(pmd, __pmd_ma(__pte_val(entry)));
+
+                               /* check to see if we have contiguous blocks */
+                               if (p_end != p || node_start != node) {
+                                       if (p_start)
+                                               printk(KERN_DEBUG " [%lx-%lx] PMD -> [%p-%p] on node %d\n",
+                                                      addr_start, addr_end-1, p_start, p_end-1, node_start);
+                                       addr_start = addr;
+                                       node_start = node;
+                                       p_start = p;
+                               }
+
+                               addr_end = addr + PMD_SIZE;
+                               p_end = p + PMD_SIZE;
+                       } else
+                               vmemmap_verify((pte_t *)pmd, node, addr, next);
                }
+
        }
        return 0;
 }
Index: head-2008-12-01/arch/x86/mm/ioremap-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/mm/ioremap-xen.c      2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/mm/ioremap-xen.c   2008-12-01 11:49:07.000000000 +0100
@@ -13,6 +13,7 @@
 #include <linux/pfn.h>
 #include <linux/slab.h>
 #include <linux/vmalloc.h>
+#include <linux/mmiotrace.h>
 
 #include <asm/cacheflush.h>
 #include <asm/e820.h>
@@ -274,7 +275,8 @@ int ioremap_check_change_attr(unsigned l
        for (sz = rc = 0; sz < size && !rc; ++mfn, sz += PAGE_SIZE) {
                unsigned long pfn = mfn_to_local_pfn(mfn);
 
-               if (pfn >= max_pfn_mapped)
+               if (pfn >= max_low_pfn_mapped &&
+                   (pfn < (1UL<<(32 - PAGE_SHIFT)) || pfn >= max_pfn_mapped))
                        continue;
                rc = ioremap_change_attr((unsigned long)__va(pfn << PAGE_SHIFT),
                                         PAGE_SIZE, prot_val);
@@ -297,11 +299,14 @@ static void __iomem *__ioremap_caller(re
 {
        unsigned long mfn, offset, vaddr;
        resource_size_t last_addr;
+       const resource_size_t unaligned_phys_addr = phys_addr;
+       const unsigned long unaligned_size = size;
        struct vm_struct *area;
        unsigned long new_prot_val;
        pgprot_t prot;
        int retval;
        domid_t domid = DOMID_IO;
+       void __iomem *ret_addr;
 
        /* Don't allow wraparound or zero size */
        last_addr = phys_addr + size - 1;
@@ -318,7 +323,7 @@ static void __iomem *__ioremap_caller(re
        /*
         * Don't remap the low PCI/ISA area, it's always mapped..
         */
-       if (is_initial_xendomain() && last_addr < ISA_END_ADDRESS)
+       if (is_initial_xendomain() && is_ISA_range(phys_addr, last_addr))
                return (__force void __iomem *)isa_bus_to_virt((unsigned long)phys_addr);
 
        /*
@@ -342,7 +347,7 @@ static void __iomem *__ioremap_caller(re
        phys_addr &= PAGE_MASK;
        size = PAGE_ALIGN(last_addr+1) - phys_addr;
 
-       retval = reserve_memtype(phys_addr, phys_addr + size,
+       retval = reserve_memtype(phys_addr, (u64)phys_addr + size,
                                                prot_val, &new_prot_val);
        if (retval) {
                pr_debug("Warning: reserve_memtype returned %d\n", retval);
@@ -410,7 +415,10 @@ static void __iomem *__ioremap_caller(re
                return NULL;
        }
 
-       return (void __iomem *) (vaddr + offset);
+       ret_addr = (void __iomem *) (vaddr + offset);
+       mmiotrace_ioremap(unaligned_phys_addr, unaligned_size, ret_addr);
+
+       return ret_addr;
 }
 
 /**
@@ -438,7 +446,7 @@ void __iomem *ioremap_nocache(resource_s
 {
        /*
         * Ideally, this should be:
-        *      pat_wc_enabled ? _PAGE_CACHE_UC : _PAGE_CACHE_UC_MINUS;
+        *      pat_enabled ? _PAGE_CACHE_UC : _PAGE_CACHE_UC_MINUS;
         *
         * Till we fix all X drivers to use ioremap_wc(), we will use
         * UC MINUS.
@@ -462,7 +470,7 @@ EXPORT_SYMBOL(ioremap_nocache);
  */
 void __iomem *ioremap_wc(unsigned long phys_addr, unsigned long size)
 {
-       if (pat_wc_enabled)
+       if (pat_enabled)
                return __ioremap_caller(phys_addr, size, _PAGE_CACHE_WC,
                                        __builtin_return_address(0));
        else
@@ -502,6 +510,14 @@ static void __iomem *ioremap_default(res
 }
 #endif
 
+void __iomem *ioremap_prot(resource_size_t phys_addr, unsigned long size,
+                               unsigned long prot_val)
+{
+       return __ioremap_caller(phys_addr, size, (prot_val & _PAGE_CACHE_MASK),
+                               __builtin_return_address(0));
+}
+EXPORT_SYMBOL(ioremap_prot);
+
 /**
  * iounmap - Free a IO remapping
  * @addr: virtual address from ioremap_*
@@ -526,6 +542,8 @@ void iounmap(volatile void __iomem *addr
        addr = (volatile void __iomem *)
                (PAGE_MASK & (unsigned long __force)addr);
 
+       mmiotrace_iounmap(addr);
+
        /* Use the vm area unlocked, assuming the caller
           ensures there isn't another iounmap for the same address
           in parallel. Reuse of the virtual address is prevented by
@@ -533,7 +551,7 @@ void iounmap(volatile void __iomem *addr
           cpa takes care of the direct mappings. */
        read_lock(&vmlist_lock);
        for (p = vmlist; p; p = p->next) {
-               if (p->addr == addr)
+               if (p->addr == (void __force *)addr)
                        break;
        }
        read_unlock(&vmlist_lock);
@@ -547,7 +565,7 @@ void iounmap(volatile void __iomem *addr
        free_memtype(p->phys_addr, p->phys_addr + get_vm_area_size(p));
 
        /* Finally remove it */
-       o = remove_vm_area((void *)addr);
+       o = remove_vm_area((void __force *)addr);
        BUG_ON(p != o || o == NULL);
        kfree(p);
 }
@@ -567,7 +585,7 @@ void *xlate_dev_mem_ptr(unsigned long ph
        if (page_is_ram(start >> PAGE_SHIFT))
                return __va(phys);
 
-       addr = (void *)ioremap_default(start, PAGE_SIZE);
+       addr = (void __force *)ioremap_default(start, PAGE_SIZE);
        if (addr)
                addr = (void *)((unsigned long)addr | (phys & ~PAGE_MASK));
 
@@ -595,8 +613,7 @@ static int __init early_ioremap_debug_se
 early_param("early_ioremap_debug", early_ioremap_debug_setup);
 
 static __initdata int after_paging_init;
-static pte_t bm_pte[PAGE_SIZE/sizeof(pte_t)]
-               __section(.bss.page_aligned);
+static pte_t bm_pte[PAGE_SIZE/sizeof(pte_t)] __page_aligned_bss;
 
 #ifdef CONFIG_X86_32
 static inline pmd_t * __init early_ioremap_pmd(unsigned long addr)
@@ -693,10 +710,11 @@ static void __init __early_set_fixmap(en
                return;
        }
        pte = early_ioremap_pte(addr);
+
        if (pgprot_val(flags))
                set_pte(pte, pfn_pte_ma(phys >> PAGE_SHIFT, flags));
        else
-               pte_clear(NULL, addr, pte);
+               pte_clear(&init_mm, addr, pte);
        __flush_tlb_one(addr);
 }
 
@@ -724,13 +742,11 @@ static int __init check_early_ioremap_le
 {
        if (!early_ioremap_nested)
                return 0;
-
-       printk(KERN_WARNING
+       WARN(1, KERN_WARNING
               "Debug warning: early ioremap leak of %d areas detected.\n",
-              early_ioremap_nested);
+               early_ioremap_nested);
        printk(KERN_WARNING
-              "please boot with early_ioremap_debug and report the dmesg.\n");
-       WARN_ON(1);
+               "please boot with early_ioremap_debug and report the dmesg.\n");
 
        return 1;
 }
Index: head-2008-12-01/arch/x86/mm/pageattr-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/mm/pageattr-xen.c     2008-12-01 12:19:27.000000000 +0100
+++ head-2008-12-01/arch/x86/mm/pageattr-xen.c  2008-12-01 11:49:07.000000000 +0100
@@ -34,6 +34,47 @@ struct cpa_data {
        unsigned        force_split : 1;
 };
 
+#ifdef CONFIG_PROC_FS
+static unsigned long direct_pages_count[PG_LEVEL_NUM];
+
+void update_page_count(int level, unsigned long pages)
+{
+       unsigned long flags;
+
+       /* Protect against CPA */
+       spin_lock_irqsave(&pgd_lock, flags);
+       direct_pages_count[level] += pages;
+       spin_unlock_irqrestore(&pgd_lock, flags);
+}
+
+static void split_page_count(int level)
+{
+       direct_pages_count[level]--;
+       direct_pages_count[level - 1] += PTRS_PER_PTE;
+}
+
+int arch_report_meminfo(char *page)
+{
+       int n = sprintf(page, "DirectMap4k:  %8lu kB\n",
+                       direct_pages_count[PG_LEVEL_4K] << 2);
+#if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE)
+       n += sprintf(page + n, "DirectMap2M:  %8lu kB\n",
+                       direct_pages_count[PG_LEVEL_2M] << 11);
+#else
+       n += sprintf(page + n, "DirectMap4M:  %8lu kB\n",
+                       direct_pages_count[PG_LEVEL_2M] << 12);
+#endif
+#ifdef CONFIG_X86_64
+       if (direct_gbpages)
+               n += sprintf(page + n, "DirectMap1G:  %8lu kB\n",
+                       direct_pages_count[PG_LEVEL_1G] << 20);
+#endif
+       return n;
+}
+#else
+static inline void split_page_count(int level) { }
+#endif
+
 #ifdef CONFIG_X86_64
 
 static inline unsigned long highmap_start_pfn(void)
@@ -106,7 +147,7 @@ static void cpa_flush_all(unsigned long 
 {
        BUG_ON(irqs_disabled());
 
-       on_each_cpu(__cpa_flush_all, (void *) cache, 1, 1);
+       on_each_cpu(__cpa_flush_all, (void *) cache, 1);
 }
 
 static void __cpa_flush_range(void *arg)
@@ -127,7 +168,7 @@ static void cpa_flush_range(unsigned lon
        BUG_ON(irqs_disabled());
        WARN_ON(PAGE_ALIGN(start) != start);
 
-       on_each_cpu(__cpa_flush_range, NULL, 1, 1);
+       on_each_cpu(__cpa_flush_range, NULL, 1);
 
        if (!cache)
                return;
@@ -229,6 +270,7 @@ pte_t *lookup_address(unsigned long addr
 
        return pte_offset_kernel(pmd, address);
 }
+EXPORT_SYMBOL_GPL(lookup_address);
 
 /*
  * Set the new pmd in all the pgds we know about:
@@ -509,6 +551,16 @@ static int split_large_page(pte_t *kpte,
        }
 #endif
 
+       if (address >= (unsigned long)__va(0) &&
+               address < (unsigned long)__va(max_low_pfn_mapped << PAGE_SHIFT))
+               split_page_count(level);
+
+#ifdef CONFIG_X86_64
+       if (address >= (unsigned long)__va(1UL<<32) &&
+               address < (unsigned long)__va(max_pfn_mapped << PAGE_SHIFT))
+               split_page_count(level);
+#endif
+
        /*
         * Get the target mfn from the original entry:
         */
@@ -565,10 +617,9 @@ repeat:
        if (!__pte_val(old_pte)) {
                if (!primary)
                        return 0;
-               printk(KERN_WARNING "CPA: called for zero pte. "
+               WARN(1, KERN_WARNING "CPA: called for zero pte. "
                       "vaddr = %lx cpa->vaddr = %lx\n", address,
                       cpa->vaddr);
-               WARN_ON(1);
                return -EINVAL;
        }
 
@@ -633,15 +684,24 @@ static int cpa_process_alias(struct cpa_
        struct cpa_data alias_cpa;
        int ret = 0;
 
-       if (cpa->pfn > max_pfn_mapped)
+       if (cpa->pfn >= max_pfn_mapped)
                return 0;
 
+#ifdef CONFIG_X86_64
+       if (cpa->pfn >= max_low_pfn_mapped && cpa->pfn < (1UL<<(32-PAGE_SHIFT)))
+               return 0;
+#endif
        /*
         * No need to redo, when the primary call touched the direct
         * mapping already:
         */
-       if (!within(cpa->vaddr, PAGE_OFFSET,
-                   PAGE_OFFSET + (max_pfn_mapped << PAGE_SHIFT))) {
+       if (!(within(cpa->vaddr, PAGE_OFFSET,
+                   PAGE_OFFSET + (max_low_pfn_mapped << PAGE_SHIFT))
+#ifdef CONFIG_X86_64
+               || within(cpa->vaddr, PAGE_OFFSET + (1UL<<32),
+                   PAGE_OFFSET + (max_pfn_mapped << PAGE_SHIFT))
+#endif
+       )) {
 
                alias_cpa = *cpa;
                alias_cpa.vaddr = (unsigned long) __va(cpa->pfn << PAGE_SHIFT);
@@ -809,7 +869,7 @@ int set_memory_uc(unsigned long addr, in
        /*
         * for now UC MINUS. see comments in ioremap_nocache()
         */
-       if (reserve_memtype(addr, addr + numpages * PAGE_SIZE,
+       if (reserve_memtype(__pa(addr), __pa(addr) + numpages * PAGE_SIZE,
                            _PAGE_CACHE_UC_MINUS, NULL))
                return -EINVAL;
 
@@ -825,10 +885,10 @@ int _set_memory_wc(unsigned long addr, i
 
 int set_memory_wc(unsigned long addr, int numpages)
 {
-       if (!pat_wc_enabled)
+       if (!pat_enabled)
                return set_memory_uc(addr, numpages);
 
-       if (reserve_memtype(addr, addr + numpages * PAGE_SIZE,
+       if (reserve_memtype(__pa(addr), __pa(addr) + numpages * PAGE_SIZE,
                _PAGE_CACHE_WC, NULL))
                return -EINVAL;
 
@@ -844,7 +904,7 @@ int _set_memory_wb(unsigned long addr, i
 
 int set_memory_wb(unsigned long addr, int numpages)
 {
-       free_memtype(addr, addr + numpages * PAGE_SIZE);
+       free_memtype(__pa(addr), __pa(addr) + numpages * PAGE_SIZE);
 
        return _set_memory_wb(addr, numpages);
 }
Index: head-2008-12-01/arch/x86/mm/pat-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/mm/pat-xen.c  2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/mm/pat-xen.c       2008-12-01 11:49:07.000000000 +0100
@@ -12,6 +12,8 @@
 #include <linux/gfp.h>
 #include <linux/fs.h>
 #include <linux/bootmem.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
 
 #include <asm/msr.h>
 #include <asm/tlbflush.h>
@@ -26,11 +28,11 @@
 #include <asm/io.h>
 
 #ifdef CONFIG_X86_PAT
-int __read_mostly pat_wc_enabled = 1;
+int __read_mostly pat_enabled = 1;
 
 void __cpuinit pat_disable(char *reason)
 {
-       pat_wc_enabled = 0;
+       pat_enabled = 0;
        printk(KERN_INFO "%s\n", reason);
 }
 
@@ -42,6 +44,19 @@ static int __init nopat(char *str)
 early_param("nopat", nopat);
 #endif
 
+
+static int debug_enable;
+static int __init pat_debug_setup(char *str)
+{
+       debug_enable = 1;
+       return 0;
+}
+__setup("debugpat", pat_debug_setup);
+
+#define dprintk(fmt, arg...) \
+       do { if (debug_enable) printk(KERN_INFO fmt, ##arg); } while (0)
+
+
 static u64 __read_mostly boot_pat_state;
 
 enum {
@@ -53,24 +68,25 @@ enum {
        PAT_UC_MINUS = 7,       /* UC, but can be overriden by MTRR */
 };
 
-#define PAT(x,y)       ((u64)PAT_ ## y << ((x)*8))
+#define PAT(x, y)      ((u64)PAT_ ## y << ((x)*8))
 
 void pat_init(void)
 {
        u64 pat;
 
-       if (!pat_wc_enabled)
+       if (!pat_enabled)
                return;
 
        /* Paranoia check. */
-       if (!cpu_has_pat) {
-               printk(KERN_ERR "PAT enabled, but CPU feature cleared\n");
+       if (!cpu_has_pat && boot_pat_state) {
                /*
-                * Panic if this happens on the secondary CPU, and we
+                * If this happens we are on a secondary CPU, but
                 * switched to PAT on the boot CPU. We have no way to
                 * undo PAT.
-               */
-               BUG_ON(boot_pat_state);
+                */
+               printk(KERN_ERR "PAT enabled, "
+                      "but not supported by secondary CPU\n");
+               BUG();
        }
 
 #ifndef CONFIG_XEN
@@ -87,8 +103,8 @@ void pat_init(void)
         *      011 UC          _PAGE_CACHE_UC
         * PAT bit unused
         */
-       pat = PAT(0,WB) | PAT(1,WC) | PAT(2,UC_MINUS) | PAT(3,UC) |
-             PAT(4,WB) | PAT(5,WC) | PAT(6,UC_MINUS) | PAT(7,UC);
+       pat = PAT(0, WB) | PAT(1, WC) | PAT(2, UC_MINUS) | PAT(3, UC) |
+             PAT(4, WB) | PAT(5, WC) | PAT(6, UC_MINUS) | PAT(7, UC);
 
        /* Boot CPU check */
        if (!boot_pat_state)
@@ -113,13 +129,13 @@ void pat_init(void)
 static char *cattr_name(unsigned long flags)
 {
        switch (flags & _PAGE_CACHE_MASK) {
-               case _PAGE_CACHE_UC:            return "uncached";
-               case _PAGE_CACHE_UC_MINUS:      return "uncached-minus";
-               case _PAGE_CACHE_WB:            return "write-back";
-               case _PAGE_CACHE_WC:            return "write-combining";
-               case _PAGE_CACHE_WP:            return "write-protected";
-               case _PAGE_CACHE_WT:            return "write-through";
-               default:                        return "broken";
+       case _PAGE_CACHE_UC:            return "uncached";
+       case _PAGE_CACHE_UC_MINUS:      return "uncached-minus";
+       case _PAGE_CACHE_WB:            return "write-back";
+       case _PAGE_CACHE_WC:            return "write-combining";
+       case _PAGE_CACHE_WP:            return "write-protected";
+       case _PAGE_CACHE_WT:            return "write-through";
+       default:                        return "broken";
        }
 }
 
@@ -157,49 +173,55 @@ static DEFINE_SPINLOCK(memtype_lock);     /
  * The intersection is based on "Effective Memory Type" tables in IA-32
  * SDM vol 3a
  */
-static int pat_x_mtrr_type(u64 start, u64 end, unsigned long prot,
-                               unsigned long *ret_prot)
+static unsigned long pat_x_mtrr_type(u64 start, u64 end, unsigned long req_type)
 {
-       unsigned long pat_type;
-       u8 mtrr_type;
-
-       pat_type = prot & _PAGE_CACHE_MASK;
-       prot &= (~_PAGE_CACHE_MASK);
-
-       /*
-        * We return the PAT request directly for types where PAT takes
-        * precedence with respect to MTRR and for UC_MINUS.
-        * Consistency checks with other PAT requests is done later
-        * while going through memtype list.
-        */
-       if (pat_type == _PAGE_CACHE_WC) {
-               *ret_prot = prot | _PAGE_CACHE_WC;
-               return 0;
-       } else if (pat_type == _PAGE_CACHE_UC_MINUS) {
-               *ret_prot = prot | _PAGE_CACHE_UC_MINUS;
-               return 0;
-       } else if (pat_type == _PAGE_CACHE_UC) {
-               *ret_prot = prot | _PAGE_CACHE_UC;
-               return 0;
-       }
-
        /*
         * Look for MTRR hint to get the effective type in case where PAT
         * request is for WB.
         */
-       mtrr_type = mtrr_type_lookup(start, end);
+       if (req_type == _PAGE_CACHE_WB) {
+               u8 mtrr_type;
 
-       if (mtrr_type == MTRR_TYPE_UNCACHABLE) {
-               *ret_prot = prot | _PAGE_CACHE_UC;
-       } else if (mtrr_type == MTRR_TYPE_WRCOMB) {
-               *ret_prot = prot | _PAGE_CACHE_WC;
-       } else {
-               *ret_prot = prot | _PAGE_CACHE_WB;
+               mtrr_type = mtrr_type_lookup(start, end);
+               if (mtrr_type == MTRR_TYPE_UNCACHABLE)
+                       return _PAGE_CACHE_UC;
+               if (mtrr_type == MTRR_TYPE_WRCOMB)
+                       return _PAGE_CACHE_WC;
+       }
+
+       return req_type;
+}
+
+static int chk_conflict(struct memtype *new, struct memtype *entry,
+                       unsigned long *type)
+{
+       if (new->type != entry->type) {
+               if (type) {
+                       new->type = entry->type;
+                       *type = entry->type;
+               } else
+                       goto conflict;
        }
 
+        /* check overlaps with more than one entry in the list */
+       list_for_each_entry_continue(entry, &memtype_list, nd) {
+               if (new->end <= entry->start)
+                       break;
+               else if (new->type != entry->type)
+                       goto conflict;
+       }
        return 0;
+
+ conflict:
+       printk(KERN_INFO "%s:%d conflicting memory types "
+              "%Lx-%Lx %s<->%s\n", current->comm, current->pid, new->start,
+              new->end, cattr_name(new->type), cattr_name(entry->type));
+       return -EBUSY;
 }
 
+static struct memtype *cached_entry;
+static u64 cached_start;
+
 /*
  * req_type typically has one of the:
  * - _PAGE_CACHE_WB
@@ -210,37 +232,36 @@ static int pat_x_mtrr_type(u64 start, u6
  * req_type will have a special case value '-1', when requester want to inherit
  * the memory type from mtrr (if WB), existing PAT, defaulting to UC_MINUS.
  *
- * If ret_type is NULL, function will return an error if it cannot reserve the
- * region with req_type. If ret_type is non-null, function will return
- * available type in ret_type in case of no error. In case of any error
+ * If new_type is NULL, function will return an error if it cannot reserve the
+ * region with req_type. If new_type is non-NULL, function will return
+ * available type in new_type in case of no error. In case of any error
  * it will return a negative return value.
  */
 int reserve_memtype(u64 start, u64 end, unsigned long req_type,
-                       unsigned long *ret_type)
+                       unsigned long *new_type)
 {
-       struct memtype *new_entry = NULL;
-       struct memtype *parse;
+       struct memtype *new, *entry;
        unsigned long actual_type;
+       struct list_head *where;
        int err = 0;
 
-       /* Only track when pat_wc_enabled */
-       if (!pat_wc_enabled) {
+       BUG_ON(start >= end); /* end is exclusive */
+
+       if (!pat_enabled) {
                /* This is identical to page table setting without PAT */
-               if (ret_type) {
-                       if (req_type == -1) {
-                               *ret_type = _PAGE_CACHE_WB;
-                       } else {
-                               *ret_type = req_type;
-                       }
+               if (new_type) {
+                       if (req_type == -1)
+                               *new_type = _PAGE_CACHE_WB;
+                       else
+                               *new_type = req_type & _PAGE_CACHE_MASK;
                }
                return 0;
        }
 
        /* Low ISA region is always mapped WB in page table. No need to track */
-       if (start >= ISA_START_ADDRESS && (end - 1) <= ISA_END_ADDRESS) {
-               if (ret_type)
-                       *ret_type = _PAGE_CACHE_WB;
-
+       if (is_ISA_range(start, end - 1)) {
+               if (new_type)
+                       *new_type = _PAGE_CACHE_WB;
                return 0;
        }
 
@@ -253,206 +274,118 @@ int reserve_memtype(u64 start, u64 end, 
                 */
                u8 mtrr_type = mtrr_type_lookup(start, end);
 
-               if (mtrr_type == MTRR_TYPE_WRBACK) {
-                       req_type = _PAGE_CACHE_WB;
+               if (mtrr_type == MTRR_TYPE_WRBACK)
                        actual_type = _PAGE_CACHE_WB;
-               } else {
-                       req_type = _PAGE_CACHE_UC_MINUS;
+               else
                        actual_type = _PAGE_CACHE_UC_MINUS;
-               }
-       } else {
-               req_type &= _PAGE_CACHE_MASK;
-               err = pat_x_mtrr_type(start, end, req_type, &actual_type);
-       }
-
-       if (err) {
-               if (ret_type)
-                       *ret_type = actual_type;
+       } else
+               actual_type = pat_x_mtrr_type(start, end,
+                                             req_type & _PAGE_CACHE_MASK);
 
-               return -EINVAL;
-       }
-
-       new_entry  = kmalloc(sizeof(struct memtype), GFP_KERNEL);
-       if (!new_entry)
+       new  = kmalloc(sizeof(struct memtype), GFP_KERNEL);
+       if (!new)
                return -ENOMEM;
 
-       new_entry->start = start;
-       new_entry->end = end;
-       new_entry->type = actual_type;
+       new->start = start;
+       new->end = end;
+       new->type = actual_type;
 
-       if (ret_type)
-               *ret_type = actual_type;
+       if (new_type)
+               *new_type = actual_type;
 
        spin_lock(&memtype_lock);
 
-       /* Search for existing mapping that overlaps the current range */
-       list_for_each_entry(parse, &memtype_list, nd) {
-               struct memtype *saved_ptr;
+       if (cached_entry && start >= cached_start)
+               entry = cached_entry;
+       else
+               entry = list_entry(&memtype_list, struct memtype, nd);
 
-               if (parse->start >= end) {
-                       pr_debug("New Entry\n");
-                       list_add(&new_entry->nd, parse->nd.prev);
-                       new_entry = NULL;
+       /* Search for existing mapping that overlaps the current range */
+       where = NULL;
+       list_for_each_entry_continue(entry, &memtype_list, nd) {
+               if (end <= entry->start) {
+                       where = entry->nd.prev;
+                       cached_entry = list_entry(where, struct memtype, nd);
                        break;
-               }
-
-               if (start <= parse->start && end >= parse->start) {
-                       if (actual_type != parse->type && ret_type) {
-                               actual_type = parse->type;
-                               *ret_type = actual_type;
-                               new_entry->type = actual_type;
-                       }
-
-                       if (actual_type != parse->type) {
-                               printk(
-               KERN_INFO "%s:%d conflicting memory types %Lx-%Lx %s<->%s\n",
-                                       current->comm, current->pid,
-                                       start, end,
-                                       cattr_name(actual_type),
-                                       cattr_name(parse->type));
-                               err = -EBUSY;
-                               break;
-                       }
-
-                       saved_ptr = parse;
-                       /*
-                        * Check to see whether the request overlaps more
-                        * than one entry in the list
-                        */
-                       list_for_each_entry_continue(parse, &memtype_list, nd) {
-                               if (end <= parse->start) {
-                                       break;
-                               }
-
-                               if (actual_type != parse->type) {
-                                       printk(
-               KERN_INFO "%s:%d conflicting memory types %Lx-%Lx %s<->%s\n",
-                                               current->comm, current->pid,
-                                               start, end,
-                                               cattr_name(actual_type),
-                                               cattr_name(parse->type));
-                                       err = -EBUSY;
-                                       break;
-                               }
-                       }
-
-                       if (err) {
-                               break;
+               } else if (start <= entry->start) { /* end > entry->start */
+                       err = chk_conflict(new, entry, new_type);
+                       if (!err) {
+                               dprintk("Overlap at 0x%Lx-0x%Lx\n",
+                                       entry->start, entry->end);
+                               where = entry->nd.prev;
+                               cached_entry = list_entry(where,
+                                                       struct memtype, nd);
                        }
-
-                       pr_debug("Overlap at 0x%Lx-0x%Lx\n",
-                              saved_ptr->start, saved_ptr->end);
-                       /* No conflict. Go ahead and add this new entry */
-                       list_add(&new_entry->nd, saved_ptr->nd.prev);
-                       new_entry = NULL;
                        break;
-               }
-
-               if (start < parse->end) {
-                       if (actual_type != parse->type && ret_type) {
-                               actual_type = parse->type;
-                               *ret_type = actual_type;
-                               new_entry->type = actual_type;
-                       }
-
-                       if (actual_type != parse->type) {
-                               printk(
-               KERN_INFO "%s:%d conflicting memory types %Lx-%Lx %s<->%s\n",
-                                       current->comm, current->pid,
-                                       start, end,
-                                       cattr_name(actual_type),
-                                       cattr_name(parse->type));
-                               err = -EBUSY;
-                               break;
-                       }
-
-                       saved_ptr = parse;
-                       /*
-                        * Check to see whether the request overlaps more
-                        * than one entry in the list
-                        */
-                       list_for_each_entry_continue(parse, &memtype_list, nd) {
-                               if (end <= parse->start) {
-                                       break;
-                               }
-
-                               if (actual_type != parse->type) {
-                                       printk(
-               KERN_INFO "%s:%d conflicting memory types %Lx-%Lx %s<->%s\n",
-                                               current->comm, current->pid,
-                                               start, end,
-                                               cattr_name(actual_type),
-                                               cattr_name(parse->type));
-                                       err = -EBUSY;
-                                       break;
+               } else if (start < entry->end) { /* start > entry->start */
+                       err = chk_conflict(new, entry, new_type);
+                       if (!err) {
+                               dprintk("Overlap at 0x%Lx-0x%Lx\n",
+                                       entry->start, entry->end);
+                               cached_entry = list_entry(entry->nd.prev,
+                                                       struct memtype, nd);
+
+                               /*
+                                * Move to right position in the linked
+                                * list to add this new entry
+                                */
+                               list_for_each_entry_continue(entry,
+                                                       &memtype_list, nd) {
+                                       if (start <= entry->start) {
+                                               where = entry->nd.prev;
+                                               break;
+                                       }
                                }
                        }
-
-                       if (err) {
-                               break;
-                       }
-
-                       pr_debug(KERN_INFO "Overlap at 0x%Lx-0x%Lx\n",
-                                saved_ptr->start, saved_ptr->end);
-                       /* No conflict. Go ahead and add this new entry */
-                       list_add(&new_entry->nd, &saved_ptr->nd);
-                       new_entry = NULL;
                        break;
                }
        }
 
        if (err) {
-               printk(KERN_INFO
-       "reserve_memtype failed 0x%Lx-0x%Lx, track %s, req %s\n",
-                       start, end, cattr_name(new_entry->type),
-                       cattr_name(req_type));
-               kfree(new_entry);
+               printk(KERN_INFO "reserve_memtype failed 0x%Lx-0x%Lx, "
+                      "track %s, req %s\n",
+                      start, end, cattr_name(new->type), cattr_name(req_type));
+               kfree(new);
                spin_unlock(&memtype_lock);
                return err;
        }
 
-       if (new_entry) {
-               /* No conflict. Not yet added to the list. Add to the tail */
-               list_add_tail(&new_entry->nd, &memtype_list);
-               pr_debug("New Entry\n");
-       }
+       cached_start = start;
 
-       if (ret_type) {
-               pr_debug(
-       "reserve_memtype added 0x%Lx-0x%Lx, track %s, req %s, ret %s\n",
-                       start, end, cattr_name(actual_type),
-                       cattr_name(req_type), cattr_name(*ret_type));
-       } else {
-               pr_debug(
-       "reserve_memtype added 0x%Lx-0x%Lx, track %s, req %s\n",
-                       start, end, cattr_name(actual_type),
-                       cattr_name(req_type));
-       }
+       if (where)
+               list_add(&new->nd, where);
+       else
+               list_add_tail(&new->nd, &memtype_list);
 
        spin_unlock(&memtype_lock);
+
+       dprintk("reserve_memtype added 0x%Lx-0x%Lx, track %s, req %s, ret %s\n",
+               start, end, cattr_name(new->type), cattr_name(req_type),
+               new_type ? cattr_name(*new_type) : "-");
+
        return err;
 }
 
 int free_memtype(u64 start, u64 end)
 {
-       struct memtype *ml;
+       struct memtype *entry;
        int err = -EINVAL;
 
-       /* Only track when pat_wc_enabled */
-       if (!pat_wc_enabled) {
+       if (!pat_enabled)
                return 0;
-       }
 
        /* Low ISA region is always mapped WB. No need to track */
-       if (start >= ISA_START_ADDRESS && end <= ISA_END_ADDRESS) {
+       if (is_ISA_range(start, end - 1))
                return 0;
-       }
 
        spin_lock(&memtype_lock);
-       list_for_each_entry(ml, &memtype_list, nd) {
-               if (ml->start == start && ml->end == end) {
-                       list_del(&ml->nd);
-                       kfree(ml);
+       list_for_each_entry(entry, &memtype_list, nd) {
+               if (entry->start == start && entry->end == end) {
+                       if (cached_entry == entry || cached_start == start)
+                               cached_entry = NULL;
+
+                       list_del(&entry->nd);
+                       kfree(entry);
                        err = 0;
                        break;
                }
@@ -464,27 +397,19 @@ int free_memtype(u64 start, u64 end)
                        current->comm, current->pid, start, end);
        }
 
-       pr_debug("free_memtype request 0x%Lx-0x%Lx\n", start, end);
+       dprintk("free_memtype request 0x%Lx-0x%Lx\n", start, end);
        return err;
 }
 
 
-/*
- * /dev/mem mmap interface. The memtype used for mapping varies:
- * - Use UC for mappings with O_SYNC flag
- * - Without O_SYNC flag, if there is any conflict in reserve_memtype,
- *   inherit the memtype from existing mapping.
- * - Else use UC_MINUS memtype (for backward compatibility with existing
- *   X drivers.
- */
 pgprot_t phys_mem_access_prot(struct file *file, unsigned long mfn,
                                unsigned long size, pgprot_t vma_prot)
 {
        return vma_prot;
 }
 
-#ifdef CONFIG_NONPROMISC_DEVMEM
-/* This check is done in drivers/char/mem.c in case of NONPROMISC_DEVMEM*/
+#ifdef CONFIG_STRICT_DEVMEM
+/* This check is done in drivers/char/mem.c in case of STRICT_DEVMEM*/
 static inline int range_is_allowed(unsigned long mfn, unsigned long size)
 {
        return 1;
@@ -508,20 +433,20 @@ static inline int range_is_allowed(unsig
        }
        return 1;
 }
-#endif /* CONFIG_NONPROMISC_DEVMEM */
+#endif /* CONFIG_STRICT_DEVMEM */
 
 int phys_mem_access_prot_allowed(struct file *file, unsigned long mfn,
                                unsigned long size, pgprot_t *vma_prot)
 {
        u64 addr = (u64)mfn << PAGE_SHIFT;
-       unsigned long flags = _PAGE_CACHE_UC_MINUS;
+       unsigned long flags = -1;
        int retval;
 
        if (!range_is_allowed(mfn, size))
                return 0;
 
        if (file->f_flags & O_SYNC) {
-               flags = _PAGE_CACHE_UC;
+               flags = _PAGE_CACHE_UC_MINUS;
        }
 
 #ifndef CONFIG_X86_32
@@ -534,25 +459,26 @@ int phys_mem_access_prot_allowed(struct 
         * caching for the high addresses through the KEN pin, but
         * we maintain the tradition of paranoia in this code.
         */
-       if (!pat_wc_enabled &&
-           ! ( test_bit(X86_FEATURE_MTRR, boot_cpu_data.x86_capability) ||
-               test_bit(X86_FEATURE_K6_MTRR, boot_cpu_data.x86_capability) ||
-               test_bit(X86_FEATURE_CYRIX_ARR, boot_cpu_data.x86_capability) ||
-               test_bit(X86_FEATURE_CENTAUR_MCR, boot_cpu_data.x86_capability)) &&
-          (pfn << PAGE_SHIFT) >= __pa(high_memory)) {
+       if (!pat_enabled &&
+           !(boot_cpu_has(X86_FEATURE_MTRR) ||
+             boot_cpu_has(X86_FEATURE_K6_MTRR) ||
+             boot_cpu_has(X86_FEATURE_CYRIX_ARR) ||
+             boot_cpu_has(X86_FEATURE_CENTAUR_MCR)) &&
+           (pfn << PAGE_SHIFT) >= __pa(high_memory)) {
                flags = _PAGE_CACHE_UC;
        }
 #endif
 #endif
 
        /*
-        * With O_SYNC, we can only take UC mapping. Fail if we cannot.
+        * With O_SYNC, we can only take UC_MINUS mapping. Fail if we cannot.
+        *
         * Without O_SYNC, we want to get
         * - WB for WB-able memory and no other conflicting mappings
         * - UC_MINUS for non-WB-able memory with no other conflicting mappings
         * - Inherit from confliting mappings otherwise
         */
-       if (flags != _PAGE_CACHE_UC_MINUS) {
+       if (flags != -1) {
                retval = reserve_memtype(addr, addr + size, flags, NULL);
        } else {
                retval = reserve_memtype(addr, addr + size, -1, &flags);
@@ -600,3 +526,88 @@ void unmap_devmem(unsigned long mfn, uns
        free_memtype(addr, addr + size);
 }
 
+#if defined(CONFIG_DEBUG_FS)
+
+/* get Nth element of the linked list */
+static struct memtype *memtype_get_idx(loff_t pos)
+{
+       struct memtype *list_node, *print_entry;
+       int i = 1;
+
+       print_entry  = kmalloc(sizeof(struct memtype), GFP_KERNEL);
+       if (!print_entry)
+               return NULL;
+
+       spin_lock(&memtype_lock);
+       list_for_each_entry(list_node, &memtype_list, nd) {
+               if (pos == i) {
+                       *print_entry = *list_node;
+                       spin_unlock(&memtype_lock);
+                       return print_entry;
+               }
+               ++i;
+       }
+       spin_unlock(&memtype_lock);
+       kfree(print_entry);
+       return NULL;
+}
+
+static void *memtype_seq_start(struct seq_file *seq, loff_t *pos)
+{
+       if (*pos == 0) {
+               ++*pos;
+               seq_printf(seq, "PAT memtype list:\n");
+       }
+
+       return memtype_get_idx(*pos);
+}
+
+static void *memtype_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+       ++*pos;
+       return memtype_get_idx(*pos);
+}
+
+static void memtype_seq_stop(struct seq_file *seq, void *v)
+{
+}
+
+static int memtype_seq_show(struct seq_file *seq, void *v)
+{
+       struct memtype *print_entry = (struct memtype *)v;
+
+       seq_printf(seq, "%s @ 0x%Lx-0x%Lx\n", cattr_name(print_entry->type),
+                       print_entry->start, print_entry->end);
+       kfree(print_entry);
+       return 0;
+}
+
+static struct seq_operations memtype_seq_ops = {
+       .start = memtype_seq_start,
+       .next  = memtype_seq_next,
+       .stop  = memtype_seq_stop,
+       .show  = memtype_seq_show,
+};
+
+static int memtype_seq_open(struct inode *inode, struct file *file)
+{
+       return seq_open(file, &memtype_seq_ops);
+}
+
+static const struct file_operations memtype_fops = {
+       .open    = memtype_seq_open,
+       .read    = seq_read,
+       .llseek  = seq_lseek,
+       .release = seq_release,
+};
+
+static int __init pat_memtype_list_init(void)
+{
+       debugfs_create_file("pat_memtype_list", S_IRUSR, arch_debugfs_dir,
+                               NULL, &memtype_fops);
+       return 0;
+}
+
+late_initcall(pat_memtype_list_init);
+
+#endif /* CONFIG_DEBUG_FS */
Index: head-2008-12-01/arch/x86/mm/pgtable-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/mm/pgtable-xen.c      2008-12-01 11:46:22.000000000 +0100
+++ head-2008-12-01/arch/x86/mm/pgtable-xen.c   2008-12-01 11:49:07.000000000 +0100
@@ -4,6 +4,7 @@
 #include <asm/pgalloc.h>
 #include <asm/pgtable.h>
 #include <asm/tlb.h>
+#include <asm/fixmap.h>
 #include <asm/hypervisor.h>
 #include <asm/mmu_context.h>
 
@@ -410,15 +411,9 @@ static inline void pgd_list_del(pgd_t *p
 static void pgd_ctor(void *p)
 {
        pgd_t *pgd = p;
-       unsigned long flags;
 
        pgd_test_and_unpin(pgd);
 
-       /* Clear usermode parts of PGD */
-       memset(pgd, 0, KERNEL_PGD_BOUNDARY*sizeof(pgd_t));
-
-       spin_lock_irqsave(&pgd_lock, flags);
-
        /* If the pgd points to a shared pagetable level (either the
           ptes in non-PAE, or shared PMD in PAE), then just copy the
           references from swapper_pg_dir. */
@@ -440,13 +435,9 @@ static void pgd_ctor(void *p)
                __pgd(__pa_symbol(level3_user_pgt) | _PAGE_TABLE);
 #endif
 
-#ifndef CONFIG_X86_PAE
        /* list required to sync kernel mapping updates */
        if (!SHARED_KERNEL_PMD)
                pgd_list_add(pgd);
-#endif
-
-       spin_unlock_irqrestore(&pgd_lock, flags);
 }
 
 static void pgd_dtor(void *pgd)
@@ -475,33 +466,6 @@ static void pgd_dtor(void *pgd)
 
 #ifdef CONFIG_X86_PAE
 /*
- * Mop up any pmd pages which may still be attached to the pgd.
- * Normally they will be freed by munmap/exit_mmap, but any pmd we
- * preallocate which never got a corresponding vma will need to be
- * freed manually.
- */
-static void pgd_mop_up_pmds(struct mm_struct *mm, pgd_t *pgdp)
-{
-       int i;
-
-       for(i = 0; i < UNSHARED_PTRS_PER_PGD; i++) {
-               pgd_t pgd = pgdp[i];
-
-               if (__pgd_val(pgd) != 0) {
-                       pmd_t *pmd = (pmd_t *)pgd_page_vaddr(pgd);
-
-                       pgdp[i] = xen_make_pgd(0);
-
-                       paravirt_release_pmd(pgd_val(pgd) >> PAGE_SHIFT);
-                       pmd_free(mm, pmd);
-               }
-       }
-
-       if (!xen_feature(XENFEAT_pae_pgdir_above_4gb))
-               xen_destroy_contiguous_region((unsigned long)pgdp, 0);
-}
-
-/*
  * In PAE mode, we need to do a cr3 reload (=tlb flush) when
  * updating the top-level pagetable entries to guarantee the
  * processor notices the update.  Since this is expensive, and
@@ -512,61 +476,7 @@ static void pgd_mop_up_pmds(struct mm_st
  * not shared between pagetables (!SHARED_KERNEL_PMDS), we allocate
  * and initialize the kernel pmds here.
  */
-static int pgd_prepopulate_pmd(struct mm_struct *mm, pgd_t *pgd)
-{
-       pud_t *pud;
-       pmd_t *pmds[UNSHARED_PTRS_PER_PGD];
-       unsigned long addr, flags;
-       int i;
-
-       /*
-        * We can race save/restore (if we sleep during a GFP_KERNEL memory
-        * allocation). We therefore store virtual addresses of pmds as they
-        * do not change across save/restore, and poke the machine addresses
-        * into the pgdir under the pgd_lock.
-        */
-       for (addr = i = 0; i < UNSHARED_PTRS_PER_PGD; i++, addr += PUD_SIZE) {
-               pmds[i] = pmd_alloc_one(mm, addr);
-               if (!pmds[i])
-                       goto out_oom;
-       }
-
-       spin_lock_irqsave(&pgd_lock, flags);
-
-       /* Protect against save/restore: move below 4GB under pgd_lock. */
-       if (!xen_feature(XENFEAT_pae_pgdir_above_4gb)
-           && xen_create_contiguous_region((unsigned long)pgd, 0, 32)) {
-               spin_unlock_irqrestore(&pgd_lock, flags);
-out_oom:
-               while (i--)
-                       pmd_free(mm, pmds[i]);
-               return 0;
-       }
-
-       /* Copy kernel pmd contents and write-protect the new pmds. */
-       pud = pud_offset(pgd, 0);
-       for (addr = i = 0; i < UNSHARED_PTRS_PER_PGD;
-            i++, pud++, addr += PUD_SIZE) {
-               if (i >= KERNEL_PGD_BOUNDARY) {
-                       memcpy(pmds[i],
-                              (pmd_t *)pgd_page_vaddr(swapper_pg_dir[i]),
-                              sizeof(pmd_t) * PTRS_PER_PMD);
-                       make_lowmem_page_readonly(
-                               pmds[i], XENFEAT_writable_page_tables);
-               }
-
-               /* It is safe to poke machine addresses of pmds under the pgd_lock. */
-               pud_populate(mm, pud, pmds[i]);
-       }
-
-       /* List required to sync kernel mapping updates and
-        * to pin/unpin on save/restore. */
-       pgd_list_add(pgd);
-
-       spin_unlock_irqrestore(&pgd_lock, flags);
-
-       return 1;
-}
+#define PREALLOCATED_PMDS      UNSHARED_PTRS_PER_PGD
 
 void pud_populate(struct mm_struct *mm, pud_t *pudp, pmd_t *pmd)
 {
@@ -596,16 +506,97 @@ void pud_populate(struct mm_struct *mm, 
                xen_tlb_flush();
 }
 #else  /* !CONFIG_X86_PAE */
+
 /* No need to prepopulate any pagetable entries in non-PAE modes. */
-static int pgd_prepopulate_pmd(struct mm_struct *mm, pgd_t *pgd)
+#define PREALLOCATED_PMDS      0
+
+#endif /* CONFIG_X86_PAE */
+
+static void free_pmds(pmd_t *pmds[], struct mm_struct *mm, bool contig)
 {
-       return 1;
+       int i;
+
+       if (contig)
+               xen_destroy_contiguous_region((unsigned long)mm->pgd, 0);
+
+       for(i = 0; i < PREALLOCATED_PMDS; i++)
+               if (pmds[i])
+                       pmd_free(mm, pmds[i]);
 }
 
-static void pgd_mop_up_pmds(struct mm_struct *mm, pgd_t *pgd)
+static int preallocate_pmds(pmd_t *pmds[], struct mm_struct *mm)
 {
+       int i;
+       bool failed = false;
+
+       for(i = 0; i < PREALLOCATED_PMDS; i++) {
+               pmd_t *pmd = pmd_alloc_one(mm, i << PUD_SHIFT);
+               if (pmd == NULL)
+                       failed = true;
+               pmds[i] = pmd;
+       }
+
+       if (failed) {
+               free_pmds(pmds, mm, false);
+               return -ENOMEM;
+       }
+
+       return 0;
+}
+
+/*
+ * Mop up any pmd pages which may still be attached to the pgd.
+ * Normally they will be freed by munmap/exit_mmap, but any pmd we
+ * preallocate which never got a corresponding vma will need to be
+ * freed manually.
+ */
+static void pgd_mop_up_pmds(struct mm_struct *mm, pgd_t *pgdp)
+{
+       int i;
+
+       for(i = 0; i < PREALLOCATED_PMDS; i++) {
+               pgd_t pgd = pgdp[i];
+
+               if (__pgd_val(pgd) != 0) {
+                       pmd_t *pmd = (pmd_t *)pgd_page_vaddr(pgd);
+
+                       pgdp[i] = xen_make_pgd(0);
+
+                       paravirt_release_pmd(pgd_val(pgd) >> PAGE_SHIFT);
+                       pmd_free(mm, pmd);
+               }
+       }
+
+       if (!xen_feature(XENFEAT_pae_pgdir_above_4gb))
+               xen_destroy_contiguous_region((unsigned long)pgdp, 0);
+}
+
+static void pgd_prepopulate_pmd(struct mm_struct *mm, pgd_t *pgd, pmd_t *pmds[])
+{
+       pud_t *pud;
+       unsigned long addr;
+       int i;
+
+       if (PREALLOCATED_PMDS == 0) /* Work around gcc-3.4.x bug */
+               return;
+
+       pud = pud_offset(pgd, 0);
+       for (addr = i = 0; i < PREALLOCATED_PMDS;
+            i++, pud++, addr += PUD_SIZE) {
+               pmd_t *pmd = pmds[i];
+
+               if (i >= KERNEL_PGD_BOUNDARY) {
+                       memcpy(pmd,
+                              (pmd_t *)pgd_page_vaddr(swapper_pg_dir[i]),
+                              sizeof(pmd_t) * PTRS_PER_PMD);
+                       make_lowmem_page_readonly(
+                               pmd, XENFEAT_writable_page_tables);
+               }
+
+               /* It is safe to poke machine addresses of pmds under the pgd_lock. */
+               pud_populate(mm, pud, pmd);
+       }
 }
-#endif /* CONFIG_X86_PAE */
 
 #ifdef CONFIG_X86_64
 /* We allocate two contiguous pages for kernel and user. */
@@ -616,19 +607,52 @@ static void pgd_mop_up_pmds(struct mm_st
 
 pgd_t *pgd_alloc(struct mm_struct *mm)
 {
-       pgd_t *pgd = (pgd_t *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, PGD_ORDER);
+       pgd_t *pgd;
+       pmd_t *pmds[PREALLOCATED_PMDS];
+       unsigned long flags;
+
+       pgd = (pgd_t *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, PGD_ORDER);
+
+       if (pgd == NULL)
+               goto out;
 
-       /* so that alloc_pd can use it */
        mm->pgd = pgd;
-       if (pgd)
-               pgd_ctor(pgd);
 
-       if (pgd && !pgd_prepopulate_pmd(mm, pgd)) {
-               free_pages((unsigned long)pgd, PGD_ORDER);
-               pgd = NULL;
+       if (preallocate_pmds(pmds, mm) != 0)
+               goto out_free_pgd;
+
+       if (paravirt_pgd_alloc(mm) != 0)
+               goto out_free_pmds;
+
+       /*
+        * Make sure that pre-populating the pmds is atomic with
+        * respect to anything walking the pgd_list, so that they
+        * never see a partially populated pgd.
+        */
+       spin_lock_irqsave(&pgd_lock, flags);
+
+#ifdef CONFIG_X86_PAE
+       /* Protect against save/restore: move below 4GB under pgd_lock. */
+       if (!xen_feature(XENFEAT_pae_pgdir_above_4gb)
+           && xen_create_contiguous_region((unsigned long)pgd, 0, 32)) {
+               spin_unlock_irqrestore(&pgd_lock, flags);
+               goto out_free_pmds;
        }
+#endif
+
+       pgd_ctor(pgd);
+       pgd_prepopulate_pmd(mm, pgd, pmds);
+
+       spin_unlock_irqrestore(&pgd_lock, flags);
 
        return pgd;
+
+out_free_pmds:
+       free_pmds(pmds, mm, !xen_feature(XENFEAT_pae_pgdir_above_4gb));
+out_free_pgd:
+       free_pages((unsigned long)pgd, PGD_ORDER);
+out:
+       return NULL;
 }
 
 void pgd_free(struct mm_struct *mm, pgd_t *pgd)
@@ -644,6 +668,7 @@ void pgd_free(struct mm_struct *mm, pgd_
        pgd_dtor(pgd);
 
        pgd_mop_up_pmds(mm, pgd);
+       paravirt_pgd_free(mm, pgd);
        free_pages((unsigned long)pgd, PGD_ORDER);
 }
 
@@ -685,7 +710,7 @@ int ptep_test_and_clear_young(struct vm_
 
        if (pte_young(*ptep))
                ret = test_and_clear_bit(_PAGE_BIT_ACCESSED,
-                                        &ptep->pte);
+                                        (unsigned long *) &ptep->pte);
 
        if (ret)
                pte_update(vma->vm_mm, addr, ptep);
@@ -707,3 +732,42 @@ int ptep_clear_flush_young(struct vm_are
 
        return young;
 }
+
+int fixmaps_set;
+
+void xen_set_fixmap(enum fixed_addresses idx, maddr_t phys, pgprot_t flags)
+{
+       unsigned long address = __fix_to_virt(idx);
+       pte_t pte;
+
+       if (idx >= __end_of_fixed_addresses) {
+               BUG();
+               return;
+       }
+
+       switch (idx) {
+#ifdef CONFIG_X86_64
+       extern pte_t level1_fixmap_pgt[PTRS_PER_PTE];
+
+       case VSYSCALL_LAST_PAGE ... VSYSCALL_FIRST_PAGE:
+               pte = pfn_pte(phys >> PAGE_SHIFT, flags);
+               set_pte_vaddr_pud(level3_user_pgt, address, pte);
+               break;
+       case FIX_EARLYCON_MEM_BASE:
+               xen_l1_entry_update(level1_fixmap_pgt + pte_index(address),
+                                   pfn_pte_ma(phys >> PAGE_SHIFT, flags));
+               fixmaps_set++;
+               return;
+#else
+       case FIX_WP_TEST:
+       case FIX_VDSO:
+               pte = pfn_pte(phys >> PAGE_SHIFT, flags);
+               break;
+#endif
+       default:
+               pte = pfn_pte_ma(phys >> PAGE_SHIFT, flags);
+               break;
+       }
+       set_pte_vaddr(address, pte);
+       fixmaps_set++;
+}
Index: head-2008-12-01/arch/x86/mm/pgtable_32-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/mm/pgtable_32-xen.c   2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/mm/pgtable_32-xen.c        2008-12-01 11:49:07.000000000 +0100
@@ -25,51 +25,49 @@
 #include <xen/features.h>
 #include <asm/hypervisor.h>
 
-void show_mem(void)
+/*
+ * Associate a virtual page frame with a given physical page frame
+ * and protection flags for that frame.
+ */
+void set_pte_vaddr(unsigned long vaddr, pte_t pteval)
 {
-       int total = 0, reserved = 0;
-       int shared = 0, cached = 0;
-       int highmem = 0;
-       struct page *page;
-       pg_data_t *pgdat;
-       unsigned long i;
-       unsigned long flags;
-
-       printk(KERN_INFO "Mem-info:\n");
-       show_free_areas();
-       for_each_online_pgdat(pgdat) {
-               pgdat_resize_lock(pgdat, &flags);
-               for (i = 0; i < pgdat->node_spanned_pages; ++i) {
-                       if (unlikely(i % MAX_ORDER_NR_PAGES == 0))
-                               touch_nmi_watchdog();
-                       page = pgdat_page_nr(pgdat, i);
-                       total++;
-                       if (PageHighMem(page))
-                               highmem++;
-                       if (PageReserved(page))
-                               reserved++;
-                       else if (PageSwapCache(page))
-                               cached++;
-                       else if (page_count(page))
-                               shared += page_count(page) - 1;
-               }
-               pgdat_resize_unlock(pgdat, &flags);
-       }
-       printk(KERN_INFO "%d pages of RAM\n", total);
-       printk(KERN_INFO "%d pages of HIGHMEM\n", highmem);
-       printk(KERN_INFO "%d reserved pages\n", reserved);
-       printk(KERN_INFO "%d pages shared\n", shared);
-       printk(KERN_INFO "%d pages swap cached\n", cached);
-
-       printk(KERN_INFO "%lu pages dirty\n", global_page_state(NR_FILE_DIRTY));
-       printk(KERN_INFO "%lu pages writeback\n",
-                                       global_page_state(NR_WRITEBACK));
-       printk(KERN_INFO "%lu pages mapped\n", global_page_state(NR_FILE_MAPPED));
-       printk(KERN_INFO "%lu pages slab\n",
-               global_page_state(NR_SLAB_RECLAIMABLE) +
-               global_page_state(NR_SLAB_UNRECLAIMABLE));
-       printk(KERN_INFO "%lu pages pagetables\n",
-                                       global_page_state(NR_PAGETABLE));
+#ifndef CONFIG_XEN
+       pgd_t *pgd;
+       pud_t *pud;
+       pmd_t *pmd;
+       pte_t *pte;
+
+       pgd = swapper_pg_dir + pgd_index(vaddr);
+       if (pgd_none(*pgd)) {
+               BUG();
+               return;
+       }
+       pud = pud_offset(pgd, vaddr);
+       if (pud_none(*pud)) {
+               BUG();
+               return;
+       }
+       pmd = pmd_offset(pud, vaddr);
+       if (pmd_none(*pmd)) {
+               BUG();
+               return;
+       }
+       pte = pte_offset_kernel(pmd, vaddr);
+       if (pte_val(pteval))
+               set_pte_present(&init_mm, vaddr, pte, pteval);
+       else
+               pte_clear(&init_mm, vaddr, pte);
+
+       /*
+        * It's enough to flush this one mapping.
+        * (PGE mappings get flushed as well)
+        */
+       __flush_tlb_one(vaddr);
+#else
+       if (HYPERVISOR_update_va_mapping(vaddr, pteval,
+                                        UVMF_INVLPG|UVMF_ALL))
+               BUG();
+#endif
 }
 
 /*
@@ -107,35 +105,10 @@ void set_pmd_pfn(unsigned long vaddr, un
        __flush_tlb_one(vaddr);
 }
 
-static int fixmaps;
 unsigned long hypervisor_virt_start = HYPERVISOR_VIRT_START;
 unsigned long __FIXADDR_TOP = (HYPERVISOR_VIRT_START - PAGE_SIZE);
 EXPORT_SYMBOL(__FIXADDR_TOP);
 
-void __set_fixmap (enum fixed_addresses idx, maddr_t phys, pgprot_t flags)
-{
-       unsigned long address = __fix_to_virt(idx);
-       pte_t pte;
-
-       if (idx >= __end_of_fixed_addresses) {
-               BUG();
-               return;
-       }
-       switch (idx) {
-       case FIX_WP_TEST:
-       case FIX_VDSO:
-               pte = pfn_pte(phys >> PAGE_SHIFT, flags);
-               break;
-       default:
-               pte = pfn_pte_ma(phys >> PAGE_SHIFT, flags);
-               break;
-       }
-       if (HYPERVISOR_update_va_mapping(address, pte,
-                                        UVMF_INVLPG|UVMF_ALL))
-               BUG();
-       fixmaps++;
-}
-
 /**
  * reserve_top_address - reserves a hole in the top of kernel address space
  * @reserve - size of hole to reserve
@@ -145,13 +118,48 @@ void __set_fixmap (enum fixed_addresses 
  */
 void __init reserve_top_address(unsigned long reserve)
 {
-       BUG_ON(fixmaps > 0);
+       BUG_ON(fixmaps_set > 0);
        printk(KERN_INFO "Reserving virtual address space above 0x%08x\n",
               (int)-reserve);
        __FIXADDR_TOP = -reserve - PAGE_SIZE;
        __VMALLOC_RESERVE += reserve;
 }
 
+/*
+ * vmalloc=size forces the vmalloc area to be exactly 'size'
+ * bytes. This can be used to increase (or decrease) the
+ * vmalloc area - the default is 128m.
+ */
+static int __init parse_vmalloc(char *arg)
+{
+       if (!arg)
+               return -EINVAL;
+
+       __VMALLOC_RESERVE = memparse(arg, &arg);
+       return 0;
+}
+early_param("vmalloc", parse_vmalloc);
+
+#ifndef CONFIG_XEN
+/*
+ * reservetop=size reserves a hole at the top of the kernel address space which
+ * a hypervisor can load into later.  Needed for dynamically loaded hypervisors,
+ * so relocating the fixmap can be done before paging initialization.
+ */
+static int __init parse_reservetop(char *arg)
+{
+       unsigned long address;
+
+       if (!arg)
+               return -EINVAL;
+
+       address = memparse(arg, &arg);
+       reserve_top_address(address);
+       return 0;
+}
+early_param("reservetop", parse_reservetop);
+#endif
+
 void make_lowmem_page_readonly(void *va, unsigned int feature)
 {
        pte_t *pte;
Index: head-2008-12-01/arch/x86/pci/amd_bus.c
===================================================================
--- head-2008-12-01.orig/arch/x86/pci/amd_bus.c 2008-12-03 15:48:43.000000000 +0100
+++ head-2008-12-01/arch/x86/pci/amd_bus.c      2008-12-01 11:49:07.000000000 +0100
@@ -607,6 +607,14 @@ static int __init pci_io_ecs_init(void)
        for_each_online_cpu(cpu)
                amd_cpu_notify(&amd_cpu_notifier, (unsigned long)CPU_ONLINE,
                               (void *)(long)cpu);
+#ifdef CONFIG_XEN
+       {
+               u64 reg;
+               rdmsrl(MSR_AMD64_NB_CFG, reg);
+               if (!(reg & ENABLE_CF8_EXT_CFG))
+                       return 0;
+       }
+#endif
        pci_probe |= PCI_HAS_IO_ECS;
 
        return 0;
@@ -614,6 +622,10 @@ static int __init pci_io_ecs_init(void)
 
 static int __init amd_postcore_init(void)
 {
+#ifdef CONFIG_XEN
+       if (!is_initial_xendomain())
+               return 0;
+#endif
        if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD)
                return 0;
 
Index: head-2008-12-01/arch/x86/pci/irq-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/pci/irq-xen.c 2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/pci/irq-xen.c      2008-12-01 11:49:07.000000000 +0100
@@ -11,8 +11,8 @@
 #include <linux/slab.h>
 #include <linux/interrupt.h>
 #include <linux/dmi.h>
-#include <asm/io.h>
-#include <asm/smp.h>
+#include <linux/io.h>
+#include <linux/smp.h>
 #include <asm/io_apic.h>
 #include <linux/irq.h>
 #include <linux/acpi.h>
@@ -45,7 +45,8 @@ struct irq_router {
        char *name;
        u16 vendor, device;
        int (*get)(struct pci_dev *router, struct pci_dev *dev, int pirq);
-       int (*set)(struct pci_dev *router, struct pci_dev *dev, int pirq, int new);
+       int (*set)(struct pci_dev *router, struct pci_dev *dev, int pirq,
+               int new);
 };
 
 struct irq_router_handler {
@@ -61,7 +62,7 @@ void (*pcibios_disable_irq)(struct pci_d
  *  and perform checksum verification.
  */
 
-static inline struct irq_routing_table * pirq_check_routing_table(u8 *addr)
+static inline struct irq_routing_table *pirq_check_routing_table(u8 *addr)
 {
        struct irq_routing_table *rt;
        int i;
@@ -74,10 +75,11 @@ static inline struct irq_routing_table *
            rt->size < sizeof(struct irq_routing_table))
                return NULL;
        sum = 0;
-       for (i=0; i < rt->size; i++)
+       for (i = 0; i < rt->size; i++)
                sum += addr[i];
        if (!sum) {
-               DBG(KERN_DEBUG "PCI: Interrupt Routing Table found at 0x%p\n", rt);
+               DBG(KERN_DEBUG "PCI: Interrupt Routing Table found at 0x%p\n",
+                       rt);
                return rt;
        }
        return NULL;
@@ -104,7 +106,9 @@ static struct irq_routing_table * __init
                        return rt;
                printk(KERN_WARNING "PCI: PIRQ table NOT found at pirqaddr\n");
        }
-       for(addr = (u8 *) isa_bus_to_virt(0xf0000); addr < (u8 *) isa_bus_to_virt(0x100000); addr += 16) {
+       for (addr = (u8 *) isa_bus_to_virt(0xf0000);
+            addr < (u8 *) isa_bus_to_virt(0x100000);
+            addr += 16) {
                rt = pirq_check_routing_table(addr);
                if (rt)
                        return rt;
@@ -126,20 +130,20 @@ static void __init pirq_peer_trick(void)
        struct irq_info *e;
 
        memset(busmap, 0, sizeof(busmap));
-       for(i=0; i < (rt->size - sizeof(struct irq_routing_table)) / sizeof(struct irq_info); i++) {
+       for (i = 0; i < (rt->size - sizeof(struct irq_routing_table)) / sizeof(struct irq_info); i++) {
                e = &rt->slots[i];
 #ifdef DEBUG
                {
                        int j;
                        DBG(KERN_DEBUG "%02x:%02x slot=%02x", e->bus, e->devfn/8, e->slot);
-                       for(j=0; j<4; j++)
+                       for (j = 0; j < 4; j++)
                                DBG(" %d:%02x/%04x", j, e->irq[j].link, e->irq[j].bitmap);
                        DBG("\n");
                }
 #endif
                busmap[e->bus] = 1;
        }
-       for(i = 1; i < 256; i++) {
+       for (i = 1; i < 256; i++) {
                int node;
                if (!busmap[i] || pci_find_bus(0, i))
                        continue;
@@ -187,7 +191,8 @@ static unsigned int read_config_nybble(s
        return (nr & 1) ? (x >> 4) : (x & 0xf);
 }
 
-static void write_config_nybble(struct pci_dev *router, unsigned offset, unsigned nr, unsigned int val)
+static void write_config_nybble(struct pci_dev *router, unsigned offset,
+       unsigned nr, unsigned int val)
 {
        u8 x;
        unsigned reg = offset + (nr >> 1);
@@ -289,7 +294,7 @@ static int pirq_ite_get(struct pci_dev *
        static const unsigned char pirqmap[4] = { 1, 0, 2, 3 };
 
        WARN_ON_ONCE(pirq > 4);
-       return read_config_nybble(router,0x43, pirqmap[pirq-1]);
+       return read_config_nybble(router, 0x43, pirqmap[pirq-1]);
 }
 
 static int pirq_ite_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
@@ -318,7 +323,7 @@ static int pirq_opti_set(struct pci_dev 
 
 /*
  * Cyrix: nibble offset 0x5C
- * 0x5C bits 7:4 is INTB bits 3:0 is INTA 
+ * 0x5C bits 7:4 is INTB bits 3:0 is INTA
  * 0x5D bits 7:4 is INTD bits 3:0 is INTC
  */
 static int pirq_cyrix_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
@@ -354,7 +359,7 @@ static int pirq_cyrix_set(struct pci_dev
  *     Apparently there are systems implementing PCI routing table using
  *     link values 0x01-0x04 and others using 0x41-0x44 for PCI INTA..D.
  *     We try our best to handle both link mappings.
- *     
+ *
  *     Currently (2003-05-21) it appears most SiS chipsets follow the
  *     definition of routing registers from the SiS-5595 southbridge.
  *     According to the SiS 5595 datasheets the revision id's of the
@@ -374,7 +379,7 @@ static int pirq_cyrix_set(struct pci_dev
  *
  *     0x62:   USBIRQ:
  *             bit 6 OHCI function disabled (0), enabled (1)
- *     
+ *
  *     0x6a:   ACPI/SCI IRQ: bits 4-6 reserved
  *
  *     0x7e:   Data Acq. Module IRQ - bits 4-6 reserved
@@ -437,7 +442,7 @@ static int pirq_vlsi_get(struct pci_dev 
 {
        WARN_ON_ONCE(pirq >= 9);
        if (pirq > 8) {
-               printk(KERN_INFO "VLSI router pirq escape (%d)\n", pirq);
+               dev_info(&dev->dev, "VLSI router PIRQ escape (%d)\n", pirq);
                return 0;
        }
        return read_config_nybble(router, 0x74, pirq-1);
@@ -447,7 +452,7 @@ static int pirq_vlsi_set(struct pci_dev 
 {
        WARN_ON_ONCE(pirq >= 9);
        if (pirq > 8) {
-               printk(KERN_INFO "VLSI router pirq escape (%d)\n", pirq);
+               dev_info(&dev->dev, "VLSI router PIRQ escape (%d)\n", pirq);
                return 0;
        }
        write_config_nybble(router, 0x74, pirq-1, irq);
@@ -471,7 +476,8 @@ static int pirq_serverworks_get(struct p
        return inb(0xc01) & 0xf;
 }
 
-static int pirq_serverworks_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
+static int pirq_serverworks_set(struct pci_dev *router, struct pci_dev *dev,
+       int pirq, int irq)
 {
        outb(pirq, 0xc00);
        outb(irq, 0xc01);
@@ -491,22 +497,20 @@ static int pirq_amd756_get(struct pci_de
        u8 irq;
        irq = 0;
        if (pirq <= 4)
-       {
                irq = read_config_nybble(router, 0x56, pirq - 1);
-       }
-       printk(KERN_INFO "AMD756: dev %04x:%04x, router pirq : %d get irq : %2d\n",
-               dev->vendor, dev->device, pirq, irq);
+       dev_info(&dev->dev,
+                "AMD756: dev [%04x/%04x], router PIRQ %d get IRQ %d\n",
+                dev->vendor, dev->device, pirq, irq);
        return irq;
 }
 
 static int pirq_amd756_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
 {
-       printk(KERN_INFO "AMD756: dev %04x:%04x, router pirq : %d SET irq : %2d\n", 
-               dev->vendor, dev->device, pirq, irq);
+       dev_info(&dev->dev,
+                "AMD756: dev [%04x/%04x], router PIRQ %d set IRQ %d\n",
+                dev->vendor, dev->device, pirq, irq);
        if (pirq <= 4)
-       {
                write_config_nybble(router, 0x56, pirq - 1, irq);
-       }
        return 1;
 }
 
@@ -553,50 +557,51 @@ static __init int intel_router_probe(str
        if (pci_dev_present(pirq_440gx))
                return 0;
 
-       switch(device)
-       {
-               case PCI_DEVICE_ID_INTEL_82371FB_0:
-               case PCI_DEVICE_ID_INTEL_82371SB_0:
-               case PCI_DEVICE_ID_INTEL_82371AB_0:
-               case PCI_DEVICE_ID_INTEL_82371MX:
-               case PCI_DEVICE_ID_INTEL_82443MX_0:
-               case PCI_DEVICE_ID_INTEL_82801AA_0:
-               case PCI_DEVICE_ID_INTEL_82801AB_0:
-               case PCI_DEVICE_ID_INTEL_82801BA_0:
-               case PCI_DEVICE_ID_INTEL_82801BA_10:
-               case PCI_DEVICE_ID_INTEL_82801CA_0:
-               case PCI_DEVICE_ID_INTEL_82801CA_12:
-               case PCI_DEVICE_ID_INTEL_82801DB_0:
-               case PCI_DEVICE_ID_INTEL_82801E_0:
-               case PCI_DEVICE_ID_INTEL_82801EB_0:
-               case PCI_DEVICE_ID_INTEL_ESB_1:
-               case PCI_DEVICE_ID_INTEL_ICH6_0:
-               case PCI_DEVICE_ID_INTEL_ICH6_1:
-               case PCI_DEVICE_ID_INTEL_ICH7_0:
-               case PCI_DEVICE_ID_INTEL_ICH7_1:
-               case PCI_DEVICE_ID_INTEL_ICH7_30:
-               case PCI_DEVICE_ID_INTEL_ICH7_31:
-               case PCI_DEVICE_ID_INTEL_ESB2_0:
-               case PCI_DEVICE_ID_INTEL_ICH8_0:
-               case PCI_DEVICE_ID_INTEL_ICH8_1:
-               case PCI_DEVICE_ID_INTEL_ICH8_2:
-               case PCI_DEVICE_ID_INTEL_ICH8_3:
-               case PCI_DEVICE_ID_INTEL_ICH8_4:
-               case PCI_DEVICE_ID_INTEL_ICH9_0:
-               case PCI_DEVICE_ID_INTEL_ICH9_1:
-               case PCI_DEVICE_ID_INTEL_ICH9_2:
-               case PCI_DEVICE_ID_INTEL_ICH9_3:
-               case PCI_DEVICE_ID_INTEL_ICH9_4:
-               case PCI_DEVICE_ID_INTEL_ICH9_5:
-               case PCI_DEVICE_ID_INTEL_TOLAPAI_0:
-               case PCI_DEVICE_ID_INTEL_ICH10_0:
-               case PCI_DEVICE_ID_INTEL_ICH10_1:
-               case PCI_DEVICE_ID_INTEL_ICH10_2:
-               case PCI_DEVICE_ID_INTEL_ICH10_3:
-                       r->name = "PIIX/ICH";
-                       r->get = pirq_piix_get;
-                       r->set = pirq_piix_set;
-                       return 1;
+       switch (device) {
+       case PCI_DEVICE_ID_INTEL_82371FB_0:
+       case PCI_DEVICE_ID_INTEL_82371SB_0:
+       case PCI_DEVICE_ID_INTEL_82371AB_0:
+       case PCI_DEVICE_ID_INTEL_82371MX:
+       case PCI_DEVICE_ID_INTEL_82443MX_0:
+       case PCI_DEVICE_ID_INTEL_82801AA_0:
+       case PCI_DEVICE_ID_INTEL_82801AB_0:
+       case PCI_DEVICE_ID_INTEL_82801BA_0:
+       case PCI_DEVICE_ID_INTEL_82801BA_10:
+       case PCI_DEVICE_ID_INTEL_82801CA_0:
+       case PCI_DEVICE_ID_INTEL_82801CA_12:
+       case PCI_DEVICE_ID_INTEL_82801DB_0:
+       case PCI_DEVICE_ID_INTEL_82801E_0:
+       case PCI_DEVICE_ID_INTEL_82801EB_0:
+       case PCI_DEVICE_ID_INTEL_ESB_1:
+       case PCI_DEVICE_ID_INTEL_ICH6_0:
+       case PCI_DEVICE_ID_INTEL_ICH6_1:
+       case PCI_DEVICE_ID_INTEL_ICH7_0:
+       case PCI_DEVICE_ID_INTEL_ICH7_1:
+       case PCI_DEVICE_ID_INTEL_ICH7_30:
+       case PCI_DEVICE_ID_INTEL_ICH7_31:
+       case PCI_DEVICE_ID_INTEL_ESB2_0:
+       case PCI_DEVICE_ID_INTEL_ICH8_0:
+       case PCI_DEVICE_ID_INTEL_ICH8_1:
+       case PCI_DEVICE_ID_INTEL_ICH8_2:
+       case PCI_DEVICE_ID_INTEL_ICH8_3:
+       case PCI_DEVICE_ID_INTEL_ICH8_4:
+       case PCI_DEVICE_ID_INTEL_ICH9_0:
+       case PCI_DEVICE_ID_INTEL_ICH9_1:
+       case PCI_DEVICE_ID_INTEL_ICH9_2:
+       case PCI_DEVICE_ID_INTEL_ICH9_3:
+       case PCI_DEVICE_ID_INTEL_ICH9_4:
+       case PCI_DEVICE_ID_INTEL_ICH9_5:
+       case PCI_DEVICE_ID_INTEL_TOLAPAI_0:
+       case PCI_DEVICE_ID_INTEL_ICH10_0:
+       case PCI_DEVICE_ID_INTEL_ICH10_1:
+       case PCI_DEVICE_ID_INTEL_ICH10_2:
+       case PCI_DEVICE_ID_INTEL_ICH10_3:
+       case PCI_DEVICE_ID_INTEL_PCH_0:
+       case PCI_DEVICE_ID_INTEL_PCH_1:
+               r->name = "PIIX/ICH";
+               r->get = pirq_piix_get;
+               r->set = pirq_piix_set;
+               return 1;
        }
        return 0;
 }
@@ -610,7 +615,7 @@ static __init int via_router_probe(struc
         * workarounds for some buggy BIOSes
         */
        if (device == PCI_DEVICE_ID_VIA_82C586_0) {
-               switch(router->device) {
+               switch (router->device) {
                case PCI_DEVICE_ID_VIA_82C686:
                        /*
                         * Asus k7m bios wrongly reports 82C686A
@@ -635,7 +640,7 @@ static __init int via_router_probe(struc
                }
        }
 
-       switch(device) {
+       switch (device) {
        case PCI_DEVICE_ID_VIA_82C586_0:
                r->name = "VIA";
                r->get = pirq_via586_get;
@@ -658,28 +663,27 @@ static __init int via_router_probe(struc
 
 static __init int vlsi_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
 {
-       switch(device)
-       {
-               case PCI_DEVICE_ID_VLSI_82C534:
-                       r->name = "VLSI 82C534";
-                       r->get = pirq_vlsi_get;
-                       r->set = pirq_vlsi_set;
-                       return 1;
+       switch (device) {
+       case PCI_DEVICE_ID_VLSI_82C534:
+               r->name = "VLSI 82C534";
+               r->get = pirq_vlsi_get;
+               r->set = pirq_vlsi_set;
+               return 1;
        }
        return 0;
 }
 
 
-static __init int serverworks_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
+static __init int serverworks_router_probe(struct irq_router *r,
+               struct pci_dev *router, u16 device)
 {
-       switch(device)
-       {
-               case PCI_DEVICE_ID_SERVERWORKS_OSB4:
-               case PCI_DEVICE_ID_SERVERWORKS_CSB5:
-                       r->name = "ServerWorks";
-                       r->get = pirq_serverworks_get;
-                       r->set = pirq_serverworks_set;
-                       return 1;
+       switch (device) {
+       case PCI_DEVICE_ID_SERVERWORKS_OSB4:
+       case PCI_DEVICE_ID_SERVERWORKS_CSB5:
+               r->name = "ServerWorks";
+               r->get = pirq_serverworks_get;
+               r->set = pirq_serverworks_set;
+               return 1;
        }
        return 0;
 }
@@ -688,7 +692,7 @@ static __init int sis_router_probe(struc
 {
        if (device != PCI_DEVICE_ID_SI_503)
                return 0;
-               
+
        r->name = "SIS";
        r->get = pirq_sis_get;
        r->set = pirq_sis_set;
@@ -697,50 +701,45 @@ static __init int sis_router_probe(struc
 
 static __init int cyrix_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
 {
-       switch(device)
-       {
-               case PCI_DEVICE_ID_CYRIX_5520:
-                       r->name = "NatSemi";
-                       r->get = pirq_cyrix_get;
-                       r->set = pirq_cyrix_set;
-                       return 1;
+       switch (device) {
+       case PCI_DEVICE_ID_CYRIX_5520:
+               r->name = "NatSemi";
+               r->get = pirq_cyrix_get;
+               r->set = pirq_cyrix_set;
+               return 1;
        }
        return 0;
 }
 
 static __init int opti_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
 {
-       switch(device)
-       {
-               case PCI_DEVICE_ID_OPTI_82C700:
-                       r->name = "OPTI";
-                       r->get = pirq_opti_get;
-                       r->set = pirq_opti_set;
-                       return 1;
+       switch (device) {
+       case PCI_DEVICE_ID_OPTI_82C700:
+               r->name = "OPTI";
+               r->get = pirq_opti_get;
+               r->set = pirq_opti_set;
+               return 1;
        }
        return 0;
 }
 
 static __init int ite_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
 {
-       switch(device)
-       {
-               case PCI_DEVICE_ID_ITE_IT8330G_0:
-                       r->name = "ITE";
-                       r->get = pirq_ite_get;
-                       r->set = pirq_ite_set;
-                       return 1;
+       switch (device) {
+       case PCI_DEVICE_ID_ITE_IT8330G_0:
+               r->name = "ITE";
+               r->get = pirq_ite_get;
+               r->set = pirq_ite_set;
+               return 1;
        }
        return 0;
 }
 
 static __init int ali_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
 {
-       switch(device)
-       {
+       switch (device) {
        case PCI_DEVICE_ID_AL_M1533:
        case PCI_DEVICE_ID_AL_M1563:
-               printk(KERN_DEBUG "PCI: Using ALI IRQ Router\n");
                r->name = "ALI";
                r->get = pirq_ali_get;
                r->set = pirq_ali_set;
@@ -751,25 +750,24 @@ static __init int ali_router_probe(struc
 
 static __init int amd_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
 {
-       switch(device)
-       {
-               case PCI_DEVICE_ID_AMD_VIPER_740B:
-                       r->name = "AMD756";
-                       break;
-               case PCI_DEVICE_ID_AMD_VIPER_7413:
-                       r->name = "AMD766";
-                       break;
-               case PCI_DEVICE_ID_AMD_VIPER_7443:
-                       r->name = "AMD768";
-                       break;
-               default:
-                       return 0;
+       switch (device) {
+       case PCI_DEVICE_ID_AMD_VIPER_740B:
+               r->name = "AMD756";
+               break;
+       case PCI_DEVICE_ID_AMD_VIPER_7413:
+               r->name = "AMD766";
+               break;
+       case PCI_DEVICE_ID_AMD_VIPER_7443:
+               r->name = "AMD768";
+               break;
+       default:
+               return 0;
        }
        r->get = pirq_amd756_get;
        r->set = pirq_amd756_set;
        return 1;
 }
-               
+
 static __init int pico_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
 {
        switch (device) {
@@ -811,7 +809,7 @@ static struct pci_dev *pirq_router_dev;
  *     FIXME: should we have an option to say "generic for
  *     chipset" ?
  */
- 
+
 static void __init pirq_find_router(struct irq_router *r)
 {
        struct irq_routing_table *rt = pirq_table;
@@ -830,7 +828,7 @@ static void __init pirq_find_router(stru
        r->name = "default";
        r->get = NULL;
        r->set = NULL;
-       
+
        DBG(KERN_DEBUG "PCI: Attempting to find IRQ router for %04x:%04x\n",
            rt->rtr_vendor, rt->rtr_device);
 
@@ -841,19 +839,19 @@ static void __init pirq_find_router(stru
                return;
        }
 
-       for( h = pirq_routers; h->vendor; h++) {
+       for (h = pirq_routers; h->vendor; h++) {
                /* First look for a router match */
-               if (rt->rtr_vendor == h->vendor && h->probe(r, pirq_router_dev, rt->rtr_device))
+               if (rt->rtr_vendor == h->vendor &&
+                       h->probe(r, pirq_router_dev, rt->rtr_device))
                        break;
                /* Fall back to a device match */
-               if (pirq_router_dev->vendor == h->vendor && h->probe(r, pirq_router_dev, pirq_router_dev->device))
+               if (pirq_router_dev->vendor == h->vendor &&
+                       h->probe(r, pirq_router_dev, pirq_router_dev->device))
                        break;
        }
-       printk(KERN_INFO "PCI: Using IRQ router %s [%04x/%04x] at %s\n",
-               pirq_router.name,
-               pirq_router_dev->vendor,
-               pirq_router_dev->device,
-               pci_name(pirq_router_dev));
+       dev_info(&pirq_router_dev->dev, "%s IRQ router [%04x/%04x]\n",
+                pirq_router.name,
+                pirq_router_dev->vendor, pirq_router_dev->device);
 
        /* The device remains referenced for the kernel lifetime */
 }
@@ -861,11 +859,13 @@ static void __init pirq_find_router(stru
 static struct irq_info *pirq_get_info(struct pci_dev *dev)
 {
        struct irq_routing_table *rt = pirq_table;
-       int entries = (rt->size - sizeof(struct irq_routing_table)) / sizeof(struct irq_info);
+       int entries = (rt->size - sizeof(struct irq_routing_table)) /
+               sizeof(struct irq_info);
        struct irq_info *info;
 
        for (info = rt->slots; entries--; info++)
-               if (info->bus == dev->bus->number && PCI_SLOT(info->devfn) == PCI_SLOT(dev->devfn))
+               if (info->bus == dev->bus->number &&
+                       PCI_SLOT(info->devfn) == PCI_SLOT(dev->devfn))
                        return info;
        return NULL;
 }
@@ -884,7 +884,7 @@ static int pcibios_lookup_irq(struct pci
        /* Find IRQ pin */
        pci_read_config_byte(dev, PCI_INTERRUPT_PIN, &pin);
        if (!pin) {
-               DBG(KERN_DEBUG " -> no interrupt pin\n");
+               dev_dbg(&dev->dev, "no interrupt pin\n");
                return 0;
        }
        pin = pin - 1;
@@ -893,20 +893,21 @@ static int pcibios_lookup_irq(struct pci
 
        if (!pirq_table)
                return 0;
-       
-       DBG(KERN_DEBUG "IRQ for %s[%c]", pci_name(dev), 'A' + pin);
+
        info = pirq_get_info(dev);
        if (!info) {
-               DBG(" -> not found in routing table\n" KERN_DEBUG);
+               dev_dbg(&dev->dev, "PCI INT %c not found in routing table\n",
+                       'A' + pin);
                return 0;
        }
        pirq = info->irq[pin].link;
        mask = info->irq[pin].bitmap;
        if (!pirq) {
-               DBG(" -> not routed\n" KERN_DEBUG);
+               dev_dbg(&dev->dev, "PCI INT %c not routed\n", 'A' + pin);
                return 0;
        }
-       DBG(" -> PIRQ %02x, mask %04x, excl %04x", pirq, mask, pirq_table->exclusive_irqs);
+       dev_dbg(&dev->dev, "PCI INT %c -> PIRQ %02x, mask %04x, excl %04x",
+               'A' + pin, pirq, mask, pirq_table->exclusive_irqs);
        mask &= pcibios_irq_mask;
 
        /* Work around broken HP Pavilion Notebooks which assign USB to
@@ -919,7 +920,8 @@ static int pcibios_lookup_irq(struct pci
        }
 
        /* same for Acer Travelmate 360, but with CB and irq 11 -> 10 */
-       if (acer_tm360_irqrouting && dev->irq == 11 && dev->vendor == PCI_VENDOR_ID_O2) {
+       if (acer_tm360_irqrouting && dev->irq == 11 &&
+               dev->vendor == PCI_VENDOR_ID_O2) {
                pirq = 0x68;
                mask = 0x400;
                dev->irq = r->get(pirq_router_dev, dev, pirq);
@@ -932,51 +934,50 @@ static int pcibios_lookup_irq(struct pci
         */
        newirq = dev->irq;
        if (newirq && !((1 << newirq) & mask)) {
-               if ( pci_probe & PCI_USE_PIRQ_MASK) newirq = 0;
-               else printk("\n" KERN_WARNING
-                       "PCI: IRQ %i for device %s doesn't match PIRQ mask "
-                       "- try pci=usepirqmask\n" KERN_DEBUG, newirq,
-                       pci_name(dev));
+               if (pci_probe & PCI_USE_PIRQ_MASK)
+                       newirq = 0;
+               else
+                       dev_warn(&dev->dev, "IRQ %d doesn't match PIRQ mask "
+                                "%#x; try pci=usepirqmask\n", newirq, mask);
        }
        if (!newirq && assign) {
                for (i = 0; i < 16; i++) {
                        if (!(mask & (1 << i)))
                                continue;
-                       if (pirq_penalty[i] < pirq_penalty[newirq] && can_request_irq(i, IRQF_SHARED))
+                       if (pirq_penalty[i] < pirq_penalty[newirq] &&
+                               can_request_irq(i, IRQF_SHARED))
                                newirq = i;
                }
        }
-       DBG(" -> newirq=%d", newirq);
+       dev_dbg(&dev->dev, "PCI INT %c -> newirq %d", 'A' + pin, newirq);
 
        /* Check if it is hardcoded */
        if ((pirq & 0xf0) == 0xf0) {
                irq = pirq & 0xf;
-               DBG(" -> hardcoded IRQ %d\n", irq);
-               msg = "Hardcoded";
-       } else if ( r->get && (irq = r->get(pirq_router_dev, dev, pirq)) && \
-       ((!(pci_probe & PCI_USE_PIRQ_MASK)) || ((1 << irq) & mask)) ) {
-               DBG(" -> got IRQ %d\n", irq);
-               msg = "Found";
+               msg = "hardcoded";
+       } else if (r->get && (irq = r->get(pirq_router_dev, dev, pirq)) && \
+       ((!(pci_probe & PCI_USE_PIRQ_MASK)) || ((1 << irq) & mask))) {
+               msg = "found";
                eisa_set_level_irq(irq);
-       } else if (newirq && r->set && (dev->class >> 8) != PCI_CLASS_DISPLAY_VGA) {
-               DBG(" -> assigning IRQ %d", newirq);
+       } else if (newirq && r->set &&
+               (dev->class >> 8) != PCI_CLASS_DISPLAY_VGA) {
                if (r->set(pirq_router_dev, dev, pirq, newirq)) {
                        eisa_set_level_irq(newirq);
-                       DBG(" ... OK\n");
-                       msg = "Assigned";
+                       msg = "assigned";
                        irq = newirq;
                }
        }
 
        if (!irq) {
-               DBG(" ... failed\n");
                if (newirq && mask == (1 << newirq)) {
-                       msg = "Guessed";
+                       msg = "guessed";
                        irq = newirq;
-               } else
+               } else {
+                       dev_dbg(&dev->dev, "can't route interrupt\n");
                        return 0;
+               }
        }
-       printk(KERN_INFO "PCI: %s IRQ %d for device %s\n", msg, irq, pci_name(dev));
+       dev_info(&dev->dev, "%s PCI INT %c -> IRQ %d\n", msg, 'A' + pin, irq);
 
        /* Update IRQ for all devices with the same pirq value */
        while ((dev2 = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev2)) != NULL) {
@@ -988,20 +989,25 @@ static int pcibios_lookup_irq(struct pci
                if (!info)
                        continue;
                if (info->irq[pin].link == pirq) {
-                       /* We refuse to override the dev->irq information. Give a warning! */
-                       if ( dev2->irq && dev2->irq != irq && \
+                       /*
+                        * We refuse to override the dev->irq
+                        * information. Give a warning!
+                        */
+                       if (dev2->irq && dev2->irq != irq && \
                        (!(pci_probe & PCI_USE_PIRQ_MASK) || \
-                       ((1 << dev2->irq) & mask)) ) {
+                       ((1 << dev2->irq) & mask))) {
 #ifndef CONFIG_PCI_MSI
-                               printk(KERN_INFO "IRQ routing conflict for %s, have irq %d, want irq %d\n",
-                                      pci_name(dev2), dev2->irq, irq);
+                               dev_info(&dev2->dev, "IRQ routing conflict: "
+                                        "have IRQ %d, want IRQ %d\n",
+                                        dev2->irq, irq);
 #endif
-                               continue;
-                       }
+                               continue;
+                       }
                        dev2->irq = irq;
                        pirq_penalty[irq]++;
                        if (dev != dev2)
-                               printk(KERN_INFO "PCI: Sharing IRQ %d with %s\n", irq, pci_name(dev2));
+                               dev_info(&dev->dev, "sharing IRQ %d with %s\n",
+                                        irq, pci_name(dev2));
                }
        }
        return 1;
@@ -1015,15 +1021,20 @@ static void __init pcibios_fixup_irqs(vo
        DBG(KERN_DEBUG "PCI: IRQ fixup\n");
        while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) {
                /*
-                * If the BIOS has set an out of range IRQ number, just ignore it.
-                * Also keep track of which IRQ's are already in use.
+                * If the BIOS has set an out of range IRQ number, just
+                * ignore it.  Also keep track of which IRQ's are
+                * already in use.
                 */
                if (dev->irq >= 16) {
-                       DBG(KERN_DEBUG "%s: ignoring bogus IRQ %d\n", pci_name(dev), dev->irq);
+                       dev_dbg(&dev->dev, "ignoring bogus IRQ %d\n", dev->irq);
                        dev->irq = 0;
                }
-               /* If the IRQ is already assigned to a PCI device, ignore its ISA use penalty */
-               if (pirq_penalty[dev->irq] >= 100 && pirq_penalty[dev->irq] < 100000)
+               /*
+                * If the IRQ is already assigned to a PCI device,
+                * ignore its ISA use penalty
+                */
+               if (pirq_penalty[dev->irq] >= 100 &&
+                               pirq_penalty[dev->irq] < 100000)
                        pirq_penalty[dev->irq] = 0;
                pirq_penalty[dev->irq]++;
        }
@@ -1035,13 +1046,17 @@ static void __init pcibios_fixup_irqs(vo
                /*
                 * Recalculate IRQ numbers if we use the I/O APIC.
                 */
-               if (io_apic_assign_pci_irqs)
-               {
+               if (io_apic_assign_pci_irqs) {
                        int irq;
 
                        if (pin) {
-                               pin--;          /* interrupt pins are numbered starting from 1 */
-                               irq = IO_APIC_get_PCI_irq_vector(dev->bus->number, PCI_SLOT(dev->devfn), pin);
+                               /*
+                                * interrupt pins are numbered starting
+                                * from 1
+                                */
+                               pin--;
+                               irq = IO_APIC_get_PCI_irq_vector(dev->bus->number,
+                                       PCI_SLOT(dev->devfn), pin);
        /*
         * Busses behind bridges are typically not listed in the MP-table.
         * In this case we have to look up the IRQ based on the parent bus,
@@ -1049,18 +1064,18 @@ static void __init pcibios_fixup_irqs(vo
         * busses itself so we should get into this branch reliably.
         */
                                if (irq < 0 && dev->bus->parent) { /* go back to the bridge */
-                                       struct pci_dev * bridge = dev->bus->self;
+                                       struct pci_dev *bridge = dev->bus->self;
 
                                        pin = (pin + PCI_SLOT(dev->devfn)) % 4;
-                                       irq = IO_APIC_get_PCI_irq_vector(bridge->bus->number, 
+                                       irq = IO_APIC_get_PCI_irq_vector(bridge->bus->number,
                                                        PCI_SLOT(bridge->devfn), pin);
                                        if (irq >= 0)
-                                               printk(KERN_WARNING "PCI: using PPB %s[%c] to get irq %d\n",
-                                                       pci_name(bridge), 'A' + pin, irq);
+                                               dev_warn(&dev->dev, "using bridge %s INT %c to get IRQ %d\n",
+                                                        pci_name(bridge),
+                                                        'A' + pin, irq);
                                }
                                if (irq >= 0) {
-                                       printk(KERN_INFO "PCI->APIC IRQ transform: %s[%c] -> IRQ %d\n",
-                                               pci_name(dev), 'A' + pin, irq);
+                                       dev_info(&dev->dev, "PCI->APIC IRQ transform: INT %c -> IRQ %d\n", 'A' + pin, irq);
                                        dev->irq = irq;
                                }
                        }
@@ -1082,7 +1097,8 @@ static int __init fix_broken_hp_bios_irq
 {
        if (!broken_hp_bios_irq9) {
                broken_hp_bios_irq9 = 1;
-               printk(KERN_INFO "%s detected - fixing broken IRQ routing\n", d->ident);
+               printk(KERN_INFO "%s detected - fixing broken IRQ routing\n",
+                       d->ident);
        }
        return 0;
 }
@@ -1095,7 +1111,8 @@ static int __init fix_acer_tm360_irqrout
 {
        if (!acer_tm360_irqrouting) {
                acer_tm360_irqrouting = 1;
-               printk(KERN_INFO "%s detected - fixing broken IRQ routing\n", d->ident);
+               printk(KERN_INFO "%s detected - fixing broken IRQ routing\n",
+                       d->ident);
        }
        return 0;
 }
@@ -1107,7 +1124,8 @@ static struct dmi_system_id __initdata p
                .matches = {
                        DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"),
                        DMI_MATCH(DMI_BIOS_VERSION, "GE.M1.03"),
-                       DMI_MATCH(DMI_PRODUCT_VERSION, "HP Pavilion Notebook Model GE"),
+                       DMI_MATCH(DMI_PRODUCT_VERSION,
+                               "HP Pavilion Notebook Model GE"),
                        DMI_MATCH(DMI_BOARD_VERSION, "OmniBook N32N-736"),
                },
        },
@@ -1122,7 +1140,7 @@ static struct dmi_system_id __initdata p
        { }
 };
 
-static int __init pcibios_irq_init(void)
+int __init pcibios_irq_init(void)
 {
        DBG(KERN_DEBUG "PCI: IRQ init\n");
 
@@ -1142,11 +1160,14 @@ static int __init pcibios_irq_init(void)
                pirq_find_router(&pirq_router);
                if (pirq_table->exclusive_irqs) {
                        int i;
-                       for (i=0; i<16; i++)
+                       for (i = 0; i < 16; i++)
                                if (!(pirq_table->exclusive_irqs & (1 << i)))
                                        pirq_penalty[i] += 100;
                }
-               /* If we're using the I/O APIC, avoid using the PCI IRQ routing table */
+               /*
+                * If we're using the I/O APIC, avoid using the PCI IRQ
+                * routing table
+                */
                if (io_apic_assign_pci_irqs)
                        pirq_table = NULL;
        }
@@ -1157,9 +1178,6 @@ static int __init pcibios_irq_init(void)
        return 0;
 }
 
-subsys_initcall(pcibios_irq_init);
-
-
 static void pirq_penalize_isa_irq(int irq, int active)
 {
        /*
@@ -1193,7 +1211,7 @@ static int pirq_enable_irq(struct pci_de
        if (pin && !pcibios_lookup_irq(dev, 1) && !dev->irq) {
                char *msg = "";
 
-               pin--;          /* interrupt pins are numbered starting from 1 */
+               pin--; /* interrupt pins are numbered starting from 1 */
 
                if (io_apic_assign_pci_irqs) {
                        int irq;
@@ -1207,35 +1225,41 @@ static int pirq_enable_irq(struct pci_de
                         */
                        temp_dev = dev;
                        while (irq < 0 && dev->bus->parent) { /* go back to the bridge */
-                               struct pci_dev * bridge = dev->bus->self;
+                               struct pci_dev *bridge = dev->bus->self;
 
                                pin = (pin + PCI_SLOT(dev->devfn)) % 4;
-                               irq = IO_APIC_get_PCI_irq_vector(bridge->bus->number, 
+                               irq = IO_APIC_get_PCI_irq_vector(bridge->bus->number,
                                                PCI_SLOT(bridge->devfn), pin);
                                if (irq >= 0)
-                                       printk(KERN_WARNING "PCI: using PPB %s[%c] to get irq %d\n",
-                                               pci_name(bridge), 'A' + pin, irq);
+                                       dev_warn(&dev->dev, "using bridge %s "
+                                                "INT %c to get IRQ %d\n",
+                                                pci_name(bridge), 'A' + pin,
+                                                irq);
                                dev = bridge;
                        }
                        dev = temp_dev;
                        if (irq >= 0) {
-                               printk(KERN_INFO "PCI->APIC IRQ transform: %s[%c] -> IRQ %d\n",
-                                       pci_name(dev), 'A' + pin, irq);
+                               dev_info(&dev->dev, "PCI->APIC IRQ transform: "
+                                        "INT %c -> IRQ %d\n", 'A' + pin, irq);
                                dev->irq = irq;
                                return 0;
                        } else
-                               msg = " Probably buggy MP table.";
+                               msg = "; probably buggy MP table";
                } else if (pci_probe & PCI_BIOS_IRQ_SCAN)
                        msg = "";
                else
-                       msg = " Please try using pci=biosirq.";
+                       msg = "; please try using pci=biosirq";
 
-               /* With IDE legacy devices the IRQ lookup failure is not a problem.. */
-               if (dev->class >> 8 == PCI_CLASS_STORAGE_IDE && !(dev->class & 0x5))
+               /*
+                * With IDE legacy devices the IRQ lookup failure is not
+                * a problem..
+                */
+               if (dev->class >> 8 == PCI_CLASS_STORAGE_IDE &&
+                               !(dev->class & 0x5))
                        return 0;
 
-               printk(KERN_WARNING "PCI: No IRQ known for interrupt pin %c of device %s.%s\n",
-                      'A' + pin, pci_name(dev), msg);
+               dev_warn(&dev->dev, "can't find IRQ for PCI INT %c%s\n",
+                        'A' + pin, msg);
        }
        return 0;
 }
Index: head-2008-12-01/arch/x86/vdso/Makefile
===================================================================
--- head-2008-12-01.orig/arch/x86/vdso/Makefile 2008-12-01 11:37:10.000000000 +0100
+++ head-2008-12-01/arch/x86/vdso/Makefile      2008-12-01 11:49:07.000000000 +0100
@@ -65,9 +65,7 @@ obj-$(VDSO32-y)                       += vdso32-syms.lds
 vdso32.so-$(VDSO32-y)          += int80
 vdso32.so-$(CONFIG_COMPAT)     += syscall
 vdso32.so-$(VDSO32-y)          += sysenter
-xen-vdso32-$(subst 1,$(CONFIG_COMPAT),$(shell expr $(CONFIG_XEN_COMPAT)0 '<' 0x0302000)) += int80
-xen-vdso32-$(CONFIG_X86_32)    += syscall
-vdso32.so-$(CONFIG_XEN)                += $(xen-vdso32-y)
+vdso32.so-$(CONFIG_X86_XEN)    += syscall
 
 vdso32-images                  = $(vdso32.so-y:%=vdso32-%.so)
 
Index: head-2008-12-01/arch/x86/vdso/vdso32.S
===================================================================
--- head-2008-12-01.orig/arch/x86/vdso/vdso32.S 2008-12-01 11:37:10.000000000 +0100
+++ head-2008-12-01/arch/x86/vdso/vdso32.S      2008-12-01 11:49:07.000000000 +0100
@@ -9,7 +9,7 @@ vdso32_int80_end:
 
        .globl vdso32_syscall_start, vdso32_syscall_end
 vdso32_syscall_start:
-#ifdef CONFIG_COMPAT
+#if defined(CONFIG_COMPAT) || defined(CONFIG_X86_XEN)
        .incbin "arch/x86/vdso/vdso32-syscall.so"
 #endif
 vdso32_syscall_end:
@@ -19,16 +19,4 @@ vdso32_sysenter_start:
        .incbin "arch/x86/vdso/vdso32-sysenter.so"
 vdso32_sysenter_end:
 
-#if defined(CONFIG_X86_64_XEN) && CONFIG_XEN_COMPAT < 0x030200
-       .globl vdso32_int80_start, vdso32_int80_end
-vdso32_int80_start:
-       .incbin "arch/x86/vdso/vdso32-int80.so"
-vdso32_int80_end:
-#elif defined(CONFIG_X86_XEN)
-       .globl vdso32_syscall_start, vdso32_syscall_end
-vdso32_syscall_start:
-       .incbin "arch/x86/vdso/vdso32-syscall.so"
-vdso32_syscall_end:
-#endif
-
 __FINIT
Index: head-2008-12-01/arch/x86/vdso/vdso32-setup-xen.c
===================================================================
--- head-2008-12-01.orig/arch/x86/vdso/vdso32-setup-xen.c       2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/arch/x86/vdso/vdso32-setup-xen.c    2008-12-01 11:49:07.000000000 +0100
@@ -195,50 +195,28 @@ static __init void relocate_vdso(Elf32_E
        }
 }
 
-/*
- * These symbols are defined by vdso32.S to mark the bounds
- * of the ELF DSO images included therein.
- */
-extern const char vdso32_default_start, vdso32_default_end;
-extern const char vdso32_sysenter_start, vdso32_sysenter_end;
 static struct page *vdso32_pages[1];
 
 #ifdef CONFIG_X86_64
 
-#if CONFIG_XEN_COMPAT < 0x030200
-static int use_int80 = 1;
-#endif
-static int use_sysenter __read_mostly = -1;
-
-#define        vdso32_sysenter()       (use_sysenter > 0)
+#define        vdso32_sysenter()       (boot_cpu_has(X86_FEATURE_SYSENTER32))
+#define        vdso32_syscall()        (boot_cpu_has(X86_FEATURE_SYSCALL32))
 
-/* May not be __init: called during resume */
-void syscall32_cpu_init(void)
+void __cpuinit syscall32_cpu_init(void)
 {
-       static const struct callback_register cstar = {
+       static /*const*/ struct callback_register __cpuinitdata cstar = {
                .type = CALLBACKTYPE_syscall32,
                .address = (unsigned long)ia32_cstar_target
        };
-       static const struct callback_register sysenter = {
+       static /*const*/ struct callback_register __cpuinitdata sysenter = {
                .type = CALLBACKTYPE_sysenter,
                .address = (unsigned long)ia32_sysenter_target
        };
 
-       if ((HYPERVISOR_callback_op(CALLBACKOP_register, &sysenter) < 0) ||
-           (HYPERVISOR_callback_op(CALLBACKOP_register, &cstar) < 0))
-#if CONFIG_XEN_COMPAT < 0x030200
-               return;
-       use_int80 = 0;
-#else
-               BUG();
-#endif
-
-       if (use_sysenter < 0) {
-               if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL)
-                       use_sysenter = 1;
-               if (boot_cpu_data.x86_vendor == X86_VENDOR_CENTAUR)
-                       use_sysenter = 1;
-       }
+       if (HYPERVISOR_callback_op(CALLBACKOP_register, &sysenter) < 0)
+               setup_clear_cpu_cap(X86_FEATURE_SYSENTER32);
+       if (HYPERVISOR_callback_op(CALLBACKOP_register, &cstar) < 0)
+               setup_clear_cpu_cap(X86_FEATURE_SYSCALL32);
 }
 
 #define compat_uses_vma                1
@@ -250,6 +228,7 @@ static inline void map_compat_vdso(int m
 #else  /* CONFIG_X86_32 */
 
 #define vdso32_sysenter()      (boot_cpu_has(X86_FEATURE_SEP))
+#define vdso32_syscall()       (boot_cpu_has(X86_FEATURE_SYSCALL32))
 
 extern asmlinkage void ia32pv_cstar_target(void);
 static /*const*/ struct callback_register __cpuinitdata cstar = {
@@ -265,13 +244,13 @@ void __cpuinit enable_sep_cpu(void)
                .address = { __KERNEL_CS, (unsigned long)ia32pv_sysenter_target },
        };
 
-       if (boot_cpu_has(X86_FEATURE_SYSCALL)) {
+       if (vdso32_syscall()) {
                if (HYPERVISOR_callback_op(CALLBACKOP_register, &cstar) != 0)
                        BUG();
                return;
        }
 
-       if (!boot_cpu_has(X86_FEATURE_SEP))
+       if (!vdso32_sysenter())
                return;
 
        if (xen_feature(XENFEAT_supervisor_mode_kernel))
@@ -341,34 +320,26 @@ int __init sysenter_setup(void)
 
 #ifdef CONFIG_X86_32
        gate_vma_init();
-#endif
 
-#if defined(CONFIG_X86_64) && CONFIG_XEN_COMPAT < 0x030200
-       if (use_int80) {
-               extern const char vdso32_int80_start, vdso32_int80_end;
-
-               vsyscall = &vdso32_int80_start;
-               vsyscall_len = &vdso32_int80_end - &vdso32_int80_start;
-       } else
-#elif defined(CONFIG_X86_32)
-       if (boot_cpu_has(X86_FEATURE_SYSCALL)
-           && (boot_cpu_data.x86_vendor != X86_VENDOR_AMD
-               || HYPERVISOR_callback_op(CALLBACKOP_register, &cstar) != 0))
-               setup_clear_cpu_cap(X86_FEATURE_SYSCALL);
-       barrier(); /* until clear_bit()'s constraints are correct ... */
        if (boot_cpu_has(X86_FEATURE_SYSCALL)) {
-               extern const char vdso32_syscall_start, vdso32_syscall_end;
-
+               if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD
+                   && HYPERVISOR_callback_op(CALLBACKOP_register, &cstar) == 0)
+                       setup_force_cpu_cap(X86_FEATURE_SYSCALL32);
+               else {
+                       setup_clear_cpu_cap(X86_FEATURE_SYSCALL);
+                       setup_clear_cpu_cap(X86_FEATURE_SYSCALL32);
+               }
+       }
+#endif
+       if (vdso32_syscall()) {
                vsyscall = &vdso32_syscall_start;
                vsyscall_len = &vdso32_syscall_end - &vdso32_syscall_start;
-       } else
-#endif
-       if (!vdso32_sysenter()) {
-               vsyscall = &vdso32_default_start;
-               vsyscall_len = &vdso32_default_end - &vdso32_default_start;
-       } else {
+       } else if (vdso32_sysenter()){
                vsyscall = &vdso32_sysenter_start;
                vsyscall_len = &vdso32_sysenter_end - &vdso32_sysenter_start;
+       } else {
+               vsyscall = &vdso32_int80_start;
+               vsyscall_len = &vdso32_int80_end - &vdso32_int80_start;
        }
 
        memcpy(syscall_page, vsyscall, vsyscall_len);
Index: head-2008-12-01/arch/x86/xen/Kconfig
===================================================================
--- head-2008-12-01.orig/arch/x86/xen/Kconfig   2008-12-01 11:36:47.000000000 +0100
+++ head-2008-12-01/arch/x86/xen/Kconfig        2008-12-01 11:49:07.000000000 +0100
@@ -17,7 +17,7 @@ config XEN_MAX_DOMAIN_MEMORY
        int "Maximum allowed size of a domain in gigabytes"
        default 8 if X86_32
        default 32 if X86_64
-       depends on XEN
+       depends on PARAVIRT_XEN
        help
          The pseudo-physical to machine address array is sized
          according to the maximum possible memory size of a Xen
@@ -26,5 +26,5 @@ config XEN_MAX_DOMAIN_MEMORY
 
 config XEN_SAVE_RESTORE
        bool
-       depends on PM
+       depends on PARAVIRT_XEN && PM
        default y
\ No newline at end of file
Index: head-2008-12-01/drivers/acpi/processor_core.c
===================================================================
--- head-2008-12-01.orig/drivers/acpi/processor_core.c  2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/drivers/acpi/processor_core.c       2008-12-01 11:49:07.000000000 +0100
@@ -721,9 +721,11 @@ static int __cpuinit acpi_processor_star
        if (result)
                goto end;
 
-       sysdev = get_cpu_sysdev(pr->id);
-       if (sysfs_create_link(&device->dev.kobj, &sysdev->kobj, "sysdev"))
-               return -EFAULT;
+       if (pr->id != -1) {
+               sysdev = get_cpu_sysdev(pr->id);
+               if (sysfs_create_link(&device->dev.kobj, &sysdev->kobj, "sysdev"))
+                       return -EFAULT;
+       }
 
        status = acpi_install_notify_handler(pr->handle, ACPI_DEVICE_NOTIFY,
                                             acpi_processor_notify, pr);
@@ -895,7 +897,8 @@ static int acpi_processor_remove(struct 
        status = acpi_remove_notify_handler(pr->handle, ACPI_DEVICE_NOTIFY,
                                            acpi_processor_notify);
 
-       sysfs_remove_link(&device->dev.kobj, "sysdev");
+       if (pr->id != -1)
+               sysfs_remove_link(&device->dev.kobj, "sysdev");
 
        acpi_processor_remove_fs(device);
 
Index: head-2008-12-01/drivers/char/tpm/tpm_vtpm.c
===================================================================
--- head-2008-12-01.orig/drivers/char/tpm/tpm_vtpm.c    2008-12-03 15:48:43.000000000 +0100
+++ head-2008-12-01/drivers/char/tpm/tpm_vtpm.c 2008-12-01 11:49:07.000000000 +0100
@@ -347,7 +347,7 @@ static int _vtpm_send_queued(struct tpm_
 {
        int rc;
        int error = 0;
-       long flags;
+       unsigned long flags;
        unsigned char buffer[1];
        struct vtpm_state *vtpms;
        vtpms = (struct vtpm_state *)chip_get_private(chip);
Index: head-2008-12-01/drivers/misc/Kconfig
===================================================================
--- head-2008-12-01.orig/drivers/misc/Kconfig   2008-12-03 15:48:43.000000000 +0100
+++ head-2008-12-01/drivers/misc/Kconfig        2008-12-01 11:49:07.000000000 +0100
@@ -438,7 +438,7 @@ config ENCLOSURE_SERVICES
 config SGI_XP
        tristate "Support communication between SGI SSIs"
        depends on NET
-       depends on (IA64_GENERIC || IA64_SGI_SN2 || IA64_SGI_UV || X86_64) && SMP
+       depends on (IA64_GENERIC || IA64_SGI_SN2 || IA64_SGI_UV || X86_64) && SMP && !XEN
        select IA64_UNCACHED_ALLOCATOR if IA64_GENERIC || IA64_SGI_SN2
        select GENERIC_ALLOCATOR if IA64_GENERIC || IA64_SGI_SN2
        select SGI_GRU if (IA64_GENERIC || IA64_SGI_UV || X86_64) && SMP
@@ -465,7 +465,7 @@ config HP_ILO
 
 config SGI_GRU
        tristate "SGI GRU driver"
-       depends on (X86_64 || IA64_SGI_UV || IA64_GENERIC) && SMP
+       depends on (X86_64 || IA64_SGI_UV || IA64_GENERIC) && SMP && !XEN
        default n
        select MMU_NOTIFIER
        ---help---
Index: head-2008-12-01/drivers/pci/msi-xen.c
===================================================================
--- head-2008-12-01.orig/drivers/pci/msi-xen.c  2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/drivers/pci/msi-xen.c       2008-12-01 11:49:07.000000000 +0100
@@ -90,12 +90,10 @@ arch_teardown_msi_irqs(struct pci_dev *d
 }
 #endif
 
-static void msi_set_enable(struct pci_dev *dev, int enable)
+static void __msi_set_enable(struct pci_dev *dev, int pos, int enable)
 {
-       int pos;
        u16 control;
 
-       pos = pci_find_capability(dev, PCI_CAP_ID_MSI);
        if (pos) {
                pci_read_config_word(dev, pos + PCI_MSI_FLAGS, &control);
                control &= ~PCI_MSI_FLAGS_ENABLE;
@@ -105,6 +103,11 @@ static void msi_set_enable(struct pci_de
        }
 }
 
+static void msi_set_enable(struct pci_dev *dev, int enable)
+{
+       __msi_set_enable(dev, pci_find_capability(dev, PCI_CAP_ID_MSI), enable);
+}
+
 static void msix_set_enable(struct pci_dev *dev, int enable)
 {
        int pos;
@@ -573,9 +576,8 @@ int pci_enable_msi(struct pci_dev* dev)
 
        /* Check whether driver already requested for MSI-X irqs */
        if (dev->msix_enabled) {
-               printk(KERN_INFO "PCI: %s: Can't enable MSI.  "
-                       "Device already has MSI-X enabled\n",
-                       pci_name(dev));
+               dev_info(&dev->dev, "can't enable MSI "
+                        "(MSI-X already enabled)\n");
                return -EINVAL;
        }
 
@@ -707,9 +709,8 @@ int pci_enable_msix(struct pci_dev* dev,
        temp = dev->irq;
        /* Check whether driver already requested for MSI vector */
        if (dev->msi_enabled) {
-               printk(KERN_INFO "PCI: %s: Can't enable MSI-X.  "
-                      "Device already has an MSI irq assigned\n",
-                      pci_name(dev));
+               dev_info(&dev->dev, "can't enable MSI-X "
+                      "(MSI IRQ already assigned)\n");
                return -EINVAL;
        }
 
Index: head-2008-12-01/drivers/pci/quirks.c
===================================================================
--- head-2008-12-01.orig/drivers/pci/quirks.c   2008-12-03 15:48:43.000000000 +0100
+++ head-2008-12-01/drivers/pci/quirks.c        2008-12-01 11:49:07.000000000 +0100
@@ -42,9 +42,7 @@ static void __devinit quirk_release_reso
                        /* PCI Host Bridge isn't a target device */
                        return;
                }
-               printk(KERN_INFO
-                       "PCI: Disable device and release resources [%s].\n",
-                       pci_name(dev));
+               dev_info(&dev->dev, "disable device and release resources\n");
                pci_disable_device(dev);
 
                for (i=0; i < PCI_NUM_RESOURCES; i++) {
Index: head-2008-12-01/drivers/pci/setup-res.c
===================================================================
--- head-2008-12-01.orig/drivers/pci/setup-res.c        2008-12-03 15:48:43.000000000 +0100
+++ head-2008-12-01/drivers/pci/setup-res.c     2008-12-01 11:50:17.000000000 +0100
@@ -129,7 +129,7 @@ int pci_claim_resource(struct pci_dev *d
 #ifdef CONFIG_PCI_REASSIGN
 void pci_disable_bridge_window(struct pci_dev *dev)
 {
-       printk(KERN_DEBUG "PCI: Disable bridge window on %s\n", pci_name(dev));
+       dev_dbg(&dev->dev, "disable bridge window\n");
 
        /* MMIO Base/Limit */
        pci_write_config_dword(dev, PCI_MEMORY_BASE, 0x0000fff0);
@@ -190,8 +190,8 @@ int pci_assign_resource(struct pci_dev *
                res->flags &= ~IORESOURCE_STARTALIGN;
                if (resno < PCI_BRIDGE_RESOURCES) {
 #ifdef CONFIG_PCI_REASSIGN
-                       printk(KERN_DEBUG "PCI: Assign resource(%d) on %s "
-                               "%016llx - %016llx\n", resno, pci_name(dev),
+                       dev_dbg(&dev->dev, "assign resource(%d) "
+                               "%016llx - %016llx\n", resno,
                                (unsigned long long)res->start,
                                (unsigned long long)res->end);
 #endif
@@ -235,8 +235,8 @@ int pci_assign_resource_fixed(struct pci
                        (unsigned long long)res->end);
        } else if (resno < PCI_BRIDGE_RESOURCES) {
 #ifdef CONFIG_PCI_REASSIGN
-               printk(KERN_DEBUG "PCI: Assign resource(%d) on %s "
-                       "%016llx - %016llx\n", resno, pci_name(dev),
+               dev_dbg(&dev->dev, "assign resource(%d) "
+                       "%016llx - %016llx\n", resno,
                        (unsigned long long)res->start,
                        (unsigned long long)res->end);
 #endif
Index: head-2008-12-01/drivers/xen/Makefile
===================================================================
--- head-2008-12-01.orig/drivers/xen/Makefile   2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/drivers/xen/Makefile        2008-12-01 11:49:07.000000000 +0100
@@ -1,4 +1,4 @@
-obj-$(CONFIG_PARAVIRT_XEN)     += grant-table.o features.o events.o
+obj-$(CONFIG_PARAVIRT_XEN)     += grant-table.o features.o events.o manage.o
 xen-xencomm-$(CONFIG_PARAVIRT_XEN) := xencomm.o
 xen-balloon-$(CONFIG_PARAVIRT_XEN) := balloon.o
 
Index: head-2008-12-01/drivers/xen/balloon/sysfs.c
===================================================================
--- head-2008-12-01.orig/drivers/xen/balloon/sysfs.c    2008-12-01 11:37:10.000000000 +0100
+++ head-2008-12-01/drivers/xen/balloon/sysfs.c 2008-12-01 11:49:07.000000000 +0100
@@ -45,6 +45,7 @@
 
 #define BALLOON_SHOW(name, format, args...)                    \
        static ssize_t show_##name(struct sys_device *dev,      \
+                                  struct sysdev_attribute *attr, \
                                   char *buf)                   \
        {                                                       \
                return sprintf(buf, format, ##args);            \
@@ -59,14 +60,15 @@ BALLOON_SHOW(hard_limit_kb,
             (bs.hard_limit!=~0UL) ? PAGES2KB(bs.hard_limit) : 0);
 BALLOON_SHOW(driver_kb, "%lu\n", PAGES2KB(bs.driver_pages));
 
-static ssize_t show_target_kb(struct sys_device *dev, char *buf)
+static ssize_t show_target_kb(struct sys_device *dev,
+                             struct sysdev_attribute *attr, char *buf)
 {
        return sprintf(buf, "%lu\n", PAGES2KB(bs.target_pages));
 }
 
 static ssize_t store_target_kb(struct sys_device *dev,
-                              const char *buf,
-                              size_t count)
+                              struct sysdev_attribute *attr,
+                              const char *buf, size_t count)
 {
        char memstring[64], *endchar;
        unsigned long long target_bytes;
Index: head-2008-12-01/drivers/xen/blktap/blktap.c
===================================================================
--- head-2008-12-01.orig/drivers/xen/blktap/blktap.c    2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/drivers/xen/blktap/blktap.c 2008-12-01 11:49:07.000000000 +0100
@@ -54,6 +54,7 @@
 #include <linux/gfp.h>
 #include <linux/poll.h>
 #include <linux/delay.h>
+#include <linux/nsproxy.h>
 #include <asm/tlbflush.h>
 
 #define MAX_TAP_DEV 256     /*the maximum number of tapdisk ring devices    */
@@ -483,7 +484,7 @@ found:
 
                if ((class = get_xen_class()) != NULL)
                        device_create(class, NULL, MKDEV(blktap_major, minor),
-                                     "blktap%d", minor);
+                                     NULL, "blktap%d", minor);
        }
 
 out:
@@ -1686,7 +1687,8 @@ static int __init blkif_init(void)
                 * We only create the device when a request of a new device is
                 * made.
                 */
-               device_create(class, NULL, MKDEV(blktap_major, 0), "blktap0");
+               device_create(class, NULL, MKDEV(blktap_major, 0), NULL,
+                             "blktap0");
        } else {
                /* this is bad, but not fatal */
                WPRINTK("blktap: sysfs xen_class not created\n");
Index: head-2008-12-01/drivers/xen/char/mem.c
===================================================================
--- head-2008-12-01.orig/drivers/xen/char/mem.c 2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/drivers/xen/char/mem.c      2008-12-01 11:49:07.000000000 +0100
@@ -35,7 +35,7 @@ static inline int uncached_access(struct
 
 static inline int range_is_allowed(unsigned long pfn, unsigned long size)
 {
-#ifdef CONFIG_NONPROMISC_DEVMEM
+#ifdef CONFIG_STRICT_DEVMEM
        u64 from = ((u64)pfn) << PAGE_SHIFT;
        u64 to = from + size;
        u64 cursor = from;
@@ -172,7 +172,10 @@ static void mmap_mem_close(struct vm_are
 
 static struct vm_operations_struct mmap_mem_ops = {
        .open  = mmap_mem_open,
-       .close = mmap_mem_close
+       .close = mmap_mem_close,
+#ifdef CONFIG_HAVE_IOREMAP_PROT
+       .access = generic_access_phys
+#endif
 };
 
 static int xen_mmap_mem(struct file * file, struct vm_area_struct * vma)
Index: head-2008-12-01/drivers/xen/console/console.c
===================================================================
--- head-2008-12-01.orig/drivers/xen/console/console.c  2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/drivers/xen/console/console.c       2008-12-01 11:49:07.000000000 +0100
@@ -416,9 +416,7 @@ static void __xencons_tx_flush(void)
 
        if (work_done && (xencons_tty != NULL)) {
                wake_up_interruptible(&xencons_tty->write_wait);
-               if ((xencons_tty->flags & (1 << TTY_DO_WRITE_WAKEUP)) &&
-                   (xencons_tty->ldisc.write_wakeup != NULL))
-                       (xencons_tty->ldisc.write_wakeup)(xencons_tty);
+               tty_wakeup(xencons_tty);
        }
 }
 
@@ -619,8 +617,8 @@ static void xencons_close(struct tty_str
        tty->closing = 1;
        tty_wait_until_sent(tty, 0);
        tty_driver_flush_buffer(tty);
-       if (tty->ldisc.flush_buffer != NULL)
-               tty->ldisc.flush_buffer(tty);
+       if (tty->ldisc.ops->flush_buffer != NULL)
+               tty->ldisc.ops->flush_buffer(tty);
        tty->closing = 0;
        spin_lock_irqsave(&xencons_lock, flags);
        xencons_tty = NULL;
Index: head-2008-12-01/drivers/xen/core/evtchn.c
===================================================================
--- head-2008-12-01.orig/drivers/xen/core/evtchn.c      2008-12-01 11:37:10.000000000 +0100
+++ head-2008-12-01/drivers/xen/core/evtchn.c   2008-12-03 15:53:53.000000000 +0100
@@ -744,9 +744,9 @@ static struct irq_chip dynirq_chip = {
 };
 
 /* Bitmap indicating which PIRQs require Xen to be notified on unmask. */
-static int pirq_eoi_does_unmask;
+static bool pirq_eoi_does_unmask;
 static DECLARE_BITMAP(pirq_needs_eoi, ALIGN(NR_PIRQS, PAGE_SIZE * 8))
-       __attribute__ ((__section__(".bss.page_aligned"), __aligned__(PAGE_SIZE)));
+       __page_aligned_bss;
 
 static void pirq_unmask_and_notify(unsigned int evtchn, unsigned int irq)
 {
@@ -1002,6 +1002,7 @@ void xen_poll_irq(int irq)
                BUG();
 }
 
+#ifdef CONFIG_PM_SLEEP
 static void restore_cpu_virqs(unsigned int cpu)
 {
        struct evtchn_bind_virq bind_virq;
@@ -1094,6 +1095,7 @@ void irq_resume(void)
        }
 
 }
+#endif
 
 #if defined(CONFIG_X86_IO_APIC)
 #define identity_mapped_irq(irq) (!IO_APIC_IRQ((irq) - PIRQ_BASE))
@@ -1175,7 +1177,7 @@ void __init xen_init_IRQ(void)
        BUG_ON(!bitmap_empty(pirq_needs_eoi, PAGE_SIZE * 8));
        eoi_mfn.mfn = virt_to_bus(pirq_needs_eoi) >> PAGE_SHIFT;
        if (HYPERVISOR_physdev_op(PHYSDEVOP_pirq_eoi_mfn, &eoi_mfn) == 0)
-               pirq_eoi_does_unmask = 1;
+               pirq_eoi_does_unmask = true;
 
        /* No event channels are 'live' right now. */
        for (i = 0; i < NR_EVENT_CHANNELS; i++)
Index: head-2008-12-01/drivers/xen/core/gnttab.c
===================================================================
--- head-2008-12-01.orig/drivers/xen/core/gnttab.c      2008-12-03 15:48:43.000000000 +0100
+++ head-2008-12-01/drivers/xen/core/gnttab.c   2008-12-02 09:26:17.000000000 +0100
@@ -449,6 +449,7 @@ static int map_pte_fn(pte_t *pte, struct
        return 0;
 }
 
+#ifdef CONFIG_PM_SLEEP
 static int unmap_pte_fn(pte_t *pte, struct page *pmd_page,
                        unsigned long addr, void *data)
 {
@@ -456,6 +457,7 @@ static int unmap_pte_fn(pte_t *pte, stru
        set_pte_at(&init_mm, addr, pte, __pte(0));
        return 0;
 }
+#endif
 
 void *arch_gnttab_alloc_shared(unsigned long *frames)
 {
@@ -633,6 +635,75 @@ void __gnttab_dma_map_page(struct page *
        } while (unlikely(read_seqretry(&gnttab_dma_lock, seq)));
 }
 
+#ifdef __HAVE_ARCH_PTE_SPECIAL
+
+static unsigned int GNTMAP_pte_special;
+
+bool gnttab_pre_map_adjust(unsigned int cmd, struct gnttab_map_grant_ref *map,
+                          unsigned int count)
+{
+       unsigned int i;
+
+       if (unlikely(cmd != GNTTABOP_map_grant_ref))
+               count = 0;
+
+       for (i = 0; i < count; ++i, ++map) {
+               if (!(map->flags & GNTMAP_host_map)
+                   || !(map->flags & GNTMAP_application_map))
+                       continue;
+               if (GNTMAP_pte_special)
+                       map->flags |= GNTMAP_pte_special;
+               else {
+                       BUG_ON(xen_feature(XENFEAT_auto_translated_physmap));
+                       return true;
+               }
+       }
+
+       return false;
+}
+EXPORT_SYMBOL(gnttab_pre_map_adjust);
+
+#if CONFIG_XEN_COMPAT < 0x030400
+int gnttab_post_map_adjust(const struct gnttab_map_grant_ref *map, unsigned int count)
+{
+       unsigned int i;
+       int rc = 0;
+
+       for (i = 0; i < count && rc == 0; ++i, ++map) {
+               pte_t pte;
+
+               if (!(map->flags & GNTMAP_host_map)
+                   || !(map->flags & GNTMAP_application_map))
+                       continue;
+
+#ifdef CONFIG_X86
+               pte = __pte_ma((map->dev_bus_addr | _PAGE_PRESENT | _PAGE_USER
+                               | _PAGE_ACCESSED | _PAGE_DIRTY | _PAGE_NX
+                               | _PAGE_SPECIAL)
+                              & __supported_pte_mask);
+#else
+#error Architecture not yet supported.
+#endif
+               if (!(map->flags & GNTMAP_readonly))
+                       pte = pte_mkwrite(pte);
+
+               if (map->flags & GNTMAP_contains_pte) {
+                       mmu_update_t u;
+
+                       u.ptr = map->host_addr;
+                       u.val = __pte_val(pte);
+                       rc = HYPERVISOR_mmu_update(&u, 1, NULL, DOMID_SELF);
+               } else
+                       rc = HYPERVISOR_update_va_mapping(map->host_addr, pte, 0);
+       }
+
+       return rc;
+}
+EXPORT_SYMBOL(gnttab_post_map_adjust);
+#endif
+
+#endif /* __HAVE_ARCH_PTE_SPECIAL */
+
 int gnttab_resume(void)
 {
        if (max_nr_grant_frames() < nr_grant_frames)
@@ -640,6 +711,7 @@ int gnttab_resume(void)
        return gnttab_map(0, nr_grant_frames - 1);
 }
 
+#ifdef CONFIG_PM_SLEEP
 int gnttab_suspend(void)
 {
 #ifdef CONFIG_X86
@@ -649,6 +721,7 @@ int gnttab_suspend(void)
 #endif
        return 0;
 }
+#endif
 
 #else /* !CONFIG_XEN */
 
@@ -759,6 +832,18 @@ int __devinit gnttab_init(void)
        gnttab_free_count = nr_init_grefs - NR_RESERVED_ENTRIES;
        gnttab_free_head  = NR_RESERVED_ENTRIES;
 
+#if defined(CONFIG_XEN) && defined(__HAVE_ARCH_PTE_SPECIAL)
+       if (!xen_feature(XENFEAT_auto_translated_physmap)
+           && xen_feature(XENFEAT_gnttab_map_avail_bits)) {
+#ifdef CONFIG_X86
+               GNTMAP_pte_special = (__pte_val(pte_mkspecial(__pte_ma(0)))
+                                     >> _PAGE_BIT_UNUSED1) << _GNTMAP_guest_avail0;
+#else
+#error Architecture not yet supported.
+#endif
+       }
+#endif
+
        return 0;
 
  ini_nomem:
Index: head-2008-12-01/drivers/xen/core/machine_kexec.c
===================================================================
--- head-2008-12-01.orig/drivers/xen/core/machine_kexec.c       2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/drivers/xen/core/machine_kexec.c    2008-12-01 11:49:07.000000000 +0100
@@ -90,7 +90,7 @@ void __init xen_machine_kexec_setup_reso
        xen_hypervisor_res.start = range.start;
        xen_hypervisor_res.end = range.start + range.size - 1;
        xen_hypervisor_res.flags = IORESOURCE_BUSY | IORESOURCE_MEM;
-#ifdef CONFIG_X86_64
+#ifdef CONFIG_X86
        insert_resource(&iomem_resource, &xen_hypervisor_res);
 #endif
 
@@ -105,7 +105,7 @@ void __init xen_machine_kexec_setup_reso
        if (range.size) {
                crashk_res.start = range.start;
                crashk_res.end = range.start + range.size - 1;
-#ifdef CONFIG_X86_64
+#ifdef CONFIG_X86
                insert_resource(&iomem_resource, &crashk_res);
 #endif
        }
@@ -152,7 +152,7 @@ void __init xen_machine_kexec_setup_reso
        return;
 }
 
-#ifndef CONFIG_X86_64
+#ifndef CONFIG_X86
 void __init xen_machine_kexec_register_resources(struct resource *res)
 {
        request_resource(res, &xen_hypervisor_res);
Index: head-2008-12-01/drivers/xen/core/machine_reboot.c
===================================================================
--- head-2008-12-01.orig/drivers/xen/core/machine_reboot.c      2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/drivers/xen/core/machine_reboot.c   2008-12-01 11:49:07.000000000 +0100
@@ -65,6 +65,7 @@ EXPORT_SYMBOL(machine_restart);
 EXPORT_SYMBOL(machine_halt);
 EXPORT_SYMBOL(machine_power_off);
 
+#ifdef CONFIG_PM_SLEEP
 static void pre_suspend(void)
 {
        HYPERVISOR_shared_info = (shared_info_t *)empty_zero_page;
@@ -119,6 +120,7 @@ static void post_suspend(int suspend_can
        HYPERVISOR_shared_info->arch.pfn_to_mfn_frame_list_list =
                virt_to_mfn(pfn_to_mfn_frame_list_list);
 }
+#endif
 
 #else /* !(defined(__i386__) || defined(__x86_64__)) */
 
@@ -137,6 +139,7 @@ static void post_suspend(int suspend_can
 
 #endif
 
+#ifdef CONFIG_PM_SLEEP
 struct suspend {
        int fast_suspend;
        void (*resume_notifier)(int);
@@ -230,7 +233,8 @@ int __xen_suspend(int fast_suspend, void
 
        if (fast_suspend) {
                xenbus_suspend();
-               err = stop_machine_run(take_machine_down, &suspend, 0);
+               err = stop_machine(take_machine_down, &suspend,
+                                  &cpumask_of_cpu(0));
                if (err < 0)
                        xenbus_suspend_cancel();
        } else {
@@ -253,3 +257,4 @@ int __xen_suspend(int fast_suspend, void
 
        return 0;
 }
+#endif
Index: head-2008-12-01/drivers/xen/core/reboot.c
===================================================================
--- head-2008-12-01.orig/drivers/xen/core/reboot.c      2008-12-01 11:36:47.000000000 +0100
+++ head-2008-12-01/drivers/xen/core/reboot.c   2008-12-01 11:49:07.000000000 +0100
@@ -29,17 +29,12 @@ MODULE_LICENSE("Dual BSD/GPL");
 /* Ignore multiple shutdown requests. */
 static int shutting_down = SHUTDOWN_INVALID;
 
-/* Was last suspend request cancelled? */
-static int suspend_cancelled;
-
 /* Can we leave APs online when we suspend? */
 static int fast_suspend;
 
 static void __shutdown_handler(struct work_struct *unused);
 static DECLARE_DELAYED_WORK(shutdown_work, __shutdown_handler);
 
-static int setup_suspend_evtchn(void);
-
 int __xen_suspend(int fast_suspend, void (*resume_notifier)(int));
 
 static int shutdown_process(void *__unused)
@@ -69,6 +64,13 @@ static int shutdown_process(void *__unus
        return 0;
 }
 
+#ifdef CONFIG_PM_SLEEP
+
+static int setup_suspend_evtchn(void);
+
+/* Was last suspend request cancelled? */
+static int suspend_cancelled;
+
 static void xen_resume_notifier(int _suspend_cancelled)
 {
        int old_state = xchg(&shutting_down, SHUTDOWN_RESUMING);
@@ -118,6 +120,10 @@ static int xen_suspend(void *__unused)
        return 0;
 }
 
+#else
+# define xen_suspend NULL
+#endif
+
 static void switch_shutdown_state(int new_state)
 {
        int prev_state, old_state = SHUTDOWN_INVALID;
@@ -194,8 +200,10 @@ static void shutdown_handler(struct xenb
                new_state = SHUTDOWN_POWEROFF;
        else if (strcmp(str, "reboot") == 0)
                ctrl_alt_del();
+#ifdef CONFIG_PM_SLEEP
        else if (strcmp(str, "suspend") == 0)
                new_state = SHUTDOWN_SUSPEND;
+#endif
        else if (strcmp(str, "halt") == 0)
                new_state = SHUTDOWN_HALT;
        else
@@ -247,6 +255,7 @@ static struct xenbus_watch sysrq_watch =
        .callback = sysrq_handler
 };
 
+#ifdef CONFIG_PM_SLEEP
 static irqreturn_t suspend_int(int irq, void* dev_id)
 {
        switch_shutdown_state(SHUTDOWN_SUSPEND);
@@ -274,6 +283,9 @@ static int setup_suspend_evtchn(void)
 
        return 0;
 }
+#else
+#define setup_suspend_evtchn() 0
+#endif
 
 static int setup_shutdown_watcher(void)
 {
Index: head-2008-12-01/drivers/xen/core/smpboot.c
===================================================================
--- head-2008-12-01.orig/drivers/xen/core/smpboot.c     2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/drivers/xen/core/smpboot.c  2008-12-01 11:49:07.000000000 +0100
@@ -27,6 +27,7 @@
 
 extern irqreturn_t smp_reschedule_interrupt(int, void *);
 extern irqreturn_t smp_call_function_interrupt(int, void *);
+extern irqreturn_t smp_call_function_single_interrupt(int, void *);
 
 extern int local_setup_timer(unsigned int cpu);
 extern void local_teardown_timer(unsigned int cpu);
@@ -54,8 +55,10 @@ DEFINE_PER_CPU(int, cpu_state) = { 0 };
 
 static DEFINE_PER_CPU(int, resched_irq);
 static DEFINE_PER_CPU(int, callfunc_irq);
+static DEFINE_PER_CPU(int, call1func_irq);
 static char resched_name[NR_CPUS][15];
 static char callfunc_name[NR_CPUS][15];
+static char call1func_name[NR_CPUS][15];
 
 #ifdef CONFIG_X86_LOCAL_APIC
 #define set_cpu_to_apicid(cpu, apicid) (per_cpu(x86_cpu_to_apicid, cpu) = (apicid))
@@ -77,8 +80,10 @@ void __init prefill_possible_map(void)
 
        for (i = 0; i < NR_CPUS; i++) {
                rc = HYPERVISOR_vcpu_op(VCPUOP_is_up, i, NULL);
-               if (rc >= 0)
+               if (rc >= 0) {
                        cpu_set(i, cpu_possible_map);
+                       nr_cpu_ids = i + 1;
+               }
        }
 }
 
@@ -114,7 +119,8 @@ static int __cpuinit xen_smp_intr_init(u
 {
        int rc;
 
-       per_cpu(resched_irq, cpu) = per_cpu(callfunc_irq, cpu) = -1;
+       per_cpu(resched_irq, cpu) = per_cpu(callfunc_irq, cpu) =
+               per_cpu(call1func_irq, cpu) = -1;
 
        sprintf(resched_name[cpu], "resched%u", cpu);
        rc = bind_ipi_to_irqhandler(RESCHEDULE_VECTOR,
@@ -138,6 +144,17 @@ static int __cpuinit xen_smp_intr_init(u
                goto fail;
        per_cpu(callfunc_irq, cpu) = rc;
 
+       sprintf(call1func_name[cpu], "call1func%u", cpu);
+       rc = bind_ipi_to_irqhandler(CALL_FUNC_SINGLE_VECTOR,
+                                   cpu,
+                                   smp_call_function_single_interrupt,
+                                   IRQF_DISABLED|IRQF_NOBALANCING,
+                                   call1func_name[cpu],
+                                   NULL);
+       if (rc < 0)
+               goto fail;
+       per_cpu(call1func_irq, cpu) = rc;
+
        rc = xen_spinlock_init(cpu);
        if (rc < 0)
                goto fail;
@@ -152,6 +169,8 @@ static int __cpuinit xen_smp_intr_init(u
                unbind_from_irqhandler(per_cpu(resched_irq, cpu), NULL);
        if (per_cpu(callfunc_irq, cpu) >= 0)
                unbind_from_irqhandler(per_cpu(callfunc_irq, cpu), NULL);
+       if (per_cpu(call1func_irq, cpu) >= 0)
+               unbind_from_irqhandler(per_cpu(call1func_irq, cpu), NULL);
        xen_spinlock_cleanup(cpu);
        return rc;
 }
@@ -164,6 +183,7 @@ static void __cpuexit xen_smp_intr_exit(
 
        unbind_from_irqhandler(per_cpu(resched_irq, cpu), NULL);
        unbind_from_irqhandler(per_cpu(callfunc_irq, cpu), NULL);
+       unbind_from_irqhandler(per_cpu(call1func_irq, cpu), NULL);
        xen_spinlock_cleanup(cpu);
 }
 #endif
@@ -171,11 +191,7 @@ static void __cpuexit xen_smp_intr_exit(
 void __cpuinit cpu_bringup(void)
 {
        cpu_init();
-#ifdef __i386__
        identify_secondary_cpu(&current_cpu_data);
-#else
-       identify_cpu(&current_cpu_data);
-#endif
        touch_softlockup_watchdog();
        preempt_disable();
        local_irq_enable();
@@ -255,9 +271,6 @@ void __init smp_prepare_cpus(unsigned in
        struct task_struct *idle;
        int apicid;
        struct vcpu_get_physid cpu_id;
-#ifdef __x86_64__
-       struct desc_ptr *gdt_descr;
-#endif
        void *gdt_addr;
 
        apicid = 0;
@@ -270,7 +283,7 @@ void __init smp_prepare_cpus(unsigned in
 
        current_thread_info()->cpu = 0;
 
-       for (cpu = 0; cpu < NR_CPUS; cpu++) {
+       for_each_possible_cpu (cpu) {
                cpus_clear(per_cpu(cpu_sibling_map, cpu));
                cpus_clear(per_cpu(cpu_core_map, cpu));
        }
@@ -297,21 +310,10 @@ void __init smp_prepare_cpus(unsigned in
                if (IS_ERR(idle))
                        panic("failed fork for CPU %d", cpu);
 
-#ifdef __x86_64__
-               gdt_descr = &cpu_gdt_descr[cpu];
-               gdt_descr->address = get_zeroed_page(GFP_KERNEL);
-               if (unlikely(!gdt_descr->address)) {
-                       printk(KERN_CRIT "CPU%d failed to allocate GDT\n",
-                              cpu);
-                       continue;
-               }
-               gdt_descr->size = GDT_SIZE;
-               memcpy((void *)gdt_descr->address, cpu_gdt_table, GDT_SIZE);
-               gdt_addr = (void *)gdt_descr->address;
-#else
+#ifdef __i386__
                init_gdt(cpu);
-               gdt_addr = get_cpu_gdt_table(cpu);
 #endif
+               gdt_addr = get_cpu_gdt_table(cpu);
                make_page_readonly(gdt_addr, XENFEAT_writable_descriptor_tables);
 
                apicid = cpu;
Index: head-2008-12-01/drivers/xen/core/spinlock.c
===================================================================
--- head-2008-12-01.orig/drivers/xen/core/spinlock.c    2008-12-01 11:37:10.000000000 +0100
+++ head-2008-12-01/drivers/xen/core/spinlock.c 2008-12-01 11:51:53.000000000 +0100
@@ -73,9 +73,9 @@ int xen_spin_wait(raw_spinlock_t *lock, 
        /* announce we're spinning */
        spinning.ticket = token;
        spinning.lock = lock;
-       spinning.prev = __get_cpu_var(spinning);
+       spinning.prev = x86_read_percpu(spinning);
        smp_wmb();
-       __get_cpu_var(spinning) = &spinning;
+       x86_write_percpu(spinning, &spinning);
 
        /* clear pending */
        xen_clear_irq_pending(irq);
@@ -102,7 +102,7 @@ int xen_spin_wait(raw_spinlock_t *lock, 
        kstat_this_cpu.irqs[irq] += !rc;
 
        /* announce we're done */
-       __get_cpu_var(spinning) = spinning.prev;
+       x86_write_percpu(spinning, spinning.prev);
        rm_lock = &__get_cpu_var(spinning_rm_lock);
        raw_local_irq_save(flags);
        __raw_write_lock(rm_lock);
Index: head-2008-12-01/drivers/xen/fbfront/xenfb.c
===================================================================
--- head-2008-12-01.orig/drivers/xen/fbfront/xenfb.c    2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/drivers/xen/fbfront/xenfb.c 2008-12-01 11:49:07.000000000 +0100
@@ -18,6 +18,7 @@
  * frame buffer.
  */
 
+#include <linux/console.h>
 #include <linux/kernel.h>
 #include <linux/errno.h>
 #include <linux/fb.h>
@@ -544,6 +545,28 @@ static unsigned long vmalloc_to_mfn(void
        return pfn_to_mfn(vmalloc_to_pfn(address));
 }
 
+static __devinit void
+xenfb_make_preferred_console(void)
+{
+       struct console *c;
+
+       if (console_set_on_cmdline)
+               return;
+
+       acquire_console_sem();
+       for (c = console_drivers; c; c = c->next) {
+               if (!strcmp(c->name, "tty") && c->index == 0)
+                       break;
+       }
+       release_console_sem();
+       if (c) {
+               unregister_console(c);
+               c->flags |= CON_CONSDEV;
+               c->flags &= ~CON_PRINTBUFFER; /* don't print again */
+               register_console(c);
+       }
+}
+
 static int __devinit xenfb_probe(struct xenbus_device *dev,
                                 const struct xenbus_device_id *id)
 {
@@ -673,6 +696,7 @@ static int __devinit xenfb_probe(struct 
                goto error;
        }
 
+       xenfb_make_preferred_console();
        return 0;
 
  error_nomem:
@@ -881,4 +905,5 @@ static void __exit xenfb_cleanup(void)
 module_init(xenfb_init);
 module_exit(xenfb_cleanup);
 
+MODULE_DESCRIPTION("Xen virtual framebuffer device frontend");
 MODULE_LICENSE("GPL");
Index: head-2008-12-01/drivers/xen/fbfront/xenkbd.c
===================================================================
--- head-2008-12-01.orig/drivers/xen/fbfront/xenkbd.c   2008-12-01 11:36:07.000000000 +0100
+++ head-2008-12-01/drivers/xen/fbfront/xenkbd.c        2008-12-01 11:49:07.000000000 +0100
@@ -350,4 +350,5 @@ static void __exit xenkbd_cleanup(void)
 module_init(xenkbd_init);
 module_exit(xenkbd_cleanup);
 
+MODULE_DESCRIPTION("Xen virtual keyboard/pointer device frontend");
 MODULE_LICENSE("GPL");
Index: head-2008-12-01/drivers/xen/gntdev/gntdev.c
===================================================================
--- head-2008-12-01.orig/drivers/xen/gntdev/gntdev.c    2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/drivers/xen/gntdev/gntdev.c 2008-12-01 11:49:07.000000000 +0100
@@ -418,7 +418,7 @@ static int __init gntdev_init(void)
        }
 
        device = device_create(class, NULL, MKDEV(gntdev_major, 0),
-                              GNTDEV_NAME);
+                              NULL, GNTDEV_NAME);
        if (IS_ERR(device)) {
                printk(KERN_ERR "Error creating gntdev device in xen_class\n");
                printk(KERN_ERR "gntdev created with major number = %d\n",
Index: head-2008-12-01/drivers/xen/netfront/accel.c
===================================================================
--- head-2008-12-01.orig/drivers/xen/netfront/accel.c   2008-12-01 11:36:55.000000000 +0100
+++ head-2008-12-01/drivers/xen/netfront/accel.c        2008-12-01 11:49:07.000000000 +0100
@@ -28,6 +28,7 @@
  * IN THE SOFTWARE.
  */
 
+#include <linux/version.h>
 #include <linux/netdevice.h>
 #include <linux/skbuff.h>
 #include <linux/list.h>
Index: head-2008-12-01/drivers/xen/netfront/netfront.c
===================================================================
--- head-2008-12-01.orig/drivers/xen/netfront/netfront.c        2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/drivers/xen/netfront/netfront.c     2008-12-01 11:49:07.000000000 +0100
@@ -640,7 +640,7 @@ static int network_open(struct net_devic
        }
        spin_unlock_bh(&np->rx_lock);
 
-       network_maybe_wake_tx(dev);
+       netif_start_queue(dev);
 
        return 0;
 }
Index: head-2008-12-01/drivers/xen/sfc_netback/accel.h
===================================================================
--- head-2008-12-01.orig/drivers/xen/sfc_netback/accel.h        2008-12-03 15:48:43.000000000 +0100
+++ head-2008-12-01/drivers/xen/sfc_netback/accel.h     2008-12-01 11:49:07.000000000 +0100
@@ -25,6 +25,7 @@
 #ifndef NETBACK_ACCEL_H
 #define NETBACK_ACCEL_H
 
+#include <linux/version.h>
 #include <linux/slab.h>
 #include <linux/ip.h>
 #include <linux/tcp.h>
Index: head-2008-12-01/drivers/xen/sfc_netfront/accel.h
===================================================================
--- head-2008-12-01.orig/drivers/xen/sfc_netfront/accel.h       2008-12-01 11:29:05.000000000 +0100
+++ head-2008-12-01/drivers/xen/sfc_netfront/accel.h    2008-12-01 11:49:07.000000000 +0100
@@ -35,6 +35,7 @@
 #include <xen/evtchn.h>
 
 #include <linux/kernel.h>
+#include <linux/version.h>
 #include <linux/list.h>
 
 enum netfront_accel_post_status {
Index: head-2008-12-01/drivers/xen/xenbus/xenbus_comms.c
===================================================================
--- head-2008-12-01.orig/drivers/xen/xenbus/xenbus_comms.c      2008-12-01 11:36:47.000000000 +0100
+++ head-2008-12-01/drivers/xen/xenbus/xenbus_comms.c   2008-12-01 11:49:07.000000000 +0100
@@ -228,14 +228,11 @@ int xb_init_comms(void)
                intf->rsp_cons = intf->rsp_prod;
        }
 
+#if defined(CONFIG_XEN) || defined(MODULE)
        if (xenbus_irq)
                unbind_from_irqhandler(xenbus_irq, &xb_waitq);
 
-#if defined(CONFIG_XEN) || defined(MODULE)
        err = bind_caller_port_to_irqhandler(
-#else
-       err = bind_evtchn_to_irqhandler(
-#endif
                xen_store_evtchn, wake_waiting,
                0, "xenbus", &xb_waitq);
        if (err <= 0) {
@@ -244,6 +241,20 @@ int xb_init_comms(void)
        }
 
        xenbus_irq = err;
+#else
+       if (xenbus_irq) {
+               /* Already have an irq; assume we're resuming */
+               rebind_evtchn_irq(xen_store_evtchn, xenbus_irq);
+       } else {
+               err = bind_evtchn_to_irqhandler(xen_store_evtchn, wake_waiting,
+                                               0, "xenbus", &xb_waitq);
+               if (err <= 0) {
+                       printk(KERN_ERR "XENBUS request irq failed %i\n", err);
+                       return err;
+               }
+               xenbus_irq = err;
+       }
+#endif
 
        return 0;
 }
Index: head-2008-12-01/drivers/xen/xenbus/xenbus_probe.c
===================================================================
--- head-2008-12-01.orig/drivers/xen/xenbus/xenbus_probe.c      2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/drivers/xen/xenbus/xenbus_probe.c   2008-12-01 11:49:07.000000000 +0100
@@ -36,6 +36,7 @@
                 __FUNCTION__, __LINE__, ##args)
 
 #include <linux/kernel.h>
+#include <linux/version.h>
 #include <linux/err.h>
 #include <linux/string.h>
 #include <linux/ctype.h>
Index: head-2008-12-01/fs/aio.c
===================================================================
--- head-2008-12-01.orig/fs/aio.c       2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/fs/aio.c    2008-12-01 11:49:07.000000000 +0100
@@ -1319,7 +1319,7 @@ static int make_aio_fd(struct kioctx *io
        int fd;
        struct file *file;
 
-       fd = anon_inode_getfd("[aioq]", &aioq_fops, ioctx);
+       fd = anon_inode_getfd("[aioq]", &aioq_fops, ioctx, 0);
        if (fd < 0)
                return fd;
 
Index: head-2008-12-01/include/asm-generic/pgtable.h
===================================================================
--- head-2008-12-01.orig/include/asm-generic/pgtable.h  2008-12-01 11:29:05.000000000 +0100
+++ head-2008-12-01/include/asm-generic/pgtable.h       2008-12-01 11:49:07.000000000 +0100
@@ -99,10 +99,6 @@ static inline void ptep_set_wrprotect(st
 }
 #endif
 
-#ifndef arch_change_pte_range
-#define arch_change_pte_range(mm, pmd, addr, end, newprot, dirty_accountable) 0
-#endif
-
 #ifndef __HAVE_ARCH_PTE_SAME
 #define pte_same(A,B)  (pte_val(A) == pte_val(B))
 #endif
Index: head-2008-12-01/include/asm-x86/dma-mapping.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/dma-mapping.h  2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/dma-mapping.h       2008-12-01 11:49:07.000000000 +0100
@@ -74,7 +74,7 @@ static inline struct dma_mapping_ops *ge
 /* Make sure we keep the same behaviour */
 static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
 {
-#ifdef CONFIG_X86_32
+#if defined(CONFIG_X86_32) && !defined(CONFIG_XEN)
        return 0;
 #else
        struct dma_mapping_ops *ops = get_dma_ops(dev);
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/desc.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/desc.h    2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/desc.h 2008-12-01 11:49:07.000000000 +0100
@@ -31,11 +31,17 @@ extern struct desc_ptr idt_descr;
 extern gate_desc idt_table[];
 #endif
 
+struct gdt_page {
+       struct desc_struct gdt[GDT_ENTRIES];
+} __attribute__((aligned(PAGE_SIZE)));
+DECLARE_PER_CPU(struct gdt_page, gdt_page);
+
+static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu)
+{
+       return per_cpu(gdt_page, cpu).gdt;
+}
+
 #ifdef CONFIG_X86_64
-extern struct desc_struct cpu_gdt_table[GDT_ENTRIES];
-extern struct desc_ptr cpu_gdt_descr[];
-/* the cpu gdt accessor */
-#define get_cpu_gdt_table(x) ((struct desc_struct *)cpu_gdt_descr[x].address)
 
 static inline void pack_gate(gate_desc *gate, unsigned type, unsigned long func,
                             unsigned dpl, unsigned ist, unsigned seg)
@@ -53,16 +59,6 @@ static inline void pack_gate(gate_desc *
 }
 
 #else
-struct gdt_page {
-       struct desc_struct gdt[GDT_ENTRIES];
-} __attribute__((aligned(PAGE_SIZE)));
-DECLARE_PER_CPU(struct gdt_page, gdt_page);
-
-static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu)
-{
-       return per_cpu(gdt_page, cpu).gdt;
-}
-
 static inline void pack_gate(gate_desc *gate, unsigned char type,
                             unsigned long base, unsigned dpl, unsigned flags,
                             unsigned short seg)
@@ -333,6 +329,28 @@ static inline void set_intr_gate(unsigne
        _set_gate(n, GATE_INTERRUPT, addr, 0, 0, __KERNEL_CS);
 }
 
+#define SYS_VECTOR_FREE                0
+#define SYS_VECTOR_ALLOCED     1
+
+extern int first_system_vector;
+extern char system_vectors[];
+
+static inline void alloc_system_vector(int vector)
+{
+       if (system_vectors[vector] == SYS_VECTOR_FREE) {
+               system_vectors[vector] = SYS_VECTOR_ALLOCED;
+               if (first_system_vector > vector)
+                       first_system_vector = vector;
+       } else
+               BUG();
+}
+
+static inline void alloc_intr_gate(unsigned int n, void *addr)
+{
+       alloc_system_vector(n);
+       set_intr_gate(n, addr);
+}
+
 /*
  * This routine sets up an interrupt gate at directory privilege level 3.
  */
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/fixmap.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/fixmap.h  2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/fixmap.h       2008-12-01 11:49:07.000000000 +0100
@@ -7,7 +7,58 @@
 # include "fixmap_64.h"
 #endif
 
+extern int fixmaps_set;
+
+void xen_set_fixmap(enum fixed_addresses, maddr_t, pgprot_t);
+
+static inline void __set_fixmap(enum fixed_addresses idx,
+                               maddr_t phys, pgprot_t flags)
+{
+       xen_set_fixmap(idx, phys, flags);
+}
+
+#define set_fixmap(idx, phys)                          \
+       __set_fixmap(idx, phys, PAGE_KERNEL)
+
+/*
+ * Some hardware wants to get fixmapped without caching.
+ */
+#define set_fixmap_nocache(idx, phys)                  \
+       __set_fixmap(idx, phys, PAGE_KERNEL_NOCACHE)
+
 #define clear_fixmap(idx)                      \
        __set_fixmap(idx, 0, __pgprot(0))
 
+#define __fix_to_virt(x)       (FIXADDR_TOP - ((x) << PAGE_SHIFT))
+#define __virt_to_fix(x)       ((FIXADDR_TOP - ((x)&PAGE_MASK)) >> PAGE_SHIFT)
+
+extern void __this_fixmap_does_not_exist(void);
+
+/*
+ * 'index to address' translation. If anyone tries to use the idx
+ * directly without translation, we catch the bug with a NULL-deference
+ * kernel oops. Illegal ranges of incoming indices are caught too.
+ */
+static __always_inline unsigned long fix_to_virt(const unsigned int idx)
+{
+       /*
+        * this branch gets completely eliminated after inlining,
+        * except when someone tries to use fixaddr indices in an
+        * illegal way. (such as mixing up address types or using
+        * out-of-range indices).
+        *
+        * If it doesn't get removed, the linker will complain
+        * loudly with a reasonably clear error message..
+        */
+       if (idx >= __end_of_fixed_addresses)
+               __this_fixmap_does_not_exist();
+
+       return __fix_to_virt(idx);
+}
+
+static inline unsigned long virt_to_fix(const unsigned long vaddr)
+{
+       BUG_ON(vaddr >= FIXADDR_TOP || vaddr < FIXADDR_START);
+       return __virt_to_fix(vaddr);
+}
 #endif
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/fixmap_32.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/fixmap_32.h       2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/fixmap_32.h    2008-12-01 11:49:07.000000000 +0100
@@ -58,10 +58,17 @@ enum fixed_addresses {
 #ifdef CONFIG_X86_LOCAL_APIC
        FIX_APIC_BASE,  /* local (CPU) APIC) -- required for SMP or not */
 #endif
-#if defined(CONFIG_X86_IO_APIC) && !defined(CONFIG_XEN)
+#ifndef CONFIG_XEN
+#ifdef CONFIG_X86_IO_APIC
        FIX_IO_APIC_BASE_0,
        FIX_IO_APIC_BASE_END = FIX_IO_APIC_BASE_0 + MAX_IO_APICS-1,
 #endif
+#else
+       FIX_SHARED_INFO,
+#define NR_FIX_ISAMAPS 256
+       FIX_ISAMAP_END,
+       FIX_ISAMAP_BEGIN = FIX_ISAMAP_END + NR_FIX_ISAMAPS - 1,
+#endif
 #ifdef CONFIG_X86_VISWS_APIC
        FIX_CO_CPU,     /* Cobalt timer */
        FIX_CO_APIC,    /* Cobalt APIC Redirection Table */
@@ -78,51 +85,38 @@ enum fixed_addresses {
        FIX_KMAP_BEGIN, /* reserved pte's for temporary kernel mappings */
        FIX_KMAP_END = FIX_KMAP_BEGIN+(KM_TYPE_NR*NR_CPUS)-1,
 #endif
-#ifdef CONFIG_ACPI
-       FIX_ACPI_BEGIN,
-       FIX_ACPI_END = FIX_ACPI_BEGIN + FIX_ACPI_PAGES - 1,
-#endif
 #ifdef CONFIG_PCI_MMCONFIG
        FIX_PCIE_MCFG,
 #endif
 #ifdef CONFIG_PARAVIRT
        FIX_PARAVIRT_BOOTMAP,
 #endif
-       FIX_SHARED_INFO,
-#define NR_FIX_ISAMAPS 256
-       FIX_ISAMAP_END,
-       FIX_ISAMAP_BEGIN = FIX_ISAMAP_END + NR_FIX_ISAMAPS - 1,
        __end_of_permanent_fixed_addresses,
        /*
         * 256 temporary boot-time mappings, used by early_ioremap(),
         * before ioremap() is functional.
         *
-        * We round it up to the next 512 pages boundary so that we
+        * We round it up to the next 256 pages boundary so that we
         * can have a single pgd entry and a single pte table:
         */
 #define NR_FIX_BTMAPS          64
 #define FIX_BTMAPS_NESTING     4
-       FIX_BTMAP_END = __end_of_permanent_fixed_addresses + 512 -
-                       (__end_of_permanent_fixed_addresses & 511),
+       FIX_BTMAP_END = __end_of_permanent_fixed_addresses + 256 -
+                       (__end_of_permanent_fixed_addresses & 255),
        FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_NESTING - 1,
        FIX_WP_TEST,
+#ifdef CONFIG_ACPI
+       FIX_ACPI_BEGIN,
+       FIX_ACPI_END = FIX_ACPI_BEGIN + FIX_ACPI_PAGES - 1,
+#endif
 #ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT
        FIX_OHCI1394_BASE,
 #endif
        __end_of_fixed_addresses
 };
 
-extern void __set_fixmap(enum fixed_addresses idx,
-                        maddr_t phys, pgprot_t flags);
 extern void reserve_top_address(unsigned long reserve);
 
-#define set_fixmap(idx, phys)                          \
-       __set_fixmap(idx, phys, PAGE_KERNEL)
-/*
- * Some hardware wants to get fixmapped without caching.
- */
-#define set_fixmap_nocache(idx, phys)                  \
-       __set_fixmap(idx, phys, PAGE_KERNEL_NOCACHE)
 
 #define FIXADDR_TOP    ((unsigned long)__FIXADDR_TOP)
 
@@ -131,38 +125,5 @@ extern void reserve_top_address(unsigned
 #define FIXADDR_START          (FIXADDR_TOP - __FIXADDR_SIZE)
 #define FIXADDR_BOOT_START     (FIXADDR_TOP - __FIXADDR_BOOT_SIZE)
 
-#define __fix_to_virt(x)       (FIXADDR_TOP - ((x) << PAGE_SHIFT))
-#define __virt_to_fix(x)       ((FIXADDR_TOP - ((x)&PAGE_MASK)) >> PAGE_SHIFT)
-
-extern void __this_fixmap_does_not_exist(void);
-
-/*
- * 'index to address' translation. If anyone tries to use the idx
- * directly without tranlation, we catch the bug with a NULL-deference
- * kernel oops. Illegal ranges of incoming indices are caught too.
- */
-static __always_inline unsigned long fix_to_virt(const unsigned int idx)
-{
-       /*
-        * this branch gets completely eliminated after inlining,
-        * except when someone tries to use fixaddr indices in an
-        * illegal way. (such as mixing up address types or using
-        * out-of-range indices).
-        *
-        * If it doesn't get removed, the linker will complain
-        * loudly with a reasonably clear error message..
-        */
-       if (idx >= __end_of_fixed_addresses)
-               __this_fixmap_does_not_exist();
-
-       return __fix_to_virt(idx);
-}
-
-static inline unsigned long virt_to_fix(const unsigned long vaddr)
-{
-       BUG_ON(vaddr >= FIXADDR_TOP || vaddr < FIXADDR_START);
-       return __virt_to_fix(vaddr);
-}
-
 #endif /* !__ASSEMBLY__ */
 #endif
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/fixmap_64.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/fixmap_64.h       2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/fixmap_64.h    2008-12-01 11:49:07.000000000 +0100
@@ -12,6 +12,7 @@
 #define _ASM_FIXMAP_64_H
 
 #include <linux/kernel.h>
+#include <asm/acpi.h>
 #include <asm/apicdef.h>
 #include <asm/page.h>
 #include <asm/vsyscall.h>
@@ -40,7 +41,6 @@ enum fixed_addresses {
        VSYSCALL_HPET,
        FIX_DBGP_BASE,
        FIX_EARLYCON_MEM_BASE,
-       FIX_HPET_BASE,
 #ifdef CONFIG_X86_LOCAL_APIC
        FIX_APIC_BASE,  /* local (CPU) APIC) -- required for SMP or not */
 #endif
@@ -53,14 +53,21 @@ enum fixed_addresses {
        FIX_EFI_IO_MAP_FIRST_PAGE = FIX_EFI_IO_MAP_LAST_PAGE
                                  + MAX_EFI_IO_PAGES - 1,
 #endif
+#ifdef CONFIG_PARAVIRT
+       FIX_PARAVIRT_BOOTMAP,
+#else
+       FIX_SHARED_INFO,
+#endif
 #ifdef CONFIG_ACPI
        FIX_ACPI_BEGIN,
        FIX_ACPI_END = FIX_ACPI_BEGIN + FIX_ACPI_PAGES - 1,
 #endif
-       FIX_SHARED_INFO,
 #define NR_FIX_ISAMAPS 256
        FIX_ISAMAP_END,
        FIX_ISAMAP_BEGIN = FIX_ISAMAP_END + NR_FIX_ISAMAPS - 1,
+#ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT
+       FIX_OHCI1394_BASE,
+#endif
        __end_of_permanent_fixed_addresses,
        /*
         * 256 temporary boot-time mappings, used by early_ioremap(),
@@ -71,27 +78,12 @@ enum fixed_addresses {
         */
 #define NR_FIX_BTMAPS          64
 #define FIX_BTMAPS_NESTING     4
-       FIX_BTMAP_END =
-               __end_of_permanent_fixed_addresses + 512 -
+       FIX_BTMAP_END = __end_of_permanent_fixed_addresses + 512 -
                        (__end_of_permanent_fixed_addresses & 511),
        FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_NESTING - 1,
-#ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT
-       FIX_OHCI1394_BASE,
-#endif
        __end_of_fixed_addresses
 };
 
-extern void __set_fixmap(enum fixed_addresses idx,
-                        unsigned long phys, pgprot_t flags);
-
-#define set_fixmap(idx, phys)                  \
-       __set_fixmap(idx, phys, PAGE_KERNEL)
-/*
- * Some hardware wants to get fixmapped without caching.
- */
-#define set_fixmap_nocache(idx, phys)                  \
-       __set_fixmap(idx, phys, PAGE_KERNEL_NOCACHE)
-
 #define FIXADDR_TOP    (VSYSCALL_END-PAGE_SIZE)
 #define FIXADDR_SIZE   (__end_of_fixed_addresses << PAGE_SHIFT)
 #define FIXADDR_START  (FIXADDR_TOP - FIXADDR_SIZE)
@@ -100,30 +92,4 @@ extern void __set_fixmap(enum fixed_addr
 #define FIXADDR_USER_START     ((unsigned long)VSYSCALL32_VSYSCALL)
 #define FIXADDR_USER_END       (FIXADDR_USER_START + PAGE_SIZE)
 
-#define __fix_to_virt(x)       (FIXADDR_TOP - ((x) << PAGE_SHIFT))
-
-extern void __this_fixmap_does_not_exist(void);
-
-/*
- * 'index to address' translation. If anyone tries to use the idx
- * directly without translation, we catch the bug with a NULL-deference
- * kernel oops. Illegal ranges of incoming indices are caught too.
- */
-static __always_inline unsigned long fix_to_virt(const unsigned int idx)
-{
-       /*
-        * this branch gets completely eliminated after inlining,
-        * except when someone tries to use fixaddr indices in an
-        * illegal way. (such as mixing up address types or using
-        * out-of-range indices).
-        *
-        * If it doesn't get removed, the linker will complain
-        * loudly with a reasonably clear error message..
-        */
-       if (idx >= __end_of_fixed_addresses)
-               __this_fixmap_does_not_exist();
-
-       return __fix_to_virt(idx);
-}
-
 #endif
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/highmem.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/highmem.h 2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/highmem.h      2008-12-01 11:49:07.000000000 +0100
@@ -74,6 +74,9 @@ struct page *kmap_atomic_to_page(void *p
 
 #define flush_cache_kmaps()    do { } while (0)
 
+extern void add_highpages_with_active_regions(int nid, unsigned long start_pfn,
+                                       unsigned long end_pfn);
+
 void clear_highpage(struct page *);
 static inline void clear_user_highpage(struct page *page, unsigned long vaddr)
 {
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/hypercall.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/hypercall.h       2008-12-01 11:36:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/hypercall.h    2008-12-01 11:49:07.000000000 +0100
@@ -323,9 +323,19 @@ static inline int __must_check
 HYPERVISOR_grant_table_op(
        unsigned int cmd, void *uop, unsigned int count)
 {
+       bool fixup = false;
+       int rc;
+
        if (arch_use_lazy_mmu_mode())
                xen_multicall_flush(false);
-       return _hypercall3(int, grant_table_op, cmd, uop, count);
+#ifdef GNTTABOP_map_grant_ref
+       if (cmd == GNTTABOP_map_grant_ref)
+#endif
+               fixup = gnttab_pre_map_adjust(cmd, uop, count);
+       rc = _hypercall3(int, grant_table_op, cmd, uop, count);
+       if (rc == 0 && fixup)
+               rc = gnttab_post_map_adjust(uop, count);
+       return rc;
 }
 
 static inline int __must_check
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/hypervisor.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/hypervisor.h      2008-12-01 11:37:10.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/hypervisor.h   2008-12-01 11:49:07.000000000 +0100
@@ -35,7 +35,6 @@
 
 #include <linux/types.h>
 #include <linux/kernel.h>
-#include <linux/version.h>
 #include <linux/errno.h>
 #include <xen/interface/xen.h>
 #include <xen/interface/platform.h>
@@ -171,6 +170,20 @@ static inline void arch_flush_lazy_mmu_m
 }
 #endif
 
+struct gnttab_map_grant_ref;
+bool gnttab_pre_map_adjust(unsigned int cmd, struct gnttab_map_grant_ref *,
+                          unsigned int count);
+#if CONFIG_XEN_COMPAT < 0x030400
+int gnttab_post_map_adjust(const struct gnttab_map_grant_ref *, unsigned int);
+#else
+static inline int gnttab_post_map_adjust(const struct gnttab_map_grant_ref *m,
+                                        unsigned int count)
+{
+       BUG();
+       return -ENOSYS;
+}
+#endif
+
 #else /* CONFIG_XEN */
 
 static inline void xen_multicall_flush(bool ignore) {}
@@ -179,6 +192,9 @@ static inline void xen_multicall_flush(b
 #define xen_multi_mmu_update(...) ({ BUG(); -ENOSYS; })
 #define xen_multi_mmuext_op(...) ({ BUG(); -ENOSYS; })
 
+#define gnttab_pre_map_adjust(...) false
+#define gnttab_post_map_adjust(...) ({ BUG(); -ENOSYS; })
+
 #endif /* CONFIG_XEN */
 
 #if defined(CONFIG_X86_64)
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/io.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/io.h      2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/io.h   2008-12-01 11:49:07.000000000 +0100
@@ -3,6 +3,76 @@
 
 #define ARCH_HAS_IOREMAP_WC
 
+#include <linux/compiler.h>
+
+/*
+ * early_ioremap() and early_iounmap() are for temporary early boot-time
+ * mappings, before the real ioremap() is functional.
+ * A boot-time mapping is currently limited to at most 16 pages.
+ */
+#ifndef __ASSEMBLY__
+extern void early_ioremap_init(void);
+extern void early_ioremap_clear(void);
+extern void early_ioremap_reset(void);
+extern void *early_ioremap(unsigned long offset, unsigned long size);
+extern void early_iounmap(void *addr, unsigned long size);
+extern void __iomem *fix_ioremap(unsigned idx, unsigned long phys);
+#endif
+
+#define build_mmio_read(name, size, type, reg, barrier) \
+static inline type name(const volatile void __iomem *addr) \
+{ type ret; asm volatile("mov" size " %1,%0":reg (ret) \
+:"m" (*(volatile type __force *)addr) barrier); return ret; }
+
+#define build_mmio_write(name, size, type, reg, barrier) \
+static inline void name(type val, volatile void __iomem *addr) \
+{ asm volatile("mov" size " %0,%1": :reg (val), \
+"m" (*(volatile type __force *)addr) barrier); }
+
+build_mmio_read(readb, "b", unsigned char, "=q", :"memory")
+build_mmio_read(readw, "w", unsigned short, "=r", :"memory")
+build_mmio_read(readl, "l", unsigned int, "=r", :"memory")
+
+build_mmio_read(__readb, "b", unsigned char, "=q", )
+build_mmio_read(__readw, "w", unsigned short, "=r", )
+build_mmio_read(__readl, "l", unsigned int, "=r", )
+
+build_mmio_write(writeb, "b", unsigned char, "q", :"memory")
+build_mmio_write(writew, "w", unsigned short, "r", :"memory")
+build_mmio_write(writel, "l", unsigned int, "r", :"memory")
+
+build_mmio_write(__writeb, "b", unsigned char, "q", )
+build_mmio_write(__writew, "w", unsigned short, "r", )
+build_mmio_write(__writel, "l", unsigned int, "r", )
+
+#define readb_relaxed(a) __readb(a)
+#define readw_relaxed(a) __readw(a)
+#define readl_relaxed(a) __readl(a)
+#define __raw_readb __readb
+#define __raw_readw __readw
+#define __raw_readl __readl
+
+#define __raw_writeb __writeb
+#define __raw_writew __writew
+#define __raw_writel __writel
+
+#define mmiowb() barrier()
+
+#ifdef CONFIG_X86_64
+build_mmio_read(readq, "q", unsigned long, "=r", :"memory")
+build_mmio_read(__readq, "q", unsigned long, "=r", )
+build_mmio_write(writeq, "q", unsigned long, "r", :"memory")
+build_mmio_write(__writeq, "q", unsigned long, "r", )
+
+#define readq_relaxed(a) __readq(a)
+#define __raw_readq __readq
+#define __raw_writeq writeq
+
+/* Let people know we have them */
+#define readq readq
+#define writeq writeq
+#endif
+
 #ifdef CONFIG_X86_32
 # include "io_32.h"
 #else
@@ -19,4 +89,17 @@ extern int ioremap_check_change_attr(uns
                                     unsigned long prot_val);
 extern void __iomem *ioremap_wc(unsigned long offset, unsigned long size);
 
+/*
+ * early_ioremap() and early_iounmap() are for temporary early boot-time
+ * mappings, before the real ioremap() is functional.
+ * A boot-time mapping is currently limited to at most 16 pages.
+ */
+extern void early_ioremap_init(void);
+extern void early_ioremap_clear(void);
+extern void early_ioremap_reset(void);
+extern void *early_ioremap(unsigned long offset, unsigned long size);
+extern void early_iounmap(void *addr, unsigned long size);
+extern void __iomem *fix_ioremap(unsigned idx, unsigned long phys);
+
+
 #endif /* _ASM_X86_IO_H */
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/io_32.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/io_32.h   2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/io_32.h        2008-12-01 11:49:07.000000000 +0100
@@ -123,6 +123,8 @@ static inline void *phys_to_virt(unsigne
  */
 extern void __iomem *ioremap_nocache(resource_size_t offset, unsigned long size);
 extern void __iomem *ioremap_cache(resource_size_t offset, unsigned long size);
+extern void __iomem *ioremap_prot(resource_size_t offset, unsigned long size,
+                               unsigned long prot_val);
 
 /*
  * The default ioremap() behavior is non-cached:
@@ -135,18 +137,6 @@ static inline void __iomem *ioremap(reso
 extern void iounmap(volatile void __iomem *addr);
 
 /*
- * early_ioremap() and early_iounmap() are for temporary early boot-time
- * mappings, before the real ioremap() is functional.
- * A boot-time mapping is currently limited to at most 16 pages.
- */
-extern void early_ioremap_init(void);
-extern void early_ioremap_clear(void);
-extern void early_ioremap_reset(void);
-extern void *early_ioremap(unsigned long offset, unsigned long size);
-extern void early_iounmap(void *addr, unsigned long size);
-extern void __iomem *fix_ioremap(unsigned idx, unsigned long phys);
-
-/*
  * ISA I/O bus memory addresses are 1:1 with the physical address.
  */
 #define isa_virt_to_bus(_x) ({ BUG(); virt_to_bus(_x); })
@@ -162,55 +152,6 @@ extern void __iomem *fix_ioremap(unsigne
 #define virt_to_bus(_x) phys_to_machine(__pa(_x))
 #define bus_to_virt(_x) __va(machine_to_phys(_x))
 
-/*
- * readX/writeX() are used to access memory mapped devices. On some
- * architectures the memory mapped IO stuff needs to be accessed
- * differently. On the x86 architecture, we just read/write the
- * memory location directly.
- */
-
-static inline unsigned char readb(const volatile void __iomem *addr)
-{
-       return *(volatile unsigned char __force *)addr;
-}
-
-static inline unsigned short readw(const volatile void __iomem *addr)
-{
-       return *(volatile unsigned short __force *)addr;
-}
-
-static inline unsigned int readl(const volatile void __iomem *addr)
-{
-       return *(volatile unsigned int __force *) addr;
-}
-
-#define readb_relaxed(addr) readb(addr)
-#define readw_relaxed(addr) readw(addr)
-#define readl_relaxed(addr) readl(addr)
-#define __raw_readb readb
-#define __raw_readw readw
-#define __raw_readl readl
-
-static inline void writeb(unsigned char b, volatile void __iomem *addr)
-{
-       *(volatile unsigned char __force *)addr = b;
-}
-
-static inline void writew(unsigned short b, volatile void __iomem *addr)
-{
-       *(volatile unsigned short __force *)addr = b;
-}
-
-static inline void writel(unsigned int b, volatile void __iomem *addr)
-{
-       *(volatile unsigned int __force *)addr = b;
-}
-#define __raw_writeb writeb
-#define __raw_writew writew
-#define __raw_writel writel
-
-#define mmiowb()
-
 static inline void
 memset_io(volatile void __iomem *addr, unsigned char val, int count)
 {
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/io_64.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/io_64.h   2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/io_64.h        2008-12-01 11:49:07.000000000 +0100
@@ -190,6 +190,8 @@ extern void early_iounmap(void *addr, un
  */
 extern void __iomem *ioremap_nocache(resource_size_t offset, unsigned long size);
 extern void __iomem *ioremap_cache(resource_size_t offset, unsigned long size);
+extern void __iomem *ioremap_prot(resource_size_t offset, unsigned long size,
+                               unsigned long prot_val);
 
 /*
  * The default ioremap() behavior is non-cached:
@@ -220,77 +222,6 @@ extern void __iomem *fix_ioremap(unsigne
 #define virt_to_bus(_x) phys_to_machine(__pa(_x))
 #define bus_to_virt(_x) __va(machine_to_phys(_x))
 
-/*
- * readX/writeX() are used to access memory mapped devices. On some
- * architectures the memory mapped IO stuff needs to be accessed
- * differently. On the x86 architecture, we just read/write the
- * memory location directly.
- */
-
-static inline __u8 __readb(const volatile void __iomem *addr)
-{
-       return *(__force volatile __u8 *)addr;
-}
-
-static inline __u16 __readw(const volatile void __iomem *addr)
-{
-       return *(__force volatile __u16 *)addr;
-}
-
-static __always_inline __u32 __readl(const volatile void __iomem *addr)
-{
-       return *(__force volatile __u32 *)addr;
-}
-
-static inline __u64 __readq(const volatile void __iomem *addr)
-{
-       return *(__force volatile __u64 *)addr;
-}
-
-#define readb(x) __readb(x)
-#define readw(x) __readw(x)
-#define readl(x) __readl(x)
-#define readq(x) __readq(x)
-#define readb_relaxed(a) readb(a)
-#define readw_relaxed(a) readw(a)
-#define readl_relaxed(a) readl(a)
-#define readq_relaxed(a) readq(a)
-#define __raw_readb readb
-#define __raw_readw readw
-#define __raw_readl readl
-#define __raw_readq readq
-
-#define mmiowb()
-
-static inline void __writel(__u32 b, volatile void __iomem *addr)
-{
-       *(__force volatile __u32 *)addr = b;
-}
-
-static inline void __writeq(__u64 b, volatile void __iomem *addr)
-{
-       *(__force volatile __u64 *)addr = b;
-}
-
-static inline void __writeb(__u8 b, volatile void __iomem *addr)
-{
-       *(__force volatile __u8 *)addr = b;
-}
-
-static inline void __writew(__u16 b, volatile void __iomem *addr)
-{
-       *(__force volatile __u16 *)addr = b;
-}
-
-#define writeq(val, addr) __writeq((val), (addr))
-#define writel(val, addr) __writel((val), (addr))
-#define writew(val, addr) __writew((val), (addr))
-#define writeb(val, addr) __writeb((val), (addr))
-#define __raw_writeb writeb
-#define __raw_writew writew
-#define __raw_writel writel
-#define __raw_writeq writeq
-
 void __memcpy_fromio(void *, unsigned long, unsigned);
 void __memcpy_toio(unsigned long, const void *, unsigned);
 
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/irq_vectors.h
===================================================================
--- /dev/null   1970-01-01 00:00:00.000000000 +0000
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/irq_vectors.h  2008-12-01 11:49:07.000000000 +0100
@@ -0,0 +1,52 @@
+#ifndef _ASM_IRQ_VECTORS_H
+#define _ASM_IRQ_VECTORS_H
+
+#ifdef CONFIG_X86_32
+# define SYSCALL_VECTOR                0x80
+#else
+# define IA32_SYSCALL_VECTOR   0x80
+#endif
+
+#define RESCHEDULE_VECTOR      0
+#define CALL_FUNCTION_VECTOR   1
+#define CALL_FUNC_SINGLE_VECTOR 2
+#define SPIN_UNLOCK_VECTOR     3
+#define NR_IPIS                        4
+
+/*
+ * The maximum number of vectors supported by i386 processors
+ * is limited to 256. For processors other than i386, NR_VECTORS
+ * should be changed accordingly.
+ */
+#define NR_VECTORS             256
+
+#define        FIRST_VM86_IRQ          3
+#define LAST_VM86_IRQ          15
+#define invalid_vm86_irq(irq)  ((irq) < 3 || (irq) > 15)
+
+/*
+ * The flat IRQ space is divided into two regions:
+ *  1. A one-to-one mapping of real physical IRQs. This space is only used
+ *     if we have physical device-access privilege. This region is at the
+ *     start of the IRQ space so that existing device drivers do not need
+ *     to be modified to translate physical IRQ numbers into our IRQ space.
+ *  3. A dynamic mapping of inter-domain and Xen-sourced virtual IRQs. These
+ *     are bound using the provided bind/unbind functions.
+ */
+
+#define PIRQ_BASE              0
+#if !defined(MAX_IO_APICS)
+# define NR_PIRQS              (NR_VECTORS + 32 * NR_CPUS)
+#elif NR_CPUS < MAX_IO_APICS
+# define NR_PIRQS              (NR_VECTORS + 32 * NR_CPUS)
+#else
+# define NR_PIRQS              (NR_VECTORS + 32 * MAX_IO_APICS)
+#endif
+
+#define DYNIRQ_BASE            (PIRQ_BASE + NR_PIRQS)
+#define NR_DYNIRQS             256
+
+#define NR_IRQS                        (NR_PIRQS + NR_DYNIRQS)
+#define NR_IRQ_VECTORS         NR_IRQS
+
+#endif /* _ASM_IRQ_VECTORS_H */
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/irqflags.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/irqflags.h        2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/irqflags.h     2008-12-01 11:49:07.000000000 +0100
@@ -118,7 +118,7 @@ static inline void halt(void)
 
 #ifndef CONFIG_X86_64
 #define INTERRUPT_RETURN               iret
-#define ENABLE_INTERRUPTS_SYSCALL_RET __ENABLE_INTERRUPTS              ; \
+#define ENABLE_INTERRUPTS_SYSEXIT      __ENABLE_INTERRUPTS             ; \
 sysexit_scrit: /**** START OF SYSEXIT CRITICAL REGION ****/            ; \
        __TEST_PENDING                                                  ; \
        jnz  14f        /* process more events if necessary... */       ; \
@@ -175,18 +175,6 @@ static inline void trace_hardirqs_fixup_
 #else
 
 #ifdef CONFIG_X86_64
-/*
- * Currently paravirt can't handle swapgs nicely when we
- * don't have a stack we can rely on (such as a user space
- * stack).  So we either find a way around these or just fault
- * and emulate if a guest tries to call swapgs directly.
- *
- * Either way, this is a good way to document that we don't
- * have a reliable stack. x86_64 only.
- */
-#define SWAPGS_UNSAFE_STACK    swapgs
-#define ARCH_TRACE_IRQS_ON             call trace_hardirqs_on_thunk
-#define ARCH_TRACE_IRQS_OFF            call trace_hardirqs_off_thunk
 #define ARCH_LOCKDEP_SYS_EXIT          call lockdep_sys_exit_thunk
 #define ARCH_LOCKDEP_SYS_EXIT_IRQ      \
        TRACE_IRQS_ON; \
@@ -198,24 +186,6 @@ static inline void trace_hardirqs_fixup_
        TRACE_IRQS_OFF;
 
 #else
-#define ARCH_TRACE_IRQS_ON                     \
-       pushl %eax;                             \
-       pushl %ecx;                             \
-       pushl %edx;                             \
-       call trace_hardirqs_on;                 \
-       popl %edx;                              \
-       popl %ecx;                              \
-       popl %eax;
-
-#define ARCH_TRACE_IRQS_OFF                    \
-       pushl %eax;                             \
-       pushl %ecx;                             \
-       pushl %edx;                             \
-       call trace_hardirqs_off;                \
-       popl %edx;                              \
-       popl %ecx;                              \
-       popl %eax;
-
 #define ARCH_LOCKDEP_SYS_EXIT                  \
        pushl %eax;                             \
        pushl %ecx;                             \
@@ -229,8 +199,8 @@ static inline void trace_hardirqs_fixup_
 #endif
 
 #ifdef CONFIG_TRACE_IRQFLAGS
-#  define TRACE_IRQS_ON                ARCH_TRACE_IRQS_ON
-#  define TRACE_IRQS_OFF       ARCH_TRACE_IRQS_OFF
+#  define TRACE_IRQS_ON                call trace_hardirqs_on_thunk;
+#  define TRACE_IRQS_OFF       call trace_hardirqs_off_thunk;
 #else
 #  define TRACE_IRQS_ON
 #  define TRACE_IRQS_OFF
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/mmu_context.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/mmu_context.h     2008-12-01 11:36:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/mmu_context.h  2008-12-01 11:49:07.000000000 +0100
@@ -1,5 +1,42 @@
+#ifndef __ASM_X86_MMU_CONTEXT_H
+#define __ASM_X86_MMU_CONTEXT_H
+
+#include <asm/desc.h>
+#include <asm/atomic.h>
+#include <asm/pgalloc.h>
+#include <asm/tlbflush.h>
+
+void arch_exit_mmap(struct mm_struct *mm);
+void arch_dup_mmap(struct mm_struct *oldmm, struct mm_struct *mm);
+
+void mm_pin(struct mm_struct *mm);
+void mm_unpin(struct mm_struct *mm);
+void mm_pin_all(void);
+
+static inline void xen_activate_mm(struct mm_struct *prev,
+                                  struct mm_struct *next)
+{
+       if (!PagePinned(virt_to_page(next->pgd)))
+               mm_pin(next);
+}
+
+/*
+ * Used for LDT copy/destruction.
+ */
+int init_new_context(struct task_struct *tsk, struct mm_struct *mm);
+void destroy_context(struct mm_struct *mm);
+
 #ifdef CONFIG_X86_32
 # include "mmu_context_32.h"
 #else
 # include "mmu_context_64.h"
 #endif
+
+#define activate_mm(prev, next)                        \
+do {                                           \
+       xen_activate_mm(prev, next);            \
+       switch_mm((prev), (next), NULL);        \
+} while (0);
+
+
+#endif /* __ASM_X86_MMU_CONTEXT_H */
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/mmu_context_32.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/mmu_context_32.h  2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/mmu_context_32.h       2008-12-01 11:49:07.000000000 +0100
@@ -1,32 +1,6 @@
 #ifndef __I386_SCHED_H
 #define __I386_SCHED_H
 
-#include <asm/desc.h>
-#include <asm/atomic.h>
-#include <asm/pgalloc.h>
-#include <asm/tlbflush.h>
-
-void arch_exit_mmap(struct mm_struct *mm);
-void arch_dup_mmap(struct mm_struct *oldmm, struct mm_struct *mm);
-
-void mm_pin(struct mm_struct *mm);
-void mm_unpin(struct mm_struct *mm);
-void mm_pin_all(void);
-
-static inline void xen_activate_mm(struct mm_struct *prev,
-                                  struct mm_struct *next)
-{
-       if (!PagePinned(virt_to_page(next->pgd)))
-               mm_pin(next);
-}
-
-/*
- * Used for LDT copy/destruction.
- */
-int init_new_context(struct task_struct *tsk, struct mm_struct *mm);
-void destroy_context(struct mm_struct *mm);
-
-
 static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
 {
 #if 0 /* XEN: no lazy tlb */
@@ -107,10 +81,4 @@ static inline void switch_mm(struct mm_s
 #define deactivate_mm(tsk, mm)                 \
        asm("movl %0,%%gs": :"r" (0));
 
-#define activate_mm(prev, next)                        \
-do {                                           \
-       xen_activate_mm(prev, next);            \
-       switch_mm((prev), (next), NULL);        \
-} while (0)
-
 #endif
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/mmu_context_64.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/mmu_context_64.h  2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/mmu_context_64.h       2008-12-01 11:49:07.000000000 +0100
@@ -1,23 +1,6 @@
 #ifndef __X86_64_MMU_CONTEXT_H
 #define __X86_64_MMU_CONTEXT_H
 
-#include <asm/desc.h>
-#include <asm/atomic.h>
-#include <asm/pgalloc.h>
-#include <asm/page.h>
-#include <asm/pda.h>
-#include <asm/pgtable.h>
-#include <asm/tlbflush.h>
-
-void arch_exit_mmap(struct mm_struct *mm);
-void arch_dup_mmap(struct mm_struct *oldmm, struct mm_struct *mm);
-
-/*
- * possibly do the LDT unload here?
- */
-int init_new_context(struct task_struct *tsk, struct mm_struct *mm);
-void destroy_context(struct mm_struct *mm);
-
 static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
 {
 #if defined(CONFIG_SMP) && !defined(CONFIG_XEN)
@@ -58,10 +41,6 @@ static inline void __prepare_arch_switch
        }
 }
 
-extern void mm_pin(struct mm_struct *mm);
-extern void mm_unpin(struct mm_struct *mm);
-void mm_pin_all(void);
-
 static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
                             struct task_struct *tsk)
 {
@@ -124,11 +103,4 @@ do {                                               \
        asm volatile("movl %0,%%fs"::"r"(0));   \
 } while (0)
 
-static inline void activate_mm(struct mm_struct *prev, struct mm_struct *next)
-{
-       if (!PagePinned(virt_to_page(next->pgd)))
-               mm_pin(next);
-       switch_mm(prev, next, NULL);
-}
-
 #endif
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/page.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/page.h    2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/page.h 2008-12-01 11:49:07.000000000 +0100
@@ -16,9 +16,9 @@
  * below. The preprocessor will warn if the two definitions aren't identical.
  */
 #define _PAGE_BIT_PRESENT      0
-#define _PAGE_PRESENT          (_AC(1, L)<<_PAGE_BIT_PRESENT)
-#define _PAGE_BIT_IO           9
-#define _PAGE_IO               (_AC(1, L)<<_PAGE_BIT_IO)
+#define _PAGE_PRESENT          (_AT(pteval_t, 1) << _PAGE_BIT_PRESENT)
+#define _PAGE_BIT_IO           11
+#define _PAGE_IO               (_AT(pteval_t, 1) << _PAGE_BIT_IO)
 
 #define __PHYSICAL_MASK                ((phys_addr_t)(1ULL << __PHYSICAL_MASK_SHIFT) - 1)
 #define __VIRTUAL_MASK         ((1UL << __VIRTUAL_MASK_SHIFT) - 1)
@@ -28,8 +28,11 @@
    (ie, 32-bit PAE). */
 #define PHYSICAL_PAGE_MASK     (((signed long)PAGE_MASK) & __PHYSICAL_MASK)
 
-/* PTE_MASK extracts the PFN from a (pte|pmd|pud|pgd)val_t */
-#define PTE_MASK               ((pteval_t)PHYSICAL_PAGE_MASK)
+/* PTE_PFN_MASK extracts the PFN from a (pte|pmd|pud|pgd)val_t */
+#define PTE_PFN_MASK           ((pteval_t)PHYSICAL_PAGE_MASK)
+
+/* PTE_FLAGS_MASK extracts the flags from a (pte|pmd|pud|pgd)val_t */
+#define PTE_FLAGS_MASK         (~PTE_PFN_MASK)
 
 #define PMD_PAGE_SIZE          (_AC(1, UL) << PMD_SHIFT)
 #define PMD_PAGE_MASK          (~(PMD_PAGE_SIZE-1))
@@ -39,8 +42,7 @@
 #define HPAGE_MASK             (~(HPAGE_SIZE - 1))
 #define HUGETLB_PAGE_ORDER     (HPAGE_SHIFT - PAGE_SHIFT)
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)       (((addr)+PAGE_SIZE-1)&PAGE_MASK)
+#define HUGE_MAX_HSTATE 2
 
 #ifndef __ASSEMBLY__
 #include <linux/types.h>
@@ -61,9 +63,17 @@
 
 #ifndef __ASSEMBLY__
 
+typedef struct { pgdval_t pgd; } pgd_t;
+typedef struct { pgprotval_t pgprot; } pgprot_t;
+
 extern int page_is_ram(unsigned long pagenr);
 extern int devmem_is_allowed(unsigned long pagenr);
+extern void map_devmem(unsigned long pfn, unsigned long size,
+                      pgprot_t vma_prot);
+extern void unmap_devmem(unsigned long pfn, unsigned long size,
+                        pgprot_t vma_prot);
 
+extern unsigned long max_low_pfn_mapped;
 extern unsigned long max_pfn_mapped;
 
 struct page;
@@ -84,15 +94,11 @@ static inline void copy_user_page(void *
        alloc_page_vma(GFP_HIGHUSER | __GFP_ZERO | movableflags, vma, vaddr)
 #define __HAVE_ARCH_ALLOC_ZEROED_USER_HIGHPAGE
 
-typedef struct { pgprotval_t pgprot; } pgprot_t;
-
 #define pgprot_val(x)  ((x).pgprot)
 #define __pgprot(x)    ((pgprot_t) { (x) } )
 
 #include <asm/maddr.h>
 
-typedef struct { pgdval_t pgd; } pgd_t;
-
 #define __pgd_ma(x) ((pgd_t) { (x) } )
 static inline pgd_t xen_make_pgd(pgdval_t val)
 {
@@ -196,6 +202,11 @@ static inline pteval_t xen_pte_val(pte_t
        return ret;
 }
 
+static inline pteval_t xen_pte_flags(pte_t pte)
+{
+       return __pte_val(pte) & PTE_FLAGS_MASK;
+}
+
 #define pgd_val(x)     xen_pgd_val(x)
 #define __pgd(x)       xen_make_pgd(x)
 
@@ -210,6 +221,7 @@ static inline pteval_t xen_pte_val(pte_t
 #endif
 
 #define pte_val(x)     xen_pte_val(x)
+#define pte_flags(x)   xen_pte_flags(x)
 #define __pte(x)       xen_make_pte(x)
 
 #define __pa(x)                __phys_addr((unsigned long)(x))
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/page_64.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/page_64.h 2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/page_64.h      2008-12-01 11:49:07.000000000 +0100
@@ -26,6 +26,12 @@
 #define PUD_PAGE_SIZE          (_AC(1, UL) << PUD_SHIFT)
 #define PUD_PAGE_MASK          (~(PUD_PAGE_SIZE-1))
 
+/*
+ * Set __PAGE_OFFSET to the most negative possible address +
+ * PGDIR_SIZE*16 (pgd slot 272).  The gap is to allow a space for a
+ * hypervisor to fit.  Choosing 16 slots here is arbitrary, but it's
+ * what Xen requires.
+ */
 #define __PAGE_OFFSET           _AC(0xffff880000000000, UL)
 
 #define __PHYSICAL_START       CONFIG_PHYSICAL_START
@@ -63,7 +69,8 @@
 void clear_page(void *page);
 void copy_page(void *to, void *from);
 
-extern unsigned long end_pfn;
+/* duplicated to the one in bootmem.h */
+extern unsigned long max_pfn;
 
 static inline unsigned long __phys_addr(unsigned long x)
 {
@@ -91,6 +98,11 @@ typedef union { pteval_t pte; unsigned i
 extern unsigned long init_memory_mapping(unsigned long start,
                                         unsigned long end);
 
+extern void initmem_init(unsigned long start_pfn, unsigned long end_pfn);
+
+extern void init_extra_mapping_uc(unsigned long phys, unsigned long size);
+extern void init_extra_mapping_wb(unsigned long phys, unsigned long size);
+
 #endif /* !__ASSEMBLY__ */
 
 #ifdef CONFIG_FLATMEM
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/pci.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/pci.h     2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/pci.h  2008-12-01 11:49:07.000000000 +0100
@@ -21,6 +21,8 @@ struct pci_sysdata {
 #endif
 };
 
+extern int pci_routeirq;
+
 /* scan a bus after allocating a pci_sysdata for it */
 extern struct pci_bus *pci_scan_bus_on_node(int busno, struct pci_ops *ops,
                                            int node);
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/pci_32.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/pci_32.h  2008-12-01 11:36:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/pci_32.h       2008-12-01 11:49:07.000000000 +0100
@@ -38,12 +38,14 @@ struct pci_dev;
 #define PCI_DMA_BUS_IS_PHYS    (1)
 
 /* pci_unmap_{page,single} is a nop so... */
-#define DECLARE_PCI_UNMAP_ADDR(ADDR_NAME)
-#define DECLARE_PCI_UNMAP_LEN(LEN_NAME)
-#define pci_unmap_addr(PTR, ADDR_NAME)         (0)
-#define pci_unmap_addr_set(PTR, ADDR_NAME, VAL)        do { } while (0)
-#define pci_unmap_len(PTR, LEN_NAME)           (0)
-#define pci_unmap_len_set(PTR, LEN_NAME, VAL)  do { } while (0)
+#define DECLARE_PCI_UNMAP_ADDR(ADDR_NAME)      dma_addr_t ADDR_NAME[0];
+#define DECLARE_PCI_UNMAP_LEN(LEN_NAME)        unsigned LEN_NAME[0];
+#define pci_unmap_addr(PTR, ADDR_NAME) sizeof((PTR)->ADDR_NAME)
+#define pci_unmap_addr_set(PTR, ADDR_NAME, VAL) \
+       do { break; } while (pci_unmap_addr(PTR, ADDR_NAME))
+#define pci_unmap_len(PTR, LEN_NAME)           sizeof((PTR)->LEN_NAME)
+#define pci_unmap_len_set(PTR, LEN_NAME, VAL) \
+       do { break; } while (pci_unmap_len(PTR, LEN_NAME))
 
 #endif
 
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/pgalloc.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/pgalloc.h 2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/pgalloc.h      2008-12-01 11:49:07.000000000 +0100
@@ -7,6 +7,9 @@
 
 #include <asm/io.h>            /* for phys_to_virt and page_to_pseudophys */
 
+static inline int paravirt_pgd_alloc(struct mm_struct *mm) { return 0; }
+static inline void paravirt_pgd_free(struct mm_struct *mm, pgd_t *pgd) {}
+
 static inline void paravirt_alloc_pte(struct mm_struct *mm, unsigned long pfn) {}
 static inline void paravirt_alloc_pmd(struct mm_struct *mm, unsigned long pfn) {}
 static inline void paravirt_alloc_pmd_clone(unsigned long pfn, unsigned long clonepfn,
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/pgtable.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/pgtable.h 2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/pgtable.h      2008-12-01 11:49:07.000000000 +0100
@@ -13,11 +13,12 @@
 #define _PAGE_BIT_PSE          7       /* 4 MB (or 2MB) page */
 #define _PAGE_BIT_PAT          7       /* on 4KB pages */
 #define _PAGE_BIT_GLOBAL       8       /* Global TLB entry PPro+ */
-#define _PAGE_BIT_IO           9       /* Mapped page is I/O or foreign and
+#define _PAGE_BIT_UNUSED1      9       /* available for programmer */
+#define _PAGE_BIT_UNUSED2      10
+#define _PAGE_BIT_IO           11      /* Mapped page is I/O or foreign and
                                         * has no associated page struct. */
-#define _PAGE_BIT_UNUSED2      10      /* available for programmer */
-#define _PAGE_BIT_UNUSED3      11
 #define _PAGE_BIT_PAT_LARGE    12      /* On 2MB or 1GB pages */
+#define _PAGE_BIT_SPECIAL      _PAGE_BIT_UNUSED1
 #define _PAGE_BIT_NX           63       /* No execute: only valid after cpuid check */
 
 /* If _PAGE_BIT_PRESENT is clear, we use these: */
@@ -28,34 +29,31 @@
 /* if the user mapped it with PROT_NONE; pte_present gives true */
 #define _PAGE_BIT_PROTNONE     _PAGE_BIT_GLOBAL
 
-/*
- * Note: we use _AC(1, L) instead of _AC(1, UL) so that we get a
- * sign-extended value on 32-bit with all 1's in the upper word,
- * which preserves the upper pte values on 64-bit ptes:
- */
-#define _PAGE_PRESENT  (_AC(1, L)<<_PAGE_BIT_PRESENT)
-#define _PAGE_RW       (_AC(1, L)<<_PAGE_BIT_RW)
-#define _PAGE_USER     (_AC(1, L)<<_PAGE_BIT_USER)
-#define _PAGE_PWT      (_AC(1, L)<<_PAGE_BIT_PWT)
-#define _PAGE_PCD      (_AC(1, L)<<_PAGE_BIT_PCD)
-#define _PAGE_ACCESSED (_AC(1, L)<<_PAGE_BIT_ACCESSED)
-#define _PAGE_DIRTY    (_AC(1, L)<<_PAGE_BIT_DIRTY)
-#define _PAGE_PSE      (_AC(1, L)<<_PAGE_BIT_PSE)      /* 2MB page */
-#define _PAGE_GLOBAL   (_AC(1, L)<<_PAGE_BIT_GLOBAL)   /* Global TLB entry */
-#define _PAGE_IO       (_AC(1, L)<<_PAGE_BIT_IO)
-#define _PAGE_UNUSED2  (_AC(1, L)<<_PAGE_BIT_UNUSED2)
-#define _PAGE_UNUSED3  (_AC(1, L)<<_PAGE_BIT_UNUSED3)
-#define _PAGE_PAT      (_AC(1, L)<<_PAGE_BIT_PAT)
-#define _PAGE_PAT_LARGE (_AC(1, L)<<_PAGE_BIT_PAT_LARGE)
+#define _PAGE_PRESENT  (_AT(pteval_t, 1) << _PAGE_BIT_PRESENT)
+#define _PAGE_RW       (_AT(pteval_t, 1) << _PAGE_BIT_RW)
+#define _PAGE_USER     (_AT(pteval_t, 1) << _PAGE_BIT_USER)
+#define _PAGE_PWT      (_AT(pteval_t, 1) << _PAGE_BIT_PWT)
+#define _PAGE_PCD      (_AT(pteval_t, 1) << _PAGE_BIT_PCD)
+#define _PAGE_ACCESSED (_AT(pteval_t, 1) << _PAGE_BIT_ACCESSED)
+#define _PAGE_DIRTY    (_AT(pteval_t, 1) << _PAGE_BIT_DIRTY)
+#define _PAGE_PSE      (_AT(pteval_t, 1) << _PAGE_BIT_PSE)
+#define _PAGE_GLOBAL   (_AT(pteval_t, 1) << _PAGE_BIT_GLOBAL)
+#define _PAGE_UNUSED1  (_AT(pteval_t, 1) << _PAGE_BIT_UNUSED1)
+#define _PAGE_UNUSED2  (_AT(pteval_t, 1) << _PAGE_BIT_UNUSED2)
+#define _PAGE_IO       (_AT(pteval_t, 1) << _PAGE_BIT_IO)
+#define _PAGE_PAT      (_AT(pteval_t, 1) << _PAGE_BIT_PAT)
+#define _PAGE_PAT_LARGE (_AT(pteval_t, 1) << _PAGE_BIT_PAT_LARGE)
+#define _PAGE_SPECIAL  (_AT(pteval_t, 1) << _PAGE_BIT_SPECIAL)
+#define __HAVE_ARCH_PTE_SPECIAL
 
 #if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE)
-#define _PAGE_NX       (_AC(1, ULL) << _PAGE_BIT_NX)
+#define _PAGE_NX       (_AT(pteval_t, 1) << _PAGE_BIT_NX)
 #else
-#define _PAGE_NX       0
+#define _PAGE_NX       (_AT(pteval_t, 0))
 #endif
 
-#define _PAGE_FILE     (_AC(1, L)<<_PAGE_BIT_FILE)
-#define _PAGE_PROTNONE (_AC(1, L)<<_PAGE_BIT_PROTNONE)
+#define _PAGE_FILE     (_AT(pteval_t, 1) << _PAGE_BIT_FILE)
+#define _PAGE_PROTNONE (_AT(pteval_t, 1) << _PAGE_BIT_PROTNONE)
 
 #ifndef __ASSEMBLY__
 #if defined(CONFIG_X86_64) && CONFIG_XEN_COMPAT <= 0x030002
@@ -71,8 +69,8 @@ extern unsigned int __kernel_page_user;
                         _PAGE_DIRTY | __kernel_page_user)
 
 /* Set of bits not changed in pte_modify */
-#define _PAGE_CHG_MASK (PTE_MASK | _PAGE_CACHE_MASK | _PAGE_IO |       \
-                        _PAGE_ACCESSED | _PAGE_DIRTY)
+#define _PAGE_CHG_MASK (PTE_PFN_MASK | _PAGE_CACHE_MASK | _PAGE_IO |   \
+                        _PAGE_SPECIAL | _PAGE_ACCESSED | _PAGE_DIRTY)
 
 /*
  * PAT settings are part of the hypervisor interface, which sets the
@@ -102,19 +100,9 @@ extern unsigned int __kernel_page_user;
 #define PAGE_READONLY_EXEC     __pgprot(_PAGE_PRESENT | _PAGE_USER |   \
                                         _PAGE_ACCESSED)
 
-#ifdef CONFIG_X86_32
-#define _PAGE_KERNEL_EXEC \
-       (_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED)
-#define _PAGE_KERNEL (_PAGE_KERNEL_EXEC | _PAGE_NX)
-
-#ifndef __ASSEMBLY__
-extern pteval_t __PAGE_KERNEL, __PAGE_KERNEL_EXEC;
-#endif /* __ASSEMBLY__ */
-#else
 #define __PAGE_KERNEL_EXEC                                             \
        (_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED | __kernel_page_user)
 #define __PAGE_KERNEL          (__PAGE_KERNEL_EXEC | _PAGE_NX)
-#endif
 
 #define __PAGE_KERNEL_RO               (__PAGE_KERNEL & ~_PAGE_RW)
 #define __PAGE_KERNEL_RX               (__PAGE_KERNEL_EXEC & ~_PAGE_RW)
@@ -125,25 +113,22 @@ extern pteval_t __PAGE_KERNEL, __PAGE_KE
 #define __PAGE_KERNEL_VSYSCALL         (__PAGE_KERNEL_RX | _PAGE_USER)
 #define __PAGE_KERNEL_VSYSCALL_NOCACHE (__PAGE_KERNEL_VSYSCALL | _PAGE_PCD | _PAGE_PWT)
 #define __PAGE_KERNEL_LARGE            (__PAGE_KERNEL | _PAGE_PSE)
+#define __PAGE_KERNEL_LARGE_NOCACHE    (__PAGE_KERNEL | _PAGE_CACHE_UC | _PAGE_PSE)
 #define __PAGE_KERNEL_LARGE_EXEC       (__PAGE_KERNEL_EXEC | _PAGE_PSE)
 
-/*
- * We don't support GLOBAL page in xenolinux64
- */
-#define MAKE_GLOBAL(x)                 __pgprot((x))
-
-#define PAGE_KERNEL                    MAKE_GLOBAL(__PAGE_KERNEL)
-#define PAGE_KERNEL_RO                 MAKE_GLOBAL(__PAGE_KERNEL_RO)
-#define PAGE_KERNEL_EXEC               MAKE_GLOBAL(__PAGE_KERNEL_EXEC)
-#define PAGE_KERNEL_RX                 MAKE_GLOBAL(__PAGE_KERNEL_RX)
-#define PAGE_KERNEL_WC                 MAKE_GLOBAL(__PAGE_KERNEL_WC)
-#define PAGE_KERNEL_NOCACHE            MAKE_GLOBAL(__PAGE_KERNEL_NOCACHE)
-#define PAGE_KERNEL_UC_MINUS           MAKE_GLOBAL(__PAGE_KERNEL_UC_MINUS)
-#define PAGE_KERNEL_EXEC_NOCACHE       MAKE_GLOBAL(__PAGE_KERNEL_EXEC_NOCACHE)
-#define PAGE_KERNEL_LARGE              MAKE_GLOBAL(__PAGE_KERNEL_LARGE)
-#define PAGE_KERNEL_LARGE_EXEC         MAKE_GLOBAL(__PAGE_KERNEL_LARGE_EXEC)
-#define PAGE_KERNEL_VSYSCALL           MAKE_GLOBAL(__PAGE_KERNEL_VSYSCALL)
-#define PAGE_KERNEL_VSYSCALL_NOCACHE   MAKE_GLOBAL(__PAGE_KERNEL_VSYSCALL_NOCACHE)
+#define PAGE_KERNEL                    __pgprot(__PAGE_KERNEL)
+#define PAGE_KERNEL_RO                 __pgprot(__PAGE_KERNEL_RO)
+#define PAGE_KERNEL_EXEC               __pgprot(__PAGE_KERNEL_EXEC)
+#define PAGE_KERNEL_RX                 __pgprot(__PAGE_KERNEL_RX)
+#define PAGE_KERNEL_WC                 __pgprot(__PAGE_KERNEL_WC)
+#define PAGE_KERNEL_NOCACHE            __pgprot(__PAGE_KERNEL_NOCACHE)
+#define PAGE_KERNEL_UC_MINUS           __pgprot(__PAGE_KERNEL_UC_MINUS)
+#define PAGE_KERNEL_EXEC_NOCACHE       __pgprot(__PAGE_KERNEL_EXEC_NOCACHE)
+#define PAGE_KERNEL_LARGE              __pgprot(__PAGE_KERNEL_LARGE)
+#define PAGE_KERNEL_LARGE_NOCACHE      __pgprot(__PAGE_KERNEL_LARGE_NOCACHE)
+#define PAGE_KERNEL_LARGE_EXEC         __pgprot(__PAGE_KERNEL_LARGE_EXEC)
+#define PAGE_KERNEL_VSYSCALL           __pgprot(__PAGE_KERNEL_VSYSCALL)
+#define PAGE_KERNEL_VSYSCALL_NOCACHE   __pgprot(__PAGE_KERNEL_VSYSCALL_NOCACHE)
 
 /*         xwr */
 #define __P000 PAGE_NONE
@@ -182,27 +167,27 @@ extern struct list_head pgd_list;
  */
 static inline int pte_dirty(pte_t pte)
 {
-       return __pte_val(pte) & _PAGE_DIRTY;
+       return pte_flags(pte) & _PAGE_DIRTY;
 }
 
 static inline int pte_young(pte_t pte)
 {
-       return __pte_val(pte) & _PAGE_ACCESSED;
+       return pte_flags(pte) & _PAGE_ACCESSED;
 }
 
 static inline int pte_write(pte_t pte)
 {
-       return __pte_val(pte) & _PAGE_RW;
+       return pte_flags(pte) & _PAGE_RW;
 }
 
 static inline int pte_file(pte_t pte)
 {
-       return __pte_val(pte) & _PAGE_FILE;
+       return pte_flags(pte) & _PAGE_FILE;
 }
 
 static inline int pte_huge(pte_t pte)
 {
-       return __pte_val(pte) & _PAGE_PSE;
+       return pte_flags(pte) & _PAGE_PSE;
 }
 
 static inline int pte_global(pte_t pte)
@@ -212,12 +197,12 @@ static inline int pte_global(pte_t pte)
 
 static inline int pte_exec(pte_t pte)
 {
-       return !(__pte_val(pte) & _PAGE_NX);
+       return !(pte_flags(pte) & _PAGE_NX);
 }
 
 static inline int pte_special(pte_t pte)
 {
-       return 0;
+       return pte_flags(pte) & _PAGE_SPECIAL;
 }
 
 static inline int pmd_large(pmd_t pte)
@@ -228,22 +213,22 @@ static inline int pmd_large(pmd_t pte)
 
 static inline pte_t pte_mkclean(pte_t pte)
 {
-       return __pte_ma(__pte_val(pte) & ~(pteval_t)_PAGE_DIRTY);
+       return __pte_ma(__pte_val(pte) & ~_PAGE_DIRTY);
 }
 
 static inline pte_t pte_mkold(pte_t pte)
 {
-       return __pte_ma(__pte_val(pte) & ~(pteval_t)_PAGE_ACCESSED);
+       return __pte_ma(__pte_val(pte) & ~_PAGE_ACCESSED);
 }
 
 static inline pte_t pte_wrprotect(pte_t pte)
 {
-       return __pte_ma(__pte_val(pte) & ~(pteval_t)_PAGE_RW);
+       return __pte_ma(__pte_val(pte) & ~_PAGE_RW);
 }
 
 static inline pte_t pte_mkexec(pte_t pte)
 {
-       return __pte_ma(__pte_val(pte) & ~(pteval_t)_PAGE_NX);
+       return __pte_ma(__pte_val(pte) & ~_PAGE_NX);
 }
 
 static inline pte_t pte_mkdirty(pte_t pte)
@@ -268,7 +253,7 @@ static inline pte_t pte_mkhuge(pte_t pte
 
 static inline pte_t pte_clrhuge(pte_t pte)
 {
-       return __pte_ma(__pte_val(pte) & ~(pteval_t)_PAGE_PSE);
+       return __pte_ma(__pte_val(pte) & ~_PAGE_PSE);
 }
 
 static inline pte_t pte_mkglobal(pte_t pte)
@@ -283,35 +268,46 @@ static inline pte_t pte_clrglobal(pte_t 
 
 static inline pte_t pte_mkspecial(pte_t pte)
 {
-       return pte;
+       return __pte_ma(__pte_val(pte) | _PAGE_SPECIAL);
 }
 
 extern pteval_t __supported_pte_mask;
 
 static inline pte_t pfn_pte(unsigned long page_nr, pgprot_t pgprot)
 {
-       return __pte((((phys_addr_t)page_nr << PAGE_SHIFT) |
-                     pgprot_val(pgprot)) & __supported_pte_mask);
+       pgprotval_t prot = pgprot_val(pgprot);
+
+       if (prot & _PAGE_PRESENT)
+               prot &= __supported_pte_mask;
+       return __pte(((phys_addr_t)page_nr << PAGE_SHIFT) | prot);
 }
 
 static inline pte_t pfn_pte_ma(unsigned long page_nr, pgprot_t pgprot)
 {
-       return __pte_ma((((phys_addr_t)page_nr << PAGE_SHIFT) |
-                        pgprot_val(pgprot)) & __supported_pte_mask);
+       pgprotval_t prot = pgprot_val(pgprot);
+
+       if (prot & _PAGE_PRESENT)
+               prot &= __supported_pte_mask;
+       return __pte_ma(((phys_addr_t)page_nr << PAGE_SHIFT) | prot);
 }
 
 static inline pmd_t pfn_pmd(unsigned long page_nr, pgprot_t pgprot)
 {
-       return __pmd((((phys_addr_t)page_nr << PAGE_SHIFT) |
-                     pgprot_val(pgprot)) & __supported_pte_mask);
+       pgprotval_t prot = pgprot_val(pgprot);
+
+       if (prot & _PAGE_PRESENT)
+               prot &= __supported_pte_mask;
+       return __pmd(((phys_addr_t)page_nr << PAGE_SHIFT) | prot);
 }
 
 static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 {
-       pteval_t val = pte_val(pte);
+       pgprotval_t prot = pgprot_val(newprot);
+       pteval_t val = pte_val(pte) & _PAGE_CHG_MASK;
 
-       val &= _PAGE_CHG_MASK;
-       val |= pgprot_val(newprot) & (~_PAGE_CHG_MASK) & __supported_pte_mask;
+       if (prot & _PAGE_PRESENT)
+               prot &= __supported_pte_mask;
+       val |= prot & ~_PAGE_CHG_MASK;
 
        return __pte(val);
 }
@@ -325,9 +321,11 @@ static inline pgprot_t pgprot_modify(pgp
        return __pgprot(preservebits | addbits);
 }
 
-#define pte_pgprot(x) __pgprot(__pte_val(x) & ~PTE_MASK)
+#define pte_pgprot(x) __pgprot(pte_flags(x) & PTE_FLAGS_MASK)
 
-#define canon_pgprot(p) __pgprot(pgprot_val(p) & __supported_pte_mask)
+#define canon_pgprot(p) __pgprot(pgprot_val(p) & _PAGE_PRESENT \
+                                ? pgprot_val(p) & __supported_pte_mask \
+                                : pgprot_val(p))
 
 #ifndef __ASSEMBLY__
 #define __HAVE_PHYS_MEM_ACCESS_PROT
@@ -338,6 +336,17 @@ int phys_mem_access_prot_allowed(struct 
                               unsigned long size, pgprot_t *vma_prot);
 #endif
 
+/* Install a pte for a particular vaddr in kernel space. */
+void set_pte_vaddr(unsigned long vaddr, pte_t pte);
+
+#ifndef CONFIG_XEN
+extern void native_pagetable_setup_start(pgd_t *base);
+extern void native_pagetable_setup_done(pgd_t *base);
+#else
+static inline void xen_pagetable_setup_start(pgd_t *base) {}
+static inline void xen_pagetable_setup_done(pgd_t *base) {}
+#endif
+
 #define set_pte(ptep, pte)             xen_set_pte(ptep, pte)
 #define set_pte_at(mm, addr, ptep, pte)        xen_set_pte_at(mm, addr, ptep, pte)
 
@@ -373,6 +382,26 @@ int phys_mem_access_prot_allowed(struct 
 # include "pgtable_64.h"
 #endif
 
+/*
+ * the pgd page can be thought of an array like this: pgd_t[PTRS_PER_PGD]
+ *
+ * this macro returns the index of the entry in the pgd page which would
+ * control the given virtual address
+ */
+#define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD - 1))
+
+/*
+ * pgd_offset() returns a (pgd_t *)
+ * pgd_index() is used get the offset into the pgd page's array of pgd_t's;
+ */
+#define pgd_offset(mm, address) ((mm)->pgd + pgd_index((address)))
+/*
+ * a shortcut which implies the use of the kernel's pgd, instead
+ * of a process's
+ */
+#define pgd_offset_k(address) pgd_offset(&init_mm, (address))
+
+
 #define KERNEL_PGD_BOUNDARY    pgd_index(PAGE_OFFSET)
 #define KERNEL_PGD_PTRS                (PTRS_PER_PGD - KERNEL_PGD_BOUNDARY)
 
@@ -383,8 +412,15 @@ enum {
        PG_LEVEL_4K,
        PG_LEVEL_2M,
        PG_LEVEL_1G,
+       PG_LEVEL_NUM
 };
 
+#ifdef CONFIG_PROC_FS
+extern void update_page_count(int level, unsigned long pages);
+#else
+static inline void update_page_count(int level, unsigned long pages) { }
+#endif
+
 /*
  * Helper function that returns the kernel pagetable entry controlling
  * the virtual address 'address'. NULL means no pagetable entry present.
@@ -441,6 +477,8 @@ static inline void xen_pte_clear(struct 
  * race with other CPU's that might be updating the dirty
  * bit at the same time.
  */
+struct vm_area_struct;
+
 #define  __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
 extern int ptep_set_access_flags(struct vm_area_struct *vma,
                                 unsigned long address, pte_t *ptep,
@@ -523,9 +561,6 @@ static inline void clone_pgd_range(pgd_t
        memcpy(dst, src, count * sizeof(pgd_t));
 }
 
-#define arch_change_pte_range(mm, pmd, addr, end, newprot, dirty_accountable) \
-       xen_change_pte_range(mm, pmd, addr, end, newprot, dirty_accountable)
-
 #define arbitrary_virt_to_machine(va)                                  \
 ({                                                                     \
        unsigned int __lvl;                                             \
@@ -535,6 +570,44 @@ static inline void clone_pgd_range(pgd_t
         | ((unsigned long)(va) & (PAGE_SIZE - 1)));                    \
 })
 
+#define __HAVE_ARCH_PTEP_MODIFY_PROT_TRANSACTION
+static inline pte_t ptep_modify_prot_start(struct mm_struct *mm, unsigned long addr,
+                                          pte_t *ptep)
+{
+#if CONFIG_XEN_COMPAT < 0x030300
+       if (unlikely(!xen_feature(XENFEAT_mmu_pt_update_preserve_ad)))
+               return ptep_get_and_clear(mm, addr, ptep);
+#endif
+       return *ptep;
+}
+
+#ifdef CONFIG_HIGHPTE
+extern void *high_memory;
+#endif
+
+static inline void ptep_modify_prot_commit(struct mm_struct *mm, unsigned long addr,
+                                          pte_t *ptep, pte_t pte)
+{
+       mmu_update_t u;
+
+#if CONFIG_XEN_COMPAT < 0x030300
+       if (unlikely(!xen_feature(XENFEAT_mmu_pt_update_preserve_ad))) {
+               set_pte_at(mm, addr, ptep, pte);
+               return;
+       }
+#endif
+#ifdef CONFIG_HIGHPTE
+       if ((void *)ptep > high_memory)
+               u.ptr = arbitrary_virt_to_machine(ptep)
+                       | MMU_PT_UPDATE_PRESERVE_AD;
+       else
+#endif
+               u.ptr = virt_to_machine(ptep) | MMU_PT_UPDATE_PRESERVE_AD;
+       u.val = __pte_val(pte);
+       if (HYPERVISOR_mmu_update(&u, 1, NULL, DOMID_SELF))
+               BUG();
+}
+
 #include <asm-generic/pgtable.h>
 
 #include <xen/features.h>
@@ -563,10 +636,6 @@ int touch_pte_range(struct mm_struct *mm
                     unsigned long address,
                     unsigned long size);
 
-int xen_change_pte_range(struct mm_struct *mm, pmd_t *pmd,
-               unsigned long addr, unsigned long end, pgprot_t newprot,
-               int dirty_accountable);
-
 #endif /* __ASSEMBLY__ */
 
 #endif /* _ASM_X86_PGTABLE_H */
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/pgtable-3level.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/pgtable-3level.h  2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/pgtable-3level.h       2008-12-01 11:49:07.000000000 +0100
@@ -14,11 +14,11 @@
 #define pmd_ERROR(e)                                                   \
        printk("%s:%d: bad pmd %p(%016Lx pfn %08Lx).\n",                \
               __FILE__, __LINE__, &(e), __pmd_val(e),                  \
-              (pmd_val(e) & PTE_MASK) >> PAGE_SHIFT)
+              (pmd_val(e) & PTE_PFN_MASK) >> PAGE_SHIFT)
 #define pgd_ERROR(e)                                                   \
        printk("%s:%d: bad pgd %p(%016Lx pfn %08Lx).\n",                \
               __FILE__, __LINE__, &(e), __pgd_val(e),                  \
-              (pgd_val(e) & PTE_MASK) >> PAGE_SHIFT)
+              (pgd_val(e) & PTE_PFN_MASK) >> PAGE_SHIFT)
 
 static inline int pud_none(pud_t pud)
 {
@@ -27,7 +27,7 @@ static inline int pud_none(pud_t pud)
 }
 static inline int pud_bad(pud_t pud)
 {
-       return (__pud_val(pud) & ~(PTE_MASK | _KERNPG_TABLE | _PAGE_USER)) != 0;
+       return (__pud_val(pud) & ~(PTE_PFN_MASK | _KERNPG_TABLE | _PAGE_USER)) != 0;
 }
 
 static inline int pud_present(pud_t pud)
@@ -102,9 +102,9 @@ static inline void pud_clear(pud_t *pudp
                xen_tlb_flush();
 }
 
-#define pud_page(pud) ((struct page *) __va(pud_val(pud) & PTE_MASK))
+#define pud_page(pud) ((struct page *) __va(pud_val(pud) & PTE_PFN_MASK))
 
-#define pud_page_vaddr(pud) ((unsigned long) __va(pud_val(pud) & PTE_MASK))
+#define pud_page_vaddr(pud) ((unsigned long) __va(pud_val(pud) & PTE_PFN_MASK))
 
 
 /* Find an entry in the second-level page table.. */
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/pgtable_32.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/pgtable_32.h      2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/pgtable_32.h   2008-12-01 11:49:07.000000000 +0100
@@ -89,10 +89,10 @@ extern unsigned long pg0[];
 /* pmd_present doesn't just test the _PAGE_PRESENT bit since wr.p.t.
    can temporarily clear it. */
 #define pmd_present(x) (__pmd_val(x))
-#define pmd_bad(x)     ((__pmd_val(x) & (~PTE_MASK & ~_PAGE_USER & ~_PAGE_PRESENT)) != (_KERNPG_TABLE & ~_PAGE_PRESENT))
+#define pmd_bad(x)     ((__pmd_val(x) & (PTE_FLAGS_MASK & ~_PAGE_USER & ~_PAGE_PRESENT)) != (_KERNPG_TABLE & ~_PAGE_PRESENT))
 #else
 #define pmd_present(x) (__pmd_val(x) & _PAGE_PRESENT)
-#define pmd_bad(x)     ((__pmd_val(x) & (~PTE_MASK & ~_PAGE_USER)) != _KERNPG_TABLE)
+#define pmd_bad(x)     ((__pmd_val(x) & (PTE_FLAGS_MASK & ~_PAGE_USER)) != _KERNPG_TABLE)
 #endif
 
 
@@ -119,26 +119,6 @@ extern unsigned long pg0[];
  */
 #define mk_pte(page, pgprot)   pfn_pte(page_to_pfn(page), (pgprot))
 
-/*
- * the pgd page can be thought of an array like this: pgd_t[PTRS_PER_PGD]
- *
- * this macro returns the index of the entry in the pgd page which would
- * control the given virtual address
- */
-#define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD - 1))
-#define pgd_index_k(addr) pgd_index((addr))
-
-/*
- * pgd_offset() returns a (pgd_t *)
- * pgd_index() is used get the offset into the pgd page's array of pgd_t's;
- */
-#define pgd_offset(mm, address) ((mm)->pgd + pgd_index((address)))
-
-/*
- * a shortcut which implies the use of the kernel's pgd, instead
- * of a process's
- */
-#define pgd_offset_k(address) pgd_offset(&init_mm, (address))
 
 static inline int pud_large(pud_t pud) { return 0; }
 
@@ -165,7 +145,7 @@ static inline int pud_large(pud_t pud) {
 #define pmd_page(pmd) (pfn_to_page(pmd_val((pmd)) >> PAGE_SHIFT))
 
 #define pmd_page_vaddr(pmd)                                    \
-       ((unsigned long)__va(pmd_val((pmd)) & PTE_MASK))
+       ((unsigned long)__va(pmd_val((pmd)) & PTE_PFN_MASK))
 
 #if defined(CONFIG_HIGHPTE)
 #define pte_offset_map(dir, address)                                   \
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/pgtable_64.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/pgtable_64.h      2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/pgtable_64.h   2008-12-01 11:49:07.000000000 +0100
@@ -23,6 +23,8 @@ extern void xen_init_pt(void);
 extern pud_t level3_kernel_pgt[512];
 extern pud_t level3_ident_pgt[512];
 extern pmd_t level2_kernel_pgt[512];
+extern pmd_t level2_fixmap_pgt[512];
+extern pmd_t level2_ident_pgt[512];
 extern pgd_t init_level4_pgt[];
 
 #define swapper_pg_dir init_level4_pgt
@@ -79,6 +81,9 @@ extern void paging_init(void);
 
 struct mm_struct;
 
+void set_pte_vaddr_pud(pud_t *pud_page, unsigned long vaddr, pte_t new_pte);
+
+
 #define __xen_pte_clear(ptep) xen_set_pte(ptep, __pte(0))
 
 static inline void xen_set_pte(pte_t *ptep, pte_t pte)
@@ -150,24 +155,24 @@ static inline void xen_pgd_clear(pgd_t *
 #define VMALLOC_END      _AC(0xffffe1ffffffffff, UL)
 #define VMEMMAP_START   _AC(0xffffe20000000000, UL)
 #define MODULES_VADDR    _AC(0xffffffffa0000000, UL)
-#define MODULES_END      _AC(0xfffffffffff00000, UL)
+#define MODULES_END      _AC(0xffffffffff000000, UL)
 #define MODULES_LEN   (MODULES_END - MODULES_VADDR)
 
 #ifndef __ASSEMBLY__
 
 static inline int pgd_bad(pgd_t pgd)
 {
-       return (__pgd_val(pgd) & ~(PTE_MASK | _PAGE_USER)) != _KERNPG_TABLE;
+       return (__pgd_val(pgd) & ~(PTE_PFN_MASK | _PAGE_USER)) != _KERNPG_TABLE;
 }
 
 static inline int pud_bad(pud_t pud)
 {
-       return (__pud_val(pud) & ~(PTE_MASK | _PAGE_USER)) != _KERNPG_TABLE;
+       return (__pud_val(pud) & ~(PTE_PFN_MASK | _PAGE_USER)) != _KERNPG_TABLE;
 }
 
 static inline int pmd_bad(pmd_t pmd)
 {
-       return (__pmd_val(pmd) & ~(PTE_MASK | _PAGE_USER)) != _KERNPG_TABLE;
+       return (__pmd_val(pmd) & ~(PTE_PFN_MASK | _PAGE_USER)) != _KERNPG_TABLE;
 }
 
 #define pte_none(x)    (!(x).pte)
@@ -175,7 +180,7 @@ static inline int pmd_bad(pmd_t pmd)
 
 #define pages_to_mb(x) ((x) >> (20 - PAGE_SHIFT))   /* FIXME: is this right? */
 
-#define __pte_mfn(_pte) (((_pte).pte & PTE_MASK) >> PAGE_SHIFT)
+#define __pte_mfn(_pte) (((_pte).pte & PTE_PFN_MASK) >> PAGE_SHIFT)
 #define pte_mfn(_pte) ((_pte).pte & _PAGE_PRESENT ? \
        __pte_mfn(_pte) : pfn_to_mfn(__pte_mfn(_pte)))
 #define pte_pfn(_pte) ((_pte).pte & _PAGE_IO ? max_mapnr :     \
@@ -200,11 +205,8 @@ static inline int pmd_bad(pmd_t pmd)
  * Level 4 access.
  */
 #define pgd_page_vaddr(pgd)                                            \
-       ((unsigned long)__va((unsigned long)pgd_val((pgd)) & PTE_MASK))
+       ((unsigned long)__va((unsigned long)pgd_val((pgd)) & PTE_PFN_MASK))
 #define pgd_page(pgd)          (pfn_to_page(pgd_val((pgd)) >> PAGE_SHIFT))
-#define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD - 1))
-#define pgd_offset(mm, address)        ((mm)->pgd + pgd_index((address)))
-#define pgd_offset_k(address) (init_level4_pgt + pgd_index((address)))
 #define pgd_present(pgd) (__pgd_val(pgd) & _PAGE_PRESENT)
 static inline int pgd_large(pgd_t pgd) { return 0; }
 #define mk_kernel_pgd(address) __pgd((address) | _KERNPG_TABLE)
@@ -226,7 +228,7 @@ static inline int pud_large(pud_t pte)
 }
 
 /* PMD  - Level 2 access */
-#define pmd_page_vaddr(pmd) ((unsigned long) __va(pmd_val((pmd)) & PTE_MASK))
+#define pmd_page_vaddr(pmd) ((unsigned long) __va(pmd_val((pmd)) & PTE_PFN_MASK))
 #define pmd_page(pmd)          (pfn_to_page(pmd_val((pmd)) >> PAGE_SHIFT))
 
 #define pmd_index(address) (((address) >> PMD_SHIFT) & (PTRS_PER_PMD - 1))
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/processor.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/processor.h       2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/processor.h    2008-12-01 11:49:07.000000000 +0100
@@ -134,7 +134,7 @@ extern __u32                        cleared_cpu_caps[NCAPINTS
 #ifdef CONFIG_SMP
 DECLARE_PER_CPU(struct cpuinfo_x86, cpu_info);
 #define cpu_data(cpu)          per_cpu(cpu_info, cpu)
-#define current_cpu_data       cpu_data(smp_processor_id())
+#define current_cpu_data       __get_cpu_var(cpu_info)
 #else
 #define cpu_data(cpu)          boot_cpu_data
 #define current_cpu_data       boot_cpu_data
@@ -153,7 +153,7 @@ static inline int hlt_works(int cpu)
 
 extern void cpu_detect(struct cpuinfo_x86 *c);
 
-extern void identify_cpu(struct cpuinfo_x86 *);
+extern void early_cpu_init(void);
 extern void identify_boot_cpu(void);
 extern void identify_secondary_cpu(struct cpuinfo_x86 *);
 extern void print_cpu_info(struct cpuinfo_x86 *);
@@ -267,15 +267,11 @@ struct tss_struct {
        struct thread_struct    *io_bitmap_owner;
 
        /*
-        * Pad the TSS to be cacheline-aligned (size is 0x100):
-        */
-       unsigned long           __cacheline_filler[35];
-       /*
         * .. and then another 0x100 bytes for the emergency kernel stack:
         */
        unsigned long           stack[64];
 
-} __attribute__((packed));
+} ____cacheline_aligned;
 
 DECLARE_PER_CPU(struct tss_struct, init_tss);
 
@@ -668,11 +664,36 @@ static inline void __sti_mwait(unsigned 
 
 extern void mwait_idle_with_hints(unsigned long eax, unsigned long ecx);
 
-extern int                     force_mwait;
-
 extern void select_idle_routine(const struct cpuinfo_x86 *c);
 
 extern unsigned long           boot_option_idle_override;
+extern unsigned long           idle_halt;
+extern unsigned long           idle_nomwait;
+
+#ifndef CONFIG_XEN
+/*
+ * on systems with caches, caches must be flashed as the absolute
+ * last instruction before going into a suspended halt.  Otherwise,
+ * dirty data can linger in the cache and become stale on resume,
+ * leading to strange errors.
+ *
+ * perform a variety of operations to guarantee that the compiler
+ * will not reorder instructions.  wbinvd itself is serializing
+ * so the processor will not reorder.
+ *
+ * Systems without cache can just go into halt.
+ */
+static inline void wbinvd_halt(void)
+{
+       mb();
+       /* check for clflush to determine if wbinvd is legal */
+       if (cpu_has_clflush)
+               asm volatile("cli; wbinvd; 1: hlt; jmp 1b" : : : "memory");
+       else
+               while (1)
+                       halt();
+}
+#endif
 
 extern void enable_sep_cpu(void);
 extern int sysenter_setup(void);
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/segment.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/segment.h 2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/segment.h      2008-12-01 11:49:07.000000000 +0100
@@ -1,6 +1,15 @@
 #ifndef _ASM_X86_SEGMENT_H_
 #define _ASM_X86_SEGMENT_H_
 
+/* Constructor for a conventional segment GDT (or LDT) entry */
+/* This is a macro so it can be used in initializers */
+#define GDT_ENTRY(flags, base, limit)                  \
+       ((((base)  & 0xff000000ULL) << (56-24)) |       \
+        (((flags) & 0x0000f0ffULL) << 40) |            \
+        (((limit) & 0x000f0000ULL) << (48-16)) |       \
+        (((base)  & 0x00ffffffULL) << 16) |            \
+        (((limit) & 0x0000ffffULL)))
+
 /* Simple and small GDT entries for booting only */
 
 #define GDT_ENTRY_BOOT_CS      2
@@ -61,18 +70,14 @@
 #define GDT_ENTRY_TLS_MAX      (GDT_ENTRY_TLS_MIN + GDT_ENTRY_TLS_ENTRIES - 1)
 
 #define GDT_ENTRY_DEFAULT_USER_CS      14
-#define __USER_CS (GDT_ENTRY_DEFAULT_USER_CS * 8 + 3)
 
 #define GDT_ENTRY_DEFAULT_USER_DS      15
-#define __USER_DS (GDT_ENTRY_DEFAULT_USER_DS * 8 + 3)
 
 #define GDT_ENTRY_KERNEL_BASE  12
 
 #define GDT_ENTRY_KERNEL_CS            (GDT_ENTRY_KERNEL_BASE + 0)
-#define __KERNEL_CS (GDT_ENTRY_KERNEL_CS * 8)
 
 #define GDT_ENTRY_KERNEL_DS            (GDT_ENTRY_KERNEL_BASE + 1)
-#define __KERNEL_DS (GDT_ENTRY_KERNEL_DS * 8)
 
 #define GDT_ENTRY_TSS                  (GDT_ENTRY_KERNEL_BASE + 4)
 #define GDT_ENTRY_LDT                  (GDT_ENTRY_KERNEL_BASE + 5)
@@ -143,10 +148,11 @@
 #else
 #include <asm/cache.h>
 
-#define __KERNEL_CS    0x10
-#define __KERNEL_DS    0x18
+#define GDT_ENTRY_KERNEL32_CS 1
+#define GDT_ENTRY_KERNEL_CS 2
+#define GDT_ENTRY_KERNEL_DS 3
 
-#define __KERNEL32_CS   0x08
+#define __KERNEL32_CS   (GDT_ENTRY_KERNEL32_CS * 8)
 
 /*
  * we cannot use the same code segment descriptor for user and kernel
@@ -154,10 +160,10 @@
  * The segment offset needs to contain a RPL. Grr. -AK
  * GDT layout to get 64bit syscall right (sysret hardcodes gdt offsets)
  */
-
-#define __USER32_CS   0x23   /* 4*8+3 */
-#define __USER_DS     0x2b   /* 5*8+3 */
-#define __USER_CS     0x33   /* 6*8+3 */
+#define GDT_ENTRY_DEFAULT_USER32_CS 4
+#define GDT_ENTRY_DEFAULT_USER_DS 5
+#define GDT_ENTRY_DEFAULT_USER_CS 6
+#define __USER32_CS   (GDT_ENTRY_DEFAULT_USER32_CS * 8 + 3)
 #define __USER32_DS    __USER_DS
 
 #define GDT_ENTRY_TSS 8        /* needs two entries */
@@ -179,6 +185,11 @@
 
 #endif
 
+#define __KERNEL_CS    (GDT_ENTRY_KERNEL_CS * 8)
+#define __KERNEL_DS    (GDT_ENTRY_KERNEL_DS * 8)
+#define __USER_DS     (GDT_ENTRY_DEFAULT_USER_DS* 8 + 3)
+#define __USER_CS     (GDT_ENTRY_DEFAULT_USER_CS* 8 + 3)
+
 /* User mode is privilege level 3 */
 #define USER_RPL               0x3
 /* LDT segment has TI set, GDT has it cleared */
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/smp.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/smp.h     2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/smp.h  2008-12-01 11:49:07.000000000 +0100
@@ -25,23 +25,16 @@ extern cpumask_t cpu_initialized;
 extern void (*mtrr_hook)(void);
 extern void zap_low_mappings(void);
 
+extern int __cpuinit get_local_pda(int cpu);
+
 extern int smp_num_siblings;
 extern unsigned int num_processors;
 extern cpumask_t cpu_initialized;
 
-#if defined(CONFIG_SMP) && !defined(CONFIG_XEN)
-extern u16 x86_cpu_to_apicid_init[];
-extern u16 x86_bios_cpu_apicid_init[];
-extern void *x86_cpu_to_apicid_early_ptr;
-extern void *x86_bios_cpu_apicid_early_ptr;
-#else
-#define x86_cpu_to_apicid_early_ptr NULL
-#define x86_bios_cpu_apicid_early_ptr NULL
-#endif
-
 DECLARE_PER_CPU(cpumask_t, cpu_sibling_map);
 DECLARE_PER_CPU(cpumask_t, cpu_core_map);
 DECLARE_PER_CPU(u16, cpu_llc_id);
+
 DECLARE_PER_CPU(u16, x86_cpu_to_apicid);
 DECLARE_PER_CPU(u16, x86_bios_cpu_apicid);
 
@@ -63,9 +56,9 @@ struct smp_ops {
 
        void (*smp_send_stop)(void);
        void (*smp_send_reschedule)(int cpu);
-       int (*smp_call_function_mask)(cpumask_t mask,
-                                     void (*func)(void *info), void *info,
-                                     int wait);
+
+       void (*send_call_func_ipi)(cpumask_t mask);
+       void (*send_call_func_single_ipi)(int cpu);
 };
 
 /* Globals due to paravirt */
@@ -106,11 +99,14 @@ static inline void smp_send_reschedule(i
        smp_ops.smp_send_reschedule(cpu);
 }
 
-static inline int smp_call_function_mask(cpumask_t mask,
-                                        void (*func) (void *info), void *info,
-                                        int wait)
+static inline void arch_send_call_function_single_ipi(int cpu)
 {
-       return smp_ops.smp_call_function_mask(mask, func, info, wait);
+       smp_ops.send_call_func_single_ipi(cpu);
+}
+
+static inline void arch_send_call_function_ipi(cpumask_t mask)
+{
+       smp_ops.send_call_func_ipi(mask);
 }
 
 void native_smp_prepare_boot_cpu(void);
@@ -122,23 +118,19 @@ int native_cpu_up(unsigned int cpunum);
 
 void xen_smp_send_stop(void);
 void xen_smp_send_reschedule(int cpu);
-int xen_smp_call_function_mask(cpumask_t mask,
-                              void (*func) (void *info), void *info,
-                              int wait);
+void xen_send_call_func_ipi(cpumask_t mask);
+void xen_send_call_func_single_ipi(int cpu);
 
 #define smp_send_stop          xen_smp_send_stop
 #define smp_send_reschedule    xen_smp_send_reschedule
-#define smp_call_function_mask xen_smp_call_function_mask
-
-extern void prefill_possible_map(void);
+#define arch_send_call_function_single_ipi     xen_send_call_func_single_ipi
+#define arch_send_call_function_ipi            xen_send_call_func_ipi
 
 #endif /* CONFIG_XEN */
 
 extern int __cpu_disable(void);
 extern void __cpu_die(unsigned int cpu);
 
-extern void prefill_possible_map(void);
-
 void smp_store_cpu_info(int id);
 #define cpu_physical_id(cpu)   per_cpu(x86_cpu_to_apicid, cpu)
 
@@ -149,6 +141,14 @@ static inline int num_booting_cpus(void)
 }
 #endif /* CONFIG_SMP */
 
+#if defined(CONFIG_SMP) && (defined(CONFIG_HOTPLUG_CPU) || defined(CONFIG_XEN))
+extern void prefill_possible_map(void);
+#else
+static inline void prefill_possible_map(void)
+{
+}
+#endif
+
 extern unsigned disabled_cpus __cpuinitdata;
 
 #ifdef CONFIG_X86_32_SMP
@@ -216,12 +216,8 @@ static inline int hard_smp_processor_id(
 #endif /* CONFIG_X86_LOCAL_APIC */
 
 #ifdef CONFIG_HOTPLUG_CPU
-extern void cpu_exit_clear(void);
 extern void cpu_uninit(void);
 #endif
 
-extern void smp_alloc_memory(void);
-extern void lock_ipi_call_lock(void);
-extern void unlock_ipi_call_lock(void);
 #endif /* __ASSEMBLY__ */
 #endif
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/spinlock.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/spinlock.h        2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/spinlock.h     2008-12-01 11:49:07.000000000 +0100
@@ -65,14 +65,14 @@ extern void xen_spin_kick(raw_spinlock_t
  */
 #if (NR_CPUS < 256)
 #define TICKET_SHIFT 8
-#define __raw_spin_lock_preamble \
+#define __ticket_spin_lock_preamble \
        asm(LOCK_PREFIX "xaddw %w0, %2\n\t" \
            "cmpb %h0, %b0\n\t" \
            "sete %1" \
            : "=&Q" (token), "=qm" (free), "+m" (lock->slock) \
            : "0" (0x0100) \
            : "memory", "cc")
-#define __raw_spin_lock_body \
+#define __ticket_spin_lock_body \
        asm("1:\t" \
            "cmpb %h0, %b0\n\t" \
            "je 2f\n\t" \
@@ -88,7 +88,7 @@ extern void xen_spin_kick(raw_spinlock_t
            : "memory", "cc")
 
 
-static __always_inline int __raw_spin_trylock(raw_spinlock_t *lock)
+static __always_inline int __ticket_spin_trylock(raw_spinlock_t *lock)
 {
        int tmp, new;
 
@@ -107,7 +107,7 @@ static __always_inline int __raw_spin_tr
        return tmp;
 }
 
-static __always_inline void __raw_spin_unlock(raw_spinlock_t *lock)
+static __always_inline void __ticket_spin_unlock(raw_spinlock_t *lock)
 {
        unsigned int token;
        unsigned char kick;
@@ -124,7 +124,7 @@ static __always_inline void __raw_spin_u
 }
 #else
 #define TICKET_SHIFT 16
-#define __raw_spin_lock_preamble \
+#define __ticket_spin_lock_preamble \
        do { \
                unsigned int tmp; \
                asm(LOCK_PREFIX "xaddl %0, %2\n\t" \
@@ -136,7 +136,7 @@ static __always_inline void __raw_spin_u
                    : "0" (0x00010000) \
                    : "memory", "cc"); \
        } while (0)
-#define __raw_spin_lock_body \
+#define __ticket_spin_lock_body \
        do { \
                unsigned int tmp; \
                asm("shldl $16, %0, %2\n" \
@@ -155,7 +155,7 @@ static __always_inline void __raw_spin_u
                    : "memory", "cc"); \
        } while (0)
 
-static __always_inline int __raw_spin_trylock(raw_spinlock_t *lock)
+static __always_inline int __ticket_spin_trylock(raw_spinlock_t *lock)
 {
        int tmp;
        int new;
@@ -177,7 +177,7 @@ static __always_inline int __raw_spin_tr
        return tmp;
 }
 
-static __always_inline void __raw_spin_unlock(raw_spinlock_t *lock)
+static __always_inline void __ticket_spin_unlock(raw_spinlock_t *lock)
 {
        unsigned int token, tmp;
        bool kick;
@@ -195,49 +195,145 @@ static __always_inline void __raw_spin_u
 }
 #endif
 
-static inline int __raw_spin_is_locked(raw_spinlock_t *lock)
+static inline int __ticket_spin_is_locked(raw_spinlock_t *lock)
 {
        int tmp = ACCESS_ONCE(lock->slock);
 
        return !!(((tmp >> TICKET_SHIFT) ^ tmp) & ((1 << TICKET_SHIFT) - 1));
 }
 
-static inline int __raw_spin_is_contended(raw_spinlock_t *lock)
+static inline int __ticket_spin_is_contended(raw_spinlock_t *lock)
 {
        int tmp = ACCESS_ONCE(lock->slock);
 
        return (((tmp >> TICKET_SHIFT) - tmp) & ((1 << TICKET_SHIFT) - 1)) > 1;
 }
 
-static __always_inline void __raw_spin_lock(raw_spinlock_t *lock)
+static __always_inline void __ticket_spin_lock(raw_spinlock_t *lock)
 {
        unsigned int token, count;
        bool free;
 
-       __raw_spin_lock_preamble;
+       __ticket_spin_lock_preamble;
        if (unlikely(!free))
                token = xen_spin_adjust(lock, token);
        do {
                count = 1 << 10;
-               __raw_spin_lock_body;
+               __ticket_spin_lock_body;
        } while (unlikely(!count) && !xen_spin_wait(lock, token));
 }
 
-static __always_inline void __raw_spin_lock_flags(raw_spinlock_t *lock,
-                                                 unsigned long flags)
+static __always_inline void __ticket_spin_lock_flags(raw_spinlock_t *lock,
+                                                    unsigned long flags)
 {
        unsigned int token, count;
        bool free;
 
-       __raw_spin_lock_preamble;
+       __ticket_spin_lock_preamble;
        if (unlikely(!free))
                token = xen_spin_adjust(lock, token);
        do {
                count = 1 << 10;
-               __raw_spin_lock_body;
+               __ticket_spin_lock_body;
        } while (unlikely(!count) && !xen_spin_wait_flags(lock, &token, flags));
 }
 
+#ifdef CONFIG_PARAVIRT
+/*
+ * Define virtualization-friendly old-style lock byte lock, for use in
+ * pv_lock_ops if desired.
+ *
+ * This differs from the pre-2.6.24 spinlock by always using xchgb
+ * rather than decb to take the lock; this allows it to use a
+ * zero-initialized lock structure.  It also maintains a 1-byte
+ * contention counter, so that we can implement
+ * __byte_spin_is_contended.
+ */
+struct __byte_spinlock {
+       s8 lock;
+       s8 spinners;
+};
+
+static inline int __byte_spin_is_locked(raw_spinlock_t *lock)
+{
+       struct __byte_spinlock *bl = (struct __byte_spinlock *)lock;
+       return bl->lock != 0;
+}
+
+static inline int __byte_spin_is_contended(raw_spinlock_t *lock)
+{
+       struct __byte_spinlock *bl = (struct __byte_spinlock *)lock;
+       return bl->spinners != 0;
+}
+
+static inline void __byte_spin_lock(raw_spinlock_t *lock)
+{
+       struct __byte_spinlock *bl = (struct __byte_spinlock *)lock;
+       s8 val = 1;
+
+       asm("1: xchgb %1, %0\n"
+           "   test %1,%1\n"
+           "   jz 3f\n"
+           "   " LOCK_PREFIX "incb %2\n"
+           "2: rep;nop\n"
+           "   cmpb $1, %0\n"
+           "   je 2b\n"
+           "   " LOCK_PREFIX "decb %2\n"
+           "   jmp 1b\n"
+           "3:"
+           : "+m" (bl->lock), "+q" (val), "+m" (bl->spinners): : "memory");
+}
+
+static inline int __byte_spin_trylock(raw_spinlock_t *lock)
+{
+       struct __byte_spinlock *bl = (struct __byte_spinlock *)lock;
+       u8 old = 1;
+
+       asm("xchgb %1,%0"
+           : "+m" (bl->lock), "+q" (old) : : "memory");
+
+       return old == 0;
+}
+
+static inline void __byte_spin_unlock(raw_spinlock_t *lock)
+{
+       struct __byte_spinlock *bl = (struct __byte_spinlock *)lock;
+       smp_wmb();
+       bl->lock = 0;
+}
+#else  /* !CONFIG_PARAVIRT */
+static inline int __raw_spin_is_locked(raw_spinlock_t *lock)
+{
+       return __ticket_spin_is_locked(lock);
+}
+
+static inline int __raw_spin_is_contended(raw_spinlock_t *lock)
+{
+       return __ticket_spin_is_contended(lock);
+}
+
+static __always_inline void __raw_spin_lock(raw_spinlock_t *lock)
+{
+       __ticket_spin_lock(lock);
+}
+
+static __always_inline void __raw_spin_lock_flags(raw_spinlock_t *lock,
+                                                 unsigned long flags)
+{
+       __ticket_spin_lock_flags(lock, flags);
+}
+
+static __always_inline int __raw_spin_trylock(raw_spinlock_t *lock)
+{
+       return __ticket_spin_trylock(lock);
+}
+
+static __always_inline void __raw_spin_unlock(raw_spinlock_t *lock)
+{
+       __ticket_spin_unlock(lock);
+}
+#endif /* CONFIG_PARAVIRT */
+
 static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock)
 {
        while (__raw_spin_is_locked(lock))
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/system.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/system.h  2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/system.h       2008-12-01 11:49:07.000000000 +0100
@@ -137,7 +137,7 @@ __asm__ __volatile__ ("movw %%dx,%1\n\t"
 #define set_base(ldt, base) _set_base(((char *)&(ldt)) , (base))
 #define set_limit(ldt, limit) _set_limit(((char *)&(ldt)) , ((limit)-1))
 
-extern void load_gs_index(unsigned);
+extern void xen_load_gs_index(unsigned);
 
 /*
  * Load a segment. Fall back on loading the zero
@@ -154,14 +154,14 @@ extern void load_gs_index(unsigned);
                     "jmp 2b\n"                 \
                     ".previous\n"              \
                     _ASM_EXTABLE(1b,3b)        \
-                    : :"r" (value), "r" (0))
+                    : :"r" (value), "r" (0) : "memory")
 
 
 /*
  * Save a segment register away
  */
 #define savesegment(seg, value)                                \
-       asm volatile("mov %%" #seg ",%0":"=rm" (value))
+       asm("mov %%" #seg ",%0":"=r" (value) : : "memory")
 
 static inline unsigned long get_limit(unsigned long segment)
 {
@@ -269,6 +269,7 @@ static inline void xen_wbinvd(void)
 #ifdef CONFIG_X86_64
 #define read_cr8()     (xen_read_cr8())
 #define write_cr8(x)   (xen_write_cr8(x))
+#define load_gs_index   xen_load_gs_index
 #endif
 
 /* Clear the 'TS' bit */
@@ -287,13 +288,12 @@ static inline void clflush(volatile void
 void disable_hlt(void);
 void enable_hlt(void);
 
-extern int es7000_plat;
 void cpu_idle_wait(void);
 
 extern unsigned long arch_align_stack(unsigned long sp);
 extern void free_init_pages(char *what, unsigned long begin, unsigned long end);
 
-void default_idle(void);
+void xen_idle(void);
 
 /*
  * Force strict CPU ordering.
Index: head-2008-12-01/include/asm-x86/mach-xen/asm/xor_64.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/asm/xor_64.h  2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/asm-x86/mach-xen/asm/xor_64.h       2008-12-01 11:49:07.000000000 +0100
@@ -1,3 +1,6 @@
+#ifndef ASM_X86__XOR_64_H
+#define ASM_X86__XOR_64_H
+
 /*
  * x86-64 changes / gcc fixes from Andi Kleen.
  * Copyright 2002 Andi Kleen, SuSE Labs.
@@ -330,3 +333,5 @@ do {                                                \
    We may also be able to load into the L1 only depending on how the cpu
    deals with a load to a line that is being prefetched.  */
 #define XOR_SELECT_TEMPLATE(FASTEST) (&xor_block_sse)
+
+#endif /* ASM_X86__XOR_64_H */
Index: head-2008-12-01/include/asm-x86/mach-xen/irq_vectors.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/irq_vectors.h 2008-12-01 11:37:10.000000000 +0100
+++ /dev/null   1970-01-01 00:00:00.000000000 +0000
@@ -1,126 +0,0 @@
-/*
- * This file should contain #defines for all of the interrupt vector
- * numbers used by this architecture.
- *
- * In addition, there are some standard defines:
- *
- *     FIRST_EXTERNAL_VECTOR:
- *             The first free place for external interrupts
- *
- *     SYSCALL_VECTOR:
- *             The IRQ vector a syscall makes the user to kernel transition
- *             under.
- *
- *     TIMER_IRQ:
- *             The IRQ number the timer interrupt comes in at.
- *
- *     NR_IRQS:
- *             The total number of interrupt vectors (including all the
- *             architecture specific interrupts) needed.
- *
- */                    
-#ifndef _ASM_IRQ_VECTORS_H
-#define _ASM_IRQ_VECTORS_H
-
-/*
- * IDT vectors usable for external interrupt sources start
- * at 0x20:
- */
-#define FIRST_EXTERNAL_VECTOR  0x20
-
-#define SYSCALL_VECTOR         0x80
-
-/*
- * Vectors 0x20-0x2f are used for ISA interrupts.
- */
-
-#if 0
-/*
- * Special IRQ vectors used by the SMP architecture, 0xf0-0xff
- *
- *  some of the following vectors are 'rare', they are merged
- *  into a single vector (CALL_FUNCTION_VECTOR) to save vector space.
- *  TLB, reschedule and local APIC vectors are performance-critical.
- *
- *  Vectors 0xf0-0xfa are free (reserved for future Linux use).
- */
-#define SPURIOUS_APIC_VECTOR   0xff
-#define ERROR_APIC_VECTOR      0xfe
-#define INVALIDATE_TLB_VECTOR  0xfd
-#define RESCHEDULE_VECTOR      0xfc
-#define CALL_FUNCTION_VECTOR   0xfb
-
-#define THERMAL_APIC_VECTOR    0xf0
-/*
- * Local APIC timer IRQ vector is on a different priority level,
- * to work around the 'lost local interrupt if more than 2 IRQ
- * sources per level' errata.
- */
-#define LOCAL_TIMER_VECTOR     0xef
-#endif
-
-#define SPURIOUS_APIC_VECTOR   0xff
-#define ERROR_APIC_VECTOR      0xfe
-
-/*
- * First APIC vector available to drivers: (vectors 0x30-0xee)
- * we start at 0x31 to spread out vectors evenly between priority
- * levels. (0x80 is the syscall vector)
- */
-#define FIRST_DEVICE_VECTOR    0x31
-#define FIRST_SYSTEM_VECTOR    0xef
-
-/*
- * 16 8259A IRQ's, 208 potential APIC interrupt sources.
- * Right now the APIC is mostly only used for SMP.
- * 256 vectors is an architectural limit. (we can have
- * more than 256 devices theoretically, but they will
- * have to use shared interrupts)
- * Since vectors 0x00-0x1f are used/reserved for the CPU,
- * the usable vector space is 0x20-0xff (224 vectors)
- */
-
-#define RESCHEDULE_VECTOR      0
-#define CALL_FUNCTION_VECTOR   1
-#define SPIN_UNLOCK_VECTOR     2
-#define NR_IPIS                        3
-
-/*
- * The maximum number of vectors supported by i386 processors
- * is limited to 256. For processors other than i386, NR_VECTORS
- * should be changed accordingly.
- */
-#define NR_VECTORS 256
-
-#define FPU_IRQ                        13
-
-#define        FIRST_VM86_IRQ          3
-#define LAST_VM86_IRQ          15
-#define invalid_vm86_irq(irq)  ((irq) < 3 || (irq) > 15)
-
-/*
- * The flat IRQ space is divided into two regions:
- *  1. A one-to-one mapping of real physical IRQs. This space is only used
- *     if we have physical device-access privilege. This region is at the 
- *     start of the IRQ space so that existing device drivers do not need
- *     to be modified to translate physical IRQ numbers into our IRQ space.
- *  3. A dynamic mapping of inter-domain and Xen-sourced virtual IRQs. These
- *     are bound using the provided bind/unbind functions.
- */
-
-#define PIRQ_BASE              0
-#if !defined(MAX_IO_APICS)
-# define NR_PIRQS              (NR_VECTORS + 32 * NR_CPUS)
-#elif NR_CPUS < MAX_IO_APICS
-# define NR_PIRQS              (NR_VECTORS + 32 * NR_CPUS)
-#else
-# define NR_PIRQS              (NR_VECTORS + 32 * MAX_IO_APICS)
-#endif
-
-#define DYNIRQ_BASE            (PIRQ_BASE + NR_PIRQS)
-#define NR_DYNIRQS             256
-
-#define NR_IRQS                        (NR_PIRQS + NR_DYNIRQS)
-#define NR_IRQ_VECTORS         NR_IRQS
-
-#endif /* _ASM_IRQ_VECTORS_H */
Index: head-2008-12-01/include/asm-x86/mach-xen/setup_arch_post.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/setup_arch_post.h     2008-12-03 15:48:43.000000000 +0100
+++ /dev/null   1970-01-01 00:00:00.000000000 +0000
@@ -1,63 +0,0 @@
-/**
- * machine_specific_* - Hooks for machine specific setup.
- *
- * Description:
- *     This is included late in kernel/setup.c so that it can make
- *     use of all of the static functions.
- **/
-
-#include <xen/interface/callback.h>
-
-extern void hypervisor_callback(void);
-extern void failsafe_callback(void);
-extern void nmi(void);
-
-static void __init machine_specific_arch_setup(void)
-{
-       int ret;
-       static struct callback_register __initdata event = {
-               .type = CALLBACKTYPE_event,
-               .address = (unsigned long) hypervisor_callback,
-       };
-       static struct callback_register __initdata failsafe = {
-               .type = CALLBACKTYPE_failsafe,
-               .address = (unsigned long)failsafe_callback,
-       };
-       static struct callback_register __initdata syscall = {
-               .type = CALLBACKTYPE_syscall,
-               .address = (unsigned long)system_call,
-       };
-#ifdef CONFIG_X86_LOCAL_APIC
-       static struct callback_register __initdata nmi_cb = {
-               .type = CALLBACKTYPE_nmi,
-               .address = (unsigned long)nmi,
-       };
-#endif
-
-       ret = HYPERVISOR_callback_op(CALLBACKOP_register, &event);
-       if (ret == 0)
-               ret = HYPERVISOR_callback_op(CALLBACKOP_register, &failsafe);
-       if (ret == 0)
-               ret = HYPERVISOR_callback_op(CALLBACKOP_register, &syscall);
-#if CONFIG_XEN_COMPAT <= 0x030002
-       if (ret == -ENOSYS)
-               ret = HYPERVISOR_set_callbacks(
-                       event.address,
-                       failsafe.address,
-                       syscall.address);
-#endif
-       BUG_ON(ret);
-
-#ifdef CONFIG_X86_LOCAL_APIC
-       ret = HYPERVISOR_callback_op(CALLBACKOP_register, &nmi_cb);
-#if CONFIG_XEN_COMPAT <= 0x030002
-       if (ret == -ENOSYS) {
-               static struct xennmi_callback __initdata cb = {
-                       .handler_address = (unsigned long)nmi
-               };
-
-               HYPERVISOR_nmi_op(XENNMI_register_callback, &cb);
-       }
-#endif
-#endif
-}
Index: head-2008-12-01/include/asm-x86/mach-xen/setup_arch_pre.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/mach-xen/setup_arch_pre.h      2008-12-03 15:48:43.000000000 +0100
+++ /dev/null   1970-01-01 00:00:00.000000000 +0000
@@ -1,5 +0,0 @@
-/* Hook to call BIOS initialisation function */
-
-#define ARCH_SETUP machine_specific_arch_setup();
-
-static void __init machine_specific_arch_setup(void);
Index: head-2008-12-01/include/asm-x86/traps.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/traps.h        2008-12-03 15:48:43.000000000 +0100
+++ head-2008-12-01/include/asm-x86/traps.h     2008-12-01 11:49:07.000000000 +0100
@@ -23,6 +23,9 @@ asmlinkage void spurious_interrupt_bug(v
 #ifdef CONFIG_X86_MCE
 asmlinkage void machine_check(void);
 #endif /* CONFIG_X86_MCE */
+#ifdef CONFIG_X86_XEN
+asmlinkage void fixup_4gb_segment(void);
+#endif
 
 void do_divide_error(struct pt_regs *, long);
 void do_overflow(struct pt_regs *, long);
@@ -48,6 +51,9 @@ void math_error(void __user *);
 void do_coprocessor_error(struct pt_regs *, long);
 void do_simd_coprocessor_error(struct pt_regs *, long);
 void do_spurious_interrupt_bug(struct pt_regs *, long);
+#ifdef CONFIG_XEN
+void do_fixup_4gb_segment(struct pt_regs *, long);
+#endif
 unsigned long patch_espfix_desc(unsigned long, unsigned long);
 asmlinkage void math_emulate(long);
 
Index: head-2008-12-01/include/asm-x86/xen/hypercall.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/xen/hypercall.h        2008-12-03 15:48:43.000000000 +0100
+++ head-2008-12-01/include/asm-x86/xen/hypercall.h     2008-12-01 11:49:07.000000000 +0100
@@ -264,7 +264,7 @@ HYPERVISOR_fpu_taskswitch(int set)
 static inline int
 HYPERVISOR_sched_op(int cmd, void *arg)
 {
-       return _hypercall2(int, sched_op_new, cmd, arg);
+       return _hypercall2(int, sched_op, cmd, arg);
 }
 
 static inline long
Index: head-2008-12-01/include/asm-x86/xen/interface_64.h
===================================================================
--- head-2008-12-01.orig/include/asm-x86/xen/interface_64.h     2008-12-03 15:48:43.000000000 +0100
+++ head-2008-12-01/include/asm-x86/xen/interface_64.h  2008-12-01 11:49:07.000000000 +0100
@@ -136,7 +136,7 @@ struct cpu_user_regs {
     uint16_t fs, _pad5[3]; /* Non-zero => takes precedence over fs_base.     */
     uint16_t gs, _pad6[3]; /* Non-zero => takes precedence over gs_base_usr. */
 };
-DEFINE_GUEST_HANDLE_STRUCT(cpu_user_regs);
+DEFINE_XEN_GUEST_HANDLE_STRUCT(cpu_user_regs);
 
 #undef __DECL_REG
 
Index: head-2008-12-01/include/linux/page-flags.h
===================================================================
--- head-2008-12-01.orig/include/linux/page-flags.h     2008-12-01 11:48:52.000000000 +0100
+++ head-2008-12-01/include/linux/page-flags.h  2008-12-01 11:49:07.000000000 +0100
@@ -109,9 +109,11 @@ enum pageflags {
        /* Filesystems */
        PG_checked = PG_owner_priv_1,
 
+#ifdef CONFIG_PARAVIRT_XEN
        /* XEN */
        PG_pinned = PG_owner_priv_1,
        PG_savepinned = PG_dirty,
+#endif
 
        /* SLOB */
        PG_slob_page = PG_active,
@@ -185,8 +187,12 @@ PAGEFLAG(LRU, lru) __CLEARPAGEFLAG(LRU, 
 PAGEFLAG(Active, active) __CLEARPAGEFLAG(Active, active)
 __PAGEFLAG(Slab, slab)
 PAGEFLAG(Checked, checked)             /* Used by some filesystems */
+#if defined(CONFIG_XEN) || defined(CONFIG_PARAVIRT_XEN)
 PAGEFLAG(Pinned, pinned) TESTSCFLAG(Pinned, pinned)    /* Xen */
+#endif
+#ifdef CONFIG_PARAVIRT_XEN
 PAGEFLAG(SavePinned, savepinned);                      /* Xen */
+#endif
 PAGEFLAG(Reserved, reserved) __CLEARPAGEFLAG(Reserved, reserved)
 PAGEFLAG(Private, private) __CLEARPAGEFLAG(Private, private)
        __SETPAGEFLAG(Private, private)
Index: head-2008-12-01/include/xen/interface/memory.h
===================================================================
--- head-2008-12-01.orig/include/xen/interface/memory.h 2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/include/xen/interface/memory.h      2008-12-01 11:49:07.000000000 +0100
@@ -82,6 +82,7 @@ struct xen_memory_reservation {
     domid_t        domid;
 
 };
+DEFINE_XEN_GUEST_HANDLE_STRUCT(xen_memory_reservation);
 typedef struct xen_memory_reservation xen_memory_reservation_t;
 DEFINE_XEN_GUEST_HANDLE(xen_memory_reservation_t);
 
@@ -167,11 +168,7 @@ struct xen_machphys_mfn_list {
      * any large discontiguities in the machine address space, 2MB gaps in
      * the machphys table will be represented by an MFN base of zero.
      */
-#ifndef CONFIG_PARAVIRT_XEN
     XEN_GUEST_HANDLE(xen_pfn_t) extent_start;
-#else
-    ulong extent_start;
-#endif
 
     /*
      * Number of extents written to the above array. This will be smaller
@@ -179,6 +176,7 @@ struct xen_machphys_mfn_list {
      */
     unsigned int nr_extents;
 };
+DEFINE_XEN_GUEST_HANDLE_STRUCT(xen_machphys_mfn_list);
 typedef struct xen_machphys_mfn_list xen_machphys_mfn_list_t;
 DEFINE_XEN_GUEST_HANDLE(xen_machphys_mfn_list_t);
 
@@ -218,6 +216,7 @@ struct xen_add_to_physmap {
     /* GPFN where the source mapping page should appear. */
     xen_pfn_t     gpfn;
 };
+DEFINE_XEN_GUEST_HANDLE_STRUCT(xen_add_to_physmap);
 typedef struct xen_add_to_physmap xen_add_to_physmap_t;
 DEFINE_XEN_GUEST_HANDLE(xen_add_to_physmap_t);
 
@@ -250,21 +249,13 @@ struct xen_translate_gpfn_list {
     xen_ulong_t nr_gpfns;
 
     /* List of GPFNs to translate. */
-#ifndef CONFIG_PARAVIRT_XEN
     XEN_GUEST_HANDLE(xen_pfn_t) gpfn_list;
-#else
-    ulong gpfn_list;
-#endif
 
     /*
      * Output list to contain MFN translations. May be the same as the input
      * list (in which case each input GPFN is overwritten with the output MFN).
      */
-#ifndef CONFIG_PARAVIRT_XEN
     XEN_GUEST_HANDLE(xen_pfn_t) mfn_list;
-#else
-    ulong mfn_list;
-#endif
 };
 DEFINE_XEN_GUEST_HANDLE_STRUCT(xen_translate_gpfn_list);
 typedef struct xen_translate_gpfn_list xen_translate_gpfn_list_t;
Index: head-2008-12-01/kernel/hrtimer.c
===================================================================
--- head-2008-12-01.orig/kernel/hrtimer.c       2008-12-03 15:48:43.000000000 +0100
+++ head-2008-12-01/kernel/hrtimer.c    2008-12-01 11:49:07.000000000 +0100
@@ -1084,7 +1084,7 @@ ktime_t hrtimer_get_remaining(const stru
 }
 EXPORT_SYMBOL_GPL(hrtimer_get_remaining);
 
-#ifdef CONFIG_NO_HZ
+#if defined(CONFIG_NO_HZ) || defined(CONFIG_NO_IDLE_HZ)
 /**
  * hrtimer_get_next_event - get the time until next expiry event
  *
Index: head-2008-12-01/kernel/timer.c
===================================================================
--- head-2008-12-01.orig/kernel/timer.c 2008-12-03 15:48:43.000000000 +0100
+++ head-2008-12-01/kernel/timer.c      2008-12-01 11:49:07.000000000 +0100
@@ -815,7 +815,7 @@ static inline void __run_timers(struct t
        spin_unlock_irq(&base->lock);
 }
 
-#ifdef CONFIG_NO_HZ
+#if defined(CONFIG_NO_HZ) || defined(CONFIG_NO_IDLE_HZ)
 /*
  * Find out when the next timer event is due to happen. This
  * is used on S/390 to stop all activity when a cpus is idle.
Index: head-2008-12-01/lib/swiotlb-xen.c
===================================================================
--- head-2008-12-01.orig/lib/swiotlb-xen.c      2008-12-01 11:44:55.000000000 +0100
+++ head-2008-12-01/lib/swiotlb-xen.c   2008-12-01 11:49:07.000000000 +0100
@@ -750,7 +750,7 @@ swiotlb_sync_sg_for_device(struct device
 }
 
 int
-swiotlb_dma_mapping_error(dma_addr_t dma_addr)
+swiotlb_dma_mapping_error(struct device *hwdev, dma_addr_t dma_addr)
 {
        return (dma_addr == virt_to_bus(io_tlb_overflow_buffer));
 }
Index: head-2008-12-01/mm/mprotect.c
===================================================================
--- head-2008-12-01.orig/mm/mprotect.c  2008-12-01 11:29:05.000000000 +0100
+++ head-2008-12-01/mm/mprotect.c       2008-12-01 11:49:07.000000000 +0100
@@ -92,8 +92,6 @@ static inline void change_pmd_range(stru
                next = pmd_addr_end(addr, end);
                if (pmd_none_or_clear_bad(pmd))
                        continue;
-               if (arch_change_pte_range(mm, pmd, addr, next, newprot, dirty_accountable))
-                       continue;
                change_pte_range(mm, pmd, addr, next, newprot, dirty_accountable);
        } while (pmd++, addr = next, addr != end);
 }