powerpc: Remaining 64-bit Book3E support

This contains all the bits that didn't fit in previous patches :-) This
includes the actual exception handlers assembly, the changes to the
kernel entry, other misc bits and wiring it all up in Kconfig.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>

6 files changed, 951 insertions, 17 deletions

diff --git a/arch/powerpc/kernel/Makefile b/arch/powerpc/kernel/Makefile
index b73396b93905..035946f9d5fb 100644
--- a/arch/powerpc/kernel/Makefile
+++ b/arch/powerpc/kernel/Makefile
@@ -33,10 +33,10 @@ obj-y				:= cputable.o ptrace.o syscalls.o \
 obj-y                           += vdso32/
 obj-$(CONFIG_PPC64)             += setup_64.o sys_ppc32.o \
                                    signal_64.o ptrace32.o \
-                                   paca.o cpu_setup_ppc970.o \
-                                   cpu_setup_pa6t.o \
-                                   firmware.o nvram_64.o
+                                   paca.o nvram_64.o firmware.o
+obj-$(CONFIG_PPC_BOOK3S_64)     += cpu_setup_ppc970.o cpu_setup_pa6t.o
 obj64-$(CONFIG_RELOCATABLE)     += reloc_64.o
+obj-$(CONFIG_PPC_BOOK3E_64)     += exceptions-64e.o
 obj-$(CONFIG_PPC64)             += vdso64/
 obj-$(CONFIG_ALTIVEC)           += vecemu.o
 obj-$(CONFIG_PPC_970_NAP)       += idle_power4.o
@@ -63,8 +63,8 @@ obj-$(CONFIG_MODULES)		+= module.o module_$(CONFIG_WORD_SIZE).o
 obj-$(CONFIG_44x)               += cpu_setup_44x.o
 obj-$(CONFIG_FSL_BOOKE)         += cpu_setup_fsl_booke.o dbell.o
 
-extra-$(CONFIG_PPC_STD_MMU)     := head_32.o
-extra-$(CONFIG_PPC64)           := head_64.o
+extra-y                         := head_$(CONFIG_WORD_SIZE).o
+extra-$(CONFIG_PPC_BOOK3E_32)   := head_new_booke.o
 extra-$(CONFIG_40x)             := head_40x.o
 extra-$(CONFIG_44x)             := head_44x.o
 extra-$(CONFIG_FSL_BOOKE)       := head_fsl_booke.o
diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c
index 4a24a2fc4574..f34ea37079b5 100644
--- a/arch/powerpc/kernel/cputable.c
+++ b/arch/powerpc/kernel/cputable.c
@@ -93,7 +93,7 @@ extern void __restore_cpu_power7(void);
                                  PPC_FEATURE_BOOKE)
 
 static struct cpu_spec __initdata cpu_specs[] = {
-#ifdef CONFIG_PPC64
+#ifdef CONFIG_PPC_BOOK3S_64
         {       /* Power3 */
                 .pvr_mask               = 0xffff0000,
                 .pvr_value              = 0x00400000,
@@ -508,7 +508,30 @@ static struct cpu_spec __initdata cpu_specs[] = {
                 .machine_check          = machine_check_generic,
                 .platform               = "power4",
         }
-#endif  /* CONFIG_PPC64 */
+#endif  /* CONFIG_PPC_BOOK3S_64 */
+#ifdef CONFIG_PPC_BOOK3E_64
+        {       /* This is a default entry to get going, to be replaced by
+                 * a real one at some stage
+                 */
+#define CPU_FTRS_BASE_BOOK3E    (CPU_FTR_USE_TB | \
+            CPU_FTR_PPCAS_ARCH_V2 | CPU_FTR_SMT | \
+            CPU_FTR_NODSISRALIGN | CPU_FTR_NOEXECUTE)
+                .pvr_mask               = 0x00000000,
+                .pvr_value              = 0x00000000,
+                .cpu_name               = "Book3E",
+                .cpu_features           = CPU_FTRS_BASE_BOOK3E,
+                .cpu_user_features      = COMMON_USER_PPC64,
+                .mmu_features           = MMU_FTR_TYPE_3E | MMU_FTR_USE_TLBILX |
+                                          MMU_FTR_USE_TLBIVAX_BCAST |
+                                          MMU_FTR_LOCK_BCAST_INVAL,
+                .icache_bsize           = 64,
+                .dcache_bsize           = 64,
+                .num_pmcs               = 0,
+                .machine_check          = machine_check_generic,
+                .platform               = "power6",
+        },
+#endif
+
 #ifdef CONFIG_PPC32
 #if CLASSIC_PPC
         {       /* 601 */
diff --git a/arch/powerpc/kernel/entry_64.S b/arch/powerpc/kernel/entry_64.S
index 1cb0f3d1714b..66bcda34a6bb 100644
--- a/arch/powerpc/kernel/entry_64.S
+++ b/arch/powerpc/kernel/entry_64.S
@@ -120,9 +120,15 @@ BEGIN_FW_FTR_SECTION
 2:
 END_FW_FTR_SECTION_IFSET(FW_FEATURE_ISERIES)
 #endif /* CONFIG_PPC_ISERIES */
+
+        /* Hard enable interrupts */
+#ifdef CONFIG_PPC_BOOK3E
+        wrteei  1
+#else
         mfmsr   r11
         ori     r11,r11,MSR_EE
         mtmsrd  r11,1
+#endif /* CONFIG_PPC_BOOK3E */
 
 #ifdef SHOW_SYSCALLS
         bl      .do_show_syscall
@@ -168,15 +174,25 @@ syscall_exit:
 #endif
         clrrdi  r12,r1,THREAD_SHIFT
 
-        /* disable interrupts so current_thread_info()->flags can't change,
-           and so that we don't get interrupted after loading SRR0/1. */
         ld      r8,_MSR(r1)
+#ifdef CONFIG_PPC_BOOK3S
+        /* No MSR:RI on BookE */
         andi.   r10,r8,MSR_RI
         beq-    unrecov_restore
+#endif
+
+        /* Disable interrupts so current_thread_info()->flags can't change,
+         * and so that we don't get interrupted after loading SRR0/1.
+         */
+#ifdef CONFIG_PPC_BOOK3E
+        wrteei  0
+#else
         mfmsr   r10
         rldicl  r10,r10,48,1
         rotldi  r10,r10,16
         mtmsrd  r10,1
+#endif /* CONFIG_PPC_BOOK3E */
+
         ld      r9,TI_FLAGS(r12)
         li      r11,-_LAST_ERRNO
         andi.   r0,r9,(_TIF_SYSCALL_T_OR_A|_TIF_SINGLESTEP|_TIF_USER_WORK_MASK|_TIF_PERSYSCALL_MASK)
@@ -194,9 +210,13 @@ syscall_error_cont:
          * userspace and we take an exception after restoring r13,
          * we end up corrupting the userspace r13 value.
          */
+#ifdef CONFIG_PPC_BOOK3S
+        /* No MSR:RI on BookE */
         li      r12,MSR_RI
         andc    r11,r10,r12
         mtmsrd  r11,1                   /* clear MSR.RI */
+#endif /* CONFIG_PPC_BOOK3S */
+
         beq-    1f
         ACCOUNT_CPU_USER_EXIT(r11, r12)
         ld      r13,GPR13(r1)   /* only restore r13 if returning to usermode */
@@ -206,7 +226,7 @@ syscall_error_cont:
         mtcr    r5
         mtspr   SPRN_SRR0,r7
         mtspr   SPRN_SRR1,r8
-        rfid
+        RFI
         b       .       /* prevent speculative execution */
 
 syscall_error:  
@@ -276,9 +296,13 @@ syscall_exit_work:
         beq     .ret_from_except_lite
 
         /* Re-enable interrupts */
+#ifdef CONFIG_PPC_BOOK3E
+        wrteei  1
+#else
         mfmsr   r10
         ori     r10,r10,MSR_EE
         mtmsrd  r10,1
+#endif /* CONFIG_PPC_BOOK3E */
 
         bl      .save_nvgprs
         addi    r3,r1,STACK_FRAME_OVERHEAD
@@ -380,7 +404,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
         and.    r0,r0,r22
         beq+    1f
         andc    r22,r22,r0
-        mtmsrd  r22
+        MTMSRD(r22)
         isync
 1:      std     r20,_NIP(r1)
         mfcr    r23
@@ -399,6 +423,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
         std     r6,PACACURRENT(r13)     /* Set new 'current' */
 
         ld      r8,KSP(r4)      /* new stack pointer */
+#ifdef CONFIG_PPC_BOOK3S
 BEGIN_FTR_SECTION
   BEGIN_FTR_SECTION_NESTED(95)
         clrrdi  r6,r8,28        /* get its ESID */
@@ -445,8 +470,9 @@ END_FTR_SECTION_IFSET(CPU_FTR_1T_SEGMENT)
         slbie   r6              /* Workaround POWER5 < DD2.1 issue */
         slbmte  r7,r0
         isync
-
 2:
+#endif /* !CONFIG_PPC_BOOK3S */
+
         clrrdi  r7,r8,THREAD_SHIFT      /* base of new stack */
         /* Note: this uses SWITCH_FRAME_SIZE rather than INT_FRAME_SIZE
            because we don't need to leave the 288-byte ABI gap at the
@@ -490,10 +516,14 @@ _GLOBAL(ret_from_except_lite)
          * can't change between when we test it and when we return
          * from the interrupt.
          */
+#ifdef CONFIG_PPC_BOOK3E
+        wrteei  0
+#else
         mfmsr   r10             /* Get current interrupt state */
         rldicl  r9,r10,48,1     /* clear MSR_EE */
         rotldi  r9,r9,16
         mtmsrd  r9,1            /* Update machine state */
+#endif /* CONFIG_PPC_BOOK3E */
 
 #ifdef CONFIG_PREEMPT
         clrrdi  r9,r1,THREAD_SHIFT      /* current_thread_info() */
@@ -540,6 +570,9 @@ ALT_FW_FTR_SECTION_END_IFCLR(FW_FEATURE_ISERIES)
         rldicl  r4,r3,49,63             /* r0 = (r3 >> 15) & 1 */
         stb     r4,PACAHARDIRQEN(r13)
 
+#ifdef CONFIG_PPC_BOOK3E
+        b       .exception_return_book3e
+#else
         ld      r4,_CTR(r1)
         ld      r0,_LINK(r1)
         mtctr   r4
@@ -588,6 +621,8 @@ ALT_FW_FTR_SECTION_END_IFCLR(FW_FEATURE_ISERIES)
         rfid
         b       .       /* prevent speculative execution */
 
+#endif /* CONFIG_PPC_BOOK3E */
+
 iseries_check_pending_irqs:
 #ifdef CONFIG_PPC_ISERIES
         ld      r5,SOFTE(r1)
@@ -638,6 +673,11 @@ do_work:
         li      r0,1
         stb     r0,PACASOFTIRQEN(r13)
         stb     r0,PACAHARDIRQEN(r13)
+#ifdef CONFIG_PPC_BOOK3E
+        wrteei  1
+        bl      .preempt_schedule
+        wrteei  0
+#else
         ori     r10,r10,MSR_EE
         mtmsrd  r10,1           /* reenable interrupts */
         bl      .preempt_schedule
@@ -646,6 +686,7 @@ do_work:
         rldicl  r10,r10,48,1    /* disable interrupts again */
         rotldi  r10,r10,16
         mtmsrd  r10,1
+#endif /* CONFIG_PPC_BOOK3E */
         ld      r4,TI_FLAGS(r9)
         andi.   r0,r4,_TIF_NEED_RESCHED
         bne     1b
@@ -654,8 +695,12 @@ do_work:
 user_work:
 #endif
         /* Enable interrupts */
+#ifdef CONFIG_PPC_BOOK3E
+        wrteei  1
+#else
         ori     r10,r10,MSR_EE
         mtmsrd  r10,1
+#endif /* CONFIG_PPC_BOOK3E */
 
         andi.   r0,r4,_TIF_NEED_RESCHED
         beq     1f
@@ -837,6 +882,10 @@ _GLOBAL(enter_prom)
 
         /* Switch MSR to 32 bits mode
          */
+#ifdef CONFIG_PPC_BOOK3E
+        rlwinm  r11,r11,0,1,31
+        mtmsr   r11
+#else /* CONFIG_PPC_BOOK3E */
         mfmsr   r11
         li      r12,1
         rldicr  r12,r12,MSR_SF_LG,(63-MSR_SF_LG)
@@ -845,6 +894,7 @@ _GLOBAL(enter_prom)
         rldicr  r12,r12,MSR_ISF_LG,(63-MSR_ISF_LG)
         andc    r11,r11,r12
         mtmsrd  r11
+#endif /* CONFIG_PPC_BOOK3E */
         isync
 
         /* Enter PROM here... */
diff --git a/arch/powerpc/kernel/exceptions-64e.S b/arch/powerpc/kernel/exceptions-64e.S
new file mode 100644
index 000000000000..695d4847d228
--- /dev/null
+++ b/arch/powerpc/kernel/exceptions-64e.S
@@ -0,0 +1,784 @@
+/*
+ *  Boot code and exception vectors for Book3E processors
+ *
+ *  Copyright (C) 2007 Ben. Herrenschmidt (benh@kernel.crashing.org), IBM Corp.
+ *
+ *  This program is free software; you can redistribute it and/or
+ *  modify it under the terms of the GNU General Public License
+ *  as published by the Free Software Foundation; either version
+ *  2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/threads.h>
+#include <asm/reg.h>
+#include <asm/page.h>
+#include <asm/ppc_asm.h>
+#include <asm/asm-offsets.h>
+#include <asm/cputable.h>
+#include <asm/setup.h>
+#include <asm/thread_info.h>
+#include <asm/reg.h>
+#include <asm/exception-64e.h>
+#include <asm/bug.h>
+#include <asm/irqflags.h>
+#include <asm/ptrace.h>
+#include <asm/ppc-opcode.h>
+#include <asm/mmu.h>
+
+/* XXX This will ultimately add space for a special exception save
+ *     structure used to save things like SRR0/SRR1, SPRGs, MAS, etc...
+ *     when taking special interrupts. For now we don't support that,
+ *     special interrupts from within a non-standard level will probably
+ *     blow you up
+ */
+#define SPECIAL_EXC_FRAME_SIZE  INT_FRAME_SIZE
+
+/* Exception prolog code for all exceptions */
+#define EXCEPTION_PROLOG(n, type, addition)                                 \
+        mtspr   SPRN_SPRG_##type##_SCRATCH,r13; /* get spare registers */   \
+        mfspr   r13,SPRN_SPRG_PACA;     /* get PACA */                      \
+        std     r10,PACA_EX##type+EX_R10(r13);                              \
+        std     r11,PACA_EX##type+EX_R11(r13);                              \
+        mfcr    r10;                    /* save CR */                       \
+        addition;                       /* additional code for that exc. */ \
+        std     r1,PACA_EX##type+EX_R1(r13); /* save old r1 in the PACA */  \
+        stw     r10,PACA_EX##type+EX_CR(r13); /* save old CR in the PACA */ \
+        mfspr   r11,SPRN_##type##_SRR1;/* what are we coming from */        \
+        type##_SET_KSTACK;              /* get special stack if necessary */\
+        andi.   r10,r11,MSR_PR;         /* save stack pointer */            \
+        beq     1f;                     /* branch around if supervisor */   \
+        ld      r1,PACAKSAVE(r13);      /* get kernel stack coming from usr */\
+1:      cmpdi   cr1,r1,0;               /* check if SP makes sense */       \
+        bge-    cr1,exc_##n##_bad_stack;/* bad stack (TODO: out of line) */ \
+        mfspr   r10,SPRN_##type##_SRR0; /* read SRR0 before touching stack */
+
+/* Exception type-specific macros */
+#define GEN_SET_KSTACK                                                      \
+        subi    r1,r1,INT_FRAME_SIZE;   /* alloc frame on kernel stack */
+#define SPRN_GEN_SRR0   SPRN_SRR0
+#define SPRN_GEN_SRR1   SPRN_SRR1
+
+#define CRIT_SET_KSTACK                                                     \
+        ld      r1,PACA_CRIT_STACK(r13);                                    \
+        subi    r1,r1,SPECIAL_EXC_FRAME_SIZE;
+#define SPRN_CRIT_SRR0  SPRN_CSRR0
+#define SPRN_CRIT_SRR1  SPRN_CSRR1
+
+#define DBG_SET_KSTACK                                                      \
+        ld      r1,PACA_DBG_STACK(r13);                                     \
+        subi    r1,r1,SPECIAL_EXC_FRAME_SIZE;
+#define SPRN_DBG_SRR0   SPRN_DSRR0
+#define SPRN_DBG_SRR1   SPRN_DSRR1
+
+#define MC_SET_KSTACK                                                       \
+        ld      r1,PACA_MC_STACK(r13);                                      \
+        subi    r1,r1,SPECIAL_EXC_FRAME_SIZE;
+#define SPRN_MC_SRR0    SPRN_MCSRR0
+#define SPRN_MC_SRR1    SPRN_MCSRR1
+
+#define NORMAL_EXCEPTION_PROLOG(n, addition)                                \
+        EXCEPTION_PROLOG(n, GEN, addition##_GEN)
+
+#define CRIT_EXCEPTION_PROLOG(n, addition)                                  \
+        EXCEPTION_PROLOG(n, CRIT, addition##_CRIT)
+
+#define DBG_EXCEPTION_PROLOG(n, addition)                                   \
+        EXCEPTION_PROLOG(n, DBG, addition##_DBG)
+
+#define MC_EXCEPTION_PROLOG(n, addition)                                    \
+        EXCEPTION_PROLOG(n, MC, addition##_MC)
+
+
+/* Variants of the "addition" argument for the prolog
+ */
+#define PROLOG_ADDITION_NONE_GEN
+#define PROLOG_ADDITION_NONE_CRIT
+#define PROLOG_ADDITION_NONE_DBG
+#define PROLOG_ADDITION_NONE_MC
+
+#define PROLOG_ADDITION_MASKABLE_GEN                                        \
+        lbz     r11,PACASOFTIRQEN(r13); /* are irqs soft-disabled ? */      \
+        cmpwi   cr0,r11,0;              /* yes -> go out of line */         \
+        beq     masked_interrupt_book3e;
+
+#define PROLOG_ADDITION_2REGS_GEN                                           \
+        std     r14,PACA_EXGEN+EX_R14(r13);                                 \
+        std     r15,PACA_EXGEN+EX_R15(r13)
+
+#define PROLOG_ADDITION_1REG_GEN                                            \
+        std     r14,PACA_EXGEN+EX_R14(r13);
+
+#define PROLOG_ADDITION_2REGS_CRIT                                          \
+        std     r14,PACA_EXCRIT+EX_R14(r13);                                \
+        std     r15,PACA_EXCRIT+EX_R15(r13)
+
+#define PROLOG_ADDITION_2REGS_DBG                                           \
+        std     r14,PACA_EXDBG+EX_R14(r13);                                 \
+        std     r15,PACA_EXDBG+EX_R15(r13)
+
+#define PROLOG_ADDITION_2REGS_MC                                            \
+        std     r14,PACA_EXMC+EX_R14(r13);                                  \
+        std     r15,PACA_EXMC+EX_R15(r13)
+
+/* Core exception code for all exceptions except TLB misses.
+ * XXX: Needs to make SPRN_SPRG_GEN depend on exception type
+ */
+#define EXCEPTION_COMMON(n, excf, ints)                                     \
+        std     r0,GPR0(r1);            /* save r0 in stackframe */         \
+        std     r2,GPR2(r1);            /* save r2 in stackframe */         \
+        SAVE_4GPRS(3, r1);              /* save r3 - r6 in stackframe */    \
+        SAVE_2GPRS(7, r1);              /* save r7, r8 in stackframe */     \
+        std     r9,GPR9(r1);            /* save r9 in stackframe */         \
+        std     r10,_NIP(r1);           /* save SRR0 to stackframe */       \
+        std     r11,_MSR(r1);           /* save SRR1 to stackframe */       \
+        ACCOUNT_CPU_USER_ENTRY(r10,r11);/* accounting (uses cr0+eq) */      \
+        ld      r3,excf+EX_R10(r13);    /* get back r10 */                  \
+        ld      r4,excf+EX_R11(r13);    /* get back r11 */                  \
+        mfspr   r5,SPRN_SPRG_GEN_SCRATCH;/* get back r13 */                 \
+        std     r12,GPR12(r1);          /* save r12 in stackframe */        \
+        ld      r2,PACATOC(r13);        /* get kernel TOC into r2 */        \
+        mflr    r6;                     /* save LR in stackframe */         \
+        mfctr   r7;                     /* save CTR in stackframe */        \
+        mfspr   r8,SPRN_XER;            /* save XER in stackframe */        \
+        ld      r9,excf+EX_R1(r13);     /* load orig r1 back from PACA */   \
+        lwz     r10,excf+EX_CR(r13);    /* load orig CR back from PACA  */  \
+        lbz     r11,PACASOFTIRQEN(r13); /* get current IRQ softe */         \
+        ld      r12,exception_marker@toc(r2);                               \
+        li      r0,0;                                                       \
+        std     r3,GPR10(r1);           /* save r10 to stackframe */        \
+        std     r4,GPR11(r1);           /* save r11 to stackframe */        \
+        std     r5,GPR13(r1);           /* save it to stackframe */         \
+        std     r6,_LINK(r1);                                               \
+        std     r7,_CTR(r1);                                                \
+        std     r8,_XER(r1);                                                \
+        li      r3,(n)+1;               /* indicate partial regs in trap */ \
+        std     r9,0(r1);               /* store stack frame back link */   \
+        std     r10,_CCR(r1);           /* store orig CR in stackframe */   \
+        std     r9,GPR1(r1);            /* store stack frame back link */   \
+        std     r11,SOFTE(r1);          /* and save it to stackframe */     \
+        std     r12,STACK_FRAME_OVERHEAD-16(r1); /* mark the frame */       \
+        std     r3,_TRAP(r1);           /* set trap number              */  \
+        std     r0,RESULT(r1);          /* clear regs->result */            \
+        ints;
+
+/* Variants for the "ints" argument */
+#define INTS_KEEP
+#define INTS_DISABLE_SOFT                                                   \
+        stb     r0,PACASOFTIRQEN(r13);  /* mark interrupts soft-disabled */ \
+        TRACE_DISABLE_INTS;
+#define INTS_DISABLE_HARD                                                   \
+        stb     r0,PACAHARDIRQEN(r13); /* and hard disabled */
+#define INTS_DISABLE_ALL                                                    \
+        INTS_DISABLE_SOFT                                                   \
+        INTS_DISABLE_HARD
+
+/* This is called by exceptions that used INTS_KEEP (that is did not clear
+ * neither soft nor hard IRQ indicators in the PACA. This will restore MSR:EE
+ * to it's previous value
+ *
+ * XXX In the long run, we may want to open-code it in order to separate the
+ *     load from the wrtee, thus limiting the latency caused by the dependency
+ *     but at this point, I'll favor code clarity until we have a near to final
+ *     implementation
+ */
+#define INTS_RESTORE_HARD                                                   \
+        ld      r11,_MSR(r1);                                               \
+        wrtee   r11;
+
+/* XXX FIXME: Restore r14/r15 when necessary */
+#define BAD_STACK_TRAMPOLINE(n)                                             \
+exc_##n##_bad_stack:                                                        \
+        li      r1,(n);                 /* get exception number */          \
+        sth     r1,PACA_TRAP_SAVE(r13); /* store trap */                    \
+        b       bad_stack_book3e;       /* bad stack error */
+
+#define EXCEPTION_STUB(loc, label)                                      \
+        . = interrupt_base_book3e + loc;                                \
+        nop;    /* To make debug interrupts happy */                    \
+        b       exc_##label##_book3e;
+
+#define ACK_NONE(r)
+#define ACK_DEC(r)                                                      \
+        lis     r,TSR_DIS@h;                                            \
+        mtspr   SPRN_TSR,r
+#define ACK_FIT(r)                                                      \
+        lis     r,TSR_FIS@h;                                            \
+        mtspr   SPRN_TSR,r
+
+#define MASKABLE_EXCEPTION(trapnum, label, hdlr, ack)                   \
+        START_EXCEPTION(label);                                         \
+        NORMAL_EXCEPTION_PROLOG(trapnum, PROLOG_ADDITION_MASKABLE)      \
+        EXCEPTION_COMMON(trapnum, PACA_EXGEN, INTS_DISABLE_ALL)         \
+        ack(r8);                                                        \
+        addi    r3,r1,STACK_FRAME_OVERHEAD;                             \
+        bl      hdlr;                                                   \
+        b       .ret_from_except_lite;
+
+/* This value is used to mark exception frames on the stack. */
+        .section        ".toc","aw"
+exception_marker:
+        .tc     ID_EXC_MARKER[TC],STACK_FRAME_REGS_MARKER
+
+
+/*
+ * And here we have the exception vectors !
+ */
+
+        .text
+        .balign 0x1000
+        .globl interrupt_base_book3e
+interrupt_base_book3e:                                  /* fake trap */
+        /* Note: If real debug exceptions are supported by the HW, the vector
+         * below will have to be patched up to point to an appropriate handler
+         */
+        EXCEPTION_STUB(0x000, machine_check)            /* 0x0200 */
+        EXCEPTION_STUB(0x020, critical_input)           /* 0x0580 */
+        EXCEPTION_STUB(0x040, debug_crit)               /* 0x0d00 */
+        EXCEPTION_STUB(0x060, data_storage)             /* 0x0300 */
+        EXCEPTION_STUB(0x080, instruction_storage)      /* 0x0400 */
+        EXCEPTION_STUB(0x0a0, external_input)           /* 0x0500 */
+        EXCEPTION_STUB(0x0c0, alignment)                /* 0x0600 */
+        EXCEPTION_STUB(0x0e0, program)                  /* 0x0700 */
+        EXCEPTION_STUB(0x100, fp_unavailable)           /* 0x0800 */
+        EXCEPTION_STUB(0x120, system_call)              /* 0x0c00 */
+        EXCEPTION_STUB(0x140, ap_unavailable)           /* 0x0f20 */
+        EXCEPTION_STUB(0x160, decrementer)              /* 0x0900 */
+        EXCEPTION_STUB(0x180, fixed_interval)           /* 0x0980 */
+        EXCEPTION_STUB(0x1a0, watchdog)                 /* 0x09f0 */
+        EXCEPTION_STUB(0x1c0, data_tlb_miss)
+        EXCEPTION_STUB(0x1e0, instruction_tlb_miss)
+
+#if 0
+        EXCEPTION_STUB(0x280, processor_doorbell)
+        EXCEPTION_STUB(0x220, processor_doorbell_crit)
+#endif
+        .globl interrupt_end_book3e
+interrupt_end_book3e:
+
+/* Critical Input Interrupt */
+        START_EXCEPTION(critical_input);
+        CRIT_EXCEPTION_PROLOG(0x100, PROLOG_ADDITION_NONE)
+//      EXCEPTION_COMMON(0x100, PACA_EXCRIT, INTS_DISABLE_ALL)
+//      bl      special_reg_save_crit
+//      addi    r3,r1,STACK_FRAME_OVERHEAD
+//      bl      .critical_exception
+//      b       ret_from_crit_except
+        b       .
+
+/* Machine Check Interrupt */
+        START_EXCEPTION(machine_check);
+        CRIT_EXCEPTION_PROLOG(0x200, PROLOG_ADDITION_NONE)
+//      EXCEPTION_COMMON(0x200, PACA_EXMC, INTS_DISABLE_ALL)
+//      bl      special_reg_save_mc
+//      addi    r3,r1,STACK_FRAME_OVERHEAD
+//      bl      .machine_check_exception
+//      b       ret_from_mc_except
+        b       .
+
+/* Data Storage Interrupt */
+        START_EXCEPTION(data_storage)
+        NORMAL_EXCEPTION_PROLOG(0x300, PROLOG_ADDITION_2REGS)
+        mfspr   r14,SPRN_DEAR
+        mfspr   r15,SPRN_ESR
+        EXCEPTION_COMMON(0x300, PACA_EXGEN, INTS_KEEP)
+        b       storage_fault_common
+
+/* Instruction Storage Interrupt */
+        START_EXCEPTION(instruction_storage);
+        NORMAL_EXCEPTION_PROLOG(0x400, PROLOG_ADDITION_2REGS)
+        li      r15,0
+        mr      r14,r10
+        EXCEPTION_COMMON(0x400, PACA_EXGEN, INTS_KEEP)
+        b       storage_fault_common
+
+/* External Input Interrupt */
+        MASKABLE_EXCEPTION(0x500, external_input, .do_IRQ, ACK_NONE)
+
+/* Alignment */
+        START_EXCEPTION(alignment);
+        NORMAL_EXCEPTION_PROLOG(0x600, PROLOG_ADDITION_2REGS)
+        mfspr   r14,SPRN_DEAR
+        mfspr   r15,SPRN_ESR
+        EXCEPTION_COMMON(0x600, PACA_EXGEN, INTS_KEEP)
+        b       alignment_more  /* no room, go out of line */
+
+/* Program Interrupt */
+        START_EXCEPTION(program);
+        NORMAL_EXCEPTION_PROLOG(0x700, PROLOG_ADDITION_1REG)
+        mfspr   r14,SPRN_ESR
+        EXCEPTION_COMMON(0x700, PACA_EXGEN, INTS_DISABLE_SOFT)
+        std     r14,_DSISR(r1)
+        addi    r3,r1,STACK_FRAME_OVERHEAD
+        ld      r14,PACA_EXGEN+EX_R14(r13)
+        bl      .save_nvgprs
+        INTS_RESTORE_HARD
+        bl      .program_check_exception
+        b       .ret_from_except
+
+/* Floating Point Unavailable Interrupt */
+        START_EXCEPTION(fp_unavailable);
+        NORMAL_EXCEPTION_PROLOG(0x800, PROLOG_ADDITION_NONE)
+        /* we can probably do a shorter exception entry for that one... */
+        EXCEPTION_COMMON(0x800, PACA_EXGEN, INTS_KEEP)
+        bne     1f                      /* if from user, just load it up */
+        bl      .save_nvgprs
+        addi    r3,r1,STACK_FRAME_OVERHEAD
+        INTS_RESTORE_HARD
+        bl      .kernel_fp_unavailable_exception
+        BUG_OPCODE
+1:      ld      r12,_MSR(r1)
+        bl      .load_up_fpu
+        b       fast_exception_return
+
+/* Decrementer Interrupt */
+        MASKABLE_EXCEPTION(0x900, decrementer, .timer_interrupt, ACK_DEC)
+
+/* Fixed Interval Timer Interrupt */
+        MASKABLE_EXCEPTION(0x980, fixed_interval, .unknown_exception, ACK_FIT)
+
+/* Watchdog Timer Interrupt */
+        START_EXCEPTION(watchdog);
+        CRIT_EXCEPTION_PROLOG(0x9f0, PROLOG_ADDITION_NONE)
+//      EXCEPTION_COMMON(0x9f0, PACA_EXCRIT, INTS_DISABLE_ALL)
+//      bl      special_reg_save_crit
+//      addi    r3,r1,STACK_FRAME_OVERHEAD
+//      bl      .unknown_exception
+//      b       ret_from_crit_except
+        b       .
+
+/* System Call Interrupt */
+        START_EXCEPTION(system_call)
+        mr      r9,r13                  /* keep a copy of userland r13 */
+        mfspr   r11,SPRN_SRR0           /* get return address */
+        mfspr   r12,SPRN_SRR1           /* get previous MSR */
+        mfspr   r13,SPRN_SPRG_PACA      /* get our PACA */
+        b       system_call_common
+
+/* Auxillary Processor Unavailable Interrupt */
+        START_EXCEPTION(ap_unavailable);
+        NORMAL_EXCEPTION_PROLOG(0xf20, PROLOG_ADDITION_NONE)
+        EXCEPTION_COMMON(0xf20, PACA_EXGEN, INTS_KEEP)
+        addi    r3,r1,STACK_FRAME_OVERHEAD
+        bl      .save_nvgprs
+        INTS_RESTORE_HARD
+        bl      .unknown_exception
+        b       .ret_from_except
+
+/* Debug exception as a critical interrupt*/
+        START_EXCEPTION(debug_crit);
+        CRIT_EXCEPTION_PROLOG(0xd00, PROLOG_ADDITION_2REGS)
+
+        /*
+         * If there is a single step or branch-taken exception in an
+         * exception entry sequence, it was probably meant to apply to
+         * the code where the exception occurred (since exception entry
+         * doesn't turn off DE automatically).  We simulate the effect
+         * of turning off DE on entry to an exception handler by turning
+         * off DE in the CSRR1 value and clearing the debug status.
+         */
+
+        mfspr   r14,SPRN_DBSR           /* check single-step/branch taken */
+        andis.  r15,r14,DBSR_IC@h
+        beq+    1f
+
+        LOAD_REG_IMMEDIATE(r14,interrupt_base_book3e)
+        LOAD_REG_IMMEDIATE(r15,interrupt_end_book3e)
+        cmpld   cr0,r10,r14
+        cmpld   cr1,r10,r15
+        blt+    cr0,1f
+        bge+    cr1,1f
+
+        /* here it looks like we got an inappropriate debug exception. */
+        lis     r14,DBSR_IC@h           /* clear the IC event */
+        rlwinm  r11,r11,0,~MSR_DE       /* clear DE in the CSRR1 value */
+        mtspr   SPRN_DBSR,r14
+        mtspr   SPRN_CSRR1,r11
+        lwz     r10,PACA_EXCRIT+EX_CR(r13)      /* restore registers */
+        ld      r1,PACA_EXCRIT+EX_R1(r13)
+        ld      r14,PACA_EXCRIT+EX_R14(r13)
+        ld      r15,PACA_EXCRIT+EX_R15(r13)
+        mtcr    r10
+        ld      r10,PACA_EXCRIT+EX_R10(r13)     /* restore registers */
+        ld      r11,PACA_EXCRIT+EX_R11(r13)
+        mfspr   r13,SPRN_SPRG_CRIT_SCRATCH
+        rfci
+
+        /* Normal debug exception */
+        /* XXX We only handle coming from userspace for now since we can't
+         *     quite save properly an interrupted kernel state yet
+         */
+1:      andi.   r14,r11,MSR_PR;         /* check for userspace again */
+        beq     kernel_dbg_exc;         /* if from kernel mode */
+
+        /* Now we mash up things to make it look like we are coming on a
+         * normal exception
+         */
+        mfspr   r15,SPRN_SPRG_CRIT_SCRATCH
+        mtspr   SPRN_SPRG_GEN_SCRATCH,r15
+        mfspr   r14,SPRN_DBSR
+        EXCEPTION_COMMON(0xd00, PACA_EXCRIT, INTS_DISABLE_ALL)
+        std     r14,_DSISR(r1)
+        addi    r3,r1,STACK_FRAME_OVERHEAD
+        mr      r4,r14
+        ld      r14,PACA_EXCRIT+EX_R14(r13)
+        ld      r15,PACA_EXCRIT+EX_R15(r13)
+        bl      .save_nvgprs
+        bl      .DebugException
+        b       .ret_from_except
+
+kernel_dbg_exc:
+        b       .       /* NYI */
+
+
+/*
+ * An interrupt came in while soft-disabled; clear EE in SRR1,
+ * clear paca->hard_enabled and return.
+ */
+masked_interrupt_book3e:
+        mtcr    r10
+        stb     r11,PACAHARDIRQEN(r13)
+        mfspr   r10,SPRN_SRR1
+        rldicl  r11,r10,48,1            /* clear MSR_EE */
+        rotldi  r10,r11,16
+        mtspr   SPRN_SRR1,r10
+        ld      r10,PACA_EXGEN+EX_R10(r13);     /* restore registers */
+        ld      r11,PACA_EXGEN+EX_R11(r13);
+        mfspr   r13,SPRN_SPRG_GEN_SCRATCH;
+        rfi
+        b       .
+
+/*
+ * This is called from 0x300 and 0x400 handlers after the prologs with
+ * r14 and r15 containing the fault address and error code, with the
+ * original values stashed away in the PACA
+ */
+storage_fault_common:
+        std     r14,_DAR(r1)
+        std     r15,_DSISR(r1)
+        addi    r3,r1,STACK_FRAME_OVERHEAD
+        mr      r4,r14
+        mr      r5,r15
+        ld      r14,PACA_EXGEN+EX_R14(r13)
+        ld      r15,PACA_EXGEN+EX_R15(r13)
+        INTS_RESTORE_HARD
+        bl      .do_page_fault
+        cmpdi   r3,0
+        bne-    1f
+        b       .ret_from_except_lite
+1:      bl      .save_nvgprs
+        mr      r5,r3
+        addi    r3,r1,STACK_FRAME_OVERHEAD
+        ld      r4,_DAR(r1)
+        bl      .bad_page_fault
+        b       .ret_from_except
+
+/*
+ * Alignment exception doesn't fit entirely in the 0x100 bytes so it
+ * continues here.
+ */
+alignment_more:
+        std     r14,_DAR(r1)
+        std     r15,_DSISR(r1)
+        addi    r3,r1,STACK_FRAME_OVERHEAD
+        ld      r14,PACA_EXGEN+EX_R14(r13)
+        ld      r15,PACA_EXGEN+EX_R15(r13)
+        bl      .save_nvgprs
+        INTS_RESTORE_HARD
+        bl      .alignment_exception
+        b       .ret_from_except
+
+/*
+ * We branch here from entry_64.S for the last stage of the exception
+ * return code path. MSR:EE is expected to be off at that point
+ */
+_GLOBAL(exception_return_book3e)
+        b       1f
+
+/* This is the return from load_up_fpu fast path which could do with
+ * less GPR restores in fact, but for now we have a single return path
+ */
+        .globl fast_exception_return
+fast_exception_return:
+        wrteei  0
+1:      mr      r0,r13
+        ld      r10,_MSR(r1)
+        REST_4GPRS(2, r1)
+        andi.   r6,r10,MSR_PR
+        REST_2GPRS(6, r1)
+        beq     1f
+        ACCOUNT_CPU_USER_EXIT(r10, r11)
+        ld      r0,GPR13(r1)
+
+1:      stdcx.  r0,0,r1         /* to clear the reservation */
+
+        ld      r8,_CCR(r1)
+        ld      r9,_LINK(r1)
+        ld      r10,_CTR(r1)
+        ld      r11,_XER(r1)
+        mtcr    r8
+        mtlr    r9
+        mtctr   r10
+        mtxer   r11
+        REST_2GPRS(8, r1)
+        ld      r10,GPR10(r1)
+        ld      r11,GPR11(r1)
+        ld      r12,GPR12(r1)
+        mtspr   SPRN_SPRG_GEN_SCRATCH,r0
+
+        std     r10,PACA_EXGEN+EX_R10(r13);
+        std     r11,PACA_EXGEN+EX_R11(r13);
+        ld      r10,_NIP(r1)
+        ld      r11,_MSR(r1)
+        ld      r0,GPR0(r1)
+        ld      r1,GPR1(r1)
+        mtspr   SPRN_SRR0,r10
+        mtspr   SPRN_SRR1,r11
+        ld      r10,PACA_EXGEN+EX_R10(r13)
+        ld      r11,PACA_EXGEN+EX_R11(r13)
+        mfspr   r13,SPRN_SPRG_GEN_SCRATCH
+        rfi
+
+/*
+ * Trampolines used when spotting a bad kernel stack pointer in
+ * the exception entry code.
+ *
+ * TODO: move some bits like SRR0 read to trampoline, pass PACA
+ * index around, etc... to handle crit & mcheck
+ */
+BAD_STACK_TRAMPOLINE(0x000)
+BAD_STACK_TRAMPOLINE(0x100)
+BAD_STACK_TRAMPOLINE(0x200)
+BAD_STACK_TRAMPOLINE(0x300)
+BAD_STACK_TRAMPOLINE(0x400)
+BAD_STACK_TRAMPOLINE(0x500)
+BAD_STACK_TRAMPOLINE(0x600)
+BAD_STACK_TRAMPOLINE(0x700)
+BAD_STACK_TRAMPOLINE(0x800)
+BAD_STACK_TRAMPOLINE(0x900)
+BAD_STACK_TRAMPOLINE(0x980)
+BAD_STACK_TRAMPOLINE(0x9f0)
+BAD_STACK_TRAMPOLINE(0xa00)
+BAD_STACK_TRAMPOLINE(0xb00)
+BAD_STACK_TRAMPOLINE(0xc00)
+BAD_STACK_TRAMPOLINE(0xd00)
+BAD_STACK_TRAMPOLINE(0xe00)
+BAD_STACK_TRAMPOLINE(0xf00)
+BAD_STACK_TRAMPOLINE(0xf20)
+
+        .globl  bad_stack_book3e
+bad_stack_book3e:
+        /* XXX: Needs to make SPRN_SPRG_GEN depend on exception type */
+        mfspr   r10,SPRN_SRR0;            /* read SRR0 before touching stack */
+        ld      r1,PACAEMERGSP(r13)
+        subi    r1,r1,64+INT_FRAME_SIZE
+        std     r10,_NIP(r1)
+        std     r11,_MSR(r1)
+        ld      r10,PACA_EXGEN+EX_R1(r13) /* FIXME for crit & mcheck */
+        lwz     r11,PACA_EXGEN+EX_CR(r13) /* FIXME for crit & mcheck */
+        std     r10,GPR1(r1)
+        std     r11,_CCR(r1)
+        mfspr   r10,SPRN_DEAR
+        mfspr   r11,SPRN_ESR
+        std     r10,_DAR(r1)
+        std     r11,_DSISR(r1)
+        std     r0,GPR0(r1);            /* save r0 in stackframe */         \
+        std     r2,GPR2(r1);            /* save r2 in stackframe */         \
+        SAVE_4GPRS(3, r1);              /* save r3 - r6 in stackframe */    \
+        SAVE_2GPRS(7, r1);              /* save r7, r8 in stackframe */     \
+        std     r9,GPR9(r1);            /* save r9 in stackframe */         \
+        ld      r3,PACA_EXGEN+EX_R10(r13);/* get back r10 */                \
+        ld      r4,PACA_EXGEN+EX_R11(r13);/* get back r11 */                \
+        mfspr   r5,SPRN_SPRG_GEN_SCRATCH;/* get back r13 XXX can be wrong */ \
+        std     r3,GPR10(r1);           /* save r10 to stackframe */        \
+        std     r4,GPR11(r1);           /* save r11 to stackframe */        \
+        std     r12,GPR12(r1);          /* save r12 in stackframe */        \
+        std     r5,GPR13(r1);           /* save it to stackframe */         \
+        mflr    r10
+        mfctr   r11
+        mfxer   r12
+        std     r10,_LINK(r1)
+        std     r11,_CTR(r1)
+        std     r12,_XER(r1)
+        SAVE_10GPRS(14,r1)
+        SAVE_8GPRS(24,r1)
+        lhz     r12,PACA_TRAP_SAVE(r13)
+        std     r12,_TRAP(r1)
+        addi    r11,r1,INT_FRAME_SIZE
+        std     r11,0(r1)
+        li      r12,0
+        std     r12,0(r11)
+        ld      r2,PACATOC(r13)
+1:      addi    r3,r1,STACK_FRAME_OVERHEAD
+        bl      .kernel_bad_stack
+        b       1b
+
+/*
+ * Setup the initial TLB for a core. This current implementation
+ * assume that whatever we are running off will not conflict with
+ * the new mapping at PAGE_OFFSET.
+ * We also make various assumptions about the processor we run on,
+ * this might have to be made more flexible based on the content
+ * of MMUCFG and friends.
+ */
+_GLOBAL(initial_tlb_book3e)
+
+        /* Setup MAS 0,1,2,3 and 7 for tlbwe of a 1G entry that maps the
+         * kernel linear mapping. We also set MAS8 once for all here though
+         * that will have to be made dependent on whether we are running under
+         * a hypervisor I suppose.
+         */
+        li      r3,MAS0_HES | MAS0_WQ_ALLWAYS
+        mtspr   SPRN_MAS0,r3
+        lis     r3,(MAS1_VALID | MAS1_IPROT)@h
+        ori     r3,r3,BOOK3E_PAGESZ_1GB << MAS1_TSIZE_SHIFT
+        mtspr   SPRN_MAS1,r3
+        LOAD_REG_IMMEDIATE(r3, PAGE_OFFSET | MAS2_M)
+        mtspr   SPRN_MAS2,r3
+        li      r3,MAS3_SR | MAS3_SW | MAS3_SX
+        mtspr   SPRN_MAS7_MAS3,r3
+        li      r3,0
+        mtspr   SPRN_MAS8,r3
+
+        /* Write the TLB entry */
+        tlbwe
+
+        /* Now we branch the new virtual address mapped by this entry */
+        LOAD_REG_IMMEDIATE(r3,1f)
+        mtctr   r3
+        bctr
+
+1:      /* We are now running at PAGE_OFFSET, clean the TLB of everything
+         * else (XXX we should scan for bolted crap from the firmware too)
+         */
+        PPC_TLBILX(0,0,0)
+        sync
+        isync
+
+        /* We translate LR and return */
+        mflr    r3
+        tovirt(r3,r3)
+        mtlr    r3
+        blr
+
+/*
+ * Main entry (boot CPU, thread 0)
+ *
+ * We enter here from head_64.S, possibly after the prom_init trampoline
+ * with r3 and r4 already saved to r31 and 30 respectively and in 64 bits
+ * mode. Anything else is as it was left by the bootloader
+ *
+ * Initial requirements of this port:
+ *
+ * - Kernel loaded at 0 physical
+ * - A good lump of memory mapped 0:0 by UTLB entry 0
+ * - MSR:IS & MSR:DS set to 0
+ *
+ * Note that some of the above requirements will be relaxed in the future
+ * as the kernel becomes smarter at dealing with different initial conditions
+ * but for now you have to be careful
+ */
+_GLOBAL(start_initialization_book3e)
+        mflr    r28
+
+        /* First, we need to setup some initial TLBs to map the kernel
+         * text, data and bss at PAGE_OFFSET. We don't have a real mode
+         * and always use AS 0, so we just set it up to match our link
+         * address and never use 0 based addresses.
+         */
+        bl      .initial_tlb_book3e
+
+        /* Init global core bits */
+        bl      .init_core_book3e
+
+        /* Init per-thread bits */
+        bl      .init_thread_book3e
+
+        /* Return to common init code */
+        tovirt(r28,r28)
+        mtlr    r28
+        blr
+
+
+/*
+ * Secondary core/processor entry
+ *
+ * This is entered for thread 0 of a secondary core, all other threads
+ * are expected to be stopped. It's similar to start_initialization_book3e
+ * except that it's generally entered from the holding loop in head_64.S
+ * after CPUs have been gathered by Open Firmware.
+ *
+ * We assume we are in 32 bits mode running with whatever TLB entry was
+ * set for us by the firmware or POR engine.
+ */
+_GLOBAL(book3e_secondary_core_init_tlb_set)
+        li      r4,1
+        b       .generic_secondary_smp_init
+
+_GLOBAL(book3e_secondary_core_init)
+        mflr    r28
+
+        /* Do we need to setup initial TLB entry ? */
+        cmplwi  r4,0
+        bne     2f
+
+        /* Setup TLB for this core */
+        bl      .initial_tlb_book3e
+
+        /* We can return from the above running at a different
+         * address, so recalculate r2 (TOC)
+         */
+        bl      .relative_toc
+
+        /* Init global core bits */
+2:      bl      .init_core_book3e
+
+        /* Init per-thread bits */
+3:      bl      .init_thread_book3e
+
+        /* Return to common init code at proper virtual address.
+         *
+         * Due to various previous assumptions, we know we entered this
+         * function at either the final PAGE_OFFSET mapping or using a
+         * 1:1 mapping at 0, so we don't bother doing a complicated check
+         * here, we just ensure the return address has the right top bits.
+         *
+         * Note that if we ever want to be smarter about where we can be
+         * started from, we have to be careful that by the time we reach
+         * the code below we may already be running at a different location
+         * than the one we were called from since initial_tlb_book3e can
+         * have moved us already.
+         */
+        cmpdi   cr0,r28,0
+        blt     1f
+        lis     r3,PAGE_OFFSET@highest
+        sldi    r3,r3,32
+        or      r28,r28,r3
+1:      mtlr    r28
+        blr
+
+_GLOBAL(book3e_secondary_thread_init)
+        mflr    r28
+        b       3b
+
+_STATIC(init_core_book3e)
+        /* Establish the interrupt vector base */
+        LOAD_REG_IMMEDIATE(r3, interrupt_base_book3e)
+        mtspr   SPRN_IVPR,r3
+        sync
+        blr
+
+_STATIC(init_thread_book3e)
+        lis     r3,(SPRN_EPCR_ICM | SPRN_EPCR_GICM)@h
+        mtspr   SPRN_EPCR,r3
+
+        /* Make sure interrupts are off */
+        wrteei  0
+
+        /* disable watchdog and FIT and enable DEC interrupts */
+        lis     r3,TCR_DIE@h
+        mtspr   SPRN_TCR,r3
+
+        blr
+
+
+
diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S
index 0552f01041ab..c38afdb45d7b 100644
--- a/arch/powerpc/kernel/head_64.S
+++ b/arch/powerpc/kernel/head_64.S
@@ -121,10 +121,11 @@ __run_at_load:
  */
         .globl  __secondary_hold
 __secondary_hold:
+#ifndef CONFIG_PPC_BOOK3E
         mfmsr   r24
         ori     r24,r24,MSR_RI
         mtmsrd  r24                     /* RI on */
-
+#endif
         /* Grab our physical cpu number */
         mr      r24,r3
 
@@ -143,6 +144,7 @@ __secondary_hold:
         ld      r4,0(r4)                /* deref function descriptor */
         mtctr   r4
         mr      r3,r24
+        li      r4,0
         bctr
 #else
         BUG_OPCODE
@@ -163,21 +165,49 @@ exception_marker:
 #include "exceptions-64s.S"
 #endif
 
+_GLOBAL(generic_secondary_thread_init)
+        mr      r24,r3
+
+        /* turn on 64-bit mode */
+        bl      .enable_64b_mode
+
+        /* get a valid TOC pointer, wherever we're mapped at */
+        bl      .relative_toc
+
+#ifdef CONFIG_PPC_BOOK3E
+        /* Book3E initialization */
+        mr      r3,r24
+        bl      .book3e_secondary_thread_init
+#endif
+        b       generic_secondary_common_init
 
 /*
  * On pSeries and most other platforms, secondary processors spin
  * in the following code.
  * At entry, r3 = this processor's number (physical cpu id)
+ *
+ * On Book3E, r4 = 1 to indicate that the initial TLB entry for
+ * this core already exists (setup via some other mechanism such
+ * as SCOM before entry).
  */
 _GLOBAL(generic_secondary_smp_init)
         mr      r24,r3
-        
+        mr      r25,r4
+
         /* turn on 64-bit mode */
         bl      .enable_64b_mode
 
-        /* get the TOC pointer (real address) */
+        /* get a valid TOC pointer, wherever we're mapped at */
         bl      .relative_toc
 
+#ifdef CONFIG_PPC_BOOK3E
+        /* Book3E initialization */
+        mr      r3,r24
+        mr      r4,r25
+        bl      .book3e_secondary_core_init
+#endif
+
+generic_secondary_common_init:
         /* Set up a paca value for this processor. Since we have the
          * physical cpu id in r24, we need to search the pacas to find
          * which logical id maps to our physical one.
@@ -196,6 +226,11 @@ _GLOBAL(generic_secondary_smp_init)
         b       .kexec_wait             /* next kernel might do better   */
 
 2:      mtspr   SPRN_SPRG_PACA,r13      /* Save vaddr of paca in an SPRG */
+#ifdef CONFIG_PPC_BOOK3E
+        addi    r12,r13,PACA_EXTLB      /* and TLB exc frame in another  */
+        mtspr   SPRN_SPRG_TLB_EXFRAME,r12
+#endif
+
         /* From now on, r24 is expected to be logical cpuid */
         mr      r24,r5
 3:      HMT_LOW
@@ -231,6 +266,7 @@ _GLOBAL(generic_secondary_smp_init)
  * Turn the MMU off.
  * Assumes we're mapped EA == RA if the MMU is on.
  */
+#ifdef CONFIG_PPC_BOOK3S
 _STATIC(__mmu_off)
         mfmsr   r3
         andi.   r0,r3,MSR_IR|MSR_DR
@@ -242,6 +278,7 @@ _STATIC(__mmu_off)
         sync
         rfid
         b       .       /* prevent speculative execution */
+#endif
 
 
 /*
@@ -279,6 +316,10 @@ _GLOBAL(__start_initialization_multiplatform)
         mr      r31,r3
         mr      r30,r4
 
+#ifdef CONFIG_PPC_BOOK3E
+        bl      .start_initialization_book3e
+        b       .__after_prom_start
+#else
         /* Setup some critical 970 SPRs before switching MMU off */
         mfspr   r0,SPRN_PVR
         srwi    r0,r0,16
@@ -296,6 +337,7 @@ _GLOBAL(__start_initialization_multiplatform)
         /* Switch off MMU if not already off */
         bl      .__mmu_off
         b       .__after_prom_start
+#endif /* CONFIG_PPC_BOOK3E */
 
 _INIT_STATIC(__boot_from_prom)
 #ifdef CONFIG_PPC_OF_BOOT_TRAMPOLINE
@@ -358,10 +400,16 @@ _STATIC(__after_prom_start)
  * Note: This process overwrites the OF exception vectors.
  */
         li      r3,0                    /* target addr */
+#ifdef CONFIG_PPC_BOOK3E
+        tovirt(r3,r3)                   /* on booke, we already run at PAGE_OFFSET */
+#endif
         mr.     r4,r26                  /* In some cases the loader may  */
         beq     9f                      /* have already put us at zero */
         li      r6,0x100                /* Start offset, the first 0x100 */
                                         /* bytes were copied earlier.    */
+#ifdef CONFIG_PPC_BOOK3E
+        tovirt(r6,r6)                   /* on booke, we already run at PAGE_OFFSET */
+#endif
 
 #ifdef CONFIG_CRASH_DUMP
 /*
@@ -507,6 +555,9 @@ _GLOBAL(pmac_secondary_start)
  *   r13       = paca virtual address
  *   SPRG_PACA = paca virtual address
  */
+        .section ".text";
+        .align 2 ;
+
         .globl  __secondary_start
 __secondary_start:
         /* Set thread priority to MEDIUM */
@@ -543,7 +594,7 @@ END_FW_FTR_SECTION_IFCLR(FW_FEATURE_ISERIES)
 
         mtspr   SPRN_SRR0,r3
         mtspr   SPRN_SRR1,r4
-        rfid
+        RFI
         b       .       /* prevent speculative execution */
 
 /* 
@@ -564,11 +615,16 @@ _GLOBAL(start_secondary_prolog)
  */
 _GLOBAL(enable_64b_mode)
         mfmsr   r11                     /* grab the current MSR */
+#ifdef CONFIG_PPC_BOOK3E
+        oris    r11,r11,0x8000          /* CM bit set, we'll set ICM later */
+        mtmsr   r11
+#else /* CONFIG_PPC_BOOK3E */
         li      r12,(MSR_SF | MSR_ISF)@highest
         sldi    r12,r12,48
         or      r11,r11,r12
         mtmsrd  r11
         isync
+#endif
         blr
 
 /*
@@ -612,9 +668,11 @@ _INIT_STATIC(start_here_multiplatform)
         bdnz    3b
 4:
 
+#ifndef CONFIG_PPC_BOOK3E
         mfmsr   r6
         ori     r6,r6,MSR_RI
         mtmsrd  r6                      /* RI on */
+#endif
 
 #ifdef CONFIG_RELOCATABLE
         /* Save the physical address we're running at in kernstart_addr */
@@ -647,7 +705,7 @@ _INIT_STATIC(start_here_multiplatform)
         ld      r4,PACAKMSR(r13)
         mtspr   SPRN_SRR0,r3
         mtspr   SPRN_SRR1,r4
-        rfid
+        RFI
         b       .       /* prevent speculative execution */
         
         /* This is where all platforms converge execution */
diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c
index 65aced7b833a..87df51720641 100644
--- a/arch/powerpc/kernel/setup_64.c
+++ b/arch/powerpc/kernel/setup_64.c
@@ -454,6 +454,24 @@ static void __init irqstack_early_init(void)
 #define irqstack_early_init()
 #endif
 
+#ifdef CONFIG_PPC_BOOK3E
+static void __init exc_lvl_early_init(void)
+{
+        unsigned int i;
+
+        for_each_possible_cpu(i) {
+                critirq_ctx[i] = (struct thread_info *)
+                        __va(lmb_alloc(THREAD_SIZE, THREAD_SIZE));
+                dbgirq_ctx[i] = (struct thread_info *)
+                        __va(lmb_alloc(THREAD_SIZE, THREAD_SIZE));
+                mcheckirq_ctx[i] = (struct thread_info *)
+                        __va(lmb_alloc(THREAD_SIZE, THREAD_SIZE));
+        }
+}
+#else
+#define exc_lvl_early_init()
+#endif
+
 /*
  * Stack space used when we detect a bad kernel stack pointer, and
  * early in SMP boots before relocation is enabled.
@@ -513,6 +531,7 @@ void __init setup_arch(char **cmdline_p)
         init_mm.brk = klimit;
         
         irqstack_early_init();
+        exc_lvl_early_init();
         emergency_stack_init();
 
 #ifdef CONFIG_PPC_STD_MMU_64