Merge branch 'master'

68 files changed, 4981 insertions, 1173 deletions

diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 1493c7896fe..ed31062029f 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -599,6 +599,10 @@ config HAVE_ARCH_EARLY_PFN_TO_NID
 	def_bool y
 	depends on NEED_MULTIPLE_NODES
 
+config ARCH_MEMORY_PROBE
+	def_bool y
+	depends on MEMORY_HOTPLUG
+
 # Some NUMA nodes have memory ranges that span
 # other nodes.  Even though a pfn is valid and
 # between a node's start and end pfns, it may not
diff --git a/arch/powerpc/kernel/Makefile b/arch/powerpc/kernel/Makefile
index b3ae2993efb..c04bbd32059 100644
--- a/arch/powerpc/kernel/Makefile
+++ b/arch/powerpc/kernel/Makefile
@@ -4,6 +4,7 @@
 
 ifeq ($(CONFIG_PPC64),y)
 EXTRA_CFLAGS	+= -mno-minimal-toc
+CFLAGS_ioctl32.o += -Ifs/
 endif
 ifeq ($(CONFIG_PPC32),y)
 CFLAGS_prom_init.o      += -fPIC
@@ -11,15 +12,21 @@ CFLAGS_btext.o		+= -fPIC
 endif
 
 obj-y				:= semaphore.o cputable.o ptrace.o syscalls.o \
-				   signal_32.o pmc.o
+				   irq.o signal_32.o pmc.o
 obj-$(CONFIG_PPC64)		+= setup_64.o binfmt_elf32.o sys_ppc32.o \
-				   signal_64.o ptrace32.o systbl.o
+				   signal_64.o ptrace32.o systbl.o \
+				   paca.o ioctl32.o cpu_setup_power4.o \
+				   firmware.o sysfs.o
 obj-$(CONFIG_ALTIVEC)		+= vecemu.o vector.o
 obj-$(CONFIG_POWER4)		+= idle_power4.o
 obj-$(CONFIG_PPC_OF)		+= of_device.o
-obj-$(CONFIG_PPC_RTAS)		+= rtas.o
+procfs-$(CONFIG_PPC64)		:= proc_ppc64.o
+obj-$(CONFIG_PROC_FS)		+= $(procfs-y)
+rtaspci-$(CONFIG_PPC64)		:= rtas_pci.o
+obj-$(CONFIG_PPC_RTAS)		+= rtas.o $(rtaspci-y)
 obj-$(CONFIG_RTAS_FLASH)	+= rtas_flash.o
 obj-$(CONFIG_RTAS_PROC)		+= rtas-proc.o
+obj-$(CONFIG_LPARCFG)		+= lparcfg.o
 obj-$(CONFIG_IBMVIO)		+= vio.o
 obj-$(CONFIG_GENERIC_TBSYNC)	+= smp-tbsync.o
 
diff --git a/arch/powerpc/kernel/asm-offsets.c b/arch/powerpc/kernel/asm-offsets.c
index b7575725199..8793102711a 100644
--- a/arch/powerpc/kernel/asm-offsets.c
+++ b/arch/powerpc/kernel/asm-offsets.c
@@ -106,7 +106,6 @@ int main(void)
 	DEFINE(ICACHEL1LINESIZE, offsetof(struct ppc64_caches, iline_size));
 	DEFINE(ICACHEL1LOGLINESIZE, offsetof(struct ppc64_caches, log_iline_size));
 	DEFINE(ICACHEL1LINESPERPAGE, offsetof(struct ppc64_caches, ilines_per_page));
-	DEFINE(PLATFORM, offsetof(struct systemcfg, platform));
 	DEFINE(PLATFORM_LPAR, PLATFORM_LPAR);
 
 	/* paca */
diff --git a/arch/powerpc/kernel/cpu_setup_power4.S b/arch/powerpc/kernel/cpu_setup_power4.S
new file mode 100644
index 00000000000..cca942fe611
--- /dev/null
+++ b/arch/powerpc/kernel/cpu_setup_power4.S
@@ -0,0 +1,233 @@
+/*
+ * This file contains low level CPU setup functions.
+ *    Copyright (C) 2003 Benjamin Herrenschmidt (benh@kernel.crashing.org)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ */
+
+#include <linux/config.h>
+#include <asm/processor.h>
+#include <asm/page.h>
+#include <asm/cputable.h>
+#include <asm/ppc_asm.h>
+#include <asm/asm-offsets.h>
+#include <asm/cache.h>
+
+_GLOBAL(__970_cpu_preinit)
+	/*
+	 * Do nothing if not running in HV mode
+	 */
+	mfmsr	r0
+	rldicl.	r0,r0,4,63
+	beqlr
+
+	/*
+	 * Deal only with PPC970 and PPC970FX.
+	 */
+	mfspr	r0,SPRN_PVR
+	srwi	r0,r0,16
+	cmpwi	r0,0x39
+	beq	1f
+	cmpwi	r0,0x3c
+	beq	1f
+	cmpwi	r0,0x44
+	bnelr
+1:
+
+	/* Make sure HID4:rm_ci is off before MMU is turned off, that large
+	 * pages are enabled with HID4:61 and clear HID5:DCBZ_size and
+	 * HID5:DCBZ32_ill
+	 */
+	li	r0,0
+	mfspr	r3,SPRN_HID4
+	rldimi	r3,r0,40,23	/* clear bit 23 (rm_ci) */
+	rldimi	r3,r0,2,61	/* clear bit 61 (lg_pg_en) */
+	sync
+	mtspr	SPRN_HID4,r3
+	isync
+	sync
+	mfspr	r3,SPRN_HID5
+	rldimi	r3,r0,6,56	/* clear bits 56 & 57 (DCBZ*) */
+	sync
+	mtspr	SPRN_HID5,r3
+	isync
+	sync
+
+	/* Setup some basic HID1 features */
+	mfspr	r0,SPRN_HID1
+	li	r3,0x1200		/* enable i-fetch cacheability */
+	sldi	r3,r3,44		/* and prefetch */
+	or	r0,r0,r3
+	mtspr	SPRN_HID1,r0
+	mtspr	SPRN_HID1,r0
+	isync
+
+	/* Clear HIOR */
+	li	r0,0
+	sync
+	mtspr	SPRN_HIOR,0		/* Clear interrupt prefix */
+	isync
+	blr
+
+_GLOBAL(__setup_cpu_power4)
+	blr
+
+_GLOBAL(__setup_cpu_be)
+        /* Set large page sizes LP=0: 16MB, LP=1: 64KB */
+        addi    r3, 0,  0
+        ori     r3, r3, HID6_LB
+        sldi    r3, r3, 32
+        nor     r3, r3, r3
+        mfspr   r4, SPRN_HID6
+        and     r4, r4, r3
+        addi    r3, 0, 0x02000
+        sldi    r3, r3, 32
+        or      r4, r4, r3
+        mtspr   SPRN_HID6, r4
+	blr
+
+_GLOBAL(__setup_cpu_ppc970)
+	mfspr	r0,SPRN_HID0
+	li	r11,5			/* clear DOZE and SLEEP */
+	rldimi	r0,r11,52,8		/* set NAP and DPM */
+	mtspr	SPRN_HID0,r0
+	mfspr	r0,SPRN_HID0
+	mfspr	r0,SPRN_HID0
+	mfspr	r0,SPRN_HID0
+	mfspr	r0,SPRN_HID0
+	mfspr	r0,SPRN_HID0
+	mfspr	r0,SPRN_HID0
+	sync
+	isync
+	blr
+
+/* Definitions for the table use to save CPU states */
+#define CS_HID0		0
+#define CS_HID1		8
+#define	CS_HID4		16
+#define CS_HID5		24
+#define CS_SIZE		32
+
+	.data
+	.balign	L1_CACHE_BYTES,0
+cpu_state_storage:
+	.space	CS_SIZE
+	.balign	L1_CACHE_BYTES,0
+	.text
+
+/* Called in normal context to backup CPU 0 state. This
+ * does not include cache settings. This function is also
+ * called for machine sleep. This does not include the MMU
+ * setup, BATs, etc... but rather the "special" registers
+ * like HID0, HID1, HID4, etc...
+ */
+_GLOBAL(__save_cpu_setup)
+	/* Some CR fields are volatile, we back it up all */
+	mfcr	r7
+
+	/* Get storage ptr */
+	LOADADDR(r5,cpu_state_storage)
+
+	/* We only deal with 970 for now */
+	mfspr	r0,SPRN_PVR
+	srwi	r0,r0,16
+	cmpwi	r0,0x39
+	beq	1f
+	cmpwi	r0,0x3c
+	beq	1f
+	cmpwi	r0,0x44
+	bne	2f
+
+1:	/* Save HID0,1,4 and 5 */
+	mfspr	r3,SPRN_HID0
+	std	r3,CS_HID0(r5)
+	mfspr	r3,SPRN_HID1
+	std	r3,CS_HID1(r5)
+	mfspr	r3,SPRN_HID4
+	std	r3,CS_HID4(r5)
+	mfspr	r3,SPRN_HID5
+	std	r3,CS_HID5(r5)
+
+2:
+	mtcr	r7
+	blr
+
+/* Called with no MMU context (typically MSR:IR/DR off) to
+ * restore CPU state as backed up by the previous
+ * function. This does not include cache setting
+ */
+_GLOBAL(__restore_cpu_setup)
+	/* Get storage ptr (FIXME when using anton reloc as we
+	 * are running with translation disabled here
+	 */
+	LOADADDR(r5,cpu_state_storage)
+
+	/* We only deal with 970 for now */
+	mfspr	r0,SPRN_PVR
+	srwi	r0,r0,16
+	cmpwi	r0,0x39
+	beq	1f
+	cmpwi	r0,0x3c
+	beq	1f
+	cmpwi	r0,0x44
+	bnelr
+
+1:	/* Before accessing memory, we make sure rm_ci is clear */
+	li	r0,0
+	mfspr	r3,SPRN_HID4
+	rldimi	r3,r0,40,23	/* clear bit 23 (rm_ci) */
+	sync
+	mtspr	SPRN_HID4,r3
+	isync
+	sync
+
+	/* Clear interrupt prefix */
+	li	r0,0
+	sync
+	mtspr	SPRN_HIOR,0
+	isync
+
+	/* Restore HID0 */
+	ld	r3,CS_HID0(r5)
+	sync
+	isync
+	mtspr	SPRN_HID0,r3
+	mfspr	r3,SPRN_HID0
+	mfspr	r3,SPRN_HID0
+	mfspr	r3,SPRN_HID0
+	mfspr	r3,SPRN_HID0
+	mfspr	r3,SPRN_HID0
+	mfspr	r3,SPRN_HID0
+	sync
+	isync
+
+	/* Restore HID1 */
+	ld	r3,CS_HID1(r5)
+	sync
+	isync
+	mtspr	SPRN_HID1,r3
+	mtspr	SPRN_HID1,r3
+	sync
+	isync
+
+	/* Restore HID4 */
+	ld	r3,CS_HID4(r5)
+	sync
+	isync
+	mtspr	SPRN_HID4,r3
+	sync
+	isync
+
+	/* Restore HID5 */
+	ld	r3,CS_HID5(r5)
+	sync
+	isync
+	mtspr	SPRN_HID5,r3
+	sync
+	isync
+	blr
+
diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c
index cc4e9eb1c13..1d85cedbbb7 100644
--- a/arch/powerpc/kernel/cputable.c
+++ b/arch/powerpc/kernel/cputable.c
@@ -52,6 +52,9 @@ extern void __setup_cpu_ppc970(unsigned long offset, struct cpu_spec* spec);
 #define COMMON_USER		(PPC_FEATURE_32 | PPC_FEATURE_HAS_FPU | \
 				 PPC_FEATURE_HAS_MMU)
 #define COMMON_USER_PPC64	(COMMON_USER | PPC_FEATURE_64)
+#define COMMON_USER_POWER4	(COMMON_USER_PPC64 | PPC_FEATURE_POWER4)
+#define COMMON_USER_POWER5	(COMMON_USER_PPC64 | PPC_FEATURE_POWER5)
+#define COMMON_USER_POWER5_PLUS	(COMMON_USER_PPC64 | PPC_FEATURE_POWER5_PLUS)
 
 
 /* We only set the spe features if the kernel was compiled with
@@ -160,7 +163,7 @@ struct cpu_spec	cpu_specs[] = {
 		.pvr_value		= 0x00350000,
 		.cpu_name		= "POWER4 (gp)",
 		.cpu_features		= CPU_FTRS_POWER4,
-		.cpu_user_features	= COMMON_USER_PPC64,
+		.cpu_user_features	= COMMON_USER_POWER4,
 		.icache_bsize		= 128,
 		.dcache_bsize		= 128,
 		.num_pmcs		= 8,
@@ -175,7 +178,7 @@ struct cpu_spec	cpu_specs[] = {
 		.pvr_value		= 0x00380000,
 		.cpu_name		= "POWER4+ (gq)",
 		.cpu_features		= CPU_FTRS_POWER4,
-		.cpu_user_features	= COMMON_USER_PPC64,
+		.cpu_user_features	= COMMON_USER_POWER4,
 		.icache_bsize		= 128,
 		.dcache_bsize		= 128,
 		.num_pmcs		= 8,
@@ -190,7 +193,7 @@ struct cpu_spec	cpu_specs[] = {
 		.pvr_value		= 0x00390000,
 		.cpu_name		= "PPC970",
 		.cpu_features		= CPU_FTRS_PPC970,
-		.cpu_user_features	= COMMON_USER_PPC64 |
+		.cpu_user_features	= COMMON_USER_POWER4 |
 			PPC_FEATURE_HAS_ALTIVEC_COMP,
 		.icache_bsize		= 128,
 		.dcache_bsize		= 128,
@@ -212,7 +215,7 @@ struct cpu_spec	cpu_specs[] = {
 #else
 		.cpu_features		= CPU_FTRS_PPC970,
 #endif
-		.cpu_user_features	= COMMON_USER_PPC64 |
+		.cpu_user_features	= COMMON_USER_POWER4 |
 			PPC_FEATURE_HAS_ALTIVEC_COMP,
 		.icache_bsize		= 128,
 		.dcache_bsize		= 128,
@@ -230,7 +233,7 @@ struct cpu_spec	cpu_specs[] = {
 		.pvr_value		= 0x00440000,
 		.cpu_name		= "PPC970MP",
 		.cpu_features		= CPU_FTRS_PPC970,
-		.cpu_user_features	= COMMON_USER_PPC64 |
+		.cpu_user_features	= COMMON_USER_POWER4 |
 			PPC_FEATURE_HAS_ALTIVEC_COMP,
 		.icache_bsize		= 128,
 		.dcache_bsize		= 128,
@@ -245,7 +248,7 @@ struct cpu_spec	cpu_specs[] = {
 		.pvr_value		= 0x003a0000,
 		.cpu_name		= "POWER5 (gr)",
 		.cpu_features		= CPU_FTRS_POWER5,
-		.cpu_user_features	= COMMON_USER_PPC64,
+		.cpu_user_features	= COMMON_USER_POWER5,
 		.icache_bsize		= 128,
 		.dcache_bsize		= 128,
 		.num_pmcs		= 6,
@@ -260,7 +263,7 @@ struct cpu_spec	cpu_specs[] = {
 		.pvr_value		= 0x003b0000,
 		.cpu_name		= "POWER5 (gs)",
 		.cpu_features		= CPU_FTRS_POWER5,
-		.cpu_user_features	= COMMON_USER_PPC64,
+		.cpu_user_features	= COMMON_USER_POWER5_PLUS,
 		.icache_bsize		= 128,
 		.dcache_bsize		= 128,
 		.num_pmcs		= 6,
@@ -276,7 +279,7 @@ struct cpu_spec	cpu_specs[] = {
 		.cpu_name		= "Cell Broadband Engine",
 		.cpu_features		= CPU_FTRS_CELL,
 		.cpu_user_features	= COMMON_USER_PPC64 |
-			PPC_FEATURE_HAS_ALTIVEC_COMP,
+			PPC_FEATURE_CELL | PPC_FEATURE_HAS_ALTIVEC_COMP,
 		.icache_bsize		= 128,
 		.dcache_bsize		= 128,
 		.cpu_setup		= __setup_cpu_be,
diff --git a/arch/powerpc/kernel/firmware.c b/arch/powerpc/kernel/firmware.c
new file mode 100644
index 00000000000..65eae752a52
--- /dev/null
+++ b/arch/powerpc/kernel/firmware.c
@@ -0,0 +1,45 @@
+/*
+ *  Extracted from cputable.c
+ *
+ *  Copyright (C) 2001 Ben. Herrenschmidt (benh@kernel.crashing.org)
+ *
+ *  Modifications for ppc64:
+ *      Copyright (C) 2003 Dave Engebretsen <engebret@us.ibm.com>
+ *  Copyright (C) 2005 Stephen Rothwell, IBM Corporation
+ *
+ *  This program is free software; you can redistribute it and/or
+ *  modify it under the terms of the GNU General Public License
+ *  as published by the Free Software Foundation; either version
+ *  2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/config.h>
+
+#include <asm/firmware.h>
+
+unsigned long ppc64_firmware_features;
+
+#ifdef CONFIG_PPC_PSERIES
+firmware_feature_t firmware_features_table[FIRMWARE_MAX_FEATURES] = {
+	{FW_FEATURE_PFT,		"hcall-pft"},
+	{FW_FEATURE_TCE,		"hcall-tce"},
+	{FW_FEATURE_SPRG0,		"hcall-sprg0"},
+	{FW_FEATURE_DABR,		"hcall-dabr"},
+	{FW_FEATURE_COPY,		"hcall-copy"},
+	{FW_FEATURE_ASR,		"hcall-asr"},
+	{FW_FEATURE_DEBUG,		"hcall-debug"},
+	{FW_FEATURE_PERF,		"hcall-perf"},
+	{FW_FEATURE_DUMP,		"hcall-dump"},
+	{FW_FEATURE_INTERRUPT,		"hcall-interrupt"},
+	{FW_FEATURE_MIGRATE,		"hcall-migrate"},
+	{FW_FEATURE_PERFMON,		"hcall-perfmon"},
+	{FW_FEATURE_CRQ,		"hcall-crq"},
+	{FW_FEATURE_VIO,		"hcall-vio"},
+	{FW_FEATURE_RDMA,		"hcall-rdma"},
+	{FW_FEATURE_LLAN,		"hcall-lLAN"},
+	{FW_FEATURE_BULK,		"hcall-bulk"},
+	{FW_FEATURE_XDABR,		"hcall-xdabr"},
+	{FW_FEATURE_MULTITCE,		"hcall-multi-tce"},
+	{FW_FEATURE_SPLPAR,		"hcall-splpar"},
+};
+#endif
diff --git a/arch/powerpc/kernel/fpu.S b/arch/powerpc/kernel/fpu.S
index 4d6001fa1cf..b780b42c95f 100644
--- a/arch/powerpc/kernel/fpu.S
+++ b/arch/powerpc/kernel/fpu.S
@@ -41,20 +41,20 @@ _GLOBAL(load_up_fpu)
 #ifndef CONFIG_SMP
 	LOADBASE(r3, last_task_used_math)
 	toreal(r3)
-	LDL	r4,OFF(last_task_used_math)(r3)
-	CMPI	0,r4,0
+	PPC_LL	r4,OFF(last_task_used_math)(r3)
+	PPC_LCMPI	0,r4,0
 	beq	1f
 	toreal(r4)
 	addi	r4,r4,THREAD		/* want last_task_used_math->thread */
 	SAVE_32FPRS(0, r4)
 	mffs	fr0
 	stfd	fr0,THREAD_FPSCR(r4)
-	LDL	r5,PT_REGS(r4)
+	PPC_LL	r5,PT_REGS(r4)
 	toreal(r5)
-	LDL	r4,_MSR-STACK_FRAME_OVERHEAD(r5)
+	PPC_LL	r4,_MSR-STACK_FRAME_OVERHEAD(r5)
 	li	r10,MSR_FP|MSR_FE0|MSR_FE1
 	andc	r4,r4,r10		/* disable FP for previous task */
-	STL	r4,_MSR-STACK_FRAME_OVERHEAD(r5)
+	PPC_STL	r4,_MSR-STACK_FRAME_OVERHEAD(r5)
 1:
 #endif /* CONFIG_SMP */
 	/* enable use of FP after return */
@@ -77,7 +77,7 @@ _GLOBAL(load_up_fpu)
 #ifndef CONFIG_SMP
 	subi	r4,r5,THREAD
 	fromreal(r4)
-	STL	r4,OFF(last_task_used_math)(r3)
+	PPC_STL	r4,OFF(last_task_used_math)(r3)
 #endif /* CONFIG_SMP */
 	/* restore registers and return */
 	/* we haven't used ctr or xer or lr */
@@ -97,24 +97,24 @@ _GLOBAL(giveup_fpu)
 	MTMSRD(r5)			/* enable use of fpu now */
 	SYNC_601
 	isync
-	CMPI	0,r3,0
+	PPC_LCMPI	0,r3,0
 	beqlr-				/* if no previous owner, done */
 	addi	r3,r3,THREAD	        /* want THREAD of task */
-	LDL	r5,PT_REGS(r3)
-	CMPI	0,r5,0
+	PPC_LL	r5,PT_REGS(r3)
+	PPC_LCMPI	0,r5,0
 	SAVE_32FPRS(0, r3)
 	mffs	fr0
 	stfd	fr0,THREAD_FPSCR(r3)
 	beq	1f
-	LDL	r4,_MSR-STACK_FRAME_OVERHEAD(r5)
+	PPC_LL	r4,_MSR-STACK_FRAME_OVERHEAD(r5)
 	li	r3,MSR_FP|MSR_FE0|MSR_FE1
 	andc	r4,r4,r3		/* disable FP for previous task */
-	STL	r4,_MSR-STACK_FRAME_OVERHEAD(r5)
+	PPC_STL	r4,_MSR-STACK_FRAME_OVERHEAD(r5)
 1:
 #ifndef CONFIG_SMP
 	li	r5,0
 	LOADBASE(r4,last_task_used_math)
-	STL	r5,OFF(last_task_used_math)(r4)
+	PPC_STL	r5,OFF(last_task_used_math)(r4)
 #endif /* CONFIG_SMP */
 	blr
 
diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S
index 16ab40daa73..8a8bf79ef04 100644
--- a/arch/powerpc/kernel/head_64.S
+++ b/arch/powerpc/kernel/head_64.S
@@ -28,7 +28,6 @@
 #include <asm/reg.h>
 #include <asm/page.h>
 #include <asm/mmu.h>
-#include <asm/systemcfg.h>
 #include <asm/ppc_asm.h>
 #include <asm/asm-offsets.h>
 #include <asm/bug.h>
@@ -1697,25 +1696,14 @@ _GLOBAL(pmac_secondary_start)
  *   SPRG3 = paca virtual address
  */
 _GLOBAL(__secondary_start)
+	/* Set thread priority to MEDIUM */
+	HMT_MEDIUM
 
-	HMT_MEDIUM			/* Set thread priority to MEDIUM */
-
+	/* Load TOC */
 	ld	r2,PACATOC(r13)
-	li	r6,0
-	stb	r6,PACAPROCENABLED(r13)
-
-#ifndef CONFIG_PPC_ISERIES
-	/* Initialize the page table pointer register. */
-	LOADADDR(r6,_SDR1)
-	ld	r6,0(r6)		/* get the value of _SDR1	 */
-	mtspr	SPRN_SDR1,r6			/* set the htab location	 */
-#endif
-	/* Initialize the first segment table (or SLB) entry		 */
-	ld	r3,PACASTABVIRT(r13)	/* get addr of segment table	 */
-BEGIN_FTR_SECTION
-	bl	.stab_initialize
-END_FTR_SECTION_IFCLR(CPU_FTR_SLB)
-	bl	.slb_initialize
+
+	/* Do early setup for that CPU (stab, slb, hash table pointer) */
+	bl	.early_setup_secondary
 
 	/* Initialize the kernel stack.  Just a repeat for iSeries.	 */
 	LOADADDR(r3,current_set)
@@ -1724,37 +1712,7 @@ END_FTR_SECTION_IFCLR(CPU_FTR_SLB)
 	addi	r1,r1,THREAD_SIZE-STACK_FRAME_OVERHEAD
 	std	r1,PACAKSAVE(r13)
 
-	ld	r3,PACASTABREAL(r13)	/* get raddr of segment table	 */
-	ori	r4,r3,1			/* turn on valid bit		 */
-
-#ifdef CONFIG_PPC_ISERIES
-	li	r0,-1			/* hypervisor call */
-	li	r3,1
-	sldi	r3,r3,63		/* 0x8000000000000000 */
-	ori	r3,r3,4			/* 0x8000000000000004 */
-	sc				/* HvCall_setASR */
-#else
-	/* set the ASR */
-	ld	r3,systemcfg@got(r2)	/* r3 = ptr to systemcfg	 */
-	ld	r3,0(r3)
-	lwz	r3,PLATFORM(r3)		/* r3 = platform flags		 */
-	andi.	r3,r3,PLATFORM_LPAR	/* Test if bit 0 is set (LPAR bit) */
-	beq	98f			/* branch if result is 0  */
-	mfspr	r3,SPRN_PVR
-	srwi	r3,r3,16
-	cmpwi	r3,0x37			/* SStar  */
-	beq	97f
-	cmpwi	r3,0x36			/* IStar  */
-	beq	97f
-	cmpwi	r3,0x34			/* Pulsar */
-	bne	98f
-97:	li	r3,H_SET_ASR		/* hcall = H_SET_ASR */
-	HVSC				/* Invoking hcall */
-	b	99f
-98:					/* !(rpa hypervisor) || !(star)  */
-	mtasr	r4			/* set the stab location	 */
-99:
-#endif
+	/* Clear backchain so we get nice backtraces */
 	li	r7,0
 	mtlr	r7
 
@@ -1777,6 +1735,7 @@ _GLOBAL(start_secondary_prolog)
 	li	r3,0
 	std	r3,0(r1)		/* Zero the stack frame pointer	*/
 	bl	.start_secondary
+	b	.
 #endif
 
 /*
@@ -1896,40 +1855,6 @@ _STATIC(start_here_multiplatform)
 	mr	r3,r31
  	bl	.early_setup
 
-	/* set the ASR */
-	ld	r3,PACASTABREAL(r13)
-	ori	r4,r3,1			/* turn on valid bit		 */
-	ld	r3,systemcfg@got(r2)	/* r3 = ptr to systemcfg */
-	ld	r3,0(r3)
-	lwz	r3,PLATFORM(r3)		/* r3 = platform flags */
-	andi.	r3,r3,PLATFORM_LPAR	/* Test if bit 0 is set (LPAR bit) */
-	beq	98f			/* branch if result is 0  */
-	mfspr	r3,SPRN_PVR
-	srwi	r3,r3,16
-	cmpwi	r3,0x37			/* SStar */
-	beq	97f
-	cmpwi	r3,0x36			/* IStar  */
-	beq	97f
-	cmpwi	r3,0x34			/* Pulsar */
-	bne	98f
-97:	li	r3,H_SET_ASR		/* hcall = H_SET_ASR */
-	HVSC				/* Invoking hcall */
-	b	99f
-98:					/* !(rpa hypervisor) || !(star) */
-	mtasr	r4			/* set the stab location	*/
-99:
-	/* Set SDR1 (hash table pointer) */
-	ld	r3,systemcfg@got(r2)	/* r3 = ptr to systemcfg */
-	ld	r3,0(r3)
-	lwz	r3,PLATFORM(r3)		/* r3 = platform flags */
-	/* Test if bit 0 is set (LPAR bit) */
-	andi.	r3,r3,PLATFORM_LPAR
-	bne	98f			/* branch if result is !0  */
-	LOADADDR(r6,_SDR1)		/* Only if NOT LPAR */
-	add	r6,r6,r26
-	ld	r6,0(r6)		/* get the value of _SDR1 */
-	mtspr	SPRN_SDR1,r6			/* set the htab location  */
-98: 
 	LOADADDR(r3,.start_here_common)
 	SET_REG_TO_CONST(r4, MSR_KERNEL)
 	mtspr	SPRN_SRR0,r3
diff --git a/arch/powerpc/kernel/ioctl32.c b/arch/powerpc/kernel/ioctl32.c
new file mode 100644
index 00000000000..3fa6a93adbd
--- /dev/null
+++ b/arch/powerpc/kernel/ioctl32.c
@@ -0,0 +1,49 @@
+/*
+ * ioctl32.c: Conversion between 32bit and 64bit native ioctls.
+ *
+ * Based on sparc64 ioctl32.c by:
+ *
+ * Copyright (C) 1997-2000  Jakub Jelinek  (jakub@redhat.com)
+ * Copyright (C) 1998  Eddie C. Dost  (ecd@skynet.be)
+ *
+ * ppc64 changes:
+ *
+ * Copyright (C) 2000  Ken Aaker (kdaaker@rchland.vnet.ibm.com)
+ * Copyright (C) 2001  Anton Blanchard (antonb@au.ibm.com)
+ *
+ * These routines maintain argument size conversion between 32bit and 64bit
+ * ioctls.
+ *
+ * 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.
+ */
+
+#define INCLUDES
+#include "compat_ioctl.c"
+#include <linux/syscalls.h>
+
+#define CODE
+#include "compat_ioctl.c"
+
+#define HANDLE_IOCTL(cmd,handler) { cmd, (ioctl_trans_handler_t)handler, NULL },
+#define COMPATIBLE_IOCTL(cmd) HANDLE_IOCTL(cmd,sys_ioctl)
+
+#define IOCTL_TABLE_START \
+	struct ioctl_trans ioctl_start[] = {
+#define IOCTL_TABLE_END \
+	};
+
+IOCTL_TABLE_START
+#include <linux/compat_ioctl.h>
+#define DECLARES
+#include "compat_ioctl.c"
+
+/* Little p (/dev/rtc, /dev/envctrl, etc.) */
+COMPATIBLE_IOCTL(_IOR('p', 20, int[7])) /* RTCGET */
+COMPATIBLE_IOCTL(_IOW('p', 21, int[7])) /* RTCSET */
+
+IOCTL_TABLE_END
+
+int ioctl_table_size = ARRAY_SIZE(ioctl_start);
diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c
new file mode 100644
index 00000000000..4b7940693f3
--- /dev/null
+++ b/arch/powerpc/kernel/irq.c
@@ -0,0 +1,478 @@
+/*
+ *  arch/ppc/kernel/irq.c
+ *
+ *  Derived from arch/i386/kernel/irq.c
+ *    Copyright (C) 1992 Linus Torvalds
+ *  Adapted from arch/i386 by Gary Thomas
+ *    Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org)
+ *  Updated and modified by Cort Dougan <cort@fsmlabs.com>
+ *    Copyright (C) 1996-2001 Cort Dougan
+ *  Adapted for Power Macintosh by Paul Mackerras
+ *    Copyright (C) 1996 Paul Mackerras (paulus@cs.anu.edu.au)
+ *  Amiga/APUS changes by Jesper Skov (jskov@cygnus.co.uk).
+ *
+ * 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.
+ *
+ * This file contains the code used by various IRQ handling routines:
+ * asking for different IRQ's should be done through these routines
+ * instead of just grabbing them. Thus setups with different IRQ numbers
+ * shouldn't result in any weird surprises, and installing new handlers
+ * should be easier.
+ *
+ * The MPC8xx has an interrupt mask in the SIU.  If a bit is set, the
+ * interrupt is _enabled_.  As expected, IRQ0 is bit 0 in the 32-bit
+ * mask register (of which only 16 are defined), hence the weird shifting
+ * and complement of the cached_irq_mask.  I want to be able to stuff
+ * this right into the SIU SMASK register.
+ * Many of the prep/chrp functions are conditional compiled on CONFIG_8xx
+ * to reduce code space and undefined function references.
+ */
+
+#include <linux/errno.h>
+#include <linux/module.h>
+#include <linux/threads.h>
+#include <linux/kernel_stat.h>
+#include <linux/signal.h>
+#include <linux/sched.h>
+#include <linux/ptrace.h>
+#include <linux/ioport.h>
+#include <linux/interrupt.h>
+#include <linux/timex.h>
+#include <linux/config.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include <linux/irq.h>
+#include <linux/proc_fs.h>
+#include <linux/random.h>
+#include <linux/seq_file.h>
+#include <linux/cpumask.h>
+#include <linux/profile.h>
+#include <linux/bitops.h>
+#ifdef CONFIG_PPC64
+#include <linux/kallsyms.h>
+#endif
+
+#include <asm/uaccess.h>
+#include <asm/system.h>
+#include <asm/io.h>
+#include <asm/pgtable.h>
+#include <asm/irq.h>
+#include <asm/cache.h>
+#include <asm/prom.h>
+#include <asm/ptrace.h>
+#include <asm/machdep.h>
+#ifdef CONFIG_PPC64
+#include <asm/iseries/it_lp_queue.h>
+#include <asm/paca.h>
+#endif
+
+static int ppc_spurious_interrupts;
+
+#if defined(CONFIG_PPC_ISERIES) && defined(CONFIG_SMP)
+extern void iSeries_smp_message_recv(struct pt_regs *);
+#endif
+
+#ifdef CONFIG_PPC32
+#define NR_MASK_WORDS	((NR_IRQS + 31) / 32)
+
+unsigned long ppc_cached_irq_mask[NR_MASK_WORDS];
+atomic_t ppc_n_lost_interrupts;
+
+#ifdef CONFIG_TAU_INT
+extern int tau_initialized;
+extern int tau_interrupts(int);
+#endif
+
+#if defined(CONFIG_SMP) && !defined(CONFIG_PPC_MERGE)
+extern atomic_t ipi_recv;
+extern atomic_t ipi_sent;
+#endif
+#endif /* CONFIG_PPC32 */
+
+#ifdef CONFIG_PPC64
+EXPORT_SYMBOL(irq_desc);
+
+int distribute_irqs = 1;
+int __irq_offset_value;
+u64 ppc64_interrupt_controller;
+#endif /* CONFIG_PPC64 */
+
+int show_interrupts(struct seq_file *p, void *v)
+{
+	int i = *(loff_t *)v, j;
+	struct irqaction *action;
+	irq_desc_t *desc;
+	unsigned long flags;
+
+	if (i == 0) {
+		seq_puts(p, "           ");
+		for_each_online_cpu(j)
+			seq_printf(p, "CPU%d       ", j);
+		seq_putc(p, '\n');
+	}
+
+	if (i < NR_IRQS) {
+		desc = get_irq_desc(i);
+		spin_lock_irqsave(&desc->lock, flags);
+		action = desc->action;
+		if (!action || !action->handler)
+			goto skip;
+		seq_printf(p, "%3d: ", i);
+#ifdef CONFIG_SMP
+		for_each_online_cpu(j)
+			seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
+#else
+		seq_printf(p, "%10u ", kstat_irqs(i));
+#endif /* CONFIG_SMP */
+		if (desc->handler)
+			seq_printf(p, " %s ", desc->handler->typename);
+		else
+			seq_puts(p, "  None      ");
+		seq_printf(p, "%s", (desc->status & IRQ_LEVEL) ? "Level " : "Edge  ");
+		seq_printf(p, "    %s", action->name);
+		for (action = action->next; action; action = action->next)
+			seq_printf(p, ", %s", action->name);
+		seq_putc(p, '\n');
+skip:
+		spin_unlock_irqrestore(&desc->lock, flags);
+	} else if (i == NR_IRQS) {
+#ifdef CONFIG_PPC32
+#ifdef CONFIG_TAU_INT
+		if (tau_initialized){
+			seq_puts(p, "TAU: ");
+			for (j = 0; j < NR_CPUS; j++)
+				if (cpu_online(j))
+					seq_printf(p, "%10u ", tau_interrupts(j));
+			seq_puts(p, "  PowerPC             Thermal Assist (cpu temp)\n");
+		}
+#endif
+#if defined(CONFIG_SMP) && !defined(CONFIG_PPC_MERGE)
+		/* should this be per processor send/receive? */
+		seq_printf(p, "IPI (recv/sent): %10u/%u\n",
+				atomic_read(&ipi_recv), atomic_read(&ipi_sent));
+#endif
+#endif /* CONFIG_PPC32 */
+		seq_printf(p, "BAD: %10u\n", ppc_spurious_interrupts);
+	}
+	return 0;
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+void fixup_irqs(cpumask_t map)
+{
+	unsigned int irq;
+	static int warned;
+
+	for_each_irq(irq) {
+		cpumask_t mask;
+
+		if (irq_desc[irq].status & IRQ_PER_CPU)
+			continue;
+
+		cpus_and(mask, irq_affinity[irq], map);
+		if (any_online_cpu(mask) == NR_CPUS) {
+			printk("Breaking affinity for irq %i\n", irq);
+			mask = map;
+		}
+		if (irq_desc[irq].handler->set_affinity)
+			irq_desc[irq].handler->set_affinity(irq, mask);
+		else if (irq_desc[irq].action && !(warned++))
+			printk("Cannot set affinity for irq %i\n", irq);
+	}
+
+	local_irq_enable();
+	mdelay(1);
+	local_irq_disable();
+}
+#endif
+
+#ifdef CONFIG_PPC_ISERIES
+void do_IRQ(struct pt_regs *regs)
+{
+	struct paca_struct *lpaca;
+
+	irq_enter();
+
+#ifdef CONFIG_DEBUG_STACKOVERFLOW
+	/* Debugging check for stack overflow: is there less than 2KB free? */
+	{
+		long sp;
+
+		sp = __get_SP() & (THREAD_SIZE-1);
+
+		if (unlikely(sp < (sizeof(struct thread_info) + 2048))) {
+			printk("do_IRQ: stack overflow: %ld\n",
+				sp - sizeof(struct thread_info));
+			dump_stack();
+		}
+	}
+#endif
+
+	lpaca = get_paca();
+#ifdef CONFIG_SMP
+	if (lpaca->lppaca.int_dword.fields.ipi_cnt) {
+		lpaca->lppaca.int_dword.fields.ipi_cnt = 0;
+		iSeries_smp_message_recv(regs);
+	}
+#endif /* CONFIG_SMP */
+	if (hvlpevent_is_pending())
+		process_hvlpevents(regs);
+
+	irq_exit();
+
+	if (lpaca->lppaca.int_dword.fields.decr_int) {
+		lpaca->lppaca.int_dword.fields.decr_int = 0;
+		/* Signal a fake decrementer interrupt */
+		timer_interrupt(regs);
+	}
+}
+
+#else	/* CONFIG_PPC_ISERIES */
+
+void do_IRQ(struct pt_regs *regs)
+{
+	int irq;
+#ifdef CONFIG_IRQSTACKS
+	struct thread_info *curtp, *irqtp;
+#endif
+
+        irq_enter();
+
+#ifdef CONFIG_DEBUG_STACKOVERFLOW
+	/* Debugging check for stack overflow: is there less than 2KB free? */
+	{
+		long sp;
+
+		sp = __get_SP() & (THREAD_SIZE-1);
+
+		if (unlikely(sp < (sizeof(struct thread_info) + 2048))) {
+			printk("do_IRQ: stack overflow: %ld\n",
+				sp - sizeof(struct thread_info));
+			dump_stack();
+		}
+	}
+#endif
+
+	/*
+	 * Every platform is required to implement ppc_md.get_irq.
+	 * This function will either return an irq number or -1 to
+	 * indicate there are no more pending.
+	 * The value -2 is for buggy hardware and means that this IRQ
+	 * has already been handled. -- Tom
+	 */
+	irq = ppc_md.get_irq(regs);
+
+	if (irq >= 0) {
+#ifdef CONFIG_IRQSTACKS
+		/* Switch to the irq stack to handle this */
+		curtp = current_thread_info();
+		irqtp = hardirq_ctx[smp_processor_id()];
+		if (curtp != irqtp) {
+			irqtp->task = curtp->task;
+			irqtp->flags = 0;
+			call___do_IRQ(irq, regs, irqtp);
+			irqtp->task = NULL;
+			if (irqtp->flags)
+				set_bits(irqtp->flags, &curtp->flags);
+		} else
+#endif
+			__do_IRQ(irq, regs);
+	} else
+#ifdef CONFIG_PPC32
+		if (irq != -2)
+#endif
+			/* That's not SMP safe ... but who cares ? */
+			ppc_spurious_interrupts++;
+        irq_exit();
+}
+
+#endif	/* CONFIG_PPC_ISERIES */
+
+void __init init_IRQ(void)
+{
+#ifdef CONFIG_PPC64
+	static int once = 0;
+
+	if (once)
+		return;
+
+	once++;
+
+#endif
+	ppc_md.init_IRQ();
+#ifdef CONFIG_PPC64
+	irq_ctx_init();
+#endif
+}
+
+#ifdef CONFIG_PPC64
+#ifndef CONFIG_PPC_ISERIES
+/*
+ * Virtual IRQ mapping code, used on systems with XICS interrupt controllers.
+ */
+
+#define UNDEFINED_IRQ 0xffffffff
+unsigned int virt_irq_to_real_map[NR_IRQS];
+
+/*
+ * Don't use virtual irqs 0, 1, 2 for devices.
+ * The pcnet32 driver considers interrupt numbers < 2 to be invalid,
+ * and 2 is the XICS IPI interrupt.
+ * We limit virtual irqs to 17 less than NR_IRQS so that when we
+ * offset them by 16 (to reserve the first 16 for ISA interrupts)
+ * we don't end up with an interrupt number >= NR_IRQS.
+ */
+#define MIN_VIRT_IRQ	3
+#define MAX_VIRT_IRQ	(NR_IRQS - NUM_ISA_INTERRUPTS - 1)
+#define NR_VIRT_IRQS	(MAX_VIRT_IRQ - MIN_VIRT_IRQ + 1)
+
+void
+virt_irq_init(void)
+{
+	int i;
+	for (i = 0; i < NR_IRQS; i++)
+		virt_irq_to_real_map[i] = UNDEFINED_IRQ;
+}
+
+/* Create a mapping for a real_irq if it doesn't already exist.
+ * Return the virtual irq as a convenience.
+ */
+int virt_irq_create_mapping(unsigned int real_irq)
+{
+	unsigned int virq, first_virq;
+	static int warned;
+
+	if (ppc64_interrupt_controller == IC_OPEN_PIC)
+		return real_irq;	/* no mapping for openpic (for now) */
+
+	if (ppc64_interrupt_controller == IC_CELL_PIC)
+		return real_irq;	/* no mapping for iic either */
+
+	/* don't map interrupts < MIN_VIRT_IRQ */
+	if (real_irq < MIN_VIRT_IRQ) {
+		virt_irq_to_real_map[real_irq] = real_irq;
+		return real_irq;
+	}
+
+	/* map to a number between MIN_VIRT_IRQ and MAX_VIRT_IRQ */
+	virq = real_irq;
+	if (virq > MAX_VIRT_IRQ)
+		virq = (virq % NR_VIRT_IRQS) + MIN_VIRT_IRQ;
+
+	/* search for this number or a free slot */
+	first_virq = virq;
+	while (virt_irq_to_real_map[virq] != UNDEFINED_IRQ) {
+		if (virt_irq_to_real_map[virq] == real_irq)
+			return virq;
+		if (++virq > MAX_VIRT_IRQ)
+			virq = MIN_VIRT_IRQ;
+		if (virq == first_virq)
+			goto nospace;	/* oops, no free slots */
+	}
+
+	virt_irq_to_real_map[virq] = real_irq;
+	return virq;
+
+ nospace:
+	if (!warned) {
+		printk(KERN_CRIT "Interrupt table is full\n");
+		printk(KERN_CRIT "Increase NR_IRQS (currently %d) "
+		       "in your kernel sources and rebuild.\n", NR_IRQS);
+		warned = 1;
+	}
+	return NO_IRQ;
+}
+
+/*
+ * In most cases will get a hit on the very first slot checked in the
+ * virt_irq_to_real_map.  Only when there are a large number of
+ * IRQs will this be expensive.
+ */
+unsigned int real_irq_to_virt_slowpath(unsigned int real_irq)
+{
+	unsigned int virq;
+	unsigned int first_virq;
+
+	virq = real_irq;
+
+	if (virq > MAX_VIRT_IRQ)
+		virq = (virq % NR_VIRT_IRQS) + MIN_VIRT_IRQ;
+
+	first_virq = virq;
+
+	do {
+		if (virt_irq_to_real_map[virq] == real_irq)
+			return virq;
+
+		virq++;
+
+		if (virq >= MAX_VIRT_IRQ)
+			virq = 0;
+
+	} while (first_virq != virq);
+
+	return NO_IRQ;
+
+}
+
+#endif /* CONFIG_PPC_ISERIES */
+
+#ifdef CONFIG_IRQSTACKS
+struct thread_info *softirq_ctx[NR_CPUS];
+struct thread_info *hardirq_ctx[NR_CPUS];
+
+void irq_ctx_init(void)
+{
+	struct thread_info *tp;
+	int i;
+
+	for_each_cpu(i) {
+		memset((void *)softirq_ctx[i], 0, THREAD_SIZE);
+		tp = softirq_ctx[i];
+		tp->cpu = i;
+		tp->preempt_count = SOFTIRQ_OFFSET;
+
+		memset((void *)hardirq_ctx[i], 0, THREAD_SIZE);
+		tp = hardirq_ctx[i];
+		tp->cpu = i;
+		tp->preempt_count = HARDIRQ_OFFSET;
+	}
+}
+
+void do_softirq(void)
+{
+	unsigned long flags;
+	struct thread_info *curtp, *irqtp;
+
+	if (in_interrupt())
+		return;
+
+	local_irq_save(flags);
+
+	if (local_softirq_pending()) {
+		curtp = current_thread_info();
+		irqtp = softirq_ctx[smp_processor_id()];
+		irqtp->task = curtp->task;
+		call_do_softirq(irqtp);
+		irqtp->task = NULL;
+	}
+
+	local_irq_restore(flags);
+}
+EXPORT_SYMBOL(do_softirq);
+
+#endif /* CONFIG_IRQSTACKS */
+
+static int __init setup_noirqdistrib(char *str)
+{
+	distribute_irqs = 0;
+	return 1;
+}
+
+__setup("noirqdistrib", setup_noirqdistrib);
+#endif /* CONFIG_PPC64 */
diff --git a/arch/powerpc/kernel/lparcfg.c b/arch/powerpc/kernel/lparcfg.c
new file mode 100644
index 00000000000..5e954fae031
--- /dev/null
+++ b/arch/powerpc/kernel/lparcfg.c
@@ -0,0 +1,612 @@
+/*
+ * PowerPC64 LPAR Configuration Information Driver
+ *
+ * Dave Engebretsen engebret@us.ibm.com
+ *    Copyright (c) 2003 Dave Engebretsen
+ * Will Schmidt willschm@us.ibm.com
+ *    SPLPAR updates, Copyright (c) 2003 Will Schmidt IBM Corporation.
+ *    seq_file updates, Copyright (c) 2004 Will Schmidt IBM Corporation.
+ * Nathan Lynch nathanl@austin.ibm.com
+ *    Added lparcfg_write, Copyright (C) 2004 Nathan Lynch IBM Corporation.
+ *
+ *      This program is free software; you can redistribute it and/or
+ *      modify it under the terms of the GNU General Public License
+ *      as published by the Free Software Foundation; either version
+ *      2 of the License, or (at your option) any later version.
+ *
+ * This driver creates a proc file at /proc/ppc64/lparcfg which contains
+ * keyword - value pairs that specify the configuration of the partition.
+ */
+
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/errno.h>
+#include <linux/proc_fs.h>
+#include <linux/init.h>
+#include <linux/seq_file.h>
+#include <asm/uaccess.h>
+#include <asm/iseries/hv_lp_config.h>
+#include <asm/lppaca.h>
+#include <asm/hvcall.h>
+#include <asm/firmware.h>
+#include <asm/rtas.h>
+#include <asm/system.h>
+#include <asm/time.h>
+#include <asm/iseries/it_exp_vpd_panel.h>
+#include <asm/prom.h>
+#include <asm/systemcfg.h>
+
+#define MODULE_VERS "1.6"
+#define MODULE_NAME "lparcfg"
+
+/* #define LPARCFG_DEBUG */
+
+/* find a better place for this function... */
+void log_plpar_hcall_return(unsigned long rc, char *tag)
+{
+	if (rc == 0)		/* success, return */
+		return;
+/* check for null tag ? */
+	if (rc == H_Hardware)
+		printk(KERN_INFO
+		       "plpar-hcall (%s) failed with hardware fault\n", tag);
+	else if (rc == H_Function)
+		printk(KERN_INFO
+		       "plpar-hcall (%s) failed; function not allowed\n", tag);
+	else if (rc == H_Authority)
+		printk(KERN_INFO
+		       "plpar-hcall (%s) failed; not authorized to this function\n",
+		       tag);
+	else if (rc == H_Parameter)
+		printk(KERN_INFO "plpar-hcall (%s) failed; Bad parameter(s)\n",
+		       tag);
+	else
+		printk(KERN_INFO
+		       "plpar-hcall (%s) failed with unexpected rc(0x%lx)\n",
+		       tag, rc);
+
+}
+
+static struct proc_dir_entry *proc_ppc64_lparcfg;
+#define LPARCFG_BUFF_SIZE 4096
+
+#ifdef CONFIG_PPC_ISERIES
+
+/*
+ * For iSeries legacy systems, the PPA purr function is available from the
+ * emulated_time_base field in the paca.
+ */
+static unsigned long get_purr(void)
+{
+	unsigned long sum_purr = 0;
+	int cpu;
+	struct paca_struct *lpaca;
+
+	for_each_cpu(cpu) {
+		lpaca = paca + cpu;
+		sum_purr += lpaca->lppaca.emulated_time_base;
+
+#ifdef PURR_DEBUG
+		printk(KERN_INFO "get_purr for cpu (%d) has value (%ld) \n",
+			cpu, lpaca->lppaca.emulated_time_base);
+#endif
+	}
+	return sum_purr;
+}
+
+#define lparcfg_write NULL
+
+/*
+ * Methods used to fetch LPAR data when running on an iSeries platform.
+ */
+static int lparcfg_data(struct seq_file *m, void *v)
+{
+	unsigned long pool_id, lp_index;
+	int shared, entitled_capacity, max_entitled_capacity;
+	int processors, max_processors;
+	struct paca_struct *lpaca = get_paca();
+	unsigned long purr = get_purr();
+
+	seq_printf(m, "%s %s \n", MODULE_NAME, MODULE_VERS);
+
+	shared = (int)(lpaca->lppaca_ptr->shared_proc);
+	seq_printf(m, "serial_number=%c%c%c%c%c%c%c\n",
+		   e2a(xItExtVpdPanel.mfgID[2]),
+		   e2a(xItExtVpdPanel.mfgID[3]),
+		   e2a(xItExtVpdPanel.systemSerial[1]),
+		   e2a(xItExtVpdPanel.systemSerial[2]),
+		   e2a(xItExtVpdPanel.systemSerial[3]),
+		   e2a(xItExtVpdPanel.systemSerial[4]),
+		   e2a(xItExtVpdPanel.systemSerial[5]));
+
+	seq_printf(m, "system_type=%c%c%c%c\n",
+		   e2a(xItExtVpdPanel.machineType[0]),
+		   e2a(xItExtVpdPanel.machineType[1]),
+		   e2a(xItExtVpdPanel.machineType[2]),
+		   e2a(xItExtVpdPanel.machineType[3]));
+
+	lp_index = HvLpConfig_getLpIndex();
+	seq_printf(m, "partition_id=%d\n", (int)lp_index);
+
+	seq_printf(m, "system_active_processors=%d\n",
+		   (int)HvLpConfig_getSystemPhysicalProcessors());
+
+	seq_printf(m, "system_potential_processors=%d\n",
+		   (int)HvLpConfig_getSystemPhysicalProcessors());
+
+	processors = (int)HvLpConfig_getPhysicalProcessors();
+	seq_printf(m, "partition_active_processors=%d\n", processors);
+
+	max_processors = (int)HvLpConfig_getMaxPhysicalProcessors();
+	seq_printf(m, "partition_potential_processors=%d\n", max_processors);
+
+	if (shared) {
+		entitled_capacity = HvLpConfig_getSharedProcUnits();
+		max_entitled_capacity = HvLpConfig_getMaxSharedProcUnits();
+	} else {
+		entitled_capacity = processors * 100;
+		max_entitled_capacity = max_processors * 100;
+	}
+	seq_printf(m, "partition_entitled_capacity=%d\n", entitled_capacity);
+
+	seq_printf(m, "partition_max_entitled_capacity=%d\n",
+		   max_entitled_capacity);
+
+	if (shared) {
+		pool_id = HvLpConfig_getSharedPoolIndex();
+		seq_printf(m, "pool=%d\n", (int)pool_id);
+		seq_printf(m, "pool_capacity=%d\n",
+			   (int)(HvLpConfig_getNumProcsInSharedPool(pool_id) *
+				 100));
+		seq_printf(m, "purr=%ld\n", purr);
+	}
+
+	seq_printf(m, "shared_processor_mode=%d\n", shared);
+
+	return 0;
+}
+#endif				/* CONFIG_PPC_ISERIES */
+
+#ifdef CONFIG_PPC_PSERIES
+/*
+ * Methods used to fetch LPAR data when running on a pSeries platform.
+ */
+
+/*
+ * H_GET_PPP hcall returns info in 4 parms.
+ *  entitled_capacity,unallocated_capacity,
+ *  aggregation, resource_capability).
+ *
+ *  R4 = Entitled Processor Capacity Percentage.
+ *  R5 = Unallocated Processor Capacity Percentage.
+ *  R6 (AABBCCDDEEFFGGHH).
+ *      XXXX - reserved (0)
+ *          XXXX - reserved (0)
+ *              XXXX - Group Number
+ *                  XXXX - Pool Number.
+ *  R7 (IIJJKKLLMMNNOOPP).
+ *      XX - reserved. (0)
+ *        XX - bit 0-6 reserved (0).   bit 7 is Capped indicator.
+ *          XX - variable processor Capacity Weight
+ *            XX - Unallocated Variable Processor Capacity Weight.
+ *              XXXX - Active processors in Physical Processor Pool.
+ *                  XXXX  - Processors active on platform.
+ */
+static unsigned int h_get_ppp(unsigned long *entitled,
+			      unsigned long *unallocated,
+			      unsigned long *aggregation,
+			      unsigned long *resource)
+{
+	unsigned long rc;
+	rc = plpar_hcall_4out(H_GET_PPP, 0, 0, 0, 0, entitled, unallocated,
+			      aggregation, resource);
+
+	log_plpar_hcall_return(rc, "H_GET_PPP");
+
+	return rc;
+}
+
+static void h_pic(unsigned long *pool_idle_time, unsigned long *num_procs)
+{
+	unsigned long rc;
+	unsigned long dummy;
+	rc = plpar_hcall(H_PIC, 0, 0, 0, 0, pool_idle_time, num_procs, &dummy);
+
+	log_plpar_hcall_return(rc, "H_PIC");
+}
+
+static unsigned long get_purr(void);
+
+/* Track sum of all purrs across all processors. This is used to further */
+/* calculate usage values by different applications                       */
+
+static unsigned long get_purr(void)
+{
+	unsigned long sum_purr = 0;
+	int cpu;
+	struct cpu_usage *cu;
+
+	for_each_cpu(cpu) {
+		cu = &per_cpu(cpu_usage_array, cpu);
+		sum_purr += cu->current_tb;
+	}
+	return sum_purr;
+}
+
+#define SPLPAR_CHARACTERISTICS_TOKEN 20
+#define SPLPAR_MAXLENGTH 1026*(sizeof(char))
+
+/*
+ * parse_system_parameter_string()
+ * Retrieve the potential_processors, max_entitled_capacity and friends
+ * through the get-system-parameter rtas call.  Replace keyword strings as
+ * necessary.
+ */
+static void parse_system_parameter_string(struct seq_file *m)
+{
+	int call_status;
+
+	char *local_buffer = kmalloc(SPLPAR_MAXLENGTH, GFP_KERNEL);
+	if (!local_buffer) {
+		printk(KERN_ERR "%s %s kmalloc failure at line %d \n",
+		       __FILE__, __FUNCTION__, __LINE__);
+		return;
+	}
+
+	spin_lock(&rtas_data_buf_lock);
+	memset(rtas_data_buf, 0, SPLPAR_MAXLENGTH);
+	call_status = rtas_call(rtas_token("ibm,get-system-parameter"), 3, 1,
+				NULL,
+				SPLPAR_CHARACTERISTICS_TOKEN,
+				__pa(rtas_data_buf));
+	memcpy(local_buffer, rtas_data_buf, SPLPAR_MAXLENGTH);
+	spin_unlock(&rtas_data_buf_lock);
+
+	if (call_status != 0) {
+		printk(KERN_INFO
+		       "%s %s Error calling get-system-parameter (0x%x)\n",
+		       __FILE__, __FUNCTION__, call_status);
+	} else {
+		int splpar_strlen;
+		int idx, w_idx;
+		char *workbuffer = kmalloc(SPLPAR_MAXLENGTH, GFP_KERNEL);
+		if (!workbuffer) {
+			printk(KERN_ERR "%s %s kmalloc failure at line %d \n",
+			       __FILE__, __FUNCTION__, __LINE__);
+			kfree(local_buffer);
+			return;
+		}
+#ifdef LPARCFG_DEBUG
+		printk(KERN_INFO "success calling get-system-parameter \n");
+#endif
+		splpar_strlen = local_buffer[0] * 16 + local_buffer[1];
+		local_buffer += 2;	/* step over strlen value */
+
+		memset(workbuffer, 0, SPLPAR_MAXLENGTH);
+		w_idx = 0;
+		idx = 0;
+		while ((*local_buffer) && (idx < splpar_strlen)) {
+			workbuffer[w_idx++] = local_buffer[idx++];
+			if ((local_buffer[idx] == ',')
+			    || (local_buffer[idx] == '\0')) {
+				workbuffer[w_idx] = '\0';
+				if (w_idx) {
+					/* avoid the empty string */
+					seq_printf(m, "%s\n", workbuffer);
+				}
+				memset(workbuffer, 0, SPLPAR_MAXLENGTH);
+				idx++;	/* skip the comma */
+				w_idx = 0;
+			} else if (local_buffer[idx] == '=') {
+				/* code here to replace workbuffer contents
+				   with different keyword strings */
+				if (0 == strcmp(workbuffer, "MaxEntCap")) {
+					strcpy(workbuffer,
+					       "partition_max_entitled_capacity");
+					w_idx = strlen(workbuffer);
+				}
+				if (0 == strcmp(workbuffer, "MaxPlatProcs")) {
+					strcpy(workbuffer,
+					       "system_potential_processors");
+					w_idx = strlen(workbuffer);
+				}
+			}
+		}
+		kfree(workbuffer);
+		local_buffer -= 2;	/* back up over strlen value */
+	}
+	kfree(local_buffer);
+}
+
+static int lparcfg_count_active_processors(void);
+
+/* Return the number of processors in the system.
+ * This function reads through the device tree and counts
+ * the virtual processors, this does not include threads.
+ */
+static int lparcfg_count_active_processors(void)
+{
+	struct device_node *cpus_dn = NULL;
+	int count = 0;
+
+	while ((cpus_dn = of_find_node_by_type(cpus_dn, "cpu"))) {
+#ifdef LPARCFG_DEBUG
+		printk(KERN_ERR "cpus_dn %p \n", cpus_dn);
+#endif
+		count++;
+	}
+	return count;
+}
+
+static int lparcfg_data(struct seq_file *m, void *v)
+{
+	int partition_potential_processors;
+	int partition_active_processors;
+	struct device_node *rootdn;
+	const char *model = "";
+	const char *system_id = "";
+	unsigned int *lp_index_ptr, lp_index = 0;
+	struct device_node *rtas_node;
+	int *lrdrp;
+
+	rootdn = find_path_device("/");
+	if (rootdn) {
+		model = get_property(rootdn, "model", NULL);
+		system_id = get_property(rootdn, "system-id", NULL);
+		lp_index_ptr = (unsigned int *)
+		    get_property(rootdn, "ibm,partition-no", NULL);
+		if (lp_index_ptr)
+			lp_index = *lp_index_ptr;
+	}
+
+	seq_printf(m, "%s %s \n", MODULE_NAME, MODULE_VERS);
+
+	seq_printf(m, "serial_number=%s\n", system_id);
+
+	seq_printf(m, "system_type=%s\n", model);
+
+	seq_printf(m, "partition_id=%d\n", (int)lp_index);
+
+	rtas_node = find_path_device("/rtas");
+	lrdrp = (int *)get_property(rtas_node, "ibm,lrdr-capacity", NULL);
+
+	if (lrdrp == NULL) {
+		partition_potential_processors = _systemcfg->processorCount;
+	} else {
+		partition_potential_processors = *(lrdrp + 4);
+	}
+
+	partition_active_processors = lparcfg_count_active_processors();
+
+	if (firmware_has_feature(FW_FEATURE_SPLPAR)) {
+		unsigned long h_entitled, h_unallocated;
+		unsigned long h_aggregation, h_resource;
+		unsigned long pool_idle_time, pool_procs;
+		unsigned long purr;
+
+		h_get_ppp(&h_entitled, &h_unallocated, &h_aggregation,
+			  &h_resource);
+
+		seq_printf(m, "R4=0x%lx\n", h_entitled);
+		seq_printf(m, "R5=0x%lx\n", h_unallocated);
+		seq_printf(m, "R6=0x%lx\n", h_aggregation);
+		seq_printf(m, "R7=0x%lx\n", h_resource);
+
+		purr = get_purr();
+
+		/* this call handles the ibm,get-system-parameter contents */
+		parse_system_parameter_string(m);
+
+		seq_printf(m, "partition_entitled_capacity=%ld\n", h_entitled);
+
+		seq_printf(m, "group=%ld\n", (h_aggregation >> 2 * 8) & 0xffff);
+
+		seq_printf(m, "system_active_processors=%ld\n",
+			   (h_resource >> 0 * 8) & 0xffff);
+
+		/* pool related entries are apropriate for shared configs */
+		if (paca[0].lppaca.shared_proc) {
+
+			h_pic(&pool_idle_time, &pool_procs);
+
+			seq_printf(m, "pool=%ld\n",
+				   (h_aggregation >> 0 * 8) & 0xffff);
+
+			/* report pool_capacity in percentage */
+			seq_printf(m, "pool_capacity=%ld\n",
+				   ((h_resource >> 2 * 8) & 0xffff) * 100);
+
+			seq_printf(m, "pool_idle_time=%ld\n", pool_idle_time);
+
+			seq_printf(m, "pool_num_procs=%ld\n", pool_procs);
+		}
+
+		seq_printf(m, "unallocated_capacity_weight=%ld\n",
+			   (h_resource >> 4 * 8) & 0xFF);
+
+		seq_printf(m, "capacity_weight=%ld\n",
+			   (h_resource >> 5 * 8) & 0xFF);
+
+		seq_printf(m, "capped=%ld\n", (h_resource >> 6 * 8) & 0x01);
+
+		seq_printf(m, "unallocated_capacity=%ld\n", h_unallocated);
+
+		seq_printf(m, "purr=%ld\n", purr);
+
+	} else {		/* non SPLPAR case */
+
+		seq_printf(m, "system_active_processors=%d\n",
+			   partition_potential_processors);
+
+		seq_printf(m, "system_potential_processors=%d\n",
+			   partition_potential_processors);
+
+		seq_printf(m, "partition_max_entitled_capacity=%d\n",
+			   partition_potential_processors * 100);
+
+		seq_printf(m, "partition_entitled_capacity=%d\n",
+			   partition_active_processors * 100);
+	}
+
+	seq_printf(m, "partition_active_processors=%d\n",
+		   partition_active_processors);
+
+	seq_printf(m, "partition_potential_processors=%d\n",
+		   partition_potential_processors);
+
+	seq_printf(m, "shared_processor_mode=%d\n", paca[0].lppaca.shared_proc);
+
+	return 0;
+}
+
+/*
+ * Interface for changing system parameters (variable capacity weight
+ * and entitled capacity).  Format of input is "param_name=value";
+ * anything after value is ignored.  Valid parameters at this time are
+ * "partition_entitled_capacity" and "capacity_weight".  We use
+ * H_SET_PPP to alter parameters.
+ *
+ * This function should be invoked only on systems with
+ * FW_FEATURE_SPLPAR.
+ */
+static ssize_t lparcfg_write(struct file *file, const char __user * buf,
+			     size_t count, loff_t * off)
+{
+	char *kbuf;
+	char *tmp;
+	u64 new_entitled, *new_entitled_ptr = &new_entitled;
+	u8 new_weight, *new_weight_ptr = &new_weight;
+
+	unsigned long current_entitled;	/* parameters for h_get_ppp */
+	unsigned long dummy;
+	unsigned long resource;
+	u8 current_weight;
+
+	ssize_t retval = -ENOMEM;
+
+	kbuf = kmalloc(count, GFP_KERNEL);
+	if (!kbuf)
+		goto out;
+
+	retval = -EFAULT;
+	if (copy_from_user(kbuf, buf, count))
+		goto out;
+
+	retval = -EINVAL;
+	kbuf[count - 1] = '\0';
+	tmp = strchr(kbuf, '=');
+	if (!tmp)
+		goto out;
+
+	*tmp++ = '\0';
+
+	if (!strcmp(kbuf, "partition_entitled_capacity")) {
+		char *endp;
+		*new_entitled_ptr = (u64) simple_strtoul(tmp, &endp, 10);
+		if (endp == tmp)
+			goto out;
+		new_weight_ptr = &current_weight;
+	} else if (!strcmp(kbuf, "capacity_weight")) {
+		char *endp;
+		*new_weight_ptr = (u8) simple_strtoul(tmp, &endp, 10);
+		if (endp == tmp)
+			goto out;
+		new_entitled_ptr = &current_entitled;
+	} else
+		goto out;
+
+	/* Get our current parameters */
+	retval = h_get_ppp(&current_entitled, &dummy, &dummy, &resource);
+	if (retval) {
+		retval = -EIO;
+		goto out;
+	}
+
+	current_weight = (resource >> 5 * 8) & 0xFF;
+
+	pr_debug("%s: current_entitled = %lu, current_weight = %lu\n",
+		 __FUNCTION__, current_entitled, current_weight);
+
+	pr_debug("%s: new_entitled = %lu, new_weight = %lu\n",
+		 __FUNCTION__, *new_entitled_ptr, *new_weight_ptr);
+
+	retval = plpar_hcall_norets(H_SET_PPP, *new_entitled_ptr,
+				    *new_weight_ptr);
+
+	if (retval == H_Success || retval == H_Constrained) {
+		retval = count;
+	} else if (retval == H_Busy) {
+		retval = -EBUSY;
+	} else if (retval == H_Hardware) {
+		retval = -EIO;
+	} else if (retval == H_Parameter) {
+		retval = -EINVAL;
+	} else {
+		printk(KERN_WARNING "%s: received unknown hv return code %ld",
+		       __FUNCTION__, retval);
+		retval = -EIO;
+	}
+
+      out:
+	kfree(kbuf);
+	return retval;
+}
+
+#endif				/* CONFIG_PPC_PSERIES */
+
+static int lparcfg_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, lparcfg_data, NULL);
+}
+
+struct file_operations lparcfg_fops = {
+      .owner	= THIS_MODULE,
+      .read	= seq_read,
+      .open	= lparcfg_open,
+      .release	= single_release,
+};
+
+int __init lparcfg_init(void)
+{
+	struct proc_dir_entry *ent;
+	mode_t mode = S_IRUSR | S_IRGRP | S_IROTH;
+
+	/* Allow writing if we have FW_FEATURE_SPLPAR */
+	if (firmware_has_feature(FW_FEATURE_SPLPAR)) {
+		lparcfg_fops.write = lparcfg_write;
+		mode |= S_IWUSR;
+	}
+
+	ent = create_proc_entry("ppc64/lparcfg", mode, NULL);
+	if (ent) {
+		ent->proc_fops = &lparcfg_fops;
+		ent->data = kmalloc(LPARCFG_BUFF_SIZE, GFP_KERNEL);
+		if (!ent->data) {
+			printk(KERN_ERR
+			       "Failed to allocate buffer for lparcfg\n");
+			remove_proc_entry("lparcfg", ent->parent);
+			return -ENOMEM;
+		}
+	} else {
+		printk(KERN_ERR "Failed to create ppc64/lparcfg\n");
+		return -EIO;
+	}
+
+	proc_ppc64_lparcfg = ent;
+	return 0;
+}
+
+void __exit lparcfg_cleanup(void)
+{
+	if (proc_ppc64_lparcfg) {
+		kfree(proc_ppc64_lparcfg->data);
+		remove_proc_entry("lparcfg", proc_ppc64_lparcfg->parent);
+	}
+}
+
+module_init(lparcfg_init);
+module_exit(lparcfg_cleanup);
+MODULE_DESCRIPTION("Interface for LPAR configuration data");
+MODULE_AUTHOR("Dave Engebretsen");
+MODULE_LICENSE("GPL");
diff --git a/arch/powerpc/kernel/misc_32.S b/arch/powerpc/kernel/misc_32.S
index 3bedb532aed..f6d84a75ed2 100644
--- a/arch/powerpc/kernel/misc_32.S
+++ b/arch/powerpc/kernel/misc_32.S
@@ -519,7 +519,7 @@ END_FTR_SECTION_IFCLR(CPU_FTR_SPLIT_ID_CACHE)
  *
  * flush_icache_range(unsigned long start, unsigned long stop)
  */
-_GLOBAL(flush_icache_range)
+_GLOBAL(__flush_icache_range)
 BEGIN_FTR_SECTION
 	blr				/* for 601, do nothing */
 END_FTR_SECTION_IFCLR(CPU_FTR_SPLIT_ID_CACHE)
@@ -607,27 +607,6 @@ _GLOBAL(invalidate_dcache_range)
 	sync				/* wait for dcbi's to get to ram */
 	blr
 
-#ifdef CONFIG_NOT_COHERENT_CACHE
-/*
- * 40x cores have 8K or 16K dcache and 32 byte line size.
- * 44x has a 32K dcache and 32 byte line size.
- * 8xx has 1, 2, 4, 8K variants.
- * For now, cover the worst case of the 44x.
- * Must be called with external interrupts disabled.
- */
-#define CACHE_NWAYS	64
-#define CACHE_NLINES	16
-
-_GLOBAL(flush_dcache_all)
-	li	r4, (2 * CACHE_NWAYS * CACHE_NLINES)
-	mtctr	r4
-	lis     r5, KERNELBASE@h
-1:	lwz	r3, 0(r5)		/* Load one word from every line */
-	addi	r5, r5, L1_CACHE_BYTES
-	bdnz    1b
-	blr
-#endif /* CONFIG_NOT_COHERENT_CACHE */
-
 /*
  * Flush a particular page from the data cache to RAM.
  * Note: this is necessary because the instruction cache does *not*
diff --git a/arch/powerpc/kernel/misc_64.S b/arch/powerpc/kernel/misc_64.S
index ae1433da09b..ae48a002f81 100644
--- a/arch/powerpc/kernel/misc_64.S
+++ b/arch/powerpc/kernel/misc_64.S
@@ -89,12 +89,12 @@ _GLOBAL(call_do_softirq)
 	mtlr	r0
 	blr
 
-_GLOBAL(call_handle_IRQ_event)
+_GLOBAL(call___do_IRQ)
 	mflr	r0
 	std	r0,16(r1)
-	stdu	r1,THREAD_SIZE-112(r6)
-	mr	r1,r6
-	bl	.handle_IRQ_event
+	stdu	r1,THREAD_SIZE-112(r5)
+	mr	r1,r5
+	bl	.__do_IRQ
 	ld	r1,0(r1)
 	ld	r0,16(r1)
 	mtlr	r0
diff --git a/arch/powerpc/kernel/paca.c b/arch/powerpc/kernel/paca.c
new file mode 100644
index 00000000000..3cf2517c5f9
--- /dev/null
+++ b/arch/powerpc/kernel/paca.c
@@ -0,0 +1,142 @@
+/*
+ * c 2001 PPC 64 Team, IBM Corp
+ *
+ *      This program is free software; you can redistribute it and/or
+ *      modify it under the terms of the GNU General Public License
+ *      as published by the Free Software Foundation; either version
+ *      2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/config.h>
+#include <linux/types.h>
+#include <linux/threads.h>
+#include <linux/module.h>
+
+#include <asm/processor.h>
+#include <asm/ptrace.h>
+#include <asm/page.h>
+#include <asm/systemcfg.h>
+#include <asm/lppaca.h>
+#include <asm/iseries/it_lp_queue.h>
+#include <asm/paca.h>
+
+static union {
+	struct systemcfg	data;
+	u8			page[PAGE_SIZE];
+} systemcfg_store __attribute__((__section__(".data.page.aligned")));
+struct systemcfg *_systemcfg = &systemcfg_store.data;
+
+
+/* This symbol is provided by the linker - let it fill in the paca
+ * field correctly */
+extern unsigned long __toc_start;
+
+/* The Paca is an array with one entry per processor.  Each contains an
+ * lppaca, which contains the information shared between the
+ * hypervisor and Linux.  Each also contains an ItLpRegSave area which
+ * is used by the hypervisor to save registers.
+ * On systems with hardware multi-threading, there are two threads
+ * per processor.  The Paca array must contain an entry for each thread.
+ * The VPD Areas will give a max logical processors = 2 * max physical
+ * processors.  The processor VPD array needs one entry per physical
+ * processor (not thread).
+ */
+#define PACA_INIT_COMMON(number, start, asrr, asrv)			    \
+	.lock_token = 0x8000,						    \
+	.paca_index = (number),		/* Paca Index */		    \
+	.default_decr = 0x00ff0000,	/* Initial Decr */		    \
+	.kernel_toc = (unsigned long)(&__toc_start) + 0x8000UL,		    \
+	.stab_real = (asrr), 		/* Real pointer to segment table */ \
+	.stab_addr = (asrv),		/* Virt pointer to segment table */ \
+	.cpu_start = (start),		/* Processor start */		    \
+	.hw_cpu_id = 0xffff,						    \
+	.lppaca = {							    \
+		.desc = 0xd397d781,	/* "LpPa" */			    \
+		.size = sizeof(struct lppaca),				    \
+		.dyn_proc_status = 2,					    \
+		.decr_val = 0x00ff0000,					    \
+		.fpregs_in_use = 1,					    \
+		.end_of_quantum = 0xfffffffffffffffful,			    \
+		.slb_count = 64,					    \
+		.vmxregs_in_use = 0,					    \
+	},								    \
+
+#ifdef CONFIG_PPC_ISERIES
+#define PACA_INIT_ISERIES(number)					    \
+	.lppaca_ptr = &paca[number].lppaca,				    \
+	.reg_save_ptr = &paca[number].reg_save,				    \
+	.reg_save = {							    \
+		.xDesc = 0xd397d9e2,	/* "LpRS" */			    \
+		.xSize = sizeof(struct ItLpRegSave)			    \
+	}
+
+#define PACA_INIT(number)						    \
+{									    \
+	PACA_INIT_COMMON(number, 0, 0, 0)				    \
+	PACA_INIT_ISERIES(number)					    \
+}
+
+#define BOOTCPU_PACA_INIT(number)					    \
+{									    \
+	PACA_INIT_COMMON(number, 1, 0, (u64)&initial_stab)		    \
+	PACA_INIT_ISERIES(number)					    \
+}
+
+#else
+#define PACA_INIT(number)						    \
+{									    \
+	PACA_INIT_COMMON(number, 0, 0, 0)				    \
+}
+
+#define BOOTCPU_PACA_INIT(number)					    \
+{									    \
+	PACA_INIT_COMMON(number, 1, STAB0_PHYS_ADDR, (u64)&initial_stab)    \
+}
+#endif
+
+struct paca_struct paca[] = {
+	BOOTCPU_PACA_INIT(0),
+#if NR_CPUS > 1
+	PACA_INIT(  1), PACA_INIT(  2), PACA_INIT(  3),
+#if NR_CPUS > 4
+	PACA_INIT(  4), PACA_INIT(  5), PACA_INIT(  6), PACA_INIT(  7),
+#if NR_CPUS > 8
+	PACA_INIT(  8), PACA_INIT(  9), PACA_INIT( 10), PACA_INIT( 11),
+	PACA_INIT( 12), PACA_INIT( 13), PACA_INIT( 14), PACA_INIT( 15),
+	PACA_INIT( 16), PACA_INIT( 17), PACA_INIT( 18), PACA_INIT( 19),
+	PACA_INIT( 20), PACA_INIT( 21), PACA_INIT( 22), PACA_INIT( 23),
+	PACA_INIT( 24), PACA_INIT( 25), PACA_INIT( 26), PACA_INIT( 27),
+	PACA_INIT( 28), PACA_INIT( 29), PACA_INIT( 30), PACA_INIT( 31),
+#if NR_CPUS > 32
+	PACA_INIT( 32), PACA_INIT( 33), PACA_INIT( 34), PACA_INIT( 35),
+	PACA_INIT( 36), PACA_INIT( 37), PACA_INIT( 38), PACA_INIT( 39),
+	PACA_INIT( 40), PACA_INIT( 41), PACA_INIT( 42), PACA_INIT( 43),
+	PACA_INIT( 44), PACA_INIT( 45), PACA_INIT( 46), PACA_INIT( 47),
+	PACA_INIT( 48), PACA_INIT( 49), PACA_INIT( 50), PACA_INIT( 51),
+	PACA_INIT( 52), PACA_INIT( 53), PACA_INIT( 54), PACA_INIT( 55),
+	PACA_INIT( 56), PACA_INIT( 57), PACA_INIT( 58), PACA_INIT( 59),
+	PACA_INIT( 60), PACA_INIT( 61), PACA_INIT( 62), PACA_INIT( 63),
+#if NR_CPUS > 64
+	PACA_INIT( 64), PACA_INIT( 65), PACA_INIT( 66), PACA_INIT( 67),
+	PACA_INIT( 68), PACA_INIT( 69), PACA_INIT( 70), PACA_INIT( 71),
+	PACA_INIT( 72), PACA_INIT( 73), PACA_INIT( 74), PACA_INIT( 75),
+	PACA_INIT( 76), PACA_INIT( 77), PACA_INIT( 78), PACA_INIT( 79),
+	PACA_INIT( 80), PACA_INIT( 81), PACA_INIT( 82), PACA_INIT( 83),
+	PACA_INIT( 84), PACA_INIT( 85), PACA_INIT( 86), PACA_INIT( 87),
+	PACA_INIT( 88), PACA_INIT( 89), PACA_INIT( 90), PACA_INIT( 91),
+	PACA_INIT( 92), PACA_INIT( 93), PACA_INIT( 94), PACA_INIT( 95),
+	PACA_INIT( 96), PACA_INIT( 97), PACA_INIT( 98), PACA_INIT( 99),
+	PACA_INIT(100), PACA_INIT(101), PACA_INIT(102), PACA_INIT(103),
+	PACA_INIT(104), PACA_INIT(105), PACA_INIT(106), PACA_INIT(107),
+	PACA_INIT(108), PACA_INIT(109), PACA_INIT(110), PACA_INIT(111),
+	PACA_INIT(112), PACA_INIT(113), PACA_INIT(114), PACA_INIT(115),
+	PACA_INIT(116), PACA_INIT(117), PACA_INIT(118), PACA_INIT(119),
+	PACA_INIT(120), PACA_INIT(121), PACA_INIT(122), PACA_INIT(123),
+	PACA_INIT(124), PACA_INIT(125), PACA_INIT(126), PACA_INIT(127),
+#endif
+#endif
+#endif
+#endif
+#endif
+};
+EXPORT_SYMBOL(paca);
diff --git a/arch/powerpc/kernel/ppc_ksyms.c b/arch/powerpc/kernel/ppc_ksyms.c
index 47d6f7e2ea9..5dcf4ba05ee 100644
--- a/arch/powerpc/kernel/ppc_ksyms.c
+++ b/arch/powerpc/kernel/ppc_ksyms.c
@@ -44,6 +44,7 @@
 #include <asm/cputable.h>
 #include <asm/btext.h>
 #include <asm/div64.h>
+#include <asm/signal.h>
 
 #ifdef  CONFIG_8xx
 #include <asm/commproc.h>
@@ -56,7 +57,6 @@ extern void machine_check_exception(struct pt_regs *regs);
 extern void alignment_exception(struct pt_regs *regs);
 extern void program_check_exception(struct pt_regs *regs);
 extern void single_step_exception(struct pt_regs *regs);
-extern int do_signal(sigset_t *, struct pt_regs *);
 extern int pmac_newworld;
 extern int sys_sigreturn(struct pt_regs *regs);
 
@@ -188,9 +188,6 @@ EXPORT_SYMBOL(adb_try_handler_change);
 EXPORT_SYMBOL(cuda_request);
 EXPORT_SYMBOL(cuda_poll);
 #endif /* CONFIG_ADB_CUDA */
-#if defined(CONFIG_PPC_MULTIPLATFORM) && defined(CONFIG_PPC32)
-EXPORT_SYMBOL(_machine);
-#endif
 #ifdef CONFIG_PPC_PMAC
 EXPORT_SYMBOL(sys_ctrler);
 #endif
diff --git a/arch/powerpc/kernel/proc_ppc64.c b/arch/powerpc/kernel/proc_ppc64.c
new file mode 100644
index 00000000000..a1c19502fe8
--- /dev/null
+++ b/arch/powerpc/kernel/proc_ppc64.c
@@ -0,0 +1,126 @@
+/*
+ * Copyright (C) 2001 Mike Corrigan & Dave Engebretsen IBM Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
+ */
+
+#include <linux/config.h>
+#include <linux/init.h>
+#include <linux/mm.h>
+#include <linux/proc_fs.h>
+#include <linux/slab.h>
+#include <linux/kernel.h>
+
+#include <asm/systemcfg.h>
+#include <asm/rtas.h>
+#include <asm/uaccess.h>
+#include <asm/prom.h>
+
+static loff_t  page_map_seek( struct file *file, loff_t off, int whence);
+static ssize_t page_map_read( struct file *file, char __user *buf, size_t nbytes,
+			      loff_t *ppos);
+static int     page_map_mmap( struct file *file, struct vm_area_struct *vma );
+
+static struct file_operations page_map_fops = {
+	.llseek	= page_map_seek,
+	.read	= page_map_read,
+	.mmap	= page_map_mmap
+};
+
+/*
+ * Create the ppc64 and ppc64/rtas directories early. This allows us to
+ * assume that they have been previously created in drivers.
+ */
+static int __init proc_ppc64_create(void)
+{
+	struct proc_dir_entry *root;
+
+	root = proc_mkdir("ppc64", NULL);
+	if (!root)
+		return 1;
+
+	if (!(platform_is_pseries() || _machine == PLATFORM_CELL))
+		return 0;
+
+	if (!proc_mkdir("rtas", root))
+		return 1;
+
+	if (!proc_symlink("rtas", NULL, "ppc64/rtas"))
+		return 1;
+
+	return 0;
+}
+core_initcall(proc_ppc64_create);
+
+static int __init proc_ppc64_init(void)
+{
+	struct proc_dir_entry *pde;
+
+	pde = create_proc_entry("ppc64/systemcfg", S_IFREG|S_IRUGO, NULL);
+	if (!pde)
+		return 1;
+	pde->nlink = 1;
+	pde->data = _systemcfg;
+	pde->size = PAGE_SIZE;
+	pde->proc_fops = &page_map_fops;
+
+	return 0;
+}
+__initcall(proc_ppc64_init);
+
+static loff_t page_map_seek( struct file *file, loff_t off, int whence)
+{
+	loff_t new;
+	struct proc_dir_entry *dp = PDE(file->f_dentry->d_inode);
+
+	switch(whence) {
+	case 0:
+		new = off;
+		break;
+	case 1:
+		new = file->f_pos + off;
+		break;
+	case 2:
+		new = dp->size + off;
+		break;
+	default:
+		return -EINVAL;
+	}
+	if ( new < 0 || new > dp->size )
+		return -EINVAL;
+	return (file->f_pos = new);
+}
+
+static ssize_t page_map_read( struct file *file, char __user *buf, size_t nbytes,
+			      loff_t *ppos)
+{
+	struct proc_dir_entry *dp = PDE(file->f_dentry->d_inode);
+	return simple_read_from_buffer(buf, nbytes, ppos, dp->data, dp->size);
+}
+
+static int page_map_mmap( struct file *file, struct vm_area_struct *vma )
+{
+	struct proc_dir_entry *dp = PDE(file->f_dentry->d_inode);
+
+	vma->vm_flags |= VM_SHM | VM_LOCKED;
+
+	if ((vma->vm_end - vma->vm_start) > dp->size)
+		return -EINVAL;
+
+	remap_pfn_range(vma, vma->vm_start, __pa(dp->data) >> PAGE_SHIFT,
+						dp->size, vma->vm_page_prot);
+	return 0;
+}
+
diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c
index f645adb5753..6a5b468edb4 100644
--- a/arch/powerpc/kernel/prom.c
+++ b/arch/powerpc/kernel/prom.c
@@ -48,9 +48,6 @@
 #include <asm/machdep.h>
 #include <asm/pSeries_reconfig.h>
 #include <asm/pci-bridge.h>
-#ifdef CONFIG_PPC64
-#include <asm/systemcfg.h>
-#endif
 
 #ifdef DEBUG
 #define DBG(fmt...) printk(KERN_ERR fmt)
@@ -74,10 +71,6 @@ struct isa_reg_property {
 typedef int interpret_func(struct device_node *, unsigned long *,
 			   int, int, int);
 
-extern struct rtas_t rtas;
-extern struct lmb lmb;
-extern unsigned long klimit;
-
 static int __initdata dt_root_addr_cells;
 static int __initdata dt_root_size_cells;
 
@@ -391,7 +384,7 @@ static int __devinit finish_node_interrupts(struct device_node *np,
 
 #ifdef CONFIG_PPC64
 		/* We offset irq numbers for the u3 MPIC by 128 in PowerMac */
-		if (systemcfg->platform == PLATFORM_POWERMAC && ic && ic->parent) {
+		if (_machine == PLATFORM_POWERMAC && ic && ic->parent) {
 			char *name = get_property(ic->parent, "name", NULL);
 			if (name && !strcmp(name, "u3"))
 				np->intrs[intrcount].line += 128;
@@ -1087,9 +1080,9 @@ void __init unflatten_device_tree(void)
 static int __init early_init_dt_scan_cpus(unsigned long node,
 					  const char *uname, int depth, void *data)
 {
-	char *type = of_get_flat_dt_prop(node, "device_type", NULL);
 	u32 *prop;
-	unsigned long size = 0;
+	unsigned long size;
+	char *type = of_get_flat_dt_prop(node, "device_type", &size);
 
 	/* We are scanning "cpu" nodes only */
 	if (type == NULL || strcmp(type, "cpu") != 0)
@@ -1115,7 +1108,7 @@ static int __init early_init_dt_scan_cpus(unsigned long node,
 
 #ifdef CONFIG_ALTIVEC
 	/* Check if we have a VMX and eventually update CPU features */
-	prop = (u32 *)of_get_flat_dt_prop(node, "ibm,vmx", &size);
+	prop = (u32 *)of_get_flat_dt_prop(node, "ibm,vmx", NULL);
 	if (prop && (*prop) > 0) {
 		cur_cpu_spec->cpu_features |= CPU_FTR_ALTIVEC;
 		cur_cpu_spec->cpu_user_features |= PPC_FEATURE_HAS_ALTIVEC;
@@ -1161,13 +1154,9 @@ static int __init early_init_dt_scan_chosen(unsigned long node,
 	prop = (u32 *)of_get_flat_dt_prop(node, "linux,platform", NULL);
 	if (prop == NULL)
 		return 0;
-#ifdef CONFIG_PPC64
-	systemcfg->platform = *prop;
-#else
 #ifdef CONFIG_PPC_MULTIPLATFORM
 	_machine = *prop;
 #endif
-#endif
 
 #ifdef CONFIG_PPC64
 	/* check if iommu is forced on or off */
@@ -1264,7 +1253,14 @@ static int __init early_init_dt_scan_memory(unsigned long node,
 	unsigned long l;
 
 	/* We are scanning "memory" nodes only */
-	if (type == NULL || strcmp(type, "memory") != 0)
+	if (type == NULL) {
+		/*
+		 * The longtrail doesn't have a device_type on the
+		 * /memory node, so look for the node called /memory@0.
+		 */
+		if (depth != 1 || strcmp(uname, "memory@0") != 0)
+			return 0;
+	} else if (strcmp(type, "memory") != 0)
 		return 0;
 
 	reg = (cell_t *)of_get_flat_dt_prop(node, "reg", &l);
@@ -1339,9 +1335,6 @@ void __init early_init_devtree(void *params)
 	of_scan_flat_dt(early_init_dt_scan_memory, NULL);
 	lmb_enforce_memory_limit(memory_limit);
 	lmb_analyze();
-#ifdef CONFIG_PPC64
-	systemcfg->physicalMemorySize = lmb_phys_mem_size();
-#endif
 	lmb_reserve(0, __pa(klimit));
 
 	DBG("Phys. mem: %lx\n", lmb_phys_mem_size());
@@ -1908,7 +1901,7 @@ static int of_finish_dynamic_node(struct device_node *node,
 	/* We don't support that function on PowerMac, at least
 	 * not yet
 	 */
-	if (systemcfg->platform == PLATFORM_POWERMAC)
+	if (_machine == PLATFORM_POWERMAC)
 		return -ENODEV;
 
 	/* fix up new node's linux_phandle field */
@@ -1992,9 +1985,11 @@ int prom_add_property(struct device_node* np, struct property* prop)
 	*next = prop;
 	write_unlock(&devtree_lock);
 
+#ifdef CONFIG_PROC_DEVICETREE
 	/* try to add to proc as well if it was initialized */
 	if (np->pde)
 		proc_device_tree_add_prop(np->pde, prop);
+#endif /* CONFIG_PROC_DEVICETREE */
 
 	return 0;
 }
diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c
index 6dc33d19fc2..4ce0105c308 100644
--- a/arch/powerpc/kernel/prom_init.c
+++ b/arch/powerpc/kernel/prom_init.c
@@ -94,11 +94,17 @@ extern const struct linux_logo logo_linux_clut224;
 #ifdef CONFIG_PPC64
 #define RELOC(x)        (*PTRRELOC(&(x)))
 #define ADDR(x)		(u32) add_reloc_offset((unsigned long)(x))
+#define OF_WORKAROUNDS	0
 #else
 #define RELOC(x)	(x)
 #define ADDR(x)		(u32) (x)
+#define OF_WORKAROUNDS	of_workarounds
+int of_workarounds;
 #endif
 
+#define OF_WA_CLAIM	1	/* do phys/virt claim separately, then map */
+#define OF_WA_LONGTRAIL	2	/* work around longtrail bugs */
+
 #define PROM_BUG() do {						\
         prom_printf("kernel BUG at %s line 0x%x!\n",		\
 		    RELOC(__FILE__), __LINE__);			\
@@ -111,11 +117,6 @@ extern const struct linux_logo logo_linux_clut224;
 #define prom_debug(x...)
 #endif
 
-#ifdef CONFIG_PPC32
-#define PLATFORM_POWERMAC	_MACH_Pmac
-#define PLATFORM_CHRP		_MACH_chrp
-#endif
-
 
 typedef u32 prom_arg_t;
 
@@ -128,10 +129,11 @@ struct prom_args {
 
 struct prom_t {
 	ihandle root;
-	ihandle chosen;
+	phandle chosen;
 	int cpu;
 	ihandle stdout;
 	ihandle mmumap;
+	ihandle memory;
 };
 
 struct mem_map_entry {
@@ -360,16 +362,36 @@ static void __init prom_printf(const char *format, ...)
 static unsigned int __init prom_claim(unsigned long virt, unsigned long size,
 				unsigned long align)
 {
-	int ret;
 	struct prom_t *_prom = &RELOC(prom);
 
-	ret = call_prom("claim", 3, 1, (prom_arg_t)virt, (prom_arg_t)size,
-			(prom_arg_t)align);
-	if (ret != -1 && _prom->mmumap != 0)
-		/* old pmacs need us to map as well */
+	if (align == 0 && (OF_WORKAROUNDS & OF_WA_CLAIM)) {
+		/*
+		 * Old OF requires we claim physical and virtual separately
+		 * and then map explicitly (assuming virtual mode)
+		 */
+		int ret;
+		prom_arg_t result;
+
+		ret = call_prom_ret("call-method", 5, 2, &result,
+				    ADDR("claim"), _prom->memory,
+				    align, size, virt);
+		if (ret != 0 || result == -1)
+			return -1;
+		ret = call_prom_ret("call-method", 5, 2, &result,
+				    ADDR("claim"), _prom->mmumap,
+				    align, size, virt);
+		if (ret != 0) {
+			call_prom("call-method", 4, 1, ADDR("release"),
+				  _prom->memory, size, virt);
+			return -1;
+		}
+		/* the 0x12 is M (coherence) + PP == read/write */
 		call_prom("call-method", 6, 1,
-			  ADDR("map"), _prom->mmumap, 0, size, virt, virt);
-	return ret;
+			  ADDR("map"), _prom->mmumap, 0x12, size, virt, virt);
+		return virt;
+	}
+	return call_prom("claim", 3, 1, (prom_arg_t)virt, (prom_arg_t)size,
+			 (prom_arg_t)align);
 }
 
 static void __init __attribute__((noreturn)) prom_panic(const char *reason)
@@ -415,11 +437,52 @@ static int inline prom_getproplen(phandle node, const char *pname)
 	return call_prom("getproplen", 2, 1, node, ADDR(pname));
 }
 
-static int inline prom_setprop(phandle node, const char *pname,
-			       void *value, size_t valuelen)
+static void add_string(char **str, const char *q)
 {
-	return call_prom("setprop", 4, 1, node, ADDR(pname),
-			 (u32)(unsigned long) value, (u32) valuelen);
+	char *p = *str;
+
+	while (*q)
+		*p++ = *q++;
+	*p++ = ' ';
+	*str = p;
+}
+
+static char *tohex(unsigned int x)
+{
+	static char digits[] = "0123456789abcdef";
+	static char result[9];
+	int i;
+
+	result[8] = 0;
+	i = 8;
+	do {
+		--i;
+		result[i] = digits[x & 0xf];
+		x >>= 4;
+	} while (x != 0 && i > 0);
+	return &result[i];
+}
+
+static int __init prom_setprop(phandle node, const char *nodename,
+			       const char *pname, void *value, size_t valuelen)
+{
+	char cmd[256], *p;
+
+	if (!(OF_WORKAROUNDS & OF_WA_LONGTRAIL))
+		return call_prom("setprop", 4, 1, node, ADDR(pname),
+				 (u32)(unsigned long) value, (u32) valuelen);
+
+	/* gah... setprop doesn't work on longtrail, have to use interpret */
+	p = cmd;
+	add_string(&p, "dev");
+	add_string(&p, nodename);
+	add_string(&p, tohex((u32)(unsigned long) value));
+	add_string(&p, tohex(valuelen));
+	add_string(&p, tohex(ADDR(pname)));
+	add_string(&p, tohex(strlen(RELOC(pname))));
+	add_string(&p, "property");
+	*p = 0;
+	return call_prom("interpret", 1, 1, (u32)(unsigned long) cmd);
 }
 
 /* We can't use the standard versions because of RELOC headaches. */
@@ -980,7 +1043,7 @@ static void __init prom_instantiate_rtas(void)
 
 	rtas_inst = call_prom("open", 1, 1, ADDR("/rtas"));
 	if (!IHANDLE_VALID(rtas_inst)) {
-		prom_printf("opening rtas package failed");
+		prom_printf("opening rtas package failed (%x)\n", rtas_inst);
 		return;
 	}
 
@@ -988,7 +1051,7 @@ static void __init prom_instantiate_rtas(void)
 
 	if (call_prom_ret("call-method", 3, 2, &entry,
 			  ADDR("instantiate-rtas"),
-			  rtas_inst, base) == PROM_ERROR
+			  rtas_inst, base) != 0
 	    || entry == 0) {
 		prom_printf(" failed\n");
 		return;
@@ -997,8 +1060,10 @@ static void __init prom_instantiate_rtas(void)
 
 	reserve_mem(base, size);
 
-	prom_setprop(rtas_node, "linux,rtas-base", &base, sizeof(base));
-	prom_setprop(rtas_node, "linux,rtas-entry", &entry, sizeof(entry));
+	prom_setprop(rtas_node, "/rtas", "linux,rtas-base",
+		     &base, sizeof(base));
+	prom_setprop(rtas_node, "/rtas", "linux,rtas-entry",
+		     &entry, sizeof(entry));
 
 	prom_debug("rtas base     = 0x%x\n", base);
 	prom_debug("rtas entry    = 0x%x\n", entry);
@@ -1089,10 +1154,6 @@ static void __init prom_initialize_tce_table(void)
 		if (base < local_alloc_bottom)
 			local_alloc_bottom = base;
 
-		/* Save away the TCE table attributes for later use. */
-		prom_setprop(node, "linux,tce-base", &base, sizeof(base));
-		prom_setprop(node, "linux,tce-size", &minsize, sizeof(minsize));
-
 		/* It seems OF doesn't null-terminate the path :-( */
 		memset(path, 0, sizeof(path));
 		/* Call OF to setup the TCE hardware */
@@ -1101,6 +1162,10 @@ static void __init prom_initialize_tce_table(void)
 			prom_printf("package-to-path failed\n");
 		}
 
+		/* Save away the TCE table attributes for later use. */
+		prom_setprop(node, path, "linux,tce-base", &base, sizeof(base));
+		prom_setprop(node, path, "linux,tce-size", &minsize, sizeof(minsize));
+
 		prom_debug("TCE table: %s\n", path);
 		prom_debug("\tnode = 0x%x\n", node);
 		prom_debug("\tbase = 0x%x\n", base);
@@ -1342,6 +1407,7 @@ static void __init prom_init_client_services(unsigned long pp)
 /*
  * For really old powermacs, we need to map things we claim.
  * For that, we need the ihandle of the mmu.
+ * Also, on the longtrail, we need to work around other bugs.
  */
 static void __init prom_find_mmu(void)
 {
@@ -1355,12 +1421,19 @@ static void __init prom_find_mmu(void)
 	if (prom_getprop(oprom, "model", version, sizeof(version)) <= 0)
 		return;
 	version[sizeof(version) - 1] = 0;
-	prom_printf("OF version is '%s'\n", version);
 	/* XXX might need to add other versions here */
-	if (strcmp(version, "Open Firmware, 1.0.5") != 0)
+	if (strcmp(version, "Open Firmware, 1.0.5") == 0)
+		of_workarounds = OF_WA_CLAIM;
+	else if (strncmp(version, "FirmWorks,3.", 12) == 0) {
+		of_workarounds = OF_WA_CLAIM | OF_WA_LONGTRAIL;
+		call_prom("interpret", 1, 1, "dev /memory 0 to allow-reclaim");
+	} else
 		return;
+	_prom->memory = call_prom("open", 1, 1, ADDR("/memory"));
 	prom_getprop(_prom->chosen, "mmu", &_prom->mmumap,
 		     sizeof(_prom->mmumap));
+	if (!IHANDLE_VALID(_prom->memory) || !IHANDLE_VALID(_prom->mmumap))
+		of_workarounds &= ~OF_WA_CLAIM;		/* hmmm */
 }
 #else
 #define prom_find_mmu()
@@ -1382,16 +1455,17 @@ static void __init prom_init_stdout(void)
 	memset(path, 0, 256);
 	call_prom("instance-to-path", 3, 1, _prom->stdout, path, 255);
 	val = call_prom("instance-to-package", 1, 1, _prom->stdout);
-	prom_setprop(_prom->chosen, "linux,stdout-package", &val, sizeof(val));
+	prom_setprop(_prom->chosen, "/chosen", "linux,stdout-package",
+		     &val, sizeof(val));
 	prom_printf("OF stdout device is: %s\n", RELOC(of_stdout_device));
-	prom_setprop(_prom->chosen, "linux,stdout-path",
-		     RELOC(of_stdout_device), strlen(RELOC(of_stdout_device))+1);
+	prom_setprop(_prom->chosen, "/chosen", "linux,stdout-path",
+		     path, strlen(path) + 1);
 
 	/* If it's a display, note it */
 	memset(type, 0, sizeof(type));
 	prom_getprop(val, "device_type", type, sizeof(type));
 	if (strcmp(type, RELOC("display")) == 0)
-		prom_setprop(val, "linux,boot-display", NULL, 0);
+		prom_setprop(val, path, "linux,boot-display", NULL, 0);
 }
 
 static void __init prom_close_stdin(void)
@@ -1514,7 +1588,7 @@ static void __init prom_check_displays(void)
 
 		/* Success */
 		prom_printf("done\n");
-		prom_setprop(node, "linux,opened", NULL, 0);
+		prom_setprop(node, path, "linux,opened", NULL, 0);
 
 		/* Setup a usable color table when the appropriate
 		 * method is available. Should update this to set-colors */
@@ -1884,9 +1958,11 @@ static void __init fixup_device_tree(void)
 	/* interrupt on this revision of u3 is number 0 and level */
 	interrupts[0] = 0;
 	interrupts[1] = 1;
-	prom_setprop(i2c, "interrupts", &interrupts, sizeof(interrupts));
+	prom_setprop(i2c, "/u3@0,f8000000/i2c@f8001000", "interrupts",
+		     &interrupts, sizeof(interrupts));
 	parent = (u32)mpic;
-	prom_setprop(i2c, "interrupt-parent", &parent, sizeof(parent));
+	prom_setprop(i2c, "/u3@0,f8000000/i2c@f8001000", "interrupt-parent",
+		     &parent, sizeof(parent));
 #endif
 }
 
@@ -1922,11 +1998,11 @@ static void __init prom_check_initrd(unsigned long r3, unsigned long r4)
 		RELOC(prom_initrd_end) = RELOC(prom_initrd_start) + r4;
 
 		val = RELOC(prom_initrd_start);
-		prom_setprop(_prom->chosen, "linux,initrd-start", &val,
-			     sizeof(val));
+		prom_setprop(_prom->chosen, "/chosen", "linux,initrd-start",
+			     &val, sizeof(val));
 		val = RELOC(prom_initrd_end);
-		prom_setprop(_prom->chosen, "linux,initrd-end", &val,
-			     sizeof(val));
+		prom_setprop(_prom->chosen, "/chosen", "linux,initrd-end",
+			     &val, sizeof(val));
 
 		reserve_mem(RELOC(prom_initrd_start),
 			    RELOC(prom_initrd_end) - RELOC(prom_initrd_start));
@@ -1969,14 +2045,15 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4,
 	prom_init_client_services(pp);
 
 	/*
-	 * Init prom stdout device
+	 * See if this OF is old enough that we need to do explicit maps
+	 * and other workarounds
 	 */
-	prom_init_stdout();
+	prom_find_mmu();
 
 	/*
-	 * See if this OF is old enough that we need to do explicit maps
+	 * Init prom stdout device
 	 */
-	prom_find_mmu();
+	prom_init_stdout();
 
 	/*
 	 * Check for an initrd
@@ -1989,14 +2066,15 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4,
 	 */
 	RELOC(of_platform) = prom_find_machine_type();
 	getprop_rval = RELOC(of_platform);
-	prom_setprop(_prom->chosen, "linux,platform",
+	prom_setprop(_prom->chosen, "/chosen", "linux,platform",
 		     &getprop_rval, sizeof(getprop_rval));
 
 #ifdef CONFIG_PPC_PSERIES
 	/*
 	 * On pSeries, inform the firmware about our capabilities
 	 */
-	if (RELOC(of_platform) & PLATFORM_PSERIES)
+	if (RELOC(of_platform) == PLATFORM_PSERIES ||
+	    RELOC(of_platform) == PLATFORM_PSERIES_LPAR)
 		prom_send_capabilities();
 #endif
 
@@ -2050,21 +2128,23 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4,
 	 * Fill in some infos for use by the kernel later on
 	 */
 	if (RELOC(prom_memory_limit))
-		prom_setprop(_prom->chosen, "linux,memory-limit",
+		prom_setprop(_prom->chosen, "/chosen", "linux,memory-limit",
 			     &RELOC(prom_memory_limit),
 			     sizeof(prom_memory_limit));
 #ifdef CONFIG_PPC64
 	if (RELOC(ppc64_iommu_off))
-		prom_setprop(_prom->chosen, "linux,iommu-off", NULL, 0);
+		prom_setprop(_prom->chosen, "/chosen", "linux,iommu-off",
+			     NULL, 0);
 
 	if (RELOC(iommu_force_on))
-		prom_setprop(_prom->chosen, "linux,iommu-force-on", NULL, 0);
+		prom_setprop(_prom->chosen, "/chosen", "linux,iommu-force-on",
+			     NULL, 0);
 
 	if (RELOC(prom_tce_alloc_start)) {
-		prom_setprop(_prom->chosen, "linux,tce-alloc-start",
+		prom_setprop(_prom->chosen, "/chosen", "linux,tce-alloc-start",
 			     &RELOC(prom_tce_alloc_start),
 			     sizeof(prom_tce_alloc_start));
-		prom_setprop(_prom->chosen, "linux,tce-alloc-end",
+		prom_setprop(_prom->chosen, "/chosen", "linux,tce-alloc-end",
 			     &RELOC(prom_tce_alloc_end),
 			     sizeof(prom_tce_alloc_end));
 	}
@@ -2081,8 +2161,13 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4,
 	prom_printf("copying OF device tree ...\n");
 	flatten_device_tree();
 
-	/* in case stdin is USB and still active on IBM machines... */
-	prom_close_stdin();
+	/*
+	 * in case stdin is USB and still active on IBM machines...
+	 * Unfortunately quiesce crashes on some powermacs if we have
+	 * closed stdin already (in particular the powerbook 101).
+	 */
+	if (RELOC(of_platform) != PLATFORM_POWERMAC)
+		prom_close_stdin();
 
 	/*
 	 * Call OF "quiesce" method to shut down pending DMA's from
diff --git a/arch/powerpc/kernel/rtas-proc.c b/arch/powerpc/kernel/rtas-proc.c
index 5bdd5b079d9..ae1a36449cc 100644
--- a/arch/powerpc/kernel/rtas-proc.c
+++ b/arch/powerpc/kernel/rtas-proc.c
@@ -259,7 +259,7 @@ static int __init proc_rtas_init(void)
 {
 	struct proc_dir_entry *entry;
 
-	if (!(systemcfg->platform & PLATFORM_PSERIES))
+	if (_machine != PLATFORM_PSERIES && _machine != PLATFORM_PSERIES_LPAR)
 		return 1;
 
 	rtas_node = of_find_node_by_name(NULL, "rtas");
diff --git a/arch/powerpc/kernel/rtas.c b/arch/powerpc/kernel/rtas.c
index 9d4e07f6f1e..4283fa33f78 100644
--- a/arch/powerpc/kernel/rtas.c
+++ b/arch/powerpc/kernel/rtas.c
@@ -29,9 +29,6 @@
 #include <asm/delay.h>
 #include <asm/uaccess.h>
 #include <asm/lmb.h>
-#ifdef CONFIG_PPC64
-#include <asm/systemcfg.h>
-#endif
 
 struct rtas_t rtas = {
 	.lock = SPIN_LOCK_UNLOCKED
@@ -671,7 +668,7 @@ void __init rtas_initialize(void)
 	 * the stop-self token if any
 	 */
 #ifdef CONFIG_PPC64
-	if (systemcfg->platform == PLATFORM_PSERIES_LPAR)
+	if (_machine == PLATFORM_PSERIES_LPAR)
 		rtas_region = min(lmb.rmo_size, RTAS_INSTANTIATE_MAX);
 #endif
 	rtas_rmo_buf = lmb_alloc_base(RTAS_RMOBUF_MAX, PAGE_SIZE, rtas_region);
diff --git a/arch/powerpc/kernel/rtas_pci.c b/arch/powerpc/kernel/rtas_pci.c
new file mode 100644
index 00000000000..0e5a8e11665
--- /dev/null
+++ b/arch/powerpc/kernel/rtas_pci.c
@@ -0,0 +1,513 @@
+/*
+ * arch/ppc64/kernel/rtas_pci.c
+ *
+ * Copyright (C) 2001 Dave Engebretsen, IBM Corporation
+ * Copyright (C) 2003 Anton Blanchard <anton@au.ibm.com>, IBM
+ *
+ * RTAS specific routines for PCI.
+ *
+ * Based on code from pci.c, chrp_pci.c and pSeries_pci.c
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
+ */
+
+#include <linux/kernel.h>
+#include <linux/threads.h>
+#include <linux/pci.h>
+#include <linux/string.h>
+#include <linux/init.h>
+#include <linux/bootmem.h>
+
+#include <asm/io.h>
+#include <asm/pgtable.h>
+#include <asm/irq.h>
+#include <asm/prom.h>
+#include <asm/machdep.h>
+#include <asm/pci-bridge.h>
+#include <asm/iommu.h>
+#include <asm/rtas.h>
+#include <asm/mpic.h>
+#include <asm/ppc-pci.h>
+
+/* RTAS tokens */
+static int read_pci_config;
+static int write_pci_config;
+static int ibm_read_pci_config;
+static int ibm_write_pci_config;
+
+static inline int config_access_valid(struct pci_dn *dn, int where)
+{
+	if (where < 256)
+		return 1;
+	if (where < 4096 && dn->pci_ext_config_space)
+		return 1;
+
+	return 0;
+}
+
+static int of_device_available(struct device_node * dn)
+{
+        char * status;
+
+        status = get_property(dn, "status", NULL);
+
+        if (!status)
+                return 1;
+
+        if (!strcmp(status, "okay"))
+                return 1;
+
+        return 0;
+}
+
+static int rtas_read_config(struct pci_dn *pdn, int where, int size, u32 *val)
+{
+	int returnval = -1;
+	unsigned long buid, addr;
+	int ret;
+
+	if (!pdn)
+		return PCIBIOS_DEVICE_NOT_FOUND;
+	if (!config_access_valid(pdn, where))
+		return PCIBIOS_BAD_REGISTER_NUMBER;
+
+	addr = ((where & 0xf00) << 20) | (pdn->busno << 16) |
+		(pdn->devfn << 8) | (where & 0xff);
+	buid = pdn->phb->buid;
+	if (buid) {
+		ret = rtas_call(ibm_read_pci_config, 4, 2, &returnval,
+				addr, BUID_HI(buid), BUID_LO(buid), size);
+	} else {
+		ret = rtas_call(read_pci_config, 2, 2, &returnval, addr, size);
+	}
+	*val = returnval;
+
+	if (ret)
+		return PCIBIOS_DEVICE_NOT_FOUND;
+
+	if (returnval == EEH_IO_ERROR_VALUE(size) &&
+	    eeh_dn_check_failure (pdn->node, NULL))
+		return PCIBIOS_DEVICE_NOT_FOUND;
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static int rtas_pci_read_config(struct pci_bus *bus,
+				unsigned int devfn,
+				int where, int size, u32 *val)
+{
+	struct device_node *busdn, *dn;
+
+	if (bus->self)
+		busdn = pci_device_to_OF_node(bus->self);
+	else
+		busdn = bus->sysdata;	/* must be a phb */
+
+	/* Search only direct children of the bus */
+	for (dn = busdn->child; dn; dn = dn->sibling) {
+		struct pci_dn *pdn = PCI_DN(dn);
+		if (pdn && pdn->devfn == devfn
+		    && of_device_available(dn))
+			return rtas_read_config(pdn, where, size, val);
+	}
+
+	return PCIBIOS_DEVICE_NOT_FOUND;
+}
+
+int rtas_write_config(struct pci_dn *pdn, int where, int size, u32 val)
+{
+	unsigned long buid, addr;
+	int ret;
+
+	if (!pdn)
+		return PCIBIOS_DEVICE_NOT_FOUND;
+	if (!config_access_valid(pdn, where))
+		return PCIBIOS_BAD_REGISTER_NUMBER;
+
+	addr = ((where & 0xf00) << 20) | (pdn->busno << 16) |
+		(pdn->devfn << 8) | (where & 0xff);
+	buid = pdn->phb->buid;
+	if (buid) {
+		ret = rtas_call(ibm_write_pci_config, 5, 1, NULL, addr,
+			BUID_HI(buid), BUID_LO(buid), size, (ulong) val);
+	} else {
+		ret = rtas_call(write_pci_config, 3, 1, NULL, addr, size, (ulong)val);
+	}
+
+	if (ret)
+		return PCIBIOS_DEVICE_NOT_FOUND;
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static int rtas_pci_write_config(struct pci_bus *bus,
+				 unsigned int devfn,
+				 int where, int size, u32 val)
+{
+	struct device_node *busdn, *dn;
+
+	if (bus->self)
+		busdn = pci_device_to_OF_node(bus->self);
+	else
+		busdn = bus->sysdata;	/* must be a phb */
+
+	/* Search only direct children of the bus */
+	for (dn = busdn->child; dn; dn = dn->sibling) {
+		struct pci_dn *pdn = PCI_DN(dn);
+		if (pdn && pdn->devfn == devfn
+		    && of_device_available(dn))
+			return rtas_write_config(pdn, where, size, val);
+	}
+	return PCIBIOS_DEVICE_NOT_FOUND;
+}
+
+struct pci_ops rtas_pci_ops = {
+	rtas_pci_read_config,
+	rtas_pci_write_config
+};
+
+int is_python(struct device_node *dev)
+{
+	char *model = (char *)get_property(dev, "model", NULL);
+
+	if (model && strstr(model, "Python"))
+		return 1;
+
+	return 0;
+}
+
+static int get_phb_reg_prop(struct device_node *dev,
+			    unsigned int addr_size_words,
+			    struct reg_property64 *reg)
+{
+	unsigned int *ui_ptr = NULL, len;
+
+	/* Found a PHB, now figure out where his registers are mapped. */
+	ui_ptr = (unsigned int *)get_property(dev, "reg", &len);
+	if (ui_ptr == NULL)
+		return 1;
+
+	if (addr_size_words == 1) {
+		reg->address = ((struct reg_property32 *)ui_ptr)->address;
+		reg->size    = ((struct reg_property32 *)ui_ptr)->size;
+	} else {
+		*reg = *((struct reg_property64 *)ui_ptr);
+	}
+
+	return 0;
+}
+
+static void python_countermeasures(struct device_node *dev,
+				   unsigned int addr_size_words)
+{
+	struct reg_property64 reg_struct;
+	void __iomem *chip_regs;
+	volatile u32 val;
+
+	if (get_phb_reg_prop(dev, addr_size_words, &reg_struct))
+		return;
+
+	/* Python's register file is 1 MB in size. */
+	chip_regs = ioremap(reg_struct.address & ~(0xfffffUL), 0x100000);
+
+	/*
+	 * Firmware doesn't always clear this bit which is critical
+	 * for good performance - Anton
+	 */
+
+#define PRG_CL_RESET_VALID 0x00010000
+
+	val = in_be32(chip_regs + 0xf6030);
+	if (val & PRG_CL_RESET_VALID) {
+		printk(KERN_INFO "Python workaround: ");
+		val &= ~PRG_CL_RESET_VALID;
+		out_be32(chip_regs + 0xf6030, val);
+		/*
+		 * We must read it back for changes to
+		 * take effect
+		 */
+		val = in_be32(chip_regs + 0xf6030);
+		printk("reg0: %x\n", val);
+	}
+
+	iounmap(chip_regs);
+}
+
+void __init init_pci_config_tokens (void)
+{
+	read_pci_config = rtas_token("read-pci-config");
+	write_pci_config = rtas_token("write-pci-config");
+	ibm_read_pci_config = rtas_token("ibm,read-pci-config");
+	ibm_write_pci_config = rtas_token("ibm,write-pci-config");
+}
+
+unsigned long __devinit get_phb_buid (struct device_node *phb)
+{
+	int addr_cells;
+	unsigned int *buid_vals;
+	unsigned int len;
+	unsigned long buid;
+
+	if (ibm_read_pci_config == -1) return 0;
+
+	/* PHB's will always be children of the root node,
+	 * or so it is promised by the current firmware. */
+	if (phb->parent == NULL)
+		return 0;
+	if (phb->parent->parent)
+		return 0;
+
+	buid_vals = (unsigned int *) get_property(phb, "reg", &len);
+	if (buid_vals == NULL)
+		return 0;
+
+	addr_cells = prom_n_addr_cells(phb);
+	if (addr_cells == 1) {
+		buid = (unsigned long) buid_vals[0];
+	} else {
+		buid = (((unsigned long)buid_vals[0]) << 32UL) |
+			(((unsigned long)buid_vals[1]) & 0xffffffff);
+	}
+	return buid;
+}
+
+static int phb_set_bus_ranges(struct device_node *dev,
+			      struct pci_controller *phb)
+{
+	int *bus_range;
+	unsigned int len;
+
+	bus_range = (int *) get_property(dev, "bus-range", &len);
+	if (bus_range == NULL || len < 2 * sizeof(int)) {
+		return 1;
+ 	}
+
+	phb->first_busno =  bus_range[0];
+	phb->last_busno  =  bus_range[1];
+
+	return 0;
+}
+
+static int __devinit setup_phb(struct device_node *dev,
+			       struct pci_controller *phb,
+			       unsigned int addr_size_words)
+{
+	pci_setup_pci_controller(phb);
+
+	if (is_python(dev))
+		python_countermeasures(dev, addr_size_words);
+
+	if (phb_set_bus_ranges(dev, phb))
+		return 1;
+
+	phb->arch_data = dev;
+	phb->ops = &rtas_pci_ops;
+	phb->buid = get_phb_buid(dev);
+
+	return 0;
+}
+
+static void __devinit add_linux_pci_domain(struct device_node *dev,
+					   struct pci_controller *phb,
+					   struct property *of_prop)
+{
+	memset(of_prop, 0, sizeof(struct property));
+	of_prop->name = "linux,pci-domain";
+	of_prop->length = sizeof(phb->global_number);
+	of_prop->value = (unsigned char *)&of_prop[1];
+	memcpy(of_prop->value, &phb->global_number, sizeof(phb->global_number));
+	prom_add_property(dev, of_prop);
+}
+
+static struct pci_controller * __init alloc_phb(struct device_node *dev,
+						unsigned int addr_size_words)
+{
+	struct pci_controller *phb;
+	struct property *of_prop;
+
+	phb = alloc_bootmem(sizeof(struct pci_controller));
+	if (phb == NULL)
+		return NULL;
+
+	of_prop = alloc_bootmem(sizeof(struct property) +
+				sizeof(phb->global_number));
+	if (!of_prop)
+		return NULL;
+
+	if (setup_phb(dev, phb, addr_size_words))
+		return NULL;
+
+	add_linux_pci_domain(dev, phb, of_prop);
+
+	return phb;
+}
+
+static struct pci_controller * __devinit alloc_phb_dynamic(struct device_node *dev, unsigned int addr_size_words)
+{
+	struct pci_controller *phb;
+
+	phb = (struct pci_controller *)kmalloc(sizeof(struct pci_controller),
+					       GFP_KERNEL);
+	if (phb == NULL)
+		return NULL;
+
+	if (setup_phb(dev, phb, addr_size_words))
+		return NULL;
+
+	phb->is_dynamic = 1;
+
+	/* TODO: linux,pci-domain? */
+
+ 	return phb;
+}
+
+unsigned long __init find_and_init_phbs(void)
+{
+	struct device_node *node;
+	struct pci_controller *phb;
+	unsigned int root_size_cells = 0;
+	unsigned int index;
+	unsigned int *opprop = NULL;
+	struct device_node *root = of_find_node_by_path("/");
+
+	if (ppc64_interrupt_controller == IC_OPEN_PIC) {
+		opprop = (unsigned int *)get_property(root,
+				"platform-open-pic", NULL);
+	}
+
+	root_size_cells = prom_n_size_cells(root);
+
+	index = 0;
+
+	for (node = of_get_next_child(root, NULL);
+	     node != NULL;
+	     node = of_get_next_child(root, node)) {
+		if (node->type == NULL || strcmp(node->type, "pci") != 0)
+			continue;
+
+		phb = alloc_phb(node, root_size_cells);
+		if (!phb)
+			continue;
+
+		pci_process_bridge_OF_ranges(phb, node, 0);
+		pci_setup_phb_io(phb, index == 0);
+#ifdef CONFIG_PPC_PSERIES
+		if (ppc64_interrupt_controller == IC_OPEN_PIC && pSeries_mpic) {
+			int addr = root_size_cells * (index + 2) - 1;
+			mpic_assign_isu(pSeries_mpic, index, opprop[addr]);
+		}
+#endif
+		index++;
+	}
+
+	of_node_put(root);
+	pci_devs_phb_init();
+
+	/*
+	 * pci_probe_only and pci_assign_all_buses can be set via properties
+	 * in chosen.
+	 */
+	if (of_chosen) {
+		int *prop;
+
+		prop = (int *)get_property(of_chosen, "linux,pci-probe-only",
+					   NULL);
+		if (prop)
+			pci_probe_only = *prop;
+
+		prop = (int *)get_property(of_chosen,
+					   "linux,pci-assign-all-buses", NULL);
+		if (prop)
+			pci_assign_all_buses = *prop;
+	}
+
+	return 0;
+}
+
+struct pci_controller * __devinit init_phb_dynamic(struct device_node *dn)
+{
+	struct device_node *root = of_find_node_by_path("/");
+	unsigned int root_size_cells = 0;
+	struct pci_controller *phb;
+	int primary;
+
+	root_size_cells = prom_n_size_cells(root);
+
+	primary = list_empty(&hose_list);
+	phb = alloc_phb_dynamic(dn, root_size_cells);
+	if (!phb)
+		return NULL;
+
+	pci_process_bridge_OF_ranges(phb, dn, primary);
+
+	pci_setup_phb_io_dynamic(phb, primary);
+	of_node_put(root);
+
+	pci_devs_phb_init_dynamic(phb);
+	scan_phb(phb);
+
+	return phb;
+}
+EXPORT_SYMBOL(init_phb_dynamic);
+
+/* RPA-specific bits for removing PHBs */
+int pcibios_remove_root_bus(struct pci_controller *phb)
+{
+	struct pci_bus *b = phb->bus;
+	struct resource *res;
+	int rc, i;
+
+	res = b->resource[0];
+	if (!res->flags) {
+		printk(KERN_ERR "%s: no IO resource for PHB %s\n", __FUNCTION__,
+				b->name);
+		return 1;
+	}
+
+	rc = unmap_bus_range(b);
+	if (rc) {
+		printk(KERN_ERR "%s: failed to unmap IO on bus %s\n",
+			__FUNCTION__, b->name);
+		return 1;
+	}
+
+	if (release_resource(res)) {
+		printk(KERN_ERR "%s: failed to release IO on bus %s\n",
+				__FUNCTION__, b->name);
+		return 1;
+	}
+
+	for (i = 1; i < 3; ++i) {
+		res = b->resource[i];
+		if (!res->flags && i == 0) {
+			printk(KERN_ERR "%s: no MEM resource for PHB %s\n",
+				__FUNCTION__, b->name);
+			return 1;
+		}
+		if (res->flags && release_resource(res)) {
+			printk(KERN_ERR
+			       "%s: failed to release IO %d on bus %s\n",
+				__FUNCTION__, i, b->name);
+			return 1;
+		}
+	}
+
+	list_del(&phb->list_node);
+	if (phb->is_dynamic)
+		kfree(phb);
+
+	return 0;
+}
+EXPORT_SYMBOL(pcibios_remove_root_bus);
diff --git a/arch/powerpc/kernel/setup-common.c b/arch/powerpc/kernel/setup-common.c
index e22856ecb5a..bae4bff138f 100644
--- a/arch/powerpc/kernel/setup-common.c
+++ b/arch/powerpc/kernel/setup-common.c
@@ -33,6 +33,7 @@
 #include <asm/io.h>
 #include <asm/prom.h>
 #include <asm/processor.h>
+#include <asm/systemcfg.h>
 #include <asm/pgtable.h>
 #include <asm/smp.h>
 #include <asm/elf.h>
@@ -51,6 +52,9 @@
 #include <asm/page.h>
 #include <asm/mmu.h>
 #include <asm/lmb.h>
+#include <asm/xmon.h>
+
+#include "setup.h"
 
 #undef DEBUG
 
@@ -60,6 +64,13 @@
 #define DBG(fmt...)
 #endif
 
+#ifdef CONFIG_PPC_MULTIPLATFORM
+int _machine = 0;
+EXPORT_SYMBOL(_machine);
+#endif
+
+unsigned long klimit = (unsigned long) _end;
+
 /*
  * This still seems to be needed... -- paulus
  */ 
@@ -510,8 +521,8 @@ void __init smp_setup_cpu_maps(void)
 	 * On pSeries LPAR, we need to know how many cpus
 	 * could possibly be added to this partition.
 	 */
-	if (systemcfg->platform == PLATFORM_PSERIES_LPAR &&
-				(dn = of_find_node_by_path("/rtas"))) {
+	if (_machine == PLATFORM_PSERIES_LPAR &&
+	    (dn = of_find_node_by_path("/rtas"))) {
 		int num_addr_cell, num_size_cell, maxcpus;
 		unsigned int *ireg;
 
@@ -555,7 +566,27 @@ void __init smp_setup_cpu_maps(void)
 			cpu_set(cpu ^ 0x1, cpu_sibling_map[cpu]);
 	}
 
-	systemcfg->processorCount = num_present_cpus();
+	_systemcfg->processorCount = num_present_cpus();
 #endif /* CONFIG_PPC64 */
 }
 #endif /* CONFIG_SMP */
+
+#ifdef CONFIG_XMON
+static int __init early_xmon(char *p)
+{
+	/* ensure xmon is enabled */
+	if (p) {
+		if (strncmp(p, "on", 2) == 0)
+			xmon_init(1);
+		if (strncmp(p, "off", 3) == 0)
+			xmon_init(0);
+		if (strncmp(p, "early", 5) != 0)
+			return 0;
+	}
+	xmon_init(1);
+	debugger(NULL);
+
+	return 0;
+}
+early_param("xmon", early_xmon);
+#endif
diff --git a/arch/powerpc/kernel/setup.h b/arch/powerpc/kernel/setup.h
new file mode 100644
index 00000000000..2ebba755272
--- /dev/null
+++ b/arch/powerpc/kernel/setup.h
@@ -0,0 +1,6 @@
+#ifndef _POWERPC_KERNEL_SETUP_H
+#define _POWERPC_KERNEL_SETUP_H
+
+void check_for_initrd(void);
+
+#endif /* _POWERPC_KERNEL_SETUP_H */
diff --git a/arch/powerpc/kernel/setup_32.c b/arch/powerpc/kernel/setup_32.c
index 3af2631e3fa..c98cfcc9cd9 100644
--- a/arch/powerpc/kernel/setup_32.c
+++ b/arch/powerpc/kernel/setup_32.c
@@ -40,6 +40,8 @@
 #include <asm/xmon.h>
 #include <asm/time.h>
 
+#include "setup.h"
+
 #define DBG(fmt...)
 
 #if defined CONFIG_KGDB
@@ -70,8 +72,6 @@ unsigned int DMA_MODE_WRITE;
 int have_of = 1;
 
 #ifdef CONFIG_PPC_MULTIPLATFORM
-int _machine = 0;
-
 extern void prep_init(void);
 extern void pmac_init(void);
 extern void chrp_init(void);
@@ -279,7 +279,6 @@ arch_initcall(ppc_init);
 /* Warning, IO base is not yet inited */
 void __init setup_arch(char **cmdline_p)
 {
-	extern char *klimit;
 	extern void do_init_bootmem(void);
 
 	/* so udelay does something sensible, assume <= 1000 bogomips */
@@ -303,14 +302,9 @@ void __init setup_arch(char **cmdline_p)
 		pmac_feature_init();	/* New cool way */
 #endif
 
-#ifdef CONFIG_XMON
-	xmon_map_scc();
-	if (strstr(cmd_line, "xmon")) {
-		xmon_init(1);
-		debugger(NULL);
-	}
-#endif /* CONFIG_XMON */
-	if ( ppc_md.progress ) ppc_md.progress("setup_arch: enter", 0x3eab);
+#ifdef CONFIG_XMON_DEFAULT
+	xmon_init(1);
+#endif
 
 #if defined(CONFIG_KGDB)
 	if (ppc_md.kgdb_map_scc)
@@ -343,7 +337,7 @@ void __init setup_arch(char **cmdline_p)
 	init_mm.start_code = PAGE_OFFSET;
 	init_mm.end_code = (unsigned long) _etext;
 	init_mm.end_data = (unsigned long) _edata;
-	init_mm.brk = (unsigned long) klimit;
+	init_mm.brk = klimit;
 
 	/* Save unparsed command line copy for /proc/cmdline */
 	strlcpy(saved_command_line, cmd_line, COMMAND_LINE_SIZE);
diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c
index 0471e843b6c..6791668213e 100644
--- a/arch/powerpc/kernel/setup_64.c
+++ b/arch/powerpc/kernel/setup_64.c
@@ -61,6 +61,8 @@
 #include <asm/xmon.h>
 #include <asm/udbg.h>
 
+#include "setup.h"
+
 #ifdef DEBUG
 #define DBG(fmt...) udbg_printf(fmt)
 #else
@@ -94,15 +96,6 @@ extern void udbg_init_maple_realmode(void);
 	do { udbg_putc = call_rtas_display_status_delay; } while(0)
 #endif
 
-/* extern void *stab; */
-extern unsigned long klimit;
-
-extern void mm_init_ppc64(void);
-extern void stab_initialize(unsigned long stab);
-extern void htab_initialize(void);
-extern void early_init_devtree(void *flat_dt);
-extern void unflatten_device_tree(void);
-
 int have_of = 1;
 int boot_cpuid = 0;
 int boot_cpuid_phys = 0;
@@ -254,11 +247,10 @@ void __init early_setup(unsigned long dt_ptr)
 	 * Iterate all ppc_md structures until we find the proper
 	 * one for the current machine type
 	 */
-	DBG("Probing machine type for platform %x...\n",
-	    systemcfg->platform);
+	DBG("Probing machine type for platform %x...\n", _machine);
 
 	for (mach = machines; *mach; mach++) {
-		if ((*mach)->probe(systemcfg->platform))
+		if ((*mach)->probe(_machine))
 			break;
 	}
 	/* What can we do if we didn't find ? */
@@ -290,6 +282,28 @@ void __init early_setup(unsigned long dt_ptr)
 	DBG(" <- early_setup()\n");
 }
 
+#ifdef CONFIG_SMP
+void early_setup_secondary(void)
+{
+	struct paca_struct *lpaca = get_paca();
+
+	/* Mark enabled in PACA */
+	lpaca->proc_enabled = 0;
+
+	/* Initialize hash table for that CPU */
+	htab_initialize_secondary();
+
+	/* Initialize STAB/SLB. We use a virtual address as it works
+	 * in real mode on pSeries and we want a virutal address on
+	 * iSeries anyway
+	 */
+	if (cpu_has_feature(CPU_FTR_SLB))
+		slb_initialize();
+	else
+		stab_initialize(lpaca->stab_addr);
+}
+
+#endif /* CONFIG_SMP */
 
 #if defined(CONFIG_SMP) || defined(CONFIG_KEXEC)
 void smp_release_cpus(void)
@@ -315,7 +329,8 @@ void smp_release_cpus(void)
 #endif /* CONFIG_SMP || CONFIG_KEXEC */
 
 /*
- * Initialize some remaining members of the ppc64_caches and systemcfg structures
+ * Initialize some remaining members of the ppc64_caches and systemcfg
+ * structures
  * (at least until we get rid of them completely). This is mostly some
  * cache informations about the CPU that will be used by cache flush
  * routines and/or provided to userland
@@ -340,7 +355,7 @@ static void __init initialize_cache_info(void)
 			const char *dc, *ic;
 
 			/* Then read cache informations */
-			if (systemcfg->platform == PLATFORM_POWERMAC) {
+			if (_machine == PLATFORM_POWERMAC) {
 				dc = "d-cache-block-size";
 				ic = "i-cache-block-size";
 			} else {
@@ -360,8 +375,8 @@ static void __init initialize_cache_info(void)
 				DBG("Argh, can't find dcache properties ! "
 				    "sizep: %p, lsizep: %p\n", sizep, lsizep);
 
-			systemcfg->dcache_size = ppc64_caches.dsize = size;
-			systemcfg->dcache_line_size =
+			_systemcfg->dcache_size = ppc64_caches.dsize = size;
+			_systemcfg->dcache_line_size =
 				ppc64_caches.dline_size = lsize;
 			ppc64_caches.log_dline_size = __ilog2(lsize);
 			ppc64_caches.dlines_per_page = PAGE_SIZE / lsize;
@@ -378,8 +393,8 @@ static void __init initialize_cache_info(void)
 				DBG("Argh, can't find icache properties ! "
 				    "sizep: %p, lsizep: %p\n", sizep, lsizep);
 
-			systemcfg->icache_size = ppc64_caches.isize = size;
-			systemcfg->icache_line_size =
+			_systemcfg->icache_size = ppc64_caches.isize = size;
+			_systemcfg->icache_line_size =
 				ppc64_caches.iline_size = lsize;
 			ppc64_caches.log_iline_size = __ilog2(lsize);
 			ppc64_caches.ilines_per_page = PAGE_SIZE / lsize;
@@ -387,10 +402,12 @@ static void __init initialize_cache_info(void)
 	}
 
 	/* Add an eye catcher and the systemcfg layout version number */
-	strcpy(systemcfg->eye_catcher, "SYSTEMCFG:PPC64");
-	systemcfg->version.major = SYSTEMCFG_MAJOR;
-	systemcfg->version.minor = SYSTEMCFG_MINOR;
-	systemcfg->processor = mfspr(SPRN_PVR);
+	strcpy(_systemcfg->eye_catcher, "SYSTEMCFG:PPC64");
+	_systemcfg->version.major = SYSTEMCFG_MAJOR;
+	_systemcfg->version.minor = SYSTEMCFG_MINOR;
+	_systemcfg->processor = mfspr(SPRN_PVR);
+	_systemcfg->platform = _machine;
+	_systemcfg->physicalMemorySize = lmb_phys_mem_size();
 
 	DBG(" <- initialize_cache_info()\n");
 }
@@ -479,10 +496,10 @@ void __init setup_system(void)
 	printk("-----------------------------------------------------\n");
 	printk("ppc64_pft_size                = 0x%lx\n", ppc64_pft_size);
 	printk("ppc64_interrupt_controller    = 0x%ld\n", ppc64_interrupt_controller);
-	printk("systemcfg                     = 0x%p\n", systemcfg);
-	printk("systemcfg->platform           = 0x%x\n", systemcfg->platform);
-	printk("systemcfg->processorCount     = 0x%lx\n", systemcfg->processorCount);
-	printk("systemcfg->physicalMemorySize = 0x%lx\n", systemcfg->physicalMemorySize);
+	printk("systemcfg                     = 0x%p\n", _systemcfg);
+	printk("systemcfg->platform           = 0x%x\n", _systemcfg->platform);
+	printk("systemcfg->processorCount     = 0x%lx\n", _systemcfg->processorCount);
+	printk("systemcfg->physicalMemorySize = 0x%lx\n", _systemcfg->physicalMemorySize);
 	printk("ppc64_caches.dcache_line_size = 0x%x\n",
 			ppc64_caches.dline_size);
 	printk("ppc64_caches.icache_line_size = 0x%x\n",
@@ -564,12 +581,12 @@ void __init setup_syscall_map(void)
 	for (i = 0; i < __NR_syscalls; i++) {
 		if (sys_call_table[i*2] != sys_ni_syscall) {
 			count64++;
-			systemcfg->syscall_map_64[i >> 5] |=
+			_systemcfg->syscall_map_64[i >> 5] |=
 				0x80000000UL >> (i & 0x1f);
 		}
 		if (sys_call_table[i*2+1] != sys_ni_syscall) {
 			count32++;
-			systemcfg->syscall_map_32[i >> 5] |=
+			_systemcfg->syscall_map_32[i >> 5] |=
 				0x80000000UL >> (i & 0x1f);
 		}
 	}
@@ -858,26 +875,6 @@ int check_legacy_ioport(unsigned long base_port)
 }
 EXPORT_SYMBOL(check_legacy_ioport);
 
-#ifdef CONFIG_XMON
-static int __init early_xmon(char *p)
-{
-	/* ensure xmon is enabled */
-	if (p) {
-		if (strncmp(p, "on", 2) == 0)
-			xmon_init(1);
-		if (strncmp(p, "off", 3) == 0)
-			xmon_init(0);
-		if (strncmp(p, "early", 5) != 0)
-			return 0;
-	}
-	xmon_init(1);
-	debugger(NULL);
-
-	return 0;
-}
-early_param("xmon", early_xmon);
-#endif
-
 void cpu_die(void)
 {
 	if (ppc_md.cpu_die)
diff --git a/arch/powerpc/kernel/signal_32.c b/arch/powerpc/kernel/signal_32.c
index 081d931eae4..a7c4515f320 100644
--- a/arch/powerpc/kernel/signal_32.c
+++ b/arch/powerpc/kernel/signal_32.c
@@ -42,6 +42,7 @@
 
 #include <asm/uaccess.h>
 #include <asm/cacheflush.h>
+#include <asm/sigcontext.h>
 #ifdef CONFIG_PPC64
 #include "ppc32.h"
 #include <asm/unistd.h>
diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c
index 36d67a8d7cb..e28a139c29d 100644
--- a/arch/powerpc/kernel/smp.c
+++ b/arch/powerpc/kernel/smp.c
@@ -44,6 +44,7 @@
 #include <asm/cputable.h>
 #include <asm/system.h>
 #include <asm/mpic.h>
+#include <asm/systemcfg.h>
 #ifdef CONFIG_PPC64
 #include <asm/paca.h>
 #endif
@@ -368,9 +369,11 @@ int generic_cpu_disable(void)
 	if (cpu == boot_cpuid)
 		return -EBUSY;
 
-	systemcfg->processorCount--;
 	cpu_clear(cpu, cpu_online_map);
+#ifdef CONFIG_PPC64
+	_systemcfg->processorCount--;
 	fixup_irqs(cpu_online_map);
+#endif
 	return 0;
 }
 
@@ -388,9 +391,11 @@ int generic_cpu_enable(unsigned int cpu)
 	while (!cpu_online(cpu))
 		cpu_relax();
 
+#ifdef CONFIG_PPC64
 	fixup_irqs(cpu_online_map);
 	/* counter the irq disable in fixup_irqs */
 	local_irq_enable();
+#endif
 	return 0;
 }
 
@@ -419,7 +424,9 @@ void generic_mach_cpu_die(void)
 	while (__get_cpu_var(cpu_state) != CPU_UP_PREPARE)
 		cpu_relax();
 
+#ifdef CONFIG_PPC64
 	flush_tlb_pending();
+#endif
 	cpu_set(cpu, cpu_online_map);
 	local_irq_enable();
 }
diff --git a/arch/powerpc/kernel/sys_ppc32.c b/arch/powerpc/kernel/sys_ppc32.c
index a8210ed5c68..9c921d1c408 100644
--- a/arch/powerpc/kernel/sys_ppc32.c
+++ b/arch/powerpc/kernel/sys_ppc32.c
@@ -52,7 +52,6 @@
 #include <asm/semaphore.h>
 #include <asm/time.h>
 #include <asm/mmu_context.h>
-#include <asm/systemcfg.h>
 #include <asm/ppc-pci.h>
 
 /* readdir & getdents */
diff --git a/arch/powerpc/kernel/sysfs.c b/arch/powerpc/kernel/sysfs.c
new file mode 100644
index 00000000000..850af198fb5
--- /dev/null
+++ b/arch/powerpc/kernel/sysfs.c
@@ -0,0 +1,384 @@
+#include <linux/config.h>
+#include <linux/sysdev.h>
+#include <linux/cpu.h>
+#include <linux/smp.h>
+#include <linux/percpu.h>
+#include <linux/init.h>
+#include <linux/sched.h>
+#include <linux/module.h>
+#include <linux/nodemask.h>
+#include <linux/cpumask.h>
+#include <linux/notifier.h>
+
+#include <asm/current.h>
+#include <asm/processor.h>
+#include <asm/cputable.h>
+#include <asm/firmware.h>
+#include <asm/hvcall.h>
+#include <asm/prom.h>
+#include <asm/systemcfg.h>
+#include <asm/paca.h>
+#include <asm/lppaca.h>
+#include <asm/machdep.h>
+#include <asm/smp.h>
+
+static DEFINE_PER_CPU(struct cpu, cpu_devices);
+
+/* SMT stuff */
+
+#ifdef CONFIG_PPC_MULTIPLATFORM
+/* default to snooze disabled */
+DEFINE_PER_CPU(unsigned long, smt_snooze_delay);
+
+static ssize_t store_smt_snooze_delay(struct sys_device *dev, const char *buf,
+				      size_t count)
+{
+	struct cpu *cpu = container_of(dev, struct cpu, sysdev);
+	ssize_t ret;
+	unsigned long snooze;
+
+	ret = sscanf(buf, "%lu", &snooze);
+	if (ret != 1)
+		return -EINVAL;
+
+	per_cpu(smt_snooze_delay, cpu->sysdev.id) = snooze;
+
+	return count;
+}
+
+static ssize_t show_smt_snooze_delay(struct sys_device *dev, char *buf)
+{
+	struct cpu *cpu = container_of(dev, struct cpu, sysdev);
+
+	return sprintf(buf, "%lu\n", per_cpu(smt_snooze_delay, cpu->sysdev.id));
+}
+
+static SYSDEV_ATTR(smt_snooze_delay, 0644, show_smt_snooze_delay,
+		   store_smt_snooze_delay);
+
+/* Only parse OF options if the matching cmdline option was not specified */
+static int smt_snooze_cmdline;
+
+static int __init smt_setup(void)
+{
+	struct device_node *options;
+	unsigned int *val;
+	unsigned int cpu;
+
+	if (!cpu_has_feature(CPU_FTR_SMT))
+		return 1;
+
+	options = find_path_device("/options");
+	if (!options)
+		return 1;
+
+	val = (unsigned int *)get_property(options, "ibm,smt-snooze-delay",
+					   NULL);
+	if (!smt_snooze_cmdline && val) {
+		for_each_cpu(cpu)
+			per_cpu(smt_snooze_delay, cpu) = *val;
+	}
+
+	return 1;
+}
+__initcall(smt_setup);
+
+static int __init setup_smt_snooze_delay(char *str)
+{
+	unsigned int cpu;
+	int snooze;
+
+	if (!cpu_has_feature(CPU_FTR_SMT))
+		return 1;
+
+	smt_snooze_cmdline = 1;
+
+	if (get_option(&str, &snooze)) {
+		for_each_cpu(cpu)
+			per_cpu(smt_snooze_delay, cpu) = snooze;
+	}
+
+	return 1;
+}
+__setup("smt-snooze-delay=", setup_smt_snooze_delay);
+
+#endif /* CONFIG_PPC_MULTIPLATFORM */
+
+/*
+ * Enabling PMCs will slow partition context switch times so we only do
+ * it the first time we write to the PMCs.
+ */
+
+static DEFINE_PER_CPU(char, pmcs_enabled);
+
+void ppc64_enable_pmcs(void)
+{
+	/* Only need to enable them once */
+	if (__get_cpu_var(pmcs_enabled))
+		return;
+
+	__get_cpu_var(pmcs_enabled) = 1;
+
+	if (ppc_md.enable_pmcs)
+		ppc_md.enable_pmcs();
+}
+EXPORT_SYMBOL(ppc64_enable_pmcs);
+
+/* XXX convert to rusty's on_one_cpu */
+static unsigned long run_on_cpu(unsigned long cpu,
+			        unsigned long (*func)(unsigned long),
+				unsigned long arg)
+{
+	cpumask_t old_affinity = current->cpus_allowed;
+	unsigned long ret;
+
+	/* should return -EINVAL to userspace */
+	if (set_cpus_allowed(current, cpumask_of_cpu(cpu)))
+		return 0;
+
+	ret = func(arg);
+
+	set_cpus_allowed(current, old_affinity);
+
+	return ret;
+}
+
+#define SYSFS_PMCSETUP(NAME, ADDRESS) \
+static unsigned long read_##NAME(unsigned long junk) \
+{ \
+	return mfspr(ADDRESS); \
+} \
+static unsigned long write_##NAME(unsigned long val) \
+{ \
+	ppc64_enable_pmcs(); \
+	mtspr(ADDRESS, val); \
+	return 0; \
+} \
+static ssize_t show_##NAME(struct sys_device *dev, char *buf) \
+{ \
+	struct cpu *cpu = container_of(dev, struct cpu, sysdev); \
+	unsigned long val = run_on_cpu(cpu->sysdev.id, read_##NAME, 0); \
+	return sprintf(buf, "%lx\n", val); \
+} \
+static ssize_t __attribute_used__ \
+	store_##NAME(struct sys_device *dev, const char *buf, size_t count) \
+{ \
+	struct cpu *cpu = container_of(dev, struct cpu, sysdev); \
+	unsigned long val; \
+	int ret = sscanf(buf, "%lx", &val); \
+	if (ret != 1) \
+		return -EINVAL; \
+	run_on_cpu(cpu->sysdev.id, write_##NAME, val); \
+	return count; \
+}
+
+SYSFS_PMCSETUP(mmcr0, SPRN_MMCR0);
+SYSFS_PMCSETUP(mmcr1, SPRN_MMCR1);
+SYSFS_PMCSETUP(mmcra, SPRN_MMCRA);
+SYSFS_PMCSETUP(pmc1, SPRN_PMC1);
+SYSFS_PMCSETUP(pmc2, SPRN_PMC2);
+SYSFS_PMCSETUP(pmc3, SPRN_PMC3);
+SYSFS_PMCSETUP(pmc4, SPRN_PMC4);
+SYSFS_PMCSETUP(pmc5, SPRN_PMC5);
+SYSFS_PMCSETUP(pmc6, SPRN_PMC6);
+SYSFS_PMCSETUP(pmc7, SPRN_PMC7);
+SYSFS_PMCSETUP(pmc8, SPRN_PMC8);
+SYSFS_PMCSETUP(purr, SPRN_PURR);
+
+static SYSDEV_ATTR(mmcr0, 0600, show_mmcr0, store_mmcr0);
+static SYSDEV_ATTR(mmcr1, 0600, show_mmcr1, store_mmcr1);
+static SYSDEV_ATTR(mmcra, 0600, show_mmcra, store_mmcra);
+static SYSDEV_ATTR(pmc1, 0600, show_pmc1, store_pmc1);
+static SYSDEV_ATTR(pmc2, 0600, show_pmc2, store_pmc2);
+static SYSDEV_ATTR(pmc3, 0600, show_pmc3, store_pmc3);
+static SYSDEV_ATTR(pmc4, 0600, show_pmc4, store_pmc4);
+static SYSDEV_ATTR(pmc5, 0600, show_pmc5, store_pmc5);
+static SYSDEV_ATTR(pmc6, 0600, show_pmc6, store_pmc6);
+static SYSDEV_ATTR(pmc7, 0600, show_pmc7, store_pmc7);
+static SYSDEV_ATTR(pmc8, 0600, show_pmc8, store_pmc8);
+static SYSDEV_ATTR(purr, 0600, show_purr, NULL);
+
+static void register_cpu_online(unsigned int cpu)
+{
+	struct cpu *c = &per_cpu(cpu_devices, cpu);
+	struct sys_device *s = &c->sysdev;
+
+#ifndef CONFIG_PPC_ISERIES
+	if (cpu_has_feature(CPU_FTR_SMT))
+		sysdev_create_file(s, &attr_smt_snooze_delay);
+#endif
+
+	/* PMC stuff */
+
+	sysdev_create_file(s, &attr_mmcr0);
+	sysdev_create_file(s, &attr_mmcr1);
+
+	if (cpu_has_feature(CPU_FTR_MMCRA))
+		sysdev_create_file(s, &attr_mmcra);
+
+	if (cur_cpu_spec->num_pmcs >= 1)
+		sysdev_create_file(s, &attr_pmc1);
+	if (cur_cpu_spec->num_pmcs >= 2)
+		sysdev_create_file(s, &attr_pmc2);
+	if (cur_cpu_spec->num_pmcs >= 3)
+		sysdev_create_file(s, &attr_pmc3);
+	if (cur_cpu_spec->num_pmcs >= 4)
+		sysdev_create_file(s, &attr_pmc4);
+	if (cur_cpu_spec->num_pmcs >= 5)
+		sysdev_create_file(s, &attr_pmc5);
+	if (cur_cpu_spec->num_pmcs >= 6)
+		sysdev_create_file(s, &attr_pmc6);
+	if (cur_cpu_spec->num_pmcs >= 7)
+		sysdev_create_file(s, &attr_pmc7);
+	if (cur_cpu_spec->num_pmcs >= 8)
+		sysdev_create_file(s, &attr_pmc8);
+
+	if (cpu_has_feature(CPU_FTR_SMT))
+		sysdev_create_file(s, &attr_purr);
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+static void unregister_cpu_online(unsigned int cpu)
+{
+	struct cpu *c = &per_cpu(cpu_devices, cpu);
+	struct sys_device *s = &c->sysdev;
+
+	BUG_ON(c->no_control);
+
+#ifndef CONFIG_PPC_ISERIES
+	if (cpu_has_feature(CPU_FTR_SMT))
+		sysdev_remove_file(s, &attr_smt_snooze_delay);
+#endif
+
+	/* PMC stuff */
+
+	sysdev_remove_file(s, &attr_mmcr0);
+	sysdev_remove_file(s, &attr_mmcr1);
+
+	if (cpu_has_feature(CPU_FTR_MMCRA))
+		sysdev_remove_file(s, &attr_mmcra);
+
+	if (cur_cpu_spec->num_pmcs >= 1)
+		sysdev_remove_file(s, &attr_pmc1);
+	if (cur_cpu_spec->num_pmcs >= 2)
+		sysdev_remove_file(s, &attr_pmc2);
+	if (cur_cpu_spec->num_pmcs >= 3)
+		sysdev_remove_file(s, &attr_pmc3);
+	if (cur_cpu_spec->num_pmcs >= 4)
+		sysdev_remove_file(s, &attr_pmc4);
+	if (cur_cpu_spec->num_pmcs >= 5)
+		sysdev_remove_file(s, &attr_pmc5);
+	if (cur_cpu_spec->num_pmcs >= 6)
+		sysdev_remove_file(s, &attr_pmc6);
+	if (cur_cpu_spec->num_pmcs >= 7)
+		sysdev_remove_file(s, &attr_pmc7);
+	if (cur_cpu_spec->num_pmcs >= 8)
+		sysdev_remove_file(s, &attr_pmc8);
+
+	if (cpu_has_feature(CPU_FTR_SMT))
+		sysdev_remove_file(s, &attr_purr);
+}
+#endif /* CONFIG_HOTPLUG_CPU */
+
+static int __devinit sysfs_cpu_notify(struct notifier_block *self,
+				      unsigned long action, void *hcpu)
+{
+	unsigned int cpu = (unsigned int)(long)hcpu;
+
+	switch (action) {
+	case CPU_ONLINE:
+		register_cpu_online(cpu);
+		break;
+#ifdef CONFIG_HOTPLUG_CPU
+	case CPU_DEAD:
+		unregister_cpu_online(cpu);
+		break;
+#endif
+	}
+	return NOTIFY_OK;
+}
+
+static struct notifier_block __devinitdata sysfs_cpu_nb = {
+	.notifier_call	= sysfs_cpu_notify,
+};
+
+/* NUMA stuff */
+
+#ifdef CONFIG_NUMA
+static struct node node_devices[MAX_NUMNODES];
+
+static void register_nodes(void)
+{
+	int i;
+
+	for (i = 0; i < MAX_NUMNODES; i++) {
+		if (node_online(i)) {
+			int p_node = parent_node(i);
+			struct node *parent = NULL;
+
+			if (p_node != i)
+				parent = &node_devices[p_node];
+
+			register_node(&node_devices[i], i, parent);
+		}
+	}
+}
+#else
+static void register_nodes(void)
+{
+	return;
+}
+#endif
+
+/* Only valid if CPU is present. */
+static ssize_t show_physical_id(struct sys_device *dev, char *buf)
+{
+	struct cpu *cpu = container_of(dev, struct cpu, sysdev);
+
+	return sprintf(buf, "%d\n", get_hard_smp_processor_id(cpu->sysdev.id));
+}
+static SYSDEV_ATTR(physical_id, 0444, show_physical_id, NULL);
+
+static int __init topology_init(void)
+{
+	int cpu;
+	struct node *parent = NULL;
+
+	register_nodes();
+
+	register_cpu_notifier(&sysfs_cpu_nb);
+
+	for_each_cpu(cpu) {
+		struct cpu *c = &per_cpu(cpu_devices, cpu);
+
+#ifdef CONFIG_NUMA
+		/* The node to which a cpu belongs can't be known
+		 * until the cpu is made present.
+		 */
+		parent = NULL;
+		if (cpu_present(cpu))
+			parent = &node_devices[cpu_to_node(cpu)];
+#endif
+		/*
+		 * For now, we just see if the system supports making
+		 * the RTAS calls for CPU hotplug.  But, there may be a
+		 * more comprehensive way to do this for an individual
+		 * CPU.  For instance, the boot cpu might never be valid
+		 * for hotplugging.
+		 */
+		if (!ppc_md.cpu_die)
+			c->no_control = 1;
+
+		if (cpu_online(cpu) || (c->no_control == 0)) {
+			register_cpu(c, cpu, parent);
+
+			sysdev_create_file(&c->sysdev, &attr_physical_id);
+		}
+
+		if (cpu_online(cpu))
+			register_cpu_online(cpu);
+	}
+
+	return 0;
+}
+__initcall(topology_init);
diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index a6282b625b4..260b6ecd26a 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -271,13 +271,13 @@ static inline void update_gtod(u64 new_tb_stamp, u64 new_stamp_xsec,
 	 * tb_to_xs and stamp_xsec values are consistent.  If not, then it
 	 * loops back and reads them again until this criteria is met.
 	 */
-	++(systemcfg->tb_update_count);
+	++(_systemcfg->tb_update_count);
 	smp_wmb();
-	systemcfg->tb_orig_stamp = new_tb_stamp;
-	systemcfg->stamp_xsec = new_stamp_xsec;
-	systemcfg->tb_to_xs = new_tb_to_xs;
+	_systemcfg->tb_orig_stamp = new_tb_stamp;
+	_systemcfg->stamp_xsec = new_stamp_xsec;
+	_systemcfg->tb_to_xs = new_tb_to_xs;
 	smp_wmb();
-	++(systemcfg->tb_update_count);
+	++(_systemcfg->tb_update_count);
 #endif
 }
 
@@ -357,8 +357,9 @@ static void iSeries_tb_recal(void)
 				do_gtod.tb_ticks_per_sec = tb_ticks_per_sec;
 				tb_to_xs = divres.result_low;
 				do_gtod.varp->tb_to_xs = tb_to_xs;
-				systemcfg->tb_ticks_per_sec = tb_ticks_per_sec;
-				systemcfg->tb_to_xs = tb_to_xs;
+				_systemcfg->tb_ticks_per_sec =
+					tb_ticks_per_sec;
+				_systemcfg->tb_to_xs = tb_to_xs;
 			}
 			else {
 				printk( "Titan recalibrate: FAILED (difference > 4 percent)\n"
@@ -483,6 +484,8 @@ void __init smp_space_timers(unsigned int max_cpus)
 	unsigned long offset = tb_ticks_per_jiffy / max_cpus;
 	unsigned long previous_tb = per_cpu(last_jiffy, boot_cpuid);
 
+	/* make sure tb > per_cpu(last_jiffy, cpu) for all cpus always */
+	previous_tb -= tb_ticks_per_jiffy;
 	for_each_cpu(i) {
 		if (i != boot_cpuid) {
 			previous_tb += offset;
@@ -559,8 +562,8 @@ int do_settimeofday(struct timespec *tv)
 	update_gtod(tb_last_jiffy, new_xsec, do_gtod.varp->tb_to_xs);
 
 #ifdef CONFIG_PPC64
-	systemcfg->tz_minuteswest = sys_tz.tz_minuteswest;
-	systemcfg->tz_dsttime = sys_tz.tz_dsttime;
+	_systemcfg->tz_minuteswest = sys_tz.tz_minuteswest;
+	_systemcfg->tz_dsttime = sys_tz.tz_dsttime;
 #endif
 
 	write_sequnlock_irqrestore(&xtime_lock, flags);
@@ -711,11 +714,11 @@ void __init time_init(void)
 	do_gtod.varp->tb_to_xs = tb_to_xs;
 	do_gtod.tb_to_us = tb_to_us;
 #ifdef CONFIG_PPC64
-	systemcfg->tb_orig_stamp = tb_last_jiffy;
-	systemcfg->tb_update_count = 0;
-	systemcfg->tb_ticks_per_sec = tb_ticks_per_sec;
-	systemcfg->stamp_xsec = xtime.tv_sec * XSEC_PER_SEC;
-	systemcfg->tb_to_xs = tb_to_xs;
+	_systemcfg->tb_orig_stamp = tb_last_jiffy;
+	_systemcfg->tb_update_count = 0;
+	_systemcfg->tb_ticks_per_sec = tb_ticks_per_sec;
+	_systemcfg->stamp_xsec = xtime.tv_sec * XSEC_PER_SEC;
+	_systemcfg->tb_to_xs = tb_to_xs;
 #endif
 
 	time_freq = 0;
diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c
index 0578f838760..2020bb7648f 100644
--- a/arch/powerpc/kernel/traps.c
+++ b/arch/powerpc/kernel/traps.c
@@ -129,7 +129,7 @@ int die(const char *str, struct pt_regs *regs, long err)
 	nl = 1;
 #endif
 #ifdef CONFIG_PPC64
-	switch (systemcfg->platform) {
+	switch (_machine) {
 	case PLATFORM_PSERIES:
 		printk("PSERIES ");
 		nl = 1;
diff --git a/arch/powerpc/lib/bitops.c b/arch/powerpc/lib/bitops.c
index b67ce3004eb..f68ad71a018 100644
--- a/arch/powerpc/lib/bitops.c
+++ b/arch/powerpc/lib/bitops.c
@@ -41,7 +41,7 @@ unsigned long find_next_bit(const unsigned long *addr, unsigned long size,
 	tmp = *p;
 
 found_first:
-	tmp &= (~0UL >> (64 - size));
+	tmp &= (~0UL >> (BITS_PER_LONG - size));
 	if (tmp == 0UL)		/* Are any bits set? */
 		return result + size;	/* Nope. */
 found_middle:
diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c
index 22e47487613..706e8a63ced 100644
--- a/arch/powerpc/mm/hash_utils_64.c
+++ b/arch/powerpc/mm/hash_utils_64.c
@@ -84,10 +84,11 @@
 extern unsigned long dart_tablebase;
 #endif /* CONFIG_U3_DART */
 
+static unsigned long _SDR1;
+struct mmu_psize_def mmu_psize_defs[MMU_PAGE_COUNT];
+
 hpte_t *htab_address;
 unsigned long htab_hash_mask;
-unsigned long _SDR1;
-struct mmu_psize_def mmu_psize_defs[MMU_PAGE_COUNT];
 int mmu_linear_psize = MMU_PAGE_4K;
 int mmu_virtual_psize = MMU_PAGE_4K;
 #ifdef CONFIG_HUGETLB_PAGE
@@ -165,7 +166,7 @@ int htab_bolt_mapping(unsigned long vstart, unsigned long vend,
 		 * normal insert callback here.
 		 */
 #ifdef CONFIG_PPC_ISERIES
-		if (systemcfg->platform == PLATFORM_ISERIES_LPAR)
+		if (_machine == PLATFORM_ISERIES_LPAR)
 			ret = iSeries_hpte_insert(hpteg, va,
 						  virt_to_abs(paddr),
 						  tmp_mode,
@@ -174,7 +175,7 @@ int htab_bolt_mapping(unsigned long vstart, unsigned long vend,
 		else
 #endif
 #ifdef CONFIG_PPC_PSERIES
-		if (systemcfg->platform & PLATFORM_LPAR)
+		if (_machine & PLATFORM_LPAR)
 			ret = pSeries_lpar_hpte_insert(hpteg, va,
 						       virt_to_abs(paddr),
 						       tmp_mode,
@@ -293,7 +294,7 @@ static void __init htab_init_page_sizes(void)
 	 * Not in the device-tree, let's fallback on known size
 	 * list for 16M capable GP & GR
 	 */
-	if ((systemcfg->platform != PLATFORM_ISERIES_LPAR) &&
+	if ((_machine != PLATFORM_ISERIES_LPAR) &&
 	    cpu_has_feature(CPU_FTR_16M_PAGE))
 		memcpy(mmu_psize_defs, mmu_psize_defaults_gp,
 		       sizeof(mmu_psize_defaults_gp));
@@ -364,7 +365,7 @@ static int __init htab_dt_scan_pftsize(unsigned long node,
 
 static unsigned long __init htab_get_table_size(void)
 {
-	unsigned long rnd_mem_size, pteg_count;
+	unsigned long mem_size, rnd_mem_size, pteg_count;
 
 	/* If hash size isn't already provided by the platform, we try to
 	 * retreive it from the device-tree. If it's not there neither, we
@@ -376,8 +377,9 @@ static unsigned long __init htab_get_table_size(void)
 		return 1UL << ppc64_pft_size;
 
 	/* round mem_size up to next power of 2 */
-	rnd_mem_size = 1UL << __ilog2(systemcfg->physicalMemorySize);
-	if (rnd_mem_size < systemcfg->physicalMemorySize)
+	mem_size = lmb_phys_mem_size();
+	rnd_mem_size = 1UL << __ilog2(mem_size);
+	if (rnd_mem_size < mem_size)
 		rnd_mem_size <<= 1;
 
 	/* # pages / 2 */
@@ -386,6 +388,15 @@ static unsigned long __init htab_get_table_size(void)
 	return pteg_count << 7;
 }
 
+#ifdef CONFIG_MEMORY_HOTPLUG
+void create_section_mapping(unsigned long start, unsigned long end)
+{
+		BUG_ON(htab_bolt_mapping(start, end, start,
+			_PAGE_ACCESSED | _PAGE_DIRTY | _PAGE_COHERENT | PP_RWXX,
+			mmu_linear_psize));
+}
+#endif /* CONFIG_MEMORY_HOTPLUG */
+
 void __init htab_initialize(void)
 {
 	unsigned long table, htab_size_bytes;
@@ -410,7 +421,7 @@ void __init htab_initialize(void)
 
 	htab_hash_mask = pteg_count - 1;
 
-	if (systemcfg->platform & PLATFORM_LPAR) {
+	if (platform_is_lpar()) {
 		/* Using a hypervisor which owns the htab */
 		htab_address = NULL;
 		_SDR1 = 0; 
@@ -431,6 +442,9 @@ void __init htab_initialize(void)
 
 		/* Initialize the HPT with no entries */
 		memset((void *)table, 0, htab_size_bytes);
+
+		/* Set SDR1 */
+		mtspr(SPRN_SDR1, _SDR1);
 	}
 
 	mode_rw = _PAGE_ACCESSED | _PAGE_DIRTY | _PAGE_COHERENT | PP_RWXX;
@@ -500,6 +514,12 @@ void __init htab_initialize(void)
 #undef KB
 #undef MB
 
+void __init htab_initialize_secondary(void)
+{
+	if (!platform_is_lpar())
+		mtspr(SPRN_SDR1, _SDR1);
+}
+
 /*
  * Called by asm hashtable.S for doing lazy icache flush
  */
diff --git a/arch/powerpc/mm/init_32.c b/arch/powerpc/mm/init_32.c
index 4612a79dfb6..7d4b8b5f060 100644
--- a/arch/powerpc/mm/init_32.c
+++ b/arch/powerpc/mm/init_32.c
@@ -84,9 +84,6 @@ void MMU_init(void);
 /* XXX should be in current.h  -- paulus */
 extern struct task_struct *current_set[NR_CPUS];
 
-char *klimit = _end;
-struct device_node *memory_node;
-
 extern int init_bootmem_done;
 
 /*
diff --git a/arch/powerpc/mm/init_64.c b/arch/powerpc/mm/init_64.c
index ce974c83d88..1134f70f231 100644
--- a/arch/powerpc/mm/init_64.c
+++ b/arch/powerpc/mm/init_64.c
@@ -20,6 +20,8 @@
  *
  */
 
+#undef DEBUG
+
 #include <linux/config.h>
 #include <linux/signal.h>
 #include <linux/sched.h>
@@ -64,6 +66,12 @@
 #include <asm/vdso.h>
 #include <asm/imalloc.h>
 
+#ifdef DEBUG
+#define DBG(fmt...) printk(fmt)
+#else
+#define DBG(fmt...)
+#endif
+
 #if PGTABLE_RANGE > USER_VSID_RANGE
 #warning Limited user VSID range means pagetable space is wasted
 #endif
@@ -72,8 +80,6 @@
 #warning TASK_SIZE is smaller than it needs to be.
 #endif
 
-unsigned long klimit = (unsigned long)_end;
-
 /* max amount of RAM to use */
 unsigned long __max_memory;
 
@@ -188,14 +194,14 @@ static void zero_ctor(void *addr, kmem_cache_t *cache, unsigned long flags)
 }
 
 #ifdef CONFIG_PPC_64K_PAGES
-static const int pgtable_cache_size[2] = {
-	PTE_TABLE_SIZE, PGD_TABLE_SIZE
+static const unsigned int pgtable_cache_size[3] = {
+	PTE_TABLE_SIZE, PMD_TABLE_SIZE, PGD_TABLE_SIZE
 };
 static const char *pgtable_cache_name[ARRAY_SIZE(pgtable_cache_size)] = {
-	"pte_pmd_cache", "pgd_cache",
+	"pte_pmd_cache", "pmd_cache", "pgd_cache",
 };
 #else
-static const int pgtable_cache_size[2] = {
+static const unsigned int pgtable_cache_size[2] = {
 	PTE_TABLE_SIZE, PMD_TABLE_SIZE
 };
 static const char *pgtable_cache_name[ARRAY_SIZE(pgtable_cache_size)] = {
@@ -213,6 +219,8 @@ void pgtable_cache_init(void)
 		int size = pgtable_cache_size[i];
 		const char *name = pgtable_cache_name[i];
 
+		DBG("Allocating page table cache %s (#%d) "
+		    "for size: %08x...\n", name, i, size);
 		pgtable_cache[i] = kmem_cache_create(name,
 						     size, size,
 						     SLAB_HWCACHE_ALIGN |
diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c
index 6f55efd9be9..1dd3cc69a49 100644
--- a/arch/powerpc/mm/mem.c
+++ b/arch/powerpc/mm/mem.c
@@ -110,6 +110,7 @@ EXPORT_SYMBOL(phys_mem_access_prot);
 void online_page(struct page *page)
 {
 	ClearPageReserved(page);
+	set_page_count(page, 0);
 	free_cold_page(page);
 	totalram_pages++;
 	num_physpages++;
@@ -127,6 +128,9 @@ int __devinit add_memory(u64 start, u64 size)
 	unsigned long start_pfn = start >> PAGE_SHIFT;
 	unsigned long nr_pages = size >> PAGE_SHIFT;
 
+	start += KERNELBASE;
+	create_section_mapping(start, start + size);
+
 	/* this should work for most non-highmem platforms */
 	zone = pgdata->node_zones;
 
diff --git a/arch/powerpc/mm/pgtable_64.c b/arch/powerpc/mm/pgtable_64.c
index 900842451bd..c7f7bb6f30b 100644
--- a/arch/powerpc/mm/pgtable_64.c
+++ b/arch/powerpc/mm/pgtable_64.c
@@ -122,8 +122,11 @@ static int map_io_page(unsigned long ea, unsigned long pa, int flags)
 		 *
 		 */
 		if (htab_bolt_mapping(ea, ea + PAGE_SIZE, pa, flags,
-				      mmu_virtual_psize))
-			panic("Can't map bolted IO mapping");
+				      mmu_virtual_psize)) {
+			printk(KERN_ERR "Failed to do bolted mapping IO "
+			       "memory at %016lx !\n", pa);
+			return -ENOMEM;
+		}
 	}
 	return 0;
 }
diff --git a/arch/powerpc/mm/stab.c b/arch/powerpc/mm/stab.c
index fa325dbf98f..cfbb4e1f966 100644
--- a/arch/powerpc/mm/stab.c
+++ b/arch/powerpc/mm/stab.c
@@ -20,6 +20,7 @@
 #include <asm/cputable.h>
 #include <asm/lmb.h>
 #include <asm/abs_addr.h>
+#include <asm/firmware.h>
 
 struct stab_entry {
 	unsigned long esid_data;
@@ -256,7 +257,7 @@ void stabs_alloc(void)
 
 		paca[cpu].stab_addr = newstab;
 		paca[cpu].stab_real = virt_to_abs(newstab);
-		printk(KERN_DEBUG "Segment table for CPU %d at 0x%lx "
+		printk(KERN_INFO "Segment table for CPU %d at 0x%lx "
 		       "virtual, 0x%lx absolute\n",
 		       cpu, paca[cpu].stab_addr, paca[cpu].stab_real);
 	}
@@ -270,10 +271,28 @@ void stabs_alloc(void)
 void stab_initialize(unsigned long stab)
 {
 	unsigned long vsid = get_kernel_vsid(KERNELBASE);
+	unsigned long stabreal;
 
 	asm volatile("isync; slbia; isync":::"memory");
 	make_ste(stab, GET_ESID(KERNELBASE), vsid);
 
 	/* Order update */
 	asm volatile("sync":::"memory");
+
+	/* Set ASR */
+	stabreal = get_paca()->stab_real | 0x1ul;
+
+#ifdef CONFIG_PPC_ISERIES
+	if (firmware_has_feature(FW_FEATURE_ISERIES)) {
+		HvCall1(HvCallBaseSetASR, stabreal);
+		return;
+	}
+#endif /* CONFIG_PPC_ISERIES */
+#ifdef CONFIG_PPC_PSERIES
+	if (platform_is_lpar()) {
+		plpar_hcall_norets(H_SET_ASR, stabreal);
+		return;
+	}
+#endif
+	mtspr(SPRN_ASR, stabreal);
 }
diff --git a/arch/powerpc/oprofile/op_model_power4.c b/arch/powerpc/oprofile/op_model_power4.c
index c4ee5478427..e3a024e324b 100644
--- a/arch/powerpc/oprofile/op_model_power4.c
+++ b/arch/powerpc/oprofile/op_model_power4.c
@@ -233,8 +233,7 @@ static unsigned long get_pc(struct pt_regs *regs)
 	mmcra = mfspr(SPRN_MMCRA);
 
 	/* Were we in the hypervisor? */
-	if ((systemcfg->platform == PLATFORM_PSERIES_LPAR) &&
-	    (mmcra & MMCRA_SIHV))
+	if (platform_is_lpar() && (mmcra & MMCRA_SIHV))
 		/* function descriptor madness */
 		return *((unsigned long *)hypervisor_bucket);
 
diff --git a/arch/powerpc/platforms/chrp/setup.c b/arch/powerpc/platforms/chrp/setup.c
index ecd32d5d85f..4099ddab920 100644
--- a/arch/powerpc/platforms/chrp/setup.c
+++ b/arch/powerpc/platforms/chrp/setup.c
@@ -361,7 +361,9 @@ static void __init chrp_find_openpic(void)
 	printk(KERN_INFO "OpenPIC at %lx\n", opaddr);
 
 	irq_count = NR_IRQS - NUM_ISA_INTERRUPTS - 4; /* leave room for IPIs */
-	prom_get_irq_senses(init_senses, NUM_8259_INTERRUPTS, NR_IRQS - 4);
+	prom_get_irq_senses(init_senses, NUM_ISA_INTERRUPTS, NR_IRQS - 4);
+	/* i8259 cascade is always positive level */
+	init_senses[0] = IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE;
 
 	iranges = (unsigned int *) get_property(np, "interrupt-ranges", &len);
 	if (iranges == NULL)
diff --git a/arch/powerpc/platforms/iseries/irq.c b/arch/powerpc/platforms/iseries/irq.c
index a06603d84a4..01090e9ce0c 100644
--- a/arch/powerpc/platforms/iseries/irq.c
+++ b/arch/powerpc/platforms/iseries/irq.c
@@ -103,6 +103,9 @@ static void intReceived(struct XmPciLpEvent *eventParm,
 		struct pt_regs *regsParm)
 {
 	int irq;
+#ifdef CONFIG_IRQSTACKS
+	struct thread_info *curtp, *irqtp;
+#endif
 
 	++Pci_Interrupt_Count;
 
@@ -110,7 +113,20 @@ static void intReceived(struct XmPciLpEvent *eventParm,
 	case XmPciLpEvent_SlotInterrupt:
 		irq = eventParm->hvLpEvent.xCorrelationToken;
 		/* Dispatch the interrupt handlers for this irq */
-		ppc_irq_dispatch_handler(regsParm, irq);
+#ifdef CONFIG_IRQSTACKS
+		/* Switch to the irq stack to handle this */
+		curtp = current_thread_info();
+		irqtp = hardirq_ctx[smp_processor_id()];
+		if (curtp != irqtp) {
+			irqtp->task = curtp->task;
+			irqtp->flags = 0;
+			call___do_IRQ(irq, regsParm, irqtp);
+			irqtp->task = NULL;
+			if (irqtp->flags)
+				set_bits(irqtp->flags, &curtp->flags);
+		} else
+#endif
+			__do_IRQ(irq, regsParm);
 		HvCallPci_eoi(eventParm->eventData.slotInterrupt.busNumber,
 			eventParm->eventData.slotInterrupt.subBusNumber,
 			eventParm->eventData.slotInterrupt.deviceId);
@@ -310,10 +326,8 @@ static void iSeries_disable_IRQ(unsigned int irq)
 }
 
 /*
- * Need to define this so ppc_irq_dispatch_handler will NOT call
- * enable_IRQ at the end of interrupt handling.  However, this does
- * nothing because there is not enough information provided to do
- * the EOI HvCall.  This is done by XmPciLpEvent.c
+ * This does nothing because there is not enough information
+ * provided to do the EOI HvCall.  This is done by XmPciLpEvent.c
  */
 static void iSeries_end_IRQ(unsigned int irq)
 {
diff --git a/arch/powerpc/platforms/iseries/misc.S b/arch/powerpc/platforms/iseries/misc.S
index 09f14522e17..dfe7aa1ba09 100644
--- a/arch/powerpc/platforms/iseries/misc.S
+++ b/arch/powerpc/platforms/iseries/misc.S
@@ -15,6 +15,7 @@
 
 #include <asm/processor.h>
 #include <asm/asm-offsets.h>
+#include <asm/ppc_asm.h>
 
 	.text
 
diff --git a/arch/powerpc/platforms/iseries/setup.c b/arch/powerpc/platforms/iseries/setup.c
index 7f8f0cda6a7..6a29f301436 100644
--- a/arch/powerpc/platforms/iseries/setup.c
+++ b/arch/powerpc/platforms/iseries/setup.c
@@ -39,7 +39,8 @@
 #include <asm/sections.h>
 #include <asm/iommu.h>
 #include <asm/firmware.h>
-
+#include <asm/systemcfg.h>
+#include <asm/system.h>
 #include <asm/time.h>
 #include <asm/paca.h>
 #include <asm/cache.h>
@@ -71,7 +72,7 @@ extern void hvlog(char *fmt, ...);
 #endif
 
 /* Function Prototypes */
-static void build_iSeries_Memory_Map(void);
+static unsigned long build_iSeries_Memory_Map(void);
 static void iseries_shared_idle(void);
 static void iseries_dedicated_idle(void);
 #ifdef CONFIG_PCI
@@ -84,7 +85,6 @@ static void iSeries_pci_final_fixup(void) { }
 int piranha_simulator;
 
 extern int rd_size;		/* Defined in drivers/block/rd.c */
-extern unsigned long klimit;
 extern unsigned long embedded_sysmap_start;
 extern unsigned long embedded_sysmap_end;
 
@@ -403,9 +403,11 @@ void mschunks_alloc(unsigned long num_chunks)
  * a table used to translate Linux's physical addresses to these
  * absolute addresses.  Absolute addresses are needed when
  * communicating with the hypervisor (e.g. to build HPT entries)
+ *
+ * Returns the physical memory size
  */
 
-static void __init build_iSeries_Memory_Map(void)
+static unsigned long __init build_iSeries_Memory_Map(void)
 {
 	u32 loadAreaFirstChunk, loadAreaLastChunk, loadAreaSize;
 	u32 nextPhysChunk;
@@ -538,7 +540,7 @@ static void __init build_iSeries_Memory_Map(void)
 	 * which should be equal to
 	 *   nextPhysChunk
 	 */
-	systemcfg->physicalMemorySize = chunk_to_addr(nextPhysChunk);
+	return chunk_to_addr(nextPhysChunk);
 }
 
 /*
@@ -564,8 +566,8 @@ static void __init iSeries_setup_arch(void)
 	printk("Max physical processors = %d\n",
 			itVpdAreas.xSlicMaxPhysicalProcs);
 
-	systemcfg->processor = xIoHriProcessorVpd[procIx].xPVR;
-	printk("Processor version = %x\n", systemcfg->processor);
+	_systemcfg->processor = xIoHriProcessorVpd[procIx].xPVR;
+	printk("Processor version = %x\n", _systemcfg->processor);
 }
 
 static void iSeries_show_cpuinfo(struct seq_file *m)
@@ -702,7 +704,6 @@ static void iseries_shared_idle(void)
 
 static void iseries_dedicated_idle(void)
 {
-	long oldval;
 	set_thread_flag(TIF_POLLING_NRFLAG);
 
 	while (1) {
@@ -929,7 +930,7 @@ void dt_cpus(struct iseries_flat_dt *dt)
 	dt_end_node(dt);
 }
 
-void build_flat_dt(struct iseries_flat_dt *dt)
+void build_flat_dt(struct iseries_flat_dt *dt, unsigned long phys_mem_size)
 {
 	u64 tmp[2];
 
@@ -945,7 +946,7 @@ void build_flat_dt(struct iseries_flat_dt *dt)
 	dt_prop_str(dt, "name", "memory");
 	dt_prop_str(dt, "device_type", "memory");
 	tmp[0] = 0;
-	tmp[1] = systemcfg->physicalMemorySize;
+	tmp[1] = phys_mem_size;
 	dt_prop_u64_list(dt, "reg", tmp, 2);
 	dt_end_node(dt);
 
@@ -965,13 +966,15 @@ void build_flat_dt(struct iseries_flat_dt *dt)
 
 void * __init iSeries_early_setup(void)
 {
+	unsigned long phys_mem_size;
+
 	iSeries_fixup_klimit();
 
 	/*
 	 * Initialize the table which translate Linux physical addresses to
 	 * AS/400 absolute addresses
 	 */
-	build_iSeries_Memory_Map();
+	phys_mem_size = build_iSeries_Memory_Map();
 
 	iSeries_get_cmdline();
 
@@ -981,7 +984,7 @@ void * __init iSeries_early_setup(void)
 	/* Parse early parameters, in particular mem=x */
 	parse_early_param();
 
-	build_flat_dt(&iseries_dt);
+	build_flat_dt(&iseries_dt, phys_mem_size);
 
 	return (void *) __pa(&iseries_dt);
 }
diff --git a/arch/powerpc/platforms/maple/pci.c b/arch/powerpc/platforms/maple/pci.c
index 340c21caeae..895aeb3f75d 100644
--- a/arch/powerpc/platforms/maple/pci.c
+++ b/arch/powerpc/platforms/maple/pci.c
@@ -380,9 +380,6 @@ void __init maple_pcibios_fixup(void)
 	for_each_pci_dev(dev)
 		pci_read_irq_line(dev);
 
-	/* Do the mapping of the IO space */
-	phbs_remap_io();
-
 	DBG(" <- maple_pcibios_fixup\n");
 }
 
diff --git a/arch/powerpc/platforms/powermac/pci.c b/arch/powerpc/platforms/powermac/pci.c
index 8f818d092e2..dfd41b9781a 100644
--- a/arch/powerpc/platforms/powermac/pci.c
+++ b/arch/powerpc/platforms/powermac/pci.c
@@ -918,9 +918,6 @@ void __init pmac_pci_init(void)
 			PCI_DN(np)->busno = 0xf0;
 	}
 
-	/* map in PCI I/O space */
-	phbs_remap_io();
-
 	/* pmac_check_ht_link(); */
 
 	/* Tell pci.c to not use the common resource allocation mechanism */
diff --git a/arch/powerpc/platforms/powermac/pic.c b/arch/powerpc/platforms/powermac/pic.c
index 83a49e80ac2..90040c49494 100644
--- a/arch/powerpc/platforms/powermac/pic.c
+++ b/arch/powerpc/platforms/powermac/pic.c
@@ -74,6 +74,9 @@ static DEFINE_SPINLOCK(pmac_pic_lock);
 #define GATWICK_IRQ_POOL_SIZE        10
 static struct interrupt_info gatwick_int_pool[GATWICK_IRQ_POOL_SIZE];
 
+#define NR_MASK_WORDS	((NR_IRQS + 31) / 32)
+static unsigned long ppc_lost_interrupts[NR_MASK_WORDS];
+
 /*
  * Mark an irq as "lost".  This is only used on the pmac
  * since it can lose interrupts (see pmac_set_irq_mask).
diff --git a/arch/powerpc/platforms/powermac/smp.c b/arch/powerpc/platforms/powermac/smp.c
index e1f9443cc87..957b0910342 100644
--- a/arch/powerpc/platforms/powermac/smp.c
+++ b/arch/powerpc/platforms/powermac/smp.c
@@ -305,9 +305,19 @@ static int __init smp_psurge_probe(void)
 	psurge_start = ioremap(PSURGE_START, 4);
 	psurge_pri_intr = ioremap(PSURGE_PRI_INTR, 4);
 
-	/* this is not actually strictly necessary -- paulus. */
-	for (i = 1; i < ncpus; ++i)
-		smp_hw_index[i] = i;
+	/*
+	 * This is necessary because OF doesn't know about the
+	 * secondary cpu(s), and thus there aren't nodes in the
+	 * device tree for them, and smp_setup_cpu_maps hasn't
+	 * set their bits in cpu_possible_map and cpu_present_map.
+	 */
+	if (ncpus > NR_CPUS)
+		ncpus = NR_CPUS;
+	for (i = 1; i < ncpus ; ++i) {
+		cpu_set(i, cpu_present_map);
+		cpu_set(i, cpu_possible_map);
+		set_hard_smp_processor_id(i, i);
+	}
 
 	if (ppc_md.progress) ppc_md.progress("smp_psurge_probe - done", 0x352);
 
@@ -348,6 +358,7 @@ static void __init psurge_dual_sync_tb(int cpu_nr)
 	int t;
 
 	set_dec(tb_ticks_per_jiffy);
+	/* XXX fixme */
 	set_tb(0, 0);
 	last_jiffy_stamp(cpu_nr) = 0;
 
@@ -363,8 +374,6 @@ static void __init psurge_dual_sync_tb(int cpu_nr)
 
 	/* now interrupt the secondary, starting both TBs */
 	psurge_set_ipi(1);
-
-	smp_tb_synchronized = 1;
 }
 
 static struct irqaction psurge_irqaction = {
@@ -625,9 +634,8 @@ void smp_core99_give_timebase(void)
 	for (t = 100000; t > 0 && sec_tb_reset; --t)
 		udelay(10);
 	if (sec_tb_reset)
+		/* XXX BUG_ON here? */
 		printk(KERN_WARNING "Timeout waiting sync(2) on second CPU\n");
-	else
-		smp_tb_synchronized = 1;
 
 	/* Now, restart the timebase by leaving the GPIO to an open collector */
        	pmac_call_feature(PMAC_FTR_WRITE_GPIO, NULL, core99_tb_gpio, 0);
@@ -810,19 +818,9 @@ static void __devinit smp_core99_setup_cpu(int cpu_nr)
 }
 
 
-/* Core99 Macs (dual G4s and G5s) */
-struct smp_ops_t core99_smp_ops = {
-	.message_pass	= smp_mpic_message_pass,
-	.probe		= smp_core99_probe,
-	.kick_cpu	= smp_core99_kick_cpu,
-	.setup_cpu	= smp_core99_setup_cpu,
-	.give_timebase	= smp_core99_give_timebase,
-	.take_timebase	= smp_core99_take_timebase,
-};
-
 #if defined(CONFIG_HOTPLUG_CPU) && defined(CONFIG_PPC32)
 
-int __cpu_disable(void)
+int smp_core99_cpu_disable(void)
 {
 	cpu_clear(smp_processor_id(), cpu_online_map);
 
@@ -846,7 +844,7 @@ void cpu_die(void)
 	low_cpu_die();
 }
 
-void __cpu_die(unsigned int cpu)
+void smp_core99_cpu_die(unsigned int cpu)
 {
 	int timeout;
 
@@ -858,8 +856,21 @@ void __cpu_die(unsigned int cpu)
 		}
 		msleep(1);
 	}
-	cpu_callin_map[cpu] = 0;
 	cpu_dead[cpu] = 0;
 }
 
 #endif
+
+/* Core99 Macs (dual G4s and G5s) */
+struct smp_ops_t core99_smp_ops = {
+	.message_pass	= smp_mpic_message_pass,
+	.probe		= smp_core99_probe,
+	.kick_cpu	= smp_core99_kick_cpu,
+	.setup_cpu	= smp_core99_setup_cpu,
+	.give_timebase	= smp_core99_give_timebase,
+	.take_timebase	= smp_core99_take_timebase,
+#if defined(CONFIG_HOTPLUG_CPU) && defined(CONFIG_PPC32)
+	.cpu_disable	= smp_core99_cpu_disable,
+	.cpu_die	= smp_core99_cpu_die,
+#endif
+};
diff --git a/arch/powerpc/platforms/pseries/Makefile b/arch/powerpc/platforms/pseries/Makefile
index b9938fece78..e7ca5b1f591 100644
--- a/arch/powerpc/platforms/pseries/Makefile
+++ b/arch/powerpc/platforms/pseries/Makefile
@@ -3,3 +3,5 @@ obj-y			:= pci.o lpar.o hvCall.o nvram.o reconfig.o \
 obj-$(CONFIG_SMP)	+= smp.o
 obj-$(CONFIG_IBMVIO)	+= vio.o
 obj-$(CONFIG_XICS)	+= xics.o
+obj-$(CONFIG_SCANLOG)	+= scanlog.o
+obj-$(CONFIG_EEH)    += eeh.o eeh_event.o
diff --git a/arch/powerpc/platforms/pseries/eeh.c b/arch/powerpc/platforms/pseries/eeh.c
new file mode 100644
index 00000000000..79de2310e70
--- /dev/null
+++ b/arch/powerpc/platforms/pseries/eeh.c
@@ -0,0 +1,1212 @@
+/*
+ * eeh.c
+ * Copyright (C) 2001 Dave Engebretsen & Todd Inglett IBM Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
+ */
+
+#include <linux/delay.h>
+#include <linux/init.h>
+#include <linux/list.h>
+#include <linux/pci.h>
+#include <linux/proc_fs.h>
+#include <linux/rbtree.h>
+#include <linux/seq_file.h>
+#include <linux/spinlock.h>
+#include <asm/atomic.h>
+#include <asm/eeh.h>
+#include <asm/eeh_event.h>
+#include <asm/io.h>
+#include <asm/machdep.h>
+#include <asm/ppc-pci.h>
+#include <asm/rtas.h>
+
+#undef DEBUG
+
+/** Overview:
+ *  EEH, or "Extended Error Handling" is a PCI bridge technology for
+ *  dealing with PCI bus errors that can't be dealt with within the
+ *  usual PCI framework, except by check-stopping the CPU.  Systems
+ *  that are designed for high-availability/reliability cannot afford
+ *  to crash due to a "mere" PCI error, thus the need for EEH.
+ *  An EEH-capable bridge operates by converting a detected error
+ *  into a "slot freeze", taking the PCI adapter off-line, making
+ *  the slot behave, from the OS'es point of view, as if the slot
+ *  were "empty": all reads return 0xff's and all writes are silently
+ *  ignored.  EEH slot isolation events can be triggered by parity
+ *  errors on the address or data busses (e.g. during posted writes),
+ *  which in turn might be caused by low voltage on the bus, dust,
+ *  vibration, humidity, radioactivity or plain-old failed hardware.
+ *
+ *  Note, however, that one of the leading causes of EEH slot
+ *  freeze events are buggy device drivers, buggy device microcode,
+ *  or buggy device hardware.  This is because any attempt by the
+ *  device to bus-master data to a memory address that is not
+ *  assigned to the device will trigger a slot freeze.   (The idea
+ *  is to prevent devices-gone-wild from corrupting system memory).
+ *  Buggy hardware/drivers will have a miserable time co-existing
+ *  with EEH.
+ *
+ *  Ideally, a PCI device driver, when suspecting that an isolation
+ *  event has occured (e.g. by reading 0xff's), will then ask EEH
+ *  whether this is the case, and then take appropriate steps to
+ *  reset the PCI slot, the PCI device, and then resume operations.
+ *  However, until that day,  the checking is done here, with the
+ *  eeh_check_failure() routine embedded in the MMIO macros.  If
+ *  the slot is found to be isolated, an "EEH Event" is synthesized
+ *  and sent out for processing.
+ */
+
+/* If a device driver keeps reading an MMIO register in an interrupt
+ * handler after a slot isolation event has occurred, we assume it
+ * is broken and panic.  This sets the threshold for how many read
+ * attempts we allow before panicking.
+ */
+#define EEH_MAX_FAILS	100000
+
+/* Misc forward declaraions */
+static void eeh_save_bars(struct pci_dev * pdev, struct pci_dn *pdn);
+
+/* RTAS tokens */
+static int ibm_set_eeh_option;
+static int ibm_set_slot_reset;
+static int ibm_read_slot_reset_state;
+static int ibm_read_slot_reset_state2;
+static int ibm_slot_error_detail;
+
+static int eeh_subsystem_enabled;
+
+/* Lock to avoid races due to multiple reports of an error */
+static DEFINE_SPINLOCK(confirm_error_lock);
+
+/* Buffer for reporting slot-error-detail rtas calls */
+static unsigned char slot_errbuf[RTAS_ERROR_LOG_MAX];
+static DEFINE_SPINLOCK(slot_errbuf_lock);
+static int eeh_error_buf_size;
+
+/* System monitoring statistics */
+static DEFINE_PER_CPU(unsigned long, no_device);
+static DEFINE_PER_CPU(unsigned long, no_dn);
+static DEFINE_PER_CPU(unsigned long, no_cfg_addr);
+static DEFINE_PER_CPU(unsigned long, ignored_check);
+static DEFINE_PER_CPU(unsigned long, total_mmio_ffs);
+static DEFINE_PER_CPU(unsigned long, false_positives);
+static DEFINE_PER_CPU(unsigned long, ignored_failures);
+static DEFINE_PER_CPU(unsigned long, slot_resets);
+
+/**
+ * The pci address cache subsystem.  This subsystem places
+ * PCI device address resources into a red-black tree, sorted
+ * according to the address range, so that given only an i/o
+ * address, the corresponding PCI device can be **quickly**
+ * found. It is safe to perform an address lookup in an interrupt
+ * context; this ability is an important feature.
+ *
+ * Currently, the only customer of this code is the EEH subsystem;
+ * thus, this code has been somewhat tailored to suit EEH better.
+ * In particular, the cache does *not* hold the addresses of devices
+ * for which EEH is not enabled.
+ *
+ * (Implementation Note: The RB tree seems to be better/faster
+ * than any hash algo I could think of for this problem, even
+ * with the penalty of slow pointer chases for d-cache misses).
+ */
+struct pci_io_addr_range
+{
+	struct rb_node rb_node;
+	unsigned long addr_lo;
+	unsigned long addr_hi;
+	struct pci_dev *pcidev;
+	unsigned int flags;
+};
+
+static struct pci_io_addr_cache
+{
+	struct rb_root rb_root;
+	spinlock_t piar_lock;
+} pci_io_addr_cache_root;
+
+static inline struct pci_dev *__pci_get_device_by_addr(unsigned long addr)
+{
+	struct rb_node *n = pci_io_addr_cache_root.rb_root.rb_node;
+
+	while (n) {
+		struct pci_io_addr_range *piar;
+		piar = rb_entry(n, struct pci_io_addr_range, rb_node);
+
+		if (addr < piar->addr_lo) {
+			n = n->rb_left;
+		} else {
+			if (addr > piar->addr_hi) {
+				n = n->rb_right;
+			} else {
+				pci_dev_get(piar->pcidev);
+				return piar->pcidev;
+			}
+		}
+	}
+
+	return NULL;
+}
+
+/**
+ * pci_get_device_by_addr - Get device, given only address
+ * @addr: mmio (PIO) phys address or i/o port number
+ *
+ * Given an mmio phys address, or a port number, find a pci device
+ * that implements this address.  Be sure to pci_dev_put the device
+ * when finished.  I/O port numbers are assumed to be offset
+ * from zero (that is, they do *not* have pci_io_addr added in).
+ * It is safe to call this function within an interrupt.
+ */
+static struct pci_dev *pci_get_device_by_addr(unsigned long addr)
+{
+	struct pci_dev *dev;
+	unsigned long flags;
+
+	spin_lock_irqsave(&pci_io_addr_cache_root.piar_lock, flags);
+	dev = __pci_get_device_by_addr(addr);
+	spin_unlock_irqrestore(&pci_io_addr_cache_root.piar_lock, flags);
+	return dev;
+}
+
+#ifdef DEBUG
+/*
+ * Handy-dandy debug print routine, does nothing more
+ * than print out the contents of our addr cache.
+ */
+static void pci_addr_cache_print(struct pci_io_addr_cache *cache)
+{
+	struct rb_node *n;
+	int cnt = 0;
+
+	n = rb_first(&cache->rb_root);
+	while (n) {
+		struct pci_io_addr_range *piar;
+		piar = rb_entry(n, struct pci_io_addr_range, rb_node);
+		printk(KERN_DEBUG "PCI: %s addr range %d [%lx-%lx]: %s\n",
+		       (piar->flags & IORESOURCE_IO) ? "i/o" : "mem", cnt,
+		       piar->addr_lo, piar->addr_hi, pci_name(piar->pcidev));
+		cnt++;
+		n = rb_next(n);
+	}
+}
+#endif
+
+/* Insert address range into the rb tree. */
+static struct pci_io_addr_range *
+pci_addr_cache_insert(struct pci_dev *dev, unsigned long alo,
+		      unsigned long ahi, unsigned int flags)
+{
+	struct rb_node **p = &pci_io_addr_cache_root.rb_root.rb_node;
+	struct rb_node *parent = NULL;
+	struct pci_io_addr_range *piar;
+
+	/* Walk tree, find a place to insert into tree */
+	while (*p) {
+		parent = *p;
+		piar = rb_entry(parent, struct pci_io_addr_range, rb_node);
+		if (ahi < piar->addr_lo) {
+			p = &parent->rb_left;
+		} else if (alo > piar->addr_hi) {
+			p = &parent->rb_right;
+		} else {
+			if (dev != piar->pcidev ||
+			    alo != piar->addr_lo || ahi != piar->addr_hi) {
+				printk(KERN_WARNING "PIAR: overlapping address range\n");
+			}
+			return piar;
+		}
+	}
+	piar = (struct pci_io_addr_range *)kmalloc(sizeof(struct pci_io_addr_range), GFP_ATOMIC);
+	if (!piar)
+		return NULL;
+
+	piar->addr_lo = alo;
+	piar->addr_hi = ahi;
+	piar->pcidev = dev;
+	piar->flags = flags;
+
+#ifdef DEBUG
+	printk(KERN_DEBUG "PIAR: insert range=[%lx:%lx] dev=%s\n",
+	                  alo, ahi, pci_name (dev));
+#endif
+
+	rb_link_node(&piar->rb_node, parent, p);
+	rb_insert_color(&piar->rb_node, &pci_io_addr_cache_root.rb_root);
+
+	return piar;
+}
+
+static void __pci_addr_cache_insert_device(struct pci_dev *dev)
+{
+	struct device_node *dn;
+	struct pci_dn *pdn;
+	int i;
+	int inserted = 0;
+
+	dn = pci_device_to_OF_node(dev);
+	if (!dn) {
+		printk(KERN_WARNING "PCI: no pci dn found for dev=%s\n", pci_name(dev));
+		return;
+	}
+
+	/* Skip any devices for which EEH is not enabled. */
+	pdn = PCI_DN(dn);
+	if (!(pdn->eeh_mode & EEH_MODE_SUPPORTED) ||
+	    pdn->eeh_mode & EEH_MODE_NOCHECK) {
+#ifdef DEBUG
+		printk(KERN_INFO "PCI: skip building address cache for=%s - %s\n",
+		       pci_name(dev), pdn->node->full_name);
+#endif
+		return;
+	}
+
+	/* The cache holds a reference to the device... */
+	pci_dev_get(dev);
+
+	/* Walk resources on this device, poke them into the tree */
+	for (i = 0; i < DEVICE_COUNT_RESOURCE; i++) {
+		unsigned long start = pci_resource_start(dev,i);
+		unsigned long end = pci_resource_end(dev,i);
+		unsigned int flags = pci_resource_flags(dev,i);
+
+		/* We are interested only bus addresses, not dma or other stuff */
+		if (0 == (flags & (IORESOURCE_IO | IORESOURCE_MEM)))
+			continue;
+		if (start == 0 || ~start == 0 || end == 0 || ~end == 0)
+			 continue;
+		pci_addr_cache_insert(dev, start, end, flags);
+		inserted = 1;
+	}
+
+	/* If there was nothing to add, the cache has no reference... */
+	if (!inserted)
+		pci_dev_put(dev);
+}
+
+/**
+ * pci_addr_cache_insert_device - Add a device to the address cache
+ * @dev: PCI device whose I/O addresses we are interested in.
+ *
+ * In order to support the fast lookup of devices based on addresses,
+ * we maintain a cache of devices that can be quickly searched.
+ * This routine adds a device to that cache.
+ */
+static void pci_addr_cache_insert_device(struct pci_dev *dev)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&pci_io_addr_cache_root.piar_lock, flags);
+	__pci_addr_cache_insert_device(dev);
+	spin_unlock_irqrestore(&pci_io_addr_cache_root.piar_lock, flags);
+}
+
+static inline void __pci_addr_cache_remove_device(struct pci_dev *dev)
+{
+	struct rb_node *n;
+	int removed = 0;
+
+restart:
+	n = rb_first(&pci_io_addr_cache_root.rb_root);
+	while (n) {
+		struct pci_io_addr_range *piar;
+		piar = rb_entry(n, struct pci_io_addr_range, rb_node);
+
+		if (piar->pcidev == dev) {
+			rb_erase(n, &pci_io_addr_cache_root.rb_root);
+			removed = 1;
+			kfree(piar);
+			goto restart;
+		}
+		n = rb_next(n);
+	}
+
+	/* The cache no longer holds its reference to this device... */
+	if (removed)
+		pci_dev_put(dev);
+}
+
+/**
+ * pci_addr_cache_remove_device - remove pci device from addr cache
+ * @dev: device to remove
+ *
+ * Remove a device from the addr-cache tree.
+ * This is potentially expensive, since it will walk
+ * the tree multiple times (once per resource).
+ * But so what; device removal doesn't need to be that fast.
+ */
+static void pci_addr_cache_remove_device(struct pci_dev *dev)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&pci_io_addr_cache_root.piar_lock, flags);
+	__pci_addr_cache_remove_device(dev);
+	spin_unlock_irqrestore(&pci_io_addr_cache_root.piar_lock, flags);
+}
+
+/**
+ * pci_addr_cache_build - Build a cache of I/O addresses
+ *
+ * Build a cache of pci i/o addresses.  This cache will be used to
+ * find the pci device that corresponds to a given address.
+ * This routine scans all pci busses to build the cache.
+ * Must be run late in boot process, after the pci controllers
+ * have been scaned for devices (after all device resources are known).
+ */
+void __init pci_addr_cache_build(void)
+{
+	struct device_node *dn;
+	struct pci_dev *dev = NULL;
+
+	if (!eeh_subsystem_enabled)
+		return;
+
+	spin_lock_init(&pci_io_addr_cache_root.piar_lock);
+
+	while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) {
+		/* Ignore PCI bridges ( XXX why ??) */
+		if ((dev->class >> 16) == PCI_BASE_CLASS_BRIDGE) {
+			continue;
+		}
+		pci_addr_cache_insert_device(dev);
+
+		/* Save the BAR's; firmware doesn't restore these after EEH reset */
+		dn = pci_device_to_OF_node(dev);
+		eeh_save_bars(dev, PCI_DN(dn));
+	}
+
+#ifdef DEBUG
+	/* Verify tree built up above, echo back the list of addrs. */
+	pci_addr_cache_print(&pci_io_addr_cache_root);
+#endif
+}
+
+/* --------------------------------------------------------------- */
+/* Above lies the PCI Address Cache. Below lies the EEH event infrastructure */
+
+void eeh_slot_error_detail (struct pci_dn *pdn, int severity)
+{
+	unsigned long flags;
+	int rc;
+
+	/* Log the error with the rtas logger */
+	spin_lock_irqsave(&slot_errbuf_lock, flags);
+	memset(slot_errbuf, 0, eeh_error_buf_size);
+
+	rc = rtas_call(ibm_slot_error_detail,
+	               8, 1, NULL, pdn->eeh_config_addr,
+	               BUID_HI(pdn->phb->buid),
+	               BUID_LO(pdn->phb->buid), NULL, 0,
+	               virt_to_phys(slot_errbuf),
+	               eeh_error_buf_size,
+	               severity);
+
+	if (rc == 0)
+		log_error(slot_errbuf, ERR_TYPE_RTAS_LOG, 0);
+	spin_unlock_irqrestore(&slot_errbuf_lock, flags);
+}
+
+/**
+ * read_slot_reset_state - Read the reset state of a device node's slot
+ * @dn: device node to read
+ * @rets: array to return results in
+ */
+static int read_slot_reset_state(struct pci_dn *pdn, int rets[])
+{
+	int token, outputs;
+
+	if (ibm_read_slot_reset_state2 != RTAS_UNKNOWN_SERVICE) {
+		token = ibm_read_slot_reset_state2;
+		outputs = 4;
+	} else {
+		token = ibm_read_slot_reset_state;
+		rets[2] = 0; /* fake PE Unavailable info */
+		outputs = 3;
+	}
+
+	return rtas_call(token, 3, outputs, rets, pdn->eeh_config_addr,
+			 BUID_HI(pdn->phb->buid), BUID_LO(pdn->phb->buid));
+}
+
+/**
+ * eeh_token_to_phys - convert EEH address token to phys address
+ * @token i/o token, should be address in the form 0xA....
+ */
+static inline unsigned long eeh_token_to_phys(unsigned long token)
+{
+	pte_t *ptep;
+	unsigned long pa;
+
+	ptep = find_linux_pte(init_mm.pgd, token);
+	if (!ptep)
+		return token;
+	pa = pte_pfn(*ptep) << PAGE_SHIFT;
+
+	return pa | (token & (PAGE_SIZE-1));
+}
+
+/** 
+ * Return the "partitionable endpoint" (pe) under which this device lies
+ */
+static struct device_node * find_device_pe(struct device_node *dn)
+{
+	while ((dn->parent) && PCI_DN(dn->parent) &&
+	      (PCI_DN(dn->parent)->eeh_mode & EEH_MODE_SUPPORTED)) {
+		dn = dn->parent;
+	}
+	return dn;
+}
+
+/** Mark all devices that are peers of this device as failed.
+ *  Mark the device driver too, so that it can see the failure
+ *  immediately; this is critical, since some drivers poll
+ *  status registers in interrupts ... If a driver is polling,
+ *  and the slot is frozen, then the driver can deadlock in
+ *  an interrupt context, which is bad.
+ */
+
+static void __eeh_mark_slot (struct device_node *dn, int mode_flag)
+{
+	while (dn) {
+		if (PCI_DN(dn)) {
+			PCI_DN(dn)->eeh_mode |= mode_flag;
+
+			if (dn->child)
+				__eeh_mark_slot (dn->child, mode_flag);
+		}
+		dn = dn->sibling;
+	}
+}
+
+void eeh_mark_slot (struct device_node *dn, int mode_flag)
+{
+	dn = find_device_pe (dn);
+	PCI_DN(dn)->eeh_mode |= mode_flag;
+	__eeh_mark_slot (dn->child, mode_flag);
+}
+
+static void __eeh_clear_slot (struct device_node *dn, int mode_flag)
+{
+	while (dn) {
+		if (PCI_DN(dn)) {
+			PCI_DN(dn)->eeh_mode &= ~mode_flag;
+			PCI_DN(dn)->eeh_check_count = 0;
+			if (dn->child)
+				__eeh_clear_slot (dn->child, mode_flag);
+		}
+		dn = dn->sibling;
+	}
+}
+
+void eeh_clear_slot (struct device_node *dn, int mode_flag)
+{
+	unsigned long flags;
+	spin_lock_irqsave(&confirm_error_lock, flags);
+	dn = find_device_pe (dn);
+	PCI_DN(dn)->eeh_mode &= ~mode_flag;
+	PCI_DN(dn)->eeh_check_count = 0;
+	__eeh_clear_slot (dn->child, mode_flag);
+	spin_unlock_irqrestore(&confirm_error_lock, flags);
+}
+
+/**
+ * eeh_dn_check_failure - check if all 1's data is due to EEH slot freeze
+ * @dn device node
+ * @dev pci device, if known
+ *
+ * Check for an EEH failure for the given device node.  Call this
+ * routine if the result of a read was all 0xff's and you want to
+ * find out if this is due to an EEH slot freeze.  This routine
+ * will query firmware for the EEH status.
+ *
+ * Returns 0 if there has not been an EEH error; otherwise returns
+ * a non-zero value and queues up a slot isolation event notification.
+ *
+ * It is safe to call this routine in an interrupt context.
+ */
+int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev)
+{
+	int ret;
+	int rets[3];
+	unsigned long flags;
+	struct pci_dn *pdn;
+	int rc = 0;
+
+	__get_cpu_var(total_mmio_ffs)++;
+
+	if (!eeh_subsystem_enabled)
+		return 0;
+
+	if (!dn) {
+		__get_cpu_var(no_dn)++;
+		return 0;
+	}
+	pdn = PCI_DN(dn);
+
+	/* Access to IO BARs might get this far and still not want checking. */
+	if (!(pdn->eeh_mode & EEH_MODE_SUPPORTED) ||
+	    pdn->eeh_mode & EEH_MODE_NOCHECK) {
+		__get_cpu_var(ignored_check)++;
+#ifdef DEBUG
+		printk ("EEH:ignored check (%x) for %s %s\n", 
+		        pdn->eeh_mode, pci_name (dev), dn->full_name);
+#endif
+		return 0;
+	}
+
+	if (!pdn->eeh_config_addr) {
+		__get_cpu_var(no_cfg_addr)++;
+		return 0;
+	}
+
+	/* If we already have a pending isolation event for this
+	 * slot, we know it's bad already, we don't need to check.
+	 * Do this checking under a lock; as multiple PCI devices
+	 * in one slot might report errors simultaneously, and we
+	 * only want one error recovery routine running.
+	 */
+	spin_lock_irqsave(&confirm_error_lock, flags);
+	rc = 1;
+	if (pdn->eeh_mode & EEH_MODE_ISOLATED) {
+		pdn->eeh_check_count ++;
+		if (pdn->eeh_check_count >= EEH_MAX_FAILS) {
+			printk (KERN_ERR "EEH: Device driver ignored %d bad reads, panicing\n",
+			        pdn->eeh_check_count);
+			dump_stack();
+			
+			/* re-read the slot reset state */
+			if (read_slot_reset_state(pdn, rets) != 0)
+				rets[0] = -1;	/* reset state unknown */
+
+			/* If we are here, then we hit an infinite loop. Stop. */
+			panic("EEH: MMIO halt (%d) on device:%s\n", rets[0], pci_name(dev));
+		}
+		goto dn_unlock;
+	}
+
+	/*
+	 * Now test for an EEH failure.  This is VERY expensive.
+	 * Note that the eeh_config_addr may be a parent device
+	 * in the case of a device behind a bridge, or it may be
+	 * function zero of a multi-function device.
+	 * In any case they must share a common PHB.
+	 */
+	ret = read_slot_reset_state(pdn, rets);
+
+	/* If the call to firmware failed, punt */
+	if (ret != 0) {
+		printk(KERN_WARNING "EEH: read_slot_reset_state() failed; rc=%d dn=%s\n",
+		       ret, dn->full_name);
+		__get_cpu_var(false_positives)++;
+		rc = 0;
+		goto dn_unlock;
+	}
+
+	/* If EEH is not supported on this device, punt. */
+	if (rets[1] != 1) {
+		printk(KERN_WARNING "EEH: event on unsupported device, rc=%d dn=%s\n",
+		       ret, dn->full_name);
+		__get_cpu_var(false_positives)++;
+		rc = 0;
+		goto dn_unlock;
+	}
+
+	/* If not the kind of error we know about, punt. */
+	if (rets[0] != 2 && rets[0] != 4 && rets[0] != 5) {
+		__get_cpu_var(false_positives)++;
+		rc = 0;
+		goto dn_unlock;
+	}
+
+	/* Note that config-io to empty slots may fail;
+	 * we recognize empty because they don't have children. */
+	if ((rets[0] == 5) && (dn->child == NULL)) {
+		__get_cpu_var(false_positives)++;
+		rc = 0;
+		goto dn_unlock;
+	}
+
+	__get_cpu_var(slot_resets)++;
+ 
+	/* Avoid repeated reports of this failure, including problems
+	 * with other functions on this device, and functions under
+	 * bridges. */
+	eeh_mark_slot (dn, EEH_MODE_ISOLATED);
+	spin_unlock_irqrestore(&confirm_error_lock, flags);
+
+	eeh_send_failure_event (dn, dev, rets[0], rets[2]);
+	
+	/* Most EEH events are due to device driver bugs.  Having
+	 * a stack trace will help the device-driver authors figure
+	 * out what happened.  So print that out. */
+	if (rets[0] != 5) dump_stack();
+	return 1;
+
+dn_unlock:
+	spin_unlock_irqrestore(&confirm_error_lock, flags);
+	return rc;
+}
+
+EXPORT_SYMBOL_GPL(eeh_dn_check_failure);
+
+/**
+ * eeh_check_failure - check if all 1's data is due to EEH slot freeze
+ * @token i/o token, should be address in the form 0xA....
+ * @val value, should be all 1's (XXX why do we need this arg??)
+ *
+ * Check for an EEH failure at the given token address.  Call this
+ * routine if the result of a read was all 0xff's and you want to
+ * find out if this is due to an EEH slot freeze event.  This routine
+ * will query firmware for the EEH status.
+ *
+ * Note this routine is safe to call in an interrupt context.
+ */
+unsigned long eeh_check_failure(const volatile void __iomem *token, unsigned long val)
+{
+	unsigned long addr;
+	struct pci_dev *dev;
+	struct device_node *dn;
+
+	/* Finding the phys addr + pci device; this is pretty quick. */
+	addr = eeh_token_to_phys((unsigned long __force) token);
+	dev = pci_get_device_by_addr(addr);
+	if (!dev) {
+		__get_cpu_var(no_device)++;
+		return val;
+	}
+
+	dn = pci_device_to_OF_node(dev);
+	eeh_dn_check_failure (dn, dev);
+
+	pci_dev_put(dev);
+	return val;
+}
+
+EXPORT_SYMBOL(eeh_check_failure);
+
+/* ------------------------------------------------------------- */
+/* The code below deals with error recovery */
+
+/** Return negative value if a permanent error, else return
+ * a number of milliseconds to wait until the PCI slot is
+ * ready to be used.
+ */
+static int
+eeh_slot_availability(struct pci_dn *pdn)
+{
+	int rc;
+	int rets[3];
+
+	rc = read_slot_reset_state(pdn, rets);
+
+	if (rc) return rc;
+
+	if (rets[1] == 0) return -1;  /* EEH is not supported */
+	if (rets[0] == 0)  return 0;  /* Oll Korrect */
+	if (rets[0] == 5) {
+		if (rets[2] == 0) return -1; /* permanently unavailable */
+		return rets[2]; /* number of millisecs to wait */
+	}
+	return -1;
+}
+
+/** rtas_pci_slot_reset raises/lowers the pci #RST line
+ *  state: 1/0 to raise/lower the #RST
+ *
+ * Clear the EEH-frozen condition on a slot.  This routine
+ * asserts the PCI #RST line if the 'state' argument is '1',
+ * and drops the #RST line if 'state is '0'.  This routine is
+ * safe to call in an interrupt context.
+ *
+ */
+
+static void
+rtas_pci_slot_reset(struct pci_dn *pdn, int state)
+{
+	int rc;
+
+	BUG_ON (pdn==NULL); 
+
+	if (!pdn->phb) {
+		printk (KERN_WARNING "EEH: in slot reset, device node %s has no phb\n",
+		        pdn->node->full_name);
+		return;
+	}
+
+	rc = rtas_call(ibm_set_slot_reset,4,1, NULL,
+	               pdn->eeh_config_addr,
+	               BUID_HI(pdn->phb->buid),
+	               BUID_LO(pdn->phb->buid),
+	               state);
+	if (rc) {
+		printk (KERN_WARNING "EEH: Unable to reset the failed slot, (%d) #RST=%d dn=%s\n", 
+		        rc, state, pdn->node->full_name);
+		return;
+	}
+}
+
+/** rtas_set_slot_reset -- assert the pci #RST line for 1/4 second
+ *  dn -- device node to be reset.
+ */
+
+void
+rtas_set_slot_reset(struct pci_dn *pdn)
+{
+	int i, rc;
+
+	rtas_pci_slot_reset (pdn, 1);
+
+	/* The PCI bus requires that the reset be held high for at least
+	 * a 100 milliseconds. We wait a bit longer 'just in case'.  */
+
+#define PCI_BUS_RST_HOLD_TIME_MSEC 250
+	msleep (PCI_BUS_RST_HOLD_TIME_MSEC);
+	
+	/* We might get hit with another EEH freeze as soon as the 
+	 * pci slot reset line is dropped. Make sure we don't miss
+	 * these, and clear the flag now. */
+	eeh_clear_slot (pdn->node, EEH_MODE_ISOLATED);
+
+	rtas_pci_slot_reset (pdn, 0);
+
+	/* After a PCI slot has been reset, the PCI Express spec requires
+	 * a 1.5 second idle time for the bus to stabilize, before starting
+	 * up traffic. */
+#define PCI_BUS_SETTLE_TIME_MSEC 1800
+	msleep (PCI_BUS_SETTLE_TIME_MSEC);
+
+	/* Now double check with the firmware to make sure the device is
+	 * ready to be used; if not, wait for recovery. */
+	for (i=0; i<10; i++) {
+		rc = eeh_slot_availability (pdn);
+		if (rc <= 0) break;
+
+		msleep (rc+100);
+	}
+}
+
+/* ------------------------------------------------------- */
+/** Save and restore of PCI BARs
+ *
+ * Although firmware will set up BARs during boot, it doesn't
+ * set up device BAR's after a device reset, although it will,
+ * if requested, set up bridge configuration. Thus, we need to
+ * configure the PCI devices ourselves.  
+ */
+
+/**
+ * __restore_bars - Restore the Base Address Registers
+ * Loads the PCI configuration space base address registers,
+ * the expansion ROM base address, the latency timer, and etc.
+ * from the saved values in the device node.
+ */
+static inline void __restore_bars (struct pci_dn *pdn)
+{
+	int i;
+
+	if (NULL==pdn->phb) return;
+	for (i=4; i<10; i++) {
+		rtas_write_config(pdn, i*4, 4, pdn->config_space[i]);
+	}
+
+	/* 12 == Expansion ROM Address */
+	rtas_write_config(pdn, 12*4, 4, pdn->config_space[12]);
+
+#define BYTE_SWAP(OFF) (8*((OFF)/4)+3-(OFF))
+#define SAVED_BYTE(OFF) (((u8 *)(pdn->config_space))[BYTE_SWAP(OFF)])
+
+	rtas_write_config (pdn, PCI_CACHE_LINE_SIZE, 1,
+	            SAVED_BYTE(PCI_CACHE_LINE_SIZE));
+
+	rtas_write_config (pdn, PCI_LATENCY_TIMER, 1,
+	            SAVED_BYTE(PCI_LATENCY_TIMER));
+
+	/* max latency, min grant, interrupt pin and line */
+	rtas_write_config(pdn, 15*4, 4, pdn->config_space[15]);
+}
+
+/**
+ * eeh_restore_bars - restore the PCI config space info
+ *
+ * This routine performs a recursive walk to the children
+ * of this device as well.
+ */
+void eeh_restore_bars(struct pci_dn *pdn)
+{
+	struct device_node *dn;
+	if (!pdn) 
+		return;
+	
+	if (! pdn->eeh_is_bridge)
+		__restore_bars (pdn);
+
+	dn = pdn->node->child;
+	while (dn) {
+		eeh_restore_bars (PCI_DN(dn));
+		dn = dn->sibling;
+	}
+}
+
+/**
+ * eeh_save_bars - save device bars
+ *
+ * Save the values of the device bars. Unlike the restore
+ * routine, this routine is *not* recursive. This is because
+ * PCI devices are added individuallly; but, for the restore,
+ * an entire slot is reset at a time.
+ */
+static void eeh_save_bars(struct pci_dev * pdev, struct pci_dn *pdn)
+{
+	int i;
+
+	if (!pdev || !pdn )
+		return;
+	
+	for (i = 0; i < 16; i++)
+		pci_read_config_dword(pdev, i * 4, &pdn->config_space[i]);
+
+	if (pdev->hdr_type == PCI_HEADER_TYPE_BRIDGE)
+		pdn->eeh_is_bridge = 1;
+}
+
+void
+rtas_configure_bridge(struct pci_dn *pdn)
+{
+	int token = rtas_token ("ibm,configure-bridge");
+	int rc;
+
+	if (token == RTAS_UNKNOWN_SERVICE)
+		return;
+	rc = rtas_call(token,3,1, NULL,
+	               pdn->eeh_config_addr,
+	               BUID_HI(pdn->phb->buid),
+	               BUID_LO(pdn->phb->buid));
+	if (rc) {
+		printk (KERN_WARNING "EEH: Unable to configure device bridge (%d) for %s\n",
+		        rc, pdn->node->full_name);
+	}
+}
+
+/* ------------------------------------------------------------- */
+/* The code below deals with enabling EEH for devices during  the
+ * early boot sequence.  EEH must be enabled before any PCI probing
+ * can be done.
+ */
+
+#define EEH_ENABLE 1
+
+struct eeh_early_enable_info {
+	unsigned int buid_hi;
+	unsigned int buid_lo;
+};
+
+/* Enable eeh for the given device node. */
+static void *early_enable_eeh(struct device_node *dn, void *data)
+{
+	struct eeh_early_enable_info *info = data;
+	int ret;
+	char *status = get_property(dn, "status", NULL);
+	u32 *class_code = (u32 *)get_property(dn, "class-code", NULL);
+	u32 *vendor_id = (u32 *)get_property(dn, "vendor-id", NULL);
+	u32 *device_id = (u32 *)get_property(dn, "device-id", NULL);
+	u32 *regs;
+	int enable;
+	struct pci_dn *pdn = PCI_DN(dn);
+
+	pdn->eeh_mode = 0;
+	pdn->eeh_check_count = 0;
+	pdn->eeh_freeze_count = 0;
+
+	if (status && strcmp(status, "ok") != 0)
+		return NULL;	/* ignore devices with bad status */
+
+	/* Ignore bad nodes. */
+	if (!class_code || !vendor_id || !device_id)
+		return NULL;
+
+	/* There is nothing to check on PCI to ISA bridges */
+	if (dn->type && !strcmp(dn->type, "isa")) {
+		pdn->eeh_mode |= EEH_MODE_NOCHECK;
+		return NULL;
+	}
+
+	/*
+	 * Now decide if we are going to "Disable" EEH checking
+	 * for this device.  We still run with the EEH hardware active,
+	 * but we won't be checking for ff's.  This means a driver
+	 * could return bad data (very bad!), an interrupt handler could
+	 * hang waiting on status bits that won't change, etc.
+	 * But there are a few cases like display devices that make sense.
+	 */
+	enable = 1;	/* i.e. we will do checking */
+	if ((*class_code >> 16) == PCI_BASE_CLASS_DISPLAY)
+		enable = 0;
+
+	if (!enable)
+		pdn->eeh_mode |= EEH_MODE_NOCHECK;
+
+	/* Ok... see if this device supports EEH.  Some do, some don't,
+	 * and the only way to find out is to check each and every one. */
+	regs = (u32 *)get_property(dn, "reg", NULL);
+	if (regs) {
+		/* First register entry is addr (00BBSS00)  */
+		/* Try to enable eeh */
+		ret = rtas_call(ibm_set_eeh_option, 4, 1, NULL,
+		                regs[0], info->buid_hi, info->buid_lo,
+		                EEH_ENABLE);
+
+		if (ret == 0) {
+			eeh_subsystem_enabled = 1;
+			pdn->eeh_mode |= EEH_MODE_SUPPORTED;
+			pdn->eeh_config_addr = regs[0];
+#ifdef DEBUG
+			printk(KERN_DEBUG "EEH: %s: eeh enabled\n", dn->full_name);
+#endif
+		} else {
+
+			/* This device doesn't support EEH, but it may have an
+			 * EEH parent, in which case we mark it as supported. */
+			if (dn->parent && PCI_DN(dn->parent)
+			    && (PCI_DN(dn->parent)->eeh_mode & EEH_MODE_SUPPORTED)) {
+				/* Parent supports EEH. */
+				pdn->eeh_mode |= EEH_MODE_SUPPORTED;
+				pdn->eeh_config_addr = PCI_DN(dn->parent)->eeh_config_addr;
+				return NULL;
+			}
+		}
+	} else {
+		printk(KERN_WARNING "EEH: %s: unable to get reg property.\n",
+		       dn->full_name);
+	}
+
+	return NULL;
+}
+
+/*
+ * Initialize EEH by trying to enable it for all of the adapters in the system.
+ * As a side effect we can determine here if eeh is supported at all.
+ * Note that we leave EEH on so failed config cycles won't cause a machine
+ * check.  If a user turns off EEH for a particular adapter they are really
+ * telling Linux to ignore errors.  Some hardware (e.g. POWER5) won't
+ * grant access to a slot if EEH isn't enabled, and so we always enable
+ * EEH for all slots/all devices.
+ *
+ * The eeh-force-off option disables EEH checking globally, for all slots.
+ * Even if force-off is set, the EEH hardware is still enabled, so that
+ * newer systems can boot.
+ */
+void __init eeh_init(void)
+{
+	struct device_node *phb, *np;
+	struct eeh_early_enable_info info;
+
+	spin_lock_init(&confirm_error_lock);
+	spin_lock_init(&slot_errbuf_lock);
+
+	np = of_find_node_by_path("/rtas");
+	if (np == NULL)
+		return;
+
+	ibm_set_eeh_option = rtas_token("ibm,set-eeh-option");
+	ibm_set_slot_reset = rtas_token("ibm,set-slot-reset");
+	ibm_read_slot_reset_state2 = rtas_token("ibm,read-slot-reset-state2");
+	ibm_read_slot_reset_state = rtas_token("ibm,read-slot-reset-state");
+	ibm_slot_error_detail = rtas_token("ibm,slot-error-detail");
+
+	if (ibm_set_eeh_option == RTAS_UNKNOWN_SERVICE)
+		return;
+
+	eeh_error_buf_size = rtas_token("rtas-error-log-max");
+	if (eeh_error_buf_size == RTAS_UNKNOWN_SERVICE) {
+		eeh_error_buf_size = 1024;
+	}
+	if (eeh_error_buf_size > RTAS_ERROR_LOG_MAX) {
+		printk(KERN_WARNING "EEH: rtas-error-log-max is bigger than allocated "
+		      "buffer ! (%d vs %d)", eeh_error_buf_size, RTAS_ERROR_LOG_MAX);
+		eeh_error_buf_size = RTAS_ERROR_LOG_MAX;
+	}
+
+	/* Enable EEH for all adapters.  Note that eeh requires buid's */
+	for (phb = of_find_node_by_name(NULL, "pci"); phb;
+	     phb = of_find_node_by_name(phb, "pci")) {
+		unsigned long buid;
+
+		buid = get_phb_buid(phb);
+		if (buid == 0 || PCI_DN(phb) == NULL)
+			continue;
+
+		info.buid_lo = BUID_LO(buid);
+		info.buid_hi = BUID_HI(buid);
+		traverse_pci_devices(phb, early_enable_eeh, &info);
+	}
+
+	if (eeh_subsystem_enabled)
+		printk(KERN_INFO "EEH: PCI Enhanced I/O Error Handling Enabled\n");
+	else
+		printk(KERN_WARNING "EEH: No capable adapters found\n");
+}
+
+/**
+ * eeh_add_device_early - enable EEH for the indicated device_node
+ * @dn: device node for which to set up EEH
+ *
+ * This routine must be used to perform EEH initialization for PCI
+ * devices that were added after system boot (e.g. hotplug, dlpar).
+ * This routine must be called before any i/o is performed to the
+ * adapter (inluding any config-space i/o).
+ * Whether this actually enables EEH or not for this device depends
+ * on the CEC architecture, type of the device, on earlier boot
+ * command-line arguments & etc.
+ */
+void eeh_add_device_early(struct device_node *dn)
+{
+	struct pci_controller *phb;
+	struct eeh_early_enable_info info;
+
+	if (!dn || !PCI_DN(dn))
+		return;
+	phb = PCI_DN(dn)->phb;
+	if (NULL == phb || 0 == phb->buid) {
+		printk(KERN_WARNING "EEH: Expected buid but found none for %s\n",
+		       dn->full_name);
+		dump_stack();
+		return;
+	}
+
+	info.buid_hi = BUID_HI(phb->buid);
+	info.buid_lo = BUID_LO(phb->buid);
+	early_enable_eeh(dn, &info);
+}
+EXPORT_SYMBOL_GPL(eeh_add_device_early);
+
+/**
+ * eeh_add_device_late - perform EEH initialization for the indicated pci device
+ * @dev: pci device for which to set up EEH
+ *
+ * This routine must be used to complete EEH initialization for PCI
+ * devices that were added after system boot (e.g. hotplug, dlpar).
+ */
+void eeh_add_device_late(struct pci_dev *dev)
+{
+	struct device_node *dn;
+	struct pci_dn *pdn;
+
+	if (!dev || !eeh_subsystem_enabled)
+		return;
+
+#ifdef DEBUG
+	printk(KERN_DEBUG "EEH: adding device %s\n", pci_name(dev));
+#endif
+
+	pci_dev_get (dev);
+	dn = pci_device_to_OF_node(dev);
+	pdn = PCI_DN(dn);
+	pdn->pcidev = dev;
+
+	pci_addr_cache_insert_device (dev);
+	eeh_save_bars(dev, pdn);
+}
+EXPORT_SYMBOL_GPL(eeh_add_device_late);
+
+/**
+ * eeh_remove_device - undo EEH setup for the indicated pci device
+ * @dev: pci device to be removed
+ *
+ * This routine should be when a device is removed from a running
+ * system (e.g. by hotplug or dlpar).
+ */
+void eeh_remove_device(struct pci_dev *dev)
+{
+	struct device_node *dn;
+	if (!dev || !eeh_subsystem_enabled)
+		return;
+
+	/* Unregister the device with the EEH/PCI address search system */
+#ifdef DEBUG
+	printk(KERN_DEBUG "EEH: remove device %s\n", pci_name(dev));
+#endif
+	pci_addr_cache_remove_device(dev);
+
+	dn = pci_device_to_OF_node(dev);
+	PCI_DN(dn)->pcidev = NULL;
+	pci_dev_put (dev);
+}
+EXPORT_SYMBOL_GPL(eeh_remove_device);
+
+static int proc_eeh_show(struct seq_file *m, void *v)
+{
+	unsigned int cpu;
+	unsigned long ffs = 0, positives = 0, failures = 0;
+	unsigned long resets = 0;
+	unsigned long no_dev = 0, no_dn = 0, no_cfg = 0, no_check = 0;
+
+	for_each_cpu(cpu) {
+		ffs += per_cpu(total_mmio_ffs, cpu);
+		positives += per_cpu(false_positives, cpu);
+		failures += per_cpu(ignored_failures, cpu);
+		resets += per_cpu(slot_resets, cpu);
+		no_dev += per_cpu(no_device, cpu);
+		no_dn += per_cpu(no_dn, cpu);
+		no_cfg += per_cpu(no_cfg_addr, cpu);
+		no_check += per_cpu(ignored_check, cpu);
+	}
+
+	if (0 == eeh_subsystem_enabled) {
+		seq_printf(m, "EEH Subsystem is globally disabled\n");
+		seq_printf(m, "eeh_total_mmio_ffs=%ld\n", ffs);
+	} else {
+		seq_printf(m, "EEH Subsystem is enabled\n");
+		seq_printf(m,
+				"no device=%ld\n"
+				"no device node=%ld\n"
+				"no config address=%ld\n"
+				"check not wanted=%ld\n"
+				"eeh_total_mmio_ffs=%ld\n"
+				"eeh_false_positives=%ld\n"
+				"eeh_ignored_failures=%ld\n"
+				"eeh_slot_resets=%ld\n",
+				no_dev, no_dn, no_cfg, no_check,
+				ffs, positives, failures, resets);
+	}
+
+	return 0;
+}
+
+static int proc_eeh_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, proc_eeh_show, NULL);
+}
+
+static struct file_operations proc_eeh_operations = {
+	.open      = proc_eeh_open,
+	.read      = seq_read,
+	.llseek    = seq_lseek,
+	.release   = single_release,
+};
+
+static int __init eeh_init_proc(void)
+{
+	struct proc_dir_entry *e;
+
+	if (platform_is_pseries()) {
+		e = create_proc_entry("ppc64/eeh", 0, NULL);
+		if (e)
+			e->proc_fops = &proc_eeh_operations;
+	}
+
+	return 0;
+}
+__initcall(eeh_init_proc);
diff --git a/arch/powerpc/platforms/pseries/eeh_event.c b/arch/powerpc/platforms/pseries/eeh_event.c
new file mode 100644
index 00000000000..92497333c2b
--- /dev/null
+++ b/arch/powerpc/platforms/pseries/eeh_event.c
@@ -0,0 +1,155 @@
+/*
+ * eeh_event.c
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
+ *
+ * Copyright (c) 2005 Linas Vepstas <linas@linas.org>
+ */
+
+#include <linux/list.h>
+#include <linux/pci.h>
+#include <asm/eeh_event.h>
+
+/** Overview:
+ *  EEH error states may be detected within exception handlers;
+ *  however, the recovery processing needs to occur asynchronously
+ *  in a normal kernel context and not an interrupt context.
+ *  This pair of routines creates an event and queues it onto a
+ *  work-queue, where a worker thread can drive recovery.
+ */
+
+/* EEH event workqueue setup. */
+static spinlock_t eeh_eventlist_lock = SPIN_LOCK_UNLOCKED;
+LIST_HEAD(eeh_eventlist);
+static void eeh_thread_launcher(void *);
+DECLARE_WORK(eeh_event_wq, eeh_thread_launcher, NULL);
+
+/**
+ * eeh_panic - call panic() for an eeh event that cannot be handled.
+ * The philosophy of this routine is that it is better to panic and
+ * halt the OS than it is to risk possible data corruption by
+ * oblivious device drivers that don't know better.
+ *
+ * @dev pci device that had an eeh event
+ * @reset_state current reset state of the device slot
+ */
+static void eeh_panic(struct pci_dev *dev, int reset_state)
+{
+	/*
+	 * Since the panic_on_oops sysctl is used to halt the system
+	 * in light of potential corruption, we can use it here.
+	 */
+	if (panic_on_oops) {
+		panic("EEH: MMIO failure (%d) on device:%s\n", reset_state,
+		      pci_name(dev));
+	}
+	else {
+		printk(KERN_INFO "EEH: Ignored MMIO failure (%d) on device:%s\n",
+		       reset_state, pci_name(dev));
+	}
+}
+
+/**
+ * eeh_event_handler - dispatch EEH events.  The detection of a frozen
+ * slot can occur inside an interrupt, where it can be hard to do
+ * anything about it.  The goal of this routine is to pull these
+ * detection events out of the context of the interrupt handler, and
+ * re-dispatch them for processing at a later time in a normal context.
+ *
+ * @dummy - unused
+ */
+static int eeh_event_handler(void * dummy)
+{
+	unsigned long flags;
+	struct eeh_event	*event;
+
+	daemonize ("eehd");
+
+	while (1) {
+		set_current_state(TASK_INTERRUPTIBLE);
+
+		spin_lock_irqsave(&eeh_eventlist_lock, flags);
+		event = NULL;
+		if (!list_empty(&eeh_eventlist)) {
+			event = list_entry(eeh_eventlist.next, struct eeh_event, list);
+			list_del(&event->list);
+		}
+		spin_unlock_irqrestore(&eeh_eventlist_lock, flags);
+		if (event == NULL)
+			break;
+
+		printk(KERN_INFO "EEH: Detected PCI bus error on device %s\n",
+		       pci_name(event->dev));
+
+		eeh_panic (event->dev, event->state);
+
+		kfree(event);
+	}
+
+	return 0;
+}
+
+/**
+ * eeh_thread_launcher
+ *
+ * @dummy - unused
+ */
+static void eeh_thread_launcher(void *dummy)
+{
+	if (kernel_thread(eeh_event_handler, NULL, CLONE_KERNEL) < 0)
+		printk(KERN_ERR "Failed to start EEH daemon\n");
+}
+
+/**
+ * eeh_send_failure_event - generate a PCI error event
+ * @dev pci device
+ *
+ * This routine can be called within an interrupt context;
+ * the actual event will be delivered in a normal context
+ * (from a workqueue).
+ */
+int eeh_send_failure_event (struct device_node *dn,
+                            struct pci_dev *dev,
+                            int state,
+                            int time_unavail)
+{
+	unsigned long flags;
+	struct eeh_event *event;
+
+	event = kmalloc(sizeof(*event), GFP_ATOMIC);
+	if (event == NULL) {
+		printk (KERN_ERR "EEH: out of memory, event not handled\n");
+		return 1;
+ 	}
+
+	if (dev)
+		pci_dev_get(dev);
+
+	event->dn = dn;
+	event->dev = dev;
+	event->state = state;
+	event->time_unavail = time_unavail;
+
+	/* We may or may not be called in an interrupt context */
+	spin_lock_irqsave(&eeh_eventlist_lock, flags);
+	list_add(&event->list, &eeh_eventlist);
+	spin_unlock_irqrestore(&eeh_eventlist_lock, flags);
+
+	schedule_work(&eeh_event_wq);
+
+	return 0;
+}
+
+/********************** END OF FILE ******************************/
diff --git a/arch/powerpc/platforms/pseries/iommu.c b/arch/powerpc/platforms/pseries/iommu.c
index fcc50bfd43f..97ba5214417 100644
--- a/arch/powerpc/platforms/pseries/iommu.c
+++ b/arch/powerpc/platforms/pseries/iommu.c
@@ -42,7 +42,6 @@
 #include <asm/machdep.h>
 #include <asm/abs_addr.h>
 #include <asm/pSeries_reconfig.h>
-#include <asm/systemcfg.h>
 #include <asm/firmware.h>
 #include <asm/tce.h>
 #include <asm/ppc-pci.h>
@@ -582,7 +581,7 @@ void iommu_init_early_pSeries(void)
 		return;
 	}
 
-	if (systemcfg->platform & PLATFORM_LPAR) {
+	if (platform_is_lpar()) {
 		if (firmware_has_feature(FW_FEATURE_MULTITCE)) {
 			ppc_md.tce_build = tce_buildmulti_pSeriesLP;
 			ppc_md.tce_free	 = tce_freemulti_pSeriesLP;
diff --git a/arch/powerpc/platforms/pseries/pci.c b/arch/powerpc/platforms/pseries/pci.c
index c198656a3bb..999a9620b5c 100644
--- a/arch/powerpc/platforms/pseries/pci.c
+++ b/arch/powerpc/platforms/pseries/pci.c
@@ -107,7 +107,6 @@ static void __init pSeries_request_regions(void)
 
 void __init pSeries_final_fixup(void)
 {
-	phbs_remap_io();
 	pSeries_request_regions();
 
 	pci_addr_cache_build();
@@ -123,7 +122,7 @@ static void fixup_winbond_82c105(struct pci_dev* dev)
 	int i;
 	unsigned int reg;
 
-	if (!(systemcfg->platform & PLATFORM_PSERIES))
+	if (!platform_is_pseries())
 		return;
 
 	printk("Using INTC for W82c105 IDE controller.\n");
diff --git a/arch/powerpc/platforms/pseries/reconfig.c b/arch/powerpc/platforms/pseries/reconfig.c
index d7d40033945..d8864164dbe 100644
--- a/arch/powerpc/platforms/pseries/reconfig.c
+++ b/arch/powerpc/platforms/pseries/reconfig.c
@@ -408,7 +408,7 @@ static int proc_ppc64_create_ofdt(void)
 {
 	struct proc_dir_entry *ent;
 
-	if (!(systemcfg->platform & PLATFORM_PSERIES))
+	if (!platform_is_pseries())
 		return 0;
 
 	ent = create_proc_entry("ppc64/ofdt", S_IWUSR, NULL);
diff --git a/arch/powerpc/platforms/pseries/rtasd.c b/arch/powerpc/platforms/pseries/rtasd.c
index e26b0420b6d..00cf331a1dc 100644
--- a/arch/powerpc/platforms/pseries/rtasd.c
+++ b/arch/powerpc/platforms/pseries/rtasd.c
@@ -482,10 +482,12 @@ static int __init rtas_init(void)
 {
 	struct proc_dir_entry *entry;
 
-	/* No RTAS, only warn if we are on a pSeries box  */
+	if (!platform_is_pseries())
+		return 0;
+
+	/* No RTAS */
 	if (rtas_token("event-scan") == RTAS_UNKNOWN_SERVICE) {
-		if (systemcfg->platform & PLATFORM_PSERIES)
-			printk(KERN_INFO "rtasd: no event-scan on system\n");
+		printk(KERN_INFO "rtasd: no event-scan on system\n");
 		return 1;
 	}
 
diff --git a/arch/powerpc/platforms/pseries/scanlog.c b/arch/powerpc/platforms/pseries/scanlog.c
new file mode 100644
index 00000000000..2edc947f7c4
--- /dev/null
+++ b/arch/powerpc/platforms/pseries/scanlog.c
@@ -0,0 +1,235 @@
+/*
+ *  c 2001 PPC 64 Team, IBM Corp
+ *
+ *      This program is free software; you can redistribute it and/or
+ *      modify it under the terms of the GNU General Public License
+ *      as published by the Free Software Foundation; either version
+ *      2 of the License, or (at your option) any later version.
+ *
+ * scan-log-data driver for PPC64  Todd Inglett <tinglett@vnet.ibm.com>
+ *
+ * When ppc64 hardware fails the service processor dumps internal state
+ * of the system.  After a reboot the operating system can access a dump
+ * of this data using this driver.  A dump exists if the device-tree
+ * /chosen/ibm,scan-log-data property exists.
+ *
+ * This driver exports /proc/ppc64/scan-log-dump which can be read.
+ * The driver supports only sequential reads.
+ *
+ * The driver looks at a write to the driver for the single word "reset".
+ * If given, the driver will reset the scanlog so the platform can free it.
+ */
+
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/errno.h>
+#include <linux/proc_fs.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <asm/uaccess.h>
+#include <asm/rtas.h>
+#include <asm/prom.h>
+
+#define MODULE_VERS "1.0"
+#define MODULE_NAME "scanlog"
+
+/* Status returns from ibm,scan-log-dump */
+#define SCANLOG_COMPLETE 0
+#define SCANLOG_HWERROR -1
+#define SCANLOG_CONTINUE 1
+
+#define DEBUG(A...) do { if (scanlog_debug) printk(KERN_ERR "scanlog: " A); } while (0)
+
+static int scanlog_debug;
+static unsigned int ibm_scan_log_dump;			/* RTAS token */
+static struct proc_dir_entry *proc_ppc64_scan_log_dump;	/* The proc file */
+
+static ssize_t scanlog_read(struct file *file, char __user *buf,
+			    size_t count, loff_t *ppos)
+{
+        struct inode * inode = file->f_dentry->d_inode;
+	struct proc_dir_entry *dp;
+	unsigned int *data;
+	int status;
+	unsigned long len, off;
+	unsigned int wait_time;
+
+        dp = PDE(inode);
+ 	data = (unsigned int *)dp->data;
+
+	if (!data) {
+		printk(KERN_ERR "scanlog: read failed no data\n");
+		return -EIO;
+	}
+
+	if (count > RTAS_DATA_BUF_SIZE)
+		count = RTAS_DATA_BUF_SIZE;
+
+	if (count < 1024) {
+		/* This is the min supported by this RTAS call.  Rather
+		 * than do all the buffering we insist the user code handle
+		 * larger reads.  As long as cp works... :)
+		 */
+		printk(KERN_ERR "scanlog: cannot perform a small read (%ld)\n", count);
+		return -EINVAL;
+	}
+
+	if (!access_ok(VERIFY_WRITE, buf, count))
+		return -EFAULT;
+
+	for (;;) {
+		wait_time = 500;	/* default wait if no data */
+		spin_lock(&rtas_data_buf_lock);
+		memcpy(rtas_data_buf, data, RTAS_DATA_BUF_SIZE);
+		status = rtas_call(ibm_scan_log_dump, 2, 1, NULL,
+				   (u32) __pa(rtas_data_buf), (u32) count);
+		memcpy(data, rtas_data_buf, RTAS_DATA_BUF_SIZE);
+		spin_unlock(&rtas_data_buf_lock);
+
+		DEBUG("status=%d, data[0]=%x, data[1]=%x, data[2]=%x\n",
+		      status, data[0], data[1], data[2]);
+		switch (status) {
+		    case SCANLOG_COMPLETE:
+			DEBUG("hit eof\n");
+			return 0;
+		    case SCANLOG_HWERROR:
+			DEBUG("hardware error reading scan log data\n");
+			return -EIO;
+		    case SCANLOG_CONTINUE:
+			/* We may or may not have data yet */
+			len = data[1];
+			off = data[2];
+			if (len > 0) {
+				if (copy_to_user(buf, ((char *)data)+off, len))
+					return -EFAULT;
+				return len;
+			}
+			/* Break to sleep default time */
+			break;
+		    default:
+			if (status > 9900 && status <= 9905) {
+				wait_time = rtas_extended_busy_delay_time(status);
+			} else {
+				printk(KERN_ERR "scanlog: unknown error from rtas: %d\n", status);
+				return -EIO;
+			}
+		}
+		/* Apparently no data yet.  Wait and try again. */
+		msleep_interruptible(wait_time);
+	}
+	/*NOTREACHED*/
+}
+
+static ssize_t scanlog_write(struct file * file, const char __user * buf,
+			     size_t count, loff_t *ppos)
+{
+	char stkbuf[20];
+	int status;
+
+	if (count > 19) count = 19;
+	if (copy_from_user (stkbuf, buf, count)) {
+		return -EFAULT;
+	}
+	stkbuf[count] = 0;
+
+	if (buf) {
+		if (strncmp(stkbuf, "reset", 5) == 0) {
+			DEBUG("reset scanlog\n");
+			status = rtas_call(ibm_scan_log_dump, 2, 1, NULL, 0, 0);
+			DEBUG("rtas returns %d\n", status);
+		} else if (strncmp(stkbuf, "debugon", 7) == 0) {
+			printk(KERN_ERR "scanlog: debug on\n");
+			scanlog_debug = 1;
+		} else if (strncmp(stkbuf, "debugoff", 8) == 0) {
+			printk(KERN_ERR "scanlog: debug off\n");
+			scanlog_debug = 0;
+		}
+	}
+	return count;
+}
+
+static int scanlog_open(struct inode * inode, struct file * file)
+{
+	struct proc_dir_entry *dp = PDE(inode);
+	unsigned int *data = (unsigned int *)dp->data;
+
+	if (!data) {
+		printk(KERN_ERR "scanlog: open failed no data\n");
+		return -EIO;
+	}
+
+	if (data[0] != 0) {
+		/* This imperfect test stops a second copy of the
+		 * data (or a reset while data is being copied)
+		 */
+		return -EBUSY;
+	}
+
+	data[0] = 0;	/* re-init so we restart the scan */
+
+	return 0;
+}
+
+static int scanlog_release(struct inode * inode, struct file * file)
+{
+	struct proc_dir_entry *dp = PDE(inode);
+	unsigned int *data = (unsigned int *)dp->data;
+
+	if (!data) {
+		printk(KERN_ERR "scanlog: release failed no data\n");
+		return -EIO;
+	}
+	data[0] = 0;
+
+	return 0;
+}
+
+struct file_operations scanlog_fops = {
+	.owner		= THIS_MODULE,
+	.read		= scanlog_read,
+	.write		= scanlog_write,
+	.open		= scanlog_open,
+	.release	= scanlog_release,
+};
+
+int __init scanlog_init(void)
+{
+	struct proc_dir_entry *ent;
+
+	ibm_scan_log_dump = rtas_token("ibm,scan-log-dump");
+	if (ibm_scan_log_dump == RTAS_UNKNOWN_SERVICE) {
+		printk(KERN_ERR "scan-log-dump not implemented on this system\n");
+		return -EIO;
+	}
+
+        ent = create_proc_entry("ppc64/rtas/scan-log-dump",  S_IRUSR, NULL);
+	if (ent) {
+		ent->proc_fops = &scanlog_fops;
+		/* Ideally we could allocate a buffer < 4G */
+		ent->data = kmalloc(RTAS_DATA_BUF_SIZE, GFP_KERNEL);
+		if (!ent->data) {
+			printk(KERN_ERR "Failed to allocate a buffer\n");
+			remove_proc_entry("scan-log-dump", ent->parent);
+			return -ENOMEM;
+		}
+		((unsigned int *)ent->data)[0] = 0;
+	} else {
+		printk(KERN_ERR "Failed to create ppc64/scan-log-dump proc entry\n");
+		return -EIO;
+	}
+	proc_ppc64_scan_log_dump = ent;
+
+	return 0;
+}
+
+void __exit scanlog_cleanup(void)
+{
+	if (proc_ppc64_scan_log_dump) {
+		kfree(proc_ppc64_scan_log_dump->data);
+		remove_proc_entry("scan-log-dump", proc_ppc64_scan_log_dump->parent);
+	}
+}
+
+module_init(scanlog_init);
+module_exit(scanlog_cleanup);
+MODULE_LICENSE("GPL");
diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c
index a093a0d4dd6..e94247c28d4 100644
--- a/arch/powerpc/platforms/pseries/setup.c
+++ b/arch/powerpc/platforms/pseries/setup.c
@@ -249,7 +249,7 @@ static void __init pSeries_setup_arch(void)
 		ppc_md.idle_loop = default_idle;
 	}
 
-	if (systemcfg->platform & PLATFORM_LPAR)
+	if (platform_is_lpar())
 		ppc_md.enable_pmcs = pseries_lpar_enable_pmcs;
 	else
 		ppc_md.enable_pmcs = power4_enable_pmcs;
@@ -378,7 +378,7 @@ static void __init pSeries_init_early(void)
 
 	fw_feature_init();
 	
-	if (systemcfg->platform & PLATFORM_LPAR)
+	if (platform_is_lpar())
 		hpte_init_lpar();
 	else {
 		hpte_init_native();
@@ -388,7 +388,7 @@ static void __init pSeries_init_early(void)
 
 	generic_find_legacy_serial_ports(&physport, &default_speed);
 
-	if (systemcfg->platform & PLATFORM_LPAR)
+	if (platform_is_lpar())
 		find_udbg_vterm();
 	else if (physport) {
 		/* Map the uart for udbg. */
@@ -592,7 +592,7 @@ static void pseries_shared_idle(void)
 
 static int pSeries_pci_probe_mode(struct pci_bus *bus)
 {
-	if (systemcfg->platform & PLATFORM_LPAR)
+	if (platform_is_lpar())
 		return PCI_PROBE_DEVTREE;
 	return PCI_PROBE_NORMAL;
 }
diff --git a/arch/powerpc/platforms/pseries/smp.c b/arch/powerpc/platforms/pseries/smp.c
index 7a243e8ccd7..3ba794ca328 100644
--- a/arch/powerpc/platforms/pseries/smp.c
+++ b/arch/powerpc/platforms/pseries/smp.c
@@ -46,6 +46,7 @@
 #include <asm/rtas.h>
 #include <asm/pSeries_reconfig.h>
 #include <asm/mpic.h>
+#include <asm/systemcfg.h>
 
 #include "plpar_wrappers.h"
 
@@ -96,7 +97,7 @@ int pSeries_cpu_disable(void)
 	int cpu = smp_processor_id();
 
 	cpu_clear(cpu, cpu_online_map);
-	systemcfg->processorCount--;
+	_systemcfg->processorCount--;
 
 	/*fix boot_cpuid here*/
 	if (cpu == boot_cpuid)
@@ -441,7 +442,7 @@ void __init smp_init_pSeries(void)
 	smp_ops->cpu_die = pSeries_cpu_die;
 
 	/* Processors can be added/removed only on LPAR */
-	if (systemcfg->platform == PLATFORM_PSERIES_LPAR)
+	if (platform_is_lpar())
 		pSeries_reconfig_notifier_register(&pSeries_smp_nb);
 #endif
 
diff --git a/arch/powerpc/platforms/pseries/xics.c b/arch/powerpc/platforms/pseries/xics.c
index c72c86f05cb..72ac18067ec 100644
--- a/arch/powerpc/platforms/pseries/xics.c
+++ b/arch/powerpc/platforms/pseries/xics.c
@@ -545,7 +545,9 @@ nextnode:
 		of_node_put(np);
 	}
 
-	if (systemcfg->platform == PLATFORM_PSERIES) {
+	if (platform_is_lpar())
+		ops = &pSeriesLP_ops;
+	else {
 #ifdef CONFIG_SMP
 		for_each_cpu(i) {
 			int hard_id;
@@ -561,12 +563,11 @@ nextnode:
 #else
 		xics_per_cpu[0] = ioremap(intr_base, intr_size);
 #endif /* CONFIG_SMP */
-	} else if (systemcfg->platform == PLATFORM_PSERIES_LPAR) {
-		ops = &pSeriesLP_ops;
 	}
 
 	xics_8259_pic.enable = i8259_pic.enable;
 	xics_8259_pic.disable = i8259_pic.disable;
+	xics_8259_pic.end = i8259_pic.end;
 	for (i = 0; i < 16; ++i)
 		get_irq_desc(i)->handler = &xics_8259_pic;
 	for (; i < NR_IRQS; ++i)
diff --git a/arch/powerpc/sysdev/u3_iommu.c b/arch/powerpc/sysdev/u3_iommu.c
index 543d6590981..f32baf7f469 100644
--- a/arch/powerpc/sysdev/u3_iommu.c
+++ b/arch/powerpc/sysdev/u3_iommu.c
@@ -226,7 +226,7 @@ static void iommu_table_u3_setup(void)
 	iommu_table_u3.it_busno = 0;
 	iommu_table_u3.it_offset = 0;
 	/* it_size is in number of entries */
-	iommu_table_u3.it_size = dart_tablesize / sizeof(u32);
+	iommu_table_u3.it_size = (dart_tablesize / sizeof(u32)) >> DART_PAGE_FACTOR;
 
 	/* Initialize the common IOMMU code */
 	iommu_table_u3.it_base = (unsigned long)dart_vbase;
diff --git a/arch/powerpc/xmon/Makefile b/arch/powerpc/xmon/Makefile
index 79a784f0e7a..b20312e5ed2 100644
--- a/arch/powerpc/xmon/Makefile
+++ b/arch/powerpc/xmon/Makefile
@@ -8,4 +8,4 @@ obj-$(CONFIG_8xx)	+= start_8xx.o
 obj-$(CONFIG_6xx)	+= start_32.o
 obj-$(CONFIG_4xx)	+= start_32.o
 obj-$(CONFIG_PPC64)	+= start_64.o
-obj-y			+= xmon.o ppc-dis.o ppc-opc.o subr_prf.o setjmp.o
+obj-y			+= xmon.o ppc-dis.o ppc-opc.o setjmp.o nonstdio.o
diff --git a/arch/powerpc/xmon/nonstdio.c b/arch/powerpc/xmon/nonstdio.c
new file mode 100644
index 00000000000..78765833f4c
--- /dev/null
+++ b/arch/powerpc/xmon/nonstdio.c
@@ -0,0 +1,134 @@
+/*
+ * Copyright (C) 1996-2005 Paul Mackerras.
+ *
+ *      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/string.h>
+#include <asm/time.h>
+#include "nonstdio.h"
+
+int xmon_putchar(int c)
+{
+	char ch = c;
+
+	if (c == '\n')
+		xmon_putchar('\r');
+	return xmon_write(&ch, 1) == 1? c: -1;
+}
+
+static char line[256];
+static char *lineptr;
+static int lineleft;
+
+int xmon_expect(const char *str, unsigned long timeout)
+{
+	int c;
+	unsigned long t0;
+
+	/* assume 25MHz default timebase if tb_ticks_per_sec not set yet */
+	timeout *= tb_ticks_per_sec? tb_ticks_per_sec: 25000000;
+	t0 = get_tbl();
+	do {
+		lineptr = line;
+		for (;;) {
+			c = xmon_read_poll();
+			if (c == -1) {
+				if (get_tbl() - t0 > timeout)
+					return 0;
+				continue;
+			}
+			if (c == '\n')
+				break;
+			if (c != '\r' && lineptr < &line[sizeof(line) - 1])
+				*lineptr++ = c;
+		}
+		*lineptr = 0;
+	} while (strstr(line, str) == NULL);
+	return 1;
+}
+
+int xmon_getchar(void)
+{
+	int c;
+
+	if (lineleft == 0) {
+		lineptr = line;
+		for (;;) {
+			c = xmon_readchar();
+			if (c == -1 || c == 4)
+				break;
+			if (c == '\r' || c == '\n') {
+				*lineptr++ = '\n';
+				xmon_putchar('\n');
+				break;
+			}
+			switch (c) {
+			case 0177:
+			case '\b':
+				if (lineptr > line) {
+					xmon_putchar('\b');
+					xmon_putchar(' ');
+					xmon_putchar('\b');
+					--lineptr;
+				}
+				break;
+			case 'U' & 0x1F:
+				while (lineptr > line) {
+					xmon_putchar('\b');
+					xmon_putchar(' ');
+					xmon_putchar('\b');
+					--lineptr;
+				}
+				break;
+			default:
+				if (lineptr >= &line[sizeof(line) - 1])
+					xmon_putchar('\a');
+				else {
+					xmon_putchar(c);
+					*lineptr++ = c;
+				}
+			}
+		}
+		lineleft = lineptr - line;
+		lineptr = line;
+	}
+	if (lineleft == 0)
+		return -1;
+	--lineleft;
+	return *lineptr++;
+}
+
+char *xmon_gets(char *str, int nb)
+{
+	char *p;
+	int c;
+
+	for (p = str; p < str + nb - 1; ) {
+		c = xmon_getchar();
+		if (c == -1) {
+			if (p == str)
+				return NULL;
+			break;
+		}
+		*p++ = c;
+		if (c == '\n')
+			break;
+	}
+	*p = 0;
+	return str;
+}
+
+void xmon_printf(const char *format, ...)
+{
+	va_list args;
+	int n;
+	static char xmon_outbuf[1024];
+
+	va_start(args, format);
+	n = vsnprintf(xmon_outbuf, sizeof(xmon_outbuf), format, args);
+	va_end(args);
+	xmon_write(xmon_outbuf, n);
+}
diff --git a/arch/powerpc/xmon/nonstdio.h b/arch/powerpc/xmon/nonstdio.h
index 84211a21c6f..47cebbd2b1b 100644
--- a/arch/powerpc/xmon/nonstdio.h
+++ b/arch/powerpc/xmon/nonstdio.h
@@ -1,22 +1,14 @@
-typedef int	FILE;
-extern FILE *xmon_stdin, *xmon_stdout;
 #define EOF	(-1)
-#define stdin	xmon_stdin
-#define stdout	xmon_stdout
+
 #define printf	xmon_printf
-#define fprintf	xmon_fprintf
-#define fputs	xmon_fputs
-#define fgets	xmon_fgets
 #define putchar	xmon_putchar
-#define getchar	xmon_getchar
-#define putc	xmon_putc
-#define getc	xmon_getc
-#define fopen(n, m)	NULL
-#define fflush(f)	do {} while (0)
-#define fclose(f)	do {} while (0)
-extern char *fgets(char *, int, void *);
-extern void xmon_printf(const char *, ...);
-extern void xmon_fprintf(void *, const char *, ...);
-extern void xmon_sprintf(char *, const char *, ...);
 
-#define perror(s)	printf("%s: no files!\n", (s))
+extern int xmon_putchar(int c);
+extern int xmon_getchar(void);
+extern char *xmon_gets(char *, int);
+extern void xmon_printf(const char *, ...);
+extern void xmon_map_scc(void);
+extern int xmon_expect(const char *str, unsigned long timeout);
+extern int xmon_write(void *ptr, int nb);
+extern int xmon_readchar(void);
+extern int xmon_read_poll(void);
diff --git a/arch/powerpc/xmon/setjmp.S b/arch/powerpc/xmon/setjmp.S
index f8e40dfd2bf..96a91f10e2e 100644
--- a/arch/powerpc/xmon/setjmp.S
+++ b/arch/powerpc/xmon/setjmp.S
@@ -14,61 +14,61 @@
 
 _GLOBAL(xmon_setjmp)
 	mflr	r0
-	STL	r0,0(r3)
-	STL	r1,SZL(r3)
-	STL	r2,2*SZL(r3)
+	PPC_STL	r0,0(r3)
+	PPC_STL	r1,SZL(r3)
+	PPC_STL	r2,2*SZL(r3)
 	mfcr	r0
-	STL	r0,3*SZL(r3)
-	STL	r13,4*SZL(r3)
-	STL	r14,5*SZL(r3)
-	STL	r15,6*SZL(r3)
-	STL	r16,7*SZL(r3)
-	STL	r17,8*SZL(r3)
-	STL	r18,9*SZL(r3)
-	STL	r19,10*SZL(r3)
-	STL	r20,11*SZL(r3)
-	STL	r21,12*SZL(r3)
-	STL	r22,13*SZL(r3)
-	STL	r23,14*SZL(r3)
-	STL	r24,15*SZL(r3)
-	STL	r25,16*SZL(r3)
-	STL	r26,17*SZL(r3)
-	STL	r27,18*SZL(r3)
-	STL	r28,19*SZL(r3)
-	STL	r29,20*SZL(r3)
-	STL	r30,21*SZL(r3)
-	STL	r31,22*SZL(r3)
+	PPC_STL	r0,3*SZL(r3)
+	PPC_STL	r13,4*SZL(r3)
+	PPC_STL	r14,5*SZL(r3)
+	PPC_STL	r15,6*SZL(r3)
+	PPC_STL	r16,7*SZL(r3)
+	PPC_STL	r17,8*SZL(r3)
+	PPC_STL	r18,9*SZL(r3)
+	PPC_STL	r19,10*SZL(r3)
+	PPC_STL	r20,11*SZL(r3)
+	PPC_STL	r21,12*SZL(r3)
+	PPC_STL	r22,13*SZL(r3)
+	PPC_STL	r23,14*SZL(r3)
+	PPC_STL	r24,15*SZL(r3)
+	PPC_STL	r25,16*SZL(r3)
+	PPC_STL	r26,17*SZL(r3)
+	PPC_STL	r27,18*SZL(r3)
+	PPC_STL	r28,19*SZL(r3)
+	PPC_STL	r29,20*SZL(r3)
+	PPC_STL	r30,21*SZL(r3)
+	PPC_STL	r31,22*SZL(r3)
 	li	r3,0
 	blr
 
 _GLOBAL(xmon_longjmp)
-	CMPI	r4,0
+	PPC_LCMPI r4,0
 	bne	1f
 	li	r4,1
-1:	LDL	r13,4*SZL(r3)
-	LDL	r14,5*SZL(r3)
-	LDL	r15,6*SZL(r3)
-	LDL	r16,7*SZL(r3)
-	LDL	r17,8*SZL(r3)
-	LDL	r18,9*SZL(r3)
-	LDL	r19,10*SZL(r3)
-	LDL	r20,11*SZL(r3)
-	LDL	r21,12*SZL(r3)
-	LDL	r22,13*SZL(r3)
-	LDL	r23,14*SZL(r3)
-	LDL	r24,15*SZL(r3)
-	LDL	r25,16*SZL(r3)
-	LDL	r26,17*SZL(r3)
-	LDL	r27,18*SZL(r3)
-	LDL	r28,19*SZL(r3)
-	LDL	r29,20*SZL(r3)
-	LDL	r30,21*SZL(r3)
-	LDL	r31,22*SZL(r3)
-	LDL	r0,3*SZL(r3)
+1:	PPC_LL	r13,4*SZL(r3)
+	PPC_LL	r14,5*SZL(r3)
+	PPC_LL	r15,6*SZL(r3)
+	PPC_LL	r16,7*SZL(r3)
+	PPC_LL	r17,8*SZL(r3)
+	PPC_LL	r18,9*SZL(r3)
+	PPC_LL	r19,10*SZL(r3)
+	PPC_LL	r20,11*SZL(r3)
+	PPC_LL	r21,12*SZL(r3)
+	PPC_LL	r22,13*SZL(r3)
+	PPC_LL	r23,14*SZL(r3)
+	PPC_LL	r24,15*SZL(r3)
+	PPC_LL	r25,16*SZL(r3)
+	PPC_LL	r26,17*SZL(r3)
+	PPC_LL	r27,18*SZL(r3)
+	PPC_LL	r28,19*SZL(r3)
+	PPC_LL	r29,20*SZL(r3)
+	PPC_LL	r30,21*SZL(r3)
+	PPC_LL	r31,22*SZL(r3)
+	PPC_LL	r0,3*SZL(r3)
 	mtcrf	0x38,r0
-	LDL	r0,0(r3)
-	LDL	r1,SZL(r3)
-	LDL	r2,2*SZL(r3)
+	PPC_LL	r0,0(r3)
+	PPC_LL	r1,SZL(r3)
+	PPC_LL	r2,2*SZL(r3)
 	mtlr	r0
 	mr	r3,r4
 	blr
@@ -84,52 +84,52 @@ _GLOBAL(xmon_longjmp)
  * different ABIs, though).
  */
 _GLOBAL(xmon_save_regs)
-	STL	r0,0*SZL(r3)
-	STL	r2,2*SZL(r3)
-	STL	r3,3*SZL(r3)
-	STL	r4,4*SZL(r3)
-	STL	r5,5*SZL(r3)
-	STL	r6,6*SZL(r3)
-	STL	r7,7*SZL(r3)
-	STL	r8,8*SZL(r3)
-	STL	r9,9*SZL(r3)
-	STL	r10,10*SZL(r3)
-	STL	r11,11*SZL(r3)
-	STL	r12,12*SZL(r3)
-	STL	r13,13*SZL(r3)
-	STL	r14,14*SZL(r3)
-	STL	r15,15*SZL(r3)
-	STL	r16,16*SZL(r3)
-	STL	r17,17*SZL(r3)
-	STL	r18,18*SZL(r3)
-	STL	r19,19*SZL(r3)
-	STL	r20,20*SZL(r3)
-	STL	r21,21*SZL(r3)
-	STL	r22,22*SZL(r3)
-	STL	r23,23*SZL(r3)
-	STL	r24,24*SZL(r3)
-	STL	r25,25*SZL(r3)
-	STL	r26,26*SZL(r3)
-	STL	r27,27*SZL(r3)
-	STL	r28,28*SZL(r3)
-	STL	r29,29*SZL(r3)
-	STL	r30,30*SZL(r3)
-	STL	r31,31*SZL(r3)
+	PPC_STL	r0,0*SZL(r3)
+	PPC_STL	r2,2*SZL(r3)
+	PPC_STL	r3,3*SZL(r3)
+	PPC_STL	r4,4*SZL(r3)
+	PPC_STL	r5,5*SZL(r3)
+	PPC_STL	r6,6*SZL(r3)
+	PPC_STL	r7,7*SZL(r3)
+	PPC_STL	r8,8*SZL(r3)
+	PPC_STL	r9,9*SZL(r3)
+	PPC_STL	r10,10*SZL(r3)
+	PPC_STL	r11,11*SZL(r3)
+	PPC_STL	r12,12*SZL(r3)
+	PPC_STL	r13,13*SZL(r3)
+	PPC_STL	r14,14*SZL(r3)
+	PPC_STL	r15,15*SZL(r3)
+	PPC_STL	r16,16*SZL(r3)
+	PPC_STL	r17,17*SZL(r3)
+	PPC_STL	r18,18*SZL(r3)
+	PPC_STL	r19,19*SZL(r3)
+	PPC_STL	r20,20*SZL(r3)
+	PPC_STL	r21,21*SZL(r3)
+	PPC_STL	r22,22*SZL(r3)
+	PPC_STL	r23,23*SZL(r3)
+	PPC_STL	r24,24*SZL(r3)
+	PPC_STL	r25,25*SZL(r3)
+	PPC_STL	r26,26*SZL(r3)
+	PPC_STL	r27,27*SZL(r3)
+	PPC_STL	r28,28*SZL(r3)
+	PPC_STL	r29,29*SZL(r3)
+	PPC_STL	r30,30*SZL(r3)
+	PPC_STL	r31,31*SZL(r3)
 	/* go up one stack frame for SP */
-	LDL	r4,0(r1)
-	STL	r4,1*SZL(r3)
+	PPC_LL	r4,0(r1)
+	PPC_STL	r4,1*SZL(r3)
 	/* get caller's LR */
-	LDL	r0,LRSAVE(r4)
-	STL	r0,_NIP-STACK_FRAME_OVERHEAD(r3)
-	STL	r0,_LINK-STACK_FRAME_OVERHEAD(r3)
+	PPC_LL	r0,LRSAVE(r4)
+	PPC_STL	r0,_NIP-STACK_FRAME_OVERHEAD(r3)
+	PPC_STL	r0,_LINK-STACK_FRAME_OVERHEAD(r3)
 	mfmsr	r0
-	STL	r0,_MSR-STACK_FRAME_OVERHEAD(r3)
+	PPC_STL	r0,_MSR-STACK_FRAME_OVERHEAD(r3)
 	mfctr	r0
-	STL	r0,_CTR-STACK_FRAME_OVERHEAD(r3)
+	PPC_STL	r0,_CTR-STACK_FRAME_OVERHEAD(r3)
 	mfxer	r0
-	STL	r0,_XER-STACK_FRAME_OVERHEAD(r3)
+	PPC_STL	r0,_XER-STACK_FRAME_OVERHEAD(r3)
 	mfcr	r0
-	STL	r0,_CCR-STACK_FRAME_OVERHEAD(r3)
+	PPC_STL	r0,_CCR-STACK_FRAME_OVERHEAD(r3)
 	li	r0,0
-	STL	r0,_TRAP-STACK_FRAME_OVERHEAD(r3)
+	PPC_STL	r0,_TRAP-STACK_FRAME_OVERHEAD(r3)
 	blr
diff --git a/arch/powerpc/xmon/start_32.c b/arch/powerpc/xmon/start_32.c
index 69b658c0f76..c2464df4217 100644
--- a/arch/powerpc/xmon/start_32.c
+++ b/arch/powerpc/xmon/start_32.c
@@ -11,7 +11,6 @@
 #include <linux/cuda.h>
 #include <linux/kernel.h>
 #include <linux/errno.h>
-#include <linux/sysrq.h>
 #include <linux/bitops.h>
 #include <asm/xmon.h>
 #include <asm/prom.h>
@@ -22,10 +21,11 @@
 #include <asm/processor.h>
 #include <asm/delay.h>
 #include <asm/btext.h>
+#include <asm/time.h>
+#include "nonstdio.h"
 
 static volatile unsigned char __iomem *sccc, *sccd;
 unsigned int TXRDY, RXRDY, DLAB;
-static int xmon_expect(const char *str, unsigned int timeout);
 
 static int use_serial;
 static int use_screen;
@@ -33,16 +33,6 @@ static int via_modem;
 static int xmon_use_sccb;
 static struct device_node *channel_node;
 
-#define TB_SPEED	25000000
-
-static inline unsigned int readtb(void)
-{
-	unsigned int ret;
-
-	asm volatile("mftb %0" : "=r" (ret) :);
-	return ret;
-}
-
 void buf_access(void)
 {
 	if (DLAB)
@@ -91,23 +81,7 @@ static unsigned long chrp_find_phys_io_base(void)
 }
 #endif /* CONFIG_PPC_CHRP */
 
-#ifdef CONFIG_MAGIC_SYSRQ
-static void sysrq_handle_xmon(int key, struct pt_regs *regs,
-			      struct tty_struct *tty)
-{
-	xmon(regs);
-}
-
-static struct sysrq_key_op sysrq_xmon_op =
-{
-	.handler =	sysrq_handle_xmon,
-	.help_msg =	"Xmon",
-	.action_msg =	"Entering xmon",
-};
-#endif
-
-void
-xmon_map_scc(void)
+void xmon_map_scc(void)
 {
 #ifdef CONFIG_PPC_MULTIPLATFORM
 	volatile unsigned char __iomem *base;
@@ -217,8 +191,6 @@ xmon_map_scc(void)
 	RXRDY = 1;
 	DLAB = 0x80;
 #endif /* platform */
-
-	register_sysrq_key('x', &sysrq_xmon_op);
 }
 
 static int scc_initialized = 0;
@@ -238,8 +210,7 @@ static inline void do_poll_adb(void)
 #endif /* CONFIG_ADB_CUDA */
 }
 
-int
-xmon_write(void *handle, void *ptr, int nb)
+int xmon_write(void *ptr, int nb)
 {
 	char *p = ptr;
 	int i, c, ct;
@@ -311,8 +282,7 @@ static unsigned char xmon_shift_keytab[128] =
 	"\0.\0*\0+\0\0\0\0\0/\r\0-\0"			/* 0x40 - 0x4f */
 	"\0\0000123456789\0\0\0";			/* 0x50 - 0x5f */
 
-static int
-xmon_get_adb_key(void)
+static int xmon_get_adb_key(void)
 {
 	int k, t, on;
 
@@ -350,32 +320,21 @@ xmon_get_adb_key(void)
 }
 #endif /* CONFIG_BOOTX_TEXT */
 
-int
-xmon_read(void *handle, void *ptr, int nb)
+int xmon_readchar(void)
 {
-    char *p = ptr;
-    int i;
-
 #ifdef CONFIG_BOOTX_TEXT
-    if (use_screen) {
-	for (i = 0; i < nb; ++i)
-	    *p++ = xmon_get_adb_key();
-	return i;
-    }
+	if (use_screen)
+		return xmon_get_adb_key();
 #endif
-    if (!scc_initialized)
-	xmon_init_scc();
-    for (i = 0; i < nb; ++i) {
+	if (!scc_initialized)
+		xmon_init_scc();
 	while ((*sccc & RXRDY) == 0)
-	    do_poll_adb();
+		do_poll_adb();
 	buf_access();
-	*p++ = *sccd;
-    }
-    return i;
+	return *sccd;
 }
 
-int
-xmon_read_poll(void)
+int xmon_read_poll(void)
 {
 	if ((*sccc & RXRDY) == 0) {
 		do_poll_adb();
@@ -395,8 +354,7 @@ static unsigned char scc_inittab[] = {
     3,  0xc1,		/* rx enable, 8 bits */
 };
 
-void
-xmon_init_scc(void)
+void xmon_init_scc(void)
 {
 	if ( _machine == _MACH_chrp )
 	{
@@ -410,6 +368,7 @@ xmon_init_scc(void)
 	else if ( _machine == _MACH_Pmac )
 	{
 		int i, x;
+		unsigned long timeout;
 
 		if (channel_node != 0)
 			pmac_call_feature(
@@ -424,8 +383,12 @@ xmon_init_scc(void)
 				PMAC_FTR_MODEM_ENABLE,
 				channel_node, 0, 1);
 			printk(KERN_INFO "Modem powered up by debugger !\n");
-			t0 = readtb();
-			while (readtb() - t0 < 3*TB_SPEED)
+			t0 = get_tbl();
+			timeout = 3 * tb_ticks_per_sec;
+			if (timeout == 0)
+				/* assume 25MHz if tb_ticks_per_sec not set */
+				timeout = 75000000;
+			while (get_tbl() - t0 < timeout)
 				eieio();
 		}
 		/* use the B channel if requested */
@@ -447,164 +410,19 @@ xmon_init_scc(void)
 	scc_initialized = 1;
 	if (via_modem) {
 		for (;;) {
-			xmon_write(NULL, "ATE1V1\r", 7);
+			xmon_write("ATE1V1\r", 7);
 			if (xmon_expect("OK", 5)) {
-				xmon_write(NULL, "ATA\r", 4);
+				xmon_write("ATA\r", 4);
 				if (xmon_expect("CONNECT", 40))
 					break;
 			}
-			xmon_write(NULL, "+++", 3);
+			xmon_write("+++", 3);
 			xmon_expect("OK", 3);
 		}
 	}
 }
 
-void *xmon_stdin;
-void *xmon_stdout;
-void *xmon_stderr;
-
-int xmon_putc(int c, void *f)
-{
-    char ch = c;
-
-    if (c == '\n')
-	xmon_putc('\r', f);
-    return xmon_write(f, &ch, 1) == 1? c: -1;
-}
-
-int xmon_putchar(int c)
-{
-	return xmon_putc(c, xmon_stdout);
-}
-
-int xmon_fputs(char *str, void *f)
-{
-	int n = strlen(str);
-
-	return xmon_write(f, str, n) == n? 0: -1;
-}
-
-int
-xmon_readchar(void)
-{
-	char ch;
-
-	for (;;) {
-		switch (xmon_read(xmon_stdin, &ch, 1)) {
-		case 1:
-			return ch;
-		case -1:
-			xmon_printf("read(stdin) returned -1\r\n", 0, 0);
-			return -1;
-		}
-	}
-}
-
-static char line[256];
-static char *lineptr;
-static int lineleft;
-
-int xmon_expect(const char *str, unsigned int timeout)
-{
-	int c;
-	unsigned int t0;
-
-	timeout *= TB_SPEED;
-	t0 = readtb();
-	do {
-		lineptr = line;
-		for (;;) {
-			c = xmon_read_poll();
-			if (c == -1) {
-				if (readtb() - t0 > timeout)
-					return 0;
-				continue;
-			}
-			if (c == '\n')
-				break;
-			if (c != '\r' && lineptr < &line[sizeof(line) - 1])
-				*lineptr++ = c;
-		}
-		*lineptr = 0;
-	} while (strstr(line, str) == NULL);
-	return 1;
-}
-
-int
-xmon_getchar(void)
-{
-    int c;
-
-    if (lineleft == 0) {
-	lineptr = line;
-	for (;;) {
-	    c = xmon_readchar();
-	    if (c == -1 || c == 4)
-		break;
-	    if (c == '\r' || c == '\n') {
-		*lineptr++ = '\n';
-		xmon_putchar('\n');
-		break;
-	    }
-	    switch (c) {
-	    case 0177:
-	    case '\b':
-		if (lineptr > line) {
-		    xmon_putchar('\b');
-		    xmon_putchar(' ');
-		    xmon_putchar('\b');
-		    --lineptr;
-		}
-		break;
-	    case 'U' & 0x1F:
-		while (lineptr > line) {
-		    xmon_putchar('\b');
-		    xmon_putchar(' ');
-		    xmon_putchar('\b');
-		    --lineptr;
-		}
-		break;
-	    default:
-		if (lineptr >= &line[sizeof(line) - 1])
-		    xmon_putchar('\a');
-		else {
-		    xmon_putchar(c);
-		    *lineptr++ = c;
-		}
-	    }
-	}
-	lineleft = lineptr - line;
-	lineptr = line;
-    }
-    if (lineleft == 0)
-	return -1;
-    --lineleft;
-    return *lineptr++;
-}
-
-char *
-xmon_fgets(char *str, int nb, void *f)
-{
-    char *p;
-    int c;
-
-    for (p = str; p < str + nb - 1; ) {
-	c = xmon_getchar();
-	if (c == -1) {
-	    if (p == str)
-		return NULL;
-	    break;
-	}
-	*p++ = c;
-	if (c == '\n')
-	    break;
-    }
-    *p = 0;
-    return str;
-}
-
-void
-xmon_enter(void)
+void xmon_enter(void)
 {
 #ifdef CONFIG_ADB_PMU
 	if (_machine == _MACH_Pmac) {
@@ -613,8 +431,7 @@ xmon_enter(void)
 #endif
 }
 
-void
-xmon_leave(void)
+void xmon_leave(void)
 {
 #ifdef CONFIG_ADB_PMU
 	if (_machine == _MACH_Pmac) {
diff --git a/arch/powerpc/xmon/start_64.c b/arch/powerpc/xmon/start_64.c
index e50c158191e..712552c4f24 100644
--- a/arch/powerpc/xmon/start_64.c
+++ b/arch/powerpc/xmon/start_64.c
@@ -6,182 +6,29 @@
  *      as published by the Free Software Foundation; either version
  *      2 of the License, or (at your option) any later version.
  */
-#include <linux/config.h>
-#include <linux/string.h>
-#include <linux/kernel.h>
-#include <linux/errno.h>
-#include <linux/sysrq.h>
-#include <linux/init.h>
 #include <asm/machdep.h>
-#include <asm/io.h>
-#include <asm/page.h>
-#include <asm/prom.h>
-#include <asm/processor.h>
 #include <asm/udbg.h>
-#include <asm/system.h>
 #include "nonstdio.h"
 
-#ifdef CONFIG_MAGIC_SYSRQ
-
-static void sysrq_handle_xmon(int key, struct pt_regs *pt_regs,
-			      struct tty_struct *tty) 
-{
-	/* ensure xmon is enabled */
-	xmon_init(1);
-	debugger(pt_regs);
-}
-
-static struct sysrq_key_op sysrq_xmon_op = 
+void xmon_map_scc(void)
 {
-	.handler =	sysrq_handle_xmon,
-	.help_msg =	"Xmon",
-	.action_msg =	"Entering xmon",
-};
-
-static int __init setup_xmon_sysrq(void)
-{
-	register_sysrq_key('x', &sysrq_xmon_op);
-	return 0;
 }
-__initcall(setup_xmon_sysrq);
-#endif /* CONFIG_MAGIC_SYSRQ */
 
-int
-xmon_write(void *handle, void *ptr, int nb)
+int xmon_write(void *ptr, int nb)
 {
 	return udbg_write(ptr, nb);
 }
 
-int
-xmon_read(void *handle, void *ptr, int nb)
+int xmon_readchar(void)
 {
-	return udbg_read(ptr, nb);
+	if (udbg_getc)
+		return udbg_getc();
+	return -1;
 }
 
-int
-xmon_read_poll(void)
+int xmon_read_poll(void)
 {
 	if (udbg_getc_poll)
 		return udbg_getc_poll();
 	return -1;
 }
- 
-FILE *xmon_stdin;
-FILE *xmon_stdout;
-
-int
-xmon_putc(int c, void *f)
-{
-	char ch = c;
-
-	if (c == '\n')
-		xmon_putc('\r', f);
-	return xmon_write(f, &ch, 1) == 1? c: -1;
-}
-
-int
-xmon_putchar(int c)
-{
-	return xmon_putc(c, xmon_stdout);
-}
-
-int
-xmon_fputs(char *str, void *f)
-{
-	int n = strlen(str);
-
-	return xmon_write(f, str, n) == n? 0: -1;
-}
-
-int
-xmon_readchar(void)
-{
-	char ch;
-
-	for (;;) {
-		switch (xmon_read(xmon_stdin, &ch, 1)) {
-		case 1:
-			return ch;
-		case -1:
-			xmon_printf("read(stdin) returned -1\r\n", 0, 0);
-			return -1;
-		}
-	}
-}
-
-static char line[256];
-static char *lineptr;
-static int lineleft;
-
-int
-xmon_getchar(void)
-{
-	int c;
-
-	if (lineleft == 0) {
-		lineptr = line;
-		for (;;) {
-			c = xmon_readchar();
-			if (c == -1 || c == 4)
-				break;
-			if (c == '\r' || c == '\n') {
-				*lineptr++ = '\n';
-				xmon_putchar('\n');
-				break;
-			}
-			switch (c) {
-			case 0177:
-			case '\b':
-				if (lineptr > line) {
-					xmon_putchar('\b');
-					xmon_putchar(' ');
-					xmon_putchar('\b');
-					--lineptr;
-				}
-				break;
-			case 'U' & 0x1F:
-				while (lineptr > line) {
-					xmon_putchar('\b');
-					xmon_putchar(' ');
-					xmon_putchar('\b');
-					--lineptr;
-				}
-				break;
-			default:
-				if (lineptr >= &line[sizeof(line) - 1])
-					xmon_putchar('\a');
-				else {
-					xmon_putchar(c);
-					*lineptr++ = c;
-				}
-			}
-		}
-		lineleft = lineptr - line;
-		lineptr = line;
-	}
-	if (lineleft == 0)
-		return -1;
-	--lineleft;
-	return *lineptr++;
-}
-
-char *
-xmon_fgets(char *str, int nb, void *f)
-{
-	char *p;
-	int c;
-
-	for (p = str; p < str + nb - 1; ) {
-		c = xmon_getchar();
-		if (c == -1) {
-			if (p == str)
-				return NULL;
-			break;
-		}
-		*p++ = c;
-		if (c == '\n')
-			break;
-	}
-	*p = 0;
-	return str;
-}
diff --git a/arch/powerpc/xmon/start_8xx.c b/arch/powerpc/xmon/start_8xx.c
index a48bd594cf6..4c17b0486ad 100644
--- a/arch/powerpc/xmon/start_8xx.c
+++ b/arch/powerpc/xmon/start_8xx.c
@@ -15,273 +15,30 @@
 #include <asm/8xx_immap.h>
 #include <asm/mpc8xx.h>
 #include <asm/commproc.h>
+#include "nonstdio.h"
 
-extern void xmon_printf(const char *fmt, ...);
 extern int xmon_8xx_write(char *str, int nb);
 extern int xmon_8xx_read_poll(void);
 extern int xmon_8xx_read_char(void);
-void prom_drawhex(uint);
-void prom_drawstring(const char *str);
 
-static int use_screen = 1; /* default */
-
-#define TB_SPEED	25000000
-
-static inline unsigned int readtb(void)
-{
-	unsigned int ret;
-
-	asm volatile("mftb %0" : "=r" (ret) :);
-	return ret;
-}
-
-void buf_access(void)
-{
-}
-
-void
-xmon_map_scc(void)
+void xmon_map_scc(void)
 {
-
 	cpmp = (cpm8xx_t *)&(((immap_t *)IMAP_ADDR)->im_cpm);
-	use_screen = 0;
-	
-	prom_drawstring("xmon uses serial port\n");
 }
 
-static int scc_initialized = 0;
-
 void xmon_init_scc(void);
 
-int
-xmon_write(void *handle, void *ptr, int nb)
+int xmon_write(void *ptr, int nb)
 {
-	char *p = ptr;
-	int i, c, ct;
-
-	if (!scc_initialized)
-		xmon_init_scc();
-
 	return(xmon_8xx_write(ptr, nb));
 }
 
-int xmon_wants_key;
-
-int
-xmon_read(void *handle, void *ptr, int nb)
+int xmon_readchar(void)
 {
-	char *p = ptr;
-	int i;
-
-	if (!scc_initialized)
-		xmon_init_scc();
-
-	for (i = 0; i < nb; ++i) {
-		*p++ = xmon_8xx_read_char();
-	}
-	return i;
+	return xmon_8xx_read_char();
 }
 
-int
-xmon_read_poll(void)
+int xmon_read_poll(void)
 {
 	return(xmon_8xx_read_poll());
 }
-
-void
-xmon_init_scc()
-{
-	scc_initialized = 1;
-}
-
-#if 0
-extern int (*prom_entry)(void *);
-
-int
-xmon_exit(void)
-{
-    struct prom_args {
-	char *service;
-    } args;
-
-    for (;;) {
-	args.service = "exit";
-	(*prom_entry)(&args);
-    }
-}
-#endif
-
-void *xmon_stdin;
-void *xmon_stdout;
-void *xmon_stderr;
-
-void
-xmon_init(void)
-{
-}
-
-int
-xmon_putc(int c, void *f)
-{
-    char ch = c;
-
-    if (c == '\n')
-	xmon_putc('\r', f);
-    return xmon_write(f, &ch, 1) == 1? c: -1;
-}
-
-int
-xmon_putchar(int c)
-{
-    return xmon_putc(c, xmon_stdout);
-}
-
-int
-xmon_fputs(char *str, void *f)
-{
-    int n = strlen(str);
-
-    return xmon_write(f, str, n) == n? 0: -1;
-}
-
-int
-xmon_readchar(void)
-{
-    char ch;
-
-    for (;;) {
-	switch (xmon_read(xmon_stdin, &ch, 1)) {
-	case 1:
-	    return ch;
-	case -1:
-	    xmon_printf("read(stdin) returned -1\r\n", 0, 0);
-	    return -1;
-	}
-    }
-}
-
-static char line[256];
-static char *lineptr;
-static int lineleft;
-
-#if 0
-int xmon_expect(const char *str, unsigned int timeout)
-{
-	int c;
-	unsigned int t0;
-
-	timeout *= TB_SPEED;
-	t0 = readtb();
-	do {
-		lineptr = line;
-		for (;;) {
-			c = xmon_read_poll();
-			if (c == -1) {
-				if (readtb() - t0 > timeout)
-					return 0;
-				continue;
-			}
-			if (c == '\n')
-				break;
-			if (c != '\r' && lineptr < &line[sizeof(line) - 1])
-				*lineptr++ = c;
-		}
-		*lineptr = 0;
-	} while (strstr(line, str) == NULL);
-	return 1;
-}
-#endif
-
-int
-xmon_getchar(void)
-{
-    int c;
-
-    if (lineleft == 0) {
-	lineptr = line;
-	for (;;) {
-	    c = xmon_readchar();
-	    if (c == -1 || c == 4)
-		break;
-	    if (c == '\r' || c == '\n') {
-		*lineptr++ = '\n';
-		xmon_putchar('\n');
-		break;
-	    }
-	    switch (c) {
-	    case 0177:
-	    case '\b':
-		if (lineptr > line) {
-		    xmon_putchar('\b');
-		    xmon_putchar(' ');
-		    xmon_putchar('\b');
-		    --lineptr;
-		}
-		break;
-	    case 'U' & 0x1F:
-		while (lineptr > line) {
-		    xmon_putchar('\b');
-		    xmon_putchar(' ');
-		    xmon_putchar('\b');
-		    --lineptr;
-		}
-		break;
-	    default:
-		if (lineptr >= &line[sizeof(line) - 1])
-		    xmon_putchar('\a');
-		else {
-		    xmon_putchar(c);
-		    *lineptr++ = c;
-		}
-	    }
-	}
-	lineleft = lineptr - line;
-	lineptr = line;
-    }
-    if (lineleft == 0)
-	return -1;
-    --lineleft;
-    return *lineptr++;
-}
-
-char *
-xmon_fgets(char *str, int nb, void *f)
-{
-    char *p;
-    int c;
-
-    for (p = str; p < str + nb - 1; ) {
-	c = xmon_getchar();
-	if (c == -1) {
-	    if (p == str)
-		return 0;
-	    break;
-	}
-	*p++ = c;
-	if (c == '\n')
-	    break;
-    }
-    *p = 0;
-    return str;
-}
-
-void
-prom_drawhex(uint val)
-{
-	unsigned char buf[10];
-
-	int i;
-	for (i = 7;  i >= 0;  i--)
-	{
-		buf[i] = "0123456789abcdef"[val & 0x0f];
-		val >>= 4;
-	}
-	buf[8] = '\0';
-	xmon_fputs(buf, xmon_stdout);
-}
-
-void
-prom_drawstring(const char *str)
-{
-	xmon_fputs(str, xmon_stdout);
-}
diff --git a/arch/powerpc/xmon/subr_prf.c b/arch/powerpc/xmon/subr_prf.c
deleted file mode 100644
index b48738c6dd3..00000000000
--- a/arch/powerpc/xmon/subr_prf.c
+++ /dev/null
@@ -1,54 +0,0 @@
-/*
- * Written by Cort Dougan to replace the version originally used
- * by Paul Mackerras, which came from NetBSD and thus had copyright
- * conflicts with Linux.
- *
- * This file makes liberal use of the standard linux utility
- * routines to reduce the size of the binary.  We assume we can
- * trust some parts of Linux inside the debugger.
- *   -- Cort (cort@cs.nmt.edu)
- *
- * Copyright (C) 1999 Cort Dougan.
- *
- *      This program is free software; you can redistribute it and/or
- *      modify it under the terms of the GNU General Public License
- *      as published by the Free Software Foundation; either version
- *      2 of the License, or (at your option) any later version.
- */
-
-#include <linux/kernel.h>
-#include <linux/string.h>
-#include <linux/module.h>
-#include <stdarg.h>
-#include "nonstdio.h"
-
-extern int xmon_write(void *, void *, int);
-
-void xmon_vfprintf(void *f, const char *fmt, va_list ap)
-{
-	static char xmon_buf[2048];
-	int n;
-
-	n = vsprintf(xmon_buf, fmt, ap);
-	xmon_write(f, xmon_buf, n);
-}
-
-void xmon_printf(const char *fmt, ...)
-{
-	va_list ap;
-
-	va_start(ap, fmt);
-	xmon_vfprintf(stdout, fmt, ap);
-	va_end(ap);
-}
-EXPORT_SYMBOL(xmon_printf);
-
-void xmon_fprintf(void *f, const char *fmt, ...)
-{
-	va_list ap;
-
-	va_start(ap, fmt);
-	xmon_vfprintf(f, fmt, ap);
-	va_end(ap);
-}
-
diff --git a/arch/powerpc/xmon/xmon.c b/arch/powerpc/xmon/xmon.c
index 1124f114620..cfcb2a56d66 100644
--- a/arch/powerpc/xmon/xmon.c
+++ b/arch/powerpc/xmon/xmon.c
@@ -1,7 +1,7 @@
 /*
  * Routines providing a simple monitor for use on the PowerMac.
  *
- * Copyright (C) 1996 Paul Mackerras.
+ * Copyright (C) 1996-2005 Paul Mackerras.
  *
  *      This program is free software; you can redistribute it and/or
  *      modify it under the terms of the GNU General Public License
@@ -18,6 +18,7 @@
 #include <linux/kallsyms.h>
 #include <linux/cpumask.h>
 #include <linux/module.h>
+#include <linux/sysrq.h>
 
 #include <asm/ptrace.h>
 #include <asm/string.h>
@@ -144,15 +145,10 @@ static void xmon_print_symbol(unsigned long address, const char *mid,
 static const char *getvecname(unsigned long vec);
 
 extern int print_insn_powerpc(unsigned long, unsigned long, int);
-extern void printf(const char *fmt, ...);
-extern void xmon_vfprintf(void *f, const char *fmt, va_list ap);
-extern int xmon_putc(int c, void *f);
-extern int putchar(int ch);
 
 extern void xmon_enter(void);
 extern void xmon_leave(void);
 
-extern int xmon_read_poll(void);
 extern long setjmp(long *);
 extern void longjmp(long *, long);
 extern void xmon_save_regs(struct pt_regs *);
@@ -748,7 +744,6 @@ cmds(struct pt_regs *excp)
 		printf("%x:", smp_processor_id());
 #endif /* CONFIG_SMP */
 		printf("mon> ");
-		fflush(stdout);
 		flush_input();
 		termch = 0;
 		cmd = skipbl();
@@ -1797,7 +1792,7 @@ memex(void)
 	for(;;){
 		if (!mnoread)
 			n = mread(adrs, val, size);
-		printf("%.16x%c", adrs, brev? 'r': ' ');
+		printf(REG"%c", adrs, brev? 'r': ' ');
 		if (!mnoread) {
 			if (brev)
 				byterev(val, size);
@@ -1976,17 +1971,18 @@ prdump(unsigned long adrs, long ndump)
 		nr = mread(adrs, temp, r);
 		adrs += nr;
 		for (m = 0; m < r; ++m) {
-		        if ((m & 7) == 0 && m > 0)
-			    putchar(' ');
+		        if ((m & (sizeof(long) - 1)) == 0 && m > 0)
+				putchar(' ');
 			if (m < nr)
 				printf("%.2x", temp[m]);
 			else
 				printf("%s", fault_chars[fault_type]);
 		}
-		if (m <= 8)
-			printf(" ");
-		for (; m < 16; ++m)
+		for (; m < 16; ++m) {
+		        if ((m & (sizeof(long) - 1)) == 0)
+				putchar(' ');
 			printf("  ");
+		}
 		printf("  |");
 		for (m = 0; m < r; ++m) {
 			if (m < nr) {
@@ -2151,7 +2147,6 @@ memzcan(void)
 		ok = mread(a, &v, 1);
 		if (ok && !ook) {
 			printf("%.8x .. ", a);
-			fflush(stdout);
 		} else if (!ok && ook)
 			printf("%.8x\n", a - mskip);
 		ook = ok;
@@ -2372,7 +2367,7 @@ int
 inchar(void)
 {
 	if (lineptr == NULL || *lineptr == 0) {
-		if (fgets(line, sizeof(line), stdin) == NULL) {
+		if (xmon_gets(line, sizeof(line)) == NULL) {
 			lineptr = NULL;
 			return EOF;
 		}
@@ -2526,4 +2521,29 @@ void xmon_init(int enable)
 		__debugger_dabr_match = NULL;
 		__debugger_fault_handler = NULL;
 	}
+	xmon_map_scc();
+}
+
+#ifdef CONFIG_MAGIC_SYSRQ
+static void sysrq_handle_xmon(int key, struct pt_regs *pt_regs,
+			      struct tty_struct *tty) 
+{
+	/* ensure xmon is enabled */
+	xmon_init(1);
+	debugger(pt_regs);
+}
+
+static struct sysrq_key_op sysrq_xmon_op = 
+{
+	.handler =	sysrq_handle_xmon,
+	.help_msg =	"Xmon",
+	.action_msg =	"Entering xmon",
+};
+
+static int __init setup_xmon_sysrq(void)
+{
+	register_sysrq_key('x', &sysrq_xmon_op);
+	return 0;
 }
+__initcall(setup_xmon_sysrq);
+#endif /* CONFIG_MAGIC_SYSRQ */